Variation in body size of terrestrial isopods (crustacea: oniscidea) and application of ecogeographical rules abstract




старонка2/3
Дата канвертавання25.04.2016
Памер201.9 Kb.
1   2   3

The number of isopod species decreases above and below 50 degrees of latitude. The greatest number of species is recorded in temperate climates. This observation may be a result of recording bias, because species in temperate areas are more thoroughly examined and more easily accessible in comparison to those existing in deserts and Polar Regions. In any case, though, in our case there is no general pattern of increase in species richness with decreasing latitude. Species richness increases with latitude from the tropics to temperate climates and so does body size. We found no inverse pattern between species-poor and species-rich regions, as small-bodied species were always more abundant. Nevertheless, there was a tendency towards equalization of small and large-bodied species from the tropics to cold climates, where the numbers of small- and large-bodied species were almost equal.




Figure 1: Principal Component Analysis showing a clear tendency of the species residing in climates with dry summers and in steppes to be larger compared to those residing in warm and moist climates (rainforests, monsoon and savannah). Body size increases towards the positive values on the X axis.



Figure 2. Coplot: length of males in relation to absolute values of latitude given the climate1. Note that although in arid and temperate climates (down left and upper left diagrams) there is a clear pattern of increase in body length for males with latitude, for cold and tropical climates (middle down and upper right diagram), there is a small tendency of reduction of body length with latitude.

Figure 3. Coplot: Width of males in relation to absolute values of latitude given the climate1. Note that although in arid and temperate climates (down left and upper left diagrams) there is a clear pattern of increase in body width for males with latitude, for cold and tropical climates (middle down and upper right diagrams), there is a small tendency of reduction of body width with latitude.


Body size and insularity

As far as the island rule is concerned, the slope of insular on mainland length and width is always smaller than one. Even when we control for phylogenetic signal, the slope is again smaller for pairs within the same genus (Table 3). The effect is not present for species belonging to the same family, but this is probably due to the small number of pairs tested. Thus, the null hypothesis of no consistent differences between insular and mainland body size can be rejected for the cases referred before. More specifically, maximum length and width seems to evolve in the direction predicted by the Island Rule for large mainland species, which evolve smaller on islands for all isopods tested, as well as for those belonging to the same genus. Small mainland species evolve larger, following the Island Rule only among species belonging to the same genus. Pairs of mainland-island species belonging to the same climatic zone were examined. No significant difference in mean body length and width was found for 44 species pairs belonging to the tropics. The same was true for 35 pairs belonging to arid and 7 pairs belonging to cold climates. Only the 109 pairs that reside in temperate climates showed a significant difference in mean body length and width between insular and mainland species for both sexes. Mainland species are always larger (Table 4 and Figures 4-5).



Table 3. Results from least squares regression of insular body length and width on mainland body length and width for all isopods and within species belonging to the same genus and family.


Groups

N (pairs)

Intercept (bo)

SE

Slope (b1)

SE

P

R2

Regression model: Length/ Width island = bo + b1 (length/ width mainland)

All species






















Length of males

230

-0.181

0.199

0.845

0.019

<0.001

89.29

Length of females

230

0.350

0.210

0.817

0.019

<0.001

88.56

Width of males

230

0.067

0.058

0.752

0.011

<0.001

95.40

Width of females

230

-0.114

0.079

0.840

0.015

<0.001

93.06

Species within family






















Length of males

15













n.s.




Length of females

15













n.s.




Width of males

15













n.s.




Width of females

15













n.s.




