Printer Friendly

Life History Variation Within and Among Populations of the Ninespine Stickleback (Pungitius pungitius) in Alaska: Lake-Stream Contrasts.


The study of natural history, "fundamental properties of organisms," is vital to science (Tewksbury et al, 2014). The changing global climate increases the need for research into natural history. Although research on and the importance of the ninespine stickleback (Pungitius pungitius) in evolutionary biology has lagged behind that of its relative, the threespine stickleback (Gasterosteus aculeatus), the ninespine stickleback has increasingly been the focus of ecological and evolutionary research in becoming a popular model in evolutionary biology (Gibson, 2005; Merila, 2013). A number of examples demonstrating what has been learned from comparisons between threespine and ninespine sticklebacks has been provided by Merila (2013). Both the ninespine stickleback and the threespine stickleback have similarly wide distributions encircling the North Pole; however, ninespine stickleback tend to be distributed further north and largely in freshwater environments (Wootton, 1976). The ninespine stickleback may have evolved in freshwater, in contrast to the marine origin of the threespine stickleback; however, the origin of the ninespine stickleback remains unresolved (Bell and Foster, 1994; Shikano et al, 2010; DeFaveri et al, 2012).

Notwithstanding the growing research interest in the ninespine stickleback, we know little about its natural history, especially its reproductive life history. The most recent reports addressing the natural history of the ninespine stickleback include a study of reproduction in a population from Alaska with a summary of prior studies (Heins et al., 2003), an investigation of variation in population age structure and longevity among populations in lakes, ponds, and marine environments (DeFaveri et al., 2014), and an overview of evolutionary biology (Merila, 2013). Life history data, including clutch size and egg size, have been included in reports addressing genetics of inter-population differences in body size, compensatory growth, and sexual size dimorphism in ninespine stickleback (Herczeg et al., 2010; Ab Ghani et al., 2012; Ab Ghani and Merila, 2014).

Intraspecific comparisons of the qualities of populations can provide insights into evolutionary patterns and processes (Endler, 1977, 1986). In an earlier study, Baker et al (1998) found a number of life-history differences between stream and lake populations of the threespine stickleback in Alaska. Similarly, Moser et al (2012) demonstrated divergent life history patterns between stream and lake populations of the threespine stickleback in Central Europe. Therefore, one might expect to find similar differences between populations of ninespine stickleback living in these different environments. This investigation was motivated by questions about potential differences among lake and stream populations of the ninespine stickleback. Herein, I present results of a multi-year study of life history traits within and between two populations of ninespine stickleback, one stream and one lake. I also use published and unpublished data from a third population (Airolo Lake; Heins et al, 2003, 2005, pers. obs.) to provide comparisons among three populations (two lake, one stream) from Alaska. Finally, I compare published knowledge of the natural history of threespine stickleback with what is known for the ninespine stickleback to ask if stream and lake populations of both species might show similar patterns for life-history traits.



Two populations from which I was able to obtain large samples for study were used in this investigation, one creek and one lake. Unnamed Creek (61[degrees]36'17.6"N; -149[degrees]30'38.3"W) is a small stream located along Church Road west of Wasilla in the Matanuska-Susitna Valley. Dog Bone Lake (60[degrees]41'36.0"N; 151[degrees]7'44.6"W) is located in the northern part of the Kenai Peninsula which borders the Cook Inlet of south-central Alaska. The lake is about 10 ha in size and 3.7 m deep. Lakes in the region are usually covered with ice from October into May (Woods, 1985).


Annual collections were made from 2000 to 2010 and in 2017 between 26 May and 5 June, within the breeding season of ninespine stickleback in this region of Alaska (Heins et al, 2003). Gee-type, metal traps having funnels at both ends, with either 3 mm or 6 mm mesh, were set near the shoreline. Fish were anesthetized in MS222 until quiescent, hence euthanized; and the samples were preserved, then stored, in buffered 10% formalin until examination.

