Spines on diatoms:Do copepods care?

Limnol. Oceanogr., 26(6), 1981, 1057-1061 @ 1981, by the American Society of Limnology

Spines on diatoms: Dian J. Gifford, Department

Richard

of Oceanography,

and Ckcanography,

Inc.

Do copepods

care?l

N. Bohrer,2 and Carl M. Boyd Dalhousic

University,

Halifax,

Nova Scotia B3H 451

Abstract CaZanus$nmarchicus was allowed to feed on two forms of the diatom Thalassiosira weissflogii (=T. fluviatilis): normal cells, which are spinose, and cells which did not have spines. Filtration rates of Calanus on the spinose form were on average 1.7 times higher than on cells with no spines.

‘son and discussions with J. McLachlan The diatom Thalassiosira weissflogii and F. Reid. (Grunow) G. Fryxell and Hasle Comb. nov. (=Thalassiosira fluviatilis Hustedt) has often been used as food in studies of Materials and methods Spined cells of T. weissflogii (clone copepod feeding and feeding behavior Actin) were grown axenically in f/2 me(e.g. Mullin 1963; Frost 1975; Donaghay dium (Guillard and Ryther 1962) without and Small 1979; Fernandez 1979). In such studies the size of the algal cell is agitation at 20°C under constant illumination of 100 quanta. Several workers usually measured with an electronic par(e.g. McLachlan et al. 1965; Herth 1978, ticle counter and is presented as equiv1979) indicated that cells grown under alent spherical diameter. Thalassiosira these conditions, which are usual in this weissflogii is a cylindrical cell laboratory and many others, are spinosc. (diam : length = 1: 1.5) with numerous long threadlike spines composed of p- We used an inoculum from the same culture for our “unspined” cells; these cells chitin (chitan) that extend radially from were grown identically, except that after its valves (McLachlan et al. 1965). The inoculation culture vessels were agitated spines (more properly called fibrils or filat 300 rpm on a rotary shaker. Agitation aments) have lengths that are typically 6prevents daughter cells in a culture from 10 times the diameter of the diatom. producing fibrils or, on some cells, reThey are included in the equivalent duces the length of the fibrils. Although spherical diameter determined by elecwe designate these cells “unspined,” tronic resistive-type particle counters (Boyd and Johnson 1981), but because of many cells possessed short stubby fibrils their thin shape they contribute ~5% to (see Fig. 1) that nevertheless gave them cell volume. We present the results of a mean diameter (cells plus fibrils), measured microscopically, significantly less two sets of grazing experiments designed to test the hypothesis that spined cells than that of fully spined cells (t-test, are ingested at greater rates than unP < 0.01). The mean cell diameter (excluspined cells and propose that the copesive of spines) of the T. weissjlogii culpod captures the spined cells by somcture used in grazing experiment 1 (17 how perceiving them as larger than pm) was greater than that of the culture unspined cells. used in grazing experiment 2 (14 pm). We appreciate the assistance of H. Log-phase cultures were used in both exRueggebcrg, P. Wilkinson, and J. S. Wilperiments, but that of experiment 1 was older than that of experiment 2 (Table 1). Twenty-four hours before the beginl This research was supported by grants from the ning of each grazing experiment, T. Natural Sciences and Engineering Research Counweissflogii cultures were placed in a 5°C cil of Canada to C.M.B.; D.J.G. was supported by coldroom under conditions of illuminaa Dalhousie Graduate Fellowship. tion and agitation similar to those de’ Present address: Max-Planck Institut ftir Limnologie, Postfach 165, 2320 Pliin, West Germany. scribed above. The possibility that fibril 1057

1058

Gifford

Table 1. Thalassiosira Data are 99% confidence lent spherical diameter). Cell diam

weissjogii cultures. intervals (ESD-equiva-

km)