Species within genus






















Length of males

26

1.150

1.186

0.792

0.148

<0.001

54.27

Length of females

26

1.822

1.335

0.780

0.154

<0.001

49.60

Width of males

26

0.143

0.406

0.880

0.105

<0.001

74.54

Width of females

26

0.460

0.522

0.875

0.128

<0.001

65.97






Figures 4 and 5. Bwplot: length and width of males determined by the distribution of species given the climate
Island area

In general, there was no consistent trait that indicates a significant correlation between body size and island area. A significant correlation was found only for small isopods concerning the length of females for islands below 5,000 km2 (r = 0.172, P = 0.031), so in this case a decreasing island surface did not promote gigantism. Only the width of small females that reside on islands smaller than 50,000 km2 (r = -0.206, P = 0.024), as well as the width of large males residing on the same islands (r = -0.261, P = 0.014), show consistency with the Island Rule. Therefore, we cannot deduct general consistency with the Island Rule using island area as a predictor variable.

Table 4. Results from regressions of island species against mainland species belonging to the same climatic zone.


Climate




N (pairs)

Mean ± SE

T

P

Tropical


Length of males

44

Main: 7.256 +/-1.437

1.752

0.083

Length of females

44

Main: 7.423 +/- 1.415

1.586

0.116







Isl: 6.089 +/- 0.936







Width of males

44

Main: 3.251 +/- 0.721

1.876

0.064







Isl: 2.451 +/- 0.467







Width of females

44

Main: 3.444 +/- 0.704

2.052

0.043







Isl: 2.581 +/- 0.471






Arid


Length of males

35

Main: 8.888 +/- 2.00

0.447

0.656







Isl: 8.343 +/- 1.461







Length of females

35

Main: 9.528 +/- 1.845

0.656

0.514







Isl: 8.774 +/- 1.435







Width of males

35

Main: 4.516 +/- 1.396

1.345

0.183







Isl: 3.450 +/- 0.800







Width of females

35

Main: 4.410 +/- 0.970

1.194

0.236







Isl: 3.673 +/- 0.795






Temperate



Length of males

109

Main: 9.628 +/- 0.890

2.83

0.005







Isl: 7.641 +/- 1.069







Length of females

109

Main: 10.523 +/- 0.966

2.726

0.007







Isl: 8.551 +/- 1.059







Width of males

109

Main: 4.575 +/- 0.492

2.669

0.008







Isl: 3.578 +/- 0.553







Width of females

109

Main: 4.908 +/- 0.493

2.657

0.008







Isl: 3.919 +/- 0.549







Cold

Length of males

7

Main: 9.086 +/- 6.241

0.815

0.431







Isl: 6.728 +/- 3.338







Length of females

7

Main: 9.300 +/- 6.469

0.419

0.682







Isl: 8.028 +/- 3.639







Width of males

7

Main: 4.714 +/- 5.244

0.678

0.511







Isl: 3.200 +/- 1.543







Width of females

7

Main: 5.043 +/- 4.043

0.573

0.577







Isl: 4.000 +/- 1.863








Multiple regression models for maximum body size

Pearson correlation tests showed a strong correlation between the length and width of males (r= 0.902) and between the length and width of females (r= 0.898). So, we built separate models to describe these size parameters, in order to avoid using them in the same model. The best multiple regression model for the length of males included climate1, climate2, distribution, family and the interaction between climate1 and family, and explained 51.22% of variance. The best model for the length of females included the same predictors, and explained 45.86% of variance. The best model for the width of males did not include the interaction between climate1 and family neither the climatic variable 2, and explained 47.73% of variance. Finally, for the width of females the model did not include the interaction between climate1 and family, plus the climatic variable 1 this time, and explained 45.03% of variance.

