34 J- H. Powers 



in coolness the Minnesota summer. Yet the metamorphosis ran 

 its course with full completeness, leaving no obvious trace of "lar- 

 val tendency," and producing adults corresponding to those which 

 Cope deemed had undergone the maximum degree of metamor- 

 phic development. Thus the key to the special problems which 

 this polymorphic species offers is not to be sought in the broad 

 differences of general climatic conditions, but in the special pe- 

 culiarities of purely local habitats. 



There remains but a single character in the head of the larva 

 of which I wish to speak at present. This is the form of the muz- 

 zle. In all ordinary larvae, larvae during their first summer, the 

 muzzle is rounded in contour. It is narrower or broader accord- 

 ing to various circumstances that modify the shape of the head, 

 the width of the gape, etc., but in all such cases it presents a 

 smooth, continuous curve. I have handled thousands of individ- 

 uals and have never yet seen an exception. The adults, too, in all 

 save the rarest instances, present a rounded muzzle. Figures i 

 and 2, plate II, show broad but otherwise typical muzzles of lar- 

 vae; so do figures 2 and 6 on plate VIII. Figures 2. 4, and 5, on 

 plate IV, show normal muzzles, although the first two are unusu-' 

 ally pointed muzzles of adults. Figure 5 is a thoroughly typical 

 wild specimen, while with these may be compared the exception- 

 ally broad, though still rounded muzzles, larval and adult, in fig- 

 ures I and 2 on plate IX. Considering hov/ universal is this char- 

 acter, it is surprising to find it readily and regularly modifiable 

 by appropriately changed conditions of the simplest nature. The 

 modification consists in the transformation of the rounded into a 

 sharply truncated muzzle (see fig. 3, plate VI). The conditions 

 which produce it are the prolongation of larval life by very low 

 nutrition,' followed by periods of at least moderately rapid growth. 

 Mere neoteny, or wintering, of ordinary larvae does not usually 

 produce the truncation, although it does in some few cases, evi- 

 dently those where the fall nutrition is also light. Nearly all of 

 my cistern specimens which have undergone periods of more or 

 less intermittent feeding develop this character by the second sea- 

 son, and all of them do so if kept under such conditions for sev- 

 eral years. Laboratory specimens, too, have usually developed 



230 



