44 Dr. F. A. Bather on Protoechinus, Austin. 



the third row of double plates there is some irregularity, 

 suggestive of a change to three columns instead of two; but 

 this does not seem to continue, and probably should be 

 regarded as an accidental abnormality. 



The inner surface of the ambulacrals is divided into two 

 fields by a transverse elevation, which arises gradually at 

 the outer end (next the interradius), but becomes more pro- 

 minent at its inner end where it bends round to meet the 

 corresponding ridge of the adjacent ambulacral in the next 

 column. These ridges are stouter in the ambulacrals nearer 

 the peristome, and in ambulacrum B those of column a are 

 seen bending adapically so as to meet the adorally bending 

 ridges of the plates in column h. This increased stoutness 

 of the ambulacrals and the apposition of their ridges afforded 

 additional support to the peristomial frame. 



May we not see here the beginnings of a regular perignathic 

 girdle? The low ridges bordering the adoral notches of the 

 primordial interambulacral suggest incipient apophyses, and 

 these elevations of the adoral ambulacrals may have served 

 for the attachment of the retractor muscles. They are con- 

 spicuous structures even in the fossil, where they are broken, 

 but in a perfect specimen they would have been still more 

 conspicuous. The auricles of later echinoids are separated 

 from the ambulacrals on which they rest by a suture, but 

 these processes are part of the ambulacrals. That, perhaps, 

 does not forbid the hypothesis of their subsequent separa- 

 tion. If, as in Lepidesthes (Jackson, 1912, pi. 68. fig. 3), 

 the ambulacrals flowed down on to the peristome, then the 

 attachment of the retractors must have kept moving from 

 the processes of one row of plates to those of the succeeding 

 row. So awkward an arrangement may have been super- 

 seded by the separation of the processes and their conversion 

 into true auricles. It is hard to believe that the auricles 

 originated later as independent elements, and the suggestion 

 that they were modified from pre-existing subambulacral 

 elements (e. (/., floor-plates) does not appear to me to be 

 supported by adequate evidence from the fossils. 



From the transverse ridge the ambulacral plate slopes to 

 its adapical margin, which is flattened out in a slight flange- 

 like rim. On the other side the plate slopes to its adoral 

 margin, passing under the next plate (as seen from the 

 inside). 



The Ambulacral Pores lie on the adoral side of the trans- 

 verse ridge, a little to the outer side of the median meridional 

 line of the plate. They appear in many cases to be very 

 close to the adoral margin of the plate ; this, however, is 



