these two avenues of water loss should be in the absence of strong selection for water savings. Second, if selection were to act in concert to reduce cuticular and respiratory transpiration, then no proportional change in the two components might be detected at all. If respiratory water loss constitutes a small, but important, proportion of overall water loss initially, then even a considerable reduction in the latter is unlikely to translate to a particularly large proportional increase in the former. For example, if respiratory transpiration represents 5 units of a total of 100, this amounts to 5 per cent, whereas if it constitutes 5 units of a total of 50 (after a c. 50 per cent reduction in cuticular transpiration and overall water loss), the percentage contribution rises to only 10 per cent. Indeed, expressing the relationship between cuticular and respiratory transpiration as a proportion seems more likely to obscure than to clarify investigations of the importance of respiratory transpiration. As is the case in other fields of physiology (e.g. Packard and Boardman 1988; Raubenheimer 1995), investigations of the components of water loss would benefit substantially from construction and statistical analysis of bivariate plots of the variables involved.
Many of the misgivings regarding the importance of respiratory transpiration have arisen as a consequence of recent work examining the partitioning of water loss in species showing discontinuous ventilation or DGCs. Water conservation was long thought to be the chief adaptive function of DGC in insects. Support for this hypothesis comes from studies such as that of Lighton et al. (1993b), who measured real-time water loss rates in female harvester ants Pogonomyrmex rugosus by both IR absorbance and gravimetric means, and found that water loss during the open phase was 2.8 times higher than cuticular water loss alone. However, the water conservation hypothesis has been questioned for two major reasons. First, as we have seen, respiratory water loss is often too small a component of total water loss to have selective significance. Second, the distribution of DGC among insects is patchy and many species from xeric regions, in particular, apparently do not exhibit cyclic patterns of gas exchange (Lighton 1994; Lighton and Berrigan 1995; Lighton 1996).
An alternative to the water conservation hypothesis was proposed by Lighton and Berrigan (1995), who suggested the DGC may be at least as important in facilitating gas exchange in hyper-capnic (high CO2) and hypoxic (low O2) environments (Lighton 1996, 1998). Such environments may be encountered by psammophilous and subterranean beetles and ants, taxa which often exhibit DGC (Bosch et al. 2000). In such conditions, the DGC leads to increased concentration gradients for both O2 and CO2, thus increasing diffusional gas fluxes, and water conservation might be a secondary benefit (Chappell and Rogowitz 2000; Duncan and Byrne 2000). Other possible functions (not necessarily mutually exclusive) have been discussed in Chapter 3 (Section 3.4.4). The topic remains controversial: a recent analysis of water loss in five species of Scarabaeus dung beetles exhibiting DGC (Chown and Davis 2003) supports the idea that modulation of DGC characteristics and metabolic rate can be used to alter water loss rate, and the changes are consistent with differences in habitat of the five species.
Flightless insects: does the subelytral cavity conserve water?
Many arid-adapted beetles are flightless and their fused elytra enclose a subelytral cavity above the abdomen. Because the abdominal spiracles open into this space rather than directly to the exterior, the reduced gradient of water vapour pressure is widely assumed to aid in water conservation (for references see Chown et al. 1998). However, the vapour pressure of air in the subelytral cavity has never been directly measured. Tracheal interconnections mean that it is possible for the whole respiratory system of an insect to be supplied by tidal air flow via a single open spiracle. In the dung beetle Circellium bacchus (Scarabaeidae) the main route for respiratory gas exchange at rest is the right mesothoracic spiracle, which lies outside the subelytral cavity (Duncan and Byrne 2000, 2002) (presumably the left spiracle is involved in other individuals, and both during activity). Similarly, the Namib Desert tenebrionid beetle Onymacris multistriata uses mesothoracic spiracles and not the subelytral cavity for gas exchange with the atmosphere (Duncan and Byrne 2000). The subelytral spiracles do, however, play a role in respiration: sampling of air from inside the subelytral cavity of C. bacchus shows sequestration of CO2 and water, which are then expelled through the mesothoracic spiracle, and the DGC patterns of anterior and posterior spiracles are synchronized (Byrne and Duncan 2003).
Respiratory water loss during flight Respiratory water loss increases dramatically during activity and especially flight (Harrison and Roberts 2000). High metabolic rates during flight require increased ventilation and increased respiratory water loss, but also generate metabolic water. Weis-Fogh's classic study of ventilation in tethered flying locusts (Schistocerca gregaria) demonstrated a 10-fold increase in water loss to 8 mgg_1h_1 at 30°C, much of which can be attributed to increased respiratory water loss, although this is offset by metabolic water production. A balance between respiratory losses and metabolic water production can be maintained during many hours of sustained flight, depending on ambient temperature and relative humidity: locusts migrating under desert conditions must fly at high altitude to avoid dehydration (Weis-Fogh 1967b).
Studies of respiratory water loss during flight are sparse, but they include work on challengingly small insects. Evaporative water loss has been measured during tethered flight in aphids: Cockbain (1961) mounted Aphis fabae (0.8 mg) on fine pins in a wind tunnel, and estimated that only 1 per cent body mass per hour was lost by evaporation during a 6 h flight at 25°C. Forty years later, Lehmann et al. (2000) investigated scaling effects on respiration and transpiration during tethered flight in four species of Drosophila (0.65-3.10 mg). They used moving visual stimuli to induce the flies to alter their energy expenditure in flight, and measured regular changes in CO2 production and respiratory water loss. Metabolic water production from the oxidation of glycogen compensated for 23 - 73 per cent of total water loss, depending on species, but this was much greater than its contribution to water balance in resting flies (Fig. 4.4). An exciting finding is that the fruit flies minimized the risk of desiccation during flight by modulating
Was this article helpful?