NO. 1706. CCELENTERATE8 FROM LABRADOR, ETC.—BIGELOW. 309 



At a sliglitly later stage the adradials in seven octants have attained 

 their definitive tentacular form, though they are still smaller tlian 

 the interradials. In the eighth octant, however, one of the radials 

 is still a mere knob. In seven octants a fourth series of marginal 

 structures, subradials, ranging from minute knobs to fully formed, 

 though small, tentacles, has Hkewise appeared. In the eighth octant, 

 however, no subradials are yet present. 



From tliis stage onward new members of the tentacular series are 

 formed in irregular succession. In an individual 5.5 mm. in diameter 

 by 6 mm. high there ^ are, besides the radials, tliirty-six tentacles, 

 respectively 5, 4, 4, 4, 4, 6, 4, 5, to the octant, and these show all stages 

 in development. In a slightly larger individual (6.5 mm. in diameter 

 by 6 mm. high) the six subradials in one octant have all attained 

 tentacular form, although in other octants both knobs and cirri are 

 present. 



In the most advanced specimen (15 mm. high by 12 mm. in diam- 

 eter) there is a total of 129 tentacular structures, of which 72 have 

 attained full tentacular form, the remainder showing early stages in 

 growth. In one octant of this specimen (pi. 31, fig. 1) there are, 

 between the two radial canals, nine large tentacles (T^), four small 

 tentacles (T^), cirri (ci), and knobs. 



This series shows that there is no morphologic distinction in this 

 species, between the various marginal structures, knobs and cirri 

 being merely early stages, partly perhaps contraction phases, in the 

 growth of tentacles. But, inasmuch as even in mature specimens 

 many such early stages are present, and since additional members 

 of the tentacular series are formed continuously, I doubt whether a 

 stage is ever reached in which only fully developed tentacles are 

 present. However, although we can draw no sound distinction 

 between knobs and cirri on the one hand and tentacles on the other, 

 I agree with Maas that the distinction between the two genera 

 Melicertum and Melicertidium is invahd, though on a different ground, 

 namely, that the knobs and cirri in Melicertidium and in Melicertum 

 jyrohoscifera (Maas '97) are probably notliing more than early stages 

 in the growth of tentacles, just- as they are in M. campanula. 



M. campanula is an abundant species in American waters from 

 Labrador to Cape Cod, and it is known to occur as far south as 

 Woods Hole, Massachusetts. A closely allied form, M. octocostata 

 Sars, is known along the European coast from Norway to England. 

 The latter was made by Haeckel ('79), the type of his genus Melicer- 

 tidium, but from the brief account of Browne ('95) it is evident that 

 the ''knobs" are notliing more than young tentacles. It is not 

 improbable that M. octocostata. may finally prove identical with M. 

 campanula, but until it is better known it is wisest to retain both 

 species. 



