INSECT SURVIVAL IN RELATION TO THE RATE OF WATER LOSS

Summary 1. If the relationship between rate of water loss from an insect and saturation deficiency is linear and survival is limited only by water loss, then the curve of longevity against saturation deficiency is hyperbola‐shaped. The position of this curve with respect to the co‐ordinates is, however, affected by other factors, which are discussed. 2. The work of Bacot & Martin (1924) on the longevity of the rat flea, Xenopsylla cheopis , is reviewed in the light of the above hypothesis: errors in curve fitting are revealed which make it necessary to modify some of Bacot & Martin's conclusions. 3. Selected data from various authors subsequent to Bacot & Martin are considered. The discrepancies, real and apparent, between Leeson's results with rat fleas and those of Bacot & Martin are analysed, and it is pointed out that these were to some extent due to errors in curve fitting. The hypothetical relationship between longevity, rate of water loss and saturation deficiency set out in the first part of this article are illustrated by Ludwig's (1937) data for Popillia japonica; the linear relationship between 1/longevity and the rate of water loss demonstrates the hypothesis that water loss limits survival with this insect. Other less complete data are also discussed. 4. Various types of longevity‐saturation deficiency curves are described and the causes which probably account for such variations. 5. The relationship of pupal or egg mortality to saturation deficiency cannot by itself throw light on the variation of rate of water loss with saturation deficiency; the time factor must also be taken into account. Lines of equal mortality may coincide with lines of equal saturation deficiency (the co‐ordinates being relative humidity and temperature), so that the product of mortality and saturation deficiency is constant at any one temperature; but this indicates only that temperature is without effect on mortality and does not throw light on the relation of mortality to different saturation deficiencies. Such coincidence of lines cannot, therefore, be taken as proof that Dalton's law operates. This work was done while I held the Avebury Studentship in the Department of Entomology, London School of Hygiene and Tropical Medicine, and I am greatly indebted to Prof. P. A. Buxton for his great kindness and ready help. I also thank Drs W. J. Martin and D. L. Gunn for their advice and criticism.