Total cross section + fibrils) (m-4

(cells Direct

ESD

% Spined cells in culture

Spined Unspined

Experiment 17.81k0.68 17.2521.26

1, 15 May 1980 22.0 168.7Ok20.44 20.7 40.03+11.82

64.1 1.4

Spined Unspined

Experiment 13.48rt1.20 13.9720.90

2,3 July 1980 15.4 175.3121.25 15.5 32.23k2.25

91.7 8.3

regeneration occurred under our grazing experimental conditions was assayed by removing unspined cultures from the rotary shaker and placing them in total darkness at 5°C. Periodic examination of live cells showed that after 48 h there was no fibril regeneration. Although the genus Thalassiosira is characteristically spinose (Hasle 1961; Herth and Barthlott 1979; Herth 1978, 1979), the fibrils of T. weissflogii are extremely fine (0.1-0.2 pm wide; McLachlan et al. 1965) and are not easily detected except by high magnification under the compound microscope or by scanning electron microscopy. We find the best resolution of fibrils on live and preserved cells to be provided by the Zeiss photomicroscope II adjusted for phase contrast, Koehler illumination, oil immersion, and with a polarizing filter inserted into the light path. With the exception of glutaraldehyde, most fixatives dissolve or disrupt chitan fibrils (J. McLachlan pers. comm.). Preliminary investigation revealed that spined cells preserved in basic Lugol’s solution (Throndsen 1978) lost their fibrils completely within 10 days. Consequently, we preserved all samples in 2.5% glutaraldehyde, a treatment that had no visible effect on the fibrils for at least 3 months. Glutaraldehyde-fixed cells were prepared for scanning electron microscopy by filtration onto a 25-mm Nuclepore polycarbonate l.O-pm filter pad which had been coated with 0.1% polylysine to enhance adhesion of the cells. The pad was then dehydrated in an ethanol series,

et al. transferred to amyl acetate, and criticalpoint-dried in liquid COz. Portions of the pad, affixed to stubs, were sputter-coated with gold-palladium and examined in a Cambridge 600 Stereoscan SEM. Photographs were taken with a Cambridge 150 SEM (Fig. 1). CIV (experiment 1) and CV (experiment 2) Calanus finmarchicus (Gunnerus) were collected by net tows from the coastal northwest Atlantic Ocean off Chebucto Head, Nova Scotia (44”32’N, 63’33’W). On the day of the experiment copepods were sorted, then preconditioned for 24 h in the dark at 5°C on spined (exp. 1) and unspined (exp. 2) T. weissflogii cells. Densities of diatoms and copepods were 4,000 cells *ml-’ and 25 individuals. liter-l. After 24 h, the copepods were transferred to l-liter glass BOD bottles containing 1,700 (exp. 1) or 1,500 (exp. 2) T. weissflogii cells *ml-l. Copepod densities were 25-30 (exp. 1) and 20-35 individuals *liter-’ (exp. 2). Aliquots (80 ml) of T. weissflogii cells were siphoned from each bottle at the beginning and end of each experiment and counted on a model TAIr Coulter Counter (Coulter Electronics) equipped with a 100~pm aperture tube. Bottles were rotated in the vertical plane at 2 r-pm on a plankton wheel for about 12 h at 5°C in the dark. The final fluid volume of each bottle was measured at the end of each experiment. All treatments of diatoms plus copepods were replicated 10 (exp. 1) and 8 times (exp. 2). Control bottles, containing spined or unspined T. weissflogii only, were replicated twice in both experiments. Copepods were examined at the end of each experiment for dead individuals, then preserved in 4% Formalin and measured immediately using a calibrated ocular micrometer. Phytoplankton cell concentrations (C) and filtration rates (F) were calculated from Frost’s (1972) equations. Because body length varied within each copepodite stage, the data were normalized to wet weight by the relationship w = o.015L3*288

(1)

where W is wet weight (mg) and L is total length (mm). Equation 1 was derived

Gifford

1060

Table 2. Grazing experiments. c is average diatom concentration (No. *ml-l), F is filtration rate (ml *ind-’ *h-l), and F, is weight-corrected filtration rate (ml *mg-’ *h-l). Values for F and F, are 95% confidence intervals.

c Spined Unspined Spined Unspined

F

Experiment 1 1,164 2.15kO.49 1.14kO.25 1,315 Experiment 2 619 7.272 1.40 940 4.53+0.90

