Vol.

110: 293-299,1994
MARINE ECOLOGY PROGRESS SERIES
Mar. Ecol. Prog. Ser. l Published July 21

Preservation of marine planktonic ciliates: losses

and cell shrinkage during fixation
Diane K. Stoecker', Dian J. iff ford', Mary putt3
'
Horn Point Environmental Laboratory, PO Box 775, Cambridge, Maryland 21613. USA
Graduate School of Oceanography, University of Rhode Island, Narragansett. Rhode Island 02882. USA
713 Cricklewood Drive, State College, Pennsylvania 16803, USA

ABSTRACT: For enumeration of microzooplankton (20-200 pm size fraction), including planktonic

ciliates, water samples are usually fixed and preserved, then concentrated by sedimentation in Uter-
mohl chambers and exanlined with an inverted microscope. However, losses of ciliates may occur
during fixation and handling, and fixation may shrink cells. Estimates of abundance with several com-
monly used fixatives were compared In samples from the North Atlantic and subarctic Pacific Oceans
and in samples from cultures. Buffered formaldehyde has the advantage that it allows epffluorescence
microscopy to be used, but, on average, ciliate counts from the North Atlantic were 56% higher (95 %
C1 30 to 82 %) in samples fixed with 10 % acid Lugol's solution than in samples fixed with 2 % formalde-
hyde. Ciliate counts from the subarctic Pacific were 23 to 49 % higher in samples fixed with 10 or 20 %
acid Lugol's solution than in samples fixed with 5 % acid Lugol's solution. Fixation with 10 or 20% acid
Lugol's solution results in significantly higher cell counts than fixation with formaldehyde or dilute acid
Lugol's solution, but shrinks cells severely and often distorts their morphology. Bouin's solution yields
cell counts that are usually similar to those with 10% acid Lugol's solution, but with less shrinkage. No
single fixation method is ideal for all purposes; fixative-specific and assemblage- or taxon-specific
correction factors are necessary for accurate estimates of cell numbers and cell volumes/biomass.

KEY WORDS: Microzooplankton . Ciliates . Oligotrichs - Fixation . Shrinkage - Cell losses . Inverted
microscope technique

INTRODUCTION served, concentrated by sedimentation, and then enu-

merated using the inverted microscope method (Hasle
Ciliates are an important component of the protistan 1978). Filtration and staining techniques suitable for
plankton in marine, estuarine and fresh waters (e.g. nanoplankton (< 20 pm protists) (reviewed by Sherr &
Beers & Stewart 1969, Rassoulzadegan & Gostan 1976, Sherr 1993) are generally not suitable for enumerating
Pace & Orcutt 1981, Smetacek 1981, Sorokin 1981, the larger microzooplankton that typically occur at
Revelante & Gilmartin 1983).In open waters, the ciliate lower densities than nanoplankton in open, coastal
assemblage is usually dominated by non-loricate, olig- waters and oceanic waters. An exception is the Quanti-
otrichous taxa (Sorokin 1991). Planktonic ciliates are tive Protargol Technique (QPS) which has been used to
important as consumers of pico- and nanoplankton (Ras- enumerate and identify both nano- and microplanktonic
soulzadegan et al. 1988). Autotrophic and mixotrophic ciliates and dinoflagellates from coastal waters (Mon-
ciliates also contribute to primary production in the mi- tagnes & Lynn 1987, Bockstahler & Coats 1993).
croplankton size fraction (Jonsson 1987, Stoecker et al. A number of fixatives/preservatives have been used
1987a). Planktonic ciliates are, at times, important com- in conjunction with the inverted microscope technique
ponents of the diet of copepods and other macrozoo- to enumerate microplankton in seawater samples,
plankton (Stoecker & Capuzzo 1990, Gifford 1993b). including dilute formaldehyde (Beers 1978, Dale &
Microplankton, including 2 20 pm ciliates, are usually Burkill 1982, Stoecker et al. 1987a), a modified Bouin's
collected as part of whole water samples, fixed and pre- solution (Dolan & Coats 1990), and various concentra-