Sticklebacks in Dog Bone Lake are infected by the cestode parasite Schistocephalus pungitii, which causes severe pathology, including host reproductive impairment and death (Heins, 2017). Epizootics can cause large changes in population size and structure as a result of the impact on host reproduction and mortality (Heins and Ecke, 2012). Therefore, all fish infected with the parasite were excluded from this study. Infections of S. pungitii were very rare in the stream population, few of these infected fish having been excluded from the data set.

Measurements of standard length (SL) were made to the nearest 0.1 mm with a digital caliper. Following length measurements, fish were dissected to determine infection status and to remove the viscera. The wet body mass (BM) of eviscerated and blotted specimens was made to the nearest 0.001 g on an electronic balance.

Ovaries were separated from the rest of the viscera. Female ninespine stickleback were classified into six reproductive stages of ovarian condition following Baker et al. (1998) and Heins et al. (1999): latent (LA), early maturing (EM), late maturing (LM), mature (MA), ripening (MR), and ripe (RE). During the breeding season, ninespine stickleback females produce multiple clutches (Heins et al, 2003). The classification scheme incorporates the "clutch-production cycle" (Heins and Baker, 1993), as sexually mature females repeatedly cycle among the LM, MA, MR, and RE stages. Females with ovaries classified as LA are sexually immature. Females with ovaries that were EM or LM were considered sexually mature given they apparently were producing a clutch to be spawned that season, but did not possess a discernable clutch at the time of capture. Females with MA, MR, and RE ovaries had easily identified clutches and were considered both sexually and reproductively mature. Herein, clutch-bearing females are referred to as being reproductive.

Clutch size (CS) was counted directly after separating out all of the enlarged oocytes or eggs comprising clutches of MA, MR, and RE females (Heins and Baker, 1993). Clutch mass (CM) of MR- and RE-stage females (oocytes or eggs fully developed) was measured to the nearest 0.00001 g after drying for 24 h at 40[degrees]C. Mean egg mass (EW) of each female was calculated to the nearest microgram by dividing CM by CS.


To estimate the ages at which females were reproducing, length-frequency plots were created using 1-mm SL intervals. The strong seasonal nature of ninespine stickleback reproduction in Alaska and other high latitude environments often creates relatively distinct size modes, allowing size-frequency plots to be an acceptable method of age estimation for the species (Baker et al., 2008).

Statistical analyses of data within and between lakes were conducted using a General Linear Model, with or without a covariate as appropriate for each analysis. Means were compared using the Bonferroni method. Response variables and covariates were transformed to [log.sub.10] values prior to analyses: standard length, LSL; somatic body mass, LBM; clutch size, LCS; egg weight, LEW. Intra-population analyses included SL, CS, and EW; analyses between populations additionally included BM. SYSTAT (Systat Software Inc.) was used for the statistical computations. For inter-population analyses, partial eta-squared values were calculated to provide effect sizes (Ferguson, 2009).

Clutch size is strongly correlated with female body size in ninespine stickleback (Heins et al, 2003, 2005), as is also the case for threespine stickleback (Baker et al., 2008). Therefore, I used [log.sub.10]-transformed variables of body mass (LBM) as a covariate to standardize estimates of population parameters to a common female size for each analysis. Notwithstanding some significant interactions between LBM and year of capture in analyses, I justified the inclusion of LBM as a covariate given the covariate explained a large, significant amount of variation, the interaction was a minor source of variation, and simulation tests have established that ANCOVA is sufficiently robust to violations of this assumption under many circumstances (Wu, 1984; Reist, 1985; Sullivan and D Agostino, 2002).

Egg size (mass) typically is not as strongly correlated with body size and shows more variation in relation to body mass in threespine stickleback (Baker et al., 1998, 2008) and ninespine stickleback as well (Heins, pers. obs.). Adjustments to a common female size were not necessary. Moreover, selection likely acts on the size of fry resulting from eggs, the size of which I assume to be related to the size of fry arising from them. Therefore, raw data for egg mass are provided.

