MIOCIDARIS. 249 



excepting that one plate in area E shows the remains of the perforation of -a primary tubercle. 

 In the mold the horizontal boundaries of the interambulacral plates are fairly well denned as 

 depressed furrows extending from the adradial suture to near the middle of the interambula- 

 crum, but the median sutures are mostly indistinct. On examining the interior of a half- 

 grown Eucidaris tribuloides it is seen that the horizontal boundaries of the interambulacral 

 plates stand out as prominent ridges, whereas the median boundaries do not, and this may 

 well account for the differences seen in the mold. The mold of each plate presents a slightly 

 convex form with a small mammillate elevation (text-figs. 239a, b) which corresponds with the 

 concave form and deeper impression beneath the tubercle on the interior of the plate of a cida- 

 rid, as seen well in an immature Eucidaris tribuloides. 



This species is of exceptional interest, as it is the first sea-urchin recorded from the Palaeo- 

 zoic of America with two columns of plates in an interambulacrum ; also it extends the range 

 of the genus Miocidaris and the order Cidaroida to the Lower Carboniferous, whereas previously 

 they were not known below the Permian. The occurrence of this type with two columns of 

 interambulacral plates in the Lower Carboniferous adds support to my view, as earlier pub- 

 lished (1896, p. 237, table facing p. 242) and maintained in this memoir, that two columns of 

 plates in an interambulacral area is a primitive structure and types having this structure 

 are to be derived from an ancestor with one column of plates in an area, and not to be derived 

 from an ancestor with many columns in an interambulacral area, as has been usually held 

 (p. 211). 



Miocidaris cannoni, which I take pleasure in naming for Mr. Cannon, from whom I received 

 the specimen, differs from other species of the genus in its large size and numerous interambula- 

 cral plates in a column. The number of plates is perhaps sufficient to justify generic distinction, 

 but as the surface characters are unknown, it seems best to place the species in the genus Mio- 

 cidaris, to which at least it is very closely allied. While this species occurs in the Lower Carboni- 

 ferous and is therefore very much older than Miocidaris keyserlingi of the Permian, it is placed 

 higher than that species because, on account of having a large number of plates in an inter- 

 ambulacral column, it is farther removed from the primitive than is keyserlingi with a smaller 

 number of plates in a column. 



Mr. Cannon informs me that he recently obtained the specimen, and that it is from the 

 Millsap Formation of the Rocky Mountain Lower Carboniferous, from near Denver, Colorado. 

 The specimen was collected by Mr. George Day, and was found in one of the loose chert concre- 

 tions in the Platte River drift. While not in place, material of this lithological character 

 according to Mr. Cannon and other authorities is known in Colorado only from the Millsap 

 Formation, which occurs as outcrops in the Front Range. Girty (1903, pp. 162, 167) ascribes 

 the Millsap Formation to the Mississippian, or Lower Carboniferous. The specimen is now 

 in the collection of the Museum of Comparative Zoology 3,201. 



