SEARCH

SEARCH BY CITATION

Keywords:

  • effective population size;
  • Eudiaptomus sp.;
  • microsatellites;
  • Ne/Nc ratio;
  • temporal approach

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In small planktonic organisms, large census sizes (Nc) suggest large effective population sizes (Ne), but reliable estimates are rare. Here, we present Ne/Nc ratios for two freshwater copepod species (Eudiaptomus sp.) using temporal samples of multilocus microsatellite genotypes and a pseudo-likelihood approach. Ne/Nc ratios were very small in both Eudiaptomus species (10−7–10−8). Although we hypothesized that the species producing resting eggs (E. graciloides) had a larger Ne than the other (E. gracilis), estimates were not statistically different (E. graciloides: Ne = 672.7, CI: 276–1949; E. gracilis: Ne = 1027.4, CI: 449–2495), suggesting that the propagule bank of E. graciloides had no detectable influence on Ne.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The effective population size, Ne, summarizes the relative magnitude of random processes (genetic drift) as opposed to deterministic processes (selection) on allele frequencies in a single variable (e.g. Wright, 1931, 1938; Hedrick, 1999). Estimations of Ne from life-history data are difficult if not impossible to obtain in wild populations (Schwartz et al., 1998; Wang & Caballero, 1999; Waples, 2002). Often, estimations from genetic marker data provide the only alternative. From these data, the so-called variance effective size can be obtained by relating the measured variance in allele frequency of a ‘real’ population to that of an ideal population of infinite population size (Hedrick, 1999).

In various species, there are pronounced discrepancies between the ecological size of a population (census size, Nc) and its population-genetic, effective population size (Frankham, 1995). In planktonic organisms, Ne/Nc ratios have rarely been studied so far. However, even if we expect much lower Ne compared with Nc, effective population sizes should be considerable in planktonic species, as specimens are small and numerous, individuals are passively transported by movements of the water, the combination of mating pairs should be near random and dispersal should effectively prevent populations from spatial genetic isolation.

Several ‘temporal methods’ have been developed to estimate Ne from changes in allele frequencies over time (e.g. Krimbas & Tsakas, 1971; Waples, 1989; Williamson & Slatkin, 1999; Wang, 2001). A common assumption of all these approaches is that allele frequencies only change due to genetic drift, whereas the role of migration, mutation and selection is negligible. Temporal methods either follow a moment-based or maximum-likelihood-based approach (ML). The latter was shown to have higher accuracy and precision compared with moment-based approaches (e.g. Williamson & Slatkin, 1999; Berthier et al., 2002; Wang & Whitlock, 2003) and Wang (2001) found that rare alleles are the main cause for the differences in performance. As most natural populations are not completely isolated but open to gene flow, emigration or immigration of alleles by propagules or migrant individuals will change local allele frequencies over time. Recently, Wang & Whitlock (2003) developed a pseudo-likelihood approach to estimate Ne that accounts for both migration and drift, rather than drift alone. Their approach allows us to estimate Ne and the fraction of migrants (m) jointly from temporal samples (Wang & Whitlock, 2003), and will therefore be employed in this study.

Here, we examined Ne of the two planktonic freshwater copepods Eudiaptomus graciloides Lilljeborg and E. gracilis Sars. We estimated Ne from four temporal samples taken from a lake in northern Germany and compared these with the estimates of Nc, which we calculated on the basis of abundances data from earlier investigations (Fußmann, 1996; Hofmann, 1979). Further, we compared Ne between E. graciloides and E. gracilis. Although both species co-occur in a number of lakes in northern Europe and reveal strong morphological similarities (Kiefer, 1978; Nauwerck, 1980; Einsle, 1993), they exhibit marked differences in their life cycle, namely the presence of a diapausing egg bank in one, but not in the other species (Santer et al., 2000). Long-term diapausing eggs can remain viable over decades and establish resting propagule banks (Hairston et al., 1995; Cáceres, 1998), which can slow down the rate of microevolution (Hairston & DeStasio, 1988). Hence, ‘old’ genotypes are stored in the sediments and the recruitment of stages that have remained dormant over longer time periods should contribute to the maintenance of genetic diversity of today’s populations (reviewed in Bilton et al., 2001). Here, we tested the hypothesis that E. graciloides has a larger effective population size compared with its cogener E. gracilis due to the existence of a resting propagule bank in the former species. Additionally, we explored different scenarios of likely generation times in Eudiaptomus species combined with different source populations, from which migrants were received, to investigate how these differences might influence the estimates of Ne and m.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Sampling

