1 1 



and C. americanus (Mc Murrich, 1890). Within recent years, however, a number of forms that 

 show a distinctly different arrangement of these mesenteries have been described, such as 

 Pachycerianthus bcnedeni (Roule, 1904), P. oligopodus (Cerfontaine, 1909) and P. cestuari, P. 

 johnsoni and Botryanthus benedeni (Torrey and Kleeberger, 1 909) v ). Since the publication of 

 von Heider's paper (1879) it has been recognized that the most ventral couple of mesenteries 

 of C. memèranaceus is characterized by being short, sterile and destitute of mesenterial filaments, 

 and that the second couple extends to the aboral pole, is fertile and is provided with mesenterial 

 filaments. The same conditions obtain for the corresponding mesenteries of the other three 

 species of Cerianthus mentioned above, and to the most ventral couple the term directives 

 has been applied, while the second couple of mesenteries have been termed the continuous 

 mesenteries, a term which may be conveniently replaced by telocnemes. The third couple resemble 

 the directives in being sterile, except, apparently, in C. verrillii, but they are provided with 

 mesenterial filaments, much coiled in a portion of their course (data with regard to this point 

 being lacking for C. verrillii); their length varies in the different species, but they are always 

 considerably shorter than the mesenteries on either side of them. The fourth couple resembles 

 the telocnemes in being fertile and provided with mesenterial filaments, which, however, do not 

 show the extensive coiling presented by those of the third couple; considerable variation in 

 length is shown by these mesenteries in different species, since in C. lloydii and C. verrillii 

 they are almost as long as the telocnemes, while in C. monbranaceus and C. americanus they 

 are much shorter, in the former species, however, being longer than any of the cleuterocnemes, 

 while in the latter they are surpassed by a number of them. 



In these four species, therefore, there is a general uniformity in the character of the 

 protocnemes, which may be expressed in a few words by saying that the telocnemes are the 

 second couple. In 1904, however, Roule described an arrangement of the mesenteries in a form 

 which he named Pachycerianthus benedeni, which differs markedly from that just described in 

 that it is the fourth couple instead of the second which become the telocnemes. He finds that 

 the mesenteries are nearly all rather short, extending but a short distance beyond the lower 

 edge of the stomatodaeum, but two of them, which, using van Beneden's nomenclature (1898), 

 he designates S 3 reach the aboral pole, enclosing the gouttière impaire, which contains vestiges 

 of the mesenteries D and S 1 , these being the first and second protocnemic couples. No mention 

 is made in his paper of the third couple of protocnemes, but it is evident that they too must 

 lie in the gouttière impaire. Such an arrangement is exceedingly interesting and important, and 

 it was of no little interest to fincl in the "Siboga" collection two species which also showed it. 

 One of these seems to be identical with the form described by Kwietniewski (1898) as C. 



fact that Danielssen first employed the Dame in 1860 and that it was given as a synonym of C. lloydii by Sars in 1S61. Danielssen's 

 application of the term certainly has the priority and it is Verrill's use of it that must be given up. It is interesting to note that the 

 term C. danielsseni which Kingsley proposed for C. borealis Dan., was also suggested for that form by Levinsen in 1893, and that the 

 same name has also been bestowed by Roule upon a species described by him in 1905. 



1) These last three species the authors refer to the genus Cerianthus. The arrangement of the protocnemes shows that two of 

 them must be referred to the same genus as the Pachycerianthus benedeni of Roule (1904). As to the third if we accept the validity of 

 van Beneden's division of the Cerianthida; into the two groups Acontiferae and Botrucnidifera;, it must be referred to the latter group 

 and therefore must be separated from Cerianthus. I suggest for it the name Botryanthus, unless it be shown in the future to belong to 

 one of the genera van Beneden has established for Botrucnidiferous larvae. 



