Ecological differences among clones of Daphnia pulex Leydig

Oecologia

Oecologia (Berl) (I981) 5I:162-168

,~ Springer-Verlag 1981

Ecological Differences Among Clones of Daphniapulex Leydig Jaimie M. Loaring and Paul D.N. Hebert Biology Department, Great Lakes Institute, University of Windsor, Windsor, Ontario, Canada, NgB 3P4

Summary. Natural populations of Daphnia pulex that reproduce by obligate parthenogenesis include a number of clones. Studies on two common and two rare clones from southwestern Ontario revealed significant differences in their intrinsic rates of increase, competitive abilities, rates of ephippial egg production, and lifespans. Environmental factors such as temperature and food type had large influences on the rate of increase of each clone. Differences in rates of increase among clones were most pronounced at temperatures higher than those encountered in nature. In general, the covariance of life history traits among clones was high. The outcome of competitive encounters between clones was deterministic and in most cases was unaffected by temperature. Clones with high rates of increase tended to be better competitors than those with low rates of increase.

Introduction

Competitive processes are often viewed as a potent means of limiting species diversity (MacArthur 1972). Support for this view stems from the observation that coexisting species usually show niche separation. It is well established, for instance, that coexisting bird species differ in foraging behaviour and microhabitat use (Cody 1974). Invertebrate communities provide similar instances of niche differentiation. Hutchinson (1951) argued that copepod communities were not random assemblages of species, but that cohabiting species tended to differ markedly in size and that such size differences reduced food overlap. More recent niche analyses (Miracle 1974; Lane 1975; Makarewicz and Likens 1975; Nilssen 1976) have supported the notion that zooplankton communities include species which differ either in size or microhabitat preferences. Laboratory studies have reinforced this conclusion. When closely related species of zooplankton were studied in laboratory microcosms, competitive exclusion was the result (Frank 1952, 1957; Allan 1973; Neill 1975). In the few cases where congeneric zooplankters of similar size coexist in nature, morphological differences are pronounced and the species persist because of differing susceptibilities to predation (Zaret 1972; Kerfoot 1977; Jacobs 1977a, b, 1978; Seitz 1979, 1980). If competitive interactions regularly prevent the coexistence of closely related species, one would anticipate that different clones of a species would not coexist. Certainly it has been vigorously argued that only a single clone should occur in a particular habitat. Williams (1975) concluded that ~ the competitive exclusion principle is ever to apply, it should do so

0029- 8549/81/0051/0162/$01.40

in the case of clones". If clonal diversity exists, one would on this basis expect a checkerboard distribution pattern (Diamond 1975) with a single clone per habitat. Yet in contrast, genetic studies on Daphnia populations reproducing by obligate parthenogenesis have revealed clonal coexistence. Hebert and Crease (1980) identified 22 different clones in the 11 ponds they surveyed in southwestern Ontario. As many as seven clones were found in a pond, but the average number was 2.4. Genetic distances among the clones were large, indicating that clonal diversity had not originated via mutation in each habitat. Separate clones had evidently been introduced and become established. This discovery raises questions about the extent and nature of ecological differences among clones. Two lines of indirect evidence suggest that the clones are not ecological analogues. Clorial distribution patterns are not random - certain clones were restricted to woodland ponds, while others were found in ponds situated in grassland habitats (Hebert and Crease, in press). Temporal shifts in clone frequencies within these ponds provided further evidence of fitness differences. The present study aimed to provide a more direct documentation of the life history differences that exist among clones of D. pulex. Four clones were selected for detailed study of their intrinsic rates of increase and competitive abilities. Data on lifespan and rates of ephippial egg production were also obtained. Materials and Methods

Four of the D. pulex clones recognized by Hebert and Crease (1980) were chosen for study. Two of the clones (r and r were classified as common because they were found in several different ponds. The other clones (/44 and /=/13) were found in only a single pond. Clones 1 and 13 were isolated from the Windsor-1 pond, whereas clone 4 came from the Cedar Springs pond, and clone 6 from the Charing Cross-I population. Clones 1 and 6 coexisted in a number of ponds, while clone 4 coexisted with clones 1 and 6 at Cedar Springs. All 4 clones were isolated from ponds in grassland as opposed to forest habitats. The intrinsic rate of increase for each clone was determined at three temperatures (10 ~ 20 ~ 30 ~ C). The experiments were conducted in controlled environment chambers that provided 24 h light and temperature control to 0.5 ~ C. Clones were maintained under experimental conditions for a minimum of two generations prior to initiation of an experiment. At 20 ~ C and 3 0 ~ experimental runs were begun with 10 female neonates of each clone, while at 10 ~ C 20 neonates of each clone were used. Two runs were conducted at 10~ C, four at 2 0 ~ and three runs at 30 ~ C. In most experiments neonates of the four

