© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Radiotracer determination of the diet of calanoid copepod nauplii and copepodites in a temperate estuary
a Department of Zoology, University of Guelph Guelph, Ontario N1G 2W1, Canada
b Environmental Science, Acadia University Wolfville, Nova Scotia B4P 2R6, Canada
*Corresponding author. e-mail: john.roff{at}acadiau.ca; kerrifin{at}uoguelph.ca.
Many studies have focused on the consumption of specific prey items and clearance rates on natural food particles by copepod adults and copepodites, but few have addressed the feeding habits of the youngest stages, the nauplii. Because of the difference in size and shape between copepod developmental stages, corresponding differences in diet would be expected. This subject was addressed using a modification of a new method offering radiolabelled natural food particles (with 3H for heterotrophs and 14C for autotrophs) to nauplii and copepodites of the copepods in a temperate estuary. By measuring the uptake of the radiolabel by the copepods it was possible to determine clearance rates on different sizes and types of food particles. All copepods began feeding on food sources >2 µm in size, and feeding on heterotrophs by all species and stages was considerably lower than feeding on autotrophs. Overall, it appears that copepod nauplii have a comparable diet to the later stages in terms of food type and size and therefore it is unlikely that nauplii are a more efficient link between the microbial foodweb and the classical foodweb, at least in temperate estuarine waters.
Keywords: clearance rates, copepod nauplii, copepodites, natural food assemblage, radiotracers
 |
Introduction |
|---|
 Top Introduction Materials and methodsResultsDiscussionReferences |
|---|
Copepods are generally considered to be the dominant trophic link between primary producers and fish in the oceans. Many studies have therefore examined their grazing rates on a variety of food sources. It is now apparent that copepods are capable of feeding on a wide variety of food sources, from bacteria (Turner and Tester, 1992; Roff et al., 1995), to heterotrophic protists (e.g. Stoecker and Egloff, 1987; Verity and Paffenhofer, 1996), to phytoplankton (e.g. Gifford and Dagg, 1988). The majority of these studies have focused on the later stages (copepodites and adults), although, because of their high abundance in marine systems, it is likely that nauplii must also contribute significantly to copepod grazing pressure.
The diets of nauplii need to be examined separately from the later stages, as there are several reasons for believing that they would differ from their adult and copepodite counterparts. First, the body plan of nauplii is significantly different from that of the copepodites and adults (both in terms of size and shape and in having fewer feeding appendages) and, second, they are known to have different feeding mechanisms (see Paffenhofer and Lewis (1989) for a detailed description of the feeding behaviour of copepod developmental stages). Consequently, their food resources will likely be different. Furthermore, previous work has shown that nauplii respond differently than copepodites and adults to food-limiting conditions. In tropical regions it has been demonstrated that naupliar growth rates are uncoupled from chlorophyll concentration, while the copepodites and adults become progressively food-limited with increasing body size (Hopcroft and Roff, 1998; Hopcroft et al., 1998; Richardson and Verheye, 1999). In temperate studies it has been similarly demonstrated that naupliar growth rates, although highly variable, were only weakly related to the smallest size fraction of chlorophyll, while the copepodites had a much stronger relationship with the chlorophyll concentration in the microplankton size range (Finlay and Roff, unpublished data). It is hypothesized that the difference between growth rate responses observed in these studies was due to a difference in diet between the life stages of the copepods: if nauplii were feeding on heterotrophic food particles, then their growth rates would not be strongly related to chlorophyll concentration. It was then proposed that nauplii could be a strong link between the microbial foodweb and the classical foodweb as they would be bringing primarily heterotrophic production into the classical foodweb.
There have been a handful of feeding studies which have started to answer the question of the food sources of copepod nauplii. Of these, Paffenhofer (1971) and Meyer et al. (2002) focused on feeding by all developmental stages on large (>10 µm) uni-algal food or assemblages of cultured algae and protists, but only one (Berggreen et al., 1988) examined feeding on the smallest size fractions of food (down to 2-µm diameter). These all found that the clearance rates of nauplii were roughly an order of magnitude lower than adults. These studies, however, do not represent copepod feeding on natural food sources, nor did they assess feeding on bacterioplankton. Merrell and Stoecker (1998) examined the difference in diet between nauplii and copepodites on natural food particles by measuring the difference in food concentrations before and after 24-h incubations. Their study similarly found that naupliar clearance rates were considerably lower than those of copepodites and adults, and also determined that all stages had higher clearance rates on protozoan microplankton than on chlorophyll a. Merrell and Stoecker (1998) additionally found that the lower size limit of food was the same for all stages, but that the optimal size increased with increasing developmental stage. However, their study only focused on relatively large particles (15–70-µm diameter) and lumped all smaller particles together in <10 µm chlorophyll analyses; their study therefore does not include the potential for feeding on the smaller heterotrophs and picoplankton. Roff et al. (1995) and Turner and Tester (1992) demonstrated that nauplii can feed on bacteria; however, the relative importance of this potential food source is still unclear. Thus some information regarding the naupliar feeding is present in the literature, but there is still much to be determined before the diets of nauplii and copepodites can be generalized.
