208 R. T. JACKSON — STUDIES OF PAL.EECHINOIDEA. 



the upper portion of the test in being distinctly hexagonal instead of sub- 

 hexagonal or tetragonal in form ; they also differ in being more drawn 

 out in the horizontal axis. The plates of this ventral portion of Lepi- 

 desthes are closely like the plates of the ventral border of Melonites midti- 

 porus, plate 2, figure 3, the similarity being in the number of columns 

 and the outline of the individual plates. This is a very important fact 

 in view of the S3^stematic relations of the genera. The passage from the 

 four columns at the base to the eight columns seen higher up is hidden 

 by local imperfections of the specimen. There are about two plates op- 

 posite each interambulacral plate. 



The clearest interambulacrum, which is area A of our figure, consists 

 of three columns of plates at the ambitus and tliroughout the greater 

 part of its length. At the ventral portion, however, a fourth column 

 exists. In area C the fourth column is clearly defined and terminates 

 dorsally in a plate A, above which it could not be traced. This fourth 

 column exists for such a brief period it seems reasonable to say that three 

 columns is the preponderant feature of the species. It is an unusual 

 feature not seen in any other species of Palaeozoic Echini for the greatest 

 number of interambulacral coUimns to exist so close to the ventral 

 border and then one to drop out (see page 150). It might be questioned 

 whether this is the ventral end except for the presence of buccal i)yra- 

 mids which are clearly seen on the other side of the specimen and are 

 just visible as two projecting ])oints, D D, on the side which is figured. 

 The structure of ambulacrum B also serves to orient the axes. The two 

 outer columns of interambulacral 2:>hites are nearly or quite hexagonal, 

 not pentagonal as in Melonites ; the plates of the median column are also 

 hexagonal. Most of the plates are worn so that they do not show sur- 

 face ornamentation ; but at the ventral portion of interambulacrum C 

 the plates are marked by somewhat obscure and very slight spine bosses, 

 all of one kind and scattered evenly over the surface of the j)lates. The 

 interambulacral plates imbricate aborally, while the ambulacral plates 

 imbricate also, but adorally. 



A specimen in Yale University Museum evidently belongs to this 

 species. It is from the Keokuk grouj:*, of Crawfordsville, Indiana. It is 

 of the same size and proportions as the type, and is densely clothed with 

 small si)ines about 2 millimeters in length. Jaws are well preserved at 

 the oral pole. Another specimen in Yale University Museum, also from 

 Crawfordsville, as ascribed to this species, Lejndesthes ivoriheni. It is 

 considerably smaller than the type, measuring only 1.3 centimeters in 

 height, so that it is probably a young one. The ambulacral plates, both 

 ventrally and at the ambitus, are similar to those shown in plate 9, figure 

 53, but only seven columns were made out. In the interambulacra there 



