INSECTS OF REVERSED BEHAVIOR. 399 



speed as the temperature rises, which limit may be actually 

 reached by some of the southern species on very warm days. 

 The writer 9 has made one interesting observation in this regard. 

 California and multicolor on very hot days when the temperature 

 i- from 100 to 105 take frequent rests, by hanging up in the 

 r-hade every few minutes. This was noted in the Yakima de-ert 

 at Sunnyside, Washington. Apparently, for the-e at least, 

 someuhere in the hrjier nineties a temperature is reached above 

 which increa-e is depn and no longer stimulating. How- 



i we can -.ifely say that fur all lowland species of .EsJnia an 

 increase of temperature up to 95* increases activity so that all 

 are e<|ii.illy po-i lively thermntn-; 



The therm. il distribution of the four groups of species outlined 

 abo\e mu-t (hen be conditioned through some indirect check 

 oil the life history. Some temperature condition of the water 

 ior tin nymph or the developing egg may lie the limiting factor 

 rather than the Hying temperature for the imago. Undoubtedly 

 individuals of each species continually fly beyond the limits of 

 the optimum habitat but the offspring of such do not survive or 

 we \\ould have a spreading species. This constant pressure of 

 the- species of dragon flies into surrounding but unsuitable habitats 



\\oiked out by the writer 10 on Put-in-Bay Odonata in 1922. 

 Until we know the life histories of the various species in minute 

 di tail \\e will not be able to define all of these limiting factors 



pi ,i- \\e -tumble onto them accidentally. 



A further i om lusion appears indicated from this study. As 

 all .K>hn.i pt ntvadtnsis, ha\e one type of reaction to tem- 



perature in the imago, and all the l-'.phemeras, except gnttulata, 

 ha\ e one t \ \n- < .1 reaction to light and the reactions of these t\\o 

 oild sp ne just the re\er>e of the other species in their 



respective genera, we may nun hide that a given type of in 

 ner\ou> >\>tem ran be completely reversed from its usual re- 

 action more easily than it can be modified in a lesser degree. 

 Thi> would appear logical tKnn the theoretical ground- of the 

 mechanics of the ner\ou- -\-iem. Because of it> minute >i/e. 

 the insect ner\oii- system is characterized by the relatively 



9 \Valki-i. see ref. 8, p. 33. 



10 Kcnm-ily, "The Ecoloj;: ationships of the Dragonflies of the Bass 

 Islands of Lake Km. Ecology, \'l. III., pp. 325-336. 