FC 4.332 1.05 2.74k0.51 17.2622.77 10.06k1.69

Numerous studies show that copepods fed multispecies mixtures of cultured phytoplankton selectively consume the larger cells (e.g. Mullin 1963; Frost 1977; Schnack 1979). In most such studies, several phytoplankton species of different sizes and shapes have been used, thereby introducing the possibility that variables other than cell size may affect results. Our experimental design has the advantage of using a single food species having different effective sizes. We believe that the fibrils of T. weissflogii cause its effective size to be shifted to the right in the particle size spectrum and that the equivalent spherical diameter of a food item, as measured by an electronic particle counter, is a misleading indicator of its capture efficiency by C. finmarchicus. Recent studies propose complex feeding behavior by copepods fed mixtures of plastic spheres and Thalassiosira species (Donaghay and Small 1979; Fernandez 1979; Donaghay 1980). In view of our results, the postulated rejection of plastic spheres in favor of Thalassiosira cells may be an artifact of experimental design. Such results may require reinterpretation if food particles usually considered to be 14-17-pm diameter in fact appear to be much larger to the organism consuming them. The “training” effect for food size or type reported by other workers (Harvey 1937; Donaghay and Small 1979; Skiver 1980) did not affect our results; copepods consistently consumed spined diatoms at a greater rate whether preconditioned on spined or unspined cells. As Harbison and McAlister (1980) have

et al. noted, the “typical” phytoplankton cell, particularly the equivalent sphere seen by electronic particle counters, is rarely observed in the oceans. Cell morphology, including the formation of spines and armature, is diverse in terms of shape, composition, and density (Smayda 1970). Although our observations specifically concern chitan fibrils, it is likely that our findings also apply to more heavily armored phytoplankton forms such as Chaetoceros. In the context of pelagic community structure, it is therefore of interest to consider why some diatom species have spines. Thalassiosira weissjlogii disposes 18% of its available nitrogen into chitan (McLachlan et al. 1965), and it would seem that spines are important to the diatom. These fibrils are known to increase form resistance and thereby to reduce sinking rate (Xypolyta and Walsby 1976; Walsby and Xypolyta 1977). As is the case for terrestrial plants (Harper 1969) and marine molluscs (Vermeij 1974), differential consumption of food items may be considered a selective force for adaptation. Spines and other architectural armature may additionally function as antipredator mechanisms for the planktonic prey organisms which possess them (Kerfoot 1977; Porter 1977). Large cell size, with its escape from predation by small organisms, is accompanied by lower surface-to-volume ratio, increased nutrient requirement, greater sinking rate, and longer doubling time (Smayda 1970). Fine spinose extensions constitute a more effective means of increasing a diatom’s size, as perceived by its predators, than would be realized by increasing its actual cell size. The spined form thus has the metabolic and life history advantages of a small cell in combination with the predator-avoidance characteristics of larger cells. In our experiments, spines increased the probability of T. weissflogii being eaten and thus did not function as an antipredator device for at least one group of planktonic herbivores. It has been suggested that microzooplankton are important herbivores in pelagic communities (Beers et al. 1971). In grazing experi-

Spines on diatoms ments using natural phytoplankton assemblages as food items, Capriulo and Carpenter (1980) found that tintinnids did not consume spined Thalassiosira or Chaetoceros cells, although as a group they consumed a significant fraction of the daily standing stock of Chl a. Thus, it seems that the spines of T. weissflogii can function as a protective mechanism, making the effectively larger diatom unavailable for consumption by small herbivores. The increased probability of a spinose cell being encountered and eaten by a copepod, as we have observed, would then be offset by the otherwise greater probability of consumption by microzooplankton. References STEVENSON,R.W. EPPLEY,AND E. R. BROOKS. 1971. Plankton populations and

BEERS,J.R.,M.R.

