PENNATUI/IDA. 



the genus Benthoptilum 1 } of Verrill may probably also be included; further, I think that the two 

 families Kophobelemnonidce and Umbellulidce are to be kept unaltered. Umbellula I regard contrary 

 to Delage and Herouard as a particularly specialized form, related to Kophobelemnon; an Umbellula 

 might be thought to be derived from a KophobelemnonAfas. form in which the polyps in the greater 

 part of the rhachis were checked so as to be kept in the zooid-stage; the zooids in Umbellula are 

 somewhat more developed than ordinary zooids, in so far as they are generally provided with one 

 tentacle; and a certain resemblance to Kophobelemnon is undeniably seen in such Umbellula-sipzcizs 

 as U. giintheri K. in which the polyp-bearing part of the rhachis is lengthened in a club-shaped 

 manner. The radial structure seen in certain Umbellula-s\)&c\zs (for instance U. encrinus), is, at all 

 events, a secondary development, having plainly arisen from an original bilateral formation, as is 

 sufficiently shown by the younger stages. To Umbellulidce, the Chunnellidce Kkth. are no doubt 

 related; in these latter also the radial arrangement of the polyps is plainly a secondary one; the 

 bilateral formation may unmistakably be traced in the arrangement of the zooids, but may also be 

 seen in that of the polyps. I think it superfluous to form a separate main group for this family. 

 According to this, the section Spicatce will have to be divided into two subsections: 



A. The polyps with calyx. 



Fain. FuniculidcB (Funiculina Lam.). 



Protoptilidce (Protoptilum, Disticlioptilum}. 



B. The polyps without calyx. 



Fam. Scleroptilidce (Scleroptihtm}. 



Anthoptilidce (Anthoptiluin, Benthoptilum}. 



Kophobelemnonidce (Kophobelemnon, Sclerobelemnon, Balhyptiluni]. 



Umbellulidce (Umbellula}. 



Chunellidce (Chunella, Amphianthus}. 



The sections Renillece and Veretillece must probably still remain. Even if Renilla in its 

 development, shows a distinct connection with a common primitive Pennatulid, it is so peculiar 

 when fully grown, that it does not agree with other groups. This holds good to a smaller degree 

 with regard to Veretillece ; the radial arrangement of polyps and zooids in these forms, however, is a 

 peculiarity justifying a separation, at least provisionally. I think it probable that this radial arrange- 

 ment will meantime prove to be of secondary origin, since a bilateral formation may be traced in the 

 inner structure of the stem in Veretilhim (comp. Kolliker Monogr. p. 430). It is rather remarkable 

 that no young stage of this group is as yet known, although it is represented, in the Mediterranean 

 for instance, by two rather common genera: Veretillum and Stylobelemnon. Both these groups: Renillece 

 and Veretillece, I must regard as very specialized and anything but primitive Pennatulids. 



With regard, finally, to the group Gondulece (Acaulina Del. & Her.), it must be quite omitted; 

 it has been established only on the genus Gondul, which is nothing but a mistaken fragment of 

 Pav onaria finmarchica (see under this species). 



i) B. sertum Verrill, from the depths of the Atlantic to the east of North America (Cape Hatteras). Rep. Comm. Fish 

 etc. 1883, p. 510 [8], PL II, fig. 4, and Am. Journ. Sc. (3), Vol. 29, 1885, p. 149, note. 



The Ingolf-Rxpedition. V. i. 2 