The purpose of this study is to determine whether there are differences in size, food type (autotroph vs. heterotroph), and quantity of food between nauplii and copepodites of the copepods in temperate estuarine waters. These experiments used radioisotopes as tracers of autotrophic and heterotrophic prey, thus allowing an examination of all potential food categories of all sizes for the copepods.
|
Materials and methods |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
All copepods were collected from Passamaquoddy Bay, adjacent to the Bay of Fundy, using a 64-µm WP2 plankton net with 30-cm diameter and 60-µm filtering codend in slow vertical tows from 10-m depth. Animals were immediately diluted in natural seawater and transported to the laboratory in insulated containers within 1 h of collection. All water for the feeding experiments was collected into 20-l polycarbonate carboys at the same location in the bay from 1-m depth with a bilge pump.
At the laboratory, the copepods were kept in 60-l aquaria in a temperature-controlled room (kept at bay temperatures, 14–15°C) and gently aerated under natural photoperiod. The water in the carboys to be used for food for the experiments was kept in an outdoor flow-through channel of bay water to maintain the temperature at natural levels. Copepods and the water used in the experiments were always used within 2 d of collection.
The most abundant species of copepods in the Bay during the experiments were Acartia hudsonica and Eurytemora herdmani; together, these typically constituted approximately 90% of the total copepod abundance. Other common species in this area included Temora longicornis, Oithona similes, Tortanus discaudatus, and Centropages typicus. Owing to the low abundance of these species, only A. hudsonica and E. herdmani were used in the feeding experiments.
Radiolabelled food experiments
We modified the methods presented in Baldwin and Newell (1991), who used radioisotopes as tracers of autotrophs (by labelling with sodium-14C-bicarbonate) and heterotrophs (labelled with 3H-thymidine). After feeding the labelled food to the different species and stages of copepods, the amount of label taken up by each can be measured by scintillation counting, and the clearance rate of each species and stage on the different size fractions of autotrophs and heterotrophs can be determined.
Aliquots of Bay water were filtered through 0.2-µm polycarbonate membrane filters (Poretics Co.), 2.0-µm polycarbonate filters (Osmonics Co.), and 5.0-µm, 11.0-µm, 20.0-µm, and 60.0-µm nitex mesh in order to provide the copepods with a range of food particle sizes. The 0.2-µm filter acted as a control, as this would remove even small bacteria. The different feeding treatments offered to the copepods were therefore additive, as they contained either no food (<0.2 µm filtered water) or food <2.0-µm, <5.0-µm, <11-µm, <20-µm, and <60-µm diameter. In order to ensure that fragile protists, such as the naked ciliates, were not destroyed, samples of the water collected through the bilge pump and into the carboys were examined under a dissecting microscope within a few hours of collection. Although no quantitative measures were made, visual examination indicated that the protists were still swimming actively, and did not appear to be negatively affected by this process. Furthermore, when filtering the carboy water through the 5.0-µm, 11.0-µm, 20.0-µm, and 60.0-µm nitex mesh, the filtration device was kept submerged in water to prevent excess damage to the fragile protists. The 0.2-µm and 2.0-µm water was filtered under vacuum pressure, but as the primary particles present in these treatments were bacteria, this was considered acceptable.
This water was allocated to 250-ml polystyrene tissue culture flasks and to each of these flasks (6 total, each containing a different size fraction), either 3H-thymidine (specific activity 60 Ci mmol–1, final concentration of 0.5 µCi ml–1), or sodium-14C-bicarbonate (specific activity 58 mCi mmol–1, final concentration of 0.5 µCi ml–1) was added (both radioisotopes from ICN Biomedicals Inc.). The flasks were then incubated in the outdoor channel for 4 h, between 10:00 and 18:00, where the natural assemblages of phytoplankton and heterotrophic bacteria and protists were allowed to grow and assimilate the radiolabel under natural light and temperature conditions.
During this 4-h incubation period, the 14C-bicarbonate would have been assimilated by any photosynthesizing primary producers present in the flasks. It is not possible to label the autotrophs in isolation, because any heterotrophs also present in the flask could consume the labelled autotrophs and themselves become labelled with the 14C. Similarly, the 3H-thymidine would be taken up by bacteria, and any organism that consumed the bacteria would also be labelled. Thus, the 14C experiments measured the feeding of copepods on autotrophs and larger heterotrophs (but not heterotrophic bacteria), while the 3H experiments measured feeding strictly on heterotrophs (including bacteria) but not autotrophs.