O Inter-Research 1994
Resale of fuU article not permitted
294 Mar. Ecol. Prog. Ser. 110: 293-299, 1994

tions of acid Lugol's solution (Gifford 1988, 1993a, In the oceanic subarctic Pacific experiment, seawater
Sherr & Sherr 1993).Although dilute acid Lugol's solu- was collected from the middle of the rmxed layer ( z =
tion is recommended for enumeration of flagellates 35 m) at Stn P near 50" N, 145" W. A teflon-lined 30 1
(Throndsen 1978) and has been used to fix and pre- Go-Flo bottle was used to sample from the RV 'Thomas
serve ciliates (Revelante & Gilmartin 1983, Putt & G. Thompson' as part of the SUPER ( x b a r c t i c Pacific
Stoecker 1989, Ohman & Snyder 1991, Jerome et al. Ecocystem Besearch) program. Whole water samples
-
1993) use of higher concentrations may reduce ciliate (1600 to 1900 ml, depending on fixative concentration)
losses (Gifford 1993a, b ) . Live counting of ciliates were drained through silicon tubing into jars contain-
often results in significantly higher estimates of cihate ing 3 concentrations of acid Lugol's solution: 5 % , 10%
abundance than does counting of preserved samples and 20 % (final v/v).
(Sorokin 1981). For example, Dale & Burkill (1982) Several months later, 100 m1 subsamples were con-
found that live counting of ciliates from coastal waters centrated by sedimentation using Utermohl chambers
often resulted in up to 20 % higher estimates of abun- and examined with an inverted microscope (Hasle
dance than did counting of samples preserved with 1978). All aloricate ciliates in the settling chambers
0.4 % (final conc.) neutral formaldehyde. This raises were enumerated at 200x or 250x magnification. Lori-
the question of the importance of cell losses during cate ciliates (tintinnids) were absent from the North
fixation and whether fixation method influences esti- Atlantic samples and were present, but rare, in the
mates of abundance in microzooplankton assemblages. subarctic Pacific samples. Tintinnids are not included
The biomass of ciliate assemblages is typically esti- in the total counts. In the North Atlantic samples, only
mated from numerical abundance and cell size using total counts of aloricate choreotrichs and oligotrichs
vo1ume:carbon conversion factors. Shrinkage, caused were made. In the samples from the subarctic Pacific,
by fixation and preservation may affect estimates of the aloricate ciliates were categorized on the basis of
cell biomass based on measurements of cell volume geometry as spheres or cones; oral membranelles and
(Choi & Stoecker 1989, Putt & Stoecker 1989, Ohman & tail structures were not included in the estimates of cell
Synder 1991, Jerome et al. 1993). Field and laboratory geometry. The large mixotroph Laboea strobila was
experiments were done to address the question of how enumerated separately from other cone-shaped cili-
fixation procedure affects estimates of ciliate abun- ates. Ciliate densities were corrected for the volume of
dance and biomass. Experiments were done at 2 fixative in each sample and expressed as cells 1-l.
oceanic stations in order to compare the effects of Fixation experiments with cultured ciliates. Fixation
various fixatives on estimates of cell number in natural of ciliate cultures with 2 %, 5 % and 10 % acid Lugol's
microplankton assemblages. Because it is difficult to solution, 2 % buffered formaldehyde, and 5 % Bouin's
compare fixation losses for individual species or to solution were compared in the laboratory to confirm
compare changes in cell volume in natural assem- and expand the results of the comparative fixation
blages, experiments on fixation and preservation of experiments done with natural assemblages. Three
laboratory cultures representative of the Oligotrichidia cultures representative of important ciliate taxa in the
(Strombidium) and the Choreotrichida (Strobilidium, oceans were used: Strombidium capitatum (an olig-
non-loricate and Favella, loricate) were also performed. otrich), Strobilidium spiralis (an aloricate choreotrich)
and a tintinnid, Favella sp. (a loricate choreotrich).
Culture methods are described in Stoecker et al.
METHODS (1987b).The 3 cultures were mixed together and tripli-
cate - 150 m1 samples were preserved in each fixative.
Fixation experiments with natural assemblages. In Samples were stored for several weeks before 100 m1
the North Atlantic experiment, water samples were subsamples were settled in Utermohl chambers and
collected with Niskin bottles from 4 depths in the each species enumerated using the inverted micro-
upper water column (z = 0 to 34 m) on 3 dates during scope technique.
late spring 1989. Sampling was done from the RV To compare cell volumes, 30 to 31 cells of each
'Atlantis 11' in the vicinity of 47" N , 18" W as part of the species from each replicate were measured with a
JGOFS North Atlantic Bloom Experiment (Sieracki et calibrated ocular micrometer. Cell volume was esti-
al. 1993).Whole water samples (950 to 990 ml, depend- mated using an ellipsoid approximation of cell shape
ing on fixative) from each Niskin bottle were drained for all 3 species. Oral membranelles and tail structures
through silicon tubing into 3 jars containing different were not included in the estimates of cell volume. Only
fixatives: 2 % formaldehyde buffered with hexamethy- the cell inside the lorica of Favella sp. was measured.
lamine (Throndsen 1978), 10% acid Lugol's solution Statistical analyses. Two-way analysis of variance
(Throndsen 1978),and 5 % Bouin's solution (stock solu- was done using SAS. All tests of significance were per-
tion from Sigma Diagnostics) (all final conc., v/v). formed at the p = 0.05 level. A Bonferroni (LSD) cor-
 Stoecker et al.: Losses of ciliates dunng fixation 295

