Chapter 6. Nutrition.

CH A PTER 6
NUTRITION
L INTRODUCTION
Table 6-1 lists the elements Table 6-1. Elements commonly considered as essential for plant growth. Ionic forms
essential for normal growth of available to plants are suggested. [Adapted from Salisbury and Ross, 1985, and
most plant species. In Brady, 1974]. Note that the common essential elements carbon, hydrogen and oxygen
hydroponics, these are usually are not considered. Atomic weights rounded to nearest whole numberf Forms
the only elements considered. available to plants are usually combined with water.
The micronutrients are found in Atomic Concentra
micromolar (pM) concentrations Chemical Forms weight of tion
(0 . 0 0 0 0 0 1 mole f^), whereas the Element symbol available to element in dry tissue
macronutrients are usually re plants (ppm)
quired in concentrations of Micronutrients
millimoles (mM) (0.001 mole Nickel^ Ni Ni^* 59 0.06
or 1 0 to 1 0 0 0 times the
Molybedenum Mo Mo0 4 =’- 96 0 .1
required concentration range of
Copper Cu Cu^ Cu'* 64 6
micronutrients. The relative
abundance of these essential Zinc Zn Zvi-* 65 2 0
elements, expressed as parts per Manganese Mn Mn^^ Mn’^ Mn“-" 55 50
million (ppm), may be noted as Boron B H3 B0 3 °, B0 3 ^- 11 20
the average concentrations Iron Fe Fe^^ Fe’* 56 100
found in plant tissue (Table 6 - Chlorine Cl cr 35 100
1). Under the heading of forms Macronutrients
available to plants, one notes
Sulfur S 8 0 4 ’- 32 1000
that the essential elements occur
Phosphorous P H 2 P0 4 -, HP0 4 ’-, 31 2000
in ionic forms with positive or
P0 4 ’-
negative electrical charge. In a
Magnesium Mg Mg’* 24 2000
bulk nutrient solution, electrical
neutrality is required, meaning Calcium Ca Ca’* 40 5000
that the total positively charged Potassium K K* 39 10000
cations must balance the Nitrogen N N0 3 ‘, NH4 *, NOj- 14 15000
negatively charged anions, and it ^Many of these elements may be found in other combinations in solution, or
is in the ionic form that most as neutral, undissociated molecules.
nutrients are absorbed [Hiatt and 2 From Kochian [1991]; Welch et al. [1991].
Leggett, 1974]. While there can
be direct exchange o f absorbed
ions between soil particles and
the root system [Barber, 1962], most nutrient uptake is from the soil solution [Lindsay, 1972b]. Crops require
greater amounts of nutrients than the soil solution contains at any one time [Dean, 1957]. Hydroponic systems
are an exception.
As will be noted, there may be other chemical combinations with the essential elements in solution. Whether
or not these also can be taken up by plants remains to be decided. For example, both Fe^^ and Fe^^ can react with
water (hydrolyze) so that each forms five different compounds or oxides —the latter largely determining iron
concentration in solution. Fe^^ can also react with other elements such as chloride, bromide, fluoride, nitrate,
sulfate and phosphate to form as many as 13 different complexes, some neutral, others with varying electrical
charge. These do not include the various reduction-oxidation reactions that can occur with iron [Lindsay, 1979].
387
388
Table 6-2. Other elements often found in higher plants. Plant response has been observed with
some of these but essentiality not proven. Others are essential to mammals or happen to be
present in the soil solution. Formulae to which a superscript o is attached are undissociated,
neutral molecules [From Lindsay's modified data base for MESfTEQA2/PRODEFA2, Allison
et al., 1991]‘. “Forms in solution” are usually combined with one or more water molecules.
Element Chemical Forms in solution available to Atomic

symbol plants weight
Aluminum Al Al'* 27
Antimony Sb Sb(OH),- 122
Barium Ba Ba'* 137
Bromine Br Br 80
Cadmium Cd ed'* 112
Chromium Cr Cr'*, Cr(OH)2*, CrO/' 52
Cobalt Co Co'*, Co'* 59
Fluorine F F- 19
Iodine I I 127
Lead Pb Pb'* 207
Lithium Li Li* 7
Mercury Hg 201
Rubidium Rb Rb* 85
Selenium Se HSe-, HSeOj-, Se0 4 '- 79
Silicon Si H3SÌ0 4 -, H2SÌ0 4 '-, HSÌ0 4 '-, Si0 4 ^-, 28
H4SÌO4“
Silver Ag Ag^ 108
Sodium Na Na* 23
Strontium Sr Sr'* 88
Thallium T1 Tl* 204
Uranium U U'*, U^*, U O /, UO2'* 238
Vanadium V V'*, V'*, VO'*, VO2* 51
‘ These elements may be found in other combinations in solution in ionic form or as
neutral molecules.
The common practice of referring to fertilizer application as the use of “phosphorus,” “potassium,” or
“nitrogen” is not correct, though one will find this the most common procedure in American horticultural
literature. Not one of these three macronutrients exists as elemental N, P, or K in nature. Nitrogen, as it exists
as N 2 in the air, is unavailable to plants. To be taken up by plants, nitrogen must be in the nitrate (NO3 ),
ammonium (N H /), or nitrite (NO2 ) forms. The application of “200 ppm N,” as recommended, is incomplete and
confusing. Furthermore, fertilizer analyses list the major forms (N, P, and K) as percent N, P2 O 5 , and K 2 O
present in the mixture. The percent contents listed on a fertilizer bag do not have a direct relationship to practical
plant nutrition. The statement of actual percentage N, P, and K by most authors merely confuses the issue.
A second point to be made is that the essential forms given in Table 6-1 can be found in most soils with
many other combinations as presented in Table 6-2. If the element is present in solution, it will usually be found
in the plant, whether or not the particular substance performs any useful function in plant metabolism [Salisbury
and Ross, 1985]. Essentialities for those in Table 6-2 have not been proven, but there have been many reports
on significant growth responses to silica, selenium, and sodium [e.g., Salisbury and Ross, 1985; Adams, 1987;
Welch et al., 1991]. Bear classified selenium as a toxic element in 1957 -which it is often to animals feeding
upon Se accumulators in semiarid regions of the U.S. On the other hand, Adriano et al. [1987] cite the literature
to the effect that selenium deficiencies occur in New Zealand and Scandinavian countries. In Finland, Se is added
389
to fertilizer formulations, Adriano et al. also mention titanium, nickel, and chromium as showing beneficial
effects in plant growth. The latter may be due to interactions of these metals with essential elements. Foy [1992],
reviewing the literature on factors limiting plant growth, mentioned yield responses to silicon application by sugar
cane, rice, horsetail, cucumber, and tomato. Silicon can also be related in some parts of the world to a high
incidence of esophageal cancer or a protective mechanism in plants against grazing. In regions of high rainfall,
soil acidification leads to clay mineral decomposition and loss of soluble Si from the soil profile. Asher [1991]
stated that more than half of the elements in the Periodic Table are known to occur in plant tissues. He included
so-called “beneficial” elements such as sodium, aluminum, silicon, nickel, cobalt, lanthanum and cerium.
Cobalt is required by bacteria for nitrogen fixation in root nodules on legumes. Nickel is necessary in the
enzyme that breaks down urea to CO 2 and N H /. Sodium, nickel, vanadium, fluorine, chromium, tin, iodine,
cobalt, and selenium are required by higher animals. It may be that the amounts of some forms in Table 6-2 are
necessary in concentrations too low to be detected by existing analytical methods and equipment. In recent years,
on the other hand, many of these elements have been considered as waste hazards in the biosphere. With few
exceptions, none of these elements have been adequately studied in greenhouse production. One cannot fmd any
mention in the horticultural literature of the use of the elements listed in Table 6-2 in hydroponic systems as a
regular practice. In this chapter, the necessity to deal with chemical terminology will be required if improved
techniques are to be developed along with practical application. Considerable attention will be given to
terminology and units employed along with spéciation (composition) of substrate solutions, particularly
concerning soilless culture and hydroponics. We are rapidly approaching the day when one can predict solution
composition, depending upon water quality, plus the other factors that influence solution makeup. Also note that
existing recommendations are empirically determined for the conditions of the local environment -especially
with respect to soil type and weathering characteristics.
It is my purpose in this chapter to present a different approach to plant nutrition from that found in the
present technical literature. Empirically obtained data are useful if the appropriate caveats are kept in mind. Note
the extensive citing of information provided by W.L. Lindsay and his students obtained over several years. Use
of this information requires terminology and units not often found in the applied literature. Nevertheless,
adequate nutritional control in greenhouse production represents one of the greatest opportunities for significant
advances in the future.
II. UNITS AND TERM INOLOGY
As noted above, recommendations for fertilizer application are subject to confusion as amounts may be ex
pressed in different chemical forms at different times. Proprietary, complete analysis fertilizers are readily
available and easy to use. Recommendations may suggest so many ppm, mg per unit area, or volume, of a
substrate for a fertilizer. At the risk of redundancy, the analysis required on a bag provides percentages of N,
P 2 O 5 and K 2 O. It need not mention the form in which the elements are present in the mixture. Some authors
convert the values to elemental form as N, P, and K. For example, a 20-20-20 soluble fertilizer contains 20% N,
8 .8 % P, and 16.6% K when converted by using molecular weight factors listed in Table 6-3. Unfortunately, none
of these elemental forms are available to plants, nor is there any attention to the requirement for electrical
neutrality. The possibility of conversion to other forms in the soil or hydroponic solution is not often considered.
Seldom is any attention paid to the presence of solids that may dissolve or precipitate the added nutrients in a
way that makes them unavailable, or completely change the chemistry of the substrate. We will see that the
chemical forms (stoichiometry) of the essential nutrients in solution can markedly alter growth. Many individuals
continuously refer to ammonium nitrogen (N H /) and nitrate nitrogen (NO3 ). This, I think, is an unnecessary
redundancy when one could simply say ammonium or nitrate and be done with it. One may also note that few
authors, if any, pay attention to the forms in which any element such as those listed in Tables 6-1 and 6-2 are
present in substrates and taken up by plants. Carbon in its various combinations HC 0 3 ', H2 C 0 3 °), for
example, will invariably be present in soils or hydroponic solutions despite the grower's wishes in the matter. For
these reasons, and others, I have chosen to review the basic chemical methods for stating concentrations, activi
ties, acidities-alkalinlties, and oxidation-reductions, and to use this terminology through the remainder of the
chapter.
390 Chapter 6, Nutrition
Table 6-3. Factors for converting chemical forms. Multiplication of the amount in the left column
(ppm, kg, g, etc.) by the value in the center will provide the equivalent of the material in the right
column.
Multiply this material By this number To obtain this equivalent amount

Ca^^ (calcium) 1.399 CaO (calcium oxide, lime)
Ca'^ 2.496 CaC0 3 (calcium carbonate, calcite)
CaO (lime) 0.715 Ca'^
CaO 1.784 CaC0 3
CaCOj 0.400 Ca'^
(potassium) 1.204 K2O (potash)
K2O (potash) 0.830
KCl (muriate of potash) 0.632 K2O
K 2O 1.583 KCl (potassium chloride, muriate of
potash)
K2SO4 (potassium sulfate) 0.541 K2O
K2O 1.849 K2SO4
Mg^^ (magnesium) 1.658 MgO (magnesium oxide, periclase)
MgO 0.603 Mg2"
MgS 0 4 (magnesium sulfate) 0.335 MgO
MgO 2.981 MgS04
MgO 2.092 MgC0 3 (calcium carbonate,
magnesite)
N (nitrogen) 4.425 N 0 3 ‘ (nitrate)
N 1.286 NH 4^ (ammonium)
N 6.067 NaN 0 3 (sodium nitrate)
N 7.218 KNO3 (potassium nitrate)
N 4.717 (NH4)2S0 4 (ammonium sulfate)
N0 3 - 0.226 N
NH4^ 0.778 N
(NH4)2S0 4 0 .2 1 2 N
NaN 0 3 0.165 N
P (phosphorus) 2.288 P2O5 (phosphorus pentoxide)
P2O5 0.437 P
H2P0 4 ‘ (orthophosphate) 0.320 P
P 3.131 H2P0 4 -
S (sulfur) 2.994 S0 4 ^-
S0 4 ^- 0.334 s
A. CONCENTRATION
There are a number of ways to express the amount of substance in a solvent, based upon some fundamental
principles of physics and chemistry [e.g., Sienko and Plane, 1957; Pétrucci, 1977; Crockford and Knight, 1964]:
1. M olarity (M)
This is the amount of solute per unit volume of solvent (in our case water). It is the term that will be used
391
Table 6-4. Ionic species often found in soil or hydroponic solutions with formula weights. Values are deliberately rounded
to the nearest whole number. These represent a selection from Lindsay [1979]. The list is not intended to be inclusive since
actual presence depends upon several factors. Compounds followed by a superscript zero (°) are neutral, undissociated
molecules. These may be combined with one or more water molecules.____________________________________ _______
Compound Formula Compound Formula Compound Formula
weight weight weight
Aluminum 27 Hydroxide OH' 17 Nitrogen N 0 3 ' 62
A1 (0 H)3 ° 78 Iron F e^ Fe^^ 56 NH 4^ 18
A1(0 H)2 " 61 Fe(OH)2^ 90 N0 2 ‘ 46
AlOtf^ 44 Fe(OH)3° 107 Phosphorus
PO4'- 95
Al2(OH)2^" 88 FeOtf^ 73 H3 P0 4 ° 98
A 1F3° 84 FeF'^ 75 HP0 4 ^' 96
AIF^^ 46 FeCP 92 H2 PO4 - 97
A 1F2^ 65 FeHP0 4 ^^ 88 H2 P2 O/- 176
A1 S0 4 ^ 60 FeOH^ 73 Potassium 39
A1(S0 4 )2- 221 FeS 0 4 ° 153 KS 0 4 ' 136
Boron Fe(OH)2° 90 Silicon
H3B 0 3 ° 62 H4Si04° 96
B 0 3 '- 59 FeH2P 04^ 153 Sodium Na^ 23
Calcium Ca^^ 40 Magnesium Mg^^ 24 NaSOF 119
CaS0 4 ° 137 MgS0 4 ‘^ 121 Sulfur
SO42- 96
136 MgHP0 4 ° 120 Zinc Zn^^ 65
CaP0 4 - 135 MgC03° 84 ZnOH^ 82
Copper Cu^ Cu^^ 64 Manganese Mn^^, Mn^^, 55 Zn(OH)2 ° 99
Mn^^
CuOH^ 81 MnOH-" 72 161
Cu(OH)2 ° 98 Mn2 (OH)3 ^ 161
CuS0 4 ° 161 Mn2 0 H^-^ 127
Carbon MnOH^^ 72
60
62 Molybdenum
M0 O42- 160
HC0 3 - 61 HMo0 4 ' 161
Hydrogen 1 H2 M0 O4 " 162
in this text. A 1 M solution contains one gram-molecular weight o f solute (expressed as grams) per liter of
solution. Formula weights of some more common substances with which the grower deals are given in Tables
6-4 and 6-5. Table 6-4 is a selection of those compounds and ionic forms that are commonly present in the
substrate solution. Methods for dealing with these will be discussed later. Table 6-5 lists those high analysis
chemicals used for fertilization purposes. Other sources will be noted later (Tables 6-27, 6-28). Since concentra
tions of macronutrients are in the millimole (mM) range, the formula weights are taken as milligrams (mg) rather
than grams (1 mg = 0.001 g). Thus, 80 mg of ammonium nitrate (NFÎ4NO3) dissolved in water made up to H
will provide a concentration of 1 mM NH4NO3.
*The SI convention gives the spelling of “liter” as “litre” and “meter” as “metre”. The American spelling
will be used here.
Table 6-5. Chemical fertilizers often used in greenhouse production and in hydroponics^ for supplying macronutrients. U)
Compound Formula® Analysis^ Formula Equivalent Effect on Solubility Other to
______________ _______ _______ w eighf________w eighf acidity (g 100 mH) materials'^ O
Aluminum sulfate Al2(S 0 4 )3-1 8 H 2 0 0 -0 - 0 667 222 Very acid Soluble 14% S
CD
Ammonium chloride NH 4CI 25-0-0 54 54 Acid 40
. . .
ON
Ammonium nitrate NH 4NO 3 34-0-0 80 80 Acid 118
'iz:
Ammonium polyphosphate 15-62-0 232 — Very acid Soluble . . .
(NH4)3H4P202
Ammonium sulfate (NH4)2S04 2 0 -0 - 0 132 66 Very acid 71 24% S
Basic slag Ca4P2 0 9 0 -2 1 - 0 366 . . . Basic 0+ 44% Ca
Calcium-ammonium-nitrate Used in place of NH 4NO 3 which can be explosive. 15.5 % N or higher, variable.
Calcium carbonate (Limestone, CaC 0 3 0 -0 - 0 100 50 Basic 0 +" 40% Ca
Calcite)
Calcium cyanamide CaCN 2 2 0 -0 - 0 80 Basic Decomposes —
Calcium hydroxide (hydrated lime) Ca(OH )2 0 -0 - 0 74 37 Basic 0+ 60-80%
Ca
Calcium metaphosphate 0-64-0 198 — Basic 0+ 20% Ca
Ca(P03)2
Calcium monophosphate Ca(H 2P 0 4 )2-H2 0 0-56-0 252 126 Basic Decomposes 16% Ca
Calcium nitrate Ca(N03)2-4H20 15-0-0 236 118 Basic 102 17% Ca
Calcium sulfate (gypsum) CaS 0 4 ’2 H 2 0 0 -0 - 0 172 86 Neutral 0+ 23% Ca
19% S
Dicalcium phosphate dihydrate CaHP 0 4 ' 2 H 2 0 0-41-0 172 — Basic 0+ 23% Ca
(DCPD, brushite)
Diammonium phosphate (NH4)2HP04 21-53-0 132 66 Acid 43
(Diammonphos, DMP)
Dolomite (dolomitic limestone) 0 -0 - 0 184 — Basic 0+ 22% Ca
CaMg(C03)2
13% Mg
Magnesium-ammonium-phosphate MgNH 4P 0 4 -H2 0 8-46-0 155 -- Acid 0+ 13% Mg
Magnesium sulfate (epsom salts) MgS04-7H20 0 -0 - 0 247 123 Neutral 71 10% Mg
13% S
Magnesium nitrate Mg(N 0 3 )2-6 H 2 0 1 1 -0 - 0 256 128 Neutral 42 10% Mg
Monoammonium phosphate NH 4H 2PO 4 11-48-0 115 115 Acid 23 1.4% Ca
(Ammophos A, MAP) 2.6% S
Nitric acid HNO 3 2 2 -0 - 0 63 63 Very acid Soluble —
Phosphoric acid H 3PO 4 0-52-0 98 33 Very acid 548
Potassium carbonate K 2CO 3 0 -0 - 6 6 138 69 Basic 112 . . .
Compound Formula® Analysis^ Formula Equivalent Effect on Solubility Other
weight*^ w eighf acidity (g 100 ml materials**
Potassium chloride (muriate of KCl 0-0-62 75 75 Neutral 35 47% Cl
potash)
Potassium diphosphate 0-41-54 174 87 Basic 167
Potassium metaphosphate KPO3 0-57-39 118 Acid 0 +
Potassium monophosphate KH2 PO4 8-53-34 120 120 Basic 33
Potassium nitrate (saltpeter) KNO3 13-0-44 101 101 Basic 13
Potassium sulfate K2SO4 0-0-53 174 87 Neutral 7 18%S
Pyrophosphoric acid*^ H4 P2 O2 0-80-0 178 45 Acid 709
Sodium nitrate NaN03 16-0-0 85 85 Basic 73 27% Na
Superphosphate CaH4(P04)2 0- 20-0 234 78 Neutral 2 18%Ca
(DCP) 12% S
Superphosphoric acid H3PO4+H4P2O7 0-76-0 Will solidify at low temperatures.
Sulfur S 0- 0-0 32 — Acid 0+
Treble superphosphate® CaH4(P04)2 0-42-0 234 78 Neutral 2 12% Ca
Urea C0 (NH2 ) 2 45-0-0 60 30 Acid 78
®Formulae may differ due to manufacturing process, addition or removal of water, etc.
**Percentage N, P2 O5 , and K2 O.
Formula and equivalent weights rounded to nearest whole number.
**These will vary with source, manufacturing process, etc.
®For practical use in hydroponics, solubility should exceed 10 g (100 ml)'*. Solubility at 0 C, with one or two exceptions.
^Continued removal of water can increase P2 O5 in excess o f 85% and can be combined with ammonium and potassium.
®Contains no gypsum._____________________________________________________________________________________
U)
u>
394 Chapter 6 , Nutrition
2. Molality (m)
This is the amount of solute per unit mass of solvent. A i m solution contains one gram-molecular weight
of solute in 1 kilogram (kg) of water. In both molarity and molality, a 1 M or 1 m solution of a substance that
does not dissociate when dissolved will provide 6.02 x 10^^ molecules or particles. Such a substance could be
sugar (sucrose) which has a formula weight of 342. If 342 mg of sucrose are dissolved and made up to H , the
concentration is 1 mM and contains 6.02 x 10^° particles. The same applies to a 1 mM solution in 1 kg of water.
Molality is unaffected by changes in temperature, whereas a rise in temperature will cause an increase in
volume, changing molarity. Molarity (moles of solute per H of solution) is a useful, very common unit but is not
recommended by the international unit convention. Molality (moles of substance per kilogram of solvent) is a
legitimate SI (Le Système International d'Unités) unit. The analogous SI concentration unit is mole per cubic
meter (M m‘^). One M m‘^ = 1 M per 1000 ^ = 1 mM The multiplicity of units makes it difficult to compare
results of various authorities. In this chapter, the author employs mole (M f*), meaning moles per liter, commonly
expressed as a logarithm to base ten.
Because the temperature range in greenhouses is narrow, and the concentrations of substances in the soil
solution are invariably in the mM range or lower, there is not much to choose between molality and molarity.
The number of particles in solution directly bears on the colligative properties of solution as briefly discussed
in Chapter 5. Thus, as the particle number in solution increases, the freezing point will be lowered, boiling point
elevated, osmotic pressure increased, and vapor pressure above the solution decreased. However, a 5 mM
solution of sucrose will have the same colligative properties as a 5 mM solution of formaldehyde, fructose,
glycerol, or naphthalene. They all have the same number of molecules in solution.
However, one notes that most of the chemical forms listed in Tables 6-1 through 6-4 are ionic species with
varying electrical charge. For example, potassium nitrate, a very common fertilizer used in greenhouse production
is highly soluble and dissociates nearly completely in solution. That is, the dissociation reaction goes to the right
completely in very dilute solutions:
KNO^ ^ + NO^ (6. 1)
If one wishes a common recommendation of 6 mM KNO 3 to be delivered in the irrigation supply, then (101)(6)
= 606 mg K N O 3 are to be dissolved in each H. Assuming complete dissociation, for each molecule there will be
two ionic particles, or double the number of particles, and each species will have a concentration of 6 mM. In
very dilute solutions, the effect on colligative properties will be to double, for example, the freezing point
depression.
Another common chemical is calcium nitrate, which dissociates:
Ca(NO0MH,O ^ + 2AO^ + A H p (6 .2 )
There are now three particles for each molecule of Ca(N 0 3 )2 ‘4 H2 0 , one a positive calcium ion and two negative
nitrate ions. The attached water molecules do not significantly change the concentration of water in 1 H which
is about 55.5 M. If one needs a 6 mM concentration of calcium nitrate, then (236)(6) = 1416 mg will be
required. But, there will be 6 mM of Ca^^ and (2)(6) = 12 mM of N 0 3 ’ in solution, assuming complete dissocia
tion of the 1416 mg calcium nitrate. The effects of electrical charges, concentration, and possible hydrolysis of
the ions (attached water molecules), etc. will be discussed shortly.
Often, fertilizer recommendations are given in parts per million (ppm), which are equivalent to milligrams
per Í or micrograms per gram (pg g'*). The moles per H of the substance can be calculated by dividing the mg
dissolved in H by the formula weight. Unfortunately, one will find in the literature recommendations to apply
200 mg f* nitrogen (N) without any specification as to a chemical source. As elemental N does not exist, one
must convert the recommendation to its equivalent in a physically possible compound. This can be done, using
the conversion factors in Table 6-3. This means that if 200 ppm N is from a nitrate source, then (4.425)(200) =
395
885 mg NO 3 ', or 885/62 = 14.3 mM N 0 3 ’ (Table 6-3). If the grower uses ammonium nitrate, then (14.3)(80) =
1144 mg f ' NH 4 NO 3 (Table 6-5). The grower will also add 14.3 mM N H /. If he uses calcium nitrate, 7.2 mM
of Ca^^ will be added. If the recommendation is for an ammonium source, then (200)( 1.286) = 257 mg N H / or
257/18 = 14.3 mM N H / and 1144 mg of NH4 NO3 will also provide 14.3 mM N Q ‘. These results must be adjust
ed downward to avoid excessive fertilization. If the grower decides to use ammonium sulfate ((NH 4 )2 S0 4 ), he
will need (14.3)(132/2) = 944 mg , and 7.2 mM of 8 0 4 ^’ will be provided besides ammonium. Note the two
ammonium ions per molecule when dissociated, requires division of the formula weight by two. Otherwise, when
the salt dissolves, each molecule will provide two ammonium ions, and a 14.3 mM solution of (NH 4 )2 S0 4 would
have 28.6 mM NH 4".
One fact is immediately apparent. Fertilization with any element, N, K, or P, will require an appropriate
compound, and a complementary ion of opposite charge will also be added (i.e., KCl - + Cl', NH 4 H 2 PO 4 -
NH 4 ^ + H 2 P 0 4 ’, etc.). Most existing, practical publications seldom mention these physical realities in fertilizer
recommendations. Even in recent scientific literature, the reader will find the use of ppm or mg of the element
(e.g., N) in solution. Hopefully, if the author shows the ionic form in which the element is supplied, the actual
concentration of the ion can be calculated. For example, the concentration o f nitrate to which plants may be
exposed in the root substrate can be lower than 0.02 mg f ^ nitrogen. If it is assumed that the nitrate form is the
sole source, this may be converted to molarity units: (0.02 x 4.425)/62 = 0.00143 mM = 1.43 \xM = 0.00000143
M = 10'^^^ M. In some publications, the requirement for the reader to convert to a common unit for comparison
purposes is laborious.
B. ACTIVITY
As noted, many chemicals dissolved in a
3-"
substrate solution are most often in ionic form
with an electrical charge. Salts may exist in an ion
ized form, but behave dependently with each
other. Each positive ion will be surrounded by a
cluster of negative anions, and each negative ion
by a cluster in which positive cations predominate.
In an electrical field, the mobility of an ion will be
reduced as well as reducing the magnitude of the
colligative properties of the solution. Typically,
single ions such as aluminum (AP^), iron (Fe^^,
Fe^^), manganese (Mn^^), etc. may be hydrolyzed
as suggested in Fig. 6-1. The number of water
molecules attached to the ion will vary with the
hydrogen ion activity (pH). The effect is again to
reduce ion mobility. The water molecules are
commonly omitted in reaction formulae for greater
simplicity. Thus, each ion type in solution has a
total concentration based upon the amount of Fig. 6-1. The aluminum hexahydronium ion (A 1 (H2 0 )6 ^^). As
solute dissolved, and an “effective” concentration pH decreases, hydrogen protons are removed. Below pH 5.0,
called the “activity” that takes into account free AP^ is predominant in solution [C hem ical E quilibria in
interionic attractions. If activity is used, solution Soils, Lindsay, W.L., ©1979. Reprinted by permission from
properties can usually be predicted. The value is John Wiley & Sons, Inc.].
expressed in the same terms as concentration
(moles/liter), but the ratio between activity and concentration (activity coefficient, y) is always less than 1 . An
activity coefficient can be calculated from a knowledge of the ionic strength (p), valencies (number of electrical
charges on the ion), and effective size of the ion. Only in very dilute, or “ideal,” solutions does y approach I
where activity equals concentration. The activity coefficient can be below 0.002 for ions having concentrations
near 0.1 M.
As the result of the factors outlined above, most chemical equilibria are expressed as “activity” rather than
concentration. For example, the equilibrium reaction of calcite (CaC 0 3 ), a common mineral in alkaline soils is:
CaC03 + 2H^ - + CO^g) -f (6.3)
And the equilibrium constant, K°, of this reaction at 25 C is:
(C a^^)(C 0,(g))(H ,0) ,9.74 (6.4)

= 1 0
{C a C O ,){H y
The parentheses, ( ) , denote “activity.” As activities o f CaCOj and HjO are considered fixed, they can be set
equal to 1. If the activities o f the products divided by the reactants are given in logarithms to the base 10, the
activity of calcium becomes:
log,o(CaÔ = 9.74 - 2pH - log,,(CO ,(g)) or

(6.5)
pCa^* = 9 .1 A -2 p H -p C O ^
Lowercase “p” denotes that the value is a negative logarithm, and the concentration of CO 2 is given in
atmospheres. The important thing to note is that the activity of calcium in solution depends on both pH and CO 2
concentration, where the dissolving mineral is calcite.
While a grower is unlikely ever to use the complex equations for calculating activities and chemical
equilibria, these factors are utilized in computer programs for determining soil solution, thermodynamic equilibria
based upon soil type, acidity, and reduction-oxidation potentials. In hydroponic systems, such programs open the
way to control solution concentrations automatically as illustrated later in this chapter.
C. EQUIVALENCY
Equivalency in chemical terms refers to the equality of combining capacity or equal valence -the latter
referring to the number of positive or negative charges on an ion. A potassium ion (K^) has a valency of 1,
whereas a magnesium ion (Mg^^) has a valency of 2, and aluminum's valency (AF^) is 3. For compounds to have
equal combining capacity, the formula weight is divided by the valency to obtain equivalent weight, often called
the gram-equivalent weight (Table 6-5). One gram-atomic weight of hydrogen, or any other ion that will combine
with or displace this amount of hydrogen, is another definition of equivalent in chemical terms. For monovalent
ions such as Na^, K^, NH 4 ^ and C l', the equivalent weight and atomic weight are the same since they can react
with 1 ion. Divalent cations such as Ca^^ or Mg^^ can replace two ions.
If 80 g ammonium nitrate (NH4NO3) are dissolved in H , there will be 1 equivalent per H each o f NH4^ and
N 0 3 " since the ions in solution each have 1 charge. On the other hand, the formula weight o f calcium nitrate
(Ca(N03)2*4H20) must be divided by two to give an equivalent weight o f 118 g f L The Ca^^ ion has two charges
and there are two nitrate ions, each with a valency o f 1. One equivalent weight o f nitric acid (HNO3 ) will react
with 1 equivalent weight o f calcium carbonate (CaC0 3 ). As an equivalent is much too large, milliequivalents are
employed as with millimoles for concentration. Six meq KNO3 requires (6)(101) or 606 mg in H , and w ill
provide 6 meq and 6 meq NO3*. This is the same amount as required for 6 mM calculated previously.
However, the equivalent weight o f magnesium sulfate (MgS 0 4 -7 H 2 0 ) is 247/2 or 123 mg (Table 6-5). A 1 mM
solution would require 247 mg in H . A 1 mM solution o f magnesium sulfate, would contain 2 meq Mg^^ and
2 meq f ’ 8 0 4 ^'. Therefore, a 1 mM solution o f magnesium sulfate would not be “ equivalent” to a 1 mM solution
o f potassium nitrate (KNO3) as to their chemical combining capacity.
The advantage to employing milliequivalents (meq) is the ability to consider the essential elements in the
forms required by plants and to balance the solution electrically. Steiner, in the early 1960s, employed these units
to derive physically possible solutions [Steiner, 1961; 1966; 1968]. In the same decade, students of W.D. Holley
at Colorado State University used this system in detailed studies of carnation and rose nutrition [Green et al,
1971; 1973; Hartman, 1971; Hughes, 1975; Hughes and Hanan, 1977; etc.]. The results were later summarized
397
by Hanan in a series of papers published in the 1980s [Hanan, 1982a; b; 1984; 1988]. While other authors have
also employed equivalents, the system has not achieved widespread use in the industry, although manipulation
of raw water supplies is considerably simplified.
D. ACIDITY
Water has the properties of both an acid and a base. Water molecules can donate protons (¥ t) in a reaction
with a base such as ammonia and accept protons in a reaction with acids such as hydrochloric or acetic acids.
There are a few ions even in pure water, producing the equilibrium:
H ,0 ^ H ,0 ^ ^ OH- ( 6 6 )
where H 3 0 "îs called the hydronium ion. For simplicity, it is usually written as H^. The equilibrium constant of
this reaction is:
= 10"
(6.7)
The activity o f water in Eq. 6.7 is 1. In pure Table 6-6. Relationship between hydrogen ion concentration, pH,
and pOH.
water, the activity must equal the OH'
activity so that each must have a value of concentration pH pOH
1 0 '^ moles/liter to satisfy the equilibrium (moles/liter)
constant lO'*"^. If either or OH' activity is 1 1 0 14 Acid solutions
known, the other can be determined using 0.1 1 X 10' 1 13
Eq. 6.7. Whether a solution is acidic or basic
0.01 1 X 10-^ 2 12
depends upon which ion is in greater concen
0.001 1 X 10-' 3 11
tration. The addition of an acidic compound
will result in a greater concentration of 0.0001 1 X lO“* 4 10
hydrogen ions compared to hydroxide ions. 0.00001 1 X 10-' 5 9
Conversely, the relative proportions will 0.000001 1 X 10-^ 6 8
favor OH' if, for example, sodium hydroxide 0.0000001 1 X 10-' 7 7 Neutral point
(NaOH) is added to the solution. The acidity 0.00000001 1 X 10-* 8 6
or basicity o f a solution is described by 1 X 10-’ 9
0.000000001 5
giving the hydrogen ion activity that may be
0.0000000001 1 X lO-'“ 10 4
expressed in moles/liter or pH where:
0.000000000001 1 X 10" 12 2
0.00000000000001 1 X lO-'“ 14 0 Basic solution
p H = -lo g ( i/0 = log-
( 6 .8)
In practice, the activity and concentration of the ion are not usually very different, and the pH as calculated
from either will be almost identical. The relationship between concentration, pH, and pOH is presented in Table
6 -6 , which shows that as pH decreases by 1 unit, hydrogen ion concentration increases tenfold. At pH = 0, (H^)
= 1 mole .
Two examples of the effect of pH may be examined for the present. In Fig. 6-2, iron equilibria calculated
by Lindsay [1974] shows that the major cations in solution, Fe^^ and Fe^^, can be reduced to levels approaching
1 ion per f at pHs greater than 7 due to precipitation as iron oxide. Below activities of 10'^ moles (0.000001 M
or 1 fiM) to 10'^ Fe, there will be insufficient soluble iron in solution to meet plant requirements. Another effect
of pH on the major orthophosphate ions in solution under ideal conditions is provided in Fig. 6-3. At pH 7.2, the
ions H 2 PO 4 ' and H P O /' are equal in activity. As we shall see, pH has a major influence on ion availability of
essential nutrients with which a grower must deal for successful plant production. A marked change in pH
suggests a radical modification in soil environment. However, as stated by Moore [1974], plants will generally
do well over a wide pH range (5-7). Allaway [1957] gives a pH range of 4 to 9. Note, however, that Moore
published his review nearly 20 years after Allaway. The effects of pH are often indirect, affecting nutrient
availability. It is this relationship between pH and nutrient availability that justifies frequent pH measurements
and makes tables of pH preferences of plants often useful [Allaway, 1957].
E. OXIDATION-REDUCTION
In redox reactions, electrons (e') are exchanged. Oxidation of a substance occurs when electrons are lost,
such as in rusting of metals, the release of electrons from the zinc anode in a flashlight battery, or the lead anodes
in a car battery. Simultaneously, there must be reduction, or gain of electrons by a substance, such as occurs at
the negative cathodes in batteries. Examples of redox reactions common to soils are:
+e ' -- Fe'^^ (6.9)

F eÔ ^(m agnetite) + 8//^ + 2 e ' ( 6 . 10)
Mn 4 + + e ( 6 . 11)
MnOl~ + m * + 4 e ' ^ + 4H p ( 6 . 12)
2 +
Cu ^ Cu'" (6.13)
NO ^ + 2H^ + 2e- N O ~ + H p (6.14)
The oxidation-reduction potential, or redox potential, is a measure of the electron availability in a chemical
or biological system. Respiration by root systems and microorganisms is an oxidative process. There will be an
abundance of electrons, especially close to the root surface. This is coupled with a corresponding decrease in
oxygen concentration and an increase in CO 2 level as respiration goes on, and may be exacerbated by a high
moisture content (see Section 5.III).
The term “potential” is commonly used since the method employed is to measure voltage differences between
a platinum electrode inserted into the soil and a suitable reference electrode such as a calomel or silver-silver
chloride cell. Potentials (Eh) greater than 400 mV are characteristic of well-drained substrates, whereas in
flooded soils with a high carbon content. Eh can be below -180 mV -o r the environment is highly reduced. The
use of this measurement in explaining greenhouse conditions of the substrate has seldom been employed. There
have not been any publications on the subject as applied to greenhouse culture, to my knowledge, in the last 30
years. More recently, in the agronomic Hature, efforts by Lindsay, his students, and contemporaries [e.g., Lindsay,
1979; 1983; Lindsay and Schwab, 1982; Lindsay et al., 1989; Schwab and Lindsay, 1983a; b; Sajwan and
Lindsay, 1986; Boyle and Lindsay, 1986; Norvell, 1991, etc.) have been made to use redox data fully. The
influence of redox on trace element availability is particularly important. Rather than using Eh, a voltage, these
investigators preferred using the negative logarithm of electron activity (pe), which allows one to use M in
calculating equilibria. Potential measurements can be converted to pe by means of the formula:
EhjmV)
pe (6.15)
59.2
To fully designate the redox status, pH and pe are combined to provide a value for pe + pH. The limit on the
reduced side (flooded soils and high carbon) is determined by the equilibrium of hydrogen ions with electrons
to form hydrogen gas. Where the H^ activity is 1 M and the partial pressure of hydrogen gas is 1 atmosphere,
+ pH = 0. If the pH is 6.0, then pe = -6 . On the opposite side, in a fully oxidized condition at 1 atmosphere
partial pressure for oxygen, pe + pH = 20.78. Again, if the pH is 6.0, pe = 14.78. Most soils lie in the pe + pH
range of approximately 4 to 17.
It would be expected that most well-drained greenhouse soils and NET systems will be highly oxidized, or
399
at least pe + pH would exceed 10. Lindsay [1991]

suggested a range for field conditions of 1 2 to 16 for
well-aerated soils. The platinum electrode,
unfortunately, is highly unstable in oxidized soils. A
redox measurement of many greenhouse substrates is
likely to be unreliable. As will be noted in Section
6 .IV.A, however, the region close to an active root
system also has the highest microbial activity compared

with the bulk soil or solution. The uptake of oxygen and
its consequent lower concentration as it diffuses through
this highly active region, combined with increased CO 2
levels, are likely to significantly lower pe + pH.
Examples o f the combined effect of pH and redox on
solubility o f phosphorus and iron in solution are
presented in Fig. 6-4. It will be noted that greatest
availability of phosphorus and iron occurs under
reduced conditions at low pHs. According to Lindsay
[1991], if soil-Fe alone controlled iron solubility, pe +
pH would have to be less than 10 at pH 7.0, or 8.0 at pH
8.0 in order for Fe^^ activity to exceed 10'^ M required Fig. 6-2. Concentration of Fe^^ and total soluble
by plants [Schwab and Lindsay, 1989]. As will be noted iron in a solution in equilibrium with iron oxide and 0 . 2
later, some plant species can reduce redox levels in atmospheres O2 [Lindsay, 1974].
solution below pe + pH 7. They are “iron-
efficienf ’ plants. Provision of iron, particularly,
is a serious problem in greenhouse production
where pH values are likely to range from 5.5 to
7.5 and conditions favor an oxidized environ
ment.
Usually, the oxidation states of an element
have an important bearing on its mobility and
availability. Manganese, selenium, mercury,
arsenic, chromium, copper, and iron are
sensitive to redox potential in soils [Adriano et
al.,1987]. As most greenhouse production is
still in field soils, better distribution of
agronomic information for practical application
in the industry would be worthwhile. To the
best of my knowledge, examination of the
immediate root environment regarding redox
conditions in greenhouse substrates, or Fig. 6-3. The effect of pH on the distribution of major orthophos
phate ions in solution. At pH 7.2, 50% of phosphate ions are
hydroponic systems, has never been carried out.
H2 PO4 *and 50% are HP0 4 ^’ {C h em ical E qu ilibria in Soils, Lind
Despite the measurement difficulty, this is a say, ©1979. Reprinted by permission of John Wiley & Sons,
serious lack of basic information that could Inc.l.
prevent full automation of greenhouse
production.
F. ADSORPTION-DESORPTION IN SOILS
1. Intensity versus Capacity
There are two primary factors in nutrition that determine nutrient supply: 1) intensity and 2) capacity. The
first refers to the concentration of nutrients in the substrate solution. The second is the ability of the substrate to
replenish a particular nutrient as it is removed from the solution. Cation exchange capacity (CEC) is one measure
of the capacity factor or the ability to hold exchangeable cations. The effect of processes of weathering on
Fig. 6-4. The solubility of phosphorous (left) and iron (right) as influenced by pH and pe, where pH = -log(H^) and pe =
-log(e*). Adapted from Reddy and Patrick [1983] (With permission of the Amer. Soc. of Agronomy).
intensity and capacity factors were diagrammed by Lindsay in Fig. 6-5 for three hypothetical soil minerals, A,
B, and C. For most potting substrates in greenhouse culture, the duration of the crop is generally short with rapid
growth. The capacity of most substrates to replenish the substrate solution is similar to mineral C, or less.
Certainly in hydroponics, the only storage factor is bulk solution. According to Moore [1974], the potential
quantity of any ionic species is fixed, whereas in a soil system, additional ions can be brought into solution. Ions
can be “desorbed” from the soil matrix. Fertilization causes “adsorption” of the ions back onto the soil matrix.
The capacity ranges from the hydroponic extreme to production in highly organic field soils.
In Fig. 6-5, the secondary mineral A dissolves most rapidly and maintains solution concentration at a high
level, well above the critical level required by plants growing in it. Minerals B and C, being supersaturated, will
slowly precipitate until A disappears, and the solution concentration then drops to level b. B dissolves,
maintaining the concentration at level b, still well above the critical plant level, and mineral C continues to
precipitate. When, however, B disappears, C remains as the final weathering product and the soil solution drops
well below the critical level. Plants growing in this “poor” soil grow either slowly or not at all. Fertilization will
be required.
2. The Soil System

To appreciate the importance of the soil's capacity to store and release essential nutrients, we need some
background on the colloidal soil system. The makeup of clay and humus in a soil, as influenced by primary
minerals and climate, plus prior cultivation and fertilization practices, will determine the cultural procedures
available to the grower and set the basic fertility level of the soil. The variation in soils from one geographical
region to another is seldom, if ever, discussed in applied greenhouse practice. Warrick [1981] emphasized the
problem o f soil variability, with more than 10000 soil series defined in the U.S. alone. The recommendations
401
START = m 0
C r itic a l
AT SOME FUTURE TIME K
C ritic a l laval
I [L SoU m kiaral O
(TOTAL A ^ T l
L E A ST S C X .m L E
Fig. 6-5. The influence o f soil minerals A, B, and C on solubility and availability of nutrients in a substrate. [C hem ical
E qu ilibria in Soils, Lindsay, W.L., ©1979. Reprinted by permission of John Wiley & Sons, Inc.].
from research stations deal largely with local conditions of soil genesis, type, and climate. Unless the grower has
had direct experience in different climatic regions, he is usually unaware of what might happen if, for example,
he applies a recommendation obtained from experiments in a temperate humid region to an arid or tropical soil.
The results can be startling and often expensive.
The most active portions of the soil are those in colloidal state with two distinct types -clay minerals of
various kinds and decomposed organic matter or humus. Clays, composed of secondary minerals and amorphous
materials, differ from the components of sand and silt. The clays are products of weathering and are not found
in unaltered rocks [Dean, 1957]. Two groups of clays are recognized: silicate clays characteristic of temperate
regions and iron and aluminum hydrous oxide clays found in the semitropics and tropics. The hydrous oxides
are mostly compounds of iron and aluminum. They play a role in phosphate fixation, influencing phosphorous
availability to plants.
All clay particles, because of their small size, expose a large surface area, and in some clays, a large internal
surface may be present (Fig. 6 -6 ). Comments on this surface area were made in Chapter 5. The clay micelles
usually carry negative charges. Consequently, thousands of positively charged cations are attracted to each
colloid. Water molecules, carried by the adsorbed cations, are also present. For humid region clays, the ions in
order of their numbers are H^, AP^, and Ca^^, followed by Mg^"^, and finally and Na^. For well-drained, arid
and semiarid soils, the order is usually Ca^^ and Mg^^, Na^ and K^, and last. The vertical dashed line in Fig.
6 - 6 arbitrarily distinguishes the inner adsorbed ionic layer versus the outer ions mostly free in solution.
When drainage is reduced in arid regions, alkaline salts accumulate and adsorbed sodium ions are likely to
be dominant, whereas in humid regions an aluminum-hydrogen clay results. Most silicate clays are
O
K)
op-
ADSORBED IONS SOIL SOLUTION
o ■o T
•Ir. • • “ •
.n." n » n • -J "
'o ,o
• o ° -
MICELLE
SOIL COLLOID a.o- O
ENLARGED SECTÔlf J
■ - CALCIUM+ +
• - MAGNESIUM+ +
• - POTASSIUM+
— HYDROGEN+
D— SODIUM+
0 -A L U M IN U M + + +
Fig. 6-6. Diagrammatic representation of the positive cation swarm about a negative clay micelle. An equilibrium between those ions adsorbed on the colloid and the ions in
solution exists unless: 1) the plant removes some from solution (i.e., Ca^^), and additional ions are released from the micelle to reestablish equilibrium, with perhaps ions
being adsorbed in the place of the metal cation, or 2) the farmer fertilizes with a chemical such as KCl. In that case, ions can be adsorbed on the micelle.
403
Table 6-7. Some properties of three types o f silicate clays [Brady, 1974].
Clay type
Property Montmorillonite Olite Kaolinite
Size (pm) 0 .0 1 - 1 . 0 0 . 1 -2 . 0 0.1-5.0
Specific surface (mVg) 700-800 1 0 0 -1 2 0 5-20
External surface High Medium Low
Internal surface Very high Medium None
Cohesion, plasticity High Medium Low
Swelling capacity High Medium Low
Cation exchange capacity 80-100 15-40 3-15
(meq/ 1 0 0 g)
Table 6-8. Representative selection of soil minerals with approximate “type” formulae. This list is not
meant to be complete [Adapted from Lindsay, 1979].
Mineral Approx, formula Mineral Approx, formula

Albanite MnS Albite NaAlSÌ3 0 g
Aluminum hydroxide A1(0H)3 Brucite Mg(OH) 2
(amorphous)
Brushite (DCPD) CaHP0 4 -2 H2 0 Calcite CaC0 3
Chlorite Mg5 Al2 SÌ3 0 io(OH)g Cupric ferrite a-CuFe2 0 4
Cuprous ferrite a-Cu2 pe2 0 4 Dolomite CaMg(C03)2
Fayalite Fc2 Si0 4 Franklinite ZnFc2 0 4
Fluorapatite Ca,(P0 4 )3 F Fluorite Cap2
Fluorphlogopite KMgjAlSijO.oFj Gibbsite A1(0H)3
Goethite a-FeOOH Gypsum CaS0 4 -2 H2 0
Hemitite '/2 a-Fe2 0 3 Hydroxyapatite(HA) Ca5 (P 0 4 )3 0 H
Illite K,Mg,3Al2.3SÌ350,o(OH)2
Jarosite KFe3 (S 0 4 )2 ( 0 H), Kaolinite Al2 SÌ2 0 3 (0 H) 4
K-taranakite H6 K3 Al5 (P0 4 )s-1 8 H2 0 Maghemite F2 A-Fe2 0 3
Magnesite MgC0 3 Magnetite F6 3 O4
Malachite CU2 (0 H)2 C0 3 Manganite y-MnOOH
Manganosite MnO Microcline KAISÌ3 O8
Mg-montmorillonite M g 2 (SÌ3 sAl,,Fe(III).2 Mg.3 ) 0 ,„(0 H) 2
Muscovite KAl2(AlSÌ30io)(OH)2 Octacalcium phos Ca4H(P04)3-2.5H20
phate (OCP)
Otavite CdC0 3 Periclase MgO
Pyrite FeS2 Pyrolusite ß-Mn0 2
Pyrophyllite Al2SÌ40,o(OH)2 Pyroxene CaAl2 Si0 6
Quartz SÌ0 2 Rhodochrosite MnC0 3
Siderite FeC0 3 Smithsonite ZnC0 3
Strengite FeP0 4 -2 H2 0 Talc Mg3SÌ40io(OH)2
Vermiculite (Mg„Fe(II)Fe(m).5 Ca,K,)SÌ 2 ,A l,, 0 ,o(OH) 2
Vivianite Fe3 (P0 4 )2 -8 H2 0 Wairakite CaAl2 SÌ4 0 i2 -2 H2 0
aluminosilicates. O f the clay groups in humid regions, kaolinite, montmorillonite, and illite are the most common
minerals, with some of their properties given in Table 6-7. Kaolinite represents a more advanced stage of
weathering under acid conditions. Vermiculite will also be found with a cation adsorption capacity exceeding
others. Other various clays can be found in diverse mixtures. Some clay minerals show positive as well as
negative charges, which makes possible exchange between the colloids and anions such as phosphate, sulfate,
chloride, and nitrate. Negatively charged ions, however, are often considered “mobile” in soils with the effect
that anions are more easily leached. Hydrous oxide clays are often intermixed with silicate clays in temperate
regions, and they are commonly dominant in tropic regions as the most advanced stages of weathering with
typical red and yellow soils. In the latter, gibbsite, geothite, and limonite are prevalent. A list of some more
important minerals often found in the solid state in soils is given in Table 6 - 8 .
The previous paragraphs dealt with inorganic minerals found in soils. The diversity and variation that can
be found as typified in Tables 6-7 and 6 - 8 merely illustrate the fact that a grower should thoroughly investigate
this aspect before beginning any particular cultural procedure. Local agricultural research stations are usually the
primary source of information. An example of the differences between humid and semiarid region soils is given
by Brady [1974] in Table 6-9.
Table 6-9. Data for representative mineral soils, comparing humid and semiarid temperate regions, as given by Brady
[1974].
Characteristics Humid region______ Semiarid region

Exchangeable calcium (meq/100 g) 6-9 13-16
Other exchangeable bases (meq/100 g) 2-3 6 -8
Exchangeable hydrogen and/or aluminum (meq/100 g) 4-6 1 -2
Cation exchange capacity (meq/100 g) 12-18 20-26

