1007.] NATURAL SCIENCES OF PHILADKLI'IIIA. 355 



"turrot'' colls. By the rapid growth of the anterior portion of the 

 blastoderm these organ bases arc forced far apart and posteriorly vintLl 

 they come to lie in the posterior margin of the blastoderm, antl by a 

 continuation of this movement they are carried around to the ventral 

 side of the eml^ryo, where the two halves of these organs approach 

 each other and finally unite in front of the mouth. 



7. The paired buccal ganglia arise lateral to the cerebral and prob- 

 al)ly from derivatives of 2a and 2c; they extend over the yolk in the 

 same manner as the cerebrals and finally concresce on the ventral side 

 of the embryo posterior to the mouth. 



8. All other organs (shell gland, pedal, pleural, parietal and abdomi- 

 nal ganglia, foot, intestine, kidney, gill, heart, etc.) arise from the 

 median posterior portion of the blastoderm, and chiefly, if not entirely, 

 from the two "somatoblasts," 2d and 4d. The bases of all of these 

 organs, except the last three, appear when the blastoderm is still a small 

 plate, and in the case of paired organs they are from the first connected 

 across the median plane and do not, therefore, undergo concrescence. 

 The origin of these post-oral organs from the two somatoblasts is 

 strikingly like the condition found in annelids, and the elongation of 

 the bod}' of the gasteropod through the shell gland is similar to the 

 elongation of the body of the annelid. 



9. All homologous organs arise from corresponding cleavage cells 

 in Fulgur, Crepidula, and probably all other gasteropods. Great 

 increase in yolk does not modify the type of germinal localization, 

 though it does profoundly modify gastridation and later stages. 



10. This indicates that germinal localization in the egg and in the 

 cleavage stages is more fundamental and primitive than are the later 

 processes of ontogeny. 



CYTOLOGICAL. 



11. During quartet formation the nuclei of the macromeres divide 

 at relati^•ely short intervals and they remain relatively small, the ratio 

 of maximum nuclear diameter to cell diameter ("Kernplasma-relation" 

 K/p) being 1 : 24. In later stages the resting period of the nucleus is 

 very long and the ratio rises to 1 : 12. The size of the nucleus is 

 therefore dependent not only upon the size of the cell, but also upon 

 the length of the resting period. There is in this species no fixed 

 ratio of nuclear size to cell size, and the cause of cell division cannot be 

 found in the maintenance of a constant ratio. 



12. After the formation of the fourth quartet the macromeres do 

 not again divide, although the nuclei do. In the first and second 

 cleavages the macromeres divide equally; this apparent anomaly is 