Using ANCOVA, we tested for effects of temperature and precipitation, as expressed through climatic variables, distribution and phylogenetic signal (family). After analyzing length and width, there were effects (model for male length) of family (d.f.=27, F=18.897, P<0.001), climate2 (d.f.=5, F=11.466, P<0.001), climate1 (d.f.=4, F=10.032, P<0.001), distribution (d.f.=2, F=15.457, P<0.001) and interaction between climate1:family (d.f.=36, F=1.959, P<0.001), ranked from the most to the least important. For female length, there were effects of family (d.f=27, F=14.998, P<0.001), climate2 (d.f.=5, F=10.979, P<0.001), climate1 (d.f.=4, F=12.305, P<0.001), distribution (d.f.=2, F=11.078, P<0.001), climate1:family (d.f.=36, F=1.943, P<0.002). For male width, there were effects of family (d.f.=27, F=18.998, P<0.001), distribution (d.f.= 2, F=18.858, P<0.001) and climate1 (d.f.= 4, F= 9.382, P<0.001), and for females width, the effects, ranked from the most to the least important, were: family (d.f.=27, F=15.508, P<0.001), climate2 (d.f.=8, F=12.101, P<0.001) and distribution (d.f.=2, F=16.142, P<0.001). The standardized coefficient for family indicates that this variable is the main driver of variation for all the parameters tested. It seems that for the length and width of females, as long as for the length of males, the second most important driver of variation is climatic variable 2, while for the width of males the second most important driver is distribution. Latitude is not included into the models because it is not a primary factor by itself, but a parameter through which other factors of variation are expressed.



Table 5. Multiple regression models for maximum body length and width. Models are ranked in each case by AIC from worst to best-fitting model. Codes for interactions between predictor variables are: F*C1: family and climate1, D*C1: distribution and climate1, C1*C2: climate1 and climate2, C2*F: climate2 and family, D*F*C2: distribution and family and climate2.

Parameters

Climate1

Distribution

Family

Climate2

F*C1

D*C1

C1*C2

C2*F

D*F*C2

AIC


































Length

of males

7.24

5.57

92.0

10.3

12.83

1.65

0.004







-963.5




7.24

5.57

92.0

10.3

12.82

1.65










-965.3




7.24

5.57

92.0

10.3

12.72













-966.2


Length

of females

9.02

4.06

74.2

10.0

12.88




0.51

7.606




-948.0




9.02

4.06

74.2

10.0

12.88




0.51







-958.4




9.02

4.06

74.2

10.0

12.82













-959.4

Width

of males

10.55

8.96

132.7

15.2







0.23







-731.9




10.55

8.96

132.7

15.2
















-733.0




10.55

10.59

144.1



















-734.9

Width of

of females




8.11

105.2

24.3










20.763

19.388

-919.9







8.11

105.2

24.3










20.763




- 930.6







8.11

105.2

24.3
















- 947.6

4. DISCUSSION

Isopods are peracaridan crustaceans with representatives in almost all kinds of environments. Members of the suborder Oniscidea (about 3,600 species) are by far the most successful group of crustaceans on land. They are the better adapted among crustaceans, but much worse adapted to land than other terrestrial arthropods. Given that, in general, they exhibit behavioral regulation of body temperature, they could be considered as a proper group of ectotherms for testing Bergmann’s and other related ecogeographical rules. In this study, we found that body size in terrestrial isopods follows trends that are consistent with Bergmann’s Rule, namely an increase in body size with latitude, but not with the geographic version of Rensch’s Rule, since we found greater variation in female body size with latitude, the opposite of what is expected by the latter rule. After testing the response of body size in relation to two climatic variables, we found a positive correlation for arid and temperate climates and a significant increase of body size with latitude in all but the dry climates. Moreover, those species residing in these climates were significantly smaller than the ones found in dry climates. We came to the conclusion that small-bodied species show a greater response to climatic variable one than large-bodied but, as far as the climatic variable two was concerned, there was a larger response for small-bodied males and females in length and an opposite trend in width. In addition, we tested body size variation with species richness and found that the number of isopod species decreases above and below 50 degrees of latitude, with the greatest species number being recorded in temperate regions. In addition, small-bodied species were over-represented in both species-poor and species-rich regions but, with the increase of latitude, there was a tendency for equal representation of small and large-bodied species, in a way that in cold climates the number of small- and large-bodied animals was almost equal. Finally, we found some consistent differences between insular and mainland species that were in agreement with the Island Rule, such as that large mainland species tend to evolve smaller on islands, and vice versa. Nevertheless, when using related species pairs, differences in size were found only for species residing in temperate climates, with mainland species always being larger. Moreover, decreasing island surface does not seem to promote gigantism in smaller insular species or dwarfism in larger ones. The major factor affecting size, besides phylogenetic relationship (documented by variable ‘family’), was ‘climatic variable two’, which incorporates information on both temperature and precipitation.

