116 Prof. P. M. Duncan on the 



primary plate (c). The lower component is a normal primary 



(«)■ 



If tlie ambulacral plates are separated along the ambulacral 



median sutm-e the usual knobs and sockets are seen upon the 

 opposite edges of the plates, and amongst them dark lines are 

 seen under benzule (fig. 1). The knobs and sockets in the 

 figure are in lines and groups, and most are near the outer 

 part of the test ; the lines are almost straight and some reach 

 from one surface to the other, and all are the joined sutures of 

 primary components of compound plates. 



Some lines, however, have a slant, and whilst most are simple, 

 others have a curved offshoot which starts below the outer surface 

 and, after curving, becomes straight. As the space included be- 

 tween two lines of sutures is a plate or part of one, so the surface 

 between the bend and the straight suture is a part of a plate. 

 That this is the case is easily noticed in such plates as fig. 9* 

 (a) , for the line of the adoral suture of the upper primary (c) 

 can be traced to the median edge of the plate and partly 

 upwards, but not to the surface of the test ; it is represented 

 in fig. 9 at x. In every instance of this bending of a sutural 

 line as it passes from within outwards in the test there are 

 proofs of the outward addition of material having buried the 

 suture and its plate, so that the outer markings of such a 

 plate would not tally with the inner, and these last are relics 

 of the early state of growth of the test. 



AmbJypneustes. 



The ambulacral plates of Avihlypneustes (fig. 13) are low, 

 broad, and thin, and the pairs of pores are in large peripodia; 

 the adoral pair of a triplet is not placed relatively so far 

 inwards as in Microcyphus ; but the appearance on the inside 

 of the test is very different. In Amhlypneustes the middle 

 pair of a triplet is nearest the ambulacro-interradial suture, 

 and the aboral pair is placed obliquely above and inwards to 

 the middle pair ; and this obliquity is continued to the adoral 

 pair of pores of the plate next in vertical succession. Hence 

 there is the common appearance presented of sets of oblique 

 pairs in threes, and this is shown to perfection inside the test, 

 and the appearance is intensified by the obliquity of succeeding 

 sets of three having their inner pores along the same oblique 

 line as the outer pores of triple pairs placed above and below. 



A primary plate, which is the aboral constituent of Micro- 

 cyphus (c, fig. 6), does not exist, however, in Amhlypneustes^ 

 for the aboral component (fig. 13, c) is a low broad demi- 



