284 THE BNTOMOLO&ISt. 



already stated, only intermittently applied, as they could only 

 be borne for short periods (in most of the many experiments 

 for two periods per day of two hours each for three to six 

 days) ; between the exposures and after same, and until emer- 

 gence, the pupae were kept at the ordinary daily temperature. 

 In the heat and frost experiments the greater part of the 

 material experimented with remained unchanged ; only the 

 smallest proportion of them departed in any way from the 

 normal type, and, most curiously, this portion was of equal 

 value in both heat and frost experiments. The same or almost 

 similar forms were obtained at —15° to —20° C. as at + 40° to 

 +45° C, that is, in both aberrations arose. These strange 

 results lead to the question — In what way do these frost and 

 heat experiments work ? 



Observation shows that if a large series of pupae be exposed 

 for two hours twice a day for three days in the frost apparatus 

 to a temperature of about — 12° C, the imagines were not de- 

 layed only the six hours, during which the pupse were in the frost 

 apparatus, but were delayed days. It is also remarkable, that 

 by substituting +44° C. for —12° C. analogous results were ob- 

 tained. High temperatures by no means hasten matters, as one 

 would be inclined to suppose, but mostly interrupt the develop- 

 ment, often delaying the same long over the extent of the time 

 of exposure. 



This retarded development under very high temperatures was 

 most noticeable when a species, whose one generation hybernates 

 as pupae, had the pupae of its summer generation submitted to 

 about + 44° C. Then a certain proportion of the pupae do not 

 emerge as a second generation, but hybernate, and give the 

 following spring imagines of the normal type of the first genera- 

 tion. This took place with broods, the remainder of which, left 

 under normal conditions, emerged all in the second generation. 

 The interruption in the development became in this case per- 

 manent, and the forced aestivation passed directly over into the 

 normal hybernation of the following generation. It may also be 

 mentioned here, which is under the circumstances more compre- 

 hendable, that in frost experiments with summer pupae of species 

 with two generations, a certain percentage according to the degree 

 and length of the frost action do not reach a second generation, 

 but hybernate. 



We have thus to first answer the question — How do these 

 frost and heat experiments work ? They arrest development, 

 and put the insect into a state of lethargy, whereas there were 

 no such interruptions in the warmth and cold experiments. 



But a further important difference exists between these two 

 series of experiments. Cold and warmth experiments resulted in 

 a change in the whole of the material under different treatment 

 in different directions and degrees, whereas with uniform treat- 