upwelling off the coast of Peru, June 1969. Fish. Bull. 69: 859-876. BOGOROV, V. G., AND YE. N. Preobrazhenskaya. 1934. Weight characteristics of Barents Sea plankters [in Russian]. Tr. Vses. Nauchno-Issled. Inst. Morsko. Rybn. Kohz. Okeanogr. 2: l-25. BOYD, C. M., AND G. W. JOHNSON. 1981. Precision of size determination and systematic errors associated with resistive electronic particle counters. J. Plankton Res. In press. CAPRIULO, G. M., AND E. J. CARPENTER. 1980. Grazing by 35 to 202 pm microzooplankton in Long Island Sound. Mar. Biol. 56: 319326. DONAGEIAY, P. L. 1980. Grazing interactions in the marine environment. Am. Sot. Limnol. Oceanogr. Spec. Symp. 3: 234-240. New England, -, AND L. F. SMALL. 1979. Food selection capabilities of the estuarine copepod Acartia clausi. Mar. Biol. 52: 137-146. FERNANDEZ, F. 1979. Particle selection in the nauplis of Calnnus pncificus. J. Plankton Res. 1: 3 13-328. FROST, B. W. 1972. Effects of size and concentration of food particles on the feeding behavior of the marinc planktonic copepod Calanus paci,ficus. Limnol. Oceanogr. 17: 805-815. -. 1975. A threshold feeding behavior in Calanus puci’cus. Limnol. Oceanogr. 20: 263266. -. 1977. Feeding behavior of Calanus puci$. icus in mixtures of food. Limnol. Oceanogr. 22: 472491. GUILLARD,R. R.,AND J. H.RYTHER. 1962. Studies on marinc diatoms. 1. CycZoteZZu nunu Hustedt and Detonulu confervuceu (Cleve) Gran. Can. J. Microbial. 8: 229-239.

1061

HARBISON, G. R., AND V. L. MCALISTER. 1980. Fact and artifact in copepod feeding experiments. Limnol. Oceanogr. 25: 971-981. HARPER, J. 1969. The role of predation in vcgetational diversity. Brookhaven Symp. Biol. 22: 48-62. HARVEY, H. W. 1937. Notes on selective feeding by Calanus. J. Mar. Biol. Assoc. U.K. 22: 97100. HASLE, G. R. 1961. The morphology of ThuZussiosiru fluviutilis from the polluted inner Oslofjord. Nytt Mag. Bot. 9: 151-154. HEI%TII, W. 1978. A special chitin-fibril-synthesizing apparatus in the centric diatom CycZoteZZu. Naturwissenschaftcn 65: S260. 1979. The site of P-chitin fibril formation -. in centric diatoms: 2. The chitin-forming cytoplasmic structures. J. Ultrastruct. Res. 68: 1627. -, ANI) W. BARTHLOTT. 1979. The site of /3-chitin fibril formation in centric diatoms: 1. Pores and fibril formation. J. Ultrastruct. Res. 68: 6-15. KANAEVA, I. P. 1962. The avcragc weight of copepods of the central and northern Atlantic, Norwegian and Greenland seas [in Russian]. Tr. Vscs. Naucho-Issled. Inst. Morsko. Rybn. Khoz. Okcanogr. 46: 253-266. KERFOOT, W. C. 1977. Implications of copepod predation. Limnol. Oceanogr. 22: 316-325. MCLACIILAN, J., A. J. MCINNES, AND M. FALK. 1965. Studies on chitan (chitin: poly-n-acetyl glucosamine) fibers of the diatom rThuZussiosiru fluviutilis Hustedt. Can. J. Bot. 43: 707-713. MULLIN, M. M. 1963. Some factors affecting fceding of marine copepods of the genus Culunus. Limnol. Oceanogr. 8: 239-250. PORTER, K. G. 1977. The plant-animal interface in frcshwatcr ecosystems. Am. Sci. 65: 159-170. SCIINACK, S. B. 1979. Feeding of Calanus heZgoZundicus on phytoplankton mixtures. Mar. Ecol. Prog. Ser. 1: 4147. SKIVER, J. H. 1980. Seasonal resource partitioning patterns of marinc Calanoid copepods: Species interactions. J. Exp. Mar. Biol. Ecol. 44: 229245. SMAYDA, T. J. 1970. The suspension and sinking of phytoplankton in the sea. Oceanogr. Mar. Biol. Annu. Rev. 8: 353414. THRONDSEN, J. 1978. Preservation and storage, p, 69-74. In A. Sournia [ed.], Phytoplankton manual. Monogr. Oceanogr. Mcthodol. 6. UNESco. VEHMEIJ, G.J. 1974. Marine faunal dominance and molluscan shell form. Evolution 28: 656-664. WALSBY, A. E., AND A.XYP~LYTA. 1977. The form rcsistancc of chitan fibers attached to the cells of ThuZussiosirufluviutiZis Hustedt. Br. Phycol. J. 12: 215-223. XYPOLYTA, A., AND A. E. WALSUY. 1976. Thulussiosiru’s parachute. Br. Phycol. J. 11: 201.

Submitted: 28 October 1980 Accepted: 7 May 1981