Contingencies limited my ability to make direct comparisons between lakes in all years. Therefore, I have also used data from a number of different years for each population to ask whether there are differences among the populations.



Two of three length-frequency histograms for females from Unnamed Creek revealed two groups of sexually mature females bearing clutches (Fig. 1), which appear to represent 1 and 2 y old fish respectively. These graphs suggest ninespine stickleback in Unnamed Creek typically mature at 1 y of age (1+) and live to reproduce as long as 2 y (2+).

The length-frequency histograms for Dog Bone Lake (Fig. 2) were highly variable, largely stemming from the impact of the parasite Schistocephalus pungitii on host reproduction and mortality (see Discussion). Nonetheless, they appear to show females usually live as long as 2 y to age 2+. Two age groups (1+ and 2+) are apparent, although the strength of the groups appears to vary annually. Small percentages of females mature at age 14- most years; however, large percentages of 1+ females can mature to produce egg clutches in some years, as was the case in 2010.

A test of length differences between reproductive females from Unnamed Creek and Dog Bone Lake (Tables 1 and 2) in 2005 and 2010 revealed that the mean LSL was significantly greater ([F.sub.1,393] = 12.802, P < 0.001) in Dog Bone Lake (antilog, 51.2 mm SL) than in Unnamed Creek (antilog, 48.6 mm SL); and LSL was significantly greater ([F.sub.1,393] = 40.450, P < 0.001) in 2010 (antilog, 52.5 mm SL) than in 2005 (antilog, 47.4 mm SL). Similarly, LBM was significantly greater ([F.sub.1,393] = 8.817, P = 0.003) in Dog Bone Lake (antilog, 0.885 g) than in Unnamed Creek (antilog, 0.778 g), and LBM was significantly greater ([F.sub.1,393] = 29.111, P < 0.001) than in 2010 (antilog, 0.944 g) than in 2005 (antilog, 0.728 g). Effect sizes (Table 3) show that year had a larger influence on variation than lake. These tests for specific years are supported by the overall means for all reproductive fish from Unnamed Creek (SL, 50.6 mm; BM 0.984 g, n = 638) and Dog Bone Lake (SL, 53.4 mm; BM 1.060 g, n = 939).


Overall, a total of 638 females from Unnamed Creek with clutches (Table 1) ranged in size from 37.1 to 67.1 mm SL and averaged 50.6 mm SL. Clutch size ranged from 48-386 eggs and averaged 141 eggs among all females. There were significant differences in size-adjusted mean LCS among years ([F.sub.4,632] = 51.279, P < 0.001), with least squares mean LCS significantly less (P < 0.05) in 2006 (antilog, 113) than in all other years and significantly less in 2007 (antilog, 134) than in 2010 (antilog, 148) and 2017 (antilog, 153). All other differences were nonsignificant (P > 0.05). Therefore, LCS decreased significantly in 2006 before increasing in 2007. Size-adjusted LCS did not differ between 2010 and 2017, during which time LCS was greater than 2006 and 2007, but not significantly different from 2005 (antilog, 146).

Reproductive females in Dog Bone Lake (n = 939, Table 2), ranging in length from 39.9 to 73.3 mm SL and averaging 53.4 mm SL, produced an average of 99 eggs per clutch (range, 30-354). Mean clutch size differed significantly among years ([F.sub.6,931] = 52.897, P < 0.001). Mean least squares LCS decreased significantly each year from 2000 (antilog, 103) to 2001 (antilog 88) and 2002 (antilog, 79) when LCS was significantly lower than in all other years sampled. Between 2004 and 2005, mean I.C.S increased significantly from 91 to 104 (antilogs); and means for 2005, 2009 (antilog 101) and 2010 (antilog, 105) were not significantly different. Therefore, LCS appears to have declined to a low in 2002, after which it was significantly greater in sampled years.