rection to the significance level was used when multi- employed has a significant effect on abundance esti-
ple comparisons were made (Kleinbaum et al. 1988). mates (Table 1). Abundance estimates from samples
For the North Atlantic data it was assumed that fixed with 10 % acid Lugol's solution were significantly
ciliate counts for any 2 preservatives are related in a higher than estimates from corresponding samples
linear fashion. Replicate counts were not made, how- fixed with 5 % acid Lugol's solution. Although the
ever it seems unlikely that the standard errors of the average abundance estimates for conical ciliates were
dependent variables are constant over the range of higher in the treatment with 20 % rather than 10 % acid
observed cell densities (156 to 9489 I - ' ) . Thus, ordinary Lugol's solution, the estimates of total ciliate abun-
least squares regression is inappropriate. Instead, a dance were not significantly different between these 2
quasi-likelihood approach (McCullagh & Nelder 1991) treatments. The interaction between ciliate category
was employed. Using this approach, we fit a linear (cone, sphere or Laboea) and percent acid Lugol's solu-
relationship and also included the assumption that the tion was not statistically significant (Table 1).However
variance of the dependent variable is proportional to the average abundance of Laboea sp. and other ciliates
its mean. Parameter, variance and proportionality con- with a conical morphology was greater with the higher
stant estimates were computed using STATA (Hilbe concentrations of fixative. Differences between species
1993). Approximate 95 % confidence intervals for the
mean were based on large sample normal approx- Table 1. Abundances of 3 morphological types of ciliates (cells
imations. Negative estimates of the intercept were I-') in samples from the sub-arctic Pacific preserved with 5 %,
truncated to zero. 10%, and 20 % acid Lugol's solution (L). Abundances were
corrected for the volun~eof fixative. Means of 3 replicates
(SD in parentheses). ns: non-significant (p < 0.05)

RESULTS

In the North Atlantic samples, the ciliate abundance Laboea strobila 204 (22) 263 (39) 304 (79)
estimates from water fixed with 10 % acid Lugol's solu- Conical morphology a 442 (74) 556 (80) 654 (170)
tion were significantly higher ( p = 0.006) than those of Spherical morphology 400 (104) 544 (51) 492 (108)
water fixed with 2 % buffered formaldehyde (Fig. 1A).
2-way ANOVA
On average, estimates for samples preserved with Source of variation Mean square df F value
10 % acid Lugol's solution were 56 % (95% C1 of 30 to -