Population samples of each Eudiaptomus species were taken from Lake Schöhsee (northern Germany, 54°09′N, 10°26′E) at four sampling dates: March 2002, March/April 2003, August 2003 and March 2004, representing a total of approximately four to six generations (see below). Samples were taken in vertical hauls with a plankton net and stored in 70% ethanol until further processing. As samples were taken from populations apparently much larger than the sample sizes (Waples, 1989; plan 1), we assume that our sampling did not change the available pool of reproductive individuals (Wang & Whitlock, 2003).

DNA extraction and microsatellite analysis

Female copepods (n = 177/sampling date/species) were isolated with dissecting needles and DNA was extracted with the Invisorb DNA Tissue HTS 96-Kit/C from Invitek (Berlin, Germany) following the manufacturer’s instructions. For estimating allele frequencies, we used the polymorphism displayed at three and seven microsatellite loci for E. graciloides and E. gracilis respectively (EGO2, 7 and 10, GenBank accession numbers AY547392AY547394; EGI1, 3, 8, 12, 13, 17 and 35, GenBank accession numbers AY547395AY547401, Zeller & Reusch, 2004). In the present data set, microsatellite loci represented a total of 119 alleles in E. graciloides and 292 alleles in E. gracilis. Microsatellite procedures were carried out as described in Zeller et al. (2006).

Census population size

We made educated guesses on census population sizes (Nc) for both Eudiaptomus species using two studies, one by Hofmann (1979) and a more recent one by Fußmann (1996). Hofmann (1979) sampled adult E. graciloides and E. gracilis from Lake Schöhsee from April to December 1974 biweekly and calculated their abundances. We averaged these abundances (Hofmann, 1979; Fig. 1) for each species (E. graciloides: 0.062 × 106 individuals per m2; E. gracilis: 0.025 × 106 individuals per m2). Considering a lake area of 82.3 ha for Lake Schöhsee (Rai, 1982), estimates of census population sizes were 5.1 × 1010 and 2.1 × 1010 for E. graciloides and E. gracilis respectively. Fußmann (1996) reported abundances for Eudiaptomus sp. in Lake Schöhsee of 5–60 copepods per litre in 1993/1994, corresponding to 5.4 × 1010–6.5 × 1011 copepods in the entire lake, well within the range of previous census sizes. To accommodate for spatial and temporal patchiness, we conservatively assumed census sizes in the order of magnitude 1010.

image

Figure 1.  Log-likelihood curves for effective population sizes (Ne) in Eudiaptomus graciloides and E. gracilis. Msat, microsatellite; source, numbers of populations that were pooled as source population. The 95% confidence interval for the method implemented here can be calculated as the range of support associated with a drop of two units of log-likelihood (y-axis).

Download figure to PowerPoint

Genetic differentiation between sampling dates

In order to quantify temporal genetic differentiation, we calculated pairwise FST values between sampling dates with genetix v. 4.01 (Belkhir et al., 1998), using the estimator θ (Weir & Cockerham, 1984). Allele frequency distributions and changes in allele frequencies between sampling dates are presented in the Supporting Information for all microsatellite loci.

Effective population size (Ne) and fraction of migrating copepods (m)

