The apparent decoupling of thermal limits in insects is of interest because it is unusual, at least by the standards of marine species. In many of these organisms, upper and lower lethal tolerances vary in synchrony. Before discussing this issue in detail it is worth pausing to determine just what is meant by coupled or correlated responses. Huey and Kingsolver (1993) argued that if selection results in a shift of the entire performance curve (Fig. 6.1), then performance at high and low temperatures is inversely correlated. That is, if performance improves at high temperature, it declines at low temperature. Similarly, Hercus et al. (2000) argue that a positive correlation between mean knockdown times at high temperature and greater cold resistance means an improvement in both cold resistance and heat resistance. Although this terminology appears to be straightforward, it is partly dependent on the way in which performance is measured. In both examples, the authors indicated that a positive correlation meant that an improvement in tolerance of one stress was associated with an increase in tolerance of the other. However, in some circumstances the term 'correlation' could be misleading, depending on the measure of tolerance used. Thus, if knockdown temperatures (critical minima and maxima) were simply plotted against each other, a negative correlation would imply that a decline in CTmin accompanies an increase in CTmax. This is precisely the converse of the terminology adopted by Huey and Kingsolver (1993). However, if the time to mortality for 50 per cent of the sample at each temperature was plotted, an inverse correlation would be perfectly in keeping with their terminology. In the present discussion, decoupling between high and low temperature responses is taken to mean that they are unrelated, or at best weakly inversely correlated such that an improvement in heat resistance means a decline in cold resistance (i.e. a shift of the performance curve).
In marine species it has long been known that performance at high and low temperatures is often inversely correlated among populations or species. That is, an improvement in high temperature tolerance leads to a decline in low temperature tolerance and vice versa (Prosser 1986; Cossins and Bowler 1987). Building on previous work (Ushakov 1964; Prosser 1986), Portner and his colleagues (Portner 2001, 2002; Portner et al. 1998, 2000) have provided the most extensive exploration of this relationship. In essence, they have argued that in complex metazoans, critical temperatures that affect fitness (i.e. survival and reproduction) are not set by cellular level responses (such as the stress protein response), but are rather set by a transition to anaerobic metabolism. These pejus (= deleterious) temperatures, which are less extreme than traditionally measured critical limits, result from insufficient aerobic capacity of mitochondria at low temperatures, and a mismatch between excessive oxygen demand by mitochondria and insufficient oxygen uptake and distribution by ventilation and circulation at high temperatures. In other words, whole-animal aerobic capacity is limited at both low and high temperatures, and this sets limits to animal performance. Adjustments to temperatures both seasonally, and over evolutionary time are made by altering mitochondrial densities, and this change has concomitant effects on both high and low deleterious temperatures. In consequence, there is an inverse correlation in performance at high and low temperatures when measured in either a population across seasons, or among populations from habitats differing in their thermal regimes. However, the nature of the change in deleterious temperature depends to some extent on whether stenothermal or eurythermal species are being examined. In addition, it appears that the width of the tolerance window between the upper and lower deleterious temperatures varies substantially with latitude, being broader in tropical and temperate than in polar species. That is, tropical and temperate species tend to have a broader tolerance range than do polar ones (Portner 2001).
Despite apparently limited information, Portner (2001) has argued that although terrestrial species may be more eurythermal, owing to a reduction in cost of ventilation (a result of higher oxygen levels), thermal tolerance limits in these species may be set in a way similar to those of marine taxa. However, in terrestrial insects, decoupling of upper and lower lethal limits at global and regional scales, between populations of the same species, and in selection experiments (see Chapter 5 and Chown 2001), suggests that this is not the case. This does not necessarily imply that oxygen delivery has no effect on thermal tolerance limits. Indeed, in several species anoxia and hypoxia have profound effects on tolerances, generally reducing them substantially (Yocum and Denlinger 1994; Denlinger and Yocum 1998, but see also Coulson and Bale 1991 who show that anoxia actually induces rapid cold hardening in house flies). Nonetheless, a major prediction of Portner's hypothesis—that a decline in critical temperature is expected to accompany hypoxia (Portner 2001)—which, to date, has been tested only once in insects, has not been supported. An assessment of the effects of both hyperoxia and hypoxia on critical thermal maxima, using a novel technique (thermolimit respirometry—Lighton and Turner 2004), on a tenebrionid beetle, revealed no effect of both treatments (Klok et al. 2004) (Fig. 7.1).
Was this article helpful?