TEMNODOU SALTATOR. 13 



does not vary much from .70 to .75 mm. in diameter, but we have recorded 

 instances which show ;i remarkably wide range of variation in size, the max- 

 imum and minimum limits being represented by .84 mm. and .66 mm. The 



oil-globule usually measures from .18 to .19 nun., but may vary between 

 .17 mm. and .23 mm. 



It is rather remarkable that we have never obtained stages much younger 

 than that given in Figs. 1 and 2, although we have collected in all parts of 

 Newport Bay, and ai all times of the day. It is probable that the eggs are 

 laid at some distance outside the Bay, which would account for the absence 

 of earlier stages, as well as for the small number of eggs obtained at a time. 



The egg agrees very nearly in size with our No. 15. but the peculiar 

 segmentation of the yolk makes it quite impossible to confound the two 

 eggs. In our Preliminary Report* we were probably mistaken in regarding 

 the segmentation of the yolk as an evidence of cleavage. The segments 

 perfectly resemble cleavage-spheres, and often appear to be nucleated. But 

 sections of a large number of these egg^. in different stages of development 

 and after different methods of preservation, have failed to give any satisfac- 

 tory evidence of nuclei. What appeared to be nuclei in the living egg turn 

 out to be superficial masses of protoplasm, which are usually located beneath 

 nuclei belonging to the periblastic layer. These masses stain deeply, but 

 vary too much in size to be nuclei, and in other respects show plainly that 

 they cannot be identified with nuclear bodies. Before examining sections, it 

 seemed not improbable that these yolk-segments either represented the peri- 

 blast or played some important part in its production. But we have not 

 obtained any evidence in favor of either view. The youngest stage that has 

 been obtained was taken in June of this year (1885). In this egg the wreath 

 of periblastic cells had just appeared. These cells were more clearly and 

 perfectly outlined than in any other egg we have studied, but they lay above 

 the yolk-segments and appeared to owe their origin to the marginal cells 

 of the blastoderm, precisely as in other teleostean ova. In sections of eggs in 

 the stage represented in Fig. 1, the periblastic cells already have a syncitial 

 character and form a layer in all respects like that we have before described 

 in other eggs. The stages we have obtained give no clue to the origin of 

 the yolk-segments. They are well defined, and their outlines remain clear 

 in most of our preparations. Their nature and mode of origin can only be 

 ascertained by a study of earlier stages, and perhaps a study of the ovarian 



* Agassiz and Whitman, Proc. Am. Acad. Arts and Sci., XX. p. 31. 