In order to estimate Ne in Eudiaptomus, we used the temporal approach developed by Wang & Whitlock (2003) implemented in MLNE (available at http://www.zoo.cam.ac.uk/ioz/people/wang.htm). This method calculates the variance effective population size, Ne, together with m (migration) using a pseudo-likelihood approach, and is based on the assumption that changes in allele frequencies over time are caused by drift and migration, rather than drift only. Therefore, using this method migration sources and gene flow must be known a priori. Per definition, temporal samples are taken from a ‘focal population’, which is receiving migrants from an infinitely large ‘source population’ (Wang & Whitlock, 2003). We calculated Ne for E. graciloides and E. gracilis inhabiting Lake Schöhsee in northern Germany (focal populations). To define source populations, we used information on allele frequencies, and hence pairwise genetic differentiation, from 14 and 15 populations of E. graciloides and E. gracilis, respectively, sampled in 2002/2003 from lakes situated on a longitudinal gradient between northern Germany and Lithuania (data from Zeller et al., 2006). Genetic exchange between these populations and the Schöhsee populations differed in intensity. Thus, we rather chose to define two different source populations a priori in order to investigate how these differences might influence our estimates of Ne and m. In one source population, we included only three nearby populations from lakes which were interconnected by waterways (Großer Plöner See, lake area: 2910 ha, mean depth: 16 m; Trammer See, lake area: 163 ha, mean depth: 12 m; Kellersee, lake area: 560 ha, mean depth: 14 m; Muuss et al., 1973) and in a second one we included information from all available populations (Table 1, for further information on source populations see Zeller et al., 2006). Subsequently, we combined the two differently defined source populations with different possible generation times of Eudiaptomus sp. in our estimations. As both Eudiaptomus species are likely to exhibit two to three generations per year (Bosselmann, 1975; Hofmann, 1979; Santer et al., 2000), we calculated Ne for two, three and additionally four generations per year. Accordingly, the number of elapsed generations associated with our four sampling dates (March 2002/March, April 2003/August 2003/March 2004) was 0/2/3/4, 0/3/4/6 and 0/3/5/6, respectively, and 0/4/6/8. Allele counts were analysed with the programme genepop v. 3.3 (Raymond & Rousset, 1995) and re-arranged for data input in MLNE. For the comparison between species, we also calculated Ne and m for E. gracilis with the information from three primer pairs only, which represented a similar number of independent alleles (106; independent alleles = total number of alleles − number of loci, Wang, 2001) as the three primer pairs used for E. graciloides (116). The number of independent alleles is an important parameter in the pseudo-likelihood approach, as an increase in independent alleles increases the accuracy and precision of the Ne estimates (Wang, 2001). In our study, we also performed estimations of Ne in E. gracilis using a reduced set of alleles in order to evaluate the influence of the number of loci (and alleles) on the estimation procedure.

Table 1.   Information on a priori defined source populations used in the calculations to estimate Ne and m.
 Eudiaptomus graciloidesEudiaptomus gracilis
Source 1Source 2Source 1Source 2
  1. Source 1: copepods were sampled from lakes located a few hundred metres apart from the Schöhsee (focal) population. Source 2: copepods were sampled from lakes located in a range of a few hundred metres to 1000 km to the Schöhsee (focal) population.

  2. *Values from Zeller et al. (2006).

No. of populations (n)3 (120)14 (612)3 (135)15 (674)
Mean pairwise FST* (SD)0.01 (0.008) 0.02 (0.017)0.05 (0.004) 0.08 (0.018)

Additionally, we present estimates of Ne values for both Eudiaptomus species calculated with a function implemented in the MLNE programme that takes no migration into account (after Wang, 2001; Table 3).

Table 3.   Effective population sizes (Ne) and fraction of migrating copepods (m) for Eudiaptomus graciloides and E. gracilis estimated by MLNE according to a pseudo-likelihood approach.
 GSMigration taken into account (Wang & Whitlock, 2003)No migration assumed (Wang, 2001)
Source 1Source 2
Ne95% CIm95% CINe95% CIm95% CINe95% CI
  1. GS, generation sequence assumed in the calculations. Source 1: formed from three nearby populations (n = 120 and 135 for E. graciloides and E. gracilis respectively). Source 2: formed from 14 populations in E. graciloides (n = 612) and 15 populations (n = 674) in E. gracilis. The number of microsatellite loci used is given in parantheses. Values in bold belong to nonsmoothed log-likelihood curves (see Figs 1 and 2).

E. graciloides0, 2, 3, 4452.9276.3–917.0 0.0450.020–0.083551.8317.4–1258.70.0250.010–0.0481182.2644.5–3861.8
0, 3, 4, 6601.8364.1–1190.00.0320.015–0.058745.5431.4–1609.00.0180.008–0.0331593.4870.6–4977.3
0, 3, 5, 6621.9367.0–1268.00.0310.014–0.057789.4445.5–1729.10.0160.007–0.0321701.1924.5–5338.5
0, 4, 6, 8726.0433.6–1435.60.0250.012–0.046892.5531.4–1949.10.0140.006–0.0262056.11125.7–6178.2
E. gracilis (7)0, 2, 3, 4677.3449.4–1201.50.0220.011–0.035837.3510.8–1697.20.0170.007–0.0292269.81241.9–8411.5
0, 3, 4, 6968.6617.8–1613.50.0150.008–0.0241322.1765.6–2495.30.0100.008–0.0183434.61798.5–16 746.2
0, 3, 5, 6867.3589.9–1412.60.0160.010–0.0261114.8712.5–1987.60.0120.006–0.0192912.11677.2–8152.0
0, 4, 6, 81062.9718.6–1724.60.0130.008–0.0201368.8913.2–2459.60.0090.007–0.0153915.02183.4–12 597.8
E. gracilis (3)0, 2, 3, 4455.5276.9–919.40.0320.015–0.060548.6309.9–1260.60.0210.008–0.0421258.2658.8–4860.8
0, 3, 4, 6728.3418.6–1624.30.0190.008–0.037871.2502.1–2465.40.0130.004–0.0242189.51033.1–20 262.0
0, 3, 5, 6564.8351.6–1037.60.0250.013–0.045672.9407.4–1361.90.0160.008–0.0301505.5848.4–4010.2
0, 4, 6, 8767.3463.8–1486.00.0180.009–0.032962.3558.7–2083.70.0110.005–0.0212198.91164.2–7552.9

Null alleles

Our previous studies indicated that null alleles might be abundant in some of the microsatellite loci (Zeller et al., 2006). Temporal changes in allele frequencies should be little affected by null alleles as long as their occurrence is not markedly different between sampling dates. We used three parameters (frequencies of nonamplified samples for individual loci, FIS values and calculated null allele frequencies, programme Microchecker 2.2.1, Brookfield 2, Van Oosterhout et al., 2004) as indicators for the abundance of null alleles. Null alleles were equally distributed among sampling dates in both Eudiaptomus species. In E. graciloides, FIS values (one-way anova: F3,8 = 0.24, P > 0.5), calculated null allele frequencies (one-way anova: F3,8 = 0.46, P > 0.5) and frequencies of nonamplified samples (Kruskal–Wallis anova: H3,12 = 1.55, P > 0.5) were not significantly different between sampling dates. The same was true for E. gracilis (FIS values: one-way anova: F3,24 = 0.27, P > 0.5; calculated null allele frequencies: one-way anova: F3,24 = 0.13, P > 0.5; frequencies of nonamplified samples: Kruskal–Wallis anova: H3,28 = 0.82, P > 0.5). The highest numerical differences in calculated null allele frequencies between sampling dates were in a range of 0.00–0.08 in E. graciloides and in a range of 0.02–0.06 in E. gracilis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Genetic differentiation between sampling dates

In E. graciloides, all pairwise FST values that included the March 2004 sample differed significantly from zero (Table 2). In E. gracilis, all pairwise FST values differed significantly from zero except for values for the March 2002 vs. March/April 2003 samples and the August 2003 vs. March 2004 samples, independent of the number of microsatellite loci used in the calculations (Table 2).

Table 2.   Pairwise genetic differentiation between sampling dates.
 Eudiaptomus graciloidesEudiaptomus gracilis (7)E. gracilis (3)
  1. The number of used microsatellite loci is given in brackets. FST values that differed significantly from zero are given in bold.

March 2002–March/April 2003−0.00020.00020.0005
March 2002–August 20030.00040.00150.0026
March 2002–March 20040.00270.00130.0030
March/April 2003–August 20030.00060.00580.0103
March/April 2003–March 20040.00350.00460.0097
August 2003–March 20040.00460.0002−0.0010

Effective population size (Ne) and fraction of migrating copepods (m)

For E. graciloides, we estimated Ne values in the hundreds (mean ± SD Ne for all scenarios: 672.7 ± 141.7) with confidence intervals ranging from 276 to 1949 (Table 3). For E. gracilis, mean estimates of Ne calculated with information from all seven microsatellite loci were 1027.4 (±239.1) with confidence intervals between 449 and 2495 (Table 3). Mean estimates of Ne for E. gracilis calculated with three primer pairs were 696.4 (±171.2) with confidence intervals ranging from 277 to 2084 (Table 3). Mean migration rate was 0.03 (±0.01) for E. graciloides, 0.02 (±0.01) and 0.01 (±0.004) for E. gracilis calculated with information from three and seven microsatellite loci respectively (data for single scenarios, see Table 3). For likelihood curves on which ML estimates and confidence intervals were based, see Figs 1 and 2. The occurrence of a lot of private rare alleles can lead to uneven log-likelihood curves (J. Wang personal communication), as they tend to cause difficulties in the computational process. Estimates of Ne increased consistently with increasing generation times for all scenarios within both Eudiaptomus species (Table 3). Ne estimates and the associated confidence intervals were shifted upwards considering more populations in the a priori defined source population (Table 3). In E. gracilis, values of Ne calculated with seven microsatellite loci were higher compared with that in E. graciloides, but confidence intervals were overlapping (Table 3). Estimates of effective population sizes were biased upwards approximately two- to threefold when no gene flow was assumed (Wang, 2001), Table 3.

image

Figure 2.  Log-likelihood curves for the proportion of migrating copepodes (m) in Eudiaptomus graciloides and E. gracilis. Msat, microsatellite; source, numbers of populations that were pooled as source population. The 95% confidence interval for the method implemented here can be calculated as the range of support associated with a drop of two units of log-likelihood (y-axis).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In this study, we present estimates of Ne for two species of copepods (Eudiaptomus sp.) using high-resolution microsatellite markers together with a novel temporal approach that takes migration into account. We found effective population sizes that were several orders of magnitude smaller than their likely census sizes. Although census sizes are often larger than effective population sizes as reported from a number of species (Frankham, 1995), such small Ne/Nc ratios are nevertheless surprising for plankton organisms. One would assume considerable within-lake exchange of individuals due to passive transport within a relatively small water body. Earlier studies had already pointed to small effective population sizes in Eudiaptomus spp. due to the occurrence of high FIS values and departures from Hardy–Weinberg equilibrium in a number of E. graciloides and E. gracilis populations (Zeller et al., 2006). However, it was difficult to decide whether this was caused by a real biological process or, for example, by the presence of null alleles. It is likely that null alleles occurred in the present study, but their frequencies were equally distributed between sampling dates. Therefore, the calculation of Ne and m should not have been seriously affected (see Jehle et al., 2001; Wang & Whitlock, 2003).

In aquatic environments, temporal methods to investigate Ne have mostly been applied to teleost species (e.g. Hauser et al., 2002; Heath et al., 2002; Hoarau et al., 2005). Consistent with our findings in copepods, in these studies, Ne was often surprisingly low, although restrictions to genetic exchange among fish individuals are difficult to imagine, and census sizes are large.

Census sizes (Nc) were not known for populations of E. graciloides and E. gracilis, but it is likely that they are at least in a range of 1010 for each species in the entire lake. Thus, Ne/Nc ratios in E. graciloides and E. gracilis were in a range of approximately 10−7–10−8. In planktonic organisms, Ne/Nc ratios have rarely been studied so far. Only a few investigations have been made in order to estimate Ne from single genetic samples as (N = effective population size, μ = mutation rate) or Nm (m = migration) using the polymorphism of microsatellite loci or the variation of mtDNA sequences (Bucklin & Wiebe, 1998; Pálsson, 2000; Bohonak et al., 2006). Bucklin & Wiebe (1998) measured the long-term effective population sizes for the marine copepods Calanus finmarchicus and Nannocalanus minor, using the variation in mitochondrial DNA sequences. This method relies on the scaling of mutation rates with time, and may be biased due to several historical processes such as vicariance and recolonization history. Their results of much larger Ne values in a range of approximately 108, and effective female population sizes (from nucleotide diversities) of ∼105 are therefore not necessarily at odds with our results. Bohonak et al. (2006) calculated relative population sizes (Neμ) for E. graciloides and E. gracilis from three lakes in northern Germany with information from mitochondrial DNA sequences. Phylogeographic analyses and Bayesian skyline plots resulted in Neμ being rather greater in E. gracilis compared with that in E. graciloides, although the contrast was statistically not significant for most time points, including the most recent one. Thus, these authors found a pattern congruent with our results despite the fact that the markers and methods used to estimate Neμ differed in both studies and that Bohonak et al. (2006) calculated Neμ together for populations from three lakes, only including the Schöhsee. Therefore, it is likely that a more general phenomenon was found in Eudiaptomus.

Small effective population sizes despite much larger census sizes can be caused by substructuring of populations, a skewed reproduction success, unequal sex ratios, nonrandom mating or fluctuating population sizes (e.g. Franklin, 1980; Caballero, 1994; Frankham, 1995; Hedrick, 2005). Earlier studies showed that abundances of both Eudiaptomus species in Lake Schöhsee had fluctuated over the years, what might contribute to low Ne (Hofmann, 1979; Santer et al., 2000). More life-history and demographic information for E. graciloides and E. gracilis would be needed to identify factors that are likely to reduce effective population sizes seven to eight orders of magnitude below their census sizes.

The pseudo-likelihood approach we used in our study accounts for migration and drift, rather than drift alone, and therefore considers a more realistic scenario for estimating Ne in natural populations (Wang & Whitlock, 2003). Nevertheless, source populations must be defined a priori and the reliability of Ne estimates might depend on information available on population structure. Consequently, it can be difficult to decide, which migration model provides the most reliable Ne estimates. Reassuringly for our case study, we can utilize extensive data on Eudiaptomus population structure from the area. Moreover, for both migrant pool scenarios assumed, we observe essentially identical estimates. Under the absence of migration, estimates of Ne were two- to threefold higher for both Eudiaptomus species. Ignoring migration should lead to an overestimation of Ne in cases where constant migration and genetic drift cause populations to approach an equilibrium level of genetic differentiation, and therefore migration slows down the rate of change of allele frequencies (Wang & Whitlock, 2003). However, even if the no migration case was true in our study, Ne/Nc ratios for both study species remain within the same order of magnitude regardless of which migration model was applied.

Estimates of Ne and confidence intervals were both shifted upwards when we included more populations in the a priori defined source populations and when using more generations per year in the calculations. In the short term, drift would generally be overestimated, and hence Ne underestimated, if migration was not taken into account for calculations of Ne (Wang & Whitlock, 2003). In our study, migration from only three surrounding populations seemed not to have reflected all migration events properly, leading to smaller Ne values compared with when larger source populations were taken into account. However, migration was low in both species and even lower in the case of the larger source population, which consisted of 612 individuals of E. graciloides and 674 individuals of E. gracilis. Thus, the impact of gene flow on changes in allele frequencies seemed higher compared with when only three populations are functioning as source population. On the other hand, more alleles seemed to be present in the source population that differed from alleles occurring in the focal population, and thus migration occurred to be lower compared with calculations where smaller source populations were used. Nevertheless, Ne was small and within the same order of magnitude for any of the applied scenarios.

Despite a tendency of Ne estimates to be higher in E. gracilis compared with that in E. graciloides, this difference was statistically not significant. This suggested that genotypes stored in the resting propagule banks of Lake Schöhsee did not affect the size of the water column population of E. graciloides as we initially hypothesized. Ne estimates and confidence intervals were even more congruent when comparable amounts of independent alleles were used in the calculations. Nevertheless, in E. graciloides, a prolonged generation time should be considered due to the hatching of long-term diapausing eggs from the sediments. Bohonak et al. (2006) estimated a mean generation time of 0.48 per year for E. graciloides in Lake Schöhsee from the average of copepods originating from subitaneous eggs and animals that hatched from the egg bank. This generation time approximately corresponded to the generation sequence 0, 2, 3, 4 in our calculations. Eudiaptomus gracilis is most likely to exhibit two to three generations per year (Hofmann, 1979; Santer et al., 2000; B. Santer, unpublished data). Therefore, for E. gracilis, in the present investigation, Ne probably equals an intermediate value of the estimates made for generation sequences 0, 2, 3, 4 and 0, 3, 5, 6/0, 3, 4, 6. Taking prolonged generation times for E. graciloides into account, differences in Ne estimates between both Eudiaptomus species were more pronounced, but still not statistically significant. Thus, within our model system of Eudiaptomus sp. inhabiting Lake Schöhsee, we could not confirm the general assumption of effective population sizes being greater in a species with a resting propagule bank compared with a closely related species that lacks diapause.

Our study is a further data point for an emerging consensus that, in small planktonic species, effective population sizes can be much smaller than expected. It will be interesting to see whether this pattern holds once more species have been rigorously studied using high-resolution genetic markers and temporal variation in allele frequencies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

We thank S. Carstensen, I. Dankert and R. Klein for their laboratory assistance. T.B.H.R. was partly funded by Deutsche Forschungsgemeinschaft (Re 1108/4).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Belkhir, K., Borsa, P., Goudet, J., Chikhi, L. & Bonhomme, F. 1998. genetix, Logiciel sous Windows pour la Génetique des Populations. Laboratoire Génome et Populations CNRS UPR 9060, Université de Montpellier II, France.
  • Berthier, P., Beaumont, M.A., Cornuet, J.M. & Luikart, G. 2002. Likelihood-based estimation of the effective population size using temporal changes in allele frequencies: a genealogical approach. Genetics 160: 741751.
  • Bilton, D.T., Freeland, J. & Okamura, B. 2001. Dispersal in freshwater invertebrates. Annu. Rev. Ecol. Syst. 32: 159181.
  • Bohonak, A.J., Holland, M.D., Santer, B., Zeller, M., Kearns, C.M. & Hairston, N.G. Jr 2006. The population genetic consequences of diapause in Eudiaptomus copepods. Arch. Hydrobiol. 167: 183202.
  • Bosselmann, S. 1975. Population dynamics of Eudiaptomus graciloides in Lake Esrom. Arch. Hydrobiol. 75: 329346.
  • Bucklin, A. & Wiebe, P.H. 1998. Low mitochondrial diversity and small effective population sizes of the copepods Calanus finmarchicus and Nannocalanus minor: possible impact of climatic variation during recent glaciation. J. Hered. 89: 383392.
  • Caballero, A. 1994. Developments in the prediction of effective population size. Heredity 73: 657679.
  • Cáceres, C.E. 1998. Interspecific variation in the abundance, production and emergence of Daphnia diapausing eggs. Ecology 79: 16991710.
  • Einsle, U. 1993. Crustacea: Copepoda: Calanoida und Cyclopoida. Süßwasserfauna von Mitteleuropa, Vol. 8, Gustav Fischer, Stuttgart.
  • Frankham, R. 1995. Effective population size/adult population size ratios in wildlife: a review. Genet. Res. 66: 95107.
  • Franklin, I.R. 1980. Evolutionary change in small populations. In: Conservation Biology: An Evolutionary-Ecological Perspective (M.E.Soule & B.A.Witcox, eds), pp. 1134. Sinauer, Sunderland, MA.
  • Fußmann, G. 1996. Die Kontrolle der Rotatorien im Pelagial eines mesotrophen Sees durch bottum-up- und top-down-Prozesse: Freilandbeobachtungen und Enclosure-Experimente. PhD thesis. Christian-Albrechts University Kiel, Plön.
  • Hairston, N.G. Jr & DeStasio, B.T.J. 1988. Rate of evolution slowed by a dormant propagule pool. Nature 336: 239242.
  • Hairston, N.G. Jr, Van Brunt, R.A., Kearns, C.M. & Engstrom, D.R. 1995. Age and survivorship of diapausing eggs in a sediment egg bank. Ecology 76: 17061711.
  • Hauser, L., Adcock, G.J., Smith, P.J., Bernal Ramírez, J.H. & Carvalho, G.R. 2002. Loss of microsatellite diversity and low effective population size in an overexploited population of New Zealand snapper (Pagrus auratus). Evolution 99: 1174211747.
  • Heath, D.D., Busch, C., Kelly, J. & Atagi, D.Y. 2002. Temporal change in genetic structure and effective population size in steelhead trout (Oncorhynchus mykiss). Mol. Ecol. 11: 197214.
  • Hedrick, P.W. 1999. Genetics of Populations, 2nd edn. Jones and Bartlett Publishers, Sudbury.
  • Hedrick, P. 2005. Large variance in reproductive success and the Ne/N ratio. Evolution 59: 15961599.
  • Hoarau, G., Boon, E., Jongma, D.N., Ferber, S., Palsson, J., Van der Veer, H.W., Rijnsdorp, A.D., Stam, W.T. & Olsen, J.L. 2005. Low effective population size and evidence for inbreeding in an overexploited flatfish, plaice (Pleuronectes platessa L.). Proc. R. Soc. B. 272: 497503.
