896 ON NESTING-HABITS OF PHYLLOMEDUSA SAUVAGII. [DeC. 14, 



is very striking, and still more so is the definite ari-angement of 

 the proportions in which they are produced in different periods of 

 oviposition. We must suppose that at the beginning of the process 

 the oviducts secrete a large number of capsules before the eggs 

 begin to pass down them, and again at the end must continue to 

 do so after the last egg has passed. The actual production of 

 empty capsules is only what may be found, though to a very 

 much smaller extent, in probably any Anuran. If a batch of 

 Rana temporaria spawn be looked over, a few empty capsules — 

 perhaps 2 or 3 per cent. — will generally be found. P. sauvagii, 

 however, has developed this peculiarity to an enormous extent 

 and also controlled it in the way we have seen. 



The nests are not always so perfect as the one figured. Often 

 gaps are left between the edges of the leaves, exposing some or 

 many of the eggs to the light. Such exposed eggs if they are 

 near the surface die. Sometimes also the jelly plugs at top and 

 bottom contain a few eggs. Such eggs, being exposed, also die. 

 It is significant that the actual percentage of eggs hatched is 

 thus greater in perfect nests than in the less perfect ones often 

 found, for we see that the advantage in productivity of the frogs 

 which make the best nests — i. e., frogs in which the oviducts 

 secrete a sufficient number of empty capsules, especially at the 

 beginning and end of oviposition, and which also make the best 

 vise of the leaves to cover the sides of the nest — must tend to 

 perfect the process. 



The question of how rounded egg-capsules are formed without 

 eggs as nuclei to form round, is one to which I have not found 

 any clue by the dissection of the oviducts, which were nearly 

 empty in both females I opened. I can only say that I found 

 empty capsules far up in the glandular portion of the oviduct, as 

 well marked off from one another as in the ovisacs. 



References. 



H. VON Jhering. — " On the Oviposition in Phyllomedasa 

 jheringii." Ann. & Mag. Nat. Hist. vol. xvii., 1886. 



J. S. BuDGETT. — " Notes on Batrachians of the Paraguayan 

 Chaco." Quart. Journ. Micro. Sci. vol. xlii., 1899. 



EXPLANATION OF PLATE LXXXIV. 



Both the figures were diawn from preserved specimens bj' Mr. A. K. Maxwell. 

 They are of natural size. 



Fig. 1. A nest of PJii/llotnednsa sauvagii. It is hanging in its natural position. 

 Some of the leaves have been turned aside to expose the egg-containing 

 portion of the mass of spawn, e.c.l., mass of emptj' egg-capsules, forming 

 the bottom of the nest, b.cm., empty capsules forming the roof of the nest. 

 ov., egg-containii:g portion of the mass of spawn. 



Fig. 2. Dissection of a female P. sauvagii preserved when it had about half filled 

 its nest, od., oviduct, ovs., ovisac, containing full and empty capsules 

 arranged as described in the text. 



[While the above was in the press, I have seen M. Siedlecki's 