Base saturation {% Y 6 6 .6 95 and 92
Probable pH 5.6-5.S ~7
‘ Most cations, except aluminum and hydrogen, neutralize soil acidity, and the proportion of the CEC occupied
by these bases is called the “percentage base saturation.”
The second major component of soils, organic matter, is considered to have a colloidal structure similar to
clay. However, the principal difference between organic matter and inorganic colloids is that the complex is
composed mostly of carbon, hydrogen, and oxygen as compared to aluminum, silicon, and oxygen for silicate
clays. The cation exchange capacity of the product of decomposition of organic matter, humus, far exceeds any
of the clay materials and is not as stable as clay. Organic matter is a potential source of nitrogen, phosphorous,
and sulfur. It contains more than 95% of the total nitrogen in soils, 5 to 60% of the total phosphorous, and 10
to 80% of the total sulfur. Mineral soils vary in organic matter from trace to as much as 20 or 30%. Probably 3
to 5% is the average organic matter content of mineral soils. A high humus content for greenhouse composts is
not always desirable -although organic soils are noted for intensive vegetable production. There is a wide array
of organic compounds involved and a series of yellow to black substances called humic and fulvic acids
[Stevenson, 1991]. These compounds form many complexes with cations, that is, a group of related units of
which the degree and nature o f the relationship may be imperfectly understood. Peat moss, which is the most
common organic material for greenhouse production, is only partially decomposed. The physical properties of
humus do not make desirable potting mixtures for shallow containers, although a decomposed black sphagnum
peat is utilized in The Netherlands for bedding plants [Boertje, 1980]. Organic matter and clay minerals are the
broad solid phases of a soil or compost substrate.
The colloidal system represents a storage and exchange facility between soil and substrate solution. Taken
from Brady [1974], the situation as to exchange is noted:
405
Ca,40 Ca38
A l20
[MICELLE] 5 / / 2 CO 3
A l20
[MICELLE] + 2Ca{HCO^\ + M(HCO^) (6.16)
H.20 ^25
M20
For this example, the cations are monovalent in the ratio of 40, 20, 20, 20 per micelle, with “M” representing
other metal cations. In humid regions where calcium may be leached, the reaction will tend to the right with
hydrogen ions entering and calcium and other bases (M) being forced out of the exchange into solution where
they can be taken up by root systems, or they are leached from the soil. This exchange between soil solution and
the colloidal complex is chemically equivalent and the storage size of a soil is called its cation exchange capacity
or CEC. The use of CEC is a simplification of the many reactions causing adsorption and desorption in soils. The
subject has been discussed by Harter [1991], and consists mostly of laboratory procedures to measure cation
retention. For brevity, we will deal mostly with CEC as a means to evaluate nutrient capacities.
3. Expression for CEC

CEC is expressed in equivalency or Table 6-10. Cation exchange capacities of some minerals and soil mixes
milliequivalents per 1 0 0 grams soil [Bunt, 1988; Biernbaum and Argo, 1993; Brady, 1974].
(meq/100 g). As noted under the section
Cation exchange capacity
on Equivalency, monovalent ions such
Material meg (100 g)~^ meg (100 m^)'^
as Na"^, K^, NH4‘", and Cl' are the same
since they can replace or react with 1 Humus 200 100
ion. Divalent cations such as Ca^^ and Vermiculite 150 4
Mg^'" can replace 2 ions, so their Montmorillonite 100 100
atomic weight must be divided by two Illite 30
to obtain equivalent weight. If a clay Kaolinite 10
has a CEC of 10 meq/100 g, it can
Fine clay 56-63
exchange 1 0 mg of or its equivalent,
Coarse clay 22-52
for each 100 grams of clay. For a
hectare, 15 cm deep, this clay could Silt 3-7
exchange 220 kg H^. This has practical Sand 2-5 0
implications since 1 meq can be Sandy loam 20-40 -
replaced on the colloids by 1 meq Perlite 1.5
CaC 0 3 in limestone. As the molecular Sphagnum peat 100-120 10-30
weight of CaC 0 3 = 1 0 0 , then 1 0 0 / 2 = 50% Sphagnum, 50% sand 8 5
50 mg is required to replace 1 mg
75% Sphagnum, 25% sand 18 7
1 1 0 0 kg limestone per hectare would be
50% Sedge, 50% sand 21 17
necessary. CEC is usually determined at
pH 7 since it increases with higher pH. 75% Sedge, 25% sand 40 22
As will be noted later, CEC can vary John Innés compost 9 8
widely. Three examples are given in 25% Peat, 75% perlite 11 1
Table 6-7. For greenhouse practice, 50% Vermiculite, 50% peat 141 32
some authors [e.g., Bunt, 1988; 6 6 % Pine bark, 33% perlite 24 5
Biernbaum and Argo, 1993, Joiner et
al.,1981] have suggested that the CEC
of peat mosses and related substrates
should be expressed based on 100 ml rather than 100 grams (Table 6-10). Otherwise, the CEC of such materials
will be much higher than they should be as the result of their low bulk density. Bunt also suggested that CEC for
potting mixtures should be determined at the pH in which plants are grown -about 5.0 to 5.5 for those containing
high amounts of peat moss. In effect, CEC indicates the buffering capacity of a soil as to how much fertilizer or
lime is required to change the colloidal complex significantly. It sets the basic fertility reserve
of the soil and is one of the easier tests to

make on soil and related mixtures.
In most soils, the principal cation
adsorbed on the colloidal complex will be
Ca^^ -unless the pH is low. If one has a
substrate with 6 meq exchangeable calcium,
possessing a total CEC of 8 meq, there is a
high probability of readily available calcium
for plant growth. But, if the total CEC is 30
meq, with the same amount of exchangeable
Ca^^, the situation is completely reversed.
With some crops, the base saturation of the
soil should approach 90% Ca^^. In some soils,
the presence of 2 0 % sodium on the exchange
complex represents serious saline conditions
and the loss of physical structure or
deflocculation (puddling). From this
discussion, one may conclude that the greater
the CEC of a substrate, the more resistant it
will be to chemical change. On the other hand,
substrates consisting of plastics, rockwool,
gravel, etc., or flowing hydroponic solutions
such as NET, will have no cation exchange
capacity. Chemical additions to such cultures
will change the root environment immediately. Fig. 6-7. Activity of and Fe^^ as measured in soils, and
Failure to provide adequate nutrition on a Mn^^ calculated in equilibrium with Mn0 2 [Lindsay, 1972a],
continuous and timely basis will often show
an adverse plant response in less than a week.
A grower will often decide to produce in a buffered substrate since there is room, usually, for mistakes.
G. MASS EXCHANGE EQUILIBRIA

Another approach to determining nutrient availability of individual ions is the methodology described by
Lindsay in his 1979 textbook. Under suitable experimental conditions [e.g., Norvell, 1970; Norvell and Lindsay,
1969; Lindsay, 1981; 1991; etc.], the equilibrium exchange between the soil matrix and solution was actually
determined for specific ions -although the controlling solid phase was unknown. From this information, an
equilibrium constant for the reaction could be calculated. Examples of such reactions are:
(6.17)
SiOîsoil) ^ IH Ô -- H^SiO^
(6.18)
Fe{OH)^{soil) + ^ r/r 1 rv\
(6.19)
Ca(soil) ^
Mgisoil) ^ (6 .2 0 )
Zn{soil) + 2H^ ^ (6 .2 1 )
Cu{soil) + 2H ^ ^ (6 .2 2 )
Once the appropriate equilibrium constants are known, these can be combined for given and CO2 levels
to calculate the theoretical level of the respective ions in solution. These data can be placed into software
programs such as MINTEQA2/PRODEFA2 [Allison et al., 1991] and the spéciation of a given substrate or
407
hydroponic solution calculated in a short time. Lindsay and Ajwa [1994] have discussed the various computer
programs for soil solution spéciation and their use in teaching soil chemistry. However, with present com
puterization of greenhouses, it appears to me that these programs can be used on-line to predict need, and quan
tities required, in the grower’s fertilization program.
As an example in a simple case, assume that Zn^^ in a soil solution is being controlled by the soil zinc, Eq.
6.21. The equilibrium constant of this reaction is 10^ ^ Since this is equal to (Zn^^)/(H^)^, then:
logZ/1 ^" = 5.8 - 2pH (6.23)
At a pH of 6.0, the concentration of Zn^^ in solution would be 5.8 - 12 = 10'^^ or 0.00000063 M (0.63 pM). At
pH = 8.0, (Zn^"^) = 10'^^ ^ or 6.3 x 10'^^ M (0.0063 pM). A 2 unit rise in pH (100-fold decrease in hydrogen ion
concentration) will cause a 100-fold decrease in concentration of the zinc ion in the soil solution. Several
examples, including zinc, are presented in Fig. 6-7. Note that a 10-fold decrease in hydrogen ion concentration
(1 pH unit) will cause a 1000-fold decrease in Fe^^ activity.
Table 6-11. Conductivity units presently employed for designating salinity of a solution [Adapted from Bunt, 1988].
Units Value Ways of writing Remarks

pmho cm'* (micromho mho X 1 0 '^ mho
0 .0 0 0 0 0 1 Units used in U.K. by
per cm) 1 0 0 0 pmho = 0 . 0 0 1 mho ADAS and are common in
pSiemen (pS) cm’‘ Siemen x 10'^ 0.000001 Siemen U.S. literature
(microSiemen) 1000 pS = 0.001 Siemen
dS m’’ (deciSiemen per Siemen x 10’^ 0.001 Siemen These units are coming
m) 1 dS m'* = 1000 pS cm‘‘ = 1 mmho into wider use in recent
cm'* years
mmho cm"' (millimho mho X 1 0 '^ mho
0 .0 0 1 Units are common in older
per cm) 1 mmho cm'* = 1 0 0 0 pmho U.S. literature
1 mmho cm'* = 1 dS m'* Newer models read from
0 . 1 to 1 0 mho X 1 0 '"
mho cm'‘ x 1 0 ’^ mho X 10’^ mho

0 .0 0 0 0 1 In the U.S., older models
10mhos X 1 0 '^ = 1 0 0 pmhos = 1 0 0 of equipment read from
pS = 0.1 mmhos = 0.1 dS 1 0 - 1 0 0 0 mho X 1 0 '^
SC (specific conductivity) Siemen x 10'^ 0.00001 Siemen Reported as such in Ire

land
CF (conductivity factor) mho or Siemen 0.0001 mho or Siemen Many meters in U.K. cali
X 10" CFIO = 1 mmho brated to read from 0 . 1 to
100 mho X 10 "
H. ELECTRICAL CONDUCTIVITY (EC)
EC is the reciprocal of electrical resistance. Under standard conditions, a determination of EC is one of the
simplest tests to make in solutions extracted from soils, or to make continuous measurements in hydroponic
systems. The readings, in units of conductivity (Table 6-11) provide information on the general fertility level
of the substrate solution and whether salinity problems are likely to exist. It is probably the most common, quick
test employed. The unit of electrical resistance is the ohm, and its reciprocal, conductance, is the mho. Conduc
tance, as measured between platinum electrodes in a suitably extracted solution, is proportional to the surface
Condecti¥ity ---- dedSiemens/m at 25® c.
Fig. 6 - 8 . Milliequivalents per liter concentration o f several saturated paste extracts of soils as a
function of electrical conductivity [Richards, 1954].
area of the electrodes and inversely proportional to the distance between them. In a given conductance cell, the
measurement will depend upon ionic strength and concentration, increasing with increasing solute concentration.
Temperature will also influence the reading. As noted in Table 6-11, the units can be confusing, particularly with
the shift to deciSiemens per meter in recent years. Total salinity has also been expressed as ppm. The relationship
between EC and solute concentration of extracts from a number o f soils is given in Fig. 6 - 8 as an average. In
actuality, EC will vary with the particular salt: e.g., MgS 0 4 , CaS 0 4 , and NaHC 0 3 have lower conductivities
than other salts at equivalent concentrations [Richards, 1954]. It will be noticed that soil extract solutions
generally have an EC more than 10 times that found in irrigation water (Fig. 6-9). Soils with saturated paste
extracts greater than 4 dS/m are generally considered saline. Irrigation supplies with ECs below 0.75 dS/m are
usually good, whereas above 2.25 dS/m, the water supply is seldom satisfactory [Richards, 1954]. Greater detail
will be given later in the section on salinity (Section 6 .V).
409
Fig. 6-9. Milliequivalents per liter concentration of several surface streams and well waters as a function of
electrical conductivity [Richards, 1954].
L COMMENTS ON LOGARITHMS
As noted in the first section, nutrient concentrations range from the micromole to millimole, where 1 ¡xM
= 0.000001 M or 10'^, and 1 mM = 0.001 M or 10'^ M. The negative exponent can also be written as a logarithm
to the base 10, or -6.0 and -3.0, respectively. Nearly always, the log of concentrations we deal with in plant
production will be negative. Equilibrium constants (K°) are usually expressed as logarithms. This method has
some advantages. Addition of logarithms is multiplication, subtraction is division, and raising a number to some
exponential value is multiplication of the exponent [i.e., (H^)^ = 2(-logio (ff)) = 2pHj. Addition and subtraction
are much easier than multiplication and division. Present, inexpensive pocket calculators allow easy conversion
of numbers to logs and vice versa. As an illustration, suppose one wishes to learn the activity of Fe^^ in solution
where Fe(OH)3 (soil) is the solid furnishing ferrous ions:
Fe{OK)(soil) + 3H^ + e~ ^ Fe^ + 3HÔ (6.24)

Under standard conditions, the equilibrium constant (log K°) = 15.74 [Lindsay, 1979], and the equilibrium state
can be expressed as:
15.74
= 1 0
(6.25)
If Eq. 6.25 is expressed in logarithms, then:
logiFe^^) = 15.74 - (pe + pH) - 2pH (6.26)
If we assume a pH of 6.5 and a pe o f 8 , then Eq.6.26 yields log(Fe^^ = 15.74 - ( 8 + 6.5) - 2(6.5), or 10'"^^ The
concentration of Fe^^ will be 1.74 x 10'*^ M. Without other extenuating circumstances, an iron deficiency will
result. Compare this with the previous calculations on soil zinc. In contrast to iron, pe does not affect Zn^^
concentration.
III. NUTRIENTS IN THE PLANT AND TISSUE ANALYSIS
There are three major parts to greenhouse nutrition control: 1) visual diagnosis of the crop, 2) tissue analysis,
and 3) soil analysis. Visual observation is most important in assessing the general well-being of a crop in
greenhouses. It is a difficult practice, requiring complete familiarity with the crop and good observational ability.
This can only be achieved with practice and the opportunity to observe crop growth under a variety of conditions
over a lengthy period. Classroom instruction and pictures do not provide this ability to Judge acceptable growth.
Expert systems for diagnostic purposes are not readily available. It is hard to do where the crop does not exhibit
obvious symptoms, as shown in Figs. 6-10 and 6-11. Nevertheless, I have seen crop experts, after a thorough
walk-through o f a range, make general remarks as to crop status from which recommendations can be made to
the grower for improvement. Growers may not always have the opportunity to judge growth in greenhouses
outside their immediate region. In the example of air pollution over a wide area, growers can readily fail to note
subnormal crop performance unless the damage resulting is acute.
I will not attempt to discuss visual symptomatology, nor outline the roles of the various essential elements
except as necessary. The number of species produced under greenhouse conditions requires a separate textbook
for such an undertaking, and familiarity with performance of the particular species under a variety of conditions.
One will find in the ^ ature [e.g.. Joiner, 1983; Holcomb and White, 1982; Kofranek and Lunt, 1975; White,
1976; 1987; Nelson, 1978; Mengel and Kirkby, 1982; etc.] attention given to the role of essential elements in
plant metabolism and general-to-detailed descriptions of deficiency symptoms. Micronutrient toxicities and
macronutrient excesses are more likely to be important in greenhouse production as the result of high fertilization
rates. Considerable attention is often given to these aspects. Several texts consist mostly of pictures of typical
deficiency and excess symptoms as exemplified in Figs. 6-10 and 6-11. Wallace's 1953 text, and Smilde and van
Eysinga’s [1968] small book on tomatoes, are good examples of nutrient deficiencies and toxicities. Almost every
technical publication and textbook on crop production will provide some examples from the authors' personal
files.
Item 3), soil diagnosis, will be discussed in Section 6 .IV. In this section, tissue analysis, as well as a review
of the behavior o f essential elements in the uptake process, will be discussed. The tissue concentration of
essential nutrients in the plant, as determined under rigidly controlled conditions, is one principal means to
address fertilization requirements and practices.
A. DEFINITIONS OF NUTRIENT REQUIREM ENTS

Essentiality of a nutrient, for many authors, is based upon the element's requirement for the plant to survive
and reproduce - often called the “critical” level, or range. For the greenhouse grower, this is not a desirable
criterion for profitable production. Obvious deficiency symptoms as outlined or pictured by the previous authors,
or in Figs. 6-10 and 6-11, show lost control of the fertilization program. Since fertilizer costs can be less than
411
Fig. 6-10. Upper picture could be excess salinity or nitrogen deficiency in carnations.
The middle is typical boron deficiency, whereas the lower is zinc deficiency in roses.
2% of the total production cost, failure by a grower

in this area is poor management.
Tissue analysis is a principal method for
fertilizer control, particularly in hydroponic
systems. Walworth and Sumner [1988], however,
emphasized that successful interpretation is de
pendent on the accuracy of standard values (norms)
used for comparison with analytical results.
Furthermore, one needs an appreciation of how
ionic forms are taken up and transported through
the plant to interpret results properly.
According to Loneragan [1968], there is
marked variation among plants as to growth related
to nutrient concentrations in substrate solutions or
in their tissue. Confusion occurs as the term
“nutrient requirement” is often used for two purpos
es: 1 ) the least concentration required in the J
substrate for maximum growth, and 2 ) the minimal
concentration within the plant required for
maximum growth. From their work with agronomic
crops [1969a, b], Loneragan and Snowball found
that a plant with a high tissue requirement may have
a low requirement in the solution. The uptake rate
can be fast compared with other uptake rates. Thus,
nutrient solution concentrations do not always
parallel plant tissue concentration. In any case,
“nutrient requirement” cannot be equated with
minimum concentration for survival, nor, in our
case, as Loneragan defined it [1968]: as the least
concentration at maximal growth. Loneragan
defined the “functional” nutrient requirement as the
minimal tissue concentration that does not limit
growth. Loneragan and Snowball [1969a] defined
the “minimal functional requirement” for calcium as
that concentration in plant tops that remained
constant while yield increased substantially.
Both Bates [1971] and Jones [1985] used the
term “critical” concentration. Bates considered the
critical level as the tissue concentration beyond Fig. 6-11. Upper is a severe case of HC0 3 *toxicity in rose,
which further application of a nutrient does not leading to an apparent Fe deficiency. The lower (courtesy of
return a profit. Such a definition is more applicable A.C. Bunt) is a case of Fe deficiency in chrysanthemum, which
to the field as compared to greenhouses since was corrected.
fertilizer costs in field agriculture are a much larger
part of total production cost. For greenhouse production, excess fertilization is common. Jones criticized both
“critical” and so-called “standard” values because they are single values and do not describe the range possible
for sustaining maximum growth (see Table 6-12). Expressing elemental concentration from just above the
appearance of deficiency symptoms to that point beyond which toxicity occurs due to excess is the more common
practice. Jones calls this the “sufficiency range.” Sometimes an author may call the critical value that tissue
concentration resulting in a 10% yield reduction (Fig. 6-12).
Typical o f the curves showing yield response to nutrient tissue concentration is Fig. 6-12. Differences in
detail o f these curves for a particular species should be kept in mind. Major differences, according to Jones
413
I ADEQUATE ZONE
TRANSITIO^J^-^
ZONE No symptoms
CRITICAL CONCENTRATION
10 % Growth
reduction
O EFiCfENT
! ZONE
I
‘ Symptoms
j present
Fig. 6-12. General relationship between plant growth or yield and elemental content of the plant. [Upper graph from
Ulrich, 1961 { € )\9 6 \,A m e r . Inst, o f B iological Sciences), and Ulrich and Hill, 1973 (With permission of the Am er.
Soc. o f A gron om y). Lower taken from Smith, 1962 (With permission from the Ann. Rev. o f P la n t P h ysiology,
Volume 13, ©1962, by Annual Review, Inc.)].
[1985], are in the slopes as elemental concentration increases from deficiency to sufficiency. The lower, C-shaped
curve in Fig. 6-12 is called the “Steenbjerg effect.” This has been shown to occur with very severe nutrient
deficiencies that retard physiological development. Jones concluded that the lower figure best represented the
relationship between growth and macronutrients, while the upper curve was the best for micronutrients. Bates,
however, stated that plant analysis has a very limited value unless the C-shaped curve can be avoided. He
believed that the Steenbjerg effect could be eliminated by careful selection of the tissue to be sampled. The upper
graph does not show the possible toxic effects common to micronutrients at high concentrations. The range in
sufficiency for micronutrients can be narrow. On the other hand, Walworth and Sumner [1988] bluntly state that
the Steenbjerg effect is of little theoretical value. The problem, according to the latter authors, is that the
determination of a “critical” nutrient concentration must be conducted under conditions where all other
deficiencies are absent and

plant growth conditions are
optimal. Such a condition
is seldom, if at all, possible
in the field but may be
approached in controlled
greenhouse environments
under suitable climatic
conditions. An example of
“critical” levels, below
which deficiencies appear,
for copper in
chrysanthem um s is
Nelson's 1971 publication.
An excellent article on
micronutrient toxicity
levels in geranium was
published by Lee et al. in
1996.1 emphasize that such
critical levels may be far
different from an optimal
Fig. 6-13. Proposed relationship between nutrient levels in plant tissue versus yield, and as
range for a maximal yield. modified by other limiting factors [Sumner and Farina, 1986] (With permission of Springer-
In their review of the Verlag).
literature, Walworth and
Fig. 6-14. Yield in grams

fresh weight of S a lvia splen-
dens plants as a function of the
percentage dry weight of po
tassium in the tissue. Data
taken from the dissertation by
Jeong [1990]. The heavy,
sloping lines are boundaries
drawn by hand. The points
represent plants grown in a va
riety of substrates, fertilization
solutions, and different times
of the year.
Sumner find that

tremendous of data are
required to amounts of
data establish “boundary”
conditions that will ensure
that yield maxima are
reached at various nutrient
concentrations. From a
generalized model,
presented in their Fig. 2, I
have redrawn Sumner and
Farina's model as suggesting the actual situation about nutrition and crop yield (Fig. 6-13). Using more than 6000
Table 6-12. Nutrient content values published for a number o f greenhouse species. Macronutrients in percent o f dry weight, micronutrients in parts per million.
Plant species Na Ca Mg K N NO, Fe Mn B Mo Zn Cu Reference
A dian tu m cau datu m 0.2»0.3 0.2-0.4 2.0- 1.5-
3.0 2.5 0.40-0.80 Poole et al.,1976
A ech m ea fa s c ia ta 0.5-1.0 0.4-0.8 1.5- 1.5-
2.5 2.0 0.40-0.70 Poole et al.,1976
A géra tu m houstonianum ^ 0.03 1.6 0.7 4.0 4.9 1.7 0.8 160 100 39 43 18 Jeong, 1990
A gératu m m exicanum 3.12 4.75 0.75 Dight, 1977
A lyssum m aritinu m 4.28 4.37 0.51 Dight, 1977
A ntirrhinu m m aju s (snapdragon) 1.7 0.69 5.6 5.9 0.28-0.33 Boodley, 1962
0.04*^ 1.0 0.7 3.6 5.0 0.6 120 70 18 26 Jeong, 1990
A ntirrhinu m nanum 3.54 4.81 0.52 Dight, 1977
A nthurium A n draean u m 1.2-1.7 0.7-1.0 1.0- 2.3 1. 6- 2.1 0.28-0.33 Poole et al.,1976
A pium g ra v e o le n s (celery) 4.0 0.5 0.2 Jones, 1985
A ph ela n d ra sq u a rro sa 0.2-0.4 0.5-1.0 1 0
. -2 . 0 2.0- 3.0 0.2-0.4 Poole et al.,1976
A sp a ra g u s retro fra ctu s 0.1-0.3 0.1-0.3 2.0- 1.5-2.5
3.0 0.3-0.5 Poole et al.,1976
A u b rieta h y b rid 4.84 4.37 0.26 Dight, 1977
R h od o den dro n (azalea) 0.45-1.6^ 0.2-0.5 1. 0- 2.0-
1.6 3.0 0.2-0.5 Bunt, 1988
0.22 0.17 0.80 2.0 0.19 17 Twigg and Link,
1951
B eg o n ia x sem p erflo ren s- 0.14 0.9 0.7 3.4 4.0 0.08 0.6 328 120 32 36 16 Jeong, 1990
cultorum^^
3.00 4.17 0.21 Dight, 1977
B ra ssica actin o p h ylla 1.0-1.5 0.3-0.6 2.5- 2.5-
3.5 3.5 0.20-0.35 Poole et al.,1976
B ra ssica ca p ita ta (cabbage) 2 .0 *^ 0.5 0.25 Jones, 1985
C a lce o la ria ru g o sa 3.92 4.58 0.44 Dight, 1977
C a llisteph u s ch in en sis 4.16 4.11 0.53 Dight, 1977
C eio sia cristata^ 0.05 1.8 0.9 3.6 4.2 0.72 0.5 81 80 28 19 Jeong, 1990
C h a m a ed o rea eleg a n s 0.4-1.0 0.3-0.4 1.0-2.0 2.5- 3.0 0.20-0.30 Poole et al.,1976
C h rysa lid o ca rp u s lu tescen s 1.0- 0.3-0.6
1.5 1.0-2.0 1.5- 2.5 0 . 1 0 -0 . 2 0 Poole et al.,1976
C h loroph ytu m com osum 1. 0- 0.5-1.5
2.0 3.5- 1.5-
5.0 2.5 0 . 10- 0.20 Poole et al.,1976
C o leu s blumei^ 0 .1 1.2 0.7 4.8 4.6 1.17 1.3 142 153 33 32 10 Jeong, 1990
C offea a ra b ic a 0.5-1.0 0.3-0.5 2.0-3.0 2.5- 3.5 0.15-0.25 Poole et al.,1976
C ucum is M elo (sweet melon) 0.1 Jones, 1985
3.0 0.5 0.15 Jones, 1985
C ucum is sa tivu s (cucumber) 0 .2 ^^ Jones, 1985
0.46- 0.56-0.58 3.55-4.31 6.42- 0.61-0.68 Ingestad, 1973
0.57™ 6.54
4^
Un
+:>.
Plant species Na Ca Mg K N N03 p s Fe Mn B Mo Zn Cu Reference 0'\
Dahlia variabilis 4.00 4.06 0.62 Dight, 1977
0.46 0.06 2.15 4.50 0.17 20 Kofranek and Lunt,
Dendranthema x grandiflorum
n::r
(chrysanthemum) 1975 f:l,.._
(1)
0.8 0.4 1.5-5.5 4.0-5.0 0.8 0.5-6.5 50-100 50- 20- 25- 5-10 Boodley, 1964 ....
•0'1
150 30 lOO
7 Nelson, 1971 z
f;:
1.0-2.0 0.35-0.65 4.5-6.5 4.0-6.5 0.2-1.0 Bunt, 1988 g.

0.5-4.6 0.14-1.5 3.5-10.0 4.5-6.0 0.26-1.15 195- 25- 7.3 10 Kofranek, 1980 5'
::I
260 200
Dianthus caryophyllus 1.0-1.5 0.2-0.4 2.9-3.3 2.0-4.0 0.5-0. 7 0.2-0.35 0.1+ 50-100 50- 25- 25- 5-10 Hanan, 1975
(carnation) 150 lOO 100
1.0-2.0 0.25-0.5 2.5-6.0 3.2-5.2 0.2-0.3 Bunt, 1988
1.0-2.0 0.24-0.50 2.5-6.0 3.2-5.2 0.2-0.3 50-150 100- 30- 25-75 10-30 Mastalerz, 1977
300 lOO
Dianthus gigantens 4.16 4.27 0.57 Dight, 1977
Dizygotheca elegantissima 0.5-1.0 0.2-0.3 1.5-2.5 2.0-2.5 0.40-0.80 Poole et al.,l976
Dracaena deremensis 1.5-2.0 0.3-0.6 3.0-4.0 2.0-3.0 0.2-0.3 Poole et al., 1976
Dracaena fragrans 1.0-2.0 0.5-1.0 1.0-2.0 2.0-3.0 0.15-0.25 Poole et al.,1976
Dracaena Sanderana 1.5-2.5 0.3-0.6 2.0-3.0 2.5-3.5 0.20-0.30 Poole et al.,1976
Dracaena surculosa 1.0-1.5 0.3-0.5 1.0-2.0 1.5-2.5 0.20-0.30 Poole et al.,l976
Euphorbia pulcherrima 195 236 176 40 64 8 Boodley, 1974
(poinsettia) ppm
0.7-2.0 0.4-l.O 1.5-3.5 4.0-6.0 0.3-0.7 100- 45- 30- 25-60 2-10 Ecke, 1971
300 300 300
0.7-2.0 0.4-1.0 1.5-3.5 4.0-6.0 0.3-0.7 100- 100- 30- 25-60 6-15 Mastalerz, 1977
500 200 100
Epipremnum aureum 1.0-1.5 0.3-0.6 3.0-4.5 2.5-3.5 0.20-0.35 Poole et al., 1976
Ficus benjamina 2.0-3.0 0.4-0.8 1.0-1.5 1.8-2.5 0.10-0.20 Poole et al., 1976
Ficus elastica 0.3-0.5 0.2-0.4 0.6-1.0 1.3-1.6 0.1-0.2 Poole et al.,l976
Foliage plantsd 0.6-1.5 0.35-0.5 2.5-4.5 2.5-4.0 0.15-0.3 50-100 50- 20-50 5-15 Joiner et al., 1981
lOO
General• 0.5-1.5 0.35-0.55 3.5-4.5 2.5-4.5 0.2-0.3 75-125 50- 25- 25- 5-15 Joiner, 1983
100 lOO 100
Geranium 0.81-1.2 0.2-0.52 2.5-4.5 3.3-4.8 0.40-0.67 70-268 42- 30- 8-40 7-16 Holcomb and White,
174 280 1982
Jmpatiens holstiik 0.22 1.8 0.7 3.1 4.4 0.87 0.7 191 69 24 34 25 Jeong, 1990
Impatiens sultani 3.54 5.59 0.70 Dight, 1977
Plant species Na Ca Mg K N N0 3 p s Fe Mn B Mo Zn Cu Reference
Kalanchoe 0.02 4.43 0.25 3.25 2.90 0.32 0.59 Hansen, 1978
Latuca sativa (lettuce) 2.0 0.4 0.2 Jones, 1985
1.4-2.0 0.5-0.7 6.0-8.0 2.5-4.0 0.4-0.6 25- Geraldson et aL, 1973
45
0.20" 1.62 0.54 9.64 3.87 0.47 175 53 31 58 10 Sanchez et al., 1991
1.2-2.0 15 0.4-0.8 4.8-8.0 4.0-5.0 0.35-0.60 80-200 25- 20- 30-60 5-10 Sanchez et al., 1991
130 60
Lilium longijlorum (Easter lily) 0.81-1.2 0.20-0.52 2.5-4.5 3.3-4.8 0.40-0.67 70-268 42- 30- 8-40 7-16 Mastalerz, 1977
174 280
Lobelia erinesk 0.22 1.1 0.6 4.5 4.2 0.39 0.7 316 107 24 42 20 Jeong, 1990
3.70 3.77 0.35 Dight, 1977
Lycopersicon esculentum 1.5 0.2 0.15 Jones, 1985
(tomato)
2.5-7.2 1.0-1.5 2.5-6.6 0.5-1.5 0.44-0.66 155- 27- 48-66' Smilde and Eysinga,
193' 168' 1968
94-8J9f 43- 201- Smilde and Eysinga,
239f 458f 1968
2.5 0.5 5.5 4.8 0.5 1.6 90 350 35 0.5 80 15 Winsor, 1973
0.01- 1.25-3.0 0.3-1.0 6.0-10.0 2.5-3.5 0.5-1.0 20-100 50- 20- 1-5 20- 5-25 Wittwer and Honma,
0.40 100 40 200 1979
526- Chapman, 1960
1489'
1000- Millikan, 1951
2QQQS
300- Brennan and Shive,
500 1948
65- Chapman, 1960
198
4.0-6.0 0.6-0.9 2.5-4.0 3.0-6.0 0.5-0.8 60- 40- 15-30 4-8 Geraldson et aL, 1973
100 80
Maranta leuconeura 0.5-1.5 0.5-1.0 3.0-4.5 2.0-3.0 0.20-0.30 Poole et al.,l976
Kerchoviana
Matthiola incana 3.54 3.64 0.39-0.48 Dight, 1977
Mesembryanthenum crinijlorum 4.84 4.04 0.39 Dight, 1977

Monstera deliciosa 0.4-1.0 0.3-0.6 3.0-4.5 2.5-3.5 0.20-0.35 Poole et al., 1976
Nemesia strumosa 4.58 4.32 0.37 Dight, 1977 .~'>
-...1
-
.)::>.
Plant species Na Ca Mg K N N03 p s Fe Mn B Mo Zn Cu Reference 00
-
Peperomia 2h 56 14 267 !OOh 17 15 I 0.2 0.1 Hansen, 1978
0
Pelargonium 0.77 0.14 0.62 2.4 0.28 18 Kofranek and Lunt, P"
I"
1975 '0
.....
Petunia hybrida 0.66- 0.31-0.60 6.31-3.52 3.04- 0.55-0.69 42-691 93- 22- 28-65 5-14 White, 1976 ....('l
0\
1.73; 3.83 186 36
0.39k 1.3 0.7 4.9 4.0 0.51 0.6 144 78 27 29 I! Jeong, 1990 z
c
.....
....
Petunia multiflora 4.16 4.32 0.37 Dight, 1977 :=;.·
Philodendron scandans 0.5- 1.5 0.3-0.6 3.0-4.5 2.0-3.0 0.15-0.25 Poole et al. , 1976 o·
::l
oxycardium
Phlox drummondii 4.24 4.23 0.60 Dight, 1977
Raphanus sativus (radish) 0.5 Jones, 1985
Rhododendron 0.14 0.39 0.30 1.0 0.11 15 Kofranek and Lunt,
1975
Rosa 1.0-1.5 0.25-0.35 1.8-3.0 3.0-5.0 0.2-0.3 50-150 30- 30- 15-50 5- 15 Bood1ey and White,
250 60 1969
1.0-1.6 0.28-0.32 2.0-2.5 3.0-3.5 0.1-0.3 0.28-0.34 0.16- 80-120 70- 40- 20-40 7- 15 Sadasivaiah and
0.21 120 60 Holley, 1973
Saintpaulia (African violet) 44h 56 23 302 IOOh 27 12 0.9 0.2 0.1 Hansen, 1978
Salvia splendens 11 0.27 1.5 0.7 3.6 5.1 0.65 0.8 !59 !55 32 35 7 Jeong, 1990
3.24 4.48 0.72 Dight, 1977
Sansevieria trifasciata 1.0-1.6 0.3-0.6 2.0-3.0 I. 7-3 .0 0.15-0.3 Poole et al., 1976
Stromanthe amabi/is 0.1-0.2 0.3-0.5 3.0-4.0 2.5-3.0 0.2-0.5 Poole et al. , I 976
Syngonium podophyllum 0.4-1.0 0.3-0.6 3.0-4.5 2.5-3.5 0.2-0.3 Poole et al. , 1976
Spinacia oleracea (spinach) 0.045- Jones, 1985
0. 17
Tagetes erecta 3.84 4.23 0.60 Dight, 1977
Tagetes patulak 0.01 1.7 0.7 3.6 5.3 0.37 0.8 304 147 39 35 5 Jeong, 1990
3.68 3.98 0.60 Dight, 1977
Viola tricolor 4.16 4.46 0.53 Dight, 1977
Zinnia elegansk 0.01 1.7 0.6 4.3 4.2 0.49 0.7 103 89 63 34 8 Jeong, 1990
Plant species Na Ca Mg K N NQ3 Fe Mn B Mo Zn Cu Reference
® Critical levels for deficiency symptoms.

^ Levels below which deficiency symptoms appear.
Levels for 10% growth reduction
^ B levels not determined, S deficiency never seen, F toxicity common.
® Values for plants growing in marine clay soils (The Netherlands).
^ Plants growing in river clay soils (The Netherlands).
^ Values represent toxicity levels.
^ Values are relative to N set at 100.
‘ Ranges for plants grown in four different root substrates.
^Critical levels below which growth reduction occurs.
^Values are for plants grown in peat-lite, and irrigated with complete fertilizer solutions with NH4/NO3 fixed at a ratio o f 50:50, pH at 6.5, total concentration at 50
meq/1. For plants growing best in a series of treatments o f differing pHs, substrates and NH4/NO3 ratios.
*Averages for all species and conditions from the d ature as reviewed by Joiner (1983).
Cucumber seedlings in solution, varying nutrient proportions.
" DRIS norms established from 3316 observations o f crisphead lettuce in Florida.
° Sufficiency ranges for crisphead lettuce in Florida, using outermost sound leaf
4:^
Co
analyses for maize, Sumner [1977a] plotted the resultant data to give figures remarkably similar to Fig. 6-13. In
another paper by Walworth et al.[1986], more than 8000 tissue samples for maize were used to provide boundary
conditions. Use of boundary lines in analysis of biological data can be traced to Webb [1972], based upon the
supposition that there are response limits to various factors influencing organisms. A boundary line occurs that
provides a basis forjudging population performance.
As the number of limiting factors is reduced (n - 6 ), the range of nutrient “sufficiency” in the plant
decreases, until, under optimal growing conditions, the limits can be close together. The boundary conditions,
as shown by the sloping vertical lines on either side in Fig. 6-13, may be skewed, resulting from the fact that the
data is bounded on one side by zero and unbounded on the other side. Tissue concentrations cannot drop below
zero, and there is no limit on maximum concentrations. These problems of limiting factors affecting elemental
tissue levels for maximum plant response can be expected to be highly variable under field conditions. However,
greenhouses should be close to optimum in situations where the internal environment can be controlled. Unfortu
nately, what is “high yield” with a greenhouse crop may be mostly subjective or ill-defined -especially regarding
quality of fixed yield crops.
When the boundary conditions are established as shown by Walworth and Sumner [1987; 1988], Sumner
[1977a], and Walworth et al.[1986] for crops such as alfalfa, maize, citrus, oats, peaches, soybeans, sugar cane,
etc.; one may achieve a diagram such as in Fig. 6-13, which permits designation of nutrient “norms.” To the best
of my knowledge, such norms have yet to be established in greenhouse crops -w ith the possible exception of
field-grown head lettuce [Sanchez et al.,1991]. The interesting point about these data is that the “norms” so
derived are remarkably constant regardless of a locality. The result is particularly applicable to maize that used
data from the U.S., Canada, Africa, South America, and France [Walworth et al.,1986]. There are exceptions.
The outcome opens the possibility that such norms established for greenhouse crops would be applicable no
matter whether, for example, carnations were grown in the ground in unheated greenhouses in the Mediterranean,
or in environmentally controlled structures, in rockwool, in Colorado.
The published greenhouse data on tissue levels are largely what one finds in Table 6-12. These are
insufficient to show clearly what Fig. 6-13 suggests. However, by using Jeong's data on bedding plants [1990],
I have plotted the potassium content of Salvia splendens as an example of what happens when sufficient data are
available (Fig. 6-14). In my estimate, the figure shows clearly the validity of Sumner and Farina's model in Fig.
6-13. The result shows that while a tissue analysis may suggest an optimum elemental level, that does not mean
that yield will be high. Yield may be limited by another factor such as water availability, temperature, irradiance,
etc. This emphasizes the fact that tissue analyses can only show whether nutrient levels may be limiting -not that
they are the major factor causing low yield. The possibilities of this approach to greenhouse nutrition are
interesting, with historical roots extending back to Blackman's classical paper on limiting factors [1905].
In summation, the terms “critical value,” “sufficiency range,” or “norm” require close attention to definition
and application as defined by a particular author. It will be discussed later, given present methods of expressing
nutrient content, that values in Table 6-12 are subject to many restrictions. A perusal of many fertilizer
application articles in the horticultural literature shows little change in outmoded and incorrect approaches.
B. PUBLISHED NUTRIENT VALUES IN TISSUE (TABLE 6-12)

In view o f the previous discussion, single values given in Table 6-12 require the assumption that the value
represents a “critical” level below which deficiency symptoms are likely to occur, or there will be a significant
yield reduction. Exceptions are noted in the footnotes o f that table. Where ranges for an element are given, one
may assume a “sufficiency” range for “healthy” plants -with the same exception as noted. There are at least two
“general” ranges, covering most species, and for foliage plants, presented by Joiner et al. [1981] and Joiner
[1983]. One may note that these “general” ranges for many plant species do not always agree with those
published for a particular species within that group. As there is no way of deciding when a difference is
significant, one must study the original publication to make a decision. For example, the single values cited from
Jeong [1990] were taken from the highest fresh weights produced in his Experiment One for each species. When
these concentrations were plotted for Salvia splendens in Fig. 6-14, most of them were grouped in the center of
the figure and did not represent the highest yields obtained in Jeong's experimentation. The validity of plotting
tissue analyses to establish “boundary” limits is proved.
One may also note that where several information sources are cited, they may differ from each other.
421
suggesting problems in limiting factors as drawn in Fig. 6-13. For Dendranthema, Bunt's ranges for Ca, Mg, K,
and N are higher than those given by Mastalerz. Smilde and Eysinga give two sets of micronutrient values for
Lycopersicon, depending upon whether the plants were grown in marine clay soils or river clay soils. Whether
both sets represent maximum, comparable yields is open to question. One may note the difference in range for
potassium as published by Smilde and Eysinga versus Wittwer and Honma's for tomatoes. On the other hand,
Hansen expressed all his values for tissue content o f Saintpaulia as relative to nitrogen when N equaled 100.
While we shall see that the ratio between ions is important, this method for nutrient content expression leaves
one without means of comparison with other species. In contrast, the macronutrient values for Dianthus
caryophyllus, as given by Bunt, Hanan, and Mastalerz, are not markedly different. The data attributed to Jones
[1985] represent an onset of deficiency symptoms or 10% yield reduction. This may account for some low
elemental values compared to others for the same species.
I have deliberately not cited data from field cultivation of vegetables except Sanchez et al.'s 1991 DRIS
norms and sufficiency ranges for crisphead lettuce in Florida. Environmental control is nonexistent, and the range
of suitable substrate nutrient concentrations will be much wider as compared with usual greenhouse practice.
Plotting available data as in Fig. 6-13 might prove highly Illuminating. However, Soltanpour [personal
communication, 1994] stated that attempting to correct nutrition to meet the tissue norms showed by the boundary
method for maximum yields is often uneconomic in field production. In many areas, other factors limit yields
with consequent lower sufficiency limits. The grower does not need to correct nutrient concentrations to coincide
with tissue levels for maximum yields. Certainly, the requirement to define maximum tissue nutrient levels in
field crops would be less urgent compared to the greenhouse practice.
Note that sampling procedures have not been mentioned. The time, and place, of sampling on the plant are
critical and subject to strict requirements. These will be discussed later.
C. FACTORS AFFECTING PLANT NUTRIENT CONCENTRATIONS

One may list the variables that can be limiting factors in both tissue nutrient levels and plant response. Some
may be indirect as for nutrient availability from the root substrate:
1. Individual ion concentration in the substrate
2. Capacity of a substrate to supply nutrients
3. Soil moisture content, and, in hydroponic solutions, conditions of stirring and water flow
4. Total solution concentration or salinity
5. Aeration
6 . Temperature
7. Physiological age of the plant

8 . Part of plant sampled
9. Species and cultivar

10. Transpiration rate
11. Nutrient interactions
12. Nutrient source
13. Organic anion concentration
The above list is neither in order of importance nor inclusive, suggesting complexity is sufficient, and the
problems one may encounter in interpreting the tissue levels presented in Table 6-12. Some of these will be
discussed in greater detail in the following sections o f this chapter.
D. PLANT NUTRIENT UPTAKE AND IONIC INTERACTIONS

Absorption of ions by plants is influenced by three processes: 1) diffusion of ions from the soil to the root
as the result of either concentration or electrical gradients; 2 ) mass flow of water as the result of transpiration;
this can occur in the flow from soil to root, and is the principal means of transport in the xylem; and 3) selective
ionic absorption by the plant. For the present, we assume that the ions are at the root endodermis or the plasma
membrane of the root cells, and we are concerned with the interactions and processes by which ions are taken
into the plant. One problem that is sometimes difficult to learn in the literature, is the differentiation between
uptake by individual cells of the root versus uptake that is largely transferred to the shoots via the xylem vessels.
L Systems for U ptake by Roots

Ionic uptake is not completely understood. It does
require an active metabolism by the root system with
appropriate energy expenditure. According to Clarkson
[1985], ion uptake represents about 20% of the
maintenance respiration o f plants. This does not
include costs resulting from root exudation and
association with mycorrhizas, or the energy required to
grow a root system. Nutrients are inorganic
compounds, and the incorporation of these substances
is a basic process of plant metabolism [Mengel,
1974b]. Cations and anions may be absorbed
independently and not in equal quantities. Monovalent
ions are taken up more readily than divalent ions, and
ions with small diameters will be absorbed faster than
ions with large diameters. Plants are negatively Fig. 6-15. Effect of KCl concentration on absorption in
charged in relation to the outer medium (e.g., a the range of system I [Hiatt and Leggett, 1974].
surplus of electrons), and since electroneutrality
must be maintained within reasonable limits, ionic
relationships are very important in nutrition [Hiatt
and Leggett, 1974]. As the result of the electrical
gradient, a passive force exists that acts to draw
cations (NH 4 ^, K^, Ca^^, Mg^^, etc.) into the cell
and inhibits passive entry of anions (NOj’, S O /‘,
H2 P0 4 ', etc.). Kochian's calculations [1991] show
that, for a monovalent ion, a 1 0 -fold accumulation
gradient inside the cell can be balanced by a
membrane potential of -59 mV. For a divalent
cation, the same accumulation gradient can be
balanced by a potential of -29 mV. As root cells
often maintain membrane potentials of - 1 2 0 to
-180 mV, there is a large electrical driving force
for passive divalent ion uptake. Invoking active
transport processes is not necessary.
However, nitrate (NO 3 ) is generally the ion
absorbed in largest amounts, although ammonium
(NH 4 ^) can also be an important nitrogen source.
Since a principal characteristic of nutrient
relationships is the much higher concentration of
essential elements within the plant versus that in
the root medium (in field soils particularly); then
there must be an active, metabolic process by
which anions -and some cations- are accumulated
against a concentration gradient. In particular, the
mechanism for uptake has been investigated
extensively, showing uptake to be very efficient Fig. 6-16. Linear absorption of NO3 (upper) and PO4 (lower) by
although Ca^^ concentration in the soil solution lyê plants as a ftinction of time [Olsen, 1950] (©1950, with
may be ten times higher. Similar mechanisms are permission of Physiogia Plantarum).
considered to exist for N 0 3 ' and H 2 P 0 4 ' [Mengel,
1974b]. The process varies with the external ion concentration, and at the levels generally found in field soils (<1
mM, 10'^ M), active uptake is dominant with concentrations in root cells sometimes 2 x 10"^ times higher than
423
in the external solution [Mengel, 1974a]. Clarkson

[ 1985] stated that natural soils contain the major plant
nutrients at levels of 10'^ to 10'^ M. Within the
upper range (10‘^ M), maximum growth rate, as set
by such factors as radiant intensity, temperature,
etc., can be sustained. Russell and Shorrocks
[1959] showed that when the nutrient status of
plant and external solution was low, the concentra
tion in the transpiration stream could exceed that
of the external solution by factors exceeding 1 0 0 .
The transpiration rate under such conditions did
not appear to affect nutrient transfer to the shoots.
The above is accumulation against an electri
cal gradient, sometimes called system 1 , with the
maximum rate approaching a limiting rate in the
range where the system operates. Fig. 6-15, from
Hiatt and Leggett [1974], is a typical response for
potassium. Again, Clarkson [1985] stated that
although nutrient flux continues above the
Fig. 6-17. Effect of concentration on Fe uptake by rice
concentrations that saturate system I, it is not plants [Schwab and Lindsay, 1983a] (With permission of the
necessary for growth and can cause luxury Amer. Soc. of Agronomy).
consumption. This conclusion seems at odds with
the general observations of fertilizer levels and
plant response common to greenhouse production. Another example, where particular care was taken to ensure
a minimum boundary zone at the root interface, is Fig. 6-16, resulting from a series of papers by Olsen [1950;
1953a; b; c]. Olsen concluded that rate of absorption of nitrate and phosphate was independent of solution
concentration unless below 0.003 meq The latter is equal to 10"^^ M N 0 3 ' or H 2 P 0 4 '. For calcium, Loneragan
and Snowball [1969b] found that plants did not accumulate Ca^^ at concentrations of 0.3 pM (10‘^^ M) in the
flowing solution. Increasing the solution concentration from 0.3 to 2.5 pM (10'^^ M) increased yield greatly,
while Ca levels in the tops remained the same. Uptake of micronutrients has not been investigated as much since
the total amount is almost negligible compared to the macronutrients. However, Schwab and Lindsay [1983a]
showed a distinct minimum, limiting Fe^^ concentration, below which Fe^^ was not accumulated (Fig. 6-17).
Ingestad [1982] stated that most information about nutrient uptake relates to concentrations greater than
requirements. Maintaining very low concentrations common to field soils is very difficult unless the
concentrations at the plasma membrane interface are maintained, such as in flowing nutrient solutions. If care
is taken in this aspect, Ingestad [1982] reported that uptake rates for nitrates have been found down to levels of
10'^ to M (0.001 to 0.020 mmol f ’). With flowing solutions and constant inlet concentrations, Edwards and
Asher [1974] found the uptake of nitrate sufficient to maintain maximum growth of wheat at a nitrate level of
10'^^ M (0.005 mM/f) and a flow rate of 1 f min'^ In essence, these results show that one cannot separate
concentration from the rate at which the nutrient is supplied. When one takes into account supply rates, then
concentration can be much lower than commonly encountered. This conclusion has, obviously, marked impli
cation in hydroponic systems, especially NFT.
At solution concentrations above 1 mM, uptake may be passive, sometimes called system II (see comments
above on electrical potentials), which operates with system I. Absorption by system I is considered to operate
independently of the accompanying “counter” ion (e.g., K 2 SO 4 where the counter ion is 8 0 4 ^'). Whereas with
system II, uptake is strongly influenced by the counter ion with the requirement that cations and anions be ab
sorbed in equivalent quantities. Uptake by system II, according to Hiatt and Leggett [1974], does not result in
accumulation of ions against a concentration gradient. While the range of solution concentrations in field soils
(i.e., greenhouse production in the ground) may be predominantly system I, it is likely that system II operates in
limited substrate volumes and hydroponic systems for greenhouses. Recommended application rates of the
macronutrients for greenhouse culture exceed 1 mM, and are often 10 to 15 mM (10'^^ M) for N 0 3 ’ levels. Under
these high nutrient conditions, ion transfer

to the shoots varies closely with the tran
spiration rate, and the internal con
centration may be dose to that of the
external solution [Russell and Shorrocks,
1959]. In many species, Loneragan and
Snowball [1969a] found that Ca solution
concentrations from 10 to 1000 ¡xM (10'^
M) markedly increased Ca levels in the
plant tops but did not affect yield.
2. Ion Competition and Antagonism