During the 4-h incubation period, copepods were pipetted from the aquaria and sorted under a dissecting microscope. Approximately 200 copepods of all species and stages were collected into separate 250-ml tissue culture flasks and were left to acclimate for 30–60 min in the temperature-controlled room. This density of roughly one copepod ml–1 was considered appropriate due to the short duration for the experiments: the largest copepods added were late-stage copepodites, which are anticipated to clear roughly 20 ml d–1, or 0.4 ml 30 min period–1 (based on the results of A. tonsa in Berggreen et al. 1988). Since the assemblage of copepods consisted of all stages, the average clearance rate of each is expected to be much less than 0.4 ml 30 min period–1 and therefore food limitation should not be a concern in these experiments. Preliminary studies indicated that there was no increase in clearance rate as the acclimation period increased; therefore 30 min was considered sufficient.
At the end of the 30-min acclimation period (which coincided with the end of the 4-h food incubation period) the radiolabelled food was poured into the flasks containing the acclimated copepods, which were left to feed for 30 min (the approximate gut passage time), with one inversion after 15 min to re-suspend the food particles. The amount of label present in the copepods at the end of the feeding period therefore represented the amount of food ingested during that period, and the amount of radiolabelled food that would have passed through the digestive tract was minimal.
A total of four experimental replicates were performed for each radioisotope. 14C-bicarbonate experiments were performed on 3, 5, 9, and 11 September, and 3H-thymidine experiments were conducted on 4, 13, 16, and 17 September 2002. On each date, one flask for each size fraction of food was set up (for a total of 6 flasks with roughly 200 copepods in each). The intention was to pool the results from each of the four dates for each isotope.
After the 30-min feeding period, all experiments were terminated by first adding 100 µl of alkaline Lugol's solution (which anaesthetizes the copepods, thus minimizing the egestion of gut contents upon preservation), then 5 ml of formalin was added to kill all organisms in the experimental container. Within 24 h of termination of the experiment, the animals and food were sorted and counted. The contents of the flask were filtered through a 60-µm mesh and the copepods collected on this mesh were rinsed with copious amounts of filtered seawater (in the case of the 3H-thymidine experiments) or rinsed and fumed over acetic acid (for the 14C-bicarbonate experiments) to remove any residual label, and were then backwashed into separate containers for further sorting. Filtered water (1 ml) was put into a 20-ml scintillation vial for counting (to determine the total amount of label added to the water) and two replicates of 10 ml (to determine the amount of label present in the food) were filtered on to a 0.2-µm polycarbonate (Osmonics Co.) filter. The filter was rinsed and fumed over acetic acid as for the copepods noted above. The rinsed and fumed filters were then put into 20-ml scintillation vials for counting.
Subsamples of the potential food items were also collected onto filters and stained for epifluorescence microscopy in order to obtain a visual observation of the food types present in each size fraction. Samples for the enumeration of bacteria were collected by filtering 1 ml of water from the incubation flasks on to a black 0.2-µm polycarbonate filter (Osmonics) and stained for 3 min with Acridine Orange (Sherr et al., 1993). In order to observe the larger autotrophs and heterotrophs, 10 ml of water from the flasks was filtered on to a 0.8-µm black polycarbonate filter and stained for 10 min with DAPI (4',6-diamidino-2-phenylindole) (Sherr et al., 1993). Thus, when examined under epifluorescent microscopy, all particles would fluoresce a white/blue under UV light, but only the autotrophs would autofluoresce red under green light excitation.
There are some complications that arise with this method. In particular, if a particle autofluoresces it may be because it is autotrophic or it may be a heterotroph that consumed an autotroph. In an attempt to distinguish one from the other, the distribution of autofluorescence within the cell was evaluated: if it was evenly distributed throughout the cell, it was considered an autotroph; if only one or a few discrete packages within the cell were autofluorescent, it was likely that these were consumed particles and the cell was labelled a heterotroph. This difference was not easy to distinguish in many cases and therefore the results of autotrophs to heterotrophs are reported as the estimated percentage of each type in each size fraction. The only way to accurately determine the trophic status of a particle at present is to identify to species, and future experiments determining the identity of copepod diets should take this into consideration.
The copepods were sorted by species and development stage using an Erwin loop under a dissecting microscope. Our aim was to pool 30 individual copepods together in each scintillation vial, but in many cases as few as 6 individuals of one species and stage were present in one experimental flask; in these cases, the total number of each species and stage was recorded and otherwise treated the same as all other species and stages. Similarly, initial results for A. hudsonica nauplii were extremely low and the target number per vial for these nauplii was increased to 50. Once the copepods had been sorted, the total volume in the scintillation vial was topped up to 5 ml (by weight) with 0.2 µm filtered unlabelled seawater; the contents of the vials were sonicated 3 times for 5 s to break the animals apart and expose the gut contents for scintillation counting.
