IV, Trenton Limestone Edrioasters. 1914 123 



sharply up from the interradial areas, they form a rounded margin 

 on each side of the groove, and then dip almost straight down to 

 the median suture, where they meet at an angle of about 130° (PL X, 

 Fig. 6). Near the peristome this angle is slightly less, that is to 

 say the depression is more marked; but at the distal ends it becomes 

 wider and almost disappears. Where the grooves bend sharply, 

 the dip of the floor-plates on the inner side of the curve becomes 

 steeper, and the angle therefore less. All these changes can be 

 traced in PI. X, Figs. 1-3. 



The suture between two floor-plates of the same side of the groove 

 is depressed from a point just within the rounded margin right down 

 to the median suture. The depression is deepest at its outer end, 

 where also it is slightly expanded in circular form ; but thence 

 it gradually widens towards the median suture. That suture also 

 is depressed, so that there is a sinuous median channel (PI. X, 

 Figs. 4, 5). 



Further details of structure in the floor-plates are to be noted. 

 The rise above the general surface of the interradial area may be 

 strongly marked; in B it amounts to circa 1*9 mm., and about 

 half-way up this outer part of the plate is a slight depression, which 

 seems to continue the pustular ornament of the interradials. 



In the proximal half of a groove, in A, the length of each floor-plate 

 (at right angles to the radius) is about 2-7 mm., and the width 

 (measured parallel to the radius) '75 mm. In B the latter measure- 

 ment is as much as 1 14 mm.; but as the groove narrows distally 

 this also decreases, so that in the distal part of the right posterior ray 

 it is noted as *66 mm. At the extreme distal end of a ray, as seen 

 in A, the floor-plates diminish considerably in size, but continue 

 to alternate, and are arranged fan wise. There is no distinct terminal 

 (Text-fig. 4). These appearances were not clearly exposed when 

 Plate X was drawn. 



The groove along the suture between adjacent floor-plates was not 

 really so simple as it appears in most specimens. In the left anterior 

 ray of E 16172 (PI. XIV, Fig. 3) are seen two slight ridges starting 

 from near the centre of the circular expansion (pore), and diverging 

 as they pass, one on each side of the suture. On the side of each 

 ridge remote from this suture are about three slight swellings, and the 

 ridge itself may thus be broken up into a row of tubercles. Hound 

 the circular expansion is a slightly raised rim. These features, when 

 once appreciated, can also be detected in parts of specimen A. 



In specimens that have not been thoroughly cleaned, the outer 

 circular ends of the depressions between the floor-plates are filled 

 with matrix, and this, according to its amount, gives the appearance 

 of a series of pores or slits. If the matrix be also retained between 

 the divergent ridges just described, it may produce the appearance 

 that in 1858 (pi. viii, fig. 2a) misled E. Billings into the supposition 

 that there was a second inner series of pores. Careful removal of the 

 matrix entirely does away with the latter appearance, and reduces 

 the supposed outer series of pores to little more than pin -pricks, each 

 in the centre of the circular expansion (PI. X, Fig. 4). It has in 

 fact been suggested that further cleaning would do away with the 