8 2 % ) higher than the formaldehyde based estimate % Acid Lugol's 45306 2 5.52 p < 0.05
for the same samples (Fig. 1A).The estimates of ciliate Morphological type 210983 2 25.69 p < 0.001
Interaction 6093 4 0.74 ns
abundance in samples preserved with acid Lugol's
Error 8213 18
solution were, on average, about 7 % higher than the
corresponding estimates for samples preserved with A priori comparisons (LSM)
Bouin's solution, but the differences between samples 5% L v s 10% L p < 0.05
were not significant (Fig. 1B). 10% L vs 20% L ns
The fixation experiment from the subarctic Pacific Not including Laboea
indicates that the concentration of acid Lugol's solution

Fig. 1. Relationship of abundance estimates of

planktonic ciliates (cells 1 - l ) for samples from
surface waters of the North Atlantic fixed and
preserved with (A) 10% acid Lugol's solution
versus 2 % buffered formaldehyde and (B) 5 %
Bouin's solution. The linear relationship was
estimated using the model Y = bo + b,X, where
X 1s the estimate with 2 % formaldehdye (A) or
5 % Bouin's solution (B) and Y is the estimate
with 10% acid Lugol's solution. Average abun-
dance with 10% Lugol's predicted from the
estimated slope (b,) and intercept (bo) (-)
with 95 % confidence interval (- - - - - -). The pre-
dicted line which would arise if bo = 0 and b, = 1
(-----) ' IS shown in (A). Parameter estimates
(95% Cl): Lugol's on formaldehyde (A) bo = 57
(0, 318); b, = 1.56 (1.30, 1.82),Lugol's on Bouin's
bo = -8 (0, 499), b l = 1.07 (0.76, 1.37)
 Mar. Ecol. Prog. Ser. 110: 293-299, 1994

may have existed but may have Table 2. Abundances (cells 100 ml-l) of cultured ciliates preserved with 2%, 5%.
been below the 'limit of detection' 10% acid Lugol's solution (L), 2 % buffered formaldehyde (F) and 5 % Bouin's solu-
for this experiment. tion (B). Means of 3 replicates (SD in parentheses). ns: non-significant (p < 0.05)
In the laboratory experiments, in-
teractions between species and fix-
ation treatment were significant for Strobilidium spiralis 151 (5) 140 (4) 167 (9) 138 (14) 165 (9)
both abundance estimates (Table 2) Fa vella sp. 288 (12) 264 (11) 277 (20) 246 (11) 234 (24)
and cell size (Table 3). Although Strornbidium capitatum 227 (15) 272 (32) 300 (39) 193 (7) 251 (23)
abundances of all 3 species varied
with fixation treatment, between- l 2-way ANOVA
treatment differences were only sig- Source of vanation Mean square df F value
nificant in Strombidium capitatum
(Table 2 ) . For this oligotrich, the Preservative
estimates with 2 % buffered form- Species
Interaction
aldehyde were significantly lower Error
than the estimates with 5 % or 10 %
acid Lugol's solution (Table 2). For
A priori comparisons (LSM)
example, the estimate of S. capita-
S. spiralis Favella sp. S. capita turn
turn cell density with 2 % formalde-
hyde was only 64 % of the estimate
with 10 % acid Lugol's fixation.
Fixation method influenced cell
size in all 3 species (Table 3). Cell
volumes in the 10% acid Lugol's
treatment were significantly lower
than cell volumes with the 2 % acid

Table 3. Mean cell volume (pm3X 10') of cultured c h a t e s preserved with 2 % , 5 % , 10% acid Lugol's solution (L), 2 % buffered
formaldehyde (F) and 5 % Bouin's solution (B). Means of 3 replicates (SD in parentheses), except for 2 % L means of 2 replicates
(range in parentheses)

