explored in detail by Kozlowski and his colleagues (Kozlowski 1996; Kozlowski and Weiner 1997; Kozlowski and Gawelczyk 2002). Optimal size depends on mortality, assimilation, and respiration rates, all of which are size-dependent (Kozlowski and Gawelczyk 2002). Competitors can influence production rates by usurping or excluding access to resources, and predation and parasitism can alter mortality rates. These interactions also depend strongly on body size, so ultimately determining the optimal size. It is this variation in the size-dependence of mortality and production rates that ultimately produces the right-skewed body size frequency distributions which are so characteristic of assemblages (Gaston and Blackburn 2000).

Interspecific geographic variation in body size seems to be set in a major way by the spatial turnover of higher taxonomic groups (Hawkins and Lawton 1995; Chown and Klok 2003). In light of this finding, the arguments by Kozlowski and colleagues that interspecific patterns may well be an epiphenomenon of those at the intraspecific level, and the paucity of whole-assemblage species-body size frequency distributions (Chown et al. 2002a, but see Gaston et al. 2001), little attention will be given to geographic variation at the interspecific level (also known as Bergmann's Rule). Rather, most attention will be given to intraspecific geographic patterns in body size variation (also known as James' Rule—Blackburn et al. 1999).

Two robust, but contrary patterns in the intra-specific geographic variation of insect body sizes

Figure 7.4 Body length variation (mean ± SE, and 2SE) across the altitudinal gradient (m) at the relatively aseasonal sub-Antarctic Marion Island for the weevil Bothrometopus parvulus.

Source: Chown and Klok (2003). Ecography 26, 445-455, Blackwell Publishing.

have been identified. In some species, body size increases with increasing latitude (David and Bocquet 1975; Arnett and Gotelli 1999; Huey et al. 2000) or altitude (Smith et al. 2000; Chown and Klok 2003) (Fig. 7.4). The proximate cause of this variation is generally thought to be the negative relationship between rearing temperature and body size in ectothermic invertebrates (Atkinson 1994). In turn, increasing size at lower temperatures has ultimately been regarded either as adaptive (Partridge and French 1996; Atkinson and Sibly 1997; Fischer et al. 2003), or as a consequence of the differential sensitivity to temperature of growth and differentiation (van der Have and de Jong 1996). Although the causes of the latitudinal increase in body size continue to be the subject of some controversy, the general pattern appears to be well supported (Chown and Gaston 1999).

In contrast, a decrease in body size with latitude or with altitude has also been documented in several species (Mousseau and Roff 1989; Nylin and Svard 1991; Chown and Klok 2003) (Fig. 7.5). This pattern has been ascribed to changes in growing season length, such that longer seasons mean a longer growing period, and hence a larger final body size. Evidence for this causal hypothesis has come from studies of species showing sawtooth clines, where extended growing seasons allow two or more generations, resulting in a substantial decline in adult body size of the bivoltine population relative to the adjacent univoltine one (Masaki 1996).

5 20 40 50 100 140 200 250 350 Altitude (m)

Figure 7.5 Body length variation (mean ± SE, and 2SE) across the altitudinal gradient (m) at the seasonal, sub-Antarctic Heard Island for the weevil Ectemnorhinus viridis.

Source: Chown and Klok (2003). Ecography 26, 445-455, Blackwell Publishing.

The explanation for these contrary patterns lies in the relationship between growing season length and generation time (Roff 1980; Nylin and Gotthard 1998; Chown and Gaston 1999). If generation time is similar to or constitutes a significant proportion of the growing season length, then growing season length is likely to have a considerable influence on body size because of constraints on resource availability. Selection for differences in voltinism, via diapause propensity or development time, is also likely to be important. Hence, resource constraints, a consequence of the interaction between generation time and growing season length, mean that body size is likely to decline with declining temperature (and thus resource availability). In contrast, as generation time declines relative to growing season length, so resources effectively become available for longer, and selection for variation in voltinism is likely to be less important. Here, temperature influences on growth and differentiation, of the form widely seen in laboratory cultures and likely to be the physiological norm (Atkinson 1994; Ernsting and Isaaks 2000), are liable to become more pronounced, resulting in a negative relationship between temperature and body size.

These ideas were recently supported in an investigation of contrary altitudinal patterns in body size variation of sub-Antarctic weevils (Chown and Klok 2003). Species from the aseasonal Marion Island, which lies north of the cold Antarctic waters below the Antarctic Polar Frontal

Zone (APFZ), can grow and develop year-round and show an increase in size with altitude (Fig. 7.4). By contrast, those from the more seasonal Heard Island, which lies to the south of the APFZ, show a decline in body size with altitude associated with shorter season-lengths at higher altitudes (Fig. 7.5).

While the interaction between season length and generation time provides a plausible explanation for both declines and increases in body size with altitude or latitude, the mechanistic basis underlying increases in size (e.g. increases in cell size or number) remains contentious. These mechanisms can vary depending on whether natural or laboratory populations are examined, with the trait in question, and with gender (Partridge and French 1996; Chown and Gaston 1999; Van'T Land et al. 1999). Moreover, other factors might also influence the size dependence of production or mortality. For example, it has been suggested that large body size enables better survival of both drought and food shortages (Lighton et al. 1994; Kaspari and Vargo 1995; Chown and Gaston 1999). Given that more than a single environmental variable is likely to influence insect survival (Chown 2002) and production (Slansky and Rodriguez 1987), it is obvious that body size is likely to respond to more than one variable. Therefore, without a careful consideration of the environment within which an insect finds itself, the reasons for variation in body size might be difficult to ascertain (Arnett and Gotelli 1999; Chown and Gaston 1999).

The feedback between physiological variables and body size also means that the requirements of body size might constrain the options open to insects for physiological regulation. In Drosophila, populations experiencing laboratory selection respond to dry conditions by increasing their water content—the canteen strategy (Gibbs et al. 1997). However, wild xeric and mesic species do not differ in water content, presumably because the large mass, associated with high water contents, reduces predator avoidance capabilities (Gibbs and Matzkin 2001). In sub-Antarctic weevils, it appears that a reduction in food quality, as a consequence of angiosperm extinction during the Pleistocene glaciations, resulted in a general decline in body size in species that managed to survive and reproduce under these conditions. In consequence, there was considerable selection for enhanced desiccation resistance because manipulation of body size, as a means to improve dehydration tolerance, was no longer open (Chown and Klok 2003b). Subsequent recolonization of the sub-Antarctic islands by angiosperms has meant an increase in body size of several of the weevil species that have once again started utilizing this nutritious resource. The increase in size has been accompanied by a decline in desiccation resistance (Fig. 4.16).

Was this article helpful?

0 0

Post a comment