Scintillation cocktail (15 ml) (Sigmafluor Universal LSC cocktail (Sigma Co.)) for liquid samples or ScintiSafe Econo F (Fisher Scientific) for filters was added to each vial (background water, filters, and sonicated copepods). The vials were counted using a Beckman 6500 scintillation counter within 4 months of the experiments.
Clearance rates (CR) of the copepods were calculated using the following equation:
|
|
Data analysis
Since the results varied considerably between days, analysis of variance (ANOVA) was run separately for each date to compare the uptake of radiolabel by the different size fractions of food (two replicates for each size fraction of food on each date). These data were also pooled and analysed for general trends across all dates. Clearance rates for the copepods on each food source were also averaged together across all dates for each isotope. ANOVAs were run in order to determine the difference in clearance rate of each species and stage on the different size fractions of food. The differences of clearance rates between copepodites and nauplii, A. hudsonica and E. herdmani, and between different sizes of food within each species and stage were determined by t-tests. t-tests were also performed in order to determine whether the clearance rates were significantly different from zero.
|
Results |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Radiolabel uptake by food particles
It was anticipated that the total amount of radiolabel incorporated into the potential food sources would increase as the size fraction of the food increased (since the total amount of food present would increase in each larger size fraction); however, this was not observed. In all cases, the amount of label significantly increased from the controls (ANOVA, p<0.0001). In the experiments using 14C-bicarbonate, in all cases the amount of label in the food significantly increased from <2 to <5 µm size fraction (ANOVA, p<0.05, Figure 1A). From <5 to <11 µm, however, in two sets of experiments, the amount of label remained the same, while the label decreased in the other two sets of experiments (ANOVA, p<0.05, Figure 1A). In all cases, the amount of label continued to increase in all subsequent size fractions >11 µm.
|
In the 3H-thymidine experiments, there were fewer general trends across dates. The amount of label in each size fraction varied considerably within each date (Figure 1B; all ANOVA, p<0.05). For example, 4 September and 17 September demonstrated opposite patterns in radiolabel uptake by the different size fractions as the most label was present in the <11 µm size fraction on 17 September, while this size fraction contained the lowest amount of label on 4 September.
The slides containing DAPI and AO stained organisms revealed that a few large (>5 µm) heterotrophic protists present in the control (0.2 µm filtered water) were likely introduced along with the copepods used in the feeding experiments. Qualitatively, in the <2 µm size fraction, the majority of all particles consisted of free-living bacteria, few of which were cyanobacteria, and in the <5 µm size fraction there was the addition of small pennate diatoms (up to 10–15 µm in length, 3–4 µm in width, which allowed them to pass through the mesh). In the <11 µm and <20 µm size fractions, the assemblage of stained particles appeared similar to the <5 µm size fraction, and the <60 µm size fraction contained much larger particles, particularly diatoms and some ciliates.
Quantitatively, the number of all particles stained with DAPI present in the experiments is shown in Figure 2 (unfortunately the slide from September 4 was damaged and we therefore do not have any data for this date). Since DAPI stains DNA, these numbers represent all living particles present. The numbers of particles in each size fraction varied considerably between the experiments, but there was no one day that had consistently high or low numbers of food particles.
|
The September 5 slides were examined in more detail to determine the relative proportions of autotrophs and heterotrophs in the experiments and it was found that 50% of the particles in the 5–10 and 10–20 µm size fractions were autotrophs, while the other 50% were heterotrophs. The >20 µm size fraction consisted of 75% autotrophs and 25% heterotrophs, but we were unable to determine the trophic status of the 1–5 µm fraction in most cases (it was not possible to determine the distribution of autofluorescence in these small particles). As noted in Methods, this method is not ideal for determining the trophic status of the protists present, but it was sufficient to satisfy us that there were considerable numbers of both autotrophs and heterotrophs present. Furthermore, although only a small proportion of the particles could be identified down to even the coarsest taxonomic level, the fact that we could distinguish some particles as diatoms or ciliates further supported that there was a range of different food types present in the experiments.
Copepod feeding
In the 14C-bicarbonate experiments, for A. hudsonica copepodites and the E. herdmani nauplii and copepodites, feeding began in the <5 µm size fractions and did not significantly increase in the larger size fraction of food (ANOVA, p<0.0001 in all cases) (Figure 3). In contrast, the A. hudsonica nauplii exhibited positive feeding in the <2 µm size fraction and this did not significantly increase in the larger size fractions. However, this average clearance rate for A. hudsonica nauplii feeding on the <2 µm size fractions is strongly influenced by one value (of a total of 6) that was much higher than the others. When this value is removed from the analyses, A. hudsonica nauplii follow the same pattern of feeding as the other species and stages.
