160 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



inal. In one instance I was successful in making an opening at the 

 edge of the shell of an adult, by which means the animal was very 

 slightly, if at all, injured, and which enabled me to introduce a pipette 

 into the mantle cavity in order to obtain embryos and to follow the un- 

 disturbed development for a couple of days; but this method was not 

 long available, since every time embryos were detached artificially great 

 numbers would escape from the parent, so that all of the brood was 

 soon lost. It is, therefore, impossible to obtain an unbroken series of 

 the different stages of development, but it is necessary to resort to the 

 method of comparison of the observed stages, and in that way endeavor 

 to form an idea of the mode of development. It is also a fact that one 

 cannot always distinguish by external marks those adults which contain 

 brood ; the relaxation of the adductor muscle and the less energetic 

 closure of the shell consequent upon that condition is a pretty sure 

 indication that the oyster is full of embryos, but this does not remove 

 our doubts as to the age of the brood and how soon it will be set free. 

 I also obtained many more mother oysters containing old than young 

 brood, and I would state here that in consequence of this fact the first 

 stages of segmentation are in great part unknown to me, a hiatus 

 which I hope to have the opportunity to fill up next season. 



Davaine* has figured several of the first stages of the segmentation 

 of the egg of Ostrea ediilis; if these are compared with the stages ob- 

 served by me, as in Figs. 1 and 2, and with those observed by Brooks t 

 in his account of the development of Ostrea virginiana, there remains 

 little doubt that the first stages of segmentation of the egg of the oys- 

 ter take place in a manner similar to that of the eggs of other lamelli- 

 branchs. From the beginning of development onwards there is already 

 a decided difference between the lowermost (vegetative) and the upper- 

 most (animal) portion of the egg, so that after repeated cleavage the 

 lowermost pole consists of a large granular cell, from which the ento- 

 derm (and mesoderm f) develop, while at the upper pole lie numerous 

 smaller and clearer cells, which enter into the formation of the ecto- 

 derm (Fig. 1). These animal or ectoderm cells multiply by repeated 

 fission and grow down over the vegetative pole more and more, until 

 they finally close over and include it. Now the large entoderm cell or 

 sphere also begins to divide, at first into two great round cells (Fig. 2), 

 later into a number of cylindrical cells (Fig. 4) ; at the same time the 

 embryo loses its spherical form, and after an invagination of the ento- 

 derm of the lower pole it assumes a slightly rcniform shape, as seen 

 from the side in Fig. 3, in which the uppermost pole is represented 

 as directed upwards. If an older stage is now observed in longitudinal 

 section (Fig. 4), it is seen that the entoderm cells are slightly invag- 



* Recherches sur la generation des Hultres, PI. II. 



I The development of the American oyster (Ostrea virginiana List.). Studies from 

 the Biological Laboratory of the Johns Hopkins University, No. IV, 18SU, Lis. 1, 2, 

 and 3. 



