Measuring Developmental Maturity

The age of immature insects can be estimated from their developmental maturity, represented by their physical growth or by the developmental milestones they have reached. Embryogenesis and metamorphosis involved little growth but extensive and complex changes in morphology that provide a rich suite of developmental events or 'milestones' for estimating the age of eggs (since fertilization, not necessarily since oviposition) (Bourel et al. 2003) and puparia (Denlinger and Zdarek 1994; Greenberg and Kunich 2002).

On the other hand, the larval stage is dominated by growth and larvae pass through fewer physical developmental events - essentially just two ecdyses. With the exception of preserved specimens in which the new spiracles and mouthparts can be seen through the cuticle (indicating imminent ecdysis), the timing of ecdysis is observed only in live larvae because it is a brief event. The age of dead larvae can be estimated from their mass, length or width (Day and Wallman 2006a; Donovan et al. 2006; Grassberger et al. 2003; Queiroz 1996). Mass is difficult to measure, especially if specimens have been preserved (because lipids may leave the body, and water may flux), and few data are available for this variable. Length and width can be measured quickly, accurately and precisely on live or preserved larvae using the geometrical micrometer devised by (Villet 2007) or a calibrated microscope. The measurement precision possible is about 0.1 mm, but the accuracy is compromised by changes in size due to the method of killing and preservation (Adams and Hall 2003; Midgley and Villet 2009a; Tantawi and Greenberg 1993). It has been suggested that width is more reproducible than length in fly larvae (Day and Wallman 2006a), but because the relative error will be greater using the same measuring equipment, the measurement precision is lower. Midgley and Villet (2009a) found that the variation produced by preservation in a beetle species is about 25-30% of live length, depending on which statistical method is used to summarise the variation, and that the distribution is often asymmetrical (i.e. biased) relative to live length. Sample precision therefore outweighs measurement precision by about an order of magnitude, especially for older larvae.

However, the estimate precision associated with size may be even more limiting. As in humans, conspecific larvae of the same size may be very different ages (Fig. 7.3), and larvae of the same age may differ considerably in size (Fig. 7.3), so that size is not precisely correlated to physiological or chronological age (Dadour et al. 2001; Richards et al. 2008). Larvae of the same age show coefficients of variation of 10-35% for length and mass, even when large samples are measured carefully. This can be ameliorated to some extent by selecting a biased sample of only the largest larvae to measure (Fig. 7.3; (Amendt et al. 2007)), but requires larger samples and the absence of precocious larvae. Size is therefore not favoured by some investigators, who use developmental milestones instead (Dadour et al. 2001; Gaudry et al. 2001). Given the extensive suite of developmental events in embryogenesis and metamorphosis, it is better to estimate the age of eggs and larvae from material preserved as soon after discovery as possible, because the passage of more time between collection and preservation will only multiply the uncertainties of estimation. On the other hand, if one wishes to use developmental events to estimate the age of larvae, they will have to be kept alive until they reach a recognisable event. However, since larvae are commonly presented as preserved material, it is likely that their age will have to be estimated from their length or width, taking into account the instar and the methods of killing and preservation. Ideally, if specimens are to be preserved in investigations and experiments, fly larvae should be killed in hot water (Amendt et al. 2007) and beetle larvae in ethanol (Midgley and Villet 2009a) and their length measured to maximise sample and measurement precision.

A recent development is the investigation of physiochemical, as opposed to physical, developmental processes, which promise greater temporal resolution in the larval stage. These include the ontogeny of cuticular hydrocarbons in eggs (Roux et al. 2008), larvae (Roux et al. 2008; Zhu et al. 2006), puparia (Roux et al. 2008; Ye et al. 2007; Zhu et al. 2007) and adults (Roux et al. 2008; Trabalon et al. 1992), and changes in pupal steroids (Gaudry et al. 2006). The age of adults can also be estimated by the accumulation of the red pigment pteridine in their eyes, the development of the ovaries (in females!) and even to some extent by the degree of wear on the wing margins (Hayes et al. 1998; Wall et al. 1991). The use of hydrocarbons and pteridine requires calibration for geographical (and other) variation, but the factors affecting the precision and bias of these estimates are largely unquantified; (Gaudry et al. 2006) discuss the measurement and interpretation of ecdysteroid levels.

0 0

Post a comment