510 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



The upward turn of the intestine, which appears in the posterior interray 

 of Antedon shortly after the appearance of the radianal, in this species does not 

 come to lie in the posterior interradius until a very much later stage in develop- 

 ment. 



As the cup expands laterally and vertically the intestine also expands, but, 

 growing more rapidly than the calyx, soon fills the body cavity. 



The radianal, formed previous to and just below the anus, is parallel to the 

 intestine at that point, and evidently very firmly attached to it, as is shown by 

 its subsequent migration. 



The upward growth of the cup and the concomitant dextral growth of the 

 intestine carry the radianal and the arms upward and into the plane of the 

 posterior interradial, which, by reason of the retirement of the oral distalward, is 

 the path of least resistance. 



When the portion of the intestine to which the radianal is attached comes to 

 lie in a vertical position along the plane of the posterior interray the radianal 

 is completely withdrawn from the radial and rests equally upon the truncated 

 distolateral margins of the posterior radials. 



Its migration follows the upward and dextral gi'owth of the intestine, which 

 does not bend sharply upward along the posterior interray. The radianal for a 

 considerable time holds closely to the right posterior radial, thus causing inhibition 

 of its lateral growth and its peculiar asymetrical development. 



TROPIOMETRA MACRODI8CU8. 



A large series of embryos of this species, collected by Mr. Furusawa, were 

 sent by Professor Ijima to Mr. Frank Springer. 



Eggs preserved from 1 to 2 hours after fertilization, which are purplish pink 

 in color, show the cleavage stages. 



The stages between 18 and 96 hours (4 days), yellowish in color, are free- 

 swimming larvae. 



The embryos of from 5 to 48 days, fading into colorless, show the earlier fixed 

 stages. 



TBOPIOMETRA PICTA. 



At Tobago Doctor Mortensen found that the embryo of this species, which 

 has the same shape as the larva of Antedon, swims very activelj' for a period of 

 about three days, when it attaches itself; but if no suitable place of attachment 

 is found it may continue its free-SM'imming life, and in some cases it was found 

 not to attach itself until it was 8 days old. 



The eggs of Tropiometra pkfa are extruded freely into the water instead 

 of being attached to the pinnules as in the case of Antedon. 



ISOMETRA VIVIPAEA. 



Dr. Th. Mortensen writes that the larvae of this species are enveloped in 

 the egg membrane until they are ready to leave the marsupium. They have well- 

 developed ciliated bands. This leads Mortensen to believe that the free-swimming 



