The evolution of frequency distributions in species body sizes

Most species contributing to a large taxonomic rank are small. This appears to be the overwhelming message from an examination of species body size distributions (Figure 15.1). The distributions tend to be right skewed, even on a logarithmic scale, and yet, the mode is seldom in the smallest size class but slightly larger. These patterns probably hold regardless of potential biases in the data (Blackburn and Gaston 1994). The skewness tends to be less common at smaller spatial scales and when small taxonomic ranks (i.e. families instead of classes) are examined (Maurer 1998a).

What processes might account for this pattern? Dial and Marzluff (1988) and later McKinnley (1990) both developed verbal/graphical evolutionary models based on evolutionary mechanisms. Dial and Marzluff assumed that the difference between speciation and extinction (net diversification) is greater at small than large body sizes (Figure 15.2). Then one can imagine smaller-bodied lineages on the left-hand size of the frequency distribution out-proliferating those on the right. McKinnley developed a passive

Body size

Fig. 15.1 A typical frequency distribution of species body sizes. The distribution is right skewed, with the mode well to the left, but not at the smallest size class.

Body size

Fig. 15.1 A typical frequency distribution of species body sizes. The distribution is right skewed, with the mode well to the left, but not at the smallest size class.

Fig. 15.2 Possible evolutionary pathways to the right-skewed frequency distribution of species body sizes. (a) The ancestor was medium sized, and speciation occurs with equal frequency at all body sizes, but larger species go extinct more rapidly. (b), Speciation occurs more frequently at small body sizes. (c) The ancestral body size was small, and only rarely do some species become large, hence anagenesis is biased.

diffusion scenario of body size evolution. He assumed first that clades originate at small body sizes, and that there is a lower limit to body size. Anagenetic processes must also play a role if a clade is to produce variable body sizes; this might occur directionally (towards larger body sizes) or in a random direction in a process akin to drift. As a result, the clade multiplies fastest at smaller body sizes and develops a right-hand tail (Figure 15.2).

These ideas were first formalized in simulation studies of clade growth and evolution by Maurer et al. (1992), whose approach was developed still further by Kozlowski and Gawelczyk (2002). The simulation models contain the following variables, which were examined to observe if they can produce the shape and range of observed patterns. First, the body size of the clade founder; second, the size-dependence of speciation rate; third, the size-dependence of extinction rate, and fourth, the presence of a 'reflecting barrier' at small body sizes. The latter represents a size limit below which species cannot, or find it difficult to persist due to physiological constraints. Kozlowski and Gawelczyk (2002) made this a gradual constraint such that anagenesis to smaller body sizes becomes increasingly less likely below a certain size. Fifth, body size change can occur either at speciation events, or between them (anagenesis), although the direction of change was assumed to conform to a random drift process (i.e. no trend).

The results of the models are mostly intuitive. The presence of the graded reflecting barrier produces the short left-hand tail rather than a truncated distribution. The simulations produce a range of skewness as seen in nature due to the stochastic nature of all the processes assumed.When the net diversification rate favours small species, more right skewed distributions are produced and they tend to become more heavily right skewed. When the reflecting barrier is present, small founder body size tends to increase skewness and proportion of skew, although without it the opposite is the case. The timing of body size change affects the relative importance of size-biased extinction and speciation.

In general then, these evolutionary processes are capable of producing the distributions seen in nature and certain combinations are more likely to do so than others. For expiricists the challenge is now to quantify them in nature and compare to theoretical predictions. Most work on this has been directed at the macroevolutionary processes of speciation, extinction, and net cladogenesis. Orme et al. (2002) have studied size-related trends in net cladogenesis by observing the correlation between body size and species richness across 38 species-level phylogenies across range of vertebrates and invertebrates. Unfortunately, the results pose more questions than they answer. There is no overall trend towards an association between body size and net rate of cladogenesis, and in fact only one study does show this—a genus of flies (Bitheca), where small body size increases the net rate of cladogenesis. Furthermore, there was no relationship between body size and distance from the root of the tree, so it appears that there are no strong and consistent anagenetic trends. These do sometimes appear, however, in particular clades; significant relationships are found in eight clades; in primates,body size increases with distance from the root,while in carnivores it decreases; overall, three of these cases show a positive relationship and five are negative.

