MONOGRAPH OF THE EXISTING CBINOIDS. 375 



It may be that the spawning season of Compsometra loveni, of southeastern 

 Australia, is circumscribed in the same way. 



But for the remaining species, other than Antedon mediterranea, we have no 

 data upon which to base deductions of any kind; judging from the known condi- 

 tions in Antedon mediterranea and in Leptometra phalangium we might suggest 

 that probably comatulids spawn individually throughout the year, breeding ex- 

 amples being rather more common at certain times, dependent upon local causes 

 as well as upon seasonal changes, than at others. 



COPULATION. 



Jickeli described what he believed to be a case of copulation in Antedon medi- 

 terranea, but no other investigator has confirmed his observations, which appear 

 to have been based upon close association of the animals as a result of lack of 

 oxygen or some other unfavorable condition. 



More recently both K. A. Andersson and Th. Mortensen have assumed the 

 occurrence of copulation in Isometra vivipara to account for the otherwise ap- 

 parently inexplicable transfer of the sperm from the male to the receptaculum 

 seminis of the female. 



PASSAGE OF THE SPERMATOZOA INTO THE RECEPTACULUM SEMINIS IN ISOMETRA VIVIPARA. 



In Isometra vivipara the spermatozoa apparently first enter the marsupium, 

 thence passing through the very thin wall between the marsupium and the recep- 

 taculum seminis, for in several cases Andersson found spermatozoa not only in the 

 dorsal portion of the marsupium, but also in small interstices in the connective tissue 

 of the wall itself, evidently on their way to the receptaculum seminis. 



It is easy to see, therefore, why the receptaculum seminis extends to the outer 

 border of the ovary and is separated from the marsupium by only a thin layer of 

 cells composed of the ovarial wall and the adjacent wall of the marsupium, here 

 completely fused into a thin partition. 



DEVELOPMENT OF THE EGOS. 



So far as we know it is normally only within the pinnules of the crinoids that 

 the sexual products come to maturity, though our information is not as yet suffi- 

 cient to enable us to speak with absolute certainty. 



The fertile intrapinnular portions of the genital cords vary considerably in 

 shape; in such types as Poecilometra, Strotometra. and the genera of the Penta- 

 metrocrinidse they are short, thick, and rounded; they sometimes terminate in 

 rounded ends, and are sometimes continued as slender cords through two or three 

 pinnulars. In Heliometra and other genera they are long, fusiform, and relatively 

 inconspicuous. Both the short rounded type and the long fusiform type occur in 

 the genus Antedon. 



Professor Semper found that the ovaries of Comanthus parvicirra are not 



attached to the genital cord by their ends in the usual way. A backward process 



is given off at the point where the short branch of the sterile genital cord expands 



into the fertile portion, lying within the ventral perisome of the arm on the 



142140 21 Bull. 82 26 