Comparison of LCS between Unnamed Creek and Dog Bone Lake in 2005 and 2010, revealed that size-adjusted LCS was significantly ([F.sub.1,392] = 367.984, P < 0.001) greater in Unnamed Creek (antilog, 143) than in Dog Bone Lake (antilog, 93). Annual variation was not a significant ([F.sub.1,392] = 0.903, P = 0.343) source of variation between years (antilogs; 2005, 114; 2010, 117). Effect sizes reveal that the covariate body mass had the largest effect, and the lake effect was greater than the year effect (Table 3).


Ovum mass (LEW) was significantly (P < 0.05) correlated with LBM two of 5 y in Unnamed Creek and one of 7 y in Dog Bone Lake. LBM was not a significant source of variation (P > 0.05) in analyses of data within or between lakes.

Of the clutch-bearing females from Unnamed Creek, dry egg mass (Table 1) was obtained for 189 fish that were RE and ranged in size from 37.1 to 64.8 mm SL. Mean female egg mass ranged from 314-668 pg and averaged 494 pg among all females. Most females produced eggs of a similar mass, the co-efficient of variation only ranging from 4.5 to 13.4% among all years. Mean least squares LEW was significantly different among years ([F.sub.4,184] = 69.419; P < 0.001). It increased significantly each year from 2005 (antilog, 416 [micro]g) to 2006 (antilog, 497 [micro]g), and 2007 (antilog, 564 [micro]g), showing mean egg mass increased across these 3 y. The mean LEWs for 2010 (antilog, 400 [micro]g) and 2017 (antilog, 425 [micro]g) were not significantly different and were lower than in 2006 (antilog, 497 [micro]g) and 2007 but not significantly different from that in 2005. The increase in mean LEW between the smallest and largest mean was 41%.

Females in Dog Bone Lake produced eggs weighing an average of 513 [micro]g (range, 365-665 [micro]g) among 473 RE females ranging in size from 39.9 to 73.3 mm SL (Table 2). The coefficient of variation ranged from 6.6 to 10.6% among all years sampled, again showing most females produced eggs that were similar in mass. Although there were significant differences in mean LEW among years ([F.sub.6,466] = 4.321; P < 0.001), the only annual differences were between the smallest mean LEW in 2001 (antilog, 488 [micro]g) and the three largest mean LEWs in 2004, 2009, and 2010 (antilogs, 529, 519, 519 [micro]g, respectively). The increase between the smallest and largest mean was 6%, showing that annual variation in egg size was minimal.

Tests of LEW for Unnamed Creek and Dog Bone Lake females in 2005 and 2010 showed that Unnamed Creek females produced significantly ([F.sub.1,134] = 197.493, P < 0.001) smaller eggs (398 [micro]g) than those in Dog Bone Lake (512 [micro]g); the year of sampling was not a significant ([F.sub.1,134] = 1.016, P = 0.315) factor in the observed differences (antilogs; 2005, 447 [micro]g; 2010, 456 [micro]g). The lake effect was larger than the year effect, as shown by the effect sizes (Table 3).



Variability in the length-frequency histograms for Dog Bone Lake females appears to have been influenced by infections of S. pungitii. For example adult fish were heavily infected in 2003 as a epizootic peaked in prevalence, intensity and parasite:host body mass ratios (Heins and Ecke, 2012). The impact of the parasite on reproduction and survivorship of infected host fish (not shown; Heins and Ecke, 2012; Heins, 2017) appears to have resulted in a very small recruitment class of 0+fish, which should have been caught as 1+fish in 2004. The 1+ fish in 2003 were largely uninfected and subsequently matured to reproduce in 2004. Those adults then gave rise to the 1+ fish that were caught in 2005, and few adult fish in 2004 appear to have survived to 2005. Presumably variation in the population structure of stickleback in Dog Bone Lake before and after this period of time reflects similar processes in the interplay between host fish and cestode parasite. Environmental conditions, including lake levels and the availability of nest sites at the margins (Heins et al, 2003), would be expected to influence the population structure as well.