2% L 5% L 10% L 2% F 5% B

Strobilidium spiralis 45.0 (40.1-49.9) 40.2 (3.6) 35.8 (1.9) 60.4 (2.4) 38.2 (1.8)
Favella sp. 73.5 (70.1-76.9) 74.4 (2.1) 61.3 (3.6) 108.9 (0.6) 76.2 (8.5)
Strombidium capita tum 47.3 (42.6-52.0) 43.2 (1.6) 31.4 (3.5) 77.1 (6.3) 45.8 (5.4)

2-way ANOVA
Source of variation Mean square df Fvalue

Preservative 1948 X 106 4 102.25

Species 4860 X 106 2 255.15
Interaction 71 X 106 8 3.7 l
Error 19 X 106 27

A priori comparisons (LSM)

S. spiralis Favella sp. S. cap~ta
turn

2 % L v s 10% L ns p < 0.05 p c 0.05

2%Lvs5%L ns ns ns
5 % Lvs 10% L ns p < 0.05 p < 0 05
2% F v s 2 % L p c 0.05 p < 0.05 p c 0.05
2 % FvsSO/o L p C 0.05 p < 0.05 p < 0.05
2%Fvs10%L p c 0.05 p < 0.05 p < 0.05
5%BvslO%L ns p<005 p<005
-
 Stoecker et al.: Losses of clliates during f ~ x a t ~ o n 297

