304 ARTHUR DENDY. 



separated from the integument of the arms, is very distinctly marked. 

 Each ambulacral groove is torn sharply and suddenly across ; gene- 

 rally at about the level of the base of the first (oral) pinnule. Between 

 any two adjacent arms, or pairs of arms, is a tightly stretched, trans- 

 lucent membrane, bounded by a free, incurved edge which marks the 

 line of tear interradially and radially. These membranes form an 

 important part of the floor of the visceral basin, and assist in sup- 

 porting the visceral mass. It will be seen that the line of tear forms 

 the edge of the visceral basin, This edge does not maintain an even 

 curvature all the way round, but is deeply notched. There are ten 

 of these notches, or bays, one between the two arms of every pair 

 (radial), and one between each two adjacent pairs of arms (interradial), 

 the latter being a good deal deeper than the former. 



The floor of the visceral basin is lined by an exceedingly thin, trans- 

 parent membrane, of connective tissue, through which a series of 

 muscles is very distinctly visible. These are the muscles between the 

 calcareous plates of the calyx, and they are arranged in two concentric 

 circles. The inner of these two circles is made up of five pairs of 

 muscles, connecting the first and second radial plates in pairs. The 

 outer circle is composed of ten pairs of muscles, which connect in 

 similar manner the third radials with the first brachials. The degree 

 of distinctness with which these muscles are visible, beneath the 

 overlying regenerating tissues, forms, during the earliest stages, a 

 good indication of the amount of regeneration which has taken place. 



On examining sections of a freshly eviscerated Antedon, and com- 

 paring these with sections of an uninjured specimen and of the isolated 

 visceral mass, it is fairly easy to determine the exact region in which 

 the separation of the visceral mass occurs. In sections of a freshly- 

 eviscerated Antedon the surface of the visceral basin, including the 

 muscles above described, is seen to be covered by a thin, smooth layer 

 of connective tissue, excepting in the centre, within the inner circle of 

 muscles, where this layer is incomplete. Here we have a deep pit, 

 the sides of which are formed by the first radial plates and the floor 

 by the perforated rosette. This pit is in the narrowest part a little 

 over half a millimetre broad and, in the one specimen I have mea- 

 sured, four fifths of a millimetre deep. The central plexus sticks up 

 into it through a hole in the middle of the rosette and is surrounded 

 by strands of connective tissue, which, attaching it to the walls of the 

 pit, support it. In the speoimen measured the central plexus projects 



