]ME. P. H. CAEPENTER ON THE GENUS ACTINOMETRA. 19 



deviation, and reach the mouth, after a circuitous course, parallel to the margin of the 

 disk. It is often difficult to state the number of ambulacral furrows abutting on the 

 mouth, as they frequently iinite immediately before reaching it ; in different species, and 

 in diiferent specimens of the same species, I have counted 4, 5, 6, 7, 10 stems ori- 

 ginating from the mouth ; this difference, therefore, is of no importance at all. The 

 anal tube in all these species [Actinometra, mihi] has a central or subcentral position, 

 and the anal area occupies the larger portion of the disk. Using this difference as 

 'fundamentum divisionis,' I liave never encountered a doubt whether any type should 

 be referred to Antedon or to Actinometra, although I have examined a great number of 

 specimens and species.. Moreover the lower or oral pinnules o^ Actinometra are always 

 very different from the others, being flagelliform and presenting a more or less distinct 

 seri'ature or comb (piunulae orales prehensiles) ; while in Antedon they are only slightly 

 differentiated from the others, or are transformed into strong rigid spines, forming a 

 protective covering over the disk [A.protectus, mihi]. It will, perhaps, be thoiight 

 improper to elevate these sections into genera, as the fossil Antedons would usually not 

 be generically determinable ^ ; but they are at least very good subgenera for the distri- 

 bution of the numerous species. The mode of classification here proposed is concerned 

 with the main point of that established by Miiller, but is evidently an amelioration of 

 it. In Actinometra he describes the five ambulacra as partially uniting before reaching 

 the mouth, so that their number becomes reduced to three or four [C. Solaris, PI. I. 

 figs. 2, 5, and A. WaMberghii, PL I. fig. 3], while he refers to Alecto all those specimens 

 in which five ambulacra separately reach the mouth, even though this orifice be quite 

 excentric and marginal, and the length of the ambulacra therefore exceedingly different, 

 as in Alecto inultiradiata [PI. I. fig. 4]. This mode of distinction used by Miiller is, 

 however, very unnatural, and often quite arbitrary or illusory. It is the marginal or 

 subcentral position of the mouth that is of importance ; and this character is never 

 ambiguous. C. multiradiata [PI. I. fig. 4] is not less a ivae. Actinometra than^. Solaris 

 [PI. I. figs. 2, 5] and A. WaMberghii [PI. I. fig. 3]." 



It AvQl be seen from the above note that Dr. Liitken considers Antedon and Actino- 

 metra as two subgenera of Comatula, ^w^ec^o?* having a subcentral mouth and but slightly 

 differentiated oral pinnules, while in Actinometra the mouth is excentric and the oral 

 pinnules bear a terminal comb (PL III. figs. 1-3). At the time I received Dr. Liitken's 

 note I had had no ojiportunity of examining any large collection of Comatulce ; and his 

 statement that the oral pinnules of all Actinometrce were marked by a terminal comb 

 was therefore new to me. I have since been able to examine a considerable number of 

 Comatulce, and, like Dr. Liitken, have never had the least difficvilty in determining to 

 which type any given specimen shoiild be referred, while at the same time I have always ^ 

 found that in Actinometrce, or Comatulce with an excentric mouth, the oral pinnules bear 



' It will be shown further on that there are ver)- decided differences in the shape of the calyx in the two genera 

 Antedon and Actinometra. These render the determination of fossil Comatulce (and also of recent specimens from 

 which the disk is lost) less impracticable than Dr. Liitken supposes. 



■ The above passage was written early in 1877. Since then I have examined the largo collection of Comatulce 

 brought home by the Challenger.' Out of nearly fifty species with an eccentric mouth, all but two bare a terminal 

 comb on the oral pinnules. 



3* 



