378 OVIPOSITION AND HABITS OF CERTAIN BATRACHIANS, 



spherical, 3-5 mm. in diameter ; if in contact tbey adhere to one 

 another slightly but are readily separable with a feather, and dO' 

 not fuse into a mass. The ova themselves are twice the size of 

 ordinary frog ova. 



The pigmented pole is black, the other cloudy white. The seg- 

 mentation is complete, but on account of the considerable amount 

 of food-yolk very irregular ; the pigmented pole segments much 

 more rapidly than the other ; and in some segmenting ova which 

 I observed, after the stage in which there were two vertical and 

 one transverse furrow the next two vertical furrows instead of 

 continuing round the lower pole frequently turned to one side and 

 joined one of the original vertical furrows. As I hope to give a 

 more complete account of the development hereafter I need only 

 briefly refer to subsequent events. By about the fifth or sixth 

 days the embryo is well-folded off from the large yolk-sac ; a day 

 or two later the embryo may be noticed at tiuies to exhibit 

 wriggling movements, and about this time the two developing 

 claspers or suckers begin to show prominently. But neither up till 

 now or at any subsequent stage have I been able to see any trace 

 of external gills, and I believe these are not developed. Gills and 

 tentacular reticulations such as Heron-Royer describes* in Alytes 

 obstetricajis, if present could hardly be overlooked ; moreover 

 external gills are noticeable enough in the newly hatched tadpoles 

 of the species which oviposit in water ; hence I am unable to say 

 how respiration is provided for in the early stages unless the tail 

 functions as a respiratory organ. [After a time a single spiraculum 

 is present on the left side as usual. In keeping the tadpoles in 

 aquaria individuals may sometimes be found floating at the surface 

 of the water, the ventral surface uppermost, and every now and 

 then ejecting a number of bubbles of gas sometimes from the 

 mouth and sometimes from the anal aperture, while numbers of 

 small bubbles may sometimes be seen in the spirally coiled intes- 

 tine ; at other times the tadpoles appear to be swallowing air, and 

 often accidentally re-swallow the bubbles previously ejected. 



* Bull. Soc. Zool. de France, 1883, p. 423. 



