We are writing about a paper recently published in The Journal of Experimental Biology (Yadav and Sharma, 2014). We wish to point out that (i) our work, and that of others, has been cited inappropriately in support of statements that the cited work either refutes or, at the least, does not support; (ii) one principal conclusion drawn about greater reproductive fitness in populations selected for faster development is wrong; (iii) a principal result about populations selected for faster development evolving greater fecundity per unit dry weight is not novel, as depicted, but has been shown before twice; and (iv) the authors provide no discussion of an unexpected and counter-intuitive result of fecundity increasing from young to middle age in some of their populations.
The first two sentences of the Introduction are erroneous citations of Prasad and Joshi (Prasad and Joshi, 2003). We did not state that fruitfly populations in nature are subject to directional selection for faster development. On the contrary, we stated that recent studies on the evolution of development time challenged the earlier held view that fruitflies were subject to selection for faster development in the wild, and then went on to review those studies [see p. 55 of Prasad and Joshi (Prasad and Joshi, 2003)]. Indeed, over the past 18 years, one of us (A.J.) has been consistently writing and speaking about why the earlier held view is wrong. Similarly, we (Prasad and Joshi, 2003) have nowhere claimed that studies on Drosophila led to the development of life history theory, or that this theory ‘posits that natural selection enhances organismal fitness’. Both these statements have not been made by us and, moreover, are incorrect. The fourth sentence of the Introduction is also incorrect: Chippindale et al. (Chippindale et al., 1994) do not report on any populations subjected to selection for slower development. Populations selected for postponed aging showed slower development as a correlated response (Chippindale et al. 1994). The second sentence of paragraph 3 of the Introduction is likewise incorrect. Resistance to starvation and desiccation are not fitness components under laboratory conditions and are not used to assess fitness in fruit flies. The references cited in support of this statement are also erroneously cited: those authors have not claimed that these traits are used to assess fitness. In the section headed ‘Adult lifespan’, Sheeba et al. (Sheeba et al., 2000) are cited in support of the statement that there are deleterious effects of light on lifespan and these are supported by experiments showing greater lifespan in constant darkness than in constant light or a light:dark cycle. However, this is also a poor citation: the main thesis of Sheeba et al. (Sheeba et al., 2000), as the title indicates, was that we need to be careful in determining deleterious effects on fitness through lifespan alone, because fruit flies kept under constant light had lower lifespan but higher reproductive output compared with flies kept in constant darkness or in a light:dark cycle.
The authors observed that faster developing populations produced fewer eggs overall, but more eggs per unit dry weight, as compared with ancestral controls (Abstract, fig. 1D,E and final sentence of paragraph headed ‘Higher fecundity per unit body weight in faster developing flies’). Reproductive fitness depends on total egg production, not fecundity per unit dry weight. Yet, the authors repeatedly and erroneously claim that their faster-developing flies have higher reproductive fitness than controls (Abstract, and elsewhere in the paper). What the results on fecundity suggest is that the faster-developing flies, which have higher Darwinian fitness than slower-developing flies under the authors' selection regime, are achieving this partly at the cost of reproductive fitness that is reduced as a consequence of the smaller size that accompanies rapid development, as has also been shown earlier in the context of selection for rapid development by Prasad and Joshi (Prasad and Joshi, 2003) and Chippindale et al. (Chippindale et al., 2004) (see below).
The authors' observation that faster-developing populations produced more eggs per unit dry weight compared with ancestral controls is presented as though it were a novel finding. Exactly this result has been seen twice before in fruit fly populations selected for faster development under constant light (Prasad and Joshi, 2003; Chippindale et al., 2004). Both these papers are cited by the authors, but it is nowhere mentioned that the authors' finding is, thus, a confirmatory result and not a novel one. Indeed, the fact that the authors, selecting under constant darkness, got essentially the same pattern of correlated responses to selection for faster development as Prasad and Joshi (Prasad and Joshi, 2003) and Chippindale et al. (Chippindale et al., 2004) did when selecting under constant light (when the fruitfly circadian clock is arrhythmic), clearly suggests that these correlated responses of life-history traits to selection for faster development are related to development time per se, and are unlikely to be clock-mediated, contrary to what the authors infer in the final paragraph of their Results section. Similarly, in the Abstract, the authors write ‘In order to rigorously examine correlations between pre-adult development and other life history traits…’ as if this is the first such study, when such rigorous examinations of correlated responses to selection for faster development have been done at least in four independent studies on D. melanogaster by Len Nunney, Bas Zwaan, Adam Chippindale and N. G. Prasad. In the paragraph headed ‘Reproductive output’, the authors contrast their finding of both lifespan and fecundity being depressed in their selected populations with some previous studies, but do not mention that exactly the same result was also seen by Chippindale et al. (Chippindale et al., 2004).
In their control populations, the authors observed higher fecundity on days 21 and 11 of adult life than at day 4 (fig. 1D). This is somewhat unprecedented in D. melanogaster, which has a typical triangular fecundity function, with maximum fecundity very early (days 2–5) in adult life. Strangely, there is no discussion whatsoever about this rather surprising and counter-intuitive result. We hope that the authors will take appropriate measures to set the record straight on these points.
- © 2015. Published by The Company of Biologists Ltd