The fact that isopod species residing in rainforests and in other permanently humid areas are significantly smaller compared to species in areas with dry summers (either in tropical, temperate or cold climates) suggests that body size may be associated with desiccation avoidance. Generally, species that have developed pleopodal lungs as a means for respiration are more resistant to drought and can attain larger body size (e.g., Armadillidae, Armadillidiidae, Eubelidae, Porcellionidae). Terrestrial isopods display a generally low resistance to desiccation Crawford (1992), so that moisture is a very important limiting factor for these animals. To minimize water evaporation, they frequent cool and humid sites, seek shelter under stones, and exhibit nocturnal lifestyles Horiguchi et al. (2007). The differences in activity of woodlice are due to the continuous alteration in their water content, and there is a reduction in activity with increasing relative humidity to the point of becoming totally akinetic Waloff (1941). The decreased activity of isopods in areas with high humidity (e.g., rainforests) suggests low metabolism and small body size towards the tropics, a trend consistent with our results. At the same time, small body size creates problems of desiccation in hot and dry environments that is why dry environments may select for individuals that are larger, in order to resist dehydration Stillwell et al. (2007), something that is also in consistency with our results, since isopods residing in deserts are larger in comparison to those residing in humid areas and this reduces surface-to-volume ratio and increases absolute water content Chown & Gaston (1999). The relative roles of temperature and humidity in temperate and cold climates are probably different, and large body size is promoted.

The increase of body size from the equator to the poles shows a trend consistent with Bergmann’s Rule which is confined to endotherms. Nevertheless, there is a set of mechanisms applicable to ectotherms that can explain such a pattern. The only mechanism actually tested on ectotherms, though, particularly on ants, is the Seasonality Hypothesis Lindstedt (1985) according to which the limiting factor for growth is the seasonality of resources. Since terrestrial isopods are not food limited Warburg (1984), the Seasonality Hypothesis, but also the Starvation Resistance Hypothesis that explains the positive body size correlation with latitude in ectotherms through a greater capacity for larger organisms to store food (Nekola et al. in press), are unlikely to hold for terrestrial isopods. Water may be a more important limiting resource than food, and can lead to an increase in body size with frequency and severity of moisture stress (Nekola et al. in press). Herein, we used precipitation through the variable climate2 as a proxy of the influence of water on body size, even though moisture is not a linear function of precipitation. Our results support the view that latitudinal variation of body proportions is largely dependent on a combination of water availability and temperature, since the variable climate2 (effects of temperature and precipitation) is by far more important than climate1 (effect of temperature alone). There are two conflicting mechanisms that differ in whether body size varies in response to the demands of keeping cool (Heat Dissipation Hypothesis) or keeping warm (Heat Conservation Hypothesis) Cushman (1993). According to the first hypothesis, variables related to temperature and moisture describe better average body size, because large-bodied animals need to respond to the challenge of dissipating heat and keeping cool, whereas smaller species do not Blackburn (2004). In contrary, Heat Conservation Hypothesis predicts greater response to temperature amongst small-bodied animals, because they face more challenges in keeping warm. Since the application of the latter hypothesis demands a thermoregulatory mechanism Watt et al. (2010), and due to the fact that terrestrial isopods do show behavioral, physiological and morphological regulatory mechanisms for temperature control Caubet et al. (1998), we can assume that this hypothesis could be extended also to such ectotherms. We found that covariation with climate2 (temperature and precipitation) was larger for small-bodied species, as far as length was concerned, but the trend was reversed for width. On the other hand, the greater response to temperature (climate1) found in small-bodied species for both sexes and for both parameters of body size, indicates that the Heat Conservation Hypothesis is more likely to explain the body size trends found herein.