  • Hofmann, W. 1979. Characteristics of syntopic populations of Eudiaptomus gracilis (SARS) und E. graciloides (LILLJEBORG) in three lakes of different trophic levels. Arch. Hydrobiol. 86: 112.
  • Jehle, R., Arntzen, J.W., Burke, T., Krupa, A.P. & Hödl, W. 2001. The annual number of breeding adults and the effective population size of syntopic newts (Triturus cristatus, T. marmoratus). Mol. Ecol. 10: 839850.
  • Kiefer, F. 1978. Freilebende Copepoda. Die Binnengewässer, Schweizbart’sche Verlagsbuchhandlung, Stuttgart.
  • Krimbas, C.B. & Tsakas, S. 1971. The genetics of Dacus oleae. V. Changes of esterase polymorphism in a natural population following insecticide control: selection or drift? Evolution 25: 454460.
  • Muuss, U., Petersen, M. & König, D.. 1973. Die Binnengewässer Schleswig-Holsteins. Wachholtz Verlag, Neumünster, 162 pp.
  • Nauwerck, A. 1980. Die Verbreitung der Familie Diaptomidae Sars in Nordschweden. Arch. Hydrobiol. 89: 247264.
  • Pálsson, S. 2000. Microsatellite variation in Daphnia pulex from both sides of the Baltic Sea. Mol. Ecol. 9: 10751088.
  • Rai, H. 1982. Primary production of various size fractions of natural phytoplankton communities in a North German lake. Arch. Hydrobiol. 95: 395412.
  • Raymond, M. & Rousset, F. 1995. genepop (version 1.2) population genetics software for exact tests and ecumenicism. J. Hered. 86: 248249.
  • Santer, B., Blohm-Sievers, E., Cáceres, C.E. & Hairston, N.G. Jr 2000. Life-history variation in the coexisting freshwater copepods Eudiaptomus gracilis and Eudiaptomus graciloides. Arch. Hydrobiol. 149: 441458.
  • Schwartz, M.K., Tallmon, D.A. & Luikart, G. 1998. Review of DNA-based census and effective population size estimators. Anim. Conserv. 1: 293299.
  • Van Oosterhout, C., Hutchinson, W.F., Wills, D.P.M. & Shipley, P. 2004. Microchecker: software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4: 535538.
  • Wang, J. 2001. A pseudo-likelihood method for estimating effective population size from temporally spaced samples. Genet. Res. 78: 243257.
  • Wang, J. & Caballero, A. 1999. Developments in predicting the effective size of subdivided populations. Heredity 82: 212226.
  • Wang, J. & Whitlock, M.C. 2003. Estimating effective population size and migration rates from genetic samples over space and time. Genetics 163: 429446.
  • Waples, R.S. 1989. A generalized approach for estimating effective population size from temporal changes in allele frequency. Genetics 121: 379391.
  • Waples, R.S. 2002. Evaluating the effect of stage-specific survivorship on the Ne/N ratio. Mol. Ecol. 11: 10291037.
  • Weir, B.S. & Cockerham, C.C. 1984. Estimating F-statistics for the analysis of population-structure. Evolution 38: 13581370.
  • Williamson, E.G. & Slatkin, M. 1999. Using maximum likelihood to estimate population size from temporal changes in allele frequencies. Genetics 152: 755761.
  • Wright, S. 1931. Evolution in Mendelian populations. Genetics 16: 97159.
  • Wright, S. 1938. Size of population and breeding structure in relation to evolution. Science 87: 430431.
  • Zeller, M. & Reusch, T.B.H. 2004. Identification and characterization of 10 microsatellite primers for the calanoid freshwater copepods Eudiaptomus gracilis and E. graciloides using enriched genomic libraries. Mol. Ecol. Notes 4: 355357.
  • Zeller, M., Reusch, T.B.H. & Lampert, W. 2006. A comparative population genetic study on calanoid freshwater copepods: investigation of isolation-by-distance in two Eudiaptomus species with a different potential for dispersal. Limnol. Oceanogr. 51: 117124.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Table S1 Allele frequencies at each locus for Eudiaptomus gracilioides and E. gracilis at four sampling years.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
JEB_1589_sm_Tab1.doc296KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.