76 PALEONTOLOGY OF KENTUCKY. 



rock. In this shell the dorsal valve shows a regular deep concavity, into which 

 the beak of the ventral valve overlaps. 



The greatest depth of the dorsal concavity is about equal to half the length 

 of the hinge-line ; below the beak there is a slight impression into the umbo 

 of the other valve. The surface is ornamented by low, rounded, radiating 

 striae, reaching almost to the beak. My specimen, before alluded to, shows 

 real continuous plications, not mere isolated elevations above and near the 

 bases of each spine. Even the dorsal valve has faint indications of corres- 

 ponding striae. On these radii are placed at irregular intervals the long, 

 slender, round spines, with which the ventral valve is closely covered. These 

 spines are only indicated on the silicified internal casts by their scars, and on 

 better preserved shells, by their bases and short stumps ; they are only pre- 

 served in full length, where even the surrounding matrix became silicilied at 

 the same time with the shell. Concentric lines of growth are seldom noticed 

 on our shells. 



Formation and Locality. Occurs rather abundantly in the cherty layers superimposed upon the 

 hydraulic limestone of the Devonian formation around the Falls of the Ohio, in Jefferson county, Ken- 

 tucky, and in Clark county, Indiana, where it is usually found as silicifled casts of the interior. "Well 

 preserved shells, like that illustrated, showing both valves in perfect condition, with the exception of the 

 spines, are extremely rare. 



Productella semiglobosa. . SP. 



Plate XXVI., figure 7. 



Shell of medium size, semi-globose or sub-circular ; hinge-line somewhat 

 shorter than greatest width of shell; cardinal extremities rounded. Yentral 

 valve very gibbous, regularly curved from umbo to front, and also trans- 

 versely ; umbo only moderately elevated above surrounding surface ; beak 

 incurved upon hinge-line, not overlapping it into dorsal valve. Width and 

 length of shell about equal, but sometimes the width exceeding the length. 

 Dorsal valve apparently deep concave, but its other characters are not known. 

 Surface does not show any markings, except the stumps of a few isolated 

 spines placed at irregular intervals ; the figure 7, on plate 26, shows about 

 twice as many as in reality exist. I am unable to identify it with any of the 

 species of Devonian Productella known to me, and I, therefore, place it in 

 the above named new species. This shell has some similarity with some 

 middle-sized, but very ventricose forms of Strophodonta demissa, from which 

 it is, however, easily distinguished by its smooth surface, which shows only a 

 few spine-bases, while the Strophodonta demissa is covered by radiating striae, 

 and never becomes fully as ventricose as our shell. The specimen illustrated 

 is of about average size. 



Formation and Locality. Occurs in the Corniferous limestone at and around the Falls of the Ohio, 

 in Kentucky and Indiana. Only a few specimens are so far found. 