163 clones were obtained within a 24 h period. Each neonate was placed in a beaker containing 100 ml of synthetic pond water (for composition see Hebert and Crease 1980). In the first trial at each temperature, each individual was fed 4 ml of aquarium cultured algal suspension every 2 days. This suspension was composed primarily of Scenedesmus (at a n approx, concentration of 175,000 cells/ml) and Kirschneriella (at an approximate concentration of 1,100,000 cells/ml). In subsequent trials animals were fed 4 ml of Chlamydomonas (at a n approximate concentration of 850,000 cells/ml) and 0.5 ml of a liver suspension (3 mg Difco liver extract/ml). In addition to the algae and liver, 1 ml of vitamin concentrate (Ca p a n t o t h e n a t e 700 rag, Vitamin B12 0.08 mg, Thiamin 6 0 m g , Riboflavin 40rag, Nicotinamide 130 mg, Folic acid 330 mg, Choline 500 mg, Putrescine 30 rag, H 2 0 1.01.) was added to each beaker once a week. Once a day, at approximately the same time, the reproductive state (nonreproductive, ephippial, or parthenogenetic) of each female was determined and any young released were counted. In most of the trials, females were maintained only until they h a d released 6 broods, as preliminary studies indicated that t" values were not significantly increased by including later broods. However, in 2 experimental runs on Chlamydomonas at b o t h 2 0 ~ and 30 ~ C, individuals were maintained until death so that clonal lifespans could be determined. During the course of the r experiments at 10~ and 2 0 ~ females occasionally produced an ephippium rather than a parthenogenetic brood. At 30 ~ C only parthenogenetic eggs were produced. Ephippial eggs were not included in the determinations of r. Instead an ephippial egg production rate was determined for each female by dividing the n u m b e r of ephippia she produced by her total n u m b e r of reproductive days. Reproductive days included all those from the day on which a female produced her first parthenogenetic b r o o d or ephippium until she died or the experiment was terminated. The intrinsic rate of increase (more properly the Malthusian parameter) for each female was calculated using the equation x=t

l=~'lxlnxe-"' x=0

where /x= survivorship mx=~/of young/brood t = age at which brood is released. The equation was solved iteratively using a microcomputer. Occasionally a female survived for some period of time, but failed to reproduce. Such individuals were omitted from the analysis. Skewness and kurtosis statistics were calculated for the r values observed for each clone at each temperature. Only one of the twenty-four values was significant at the 5% level indicating that the data sets approximated normal distributions. As a result, a three-way analysis of variance and the accompanying D u n c a n ' s multiple range tests were performed, A two-way analysis of variance was conducted on clonal rates of ephippial production at 10 ~ C and 20 ~ C. Clonal lifespan data were also analyzed by a two-way analysis of variance. All statistical analyses were carried out on an IBM 3031 computer using SAS programs. Competition experiments were conducted by placing ten neonates of two clones together in a glass jar containing 1 1 of synthetic pond water. Fifty ml of a q u a r i u m cultured algal suspension (primarily Scenedesmus) were added twice weekly and 1 ml of vitamin concentrate was added once per week. The regular food additions were supplemented by bacterial and algal growth within the jar as well as by the organic detritus which accumulated over the course of an experiment. Four replicates of every

pairwise c o m b i n a t i o n of the clones were set up at 10 ~ C, 23 ~ C and 30 ~ C. Experiments at 10~ and 3 0 ~ were maintained in controlled environment chambers, while the other trials were maintained at room temperature ( 2 3 + 2 ~ C). Experiments at 10 ~ C were terminated after 120 days, those at room temperature after 80 days and those at 30 ~ C after 60 days. When generation time is defined as half the median lifespan (Hebert 1978), the experiments at 10 ~ C and 23 ~ C were run for approximately two generations, while those at 30 ~ C were run for four generations. Populations were monitored immediately after establishment to ensure that survival of the neonates was complete. Populations at each temperature rapidly increased in numbers and competition for resources was strong for m u c h of the experiment, as evidenced by a m a r k e d reduction in parthenogenetic b r o o d size. U p o n termination of an experiment, clonal frequencies were determined by electrophoresing a sample of 48 individuals according to methods given by Hebert and Crease (1980). Lactate dehydrogenase and phophoglucose isomerase phenotypes were used to distinguish the clones.