|
Since feeding was not significantly different in all size fractions from <5 µm to <60 µm, these clearance rates were pooled to get an average clearance rate and 95% confidence interval (see Table 1). For both A. hudsonica and E. herdmani the clearance rates of the copepodites were significantly higher than those for the nauplii on each size fraction (all t-tests p<0.01), and when comparing the nauplii of each species and the copepodites of each species, E. herdmani had significantly higher clearance rates than A. hudsonica (all t-tests p<0.01). Over all species and stages, the E. herdmani copepodites had the highest clearance rates on the 14C labelled food of 0.119 ml h–1. All species and stages had average clearance rates on 14C labelled food that were significantly greater than 0, indicating all species and stages were feeding on >2-µm autotrophs.
|
As copepods were sorted into the scintillation vials, when large copepodites (CIV and CV) were present, these were put into separate vials from the earlier copepodite stages. Since the sample sizes for these larger copepodites are quite low, it was not possible to perform statistical analyses. However, it was generally noted that the clearance rates for these larger copepodites were consistently much higher than their earlier staged counterparts (nearly double that of the early copepodites for the 14C experiments for both E. herdmani and A. hudsonica (Figure 3A, C)).
In the 3H-thymidine experiments, none of the copepod species or stages demonstrated significantly different clearance rates on any of the different sized food fractions (ANOVA, p>0.05; Figure 4). When the clearance rates were pooled for each species and stage, as above for the 14C experiments, only the copepodites exhibited clearance rates that were significantly different from zero (Table 1). The E. herdmani copepodites consumed approximately 15 times more 14C labelled food than 3H labelled, while the A. hudsonica consumed roughly 4 times more 14C labelled food. When examined individually by size fraction, only the A. hudsonica copepodites exhibited positive feeding on the <2 µm size fraction (t-test, p=0.013, mean clearance rate = 4.0 µl h–1). Neither the A. hudsonica nor the E. herdmani nauplii exhibited clearance rates significantly different from zero (Table 1). As was seen in the 14C labelled experiments, the clearance rates for the later copepodite stages of A. hudsonica were consistently higher than the earlier stages (except on the <11 µm, where the CR of the late copepodites was slightly lower than the early stages). No E. herdmani late copepodite stages were found in any of the 3H experiments.
|
Negative values for clearance rates were initially seen for some copepods; however, these were only significantly less than zero for the E. herdmani nauplii in three treatments of the 3H experiments, and in one case for the 14C experiment. In these cases, the negative values arose because of the subtraction of the radioactivity of the control. Since there were differing numbers of copepods in each scintillation vial, the radioactivity of the vials was plotted against the number of copepods present (Figure 5). In the 14C experiments, regression was not significant (Figure 5A), and therefore the average background of all copepod control vials was subtracted from each experimental vial and was not corrected for the total number of copepods in each vial. However, for the nauplii in the 3H experimental controls, there was a significant regression (Figure 4B, r2=0.54, p=0.036). This regression only affects the clearance rates calculated for the E. herdmani nauplii as there were always >20 A. hudsonica nauplii present. Thus, in the case of the E. herdmani nauplii feeding on 3H labelled food, the average activity per animal in the control was used rather than the total activity. This only served to make the E. herdmani naupliar clearance rates on 3H food less negative and therefore likely a more appropriate approximation of their feeding on this food source. (The data presented in Figure 3D represent these corrected clearance rates.)
|
In all cases, the clearance rates of the copepods were always higher on the 14C labelled food than on the 3H food (the clearance rates were at least double on the 14C labelled foods, and in the extreme case for E. herdmani copepodites the clearance rates were over 10 times higher). Similarly, the clearance rates of the copepodites were always considerably higher than those of the nauplii from 4 to 10 times the rate of nauplii (Figures 3 and 4).
|
Discussion |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The methods used here allowed us to determine the difference in food type, size, and quantity consumed by nauplii and copepodites in Passamaquoddy Bay, New Brunswick. The copepodites of both A. hudsonica and E. herdmani were capable of feeding on both auto- and heterotrophs, although there did appear to be a preference for feeding on autotrophs. The nauplii, in contrast, exhibited significant feeding only on autotrophs. However, it is not possible to determine whether the lack of significant feeding on heterotrophs by nauplii is due to the absence of feeding on these organisms or whether this method was not sensitive enough to detect this feeding. If we extrapolate the results of the 14C feeding experiments, we would expect the E. herdmani nauplii to consume 10% of the 3H labelled food as the copepodites. This means that the nauplii would be clearing 0.8 µl h–1. Even by pooling 30 nauplii, it would not be possible to detect this low level of feeding (the radioactivity of this amount of feeding is less than 1 s.d. of the mean of the controls). It is not therefore possible to conclude here that the types of food consumed by nauplii and copepodites are significantly different, but it is possible to state that heterotrophs did not comprise a significant proportion of the diets of either nauplii or copepodites of both species examined here.