Lugol's treatment for all 3 species. For Strombidium tion with epifluorescence microscopy, and thus plastidic
capitaturn and Favella sp., cell volumes with 10 % acid and non-plastidic cells can be distinguished (Stoecker
Lugol's solution were significantly lower than with 5 % et al. 1987a).
acid Lugol's solution. For all 3 species, cell volumes Because estimates of ciliate abundance from fixed
with 2 %, 5 % and 10 % acid Lugol's solution were sig- samples were not compared to live counts, we do not
nificantly lower than with 2 % buffered formaldehyde. know the accuracy of our estimates. However, Dale &
For example, the cell volume of S. capitatum preserved Burkill's (1982) reported loss factors of up to 20%
in 2 % buffered formaldehyde was 2 . 4 5 the~ volume of with dilute buffered formaldehyde (compared to live
similar cells preserved in 10% acid Lugol's solution. counts) and our estimated loss factor of 30 to 40%
For Favella sp. and S. capitatum, but not S. spiralis, (compared to 10% acid Lugol's) with buffered form-
volumes of cells preserved with 5 % Bouin's solution aldehyde suggest that the estimates obtained with 10
were significantly higher than those preserved with or 20 % acid Lugol's solution or 5 % Bouin's are close
10 % acid Lugol's solution. In all cases, volumes of cells to maximum. However, our experiment with cultures
preserved with 2 % formaldehyde were significantly of ciliates demonstrated that cell losses can be taxon-
higher than those preserved with 5 % Bouin's solution. specific (Table 2). It is thus possible that some species
are not fixed and preserved quantitatively with either
strong acid Lugol's or Bouin's solutions.
DISCUSSION Type of fixative as well as fixative concentration and
ciliate taxa also influence biomass estimates through
Method of fixation can influence estimates of ciliate their effect on post-fixation cell volume (Table 3).
numerical abundance significantly (Pace & Orcutt Similar effects of fixation on ciliate cell volume have
1981, Revelante & Gilmartin 1983; R. Leakey, P. Burk- been reported by Choi & Stoecker (1989), Putt &
hill & M. Sleigh summarized in Rassoulzadegan 1991). Stoecker (1989), Ohman & Snyder (1991) and Jerome
The combined results of our field and laboratory et al. (1993). Volume to carbon conversion factors for
experiments indicate that higher cell counts are preserved planktonic ciliates, such as those of Putt &
obtained in either strong ( l 0 or 20 %) acid Lugol's solu- Stoecker (1989) and Ohman & Snyder (1991) are fixa-
tion or Bouin's solution than in 2 % acid Lugol's solu- tive- and probably taxon-specific. For example, if the
tion or 2 % formaldehyde. Higher non-loricate ciliate volume to carbon conversion factor of 0.14 pg C pm-3
densities in samples fixed with 1% acid Lugol's (final for oligotrichs preserved in 2 % buffered formaldehyde
conc.) than in samples fixed with formaldehyde have (Putt & Stoecker 1989) is applied to oligotrichs pre-
been reported previously by Pace & Orcutt (1981) and served with 2 % acid Lugol's solution or 5 % Bouin's
Revelante & Gilmartin (1983).However, our data indi- solution, biomass is probably underestimated by about
cate that counts of oligotrichs are significantly higher 40%. If the same ciliates are fixed with 10% acid
in strong (2 10 %) than in the dilute (12 %) acid Lugol's Lugol's solution, biomass is probably underestimated
solution. Low concentrations of acid Lugol's solution by about 59% with the formaldehyde conversion
(usually 0.5 to 2 %) have traditionally been used to pre- factor.
serve microplankton samples (Sherr & Sherr 1993) Fixation procedures can have a large effect on the
although Gifford (1993a)recommends 10 % for ciliates. estimation of planktonic ciliate abundance and bio-
The data in Table 1 suggest that for certain ciliates mass (Sime-Ngando et al. 1990, Ohman & Snyder
(Laboea and other ciliates with a conical morphology) 1991, Sime-Ngando & Groliere 1991, Jerome et al.
slightly higher counts may be obtained with 20 % than 1993). Failure to account for losses during fixation
with 10% acid Lugol's solution. However, with 20% can result in a n underestimation of ciliate numbers
acid Lugol's solution, ciliates may be difficult to iden- and biomass (Fig. lA, Tables 1 & 2). Failure to adjust
tify and size because of severe shrinkage, distortion of vo1ume:carbon conversion factors for cell shrinkage
morphology and dark coloration. The iodine color can (Table 3) may underestimate ciliate biomass by >40 %.
be bleached with sodium thiosulfate (E. Sherr & B. Because some taxa appear to be more sensitive to
Sherr summarized in Rassoulzadegan 1991, Sherr & these effects than others (Sime-Ngando & Groliere
Sherr 1993),laundry bleach (Gifford pers. obs.) or light 1991, Jerome et al. 1993) the taxonomic composition of
(Gifford pers. obs.). the ciliate assemblage probably affects the accuracy of
For cells preserved with 2 % buffered formaldehyde, estimates of d a t e abundance and biomass.
abundance estimates of oligotrichs were on average Correcting for cell losses in samples preserved in
63 to 64 % of those obtained for cells preserved with dilute acid Lugol's solution or 2 % formaldehyde based
10% acid Lugol's solution (Fig. l A , Table 2). Although on our data may not be appropriate for all samples, in
cell losses are significant with this fixative, formalde- part because of the taxon-specific effects of the differ-
hyde has the advantage that it can be used in conjunc- ent fixatives. Other factors may influence cell losses
298 Mar. Ecol. Prog. Ser. 110: 293-299, 1994

