141 



Gouldii Pourtalés) begun at the Laboratory of the U. S. Fish Com- 

 mission in 1888 yielded the following results. 



1) Individuals of both sexes full of ova or of spermatozoa were 

 about equally abundant from July 20 to September 20 and could some- 

 times be distinguished as male or female by the colour, which was 

 often more white in the male and pink in the female owing to the 

 nature of the contents of the coelom. 



2) Attempts made during that entire perrid to fertilize the eggs 

 artificially were unsuccesful though the various methods employed 

 by Keferstein were used and some ova segmented a few times as in 

 the early stages of those studied by that observer, but then passed 

 through abnormal changes. The difficulty in fertilizing eggs, appa- 

 rently ripe , taken from the coelom lies, I judge, in the want of some 

 necessary concomitant connected with their stay in the nepb ridia, in 

 which none were found at this season of the year. 



3) Careful dissection reveals two conspicuous, irregularly fim- 

 briated bands running outward from the nerve cord along the posterior 

 face of the ventral retractors very near their attachment to the body- 

 wall: each of these bands passes around onto the outer edge of the 

 muscle to which it is attached and there abruptly ends. Each is made 

 up of a common base of attachment to the muscle from which innu- 

 merable, irregular finger-like processes stand out freely into the coelom. 

 In adult specimens these bands project as much as 250 \i from the re- 

 tractor : in young specimens 70 mm long only 70 [j,, while in a very 

 young one only 13 mm long no reproductive organs could Ite found 

 at all. 



4 .Serial sections show that these two bands are continuous with 

 one another ventral to the nerve cord and are there attached to the lon- 

 gitudinal muscles of the body-wall. This union of the lateral halves of 

 the reproductive organ is indicated in the figures of Koren and 

 Danielssen, butin some cases as if it were dorsal to the nerve cord. 



5) The single reproductive organ thus found is made up of a solid 

 mass of germ cells supported by a structureless lamella projecting ho- 

 rizontally from between the retractor muscle fibres and the enveloping 

 peritonaeal membrane and is invested for the most part by a delicate 

 nucleated membrane. Branches of the supporting lamella extend into 

 the chief lobes of the organ and it is accompanied by elongated nuclei 

 similar to those of the peritonaeal membrane and to those of the mem- 

 brane investing the organ and measuring about 6 jx in length. The 

 germ nuclei have quite difi'erent staining properties from those of the 

 above nuclei and increase in size toward the distal or free ends of the 

 lobes of the organ where they are surrounded by protoplasm and this 