The presence of several ions in
solution can cause competition between
cations and anions. Cation-anion
competition can occur after absorption. For
either cations or anions, an increased
Fig. 6-18. N uptake by chrysanthemum grown 10 weeks in the summer
concentration of one will often versus a cyclamen grown over 1 year [Bunt, 1976].
competitively inhibit the uptake of
others, with the effect greater on ions of
similar size. Potassium, rubidium,
ammonium, and cesium are examples of
such ions. As and N H / are normally
present, competition will occur between
these two. Typical of such relationships
was the examination of chrysanthemum
critical levels in fresh tissue by Nelson
and Hsieh [1971]. They showed that
below NH4/K ratios of 0.025 to 0.026
meq, N H / injury never occurred. Above
these ratios, injury to chrysanthemum
was always observed. Competition
between ions may be simply competition
for the same uptake sites in the cell
membranes. Adriano et al. [1987] listed
a number of antagonistic couples such
as phosphorous-zinc, cadmium-zinc, Fig. 6-19. Effect of total soluble salts on Ca levels in carnations [Hanan,
sulfur-selenium, and phosphorous- 1975].
arsenic. Synergistic effects also occur,
such as with calcium that will enhance
absorption of small ions, and decrease absorption of ions with large hydrated diameters such as sodium.
Sodium can also replace potassium to a limited extent. On the other hand, calcium can inhibit magnesium absorp
tion in some plants such as barley and maize.
There are distinct differences in ion absorption and accumulation between species and cultivars. A good
example of differences between species is Bunfs comparison between chrysanthemum and cyclamen in Fig. 6-18
concerning nitrogen accumulation. Dendranthema is considered a “heavy” feeder compared to cyclamen. The
differences between species are so great that making general assumptions is difficult. Some plants will selectively
accumulate sodium, others selenium, still others silicon. The general effect of total ionic concentration on Ca^^
uptake by carnations is illustrated in Fig. 6-19. Most authors stress the differences to be found between species
and even cultivars within a species. In particular, much of the basic information in the agronomical i ature deals
with monocotyledons such as grasses, maize, etc. Many monocots are grown in greenhouse production (i.e.,
decorative foliage plants). However, most important species (i.e., tomato, cucumber, chrysanthemum, etc.) are
425
dicotyledons. There can be sharp

division between such groups as for
nutritional requirements and
behavior. In a study of the effect of
micronutrients on macronutrient
uptake, El Kholi [1961] concluded
that there was no specific effect on
macronutrient uptake common to the
three plant species examined (oat,
lucerne, tomato). This might have
been due simply to the much lower
concentrations of trace elements
versus macronutrients. In oats, Yo-
sida [1964] decided that both
magnesium and potassium had a
mechanism of uptake that was
specific for the ion. Uptake via these
mechanisms was unaffected by the
concentration of other ions in
solution. However, the ratios of the
ions absorbed were determined by
the composition of the solution and
not by the solution concentration.
Vlamis and Williams [1962] found
that manganese toxicity in barley
grown in Hoagland's solution could
be eliminated by: 1 ) decreasing the
manganese concentration, 2 ) adding
10 ppm silicon to the solution, or 3)
increasing the concentration of Ca^^,
Mg"^ N H /, N 0 3 -, or SO4 '-. On the
other hand, the presence of
phosphate increased Mn uptake.
Raising the pH of soils by adding
lime will also reduce Mn uptake.
The problems of iron and zinc
deficiencies in Colorado soils were Fig. 6-20. (Upper) The effect of Fe levels on Zn content, and (lower) the effect
of phosphate level on Zn uptake of maize [Lindsay et al., 1963],
examined by Lindsay et al. [1963],
showing interaction between iron,
zinc and phosphate (Fig. 6-20). Increasing phosphate or iron levels suppressed zinc uptake in maize. There was
antagonism between iron, zinc, and phosphate.
The ionic relationships to be found in the plant kingdom are numerous and complex. This complexity
requires that nutrition be adjusted for the particular cultivar. For the present, Mengel [1974b] assumed that no
active uptake mechanism exists for Ca^^ and Mg^^. Active uptake systems do exist for K^, N 0 3 ‘, N H /, H 2 P 0 4 ’,
Cl' and S0 4 ^'. Active uptake systems for other cations such as Fe?^, Mrf^, etc. are also likely to be present. How
ever, the effects of micronutrients on macronutrient uptake are highly variable [El Kohli, 1961].
3. Ammonium and N itrate Relationships

The relationships between ammonium and nitrate in nutrition are particularly important. Most plants do well
solely on NO3'. On the other hand, Jeong [1990] found that Agératum exhibited a tolerance to NH4^ and some
toxicity to N 0 3 *. A s a rule, however, a 1:1 ratio of NH 4 ^/N0 3 ' produced the highest fresh weight in the seven
bedding plant species examined. Similar results have been obtained at other ratios and with different species
iS lIililii
<•» 'V.
' » C'~M.
Fig. 6-21. General effect of N 0 3 ' supply on lettuce growth. Far right = no N 0 3 ‘ applied, with
increasing amounts toward the left.
[e.g., Winsor and Massey, 1978;

Ingestad, 1970; 1971; 1973; etc.]. With
maize, however, Blair et al. [1970]
found no difference between plants
grown with ammonium or nitrate supply
when at low concentrations. The
maximum concentration of NH 4 ^ used
by Blair et al. was 2 mM, while nitrate
was supplied at 1 mM. Such low
concentrations would result, in a
commercial greenhouse situation, in
unacceptable plant response with exist
ing application systems -despite
Clarkson's [1985] and Ingestad's [1982]
conclusions.
Fertilizers using ammonium as the
nitrogen supply are cheaper than nitrate DISTANCE FROM ROO T A PEX IN MIL L IM ETER S
fertilizers. Therefore, growers will tend
to use such sources to the point that Fig. 6-22. Amount of accumulated in various regions of cotton roots.
outright ammonium toxicity can occur. The curve is an average of 10 roots with an average length of 72 mm
For reasons of phytotoxicity to plants [Canning and Kramer, 1958] (With permission of the B o ta n ic a l Soc. o f
and humans, the presence of ammonia A m er.).
(NH3) in greenhouses is forbidden
-although this is one of the more common nitrogen sources for field culture.
The literature was cited by Hofstra and Koch-Bosma [1970] as showing NH 4 ^ to be a respiratory inhibitor.
These investigators also employed low nutrient solution concentrations on tomatoes (0.5 mM N 0 3 ', 0.5 mM NFf^^
= 10'^^ M). Their results showed, however, a distinct seasonal difference on the responses of ammonium- and
nitrate-fed plants. In the summer, leaf growth of ammonium-fed plants was inferior to nitrate-fed plants. In the
spring and autumn, ammonium did not have any harmful effect.
Plants have a high demand for nitrogen, largely through the uptake of N 0 3 '. The “classical” effect of in
creasing N 0 3 ‘ concentrations on general growth of most species may be noted in Fig. 6-21. While N 0 3 *toxicity
has been observed (whiptail in cauliflower), excessive N 0 3 ‘ is more likely to result in “overgrowth” or excessive
vegetation (Fig. 6-21), with concomitant problems of ground water contamination by nitrate (Section 6 .V.C).
Several types of ammonium toxicity, on the other hand, have been observed, often associated with ammonium
accumulation in fumigated or steam-pasteurized soils. According to Barker and Mills [1980], plants that can
427
convert N H / to organic nitrogen have a much greater tolerance to ammonium nutrition. If, however, ammonium
reaches the shoots, biochemistry and physiology are greatly disrupted. Much of the energy production of the plant
must go into carbon skeletons for ammonium incorporation and its detoxification.
While 1:1 ratios o fN 0 3 ‘:NH4 ^ are usually safe, ratios of 5:1 are usual. Greater ratios may be recommended
or, levels of 10 mM N 0 3 ’ to 2 mM N H / [e.g., Hanan and Holley, 1975; Sadasivaiah and Holley, 1973; Jones,
1983; etc.]. However, it equally common to find numerous hydroponic solutions without any ammonium [e.g.,
Steiner, 1961; 1968; Hoagland and Amon, 1950; Cooper, 1979; etc.]. Hydroponic nutrition will be discussed in
greater detail in Section 6.5.
4. Organic Anion Content

To establish equilibrium between anions and cations, N 0 3 ' must be balanced by an equivalent cation uptake
or release of anions [Mengel, 1974a]. Thus, an N 0 3 ' supply results in release of bicarbonate (HC0 3 ‘) ions,
increasing the external pH. On the other hand, uptake o fN H / causes H^ release, significantly reducing pH. NO3 '
uptake stimulates organic anion formation (oxalate, malate, citrate, etc.) within the plant [van Tuil, 1965]. Since
the organic anions synthesized within the cell, at usual pHs, are 80 to 90% dissociated, there will be a consid
erable attraction for cations [Hiatt and Leggett, 1974] (see previous comments on electrical gradients). Van Tuil
[1965] found that, with plants cultivated in nitrate-containing solutions, the accumulation of K^, Na^, Mg^^, and
Ca^"^, denoted as “C,” minus the accumulated N 0 3 ', Cf, H2 PO4 ', and S O /', or “A,” would give one the organic
salt content of the plant material (C-A). If nitrates are replaced by ammonium, the C-A content is reduced,
ammonium inhibiting uptake of other cations. Usually, the C-A value is constant over large changes in salt
content -expressed as meq kg'' dry weight. According to van Tuil, the normal C-A content for perennial ryegrass
is 1000 meq kg'', whereas with sugar beets, the C-A value is 3500 meq kg''.
C-A contents have seldom been evaluated for greenhouse crops. Noteworthy was the work by Green [1967a,
b] and Green et al. [1971] for carnations. The total organic anion concentration of carnation leaf tissue was
highly correlated with growth. Optimal growth was invariably associated with a normal C-A content of about
1700 meq kg'. Hiatt and Leggett [1974] stated that the association of inorganic ions with the internal organic ions
(i.e., malate) reduces ionic activity, so that ions may not accumulate against a concentration gradient. The inter
action of organic and inorganic ions is a major determinant of ion accumulation levels and ion distribution in
plants. The diagnostic flow chart devised by Green et al. [1971], wherein nutrition can be adjusted according to
C-A and nutrient levels, has not been applied to any great degree in greenhouse production. This is unfortunate
since evaluation of C-A content would be another parameter in tissue analysis for diagnostic purposes. In an
examination of 16 species, which included lettuce and spinach, Noggle [1966] found that a lower yield was
associated with plants that had higher total inorganic anion concentration. With one exception (lettuce), lower
yields were associated with lower organic anion concentration. Chloride-treated plants always contained more
inorganic anions than sulfate-treated plants, the former being associated with lower yields.
Nitrate, potassium, and chloride are taken up very rapidly whereas ions such as sulfate and calcium are taken
up more slowly. The difference means that plants remove cations and anions in unequal amounts from solution.
These differences are balanced by the accumulation or degradation of the internal organic anions such as malate
[Mengel and Kirkby, 1982], or the release to the external medium of protons (H^) or carbonate anions (HCO3 ).
For example, if K2 SO4 is applied, the more rapid uptake causes a release of H / and malate is accumulated.
Conversely, when CaCl2 is applied, anion uptake exceeds Ca^^ absorption, HCO3' is released and malate level
drops. Similar responses with NO 3 ' and NH 4 ^ were mentioned previously.
5. The Root and T ranspiration

A very large root surface is generally exposed to the substrate environment (Chap. 5.III.A), and how much
of this surface participates in ion uptake is important. Individual ions differ markedly in behavior. Mengel and
Kirkby [1982] stated that movement of Ca^^ is restricted to young plant parts, whereas and H2 PO 4 ' absorption
and movement to the shoots takes place readily throughout the root length. Ca^^ uptake is in the root tip region.
A typical example, using a radioactive isotope of phosphorous with cotton as the test plant, is shown in Fig. 6-22.
In the three species tested (cotton, maize and pea), heavy upward translocation from the point of supply occurred,
with much more in regions 10 to 50 mm behind the root tip [Canning and Kramer, 1958]. The peak uptake in the
rapidly dividing root apex may have been due more to rapidly growing cells than to export to the rest of the plant
(Fig. 6-22).
NO 3 ', H2 PO4 ' , and K" are usually transported rapidly to the upper parts of the plant, depending upon uptake
and transpiration rates, while Ca^^, Mg^^, and S O /' are transported at lower rates. The anions NO3', H2P04' and
S0 4 ^' are rapidly incorporated into the organic composition of the plant. They can be moved through the phloem
and redirected in the plant, either as ions or as part of the carbohydrate system, which includes amino acids and
other products. The fact that nitrogen deficiency first appears in the older parts of most species shows
mobilization and removal of nitrogen to the younger, growing parts of the plant. Cl', N a / and can also be
redirected as univalent ions in the phloem system.
Iron, and particularly calcium, on the other hand, are generally fixed so that deficiencies usually appear first
in the actively growing regions. The lower picture in Fig. 6-11, is a classical example of iron deficiency, which,
on correction, allows the plant to resume acceptable growth. Ca^^ is always transported to the growing meristems,
and, as shown by Loneragan and Snowball [1969a, b], will not be translocated from older tissue under a
deficiency situation. As these authors showed, plants transferred from 10'^ M calcium solutions to 10'^^ M devel
oped calcium deficiency though their tops had three to ten times the calcium concentration of plants grown
continuously in solutions of 10'^^ to 10'^ M. Loneragan and Snowball [1969a] suggested that variations in
calcium supply under which deficiency develops partly account for the wide range of critical values found in the
^ ature (Table 6-12). The average Ca^^ “functional requirement” found for 30 species by Loneragan and Snowball
was 0.1 to 0.2%. Table 6-12 shows critical or sufficiency levels given for many greenhouse-grown species to be
much higher.
Section 5.IV.B.2 elaborated on calcium deficiencies of many plants under high humidity and low solar
radiation. Geraldson [1954; 1955] examined problems of blossom-end rot of tomatoes and blackheart of celery,
showing that high substrate calcium levels did not always mean sufficient calcium assimilation. Appropriately
timed foliar applications of 10'^ ^ M CaCl2 to tomatoes, or 10'°^ M directly to the hearts of celery, were sufficient
to prevent the disorders. In a rockwool slab system, Sonneveld and Voogt [1991] found that Ca deficiency
symptoms could appear within 2 to 3 weeks after the start of a deficient treatment in tomatoes. They found that
tomato fruits were most sensitive to Ca deficiency 7 to 10 days after flowering. Temporary disturbances of
calcium supply in rockwool systems were likely to have both short-term and long-term effects. Similar calcium
problems have been found in ornamentals such as the poinsettia concerning stem strength and bract necrosis
[Lawton, 1989; Harbaugh and Woltz, 1989].
It is apparent that solution concentrations of the respective essential ions are not always good predictors of
sufficiency in the plant. The nutrition (uptake) can be altered not only by concentration but also by the irrigation
treatment. Increasing concentration will tend to offset a low irrigation frequency (Fig. 6-23, left). Conversely,
increasing irrigation frequency, will tend to offset a low nutrient solution concentration (Fig. 6-23, right).
Obviously, such cultural procedures will be limited by the substrate and weather conditions. In flowing nutrient
solutions, where the boundary layer is as thin as possible, quite low nutrient concentrations may be satisfactory.
Reducing transpiration by raising humidity has been shown to reduce uptake of Ca^^, and Mg^^ in tomato,
nitrogen in Nephrolepis, phosphorous in begonia, and and Ca^^ in chrysanthemum [Gislerod and Mortensen,
1991].
There is a distinction between ions that may be absorbed by the individual cells of the root and movement
through the “free space” of the root -at least up to the endodermis. The latter refers to ion diffusion through the
cell walls and the intercellular spaces. Ions can be absorbed by cells and move from cell to cell. At the root
surface, the epidermis is often covered by a polysaccharide gel material (mucigel), which can influence water and
ionic relationships in the rhizosphere [Kochian, 1991]. Also, within the cell walls, there are many organic groups
that could bind micronutrients very tightly. Kochian states that the primary site of ion entry into roots is still not
clearly outlined. There might be several ways by which ions can be transferred through the cortex, across the
endodermis, and unloaded into the xylem for transfer to shoots. These considerations emphasize that flowing
solutions that reduce the boundary layer at the root surface do not take into account the fact that diffusion pro
cesses still occur within the root itself
E. NUTRIENT DISTRIBUTION AND TISSUE SAMPLING

It can be expected from the previous discussion that nutrient contents throughout a plant will not be uniform.
An example of the variation to be found in tomato is exhibited in Fig. 6-24. Nutrient content will also vary with
429
days
IRRIGATION FREQUENCY
Fig. 6-23. Interaction of solution concentration and irrigation frequency on fresh weight of carnations grown in perlite with
nutrients injected at each irrigation [Hartman, 1971].
age [Boodley, 1974], and as the plant ages, dry weight will increase. Since nutrient content is expressed by
percent of dry weight, values at maturity will tend to be lower. According to Walworth and Sumner [1988],
nutrient content should be based upon fresh weight. However, the inconvenience in storing and transporting wet
samples has resulted in failure to adopt a fresh weight convention. Commonly, leaves are sampled for tissue
analysis, and even here. Dole and Wilkins [1991] showed the elemental content in poinsettia leaves would vary
with nodal position. One of the better expositions on plant sampling was Jones' review of soil testing and plant
analysis [1985]. He pointed out that mixing of plant parts, or the taking of whole plants, is not a recommended
procedure. Nutrients can accumulate in leaf margins, and Jones suggested removal of leaf margins to avoid this
problem. However, this greatly increases handling, and with some species, would not be feasible.With many
greenhouse crops, the short period in the greenhouse means that only one tissue analysis can be taken for future
guidance since the crop will probably be marketed before any corrective action can be taken. In other words,
tissue analysis becomes a postmortem. For most crops, sampling may be at, or just before, flowering. Plants
damaged by insects or diseases, or mechanically damaged, should not be selected. Obviously, steps should be
taken to remove any soil, dust, or chemicals that may have been applied -such as pesticides containing copper,
zinc, etc. Jones [1985] recommended washing fresh tissue with a 0.1 to 0.3% detergent solution followed by a
distilled water rinse. Iron and manganese concentrations may be most affected by washing.
Once obtained and treated as suggested above, tissue must be dried. Jones suggested drying at 80 C.
However, Section 5.III.E cited Gardner as recommending 50 C for drying organic soils to avoid loss of organic
material. I would suggest that 50 C is probably safer. One will also find investigators using 70 to 75 C for drying
in more recent ^ ature [e.g., Frick and Mitchell, 1993; Hood et al., 1993]. So, one cannot be completely sure of
investigational results. Once dried, tissue should be stored in a moisture-free atmosphere until processing by the
testing laboratory. If tissue can be taken to the laboratory immediately after sampling, the laboratory can provide
drying and remaining sample preparation. If there is any delay, however, the sample should be dried immediately
before shipping.
Sampling procedures for many greenhouse crops are given in Table 6-13. As will be noted later, problems
in variance of the physiological age and sampling position on the plant can be reduced by the manner of
diagnosis. Sumner [1977b] stated that if one tissue type is sampled, diagnosis by the DRIS system is likely to be
independent of the leaf position in maize. Diagnosis will be discussed in the following section. Usually, tissue
may be taken from 20 to 100 plants for a representative sample. This requires some judgment on the grower's
17%
13%
8%
62%
Leaves Fruits Stems Roots

Fig. 6-24. Percentages of cations (K + Na + Ca + Mg = 100%) in different parts of tomatoes [Mengel,
1974a],
part. For example, plants of the same cultivar, if grown under uniform conditions of temperature, fertilization,
substrate, etc., may only require one sample taken randomly in the crop. Single samples may not be desirable
initially. Some idea of the variability to be expected between so-called “identical” treatments should be available.
This may require as many as ten or more samples from identically treated plants. It allows calculation of a
suitable standard error. The grower may then decide the reliability of the data obtained from what he considers
a “uniform” crop. On the other hand, plants growing in the ground may require several samples to obtain a
reasonable average because of soil variability common to most field soils. Grower observations of plant behavior
in different parts of the greenhouse can suggest appropriate sampling patterns. About 5 g dried tissue are
required. Assuming a dry weight content of 10%, then 50 g fresh tissue will be necessary -which can be more
than can be comfortably held in two hands. It will be noted in Table 6-13 that some recommendations are
different ifom each other. Those underlined are suggested by the author. Usually, a general recommendation is
to sample the youngest, fully mature, and expanded leaves on a flowering stem. Unfortunately, most greenhouse
operations have failed to take full advantage of foliar analysis -in part due to the higher costs compared to soil
analysis. However, laboratory analyses of plant tissues are more standardized and reliable than many soil tests.
That is, comparison of results between laboratories for foliar analyses are more likely to agree.
F. DIAGNOSIS FROM TISSUE SAMPLING

In previous sections we found that elemental content of plant tissue will vary with age and position on the
plant and also with interactions that affect nutrient uptake and distribution. In their description of the DRIS
system, Walworth and Sumner [1987] cite the literature to the effect that N, P, K, S, Zn, Mn and B concentrations
decrease with advancing maturity in field crops such as alfalfa, whereas Ca and Mg increase to early flowering
and then decrease. In potato, P, K, and Zn decrease as the plant ages. Even assuming that the proper procedures
have been carried out on a timely basis, foliar analysis cannot provide information about fertilizer requirements
or plant response to fertilizers. The fact that there is a suboptimal level of potassium does not say how much
fertilizer must be applied or how the plant will respond [Walworth and Sumner, 1988]. The dynamic nature of
elements within a plant means that use of values presented in Table 6-12 requires strict attention to the sampling
method and time, as shown in Table 6-13. A sample must be taken at precisely the right time and from tissue that
431
Table 6-13. Plant tissue sampling procedures for greenhouse crops [adapted from Jones, 1985; also Witter and Honma,
1979; Winsor, 1973; Boodley, 1962; White, 1982; 1987; Nelson, 1978; Ecke, 1976].
Crop Growth stage Part to sample

Tomato Immediately below last developing flow- Petioles of leaves
er cluster
At fruiting 5th leaf below top
Prior to or during fruit set Young plants, leaves adjacent to 2nd and 3rd clus
1st leaf above a fruit cluster^ ters, or
old plants, leaves from 4th to 6th clusters, or the
1St leaf above a cluster
Leaf crops, lettuce, spi- Midgrowth Youngest mature leaf, or wrapper leaf on head let
nach, etc. tuce
Snapdragons At flower Most recent mature leaves on flowering stem
Geraniums At flower 1st & 2nd fully mature, expanded leaves on a
shoot
Melons (water, cucum Early stages of growth prior to fruit set Mature leaves near base of plant on main stem
ber)
Roses During flower production Upper mature leaves on flowering stem
Flower calyx just opening 2 uppermost 5-leaflet leaves including petioles
Flowering stem, pea size to when Entire 1st 5-leaflet leaf, including petiole, count
showing color ing from top of shoot
Chrysanthemum 5-6 wks after planting or after pinching Youngest fully expanded leaves. 1/3 of distance
down the stem
Prior to or at flowering Upper mature leaves on flowering stems
Carnations Unpinched plants, single stem 4th or 5th leaf pairs from base of plants
Pinched plants 5th and 6th leaf pairs from top of resulting laterals
Poinsettias Prior to or at flowering Most recently mature, fully expanded leaves
Youngest mature leaves including petiole
*Underlined recommendations are suggested by the author.
is physiologically identical to that from which the critical value or sufficiency range was determined.
A number of approaches have been made to interpret foliar analyses. Of these, three will be discussed in
detail: the Critical Value Approach (CVA), the Sufficiency Range (SR), and the Diagnosis and Recommendation
Integrated System (DRIS). The latter, although developed initially in the late 1960s, has not, to my knowledge,
been applied to crops in greenhouse production. Most of the available literature advances the use of DRIS, but
there remain problems in practicality [Soltanpour, personal communication, 1994]. It may be a system that would
advance nutritional practices in greenhouses.
1. Critical Value A pproach and Sufficiency Range

Diagnosis using the CVA and SR methods requires that the composition of the tissue be compared with a
value determined at the same growth stage (Tables 6-12, 6-13). In both the CVA and SR systems, problems with
determining the appropriate physiological age, and the variation in nutrient content on a dry matter basis with
age, is their greatest disadvantage. The SR method may decrease diagnostic precision because the limits are too
wide [Walworth and Sumner, 1988].
Some of the above problems were discussed by Holland [1966]. He pointed out that a nitrogen deficiency
may retard growth without affecting K uptake, which could result in an increased potassium level. There exists
a dynamic balance within the plant in which single elemental analyses may not successfully suggest which
Fig 6-25. Schematic diagram of data used for

DRIS evaluation. Shaded areas are values
fromunhealthy tissue due to insufficiency,
excess, or imbalance and not used for
determining a norm value. Only those results
from high yield group utilized [From Walworth
and Sumner, 1988].
element is insufficient or in excess.

Commonly, elemental concentration is
compared with some yield index such as
dry weight of seed, fresh weight of the
whole plant or fruit production, or number
of flowers produced. Typically, studies of
this type are often called fertilizer response
studies. From these, generating a
mathematical relationship by statistical
means is common. Such a curve may be
divided into deficient, transition, and
adequate zones [Ulrich and Hills, 1967]
(Fig. 6-12). However, this is applicable
only to a single element, whereas the relationship is actually affected by other elements as mentioned above, or
the fact, for example, that the critical P level depends upon the N level. The so-called “univariate” approach (a
single element), according to Holland [1966], merely produces a catalog of means, standard errors, and signifi
cance levels. It contributes nothing to interpreting the results. Holland [1966] proposed a “multivariate” approach,
capable of defining restrictions without limit to the number of elements involved. Holland also developed the use
of nutrient ratios as suggested by Beaufils [1971; 1973]. Terman and Nelson [1976], however, pointed out that
all of the simple linear (rectilinear) or curvilinear (quadratic), multiple regression models -which are calculated
by common statistical procedures- are in error if they give a positive yield intercept. (The curve crosses the yield
axis at some point above zero at 0% nutrient.) Several studies have shown that the critical level of one nutrient
can vary with levels of other nutrients. Terman and Nelson concluded that multiple rate, one- or two-variable
fertilizer experiments, summarized by the usual quadratic regression models, are not realistically or biologically
meaningful.
The problems outlined above are some principal reasons why CVA or SR methods have not achieved wide
use in the greenhouse industry -not to mention higher costs. Diagnoses of adequacy or insufficiency in crops
with CVA and SR methods have the advantage of simplicity. Walworth and Sumner [1988] concluded that since
many CVA and SR values may have been determined under suboptimal conditions, it is impossible to infer what
might have occurred had yields been higher. A CVA nutrient concentration based on yield curves from such
experiments may not be applicable elsewhere. Nevertheless, the CVA and SR methods remain the most common
practice in greenhouse nutritional control.
2. The DRIS System for Foliar Diagnosis

Two items in application of the DRIS system must be disposed o f The first is the development of “norms”
to which foliar analyses are to be compared. This generally follows the “boundary line” definition discussed
briefly in Section 6 .III.A. The second is calculation of the necessary non-dimensional “indices” used to diagnose
nutrient deficiency or sufficiency, and the relative order of their importance. This will be difficult and necessarily
brief since there is no body of information in the i ature that is applicable to greenhouse practice. Development
and description of the methods for field crops may be found in articles by Beaufils [1971; 1973], Sumner [1977a;
b; c; 1979], Sumner and Farina [1986], Walworth and Sumner [1987; 1988], and Walworth et al. [1986]. Many
other articles may be found in the d ature, but the only publication I know of in the horticultural d ature that deals
with what may be considered a greenhouse crop, employing DRIS, is the article by Sanchez et al. [1991], in
which DRIS is used to evaluate the nutritional status of head lettuce.
433
a. Norm Development
The designation of boundary lines was illustrated in Fig. 6-14. Such figures are plotted for each element
being considered (N, P, K, Ca, Mg, etc.), and for each ratio to be evaluated (e.g., N/P, N/K, P/K, etc.). Use of
ratios eliminates the need for dry weight in the denominator:
. IQQAT
7V% ^ DM ^ AmN. , DM. N (6.27)
( ----------) ^ ( ----------) = ■—
PVo DM loop P
^ DM^
As a result, the ratios are dimensionless. The variation in dry weight as the plant matures is eliminated. If the ratio
is properly chosen, variation with time and location of the elements in tissue is also reduced. Data in a survey
technique are considerable (8000+ for maize). For crisphead lettuce, 3316 observations from various field
experiments were used to develop norms [Sanchez et al.,1991]. These may be skewed so that the data do not
develop the “normal” or Gaussian distribution. To reduce the skewness, the data are divided into high-yield
groups and low-yield groups (Fig. 6-25). The high-yield data correspond to “good” growers, with a distribution
that usually allows one to calculate a mean and coefficient of variability (CV), which is used to “weight” the
mean, or define the contribution of the ratio to the index to be calculated:
(E ^ )*
E > , '-
lOOi- j 2 _
CV = ------ where s = and s
^ n n -1 (6.28)
Here; x = the average of the data, Sx^^ = the sum of each datum squared, (Sx^)^ = the total sum of all xs squared
and, n = the number o f observations (data). The symbol “s” is the standard deviation of a sample, standard error,
or standard error of a mean while s^ = the variance. This is an estimate of the variation from a mean of a
population. O f course, the mean is the sum of all observations divided by the number of observations. The CV
is a relative measure (percent) as contrasted with the standard deviation, which remains in the same units as the
observations.
The standard deviation of the mean, x, gives the reliability of the mean ( x ), but there is no indication that
the true value must be within the limits of x ± s. However, if the errors are randomly distributed, 95.4% of the
errors will lie within x ± 2s, and 99.7% of the errors within x ± 3s.
A partial listing of DRIS norms for head lettuce is given in Table 6-14. Sanchez et al.expressed the
nutritional data in all possible ratios and reciprocal ratios plus elemental as the ratios %/DM (Dry Matter). The
elemental ratios are listed in Table 6-12, along with the sufficiency ranges. Not all of the calculated DRIS norms
fall within the SA ranges. Although Sumner and others have discussed incorporation of other environmental
factors in DRIS analyses, this has not been carried out. There is the problem that crop yields are often limited
by factors other than nutrient concentrations and ratios. This would place sufficiency ranges lower on the diagram
in Fig. 6-25. The critical value for deficiency, or minimum sufficiency, would be lower. Under such conditions,
it would make little sense to attempt fertilization to bring concentrations to the optimum for maximum yield since
a maximum yield would never be obtained given the particular circumstances.
b. Index Derivation
For each pair of nutrients there are three forms: e.g., N/P, P/N, and N x P. The selection may be made by
choosing the ratio that has the largest variance ratio: that is s^/sp versus Sp/s^. For a product (NxP), a new value
of A = 1/P is defined. As N/(l/P) = N x P , then one has N/A as the product N x P . Only one expression is used
to relate each nutrient pair. Learning from a single ratio such as N/P whether N or P is too high relative to the
other or too low is not possible. A combined yield-quality index was used to separate the high yielding population
versus number of heads alone. This requirement would be necessary for most greenhouse crops since quality
would determine the number of flowers, pots, or fruits marketable. Sulfur was not included in the evaluations.
In the calculation of the indices, a simplified case of N, P, and K will be used from Sanchez et al.'s data,
although, obviously, from Table 6-14, calculation for all essential elements may be used. There was insufficient
data in Jeong's dissertation [1991] to determine such indices for the bedding plants Jeong used. Data for other
important greenhouse crops are unavailable at this time.
Since we want determination of N, P, and K, three equations are required:
N index (6.29)
2
P index - [ -A N IP ) ^ A ^ P /K )] (6.30)
2
K . m « P IIQ -A K IN )] (6.31)
2
A value must be calculated for each of the above functions N/P, lOOP/K, and K/N. This requires two
equations, depending upon whether the measured nutrient ratio (i.e., N/P) is smaller or larger than the norm value
determined from the boundary line survey (i.e., n/p):
if NIP > nip then

(6.32)
f r example
jo 1 jyNIP) = ( -----
n /p
-- 1,] -------
1 0 0 0
( nip ) CV
OR
if NIP < nip then
nip 1000 (6.33)
fo r example f{NIP) 1-
NIP c v
The appropriate equations (6.32, 6.33) are calculated for each nutrient ratio, which allows the indices to be
determined as shown in Eqs. 6.29 to 6.31. If the actual analysis for (N/P) = (n/p) = 1, then f(N/P) = 0. N/P is
optimum. However, if N/P is greater than n/p, then f(N/P) is a positive number. If less than n/p, f(N/P) will be
negative. The 1000 multiplier is 100 x 10, with value 10 being included to give the resultant indices convenient
numbers.
Assume that a foliar analysis of head lettuce provides us with N/P = 4.87/0.29 = 16.79, lOOP/K = 29/7.43
= 3.90, and K/N = 7.43/4.87 = 1.53. Then:
435
(6.34)
(1) NIP > nip-. = 0.85: CV = (10Q)(3-04) ^ 3 3 ^ 5
9.06 9.06
1000 ^
J{NIP) = 0.85 = +25
33.55
(6.35)
(2) m P I K < \0 0 p lt 1 - 1 : ^ = -0.42: CF= = 46.58
3.90 5.56
filOOPIK) = -0.42
1000^ = -9
46.58
(3) KIN < kin-. 1 = -0.71: CV = = 4 5 .4 2
(6.36)
1.53 2.62
1 0 0 0 '
AKJN) = -0.71 = -16
45.42
The values +25, -9, and -16 can now be substituted into Eqs. 6.29 through 6.31 :
2 5 + (-1 6 ) _ 3
(6.37)
N index =
2
-2 5 + (-9 ) _ j., (6.38)

P index =
2
-9 -(-1 6 ) _ , (6.39)
K index =
The above solutions show that phosphorous is deficient, whereas nitrogen and potassium are sufficient. In
more complex cases, it is necessary that the computations outlined above be carried out in a suitable software
program, and that a database of foliar analyses be built up for determination of norm values for greenhouse crops.
Sanchez et al. [1991] carried out the above process for five macronutrients and four micronutrients, concluding
that DRIS was a useful tool in diagnosing the nutritional status of lettuce. Although the norm values were
obtained outdoors, there is no reason that such values would not be applicable to crops in protected cultivation.
The application of DRIS to field and orchard crops has found increasing use in the past decade. Several
modifications have been proposed [e.g.. Parent and Dafir, 1992]. One of the more recent summaries on plant
tissue analyses has been Jones’ [1991]. It is, according to Jones, doubtful that DRIS interpretation will ever be
used exclusively. It is a complicated procedure for which there is insufficient data on greenhouse crops by which
one can determine norm values. Jones, [1991] cited several publications suggesting that DRIS provided no better
diagnoses than the usual CVA or SR methods. For Colorado field conditions, Soltanpour [personal
communication, 1994] felt that DRIS had no application. There were instances where an excess of an element,
such as zinc, indicated deficiencies of the others when that was not so. Thus, Soltanpour felt that DRIS
represented an overly complicated method where other methods were more suitable. However, the system has
been utilized by foreign governmental agencies as a means to handle chemical requirements of their cropping
systems. Not having had direct experience with DRIS, offering conclusions is difficult for me. Still, the use of
boundary defmitions as illustrated in Figs. 6-13, 6-14 and 6-25 appears to me to be an excellent development for
greenhouse use given the much higher yields usual to such production systems.
IV. NUTRIENT SUPPLY Table 6-14. Partial reproduction of Sanchez et

al.'s Table 1, showing DRIS norms for crisphead
Aung [1974] stated that the root system remains, from a lettuce with standard errors [1991].
physiological and practical standpoint, the primary site of ion Nutrient ratio Mean s
uptake. Foliar application of nutrients may be a satisfactory
N/P 9.06 3.04
method of dealing with certain problems of deficiency, but such
N/Mg 8.18 3.18
practice is minor. In this section, we deal with the complexity of
nutrient availability and the methods of soil analyses and lON/Mg 1.09 0.64
practices that allow maximum profitability. Roots, as specialized lOOP/K 5.56 2.59
organs, are in constant competition for available energy. Root P/Mg 1 .0 0 0.49
growth is synchronized with the morphological state of the P/Na 2 .8 6 2.36
shoot. In the vegetative stage, shoot, and root growth continue lOOOP/Fe 3.42 2 .0 1
linearly, but with the shoot at a faster rate. With the approach of lOOOP/Zn 9.24 3.79
flowering and fruiting, root growth slows or ceases abruptly as
lOOP/Mn 1.67 1.50
the result of energy shortage from the shoot. There is a decrease
lOOP/B 1.39 0.83
in the shoot-root ratio during reproduction. Physical and
chemical factors, plus pruning, moisture, and radiant energy can K/N 2.62 1.19
modify the shoot-root ratio. O.IK/Mg 1.95 0.73
The presence of roots in the substrate -even in hydropon lOK/Mn 2.96 2 .2 2
ics- can markedly affect nutrient availability. In Section 5.Ill, lOK/B 3.75 1 .2 1
water supply, considerable attention was given to factors limiting lOCa/N 4.41 2.09
root activity and the influences on water uptake. In this section,
lOCa/K 1.93 1 .1 2
we continue the discussion of root growth and environmental
Ca/Mg 3.23 1.34
factors as they affect nutrient supply. We shall also consider
some common practices about how nutrients are applied in lOOCa/Mn 4.62 3.98
greenhouse production. This is consideration of concentrations lOOCa/B 5.35 2.46
required in the substrate for maximal response according to a lOOMg/Mn 1.43 1.15
grower's objectives as contrasted with the previous discussion of lOOMg/B 1.63 0.60
nutrient concentrations in the foliage. lOONa/N 5.45 3.85
lOONa/K 1 .8 8 1.57
A. THE ROOT ENVIRONM ENT
lONa/Ca 1.64 0.95
The root, protected by the root cap, grows through the soil,
lONa/Mg 4.34 3.61
or substrate, so that it continuously moves into regions that have
yet to be depleted of nutrients. Wadleigh [1957] reported on a lOOONa/Fe 2 .2 1 1.56
study of a single rye plant grown in 28 ^ s soil for 4 months. This lOOONa/Zn 4.55 2.80
one plant had 13.8 million roots with a length of more than 620 lOOONa/Mn 4.62 3.76
km, and a surface area of 237 m^. The root hairs numbered about lOOONa/B 6.53 4.46
14 billion with a combined length more than 10600 km and a O.lFe/N 4.62 2 .8 8
surface area of 397 m^. A remarkable amount of linear growth O.lFe/K 2.09 1.41
occurs with an average root system unless something restricts
O.OlFe/Ca 1.18 0.70
growth. The volume intercepted varies with species and soil
O.OlFe/Mg 3.68 2 .6 6
conditions from 0.1 to 2% of the upper 15 cm soil layer [Barber,
1974] (see Section 5.III.B). A depletion shell is formed about the Fe/Zn 3.23 1.62
root as ionic uptake occurs. Depletion of, for example, phospho Fe/Mn 4.35 2.93
rous, will occur, beginning at the root tip, and extending back Fe/B 5.45 4.28
through the root hairs that begin several millimeters behind the O.lZn/N 1.56 1.38
root tip. The root apex and elongation zone will have several Zn/K 6.80 4.06
hours for nutrient uptake before root hairs begin to emerge. A
O.lZn/Ca 4.29 3.36
depletion shell is already formed by the time root hairs emerge
O.OlZn/Mg 1.24 0.85
with an outer limit between 1 0 0 to 2 0 0 pm from the root surface
[Clarkson, 1985]. This volume, close to the root, possessing Zn/Mn 1.61 1.33
characteristics often totally different from the bulk substrate, will Zn/B 1.97 1.83
437
be considered in the following paragraphs. The volume is commonly called the “rhizosphere,” and the reactions
that can occur within it often account for the various plant responses that serve to confound even knowledgeable
growers. One may note that this volume about the root is much smaller than the value given for a sphere of
influence for water uptake in Section 5.III.A.
In a nutrient-rich zone, growth response, even to highly mobile forms as N 0 3 ‘, results in the formation of
more lateral roots. Apparently, roots tend to elongate at rates commensurate with the rates at which most mobile
ions diffuse [Uren and Reisenauer, 1988; Baldwin et al.,1972]. Root growth is a principal means of nutrient
acquisition in unconfmed soils as found in ground culture.
The effect of opportunistic root growth on nutrient uptake in usually “low-level” fertility field soils is highly
important in greenhouse culture of cut flowers and some vegetables. Most production of such crops in the
Mediterranean region, Colombia, Kenya, and the U.S. is in essentially unconfmed substrates. This situation
should be taken into account when considering most of the fertilizer recommendations that deal largely with
limited volume, highly porous substrates (Section 5.III.F). Depending upon the volume to which the root system
is restricted, depletion shells will eventually overlap so that, in container production, a root system will compete
with itself for available nutrients. Even in flow-through culture systems, reducing volume available for root
growth will reduce growth, assuming equal nutrient concentrations. The effects of substrate volume were
discussed in Section 5.III.F.2. Personal observation over a period of 30 years suggests, for example, that roses
do better in a large volume as contrasted with small rockwool blocks or peat bags. It is sometimes easier to
increase volume rather than attempting to increase nutrient concentration with its consequent dangers of high
salinity. Before the 1960s, a significant portion of commercial rose growing in the U.S. was in raised, wooden
benches. That changed so that nearly all cut-flower rose production is in the ground now. There are situations,
from an economic viewpoint as well as safety, that require production in unconfmed substrates. The levels of
nutrients such as phosphorous and potassium must be higher in a pot because of competition. Bray [1963] stated
that pot experiments were valueless for establishing fertilizer requirements for field-grown crops. Nevertheless,
a significant part of greenhouse production is in small containers. The grower must be careful to separate the
different conditions in handling fertilizer adjustments.
1. Ion M ovement
With ion uptake by a root, concentration of a particular element will decrease to a minimum at the root
absorbing surfaces. Concentration gradients occur, and these, combined with mass water flow, cause cation,
anion, and soluble organic molecules to move toward the root. Movement of the root into spaces formerly
occupied by soil particles causes an increase in bulk density so that steeper concentration gradients may occur
[Barber, 1974]. The movement of ions in response to a concentration gradient has been discussed in detail by
Noble [1991]. For small molecules, the time to cross 50 pm - a typical cell width- would be about 0.6 sec. For
1 meter, however, the time for diffusion would be 8 years. One has to keep in mind the scale involved when
interpreting results. With roots in intimate contact with soil particles, and within the rhizosphere, movement of
molecules and ions will be rapid when one considers the dimensions encountered. The diffusion rate can obvious
ly be increased by increasing concentration in the bulk soil, or by reducing the distance the ion must move to
reach the root. Boundary layers in which there is no turbulent mixing occur at the interface between any fluid and
a solid. The thickness of an unstirred layer can be reduced but never eliminated. Vigorous stirring and aeration
of plants in fluid hydroponic systems are means to reduce the diffusion resistance, and this is operative in such
methods as the Nutrient Film Technique (NFT) [i.e.. Barber, 1962; Olsen, 1953b].
Increasing the moisture content of a substrate will increase the cross-sectional area for diffusion by reducing
tortuosity (Fig. 6-23). Up to a maximum, increasing bulk density will also increase diffusion rates. These
phenomena probably account for the general observation by growers of enhanced plant response when clay is
added to a high organic or porous substrate. Where the nutrients are supplied by mass flow, the transpiration rate
and solution concentration will influence the nutrient amount supplied. Where the salt concentration is low (<10'^
M), it is generally accepted that transpiration rates will not affect nutrient supply to the root. There are
insufficient ions to be supplied and additional ions must be provided by diffusion [Barber, 1962]. At high salt
levels, as generally encountered in greenhouses, mass flow can significantly affect nutrient supply. Under these
conditions, the transpiration rate will influence plant nutrition. The relationship between high and low soil salt
concentrations was illustrated by Barber [1962], reproduced in Table 6-15. When converted to equivalent N 0 3 ',
H2 P 0 4 - and SO4 , the concen Table 6-15. Relationship between ion concentration in the soil solution versus that
trations in the low salt treatments within Z ea m ays, compiled by Barber [Barber, S.A. 1962. S o il S cien ce, 93:39-49].
ranged from lO'^' (S) to lO'^^ M (K)
Concentration (ppm) Ratio of plant content to
as contrasted to 10^ ^ (S) to lO’^"^M " lowest and highest soil
(K) for the high salt levels. Plants Soil solution Zea mays solution levels
appear able to regulate ion uptake Ion content
depending upon external ion Low High Average Low High
concentration. Soil analyses, there Calcium 8 450 2200 275 4.9
fore, are not always indicative of Potassium 3 156 20000 6666 128
what is occurring in the plant. In the
Magnesium 3 204 1800 600 8.8
winter conditions of northern
Nitrogen 6 1700 15000 2500 8.8
Europe and U.S., the low transpira
tion rate is, in effect, a reduction in Phosphorus 0.3 7.2 2000 6000 278
nutrient supply as contrasted to the Sulfur 118 655 1700 155 2.6
situation in summer conditions or
arid regions such as Colorado or
southern California. Sometimes,
where the movement of ions to the root exceeds the uptake rate, accumulation at the root interface may occur.
Examples are the accumulation of iron oxides or calcium that can precipitate as CaS0 4 or CaC 0 3 [Barber, 1974].
The cation exchange capacity of the substrate, on the other hand, reduces the diffusion rate as compared to
that in water. When ions diffuse, they diffuse either as a cation-anion pair or as “counter diffusion” in which an
ion such as may diffuse in one direction and Ca^^ moves in the opposite direction. Removal of cations (e.g.,
K^, Ca^^, etc.) from the negatively charged clay or humus micelles requires replacement with an equivalent
number of ions (Fig. 6 -6 ). Inorganic ions released by a root are in exchange for cations and anions absorbed by
the root (see also Section 4.III.D.4). Significant ionic release from roots will occur, according to Barber [1974],
when cation and anion absorption are unequal. When more cations are absorbed than anions, then is released
-such as happens when NH 4 ^ is the predominant nitrogen supply. When more anions are absorbed, HC 0 3 " ions
can be released [Riley and Barber, 1969; Moore, 1974; Marschner and Rdmheld, 1983]. The pH near the root
changes significantly. The situation varies with species and zones along the root plus the fact that more than one
species may be grown together, resulting in a different pH response [Marschner and Rdmheld, 1983]. The general
effect of ammonium and nitrate uptake on pH, which has been repeated several times in this chapter, should not
be taken to be universal under all conditions. Thus, Grinsted et al. [1982] found that growing rape in a P-free
solution with only NO 3 ' as the nitrogen supply resulted in a pH decrease of 2.4 units, later increasing toward the
end of their experiment. pH changes greater than one unit in the rhizosphere are usual.
The ionic and molecular exchange in the rhizosphere, therefore, can affect acidity so that nutrient availability
can be increased or decreased (Fig. 6-7), depending upon the soil minerals available and relevant concentrations
[Barber, 1974; Moore, 1974; Lindsay, 1972b; 1984]. In fact, a lower pH in the rhizosphere can go far to increase
phosphorous and iron availability. Both elements are considered immobile in soils as compared to nitrate. There
will also be striking effects of pH on manganese, zinc, copper, calcium, etc. availability that will be examined
later. Quite often, a nutrient insufficiency may cause a corresponding plant reaction that tends to compensate for
the insufficiency. Examples of such reactions have been cited by Korcak [1987]. The work by Lang et al. [1990]
showed that Ficus benjamina could reduce iron at the cell wall that was four times greater than that of F.
marginata. This ability was stimulated by low Fe concentrations on F. benjamina but not F. marginata.
2. Microbial Population in the Rhizosphere

Included in the rhizosphere, defined above, is a population of microorganisms that are more abundant than
in the soil remote from the root. Not only may pH be different by as much as 1 to 2 units, but microscopic
examination reveals that many parts of the root can be covered by bacteria 10 to 40 cells deep [Rovira and Davey,
1974]. Fungi and bacteria are often found arranged in lines along the root axis and mycorrhizal fungi may
traverse all the defined zones from the outer bulk soil to several cells deep within the cortex. The change in
microbial density with distance from the root of blue lupine is shown in Table 6-16. As the root grows through
439
the soil, there is successive colonization of the Table 6-16. The rhizosphere microbial population of 18-dy-old
root from the soil. The root tip, according to L u p in u s a n g u stifo liu s seedlings [adapted from Rovira and Davey,
Rovira and Davey [1974], is nearly without 1974].
fungi with few bacteria, and the root proximity Microorganisms (1000s per g oven dried
stimulates spore germination of many Distance from soil)
nonpathogenic species. The major root (mm)
proliferation, as cited by Ur en and Reisenauer Bacteria Streptomyces Fungi
[1988], takes place in 2 zones near the root Root surface 159000 46700 355
hairs where lateral roots emerge and cells of
0-3 49000 15500 176
the epidermis and cortex degenerate. Bacteria,
3-6 38000 11400 170
actinomycètes, and fungi of the rhizosphere
affect the host plant through their influence on 9-12 37400 11800 130
nutrient availability, growth, root morphology, 15-18 34170 1 0 1 0 0 117
and nutrient uptake processes. For example, a 80 27300 9100 91
high microbial population means that oxygen
uptake for respiration will cause a lower O 2
concentration, and, conversely, the CO2 level
will be higher in the rhizosphere. These will affect redox potentials as well as nutrient solubilities in the soil solu
tion.
As will be noted in the next section, part of the reason for high microbial activity close to a root is the
significant root exudation of many compounds (Table 6-17). The concern with root death and disease organisms
in hydroponics, discussed in Section 5.III.G.2, is due mostly to the failure to appreciate the normal microbial root
population. Using the microorganisms might be better than to attempt sterile culture as suggested by the complex
apparatus in Fig. 5-54.
3. Root exudation Table 6-17. General summation of root exudates according to Uren and Reisenauer
The many microbes in the [1988].
rhizosphere are partly or wholly Product Compound
dependent upon carbon
Diffusâtes Sugars, organic acids, amino acids, water, inorganic ions, oxy
compounds that leak or are exuded
gen, riboflavin, etc.
from the roots. An outline of the
Excretions CO2, HCOj', protons, electrons, ethylene, etc.
materials is given in Table 6-17.
The compounds that come from Secretions Mucilage, protons, electrons, enzymes, siderophores, allelopa-
thic compounds, etc.
intact root cells include, besides
mucilage and other organic Root debris Root cap cells, cell contents, etc.
compounds, carbon dioxide,
ethylene, nutrient ions,
bicarbonate ions, protons, and
electrons. These may either leak from epidermal cells (diffusâtes), be actively excreted (excretions), or be actively
secreted (secretions) [Uren and Reisenauer, 1988]. Release of organic products is increased when high-salt roots
are suddenly exposed to distilled water, acidic solutions, low calcium concentrations, iron deficiencies, or low
phosphorous levels, etc. Generally, anything that increases plant growth will increase exudation.
The release of materials by a root represents a marked drain on plant resources. As noted by Barber and
Martin [1976], cereals growing without microbes released material equivalent to 7 to 13% of the total dry matter
production over a period of 3 weeks. In unsterilized soil, the losses increased to 18 to 25%. Barber and Martin
also suggested that roots respond to dry situations by exuding additional mucilage. Clarkson [1985] concluded
that the many carbon compounds lost by roots can range from 2 to 2 0 % of the carbon fixed in the photosynthetic
process. There is a functional equilibrium between the shoot and root system. A significant change in rate of
shoot growth is transmitted to the roots and likewise, changes in root growth affect shoot growth. Brown and
Scott [1984] conclude that cultural systems that minimize the amount of photosynthetic energy the crop must
expend on the root system should lead to higher economic yields.
The layer of mucilage, starting in the root cap zone, is about 1 to 10 pm thick. The mucilage helps ensure
good contact with the substrate and helps water and ionic transfer. Root cap cells can be sloughed off, yet remain
viable, still producing mucilage for up to 3 weeks. Although plants adequately supplied with water and nutrients
in sterile culture do not need microorganisms, few plants complete their life cycle without infestation with micro
organisms. Under commercial conditions, certainly, complete sterility is highly unlikely.
The direct effects of root exudates involve increased solubility and uptake of immobile nutrients such as Fe,
Mn, and P. Most iron is insoluble. For it to move to the relevant membranes, iron must be in solution. Protons
and electrons, under suitable conditions, can contribute to the mobilization of iron. Most manganese in soils
exists as very insoluble hydrous oxides. Again, both protons and electrons must be involved. In the case of
phosphorous, the release of organic acids, enzymes such as phosphatases, and general acidification of the
rhizosphere act to increase P availability. Indirect effects include the influences on the microbial population and
the effects of these organisms on nutrient solubility, or the release of siderophores that chelate and mobilize Fe.
These factors of ion mobility, microbial population, and root exudation also have marked effects on soil-
borne pathogens. Robson and Abbott [1989] showed that Pythium root rots are likely to be most severe in
conditions where the host plant is under greatest stress. For F.oxysporum on tomatoes, plants grown in acidic
conditions before inoculation are more severely affected than those grown in alkaline conditions before
inoculation and then transferred to acidic conditions.
Even in hydroponic systems, exudation from roots must occur. It would appear that these exudates could be
utilized as a means to improve growth if we knew more about them.
4. M ychorrizae and Symbiotic Relationships

