522 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [10] 



3. Zygodactyly Tentacles more numerous than the radial tubes. 



It is evident from what we know of the development of the Medusa 

 (gonophore) of Z. grcenlandica (?) that the relative number of tentacles 

 and radial tubes varies with age, and consequently the three subgenera 

 are difficult to separate on this feature alone. There are specimens of 

 Polycanna in the collection with characters of the first subgenus Bha- 

 costoma, to which I have already given the name P. americana. It is 

 believed that we have at least two species of Polycanna on our New 

 England coast, and provisionally these may be known as P. grcenlandica 

 and P. americana. The basis of the separation of the two is the exist- 

 ence in the former of rows of subumbral knobs between the chymiferous 

 tubes and the absence of these knobs in the latter. It happens that in 

 the latter the number of tentacles is less than the number of chymif- 

 erous tubes, while in the former, according to A. Agassiz, the number 

 of tentacles is greater than that of the radial tubes.* 



It seems to me that the presence or absence of the subumbral knobs 

 is a much safer character to rely upon in the separation of our species 

 of Polycanna than any which has yet been suggested. If new inves- 

 tigation shall show that true specimens of grcenlandica do not have 

 subumbral knobs, our New England species is possibly new. From the 

 fact that a supposed type specimen of Polycanna, labeled Z. grcenlandica, 

 in the collection of the Museum of Comparative Zoology, has these tu- 

 bercles, the name grcenlandica is retained for this species. 



There is another Zygodactyla-like Medusa in which I have not been 

 able to find these gelatinous knobs, either in a live animal or in alcoholic 

 representatives. As this species also differs from the species crassa and 

 cyanea in the relative number of tentacles and chymiferous tubes, it is 

 supposed to be the new species, americana. 



Unlike all other American Zygodactyly, as described by A. Agassiz, 

 this species has a smaller number of tentacles than of radial tubes, and 

 at the same time none of the alcoholic specimens have subumbral tuber- 

 cles. It is possible that the former feature indicates an immature 

 Medusa, but not so the latter ; for, as has been already shown, the sub- 

 umbral tubercles are present in the Medusa when very small. 



Specimens referred to P. americana were collected in the following 

 localities : 



Catalogue 

 number. 



ncr.o 



11665 

 11673 

 11674 

 11677 

 ! 11649 



Station. 



2563 



2567 



2566 



2563 >. 



2566 



2539 



* The existence of radial subumbral knobs and a larger number of tentacles than 

 radial tubes is supposed to characterize grcenlandica, although the knobs are not 

 mentioned in A. Agassiz's description. 



