ECHINODERMATA OF THE BRASSFIELD FORMATION 
5 
Ohio^ also has been added. This upper member of the West 
Union formation is distinct faunally from the Lower or Bisher 
member. The latter is correlated provisionally with the Iron- 
dequoit member of the Clinton of New York, so that the upper 
member may correspond to the lower part of the Lockport for- 
mation of that state. The so-called Niagara shales of the earlier 
reports of the Ohio Geological Survey, known as the Crab Orchard 
shales in Kentucky, contain, in their upper layers, a fauna in- 
cluding some of the characteristic species of the typical Clinton 
of the central and more eastern parts of New York, such as 
Liocalymene clintoni. 
Several of the specimens here discussed present very unusual 
features. The nodose aggregation of columnals at the top of 
the stem of Brockocystis nodosarius is one of these. The frequent 
coiling of the stem of an unknown crinoid, (plate I, fig. 6), with 
the narrow end of the stem at the center, is another. An ances- 
tral form of Myelodactylus is a third. On plate II (fig. 6) are 
figured fragments of a crinoid which would be a Dimerocrinid 
if it had sub-basal plates ; but the latter apparently do not occur. 
On plate VII are presented several figures of an echinoderm 
{Stereoaster) regarding which little is known at present beyond 
the fact that it resembles a star-fish in appearance but not in 
structure. It promises to be one of the anomalous forms of 
which the relationship remains unknown, at least for the present. 
Finally, on plates IV and V is figured a star-fish which evidently 
diverges distinctly from typical forms of Mesopalaeaster. 
Brockocystis nodosarius sp. nov. 
Plate 7, figs. 1, 2, S, 5 
Theca small, oblong in outline, 11 or 12 mm. long and 9 or 
10 mm. wide. Pectinirhombs present on plates 1-5, 12-18, 
14-15, and 10-15, but absent on plates 11-17; more or less dis- 
crete along the suture which separates the two plates forming 
each rhomb. The lower boundary of the pectinirhombs is more 
strongly defined on plates 1 and 12 than is the corresponding 
upper boundary of the same pectinirhombs on plates 5 and 18. 
