SAGENOCRINIDAE 227 



in a more or less disturbed condition, pushed out of their natural order and sequence. The 

 upper of these is no higher than the axillary primibrach, and although the two rays are pre- 

 served to considerably above the second bifurcation there is no sign of any interbrachials 

 between them above the level mentioned. By reason of the flattening of the specimen by 

 lateral pressure the interbrachials which are shown would be forced up higher than their 

 natural position, and the lines and angles of their sutures would be much altered. 



At the opposite side, where the interbrachials are similarly squeezed out of their normal 

 position, only four or five are visible pushed into a vertical row to the same height as the 

 others, with no indication of any plates between the rays above the level of the last primibrach 

 (PL XXII, fig. id). Therefore at the most this specimen shows the presence of from 4 to 8 

 or 10 interbrachials occupying the lower part of the area only. I have given a figure of the 

 posterior side, which was not done by the authors (PL XXII, fig. ib). The posterior basal is 

 perfectly plain, angular above, followed by two plates suturally united to it, with others suc- 

 ceeding them ; at the right two or three more are indistinctly shown, tending to form a series 

 bordering the right posterior ray. At one place a single intersecundibrach is seen. 



No other specimens were known to De Koninck and Le Hon ; and upon the data fur- 

 nished by the one at Brussels they constructed their generic diagram (PL XXII, fig. 2). 

 This was substantially correct as to the form of the posterior basal and one or two ranges of 

 anal plates succeeding it. It shows interbrachials in one area to the number of 10, with 3 

 more indicated by dotted lines to fill out the arrangement as it was supposed to be. This, of 

 course, was taken wholly from the right posterior interradius now shown by my figure ic; 

 and while it is possible to deduce such a number and arrangement from what appears there, I 

 should consider this somewhat doubtful when taken in connection with the much smaller 

 number just as plainly shown at the left posterior interradius. 



Since De Koninck's time three more important specimens of this species have been found 

 in the same lower limestone beds at Tournai, all of which came into my possession, and are 

 figured upon Plate XXIII. Two of them are of the same large size as the types, have the 

 same surface ornament and the same deeply rounded rays and ray divisions, with wide inter- 

 brachial spaces between them. Each of these has the same angular posterior basal with two 

 good sized anal plates suturally united to it, and two or three much smaller ones in a second 

 range ; and in both one or more interbrachial areas are preserved with all the plates in their 

 natural position. These specimens are in good condition, having come out of a soft matrix 

 readily removed by cleaning, and while somewhat broken in collecting they were easily put 

 together, so that I do not think any important solid structure is missing except possibly a few 

 small plates at the right side of the anal interradius. A third specimen, consisting only of 

 the stem, infrabasals, and two basal plates, has the posterior basal with the same angular 

 sutural margin (fig. 3). All three have the infrabasals distinctly visible. 



It will be observed that the primary interbrachial structures are the same in the two 

 complete crowns, consisting of one quite large plate followed by three smaller ones abreast, 

 with a crescentic distal margin following the curvature of the margin of the brachials with 

 which they connect, and rising to about the middle of the second primibrach. Upon com- 

 paring the figures of these specimens with those of the type as interpreted by De Koninck and 

 Le Hon, it will be seen that the only difference between them lies in the possibly greater 

 number of anal plates in the latter, and of interbrachials in one area ; the other area preserved 

 would have about the same number as in my specimens. There can therefore be no reasonable 

 doubt that these later acquired specimens, from the same horizon and locality, represent the 

 same form as De Koninck and Le Hon's types ; and we are accordingly warranted in looking 

 to them for the generic and morphological characters which they exhibit so much more clearly. 



I have given the foregoing recital of facts with what may seem a wearisome prolixity of 

 detail, but I have a definite reason for it. Beginning with Hall in 1858, a large number of 

 species from the American Lower Carboniferous have been described by several of our best 



