NIAGAHA OKOTTP. 219 



The greatest diameter is in the direction of the two parts, and the shortest diameter is in the 

 direction of the line separating the two halves. This is shown not only by measuring across 

 the summit of the fossil, but through the centre of the costal plates. In specimens which are 

 compressed or distorted, these proportions may not be preserved ; but in all well preserved 

 specimens it is a prominent feature. 



The arms vary in number, but in full grown specimens there are thirteen : in several smaller 

 specimens, I have counted only nine ; and in a very young specimen, there are not more than 

 three or four ; -while in several somewhat larger, there are six. It is evident that the number 

 depends in some degree upon the age of the individual ; and in two or three very large speci- 

 mens, there remain the bases of fourteen arms. These arms appear to be composed, from their 

 base upwards, of a double series of joints ; at the base, semicircular, and scarcely interlocking ; 

 while in the middle and upper part, the plates are more extended laterally, and there appears 

 to be a thin extension or lamina from the middle of the uniting face on each side, penetrating 

 between every two of the adjoining plates, and strongly interlacing the whole series through- 

 out. The surface is finely granulated with stronger sharp tubercles. The tentacula are sharply 

 grooved along the centre, or apparently composed of a double series of joints alternating with 

 each other. 



The most conspicuous and interesting feature of the surface, exists in the ambulacral pores 

 which penetrate the plates. In the very young specimens, these pores are sparsely scattered 

 over the surface of the costal plates, scarcely taking any definite direction, and are barely visible 

 on the lower plates : in the farther advanced stage, they have assumed the direction so con- 

 spicuous in the full grown individual*. Beginning at the base, we find there are six conspicuous 

 double rows of pores, extending from the base to the highest angles of the pelvic plates ; being 

 a single row on each of the pentagonal, and two on each of the hexagonal plates. In looking 

 at the structure, it will be perceived that these angles meet the intersection of each two of the 

 costal plates ; and these double rows of pores are continued, as it were diverging one into each 

 of two adjoining costals at their outer basal angles, and extending in single rows to the centre 

 of the plate. In addition to these, there is usually an interrupted row of pores on one or both 

 margins of the pelvic plates, along their line of junction ; and these are continued, sometimes 

 as a single and sometimes as a double row, from the angle at the base of the hexagonal and 

 heptagonal costal plates, to the centre. Sometimes there is a scattered row of pores from the 

 centre of the base of the pentagonal costal, extending to the centre ; and sometimes there is an 

 interrupted double row of pores, instead of the usual single row^, extending from the centre lo 

 the outer basal angles of the other costal plates. In many instances these pores are obsolete in 

 part ; and where a double row exists in place of what is usually a single one, a portion of the 

 whole are obsolete. On the upper part of the costal plates, these pores become more numerously 

 and perfectly developed. To each of the higher angles of these costal plates, which meet the 



' The appearance of being scattered arises in part from the small number present, there being often only one and 

 sometimes none, in a line from the centre to the angles of the costal plate. At this period of growth, also, there are 

 pores between the centre and lateral margins of these plates, which become obsolete in the more advanced stages. 