and cell shrinkage as well. Losses may occur prior to Gifford, D. J. (1993b). Protozoa in the diets of Neocalanusspp.
fixation due to the mechanics of collection or manipu- in the oceanic subarctic Pacific Ocean. Prog. Oceanogr.
32: 223-237
lation of the sample (Gifford 1985). Osmotic affects
Hasle, G. R. (1978). The inverted-microscope method. In:
(Jerome et al. 1993), due to the salinity, as well as the Sournia, A. (ed.) Phytoplankton manual. UNESCO, Paris,
feeding history of cells (Choi & Stoecker 1989) may p. 88-96
influence preservation and shrinkage. Cell losses in Hllbe, J . (1993).Generalized linear models. Stat. techn. Bull.
liquid samples may increase with storage time (Sherr & 11: 20-28
Jerome, C. A., Montagnes, D. J . S., Taylor, F. J . R. (1993).The
Sherr 1993), although this does not appear to be the effect of the quantitive protargol stain and Lugol's and
case for marine ciliates preserved in acid Lugol's Bouin's fixatives on cell size: a more accurate estimate of
solution (Ohman & Snyder 1991, Gifford unpubl. data). ciliate species biomass. J. euk. Microbiol. 40: 254-259
For all of the above reasons, carbon:volume factors and Jonsson, P. R. (1987) Photosynthetic assimilation of inorganic
carbon in marine oligotrich ciliates (Ciliophora, Olig-
factors correcting for cell loss and shrinkage should not
otrichina).Mar. microb. Food Webs 2: 55-68
be used indiscriminately. Individual investigators must Kleinbaum, D. G., Kupper, L. L., Muller, K. E. (1988).Applied
check or derive correction factors for the particular regression analysis and other multivariate methods. PWS-
microplankton assemblages and environmental condi- Kent Publishing Co., Boston
tions which apply in their own research. Further com- McCullagh, P., Nelder, J . A. (1991). Generalized linear
models. Monographs on statistics and applied probabihty,
parisons of numerical and biomass estimates obtained 37. Chapman and Hall. New York
with different fixation and preservation methods and Montagnes, D. J. S., Lynn, D. H. (1987). A quantitative pro-
live techniques are needed before correction factors targol stain (QPS) for ciliates: method descnption and
can be recommended for routine use. test of its quantitative nature. Mar. rnicrob. Food Webs 2:
83-93
Ohman, M. D., Snyder, R. A. (1991). Growth kinetics of
Acknowledgements. This research was partially supported by
the omnivorous oligotrich ciliate Strombidium sp. Limnol.
NSF grants OCE-88-l7399,OCE-86-13892 and DPP-88-16668.
Oceanogr. 36: 922-935
We thank P. G. Verity and M. E. Sieracki and the USJGOFS
Pace, M. L., Orcutt, J. D. Jr (1981).The relative importance of
program for the help in obtaining the North Atlantic samples,
protozoans, rotifers, and crustaceans in a freshwater zoo-
the SUPER program for assistance in the subarctic Pacific and
plankton community. Limnol. Oceanogr. 26: 822-831
A . E . Michaels for assistance in the culturing and enumeration
Putt, M., Stoecker, D. K. (1989). An experimentally deter-
of ciliate samples. Critical reviews by F. Rassoulzadegan and
mined carbon-volume ratio for marine oligotrichous cili-
2 anonymous readers are appreciated.
ates from estuarine and coastal waters. Limnol. Oceanogr.
34. 1097-1103
Rassoulzadegan, F. (1991) Methods for the study of marine
LITERATURE CITED microzooplankton - Session summary. In: Reid, P. C,
Turley, C. M., Burkill, P. H. (eds.) Protozoa and their role
Beers, J. R. (1978). About microzooplankton. In: Sournia, A. in marine processes. Springer-Verlag, Berlin, p. 39-57
(ed.) Phytoplankton manual. UNESCO, Paris, p. 288-296 Rassoulzadegan, F., Gostan, J. (1976). Repartition des cilies
Beers, J . R., Stewart, G. L. (1969).Microzooplankton and its pelagiques dans les eaux de Villefranche-sur-Mer. Remar-