These results do seem at odds with the evolutionary models, and there are several possible explanations; one is that there is actually no consistent skew in the body size distributions of those taxa examined. If this is the case, the studies can hardly be considered as explanations for positive skew. However, the studies are also all much smaller than those generally used to describe the patterns of interest, and, as described earlier, taxa of small taxonomic rank tend to display less skewed body size distributions. It may be therefore that we need to wait for larger phylogenies to become available. An alternative possibility, not considered in the modelling studies, however, is that small size is not a consistent correlate of species diversity but that large radiations might tend to be small bodied. In fact this pattern has been documented in mammalian clades by Gardezi and da Silva (1999). Such radiations would have large effects on species richness but represent only single data points so are unlikely to affect a general bias in cladogenesis. Another suite of studies have examined the relationship between present day extinction risk and body size in a number of groups. They generally conform to the studies of cladogenesis in that some positive relationships are found, some negative

Patterns

Proximate evolutionary processes

Ultimate ecological forces

Patterns

Proximate evolutionary processes

Ultimate ecological forces

Fig. 15.3 Processes contributing towards macroecological patterns at very large (e.g. global) scales. The patterns are the direct result of evolutionary processes, which are thus proximate forces. These forces occur in the way they do because of ecological forces, which are thus ultimate.

Fig. 15.3 Processes contributing towards macroecological patterns at very large (e.g. global) scales. The patterns are the direct result of evolutionary processes, which are thus proximate forces. These forces occur in the way they do because of ecological forces, which are thus ultimate.

ones, and in some there is no relationship (Purvis et al. 2000). In general then, there is presently scant evidence for any of the postulated evolutionary processes that might have given rise to body size distribution patterns; the reason for the mismatch must be found.

Even if we had a good knowledge of the evolutionary processes mentioned above, that would remain an incomplete explanation, for underlying these must themselves be causal processes (Figure 15.3). Three major ultimate reasons have been postulated. First, there might be more resources or niches available for smaller bodied species that might increase rates of cladogenesis or affect anagenetic trends. In particular one can describe the relationship between body size and number of individuals from measurements of the habitat at different scales. Some studies of arthropod communities (Morse etal. 1985; Shorrocks etal. 1991) show reasonably good fit between the observed and predicted number of individuals. Of course the link between amount of resource/niche space and evolutionary process remains rather implicit. A second postulated ultimate process is that organisms might be evolving towards some theoretical optimum size for the clade based on reproductive power (Brown et al. 1993). The idea here is that within a clade there will be some characteristic body size at which most energy becomes available for reproduction, due to the differential between the mass specific gains and losses of energy. Brown et al. (1993) predict that the relationship between production rate (giving reproductive power) and body size is shaped rather like species body size distributions. Other factors not in the model can combine to drive a species away from the optimum and this is more likely when the reproductive power is close to the maximum achievable. Thus, the species distribution comes to resemble the body size-production relationship.

The Brown et al.(1993) model apparently predicts the modes of some body size distributions quite well (Maurer 1998; Roy et al. 2000). However, these tests have themselves been criticized on the grounds of loose application of the necessary parameters (Kozlowski 2002). In addition the model itself has been criticized on a number of mathematical issues, points of internal consistency, and on biological and evolutionary grounds (for a review of these points see Kozlowski 2002). In addition two further tests of the model, both of which look for the proposed switch point in reproductive power at the modal body mass, fail to confirm the model prediction (Jones and Purvis 1997; Symonds 1999). Once again, it is rather implicit which evolutionary processes actually lead to the species body size distribution. Presumably ana-genetic change is involved, as the model is really about what an individual species' body size should be. Presumably though, how many species get close to that could be affected by speciation or extinction rates, which might themselves be affected by the relatively reduced reproductive power away from the optimum.

An alternative kind of optimality approach has been developed by Kozlowski and Weiner (1997), in a model that developed Charnov's (1991) model of mammalian life history evolution. In Chapter 4, we saw how the trade-off between production and mortality could combine to influence when a species should mature, hence, its optimum adult body size. Kozlowski and Weiner (1997) investigated a slightly more complex model than Charnov's in that they derive production from separate allometric relationships of assimilation and respiration. Varying the parameters of this model by drawing them at random from normal distributions happens to produce distributions of optimal species body sizes that are right skewed and shaped very like those in nature (see also Kindlmann et al. 1999; Kozlowski and Gawelczyk 2002). Exactly how this optimization process affects the required evolutionary processes is again implicit; presumably ecological conditions make some parameters more common than others, our third ultimate process, and this affects the frequency of optima and hence of anagenetic change towards those optima. Not all organisms optimize body size in this way (parasitoids and birds are obvious exceptions—see Chapter 4), but the model illustrates the general principle of how evolution on a per species basis towards individually determined optima can lead to species body size distributions.

We will now turn to another well-known pattern and see how the same sets of evolutionary principles can help explain that.

Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook


Post a comment