Many fungal organisms form highly intimate and mutually beneficial relationships with plant roots
-sym biosis- and are classified as mychorrizae. They are obligate parasites. VA mychorrizae have never been
grown separately in culture nor have they been seen to reproduce sexually. They depend upon the host plant to
give them sufficient carbon for growth. There are three general classes [Maronek et al., 1981; Gerdemann, 1974]:
1) “Ectomycorrhizae,” which are most common among forest and ornamental tree species and form a thick weft
or sheath of hyphal strands around the feeder roots known as the “mantle.” The mantle replaces root hairs with
fungal strands, greatly increasing absorptive root surfaces, and it can reach outward from the root by several
meters. Hyphae also penetrate into the intercellular spaces of the root cortex, forming an interconnecting network.
2) “Endomychorrizae,” the most common “vesicular-arbuscular” (VA) mychorrizae, called so as the result of the
specialized vesicles and arbuscules formed within the cortical root cells. These are the most widely distributed,
being found in many herbaceous shrub and tree species. There are no discernible morphological changes in the
external root structure, and a loose hyphal network radiates outward several centimeters or more from the root.
And 3) “Ectendomychorrizae,” which exhibit characteristics of the previous two types. Koide and Schreiner
[1992] cite the literature to the effect that 85 to 90% of the known seed-bearing plants (angiosperms) form this
symbiosis even though there are only 120 described species of VA mychorrizae.
VA mychorrizae have been examined on Easter lily [Ames and Linderman, 1977; 1978], strawberry
[Holevas, 1966], poinsettia [Barrows and Roncadori, 1977], tomatoes, cucumbers, etc. Orchids, especially, are
known to require mychorrizal infection for survival, and woody species such as citrus require such infections for
healthy growth. The review by Maronek et al. [1981] is particularly useful to horticulturists. Overall, the benefi
cial effects of symbiosis with higher plants occur in substrates with low fertility levels. Improved survival and
growth of plants in landscapes or newly reclaimed land areas have been the most frequent claims [e.g., Johnson
and Crews, 1979; Crews et al., 1978]. In particular, improved phosphorus nutrition has been found with the
ability of VA mychorrizae to transfer phosphorus to its host. There is evidence for marked improvement in copper
and zinc nutrition in some species [Clarkson, 1985]. The obvious principal benefit of mychorrizae is the
expansion of hyphae far beyond the nutrient-depleted shells immediately about the roots. High fertilizer
concentrations apparently reduce benefits from mychorrizae to the host plant in terms of improved nutrition, as
well as reducing mychorrizal infection. Since most greenhouse container production is in pasteurized or sterilized
substrates, at high nutrient concentrations, it is doubtful that deliberate mychorrizal infection as a standard
practice in greenhouse production will be found cost effective. To my knowledge, direct comparisons, applicable
to commercial conditions, on the benefits of deliberate mychorrizal infection have not been adequately tested.
441
B. BASIC FERTILITY RELATIONSHIPS

The wide range of conditions in greenhouse production makes it difficult to provide simple recommendations
that are applicable in all circumstances. Crop production may be: 1) over long periods in unconfined soils, 2) in
limited volumes (container production), and in substrates ranging from highly organic to essentially inert with
periods ranging from a few weeks to several months (i.e., seedling transplants versus slow growing, decorative
plants), and 3) in various hydroponic systems, ranging from those in which an inert substrate is periodically
flushed to continuous flow systems in which the solution may be expended or reused. Any division between a
true hydroponics culture and many container practices disappears.
All systems require consideration of carbonate equilibria and chelation, the latter principally with
micronutrients. The standard practice of “fertigation” or injection of nutrients into the irrigation supply or nutrient
solution can be applied to all systems with some variation. This latter practice, which serves as a semiautomatic
fertilization procedure, simplifies cultural practice. On the other hand, liming practices are important in field soils
and in container production where the substrate can be highly organic or consist partly of mineral soils.
1. N atural Fertility Levels

In an unconfined field soil, Table 6-18. Total content of various elements in soils [Abridged from Lindsay,
one deals with a large mass and 1979].
volume. A 15 cm plow layer
Selected average for soils
contains 1500 m^ ha‘^ and weighs
Common range for
more than 1 0 0 0 tons, depending Element soils ppm Log Molar concen
upon bulk density. Though a crop (ppm) tration at 1 0 %
may be restricted to longitudinal moisture
beds, we can assume that root Aluminum (Al) 10000-300000 71000 1.42
growth is essentially unrestricted
Arsenic (As) 1-50 5 -3.18
and concentration of nutrients in
Boron (B) 2 -1 0 0 10 -2.03
the soil solution is low (<10'^ M).
Given these conditions, root Carbon (C) 20000 1 .2 2
extension for nutrient acquisition, Calcium (Ca) 7000-500000 13700 0.53
at least during early crop growth Chlorine (Cl) 20-900 100 -1.55
stages, will be important. The Cobalt (Co) 1-40 8 -2.87
volume to be dealt with means that Copper (Cu) 2 -1 0 0 30 -2.33
significant physical modification
Fluorine (F) 10-4000 200 -0.98
can only be achieved over several
Iron (Fe) 7000-550000 38000 0.83
years with continuous addition of
amendments which may be bark, Iodine (I) 0.1-40 5 -3.40
wood chips, manure, leaf mold, or Potassium (K) 400-30000 8300 0.33
such others as listed in Table 5-7 Magnesium (Mg) 600-6000 5000 0.31
(Sections 5.III.C.4 and 5.III.F.3). Manganese (Mn) 20-3000 600 -0.96
Often, additions used will be those Molybdenum (Mo) 0.2-5 2 -3.68
most readily available and cheap. Nitrogen (N) 200-4000 1400 0 .0 0
This eliminates peat moss,
Sodium (Na) 750-7500 6300 0.44
vermiculite, perlite, etc. since
Nickel (Ni) 5-500 40 -2.17
many of these require special
handling and may have to be Oxygen (0 ) 490000 2.49
shipped long distances. Also, as Phosphorous (P) 200-5000 600 -0.71
the organic content is raised, the Sulfur (S) 30-10000 700 - 0 .6 6
rate at which carbon is lost also Selenium (Se) 0 . 1 -2 0.3 -4.42
increases. The maintenance of Silicon (Si) 230000-350000 320000 2.06
humus at a high level is difficult Titanium (Ti) 1 0 0 0 -1 0 0 0 0 4000 -0.08
and expensive. Brady [1974]
Zinc (Zn) 10-300 50 -2 .1 2
pH
Fig. 6-26. The effect of pH on activities of several essential nutrients in soil solution. Adapted from unpublished data
provided by Lindsay [1993] as a general summary of the more important ion equilibria relationships to be found in soils.
443
stated that it was unwise to hold organic matter above a level consistent with crop yields that pay best. Contrary
to the situation of summer crop production -at least in temperate climates- continuous production, such as with
roses, may occur, requiring a continuous fertilization program that varies with the season. In cut chrysanthemum
culture, pasteurization and replanting of three or more crops per year will put a serious drain on soil fertility.
Carnations may also be grown for more than 1 year if soil-borne pathogens are not a great problem. Numerous
other cut-flower species are commonly grown in the ground and may be present for several years.
As a result, there is a restriction upon what can be done -both in terms of physical and chemical change.
The grower should have some idea of basic natural fertility level of the soil, and means to improve fertility. If
one looks at the total amount of the various nutrients in the soil, concentrations can be quite high. The abridged
Table 6-18, from Lindsay [1979], provides an approximation of the average total elemental concentration in soils.
The final column is the maximum level in the soil solution if all the elements were to dissolve in water present
at 10% of the dry weight of the soil. Expressing concentration as a logarithm permits calculation of actual
elemental concentration and moisture content. If a test shows a concentration of 300 ppm copper, then the ratio
(300/30) = 10’ can be added to 10'^^^ M to give 10’’^^ M for the maximum concentration of Cu at 10% moisture
content. If the moisture content is 40% on a dry weight basis, then the ratio (10/40) = 10'°^° can be added to -1.33
to give 10’’^^ M. This is the maximum concentration of Cu in the soil containing 300 ppm Cu at 40% moisture,
or 0.0117 M = 11.7 mM f ’ . The formula can be written:
log[Cu^^] - - 2 .3 3 +/og(300/30)+ /og(0.1/0.4) (6.40)
Note, in Table 6-18, the high total levels of aluminum, calcium, iron, and silicon. Although the total amounts of
these and other elements are more than enough to meet plant requirements, the amounts in solution are important.
Thus, one important criterion of soil tests is to determine amounts readily available to the plant. Even though Ca
appears high for most soils in Table 6-18, a level of 10'^^ M (0.003 M) is common in solution. At low pHs, AE^
may be high enough to be toxic. Silicon leaches from soils and where a soil is highly weathered in humid regions,
Si can be very low. Responses to silicon fertilization have been observed [Foy, 1992; Takahashi and Miyake,
1982].
a. Examples o f pH Relationships
In Section 6 .II, it was explained that nutrient concentration in a natural soil will depend upon the minerals
present which vary with soil origin and climate (Fig. 6-5). If the controlling mineral is known, then solution
concentration at equilibrium can be calculated. Examples were provided in Eqs. 6.17 through 6.22 and Fig. 6-7.
Fig. 6-7 showed the activity of different micronutrients as a function of pH, whereas Fig. 6-4 showed the varia
tion in iron and phosphorous as influenced by both pH and pe (redox). Fig. 6-2 compared the plant requirement
for iron (10'^ M) with the total soluble iron to be expected in solution as influenced by pH.
The theoretical equilibrium concentrations of the ions Fe^\ Fe^"", Zn^^, Mn^^, and Cu^^ are repeated in Fig.
6-26. They differ from those in Figs. 6-2 and 6-7 as the result of different minerals and conditions. In practice,
the concentrations as a function of pH depicted in Fig. 6-26 are not exact since equilibrium conditions are not
likely. The lines do provide a starting point. Note that solution activities of Ca^^ and Mg^^ remain constant up to
pH values between 7.5 and 8.0. Between pH 7.5 and 8.0, gypsum (CaS0 4 *2 H2 0 ) and calcite (CaC0 3 ) can coexist.
Above pH 8.0, calcite precipitates and both Ca^^ (Eq. 6.4) and Mg^^ levels drop rapidly below the usual
concentrations of 10'^^ for Ca^^ and 10'^ for Mg^^. These are levels of 3 and 1 mM Ca^^ and Mg^”^, respectively.
Ca^^ and Mg^^ solution levels in calcareous soils are controlled largely by the solubility of their carbonates and
the CO 2 concentration [Lindsay and Moreno, I960]. The effect of the basic mineral on ion concentrations, for
Mn^"“, is shown by a shift upward when in equilibrium with manganite (y-MnOOH) versus pyrolusite (P-Mn0 2 ),
and a still higher level with rhodochrosite (MnC0 3 ). In the latter case, the CO2 level in the soil will be instrumen
tal in controlling Mn^^, whereas with manganite and pyrolusite, the redox potential will be important.
The formula for calculating the Fe^^ line is:
lo g F e ^ ^ = \5 .1 A - { p e + p H ) - 2 p H (6.41)
Therefore, this line will shift back and forth as determined by the redox potential p e + pH. If the pH is 7 .0 , p e
will equal 5 { p e + pH = 12). The concentration of Fe^^ at pH 7.0 is 10’’^^ or 0.00005 pM Fe^^. Iron level is
insufficient (see Fig. 6-2). On the other hand, if pH drops 2 units to 5.0 (a 100-fold increase in hydrogen ion
level), the Fe^^ concentration will be 10'^^^ M or 5 pM - a 100000-fold increase in Fe^^. Several authors state that
plants take up Fe^'*' only [e.g.. Barker and Mills, 1980]. One notes, however, that the Fe^^ concentration is
nowhere high enough to be of direct importance in iron nutrition. The fact that only Fe^^ is taken up is largely
due to the low concentration of Fe^^.
Influence of a primary mineral can also be noted by the shift in AP^ line, depending upon whether the ion
is in equilibrium with gibbsite (y-Al(OH)3 ) or quartz (a-Si 0 2 ). The effect of pH on molybdenum would be
opposite to those equilibria shown in Fig. 6-26. Concentration of Mo 0 4 ^‘ increases with increasing pH so that
at pH 5, activity will be 10'^ ^, versus 10'^ M at pH 8 (0.004 versus 1.0 pM). The relationships of boron are not
well understood. Lindsay [1978; 1991] indicated activity levels for H 3 B 0 3 ° of 10'^ in 1978 and 10'^^ in 1991, the
latter based on newer information, regardless of the pH between 5 and 11. The ions H 2 B 0 3 *, HB 0 3 ^', and B 0 3 ^'
are present in solution at pHs above 10: otherwise the neutral molecule H 3 B 0 3 ° is dominant.
One may note that the usual range of micronutrients in soil solutions as calculated by the various authorities
will vary from about lO'"^ to less than 10'^^ M fP Deficiencies will generally occur when concentrations drop to
10'^ or 10'^ M - a 100-fold concentration difference. Unless added, levels of Ca^^ and Mg^^ will be less than 10'^
M.
b. N i tr o g e n R e l a t io n s h ip s :
In most cropping systems. Baker and Mills [1980] stated that available nitrogen is the most limiting factor
in crop growth, complicated by the fact that less than 50% of the nitrogen fertilizer applied may be used. More
than 90% of the N in the surface soil layers is organically combined [Stevenson, 1991], although some authors
suggest 95% or more [i.e., Legg and Meisinger, 1982]. Nitrogen is also one of the most widely distributed
elements in nature, with the highest amount fixed in rocks and sediments.
A general perspective of the nitrogen cycle as modified in greenhouse production is given in Fig. 6-27.
About 2 to 3% of the nitrogen is released yearly from the organic nitrogen pool by the process known as “min
eralization.” The product is N H /, of which large amounts are used by the soil microbial population, some by
the crop, some by fixation in clay minerals and organic matter, or it may be lost as NH 3 to the air. In well-aerated
soils, with an abundance of basic cations, “nitrification” occurs, in which NH 4 ^ is converted to N 0 2 ' by the
bacterium N i tr o s o m o n a s and very shortly to N 0 3 ' by the bacterium N i tr o b a c t e r . Thus, NO 2 ', which is highly
phytotoxic, does not accumulate. Other organisms can oxidize the ammonium ion. Although most plants can
assimilate N H /, previous discussion (Section 6 .III.D.3) showed that N 0 3 ’ is commonly the form used. Small
amounts of many kinds of salts stimulate nitrification, whereas the application of large amounts of NH 4 ^
fertilizers to alkaline soils have been found to depress the second step, allowing toxic accumulation of N 0 2 '.
Similarly, urea may cause adverse effects since it is converted quickly to ammonium ion. All these compounds
can be lost to the atmosphere as N 2 , N 2 O, NO, and NH 3 . In the case where nitrate is reduced, the process is called
“denitrification.” Brady [1974] states that rapid losses can occur with heavy application of urea or ammonium
fertilizers, especially in alkaline soils. In greenhouse practice, particularly where soil fumigation or steam
pasteurization is practiced, buildup of NH 4 ^ can occur since the organisms in nitrification are killed. Heavy use
of manures can be particularly dangerous in this respect.
When organic matter is added to a soil, especially materials having a high carbon/nitrogen (C/N) ratio, the
microbial population, during decomposition, will claim most of the ammonium and nitrate for their own use. This
is part of the “immobilization” possible in soils. Nitrogen deficiencies can readily occur until after the
carbonaceous matter has been decomposed sufficiently to reduce the influence of the microorganisms. In fact,
I have seen severe damage from nitrogen excess when the large microbial population dies off after decomposing
a fine material such as sawdust or straw where the C/N ratio can range from 50:1 to 100:1. Well-balanced
nutritional conditions are represented by a C/N ratio of about 25 [Jansson and Persson, 1981]. Decomposition
445
Fig. 6-27. General soil nitrogen eycle. Additions to soil nitrogen in greenhouses by rainfall are unavailable. Nitrogen
fixation by bacteria unlikely to be of any importance in most greenhouse production.
Fig. 6-28. Schematic of changes that occur

in pH and nitrogen when organic-type fer
tilizers are used in peat-sand substrates. The
amount of nitrites will depend upon the
rapidity of the conversion to nitrate by bacte
ria. Excessive ammonium combined with
pasteurization can result in phytotoxicities
[Adapted from Bunt, 1976].
of organic materials, especially peat

moss, can be very rapid under
greenhouse conditions. The microbial
population may vary drastically, with
marked effect on nitrate and ammonium
concentrations in the soil solution. A
diagram showing the possible changes in
composts is presented in Fig 6-28.
The mobility of N H / in soils is
restricted since it will be attracted to
negative colloids, and it can be fixed
almost permanently on some clay
minerals such as vermiculite, illite, and
montmorillinites. Nitrate, on the other
hand, is highly mobile and readily lost
from the soil. This has led to
consideration of nitrate toxicities in
ground water in several parts of the
world. This will be addressed later.
Although there can be no rain to cause leaching in greenhouses, excess irrigation, combined with high NO 3 '
application, can drastically reduce the efficiency of N 0 3 ' uptake by a crop.
As discussed earlier (Sections 6.III.C.1, 6.III.D.3), plants can extract from solutions ions that are in very low
concentrations. One figure given for N 0 3 ' was 0.003 mM or 10'^^ M. Mengel and Kirkby [1982] summarized
the situation of nitrogen in the soil solution by stating that levels of N H / and N 0 3 ' can change rapidly
-particularly with respect to nitrate. N 0 3 ' can be as high as 20 to 30 mM after fertilizer application (10'^^ to 10'*^
M). In fertile soils, however, the range may be 2 to 20 mM, depending upon the rate of mineralization and plant
uptake. Clement et al.[1978] stated that roots of agricultural crops could be exposed to nitrogen levels ranging
from 1.43 pM to 1430 mM, or 10'^^ to 10°^ M. It will be noted later that, in greenhouse production, nitrate and
ammonium concentrations are almost always around 10'^ M or higher. This results, especially, from plant
requirements where they are grown in small volumes or in hydroponics.
c. Phosphorus Relationships
Phosphorus is the second most important soil
fertility problem in the world [Lindsay et al.,
1989] today. It is only a minor component in soils
with a large part in the organic fraction that is still
o
poorly characterized. The element has unique
solution properties, occurring as
“orthophosphate” ions (Fig. 6-29) absorbed by
plants (H 2 P 0 4 ', HP 0 4 ^'), and they are highly
reactive. The result is a very low solution concen
o
tration under usual conditions, requiring rapid re
plenishment from the solid phases. Peaslee [1981]
stated that phosphate ions may diffuse over a dis
tance of 0 . 2 to 2 mm to the roots, with solution
concentrations ranging from 1 to 2 pM (10’^ -
10'^ M), depending upon the pH. In restricted
volumes, the phosphorous application may have Fig. 6-29. The orthophosphate ion with P surrounded by four O
to be increased more than 2 0 times the usual field in a tetrahedral arrangement.
practice (see Section 5.III.F.2 and Tables 5-4
through 5-6). Soils exhibit sorption and desorption buffering properties by removing phosphate from solution
during enrichment and releasing it during depletion. When soluble sources are applied, the phosphorous is often
“fixed” or rendered unavailable to plants. As a result, the phosphorous added in fertilizers exceeds that removed
by crops by more than 24% [Brady, 1974], and in some areas, additions of P may be more than triple the removal
by crops. In contrast to nitrates, P is relatively immobile and seldom found in any significant concentrations in
the leach water.
Lindsay et al. [1989] list 65 materials formed from the reaction of phosphate fertilizers. A short list is
provided in Table 6-19 as an example of the bewildering number of minerals and compounds. The application
of phosphates (e.g., superphosphate. Table 6-5) results in interesting and complex chemical reactions as
investigated by Lindsay and others in the 1960s and 1970s [ Lindsay, 1959; Lindsay and Taylor, 1960; Lindsay
and DeMent, 1961; Watanabe et al., 1965; Lindsay and Stanford, 1960; etc.]. The situation close to a phosphate
fertilizer pellet is diagrammatically presented in Fig. 6-30. In the presence of a fertilizer granule, water is
condensed and moves into the pellet, which dissolves the chemical. The pH can be reduced below 2.0 so that
numerous aluminum, iron, calcium, ammonium, and other compounds are formed with the phosphate. Lindsay
and Taylor [1960] divided these chemicals into those formed at pHs below 4.0 and those above 6.0. Products
formed from polyphosphate fertilizers are a third group. As the solution, moving outward from the pellet, is
neutralized, complex phosphate products are precipitated. DCPD is one of the major initial compounds formed,
and where MCP (monocalcium phosphate) is applied, there is sufficient calcium to precipitate half the
phosphorous applied. Above pH 6.5, DCP (dicalcium phosphate) converts to OCP. Octacalcium phosphate is
unstable and gradually converts to apatite in neutral or alkaline soils, becoming less available. Lindsay and co
investigators have also documented many equilibria between phosphate ions and the numerous minerals to be
447
found in various soils Table 6-19. Abridged table of compounds formed from reaction of phosphate fertilizers with
[e.g., Lindsay and soils [Adapted from Lindsay et ah, 1989; Lindsay and Taylor, 1960; Lindsay, 1979].
Moreno, 1960; Boyle
Name Acronym Chemical formula
and Lindsay, 1986;
Lindsay, 1979; etc.]. The Variscite AIP 0 4 -2 H2 0
reactions are strongly NH 4-taranakite AI,(NH4)3H,(P0 4 )*-1 8 H2 0
influenced by pH and pe, K-taranakite Al5K3H,(P04)8-18H20
with some examples Monetite (dicalcium phosphate anhydrous) DCPA CaHP0 4
presented in Figs. 6-3 Brushite (dicalcium phosphate dihydrate) DCPD CaHP0 4 -2 H2 0
and 6-4. Separation as to
Octocalcium phosphate OCP CasH2(P04),-5H20
general types of
Hydroxyapatite HA Ca,„(P0 4 ),( 0 H)2
compounds can be based
on whether the soil is Fluorapatite FA Ca,„(P0 4 ),F 2
acidic or alkaline and the Strengite FeP0 4 '2 H2 0
basic reaction of the Vivianite Fe3(P0 4 )j'8 H2 0
fertilizer. Lindsay and Struvite MgNH4P0 4 '6 H2 0
Taylor [1960] classified Reddingite Mn3(P04)2-3H20
diammonium phosphate p-Tricalcium phosphate P-TCP p-Ca3(P0 4 )2
(DMP) and dipotassium
Hydrogen ammonium iron phosphate H3NH4 Fe3(P0 4 ),-6 H2 0
phosphate (DKP) as in
the neutral or alkaline
class as compared to
MCP or MAP
(monoammonium phosphate). Depending upon the relative concentrations of the respective reactants, phosphorus
can be rendered unavailable, or P can tie up other essential elements. Most of the reactions elucidated are
extremely slow [Lindsay and Moreno, 1960], with biological reactions responsible for frequent changes in phos
phate equilibria.
In general, the materials used in loamless substrates do not provide a natural phosphorus supply, nor do they
have the ability to fix or retain phosphorus to the same extent as mineral soils [Bunt, 1988]. Greater attention
must be given to maintaining phosphate levels in greenhouse mixes, with care taken to ensure solubility in the
H2 P 0 4 ' and HP 0 4 ^' forms. High concentrations encountered in mixing tanks for fertigation systems can result in
precipitation of calcium phosphates and others. This can also occur in alkaline irrigation supplies. The grower
winds up with a sludge in the bottom of his mixing tanks, or a stopped-up trickle irrigation system.
d. Potassium Relationships
The greatest part of potassium is bound on secondary clay minerals whose particle sizes are less than 2 pm
in diameter. Soils rich in clay are generally rich in potassium [Mengel and Kirkby, 1982]. Organic soils are likely
to have a much lower K content. Except for soils of a sandy nature, most mineral soils are comparatively rich in
K. The main source of comes from weathering of potassium-containing minerals such as vermiculite, illite,
montmorillonite, etc. As with other elements, the relationships between soil minerals and plant roots can be
divided into two or three major sections (Fig. 6-31): 1) K fixed as the structural element of soil minerals, 2)
adsorbed in the exchangeable form on the soil colloids, and 3) present in the soil solution. The situation with
respect to K parallels P and N by the fact that large portions are insoluble and relatively unavailable to plants.
Lime applications are likely to increase K fixation, and K in well-limed soils is not as likely to be leached out as
K in acid soils. Luxury consumption by plants can occur with K^, so that light and frequent applications are
usually superior compared to heavier and less frequent. Fertigation methods are an excellent means of potassium
fertilization. In container production, K in the tissue of pot plants is likely to be of the same amount as N, and
experiencing luxury consumption of K is usual.
Replacement of K by Na can occur -up to a point. The extent to which substitution can be made, however,
depends on the potential for Na^ uptake. There have been reports of the use of NaCl in solution culture, with an
increase in tomato yield under U.K. winter conditions [e.g., Adams, 1987]. Similar reports were cited by Mengel
and Kirkby [1982]. However, high Na^ levels have been shown to increase tipbum on lettuce
4^
00
o
p
:z:
c
B.
o'
Solution
W§t$r mm9§ m Solution d ilu tit end
mov«« out iipondf out-
word
SUPERPHOSPHATE S o il-fir fM litr

rtoetton zone
Fig. 6-30. Situation over time for a superphosphate granule in a soil. Water is condensed and moves into the granule, dissolving it within 6 to 72 hr. In acid
soils the major reaction product is DCPD at pHs below 2.0. The solution dissolves large quantities of Al, Fe, Mn, Ca, K, etc. As the front moves outward,
the pH gradually rises and much of the applied phosphorous is precipitated in combination with Al, Fe, and Ca. In alkaline soils, the fertilizer particle
dissolves to form basic calcium phosphates [Adapted from Lindsay and Stanford, I960].
449
[Sonneveld and Mook, 1983]

and accumulate in roses to
the point that leaf margins
may bum [Hughes and K + Weathering i
Hanan, 1977]. Overall, under Mineral -------------- > K Plant
semiarid conditions of Soil colloids — uptake
Colorado, minimum solution ^< - --------- K ' * '
concentration is needed "f" Adsorption Root
without sodium [Hanan, Solution system
1982b; 1984]. Schekel
[1971] recommended that if Fig. 6-31. Potassium relationships in soil.
the ratio of to Na^ was
1 :1 , carnations would
tolerate Na^ up to maximum levels of 6 mM of either ion in solution. The problem with most studies of this type
is the tendency to ignore the accompanying ion. There is considerable difference about whether Na^ is supplied
as NaCl, NaHC 0 3 , or another salt containing sodium [Sonneveld and Mook, 1983]. The behavior of the
accompanying anion can be completely different in soil versus a hydroponics system. Sodium minerals are too
soluble to precipitate in soils, and either accumulate as soluble salts in poorly drained soils or are leached,
whereas potassium can be re-fixed on clay minerals [Lindsay, 1979]. Overall, the combinations of Na^ and
with anions (e.g., S O /', Cl', HCO3', OH' and C03^') in most mineral soils are not important.
c. Sulfur Relationships
S O /' is the principal ionic sulfur form in solution and is not as strongly bound to soil particles as phosphate
-although the S content of plants is on the same order as P (Table 6-12). Also, significant amounts of S are taken
up from the air or from fertilizers containing sulfur. The atmosphere in industrialized areas is usually high in SO2 ,
and although sulfur dioxide is considered an air pollutant, it is also a source of sulfur when oxidized to SO /'.
Sulfur deficiencies have been noted in several areas, and responses to sulfur applications have been noted
[Mengel and Kirkby, 1982]. The movement to high analysis fertilizers, air pollution control, and replacement of
sulfur-containing pesticides has increased the occurrence of sulfur deficiencies. Lindsay [1979] states that the
solubility of S O /' in soils is limited by the solubility of gypsum (CaS0 4 -2 H2 0 ). When in equilibrium with soil-Ca
at 10'^^ M, S0 4 ^' is 10 M (7.9 mM). As Ca^^ activity is reduced by the precipitation of CaC 0 3 at pHs
approaching 8.0, S0 4 ^' activity can increase. Fortunately, plants are more tolerant to S0 4 ^' excess compared to
chlorides. Schekel [1971] reported that carnations would produce an acceptable yield at 20 mM/f S0 4 ^' in
hydroponic systems, as contrasted with Cl' concentrations at 2 mM. Generally, in greenhouse production, sulfate
fertilizers and elemental sulfur are used as means to reduce pH. This will be discussed later.
C. CARBONATES AND pH CONTROL

1. C arbonate and C O 2
Root substrates and also hydroponic systems can be considered as “open” with regard to carbonates since
CO 2 can escape from solution or return to precipitate carbonate materials. CO 2 dissolves in water to form
dissolved C 0 2 ° and undissociated carbonic acid, ■In solution, carbonic acid dissociates to give:
H,CO, H*+HCO; (6.42)
At pH 6.36, the ratio of (HCO3 ) to (H2C03°) is 1. The ratio increases 10-fold for each 10-fold increase in pH
(one unit), and vice versa with a decrease in pH. The bicarbonate ion also dissociates to ¥ t and C0 3 ^'. At pH
10.33, the ratio of these two is one with a change similar to that for H 2 C 0 3 ' and H 2 C 0 3 ° when pH changes. As
the result of C0 2 (g) levels in the substrate and pH, the concentrations of C 0 3 ^' and HCO3' can be calculated for
a CO 2 -H 2 Q system:
(6.43)
logC O l = -18.15 + 2pH +logCOj{g)
logHCO; = -1.% 2+pH^logCOîg) (6.44)
Table 6-20 provides examples of a Table 6-20. Distribution of carbonate species in solution as a function of
carbonate in solution at different CO^ CO^ concentration [Adapted from Lindsay, 1979].
concentrations, and Fig. 6-32 shows the CO,(g) co,(g) log log log HCO3 log CO,^
effect of pH on the fraction of various (atm) (kPa) CO,(g) H2C 0 3 “ (M) (M)
carbonate ions. The table and figure serve (atm) (M)
to emphasize that carbonate in some form 0.0003 0.030 -3.52 -4.98 pH-11.34 2pH-2L67
will always be present in solutions, 0.003 0.303 -2.52 -3.98 pH-10.34 2pH-20.67
whatever the system the grower uses.
0 .0 1 1 .0 1 -2 .0 0 -3.46 pH-9.82 2pH-20.15
Due to the release of CO 2 by roots and
0 .1 1 0 .1 - 1 .0 0 -2.46 pH-8.82 2pH-19.15
microorganisms, CO 2 levels will always
be higher in field soils compared to the 1 .0 1 0 1 .0 0 .0 0 -1.46 pH-7.82 2pH-18.15
atmosphere (0.030 kPa).
Helyar and Porter [1989]
suggest the partial pressure
of CO 2 in soils to be 5 to
2 0 times higher than in the
atmosphere. Lindsay
[1979] usually increases
CO2 ten times. In shallow,
porous greenhouse sub
strates, differences be
tween the atmosphere and
gas in the soil pores may
not be detectable [Hanan,
1964]. Nevertheless, in the
root rhizospheres, and
microsites in the substrate,
CO2 in any system must be
higher (see Fig. 5-22). CO 2
levels in hydroponic
systems have not been
thoroughly examined to
my knowledge. Hurd
[1978] did discuss the root Fig. 6-32. The effect of pH on distribution of carbonate ions in solution [C hem ical E qulibria
environment of NFT sys in Soils, Lindsay, W.L., ©1979. Reprinted by permission of John Wiley & Sons, Inc.].
tems, as did Steiner
[1968]. Neither of these authors, however, cited any ^ ature giving actual CO 2 levels to be expected close to the
roots of crop plants in hydroponic systems.
2. pH Control
a. Raising pH:
Most temperate soils are generally acid, and weathering will result in continuous acidification to the point
that aluminum, manganese, or hydrogen toxicities may occur, plus deficiencies of phosphorous, molybdenum,
and calcium [Ritchie, 1989]. Even though, in greenhouse production, the soil is not subject to rainfall, high irriga
tion rates, combined with some fertilizer types, can cause rapid acidification (Table 6-5).
451
There are exceptions in

greenhouse practice such as blu
ing of hydrangeas where highly
acidic chemicals (aluminum
sulfate) are used to reduce pH
deliberately. However, it is a com
mon practice to lime soils to
correct highly acid conditions. It is
necessary in container production
where the substrate may use media
combinations incorporating peat
moss. It is not the objective to
apply alkaline fertilizers such as
limestone to the point of neutrality
(pH = 7.0) since this can depress
yield. Limestone is primarily
applied to correct toxicities
associated with acid soils [Cregan
et al., 1989]. Liming not only
neutralizes soil acidity, but
replenishes exchangeable Ca^^
[Lindsay, 1979]:
Fig. 6-33. Lime requirement of four peat-sand substrates made with different peat
moss sources. Liming material was equal parts of CaC0 3 and CaMg(C0 3 )2 [Adapted
from Bunt, 1988].
C aC O Jc) +[ 2 H ^ -e x c h a n g e ] - [C a ^ ^ -e x c h a n g e ] + C O ^ ig ) +H ^ O (6.45)
C a C O ^ {c ) + [ 0 . 6 6 - e x c h a n g e ] [C a ^ ^ -e x c h a n g e ]
6 .6 6 A l{ O H )^ { s ) + C O ^ {g ) Hp (6.46)
The result is an equation:
lo g C a ^ ^ + 2 p H = 9 . 7 4 -lo g C O ^ ig )
(6.47)
which can be rearranged:
p H -0 .5 p C a = 4 .S l-0 .5 lo g C O ^ (g ) (6.48)
Eq. 6.48 is called the lime potential relationship, where pCa = -log Ca^"' and CO2 is given in atmospheres
pressure. According to Mengel and Kirkby [1982], the usual application rates for mineral soils are about 4 to 6
tons per hectare over 3 to 5 years. The amounts required not only depend upon soil pH (actual acidity), but also
on the H^ adsorbed to the exchangeable system (colloids) (potential acidity). Some people call the latter the
“reserve acidity.” Obviously, the higher the cation exchange capacity of a soil, the more likely that an acid soil
will have a high reserve acidity.
Limestone is a calcium carbonate (CaC 0 3 ) or calcite. Another common liming material is dolomitic
limestone (CaMg(C 0 3 )2 ) (Table 6-5). Still others that are occasionally used are burned lime (quicklime, CaO)
and hydroxide of lime (Ca(OH)2 , slaked lime). The average purity of crushed limestone is about 94% [Brady,
1974]. CaO and Ca(OH ) 2 are powder forms and readily available. Of course, other chemicals give a basic
reaction in solution, a number of them given in Table 6-5. Those with sufficient solubility to be used in fertigation
methods or hydroponics can cause the solution pH to move upward gradually, as contrasted with the reaction rate
in traditional liming materials as noted. The major changes in peat-based substrates can occur over a period of
2 days and be essentially complete in 14 days [Williams et a l, 1988]. The medium perlite, some types of
vermiculite, and many organic fertilizers (i.e., manure, wood ashes, etc.) can also contribute to liming responses
of substrates into which they are incorporated. On the other hand, peat moss in perlite, used for rooting purposes,
requires about 0.5 kg powdered calcium carbonate per 10 m^. Otherwise, the rapid pH drop will cause aluminum
to be released from the perlite with consequent phytotoxicity. Fig. 6-33 shows the lime requirement for different
peat-sand materials. Tables 5-4 through 5-6 show that some type of liming material is invariably mixed into the
substrates. Bunt [1988] presented rates of limestone application ranging from 0.9 kg m'^ for Finnish sphagnum
to 2.0 kg m'^ for English sphagnum to raise pH by 1 unit. Liming reactions, however, are not easily predictable,
requiring tests to decide desirable application rates.
The application to soil of two different liming materials in chemically equivalent quantities does not always
mean that equivalent results will be attained. If both limestones, there may be differences in hardness and particle
size. The latter are important since the finer the particle size the more rapid its reaction rate. Cregan et al. [1989]
state that ground limestone reaches 100% efficiency at a fineness of about 150 pm. Experimentally, limestone
that passes 60- to 100-mesh screens (250-150 pm) is the largest particle size that is 100% efficient. Brady [1974]
gives the requirement that all material must pass a 10-mesh screen (1.8 mm diameter) and at least 50% of the
pulverized limestone must pass a 100-mesh screen (0.15 mm diameter). Such material is called “fine” limestone.
Grinding to a smaller size is not usually economical. There have been occasions when 2- to 3-cm diameter stone
limestone has been used in gravel hydroponic beds. Applications must be mixed uniformly in the substrate. This
is particularly important in infertile, acid soils. The target pH is 5.5 to 7.0 [Cregan et al.,1989], although Bunt
[1988] recommended for organic soils a pH range of 5.0 to 5.5. Deliberate liming to neutrality (7.0) is excessive
and expensive. The idea is to retard or slow acidification.
b. Lowering p H
When calcium carbonate is present in soil, Lindsay [1979] states that it has a dominating influence on many
soil properties. Most calcareous (semiarid or arid) soils have a pH range of 7.3 to 8.5. The reaction showing the
dissolution of CaC0 3 was given in Eq. 6.3 (Section 6 .II.B), and the resulting formulae, expressing Ca^^ activity
as functions of pH and CO 2 levels, appear in Eqs. 6.47 and 6.48. What can be done, practically, depends upon
calcite-gypsum equilibria and CO 2 concentrations.
It is doubtful that saline-sodic or sodic soils will be encountered in greenhouse practice. The latter do not
contain a great amount of neutral soluble salts, the phytotoxicity arising from sodium and hydroxyl ions, resulting
in pHs above 10.0. The former usually has more than 15% of the CEC saturated with Na^, with a saturated paste
extract above 4 dS m'^ EC, and a pH below 8.5. These types of soils are extremely difficult to manage and are
uneconomical for greenhouse production requirements. Saline soils, however, may be encountered -sometimes
from the mismanagement of fertilizer programs. Brady [1974] attributes these types with high neutral soluble
salts, less than 15% of their CEC is saturated with Na^, and the electrical conductivity (saturated paste extract)
is above 4 dS m ’. With saline soils, the excess salts (mostly chlorides and sulfates of sodium, calcium and
magnesium) can be leached with no appreciable rise in pH. Obviously, the irrigation supply must be low in
sodium. High applications of gypsum are the usual means to reclaim saline or saline-sodic soils. The latter should
not be leached with water low in calcium or magnesium since the pH will rise rapidly and a tight, impervious soil
structure results. The neutral salt gypsum (CaS0 4 -2 H 2 0 ) is often recommended at several tons per hectare, mixed
into the soil, to help reclaim saline or sodic soils. If calcium is required without a change in pH, calcium sulfate
is recommended. Gypsum changes the caustic alkali carbonates into milder sulfates (e.g., Na 2 S0 4 ) that can be
leached.
In soils with pH above 7.0, calcite, as noted previously, precipitates. Fig. 6-26 showed the relationship with
pH, and Fig. 6-34 examines the relationship in greater detail, including the effect of CO 2 . As CO 2 increases, the
calcite line shifts to the left. At a very high CO2 level of 30.3 kPa, calcite could exist at pH 5.0 if (Ca^"^) = 1.0 M
f ’ . At the usual atmospheric level (30.3 Pa), the Ca^^ activity would have to be several moles. This is a situation
impossible in productive soils. One also notes that at any given CO 2 concentration, the effect of raising or
lowering pH moves along the calcite equilibrium line. Surprising to some, calcium chloride (CaCl2 ) is more
453
effective than an equal amount of hydrochloric acid (HCl) in lowering pH. CaCl2 raises Ca^^ activity, precipitates
calcite, absorbs more CO 2 from the air, forming carbonic acid that dissociates and lowers pH [Lindsay and Ajwa,
1993]. In the pH range at which gypsum and calcite can coexist (7.5-8.0), adding H 2 SO4 merely causes calcite
to dissolve and gypsum to precipitate [Lindsay, 1979]:
HjSO. + CaCOJcalcite) + HÔ ^ CaS0/2H20(gypsum) + COj (g) (6.49)
Fig. 6-34. activity versus

pH for equilibrium with
CaC0 3 at various CO2
concentrations. At (CO2) =
30.3 Pa, and at equilibrium
(dark square), the pH = 8.30,
(Ca'O = 10-'^' and (H C O 3-) =
10’^^^ M r' [Lindsay and
Ajwa, 1993]. Gypsum not in
cluded in this calculation.
I have seen growers

attempt acidification of
calcareous soils with
sulfuric acid -much to the
detriment of metal piping
and pumps. One can
calculate the equilibrium of
calcite-gypsum as pH = 7.8
for 30.3 Pa CO 2 . This will
vary due to other compo
nents in the soil, but the
addition of CaO, CaC 0 3 ,
CaS0 4 ‘H2 0 , and H2 SO4 , to
a soil containing gypsum
and calcite will not affect
the pH of that soil.
A list of materials pH
used to acidify soils is
given in Table 6-21.
Depending upon the particular situation, some of these, as suggested above, may not be desirable. Ferrous sulfate
and elemental sulfur are generally recommended. Sulfur is oxidized in the medium to H 2 SO4 by the bacterium
Thiobacillus, hence the reaction is slow. As potting substrates may have low amounts of calcite, the pH lowering
follows the curves given in Fig. 6-35 for selected substrates. Tables 5-4 through 5-6 provide suggested amounts
of CaCO^ for acid media. It can be appreciated that most published recommendations for container production
come from experience in temperate regions with acid soils. Where the grower uses a calcareous soil as part of
the substrate, the recommendations will be changed. Because conditions are so variable, the grower must exper
iment or rely on the local agricultural experiment station for suggestions.
For hydroponic systems where CEC is, to all purposes, nonexistent, any of the acids or soluble acid materials
listed in Table 6-21 can be used to adjust pH. For upward adjustment, bases such as sodium hydroxide (NaOH),
potassium hydroxide (KOH), or ammonium hydroxide (NH 4 OH) can be utilized. Cooper [1979] devoted
considerable space to pH adjustment in NFT solutions, suggesting for most crops that pH should be controlled
between 6.0 and 6.5. In actuality, controlling pH in the range suggested is extremely difficult. The previous
Section 6.IV.C.1 emphasized that CO 2 will dissolve in Table 6-21. In organ ic and organ ic m aterials u se d to
water, and remarkable buffering of the solution will red u ce pH.
occur. For example, Table 6-22 is a selected list of

Name Chemical formula
hydroponic solutions checked for electrical neutrality.
Inorganic
Hoagland's solutions are probably the most commonly
employed in biological research at one-half the A lu m in u m su lfate Al2 (S0 4 ) 3
concentrations given. If the data for each solution in A m m o n iu m nitrate NH4NO3
Table 6-22 are submitted into Allison et al.'s [1991] A m m o n iu m p o ly p h o sp h a te (NH4)3H4P202
chemical equilibrium program MINTEQA2/- A m m o n iu m su lfate (NH4)2S04
PRODEFA2, using Lindsay's modifications, the N itr ic acid HNO3
equilibrium pHs of all solutions lie between 7.0 and 7.2. P h o sp h o r ic acid H3PO4
These salts have been added to pure water and the
P y r o p h o sp h o r ic acid H4 P2 O7
equilibrium carbonic acid concentration (¥[2^ 02°), where
the gaseous CO 2 level is set to 0.0003 atm (30.3 Pa), is
S u lfu ric acid H2 SO4
1 0 - 4 992 ] \ 4 (10.2 pM). The concentration where the CO 2
E lem en tal sulfu r s
pressure is assumed 0.003 atm (303 Pa) is 10'^^^^ (101.9 Ferrous su lfate Fe
pM). These concentrations agree well with those given C a lciu m ch lo r id e C aC F
in Table 6-20. Organic
The effect of adding an acid such as HCl, or a base L e a f m o ld
such as NaOH, is depicted in Fig. 6-36. Not until almost
P in e n e e d le s
100 pM of acid has been added does pH begin to drop,
Tanbark
decreasing more rapidly so that at 800 pM HCl, the equi
S aw d u st
librium pH is almost 5.0. The buffering effect in raising
pH with NaOH is more difficult. An addition of 800 pM P eat m o ss
base does not raise pH significantly. At the point where
Fig. 6 -3 5 . C hanges in m edia pH w ith rates o f

sulfu r a p p lica tio n . A = a sp en t m u sh ro o m
co m p o st, B = o ld to m a to m o d u le s (se d g e
peat), C = o ld to m a to m o d u le s (sp h a g n u m
peat, h a lf d e c o m p o s e d ), D = o ld tom ato
m o d u le s (u n d e c o m p o se d sp h a g n u m peat)
[B unt, 1 9 8 8 ].
sufficient acid or base is added, the pH

control becomes difficult. Most authors
[e.g.. Cooper, 1979; Jones, 1983] sug
gest the use of automatic injection sys
tems for pH measurement and control.
To calculate the HCl necessary. Fig. 6 -
36 shows that about 400 pM will be
required or 0.0004 M f ’ for a pH near
6.5. At a formula weight of 55, the
amount in grams will be (55)(0.0004) =
0.022 g. The purity of the liquid acid is
90% so that an additional 10% must be
added, or 0.022 + 0.0022 = 0.0242 g.
The density of HC 1 -H2 0 is 1.48. The S u lfu r r a t e C k g /c u .m .3
required volume is 0.0242/1.48 = 0.016
ml fC In an NFT system, having a liquid
capacity of 30 m^, (30000)(0.016) = 491 ml HCl. Any of the acids or bases employed are highly caustic. Suitable
safety procedures are required such as goggles, rubber gloves, and aprons. Be sure never to add water directly
455
to a concentrated acid or base. The

opposite procedure is required, with
suitable precautions to prevent
spattering.
It should be emphasized that these
comments deal with salts dissolved in
pure water, and the assumed CO 2
partial pressure of 0.003 atm is correct. ft
Undoubtedly, the equilibrium pH will
change with different fertilizer salts
used in making up the nutrient solution.
For example, phosphorus could be
supplied in phosphoric acid. Then, the
formulae in Table 6-22 would have to
include the ion. Furthermore, plants
1000 100 10 1 0.1 0 1
would be reducing concentrations of HCl (uM) NaOH
the various ions at differing rates, and
<-----------------
(mM)
they would be releasing many
substances to the solution. Intermittent
Fig. 6-36. T he effe c t o f add ing H C l or N a O H to th e carn ation so lu tio n g iv e n
solution testing for pH, and hand in T able 6 -2 2 . T he square dot is the equilibrium pH at a C O 2 pressure o f 0 .0 0 3
calculation, would be laborious. atm (3 0 3 Pa).
3. pH M easurem ent
While there are simple systems using dye indicators to show pH, the most common instrument capable of
being used as a controller in a suitable injection system is the pH meter. Instruments with either analog or digital
output, portable, and reading to three or more significant digits can be obtained. Glass electrodes are immersed
in the sample, and the hydrogen ion concentration is balanced against a standard cell that behaves similarly to
the standard hydrogen electrode.
However, there are difficulties. The conventional method of making a pH test on a suspension of the medium
and distilled water (one part soil to 1 part water) may give pH readings that are much higher than those actually
experienced by the plants growing in the medium. Bunt [1988] suggests a 1:2.5 mixture. The amount of soil is
often 5 g in 5 ml water or CaCl2 reagent and shaken for 10 to 30 minutes. The procedure should be the same in
any series o f tests. In fact, Helyar and Porter [1989] state that water is not an appropriate liquid in which to
suspend soil for measuring pH. The deceptively simple process has been the center of many controversies
[Ritchie, 1989]. One records a higher pH when electrodes are placed in the supernatant solution than when
immersed in the sedimented soil particles. Variations in pH can be attributed to soil variability, seasonal changes,
soil-liquid ratio, and the type and concentration of the ions in the extracting solution. Jones [1985] states that
inaccurate meter calibration and electrode problems are common. Combination electrodes are not well suited for
pH determinations. To avoid some problems with distilled water, particularly salt content variability and
maintaining soil in flocculated conditions, the extracting solution should contain 0.01 M CaCl2 , which gives a
pH similar to the saturated paste method [Bunt, 1988; Jones, 1985]. A 1 Normal (equivalent) HCl solution will
also mask differences in salt concentrations. Although pH measurements can be duplicated with high precision,
attempting to read an output to more than ±0.1 unit is useless. This is contrary to Jones' [1985] recommendation
to read pH to the nearest tenth. In fact, in my estimation, one does very well to read or control pH to within one-
half pH unit in hydroponic systems, soils, or container substrates. Despite these limitations, Brady [1974] stated
that a great deal more information may be obtained from a pH determination than any other single analytical
value.
D. CHELATION AND MICRONUTRIENTS

