26o NEW YORK STATE MUSEtJM 



and the fact that males and females have been taken from beneath 

 the same log (C. & H. Thompson '12, p. 36) or stone (Moesel '18) 

 at this time of the year points to the same conclusion. 



The four-toed salamander has been taken in the spring as early 

 as April 7th (Wright and Allen '13) and this date probably approxi- 

 mately marks the beginning of general release from hibernation 

 (in the latitude of central New York) and the preparation for mating 

 and migration of the females at least, to situations favorable to 

 development of the eggs. All adult specimens taken in the bogs 

 were females, the sex being determined by dissection. As half-grown 

 specimens have also been found in bogs, it is probable that they 

 too perform the seasonal migration. 



Developing eggs were kept in sphagnum moss in jars and cans, 

 some of which in turn were surrounded by moss. With three groups 

 of the eggs a female was placed, the others were simply covered by 

 moss. Apparently the attendance of the female made no difference 

 with the development of the eggs as long as the moisture content 

 of the moss approximated conditions in the field. Eggs kept in 

 moss saturated with water became mouldy and many were lost. 



Drawings of the earlier development stages were made from 

 specimens placed in 60 to 70 per cent alcohol on successive days; 

 later stages from live embryos selected at longer intervals. Only 

 those stages showing some particular feature of development are 

 considered in this account, although a constant check was kept on 

 all stages. In all stages the envelops of the eggs are represented 

 more or less diagrammatically as the thickness of these envelops 

 varied constantly with their water content. 



Eggs 



Unlike the eggs of Desmognathus fuscus, Pletho- 

 doT^ cinereus and some others, those of Hemidactylium are 

 not stalked but cling together and to the surrounding moss by the 

 adhesiveness of their gelatinous coverings (plate 4, figures i and 2). 



In size they vary from 2.5 to 3 mm in diameter and are inclosed 

 by two, at least, gelatinous envelops which give them a total diameter 

 of 4.5-5 mm depending to a considerable extent on the amount of 

 moisture present and consequent absorption. The outer of the 

 two distinct envelops is comparatively thin, tough and elastic; the 

 inner, more fluid in its consistency like clear jelly. In some stages 

 of development it was possible to distinguish a thin vitelline mem- 