abundance relative to larger zooplankton and other seston ques sur la dispersion du microzooplankton en mer et a
components mar. Biol. 4: 182-189 I'interieur des echantillons denombres par la methode
Bockstahler, K. R., Coats, D. W. (1993).Grazinq of the mixo- d'utermohl. Annls Inst. oceanogr., Pans 52: 175-188
trophic &noflagellate ~ ~ m n o d i r sanguineurn
uu~ Hirasaka Rassoulzadegan, F., Laval-Peuto, M,, Sheldon, R. W. (1988).
-
on cihate -populations of Chesapeake Bay. Mar. Biol. 116: Partitioning of the food ration of marine ciliates between
477-487 pico- and nanoplankton. Hydrobiologia 159: 75-88
Choi, J. W., Stoecker, D. K. (1989). Effects of fixation on cell Revelante, N., Gilrnartin, M. (1983).Microzooplankton distri-
volume of marine planktonic protozoa. Appl. environ. bution in the Northern Adriatic Sea with emphasis on the
Microbiol. 55: 1761-1765 relative abundance of ciliated protozoans. Oceanologica
Dale, T., Burkill P. H. (1982). 'Live counting' - a quick and Acta 6: 407-415
simple technique for enumerating pelagic ciliates. Annls Sherr, E. B., Sherr, B. F. (1993).Preservation and storage of
Inst. oceanogr., Paris 58(S):267-276 samples for enumeration of heterotrophic protists. In:
Dolan, J. R., Coats, D. W. (1990).Seasonal abundance of plank- Kemp, P. F.. Sherr, B. F., Sherr, E. B.. Cole, J . J . (eds.)
tonic ciliates and microflagellates in mesohaline Chesa- Handbook of methods in aquatic microbial ecology. Lewis
peake Bay waters. Estuar. coast. Shelf Sci. 31: 157-175 Publishers, Boca Raton, p. 207-212
Gifford, D. J. (1985). Laboratory culture of marine planktonic Sieraclu, M. E., Verity, P. G., Stoecker, D. K. (1993).Plankton
oligotrichs (Ciliophora, Oligotrichida). Mar Ecol. Prog. community response to silicate and nitrate depletion at 47'
Ser. 23: 257-267 during the 1989 North Atlantic spring bloom. Deep Sea
Gifford, D . J. (1988).Impact of grazing by microzooplankton Res. 40: 213-225
in the northwest arm of Halifax Habour, Nova Scotia. Mar. Sime-Ngando, T., Groliere, C. A. (1991). Effets quantitatifs
Ecol. Prog. Ser. 47: 249-258 des Fixateurs sur la Conservation des Cilies Planctoniques
G~fford,D. J . (1993a).Consumption of protozoa by copepods d'Eau Douce. Arch. Protistenkd. 140: 109-120
feeding in natural microplankton assemblages. In: Kemp. Sime-Ngando, T., Hartmann, H. J., Groliere, C. A. (1990).
P. F., Sherr. B. F., Sherr, E. B., Cole, J . J. (eds.) Handbook Rapid quantification of planktonic ciliates: comparison of
of methods in aquatic microbial ecology. Lewis Publishers, improved live counting with other methods. Appl. envi-
Boca Raton, p. 723-729 ron. Microbiol. 56: 2234-2242
 Stoecker et al.: Losses of ciliates during fixation

Smetacek, V. (1981).The annual cycle of protozooplankton in Stoecker, D. K., Michaels, A. E.. Davis, L. H. (1987a). Large
the Kiel Bight. Mar. Biol. 63: 1- 11 proportion of marine planktonic ciliates found to contain
Sorokin, Y . I. (1981).Microheterotrophic organism in marine functional chloroplasts. Nature 326: 790-792
ecosystems. In: Longhurst, A R. (ed.).Analysis of marine Stoecker, D. K., Verity, P. G., Michaels, A. E., Davis, L. H.
ecosystems. Academic Press, New York, p. 293-341 (198713).Feeding by larval and post-larval ctenophores on
Stoecker, D. K., Capuzzo, J. M. (1980). Predation on Protozoa: microzooplankton. J. Plankton Res. 9: 667-683
its importance to zooplankton. J. Plankton Res. 12: Throndsen, J. (1978).Preservation and storage. In: Sournia, A.
891-908 (ed ) Phytoplankton manual. UNESCO, Paris, p. 69-74
This article was submitted to the editor Manuscript first received: December 12, 1993
Revised version accepted: April 22, 1994