FOSSILS OF THE NIAGARA GEOUP. 155 



which retains parts of five of the tlioracic segments attached. The speci- 

 mens appear to be identical in form, in the ornamentation of the sur- 

 face, and in the number of segments, with those from the Clinton group 

 of New York, Avhich have been referred to E. punctatus, but differ very 

 considerably from the figures and descriptions of that species as given by 

 European authors. Some of the Ohio specimens are considerably larger, 

 and show the markings more distinctly than those illustrated in Vol. II., 

 Pal. N. Y., as above cited. 



The form of the pygidium is triangular, as wide, or a little wider, than 

 long, measured on the internal cast, which is the condition in which it 

 occurs. The prolongation of the posterior extremity being much con- 

 tracted in this condition, the length of the plate on the perfect crust has 

 probably somewhat exceeded the greatest width. The axial lobe forms 

 less than one-third of the entire width, is strongly marked and some- 

 what flattened on the top, except at the anterior portion, and divided 

 transversely into about twenty or more segments. The center of the 

 lobe is marked by a row of five, prominent, rounded nodes, the anterior 

 one of which is situated on the second segment, the next on the fifth, 

 the others having three segments between each node. It is possible 

 that on the exterior surface of the crust there may have been an addi- 

 tional or sixth node on the terminal segment of the axis, but no evi- 

 dence of it exists on the specimens examined. The lateral lobes are 

 marked by seven segments, all of which are directed abruptly backwards 

 from their origin at the longitudinal suture, and each segment ap})ears 

 to have been marked near its inner end by a transverse node, and, also, 

 by one or more other nodes further out on the rib. 



The thoracic fragments are too imperfect to afford characters for a 

 complete description. The axial lobe is shown to be highly convex, and 

 the lateral lobes flattened for nearly half their breadth, at which point 

 they are abruptly bent downwards and slightly inclined backwards. 



Compared with specimens from the Clinton group of New York, there 

 are very few points of difference of any importance, except the more dis- 

 tant nodes on the lateral lobes of the pygidium where the specimens 

 preserve the original crust ; but compared with the figures of European 

 authors, the differences are more marked — those all representing a single 

 row of nodes along the middle of each lateral lobe, while the Ohio speci- 

 mens have had two or more — probably four — on several of the anterior 

 segments. The New York examples show even more than four on the 

 anterior segments. The figures given in Murch. Siluria, plate 10, figs. 

 4 and 5, differ in the disposition of the nodes on the axial lobe to such 