‘Bergmann’s Rule’-like patterns may in fact be causally linked to the latitudinal gradient of decreasing species diversity towards the poles. Low alpha-diversity means reduced interspecific interactions, including competition. In the absence of larger competitors, smaller members of the same guild often exhibit ecological release, increasing in size in species-poor environments, such as those at high latitudes McNab (1971); Dayan (1990); Iriarte et al. (1990). Besides the fact that there is a general tendency for richness to be greater in the tropics Orme et al. (2006), such a pattern is not reflected in our study. There is strong evidence for a broadly positive monotonic relationship between species richness and energy availability Gaston (2000). The increment of energy is thought to enable a greater biomass to be supported in an area Kerr et al. (1997). So, species richness may represent a trade-off between body size and abundance in a way that limiting resources can increase body size and decrease abundance Olson et al. (2009). Lack of energy, expressed through shortage of resources towards higher latitudes, may be an explanation of the increase in body size of terrestrial isopods towards the poles, and may account for latitudinal trends in species diversity (Energy Availability Hypothesis) Roy et al. (2001). According to the Habitat Heterogeneity Hypothesis, species richness depends on water availability, heat and light Davies et al. (2007), and due to the trade-off between body size and abundance, these abiotic factors are very likely to affect body size trends

Any latitudinal change in environmental factors that affects one sex more than the other can generate variation in male and female body size clines and induce dimorphism Blanckenhorn et al. (2006). In our study, the slope of body size increase with latitude is steeper for females than males (inverse of geographic version of Rensch’s Rule), which also means a larger body size for females (larger = steeper hypothesis) Blackenhorn et al. (2006) and that sexual size dimorphism declines with increasing latitude Chown et al. (2010). In order to identify the mechanisms that produce steeper female clines requires studies quantifying latitudinal variation in sex-specific natural and sexual selection on body size. A mechanism that induces smaller size in males is their early emergence and maturation Bidau et al. (2008). Also, the bias in sex ratio towards females, very common in Oniscidea, might give a greater probability to females to include larger body sizes, in relation to the fewer males in the population. A possible link between Bergmann’s Rule (clinal size variation) and differential size variability among sexes (Rensch’s Rule) has not yet been explicitly explored.

The Island Rule Clegg (2002) is a complex pattern not reflecting only gigantism or dwarfism towards islands, but also a graded trend towards gigantism in smaller mainland species and dwarfism in larger. This reflects the relative importance of selective forces on islands and among species of different body size Lomolino, (2005). According to Meiri et al. (2009), the Island Rule may hold within taxonomic groups, such as family or order, therefore we compared mainland and island pairs of species belonging to the same genus and to the same family. For species belonging to the same genus we found the tendency, predicted by Lomolino et al. (2006), of insular species to become dwarfs or giants according to the size of their mainland relatives. More specifically, large mainland species become smaller on islands and small mainland species become larger, given that they belong to the same genus. The same results arose when all isopod species were tested. Because climatic conditions that affect available resources and primary productivity differ on islands Lomolino (2005), affecting, in turn, the intensity and the type of intra- and interspecific relationships among species, we tried to see if the Island Rule applies in islands belonging to the same climatic zone. Only pairs belonging to the temperate zone showed a significant difference in body size, with island species always being smaller than mainland ones. The fact that there are a great number of different possible climatic conditions that confine with the general temperate and cold climate, may give a wider variability of climatic conditions within the temperate and the cold zone. So, selective forces on islands may be more vigorous and they may swift more often, driving to dwarf forms on islands Lomolino (2005). Another reason that can promote dwarfism is resource limitation that can lead to an increase of intraspecific competition Lomolino (2005). Differences on body size between mainland and island species could not be evaluated for species residing in cold climates because the number of pairs tested was very limited.

