258 beport— iS84. 



It may, in fact, be described as a nerve-sheath extending practically over 

 tbe whole animal and everywhere directly continuous with the external 

 epidermis, of which, indeed, it forms the deepest and specially modified 

 layer. Such a condition of the nervous system there is independent reason 

 for regarding as a very primitive one ; and I regard it as the type from 

 which the more specialized nervous systems of the other Echinoderms 

 have been derived. This specialization consists chiefly in separation, more 

 or less complete, of the nervous system from the epidermis, in exaggera- 

 tion of the radial nerve bands with reduction of the intervening parts of 

 the nerve sheath, and finally in sinking down of the radial nerve bands 

 into and through the dermis so that they become separated from the 

 external epidermis by a layer of connective tissue which may, as in 

 Echinids and some Ophiurids, be firmly calcified. In Echiuids the nerve 

 sheath still persists as the external nervous plexus outside the test first 

 described and figured by Loven. 



I consider that in Crinoids the subepithelial bands most certainly are 

 homologous with the radial or ambulacral nerves of a starfish : and I 

 consider that they represent a part of a continuous nerve sheath which 

 has retained permanently its primitive continuity with the epidermis. 

 The axial cords, some of the branches of which can be traced into ex- 

 tremely close proximity with the subepithelial bands, I regard as portions 

 of the antambulacral nerve sheath which, like the radial cords of Echinids, 

 Ophiurids, and Holothurids, have lost their primitive position and shifted 

 into or through the dermis. 



On this view the nervous systems of all recent groups of Echinoder- 

 mata can be reduced to one plan, and furthermore, an explanation is 

 obtained of the histological similarity or identity between the axial cords 

 and subepithelial bands, as well as of the very close relation, and pro- 

 bably continuity, between the two sets of structures in Antedon. 



It must be noted, however, that while this enables us to reconcile the 

 Crinoid with the other Echinoderm types of structure, it leaves the gap 

 between the two groups an exceedingly wide one. Antedon, on this view, 

 is very far indeed from being a primitive Echinoderm : it is, indeed, as 

 regards its nervous system, the most highly differentiated of all recent 

 Echinoderms. On the other hand, the starfish has retained an extremely 

 primitive type of nervous system, which must probably be regarded as 

 ancestral for all Echinoderms. 



A further point of interest concerning Antedon, that I observed during 

 my stay at Naples, is that not only may the visceral mass be entirely 

 removed from the living animal without causing death, or indeed, any 

 apparent inconvenience, but that such specimens very speedily regenerate 

 the whole visceral mass. I have obtained a series of specimens illustrating 

 the various stages of this very remarkable and extensive regeneration, but 

 have not yet had time to examine or describe them. 



I also devoted some time to an examination of fresh specimens of 

 Amphioxus with the object of ascertaining whether the spinal nerves have 

 single or double roots of origin. By following the methods described by 

 Rohon, I have convinced myself of the accuracy of his description of 

 the existence of anterior spinal roots in addition to the well-known and 

 much more obvious posterior roots. Rohon's attempt, however, to 

 homologize the anterior nerves of Amphioxus with certain of the cranial 

 nerves of the more typical vertebrates seems to me entirely devoid of 

 justification. 



