342 DR. R. HORST. 



simply ceasing. So Lacaze-Duthiers tells us that the oyster 

 larvce remained alive in his aquaria for longer than a month, 

 but all the time he saw no important changes take place in 

 their organisation, which is decidedly not normal. Though 

 I succeeded a few times in following the development of the 

 spat for some days undisturbedly by making a little hole in 

 the margin of the shell, by means of which the oyster was 

 not or hardly at all hurt, and in which way I could never- 

 theless reach the eggs with a glass tube ; yet even this was 

 only for a short time, because the oyster cast out at 

 every artificial delivery a great quantity of embryos, and 

 consequently soon lost all the spat. It is, therefore, not 

 possible to obtain an unbroken series of the different stages 

 of the larvae, but we must take refuge in comparing the 

 observed stages together, and try to form in that way an idea 

 of the development. 



Another difficulty is that the breeding oysters cannot 

 always be recognised externally ; the relaxation of the ad- 

 ductor muscle, and accordingly the less energetic closing of 

 the shell, is to be sure a pretty certain sign that the oyster 

 contains embryos, but this phenomenon happens most ob- 

 viously, the older the spat and the closer to the period of its 

 getting free. This may be the reason that I generally found 

 more oysters with old spat than with young, and, therefore, 

 the first stages of segmentation of the egg mostly remained 

 unknown to me. 



Davaine^ has figured some of the earliest stages of segmen- 

 tation of the egg of Ostrea edulis. These, combined with older 

 stages observed by me (figs. 1 and 2), and with Brooks'^ 

 observations on the development of Ostrea virginiana, leave 

 no doubt in my mind that the segmentation of the oyster egg 

 in the beginning is quite similar to that of the egg of other 

 Lamellibranchiata. The first stages I observed were pro- 

 vided with a large granular sphere at the lower (vegetative) 

 pole of the egg, while at the opposite (formative) pole lay 

 some smaller, clearer cells ; a segmentation cavity seems not 

 to exist (fig. 1). The smaller (epiblastic) cells are increas- 

 ing rapidly in number and embrace the great hypoblastic 

 sphere, though without closing it wholly in. The hypo- 

 blastic sphere begins to divide itself first into two great round 

 cells (fig. S) , afterwards into several cylindrical cells (fig. 4) ; 

 meanwhile the embryo has lost its spherical form, and has 



' ' Kecherches sur la Generation des Huitres,' pi. ii. 



2 "The Development of the American Oyster {Ostrea virginiana. List.)," 

 • Studies from the Biological Laboratory of John Hopkins University,' No, 

 iv, 1880, pi. 1, 2 and 3. 



