Temperature and water availability are both thought to influence metabolic rate, especially over the longer term, resulting in adaptations that apparently reflect the need either for water conservation or starvation resistance, or the response to low environmental temperatures (Chown and Gaston 1999). The influence of temperature on metabolic rate over short timescales has been called the most overconfirmed fact in insect physiology (Keister and Buck 1964), and acute modifications of metabolic rate by temperature are certainly widely known for insects, with many modern studies continuing to document them. The short-term influence of humidity on metabolic rates has also been documented in several species, though with the advent of flow-through respirometry these effects are often not investigated, largely because rate measurements are made in dry air for technical reasons. In at least some instances, increases in metabolic rate with declining humidity may be the result of increased activity, rather than any other fundamental alteration in metabolism or gas exchange (Lighton and Bartholomew 1988).
Much of the discussion of temperature effects on metabolic rate has been undertaken under the rubric of Q10, or the change in the rate of a process with a 10oC change in temperature. Whether this concept (and the Arrhenius equation, see below) should be applied to the whole organism, rather than to specific reactions is an argument that dates back at least to Krogh (Keister and Buck 1961), and continues to generate discussion. Nonetheless, it is now widely known that whole-animal Qj0 varies with the temperature range over which it is measured (Keister and Buck 1964; Cossins and Bowler 1987). In consequence, Qj0 cannot strictly be called a parameter and should rather be termed the apparent Q10 when calculated in the normal fashion (Chaui-Berlinck et al. 2001). Moreover, the apparent Q10 is also supposedly inappropriate for deriving conclusions regarding metabolic control (i.e. up- or down-regulation of the response to temperature), and only if a Qi0 is assumed in advance of an analysis can conclusions regarding Qi0 be made (Chaui-Berlinck et al. 2001). However, there are substantial problems inherent in the solutions proposed by Chaui-Berlinck et al. (2001) to the 'apparent Q10' problem.
The temperature dependence of Qi0 was also recently rediscovered by Gillooly et al. (2001), who suggested that their 'universal temperature dependence' (UTD) of biological processes should be used in the place of Qi0 to describe rate-temperature relationships. This UTD amounts to little more than a rediscovery of the Arrhenius equation, originally proposed at the turn of the last century, and extensively discussed by Cossins and Bowler (1987). Here, the relationship between rates at two temperatures is given by k1 = k2e»lR((T T2)/T1 T2); (9)
where, k1 is the rate at absolute temperature Tj, k2 is the rate at absolute temperature T2, ^ is a constant (the Arrhenius activation energy or critical thermal increment), and R is the universal gas constant (8.314 J moP1).
The novelty of the Gillooly et al. (2001) approach is that it takes a concept initially applied within organisms (and species) and applies it to a cross-species problem. However, this application has generated some concern (Clarke 2004), and its validity, especially given reasonably constant metabolic rates of insects across latitude (Addo-Bediako et al. 2002), remains an issue open for discussion.
Although the Arrhenius equation does resolve the problems associated with temperature dependence of Q10, Cossins and Bowler (1987) have noted that over the range of biologically relevant temperatures (0-40oC), the two measures scarcely differ. Given the concerns raised by Chaui-Berlinck et al. (2001) it would seem most appropriate to make use of the universal temperature dependence (or Arrhenius activation energy), which can readily be calculated from most data obtained for investigation of Q10 effects. In this context it is also worth noting that the use of analysis of variance (ANOVA) for assessing the temperature dependence of metabolic rate, as is sometimes done (Duncan and Dickman 2001), is inappropriate. ANOVAs are generally much less sensitive than regression techniques for determining the relationship between two variables (see Somerfield et al. 2002), and in consequence give spurious conclusions. In this context, ANOVAs might lead authors to conclude that there are no effects of temperature on metabolic rate, when in fact these effects are profound.
Long-term responses of insect metabolic rates to temperature have also been documented in many species and have generally been discussed in the context of the conservation of the rate of a temperature-dependent physiological process, in the face of temperature change. This temperature conservation, or metabolic cold adaptation (MCA), has been the subject of considerable controversy both in ectotherms in general (Clarke 1993), and in insects (Addo-Bediako et al. 2002; Chown et al. 2003; Hodkinson 2003).
Elevated metabolic rates of species (or populations) from cold climates relative to those from warm climates (at the same trial temperature) are thought to allow the species showing them to meet the elevated ATP costs of growth and development (Wieser 1994) necessary for completion of life cycles in the relatively short, cool growing seasons that characterize cold (high latitude or altitude) regions (Chown and Gaston 1999). Such high growth rates and their elevated metabolic costs are not maintained in species from all environments because of the likely fitness costs of rapid growth (Gotthard et al. 2000). Thus, MCA represents not only a physiological response to environmental temperature, but also a significant component of the life history of an organism, which allows it to respond, via internal alterations, to its environment. Metabolic cold adaptation in insects has been discussed extensively by Chown and Gaston (1999), and they have highlighted the importance of distinguishing between intra- and interspecific levels when addressing MCA. While the outcomes of intraspecific studies are varied, though often finding support for MCA, the ways in which interspecific studies should be undertaken and the outcome of these investigations are much more controversial.
At the intraspecific level, only a few investigations have been undertaken across latitudinal gradients, with some finding no evidence for MCA (Nylund 1991), while others have found variation in line with predictions of the MCA hypothesis (Berrigan and Partridge 1997). Presumably for reasons of species geographic range sizes (most tend to be small), studies across elevational gradients are more commonplace. In general, these have found evidence in favour of metabolic cold adaptation (Chown and Gaston 1999), although Ashby (1997) has pointed out that simultaneous changes of body size across elevational gradients (Blackburn et al. 1999), and preferred temperatures of the individuals at each site, complicate interpretation of the evidence. In her study, mass-specific metabolic rates of Xanthippus corallipes (Orthoptera, Acrididae) increased with elevation, though body mass declined, leading to conservation of rate across the gradient. In another grasshopper, Melanoplus sanguinipes, body size varied in the opposite direction to that found in X. corallipes, suggesting that size effects cannot account entirely for their higher metabolic rates at high elevations. Nonetheless, cooler activity temperatures meant that compensation was not complete (Rourke 2000) (Fig. 3.14). These studies highlight the need to take both size and temperature into account in investigations of MCA, something that, until recently, has not been routinely done. Body size can be particularly problematic in this regard because it might either increase or decline with elevation (or latitude), even within closely related taxa (see
Was this article helpful?