The lengths of reproductive females in Unnamed Creek were smaller on average than those of females in Dog Bone Lake. The size distributions of clutch-bearing females in Airolo Lake and Dog Bone Lake were similar (Heins et al, 2005), and a test of those differences for this report showed there were no significant differences overall in either LSL ([F.sub.1,349] = 0.015; P = 0.901) or LBM ([F.sub.1,349] = 2.318; P = 0.129) between the two populations (Heins, pers. data).

Female ninespine stickleback from Unnamed Creek reproduced at age 1+ and 2+, the smallest reproductive female ranging in size from 37.1 to 40.9 mm SL across 5 y. In contrast the majority of females in Dog Bone Lake did not appear to reproduce until 2+y old, with the exception of the last year of this study when both 1+ and 2+ females were reproductive. The smallest reproductive female ranged in size from 39.9 to 50.9 mm SL among 7 y. Female stickleback in Airolo Lake lived to 2+y, at which time they matured sexually and reproduced. Females age 1+ were sexually immature. The smallest clutch-bearing females were 44.7 and 48.1 mm SL in two different years. Therefore, ninespine stickleback females in Airolo Lake appear to have begun reproducing at about the same length as those in Dog Bone Lake, whereas females in Unnamed Creek tend to begin reproducing at a smaller size.

An analysis of CS for Dog Bone Lake and Airolo Lake (Heins et al, 2005) showed small but significant differences between lakes across 2 y. Females in Dog Bone produced size-adjusted clutches that were 6-9% smaller than females in Airolo Lake (Heins et al, 2005); however, females in Dog Bone Lake produced clutches that were 35% smaller than those in Unnamed Creek. There were no significant differences in egg mass between Airolo and Dog Bone lakes (Heins et al, 2005); and, as shown herein females in Unnamed Creek produced smaller eggs than those in Dog Bone Lake. Therefore, females in Unnamed Creek appear to produce larger size-adjusted clutches with a smaller mean ovum mass than females in either lake population.

In summary the results of this study suggest ninespine stickleback females in stream populations of Alaska reproduce at a smaller average size, tending toward a younger age, than their conspecifics in lake populations. The data also suggest females in streams produce larger size-adjusted clutches of smaller eggs. Further study is needed to test this life-history pattern for ninespine stickleback in Alaska, given the small number of populations on which it is based. For a broader comparison among populations worldwide, additional research is needed to provide more data than die limited information now available, which was reviewed by Heins et al (2003; see Table IV). Graphical presentations in more recent reports (e.g., Herczeg et al, 2010; Ab Ghani et al., 2012; Ab Ghani and Merila, 2014) would not allow an accurate update to the review.


Threespine stickleback in stream populations appear to be smaller than their counterparts in lakes of Alaska (Baker et al., 1998) and Central Europe (Moser et al, 2012). Populations in the two environments of Central Europe showed a conspicuous size dichotomy as stream populations mostly lived to and reproduced at age 1+, whereas lake populations largely lived to and reproduced at age 2+ (Moser et al, 2012). Stream and lake populations in Alaska overlapped in size and age distributions (Baker et al., 1998). Clutch size increased with body size, and smaller stream fish produced fewer eggs per clutch than lake fish (Baker et al., 1998; Moser et al., 1998). However, when clutch size was adjusted to a common body size, stream populations in Alaska were found to produce greater size-adjusted numbers of eggs per clutch than lacustrine populations, whether full-pelvic or reduced-pelvic populations (Baker et al., 1998). Moser et al (2012) did not compare size-adjusted differences, perhaps because the two groups did not show much overlap in length. In Alaska stream populations have lighter, hence smaller, eggs than lacustrine, reduced-pelvic lacustrine populations: however, lake populations with fully formed pelvic complexes did not differ from stream populations (Baker et al, 1998).


The results for both ninespine and threespine stickleback suggest there are consistent differences in life history tactics between stream and lake populations, notwithstanding variation among populations and habitat types. The most consistent difference appears to be size at reproduction, and differences can be conspicuous, as seen in Central Europe. After adjustment for body size, stream populations appear to produce greater numbers of smaller eggs than their conspecifics in lakes. The results of this study further suggest these differences might persist despite annual variation in life history traits, possibly reflecting genetic and environmental components (Moser, 2012; Heins and Baker, 2018). Further research is needed to test the hypothesis that lentic and lotic environments select for different suites of life history traits manifested as functional convergence in both ninespine and threespine sticklebacks (Merila, 2013). Possible environmental influences include stream-flow characteristics, stream versus lake temperatures, and food sources. In Alaska streams typically carry meltwater; and they can be colder than lakes during the growing season. Moser et al. (2012) suggested divergence between lake and stream populations of threespine stickleback might result from genetic and/or plastic responses to changes in foraging mode (i.e., pelagic versus benthic). Two hypothetical mechanisms for the lake-stream divergence based upon maturation size thresholds have been proposed by Moser et al. (2012).


This investigation suggests that lake and stream populations of ninespine stickleback show life history divergence similar to that observed for threespine stickleback. Further research should aim to test whether there is repeated divergence among ninespine stickleback populations. Illuminating the environmental factor or factors driving any repeated divergence, the mechanisms of divergence, and the role of genetic versus plastic responses to environmental differences will be an important research initiative in the future.

Acknowledgments.--Many undergraduate students at Titiane University contributed to field and laboratory work involved in this study, chiefly Kelly Barry, Emily Birden, Amanda Burr, Gayan DeSilva, Johanna Ecke, Megan Sekiya, Britt Ulinski, and Carmelia Vizza. John Baker recommended and provided advice for the use of partial eta-squared. Field trips for specimen collection were supported by the Newcomb College Foundation. This research was conducted under Fish Resource Permits of the Alaska Department of Fish and Game, following protocols approved by the Institutional Animal Care and Use Committee al Tulane University.


Department of Ecology and Evolutionary Biology, Tulane University, New Orleans, Louisiana 70118


AB CHAM, N. I., G. HERCZEG, AND J. MERILA. 2012. Body size divergence in nine-spined sticklebacks: disentangling additive genetic and maternal effects. Biol. J. Linn. Soc., 107:521-528.

--AND J. MERILA. 2014. Cross-generational costs of compensatory growth in nine-spined sticklebacks. Oikos 123:1489-1498.

BAKER, J. A., S. A. FOSTER, D. C. HEINS, M. A. BELL, AND R. W. KING. 1998. Variation in female life-history traits among Alaskan populations of the threespine stickleback, Gasterosteus aculeatush. (Pisces: Gasterosteidae). Biol. J. Linn, Soc., 63:141-159.

--, D. C. HEINS, S. A. FOSTER, AND R. W. KING. 2008. An overview of life-history variation in female three-spined stickleback. Behaviour, 145:579-602.

BELL, M. A. AND S. A. FOSTER. 1994. Introduction to the evolutionary biology of the threespine stickleback, p. 1-27. In: Bell, M. A. and S. A. Foster (eds.). The evolutionary biology of the threespine stickleback. Oxford University Press.

DEFAVERI, J., T. SHIKANO, N.L. AB GUANI, AND J. MERILA. 2012. Contrasting population structures in two sympatric fishes in the Baltic Sea basin. Mar. Biol., 159:1659-1672.

--,--, AND J. MERILA. 2014. Geographic variation in age structure and longevity in the nine spined stickleback (Pungitius pungitius). PLoS One, 9(7):el02660. doi:10.1371/journal.pone. 0102660.

ENDLER, J. A. 1977. Geographic variation, speciation, and clines. Princeton University Press, Princeton NJ.

--, 1986. Natural selection in the wild. Princeton University Press, Princeton, NJ.

FERGUSON, C. J. 2009. An effect size primen a guide for clinicians and researchers. Prof. Psychol. Res. Pr., 40:532-538.

GIBSON, G. 2005. The synthesis and evolution of a supermodel. Science, 307:1890-1891.

HEINS, D. C. 2017. The cestode parasite Schistocephahis pungitii: castrator or nutrient thief of ninespine stickleback fish? Parasitology, 144:834-840.

--AND J. A. BAKER. 1993. Clutch production in the darter Etheostoma lynceum Hay and its implications for life-history study. J Fish Biol., 42:819-829.

--AND --, 2018. Streamflow environment predicts divergent life history phenotypes among populations of the Blacktail Shiner Cyprinella venusta: Temporal stability of a large-scale pattern. Ecol. Freshw. Fish, 27:453-459.

--, --, G. DE SILVA, AND F.. L. BIRDEN. 2005. Clutch characteristics of two populations of ninespine stickleback from south-central Alaska. J. Fish Biol., 67:873-878.

--and J. K. ECKE. 2012. The rise and fall of an epizootic of the diphyllobothriidean cestode Schistocephahis pungitii infecting the ninespine stickleback. J Parasitol., 98:1-5.

--, J. M. JOHNSON, AND J. A. BAKER. 2003. Reproductive ecology of the ninespined stickleback from south-central Alaska. J Fish Biol., 63:1131-1143. doi: 10.1046/j.l095-8649.2003.00232.x

--, S. S. SINGER AND J. A. BAKER. 1999. Virulence of the cestode Schistocephahis soliclus and reproduction in infected three-spined stickleback, Gasterosteus aculeatus. Can. J. Zool., 77:1967-1974.

HERCZEG, G., A. GONDA, AND J. MERILA. 2010. Rensch's rule inverted: female-driven gigantism in ninespined stickleback, Pungitius pungitius. J. Anim. Ecol., 79:581-588.

MERILA, J. 2013. Nine-spined stickleback (Pungitius pungitius): an emerging model for evolutionary biology research. Ann. S.V. Acad. Sei.. 1289:18-35.

MOSER, D., M. ROESTI, AND D. BERNER. 2012. Repealed lake-stream divergence in stickleback life history within a Central European lake basin. PLoS One, 7(12):e50620. doi:10.1371/journal.pone. 0050620

REIST, J. D. 1985. An empirical evaluation of several univariate methods that adjust for size variation in morphometric data. Can. J. Zool.. 63:1429-1439.

SHIKANO, T., Y. SHIMADA, G. HERCZEG, AND J. MERILA. 2010. History vs. habitat type: explaining the genetic structure of European nine-spined stickleback (Pungitius pungitius) populations. Mol. Ecol, 19:1147-1161.

SULLIVAN, L. M. AND R. B. D'AGOSTINO. 2002. Robustness and power of analysis of covariance applied to data distorted from normality by floor effects: non-homogeneous regression slopes. / Stat. Comput. SimuL, 72:141-165.

TEWKSBURY. J., J. G. T. ANDERSON, J. D. BAKKER. T.J. BILLO, P. W. DI NWIDDIE, M.J. GROOM, S. E. HAMPTON, S. G. HERMAN, D.J. LEVERY, N.J. MACHNTCKI, C. M. DEL RIO, M. E. POWER, K. ROWELL, A. K. SALOMON. L. STACEY, S. C. TROMBULAK, AND T. A. WHEELER. 2014. Natural history's place in science and society. Bioscience. 64:300-310.

WOODS, P. F. 1985. Limnology of nine small lakes, Matanuska-Susitna Borough, Alaska, and the survival and growth rates of rainbow trout. Anchorage: United States Geological Survey Water-Resources Inventory Report 85-4292, Geological Survey, United Slates Department of the Interior.

