72 MR. p. H. CAEPEXTEE OX THE GENUS ACTINOMETRA. 



niclit verschliessen, das die Skeletzone, aus welchpr die Centrodorsalplattc horvorgcht, 

 unabh;in<;i<? von den cigentlicheu Stielgliederu, niehr in Auscliluss an die Basalia und 

 wohl als rudiuiontiu-e Wiederholuugen derselbcn sicli entwickelt. Besonders lelirreich 

 fiir diese Auifassung sind die gar nicht seltenen stiellosen Missbildungcn der Comatula- 

 larven welche ich beobachtet habe. An solchen finden sicb in der hinteren Korpcrhalfte, 

 welebc ihre urspriinglichen Dimensionen behiilt, statt der Stielgliedcr grosse netzformige 

 Flatten welche den Raum zwischen den Basalia und dem Endknopf ausfi'illen (pi. xxviii. 

 fig. 50). Vcrgleicht man sie mit den viel scliwuclieren Anlagcn der Centrodorsalplattc, 

 so spricht die Darstellung sehr an, dass sic durch die Stiell)ildung in ihrer Entwickelung 

 gebemmt und im umgekehrten Falle gefordert werden." 



The only normal figure given by Gotte in support of his views represents a ciliated 

 larva, very much younger than the pentacrinoid stage, and with only eight stem-seg- 

 ments, over parts of the four uppermost of which are traces of a calcareous network 

 connected with the lower end of one of the embryonic basal plates. This network, which 

 reaches a more extensive development in the malformation represented in Gottc's other 

 figure, does not appear in any one of Wyville Thomson's figures of Comatula larvaj, 

 either in the free-swimming or in the pentacrinoid condition. As his observations 

 were carried on for four years, in each of which he followed out the development of 

 several broods of embryos, it is impossible to suppose that lie can have overlooked it had 

 it been present in the larvte of the British variety investigated by him. It is possible 

 that the early-formed irregular calcareous ring, " considerably wider and broader than the 

 ordinary rings of the stem, which lies immediately beneath the basal plates, and subse- 

 quently develops into the permanent centrodorsal plate," may represent the network 

 figured and described by Gotte. But then, as the latter says, Thomson makes no mention 

 of its extending downwards around the other stem-segments ; he gives, however, a series 

 of figures which, taken in connexion with those of the later stage given by Dr. Carpenter, 

 demonstrate conclusively that the above-mentioned ring does develope into the per- 

 manent centrodorsal piece. Gotte gives no figures whatever of the pentacrinoid stage. 

 If, as I believe to be the case, the network described by him as the rudiment of the 

 centrodorsal piece really does represent the primitive centrodorsal ring of "Wy^-ille 

 Thomson, commencing, be it remembered, as a network of small curved hollow spicules, 

 then his observations arc in complete accordance Avith the views of "Wyville Thomson and 

 Dr. Carpenter. Gotte oilers no explanation of its dowuAvard extension over the remain- 

 ing stem-segments as described by him in the Mediterranean variety ; and nothing of the 

 kind is described by the two above-mentioned observers as occurring in the British 

 variety, unless, indeed, it be the deposit of calcareous material upon the outer cylintb-ical 

 surface of each stem-segment by which its diameter is increased. 



It is possible that this deposit might commence to be formed at an earlier period in the 

 Mediterranean variety than in the British one ; but it is difficult to understand its down- 

 ward extension from the rudiment of the centrodorsal plate as described by Gotte. 



(§ 50) The condition of the centrodorsal piece in Aiil. rosacea and in Actinometra 

 gives us, I believe, the means of understanding a problematical Cretaceous fossil, first 



