BRACHIOPODA. 239 



the valves ; usually, however, the septa of the brachial valve are very short 

 and rest upon the inner surface of the shell. It sometimes happens that 

 these septa unite before reaching the inner surface, and the spondylium 

 thus formed is supported by a very low axial septum. This is the case in the 

 original specimen of P. bisinuatus, McChesney, and in the Wisconsin shell refer- 

 red to that variety by Whitfield.* It is more conspicuously developed in the 

 Iowa shell which has just been mentioned as P. oblongus, var. subredus, and it 

 serves to confirm the varietal character of that form. It has been already 

 observed that the union or independence of these dorsal septa in the genera 

 Anastrophia and Parastrophia can be regarded as a feature of only secondary 

 importance. In the later pentameroids it will be found that the difference 

 becomes fixed and of more positive significance, but in the Silurian shells it is 

 still a variable feature, but not of usual occurrence. At times an exceedingly 

 faint obsolescent radial plication of the exterior is observable in P. oblongus, and 

 this feature is also occasionally apparent in P. pergibbosus, Hall and Whitfield, 

 and more noticeable in P. occidmtalis, Hall, of the Guelph fauna of the Province 

 of Ontario, and of the Niagara fauna of Ohio and Wisconsin.! 



* Geology uf Wisconsin, vol. iv, p. 290, pi. xvii, fig. 3. ltS82. 



t The shells which are currently referred to Pentaviems pei-gibbosus also vary not a little among them- 

 selves, and it would be no difficult matter to acijuire a series of forms to demonstrate that this is but another 

 extreme of development which has oi-iginated in P. oblongiis. The originals of P. payibbosus from the 

 Niagara dolomites of Darke county, Ohio, are rather small shells with long, oblique cardinal slopes, nar- 

 row umbones and very deep valves. In the limestones about Milwaukee shells of this character attain 

 gi-eat size, and in the chert of Jones county, Iowa, occui'S a very small shell which cannot be separated from 

 this species by any decisive characters. 



Mr. Whitfikld has figured J as one of the variations of P. obhngiis, a gibbous shell from the upper 

 coral beds of the Niagara group at Ashford, Wisconsin ; a similar, though persistently smaller shell 

 abounds in the dolomites of the Maquoketa region near Dubuque, and at Hopkinton, Iowa. The latter has 

 been generally identified as P. perj/ibbosus. It is, however, quite a different shell from that occuring in 

 Darke county, Ohio, its full, rotund valves, broad across the cardinal region, jjroducing an expression dis- 

 tinct from that of P. pergilboatui, while the suggestion of Irilobation of the surface which is shown on all the 

 specimens examined, indicates its neaier relations to P. oblongus; as a corrected identification of this shell, 

 the name P. oblongus, var. Maquoketa, is suggested. It is observed above that the variety of P. oblongus, 

 prevailing in Ohio (var. bisinitatus), U represented with extreme rarity among the Clinton shells of New 

 York. Similarly, the variety Majiu)keta is known to occur on this side of the Mississippi only, in the Wis- 

 consin locality cited. On Plate LXVIII, figure 13, there is given a figure of a shell of great size, probably 

 from Indiana, which is nearer to this than to any other foi-m of P. oblongus. 



The differences between P. pergibbosus and the P. occidentalis. Hall, from the Guelph fauna are also 

 obscure. The latter has the cardinal slopes very broad, the axial slopes flattened or depressed and the 



t Op. cU., pi. xvii, ngs. 8, 9. 



