492 PROCEEDINGS OF THE NATION jiX MUSEUM. vol. xxix. 



The difficult (juestion, whether, in the egg of the pipehsh, cells are 

 budded oil' from the central periblast and added to the blastomeres, 

 can not here be taken up. However, this would seem to be a legiti- 

 mate consequence of such a mode of cell formation as that shown in 

 Type 3 above, and apparently finds conlirmation in hgs, 75, 77, 79, 

 and 82, in which a perfectly definite periblast laj^er has been formed. 

 If these figures are compared with His's (1898) tigs. 10 and 12, this 

 matter will be made clearer. 



For a fuller discussion of the origin of the periblast and its nuclei, 

 and of the fate of the latter, the reader is referred to Brook (1887), 

 Kowalewski (1886), Hoffmann (1888), Fusari (1890), Berent (1896), 

 Zeigler (1887 and 1896), His (1898), and Hertwig (1903). 



At this point, the work on the development of the pipefish will have 

 to rest. It has been the intention of the writer to carry it further, at 

 least to the closure of the blastopore, and for this purpose the sections 

 have been cut, but the difficulties met with have caused so many 

 delays that it has been impossible to complete it. 



The Q,gg of the pipefish is very different from most other teleostean 

 eggs in the form of its segmentation and the dual origin of its peri- 

 blast, together with the "after-segmentation" of cells therefrom. So 

 marked are these differences that it seems proper to say that the 

 figures in this paper are representative of the sections of a thousand 

 or more eggs, obtained from thirt^^-three fishes during three summers. 



The slides containing the sections from which these figures were 

 drawn have been presented to the U. S. National Museum. 