The results here are comparable to those found for adults of E. affinis and A. bilfosa in Gasparini and Castel (1997) as it was found that both species fed on autotrophs when available, but E. affinis consumed greater quantities of heterotrophs when the concentration of autotrophic food particles decreased. In contrast, Merrell and Stoecker (1998) found that all stages of Eurytemora affinis fed preferentially on heterotrophs. Since the results presented here only represent feeding over a few weeks, it is not possible to conclude that these species always preferentially consume autotrophs, but it is clearly of interest to determine when and under what circumstances they switch from one food type to another. Diet-switching by copepods has been demonstrated previously by Kiorboe et al. (1996) and the comparison of these results indicates that both Acartia and Eurytemora are capable of consuming both autotrophic and heterotrophic prey, but it is not yet entirely clear when or why they switch from one food source to the other. The method presented here offers a less time-consuming alternative to the traditional methods of calculating diet preferences of copepods and thus may help us elucidate these trends.
In addressing the issue of size of food consumed, the copepods fed on 14C labelled food indicated that the smallest size of food particles consumed by the nauplii appears to be comparable to that of the later stages, which is consistent with the results of Berggreen et al. (1988). Although the A. hudsonica nauplii did appear initially to be feeding on <2-µm autotrophs, as noted in Results this is dependent on one observation; we are therefore reluctant to conclude that A. hudsonica nauplii have a significantly different feeding strategy from the other species and stages. Thus, there is not a simple proportional relationship between predator and prey size within copepod species (as has been noted between species; Hansen et al. 1994). Although there may be a difference in optimal size of food consumed by the nauplii and copepodites (as observed by Berggreen et al., 1988), it was not possible to determine this with the methods used in the present study. Overall, it appears that nauplii are consuming comparable food types and sizes as copepodites.
These results are not consistent with the findings of Turner and Tester (1992) and by Roff et al. (1995), who observed copepod nauplii feeding on bacteria. In the present study, only the A. hudsonica copepodites exhibited positive feeding on heterotrophic particles <2 µm in size, while no other species or stage fed on particles this small. Similarly, as noted above, although we cannot conclude that nauplii were not feeding on heterotrophic particles, the fact that this feeding was not detectable indicates that bacteria cannot be a significant component of the naupliar diet. The results of Berggreen et al. (1988), however, do support the findings here; these authors did not find significant feeding by any copepods on particles <2 µm. This discrepancy between the present experiments and Turner and Tester (1992) could be explained if nauplii are capable of picking up bacteria, but not in sufficient quantities to be detected as positive clearance rates in the present experiments. The methods used by Turner and Tester (1992) measured the disappearance of fluorescently labelled bacteria after nauplii were allowed to feed for several hours, which resulted in very high clearance rates (up to 0.2 ml h–1 for Acartia tonsa NI nauplii, which are otherwise known to be non-feeding). Roff et al.'s (1995) visual observation of feeding on bacterial-sized particles therefore does not necessarily contradict the findings here. However, the results here do suggest that bacteria are not a major food source for copepods.
Our findings raise several questions. Finlay and Roff (unpublished data) have demonstrated that the growth rates of copepod nauplii in this system (and thus presumably feeding on the same general food sources), although highly variable, were not related to chlorophyll concentration and therefore it was anticipated that this was due to naupliar feeding on heterotrophic organisms. Now that it is observed that nauplii do not appear to feed heavily on heterotrophs, how does one explain this observation? It appears contradictory that naupliar growth is independent of chlorophyll concentration while their primary food source is autotrophic. Although one explanation for this discrepancy may be that the nauplii are switching their diet, the fact that the present radioisotope experiments were performed in the same location during the same approximate time of year (late summer) as the growth rate experiments suggests that there may be some other explanation. One possibility is that the nauplii are selectively consuming one type of autotroph (i.e. chlorophytes only) and that their diet cannot be generalized by measuring Chl a concentration or autotroph abundance.
It is also possible that the nauplii may be feeding on other, non-conventional food sources. Transparent exopolymeric particles (TEP) (Prieto et al., 2001; Ling and Alldredge, 2003), fecal material (Green et al., 1992), and detritus (Roman, 1984) are all potential food sources for copepods (none of these studies examined feeding by nauplii). J. Dutz (pers. comm.) did find significant feeding of copepod nauplii on TEP (although clearance rates were much lower than those on algae). Furthermore, Dutz's experiments showed that copepod nauplii reared with only TEP survived much better than those without any food. Thus, other potential food resources for nauplii which were not considered in these experiments may be confounding the interpretation of our results.
