no NEW YORK STATE MUSEUM 



Farther out on the ambulacrum the brachioles possess this outer 

 biserial arrangement between the first and the eighth but after 

 this are uniserial. The larger drawing of the seventh brachiole 

 at a in figure i shows a transitional form. These newer brachi- 

 oles show from 17 to 30 external plates. Each brachiole, with 

 the exception of the two oldest, starts with a single large kidney- 

 shaped plate very slightly less than .5 mm in longest diameter. 

 This is surmounted by two plates in which the lower or distal 

 is much the smaller; the next two plates above this are of nearly 

 equal size and of the pair that follows these it is the upper or 

 the one nearest the axis of the theca that is the smaller. It 

 appears as if the outer smaller plate of the first pair became the 

 foundation plate of the arm. The law of biogenesis would indi- 

 cate that biserial brachioles preceded the uniserial but the change 

 in this instance may be caused by a rotation of the brachiole to 

 the left on its own longitudinal axis, as is indicated at A7 figure 

 I and by other earlier brachioles of this specimen, bringing the 

 left-hand plates of a biserial arm to the front and making the 

 arm appear uniserial. The diagonal sections of the lower 

 brachioles in figure 2 show in places a section through a pair 

 of plates and the structure near the base of the brachioles sug- 

 gests the arrangement of the side plates and outer side plates 

 of Codaster. 



Where the upper brachioles have been turned away from the 

 wing plates, as shown in plate 3, lower figure, at a, the older 

 brachioles seem to have had an additional row of plates on either 

 side and alternating with the outer or back plates. A cross- 

 section of one of these brachials would give the form of a paral- 

 lelogram with its long axis set at right angles to the surface of 

 the wing plate. The middle two fourths of the rear wall consist 

 of a double row of very small, alternating, covering pieces. The 

 cross-section of the lower parts of the newer brachioles seems 

 to show only the back or outer plates (the end of but one of them 

 seen from the outside) with greatly elongated sides and with 

 traces here and there of what may be small covering plates. 



One brachiole in the anal interradius seems to have been 

 certainly free from the others for its entire length but the 

 brachioles with apparently uniserial back plates have their 

 margins zigzagged as if they had become bound, each to its 

 neighbor, at their sides. This may not have been true of all 

 the upper brachioles but where some of the older ones have 

 fallen away from the wing plate, as in plate 3, lower figure; 



