On the Structure o/Antedon rosaceua. 181 



noticed by my son, Mr. P. H. Carpenter * (who is at present work- 

 ing in the laboratory of Professor Semper at "Wiirzburg), in two of 

 Professor Semper's Philippine species, Actinometra armata and A. 

 nigra, as also in Antedon Eschrichtii, in which it had been previ- 

 ously discovered by Ludwig. It is not nearly so distinct, however, 

 in A. rosaceus ; but its existence in that species was also indepen- 

 dently recognized by Professor Huxley, who, like Ludwig, was led 

 by his general view of the homologies of the Crinoids to regard 

 it as a neiwe. My son regards both the ventral band of Ludwig 

 and my "axial cord" as belonging to the nervous system, being 

 led to that conclusion, as regards the former, by its homology with 

 the radial nerves of other Echinoderms, and, as regards the latter, 

 by the very definite branching he has discovered in the axial cord 

 of the arms of Actinometra armata and A. nigra — two pairs of 

 branches running on each side towards the dorsal surface, and 

 two to\\'ards the ventral, where he has distinctly traced their rami- 

 fications as far as the leaflets bounding the ventral furrow. Prof. 

 Greef, on the other hand, describes the whole epithelial floor of the 

 ventral furrow as a nerve, on the ground that its histological 

 character resembles that of the nerves of other Echinoderms. 



Having recently had an opportunity of. examining at Wiirzburg 

 the very thin sections prepared by my son, I can say with certainty 

 that the fibrillar band is quite distinct from the layer of columnar 

 epithelium which it underlies ; but it appeared to me to send off 

 very min\ite fibrils that pass up between the cells of which that 

 layer is composed. 



To myself it appears by no means improbable, looking alike to 

 its position and to its histological characters, that this band is a 

 nerve; but having regard to its immediate proximity to the 

 sensory (ventral) surface, and to its separation from the muscles 

 by the interposition of the triple canal-system, I cannot but think 

 it more likely that it is functionally related rather to the former 

 than to the latter — in other words, that it is an afferent rather 

 than a motor nerve. 



As it seemed to me that important evidence might be obtained 

 on this point from experiments made on the living animal, I took 

 the opportunity afforded by my recent visit to the Zoological 

 Station at Xaples to institute such experiments; the results of 

 which I am desirous of appending to my paper, as they seem to 

 me to place the doctrine advocated in it beyond reasonable doubt. 



Every one who has had the opportunity of observing the habits 

 of the living Antedon well knows the peculiarly rhythmical and 

 Bymmetrical swimming action which it executes when it spon- 

 taneously leaves or is detached from the anchorage afforded by 

 the grasp of its dorsal cirri. Each of its five rays divaricates 

 into two arms, which may be characterized (like the two legs 

 proceeding from the human trunk) as the right and the left 

 respectively ; and the act of swimming consists in the alternate 



* " Eemarks on the Anatomy of the Arms of the Crinoids," in the Journal of 

 Anatomy and Physiology for April 1876, p. 571. 