WOOTTON, R. J. 1976. The biology of sticklebacks. Academic Press, London.

WU Y. B. 1984. The effects of heterogeneous regression slopes on the robustness of two test statistics in the analysis of covariance. Frluc. Psychol. Meas., 44:647-663.



Caption: Fig. 1.--Length-frequency histograms for ninespine stickleback females from Unnamed Creek, Alaska in 2006-2007 and 2010 (n = 225, 244, and 188, respectively). Filled portions of bars signify' clutchbearing (reproductive) females and open bars represent females without clutches of eggs

Caption: Fig. 2.--Length-frequency histograms for ninespine stickleback females from Dog Bone Lake, Alaska in 2000-2005 and 2009-2010 (n = 143, 212, 118, 46, 105, 98, 288, and 147, respectively). Filled portions of bars signify clutch-bearing (reproductive) females and open bars represent females without clutches of eggs
TABLE 1.--Descriptive statistics (untransformed data)
for clutch-bearing females of the ninespine stickleback
from Unnamed Creek, Alaska. Standard lengths for the subset
of females used for estimates of egg mass are different
from those shown here

Trait             Year    N      Mean    deviation

Standard length   2005    34     45.1       4.3
                  2006   182     50.8       6.4
                  2007   211     51.6       4.7
                  2010   156     52.1       7.9
                  2017    55     45.1       4.3
Clutch size       2005    34    118.6      38.2
                  2006   182    122.0      35.8
                  2007   211    151.7      41.4
                  2010   156    156.9      59.7
                  2017    55    126.5      42.3
Egg weight        2005     5    416.6      18.8
                  2006    62    501.0      67.0
                  2007    67    565.4      49.6
                  2010    43    402.2      41.8
                  2017    12    427.2      45.1

TABLE 2.--Descriptive statistics (untransformed data)
for clutch-bearing females of the ninespine stickleback
from Dog Bone Lake, Alaska. Standard lengths for the subset
of females used for estimates of egg mass are different
from those shown here

Trait         Year    N      Mean    deviation

Standard      2000   133     53.0       3.9
  length      2001    96     49.4       8.1
              2002   250     55.4       4.2
              2004   119     58.1       3.1
              2005    78     49.5       5.0
              2009   135     50.1       4.2
              2010   128     54.1       8.3
Clutch size   2000   133    113.5      30.4
              2001    96     81.9      52.1
              2002   250     84.9      22.9
              2004   119    122.2      33.6
              2005    78     89.0      33.6
              2009   135     96.5      31.0
              2010   128    114.8      51.9
Egg weight    2000    48    516.7      37.0
              2001    52    490.5      52.2
              2002   181    509.7      52.6
              2004    49    530.7      38.4
              2005    28    503.6      46.1
              2009    54    519.4      34.4
              2010    61    521.1      47.5

TABLE 3.--Partial eta-squared ([[eta].sub.P.sup.2]) for analyses
of standard length, body mass, clutch size and egg mass for
ninespine stickleback from Unnamed Creek and Dog Bone Lake,
Alaska, in 2005 and 2010

Trail             Factor      [[eta].sub.P

Standard length   Lake           0.038
                  Year           0.099
                  Lake*Year      0.007
Body mass         Lake           0.026
                  Year           0.073
                  Lake*Year      0.005
Clutch size       Lake           0.431
                  Year           0.001
                  Body mass      0.693
                  Lake*Year      0.004
Egg mass          Lake           0.372
                  Year           0.000
                  Lake*Year      0.017
COPYRIGHT 2019 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Heins, David C.
Publication:The American Midland Naturalist
Article Type:Report
Geographic Code:1U9AK
Date:Oct 1, 2019
Previous Article:Microhabitat Use, Spawning Behavior, and Spawning Substrate Use of the Tallapoosa Darter (Etheostoma tallapoosae).
Next Article:Spatial Ecology of Re-introduced American Martens in the Northern Lower Peninsula of Michigan.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters