H. L. Hawkins — The Lantern of Perischodonius. 301 



total length 24-4, greatest width (margin of epiphysis) 12-3 mm. 

 As is usual among Palaeozoic lanterns, the maxillse are very wide 

 (measured radially) and have the interpyramidal faces considerably 

 curved. The convexity of the outer surface (at the symphysis) is 

 such that the tooth must have projected almost at right angles to the 

 line of the upper part of the pyramid — perhaps working horizontally. 

 None of the specimens show the outer face completely, but the 

 indications point to its being similar in proportions to that of 

 ArchcBocidaris nerei (cf. Jackson, op. cit., pi. x, fig. 6). 



Epiphyses (Fig. 3).— Two of these ossicles are shown, one being 

 almost complete and exposed from the side of its articulation with 

 the maxillse. Its radial width is 12-2 mm., and its height at the 

 outer edge is 10-4 mm. Its inner height is only 4'2 mm. There is 

 a slight concavity on the free part of the surface, and a very low 

 knob at the inner and upper corner. The articulation surface is 

 irregularly and coarsely roughened. 



Compass (Figs. 4, 5). — One compass exists on the slab, exposed 

 from the side and upper surface. It is broken, apparently along the 



Perischodonius biserialis (?). 

 Fig. 3. Epiphysis, articular surface. 

 Figs. 4,5. Compass, upper and side views. 



suture by which all compasses are traversed, and has been pressed 

 into one of the interambulacral plates. A scar is left on the plate 

 where the compass has been destroyed, and this additional part is 

 indicated on the figures by shading. 



The total length, as far as can be ascertained, is 10 '5 mm. 

 (measured along the chord). From the outer-muscle prominence 

 to the circular-muscle prominence is 4 mm. From the outer 

 extremity to the fracture (? suture) is 6 mm. The height at the 

 inner (broken) end is 2 '4 mm. The compass is flattened and broad 

 (viewed from above) in the outer part, but inward from the cross- 

 fracture becomes compressed and narrow. This feature has probably 

 been exaggerated by crushing. 



The bifurcation of the outer end is not very marked, although the 

 prominences project laterally to a considerable amount. The most 

 striking features in the compass are the supports for the circular 

 muscle. These project even further outwards than the external 

 supports, as shown in Fig. 4. As far as I am aware, no such 