When examining the quantity of food consumed, the clearance rates observed here for both nauplii and copepodites are considerably lower than those previously published. The clearance rates observed in this study are generally approximately 10% of the values published by Berggreen et al. (1988) and Merrell and Stoecker (1998) for similar species. This may be due to the methods used here. Although the use of radioisotopes is expected to give accurate ingestion rates of the food sources (as it is not subject to the potential counting errors associated with the more traditional methods), the method only allowed us to calculate rates over a 30-min period. It has been demonstrated that regardless of food availability and vertical migration, the feeding rate of copepodites and adults of Calanus spp. increased at night (Huggett and Richardson, 2000; Dale and Kaartvedt, 2000) and that feeding rates can be anywhere from 2 to 10 times greater at night (Peterson et al., 1990). Although this issue has not been specifically tested for copepod nauplii, their behaviour may well be the same as for later developmental stages. Therefore, since previously published clearance rates were conducted over a 24-h period, they may more accurately represent the true clearance rates of the copepods over the long term. It is also possible that although the densities of copepod used in the feeding studies here were not expected to deplete the food resources, the copepods may have interfered with each other directly due to overcrowding. This could be an alternative or additional explanation for the low clearance rates observed here.
Clearly, some of these questions can be answered by comparing the methods here with the results obtained by conventional methods. These experiments have also been performed and will be published elsewhere. The experiments were set up by examining the difference in food particle concentration before and after a 24-h incubation. Therefore a comparison between the conventional and radioisotope methods will allow us to determine how much lower the observed clearance rates are (in order to discount the possibility that these low CR are just a function of the different species used in these experiments). However, the conventional methods are not a direct equivalent of the methods presented here. First, the radioisotope experiments were performed for only 30 min, while the conventional experiments were performed over 24 h. As noted above, the feeding of copepods fluctuates during the day, so the differences in the results of these feeding studies may be due to the different methods, or they could be a function of diel feeding cycles. Similarly, the grazing containers used in each experiment are different volumes: 250 ml was used for the radioisotope experiments to reduce the amount of radioisotope used, while 2 l was used for the conventional methods, which was necessary in order to measure Chl a concentrations accurately. Furthermore, the assemblages of copepods were different in each experiment (both copepodites and nauplii were fed together in the radioisotope experiments, while they were separated in the natural clearance rate experiments) and thus both of these factors could also confound the comparison. It is suggested here that although a comparison with traditional feeding methods is warranted, and will be published in the future, it may not provide us with all of the information we are seeking.
Our experiments demonstrate some important general trends in the feeding of nauplii and copepodites of temperate estuarine copepods. First, as observed previously in the literature, naupliar clearance rates are approximately an order of magnitude lower than the clearance rates of the later stages (Paffenhofer, 1971; Merrell and Stoecker, 1998; Meyer et al., 2002). In the present experiments, even though the E. herdmani nauplii had comparable clearance rates to the A. hudsonica copepodites, the difference in clearance rates between the developmental stages within species still follows this trend. Furthermore, it appears that clearance rate continues to increase with body size of the copepod. The absolute amount of food consumed at each developmental stage is important as this will influence the energy budgets for marine systems. If one knows the relative abundance of each developmental stage and has a good knowledge of their diets, one can estimate the relative impact of each stage on each food type.
There are two main conclusions that can be drawn from the experiments presented here. First, in terms of the type of food consumed by the different developmental stages, it appears that nauplii do not feed on significantly smaller particles, nor do they feed on different types of food from copepodites. There is no evidence here that nauplii are feeding heavily on bacteria nor does it appear that they are a stronger link between the microbial foodweb and classical foodweb than the later developmental stages of copepods.
Second, methodologically, the experiments presented here, as a modification of the methods presented in Baldwin and Newell (1991), offer a quicker and easier method than conventional methods of measuring the uptake of autotrophic and heterotrophic particles by copepods. It is suggested, however, that future experiments employing this method should explore the effects of the diel cycle of feeding, and that perhaps the sensitivity should be increased either by adding more radiolabel or by incubating the food sources for a longer period of time with the labels. This would allow us more easily to explore the seasonal and spatial diet changes of all developmental stages of copepods.
|
Acknowledgements |
|---|
We thank G. Puncher for field and laboratory assistance and the staff of the HMSC in St. Andrews, New Brunswick for logistical support and laboratory space. This work was supported by an NSERC operating grant to JCR and an NSERC PGSA to KF.
|
References |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
-
Baldwin B.S. and Newell R.I.E. (1991) Omnivorous feeding by planktotrophic larvae of the eastern oyster Cassostrea virginica. Marine Ecology Progress Series 78:285–301.[Web of Science]
Berggreen U., Hansen B., Kiorboe T. (1988) Food size spectra, ingestion and growth of the copepod Acartia tonsa during development: implications for determination of copepod production. Marine Biology 99:341–352.[CrossRef]
Dale T. and Kaartvedt S. (2000) Diel patterns in stage-specific vertical migration of Calanus finmarchicus in habitats with midnight sun. ICES Journal of Marine Science 57:1800–1818.
