356 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



ject to a severe shearing strain along the divisions which would then pass directly 

 across it. In animals in which the radial divisions are marked by lines of strength, 

 as in the coslenterates, their number is always even, for the reason that the con- 

 tinuation of these lines of strength across the entire animal give an added rigidity 

 which would be lost were these lines not continuous across the center. Now if the 

 basals were superposed directly upon the infrabasals and bore the radials, single 

 or paired, directly upon them, the entire, animal would be divided from the stem 

 outward by five sutural lines separating five solid calcareous masses. Such an 

 arrangement would greatly weaken the animal; every time an arm were struck 

 there would be great danger of tearing out an entire sector as far down as the top 

 of the stem. For this reason we find the radials alternating with the basals instead 

 of superposed directly upon them, and five instead of ten in number. The same 

 mechanical reason has induced the prolongation of the basals and infrabasals 

 into sharp angles between the bases of the succeeding plates, for a sharply zigzag 

 sutural line is not subject to the shearing strain to which a straight bine of weak- 

 ness would be liable, and thus the extremely angular line marking the union of the 

 basals and the radials, or of the infrabasals and basals, is far stronger than a straight 

 line would be in the same situation. 



Against this mechanical interpretation of the origin of zigzag arrangement of 

 the calyx plates in the crinoids it might be urged that in the echinoids, which are 

 more or less globular and rigid and therefore as a whole comparable to a crinoidal 

 calyx, ah 1 except the apical plates are arranged in columns. But the two cases are 

 not by any means the same. The ambulacral series of the echinoid are analogous 

 to the biserial crinoid arms, and the interambulacral series to the perisomic interra- 

 dial plates such as are well seen in certain comasterids in which, though of purely 

 fortuitous origin, and arising very late in life, through a segregation of the peri- 

 somic spicules into dense groups, their arrangement is strictly comparable to that 

 of the echinoid interambulacrals. Originally the echinoid was provided with 

 strong internal muscles and possessed a more or less flexible test, as we see in the 

 echinothurids to-day. This resulted in the retention of the columnar arrange- 

 ment of the plates and also induced a narrowing of the individual plates so that, 

 though they alternate in adjacent columns, the angles of the horizontal suture 

 lines are eliminated so far as possible. With the plates in vertical columns and 

 the plates in each column very narrow there is given a maximum of flexibility along 

 the axes at right angles to the longer diameter of the plates. With the deteriora- 

 tion of the muscles, though still retaining the columnar arrangement, the plates 

 became broader with much more prominent angles, approaching the hexagonal 

 in form; so that, in such forms as the cidarids, a very considerable rigidity is at- 

 tained, and in exactly the same way as in the crinoid calyx, the adjacent columns 

 of plates alternating with each other and joining by a very sharply angular line 

 resulting in a firm dove-tailing, just as the basals are joined to the radials, and the 

 plates of each column joining the plates above and below for a minimum length of 

 their edge while interlocking with the alternating plates for a maximum, just as the 

 circlet of basals is interlocked between the circlet of underbasals and the circlet of 

 radials. 