A number of micronutrient formulae are given in Table 6-22. The amounts are critical since the range
between deficiency and toxicity can be narrow compared with ranges of the macronutrients. More extensive
micronutrient formulations have been published [Hewitt, 1966; Jones, 1983]; but the value of these for commer-
o^
Table 6 -2 2 . S e le c tio n o f nutrient so lu tio n s u sed for e x p erim en ta l and c o m m ercia l ap p lica tio n s. U n its are in m ic r o m o le s per liter (p M f^). Inform ation taken from a variety
o f so u r c e s as g iv e n in fo o tn o te s. O
Ion Hoagland V Hoagland 2^ Modified Johnson’s Steiner’s NFT solution® Carnations^ Roses^ Tomatoes*’ Cucumber*’ Pepper*’
T3
_____ _______ Hoagland^ 0.25 strength® formula"*___________________________________
NH4 ^ 1000 400 500 2500 1000 500 500
K" 6000 6000 2400 1500 13151 7964 6000 4100 7000 5500 6000
3.
Ca^" 5000 4000 1600 1000 822 4250 1500 3000 3750 3500 3750 o'
o
2000 2000 800 250 822 2077 1000 500 1000 1000 125 0
H2 PO4 - 1000 1000 400 51 822 2192 1100 1000 1 500 12 5 0 125 0
NOs’ 15000 14000 5600 3500 13151 14272 10400 9000 10500 11750 12250
S0 4 ^- 2000 2000 800 300 1233 2077 1000 500 2500 1000 125 0
Fe'^ 15 15 100 25 32 15 15 10 10 10
Mn^^ 9.1 9.1 4 .5 6 5 9 36 9.1 9.1 10 10 10
Zn'^ 0.8 0.8 1 .5 4 2 0 .3 4 1.5 0.8 0.8 4 4 4
Cu'^ 0 .3 2 0 .3 2 0 .3 1 5 0 .5 0 .0 8 1.6 0 .3 2 0 .3 2 0.5 0.5 0.5
BO3 '- 18.2 18.2 2 3 .1 137 18.1 11 18.2 18.2 20 20 20
Mo0 4 ^' 0.1 0.1 0 .1 0 4 0.2 0.21 0 .3 0 .1 0.1 0.5 0.5 0 .5
Cl- 9 .1 2
C h e la tin g a g e n t 15’ 15* lOO* 2 5 J 32J 15‘ 15’ 1 0 * lOi 1 0 >
” H o a g la n d and A rn on , 1 9 5 0
^ H a lv o r s o n and L in d sa y , 19 7 2 . In c lu d e s trace e le m e n ts
®N o r v e ll, 1 9 9 1 . In c lu d e s trace e le m e n ts
^ S tein er, 1 9 6 1 . “B a la n c e d ,” “Id eal” so lu tio n
®C o o p e r , 1 9 7 9 . In c lu d e s trace e le m e n ts
^H artm an and H o lle y , 196 8
® S a d a siv a ia h and H o lle y , 1973
*’ W e lle m a n and V erw er, 1983
‘ F e N a D T P A , H o a g la n d and A rnon, 1 9 5 0 , w ith e x c e p tio n o f D T P A in p la c e o f F e S 0 4
jP e N a E D T A
*"Trace e le m e n ts from Stein er, 196 6
457
dal use remains to be Table 6-23. Inorganic compounds capable of being used as micronutrient sources
determined. Very often, a [Adapted from Mortvedt, 1991].^
sufficient trace element can be
a contaminant in the fertilizer Micronutrient Chemical name Chemical formula Solubility
(organic or inorganic). Zinc Boron Anhydrous borax Na2B4 0 7 Soluble
from galvanized pipes or Fertilizer borate Na2B407-5H20 Soluble
copper from copper piping can Borax Na2B 4O7-1 0 H2O Soluble
dissolve sufficiently to provide Boric acid H3BO3 Soluble
the necessary Zn^^ or Cu^^. Colemanite Ca2B60,,-5H20 Slightly soluble
Piping replacement with plastic
Copper Cupric sulfate CUSO4 H2O Soluble
has often resulted in zinc or
Blue vitrol CUSO 4 5 H2O Soluble
copper deficiencies in the crop.
Additions of nickel, cobalt, Cupric oxide CuO Insoluble
arsenic, barium, cadmium, etc. Iron Ferrous sulfate FeS0 4 -H2 0 Soluble
have yet to be shown necessary Ferrous sulfate FeS04’7H20 Soluble
under commercial conditions. Ferric sulfate Fe2(S04)3-9H20 Soluble
With the possible exceptions of Ferrous-ammonium sul FeS0 4 -(NH4)2 S0 4 Soluble
sodium and selenium, most of fate
the difficulty of making up Manganese Manganous sulfate MnS0 4 -xH2 0 Soluble
solutions as given by Hewitt
Manganese dichloride MnCl2‘4H20 Soluble
[1966] is not worth the effort.
Manganous carbonate MnC0 3 Insoluble
Proprietary mixtures of the
more important chemicals can Manganous oxide MnO Insoluble
be purchased. Individual Manganese oxysulfate Variable Variable
sources are noted in Table 6-23 Molybdenum Sodium molybdate Na2Mo0 4 Soluble
for the common micronutrients. Sodium molybdate Na2Mo 0 4 -2 H2 0 Soluble
Of course, most of these are Ammonium molybdate (NH4)2Mo0 4 Soluble
required in such small amounts Molybdic anhydride M0 0 3 Slightly soluble
that there can be difficulty in
Calcium molybdate CaMo0 4 Insoluble
uniform application to soils and
Zinc Zinc sulfate ZnS 0 4 -H2 0 Soluble
container substrates. Dissolving
the chemicals in water and Zinc sulfate ZnS04-7H20 Soluble
applying through the irrigation Zinc chloride ZnCf Soluble
system, or suitable injection ZnS04-4Zn(0H)2 Slightly soluble
equipment, represents the Zinc carbonate ZnC0 3 Insoluble
easiest and safest method. Zinc oxide ZnO Insoluble
Hoagland and Amon's trace
Zinc oxysulfate Variable Variable
element mixture is usually
Note comments in the text on dissolving many of these compounds, espe
made up in a concentrated
cially if making stock solutions.
solution and then an aliquot
taken for final dilution. If
chelates are used, the stock
solution requires refrigeration.
Many salts can be difficult to dissolve, boron in particular. There have been instances of a grower placing pow
dered borax in his concentrate tank and winding up with boron deficiency in carnations. That problem can be
eliminated by making sure that each salt used is dissolved in hot water before adding it to a stock solution or into
a concentrate tank.
Previous sections have made clear that provision of the necessary trace elements in solution does not mean
that they will be available to the root system. This is particularly true with respect to iron in greenhouse
production. Ferrous sulfate, recommended by Hoagland and Amon [1950] in their original publication, usually
precipitates. The effect of pH on solution concentrations of Fe^^ and other trace elements was shown by Figs.
6-2, 6-7,6-16, and 6-17. There are various means available to plants to overcome the low concentrations common
to the pHs usually recommended in the bulk soil. For example, the pH in the rhizosphere can be lower than in
10 20 30 AO 50 60
DAYS AFTER TRANSPLANTING
Fig. 6-37. The effect of alternate flooding and drying on soil redox of a mineral rice paddy soil 2 cm
below the soil-water interface [Sajwan and Lindsay, 1986] (With permission of the Amer. Soc. o f A g ro n ).
Table 6-24. Common chelating agents with names and abbreviations [Adapted from Norvell, 1991;
Lindsay, 1983].
Abbreviations and acronyms Chemical name

BPDS Bathophenanthrolinedisulfonic acid
CIT Citric acid [(COOH)CH2C(OH)(COOH)CH2COOH]
CDTA (DCTA) tr a n s - 1,2-Cyclohexylenetrinitrilotetraacetic acid
DTPA (Fe 330) Diethylenetrinitrilopentaacetic acid
EDDHA (EHPG, APCA, Fe 138) Ethylenediiminobis(2-hydroxyphenyl)acetic acid
EDMA Ethylenediaminemonoacetic acid
EDDA Ethylenediamine-N,N'-diacetic acid
EDTA (Sequestrene, versone) Ethylenedinitrilotetraacetic acid
ED3A Ethylenedinitrilotriacetic acid
EGTA Ethylenebis(oxyethylenetrinitrilo)tetraacetic acid
HBED N,N'-bis(2-hydroxybenzyl)ethylenedinitrilo-N,N'-
diacetic acid
HEDTA (HEEDTA, Versonal, Perm. N-(2-hydroxyethyl)ethylenedinitrilotracetic acid
Green)
HIDA (HEIDA) N-(2-hydroxyethyl)iminodiacetic acid
IDA Iminodiacetic acid
NTA Nitrilotriacetic acid
459
the bulk substrate. The effect of alternate

flooding and drying on the redox potential
of a mineral soil was noted by Sajwan and
Lindsay [1986] (Fig. 6-37). Both lower pH NaEDTA 10‘^fcd
and lower pe values will increase
micronutrient activities in the substrate
solution -except molybdenum. Respiratory
processes by microorganisms and roots will
provide conditions more suitable for iron
availability. Also, plants and
NaEDTA
microorganisms can produce chelating
substances that complex with the metal ion,
keeping the ion in solution and making it
more available. Such substances are called
siderophores. The term “chelate” refers to
the fact that the ion is attached to the HCI 10“2 m
chelate -o r “ligand”- at several points. The CITRIC A C ID lO'^M

C A SA M N O ACIDS O l^
affinity of the ligand for the ion is greater CITRIC A CID 1 0 '^
than competing anions that tend to precipi HoO
tate the micronutrient. Reference has
already been made to so-called iron
efficient plants that can shift metabolism
Fig. 6-38. The effect of various complexing agents and acids on diffusion
when iron is deficient, thereby enhancing
of Zn into a simulated root over 7 days [Elgawhary et al., 1970] (With
iron supply. There are also synthetic permission of the 4 mer. 5 (9 c. o f A gro n o m y).
chelates, organic molecules that have been
especially beneficial in improving plant
nutrition over the past 3 decades. A list of these is pro
vided in Table 6-24. The compounds EDTA, DTPA,
EDDHA, and HEDTA are the most common and
readily available for commercial greenhouse use. These
compounds can chelate with calcium, copper,
manganese, and zinc, as well as iron. Fe330 (DTPA) is
a green material most often purchased, while Fel38
(EDDHA) is dark red and recommended for alkaline
conditions. Lindsay and co-workers have been partic
ularly instrumental in explaining the complexities of
chelation in soils and hydroponic solutions [e.g.,
Lindsay, 1984; 1991; Halvorson and Lindsay, 1972;
Lindsay and Norvell, 1969; Lindsay and Schwab, 1982;
Schwab and Lindsay, 1983a; b; Sadiq and Lindsay,
1989; etc.]. The effect of EDTA, for example, on the
diffusion of zinc into a simulated root was examined by
Elgawhary et al. [1970], showing greatly enhanced
accumulation of the zinc complex at concentrations
ranging from 100 to 1000 pM NaEDTA (Fig. 6-38).
The development of computer programs to
calculate equilibrium levels has helped explain many
results. Halvorson and Lindsay [1972] showed the Fig. 6-39. Effect of pH on the ability of various chelates to
ability of various chelates to complex with iron (Fig. 6 - hold iron in a modified Hoagland solution [Halvorson and
39). The high equilibrium constant for the reaction Eindsay, 1972] (With permission of the A m er. Soc. o f
between Fe^^ and the ligand EDDHA means that a high A g ro n o m y).
pH will not result in displacement of the iron from
EDDHA. For EDTA, however, a pH

above 7.0 will displace Fe^"^, which then
precipitates as an unavailable hydrous-
ferric oxide. This is the reason for the
common recommendation of EDDHA
in alkaline soils. Unfortunately, the
situation is not that simple. Halvorson
[1971] showed actual results on
growing maize (Fig. 6-40). All four Fe
chelates could supply Fe at pH 5.3, but
at pH 7.5, only FeEDTA was effective.
At 7.5, Fe was displaced to form
CaEGTA^' and insoluble iron oxide.
The poor response with FeDTPA at 7.5
resulted from zinc deficiency and not
iron, since spraying the plants with Zn
corrected the low maize yield. DTP A
combines strongly with Zn above pH
7.5. Poor response from FeEDDHA at
high pH in solution culture resulted
from the presence of a free chelating
agent that depressed the Fe^^ activity.
Even when EDDHA exceeds Fe by
only 0.5%, the free Fe^^ can decrease
by some millionfold [Lindsay, 1972c].
Thus, excessive chelating agents,
particularly EDDHA can cause undesir
able results. Usually, ligands are not Fig. 6-40. The effect of different chelates and pH on growth of Z ea m ays
taken into the root. However, they are [Halvorson, 1971].
separated from Fe^^ at the root and Fe^^
is reduced to Fe^"“ [Simons et al., 1962; Lindsay, 1972c]. Uptake of the ligand can occur, and in the case of
EDDHA, the ability of this chelate to complex with iron can result in a severe deficiency although the plant may
have more than enough iron [Simonds et al., 1962]. This strong affinity of EDDHA for iron has also been im
plicated in Fusarium suppressive soils. Scher et al. [1984] suggested that this gave a selective advantage to
siderophore-producing antagonists such as Pseudomonas putida, resulting in Fe deficiency in Fusarium.
Whether in soils or nutrient solutions, the chemical relationships to be found can be highly complex. Lindsay
and Norvell [1969] developed '‘mole-fraction” diagrams for EDTA and DTPA in soils where the competing ions
were Zn^^, Fe^^, Ca^^, and H^ at 0.003 atm CO 2 (Fig. 6-41). When Zn^^ is included as a competing ion, ZnL^'
becomes a major complex between pH 6 and 7. FeL^‘ is dominant below pH 6 and CaL^' above 7. This helps ex
plain the results in Fig. 6-40 at high pH. For DTPA (right side), Zn2 L' is dominant below pH 4.5, FeL^' between
4.5 and 6.4, and ZnL^' above 6.4.
Fig. 6-42 is similar to Fig. 6-41 except that the curves show mole fractions for a modified Hoagland's solu
tion in equilibrium with hydrous-ferric oxide and 0.0003 atm CO 2 . Depending upon pH, EDTA and DTPA also
complex with manganese and copper. These two were not included in Fig. 6-41 for soils. EDTA chelates iron
effectively to about pH 6.5. The zinc is nearly all chelated above 6.5, the copper above 5.5, and Mn^^ above 6.5.
The amount of magnesium chelated is significant above pH 8.2. While not shown here, the effectiveness of
FeEDTA in supplying iron to plants may decrease with increased zinc concentration. The right diagram in Fig.
6-42, shows that FeDTPA can be an effective iron supply up to pH 7.8. What is interesting about these diagrams
is the number of complexes that may be present in soil solutions and in nutrient solutions. The relationships of
chelated ions in solution are not simple, and the addition of a synthetic chelate sets in train a whole order of
reactions that influence plant growth.
461
Fig. 6-41. The proportion of EDTA (left) and DTP A (right) attached to various metal ions in soils where CO2 gaseous
concentration is 0.003 atm [Modified from Lindsay and Norvell, 1969] (With permission of the A m er. Soc. o f A ronom y).
Fig. 6-42. Amount of EDTA (left) and DTP A (right) attached to various metal ions in Hoagland's modified solution in
equilibrium with ferric-oxide and 0.0003 atm CO2 [Modified from Halvorson and Lindsay, 1972] (With permission of the
A m er. Soc. o f A g ro n o m y).
462 Chapter 6, N utrition
In a more recent review, Norvell [1991] showed that many other metal ions (i.e., AP^, Co^^, Cd^^, Pli^,
etc.) have been found to chelate with many of the compounds listed in Table 6-24, depending upon the acidity
or alkalinity of the soil. These have not been shown to have significant impact on greenhouse practice -largely
since no one in the business has actively investigated the relationships. For nutrient solutions, Norvell used the
GEOCHEM computer program to evaluate Johnson's 0.25-strength nutrient solution (Table 6-22), producing
several diagrams similar to Figs. 6-41 and 6-42. These types of studies have not been carried forward by
greenhouse researchers to the best of my knowledge. Johnson's solution is not commonly employed in greenhouse
hydroponics, and some ofNorvell's comments, while highly interesting, have not been examined in detail under
commercial conditions.
Norvell [1991] stated that introduction of chelated Fe causes high pH sensitivity of all micronutrient ions.
Concentrations of free metal ions may change 100- or 1000-fold in poorly buffered nutrient solutions. The failure
of chelating agents to buffer micronutrient ions is, according to Norvell, because the chelate buffer is added in
one form. The insoluble hydrous Fe(III) oxides largely control Fe^^ concentration whether or not a chelate is
present -unless sufficient chelate is added to complex most of the Fe, Cu, Zn, etc. With a “modest” excess of
EDTA, the free ligand is buffered by equilibria between Ca^^ and Ca-EDTA. The activities of other ions (Fe, Cu,
Zn, and some Mn) are also buffered and freed from disturbance by the pH-dependent solubility of iron oxides.
Inclusion of 25 pM EDTA, in addition to 25 pM f ’ Fe-EDTA, produces nearly constant levels of metal EDTA
chelates and constant activities of all metal ions from about pH 4.5 to nearly 7.5 -where iron oxides begin to
precipitate despite additional EDTA. Iron phosphates may also precipitate in nutrient solutions with iron oxides.
Citing the i ature, Norvell [1991] rejected the use of chelating agents CIT, EGTA, HIDA, EDDHA, NTA, and
HBED since they could not maintain Zn^^ in the desired range. In contrast, EDTA, CDTA, HEDTA, or DTP A
appeared able to provide adequate buffering in a convenient range for Zn^^.
To provide an example of the ions in a nutrient solution for roses (Table 6-22), MINTEQA2/PRODEFA2
has been used to calculate ion distribution for three different situations at pH 6.5 and 0.003 atm CO 2 , with the
results presented in Table 6-25. Table 6-26 gives the percentage distribution for the same examples. One will
immediately note (Table 6-25) that none of the added nutrients remain at the initial concentrations. There is a
whole host of complexes that form in solution -some 65 of those printed out by MINTEQA2/PRODEFA2 were
not included in Table 6-25. Table 6-26 is more interesting. Increasing the concentration of DTP A results in
chelation with calcium, increases manganese chelation from 1 to 29%, and reduces DTP A chelation with iron
by some 40%. On the other hand, 99% of EDDHA chelates with iron. The partitioning of phosphate does not
change with chelating agent. With additional DTP A (30 jliM ), 100% of the Fe^^ and Fe^^ ions will be chelated,
whereas 100% of Fe^"" only combines with EDDHA. Better than 50% of Fe^^ remains as the free ion in solution.
Table 6 -2 5 . E x a m p les o f e ffe c t o f c h e la te s in a rose nutrient

so lu tio n (T a b le 6 -2 2 ) by the M IN T E Q A 2 /P R O D E F A 2 com p u ter
program . Ions w ith c o n cen tra tio n s b e lo w 0.01 p M ignored.
C h em ica l form u lae w ith are neutral a q u eo u s m o le c u le s.
F e(O H ) 3 a llo w ed to precipitate, in eq u ilib riu m w ith 0 .0 0 3 atm C O 2
and pH fix e d at 6 .5 .
Chemical for 15 |liM Fe^^ 15 pM Fe'^ 15 pM Fe^^
mula and 15 pM and 30 pM and 15 pM
DTPA DTPA EDDHA
NH 4^ 879 879 879
3614 3614 3614
Ca'^ 1690 1690 1690
Mg2^ 277 277 277
N0 3 ' 7943 7943 7943
SO42- 240 240 240
Mn^^ 5.3 0 .1 2 5.4
Zn^^ 0.03 < 0 .0 1 0.39
Cu'^ < 0 .0 1 < 0 .0 1 0.05
463
Chemical for 15 ¡iM 15 ¡iM 15 ¡iM Fe^^

mula and 15 ¡iM and 30 juM and 15 ¡iM
DTPA DTPA EDDHA
H3B03° 18.2 18.2 18.2
Mo0 4 ^- 0.06 0.06 0.06
MnDTPA^- 0.06 2.84 ...
MnDTPA'- <0 .0 1 0 .1 0
H2B03- 0.03 0.03 0.03
NH3° 1.58 1.58 1.58
NH4S04- 2.72 2.72 2.72
MgHC03^ 0.46 < 0 .0 1 0.46
MgS04° 11.3 11.3 11.3
MgP04- 0.17 0.17 0.17
MgH,P0 4 * 25.6 25.6 25.6
CaHG03^ 2.51 2.51 2.51
CaC03° 0.05 0.05 0.05
CaS04° 82.8 82.8 82.8
CaHP04° 106 106 106
CaP04' 0.78 0.78 0.78
CaH2P04^ 24.2 24.2 24.2
KSO4- 6 .1 1 6 .1 1 6 .1 1
KHPO4- 0.79 0.79 0.79
MnS04° 0.23 < 0 .0 1 0.23
MnHC03^ 0 .0 1 < 0 .0 1 0 .0 1
ZnOH^ <0 .0 1 < 0 .0 1 0 .0 2
ZnS04° <0 .0 1 < 0 .0 1 0 .0 2
ZnHP04‘^ < 0 .0 1 < 0 .0 1 0.09
HCO3- 144 144 144
H2C03° 102 102 102
HP04^- 114 114 114
H2PO4- 570 570 570
H3P04° 0.03 0.03 0.03
CaDTPA^- <0 .0 1 0.64
CaHDTPA^- < 0 .0 1 1.14
Ca2DTPA- <0 .0 1 0.94
MgHDTPA'- < 0 .0 1 0.04 --
FeDTPA^- 8.34 9.08 ...
FeEDDHA- 13.2
FeHDTPA- 0.03 0.03 ...
ZnDTPA'- 0.18 0 .2 0 ...
ZnHDTPA^- 0 .1 0 0 .1 1 ...
CuDTPA^- 0 .1 0 0 .1 0 ...
CuHEDDHA- 0.18
Table 6-26. Calculated percent distribution of various ions in a rose solution (Table 6-22) and treated
as in Table 6-25 to two chelates and differing concentrations when originally mixed. Conditions the same
as given in Table 6-25 legend. Data from Lindsay's modification of Allison et al.'s
' ^ a neutral, undissociated molecule.
15 fiM & 15 fiM 15 nM & 30 fiM 15 fiM Fe^^ & 15 nM
Major DTPA DTPA EDDHA
ion Complex Percent of Complex Percent of Complex Percent of
major ion major ion major ion
DTPA^- MnDTPA^- 1 MnDTPA^- 29
FeDTPA^- 92 FeDTPA^- 50
ZnDTPA^- 4 ZnDTPA^- 2
ZnHDTPA^- 1
CuDTPA^- 2 CuDTPA^- 1
CaDTPA’- 7
CaHDTPA^- 6
Ca^DTPA- 4
EDDHA“- FeEDDHA- 99
CuHEDDHA- 1
P0 4 ^- MgHP0 4 ° 3 MgHP0 4 ° 3 MgHP0 4 ° 3
CaHP0 4 ‘^ 11 CaHP0 4 ° 11 CaHP0 4 ° 11
CaH2P0 4 * 3 CaH2P0 4 * 3 CaH2P0 4 ^ 3
HP0 4 ^- 19 HPO4'- 19 HPO4'- 19
H2P0 4 - 64 H2P0 4 - 64 H2PO4- 64
Fe'^ Fe'" 48 Fe^^ 53
FeOH* 19 FeOH^ 21
FeS0 4 ° 1 FeS 0 4 ° 1
FeH2P0 4 * 10 FeH2P0 4 ^ 10
FeHP0 4 ° 13 FeHP0 4 ° 15
Fe(II)DTPA^- 7 Fe(II)DTPA’- 87
Fe(II)HDTPA^- 1 Fe(II)DTPA^- 13
Fe'^ FeDTPA^- 100 FeDTPA^- 100 FeEDDHA- 100
Mn'^ 95 2 Mn'^ 97
MnDTPA^- 2 MnDTPA^- 96 MnS04° 3
MnS0 4 ° 3 MnHDTPA^- 2
Zn'^ Zn^* 6 Zn'" 80
ZnDTPA^- 71 ZnDTPA^- 77 ZnOH^ 4
ZnHDTPA^- 21 ZnHDTPA^- 23 ZnS0 4 ° 3
ZnH2P0 4 ^ 1
ZnHP0 4 " 11
Cu'T CuDTPA’- 95 CuDTPA’- 95 Cu'^ 23
CuHDTPA^- 5 CuHDTPA^- 5 CuC0 3 ° 2
CuHP0 4 ° 3
CuOH^ 1
Cu(OH)2° 2
CuHEDDHA- 64
CuH2EDDHA^' 3
465
Adding more DTP A also increases chelation of manganese, zinc, and copper. EDDHA, however, complexes
about 67% of the copper and none of the manganese and zinc. There might be no buffering of the latter two in
solutions with EDDHA. The remainder of the macronutrients generally exist as free ions above 90%, according
to the computer program. These tables, showing possible equilibrium status of common nutrient solutions, are
the first I have seen in the horticultural i ature. My objective here is to illustrate the possibilities that exist in
modern greenhouse nutrition, and the fact that most investigators fail to acknowledge any awareness of what
goes on in their fertilization studies. Published results are, therefore, severely limited in scope and reliability.
Remember that synthetic chelates are also organic compounds. They are subject to breakdown by
microorganisms. Furthermore, microorganisms require the same micronutrients as higher plants. Competition
occurs, analogous to nitrogen immobilization when crop residues with wide C/N ratios are applied [Stevenson,
1991]. Sonneveld and Voogt [1980] reported that the sudden drop of manganese in recirculating hydroponics
that often occurs, is due to manganese-oxidizing bacteria. The Mn content of the nutrient solution is not always
a reliable measure. Clay mineral soils, in combination with organic matter, are major components involved in
micronutrient retention. Organic matter contribution is highest when the predominant clay mineral is kaoUnite
and lowest when it is montmorillonite [Stevenson, 1991]. Mooraghan and Mascagni [1991], however, point out
that when boron, for example, is released from soil minerals, it can be rapidly leached because of its nonionic
nature. In contrast, increasing soil organic matter can cause copper deficiency due to complexing of copper in
insoluble organic forms.
E. MACRONUTRIENT FERTILIZATION AND CONTROL

Readers will not see in this text specific fertilizer recommendations for particular species except as examples.
A perusal of the most recent issues of the horticultural d ature [1996] will show that most articles deal with treat
ment conditions as to N, P, and K. For reasons expressed at the start of this chapter, I do not agree with the
existing methodology. There is little uniformity. Typical of specific fertilizer recommendations for crops is
Joiner's [1983] review of ornamental fertilization in the greenhouse industry. One may also refer to publications
that deal with a specific crop or several crops for greater detail. It is my opinion that fertilizer recommendations
should be given in terms of millimoles per liter (mmol f ') or per cubic meter (mmol m'^), or millequivalents per
d (meq i •), and the recipe should deal with those ionic forms available to plants (e.g., N 0 3 ', N H /, H2 P 0 4 ',
etc.).
1. Inorganic Fertilizers
Inorganic fertilizers as listed in Table 6-5 have many advantages. First, those used in agriculture are
relatively pure, particularly compounds highly soluble and employed in hydroponics. Secondly, their high
analysis of macronutrients means that they can be shipped long distances at low cost. Thirdly, the grower can be
assured of a uniform, consistent source. Concerning the last, some care is required that handling procedures have
not contaminated the product. There have been examples of bagging machinery contaminated with weed killers,
resulting in crop damage. Or, placement of fertilizers on surfaces, or in containers, having had herbicides stored
previously can cause significant damage. Movement of 2,4-D or Ureabor into amendments or fertilizer bags can
happen. The grower finds a crop damage pattern in his greenhouses that is difficult to diagnose since there is
neither record of where the particular bags were stored, nor which bags were used on the particular damaged
location. Fertilizer and herbicide storages should be distinct and well separated. Another problem is the
application control. Uneven application can result in uneven plant response, or through miscalculation, excess
is applied, leading to salinity or specific phytotoxicities. As will be discussed further in Section 6 .III.F, soil
testing, an attempt to correct a perceived deficiency in one step is dangerous. The secret to a good fertilizer
program is one of gradual progression based upon observation, and tissue and soil analysis. Many greenhouse
cropping patterns are too short to correct nutrition -especially if readily observable- before the product is to
be sold. The grower must set up a change from one crop to the next. This requires experience and good judgment
since soil and tissue analyses are usually after the fact. Inorganic compounds are also rated according to their
contribution to salinity and equivalent acidity or basicity. For the former, the compound is compared with NaN 0 3
as a reference. KCl, for example, has a salt index of 114, as compared to superphosphate with an index of 10.
Most compounds used in hydroponics have salt indices ranging from 8 for calcium sulfate to 104 for ammonium
nitrate. Acidity is expressed as the

equivalent weight of CaC0 3 per 45 kg of
the material. In some countries, the com
parison is made to CaO. Using CaC 0 3 ,
the neutralizing value of CaO is 178,
MgO 250, and MgC 0 3 119.
a. N i tr o g e n
In field soils, nitrogen applied in
fertilizers undergoes the same kinds of
reactions as does nitrogen released from
plant residues. Fertilizer nitrogen can be
present as nitrate, ammonia, and urea
[Brady, 1974]. Urea is subject to
ammonification, nitrification, and utili
zation. Ammonium can be oxidized to
nitrate, fixed by the soil colloids, or used
(Fig. 6-27). Nitrate can be lost by
volatilization, leaching, or use by mi
crobes and plants. Ammonium-
containing fertilizers increase acidity. In
any given climatic condition, soils
develop a “normal” nitrogen content, and
attempts to raise this content materially
higher will be accompanied by waste. O f
the fertilizers given in Table 6-5, Days and Date
ammonium nitrate can be explosive and

its direct use is prohibited in some Fig. 6-43. Uptake of P by rye grass when fertilized by different P sources
[Bunt, 1980] (01980, Int. Soc. Hort. S c l, A c ta H ort., 99:25-32).
countries unless mixed with limestone
(nitrochalk). Mengel and Kirkby [1982]
cite some rates of application ranging from 90 to 135 kg N ha'^ in the field. In container production,
recommendations for fertilizer additions were given in Tables 5-4 through 5-6. Note also that some amendments
contain significant nitrogen and other macronutrients (Table 5-7). There is some danger if a mixture is stored for
lengthy periods. The salts may dissolve, increasing salinity over time. Composts to which fertilizer has been
added should be used shortly after mixing. I have observed considerable crop damage if the grower steam-
pasteurizes following the application of highly soluble, or some slow-release, fertilizers. About the only inorganic
materials that can be safely steamed are limestone and 20% superphosphate. As will be noted later,
macronutrients can be applied automatically with irrigation.
b. P h o s p h o r u s
The total amount of phosphorus in soils compares favorably with nitrogen, but it is much lower than
potassium, calcium, or magnesium, and much of it is unavailable to plants (Table 6-18). Brady [1974] states that
the tonnage of phosphorous-supplying materials exceeds all except the nitrogen carriers. Phosphorus can complex
with aluminum, calcium, iron, and many other materials that can depress P availability. According to Mengel and
Kirkby [1982], most P uptake by plants results from root growth to unexplored areas -which is unavailable in
container production or hydroponics. Soils containing iron oxides and clay are particularly prone to strong
phosphate fixation and often require very high fertilizer applications. Application rates up to 100 kg P ha'^ may
be found in the Hature. Rates of 0.5 to 1.0 kg superphosphate per 10 m^ are usual when applied to raised soil
benches. Of the phosphorus sources listed in Table 6-5, the superphosphates are most commonly employed for
greenhouse production in the ground or benches as a preplant addition. Availability is determined by granule size
plus chemical form. An example of phosphorus uptake by grass for several sources is given in Fig. 6-43. Both
ground rock phosphate (apatite) and bonemeal (organic) are generally rejected due to their low availability
467
Table 6-27. Some common organic fertilizers [Hanan et al.,1978].
Material Acidity Analysis^ Remarks^

Activated sludge Acid 4 -6 ,2 -4 ,0 Can contain heavy metals. Rates 2 kg per 10 m^. Some
are proprietary.
Beet sugar residue Basic 3-4,0, 10 Mostly ealcium, micronutrient traees.
Castor pumice Acid 5 -6 ,0 ,0 2 kg per 1 0 m^, poisonous to animals.
Cocoa shell meal Basic 5% total Used as conditioner in some fertilizers.
Cocoa tankage Basic 4, 1.5,2 May contain 20% lime.
Dried blood Acid 9 -1 4 ,0 ,0 1 kg per 10 m^, N readily available.
Steamed bone meal Basic 2 ,25-30,0 2 kg per 10 m^, 20-30% calcium.
Fish scrap Acid 9,7, trace 1 -2 kg per 1 0 m^
Garbage tankage Basic Variable
Guano Acid 12, 11,2 Contains 8 % ealeium. If leached, may have 25% P2O5 .
Hoof and horn meal — 13, 0, 0 2 kg per 1 0 m l
Kelp (seaweed) — 2 ,1,4-13 1-1.5 kg per 10 m l
Cottonseed meal Acid 7 ,2 -3 ,2 2 kg per 1 0 m l
Oyster shells Basic 31-36% calcium, liming material.
Process tankage Acid 4 -1 2 ,0 ,0 Sometimes a conditioner in fertilizer.
Rapeseed meal — 6 ,2 ,2
Ground rice hulls — 0.5, 0.2, 0.5 Very light and bulky.
Shrimp bran Basic 7 ,4 ,0 57% calcium.
Soybean meal — 6 ,1 ,2 Oil removed, mostly feed for animals.
Tobacco stems Basic 2 ,0 , 6 3%Caand 1%C1.
Animal tankage Basic 7 ,1 0 ,0 Principally in animal feeds.
Vermiculite Basic 0-2, trace, 2-3 Mostly a soil conditioner.
Wood ashes Basic 0 ,0 ,5 23% Ca, 2 kg per 10 m l
Analysis for N, P2O5,, and K2O.
^ Rates for a beneh 15 cm deep.
[Bunt, 1988]. As with other nutrients, phosphorous can be applied in the irrigation water. Suggestions will be
given later.
In general, most greenhouse species can tolerate high P levels unless the concentrations affect the availability
of other nutrients such as iron. However, the use of organic-based potting substrates, coupled with some species,
will cause toxicity problems. Among sensitive species mentioned by Bunt [1988] are the Proteaceae and some
hardy flowering shrubs such as Cytisus x praecox. Orthophosphates at application rates above 50 mg have
been mentioned as causing damage to sensitive species. Tomato and chrysanthemum, on the other hand, are
highly tolerant to excess phosphorus. P fertilizers can have sufficient fluorides present to cause phytotoxicity in
many ornamental foliage plants. This can be avoided by using phosphoric acid in place of superphosphate,
avoiding water with more than 0.25 ppm F, and raising the mix pH to 6.0 to 6.5.
c. Potassium
Of the inorganic fertilizers given in Table 6-5, potassium chloride (muriate of potash) is not generally recom
mended for greenhouse production since it contributes chloride and increases salinity compared with other
compounds. It is usually much cheaper than other sources. Vermiculite is the only material in loamless mixes
containing sufficient potassium. For others, K must usually be added. In field soils, the main source is usually
from weathering of K-containing minerals. Mengel and Kirkby [1982] suggest an application range of 40 to 250
kg K ha‘‘ yr l The response to K uptake depends to a considerable extent on the N nutrition level. Interactions
between K and N were examined in previous sections. Deficiencies often appear where has been leached from
light sandy soils or highly leached lateritic soils. Organic soils and peats are usually low in K.
2. Organic Fertilizers
The organic fertilizers listed in Table 6-27 are not widely used in modem greenhouse production -except
for vermiculite and one or two others used as soil or fertilizer conditioners. They have two distinct disadvantages:
1) A low nutrient level, which means that considerable quantities may be required to supply sufficient nutrient.
Shipping organic fertilizers any distance is expensive in terms of fertilizer content. 2). The analysis is likely to
be variable, so the grower is not assured of a consistent nutrient supply. This is particularly tme in respect to
manures. Manures can cause ammonia toxicities, especially if steamed. There is the aesthetic problem with a
product sold to the general consumer. Although salinity problems are not likely to arise, the high organic content
of most materials can fix nutrients, resulting in deficiencies. Rapid decomposition can also tie up nitrogen.
Organic substances can also be considered as slow-release fertilizers since nitrogen mineralization and release
requires microbial action. The benefits of organic fertilizers are sometimes more likely to be derived from their
effects on the substrate physical characteristics -such as better aeration and water relationships. No matter the
organic content, nutrient availability requires the ions to be present in solution. The nutrient supply is the same
as that from an inorganic source [Brady, 1974; Mengel and Kirkby, 1982].
3. Controlled or Slow-Release
Fertilizers
The possibility of applying
fertilizers that release nutrients
gradually has intrigued growers for 5000 80
many years. Once-only application,
either by mixing in the substrate prior
60
to planting, or as a top dressing, m
would certainly reduce labor costs. % 3000
Where container plants are sold, 40 QW >
m
inclusion of a nutrient supply can 2 2000 c<¡3
enhance keepability and consumer O
~ 20
satisfaction. Most organic fertilizers, Z 1000
mentioned in the previous section, and
some inorganic compounds such as LLJLJL J_ _L -J 0
su p erp h o sp h ate, p o tassium 0 3 6 12 24 36 48 72 96

metaphosphate, sulfur, etc. could be Days
considered as slow release, with
rapidity controlled by particle size.
However, materials deliberately
manufactured to provide essential 60
elements over significant periods are
given in Table 6-28.
40 £
O f the types listed in Table 6-28,
0)
magnesium-ammonium-phosphate is m
CQ
probably considered the original slow- 20
release fertilizer. Urea formaldehyde, m
a common material, is not usually CL
recommended for use on carnations in

Colorado. During winter seasons,
temperatures are too low to allow
sufficient release. As temperatures Fig. 6-44. Effect of temperature on release rates of coated fertilizers: Upper:
rise in the spring, excessive nitrogen Osmocote 14-14-14. Lower: Osmocote 18-6-12, 9-12 mo. formulation [Har-
may be released. Fig. 6-44 shows that baugh and Wilfret, 1982] (With permission of the F lo rists ’ R ev.).
Table 6-28. Some examples of slow-release fertilizers used in greenhouse culture.
Factor causing Effect of Examples

Type Name Trade names or acronyms release temperature Duration
rise on re
lease
Organic Urea formaldehyde Nitro-form, (UF): Several polymers - i.e. Microbial activity High 38% N, few weeks to
methylene-, di-, tri-, and tetramethylene months
diurea
Iso-butylidene diurea (IBDU) Particle size, Slight 32% N
dissolution
Crotonylidene diurea Triabon, Crotodur, (CDU) Microbial activity, High 28% N, 16-18-12-4^
particle size
Diamide of oxalic acid Oxamide Particle size, Slight 32% N
dissolution
Sulfur-coated urea Gold-In, (SCU): microbicide may be in Various dissolution Slight Up to 8 weeks, 6 , 5 or
cluded rates 1 % per day
Inorganic Polyolefin-coated Nutricote Membrane, pore High 100-360 dys, 13-13-
size and number 1 1 , 16-10-10, etc.
Co-polymers of Osmocote Membrane, pore High Many formulations,
dicyclopentadine and glycorol size and number 14-14-14, 15-11-13-2,
ester 18-6-12,3-9 mo.
Magnesium-ammonium- Magamp (7-40-6-12)^, Enmag (6-20-10- Particle size Slight (Table 6-5)
phosphate 8.5)*’
Dicalcium phosphate Particle size Slight see Table 6-5
Frits: K, mainly micronutrients. Potassium frit, 29% K Particle size, pH Slight Several formulations
Compounds fused with sodium FTE 253A (2-B, 2-Cu, 12-Fe, 5-Mn, 0.13- for micronutrients
silicate Mo, 4-Zn‘’), #36 (0.5-B, 2-Cu, 9-Fe, 2-Mn,
0.5-Mo, 2-Zn)‘’
^Values are percentages of N, P2O5, K2O, and MgO in the formulation.
Values are percentages of N, P2O5, K2O, and Mg in the formulation.
^Numbers are percentages of elements in the glass frit._______________
4^
os
so
Fig. 6-45. Two examples of fertigation of greenhouse crops. Upper: Tanks for mixing fertilizers
at the final concentration. Lower: System that proportions each individual fertilizer salt as required.
temperature has a marked influence on release from coated fertilizers. Thus, the grower must keep in mind the
likelihood of rapid release if substrate temperatures rise excessively. Oertli and Lunt [1962] found release from
encapsulated fertilizers to nearly double on a rise from 10 to 20 C. Soil moisture content did not appreciably
affect the rate of transfer through the membrane, with an efficiency of recovery from 25 to 45%. Release rates
are much slower if the fertilizer is applied as a top dressing. Recommendations are to mix into the substrate prior
to planting.
•■•»-• - •" - - ' ■
*? " ' •lyi
if f'i
,^;-v
$ \ -“; i
/ . > •
■ M l
Vj }"-"-V'
F .5 ^ ' C
Fig. 6-46. Several examples of proportioning systems for greenhouse production, ranging from water powered to electrical, fixed or variable ratios, and differing costs.
-ii-
Materials, such as frits, have also been employed in NFT to provide micronutrients. Glass frit 253A (Table
6-28) was recommended at a rate of 25 g per meter of gully -suitable for all trace elements except iron [Hall and
Wilson, 1980]. Dolomitic limestone has also been suggested for NFT at 125 g per meter to halt pH drift and
stabilize acidity. Use of slow-release fertilizers in container production has been found comparable to liquid
feeding [e.g., Harbaugh and Woltz, 1989; Barragry and Morgan, 1978]. According to Roude et al. [1991],
however, slow-release fertilizers such as Osmocote had no effect on the longevity of chrysanthemums per se.
Total nitrogen, regardless of a fertilizer source, had the greatest effect on longevity. In highly porous soils, such
as the sandy types in Florida, slow-release fertilizers can maintain sufficient nutrition without rapid loss in
recovery efficiency by plants and without significant ground water contamination (see Section 6 .VIII).
Nelson [1978] gives rates for a 14-14-14 Osmocote of 7.3 kg m'^ in light, course-textured media, and 6.5 kg
m'^ in medium-textured substrates. The rate is lower for a 19-6-12 formulation, while a medium application for
an 18-5-11 is 8.3 kg m‘f The supply for the first two is 3 to 4 months, and 12-14 months for the 18-5-11. About
14 g is suggested by Nelson per 15 cm pot as a top dressing. Mengel and Kirkby [1982] state, however, that
nitrogen recovery as measured by the proportion of N taken up by the crop of total N applied, is generally low
for slow-release fertilizers as compared to water-soluble compounds. A further disadvantage is their high cost
per unit weight of N. Despite these latter comments, such fertilizers are a viable means of nutrient control in
greenhouse production, especially as amounts can be added to substrates at rates that would commonly salinize
the medium if applied as soluble fertilizers.
4. Fertigation and Solution Form ulation

Up to the late 1940s and early 1950s, dry fertilizer applications, either as broadcast or mixed into the
substrate, were the principal means of nutrient control. By the late 1950s, it was usual to find some system of
placing soluble fertilizers into the irrigation water. There are many systems with varying advantages and
disadvantages. The principal advantage of fertigation itself, whatever the method, is that injection into the water
supply allows a more consistent fertilizer application that varies with crop growth and season. Thus, it is a
semiautomatic program that varies with irrigation frequency and amount. With the ability to use consistent, high-
analysis chemicals, the grower may increase or decrease an application rate according to tissue and soil analysis
with less likelihood of deficiency or excess. It is applicable to any of the cultural systems employed in
greenhouses.
a. Systems
Examples of systems are presented in Figs. 6-45 and 6-46. The upper picture in Fig. 6-45 shows storage
tanks in which soluble compounds are dissolved and pumped directly to the crop. It is common to more
undeveloped regions where specialized equipment can be difficult to obtain and maintain. Care is required to
ensure adequate mixing and complete solution of the fertilizer. In early years, it was usual for growers to modify
coal stokers that added fertilizer to the tank under agitation. The lower figure in 6-45 is a highly sophisticated
system whereby each nutrient can be added at its own regulated rate, and can be linked to a suitable computer
for automatic control. Chapter 8 covers further discussion of automatic fertilizer control systems.
Almost any type of proportional injection system can be purchased that maintains a fixed ratio of chemical
to each unit of irrigation water supplied. Fig. 6-46A shows a common system of water-powered injectors that
proportion concentrated solutions from the two tanks into the water supply at a fixed 1:200 ratio. Therefore, the
concentration in the supply tanks must be 200 times the final dilution. This can lead to problems since the high
concentrations can be difficult to fully dissolve in cold water. Greenhouses having a steam supply can inject
steam to heat the water. Another problem is that not all fertilizers can be mixed. Phosphorous-containing salts
must be mixed separately from those containing calcium. Otherwise, there will be precipitation and the grower
winds up with a sludge at the bottom of his tanks. This is the reason for two concentrate tanks in Fig. 6-46A.
Pictures 6-46B, D, E, and F are also water-powered injectors. B is common where the fertilizer is injected from
the storage tank by displacement and is adequate for small areas. E and F are inexpensive devices. E, usually
designated as a ‘‘Hoze-on,” sucks the fertilizer into the irrigation line by means of a venturi at proportional rates
of about 1;24. F is called the “Dole” valve containing a variable diaphragm that changes an orifice diameter with
pressure across the membrane. These are usually placed in the suction line of the irrigation pump. Obviously,
these (E and F) are much cheaper than the other systems depicted. However, their proportioning rates are much
473
less precise. C is a variable rate, electrically powered injector. The grower may manually set the injection ratio,
if the volume of irrigation water through the system is constant. Fig. 6-46D shows a water-powered, inexpensive,
pump that wastes part of the water that powers it. D and A can adjust to variations, within limits, in the flow rate
of the main water supply. More commercial systems, in recent years, are approaching the versatility of the bottom
picture in Fig. 6-45. These newer systems may also be linked with conductivity cells, pH meters, and ion selective
sensors for continuous measurement and adjustment of the injected solution on a real-time basis. Several specific
ion probes allow continuous measurement of nitrate, ammonium, sulfate, etc. in the irrigation water. Changes
in water composition can be automatically compensated [Alberry et al., 1985; Gieling et al., 1988; Gieling and
Schurer, 1995]. Measurement of individual ions remains a problem, however. The greenhouse environment is
“noisy” to electronic equipment. Some electrical measurements in the greenhouse are difficult to carry out with
any accuracy, unless special provisions are made to shield the sensing equipment and its attendant wiring.
b. Solutions
Injection solutions for plant culture are diverse. Where the grower is producing in ground, or in substrates
containing soil, a “complete” solution is not often employed. The grower may combine practices by preplant
fertilization with lime, superphosphate, or mixing in some slow-release fertilizers and using fertigation to
supplement the preplant with such ions as ammonium, nitrate, and potassium. A producer may not always use
fertigation with each watering, depending upon soil tests and observation to assess crop performance. Complete
solutions of micronutrients are not often employed. Location, substrate type, and plant species may require
special additions of one or more trace elements. In Colorado, for example, boron is always added in carnation
culture. Zinc is also included in more recent years as piping systems have been replaced with plastics. Many
growers purchase complete, proprietary micronutrient mixtures on the idea that if a little is good, more is better.
With pure chemicals of the major macronutrients, sufficient trace elements may not be present as contaminants.
Hydroponic solutions in good water probably require most of the micronutrients.
Any of the solutions listed in Table 6-22 may be employed to produce acceptable plants. In fact, experience
at Colorado State University has shown that the carnation solution can be used to grow any species in any cultural
medium. This does not mean that plant response will always be the best. Unfortunately, hydroponic solutions
proliferate almost like lemmings. Steiner [1968] reported that more than 300 recipes had been published. If, for
example, carnations grow well, it is published as a “carnation” solution. Jones [1983] listed 22 solutions given
by various authorities. Cooper [1979], of course, reported his own NFT solution with various modifications.
Similar recipes may be found in the publications from research institutions [e.g., Jones, 1983; Ellis et al., 1974;
Hanan and Holley, 1974; Maxwell, 1972; Hoagland and Amon, 1950; etc.]. One finds in the literature, innu
merable reports on the effects of individual fertilizers, the use of complete, soluble fertilizers, and the require
ments of suitable nutrient balance, depending upon growth stage and particular grower objectives [e.g., Lawton
et al., 1989; Harbaugh and Woltz, 1989; Roude et al., 1991; Weston and Zandstra, 1989, Melton and Dufault,
1991; Boertje, 1980; etc.].
One suspects that many of these solutions could be the “besf ’ for the particular conditions under which they
were derived. Several investigators in the field have expressed to the author that Hoagland solutions are perfectly
adequate for most purposes. Two things are usually neglected in constituting fertigation or hydroponic solutions:
1 ) individual salts added to water do not always remain the same, regardless of how consumed by plants; and 2 )
little attention is given about whether a particular solution is physically possible. For 1), reference to Tables 6-25
and 6-26 should be sufficient to be obvious. For 2), Steiner's work [1961; 1966; 1968; 1969] on electrical
neutrality and composition, and the development of his “ideal” solution (Table 6-22) was the first to receive
adequate attention -although nearly neglected in any practical application. Publication of so-called “ideal”
solutions in terms of ppm of elemental nutrients are nearly useless since neither nitrogen nor phosphorous, and
some micronutrients, exist as such in nature.
Another factor, often neglected, is the relationship between concentration and rate of supply -o r flow rates
in hydroponic solutions. It is typical to reduce Hoagland's solutions (Table 6-22) to one-half strength or more in
research problems. Typical is Letey et al.'s [1982] determination of minimum nitrate concentrations for lettuce
and tomato. The authors neglected, however, to suggest any agitation of their nutrient solutions except in passing
that the dense root systems resulting after 60 days probably resulted in insufficient stirring. Their results showed
Table 6-29. Solutions given in Table 6-22 expressed in milliequivalents per ^ . Micronutrients not considered as necessary in testing electrical neutrality„
o
Ion Hoagland 1 Hoagland 2 Modified Johnson's Steiner's NFT solution Carnations Roses Tomatoes Cucumber Pepper
Hoagland 0.25 strength formula
— —
o^
NH4^ 1.0 0.4 0.5 2.5 1.0 0 .5 0 .5
K" 6.0 6.0 2.4 1.5 13.151 7.964 6.0 4.0 7 .0 5 .5 6.0
Ca^" 10.0 8.0 3.2 2.0 1.644 8.5 3.0 5.0 7 .5 7 .0 7.5
4.0 4.0 1.6 0.5 1.644 4.154 2.0 2.0 2 .0 2 .0 2.5
H,P0 4 - EO 1.0 0.4 0.5 0.822 2.192 1.1 1.0 1 .5 1.25 1.25
NOy 15.0 14.0 5.6 3.5 13.151 14.272 10.4 9.0 10.5 11.75 12.25
S0 4 ^- 4.0 4.0 1.6 0.6 2.466 4.154 2.0 2.0 5.0 2.0 2.5
Total meq/i 40.0 38 15.2 9.1 32.878 41.236 27.0 24.0 34.0 30.0 32.0
Estimated 4.0 3.8 1.5 0.9 3.2 4.1 2.4 2.3 3.9 3.5 3.6
EC‘
^ Electrical conductivity in dS estimated from Fig. 6-8.
Table 6-30. Example of setting out a physically possible fertigation or hydroponic solution to ensure initial electrical neutrality, using the
carnation solution from Tables 6-22 and 6-29.
Milligrams per d
Soluble fertilizer salt to be Cations Anions required of each
added Na^ Ca'" Mg'" NH 4^ NO 3 H 2 PO 4 SO 4 ' cr HCQ f fertilizer salt
Potassium nitrate 6.0 6.0 606.0

