Experimental Studies en Germinal Localization. 199 



obvious gaps that they show In some directions, I therefore pub- 

 lish the results as they stand, with the hope that they may be 

 extended hereafter/ 



METHODS. 



The eggs ,of Patella cocrulea were obtained in a mature state from March 

 until June, those of Dentalium entalis during June and July. Artificial fertiliza- 

 tion is easily effected in Dentalium, but is much more difficult in Patella. In 

 the latter case I found, after many trials, that the eggs fertilized more readily 

 if first placed for half an hour in sea-water rendered slightly alkaline by the 

 addition of 4-6 drops of a 5% solution of potassium or sodium hydrate to 

 half a litre of sea-water (the slight precipitate first formed quickly dissolves 

 upon agitation). The spermatozoa were also placed in the alkalized water 

 for the same length of tiime. From 15 to 20 minutes after fertilization (in the 

 same water) the eggs were as a rule transferred to a large quantity of pure sea- 

 water brought from the open sea. 



In both forms the opaque tgg is at first surrounded by a very distinct 

 membrane, whrich, in the case of the ripe eggs, disappears as the eggs lie in 

 water before fertilization, in Patella by gradually dissolving away and disinte- 

 grating at several points, in Dentalium by suddenly bursting and being thrown 

 ofif. Double fertilization occurs rarely in Dentalium, but very frequently in 

 Patella, so that ^in the latter case it is essential to pick out the normal eggs one 

 by one with a pipette at the 2-cell stage. In both forms the blastomeres can 

 be separated with the greatest ease by means of Herbst's calcium-free sea- 

 water — indeed, the action is so energetic that better results are obtained if 

 it is restrained somewhat by mixing the artificial water with a certain amount 

 of normal sea-water. The eggs were placed in the artificial water shortly 

 after both polar bodies had formed, and after division the blastomeres were 

 carefully separated under the lens with a fine scalpel and immediately isolated 

 in normal sea-water. Even so, however, the blastomeres often continue to 

 separate in the normal water, and the best results for the earlier stages were 

 obtained by not employing the artificial water, but by separating the cells 

 with the scalpel in normal water. This is difficult in Patella, but very easy in 

 Dentalium. in the earlier stages. For somewhat later stages the artificial water 

 must be used; but this can often be successfully accomplished by transferring 

 the 2-cell stages to normal water and separating the blastomeres at the proper 

 stage. The tendency to separate after transference to normal water steadily 

 decreases as the development proceeds; hence good results for still later stages 

 are obtained by allowing the eggs to segment in the artificial water up to the 

 16-32-64-cell stages, before isolation and transfer to normal water. For greater 

 certainty of identification the best plan is to separate and isolate the blasto- 



iLike the preceding work, this was done at the Naples Zoological Station 

 between February and the end of July, 1903, on a grant from the Carnegie In 

 stitution of Washington. I would again express my great indebtedness to 

 the administration of the Station for the unremitting care and efficiency with 

 which my work was aided. 



