MUSEUM OF COMPARATIVE ZOOLOGY. 275 



PLATE IV. 



In none of the figures are the vitelline cells and the nuclei of the thin epiblastic 

 la5'er of the primitive hydrophyllium brought out with sufficient distinctness. There 

 should be two layers instead of one at pyt. in Figs. 1, 2, and 3. 



In Figs. 2 and 4 the wall of c.p. I. is too black, and does not show the thickness 

 Df the hypoblast. The rows of neraatocysts on the surface of hyph.. Fig. 8, 

 ire not well shown. The clusters of nematocysts at the margin of liyph. after the 

 bifurcation of th. are faulty. The cilia on the surface of the ovum, well seen in 

 Fig. 2 at pyt. in my drawing, are not found in the photographic reproduction. 



Fig. 1. A larva (primitive larva) in about the same age as the last and in a like 

 position, except that the primitive covering-scale or hydrophyllium is 

 turned a little more to the plane of the observer. 



" 2. The same, looking through the primitive covering-scale upon the apex of 

 the float, which lies in the geometric centre of the figure. Three days 

 old. 



" 3. Older larva, free-hand drawing, showing growth of serrated scale {scr. 

 liyi)h.). 



" 4. The same, lateral view, 



" 5-7. Successive stages, in reduction in size of the primitive hydrophyllium, 

 following its great development. 



" 8. A larval Agalma, one week old, bearing a large flat hydrophyllium (rem- 

 nant of the primitive covering- scale) through which runs a tube {th.) 

 which bifurcates and ends at the distal edge in clusters of nema- 

 tocysts. 