(KNO3)
Calcium nitrate 3.0 3.0 354.0
(Ca(N03)2'4H20)
Magnesium sulfate 2.0 2.0 246.0
(MgS04-7H20)
Ammonium nitrate 1.4 1.4 112.0
(NH4NO3)
Monoammonium phosphate 1.1 1.1 126.5
(NH4H2PO4)
Total milliequivalents 6.0 3.0 2.0 2.5 10.4 1.1 2.0
Total cations = anions 13.5 13.5
475
no effect of nitrate concentration on accumulative uptake of the ion until after the first 40 days of growth, even
at lowest concentrations of 140 pM N 0 3 '. This is expected given the previous discussion in Section 6.III.D.1
(Fig. 6-16), which showed that NO 3 ' uptake remained constant down to levels of 3 pM. The relationships between
concentration and supply rate are the thrusts of such articles on flow rates by Edwards and Asher [1974] and
Ingestad [1974; 1982]. Such studies form the impression that if an ion could be presented directly to the
absorbing membrane as required, concentrations of macronutrients in the external solution could well be less than
1 pM r f But, ensuring minimum diffusion distances even at the outer surfaces of roots is well nigh impossible
under commercial conditions. NET systems are not intended to provide fast flowing solutions to an entire root
system of a well-developed crop. In container production, the intense competition between roots in a confined
volume requires higher concentrations to ensure adequate nutrient supply.
Although I have shown that a solution for fertigation does not remain the same after chemicals are added
-certainly not after being added to a root substrate; there are means to prepare complete solutions that ensure
reasonable electrical neutrality and where the solution is physically possible. Assume that a grower is producing
carnations in rockwool, so we select the carnation solution in Table 6-22. Table 6-29 shows the same solutions
with macronutrient concentrations converted to levels given in equivalencies (meq (Section 6 .II.C).
Micronutrients are not included since their total concentrations are very small. Also, estimated from Fig. 6 - 8 is
an EC value for the respective solutions based on total meq f f A second table (6-30) shows the salts necessary
to provide the respective ions and the amount necessary to dissolve in a i of solution. In good water, at these
concentrations, precipitation should not occur and the pH is estimated to be 7.0. Columns for the ions H^, Na^,
Cl’, and HC0 3 ’ are included since we will have occasion to use a table of this type later in salinity discussion. If
a dilution system is being used such as a 1 : 1 0 0 proportioner, then the amounts necessary for 1 H are merely
multiplied by 100 times the volume of the concentrate tank. Here, as suggested above, calcium nitrate must be
dissolved in a separate tank. Some solutions in Tables 6-22 and 6-29 require the use of calcium sulfate and
calcium diphosphate, which are not highly soluble. They cannot be used in a typical proportioning system unless
added separately at their final concentration. Note that the amount per ^ is obtained by multiplying the
equivalents required for the fertilizer by its equivalent weight (Table 6-5). The assumption is that each salt will
be completely dissociated in solution so that, for example, 6 meq f ’ KNO3 will provide 6 meq f ’ and 6 meq
r ' NO 3 ’ (Eqs. 6.1 and 6.2). As Tables 6-25 and 6-26 show, this is not always the case.
Table 6-30 is highly appropriate for use in a spreadsheet, and the necessary calculations can be easily carried
out by the spreadsheet program. A table of this type also presents each macronutrient required by plants, in the
form taken up by plants and shows the total concentration in values that can be used to estimate salinity. It
eliminates the N-P-K formulae so commonly given in ppm or mg t \ and ensures practicality -i.e., the solution
is not likely to precipitate, leading to difficulties with trickle irrigation systems or leaving a part of the fertilizer
solution in the dissolving tanks to be thrown out.
5. Examples of Interactions in
Container Substrates
Application of a particular
fertilizer to supply a known amount
of nutrient does not mean that the
expected response will occur.
Especially in organic container
substrates, response may be
opposite to that needed. One of the .
best descriptions of such responses
to fertilizers and their interactions is
presented in Bunt's Media and
Mixtures for Container-Grown
Plants [1988]. My purpose here is to
Fig. 6-47. Tomato seedlings grown in the winter with equal amounts of
show some examples taken from
ammonium from three sources: Left-to-right, ammonium carbonate, urea, and
Bunt's text. ammonium sulfate [Courtesy of A.C. Bunt].
Fig. 6-48. The effect of superphosphate and nitrogen

source on snapdragon growth. From left-to-right, phos
phate application rate decreases in both pictures. The top
picture are plants with urea as the N source, the bottom
picture plants fed with calcium nitrate as the N source
[Courtesy of A.C. Bunt].
in his discussion of peat types, Bunt presented

several practical examples of the interactions be
tween fertilizer types and plant growth in peat-
sand container mixes. When it comes to chemical
or organic fertilizers, the results can be startling.
Fig. 6-47 shows tomatoes grown in the winter,
receiving equal, but from different sources,
ammonium. Plants grown with calcium nitrate (not
shown) were larger. Free ammonia can occur at
pHs close to neutral and increases as pH rises.
With an organic fertilizer (urea), the rise in pH can
Fig. 6-49. Tomato plants grown in the winter in four

peat-sand mixes. Upper left, lowN, low CaC0 3 ; Upper
right, high N, low CaC0 3 ; Lower left, low N, high
.. m M . &
CaC0 3 ; Lower right, high N, high CaC0 3 . N supply
70% from hoof-and-horn meal, 30% from NH4NO3. Same vi rW #
treatments in the summer [Courtesy of A.C. Bunt].
■
release considerable amounts of toxic ammonia. A „ :M l

second example of interaction between two
fertilizer types in peat-sand mixtures is Fig. 6-48.
The upper picture shows that greatest response to
an organic fertilizer occurs with highest superphos
phate rates, but such response to increasing
superphosphate does not occur when calcium
nitrate is employed. In peat substrates, superphos
phate will reduce pH, which slows nitrogen min
eralization rates and significantly reduces the free |
ammonia and nitrite levels.
The relationship between climate and
fertilizers was examined by Bunt as shown for the winter treatment in Fig. 6-49. During the summer, there were
essentially no differences between the treatments. The four treatments received calcium carbonate at low or high
rates while 70% of the nitrogen was supplied from hoof-and-horn meal and 30% from ammonium nitrate at low
or high levels. A part of the reason for the differences between summer and winter results from higher summer
temperatures which about double nitrogen mineralization and, with higher radiation, plant grovvTh can be greater
with less risk of toxicity. True mineral deficiencies are more quickly observed in summer than in winter. A third
example is Fig. 6-50 where urea and calcium nitrate are compared with different sources of calcium. With
calcium nitrate as the nitrogen source, growth was similar for all three liming materials. With urea, however,
calcium sulfate was superior to calcium carbonate and dolomitic limestone. Bunt stated that the higher pH
obtained with CaC 0 3 caused free ammonia and nitrite production. Dolomitic limestone gave a slightly reduced
pH with growth depression correspondingly reduced. Plants grown with ammonium-producing fertilizers are
more susceptible to boron deficiency (Fig. 6-51). Boron uptake can also be suppressed by high phosphorous
levels. Other examples of trace element interactions were presented earlier (Fig. 6-20).
Interaction between three substrates and various ratios of ammonium-to-nitrate is examined in Fig. 6-52.
The pictures are three of several bedding plant species examined by Jeong [1990]. The soil mix (S) consisted
477
Fig 6-50. The top picture shows tomato growth with the
N source urea, the bottom calcium nitrate. From left-to-
right, the calcium source is CaC0 3 , dolomitic limestone,
and calcium sulfate [Courtesy of A.C. Bunt].
of 1 part soil (local Ft. Collins clay loam), 3 parts

sphagnum peatmoss, and 2 parts by volume perlite.
Peat-lite mix (P) was a proprietary mixture
composed of 75% Canadian sphagnum peatmoss,
24% perlite, and 1% vermiculite with dolomitic
limestone and a wetting agent added. All three
media, including the rockwool (R), had no base
fertilizers. The N H /:N 0 3 ‘ ratios varied from 0:100, WBs
25:75, 50:50, 75:25, to 100:0 applied in the
irrigation water. It can be seen that the response to
treatments was much greater in rockwool media as
compared to the buffered peat- and soil-containing
media. Salvia were much more sensitive to either
extreme, particularly 1 0 0 %
NH 4 ‘ and no nitrate. Agératum
response (top), however,
suggests some tolerance to
100% ammonium supply. The
results point up the inherent
criticality of loamless, non-
buffered substrates, and the
specific responses of each
species that were significantly ^
different. O f the 11 species
tested, only agératum showed
tolerance to ammonium. Jeong
found that pH did not affect m
growth and tissue composition,
reinforcing a previous statement
that pH per se is not as im ■ifm
m
portant as the secondary effects
of hydrogen ion concentrations Fig. 6-51. Boron deficiency on snapdragons grown in peat-sand. Left, N source hoof-
on nutrient relationships. and-horn meal; Center, urea; Right, calcium nitrate [Courtesy of A.C. Bunt].
In his Experiment 3, Jeong
compared three NH 4 ^:N0 3 '
ratios of 50:50, 100:0, and 0:100 applied to four bedding plant species grown in peat-lite and rockwool. Each
ratio compared solutions applied with no chlorine versus 4 meq Cl', all solutions fixed at 50 meq total.
Again, responses to these treatments were greater in the unbuffered rockwool. In rockwool, the pictures (Fig. 6 -
53) suggest that there may have been some beneficial effects from chlorine addition. Still, the absence or presence
of Cl' was not apparent on plants grown in peat-lite. Nukaya et al. [1991] showed that increased CT reduced
blossom end rot on tomatoes, but increased gold speck injury. Shelf life decreased with increasing S0 4 ^' and Cl'.
These few examples, combined with previous discussion, serve to emphasize that nutrient interactions and
corresponding plant growth are not simple concentration manipulations. Much information on specific species
is unavailable so a grower is forced to some trial and error.
1 ,#
Fig. 6 -5 2 . E ffe c t o f fiv e N H / : N 0 3 ' ratios a p p lied in irrigation w ater to agératum

(top), c e lo sia and sa lv ia (b ottom ) grow n in a so il m ix (S), peat-lite (P), and rock w o o l
(R ) [C o u rtesy o f B.R. Jeo n g ].
479
Fig. 6-53. E ffect o f N H / ; N 0 3 ‘ ratios w ith and w ith out 4 m eq f ‘ C t on agératum (to p ) and petu nia
(b o tto m ) g ro w n in tw o substrates [C ou rtesy o f B .R . Jeon g],
F. SOIL ANALYSIS
Soil analysis forms the third leg o f nutrient control in greenhouses, the other two being observation and
tissue analysis (Section 6 .III). The purpose o f soil testing is to learn which nutrients are deficient and to estimate
how much fertilizer is required to correct a deficiency and increase yield. According to Jones [1985], soil testing
is the only means o f specifying lime and fertilizer needs. The use o f fertilizers without a soil test, and its
interpretation, is hazardous to successful crop production. Equally important from the discussion on tissue
analysis, soil analyses do not always track tissue concentrations in plants. So there are restrictions on the use o f
soil analyses to handle nutrient control, just as there are on tissue analyses. There are three aspects to be
considered: 1) obtain a representative sample from the crop substrate; 2) obtain a suitable fluid extract for
chemical analysis from the sample; and 3) interpret the result.
1. Sampling the Substrate

a. Mineral Soils
Considerable detail for obtaining a representative sample from a field is given by such authors as Jones
[1985], Peck and Soltanpour [1990], and James and Wells [1990]. However, most sampling is carried out by the
grower, and Peck [1990] states that sampling is the greatest source of error in the testing process. A particular
problem is soil heterogeneity horizontally and vertically. Random versus planned sampling has been discussed,
and one will find James and Wells' discussion worth reading. In greenhouses, however, the small size (<1 ha)
of most establishments, and the fact that a crop is usually growing in a longitudinal bench arrangement, means
that most sampling will be to provide a composite sample for each bench. An assumption here is that the soil is
uniform, and where a single crop is treated similarly for the entire area, a single composite sample from several
benches may be adequate. Although large areas of greenhouse production are in the ground, little attention has
been given to possible problems arising from heterogeneity. Since each bench may be analyzed, any variation
in soil properties from one location to another may be disguised. Thus, sampling a ground bench randomly, four
or more times to obtain a single sample for the bench, is usual for a grower. Long benches, obviously, should be
sampled more times, or until the grower has evaluated uniformity in a single bench. The first time a bench is
analyzed, individual cores should be analyzed so a grower can determine variation and what sampling method
will provide the greatest reliability.
Trickle irrigation systems, particularly if the emitters are 30 cm or more apart, are likely to cause significant
variations in nutrient concentrations with distance from the emitter. Such characteristic, concentric patterns of
pH, EC, and nutrients, extending in cone shapes through the soil profile, were found by White and Prasad [1980]
in tomato culture. Distribution patterns in peat modules were also found. At the end of the growing season, tests
in soils revealed the existence of a deep, narrow, strongly leached zone, directly under the emitter, with low pH,
Ca, and EC values. Depending upon conditions, evaporation and outward salt movement from the emitter
frequently result in salt accumulation at other locations in a bench or container. In peat modules. White and
Prasad found no evidence of leaching immediately below the emitter. Under the conditions found in this study.
White and Prasad concluded that there was no optimal sampling location. They suggested sampling 15 cm from
an emitter, at right angles to the row as the easiest compromise. Such patterns as described -not necessarily the
same- can occur in almost any container or ground bed. Problems with an inadequate irrigation system, or with
climatic extremes within the greenhouse (i.e., south ends of structures in the winter, etc.), will exacerbate
variations in nutrient content of the medium. Growers should allow for these factors when sampling for soil
analysis. That is, areas manifestly different in soil moisture or plant response should be sampled separately. Such
soil tests may point the way to modifications in the cultural program for greater efficiency and better product
quality and yield.
Most analyses for field soils use less than 100 ml soil for extraction. About 0.2 kg dry soil is more than
adequate for sending to the laboratory. Most soil testing laboratories will have special containers for the grower
to use. This means that one or more cores must be well mixed and a subsample of suitable size taken from the
mixture to provide a composite sample. The procedures for mixing and subsampling require care to prevent
contamination and to obtain a representative subsample. After mixing several cores, the soil pile is split. One
remaining half may be split in succession until the remaining half, the final split, is small enough to place in the
laboratory's package.
Depth of a soil core, commonly taken with a stainless steel, hollow tube, 2 to 3 cm diameter, with a sharp
edge, is 15 to 20 cm -o r the cultivation depth (see Fig. 5-39). Under most greenhouse conditions, such coring
tubes can be pressed into the soil, and removed, by one person. Auger-type tools should not be necessary. If the
soil is gravelly, a shovel or spade may be required. If the active root system extends below 20 cm (i.e., roses),
sampling depth should be extended to include the active root zone. An auger may be required here. In raised soil
benches, or pots, a core of the entire depth is taken after first removing the upper half centimeter or so of the soil.
Argo and Biembaum’s results [1994] showed the top 2.5 cm in pots growing subirrigated Easter lilies to have
nutrient concentrations up to 10 times higher than those measured in the remaining root zone. Use of an
evaporation barrier reduced the stratification. In peat modules, two or more samples halfway between center line
and the edge, the entire depth, should be adequate. For pots, a smaller tube of 1 to 1.5 cm diameter can be used.
Some growers make cheap tubes from rigid electrical conduit. As for sample contamination in micronutrient
testing, such sampling tubes are not desirable.
481
b. Container Mixes and Loamless Media

Most samples for mineral and organic soils are dried, ground, and sieved in preparation for extracting the
nutrients for analysis. This is not a good practice where slow-release fertilizers are employed. The general
practice is to obtain a moist sample near container capacity, sometimes in amounts up 500 ml. Whereas dry
samples are often mixed with extracting solutions on a weight-soil-to-volume-extractant ratio, moist samples,
particularly peat-based and loamless media, are extracted based on volume-to-volume. Some laboratories are
careful to bring a sample to a known matric suction (i.e., 1 0 or 60 cm water column) before extraction [van den
Ende, 1971; Sonneveld and van den Ende, 1971; Sonneveld et al., 1974]. This is laborious and not often used
where large sample numbers are encountered. Markus and Steckel [1980] found that placing moist media samples
in sealed plastic bags was adequate although the sample might be in the mail for 2 to 3 days. Individuals dealing
with field soil analysis commonly argue that significant changes in chemical composition of moist samples are
likely to occur. Extracting moist samples markedly speeds up testing.
Sampling container mixes can use the same tools as outlined above. However, with the marked movement
to sample extraction using only water, Wright [1986] discussed the “pour-through” method whereby a liquid
sample was obtained from drainage of the container. Containers should be close to, or at, container capacity, and
additional water applied sufficient to obtain, for example, a 50 ml aliquot from the drainage. The container is the
sample. The grower makes a random selection of containers from the crop for liquid extraction. This procedure
has the advantage that root systems are undamaged versus core sampling of the container. For research purposes.
Nelson and Faber [1986] discussed a displacement method where a 5 cm diameter plastic column, 60 cm tall, was
packed with a f of the sample. About 250 ml of a displacement solution, consisting of 0.5% thiocyanate in 50%
ethanol was poured through the column with several samples taken from the drainage until formation of a bright
red color from the use of a 0.5% ferric chloride solution showed mixing had occurred. Any drainage after that
was discarded. Unfortunately, this method seems laborious for quick testing on a practical basis. Nelson and
Faber, I think, did not adequately detail the packing procedure. Given the high percolation rates of some media
(Tables 5-9, 5-10), sufficient packing to prevent immediate mixing would be difficult.
A.C. Bunt discussed the problems in analysis of organic and lightweight potting substrates [1986]. He
pointed out that drying can cause ingredient separation, reduce amounts of nutrients extracted, and some organic
materials (e.g., peatmoss and bark) are difficult to re-wet. Grinding changes medium structure, increasing bulk
density. The problem with slow-release fertilizers has been mentioned. With the low bulk density of some media
components (0.1 g/cm^), large errors in analytical interpretation can occur if not corrected. Bunt discouraged the
use of displacement methods for any practical testing service.
c. Hydroponic Systems
For hydroponic systems that waste excess water, that is, the nutrients flow once through the inert substrate,
sampling for analysis, other than to check the applied nutrient solution, is unnecessary. The same feeding solution
is applied at each irrigation. As long as suitable precautions are taken to maintain equipment and ensure proper
mixing, there should be no need for numerous analyses. However, as will be noted in Section 6 .VII, some means
of rapidly measuring pH and total salts are good checks for system performance, and ensuring that the water
supply quality has not changed significantly. For recirculating solutions, regular sampling for analysis is an
important procedure. Though continuous measurement of pH and EC and some ions (i.e., nitrate) can be done,
nutrient uptake varies with each ion and over time. The nutrient solution composition will change markedly from
what the grower initially began. Cooper [1979] went into considerable detail on uptake of nutrients in the NET
system. As the crop grows, nutrient uptake increases with many increases or decreases in the main trend that can
be attributed to such factors as root death (onset of fruiting or leaf removal in tomatoes), pinching, changes in
radiant level, etc. Sampling in either case is merely obtaining a water sample to send to the laboratory. The water
sample, here, is analogous to an extracted solution from a substrate sample. The only care is the use of adequate,
clean, plastic containers.
d. Handling Samples
James and Wells [1990] emphasized the care required to prevent sample contamination. Common
contaminant sources include dirty sampling tools, dirty containers, cigarette or pipe ashes, and preparing samples
4^
oo
T a b le 6 -3 1 . E xa m p les o f a num ber o f so il testin g p roced ures in use. T he list is not inten ded to b e com p lete. P roced u res in grin d in g, sie v in g , preparation o f extractant, etc. m ay
b e in c o m p le te . R efer to fo o tn o te s as n ecessa ry .______________________ __________________ __________ _________ O
pH of ex- Reference P
Name Conditions and Extractant chemicals Dilution Shaking time Nutrients tested tracted
media solution o^
M o rg a n u n iversal A ll acid, m in eral 0 .7 3 M so d iu m acetate 2 0 m l dry sam p le, p a s 5 m in and f il P, K , C a, M g, N O 3 , 4 .8 Jones, 1990
s o ils, s o ille s s m ix e s s in g 2 m l s ie v e in 4 0 m l ter N H 4 , S O 4 , Fe, C u, M n,
c
so lu tio n Zn
s.
o'
M o rg a n -W o If u n iv ersal A ll acid, m in eral and 0 .0 7 3 M so d iu m acetate 2 0 m l dry sa m p le 5 m in and f il P, K, Ca, M g, N O 3 , 4 .8 Jones, 1990 o
o rga n ic s o ils 0 .5 2 M ac etic acid p a ssin g 2 m l s ie v e in 4 0 ter N H 4 , S O 4 , Fe, B , Cu,
0 .0 0 0 1 M D T P A m l so lu tio n M n, Zn
Modified Morgan V a r io u s p o ttin g 1.4 N so d iu m acetate 2 0 m l m o ist sa m p le, 1:5 3 0 m in and P, K , C a, M g, N O 3 , 4 .8 M arkus and
m ixtu res, p eat- 1.0 N a c etic acid ratio filter NH4 S teck el, 1980
ve rm ic u lite , etc.
M e h lic h N o . 1 A c id , san d y, m in eral 0 .0 5 N H C l 4 m l dry sa m p le p a ssin g 5 m in and f il P, K , C a, M g. N a, M n, 1 .2 Jones, 19 9 0
u n iv er sa l s o ils, C E C < 1 0 0 .0 2 5 N H 2 S O 4 2 m m s ie v e , 1:5 ratio ter Z n, N O 3 M arkus and
m e q /1 0 0 g, O .M . < 5 S teck el, 198 0
%
M e h lic h N o . 3 A c id to neutral 0 .2 N a c etic acid 5 m l dry sa m p le in 5 0 5 m in and f il P, K, Ca, M g, B , Cu, 1 .2 Jones, 19 9 0
u n iv er sa l m in eral s o ils 0 .2 5 N NH4NO3 m l, 1 : 2 0 ratio^ ter F e, M n, Z n
0 .0 1 5 NNH4F
0 .0 1 3 NHNO3
0 .0 0 1 M E D I A
AB-DTPA A lk a lin e , c a lc a re o u s, IMNH4HCO3 8 .5 m l dry sa m p le p a s 15 m in and P, K , N a , C u, Fe, M n, 7 .6 Jones, 199 0
m in eral s o ils, pH 0 .0 0 5 M D T P A s in g 2 m m s ie v e , in 2 0 filter Zn W orkm an et
> 7 .5 ml a l.,1 9 8 8
NaHC0 3 -DTPA P o ttin g m ed ia , pea t- 0 .5 M N a H C O 3 ? ? N , K , P, C u, C a ? A lt et
c la y m ix e s 0 .0 0 5 M D T P A a l.,1 9 8 8
Olsen bicarbonate A lk a lin e , m in eral 0 .0 5 N a H C 0 3 D ry sam p le, p a s sin g 2 3 0 m in and P 8.5 B ates, 1990
s o ils m m s ie v e , 1 :2 0 ratio filter
Bray-Kurtz PI A c id , m in eral s o ils 0 .0 2 5 N H C l D ry sam p le, p a s sin g 2 5 m in and fil P, K , Ca, M g, N O 3 1 .2 B ates, 19 9 0
0 .0 3 N NH4F m m s ie v e , l.TO ratio ter. 1 m in ’" M arkus, 198 0
M arkus and
S tec k e l, 1 9 8 0
Bray-Kurtz P2 A c id , m in eral s o ils 0.1 N H C l D ry sam p le, p a s sin g 2 5 m in and f il P, K , C a, M g 1 .2 B a tes, 1990
0 .0 3 N H 4 F m m s ie v e , 1 :1 0 ratio ter
DTPA-SME P o ttin g m ix e s 0 .0 0 5 M D1 Saturate m e d iu m after M ix w h ile pH , E C , P, K , Ca, M g, B er g h a g e et
ml s o if a d d in g D T P A saturating, N O 3 , F e, M n, B , C u, a l.,1 9 8 7
eq u ilib rate Zn
1.5 hr
pH of eX“ Reference
Name Conditions and Extractant chemicals Dilution Shaking time Nutrients tested tracted
media solution
S p u rw ay a cid e x tra ctio n S o ils, p e a t-b a sed 0 .0 1 8 N a c e tic acid 2 0 m l m o ist sam p le, 1 :5 1 m in and f il P, K , Ca, M g, N O 3 3.3 M arkus, 1 9 8 6
p o ttin g m ix e s ratio ter
D u tc h 1:5 S o ils and s o il-b a s e d W ater, sa m p le brough t to 63 1:5 ratio by w e ig h t 15 m in EC , C l, N , P O 4 , K, M g van den
p o ttin g m ix e s cm w ater m atric su c tio n (1:25)^^ E nde, 1971
D u tc h 1:2 S o ils, p o ttin g m ix e s. W ater S am p le brough t to fie ld E C , P O 4 , Cl, N , K, M g — S o n n e v eld

p eat- and so il-b a se d c ap acity, 1 : 2 ratio and van den
E nde, 1971
D u tc h 1 :1.5 P o ttin g m ix e s, peat. W ater, s o il brough t to 3 2 cm 100 m l so il in 150 m l 15 m in , filter EC, PO 4, N O 3, N H 4, N , — S o n n e v e ld et
bark, c la y -sa n d , etc. w ater m atric su c tio n w ater, 1:1.5 ratio K , C l, M g a l.,1 9 7 4
SM E P o ttin g m ix e s, p eat- W ater, 5 0 0 m l m ed iu m E q u ilib rate pH , E C , N O 3 , N H 4 , P, W arncke,
(saturated m e d iu m b ased , lo a m le ss brough t to saturation 1.5 hr K, Ca, M g, Fe, M n, 1986
extract) (O vernight)^ C u, Zn
P ou r-th rou gh P o ttin g m ix e s, n o W ater, 5 0 m l c o lle c te d “““ pH , E C , N O 3 , N H 4 , P, W right, 1986

sa m p le s rem oved , K , Ca, M g, Fe, M n,
leach ate from C u, Zn
co n tain ers
A D A S (U .K . a d v iso r y P o ttin g m ix e s, lo a m - Saturated C a S 0 4 so lu tio n for D ry sam p le, p a ssin g 2 7 P, E C , N O 3 , K, M g Joh n son ,
s e r v ic e ) le ss m ed ia EC and N O 3 determ ination; m m s ie v e , 2 0 m l so il, 19 8 0
N a H C 0 3 for P 1 : 6 ratio
L e v in g to n (U .K .) L o a m le ss m ix e s W ater 1 :6 d ilu tio n 6 0 m in and pH , EC , P, K, N O 3 , M g — Joh n son ,

filtered 19 8 0
2 mM DTPA P o ttin g m ix e s 0 .0 0 2 M D T P A S am p le m atric su c tio n 1.5 hr, pH , EC , N , K, P, Ca, H andreck,
brough t to 1 0 cm water. interm ittant M g, F e, C u, B , S, M n, 1991
2 5 m l D T P A to 100 m l Zn
sam p le
T race e le m e n t T ro p ica l m in eral 0 .2 5 N N a H C 0 3 D ry sam p le, to p ass 2 1 0 m in and P, M n, Z n, Cu L in d sa y and
s o ils O .O I M E D T A m m s ie v e , 1 : 1 0 ratio filter C o x , 1985
5 0 pp m S u p e rflo e 127^
0 .0 1 N N H 4 F
H o t w ater M ineral s o ils H o t w ater D ry sa m p le to pass 2 5 m in and fil B — L in d sa y and
m m s ie v e , 1 : 2 ratio ter C o x , 1985
(NH4),C,04 M ineral s o ils A jn m o n iu m ox a la te D ry sam p le to pass 2 1 0 hr and Mo 3.3 L in d sa y and
m m s ie v e , 1 : 1 0 ratio filter C o x , 1985 1^
-oo
LO
4^
00
pH of ex Reference 4^
Name Conditions and Extractant chemicals Dilution Shaking time Nutrients tested tracted 0
media solution ir
p
DTPA-TEA Mineral soils 0.005 M DTPA Dry sample 1 mm stain 2 hr and filter Zn, Fe, Mn, Cu 7.3 Lindsay and
0.01 MCaCl2 less steel sieve, 1:2 ratio Norvell, 1978 o^
0.1 MTEAs (w/v)
Boron Mineral soils 0.1 MCaH4(P04)2 Dry sample, 3 mm sieve, 10 min and B Lindsay and
1:5 ratio filter Cox, 1985 S*
0
NaCl Potting mixes 0.5 N NaCl 20 ml moist sample, 1:5 30 min and P, K, Ca, Mg, NO3, 5.9 Markus and
ratio filter NH4 Steckel, 1980
^ Use only plastic containers with this mixture. ^ Mixing time used by Markus, 1986.
LiCl2 required at 1000 ppm if emission spectroscopy used for analysis. P determination on dry sample, shaken 15 min.
See Markus and Steckel, 1980. ^ Superfloe is an organic flocculating agent manufactured by American Cyanamid.
^ TEA is triethanolamine (HOCH2CH2)3N.
485
in dusty locations. Obviously, galvanized containers cannot Table 6-32. Standard values for EC and nutrient levels
be used if testing for zinc levels is expected. Empty coffee in extracts obtained by the 1:1.5 and SME methods
cans can be a major zinc source. Samples to be tested for [Adapted from Wamcke, 1990] (With permission of the
A m er. Soc. o f A g ro n o m y).
boron should not be placed on Kraft paper since such paper
can be a B contaminant. The best preventive for Optimum levels
contamination is use of clean tools, clean plastic buckets, Analytical value 1:1.5 SME
clean plastic bags, and use of containers supplied by the
Conductivity* L3-1.8 2.0-3.5
testing laboratories. Tube corers should be stainless steel.
NO 3' 3.7-5.4 7.1-13.2
2. Analytical Procedures P 0.48-0.68 0.23-0.42
K 1 .5-2.1 4.0-6.0
There is good reason for Table 6-31. Although a Ca — 2.5-5.0
grower has no need to know detailed laboratory procedures, Mg 0.65-0.90 1.5-3.0
he must appreciate the fact that switching willy-nilly from Na — <3.0
one laboratory to another is likely to cause chaos. Nine of
Cl <3.3 <2.5
the procedures listed in Table 6-31, such as the Morgan,
Mehlich, Bray-Kurtz, and AB-DTPA were devised to test Ûnits = dS m'f
^ Units = mmol
field soils. It is well to remember that most mineral field
soil solutions have low concentrations, and the methods
allow removal of labile nutrients from the exchange
complex that replenish the soil solution. Thus, the tests also Table 6-33. Estimated optimum P values based
suggest the soil's capacity to provide nutrients over the upon cyclamen dry weight for three extracting
methods and three media types [Adapted from
growing season. Most of these procedures deal with acid,
Prasad et al.,1983].
mineral soils common to the eastern U.S., whereas the AB-
DTPA was devised for alkaline, calcareous mineral soils with Test method Substrate Optimum level*
pH above 7.0. If the grower producing tomatoes in the 1:1.5 Peat 0.26-0.28
southwest U.S. sends his samples to a laboratory using a SME Peat 0.79-0.86
Mehlich method in New York, the results are not likely to be
Spurway Peat 3.2-3.S
relevant. Interpretation by someone unfamiliar with local
conditions is circumspect.
Lindsay and Cox [1985] surveyed procedures used in 1:1.5 Bark 0.12-0.13
tropical regions for micronutrients, finding a wide range in SME Bark 0.45-0.48
practice. Some of their conclusions were that no particular Spurway Bark 1.4-L6
extractant has been superior under all conditions, although
several types can be employed successfully. The use of dilute 1:1.5 Peat + soil 0.045-0.061
acids is restricted to acid soils. Although chelates have been SME Peat + soil 0.074-0.119
used mostly on alkaline soils, studies with DTP A suggest it
Spurway Peat + soil 0.9-1.2
can be used on acid soils as well. Micronutrient testing is
Units = mmols P'
particularly difficult as the result of very low concentrations.
The preparation procedures, such as sample drying, grinding
force and time, sample quantity, shaking type (oscillation
versus rotary), and time, all influence the result [Martens and
Lindsay, 1990]. So-called “universal” extracting solutions may be misnamed. It can be appreciated that slight
variations from one laboratory to another can change the result in so-called duplicate samples. Only in
cooperative tests, such as that reported by Wamcke [1986] for the SME method, are there likely to be close
values from several laboratories.
The higher fertilization rates found in greenhouse practice are likely to result in reports of excessive
fertilization when substrate samples are subjected to extraction by procedures suitable for field soils. The
laboratory may fmd it necessary to massively dilute extracts to bring them into the range of their standards. The
Spurway system (Table 6-31), using a weak acid solution with minimum shaking, has enjoyed a long period in
the U.S. as the principal system for greenhouse substrates. Many laboratories still use it since there is a large
background of experience with the system. Because many substrates nowadays are loamless, with low buffering
capacity, more laboratories, especially European, are using pure water extraction on moist samples. Even though
peat, composted bark, and vermiculite have nutrient-holding abilities, nutrients are held less tightly compared to
mineral soils [Wamcke, 1990]. With fertigation, the lack of a capacity factor in greenhouse media is less
important. The bulk density of container substrates can be variable, and CEC of low-density materials with
peatmoss, bark, etc. can appear high when expressed on a weight basis. When expressed on a volume basis,
however, CEC values are similar or less than mineral soils. Samples measured on a weight basis require
conversion to a volume basis or separate interpretation guidelines.
in the U.S., several laboratories have switched to the saturated media extract (SME) system. This method
was first developed by the U.S. Salinity Laboratory [Richards, 1954] for testing total soluble salts. Wamcke
[ 1990] states that the method is useful in weakly buffered growth media. The saturated sample is vacuum filtered
and all subsequent analyses are performed on the filtrate. Wamcke acknowledges that results with SME have
been more variable, much of it associated with the difficulty of accurately mixing the medium to the saturation
point. With field soils, determining the end-point is relatively easy: the soil begins to flow, the surface glistens,
and the soil slides cleanly off a spatula. As the amounts of coarse peat, bark, plastic beads, etc. increase,
determining the actual endpoint becomes more difficult. While the SME is more laborious, pH and EC can also
be determined from the sample. Other procedures such as the displaced solution method described by Nelson and
Faber [1986] and the pour-through [Wamcke, 1986] have been described. The detail required for the former
procedure precludes its use in routine diagnostic analyses [Wamcke, 1990].
In Europe, saturated media extracts and displaced solutions are used in research, but water extracts are
employed by service facilities. The Dutch reduce variability by adjusting the moisture tension in the sample to
about 32 cm water column before extracting the sample solution at a 1:1.5 ratio. The Levington and ADAS
procedures, reported by Johnson [1980], use a 1:6 ratio, which apparently overcomes variability due to failure
to account for initial moisture content. Standard test values for the Dutch 1:1.5 and SME are compared in Table
6-32. Table 6-33 compares optimum P levels determined as a function of testing method and substrate
composition. As noted, P levels for optimum plant growth vary markedly with root medium composition and test
procedure.
For micronutrients, several specific procedures have been included in Table 6-31. Lindsay and Cox's [1985]
survey of tropical soil testing showed wide variation. These authors concluded that there is no way to tell from
their compiled data the reliability of the various procedures for detecting iron deficiencies in tropical soils. The
critical levels of manganese varied from 1 to 28 ppm; zinc, 0.5 to 10.0 ppm; copper, 0.2 to 10.0 ppm; boron, 0.2
to 2 . 0 ppm; and molybdenum,
0.1 to 0.3 ppm. They also stated
that although a particular soil
test is used by several
laboratories, it is no guarantee
that the test is effective.
3. Interpretation of Soil
Analyses
Dahnke and Olson [1990]
separate understanding soil
testing into three steps: 1 )
correlation, 2) calibration, and 3)
interpretation. Correlation is the
process to determine if an Critical level
extracted nutrient and crop X J __ L JL X
response to the added nutrient 20 40 60 80 100 120 140 160
are so well related that one Soil Analysis - ppm P
directly implies the other. There
are several means of correlation Fig. 6-54. A Cate-Nelson scatter diagram of percentage yield versus soil test P for
and calibration, one of which is maize [From Cate and Nelson, 1971] (With permission of the A m er. Soc. o f
the graphical Cate-Nelson A gronom y).
487
diagram for P presented in Fig. 6-54. The diagram plots percentage yield against the soil test to give visual
indication of soil test reliability. The division between negative and positive quadrants is divided to maximize
the number of points in the positive quadrants and minimize those in the negative areas. Many points in the neg
ative quadrants show the soil test is not well suited to the soils of the area, or there is no correlation between soil
test values and plant response to the added nutrient. Whether Cate-Nelson diagrams can be applied to greenhouse
conditions has not been determined to my knowledge. Calibration finds out the meaning of the soil test in terms
of crop response. This allows soils to be placed into response categories such as very low, low, medium, high,
and very high concentration ranges [Dahnke and Olson, 1990]. The Cate-Nelson procedure (Fig. 6-54) has at
least two advantages over fitting continuous curves to the data. It shows whether there is a good correlation and
separates the data into populations likely to respond to added nutrient and those unlikely to respond to added
nutrient. The point at which this occurs is the critical level. Laboratories in the U.K., using the ADAS or
Levington procedures [Johnson, 1980], assign a numerical value to a range of test values to provide an index.
The recommendations are related to the species, growth period, cultural procedures, etc. by assigning an index
number that specifies the desirable nutrient test value for each nutrient. Those assigned to the Levington method
range fi'om 0 to 9. For tomatoes in modules, Johnson [1980] reported an index of 7 for phosphorous at the start
of the crop, dropping to 5 later in the season. These indices corresponded to P levels of 56-75 and 29-40 mg
respectively.
With the background of correlation and calibration, and having selected methods that will best serve the
conditions of substrate, species, and cultural procedures, etc., the final step is interpreting and making
recommendations. The individual making the recommendations should be thoroughly familiar with the crops,
substrates, and cultural conditions of the region. The person making the recommendations must be well-versed
in general principles and soil chemistry. The values listed in Table 6-34 are mostly optima, although some are
considered as standards or critical levels below which deficiency occurs, or, with NH 4 or Cl, if a potential for
toxicity exists. Usually, where constant fertigation is practiced, my experience suggests that soil test values can
be lower than those given in Tables 6-32 through 6-34. pH is another parameter that, it appears, may be given
more emphasis in the wrong direction than is needful. Black [1992] makes the point that acidity is not the cause
of poor growth on acid soils. Furthermore, comments have been made in previous discussion that the hydrogen
ion concentration can vary quite widely without direct effects on growth. The secondary effects of
concentration are important: such as the presence of aluminum in acid media, excessive bicarbonate in
hydroponics, micronutrient availability, in particular iron availability in well-aerated media with high redox
potentials, and the effects and OH' may have on microorganisms. Salinity will be discussed in the following
section.
In addition to absolute concentrations, nutrient balance can be important. Wamcke [1990] cited Geraldson
as showing that a good nutrient balance as a percentage of total soluble salts to be: NO 3 , 8 to 10; NH 4 <3; K, 11
to 13; Ca, 14 to 16; Mg, 4 to 6 ; Na, < 10; and Cl, <10%. Expressing the nutrient levels as a percentage of total
salts helps assess the most limiting nutrient. The values in Table 6-34, with some exceptions, are difficult to use
in this fashion. The interpreter can examine: 1) absolute concentrations resulting from the test, 2) crop yields
expressed as a percentage of maximum compared with the particular test level, and 3) total soluble salts with test
levels expressed as a percentage of that total.
It has been my observation that growers, particularly cut-flower producers, often fail to make full use of soil
test analyses. For example. Table 6-35 provides an abbreviated example of results obtained bench-by-bench on
roses produced in the ground in a Central American country. Many soil analyses are presented in this fashion,
and the grower looks them over, filing them in some remote drawer. This practice is unfortunate. Given the
spreadsheets presently available, with graphic capabilities, considerable information can be obtained regardless
of whether the analytical procedure is known or understood. Simply looking at such a mass of data cannot
provide adequate answers. However, a spreadsheet program allows one to compute the average of all values -for
a house, or for a section, where all benches are treated the same- and to calculate the standard deviation of the
mean. This at least provides an idea of the average soil test value, and by multiplying the standard deviation by
two, a range can be calculated that allows outliers to be identified -e.g., 306 mg P, bench 4; 16 pM f^ Ca,
bench 1; etc. Since these samples probably came from one section of an 8 ha range, the outlying data suggest
problems in collecting the soil sample.
More important in using soil analysis is the development of a history. If soil analyses are carried out on a
oo
oo
Table 6-34. Recommended soil test values from various sources for different soil test procedures. The majority of tests must be interpreted by specialists familiar with the O
analytical and cultural procedures as well as the medium and species requirements. Units are ppm, mg f *or mg kg~^ except where noted. ______ __ ____________ _ p
Mo Cu Reference
"S
a
Method NO, NH. N K Ca Mg Na Cl Fe Mn B Zn
Dutch 1:1.5 3.7-5.4^ 15-20 1.5-2.D 1.3-1.8^ 2.3-3.3^ Bik and Boertie,
1975
c
ADAS*^ 51-80 101-150 12-18 101-175 16-25 Bunt, 1988 s,
Levington (Index = 4) 81-130 19-28 176-250 26-35 Johnson, 1980 o'
cs
Spurway^ 10-20 2 5 20 60-120 Nelson, 1978
25-100 4-6 30-50 >100 Mastalerz, 1977
20-80 4-5 25-35 150-200 Hanan et ah, 1978
Modified Morgan 76-125 126-250 0.76- 8.1-10^ 1.3-2.5^ Mastalerz, 1977
1.0"
100-400 20-45 2.0-4.0^ 7.5-40^^ 4.0- Mastalerz, 1977^
10.0^
AB-DTPA^ 8-11 61-120 >5.0 >0.5 >0.2 >1.50 Soltanpour and
Follet, 1985
SME 100-199 6-9 60-149 >200 >70 Warncke, 1980
8-10 <3 11-13 14-16 4-6 <10 <10 Warncke, 1984
15 16 0.7 9 14 Berghage et
al.,1987
Pour-through 75-100 10-15 30-50 10-15 10-15 Wright, 1986
Mehlich No. 3 3.0^ 0.3*^ Martens and Lind
say, 1990
DTPA-TEA 4.5 0.22 0.5-0.8 Martens and Lind
say, 1990
Trace elemenf 3 pg/ml 5 1 2 Lindsay and Cox,
pg/ml pg/ml pg/ml 1985
Boron 0.2 Lindsay and Cox,
pg/ml 1985
(NH4)2CA 0.2 Lindsay and Cox,
1985
Hot water >0.3- Lindsay and Cox,
0.5 1985
Method NO 3 NH 4 N K Ca Mg Na Cl Fe Mn Mo Cu Zn Reference
Units in milliequivalents per d.
Index rating = 3, lower for seedlings, higher for tomatoes in peat modules.
Medium levels of nutrients.
Units in milliequivalents per 100 grams.
Upper values for mixtures containing soils, lower for soilless mixes.
Values for irrigated production in the field.
Units in mg per dml
Value is pH variable, number given is for pH 6.4.
Critical levels for micronutrients.
4^
00
Table 6-35. An example of a bench-by-bench soil analysis on roses grown in the ground. The
average of each value, standard deviation (s^), and range (2*s^) were calculated.^
Bench No. pH P ( mg/Q Ca Mg K Fe Cu Zn Mn

5.5 110 16 0.9 0.8 69 9 11
2 5.5 150 13 0.6 0.6 63 1 8
3 5.2 265 6 0.3 0.4 90 6 10
4 4.8 306 4 0.3 0.5 68 7 15
5 5.3 172 5 0.3 0.3 40 6 20
6 5.4 210 4 0.5 0.4 49 4 30
7 5.2 270 5 0.4 0.4 84 0.2 26
8 5.6 140 6 0.4 0.5 207 1 32
9 5.7 153 4 0.8 0.8 74 1 13
10 5.4 87 10 0.7 0.6 134 9 14
11 5.4 150 7 0.5 0.5 104 2 8
2 5.5 135 8 0.5 0.5 81 5 10
13 5.5 190 6 0.3 0.5 77 3 6
14 5.7 190 8 0.4 0.5 65 7
15 5.7 123 5 0.3 0.4 78 9
Average 5.4 177 0.5 0.5 68 4 3 15

0.2 63 0.2 0.1 40 3 2 9
Range*" 5.0-5.9 51-302 0.2-14 0.1- 0.2- 0.? 5-166 0-10 0. 1-6 0-32
0.9
" With exception of P and pH, all values in pM d'\
*"Range calculated by multiplying the standard deviation (s^) by two and adding to
the average.
regular basis (i.e., monthly), than spreadsheet programs (Excel, Lotus 1-2-3) with graphics can be used to develop
lines showing actual values. These programs can use a TREND function that predicts the value at some future
time from previous results. Obviously, several values are required to establish a trend. The grower can then
predict the future concentration level, and make adjustments in his feeding program to lower the rate at which
a particular value is increasing or decreasing. Attempts to halt, or reverse, a trend are dangerous unless control
has been lost in the feeding program. With fertigation, adjustments can be made incrementally, and this provides
greater safety to avoid excesses and high salinity. Once a history has been developed, and is used, the grower can
manage his substrate fertility level for maximum profitability.
G. SALINITY AND IRRIGATION W ATER QUALITY

To this moment, nutrition has been discussed without reference to irrigation water quality and salinity
problems. For one to neglect these two factors in nutrition control can result in serious difficulty. Mistakes made
in locating appropriate water sources are difficult to correct -impossible sometimes. Regardless of how good
a grower is, or how modem and extensive growth facilities may be, a poor water source makes salinity control
difficult, reduces options in nutrient control, and limits yield and quality. A cheap raw material, which growers
often ignore, becomes a highly expensive proposition.
I. Salinity
The sum of all cations and anions in the substrate solution is the total soluble salts, which may be expressed
in terms of concentration or equivalency. While a complete solution analysis is possible in order to arrive at a
491
figure, measuring the electrical

conductivity of a suitable extract
from the medium is easier and
quicker (Section 6 .II.H). This
determination is invariably carried
out in soil analyses, and it can be
made a continuous measurement of
flowing solutions such as NFT.
The units were given in Table 6 -
11. The EC of solutions will be
employed throughout as a measure
of salinity of a substrate or
hydroponic solution.
a. E ffect o f S a lin ity on G ro w th

Total salts directly affect
osmotic potential as was discussed
in Chapter 5 (Section 5.II.A). Fig.
5- 14 showed typical examples of
excessive soluble salts on com
4.0 5.0 6.0 7.0 8.0
mercial greenhouse plants. Besides
EC of soil (dS/m)
the osmotic effect, specific ions
may be high enough in
Fig. 6-55. The effect of total soluble salts (EC) on rose growth in the ground . The
concentration to cause toxicity and
ordinate is the number of flowers cut per treatment [Reprinted from Sci. H ort.,
interference in nutrient uptake and 29:373-385, Femandez-Falcón, M. et al., ©1986. With kind permission of Elsevier
balance. Typical of the growth Sci.-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands].
response demonstrated several
times for greenhouse crops is Fig.
6- 55, showing a continual yield reduction of roses as EC of the soil increases. Such relationships are often
curvilinear, and the fact that plants can adjust to increasing salt concentration changes the relationship so ex
pressed. Such responses have been examined by Sonneveld and Mook [1983], Sonneveld and Voogt [1978], Son-
neveld and van Beusekom [1974], Maagistad et al. [1943], Hughes [1975], Hughes and Hanan [1977], Schekel
[1971], as well as several others. An EC of 4 dS m‘^ as shown in Fig. 6-55, corresponds to a total salt concentra
tion of about 50 meq whereas an EC of 8.0 is approximately 90 meq (Fig. 6-7).
Effects of high concentrations can be subtle or obvious as shown in Fig. 6-56, depicting a change in color
for gerbera and obvious damage on chrysanthemum. The work by Sonneveld and others at Naaldwijk, The
Netherlands, showed cucumbers to be highly sensitive, with yields decreasing 4,7, and 14% for lettuce, tomatoes,
and cucumbers, respectively, if the EC of the irrigation water increased 1 dS m '\ An increase of 1 dS m'^ in the
irrigation water caused an increase of 2 dS m’’ in the soil. Cucumbers showed a special sensitivity to excess
calcium and magnesium. Most detrimental was the application of sodium bicarbonate. The deleterious effects
of bicarbonate have been noted on several greenhouse crops (Fig. 6-11). Spinach, on the other hand, was not
affected by the treatments imposed [Sonneveld and van Beusekom, 1974]. Sonneveld and Mook [1983] showed
that the incidence of tipbum in lettuce could be attributed to low calcium content versus a high magnesium level.
High sodium also promotes tipbum. Thus, some crop responses may be due more to a nutrient imbalance than
to high salts.
Common is a salt effect noted in Fig. 6-57, dealing with roses produced in gravel. Treatment Ts applied
solution contained the nutrients given in Table 6-22 at an EC of 1.3 dS m'^ versus Treatment 2 with 8 meq of
HCO3 ' added and an EC of 2.5 dS m \ There were not, by observation, great differences in visuaj appearances
between treatments. But, the production showed that Treatment 2 reduced peak flowering periods and delayed
them to the extent that by early summer peak flower cycles no longer occurred. Thus, excessive and imbalanced
salt levels have the practical influence of upsetting timing cycles made by the grower -such as pinching- not
Fig. 6-56. Upper; Lighter color in gerbera flowers as the result of high total soluble salts.
Lower: Fertilizer salt damage to young chrysanthemums.
to mention quality reductions as smaller flowers, or fruit, and shorter stems.