Gasparini S. and Castel J. (1997) Autotrophic and heterotrophic nanoplankton in the diet of the estuarine copepods Eurytemora affinis and Acartia bilfosa. Journal of Plankton Research 19:877–890.
Gifford D.J. and Dagg M.J. (1988) Feeding of the estuarine copepod Acartia tonsa Dana—Carnivory vs. herbivory in natural microplankton assemblages. Bulletin of Marine Science 43:3458–468.[Web of Science]
Green E.P., Harris R., Duncan A. (1992) The production and ingestion of faecal pellets by nauplii of marine calanoid copepods. Journal of Plankton Research 14:1631–1643.
Hansen B., Bjornsen P.K., Hansen P.J. (1994) The size ratio between planktonic predators and their prey. Limnology and Oceanography 39:395–403.[Web of Science]
Hopcroft R.R. and Roff J.C. (1998) Zooplankton growth rates: the influence of size in nauplii of tropcial marine copepods. Marine Biology 132:87–96.[CrossRef]
Hopcroft R.R., Roff J.C., Webber M.K., Witt J.D.S. (1998) Zooplankton growth rates: the influence of size and resources in tropical marine copepodites. Marine Biology 132:67–77.[CrossRef]
Huggett J.A. and Richardson A.J. (2000) A review of the biology and ecology of Calanus agulhensis off South Africa. ICES Journal of Marine Science 57:1834–1849.
Kiorboe T., Saiz E., Viitasalo M. (1996) Prey switching behaviour in the planktonic copepod Acartia tonsa. Marine Ecology Progress Series 143:65–75.[Web of Science]
Ling S.C. and Alldredge A.L. (2003) Does the marine copepod Calanus pacificus consume transparent exopolymer particles (TEP)? Journal of Plankton Research 25:507–515.
Merrell J.R. and Stoecker D.K. (1998) Differential grazing on protozoan microplankton by developmental stages of the calanoid copepod Eurytemora affinis Poppe. Journal of Plankton Research 20:289–304.
Meyer B., Irigoien X., Graeve M., Head R.N., Harris R.P. (2002) Feeding rates and selectivity among nauplii, copepodites and adult females of Calanus finmarchicus and Calanus helgolandicus. Helgoland Marine Research 56:169–176.[CrossRef][Web of Science]
Paffenhofer G.-A. (1971) Grazing and ingestion rates of nauplii, copepodids and adults of the marine planktonic copepods Calanus helgolandicus. Marine Biology 11:286–298.[CrossRef]
Paffenhofer G.-A. and Lewis K.D. (1989) Feeding behavior of nauplii of the genus Eucalanus (Copepoda, Calanoida). Marine Ecology Progress Series 57:129–136.[Web of Science]
Peterson W.T., Painting S.J., Hutchings L. (1990) Diel variations in gut pigment content, diel vertical migration and estimates of grazing impact for copepods in the southern Benguela upwelling region in October 1987. Journal of Plankton Research 12:259–281.
Prieto L., Sommer F., Stibor H., Koeve W. (2001) Effects of planktonic copepods on transparent exopolymeric particles (TEP) abundance and size spectra. Journal of Plankton Research 23:515–525.
Richardson A.J. and Verheye H.M. (1999) Growth rates of copepods in the southern benguela upwelling system: the interplay between body size and food. Limnology and Oceanography 44:382–392.[Web of Science]
Roff J.C., Turner J.T., Webber M.K., Hopcroft R.R. (1995) Bacterivory by tropical copepod nauplii: extent and possible significance. Aquatic Microbial Ecology 9:165–175.[Web of Science]
Roman M.R. (1984) Utilization of detritus by the copepod Acartia tonsa. Limnology and Oceanography 29:949–959.[Web of Science]
Sherr E.B., Caron D.A., Sherr B.F. (1993) Staining of heterotrophic protists for visualization via epifluorescence microscopy. In Kemp P.F., Sherr B.F., Sherr E.B., Cole J.J. (Eds.). Handbook of Methods in Aquatic Microbial Ecology(Lewis Publishers, Boca Raton) 213 pp.
Stoecker D.K. and Egloff D.A. (1987) Predation by Acartia tonsa Dana on planktonic ciliates and rotifers. Journal of Experimental Marine Ecology 110:53–68.[CrossRef]
Turner J.T. and Tester P.A. (1992) Zooplankton feeding ecology: bacterivory by metazoan microzooplankton. Journal of Experimental Marine Biology and Ecology 160:149–167.[CrossRef][Web of Science]
Verity P.G. and Paffenhofer G.A. (1996) On assessment of prey ingestion by copepods. Journal of Plankton Research 18:101767–1779.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||