MUSEUM OF COMPARATIVE ZOOLOGY. 79 



and then with acetic acid, the radiate structure remains easily distin- 

 guishable in both membranes. 



With regard to the formation of the two membranes, Buchholz argues 

 that the outer is the older, since one often finds, in the earlier stages of 

 their formation, that the inner is thinner than the outer, whereas subse- 

 quently they are both of equal thickness. The increase in the thickness 

 of the inner membrane was observed between the middle of February 

 and the middle of April, — the spawning time, — and meanwhile the 

 outer membrane was found to be thinner over about one third of the 

 surface. It is to be assumed, according to the author, that this attenu- 

 ated portion finally disappears altogether, since the persistent portion 

 which remains attached in the region of the micropyle is too small to 

 have completely enveloped the egg. Even nearly up to the time of 

 maturity there is no fusion between the two membranes, which must, 

 therefore, take place rather late. 



The homology of the outer membrane in Osmerus is not at once 

 evident from this account by Buchholz. If one were to accept un- 

 questioned his description, it would be most natural to regard it simply 

 as a detached portion of the zona radiata, for he maintains that the two 

 are identical in structure. There are, however, two other possibilities ; 

 it may be homologous either with the villous layer of Lepidosteus, or with 

 the capsular envelope of the perch. I regret that I have not yet found 

 the opportunity to acquire from personal study additional evidence in 

 favor of one or the other of these explanations ; but there are two or 

 three things connected with the account given which incline me to 

 believe that the outer membrane is the equivalent of the villous layer. 

 The very fact of its becoming detached from the deeper layer and 

 thrown into folds after the egg has lain in the water suggests a similar 

 though less striking feature of the villous layer in Lepidosteus ; and al- 

 though there is no evidence that any such eversion of the membrane 

 takes place in the latter case, or that it eve:: becomes regularly attenu- 

 ated on one side, as in the case of Osmerus, still I can imagine that a 

 similar condition might be artificially produced in Lepidosteus, so far at 

 least as regards the peeling off and eversion of the outer covering, and 

 it is possible that a slightly different physical condition of the villi 

 would cause them to adhere to each other so persistently as to allow the 

 attenuation of the whole membrane on one side of the egg without 

 separating the individual elements. Since Buchholz asserts that the 

 pore-canals are more readily distinguished as such in this fish than in 

 other instances, I infer that the markings which he observed must have 