Size evolution is not only affected by latitude, but also by isolation and island area Meiri (2007). There is a conviction that decreasing island area promotes gigantism in smaller insular species and dwarfism in larger ones. Area did not seem to be important in determining changes in size of island species. Beside the fact that Island Rule should be manifest mostly on small islands Meiri et al. (2009), neither of the three island categories based on area exhibited a change in size of insular species. As a consequence, we found little evidence for an effect of area on body size evolution of isopods. This is consistent with the fact that most studies found area not to affect size Lawlor (1982); Angerbjörn (1986); Yom-Tov et al. (1999); Anderson & Handley (2002); Boback (2003); Meiri et al. (2005). Of course, our study used an interspecific approach and was not restricted to land-bridge islands. Moreover, we had no available data to quantify the selective forces associated with interspecific pressures (predation, parasitism, interspecific competition) and those associated with intraspecific competition for limited resources. In addition, the calculation of the mean value of body length and width for the formation of pairs and the extraction of values (species), even though it did not change the initial mean value, induces a bias for the real body size of some mainland-island pairs. As a consequence, we cannot claim our results to be conclusive at any rate regarding the Island Rule, but only that they are indicative of trends to be explored further in the future.



REFERENCES

Adams, D.C. & J.O. Church, 2007. Amphibians do not follow Bergmann’s rule. Evolution 62: 413-420.

Anderson, R.P. & C.O. Handley, 2002. Dwarfism in insular sloths: Biogeography, selection and evolutionary rate. Evolution 56: 1045-1058.

Angerbjörn, A. 1986. Gigantism in island populations of wood mice (Apodemus) in Europe. Oikos 47: 47-56.

Angilletta, M.J., T.D. Steury & M.W. Sears, 2004. Temperature, growth rate, and body size in ectotherms: Fitting pieces of a life-history puzzle. Integrative and Comparative Biology 44: 498-509.

Ashton, G.K. 2002. Patterns of within-species body size variation of birds: strong evidence for Bergmann’s rule. Global Ecology and Biogeography 11: 505-523.

Bergmann, C. 1847 Über die Verhältnisse der Wärmekönomie der Thiere zu ihrer Grösse [The relation of energy budget to body size in animals]. Göttinger Studien 1: 595-708.

Bidau, C.J. & D.A. Marti, 2008. A test of Allen’s Rule in ectotherms: The case of two South American melanopline grasshoppers (Orthoptera: Acrididae) with partially overlapping geographic ranges. Neotropical Entomology 37(4): 370-380.

Blackburn, T.M. & B.A. Hawkins, 2004. Bergmann’s rule and the mammal fauna of North America. Ecography 27: 715-724.

Blanckenhorn, W.U. & M. Demont, 2004. Bergmann and converse Bergmann latitudinal clines in Arthropods: Two ends of a continuum? Integrative and Comparative Biology 44: 413-424.

Blanckenhorn, W.U, Stillwell, R.C, Young, K.A, Fox, C.W. & K.G. Ashton, 2006. When Rensch meets Bergmann: Does sexual size dimorphism change systematically with latitude? Evolution 60(10): 2004-2011.

Blanckenhorn, W.U., Dixon, A.F., Fairbairn, D.J., Foellmer, M.W., Gibert, P., Van der Linde, K., Meier, R., Nylin, S., Pitnick, S., Schoff, C., Signorelli, M., Teder, T. & C. Wiklund, 2007. Proximate causes of Rensch’s Rule: Does sexual size dimorphism in Arthropods result from sex differences in development time? The American Naturalist 169 (2): 245-257.

Boback, S.M., 2003. Body size evolution in snakes: evidence from island populations. Copeia 81-94.

Bouchon, D., Rigaud T. & P. Juchault, 1998. Evidence for widespread Wolbachia infection in isopod crustaceans: molecular identification and host feminization. Proceedings of the Royal Society London B. 265: 1081-1090.

1   2   3


База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка