Bartholmé, silvia, laryssa samchyshyna, barbara santer, and winfried lampert. subitaneous eggs of freshwater copepods pass through fish guts: survival, hatchability, and potential ecological implications. limnol. oceanogr., 50(3), 2005, 923–929

Limnol. Oceanogr., 50(3), 2005, 923–929 ᭧ 2005, by the American Society of Limnology and Oceanography, Inc.
Subitaneous eggs of freshwater copepods pass through fish guts: Survival, hatchability,and potential ecological implications Silvia Bartholme´, Laryssa Samchyshyna, Barbara Santer,1 and Winfried LampertMax Planck Institute for Limnology, Postbox 165, D-24302 Plo¨n, Germany We measured the survival of subitaneous eggs of two calanoid (Eudiaptomus gracilis, E. graciloides) and two cyclopoid (Cyclops abyssorum, Macrocyclops albidus) freshwater copepods after they had been consumed by fish.
Unexpectedly, over 80% of the calanoid eggs and 30–59% of the cyclopoid eggs were morphologically intact infish feces. Subitaneous eggs of E. graciloides showed similar proportions of gut passage as dormant resting eggs.
About 70–80% of the calanoid eggs and 35–50% of the cyclopoid eggs released nauplii within 3 d. Hence, a totalof 50–70% of the calanoid eggs and 11–29% of the cyclopoid eggs survived ingestion and gut passage. Survivalwas slightly higher because of shorter gut passage time when the fish had been prefed natural plankton comparedwith hungry fish. We interpret digestion resistance of subitaneous eggs in copepods as an adaptation to fish predationon egg-carrying females.
Females of most freshwater copepod species do not spawn membrane (Hairston and Olds 1984). Population dynamics their eggs freely, but carry them in egg sacs attached to the models (Argentesi et al. 1974; Threlkeld 1979; Taylor and genital segment (Einsle 1993). Egg-carrying females are Slatkin 1981) assumed a close correlation between mortali- more conspicuous and possibly constrained in their escape ties of females and eggs. However, Marcus (1984) had al- ability, which renders them more prone to fish predation than ready noticed that more than 50% of the subitaneous eggs their nonovigerous conspecifics (Brooks and Dodson 1965).
of the marine copepod Labidocera aestiva that had been in- Hence, the often-observed skewed sex ratio of adult cope- gested by worms were still viable in the fecal pellets. Various pods has been suggested to be a consequence of increased qualitative studies reported the presence of intact copepod fish predation pressure on females (Sandstro¨m 1980; Flink- eggs in fish guts, although it was not always clear if these man et al. 1992). Winfield and Townsend (1983) and Hair- were subitaneous or resting eggs (reviewed in Conway et al.
ston et al. (1983) confirmed this mechanism experimentally 1994). Although the proportions of subitaneous eggs that when they showed that fish had a greater reaction distance passed through fish guts were not determined, eggs of brack- (i.e., higher predation efficiency) for egg-carrying copepod ish water copepods were found to be viable within fish feces.
females. Vuorinen et al. (1983) observed a preference of Redden and Daborn (1991) found that 90% of Eurytemora sticklebacks for ovigerous Eurytemora at high densities, and eggs that had passed through the gut of Menidia successfully inferred selective predation from the more pronounced diel were viable and able to hatch, which was not different from vertical migrations of the ovigerous copepods. Hairston et the controls. Hatching success was lower (about 60%) in al. (1983) estimated the costs in terms of population growth other studies (Flinkman et al. 1994; Saint-Jean and Pagano rate of carrying subitaneous eggs that develop immediately 1995). The most comprehensive and quantitative study with to be high, and they suggested the production of resting (dia- six marine copepod species fed to larval turbot is by Conway pausing) eggs that develop later as an alternative strategy.
et al. (1994). They found large differences between copepod Like other zooplankton (rotifers, cladocerans), some co- species. Between 21% and 94% of subitaneous eggs passed pepods can produce diapausing eggs that are resistant to un- through the fish gut undigested. After gut passage, eggs of favorable environmental conditions and can survive in dor- one group of copepods exhibited 67–92% viability (similar mancy for many years (Hairston 1996). Resting eggs have to controls), but only 1.1% and 1.5%, respectively, hatched a durable outer shell (Hairston and Olds 1984) or are en- in two of the species. Although digestion resistance of sub- closed in a chitinous case (ephippium of cladocerans). It has itaneous eggs seems to be widespread in marine and brackish been demonstrated that passage through the gut of a fish copepods, there is little information on freshwater species.
(Mellors 1975; Hairston and Olds 1984) or invertebrates Gliwicz and Rowan (1984) found viable subitaneous eggs (Marcus 1984) is among the harsh conditions that they can of Cyclops abyssorum tatricus in the hindguts of brook char.
withstand. On the contrary, subitaneous eggs were not con- They suggested that eggs passed through the fish intestines sidered to be digestion resistant, as they have a thinner outer unharmed, and this mechanism enabled coexistence of thecyclopoids with fish. Some nauplii of Eudiaptomus gracilis have been observed hatching from feces of vendace (Flink-man et al. 1994), but quantitative studies are lacking.
Digestion resistance of marine copepod eggs has been in- We thank Sonja Dembeck for technical support and Heidrun Feuchtmayr, Kirsten Kessler, Martina Zeller, Heinz Brendelberger, terpreted as an important energy loss to larval fish feeding and two anonymous reviewers for valuable comments.
on copepods (Conway et al. 1994). This may not be so im- L.S. received a fellowship of the German Academic Research portant in freshwaters, as young fish there have sufficient alternative (and preferred) prey, e.g., cladocerans. However, Table 1. Origin of the copepods. Copepods carried subitaneous (sub) or resting (rest) eggs. Origin of predators was Scho¨hsee for Perca fluviatilis and Vierer See for Gasterosteus aculeatus. if digestion resistance of subitaneous eggs is also common copepods were freshly collected by a plankton net when they among freshwater copepods, this will have consequences for were abundant and carried eggs in the field (Table 1). Plank- our understanding of zooplankton dynamics and life histo- ton samples were brought to the laboratory immediately and ries: (1) Present population dynamics models may underes- were sorted under a dissecting microscope. Egg-carrying fe- timate copepod birth rates. (2) Competitive relationships be- males were kept in 1-liter glass jars in the same temperature tween cladocerans and copepods may be altered if copepod controlled room as the fish. They were kept in the laboratory eggs survive gut passage, but cladoceran eggs do not. Gli- for a maximum of 6 d, fed cultured flagellates (Cryptomonas wicz and Rowan (1984) may be reporting an extreme ex- spec.), and transferred to fresh, filtered lake water every oth- ample for high-mountain lakes dominated by Daphnia in the er day. E. graciloides were collected from two sites at dif- absence of fish, but by Cyclops in the presence of fish. (3) ferent dates to obtain females with both subitaneous and rest- Subitaneous eggs may play a role as dispersal agents being ing eggs. An earlier study by Zeller et al. (2004) had shown transported in the guts of vertebrates.
that the eggs carried by E. graciloides in Selenter See in Fish predation is important both in the sea and in fresh- November were almost exclusively resting eggs. Although waters. Hence, egg-carrying copepods should have evolved the two egg types cannot be discriminated microscopically, similar strategies, one of them being digestion resistance of the later hatching experiments confirmed that the eggs col- their eggs. We tested the hypothesis that digestion resistance lected in November were in fact resting eggs.
of subitaneous eggs is important in freshwater copepods. We We performed two types of experiments with ‘‘hungry’’ predicted that investment into egg resistance will vary be- and ‘‘prefed’’ fish. In preparation for an experiment, fish tween copepod species in relation to their exposure to pred- were individually kept in a 16-liter aquarium with freshly ators and to other life-history traits. Therefore, we studied filtered, aerated lake water. They were fed chironomid larvae pelagic and littoral species, as well as species with small and and feces were removed frequently. After 2 d, they were large clutches. As an ‘‘internal standard’’ we compared sub- transferred again into clean water. One group (hungry) was itaneous and resting eggs of the same species.
left unfed for 3 d while the second (prefed) was offered amixture of Daphnia, frozen Artemia, copepods without eggs, and small amount of chironomid larvae to mimic a naturaldiet.
We offered egg-carrying females of two calanoid and two Before each feeding trial, copepod females were isolated cyclopoid freshwater copepods to their natural fish predators.
and their egg numbers determined. Small subsamples of the Three pelagic copepods, the calanoids E. gracilis (Sars) and copepod populations were transferred into a petri dish and E. graciloides (Lilljeborg) and the cyclopoid C. abyssorum the copepods were gently narcotized with a few drops of (Sars), were fed to juvenile (8–10 cm) perch (Perca fluvia- carbonated water. Egg-carrying females were then sorted un- tilis L.). The littoral cyclopoid copepod Macrocyclops albi- der a dissecting microscope and transferred into small glass dus (Jurine) was fed to adult (3–5 cm) three-spined stickle- beakers, taking care that they recovered quickly. As eggs per back (Gasterosteus aculeatus L.). All copepods and fish female could not be counted on live copepods, a random originated from meso- to eutrophic lakes in the vicinity of subsample of 30 females from each population to be offered Plo¨n (Schleswig-Holstein, Germany) where they coexist (Ta- to the fish was isolated, egg sacs removed by needles, and ble 1). With the exception of M. albidus, which had been eggs counted. Average (Ϯ1 SD) egg numbers per female maintained as a laboratory culture at the Max Planck Insti- varied greatly between calanoids and cyclopoids. They were tute for Limnology by I. van der Veen, all copepods were much lower for E. gracilis (6.0 Ϯ 1.4) and E. graciloides obtained from the lakes shortly before the experiments. Ju- (4.8 Ϯ 1.5) than for C. abyssorum (36.0 Ϯ 3.5) and M. venile perch were caught in May 2003 and acclimated to the albidus (27.8 Ϯ 3.6). The total number of eggs offered to experimental conditions in a 200-liter aquarium in a tem- each fish was estimated from the mean number of eggs per perature-controlled room (18ЊC, 12 : 12 h light : dark cycle) where all handling, feeding trials, and hatching tests were Resting eggs required a special treatment, as they were carried out under identical conditions. Fish were fed Chi- not recognized as such under the microscope. Of the sus- ronomus larvae ad libitum every second day. Sticklebacks pected resting eggs (collected in November), 5 times 10 were collected at the end of August 2003 and kept in a 30- clutches (200 eggs) were removed from the females and in- liter aquarium under the same conditions as perch. Pelagic cubated in a petri dish with filtered lake water. They were Gut passage of subitaneous copepod eggs monitored during the following 2 weeks for hatched nauplii that produced nauplii in treatments and controls. The differ- or decomposed eggs. Eggs that did not hatch or decay within ence of the latter was defined as egg mortality imposed by 2 weeks were considered resting eggs (Hairston and Munns the gut passage. Only gut passage could be compared for 1984; Santer et al. 2000). None of the 200 suspected resting resting eggs, as they did not hatch by definition. After check- eggs hatched and only two of them decomposed; hence we ing for homogeneity of variances and normality of residuals, considered all E. graciloides eggs collected in November we tested for differences between treatments and copepod from Selenter See to be resting eggs.
species by nested analysis of variance (ANOVA) for the arc- To start a feeding trial, fish that had been adapted to the sin proportions with species and food as fixed factors and respective food conditions were placed into 3 liters of fil- block as nested factor. The ANOVA was followed by a Tu- tered water and the egg-carrying females were added after key–Kramer post hoc test. The effect of gut passage on sur- 15 min. Each fish received either 30 females of a calanoid vival of nauplii was tested by two-way ANOVA, and the species or 20 females of a cyclopoid species. After feeding difference in gut passage of subitaneous and resting eggs by on the copepods for 1 h, each fish was placed into a small one-way ANOVA. All ANOVAs were carried out using the intermediate container while the contents of the aquaria were NCSS statistical package (Hines 2000).
filtered through a 30-␮m gauze to collect all remaining co-pepods and eggs. The aquaria were then filled with clean filtered water and the fish were put back. Hungry fish weretransferred back immediately, whereas prefed fish were of-fered the food mix again for 30 min before they returned to Rather high proportions of consumed subitaneous eggs clean water. The successive treatment was identical for both were recovered morphologically intact from the feces, both groups. To stimulate digestion, some chironomid larvae (10– in the hungry and the prefed fish (Table 2). Percentages of 15 per perch, 5–7 per stickleback) were offered after the eggs that passed through the gut ranged from 80% for cal- feeding trial. Feces appeared 5–7 h after the copepod meal anoid eggs to 30% for M. albidus. The copepod species had when the fish received the food mix, but not until after 17– a marked effect (Table 3). A Tukey–Kramer post hoc test 20 h in the ‘‘hunger’’ treatment. They were frequently col- showed that both calanoid copepods differed from the cy- lected by a pipette until 48 h after the feeding trial. Fish clopoids, but not from each other, whereas there was a sig- were then returned to the holding tank, and the contents of nificant difference between the cyclopoids. Gut passage was the experimental container were again screened through a slightly lower for hungry fish in all copepods. Resting eggs gauze to collect eggs that might have been released from of E. graciloides passed through the gut of perch signifi- feces by fish action. The feeding experiments were per- cantly better than subitaneous eggs (F formed on successive days (i.e., in blocks). There were five squares [MS] ϭ 0.14, p ϭ 0.012), but there was no effect blocks per species of five replicates each for the ‘‘hungry’’ of the food treatment. The mean (Ϯ1 SD) proportion of rest- treatment and three blocks of three replicates for the ‘‘pre- ing eggs recovered in feces (n ϭ 18) was 86% Ϯ 5.1% (cf.
Table 2 for subitaneous eggs). Within 2 weeks, no nauplii Feces were checked for intact eggs under the microscope.
hatched from 1,961 recovered and 738 control eggs, but Eggs were collected with a pipette, counted, and transferred 2.5% of the recovered and 0.4% of the control eggs decom- to a petri dish containing filtered lake water. Controls were posed. This assured us that the females fed to the fish in this set up with eggs removed from healthy females before the experiment carried exclusively resting eggs.
feeding trial. They were treated in the same way as the eggs In addition, there was a significant effect of copepod spe- removed from the feces. Eggs were monitored once or twice cies (but not food level) on the proportion of subitaneous a day for hatched nauplii that are supposed to appear from eggs that hatched after gut passage. Eggs of the two calanoid subitaneous eggs after 2–3 d at 18ЊC (Einsle 1993). Moni- species were significantly more viable than those of the cy- toring was continued until all eggs had released nauplii or clopoid species, but there was no significant difference with- had decayed. Eggs were considered viable only when nauplii in the two groups. Differential gut passage and viability re- hatched. E. gracilis and E. graciloides were used to test for sulted in an even more pronounced difference between effects of gut passage on the survival of nauplii after hatch- calanoids and cyclopoids for the proportion of ingested eggs ing. Subsamples of 30 hatched nauplii each of treatments that hatched. They ranged from ഠ70% in E. gracilis to only and controls were placed into 1-liter glass jars and fed Cryp- ഠ12% in M. albidus. The effect of food was not very strong, tomonas. Surviving nauplii were counted and transferred into fresh medium every second day for 2 weeks. This ex- However, hatching proportions differed also in the con- periment was repeated three times (i.e., with 90 nauplii trols. Over 90% of the isolated eggs hatched in both Eu- diaptomus species, whereas only ഠ70% hatched in the cy- Numbers of eggs consumed by each fish were estimated clopoids. Pooled proportions are given in Table 2, as there from the difference between the eggs offered and those was no significant difference between the controls of the two found in the aquarium after the feeding trial. They varied, food treatments. The species effect was highly significant but were always between 100 and 180 for the calanoids and ϭ 44.5, MS ϭ 0.35, p Ͻ 0.001) and was due to the between 370 and 720 for the copepods. We were then able difference between calanoids and cyclopoids. Because of the to calculate the proportions of consumed eggs that had species differences in natural mortality of eggs, the differ- passed the gut morphologically intact, the proportion of vi- ences in hatched proportions of ingested eggs alone do not able eggs in the feces, and the proportion of consumed eggs reflect the true effect of gut passage. Hence, we calculated Table 2. Gut passage and survival of subitaneous eggs of the four copepod species offered to prefed or hungry fish. Means (Ϯ1 SD) the mortality imposed by gut passage as the difference be- Saint-Jean and Pagano 1995), subitaneous eggs of all four tween proportions in treatments and respective controls.
freshwater copepod species passed through the gut of fish in There is still a significant species effect due to higher mor- considerable proportions. The range of proportions passing tality in cyclopoids, but the food effect is no longer signif- through the gut morphologically intact (83–30%) is similar icant (Table 3). The treatments were, therefore, pooled to to the results obtained by Conway et al. (1994) for marine demonstrate the impact of gut passage on hatching success species. The study by Conway et al. (1994) is the only one (Fig. 1). The smaller the negative impact of gut passage, the that estimated ingestion rates of eggs by the fish, and thus higher the digestion resistance of the eggs. Gut passage re- has quantitative data to compare with ours. All other studies duced hatching success by only 27% in E. gracilis and 37% tested the viability of recovered eggs, but did not know how in E. graciloides, but by 53% in C. abyssorum and 58% in many were lost during gut passage. To evaluate the ecolog- M. albidus. Subitaneous eggs of calanoids are clearly more ical impact of gut passage, however, it is necessary to know how many of the ingested eggs finally produce nauplii, and Once the calanoid nauplii had hatched, gut passage had if these nauplii survive as well as those hatching directly.
no further impact on their survival during the following 2 Nauplii must not only hatch, they must also be able to escape weeks. There was no significant difference between nauplii the feces that are initially covered by a peritrophic mem- hatched from eggs that had passed through the fish guts and brane. All studies so far removed the eggs from gut contents controls. In addition, survival rates of E. gracilis and E. or feces; hence, hatching rates may be overestimated. If re- graciloides nauplii did not differ significantly. Approximate-ly 70% of both species survived in treatments as well as maining inside the peritrophic membrane, hatched nauplii may be trapped and die. The peritrophic membrane dissolvesdepending on environmental conditions and bacterial colo- nization. Nauplii have been observed escaping from brokenends (Conway et al. 1994) as well as from central parts (pers.
As in marine and brackish water copepods (Redden and observ.) of the fecal pellets. It seems justified to assume that Daborn 1991; Conway et al. 1994; Flinkman et al. 1994; feces dissolve faster in the field than in an aquarium, but Table 3. Subitaneous eggs. Results of nested ANOVAs for arcsin-transformed experimental variables with copepod species (spec) and hunger level (food) of fish as fixed factors and experimental block as nested factor. Interactions not shown, as none of them was significant.
* PASS, proportion of eggs recovered after gut passage; VIABLE, viable proportion of recovered eggs; HATCH, proportion of ingested eggs that hatched after gut passage; MORT, egg mortality caused by gut passage in addition to natural (control) mortality.
Gut passage of subitaneous copepod eggs higher proportion of the recovered eggs was viable. Thisresulted in large differences in proportions of eggs that fi-nally hatched after ingestion; whereas 60–70% of the eggsof Eudiaptomus produced healthy nauplii, the figures for thecyclopoids were only between 12% and 30%. It is not clear,however, if this is really specific for the different taxonomicorders or for the larger clutches carried by the cyclopoids.
A comparison with the literature does not yield a decisiveresult. In the study of Conway et al. (1994), none of the eggsof the marine cyclopoid Corycaceus anglicus survived thegut passage, but large differences were found among thecalanoids. Whereas over 60% survived in Eurytemora, sur-vival was less than 1% in Pseudocalanus, although Pseu-docalanus had fewer eggs than Eurytemora. Only the hatch-ing rates of nauplii from eggs recovered from guts or fecescan be compared with the remaining studies. Egg viabilitiesof Eurytemora (Redden and Daborn 1991; Conway et al.
Fig. 1. Hatching success of subitaneous eggs of C. abyssorum 1994; Flinkman et al. 1994) were very similar to our Eu- (C. ab.), M. albidus (M. al.), E. gracilis (E. gi.), and E. graciloides diaptomus results, and the slightly lower values for C. abys- (E. go.) after gut passage (white bars ϭ controls, hatched bars ϭ sorum are consistent with the observations of Gliwicz and treatments) and impact of gut passage on hatching (black bars).
Means (ϮSE) of prefed and hungry fish.
M. albidus showed the lowest survival of all four species.
Only about 12% of the ingested eggs released nauplii. This there may be nevertheless some nauplii mortality, in partic- cyclopoid is a littoral species and, thus, it probably experi- ular among eggs ingested at late developmental stages. The ences a predation pressure different from the pelagic species.
feeding status of the fish may be important in that respect.
To make the predation more realistic, we confronted M. al- Fish fed natural plankton in our experiments produced feces bidus with a littoral predator, stickleback. However, the com- much faster than starving fish. Also, the peritrophic mem- parison of M. albidus with the other copepods is now ham- branes of feces from fed fish were already perforated when pered by the use of a different predator. There is the they were collected. As this is the natural condition, our possibility that sticklebacks are more efficient in digesting estimates of hatching rates are probably not too far from eggs than perch. On the other hand, fish predation may be less important as a selection factor for M. albidus as they Feeding history had a significant effect on gut passage, can hide in the littoral. The large, conspicuous clutches point but not on the viability of passed eggs. Egg recovery was to the same direction. Despite the low survival rate of cy- always lower when the fish had been starved. As eggs stayed clopoid eggs, there are still a reasonable number of nauplii much longer in the guts of hungry fish, there may have been produced after gut passage. Because of the much larger some direct egg loss due to digestion. It is possible, however, clutch sizes of the cyclopoids, the absolute numbers of nau- that more nauplii hatched from eggs at late developmental plii produced per ingested female is rather similar in calan- stages during the prolonged gut residence and were succes- oids and cyclopoids. We did not test the survival of cyclo- sively digested. Although inspection of the feces showed that poid nauplii after hatching, but we have no reason to believe egg sacs were often dislodged from the remaining chitinous that they were less healthy than the calanoid nauplii. The carapaces of the females and eggs were separated, mechan- relation between clutch size and digestion resistance should ical destruction of eggs during feeding or in the gut seems be studied in more detail within and between species. Clutch to be of minor importance, at least in calanoids. Of the four sizes vary with food abundance and the trophic state of a copepods tested, only E. graciloides produces resting eggs lake (Elster 1954; Czeczuga 1959; Santer 1994); hence a that could be compared to subitaneous eggs. The proportion trade-off between egg number and digestion resistance may of resting eggs passing through the gut was significantly be important for the distribution of copepod species.
larger for resting eggs than for subitaneous eggs, but the There are also significant differences between the hatching difference was small with prefed fish. However, there was proportions of calanoids and cyclopoids in the controls.
no significant effect of the food treatment on gut passage of Hatching success of about 70% for detached eggs has also resting eggs, which suggests that the increased losses of sub- been found for Apocyclops panamensis (Saint-Jean and Pa- itaneous eggs in hungry fish are due to longer digestion time.
gano 1995) and for C. abyssorum tatricus (Gliwicz and Although there are no quantitative data on gut survival of Rowan 1984), whereas 99% were reported for Eurytemora resting eggs, they are usually considered digestion resistant (Redden and Daborn 1991). Cyclopoids are generally not (Hairston and Munns 1984; Marcus 1984). Hence the small more sensitive to handling and egg removal than calanoids difference between resting eggs and subitaneous eggs is sur- (B. Santer pers. comm.); thus the lower hatching success must reflect the specific experimental conditions.
Subitaneous eggs of the two calanoid species were clearly Therefore, we have calculated the additional egg mortality more resistant than those of the cyclopoids. Higher propor- caused by gut passage as the difference between treatments tions of the calanoid eggs passed through the guts, and a and controls (Fig. 1). This characteristic is more important from a physiological than from an ecological point of view.
Jean and Pagano (1995) found complete digestion of the Although there is no longer a difference between food treat- eggs of Moina and Diaphanosoma. We performed prelimi- ments, the clear distinction between calanoids and cyclo- nary experiments with Daphnia to test for egg survival, but poids is retained, but the difference between C. abyssorum contrary to ephippia, not a single subitaneous Daphnia egg and M. albidus is no longer significant. Gut passage causes survived gut passage. In our main experiments, fish were fed only 29–38% mortality in the calanoids and 53–55% in the Daphnia in the prefed treatments, but no intact egg was de- cyclopoids. Considering that the cyclopoid hatching success tected in the feces. Hence, under fish predation, copepods in the controls may have been underestimated, these are con- may have an advantage over Daphnia not only because they servative results, i.e., the difference between the two groups are better evaders. In the rather extreme high-mountain lakes may even be larger. Taking all this information together, we (Gliwicz and Rowan 1984), large Daphnia monopolized the conclude that a significant part of subitaneous eggs of fresh- resources in the absence of fish, while only C. abyssorum water copepods, in particular in Eudiaptomus, the most tatricus coexisted with fish. Viable gut passage in copepods, abundant calanoid genus in European lakes, will survive gut but not in Daphnia, may also provide an explanation for the greater recolonization success of copepods compared to Although it may be important for zooplankton ecology Daphnia in lakes with heavy fish predation (Yan et al. 2004).
and evolution, this aspect has rarely been considered until Finally, gut passage of subitaneous eggs can be involved now. Whenever predation on egg-carrying copepod females in dispersal of copepods and the colonization of new habi- is taken into account in population dynamics models (e.g., tats. Resting eggs, in particular of cladocerans, have long Argentesi et al. 1974), they should be modified, as not all been suggested to be dispersal agents after they had been eggs are lost. However, one has to be cautious, as the impact found intact in the hindguts of waterfowl (Proctor 1964).
of gut passage on population dynamics can be easily over- Many copepods (cyclopoids, various calanoids), however, do estimated. Most copepods carry multiple clutches. The total not produce resting stages, but nevertheless colonize small, gain is low if a female is being consumed when it carries a isolated new water bodies quickly. This has been explained late clutch. If it is killed when carrying an early clutch, the by attachment of copepodids and adults to the plumage of actual clutch is saved, but all possible successive clutches birds (Einsle 1993). Though not explicitly stated, gut pas- sage of subitaneous eggs in birds is implied in the experi- Small fitness differences, however, may have implications ments of Proctor et al. (1967), as they report C. vernalis for the evolution of egg-carrying in copepods compared to hatching from duck feces. As birds have been observed feed- free spawning (Hairston et al. 1983; Webb and Weaver ing on zooplankton (Dodson and Egger 1980), the possible 1988). Costs of egg-carrying are associated with increased transport of viable copepod eggs in bird guts might be worth predation mortality, and resistance of eggs to digestion reinvestigation. It would be particularly interesting to know would lower these costs considerably, favoring the carrying if a double gut passage through fish and fish-eating birds is of eggs. The model will probably have to be modified with the new information, or made species specific. Hairston et In connected water bodies, fish themselves can act as dis- al. (1983) based their calculations on complete mortality of persal vector, as has been observed frequently for seeds subitaneous eggs of Diaptomus sanguineus, as they found (Chick et al. 2003). Jarnagin et al. (2000) have studied the no subitaneous but most of the diapausing eggs to pass dispersal of Bythotrephes through resting eggs in fish, but through the gut of sunfish (N. G. Hairston Jr. pers. comm.).
the same should also apply for subitaneous copepod eggs.
As they found egg-carrying in the presence of fish very cost- Using fish as vectors, copepods could easily colonize oth- ly, they suggested the production of diapausing eggs as an erwise inaccessible upstream lakes. They would even be alternative to free spawning. Hairston and Munns (1984) fol- transported across watersheds in the guts of bait fish.
lowed this line, presenting convincing evidence for timing Our experiments and comparisons to the literature suggest of diapause in D. sanguineus as an evolutionarily stable species-specific differences in digestion resistance of subi- strategy. Digestion resistance of subitaneous eggs in E. gra- taneous copepod eggs, but the phenomenon seems to be ciloides may be the reason why resting egg production in more widespread than previously believed. We have dem- this species is not related to fish predation. Instead, E. gra- onstrated various ecological topics where this might be im- ciloides produces resting eggs in late fall when fish predation portant. A survey including more copepod species should ceases (Pasternak and Arashkevich 1999; Santer et al. 2000).
explore patterns in digestion resistance related to taxonomic Costs of egg mortality may not be sufficient to make switch- units, predator type, life-history strategies, or habitat pref- ing to resting eggs profitable as a predator avoidance strat- erences. We may even find local adaptation to the predator egy. The remaining function of the resting eggs may be the avoidance of harsh environmental conditions, contribution tothe resting egg bank (Hairston 1996), and dispersal.
Our results support the idea of Gliwicz and Rowan (1984) that gut passage of subitaneous copepod eggs facilitates the ARGENTESI, F., R. DE BERNARDI, AND G. DI COLA. 1974. Mathe- balance between cladocerans and copepods. Cladocerans, es- matical models for the analysis of population dynamics in spe- pecially Daphnia, are strong competitors of copepod nauplii cies with continuous recruitment. Mem. Ist. Ital. Idrobiol. 31:
and may force copepods to enter diapause (Santer and Lam- pert 1995). To our knowledge, subitaneous cladoceran eggs BROOKS, J. L., AND S. I. DODSON. 1965. Predation, body size, and have never been reported to pass through a fish gut. Saint- composition of plankton. Science 150: 28–35.
Gut passage of subitaneous copepod eggs CHICK, J. H., R. J. COSGRIFF, AND L. S. GITTINGER. 2003. Fish as and the resistance of ephippial eggs to digestion. Ecology 56:
potential dispersal agents for floodplain plants: First evidence in North America. Can. J. Fish. Aquat. Sci. 60: 1437–1439.
PASTERNAK, A. F., AND E. G. ARASHKEVICH. 1999. Resting stages CONWAY, D. V. P., I. R. B. MCFADZEN, AND P. R. G. TRANTER.
in the life cycle of Eudiaptomus graciloides (Lill.) (Copepoda: 1994. Digestion of copepod eggs by larval turbot Scophthal- Calanoida) in Lake Glubokoe. J. Plankton Res. 21: 309–325.
mus maximus and egg viability following gut passage. Mar.
PROCTOR, V. W. 1964. Viability of crustacean eggs recovered from Ecol. Progr. Ser. 106: 303–309.
fish. Ecology 45: 656–658.
ZECZUGA, B. 1959. Oviposition in Eudiaptomus gracilis G. O. Sars ALONE, AND V. L. DEVLAMING. 1967. Dispersal of and E. graciloides Lilljeborg (Diaptomidae, Crustacea) in re- aquatic organisms: Viability of disseminules recovered from lation to season and trophic level of lakes. Bull. Pol. Acad. Sci.
the intestinal tract of captive Killdeer. Ecology 48: 672–676.
Cl. II. 7: 227–230.
REDDEN, A. M., AND G. R. DABORN. 1991. Viability of subitaneous copepod eggs following fish predation on egg-carrying calan- DODSON, S. I., AND D. L. EGGER. 1980. Selective feeding of Red oids. Mar. Ecol. Progr. Ser. 77: 307–310.
Phaloropes on zooplankton of arctic ponds. Ecology 61: 755–
SAINT-JEAN, L., AND M. PAGANO. 1995. Egg mortality through pre- dation in egg-carrying zooplankters. Studies on Heterobran- EINSLE, U. 1993. Crustacea. Copepoda. Calanoida und Cyclopoida.
chus longifilis larvae fed on copepods, cladocerans and rotifers.
Su¨ßwasserfauna von Mitteleuropa. Vol. 8. Gustav Fischer Ver- J. Plankton Res. 17: 1501–1512.
SANDSTRO¨M, O. 1980. Selective feeding by Baltic herring. Hydro- ¨ ber die Populationsdynamik von Eudiaptomus biologia 69: 199–207.
gracilis und Heterocope borealis Fisher im Bodensee-Obersee.
SANTER, B. 1994. Influences of food type and concentration on the Arch. Hydrobiol. Suppl. 20: 546–614.
development of Eudiaptomus gracilis and implications for in- FLINKMAN, J., I. VUORINEN, AND E. ARO. 1992. Planktivorous Bal- teractions between calanoid and cyclopoid copepods. Arch.
tic herring (Clupea harengus) prey selectively on reproducing Hydrobiol. 131: 141–159.
copepods and cladocerans. Can. J. Fish. Aquat. Sci. 49: 73–
, E. BLOHM-SIEVERS, C. E. CACERES, AND N. G. HAIRSTON, JR. 2000. Life-history variation in the coexisting freshwater , AND M. CHRISTIANSEN. 1994. Calanoid copepod copepods Eudiaptomus gracilis and Eudiaptomus graciloides. eggs survive passage through fish digestive tracts. ICES J. Mar.
Arch. Hydrobiol. 149: 353–364.
Sci. 51: 127–129.
, AND W. LAMPERT. 1995. Summer diapause in cyclopoid copepods: Adaptive response to a food bottleneck? J. Anim.
Ecol. 64: 600–613.
VUORINEN, I., M. RAJASILTA, AND J. SALO. 1983. Selective preda- , AND W. R. J. MUNNS. 1984. The timing of copepod dia- tion and habitat shift in a copepod species—support for the pause as an evolutionarily stable strategy. Am. Nat. 123: 733–
predation hypothesis. Oecologia 59: 62–64.
WEBB, D. G., AND A. J. WEAVER. 1988. Predation and the evolution , AND E. J. OLDS. 1984. Population differences in the timing of free spawning in marine calanoid copepods. Oikos 51: 189–
of diapause: Adaptation in a spatially heterogeneous environ- ment. Oecologia 61: 42–48.
WINFIELD, I. J., AND C. R. TOWNSEND. 1983. The cost of copepod reproduction: Increased susceptibility to fish predation. Oec- ologia 60: 406–411.
YAN, N. D., R. GIRARD, J. H. HENEBERRY, W. B. KELLER, J. M.
HINES, J. 2000. NCSS 2000. NCSS Statistical Software.
GUNN, AND P. J. DILLON. 2004. Recovery of copepod, but not JARNAGIN, S. T., B. K. SWAN, AND W. C. KERFOOT. 2000. Fish as cladoceran, zooplankton from severe and chronic effects of vectors in the dispersal of Bythotrephes cederstroemi: Dia- multiple stress. Ecol. Lett. 7: 452–460.
pausing eggs survive passage through the gut. Freshwat. Biol.
ZELLER, M., R. JIME´NEZ-MELERO, AND B. SANTER. 2004. Diapause 43: 579–589.
in the calanoid freshwater copepod Eudiaptomus graciloides. J. Plankton Res. 26: 1379–1388.
ARCUS, N. H. 1984. Recruitment of copepod nauplii into the plankton: Importance of diapause eggs and benthic processes.
Mar. Ecol. Progr. Ser. 15: 47–54.
MELLORS, W. K. 1975. Selective predation of ephippial Daphnia

Source: http://www.aslo.net/lo/toc/vol_50/issue_3/0923.pdf