Also common are the effects of internal cell water potentials as influenced by the external solution
concentrations. Brouwer’s [1963] results with beans are typical regarding transpiration rate and leaf growth as
the external concentration increases (Fig. 6-58). As noted in Chapter 5, the water potential of the cell must be
lower than the external potential if water is to move into a cell, thereby maintaining sufficient turgor to prevent
wilting. An osmotic water potential of 1.0 MPa corresponds approximately to nearly 40 dS m'* EC. A potential
of 0.3 MPa is about 8 dS m '” and 0.1 MPa to slightly less than 3 dS m'^ [Richards, 1954]. Similar studies have
been carried out by many investigators [i.e., Hayward and Spurr, 1944; Nieman and Poulsen, 1967; Bernstein,
1961, 1963; Meiri and Poljakoff-Mayber, 1967; Slatyer, 1961; Wadleigh and Ayers, 1945; etc.]. Bernstein and
Hayward [1958] summarized the situation by categorically stating that the effects of excessive salt concentrations
493
Fig. 6-57. Effect of fertigation

treatments on roses in gravel. See Table
6-22 for solution employed in
Treatment No. 1. Treatment 2 included
8 meq f * HCOs' and an EC of 2.5 dS
m'^ versus 1.3 for the former. Note the
peak flowering delay of about 2 weeks
in No. 2 for the first cycle even through
both treatments were pinched for
cycling in the late fall at the same time
[Hughes and Hanan, 1977].
are mediated by osmotic inhibition,

by specific effects of the constituent
ion, or by a combination of the two.
A particular problem with
substrates irrigated at intervals is
the increase in solution con
centration as the medium dries out.
This is illustrated in Fig. 6-59 for
selected ions in the soil. In the
range of soil matric potentials for
most greenhouse substrates, the periodic exposure of root systems to extreme ion concentrations can be frequent.
The higher the solution concentration, the more likely acute damage is to occur if the grower allows the medium
to dry out. O f course, the ability to maintain high moisture content in ground soils is severely limited by aeration
limitations that restrict options available to the grower.
In soils, container substrates, etc., high salt concentrations may also have deleterious effects on soil structure
-such as replacement of calcium by sodium on the exchange complex. Most field soil tests also evaluate the
salinity hazard with the sodium absorption ratio -which is seldom seen in greenhouse soil tests. For acid soils,
the lime requirement is commonly calculated for the grower. In hydroponics, the effects of high concentrations
on substrate relationships mentioned above can be ignored. Although high water quality has been emphasized
for recirculating culture systems [Anon., 1978b], there have been occasions when the deliberate addition of a salt
such as sodium chloride has been found beneficial [Adams, 1987]. Early vigor in tomatoes was controlled by the
addition to the NFT solution of 1200 to 1550 mg f^ NaCl (ca. 20 to 27 meq with an EC of 7.5 to 9 dS m ^
Concentrations of 17 meq (1000 ppm Na) reduced yield, whereas concentrations of about 9 meq (4.6 dS
m'*) resulted in the highest yields. Increasing the sodium level improved fruit quality. It is well to remember the
very low solar radiation under U.K. winter conditions, and increasing total soluble salts in hydroponic systems
is likely to aid in maintenance of acceptable growth and fruit quality. Several years ago, sodium nitrate was often
used as a means to increase flower quality during the winter in Colorado. Previous mention has been made of
possibilities in crop improvement through the judicious use of such elements as sodium, selenium, etc. There
have been few, if any, studies of this kind for greenhouse production. There have been articles dealing with the
use of salt water in agricultural production such as the article by Boyko [1967]. Breeding crops for salt tolerance
has been a major research thrust [e.g., Saranga et al., 1991; Jones, 1981]. Saranga et al. state that their results
suggested the existence of a genetic potential for high salt tolerance in wild tomato germplasm. In either case,
given the intensive culture found in greenhouse production, I do not feel that attempting production with salt
water, or introducing salt-tolerant species, is economically viable at this time.
b. Ranges o f Total Soluble Salts

Recommendations encourage the use of the water-saturated paste as the means to obtain a value for total
soluble salts -o r EC. However, several laboratories also use 1:2 or 1:5 dilutions to provide a sample for electrical
conductivity determination. The latter are convenient for rapid determinations, but reliability depends upon the
salts present. For chloride salts, results are only slightly affected by moisture content. However, if sulfate or
carbonate salts are present, the apparent amount will depend upon the soil:water ratio. With container mixtures
rate of le af growth
mm/day
o
1 0
o
^ \o
\
Oy
\
\
\
\
o \
©\ ©
\ O
\
\
\
\
o
__ ^
L L -L o—
-0.8 -0.9 -1.0 -1.1 -1.2 -13 -1.4 (MPa)
Water potential of leaves
rate of troniplratlon
Water potential in leaves»
Fig. 6-58. Effect of leaf water potential in beans on leaf growth rate (upper) and
transpiration rate (lower) [Brouwer, 1963] (With permission of A c ta Bot. N e e d ., Blackwell
Sci. LTD).
ising peat and soilless additives, EC readings will be higher at any given moisture content than for mineral clay
oams. Sandy loams will give the lowest EC readings [Wamcke, 1990]. The U.S. Salinity Laboratory [Richards,
[954] did not recommend such dilutions for arid region, calcareous soils. Waters et al. [1970], in an examination
)f 27 soil media, found that excessive soil moisture in light weight media reduced salinity readings with the
lilution method. In the saturated paste method, initial moisture content, sample volume, bulk density, container
495
Fig. 6-59. The change in

cation concentration of
soil solution with a
change in moisture
content as indicated by
the water:soil ratio [From
P la n t a n d S o il, 28:99-
113, Moss, P., ©1963.
With kind permission of
Kluwer Acad. Publ.l
Table 6-36. Soluble salt guidelines for growth media by three test methods [Wamcke, 1990].
Units are dS m'K (With permission of the A m er. Soc. o f A g ro n o m y).
Saturated One part medium to:

media (v/v) Interpretation
extract Two parts Five parts
water water
0.0-0.74 0.0-0.24 0.0-0.12 Low nutrient status.
0.75-1.99 0.25-0.49 0.13-0.34 Suitable for seedlings and sensitive plants.
2.00-3.49 0.50-0.99 0.35-0.64 Suitable for most established plants. Upper range
may reduce growth of some salt-sensitive plants
and seedlings.
3.50-5.00 1.00-1.49 0.65-0.89 Higher than desirable. Loss of vigor in upper
range. Okay for high nutrient-requiring plants.
5.00-6.00 1.50-1.99 0.90-1.10 Reduced growth. Wilting and marginal leaf burn.
6.00+ 2.00+ 1.10+ Severe symptoms, wilting, crop failure.
capacity, or media composition are not critical factors. Distilled or deionized water is added to a soil sample while
stirring until the conditions noted earlier are obtained. The sample should be allowed to stand an hour or more
and criteria for saturation checked. Free water should not collect on the surface, nor should the paste stiffen or
lose its glistening appearance on standing. Richards [1954] suggested a 250 g sample, although the previous
section said that 500 ml may be used in the SME analytical procedure. Volumes are better to use for lightweight
mixtures. The sample is vacuum filtered. If gypsum is present, the saturated paste should stand for several hours
before filtering. Table 6-31 shows overnight standing for some laboratories.
Thus, one finds such tables as 6-36 showing EC values using the different extraction methods and interpre
tation. Table 6-37 is another example. Optimum EC levels for different analytical extraction methods were given
in Table 6-32. Variations can be found in the ranges given and methods by which reported. Table 6-11 shows
a plethora of units, with the deciSiemens per meter (dS m'^) being recommended in recent years [e.g.. Miller et
al, 1981; Anon., 1985]. The approximate relationships between EC and concentrations in milliequivalents per
H (meq f^) were presented in Figs. 6-7 and 6 -8 . The relationship between osmotic water potential in MegaPascals
and EC can be estimated by:
Table 6-37. Interpretation of EC values obtained with a 1:2 soil-water extract [Jones, 1985] (With
permission of H ort. R ev.).
Specific conductance
at 25 C (dS m-^) Interpretation
<0.40 Salinity effects negligible except in beans and carrots.
0.40-0.80 Slightly saline. Yields of salt-sensitive crops such as pepper, lettuce, etc.
may be reduced 25-50%.
0.81-1.20 Moderately saline. Yield of salt-sensitive crops restricted. Seedlings
injured. Satisfactory for well-drained greenhouse soils.
1.21-1.60 Saline soils. For tolerant crops. Higher than desired for greenhouse soils.
1.61-3.20 Highly saline. Only salt tolerant crops. Leach greenhouse crops with 20 to
40 i s per m^ water or 0.5 i s per 15 cm pot.
>3.2 Only salt-tolerant species. Very saline.
0.36 x £ C x 0.101 (6.50)
where 0.101 is the conversion from atmospheres to MPa and EC is given in dS m‘'
c. Controlling Salinity in Substrates

If, through excessive fertilizer application or Table 6-38. Leaching requiremenf as related to the ECs <
failure to water adequately, the grower finds irrigation and drainage waters [Richards, 1954].
himself faced with high salts, about the only
Leaching requirement for the indicated
option is to wash excess salts from the soil layer maximum values of the EC of the drainage
or active root zone. Application of excess water, EC of irriga water at the bottom of the root zone
which itself contains high salts, will not reduce tion waters 4 dS m ' 8 dS m ' 12 dS m ' 16 dS m *
total soluble salts below that contained in the (dS m-')
water supply. As will be discussed later, water Percent Percent Percent Percent
quality becomes highly important in greenhouse
0.1 2.5 1.2 0.8 0.6
culture.
0.25 6.2 3.1 2.1 1.6
Saline soils are those in which the
conductivity of the saturation extract is greater 0.75 18.8 9.4 6.2 4.7
than 4 dS m’^ [Richards, 1954]. The previous 2.25 56.2 28.1 18.8 14.1
sections show that EC, under many conditions 5.0 62.5 41.7 31.2
found in greenhouses, may be much higher. Leaching requirement is the fraction of applied irrigation
Without salt accumulation from ground water, water that must be leached through the root zone ex-
the EC ranges from 2 to ten times as high as EC pressed as a percent.
of the applied water. This results from moisture
extraction by the root system and evaporation.
The leaching requirement for field soils as
suggested by the U.S. Salinity Laboratory is provided in Table 6-38. The EC of the applied water obviously has
a strong influence on water consumption where leaching is required.
Under ideal conditions, when leaching, it is desirable that the applied water not mix with the soil solution,
but to “push” the excess salts through the layer with no mixing. This is called “piston” flow since incoming water
from the surface acts as an impervious plunger. Even with highly uniform substrates and low infiltration rates,
some mixing always occurs. As the infiltration or percolation rate increases, part of the leaching water bypasses
small pores and capillaries, leaving some salts behind. Mixing between incoming and outgoing solutions can be
very high. This is the reason for the special displacement solution used by Nelson and Faber [1986] in obtaining
soil extracts. The situation for container mixes was studied by Kerr and Ffanan [1985] (Fig. 6-60), usually
497
Fig. 6-60. Leaching curves of

15 cm deep columns of
mixtures containing various
proportions of peatmoss (PM),
perlite (P), and glass beads
(GB). The surface was flooded
with water containing 4 meq p'
CaCl2 and 4 meq f* NaCl. The
columns were salinized prior
to leaching with 15 meq f*
each of CaCl2 and NaCl per ^
medium [Kerr and Hanan,
1985].
Fig. 6-61. EC of the drainage water

from the bottom of containers
containing poinsettias during the
growing period for different leaching
fractions (LF) [Ku and Hershey, 1991]
(With permission of the A m er. Soc. f o r
Hort. Sci.).
showing an initial increase in EC of

the drainage water as water was
applied to the sample surface, and
then decreasing, approaching a
minimum asymptotically as water
continued to flow through the
sample, regardless of the mixture.
The use of glass beads was an
attempt by the authors to obtain
greater reproducibility. Mixtures
with a greater proportion of glass
beads resulted in a higher bulk
density and greater salt removal rate
in initial stages, indicative of lower
mixing between soil solution and
the leaching water -o r closer to piston flow. Regardless of the leaching solution salinity, about the same amount
of salt was removed in initial stages, but the higher the EC of the water applied, the higher the final minimum
as the removal rates approached zero. Under laboratory conditions, the greatest amount of salt in a mixture was
49 8 Chapter 6, N utrition
removed when enough water equal to 1.0 to 1.5 container capacities had moved through the sample.
Some authors have examined leaching requirements with plants established [e.g., Ku and Hershey, 1991;
1992; Yelanich and Biembaum, 1993]. The approach of these authors has been to use the “leaching fraction,”
which is the volume of solution leached from the container divided by the total solution applied. That is, if 400
mi is the container capacity, an application of 400 ml with 100 ml of drainage, or 200 ml applied with 50 ml
leached, would both equal an LF of 0.25. Container capacities would vary since the drainage loss is different.
An example of drainage EC over time at different LFs is given in Fig. 6-61. The thrust of this work is to control
medium salinity with minimum wastage. Yelanich and Biembaum state that the recommended LF is 0.1, but most
growers use ratios greater than 0.5. Fig. 6-61 indicates an LF of 0.4 is required to maintain EC of the drainage
below 8 dS m'^ The latter authors reported LFs greater than 0.35 were required when the commonly recom
mended 14 mM f ' N was applied to poinsettias grown in 15 cm pots. A total of 1250 ml were required to achieve
an LF of 0.5 to 0.6. Adequate fertility levels were maintained using 7 mM N, or less than half the usually
recommended amount. If good water is available, drainage produced in poinsettia culture -and consequent
fertilizer required and lost- can be reduced significantly.
Unfortunately, water and fertilizer input in greenhouse production has been such a small proportion of total
production cost that many growers deliberately apply excess water to be safe. There is also the problem that many
container mixes are so porous that trickle irrigation systems may not wet the entire soil mass. This would be
particularly true during initial growth stages when root proliferation is least. As root density increases during
growth, better wetting from a single trickle emitter would be more likely. Flooding of the medium surface may
be necessary at times if efficient leaching is to occur. Another factor to be considered is excessive drying of the
substrate between irrigations. Upon watering, most of the water is likely to flow between the container sides and
soil mass. Though drainage is occurring, little moves through the soil mass until it has been thoroughly wetted.
In raised bench, cut-flower production, “dry” growers find it necessary to water twice over a several hour period
to ensure adequate leaching. Of course, such shrinkage may not occur with some substrates, but peat and bark
mixtures are very difficult to wet unless a wetting agent has been included in the mixture. Crops watered by sub
irrigation (i.e., capillary mat, ebb-and-flow, constant water table) may experience excessive salts on the medium
surface because of evaporation. This can be especially apparent if high fertilizers are included in the water.
Overhead irrigation may be necessary.
One cultural procedure that has apparently received insufficient attention is the need to determine container
capacity, and water commonly lost between irrigations -exclusive of any drainage from the container. To this
time, most growers always supply a surplus, which often exceeds container capacity several fold. As will be
emphasized later, some environmental agencies are placing severe restrictions on waste water and fertilizer loss
from greenhouse operations. An estimate of container capacity and water consumption for a given crop and
season would help answer these new requirements, plus increasing operational efficiency and reducing raw
material costs.
H. IRRIG A TIO N W ATER QUALITY AND M ANIPULATION

The presence of salts in the irrigation water will influence ability to manipulate nutrition. Newer systems for
automatic control can take into account variations that may occur in water supply quality. Secondly, water salinity
ultimately limits production capability and can markedly reduce profitability -even though the source is cheap
compared to labor or fuel.
1. Expectations in W ater Quality

Variations in water quality are to be expected, and water analyses are an important adjunct to greenhouse
operation. Typical of the analyses performed in Colorado are publications in the Colorado Greenhouse Growers'
Bulletin [Hanan et al.,1968; Hanan, 1973; 1979] and the discussions on methods for handling saline irrigation
supplies [Hanan, 1982a; b; 1988]. Fig. 6-62 graphically presents the concentrations of Mg, SO4 , Ca, and Cl
found in 36 shallow wells in the Denver region. Fig. 6-63 illustrates the common salt level found in such wells.
Average concentrations for Na and HCO3 were 4.7 and 5.0 meq t \ respectively, with 42% of the wells
containing sodium between 0 and 3.2 meq f ’ and 36% with bicarbonate between 4 and 6 meq fh Shallow wells
in Colorado - a semiarid region with alkaline soils- are likely to be well above the maximum desirable EC
allowable for reasonable nutrient manipulation (Fig. 6-63). Sulfate concentrations can be as high as 25.2 meq
499
Number Number
11
Average = 4.6
36 % 28 % 25 %
0 2.8 5.6 8.4 11,2 14.0

Mg (meq/1)
11
17
31 % 19 % 31 % Average = 4.1 H
Average = 1.1
47 % 31 %
14 %
14 %
---------- \----------.
0 2.2 4.4 6.6 8.8 11.0 0 1 4
Ca (meq/1) Cl (meq/1)
Fig. 6-62. Concentrations of magnesium, calcium, sulfate and chlorine in 36 shallow wells in the Denver, CO,
urban area. The vertical axes show the number of samples [Hanan, 1973].
and I have seen at least one tomato producer attempt culture of tomatoes with such irrigation waters. Calcium
and magnesium are likely to be significant in shallow wells (Fig. 6-62). Deep wells in Colorado often contain
high sodium and bicarbonate, although some are very high quality (i.e., no salts). Domestic supplies from high
mountain regions usually have ECs less than 0.5 dS m'^ Such examples can be found in almost any region of the
world. The Netherlands are plagued with high chlorides from the Rhine River. In the Mediterranean region
(Spain, Crete, Cyprus, etc.), salt water intrusion is often experienced in shallow wells close to coast lines. The
Sea of Galilee contains high chlorides.
The characteristics of irrigation water that are important in determining its quality are: 1) total soluble salt
concentration, 2 ) concentration of individual ions, 3) concentration of boron and other elements that may be
toxic, and 4) the bicarbonate concentration as related to concentrations of calcium and magnesium. For substrates
containing soil, sodium is particularly important in determining an alkali hazard. Sodium hazard of irrigation
water is usually expressed as the sodium absorption ratio (SAR), computed from the formula:
[Na*]
SAR = (6.51)
where concentrations are in meq Use of water with an SAR greater than lO must be avoided where soils are
part of the substrate. If the mixture contains appreciable amounts of gypsum, an SAR value of lO can be
exceeded. Continued use of high SAR water leads to a breakdown in the soil’s physical structure. Adsorbed
sodium results in clay dispersion, and the soil becomes hard and compact when dry, with slower water
penetration when wet [Follett and Soltanpour, 1985].
Fig. 6-63. Total soluble salts found in

36 shallow wells in the Denver, CO,
region. Vertical axis shows the number Number
of samples. The figures within the bars
12
are percentages of the total samples
found in each range [Hanan, 1973].
1 0
The U.S. Salinity Laboratory

[Richards, 1954] states that water Average = 1.169
with conductivity values below 0.75
dS is generally safe. However,
sensitive species are likely to be
affected at ECs above 0.25 dS 31% 33%
The upper limit of 0.75 was chosen
17%
by Hanan [1988] as the upper limit
11%
for greenhouse production. But,
Fig. 6-63 shows that more than 80%
of the supplies from shallow wells 0.44 0.88 132 1.76 2,64
have ECs ranging from 0.4 to 1.8 E.C. (dS/m)
dS m '\ and 50% of those are
between 0.9 to 1.8 dS m ^ Growers
using such water supplies can
expect to have difficulty in their nutrition programs to maintain reasonable productivity. Use of an irrigation
supply above 2.3 dS m '’ is likely to result in crop failure.
Fig. 6-64. Variation in total soluble salts

in water from a shallow well in the South
Platte River Valley, CO, over a year.
Samples were analyzed once monthly so
the curve is likely to conceal small
variations probably significant in
nutrition practice [Hanan, 1988].
A factor that is seldom

emphasized is variation over time in
water quality -especially from
shallow wells and surface irrigation
supplies. Total soluble salts and
individual ion concentration can vary
markedly with season. A series of
tests for a commercial range in
Colorado showed that water from a
well in the South Platte river basin
could vary from about 0.9 to over 1.4
dS m'^ (Fig. 6-63), reflecting changes
as the result of irrigation,
fluctuations in the water table, etc.
Similarly, N 0 3 ' in this supply could
fluctuate from less than 1 to more
than 2 meq Calcium and sodium showed even greater differences with season [Hanan, 1988]. Such variations
cry out for an automatic system of continuous water analysis so nutrient injection in a fertigation system can be
automatically varied as necessary. ECs approaching 1 dS m'^ preclude the use of fertigation in crop culture.
There are many laboratories to which one can send a water sample for analysis. Unfortunately, many of these
501
facilities are set up for bacteriological Table 6-39. Typical water analyses from four shallow wells with ECs
analyses for drinking water suitability, and ranging from 0.3 to 1.3 dS m‘f Analyses subjected to
the results on mineral content are likely to MINTEQA2/PRODEFA2 program to caleulate equilibrium at CO2 level
be given in terms of total carbonate, of 300 Pa.
hardness (grains), or softness. The grower Water sample number
is not interested in drinking the water. He Analysis for: 1 2 3 4
needs to know individual ionic content in
Electrical
terms that can be used in manipulating his
conductivity 0.3 0.5 0.9 1.3
nutrition program. It has been my (dS m-‘)
observation, in countries requiring
pH 7.7 7.7 7.7 7.6
mineralogical and bacteriological analyses
(meq f*) 0.9 4.0 3.8 5.1
on bottled water, that often such water
would make a very poor irrigation supply. 0 .2 1 .6 1.4 2 .0
Any water analysis should include, in parts r 0 .2 0.4
per million or milliequivalents per ^ , the Na^ 1.7 2 .0 4.0 5.9
concentrations of magnesium, calcium, NO 3- 0.5 0 .8
sodium, chloride, bicarbonate, carbonate, PO4'- 0.4 0 .1
sulfate, nitrate, and potassium. Where 8 0 4 '- 0.5 2 .8 3.2 5.2
likely, a test for boron should be included.
cr 0 .1 0.4 1.7 3.2
Presentation of results in meq f ^ allows
HCOs’ 2.7 4.4 3.6 4.1
one to check credibility of the analysis by
determining if the total cations equal the Solids precipitated None CaCOs CaC0 3 CaC0 3
total anions. The analysis should include Ca5(P 0 4 )3 0 H Ca5(P0 4 )3 0 H
electrical conductivity and pH.
2. Interpretation and M anipulation of

Salty Irrigation Supplies
Previous comments suggest that nutritional control based on a single water analysis can be dangerous. If
water quality changes slightly, any program devised with the original analysis in mind becomes improper. As
examples of how to deal with irrigation supplies, and their the limits, four water analyses were selected (Table
6-39). The software program MINTEQA2/PRODEFA2 was used to balance the analyses. The program was
allowed to calculate the equilibrium pH as a check on the pH actually measured. There were no significant
differences between pH, either calculated or measured. These calculations were, of course, highly simplified. A
complete analysis of Sample No. 3 showed traces to significant concentrations of aluminum, iron, manganese,
copper, zinc, nickel, molybdenum, cadmium, chromium, barium, ammonium, and boron. Inclusion of these in
MfNTEQA2/PRODEFA2 would have greatly increased complexity. Always, the presence of bicarbonate resulted
in alkaline pHs. Secondly, Sample 2 precipitated calcite, and Samples 3 and 4 precipitated both calcite and
hydroxyapatite. This suggests that problems might occur with blockage in trickle irrigation systems, whether or
not fertilizer is added to the water. Samples 3 and 4 contained phosphates, nitrates, and potassium, showing
shallow wells contaminated by nutrients leached from the soils above the water table. This is typical of shallow
well supplies. Traces of these nutrients will seldom be found in deep wells (e.g., more than 30 to 60 m deep).
Tables 6-40 and 6-41 show the method for modifying Samples 1 and 2 to reduce salinity and pH. Note the
use of phosphoric and nitric acids. Although these acids are highly caustic, they are much better than sulfuric,
which can greatly increase sulfate without contributing anything except to acidify the solution. The results of
calculation with MINTEQA2/PRODEFA2 are given in Table 6-42. One set of values shows the results of
modification according to the water analysis; the other shows the results if the carnation solution as devised in
Table 6-30 was simply added to the water supply. Although the estimated EC of the carnation solution in Table
6-30 resulted in a value of 2.4 dS m'^ measurements over several years showed an average of about 1.5 dS m'^
for a total of 27 meq Even with modification of the basic solution, total meq increased to 30.4 meq f ’ versus
33.6 meq f ’ if the basic solution was simply added without regard to the water analysis. For Sample No. 2, a
modified solution resulted in 32.2 meq total versus 42 meq total for the unmodified process. This would
probably result in an EC greater than 3.0 dS m \ In both cases, the use of acids caused pH to be reduced to 6 .1
o
to
O
p
o»1
C
3 .
o‘
1:3
Table 6-40. Manipulation of a water supply with an EC of 0.3 dS and 2.7 meq f* HC0 3 '. The carnation solution from Tables 6-22 and 6-29 is to
be used. Units = meq
K* Ca^* Na* NH/ Total NOj H jP 0 4 SO4" Cl HCO 3 Total mg in 1 i

cations anions required
Well water analysis 0.9 0.7 1.7 3.3 ______________ 0^5___ OJ_____2.7_ ____ 3 ^ ______________
Actual solution desired in 6.0 3.0 2.0 2.5 13.5 10.4 1.1 2.0 13.5
good water
Fertilizers required
Potassium nitrate 6.0 6.0 606
Monoammonium phosphate 1.1 1.1 127
Nitric acid^ 0 2.7 0 170
Calcium nitrate 2.1 2.1 248
Magnesium sulfate 1.3 1.3 160
Ammonium sulfate 1.4 1.4 311
"l07s " l.T ' 3 .2 0.1 0 ~ 15.2
Nitric acid was used to neutralize bicarbonate so values of and HCO^' were set to zero.
Table 6-41. Manipulation o f water supply with 0.5 dS m'^ EC, using carnation solution in Tables 6-22 and 6-29. Units = meq f'
H* K* Ca^* Na* NH/ Total NO 3- H 2PO 4- SO /- cr HCO 3- Total mg fertilizer

cations anions required per
d
Raw water analysis 4.0 1.6 2.0 7.6 _______________ 2 .S _ _ _ 0 A _ __ 4 4 ___ J ^ 6 _ _ ______ ____
Desired solution in pure 6.0 3.0 2.0 0 2.5 13.5 10.4 1.1 2.0 0 0 13.5
water
Fertilizers required
Phosphoric acid 1.1 36
Nitric acid 3.3 208
Potassium nitrate 6.0 6.0 606
Ammonium nitrate 2.5 2.5 200
Totals 6.0 \.6 ~ ~ Z 0 ^2.5 I 6 .i~ ~ 11.8 1.1 2.8 0.4 16.1
Table 6-42. Calculated ionic concentrations (pM ) in two solutions, one with total soluble salts in the
basic water supply of 0.3 dS m'^ (Sample 1) and the other with 0.5 dS m‘h With the carnation supply
solution, both samples modified to keep total salts as low as possible and neutralize bicarbonates in the
solutions. Both water supplies were calculated again with the carnation solution (Tables 6-21, 6-29)
added directly to the basic water without modification. Calculations included micronutrients with DTP A
chelate, CO2 concentration at 300 Pa. Computer program was MINTEQA2/PRODEFA2 (Allison et
al.,1991). Refer to Tables 6-40 and 6-41 for composition of modified solutions using acids to neutralize
bicarbonate. Components having concentrations less than 1 pM f' ignored. Components with a
*are neutral, undissociated molecules in solution.
Water sample No. 1 (0.3 dS m'^) Water sample No. 2 (0.5 dS m'^)
Calculation
for: Modified Unmodified Modified Unmodified
pH 6 .2 7.7 6 .1 7.6
NH 4" 2137 2137 2188 2089
5248 5248 5248 5248
Ca'" 550 676 813 912
550 724 437 759
N0 3 - 6309 9120 10232 8912
so/- 813 603 692 1096
Na^ 1479 1479 1862 1737
H3B 0 3 ° 18 18 18 18
cr 87 87 347 347
Mn^^ 1 2
NH 3° 2 56 1 50
NH 4 SO4' 23 17 19 30
MgC0 3 ° 3 3
MgHC0 3 ^ 18 17
MgS 0 4 ° 76 74 51 141
MgH2P04^ 1 1
MgHP0 4 ° 3 2
CaC0 3 ° 4 5
CaS0 4 ° 89 85 112 204
CaHP0 4 ° 2 2
CaH2P0 4 ^ 1 1
NaHC0 3 ° 2 2
NaS 0 4 - 6 5 6 10
KS0 4 - 30 22 25 40
HCO3- 66 2089 51 1905
H2C0 3 ° 102 102 102 102
HP04'- 7 5
H2P0 4 ^- 78 71
CaDTPA^- 1
FeDTPA^- 4
MnDTPA'- 2 2 3
Precipitated M nP04l.5H20 P-Mn0 2 M nP04l.5H 20 P-Mn0 2
substances a-FeOOH a-FeOOH cc-FeOOH a-FeOOH
Ca5(P0 4 )3 0 H Ca5(P0 4 )3 0 H Ca5(P 0 4 )3 0 H Ca5(P 0 4 )3 0 H
CaCÛ3
505
versus 7.7 where HCO 3 ' was not neutralized.

The total ion concentration for Sample No. 3 (0.9 dS m'^) when modified was 35.4 meq or about 2.5 dS
m■^ The total for Sample No. 4 was estimated above 3.0 dS m '\ The latter value for irrigation supplies is the
suggested limit for successful crop production [Richards, 1954]. It is doubtful that basic supplies above 0.75 dS
m \ when fertilizer is injected, will give acceptable growth.
DTP A complexed with Fe^”^, Fe^^, Cu^^, Mn^^, and Zif^. The proportions changed, depending upon pH
(see Figs. 6-41 and 6-42); 45 to 52% of the DTP A complexed with iron, and 37 to 44% with manganese in
modified solutions, using No. 1 and 2 water samples. In the unmodified, high pH solutions, 24% of the DTP A
was complexed with calcium. Formation of MnDTPA utilized 6 6 to 99% of the manganese. Nearly 100% of the
zinc, copper and iron complexed with DTPA. In the unmodified solutions, H P O /' and H 2 PO 4 ' levels dropped
below 1 pM and MgC 0 3 °, CaHC 0 3 ° and NaHC 0 3 ° levels rose above 1 pM. Thus, failure to lower the pH
markedly reduced phosphorous supply.
Note (Table 6-42) that solids were calculated to precipitate in all solutions with calcite appearing in the
unmodified Sample No. 2. Goethite (a-FeOOH), pyrolusite ((3 -Mn0 2 ), and hydroxyapatite (Ca5 (P 0 4 )3 0 H) were
generally the solids to be found.
To the best of my knowledge, these types of calculations have not been published in the technical literature
or in the scientific horticultural literature. Fertilizer recommendations in the industry do not commonly
acknowledge the influence of basic water composition. Ratios of the different nutrients are considered as though
pure water was at hand for all. Cooper [1979] stated that total EC ofNFT solutions should never drop below 2.0
dS m'^ My experience suggests the opposite is likely to be more common. Cooper spent considerable effort in
dealing with nutrient toxicities and solution control. Unfortunately, computer software, such as applied here, was
unavailable when both Cooper and Jones published their manuals on hydroponics [1979; 1983]. These software
programs could be incorporated for real-time solution of fertigation problems, given suitable water supplies and
instantaneous ion measurement.
3. Im proving W ater Supplies

When a grower faces a poor water supply, the easiest thing to do is to change to a better supply -even if
that costs money. Most domestic supplies that do not depend upon shallow wells will usually be higher quality,
i.e., fewer salts. In some countries, however, high quality is not obtainable. If the water supply contains mostly
chlorides and sodium, higher concentrations may by lived with, especially if the substrate is loamless or one is
using hydroponics. Sodium is not likely to be as dangerous.
The other route to take is directly to improve the water supply by removing salts. Processes for manipulating
water supplies are outlined in Table 6-43, except for electrodialysis. Systems such as softening to reduce hardness
are of no use in greenhouse culture. Coagulation, sedimentation, and filtration are commonly necessary with any
method such as reverse osmosis (RO) or electrodialysis (EDR). Evaporation or distillation systems are not used
for greenhouse production, but they are often employed in large-scale plants with a variety of modifications such
as vertical tube distillation, multistage flash distillation, multi-effect multistage and vapor compression distillation
[e.g., U.S. Dept. Interior, 1962; 1968; 1979; Avissar and Mahrer, 1986]. Reverse osmosis uses a semipermeable
membrane through which water is passed by pressurization, leaving the salts behind. Feedwater often requires
adjustment to eliminate scaling, CaS0 4 precipitation, and some form of biocide to control fouling. Capital costs,
given by the Office of Saline Water range from about $400 per m'^ dy'^ for an installation of about 40 m^ daily
capacity to $145 m‘^ for installations having capacities up to 19000 m^ daily. Operating costs range from about
20 to 30 0 per m^ for small installations down to 2 to 3 0 per m^ for large operations. There are several variations,
depending upon the water supply to be purified. Irreversible membrane fouling is sometimes encountered unless
the system is equipped with appropriate safety devices and the operators properly trained.
Table 6-43. Outline of some water and waste treatment proeesses [U.S. Dept. Interior, 1979]. Electrodialysis not included.
Process General use or capability Advantages Disadvantages

Coagulation, sedimentation, Reduction in suspended sol Low cost, simplicity Large area required. Does
and filtration ids by 90-98% not remove salts or organics
Softening Reduction in hardness by 95 Relatively low in cost, sim Requires frequent regenera
to 100% plicity tion. Does not remove
organics (of no use in green
houses)
Ion exchange Reduction of salts by 95 to Can reach very low levels of Does not remove organics.
100% salinity Requires regeneration
Biological treatment Reduction of organics by 50 Low cost Subject to upset and varia
to 90% tions. Limited to 90% of
organic loading
Carbon columns Reduction of organic load Good way to remove small Relatively expensive regen
ing by 95 to 100% amounts of organics eration. Removes only spe
cific organics
Evaporation Removes dissolved and sus Well demonstrated. Handles High energy use
pended solids wide range
Reverse osmosis Reduces nearly all contami Simple, low energy utiliza Membrane subject to deteri
nants by 90 to 95% tion oration. Requires some form
of pretreatment. Limited
chemical compatibility
Electrodialysis uses DC current to perform separation of charged ions and their removal across suitable
membranes [U.S. Dept. Interior, 1979]. EDA is the most advanced of the membrane processes, and it is generally
favored for brackish-water conversion. Electrical costs will increase with the feedwater salinity. Capital costs
for a polarity reversal plant, for feedwater with 2.3 dS m’’ (1500 ppm total dissolved salts), and an output of 0.24
dS m \ range from $144 to $526 per cubic meter of installed capacity. Operating costs are 10 to 13 0 U.S. per
m^ for a system processing about 270 m^ daily. A polarity reversal system requires no chemical feed or pretreat
ment.
Packaged reverse osmosis and electrodialysis equipment can be readily purchased in the U.S., in sizes
ranging from a few ^ s per day to several hundred cubic meters per day. Growers producing seedlings are
commonly forced to dimineralization when their basic water quality is salty. Cut-flower growers also find it
beneficial to use such systems for handling cut flowers after harvest [e.g., Montgomery, 1984; Anon., 1974]. The
idea is not to remove all ions, but to reduce concentrations to something near water sample No. 1 in Table 6-39.
Another problem with these systems is disposal of the concentrated brine. Government regulatory agencies
are likely to be severe if the grower merely wastes the brine to the most convenient ditch or stream. These
methods are not cheap, but acceptable production requires a good water supply.
One way to help surmount the problem is to collect rainwater from the greenhouse roof and store it in a cis
tern. Some operations have invested in this type of system, especially where high-quality water is required for
misting roses or other plants. Deposition of salts on foliage usually causes marked damage, not to mention effects
on observable quality. The same water supply can also be used to dilute the poorer water to bring it into a
desirable range. Fig. 6-64 is an example of irrigating an ornamental crop with hard water in an African country.
Each sprinkler leaves its own pattern in precipitated salts on the screen covering the crop. This obviously will
reduce quality.
Far too often, I have seen greenhouse operations started with little attention given to the water supply. The
507
■ ft'i-^ ^ ^- ......'.^-
■'*'
"” -." ...... "'
' - ' ^’ ■
', , , ^ ',' =^» '- i
.......
Fig. 6-65. Example of irrigating a cloth house with water containing high calcium and magnesium
carbonates. Each circle is a sprinkler [Courtesy of L. Edstrom].
Table 6-44. Net profits (U.S.$) of reference and closed systems for greenhouse use in The
Netherlands [van Os et al., 1991] (©1991, Int. Soc. Hort. Sci., A c ta H o r t ., 294:49-57).
System type Chrysanthe Lettuce Radish Cucumber

mum
Reference greenhouse'
Polypropene film under growing layer +0.56^ -2.31 -1.11 -0.62
Gullies
Nutrient film technique -1.38 -5.50 -4.46
Substrate slabs __ -0.45
Slabs and containers - - - +0.24
Aeroponics -1.38 -- - -0.27
Concrete floor^ -2.84 -7.40 -7.22 -2.43
Transportable benches^ -5.81 -11.68 -11.95 ?
' Crops in soil except cucumber which is grown in rockwool slabs
^ Dutch guilders converted to U.S. dollars.
^ May use sub-irrigation or ebb-and-flow.
idea is that if one can drink it safely, it is all right to grow plants with it. That is not so. I stated in Chapter 5 that
the approach to nutrition in greenhouse production is archaic, and this must change if sufficient return is to be
made on the capital investment required for greenhouses.
I. ENVIRONM ENTAL CONTAM INATION FROM GREENHOUSES

There are two problems that growers have been forced to deal with in recent years. One is the presence of
contaminants, particularly nitrate, in vegetable crops, the second is waste disposal from the greenhouse operation.
Table 6-45. U.S. Health Advisories for chemicals in parts per billion (ppb) [Shimskey, 1988].
HAs are not enforceable by federal government, but may be at the state level (With permission
of the G reen h o u se G row er).
Chemical ppb Chemical ppb Chemical ppb

Acifluorfen 9 Aldicarb 10 Ametryn 60
Ammonium sulfamate 1500 Atrazine 3 Bentazon 17.5
Bromacil 80 Butylate 50 Carbaryl 700
Carbofuran 36 Carboxin 700 Chloramben 105
Cyanazine 9 Dalapon 560 DCPA ’ 3500
Diazinon 0.63 Dicamba 9 2,4-D 70
Dimethrin 2100 Dinoseb 7 Diphenamid 200
Disulfoton 0.3 Diuron 14 Endothal 140
Endrin 0.32 Fenamiphos 1.8 Fonofos 14
Glyphosate 700 Hexazinone 210 Maleic hydrazide 3500
MCPA 3.6 Methomyl 175 Methoxychlor 340
Methyl parathion 2 Metolachlor 10 Metribuzin 175
Oxamyl 175 Paraquat 3 Pentachlorophenol 220
Picloram 490 Prometon 100 Pronamide 52
Propachlor 92 Propazine 14 Propham 120
Propoxur 3 Silvex 52 Simazine 35
2,4,5-T 21 Tebuthiuron 35 Terbacil 90
Terbufos 0.18 Trifluralin 2
In California, the recent drought and effluent polluting runoff in the Half Moon Bay area, have received
considerable attention [Hasek et al, 1986; Whitesides, 1989]. In other parts of the country, the varying require
ments imposed by states have resulted in grower liability, although their use of chemicals has followed
governmental regulations listed on the product [Schmuck and Firth, 1987; Shimskey, 1988]. The Netherlands
have imposed severe restrictions on polluting factors [van Os et al.,1991] from greenhouses. The policy plan in
The Netherlands, according to van Os et al, is to reduce nitrate and phosphate leaching to surface waters by more
than 50%, reduce use of chemical plant protection products by 50%, reduce use of soil disinfection products by
75%, increase energy efficiency by 50%, and reduce CO 2 emission by 5%. The aim is to grow 80% of the
greenhouse vegetables and pot plants separately from the soil.
Table 6-44 shows net profits of closed systems simulated by the Dutch, compared with a “reference” opera
tion that produces chrysanthemums, lettuce, and radish in soil and cucumbers in rockwool with free drainage.
It can be seen that most of these procedures to reduce pollution also reduce net profit. They are costs that will
be transferred to the consumer, and governments will be required to regulate all growers to prevent an advantage
to those refusing to carry out the new requirements. This is no different from costs of cleanup for electrical
generating, chemical, and manufacturing enterprises. Table 6-45 lists present health advisory limits for several
chemicals used in the U.S. by the industry. Regulating will be necessary since there are always a few who
attempt to increase profit by polluting the commons.
Keeny [1982] states that nitrates are relatively nontoxic to humans. Acute nitrate poisoning in an adult
requires a single oral ingestion of 1 to 2 g N 0 3 ‘ -which is far above usual exposure limits. The adverse effect
results from reduction of NO 3 ' to nitrite (N 0 2 ‘), which can occur in the intestine of some animals and in the
human infant during the first few months of life. Maynard et al. [1976] discussed problems of nitrate
accumulation in vegetables to some detail. According to these authors, a lethal nitrite dose is about 20 mg NO 2 '
per kg body weight. Fatal reactions of this type may be caused by other chemicals. Most cases of acute nitrate
toxicity have occurred largely with households having a private well supply. The standard limits in water supply
509
are 10 mg per ^ N 0 3 '. Instances of abortion in cattle have been noted when fed with plant products produced in
greenhouses.
As was discussed, nitrate concentration in plants varies widely. Accumulation occurs when the ion is
unconverted to other compounds, which may occur with excessive nitrate fertilization, low solar radiation, and
high temperatures. Plants subjected to drought tend to accumulate nitrate. There are numerous other sources of
nitrate [Maynard et al.,1976]. Comments by Rooda van Eysinga [1984] were to the effect that even if fertilizers
are omitted, which would result in 1 0 to 2 0 % yield reductions, the nitrate contents in lettuce would be reduced
by only 1 0 %.
Some investigators have attempted to reduce nitrate in lettuce by transferring the crop to solutions with
diluted N 0 3 ‘ 2 to 7 days before harvest [Shinohara and Suzuki, 1988]. Hydroponic culture in Japan uses plastic
panels floating on the solution, so moving the crop is relatively easy. Eysinga discussed a number of methods
to reduce nitrate in lettuce. Nevertheless, he stated that there are no known cases recorded in which methe
moglobinemia has been proven in adults from eating vegetables with high nitrates. A causal relationship between
vegetable consumption with high nitrate contents and incidence of stomach cancer has never been proved. Some
authors have recommended not eating greenhouse-produced vegetables, but only those produced in season and
including stinging nettles and sorrel. Eysinga noted concentrations of 6150 and 2900 mg NO 3 per kg,
respectively, for these two weeds. This is far above any nitrate levels found in commercial greenhouse vegetables.
One should also keep in mind that not only can nutrients, pesticides, herbicides, and soil chemicals leach into
the ground water, but waste materials such as styrofoam beads (used in potting mixtures), plastics, paper, flue
gases, etc. can cause pollution. Thus, a beginning can be made in reducing pollution from greenhouses by simply
cleaning up the surroundings. O f the individuals immediately concerned with such problems, Biembaum [1992]
discussed methods to reduce runoff from greenhouse property and to lower contamination. Closed systems that
recirculate the water would be the most obvious. Experience has shown, however, that such systems are not
always the practical solution [Whitesides, 1989]. Even with sterilization of the recirculated solution (Fig. 5-54),
a continual increase in salt content usually requires wastage and replenishment with fresh water. Water partially
wasted in evaporative pad systems, to prevent salt accumulation on the pads, is a common practice.
Control of salinity in greenhouse substrates has usually involved excess watering, especially where
fertigation is continually practiced. Greater care in making sure that fertilizer applications are maintained as low
as possible, commensurate with desired results, and measuring the amount of water required by the crop so that
application is not excessive, can go far to reducing excess waste to the environment. Pulsed irrigation applications
have been suggested as a means to reduce water loss [Biembaum, 1992], rather than continuous applications to
wet the entire substrate volume. Slow-release fertilizers have been found to reduce fertilizer wastage from
containers. Biembaum stated that superabsorbent polyacrylamide gels were ineffective under usual watering
practices. Sometimes, growers have placed pans under containers to catch drainage water that can be reused.
Manipulation of the irrigation system to ensure uniform watering can go far in eliminating wastage. Whitesides'
[1989] account of one California operation cited a 50% reduction in fertilizer usage, a reduction in labor of 4000
man-hours, and 30% less water consumption through nutrient and water control by computer with drip irrigation,
variable fertilizer injection, and water consumption correlated with local weather conditions. In all these, storing
m noff from greenhouse roofs is encouraged.
In summary, most of the methods to conserve and reduce pollution require greater attention to detail and less
sloppiness in cultural procedures. The grower should use all tools available to him since this will often increase
efficiency and profitability.
V. REFERENCES
Adams, P. 1987. The test of raised salinity. The Grower. 107(2).

