SECT. 2] AND WEIGHT 513 



effect of temperature on the insect embryo, and has in many 

 cases recalculated the figures of earlier observers, such as those of 



Abbe : hatching of the Rocky mountain locust [Melanoplus femur- 

 rubrum) . 



Regener : hatching of the Pine bug {Dendrolimus pini) . 



Hennings : hatching of the Typographic beetle ( Tomicus typographus) . 



Quaintaince & Brues : hatching of the bollworm {Heliothis obsoleta) . 



Girault : hatching of the bollworm {Heliothis obsoleta) . 



Jenne; Hunter & Glenn: hatching of the codling moth [Carpocapsa 

 pomonella) . 



Jenne ; Hunter & Glenn : hatching of a Braconid parasite [Lysi- 

 phlebus tritici) and of the green bug ( Toxoptera gramnium) . 



Hunter & Hooker : hatching of the Texan cattle-fever tick [Mar- 

 garopus annulatus) . 



All these data without exception, together with figures collected by 

 Sanderson himself on the incubation oiEuproctis, Samia, Tenebrio, Lepti- 

 wo^arj"a,Afa/(2coj'oma, etc., gave curves closely resembling those of Hertwig 

 for the development of the frog. Thus the Q^^q for Jenne's codling 

 moth curve would work out at approximately 2-8 between 14 and 

 24°, and that for Samia cecropia would be i-g or so between 20 and 30°. 

 In most cases the usual rise in temperature coefficient with decreasing 

 temperature was shown, and in Samia, for instance, Q^io"^ould be 2-5 

 between 10 and 20°. It is extraordinary that Sanderson made 

 no calculation of temperature coefficients, but he was particularly 

 interested in the practical application of his work and computed his 

 results according to an empirical day-degree system which, as Martini 

 showed, has no physico-chemical meaning, and even confuses in- 

 tensity with quantity of heat. It is curious that ichthyologists on 

 the one hand and entomologists on the other should have evolved 

 very similar treatments of time/temperature curves both equally 

 unsatisfactory. The temperature coefficients for these insect curves 

 would work out somewhat as shown in Table 72. Peairs later 

 tried to show that, for certain insects, the curve relating time taken 

 in embryonic growth with temperature was a hyperbola, and he did 

 indeed, over a short range, demonstrate the reciprocal to be a straight 

 line, as, if that were the case, it should be. Peairs paid no more 

 attention to the temperature coefficient question than did Sanderson. 

 Precisely analogous experiments and results were obtained by Blunk 

 working on Dytiscus marginalis and by Bodine on various grasshoppers. 



NEI 33 



