Annals of Botany 97: 611621, 2006

doi:10.1093/aob/mcl010, available online at www.aob.oxfordjournals.org

Leaf Vascular Systems in C3 and C4 Grasses: A Two-dimensional Analysis

O S A M U U E N O 1,*, Y U K I K O K A W A N O 2, M A S A T A K A W A K A Y A M A 1, and T O M O S H I R O T A K E D A 2
1
Plant Physiology Department, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki, 305-8602, Japan and
2
Faculty of Agriculture, Kyushu University, Hakozaki, Fukuoka, 812-0053, Japan
Received: 13 September 2005 Returned for revision: 20 October 2005 Accepted: 6 December 2005 Published electronically: 7 February 2006

Background and Aims It is well documented that C4 grasses have a shorter distance between longitudinal veins in
the leaves than C3 grasses. In grass leaves, however, veins with different structures and functions are differentiated:
large longitudinal veins, small longitudinal veins and transverse veins. Thus, the densities of the three types of vein
in leaves of C3 and C4 grasses were investigated from a two-dimensional perspective.

Methods Vein densities in cleared leaves of 15 C3 and 26 C4 grasses representing different taxonomic groups and
photosynthetic subtypes were analysed.

Key Results The C4 grasses had denser transverse veins and denser small longitudinal veins than the C3 grasses
(19 and 21 times in interveinal distance), but there was no significant difference in large longitudinal veins. The
total length of the three vein types per unit area in the C4 grasses was 21 times that in the C3 grasses. The ratio of
transverse vein length to total vein length was 143 % in C3 grasses and 99 % in C4 grasses. The C3 grasses
generally had greater species variation in the vascular distances than the C4 grasses. The bambusoid and panicoid
C3 grasses tended to have a denser vascular system than the festucoid C3 grasses. There were no significant
differences in the interveinal distances of the three vein types between C4 subtypes, although the NADP-malic
enzyme grasses tended to have a shorter distance between small longitudinal veins than the NAD-malic enzyme
and phosphoenolpyruvate carboxykinase grasses.

Conclusions It seems that C4 grasses have structurally a superior photosynthate translocation and water distribution
system by developing denser networks of small longitudinal and transverse veins, while keeping a constant density
of large longitudinal veins. The bambusoid and panicoid C3 grasses have a vascular system that is more similar to
that in C4 grasses than to that in the festucoid C3 grasses.
Key words: C3 and C4 photosynthesis, interveinal distance, longitudinal vein, photosynthetic type, Poaceae, transverse vein.

INTROD UCTION
The leaves of C4 plants have different anatomical features
from those of C3 plants. Usually, C4 leaves are characterized
by Kranz-type anatomy, in which the vascular bundle is
surrounded by organelle-rich bundle sheath (BS) cells,
and this tissue layer is further surrounded by radially
arranged mesophyll (M) cells. In contrast, in C3 leaves,
the M cells are well developed relative to the BS cells,
which include only a few organelles (Dengler and
Nelson, 1999). In C4 photosynthesis, atmospheric CO2 is
initially fixed in the M cells, then decarboxylation and
refixation of CO2 occur in the BS cells (Hatch, 1987).
C4 plants are divided into three C4 subtypes differing in
the process of decarboxylation of C4 acids: the NADP-malic
enzyme (NADP-ME), NAD-malic enzyme (NAD-ME) and
phosphoenolpyruvate carboxykinase (PCK) types (Hatch,
1987). The difference in biochemical function is associated
with that in structural features of leaves. In C4 grasses, in
general, the NADP-ME grasses have the BS that originated
from the mestome sheath, whereas both the NAD-ME and
PCK grasses have the BS that originated from the parenchyma sheath (Dengler and Nelson, 1999). The BS cells of
the C4 subtypes also differ in the structure, intracellular

* For correspondence. E-mail [email protected]

Permanent address: Department of Biochemistry and Molecular
Biology, Saitama University, Saitama 338-8570, Japan

position and amount of the chloroplasts and mitochondria

(Hatch, 1987; Prendergast et al., 1987; Yoshimura et al.,
2004). The quantitative balance of photosynthetic tissues
(Hattersley, 1984; Ohsugi and Murata, 1986; Dengler et al.,
1994) and organelles (Yoshimura et al., 2004) between
the M and BS cells reflects the difference in biochemical
function of the photosynthetic subtypes.
In the early stage of C4 plant studies, it was reported that
C4 leaves have a denser vascular system than C3 leaves
(Takeda and Fukuyama, 1971). This is clearly seen in leaves
of grasses, which possess parallel venation. This distinctive
difference between C3 and C4 leaves is usually expressed as
a difference in the interveinal distance (distance between
vein centres) (Takeda and Fukuyama, 1971; Crookston and
Moss, 1974). For the efficient operation of C4 photosynthesis, a short distance between the M and BS cells is a
prerequisite for rapid diffusion of photosynthetic metabolites (Hatch, 1987). The proximity of veins may also be
needed for the proper expression of photosynthetic enzymes
in the M and BS cells (Langdale and Nelson, 1991; but see
Wakayama et al., 2003).
C4 plants have higher photosynthetic rates under high
irradiance and at high temperatures than C3 plants
(Ehleringer and Monson, 1993). In general, greater photosynthetic rates would result in greater rates of photosynthate
export, to remove recently formed photosynthate from
leaves rapidly and so avoid end-product inhibition of photosynthesis (Roth-Nebelsick et al., 2001). C4 plants have
been reported to show higher export rates of photosynthate

The Author 2006. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: [email protected]

612

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

than C3 plants (Hofstra and Nelson, 1969; Gallaher et al.,

1975; Lush, 1976; Grodzinski et al., 1998; Leonardos and
Grodzinski, 2000). In support of higher rates of translocation,
some researchers have pointed to the denser vascular system
(Crookston and Moss, 1974) and larger cross-sectional area
of phloem (Gallaher et al., 1975) in C4 leaves. The vascular
bundle is composed of two kinds of conducting tissues: the
xylem and phloem. Thus, it appears that C4 leaves have a
denser hydraulic network than C3 leaves.
The CO2-concentrating mechanism of the C4 pathway
gives C4 plants an efficient photosynthetic mechanism
under low stomatal conductance and thus a higher water
use efficiency and photosynthetic ability under environments of low water availability than is the case for C3 plants
(Ehleringer and Monson, 1993; Sage, 2004). It is generally
accepted that C4 plants evolved from C3 plants, accompanied by modifications to anatomical and biochemical
features of leaves. A change in vein density of leaves
undoubtedly occurred during the evolution from C3 to C4
plants (Sage, 2004; Ueno and Sentoku, 2006).
Leaf veins show a hierarchical order and have different
structures. In grass leaves, the differentiation of transverse
and longitudinal veins of different sizes allows a division
of labour (Yamazaki, 1960; Lush, 1976; Altus and Canny,
1982, 1985; Altus et al., 1985; Fritz et al., 1989). The large
longitudinal veins run from the leaf blade into the sheath.
However, most of the small longitudinal veins in the leaf
blade terminate at the junction of the blade and sheath
(Chonan et al., 1974; Colbert and Evert, 1982; Russell
and Evert, 1985; Dannenhoffer and Evert, 1994). The
large longitudinal veins serve primarily in longitudinal
transport of photosynthate outside the leaf blade. The
small longitudinal veins serve primarily in collecting photosynthate from nearby photosynthetic cells. The transverse
veins connect the longitudinal veins, and play important
roles in the lateral transport of photosynthate from the
small to the large longitudinal veins (Altus and Canny,
1982).
Water moves in the three types of vein opposite to the
direction of photosynthate. The network system for water
movement matches the structural demands of an efficient
irrigation system (Pelletier and Turcotte, 2000; RothNebelsick et al., 2001). Water absorbed in the root rises
through the large longitudinal veins from the leaf base to the
tip of the leaf blade. In the leaf blade, water moves laterally
from the large longitudinal veins via the transverse veins
to the small longitudinal veins, and is distributed to the M
or transpired from stomata (Altus and Canny, 1985;
Altus et al., 1985; Canny, 1990).
In order to relate the vascular architecture to physiological functions of the leaves, we need to consider the
localization and functional partitioning of different
types of veins. To our knowledge, however, most previous
studies of the vascular density of C3 and C4 grass leaves
have neglected vein types: only the distance between longitudinal veins has been measured (Takeda and Fukuyama,
1971; Crookston and Moss, 1974; Kawamitsu et al., 1985;
Dengler et al., 1994). Exceptionally, Oguro et al. (1985)
investigated the densities of transverse veins in leaves of
some Panicum species.

Here, we analysed the densities of the three types of

vein in leaves of various C3 and C4 grasses from a
two-dimensional perspective. This family includes all
three C4 subtypes and has been investigated sufficiently
for the photosynthetic types (Hattersley and Watson,
1992; GPWG, 2001). In addition, C3 grasses consist of
phylogenetically different groups such as the festucoid,
bambusoid and panicoid grasses, which differ in their
temperature requirements for growth. Thus, this family
provides an ideal subject for comparative analysis of leaf
vascular systems in plants that differ in photosynthetic
types and ecological characteristics.

MATERIA LS A ND METHODS
Plant materials

Table 1 lists the grass species examined in this study: 15 C3

species and 26 C4 species. We divided the C3 grasses into
two subgroups depending on growth and flowering period:
the panicoid and bambusoid C3 grasses and the festucoid
C3 grasses. The bambusoid and panicoid C3 grasses grow
in summer, and the flowering period is late summer to early
autumn. The festucoid C3 grasses grow in spring, and
the flowering period is late spring. Taxonomically, the
panicoid, bambusoid and festucoid grasses belong to the
subfamilies Panicoideae, Bambusoideae and Pooideae,
respectively (Clayton and Renvoize, 1992). The C4 species
were divided into three subgroups depending on the C4
biochemical subtypes: eight NADP-ME-type species, nine
NAD-ME-type species and nine PCK-type species. All
NADP-ME-type grasses, the three Panicum species in the
NAD-ME-type grasses, and the four Brachiaria species,
Panicum maximum and Urochloa texana in the PCK-type
grasses belong to the subfamily Panicoideae. The other
grasses of the NAD-ME and PCK types belong to the
subfamily Chloridoideae (Clayton and Renvoize, 1992).
Seeds of 13 species (four Brachiaria species, Chloris
gayana, Eleusine coracana, Oryza sativa, Panicum coloratum, P. dichotomiflorum, P. maximum, P. miliaceum,
Sorghum sudanense and Urochloa texana) were sown in
pots filled with fertilized field soil, and plants were raised
outdoors in Tsukuba and Fukuoka, Japan, in summer.
The plants were watered daily. Plants of the remaining
29 species growing naturally in the field in Fukuoka and
Tsukuba were used.
Leaf samples were collected in May for the festucoid C3
grasses and in August to September for the panicoid and
bumbsoid C3 grasses and all C4 grasses. No significant
differences in the interveinal distances occur among different leaf positions in rice plants, except that those in the
primary leaves are somewhat smaller (Yamazaki, 1963).
In barley leaves, the flag leaf and the first and second leaves
below it tend to have similar leaf vein densities (Hanson and
Rasmusson, 1975). Therefore, we used either flag leaves or
the first leaves below the flag leaves for our experiments.
Leaf blades were cut and immediately fixed in a mixture
of formaldehyde, acetic acid and ethanol in water (FAA).
The plants collected in the field were also retained as
voucher specimens for exact identification.

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

613

T A B L E 1. The C3 and C4 grass species examined in this study, which are divided according to photosynthetic types and
phylogenetic groups
NADP-ME

NAD-ME

PCK

Panicoid
Panicum coloratum L. var. makarikariense
Goossens
P. dichotomiflorum Michaux
P. miliaceum L.
Chloridoid
Cynodon dactylon (L.) Persoon
Eleusine coracana (L.) Gaertner

Panicoid
Brachiaria brizantha (Hochst. Ex A. Rich)
Stapf
B. decumbens Stapf
B. humidicola (Rendle) Schweick.
B. mutica (Forsk.) Stapf
Panicum maximum Jacq.
Urochloa texana (Buckley) Webster

E. indica (L.) Gaertner

Eragrostis cilianensis (Allioni) Vignolo-Lutati
E. ferruginea (Thunb.) P. Beauv.
Leptochloa chinensis (L.) Nees

Chloridoid
Chloris gayana Kunth
Sporobolus indicus R. Br. var. purpureosuffusus (Ohwi) T. Koyama
Zoysia tenuifolia Willd.

Bambusoid

Panicoid

Festucoid

(b) C3 species
Leersia japonica Makino

Hymenachne indica Buse

Agropyron tsukushiense (Honda) Ohwi var.

transiens (Hackel) Ohwi
Alopecurus aequalis Sobol. var. amurensis
(Komar.) Ohwi
Avena fatua L.
Beckmannia syzigachne (Steud.) Fernald
Briza minor L.
Bromus catharticus Vahl
B. rigidus Roth
Dactylis glomerata L.
Lolium multiflorum Lam.
Poa acroleuca Steud.

(a) C4 species
Panicoid
Digitaria sanguinalis (L.) Scopoli
D. violascens Link
Echinochloa crus-galli P. Beauv.
Paspalum distichum L.
Setaria glauca (L.) P. Beauv.
S. viridis (L.) P. Beauv. var minor (Thunb.)
Ohwi
Sorghum sudanense Stapf
Spodiopogon cotulifer (Thunb.) Hackel

Oryza sativa L. cv. Reiho

Isachne globosa O. Kuntze

Panicum bisulcatum Thunberg

Preparation of cleared leaves

Cleared leaf blades were prepared by a method described

in Ueno (1995). The middle portions of fixed leaf blades
were boiled in 70 % ethanol for about 1020 min. After
washing in distilled water several times, they were transferred to boiling 85 % lactic acid for 20 min, and then stored
in chloral hydrate-saturated ethanol before analysis. The
leaf vasculature was observed without staining under a
light microscope.
Quantitative data of leaf vascular systems

Leaf veins were divided into three types: large longitudinal veins, small longitudinal veins and transverse
veins (Chonan et al., 1974). The two types of longitudinal
vein were distinguished by diameter in paradermal view
under the light microscope (Fig. 1).
The distances between small longitudinal veins, between
large longitudinal veins and between transverse veins were
represented by means of 30 measurements of middle portions of 36 leaf blades taken from three plants. The distance between longitudinal veins was measured between the
centres of adjacent veins. The transverse veins are usually
curved, unlike the parallel longitudinal veins (Fig. 1). For
the distance between transverse veins, therefore, the mean
of the minimum and maximum distances between adjacent
transverse veins running between a pair of longitudinal veins

was calculated. The distance between small longitudinal

veins was multiplied by that between transverse veins for
each species to indicate areolar area, which represents the
minimum area of photosynthetic tissue surrounded by veins.
The number of transverse veins per unit leaf area was
measured on photomicrographs (70) obtained from the
middle portions of the 36 leaf blades. The lengths of
longitudinal and transverse veins per unit leaf area were
measured with a curvimeter on the same photomicrographs.
The total vein length per unit leaf area and the ratio of
transverse vein length to total vein length were calculated
using these values.
Statistical analysis

We tested the significance at P < 005 of any differences

in mean values generated for each species between the C4
and C3 groups and between the five subgroups (the NADPME, the NAD-ME, the PCK, the C3 bambusoids and
panicoids, and the C3 festucoids) using one-way analysis
of variance (ANOVA) with Tukeys honestly significant
difference (HSD) test (statistical software R 2.1.0,
R Foundation of Statistical Computing, 2005). To test
phylogenetic differences in the C4 grasses, we also analysed any differences between the C4 panicoids, the C4
chloridioids, the C3 bambusoids and panicoids, and the
C3 festucoids using the same statistical test.

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

614

LLV

LLV

LLV

LLV

TV

TV
SLV

SLV

LLV

LLV
SLV

LLV
TV

TV

LLV

SLV

F I G . 1. Paradermal view of cleared leaf blades of C3 and C4 grasses. (A) Digitaria sanguinalis, an NADP-ME C4 species. (B) Eleusine indica, an NAD-ME
C4 species. (C) Panicum bisulcatum, a panicoid C3 species. (D) Briza minor, a festucoid C3 species. LLV, large longitudinal vein; SLV, small longitudinal
vein; TV, transverse vein. The magnification of the four photomicrographs is the same. Scale bars = 250 mm.

RESULTS
General features of leaf vascular system in grasses

In the paradermal view of cleared leaf blades, longitudinal

veins run at regular intervals in parallel (Fig. 1). The size

difference and the existence of strands of hypodermal

sclerenchyma allowed the large longitudinal veins to be
distinguished from the small longitudinal veins. The transverse veins connected two adjacent longitudinal veins,
irrespective of size. The pattern of connection varied even

615

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

Distance between small LVs (m)

y = 04x

C4 (NADP-ME)

500

C4 (NAD-ME)
y = 02x

C4 (PCK)
C3 (bambusoids

400

and panicoids)
C3 (festucoids)
300

y = 01x

200

100

0
0

500

1000

1500

2000

2500

Distance between large LVs (m)

F I G . 2. Relationship of the distance between large longitudinal veins and that between small longitudinal veins in leaf blades of C3 and C4 grasses. Slopes
of dotted lines show the ratio of the distance between small longitudinal veins (y) to that between large longitudinal veins (x). LV, longitudinal vein.

within a leaf: at a variety of angles and from straight to

curved.
Interveinal distances of longitudinal and transverse veins

Figure 2 shows the relationship between large longitudinal veins and small longitudinal veins in the C3 and C4
grasses. The C4 grasses had a significantly shorter distance
between small longitudinal veins than the C3 grasses: the
mean value of the C3 grasses was 22 times that of the
C4 grasses (Table 2; P < 005). Within the C4 grasses, the
distance between small longitudinal veins in the NADP-ME
grasses was shorter than those in the NAD-ME and PCK
grasses (Table 2), although the difference was not significant. Within the C3 grasses, there was a significant difference
between the bambusoid and panicoid C3 grasses and the
festucoid C3 grasses (Table 2). There was also a significant
difference between the festucoid C3 grasses and the three
C4 groups. When the C4 grasses were compared between
the phylogenetic groups, there was no significant difference
between the panicoid C4 and the chloridoid C4 grasses
(Table 2).
With respect to the distance between large longitudinal
veins (Fig. 2), the mean value of the C3 grasses was 11
times that of the C4 grasses, but this difference was not
significant (Table 2). Likewise, there were no significant
differences between the five subgroups (Table 2). No
correlation was found between the density of large longitudinal veins and that of small longitudinal veins in the
C3 grasses (r = 0045, NS). Low positive correlations were
found between the two densities in all grasses examined
(r = 0319, P < 005) and in the C4 grasses (r = 0526,
P < 001). On the other hand, there were high positive
correlations between the two densities in the NADP-ME

(r = 0773, P < 005), NAD-ME (r = 0794, P < 005)

and PCK (r = 0864, P < 001) grasses. The ratio of the
distance between small longitudinal veins to that between
large longitudinal veins tended to be lower in C4 grasses
than in C3 grasses (Fig. 2). Between the panicoid and
chloridoid C4 grasses, there was a tendency that chloridoid
C4 grasses had a somewhat shorter distance between
large longitudinal veins than the panicoid C4 grasses
(Table 2).
Figure 3 shows the relationship between transverse veins
and small longitudinal veins in the C3 and C4 grasses. There
was a significant difference in the distance between transverse veins between the C3 and C4 grasses: the mean value
of the C3 grasses was 19 times that of the C4 grasses
(Table 2; P < 005). The only significant difference between
subgroups was that between the festucoid C3 grasses and the
other four subgroups as a whole (Table 2); the mean value
was 20 times that of the bambusoid and panicoid C3 grasses
(Table 2). The values of all the bambusoid and panicoid C3
grasses except Hymenachne indica (1491 mm) were comparable with those of C4 grasses (Fig. 3). The bambusoid C3
grass O. sativa showed the shortest distance between transverse veins (481 mm) among all grass species examined
(Fig. 3). The festucoid C3 grass Bromus rigidus showed
the greatest distance (2574 mm; Fig. 3). There was a positive
correlation between the distance between small longitudinal
veins and that between transverse veins in all grasses (Fig. 3;
r = 0744, P < 005). There was no significant difference in
the distance between transverse veins in the panicoid C4 and
the chloridoid C4 grasses (Table 2).
The mean value of the areolar area (the minimum area
of photosynthetic tissue surrounded by veins) of the C3
grasses was 43 times that of the C4 grasses (Table 2;
P < 005). In the five subgroups, the festucoid C3 grasses

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

616

T A B L E 2. Means of various measures of vascular density in the leaf blades of grasses of different groups and subgroups
Distance
Distance
No. of between small between large
species LVs (A) (mm) LVs (mm)

Group and
subgroup
C4
C3

26
15

NADP-ME
NAD-ME
PCK
C3 bambusoids and panicoids
C3 festucoids
C4 panicoids
C4 chloridoids
C3 bambusoids and panicoids
C3 festucoids

8
9
9
5
10
17
9
5
10

118.5
255.5
86.3
138.2
127.3
195.0
285.8
115.4
124.3
195.0
285.8

6 6.4a
6 18.5b
6 4.3a
6 7.9ab
6 11.0a
6 9.4b
6 21.8c
6 8.6a
6 9.4a
6 9.4b
6 21.8c

938.4
1059.6
960.5
840.2
1017.0
1037.2
1070.8
1049.8
728.0
1037.2
1070.8

6 62.2a
6 54.5a
6 48.8a
6 106.0a
6 140.8a
6 82.7a
6 73.3a
6 75.7ab
6 69.8ac
6 82.7abc
6 73.3ab

Total vein
Areolar area length per
(A B)
unit leaf area
(103 mm2) (mm mm2)

Distance
between TVs
(B) (mm)
765.6
1460.0
761.6
725.9
808.8
883.6
1748.0
782.2
734.1
883.6
1748.0

6 29.8a
6 154.5b
6 54.1a
6 30.6a
6 66.0a
6 168.5a
6 147.1b
6 42.6a
6 31.1a
6 168.5a
6 147.1b

91.5
395.3
64.3
100.8
105.1
176.5
504.7
90.6
92.0
176.5
504.7

6 6.5a 10.6
6 56.1b 5.1
6 2.5a 14.3
6 7.7a
8.9
6 14.4a z8.7
6 42.3a 6.1
6 54.3b 4.5
6 9.1a 11.2
6 8.8a z9.2
6 42.3a 6.1
6 54.3b 4.5

6 0.6a
6 0.3b
6 0.7a
6 0.4b
6 0.8bc
6 0.3cd
6 0.3d
6 0.9ab
6 0.6abc
6 0.3bcd
6 0.3cd

Ratio of
TV length
to total vein
length (%)

9.9
14.3
7.2
10.6
z .
11 8
15.0
14.0
9.8
z .
10 2
15.0
14.0

6 0.6a
6 1.1b
6 1.4a
6 2.4ab
6 1.4ab
6 1.9b
6 1.4b
6 0.9ab
6 0.9abc
6 1.9ac
6 1.4ac

No. of TV
per unit
leaf area
(mm2)
9.8
3.0
13.0
8.0
8.8
5.6
1.8
10.2
9.1
5.6
1.8

6 0.7a
6 0.6b
6 2.6a
6 1.7b
6 1.2b
6 0.9b
6 0.3c
6 0.9ab
6 0.9abc
6 0.9bcd
6 0.3cd

Values are given as mean 6 s.e.

n = 25 species; z n = 8 species.
Values followed by the same letter are not significantly different at P < 005.
LV, longitudinal vein; TV, transverse vein.

Distance between small LVs (m)

y = 04x

C4 (NADP-ME)

500

C4 (NAD-ME)
C4 (PCK)

y = 02x

C3 (bambusoids

400

and panicoids)
C3 (festucoids)

Br

300
Os

y = 01x

200
Hi
100

0
0

500

1000

1500

2000

2500

Distance between TVs (m)

F I G . 3. Relationship of the distance between transverse veins and that between small longitudinal veins in leaf blades of C3 and C4 grasses. Slopes of dotted
lines show the ratio of the distance between small longitudinal veins (y) to that between transverse veins (x). LV, longitudinal vein; TV, transverse vein;
Br, Bromus rigidus; Hi, Hymenachne indica; Os, Oryza sativa.

had a significantly higher value than the other four subgroups (Table 2). Exceptionally, the festucoid C3 grass
Dactylis glomerata showed a low value (170
103 mm2), whereas the panicoid C3 grass Hymenachne
indica showed a high value (337 103 mm2; Fig. 4).
Although there was no statistical difference between the
other four subgroups, the mean value was lowest in the
NADP-ME grasses and highest in the bambusoid and
panicoid C3 grasses (Table 2). There was no significant
difference in the mean value between the panicoid C4
and the chloridoid C4 grasses (Table 2).

Vein densities per unit leaf area

Several measures of leaf vein densities were calculated

(Table 2). The mean value of total vein length per unit leaf
area in C4 grasses was 21 times that in C3 grasses (Table 2).
In the five subgroups, it was significantly higher in the
NADP-ME grasses than in the other four subgroups
(Table 2; Fig. 5). There was no significant difference
between the NAD-ME and PCK grasses, between the PCK
and the bambusoid and panicoid C3 grasses, or between the
bambusoid and panicoid C3 and the festucoid C3 grasses
(Table 2). There was no significant difference in the mean

617

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

600

Hi

400

a
200
a

a
0

Dg

C4
C4
NADP-ME NAD-ME

C4
PCK

C3
C3
bambusoids festucoids
and panicoids

F I G . 4. Comparison of the areolar area (the minimum photosynthetic tissue

area surrounded by veins) in leaf blades of C3 and C4 grasses. The values
were calculated from the distances between small longitudinal veins and
the distances between transverse veins. The mean and s.e. are shown for
the respective subgroups. Values followed by the same lower case letter
are not significantly different at P < 005. Dg, Dactylis glomerata; Hi,
Hymenachne indica.

value between the panicoid C4 and the chloridoid C4 grasses

(Table 2).
The ratio of transverse vein length to total vein length was
14 times higher in the C3 grasses than in the C4 grasses
(Table 2). There was no significant difference in the mean
values between the three C4 subgroups, although the
mean value in the NADP-ME grasses was significantly
lower than those in the two C3 subgroups (Table 2;
Fig. 6). There was no significant difference in the mean
value between the panicoid C4 and the chloridoid C4 grasses
(Table 2).
The number of transverse veins per unit leaf area in the C4
grasses was 33 times that in the C3 grasses (Table 2). The
mean value of the NADP-ME grasses was significantly
higher than the other five subgroups, and that of the festucoid C3 grasses was significantly the lowest (Table 2; Fig. 7).
There was no large difference in the mean value between
the panicoid C4 and the chloridoid C4 grasses (Table 2).

DISCUSSION
Functional implications of differences of leaf vascular
systems between C3 and C4 grasses

Our study confirmed that C4 grasses have a denser system

of small longitudinal veins than C3 grasses, as reported
in previous work (e.g. Takeda and Fukuyama, 1971;
Crookston and Moss, 1974; Kawamitsu et al., 1985;
Dengler et al., 1994). Our study is the first to demonstrate
that the statistically significant differences observed
between C3 and C4 grasses are due to the differences in

Total vein length per unit leaf area (mm mm2)

Areolar area (103mm2)

800
a
15

10

b
c
c
d
d

C4
C4
NADP-ME NAD-ME

C4
PCK

C3
C3
bambusoids festucoids
and panicoids

F I G . 5. Comparison of total vein length per unit leaf area in leaf blades
of C3 and C4 grasses. The mean and s.e. are shown for the respective
subgroups. Values followed by the same lower case letter are not
significantly different at P < 0.05.

small longitudinal veins, not large longitudinal veins.

Furthermore, this study demonstrates that the leaves of
C4 grasses tend to develop transverse veins more densely
than do those of C3 grasses, although some C3 grasses
also had a short distance between transverse veins. Oguro
et al. (1985) reported that in some Panicum species, C4
species tend to have shorter distance between transverse
veins than C3 species.
The large longitudinal veins contain a greater crosssectional area of phloem than the small longitudinal
veins (Altus and Canny, 1982; Colbert and Evert, 1982;
Russell and Evert, 1985). The small longitudinal vein has
about the same phloem area throughout the length of
the blade, whereas the phloem area of a large longitudinal
vein increases from the tip to the base (Altus and Canny,
1982). The transverse veins, which have a single sieve
tube (Kuo et al., 1972; Chonan et al., 1985; Tiba and
Frean, 1989), connect the longitudinal veins. The transverse
veins and small longitudinal veins play a vital role in
lateral transport of photosynthate from the small to the
large longitudinal veins (Altus and Canny, 1982). Thus,
denser development of transverse veins will probably be
effective for movement of photosynthate from the small
to the large longitudinal veins. Our study demonstrates
that C4 grasses have acquired a superior photosynthate
translocation system by developing denser networks of
small longitudinal and transverse veins, while keeping a
constant density of large longitudinal veins.
This feature is also shown by the small areolar area. If
we attempt to correlate structural features of leaves to
photosynthate translocation, the leaf thickness should also
be considered, because the volume of photosynthetic tissue
would be more strongly involved in translocation than the

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

618

20
b
b
15

a
b
a
b

Number of TV per unit leaf area (mm2)

Ratio of TV length to total vein length (%)

25
15

10
b

b
5
c

10
a

0
C4
C4
NADP-ME NAD-ME

C4
PCK

C3
C3
bambusoids festucoids
and panicoids

F I G . 7. Comparison of the number of transverse veins per unit leaf area

in leaf blades of C3 and C4 grasses. The mean and s.e. are shown for the
respective subgroups. Values followed by the same lower case letter are
not significantly different at P < 005.

C4
C4
NADP-ME NAD-ME

C4
PCK

C3
C3
bambusoids festucoids
and panicoids

F I G . 6. Comparison of the ratio of transverse vein length to total vein

length per unit leaf area in leaf blades of C3 and C4 grasses. The mean
and s.e. are shown for the respective subgroups. Values followed by the
same lower case letter are not significantly different at P < 005.

areolar area. Dengler et al. (1994) reported that the leaves

of C4 grasses are significantly thinner than those of C3
grasses, if C3 bambusoid grasses, which have the thinnest
leaves, are not included. Within C4 grasses, there were no
significant differences in the leaf thickness between NADPME, NAD-ME and PCK grasses, but NAD-ME C4 grasses
with PCK-like leaf anatomy have thinner leaves than other
C4 subtypes (Dengler et al., 1994). In the NAD-ME C4
grasses examined in our study, only P. dichotomiflorum
has this PCK-like anatomy (Ohsugi and Murata, 1986).
Thus, the festucoid C3 grasses would have greater photosynthetic tissue volume than the C4 grasses, because they
have thick leaves with high areolar area.
It should also be noted that biochemical mechanisms
such as sugar transport (Williams et al., 2000), plasmodesmatal distribution in the BS cells (Botha, 1992) and
the architecture of the conducting tissue are involved in
the process of photosynthate translocation. It has been
reported that C4 species usually show preferential localization of sucrose phosphate synthase in the M cells,
although there is a wide variation in the relative abundance
between the M and BS cells among C4 species (Lunn and
Furbank, 1999). Thus, trioses produced in the BS cells
must move first to the M cells before being converted to
sucrose. Newly formed sucrose must then be returned to

the BS cells and loaded to the phloem, suggesting a more

complicated pathway of photosynthates in C4 species as
compared with that in C3 species.
Reflecting the different conducting roles, the large longitudinal veins contain vessels of larger diameter than the
small longitudinal veins (Kuo et al., 1974). The diameter of
the largest vessel in the large longitudinal veins decreases
with distance along the blade toward the tip, but that in the
small longitudinal veins does not change (Altus et al.,
1985). The transverse veins possess only one vessel of similar diameter at all places in the leaf blade (Altus et al.,
1985). It appears that the leaves of C4 grasses have structurally a better water distribution system than those of
festucoid C3 grasses by developing denser networks of
the small longitudinal and transverse veins.
Leaf vascular system in C4 grasses

Kawamitsu et al. (1985) reported that the C4 grasses

with BS cells that originated from mestome sheath cells
(NADP-ME type) have a shorter interveinal distance in
leaves than those with BS cells that originated from
parenchyma sheath cells (NAD-ME and PCK types).
They found no difference in the interveinal distance
between the NAD-ME and PCK grasses. Other workers
have reported the following order for the interveinal distance: NADP-ME < PCK < NAD-ME in Panicum C4 grasses
(Ohsugi and Murata, 1986) and NADP-ME = PCK <
NAD-ME in more divergent C4 grasses (Dengler et al.,
1994). We found no significant differences in the distance
between small longitudinal veins between the three C4
subgroups, although the mean value of the NADP-ME
grasses was lower than those of the NAD-ME and PCK

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

grasses. Thus, our data roughly correspond to the results of
Kawamitsu et al. (1985), although the previous study did
not distinguish the two types of longitudinal vein. It is
unclear whether the features of the vascular system of C4
subgroups are associated with some biochemical functions
of photosynthesis, although it has been suggested that the
interveinal distance could affect photosynthesis in C4
grasses, resulting in differences in photon capture (quantum
yield) (Ehleringer et al., 1997; Ogle, 2003). Fisher and
Evert (1982) reported a vein density of 797 mm mm2
for Amaranthus retroflexus, an NAD-ME C4 dicot having
a reticulated vascular system. This value is comparable with
those of the NAD-ME C4 grasses we examined.
When the measures of vascular density of the C4 grasses
were compared between the panicoid C4 and the chloridoid
C4 grasses, there were no significant differences, except
that the chloridoid C4 grasses had a somewhat shorter
distance between large longitudinal veins than the panicoid
C4 grasses. Thus, it seems that a similar change in the leaf
vascular system occurred in parallel between the two phylogenetic lines, although further studies would be required
for the density of large longitudinal veins.
Leaf vascular system in C3 grasses

Considerable differences in leaf vascular systems within

the C3 grasses were revealed. The bambusoid and panicoid
C3 grasses generally had a shorter distance between small
longitudinal veins than the festucoid C3 grasses, but there
were no differences in the distance between large longitudinal veins. The distance between transverse veins was generally shorter in the bambusoid and panicoid C3 grasses than
in the festucoid C3 grasses. Thus, the bambusoid and panicoid C3 grasses have leaves with a denser vascular system
than the festucoid C3 grasses. The two C3 subgroups differ
in the seasonal growth pattern: the former subgroup grows
in summer, as do C4 grasses, but the latter subgroup grows
in spring. All the C3 grasses we examined grow in sunny
habitats; however, the bambusoid and panicoid C3 grasses
grow in wet habitats, whereas the festucoid C3 grasses grow
in mesic habitats. It remains unknown whether the gas
exchange and water physiology of leaves differ between
the two C3 subgroups. Interestingly, the festucoid C3 grasses
tend to have larger stomata in the leaf blade than the
bambusoid and panicoid C3 grasses, but a lower density
of stomata (Ashida and Sugino, 1984; Kawamitsu et al.,
1996). For instance, O. sativa leaves show the highest
density of stomata among grass species, but the stomatal
size is very small (Kawamitsu et al., 1996). Such C3 grasses,
even though they grow in wet places, may have a high
evaporative demand in order to lower the leaf temperature
in the heat of the day. It has also been reported that even
in paddy fields with enough water, rice plants are often
subject to water stress at midday on a fine day because
of intense transpiration over water absorption from roots,
which is caused by a high vapour pressure deficit, accompanied by a decrease in leaf water potential and stomatal
closure (Ishihara and Hirasawa, 1978; Ishihara and Saitoh,
1987). It would be interesting to study whether the two
C3 subgroups have developed different strategies in the

619

vascular and stomatal architectures for water movement

and transpiration in leaves.
Ecological and evolutionary implications of leaf vascular
systems in C3 and C4 grasses

The vein density of leaves is influenced by various environmental factors. One of the factors that lead to higher vein
densities is reduction of soil water availability. Likewise,
high temperature induces a similar response in leaves (Uhl
and Mosbrugger, 1999; Roth-Nebelsick et al., 2001). These
environmental conditions are generally advantageous to
the performance of C4 plants, because C4 photosynthesis
is more efficient than C3 photosynthesis under environments
that promote photorespiration (Ehleringer and Monson,
1993). However, it seems that a change in atmospheric
CO2 concentration has no significant effect on the vein
density of leaves in both the short and long term (Uhl
and Mosbrugger, 1999), unlike the response of stomatal
density, which declines as the CO2 concentration increases
(Woodward, 1987), although more data are needed to understand the effect of CO2 concentration on the vein density
(Roth-Nebelsick et al., 2001). An increase in the vein density of leaves might be an anatomical pre-conditioning to
the evolution of C4 plants from C3 plants (Sage, 2004).
Increasing vein density may initially have little effect on
the performance of an effective CO2 concentration mechanism, but may enhance the water status of leaves in hot
environments (Sage, 2004). Kocacinar and Sage (2003)
have found that C4 dicots have a stem xylem structure
and hydraulic function differing from that of C3 dicots,
reflecting their greater water use efficiency and lower
water requirements. It is unknown whether C4 grasses
also have such characteristics.
Recent molecular phylogenetic studies on the grasses
have demonstrated that C4 photosynthesis originated multiple times among several closely related subfamilies
(Kellogg, 2001). The earliest divergent branches in the
grasses are the C3 bambusoids and C3 festucoids. The
remainder of the family is in a large clade (the PACC
clade) with a mix of C3 and C4 members. It includes the
panicoids, the chloridoids, a lineage with Aristida and
Stipagrostis, and a lineage with Eriachne (Sinha and
Kellogg, 1996; Kellogg, 2001). Our study indicates that
the festucoids have a sparse leaf vascular system, whereas
a dense vascular system occurs in the two lineages, the
panicoids and chloridoids, together with evolution of C4
photosynthesis. A recent molecular phylogenetic study on
the Panicoideae has demonstrated that C3 photosynthesis
is the ancestral condition in this subfamily and that C4
photosynthesis arose at least eight times (Giussani et al.,
2001). It is interesting to note that the panicoid C3 grasses
have a vascular system that is more similar to that in
C4 grasses than to that in the festucoid C3 grasses. The
bambusoid grasses we examined also had a relatively
dense leaf vascular system, which is comparable with that
of the panicoid C3 grasses. The acquisition of this structural
characteristic may partly be related to similar ecological
features in these two groups. However, C4 grasses with a
denser leaf vascular system evolved within the panicoids

620

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

but not within the bambusoids of old origin. A more

extensive study would be required to understand the
evolution of leaf vascular systems within the grass family,
especially with respect to the bambusoids, the C3 panicoids
and the remaining two C4 lineages including Aristida
and Eriachne.
ACKNOWLEDGEMENTS
We thank Dr Y. Kawamoto (University of the Ryukyus)
for his kind gift of seeds of Brachiaria species.

LITERATURE CITED
Altus DP, Canny MJ. 1982. Loading of assimilates in wheat leaves. I.
The specialization of vein types for separate activities. Australian
Journal of Plant Physiology 9: 571581.
Altus DP, Canny MJ. 1985. Water pathways in wheat leaves. I. The division
of fluxes between different vein types. Australian Journal of Plant
Physiology 12: 173181.
Altus DP, Canny MJ, Blackman DR. 1985. Water pathways in wheat
leaves. II. Water conducting capacities and vessel diameters of
different vein types, and the behaviour of the integrated network.
Australian Journal of Plant Physiology 12: 183199.
Ashida K, Sugino M. 1984. Studies on the stomata in the leaf blades
of gramineous weeds. Weed Research (Japan) 29: 138146.
Botha CEJ. 1992. Plasmodesmatal distribution, structure and frequency
in relation to assimilation in C3 and C4 grasses in southern Africa.
Planta 187: 348358.
Canny MJ. 1990. What becomes of the transpiration stream? New
Phytologist 114: 341368.
Chonan N, Kawahara H, Matsuda T. 1974. Morphology of vascular
bundles of leaves in gramineous crops. I. Observations on vascular
bundles of leaf blades, sheaths and internodes in rice plants.
Proceedings of the Crop Science Society of Japan 43: 425432.
Chonan N, Kawahara H, Matsuda T. 1985. Ultrastructure of transverse
veins in relation to phloem loading in the rice leaf. Japanese Journal of
Crop Science 54: 160169.
Clayton WD, Renvoize SA. 1992. A system of classification for the
grasses. In: Chapman GP, ed. Grass evolution and domestication.
New York: Cambridge University Press, 338353.
Colbert JT, Evert RF. 1982. Leaf vasculature in sugarcane (Saccharum
officinarum L.). Planta 156: 136151.
Crookston RK, Moss DN. 1974. Interveinal distance for carbohydrate
transport in leaves of C3 and C4 grasses. Crop Science 14: 123125.
Dannenhoffer JM, Evert RF. 1994. Development of the vascular system
in the leaf of barley (Hordeum vulgare L.). International Journal of
Plant Sciences 155: 143157.
Dengler NG, Nelson T. 1999. Leaf structure and development in C4 plants.
In: Sage RF, Monson RK, eds. C4 plant biology. San Diego: Academic
Press, 133172.
Dengler NG, Dengler RE, Donnelly PM, Hattersley PW. 1994. Quantitative leaf anatomy of C3 and C4 grasses (Poaceae): bundle sheath and
mesophyll surface area relationships. Annals of Botany 73: 241255.
Ehleringer JR, Monson RK. 1993. Evolutionary and ecological aspects
of photosynthetic pathway variation. Annual Review of Ecology and
Systematics 24: 411439.
Ehleringer JR, Cerling TE, Helliker BR. 1997. C4 photosynthesis,
atmospheric CO2, and climate. Oecologia 112: 285299.
Fisher DG, Evert RF. 1982. Studies on the leaf of Amaranthus
retroflexus (Amaranthaceae): morphology and anatomy. American
Journal of Botany 69: 11331147.
Fritz E, Evert RF, Nasse H. 1989. Loading and transport of assimilates
in different maize leaf bundles. Digital image analysis of
14
C-microautoradiographs. Planta 178: 19.
Gallaher RN, Ashley DA, Brown RH. 1975. 14C-photosynthate translocation in C3 and C4 plants as related to leaf anatomy. Crop Science
15: 5559.
Giussani LM, Cota-Sanchez JH, Zuloaga FO, Kellogg EA. 2001. A
molecular phylogeny of the grass subfamily Panicoideae (Poaceae)

shows multiple origins of C4 photosynthesis. American Journal of

Botany 88: 19932012.
GPWG (Grass Phylogeny Working Group) 2001. Phylogeny and
subfamilial classification of the grasses (Poaceae). Annals of the
Missouri Botanical Garden 88: 373457.
Grodzinski B, Jiao J, Leonardos ED. 1998. Estimating photosynthesis
and concurrent export rates in C3 and C4 species at ambient and
elevated CO2. Plant Physiology 117: 207215
Hanson JC, Rasmusson DC. 1975. Leaf vein frequency in barley.
Crop Science 15: 248251.
Hatch MD. 1987. C4 photosynthesis: a unique blend of modified biochemistry, anatomy and ultrastructure. Biochimica et Biophysica Acta
895: 81106.
Hattersley PW. 1984. Characterization of C4 type leaf anatomy in grasses
(Poaceae). Mesophyll:bundle sheath area ratios. Annals of Botany
53: 163179.
Hattersley PW, Watson L. 1992. Diversification of photosynthesis.
In: Chapman GP, ed. Grass evolution and domestication. New York:
Cambridge University Press, 38116.
Hofstra G, Nelson CD. 1969. A comparative study of translocation of
assimilated 14C from leaves of different species. Planta 88: 103112.
Ishihara K, Hirasawa T. 1978. Relationship between leaf and xylem
water potentials in rice plants. Plant and Cell Physiology 19:
12891294.
Ishihara K, Saitoh K. 1987. Diurnal courses of photosynthesis, transpiration, and diffusive conductance in the single-leaf of the rice
plants grown in the paddy field under submerged condition. Japanese
Journal of Crop Science 56: 817.
Kawamitsu Y, Hakoyama S, Agata W, Takeda T. 1985. Leaf interveinal
distances corresponding to anatomical types in grasses. Plant and
Cell Physiology 26: 589593.
Kawamitsu Y, Agata W, Hiyane S, Murayama S, Nose A, Shinjyo C.
1996. Relation between leaf gas exchange rate and stomata. I.
Stomatal frequency and guard cell length in C3 and C4 grass species.
Japanese Journal of Crop Science 65: 626633.
Kellogg EA. 2001. Evolutionary history of the grasses. Plant Physiology
125: 11981205.
Kocacinar F, Sage RF. 2003. Photosynthetic pathway alters xylem
structure and hydraulic function in herbaceous plants. Plant, Cell and
Environment 26: 20152026.
Kuo J, OBrien TP, Zee S-Y. 1972. The transverse veins of the wheat leaf.
Australian Journal of Biological Science 25: 721737.
Kuo J, OBrien TP, Canny MJ. 1974. Pit-field distribution, plasmodesmatal frequency, and assimilate flux in the mestome sheath of
wheat leaves. Planta 121: 97118.
Langdale JA, Nelson T. 1991. Spatial regulation of photosynthetic
development in C4 plants. Trends in Genetics 7: 191196.
Leonardos ED, Grodzinski B. 2000. Photosynthesis, immediate export
and carbon partitioning in source leaves of C3, C3C4 intermediate
and C4 Panicum and Flaveria species at ambient and elevated CO2
levels. Plant, Cell and Environment 23: 839851.
Lunn JE, Furbank RT. 1999. Sucrose biosynthesis in C4 plants.
New Phytologist 143: 221237.
Lush WM. 1976. Leaf structure and translocation of dry matter in a C3
and a C4 grass. Planta 130: 235244.
Ogle K. 2003. Implications of interveinal distance for quantum yield in
C4 grasses: a modeling and meta-analysis. Oecologia 136: 532542.
Oguro H, Hinata K, Tsunoda S. 1985. Comparative anatomy and
morphology of leaves between C3 and C4 species in Panicum. Annals
of Botany 55: 859869.
Ohsugi R, Murata T. 1986. Variations in the leaf anatomy among some
C4 Panicum species. Annals of Botany 58: 443453.
Pelletier JD, Turcotte DL. 2000. Shapes of river networks and leaves: are
they statistically similar? Philosophical Transactions of the Royal
Society B: Biological Sciences 355: 307311.
Prendergast HVD, Hattersley PW, Stone NE. 1987. New structural/
biochemical associations in leaf blades of C4 grasses (Poaceae).
Australian Journal of Plant Physiology 14: 403420.
Roth-Nebelsick A, Uhl D, Mosbrugger V, Kerp H. 2001. Evolution and
function of leaf venation architecture: a review. Annals of Botany 87:
553566.
Russell SH, Evert RF. 1985. Leaf vasculature in maize (Zea mays L.).
Planta 164: 448458.

Ueno et al. Leaf Vascular Systems in C3 and C4 Grasses

Sage RF. 2004. The evolution of C4 photosynthesis. New Phytologist 161:
341370.
Sinha NR. Kellogg EA. 1996. Parallelism and diversity in multiple origins
of C4 photosynthesis in the grass family. American Journal of Botany
83: 14581470.
Takeda T, Fukuyama M. 1971. Studies on the photosynthesis of
the Gramineae. 1. Differences in photosynthesis among subfamilies and their relations with the systematics of the
Gramineae. Proceedings of the Crop Science Society of Japan 40:
1220.
Tiba SD, Frean ML. 1989. A comparative study of the structurefunction
relationship of cross veins in leaves of Digitaria eriantha and Zea mays.
Annals of Botany 63: 433439.
Ueno O. 1995. Occurrence of distinctive cells in leaves of C4 species
in Arthraxon and Microstegium (Andropogoneae-Poaceae) and the
structural and immunocytochemical characterization of these cells.
International Journal of Plant Sciences 156: 270289.
Ueno O, Sentoku N. 2006. Comparison of leaf structure and photosynthetic
characteristics of C3 and C4 Alloteropsis semialata subspecies. Plant,
Cell and Environment 28: (in press), doi: 10.1111/j.13653040.
2005.01418.x

621

Uhl D, Mosbrugger V. 1999. Leaf venation density as a climate and/or

environmental proxy: a critical review and new data. Palaeogeography, Palaeoclimatology, Palaeoecology 149: 1730.
Wakayama M, Ueno O, Ohnishi J. 2003. Photosynthetic enzyme
accumulation during leaf development of Arundinella hirta, a C4
grass having Kranz cells not associated with veins. Plant and Cell
Physiology 44: 13301340.
Williams LE, Lemoine R, Sauer N. 2000. Sugar transporters in higher
plants: diversity of roles and complex regulation. Trends in Plant
Science 5: 283290.
Woodward FI. 1987. Stomatal numbers are sensitive to increase in CO2
from pre-industrial levels. Nature 327: 617618.
Yamazaki K. 1960. Studies on the connecting strand of the vascular system
in rice leaves. Proceedings of the Crop Science Society of Japan 29:
400403.
Yamazaki K. 1963. Studies on leaf formation in rice plants. II. The
development of leaves in relation to their position on a stem. Proceedings of the Crop Science Society of Japan 32: 8188.
Yoshimura Y, Kubota F, Ueno O. 2004. Structural and biochemical bases
of photorespiration in C4 plants: quantification of organelles and
glycine decarboxylase. Planta 220: 307317.