Results

Intrinsic Rates oj Increase Table 1 documents the variation in r values observed a m o n g clones maintained both on Chlamydomonas and Scenedesmus. It is striking that on b o t h food sources inter-clonal variation in r is small at b o t h 10 ~ C and 20 ~ C, while at 30 ~ C the differences a m o n g clones are large. At 20 ~ C three experimental trials were carried out on each clone using Chlamydomonas as a food supply, while at 3 0 ~ two trials were done. These trials were made over a three m o n t h period. Despite attempts to standardize environmental conditions, variation in r of specific clones was noted. For example, the r of clone 13 was significantly higher in trial 1 than in later trials at b o t h 2 0 ~ and 30 ~ C. It is worth emphasizing that the variation in r a m o n g the ten individuals included in each trial was ordinarily very small, as evidenced by the low standard error of r values within trials. The three way analysis of variance shown in Table 2, indicates that temperature and food type had major effects on r. Interclonal variation in r was less striking but was significant. The D u n c a n ' s multiple range tests shown in Table 3 make it possible to examine the effect of a single variable on r when all remaining variables are grouped. This test indicates that r values increased significantly over the temperature range of 10 ~ C to 30 ~ C. This increase was largely due to the increased developmental rate at high temperatures which reduced the pre-adult period. In addition, the r values were, on average, significantly higher when the clones were fed Scenedesmus rather t h a n Chlamydomonas. Over the range of environments studied, clone 1 had a significantly higher r than any of the other clones. Clone 4 ranked second, but its r value was not significantly greater than that of clone 13. Clone 6 had the lowest r value.

L(fbs])an Table 4 shows the means and standard errors of the clonal lifespans at 20 ~ C and 30 ~ C when animals were fed Chlamydamonas. The results of two-way analysis of variance and D u n c a n ' s multiple range tests are presented in Table 5 and 6. These analyses show that temperature affects lifespan. Clonal differences in lifespan are also evident. Specifically, clone 1 outlives the remaining clones, none of which differ significantly in terms of lifespan.

164 Table 1, Mean r + SE of the four clones. Trial 1 was carried out using Scenedesmus as food, while trials 2-4 used Chlamydomonas

Temperature

10

Clone

1

Trial

20 4

6

13

1

30 4

6

13

1

4

6

13

1

0.153 0.121 0.088 0.121 +0.003 +0.004 +0.005 -+0.005

0.295 0.299 0.293 0.329 -+0.006 -+0.009 +0.004 -+0.005

0.416 0.367 0.334 0.232 +0.009 +0.010 -+0.023 -+0.03l

2

0.120 +0.007

0.154 0.214 0.191 _+0.012 -+0.013 •

0.281 0.222 0.064 _+0.011 +0.024

0.094 0.096 0.095 -+0.007 -+0.008 +0.007

0.288 -+0.008

3

0.264 0.233 0.191 0.208 -+0.007 _+0.009 -+0.016 +0.013

4

0.210 0.196 0.190 0.163 -+0.007 -+0.007 -+0.015 -+0.009

Pooled values for trials 2-4

0.211 0.214 0,190 0.219 -+0.010 -+0.006 +0.008 -+0.011

Table 2. Three-way analysis of variance of clonal r values

Source

F

2 1 3

1.404 0.490 0.133

401.60 * 280.33 * 25.46*

2 6 3 6

0.210 0.172 0.004 0.064

60.19" 16.43 * 0.80 6.15 *

Error

337

0.589

Total

360

3.067

Main ef]ects Temperature Food Clone

0.254 0.206 0.092 0.083 -+0.030 -+0.020 -+0.013 -+0.036

0.268 0.212 0.088 0.158 +0.016 -+0.015 -+0.011 +0.020

Table 4. Mean lifespan ( + SE) of the 4 clones at 20 ~ C and 30 ~ C when fed Chlamydomonas

Type 1 SS

Model

DF

0.195 -+0.007

Clone

20 ~ C

30~

1 4 6 13

50.85 + 5.75 34.75 _+4.10 33.30 + 6.44 31.25+2.96

19.25 + 17.75 + 13.00 + 18.75+

1.04 1.27 1.29 1.59

Interactions Temp. Temp. Food Temp.

x'food x clone x clone x food x clone

Table 5. Two-way analysis of variance of clonai lifespan values

Source

Type 1 SS

F

1 3 3

16292.99 3564.67 1838.62

63.44* 4.63** 2.39

Error

150

38528.29

Total

157

60226.68

Model *p