Adriano, D.C. et al. 1986. New results in the research of hardly known trace elements and the analytical problems
of trace element research. Proc. Int. Symp., Budapest, June, 1986. 3-33 pp.
Albery, W.J., B.G.D. Haggett and L.R. Svanberg. 1985. The development of sensors for hydroponics. Bio
sensors. 1:369-397,
Allaway, W.H. 1957. pH, soil acidity and plant growth. In Soil, The Yearbook of Agriculture 1957. A. Stefferud,
ed. U.S. Government Printing Office, Washington, D.C.
Allison, J.D., D.S. Brown and K.J. Novo-Gradac. 1991. MINTEQA2/PRODEFA2. A geochemical model for
environmental systems; Version 3.0 user's manual. EPA/600/3-91/021. U.S. EPA, ORD, Environ. Res. Lab.,
Athens, GA. 106 pp.
Alt, D., V.T. Gizewski and B. Schroer. 1988. Analysis of substrates by NAHCO 3 -DTPA method. A c ta H ort.
221:395-402.
Ames, R.N. and R.G. Linderman. 1977. Vesicular-arbuscular mycorrhizae of Easter lily in the northwestern
United States. Can. J. M ic ro b io l. 23:1663-1668.
Ames, R.N. and R.G. Linderman. 1978. The growth of Easter lily (LiHum lon gifloru m ) as influenced by
vesicular-arbuscular mycorrhizal fungi, F usarium oxysporu m , and fertility level. Can. J. B ot. 56:2773-2783.
Anon. 1974. R.O. helps the flowers grow. W ater C o n d itio n in g . June, 1974, 12-13.
Anon. 1978. Water quality vital for NET success. The G ro w e r. 89(15):880.
Anon. 1985. Soil lab tests help growers adjust soil, define nutrient needs. G reen h o u se M g r. Oct., 1985.
Argo, W.R. and J.A. Biembaum. 1994. Irrigation requirements, root-medium pH, and nutrient concentrations
of Easter Lilies grown in five peat-based media with and without an evaporation barrier. J. A m er. Soc.
H ort. Sei. 119:1151-1156.
Asher, C.J. 1991. Beneficial elements, functional nutrients, and possible new essential elements. In
Micronutrients in Agriculture. J.J. Mortvedt et al .eds. 2nd edition. No. 4. Soil Sei. Soc. Amer., Madison, WI.
Aung, L.H. 1974. Root-shoot relationships. In The Plant Root and Its Environment. E.W. Carson, ed. Univ. Press
of VA, Charlottesville. 29-61.
Avissar, R. and Y. Mahrer. 1986. Water desalination in solar earth stills; A numerical study. W ater R e so u rc e s
Res. 7:1067-1075.
Baldwin, J.P., P.B. Tinker and P.H. Nye. 1972. Uptake of solutes by multiple root systems from soil. II. The
theoretical effects of rooting density and pattern on uptake of nutrients from soil. P la n t a n d Soil. 36:693-708.
Ball, V., ed. 1985. The Ball Red Book. Reston Publ. Co., Reston, VA. 720 pp.
Barber, D.A. and J.K. Martin. 1976. The release of organic substances by cereal roots into soil. N e w P h ytol.
76:69-80.
Barber, S.A. 1962. A diffusion and mass-flow concept of soil nutrient availability. S o il Sei. 93:39-49.
Barber, S.A. 1974. Influence of the plant root on ion movement in soil. In The Plant Root and Its Environment.
E.W. Carson, ed. Univ. Press of VA., Charlottesville. 525-564.
Barker, A.V. and H.A. Mills. 1980. Ammonium and nitrate nutrition of horticultural crops. H ort. Rev. 2:395-423.
Barragry, A.R. and J.V. Morgan. 1978. Effect of mineral and slow-release nitrogen combinations on the growth
of tomato in a coniferous bark medium. A c ta H ort. 82:43-53.
Barrows, J.B. and R.W. Roncadori. 1977. Endomychorrizal synthesis by G ig a s p o r a m a r g a rita in poinsettia.
M y c o lo g ia . 69:1173-1184.
Bates, T.E. 1971. Factors affecting critical nutrient concentrations in plants and their evaluation: A review. S o il
Sei. 112:116-130.
Bates, T.E. 1990. Prediction of phosphorous availability from 8 8 Ontario soils using five phosphorous tests.
C om m un. in S o il Sei. P la n t A n al. 21:1009-1023.
Bear, F.E. 1957. Toxic elements in soils. In Soils. Yearbook of Agriculture 1957. A. Stefferud, ed. U.S. Govern.
Publ., Washington, D.C.
Beaufils, E.R. 1971. Physiological diagnosis - A guide for improving maize production based on principles
developed for rubber trees. F ert. Soc. S. Afr. 1:1-30.
Beaufils, E.R. 1973. Diagnosis and recommendation integrated system (DRIS). A general scheme for
experimentation and calibration based on principles developed from research in plant nutrition. Univ. o f
N a ta l S o il Sei. Bull. 1:1-132.
Berghage, R.D. et al. 1987. Micronutrient testing of plant growth media: Extractant identification and evaluation.
C om m un. in S o il Sei. P la n t A n al. 18:1089-1109.
Bernstein, L. 1961. Osmotic adjustment of plants to saline media. I. Steady state. A m er. J. B ot. 48:909-918.
Bernstein, L. 1963. Osmotic adjustment of plants to saline media. II. Dynamic phase. A m er. J. B ot. 50:360-370.
Bernstein, L. and H.E. Hayward. 1958. Physiology of salt tolerance. Ann. Rev. P la n t P h ysio l. 9:25-46.
511
Biembaum, J.A. 1992. Root-zone management of greenhouse container-grown crops to control water and
fertilizer use. H o rtT e c h n o lo g y . 2:127-132.
Biembaum, J.A. and W.R. Argo. 1993. Greenhouse peat-based root media: Properties and components.
Unpublished MS. Dept, of Horticulture, MI State Univ., East Lansing, MI.
Bik, R.A. and G.A. Boertje. 1975. Fertilizing standards for potting composts based on the 1:LA volume
extraction method of soil testing. A c ta H ort. 50:153-156.
Black, C.A. 1992. Soil Fertility Evaluation and Control. Lewis Publ., Boca Raton, FL. 746 pp.
Blackman, F.F. 1905. Optima and limiting factors. Ann. B ot. 15:185-194.
Blair, G.J, M.H. Miller and W.A. Mitchell. 1970. Nitrate and ammonium as sources of nitrogen for com and their
influence on the uptake of other ions. A gron. J. 62:530-532.
Boertje, G.A. 1980. Results of liquid feeding in the production of bedding plants. A c ta H ort. 99:17-23.
Boodley, J.W. 1962. Fertilization. In Snapdragons. R.W. Langhans, ed. Cornell Univ., Ithaca, NY. 185 pp.
Boodley, J.W. 1974. Nutrient content of Paul Mikkelsen poinsettias from juvenile to mature growth. F lo rists'
R ev. Dec. 26, 1974.
Boodley, J.W. and J.W. White. 1969. Fertilization. In Roses. J.W. Mastalerz and R.W. Langhans eds. PA Flower
Growers Assoc., University Park, PA.
Bowe, R. et al. 1969. Gerbera. V.J. Neumann-Neudamm, Berlin. 180 pp.
Boyko, H. 1967. Salt-water agriculture. Sci. A m er. 216:89-96.
Boyle, F.W., Jr. and W.L. Lindsay. 1986. Manganese phosphate equilibrium relationships in soils. S o il Sci. Soc.
A m er. J. 50:588-593.
Brady, N.C. 1974. The Nature and Properties of Soils. MacMillan Publ. Co., New York. 639 pp.
Bray, R.H. 1963. Confirmation of the nutrient mobility concept of soil-plant relationships. S o il Sci. 95:124-130.
Brennan, E.G. and J.W. Shive. 1948. Effect of calcium and boron nutrition of the tomato on the relation between
these elements in the tissues. S o il Sci. 66:65-75.
Brouwer, R. 1963. The influence of the suction tension of the nutrient solutions on growth, transpiration and
diffusion pressure deficit of bean leaves {P h a seo lu s v u lg a ris). A c ta Bot. N e e rl. 12:248-261.
Bunt, A.C. 1976. Modem Potting Composts. Allen and Unwin, London. 277 pp.
Bunt, A.C. 1980. Phosphorous sources for loamless substrates. A c ta H ort. 99:25-32.
Bunt, A.C. 1986. Problems in the analysis of organic and lightweight potting substrates. H o rtS cien ce. 21:229-
231.
Bunt A.C. 1988. Media and Mixes for Container-Grown Plants. Unwin Hyman, London. 309 pp.
Canning, R.E. and P.J. Kramer. 1958. Salt absorption and accumulation in various regions of roots. Am er. J. Bot.
45:378-382.
Cate, R.B. and L.A. Nelson. 1971. A simple statistical procedure for partitioning soil test correlation data into
two classes. S o il Sci. Soc. A m er. P ro c. 35:658-660.
Chapman, H.D. 1960. The diagnosis and control of zinc deficiency and excess. Bull. Res. C o u n cil o f Isra el. 8 D.
Clarkson, D.T. 1985. Factors affecting mineral nutrient acquisition by plants. Ann. Rev. P la n t P h ysiol. 36:77-115.
Clement, C.R., M.J. Hopper and L.H.P. Jones. 1978. The uptake of nitrate by L oliu m p e re n n e from flowing
nutrient solution. I. Effect of N 0 3 ‘ concentration. J. Expt. B ot. 29:453-464.
Colegrave, D. et al. 1973. The Colegrave Manual of Bedding Plants. Grower Books, London. 211 pp.
Cooper, A. 1979. The ABC of NFT. Grower Books, London. 181 pp.
Cregan, P.D., J.R. Hirth and M.K. Conyers. 1989. Amelioration of soil acidity by liming and other amdendments.
In Soil Acidity and Plant Growth. A.D. Robson, ed. Academic Press, New York. 306 pp.
Crews, C.E., C.R. Johnson and J.N. Joiner. 1978. Benefits of mycorrhizae on growth and development of three
woody ornamentals. H o rtS cien ce. 13:429-430.
Crockford, H.D. and S.B. Knight. 1964. Fundamentals of Physical Chemistry. 2nd, ed. J. Wiley & Sons, New
York. 415 pp.
Dahnke, W.C. and R.A. Olson. 1990. Soil test correlation, calibration and recommendation. In Soil Testing and
Plant Analysis. R.L. Westerman, ed. Soil Sci. Soc. Amer. No. 3. Soil Sci. Soc. Amer., Madison, WI.
Dean, L.A. 1957. Plant nutrition and soil fertility. In Soil, The Yearbook of Agriculture 1957. A. Steffemd, ed.
U.S. Government Printing Office, Washington, D.C.
De Hertogh, A. 1958. Holland Bulb Forcer’s Guide. Int. Flower Bulb Centre, Hillegom, The Netherlands. 284
PP-
Dight, R.J.W. 1977. Nutritional requirements of bedding plants. Expl Hort. 29:63-71.
Dole, J.M. and H.F. Wilkins. 1991. Relationship between nodal position and plant age on the nutrient
composition of vegetative poinsettia leaves. J. Amer. Soc. Hort. Sei. 116:248-252.
Ecke, P., Jr. 1971. The Poinsettia Manual. Paul Ecke Poinsettias, Encinitas, CA. 183 pp.
Editors. 1980. Roses Under Glass. No. 9. Grower Books, London. 78 pp.
Editors. 1980. Cucumbers. No. 15. Grower Books, London. 74 pp.
Editors. 1981. Lettuce Under Glass. No. 21. Grower Books, London. 105 pp.
Edwards, D.G. and C.J. Asher. 1974. The significance of solution flow rate in flowing culture experiments. Plant
a n d S o ilA \:\6 \A 1 5 .
Elgawhary, S.M., W.L. Lindsay and W.D.Kemper. 1970. Effect of complexing agents and acids on the diffusion
of zinc to a simulated root. Soil Sei. Soe. Amer. Proe. 34:211-214.
El Kohli, A.F. 1961. An experimental study of the influence of the microelements on the uptake of macro
elements. Versl. Landbouwk. onderz. NR. 61.A. Wageningen
Ellis, N.K. et al. 1974. Nutriculture systems. Purdue Univ. Agrie. Expt. Sta. Bull. No. 44. 20 pp.
Fernández Falcon, M. et al. 1986. The effect of chloride and bicarbonate levels in irrigation water on nutrition
content, production and quality of cut roses 'Mercedes'. Sei. Hort. 29:373-385.
Follett, R.H. and P.N. Soltanpour. 1985. Irrigation water quality criteria. Service in Action, CO State Univ. Ext.
Serviee. No. 506. 2 pp.
Foy, C.D. 1992. Soil chemical factors limiting plant root growth. Adv. in Soil Sei. 19:97-149.
Frick, J. And C.A. Mitchell. 1993. Stabilization of pH in solid-matrix hydroponic systems. HortSeienee.
28:981-984.
Geraldson, C.M. 1954. The control of blackheart of celery. Proe. Amer. Soe. Hort. Sei. 63:353-358.
Geraldson, C.M. 1955. The use of calcium for control of blossom-end rot of tomatoes. Proe. FL State Hort. Soe.
68:197-202.
Geraldson, C.M., G.R. Klacan and O.A. Lorenz. 1973. Plant analys as an aid in fertilizing vegetable crops. In
Soil Testing and Plant Analysis. L.M. Walsh and J.D. Beaton, eds. Soil Sei. Soc. Amer., Madison, WI.
Gerdemann, J.W. 1974. Mycorrhizae. In The Plant Root and Its Environment. E.W. Carson, ed. Univ. Press of
VA., Charlottesville. 205-217.
Gieling, T.H., E. van Os and A. de Jager. 1988. The application of chemo-sensors and bio-sensors for soilless
culture. Aeta Hort. 230:357-361.
Gieling, T.H. and K. Schurer. 1995. Sensors and measurement. In Greenhouse Climate Control. J.C. Bakker
et a l, eds. Wageningen Pers.
Gislerod, H.R. and L.M. Mortensen. 1990. Relative humidity and nutrient concentration affect nutrient uptake
and growth of Begonia x hiemalis. HortSeienee. 25:524-526.
Gislerod, H.R. and L.M. Mortensen. 1991. Air humidity and nutrient concentration affect nutrient uptake and
growth of some greenhouse plants. Aeta Hort. 294:230-235.
Green, J.L. 1967a. Ionic balance and growth of carnations - 1. Recent Hature. CO Flower Growers’Assoe. Res.
Bull. 210:1-4.
Green, J.L. 1967b. Ionic balance and growth of carnation. CO Flower Growers' Assoe. Res. Bull. 211:1-5.
Green, J.L., C.V. Cole and W.D. Holley. 1971. Ionic balance and growth of carnations. Soil Sei. 111:138-145.
Grinsted, M.J. et al. 1982. Plant-induced changes in the rhizosphere of rape (Brassiea napus var. Emerald)
seedlings. I. pH change and the increase of P concentration in the soil solution. New Phytol. 91:19-29.
Hall, D.A. and G. Wilson. 1980. Slow-release mix simple solution for small growers. The Grower. 93(5):37-3 8.
Halvorson, A.D. 1971. Chelation and availability of metal ions in nutrient solutions. Ph.D. Dissertation. Colo.
State Univ., Ft. Collins. 134 pp.
Halvorson, A.D. and W.L. Lindsay. 1972. Equilibrium relationships of metal chelates in hydroponic solutions.
Soil Sei. Soe. Amer. Proe. 36:755-761.
Hanan, J.J. 1964. Oxygen and carbon dioxide concentrations in greenhouse soil-air. Proe. Amer. Soe. Hort. Sei.
84:648-652.
Hanan, J.J. 1973. Denver water quality. CO Flower Growers'Assoe. Res. Bull. 280:1-3.
513
Hanan, J J . 1975. Calcium tissue levels in the carnation. CO Flower Growers' Assoc. Res. Bull. 305:1-4.
Hanan, J.J. 1979. Carnation fertilizer injection analysis for the Denver region. CO Flower Growers’Assoc. Res.
Bull 343:3-4.
Hanan, J.J. 1982a. Salinity II: A problem in terms. CO Greenhouse Growers' Assoc. Res. Bull. 383:1-3.
Hanan, J.J. 1982b. Salinity III: Handling water supplies to minimize salinity problems. CO Greenhouse Growers'
Assoc. Res. Bull. 384:1-4.
Hanan, J.J. 1984. Thirty years of nutrition studies at Colorado State University. CO Greenhouse Growers' Assoc.
Res. Bull.A\3:l-3.
Hanan, J.J. 1988. Manipulation of nutrition using a salty water. CO Greenhouse Growers' Assoc. Res. Bull.
453:1-3.
Hanan, J.J. and W.D. Holley. 1974. Hydroponics. CO State Univ. Expt. Sta. General Series 941, 21 pp.
Hanan, J.J., W.D. Holley and K.L. Goldsberry. 1978. Greenhouse Management. Springer-Verlag, Heidelburg.
530 pp.
Hanan, J.J., K. Schekel and W.D. Holley. 1968. Colorado water analysis. CO Flower Growers' Assoc. Res. Bull.
222:2-4.
Hansen, M. 1978. Plant specific nutrition and preparation of nutrient solutions. Acta Hort. 82:109-112.
Handreck, K.A. 1991. Assessment of available nutrients in potting media using dilute DTP A extractants. Acta
Hort. 294:250-258.
Harbaugh, B.K. and G.J. Wilfiret. 1982. Correct temperature is the key to successful use of Osmocote. Florists ’
Review. 170(4403):21-23.
Harbaugh, B.K. and S.S. Woltz. 1989. Fertilization practice and foliar-bract calcium sprays reduce incidence
of marginal bract necrosis of poinsettia. HortScience. 24:465-468.
Harter, R.D. 1991. Micronutrient adsorption-desorption reactions in soils. In Micronutrients in Agriculture. J.J.
Mortvedt et al., eds. 2nd edition. No. 4. Soil Sci. Soc. Amer., Madison, WI.
Hartman, L.D. 1971. Effects of irrigation frequency and nutrient solution concentration on yield of carnation.
CO Flower Growers' Assoc. Res. Bull. 250:1-5.
Hasek, R.F., R.H. Sciaroni and R.L. Branson. 1986. Water conservation and recycling in ornamentals production.
HortScience. 21:35-38.
Hayward, H.E. and W.B. Spurr. 1944. Effects of isosmotic concentrations of inorganic and organic substrates
on entry of water into com roots. Bot. Gaz. 106:131-139.
Helyar, K.R. and W.M. Porter. 1989. Soil acidification, its measurement and the processes involved. In Soil
Acidity and Plant Growth. D. Robson, ed. Academic Press, New York. 306 pp.
Hewitt, E.J. 1966. Sand and water culture methods used in the study of plant nutrition. Tech. Comm. No. 22
(Revised). Commonwealth Agric. Bur., Maidstone, Kent, England.
Hiatt, A.J. and J.E. Leggett. 1974. Ionic interactions and antagonisms in plants. In The Plant Root and Its
Environment. E.W. Carson, ed. Univ. Press of VA., Charlottesville. 101-134.
Hoagland, D.R. and D.I. Amon. 1950. The water-culture method for growing plants without soil. Calif. Agric.
Expt. Sta. Circ. 347.
Hofstra, J.J. and T. Koch-Bosma. 1970. Organic phosphate in ammonium- and nitrate-fed tomato plants. Acta
Bot. Neerl. 19:546-552.
Holcomb, E.J. and J.W. White. 1982. Fertilization. In Geraniums. J.W. Mastalerz and E.J. Holcomb, eds. PA
Flower Growers, University Park, PA. 410 pp.
Holevar, C.D. 1966. The effect of vesicular-arbuscular mycorrhiza on the uptake of soil phosphorous by
strawberry (Fragaria sp. var. Cambridge Favorite). J. Hort. Sci. 41:57-64.
Holland, D.A. 1966. The interpretation of leaf analysis. J. Hort. Sci. 41:311-329.
Holley, W.D. and R. Baker. 1991. Carnation Production II. Kendall/Hunt Publ. Co., Dubuque, lA. 156 pp..
Hood, T.M., H.A. Mills and P.A. Thomas. 1993. Developmental stage affects nutrient uptake by four
snapdragon cultivars. HortScience. 28:1008-1010.
Hughes, H. 1975. Progress report. Effect of salinity on roses. CO Flower Growers’Assoc. Res. Bull. 304:1-3.
Hughes, H. and J.J. Hanan. 1977. Effects of salinity in water supplies on rose production. Experiment 2. CO
Flower Growers' Assoc. Res. Bull. 327:1-3.
Hurd, R.G. 1978. The root and its environment in the nutrient film technique of water culture. Acta Hort. 82:87-
97.
Ingestad, T. 1970. A definition of optimum nutrient requirements in Birch seedlings. Physiol Plant. 23:1127-
1138.
Ingestad, T. 1971. A definition of optimum nutrient requirements of birch seedlings. II. Physiol Plant. 24:118-
125.
Ingestad, T. 1973b. Mineral nutrient requirements of cucumber seedlings. Plant Physiol. 52:332-338.
Ingestad, T. 1974. Towards optimum fertilization. AMBIO. 3:49-54.
Ingestad, T. 1982. Relative addition rate and external concentration; driving variables used in plant nutrition
research. Plant, Cell and Environ. 5:443-453.
James, D.W. and K.L. Wells. 1990. Soil sample collection and handling: Technique based on source and degree
of field variability. In Soil Testing and Plant Analysis. R.L. Westerman, ed. Soil Sci. Soc. Amer. Series No.
3. Soil Sci. Soc. Amer., Madison, WI.
Jansson, S.L. and J. Persson. 1982. Mineralization and immbolization of soil nitrogen. In Nitrogen in Agricultural
Soils. F.J. Stevenson, ed. No. 22, Amer. Soc. Agron., Madison, WI. 229-252.
Jeong, B.R. 1990. Ammonium and nitrate nutrition of selected bedding plants. Ph.D. Dissertation, CO State
Univ., Ft. Collins. 274 pp.
Johnson, E.W. 1980. Comparison of methods of analysis for loamless composts. Acta Hort. 99:197-204.
Johnson, C.R. and C.E. Crews, Jr. 1979. Survival of mychorrizal plants in the landscape. Amer. Nurseryman.
150:15,59.
Joiner, J.N., ed. 1981. Foliage Plant Production. Prentice-Hall, Inglewood Cliffs, NJ. 614 pp.
Joiner, J.N. 1983. Nutrition and fertilization of ornamental greenhouse crops. Hort. Rev. 5:317-403.
Joiner, J.N., C.A. Conover and R.T. Poole. 1981. Nutrition and fertilization. In Foliage Plant Production. J.N.
Joiner, ed.. Prentice-Hall, Englewood, NJ. 614 pp.
Jones, J.B., Jr. 1983. A Guide for the Hydroponic and Soilless Culture Grower. Timber Press, Portland, OR. 124
pp.
Jones, J.B., Jr. 1985. Soil testing and plant analysis: Guides to the fertilization of horticultural crops. Hort. Rev.
7:1-68.
Jones, J.B . Jr. 1990. Universal soil extractants: Their composition and use. Commun. in Soil Sci Plant Anal.
21:1091-1101.
Jones, J.B., Jr. 1991. Plant tissue analysis in micronutrients. In Micronutrients in Agriculture. J.J. Mortvedt et
al., eds. 2nd edition. No. 4. Soil Sci. Soc. Amer., Madison, WI.
Jones, R.G.W. 1981. Plant physiology. In McGraw-Hill Yearbook of Science and Technology. McGraw-Hill,
New York. 303-306.
Keeney, D.R. 1982. Nitrogen management for maximum efficiency and minimum pollution. In Nitrogen in
Agricultural Soils. F.J. Stevenson, ed. No. 22, Amer. Soc. Agron., Madison, WI. 605-649,
Kerr, G.P. and J.J. Hanan. 1985. Leaching of container media. J. Amer. Soc. Hort. Sci. 110:474-480.
Kingham, H.G., ed. 1973. The U.K. Tomato Manual. Grower Books, London. 223 pp.
Kiplinger, D.C. and R.W. Langhans, eds. 1967. Easter Lilies. N.Y. State College of Agric., Cornell Univ., Ithaca,
NY. 158 pp.
Kochian, L.V. 1991. Mechanisms of micronutrient uptake and translocation in plants. In Micronutrients in
Agriculture. J.J. Mortvedt et al., eds. 2nd edition. No. 4. Soil Sci. Soc. Amer., Madison, WI.
Kofranek, A.M. 1980. Mineral nutrition. In Introduction to Floriculture. R.A. Larson, ed.. Academic Press, New
York.
Kofranek, A.M. and R.A. Larson eds. 1975. Growing Azaleas Commercially. Publ. 4058., Univ. of CA., Davis,
CA. 108 pp.
Kofranek, A.M. and O.R. Lunt. 1975. Mineral nutrition. In Growing Azaleas Commercially. A.M. Kofranek and
R.A. Larson eds. Publ. No. 4058, Univ. of CA, Davis, CA.
Koide, R.T. and R.P. Schreiner. 1992. Regulation of the vesicular-arbuscular mychorrizal symbiosis. Ann. Rev.
Plant Physiol, and Plant Mol. Biol. 43:557-581.
Korcak, R.F. 1987. Iron deficiency chlorosis. Hort. Rev. 9:133-186.
515
Ku, C.S.M. and D.R. Hershey. 1991. Leachate electrical conductivity and growth of potted poinsettia with
leaching fractions of 0 to 0.4. J. Amer. Soc. Hort. Sci. 116:802-806.
Ku, C.S.M. and D.R. Hershey. 1992. Leachate electrical conductivity and growth of potted geranium with leach
ing fractions of 0 to 0.4. J. Amer. Soc. Hort. Sci. 117:893-897.
Lang, H.J., C.L. Rosenfield and D.W. Reed. 1990. Response of Ficus benjamina and Dracaena marginata to
iron stress. J. Amer. Soc. Hort. Sci. 115:589-592.
Langhans, R.W., ed. 1962. Snapdragons. NY State Ext. Serv., Cornell Univ., Ithaca, NY. 93 pp.
Langhans, R.W., ed. 1987. Roses. Roses, Inc., Haslett, MI. 372 pp.
Larson, R.A., ed. 1980. Introduction to Floriculture. Academic Press, New York. 607 pp.
Lawton, K.A., G.L. McDaniel and E.T. Graham. 1989. Nitrogen source and calcium supplement affect stem
strength of poinsettia. HortScience. 24:463-465.
Lee, C.W., J.M. Choi and C.H. Pak. 1996. Micronutrient toxicity in seed geranium {Pelargonium x hortorum
Bailey). J. Amer. Soc. Hort. Sci. 121:77-82.
Legg, J.O. and J.J. Meisinger. 1982. Soil nitrogen budgets. In Nitrogen in Agricultural Soils. F.J. Stevensen, ed.
No. 22, Amer. Soc. Agron., Madison, WI. 503-566.
Letey, J., W.M. Jarrell and N. Valoras. 1982. Nitrogen and water uptake and growth of plants at various
minimum solution nitrate concentrations. J. Plant Nutr. 5:73-89.
Lindsay, W.L. 1972a. Zinc in soils and plant nutrition. Adv. Agron. 24:147-186.
Lindsay, W.L. 1972b. Influence of the soil matrix on the availability of trace elements to plants. Ann. NY Acad.
Sci. 199:37-45.
Lindsay, W.L. 1974. Role of chelation in micronutrient availability. In The Plant Root and Its Environment. E.W.
Carson, ed. Univ. Press of VA., Charlottesville. 507-524.
Lindsay, W.L. 1978. Chemical reactions affecting the availability of micro-nutrients in soils. In Mineral Nutrition
of Legumes in Tropical and Subtropical Soils. C.S. Andrew and E.J. Kamprath, ed. Proc. Workshop CSIRO,
Brisbane, Australia.
Lindsay, W.L. 1979. Chemical Equilibria in Soils. J. Wiley & Sons, New York. 449 pp.
Lindsay, W.L. 1981. Soil chemistry. In McGraw-Hill Yearbook of Science and Technology. McGraw-Hill, New
York. 358-359.
Lindsay, W.L. 1983. Trace elements and chelation in fertilizers. 10th Ann. Rocky Mtn. Reg. Turfgrass Conf, CO
State Univ., Ft. Collins. 5 pp.
Lindsay, W.L. 1984. Soil and plant relationships associated with iron deficiency with emphasis on nutrient
interactions. J. Plant Nut. 7:489-500.
Lindsay, W.L. 1991. Inorganic equlibria affecting micronutrients in soils. In Micronutrients in Agriculture. J.J.
Mortvedt et al., eds. 2nd edition. No. 4. Soil Sci. Soc. Amer., Madison, WI. 89-112.
Lindsay, W.L. and H.A. Ajwa. 1994. Use of MINTEQA2 for teaching soil chemistry. Unpubl. Ms., CO State
Univ., Ft. Collins. 44 pp.
Lindsay, W.L. and F.R. Cox. 1985. Micronutrient soil testing for the tropics. Fertilizer Res. 7:169-200.
Lindsay, W.L. and J.D. DeMent. 1961. Effectiveness of some iron phosphates as sources of phosphorous for
plants. Plant and Soil. 14:118-126.
Lindsay, W.L. and E.C. Moreno. 1960. Phosphate phase equilibria in soils. Soil Sci. Soc. Amer. Proc. 24:177-
182.
Lindsay, W.L. and W.A. Norvell. 1969. Equilibrium relationships of Zn^^, Fe^^, Ca^^, and H^ with EDTA and
DTP A in soils. Soil Sci. Soc. Amer. Proc. 33:62-68.
Lindsay, W.L. and W.A. Norvell. 1978. Development of a DTP A soil test for zinc, iron, manganese and copper.
Soil Sci. Soc. Amer. Proc. 42:421-428.
Lindsay, W.L. and A.P. Schwab. 1982. The chemistry of iron in soils and its availability to plants. J. Plant
Nutrition. 5:821-840.
Lindsay, W.L. and G. Stanford. 1960. What happens to water-soluble phosphate in the soil? Crops and Soils.
12:2 .
Lindsay, W.L. and A.W. Taylor. 1960. Phosphate reaction products in soil and their availability to plants. 7th
Int. Cong. Soil Sci., Madison, WI. 3:580-589.
Lindsay, W.L., P.L.G. Vlek and S.H. Chien. 1989. Phosphate minerals. In Minerals in Soil Environments, 2nd
ed. SSSA Book Series No. 1. Soil Sci. Soc. Amer., Madison, WI. 1089-1130.
Lindsay, W.L., F.S. Watanabe and S.R. Olsen. 1963. Nutritional inter-relationships of phosphorous, zinc and
iron. Proc. 14th Ann. Fert. Conf. Pacific Northwest, Idaho Falls, ID. 109-115.
Loneragan, J.F. 1968. Nutrient requirement of plants. Nature. 220:1307-1308.
Loneragan, J.F. and K. Snowball. 1969a. Calcium requirements of plants. Austr. J. Agric. Res. 29:465-478.
Loneragan, J.F. and K. Snowball. 1969b. Rate of calcium absorption by plant roots and its relation to growth.
Austr. J. Agric. Res. 29:479-490.
Lunt, O.R. and J.J. Oertli. 1962. Controlled release of fertilizer minerals by incapsulating membranes: II.
Efficiency of recovery, influence of soil moisture, mode of application, and other considerations related to
use. Soil Sci. Soc. Amer. Proc. 26:584-587.
Maagistad, O.C. et al. 1943. Effect of salt concentration, kind of salt, and climate on plant growth in sand
cultures. Plant Physiol. 18:151-166.
Markus, D.K. 1986. Spurway/acid extraction procedures. HortScience. 21:217-222.
Markus, D.K. and J.E. Steckel. 1980. Periodical analysis of artificial rooting media and tomato leaf analysis from
New Jersey greenhouses. Acta Hort. 99:206-217.
Maronek, D.M., J.W. Hendrix and J. Kieman. 1981. Mycorrhizal fungi and their importance in horticultural crop
production. Hort. Rev. 3:172-213.
Marschner, H. and V. Romheld. 1983. In vivo measurement of root-induced pH changes at the soil-root interface:
Effect of plant species and nitrogen source. Z Pflanzenphysiol. 111:241-251.
Martens, D.C. and W.L. Lindsay. 1990. Testing soils for copper, iron, manganese and zinc. In Soil Testing and
Plant Analysis. Series No. 3. Soil Sci. Soc. Amer., Madison, WI.
Mastalerz, J.W., ed. Bedding Plants. PA Flower Growers, PA State Univ., University Park, PA. 516 pp.
Mastalerz, J.W. 1977. The Greenhouse Environment. Wiley, New York.
Mastalerz, J.W. and E.J. Holcomb, eds. 1982. Geraniums III. PA Flower Growers, PA State Univ., University
Park, PA. 410 pp.
Maxwell, M.K. 1972. Soilless culture -hydroponics. Occasional Paper No. 1. Dept, of Plant Sci., Hawkesbury
Agric. Col., Richmond, N.S.W., Australia. 10 pp.
Maynard, D.N. et al. 1976. Nitrate accumulation in vegetables. Adv. in Agron. 28:71-118.
Meiri, A. and A. Poljakoff-Mayber. 1967. The effect of chlorine salinity on growth of bean leaves in thickness
and in area. Israel J. Bot. 16:115-123.
Melton, R.R. and R.J. Dufault. 1991. Nitrogen, phosphorous and potassium fertility regimes affect tomato trans
plant growth. HortScience. 26:141-142.
Mengel, K. 1974a. Plant ionic status. In The Plant Root and Its Environment. E.W. Carson, ed. Univ. Press of
VA, Charlottesville. 63-81.
Mengel, K. 1974b. Ion uptake and translocation. In The Plant Root and Its Environment. E.W. Carson, ed. Univ.
Press of VA., Charlottesville. 83-100.
Mengel, K. and E.A. Kirkby. 1982. Principles of Plant Nutrition. International Potash Inst., Worblaufen-Bem.
655 pp.
Millar, R.O., D.C. Kiplinger and H.K. Tayama. 1981. Soluble salts: an overview of negative effects. Florists'
Rev. 168(4364): 10, 26-27.
Minotti, P.L., D.C. Williams and W.A. Jackson. 1969. Nitrate uptake by wheat as influenced by ammonium and
other cations. Crop Sci. 9:9-14.
Montgomery, G. 1984. A low-sodium diet. Florists’ Rev. 175(4528):52-54.
Moore, D.P. 1974. Physiological effects of pH on roots. In The Plant Root and Its Environment. E.W. Carson,
ed. Univ. Press of VA., Charlottesville. 135-151.
Moraghan, J.T. and H.J. Mascagni, Jr. 1991. Environmental soil factors affecting micronutrient deficiencies and
toxicities. In Micronutrients in Agriculture. J.J. Mortvedt et al., eds. No. 4. Soil Sci. Soc. Amer., Madison,
WI.
Mortvedt, J.J. 1991. Micronutrient fertilizer technology. In Micronutrients in Agriculture. J.J. Mortvedt et al.,
eds. 2nd edition, No. 4. Soil Sci. Soc. Amer., Madison, WI.
517
Moss, P. 1963. Some aspects of the cation status of soil moisture. Part I. : The ratio law and soil moisture
content. Plant and Soil 28:99-113.
Nelson, P.V. 1971. Copper deficiency in chrysanthemum: Critical level and symptoms. J. Amer. Soc. Hort. Sci.
96:661-663.
Nelson, P.V. 1978. Greenhouse Operation and Management. Reston Publ. Co., Reston, VA. 518 pp.
Nelson, P.V. and W.R. Faber. 1986. Bulk solution displacement. HortScience. 21:225-227.
Nelson, P.V. and Kuo-Hsien Hsieh. 1971. Ammonium toxicity in chrysanthemum. Soil Sci. and Plant Anal.
2:439-448.
Nieman, R.H. and L.L. Poulsen. 1967. Interactive effects of salinity and atmospheric humidity on the growth of
bean and cotton plants. Bot. Gaz. 128:69-73.
Nobel, P.S. 1991. Physicochemical and Environmental Plant Physiology. Academic Press, New York. 635 pp.
Noggle, J.C. 1966. Ionic balance and growth of sixteen plant species. Soil Sci. Soc. Amer. Proc. 30:763-766.
Norvell, W.A. 1970. Solubility of Fe^^ in soils. Ph.D. Dissertation. CO State Univ., Ft. Collins. 100 pp.
Norvell, W.A. 1991. Reactions of metal chelates in soils and nutrient solutions. In Micronutrients in Agriculture.
J.J. Mortvedt et al., eds. 2nd edition. No. 4. Soil Sci. Soc. Amer., Madison, WI.
Norvell, W.A. and W.L. Lindsay. 1969. Reactions of EDTA complexes of Fe, Zn, Mn, and Cu with soils. Soil
Sci. Soc. Amer. Proc. 33:86-91.
Nukaya, A., W. Voogt and C. Sonneveld. 1991. Effects of NO 3 , SO4 and Cl ratios on tomatoes grown in
recirculating system. Acta Hort. 294:297-304.
Oertli, J.J. and O.R. Lunt. 1962. Controlled release of fertilizer minerals by incapsulating membranes: I. Factors
influencing the rate of release. Soil Sci. Soc. Amer. Proc. 26:579-583.
Olsen, C. 1950. The significance of concentration for the rate of ion absorption by higher plants in water culture.
Physiol. Plant. 3:152-164.
Olsen, C. 1953a. The significance of concentration for the rate of ion absorption by higher plants in water culture.
II. Experiments with aquatic plants. Physiol. Plant. 6:837-843.
Olsen, C. 1953b. The significance of concentration for the rate of ion absorption by higher plants in water
culture. II. The importance of stirring. Physiol. Plant. 6:844-847.
Olsen, C. 1953c. The significance of concentration for the rate of ion absorption by higher plants in water culture.
IV. The influence of hydrogen ion concentration. Physiol. Plant. 6:848-858.
Parent, L.E. and M. Dafir. 1992. A theoretical concept of compositional nutrient diagnosis. J. Amer. Soc. Hort.
Sci. 117:239-242.
Peaslee, D.E. 1981. Soil phosphorous. In McGraw-Hill Yearbook of Science and Technology. McGraw-Hill,
New York. 359-361.
Peck, T.R. 1990. Soil testing: Past, present and future. Commun, in Soil Sci. Plant Anal. 21:1165-1186.
Peck, T.R. and P.N. Soltanpour. 1990. The principles of soil testing. In Soil Testing and Plant Analysis. R.L.
Westerman, ed. Soil Sci. Soc. Amer. No. 3. Soil Sci. Soc. Amer., Madison, WI.
Pétrucci, R.H. 1977. General Chemistry. 2nd ed. Macmillan Publ. Co., New York. 790 pp.
Poole, R.T., C.A. Connover and J.N. Joiner. 1976. Chemical composition of quality tropical foliage plants. Proc.
FL State Hort. Soc. 89:307-308.
Prasad, M., R.E. Widmer and R.R. Marshall. 1983. Soil testing of horticultural substrates for cyclamen and poin-
settia. Commun, in Soil Sci.and Plant Anal. 14:553-573.
Reddy, K.R. and W.H. Patrick, Jr. 1983. Effects of aeration on reactivity and mobility of soil constituents. In
Chemical Mobility and Reactivity in Soil Systems. D.W. Nelson et al., eds. Soil Sci. Soc. Amer. Special Publ.
11. Madison, WI. 262 pp.
Richards, L.A., ed. 1954. Diagnosis and Improvement of Saline and Alkali Soils. Agric. Handbook No. 60,
USDA, Washington, D.C. 160 pp.
Riley, D. and S.A. Barber. 1969. Bicarbonate accumulation and pH changes at the soybean {Glycine max (L.)
Merr.) root-soil interface. Soil Sci. Soc. Amer. Proc. 33:905-908.
Ritchie, G.S.P. 1989. The chemical behavior of aluminum, hydrogen and manganese in acid soils. In Soil Acidity
and Plant Growth. A.D. Robson, ed. Academic Press, New York. 306 pp.
Robson, A.D. and L.K. Abbott. 1989. The effect of soil acidity on microbial activity in soils. In Soil Acidity and
Plant Growth. A.D. Robson, ed. Academic Press, New York. 306 pp.
Roude, N., T.A. Nell and J.E. Barrett. 1991. Nitrogen source and concentration, growing medium, and cultivar
affect longevity of potted chrysanthemums. HortScience. 26:49-52.
Rovira, A.D. and C.B. Davey. 1974. Biology of the rhizosphere. In The Plant Root and Its Environment. E.W.
Carson, ed. Univ. Press of VA., Charlottesville. 153-204.
Russell, R.S. and V.M. Shorrocks. 1959. The relationship between transpiration and the absorption of inorganic
ions by intact plants. J. Expt. Bot. 10:201-316.
Sadasiviah, S.P. and W.D. Holley. 1973. Ion balance in nutrition of greenhouse roses. Roses, Inc. Bull.
(Supplement). Nov., 1973. 27 pp.
Sadiq, M. and W.L. Lindsay. 1989. Electron titration as a technique to study iron and manganese redox
transformation in soils. Soil Sci. 147:348-356.
Sajwan, K.S. and W.L. Lindsay. 1986. Effect of redox on zinc diffusion in paddy rice. Soil Sci. Soc. Amer.
Jour., 50:1264-1269.
Salisbury, F.B. and C.W. Ross. 1985. Plant Physiology. 3rd ed. Wadsworth Publ., Belmont, CA. 540 pp.
Sanchez, C.A., G.H. Snyder and H.W. Burdine. 1991. DRIS evaluation of the nutritional status of crisphead
lettuce. HortScience. 26:274-276.
Saranga, Y. et al. 1991. Breeding tomatoes for salt tolerance: Field evaluation of Ly copers icon germplasm for
yield and dry-matter production. J. Amer. Soc. Hort. Sci. 116:1067-1071.
Schekel, K.A. 1971. Effects of increased soluble salt concentrations on carnation growth. CO Flower Growers’
Assoc. Res. Bull. 251 :l-5.
Scher, F.M., M. Dupler and R. Baker. Effect of synthetic iron chelates on population densities of Fusarium
oxysporum and the biological control agent Pseudomonas putida in soil. Can. J. Microbiol. 30:1271-1275.
Schmuck, D. and K. Firth. 1987. Groundwater contamination raises liability question. Amer. Vegetable Grower.
35:14-15.
Schwab, A.P. and W.L. Lindsay. 1983a. Effect of redox on the solubility and availability of iron. Soil Sci. Soc.
Amer. J. 47:201-205.
Schwab, A.P. and W.L. Lindsay. 1983b. The effect of redox on the solubility and availability of manganese in
a calcareous soil. Soil Sci. Soc. Amer. J. 47:217-220.
Schwab, A.P. and W.L. Lindsay. 1989. A computer simulation of Fe(III) and Fe(II) complexation in nutrient
solution: I. Program development and testing. Soil Sci. Soc. Amer. Proc. 53:29-34.
Shimskey, D.S. 1988. Get the facts on groundwater. Greenhouse Grower. 6:54-62.
Shinohara, Y. and Y. Suzuki. 1988. Quality improvement of hydroponically grown leaf vegetables. Acta Hort.
230:279-286.
Sienko, M.J. and R.A. Plane. 1957. Chemistry. McGraw-Hill, New York. 621 pp.
Simonds, J.N., R. Swidler and H.M. Benedict. 1962. Absorption of chelated iron by soybean roots in nutrient
solutions. Plant Physiol. 37:460-466.
Slatyer, R.O. 1961. Effects of several osmotic substrates on the water relationships of tomato. Austr. J. Biol. Sci.
14:519-541.
Smilde, K.W. and J.P.N.L.R. van Eysinga. 1968. Nutritional Diseases in Glasshouse Tomatoes. Inst, voor
Bodemvrucht., Wageningen. 47 pp.
Smith, P.F. 1962. Mineral analysis of plant tissues. Ann. Rev. Plant Physiol. 13:81-108.
Soltanpour, P.N. and R.H. Follett. 1985. Soil test explanation. Service in Action No. 502. CO State Univ., Ft.
Collins.
Sonneveld, C. and E. Mook. 1983. Lettuce tipbum as related to the cation contents of different plant parts. Plant
and Soil. 75:29-40.
Sonneveld, C. and J. van Beusekom. 1974. The effect of saline irrigation water on some vegetables under glass.
Acta Hort. 35:75-86.
Sonneveld, C. and J. van den Ende. Soil analysis by means of a 1:2 volume extract. Plant and Soil. 35:505-516.
Sonneveld, C. and S.J. Voogt. 1978. Effects of saline irrigation water on glasshouse cucumbers. Plant and Soil.
49:595-606.
Sonnneveld, C. and W. Voogt. 1991. Effects of Ca-stress on blossom-end rot and Mg-deficiency in rockwool
grown tomato. Acta Hort. 294:81-88.
519
Sonneveld, C., J. van den Ende and P.A. van Dijk. 1974. Analysis of growing media by means of a 1:1 V2 volume
extract. Commun. in Soil Sci Plant Anal. 5:183-202.
Steiner, A. A. 1961. A universal method for preparing nutrient solutions of a certain desired composition. Plant
and S oil 15:134-154.
Steiner, A. A. 1966. The influence of the chemical composition of a nutrient solution on the production of tomato
plants. Plant and S oil 24:454-466.
Steiner, A.A. 1968. Soilless culture. Proc. 6th Colloq. Int. Potash Inst. Florence, Italy. 324-341.
Steiner, A.A. 1969. Recipe for a universal nutrient solution. Gostencild Verslag nr. 35. Vrentrum voor
Plantenfys. Oderz., Wageningen. 4 pp.
Stevenson, F.J. 1991. Organic matter-micronutrient reactions in soil. In Micronutrients in Agriculture. J.J.
Mortvedt et al., eds. 2nd edition, No. 4. Soil Sci. Soc. Amer., Madison, WI.
Sumner, M.E. 1979. Interpretation of foliar analyses for diagnostic purposes. Agron. J. 71:343-348.
Sumner, M.E. 1977a. Use of the DRIS system in foliar diagnosis of crops at high yield levels. Commun. in Soil
Sci. and Plant Analysis. 8:251-268.
Sumner, M.E. 1977b. Effect of com leaf sampled on N, P, K, Ca and Mg content and calculated DRIS indices.
Commun. in Soil Sci. and Plant Analysis. 8:269-280.
Sumner, M.E. 1977c. Application of Beaufil's diagnostic indices to maize data published in the literature
irrespective of age and conditions. Plant and Soil. 46:359-369.
Sumner, M.E. and M.P.W. Farina. 1986. Phosphorous interactions with other nutrients and lime in field cropping
systems. Adv. in Soil Sci. 5:201-236.
Takahashi, E. and Y. Miyake. 1982. The effects of silicon on the growth of cucumber plant. In Proc. 9th Intern.
Plant Nutrition Colloq., M Scaife ed.. Warwick Univ., U.K.
Terman, G.L. and L.A. Nelson. 1976. Comments on the use of multiple regression in plant analysis interpretation.
Agron. J. 68:148-150.
Tinus, R.W.and S.E. McDonald. 1979. How to grow tree seedlings in containers in greenhouses. General Tech.
Rpt.. RM-60. Rocky Mtn. Forest and Range Expt. Sta., USDA. Ft. Collins, CO. 256 pp.
Tjia, B., ed. 1984. Commercial Poinsettia Production in Florida. FL Cooper. Ext. Serv., Univ. of FL., Gainesville,
FL. 60 pp.
Twigg, M.C. and C.B. Link. 1951. Nutrient deficiency symptoms and leaf analysis of azaleas grown in sand
culture. Proc. Amer. Soc. Hort. Sci. 57:369-375.
Ulrich, A. 1960. Plant analysis in sugar beet nutrition. In Plant Analysis and Fertilizer Problems. W. Reuther,
ed. Pubi. 8 , Amer. Inst. Biol. Sci., Washington, D.C.
Ulrich, A. and F.J. Hill. 1973. Plant analysis as an aid in fertilizing sugar crops: Part 1, Sugar beets. In Soil Test
ing and Plant Analysis. L.M. Walsh and J.D. Beaton, eds. Soil Sci. Soc. Amer., Madison, WI.
Uren, N.C. and H.M. Reisenauer. 1988. The role of root exudates in nutrient acquisition. Adv. in Plant Nutrition.
B. Tinker and A. Lauchli eds. 3:79-114.
U.S. Dept, of Interior. 1962. Saline water conversion. U.S. Dept. Interior, Office of Saline Water. U.S. Gov.
Printing O ff, Washington, D.C. 32 pp.
U.S. Dept, of Interior. 1968. The A-B-Seas of desalting. U.S. Dept. Interior, Office of Saline Water. U.S. Gov.
Printing O ff, Washington, D.C.
U.S. Dept. Interior. 1979. Electrodialysis technology. U.S. Dept. Interior, Office of Water Res. and Tech. U.S.
Gov. Printing O ff, Washington, D.C. 15 pp.
U.S. Dept, of Interior. 1979. Reverse osmosis. U.S. Dept. Interior, Office of Water Res. and Tech., U.S. Gov.
Printing O ff, Washington, D.C.
van den Ende, J. 1971. Extraction methods for the determination of major elements in greenhouse soils and
potting and culture media. Acta Hort. 29:125-140.
van Eysinga, J.P.N.L.R. 1984. Nitrate and glasshouse vegetables. Fert. Res. 5:149-156.
Van Os, E.A., M.N.A. Ruijs and P.A. van Well. 1991. Closed business systems for less pollution from
greenhouses. Acta Hort. 294:49-57.
van Tuil, H.D.W. 1965. Organic salts in plants in relation to nutrition and growth. Agric. Res. Rpt. No. 657.
Wageningen.
Vlamis, J. and D .E. Williams. 1962. Ion competition in manganese uptake by barley plants. Plant Physiol 37:
650-655.
Wadleigh, C.H. 1957. Growth o f plants. In Soil, The Yearbook o f Agriculture 1957. A. Stefferud, ed. U .S.
Government Printing Office, Washington, D.C.
Wadleigh, C.H. and A.D. Ayers. 1945. Growth and biochemical composition of bean plants as conditioned by
soil moisture tension and salt concentration. Plant Physiol. 20:106-132.
Wadsworth, G.A. 1980. Methods for sampling peat growing bags. Acta Hort. 99:57-65.
Wallace, T. 1953. The Diagnosis of Mineral Deficiencies in Plants. Chemical Publ. Co., New York.
Walworth, J.L. and M.E. Sumner. 1987. The diagnosis and recommendation integrated system (DRIS). Adv. in
SoilSci. 6:149-188.
Walworth, J.L. and M.E. Sumner. 1988. Foliar diagnosis: A review. In Adv. in Plant Nutrition. B. Tinker and
A. Lauchi, eds. Volume 3, pp 193-241.
Walworth, J.L., W.S. Letzsch and M.E. Sumner. 1986. Use of boundary lines in establishing diagnostic norms.
SoilSci. Soc. Amer. J. 50:123-128.
Wamcke, D.D. 1986. Analyzing greenhouse growth media by the saturation extraction method. HortScience.
21:223-225.
Wamcke, D.D. 1990. Testing artificial growth media and interpreting the results. In Soil Testing and Plant
Analysis. R.L. Westerman, ed. Soil Sci. Soc. Amer. No. 3. Soil Sci. Soc. Amer., Madison, W I.
Warrick, A.W. 1981. Soil variability. In McGraw-Hill Yearbook of Science and Technology. McGraw-Hill, New
York. 359-363.
Watanabe, F.S., W.L. Lindsay and S.R. Olsen. 1965. Nutrient balance involving phosphorous, iron, and zinc.
Soil Sci. Soc. Amer. Proc. 29:562-565.
Waters, W.E. et al. 1973. The interpretation of soluble salt procedures as influenced by salinity testing procedure
and soil media. Proc. Tropical Region, Amer. Soc. Hort. Sci. 17:397-405.
Webb, R.A. 1972. Use of the boundary line in the analysis of biological data. J. Hort. Sci. 47:309-319.
Welch, R.M. et al. 1991, Geographic distribution of trace element problems. In Micronutrients in Agriculture.
J.J. Mortvedt et al., eds. 2nd edition. No. 4, Soil Sci. Soc. Amer. Madison, WI.
Weston, L.A. and B.H. Zandstra. 1989. Transplant age and N and P nutrition effects on growth and yield of
tomatoes. HortScience. 24:88-90.
White, J. 1987. Fertilization. In Roses. R. Langhans, ed. Roses, Inc., Haslett, MI. 372 pp.
White, J. 1976. Fertilization. In Bedding Plants. J.W. Mastalerz, ed. PA Flower Growers, University Park, PA.
516 pp.
White, R.A.J. and M. Prasad. 1980. Nutrient, salt and pH distribution in soil and in peat modules used for tomato
growing. Acta Hort. 99:167-178.
Whitesides, R. 1989. El Modeno Gardens: Innovative solutions to California's irrigation runoff restrictions.
Grower Talks. 52:28-36.
Williams, J.W., J.C. Peterson and J.D. Utzinger. 1988. Liming reactions in sphagnum peat-based growing media
J. Amer. Soc. Hort. Sci. 113:210-214.
Winsor, G.W. 1973. Nutrition. In The U.K. Tomato Manual. H.G. Kingham, ed. Grower Books, London. 223
PP-
Winsor, G.W. amd D.M. Massey. 1978. Some aspects of the nutrition of tomatoes grown in recirculating
solution. Acta Hort. 82:121-132.
Wittwer, S.H. and S. Honma. 1979. Greenhouse Tomatoes, Lettuce and Cucumbers. MI State Univ. Press., East
Lansing, MI. 225 pp.
Workman, S.M., P.N. Soltanpour and R.H. Follett. 1988. Soil testing methods used at Colorado State University
for the evaluation of fertility, salinity and trace element toxicity. CO State Univ. Agric. Expt. Sta. Tech. Bull.
LTB 8 8 -2 . 29 pp.
Wright, R.D. 1986. The pour-through nutrient extraction procedure. HortScience. 21:227-229.
Wright, R.D., K.L. Gmeber and C. Leda. 1990. Medium nutrient extraction with the pour-through and saturated
medium extract procedures for poinsettia. HortScience. 25:658-660.
Yelanich, M.V. and J.A. Biembaum. 1993. Root-medium nutrient concentration and growth of poinsettia at three
fertilizer concentrations and four leaching factors. J. Amer. Soc. Hort. Sci. 118:771-776.

Chapter 6. Nutrition.

Uploaded by

Copyright:

Available Formats

Chapter 6. Nutrition.

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 6. Nutrition.

Uploaded by

Copyright:

Available Formats

CH A PTER 6

Element Chemical Forms in solution available to Atomic

II. UNITS AND TERM INOLOGY

Multiply this material By this number To obtain this equivalent amount

KNO^ ^ + NO^ (6. 1)

CaC03 + 2H^ - + CO^g) -f (6.3)

And the equilibrium constant, K°, of this reaction at 25 C is:

(C a^^)(C 0,(g))(H ,0) ,9.74 (6.4)

log,o(Ca^O = 9.74 - 2pH - log,,(CO ,(g)) or

p H = -lo g ( i/0 = log-

+e ' -- Fe'^^ (6.9)

at least pe + pH would exceed 10. Lindsay [1991]

system also has the highest microbial activity compared

2. The Soil System

AT SOME FUTURE TIME K

ADSORBED IONS SOIL SOLUTION

Mineral Approx, formula Mineral Approx, formula

Characteristics Humid region______ Semiarid region

Exchangeable hydrogen and/or aluminum (meq/100 g) 4-6 1 -2

Cation exchange capacity (meq/100 g) 12-18 20-26

3. Expression for CEC

of the soil and is one of the easier tests to

G. MASS EXCHANGE EQUILIBRIA

logZ/1 ^" = 5.8 - 2pH (6.23)

Units Value Ways of writing Remarks

mho cm'‘ x 1 0 ’^ mho X 10’^ mho

SC (specific conductivity) Siemen x 10'^ 0.00001 Siemen Reported as such in Ire­

H. ELECTRICAL CONDUCTIVITY (EC)

Condecti¥ity ---- dedSiemens/m at 25® c.

Fe{OK)(soil) + 3H^ + e~ ^ Fe^ + 3H^O (6.24)

If Eq. 6.25 is expressed in logarithms, then:

logiFe^^) = 15.74 - (pe + pH) - 2pH (6.26)

III. NUTRIENTS IN THE PLANT AND TISSUE ANALYSIS

A. DEFINITIONS OF NUTRIENT REQUIREM ENTS

2% of the total production cost, failure by a grower

deficiencies are absent and

Fig. 6-14. Yield in grams

Sumner find that

1.0-2.0 0.35-0.65 4.5-6.5 4.0-6.5 0.2-1.0 Bunt, 1988 g.

Mesembryanthenum crinijlorum 4.84 4.04 0.39 Dight, 1977

® Critical levels for deficiency symptoms.

B. PUBLISHED NUTRIENT VALUES IN TISSUE (TABLE 6-12)

C. FACTORS AFFECTING PLANT NUTRIENT CONCENTRATIONS

7. Physiological age of the plant

9. Species and cultivar

D. PLANT NUTRIENT UPTAKE AND IONIC INTERACTIONS

L Systems for U ptake by Roots

in the external solution [Mengel, 1974a]. Clarkson

these high nutrient conditions, ion transfer

2. Ion Competition and Antagonism

dicotyledons. There can be sharp

3. Ammonium and N itrate Relationships

[e.g., Winsor and Massey, 1978;

4. Organic Anion Content

5. The Root and T ranspiration

E. NUTRIENT DISTRIBUTION AND TISSUE SAMPLING

Leaves Fruits Stems Roots

F. DIAGNOSIS FROM TISSUE SAMPLING

Crop Growth stage Part to sample

SC (specific conductivity) Siemen x 10'^ 0.00001 Siemen Reported as such in Ire

Chemical for 15 ¡iM 15 ¡iM 15 ¡iM Fe^^