90 The Wood-frog, Rana sylvatica LeConte. 



embrace, the mortality of females is low. Only once have we recorded 

 a mated pair dead in the embrace. 



For the first few years of our investigation the wood-frog males were 

 regularly isolated or kept with females of the same species, thus pre- 

 cluding any chance of cross-embraces with other species. For the last 

 two years more opportunity was given for this phenomenon. A male 

 placed in the common frog and salamander jar seized a female Rana 

 palusiris and would not release until its hold was forcibly broken. 

 Another embraced a male Rana pipiens, but not with much vigor. 

 A wood-frog embraced a female R. clamata normally and abnormall}^ 

 In the latter instance it seized the female with venters apposed and with 

 his forearms in front of those of the female. In another instance, 

 another male of R. sylvatica embraced normally a half-grown green-frog. 

 Cross-embraces would ensue with females of other species of Rana if 

 an opportunity were afforded. Of abnormal embraces, we have one 

 pecuhar record noted March 30, 1907. A male had accidentally been 

 put in the Ainbystoma jar and soon had one of the salamanders in his 

 embrace. The salamander repeatedly swam around and around in 

 the jar, but the embrace was too tenacious for it to disengage the frog. 

 This embrace was continued for an hour or more and was just in front 

 of the hind limbs. To make doubly certain, another vigorous male 

 was placed in the jar, and in the evening of the same day (March 30) a 

 second record of the same phenomenon M^as recorded. On April 21,1913, 

 another male in the laboratory was observed embracing an Amby stoma 

 just back of the fore legs. 



THE OVULATION. 



For spawning, this species usually chooses still water, rarely the 

 backwaters or bayous of streams. It prefers the leaf-laden ponds and 

 transient pools of wooded districts, 

 though not wholly restricted to such 

 localities. Occasionally we have 

 seen the frogs migrating to swamp}'^ 

 cat-tail stretches for spawning and 

 have both observed and heard tliem 

 in such situations. Two of our best 

 collecting-spots were a grassy pool 

 and a high upland pond, both of 

 which were out in the open and 

 ordinarily dried up in midsummer. 

 In wooded districts we have found 

 them even using pools no more than 

 1| by 4 feet. Our first records 

 of wood-frog spawning have been as shown in the above table. 



