Vol. 55.] DR. A. W. ROWE ON THE GENTJS MICRASTER. 507 



Peristome. — Micraster cor-bovls forms the type of the smooth, 

 flat, unbroken peristomal ring, with no trace of divisions between 

 the plates of the peristome (PI. XXXIX, fig. 1), and we have an easy 

 transition from this species into M. LesJcei^ where the ring is still 

 prominent ; but the subsequent separation of the peristomal plates 

 is indicated only by slight nodulations in the outline, and in most 

 instances even these may be wanting. The peristome is then seen 

 as an unbroken raised ring (PI. XXXVIII, fig. 3). 



From M. Leshei into the narrow ' sutured ' variety of M. prce- 

 cursor of the Holaster ijlanus-zone. is the next stage, for here we 

 find the labral tip and the peristomal plates adjoining it smooth and 

 unjointed; while the anterior segment of the peristome, for the first 

 time, shows indications of definite jointing between the neristomal 

 plates (PL XXXVIII, fig. 4). In all the other forms in the 

 H. planus-zoiae the labral tip is the only smooth part of the peristome, 

 and the jointing of the anterior segment becomes more marked. This 

 description is true, in an increased degree, for all forms in the zones of 

 M. cor-testudinarium and M. cor-cmguinum (PL XXXVIII, figs. 7-9). 

 In the M. cor-anguinum-zone, in addition, there is a tendency for 

 the surrounding granulation to invade the ends of the peristomal 

 plates which are farthest from the oral cavity (PL XXXVIII, fig. 9). 

 There is also a marked tendency for the anterior peristomal plate 

 to become longer, and there is a progressive increase in this direction 

 from the B. jolamis-zone to the M. cor-testudinanum-zoTie, and from 

 the latter to the zone of 31. cor-cmguinum. The granulation invades 

 the margins of the plates in the same manner as in the ambulacra 

 (PL XXXVI, fig. 7). 



Periplast ronal area. — This is the broad band which surrounds 

 the plastron and labral plate, separating them from the general 

 tuberculated surface of the base. The distinction between M. cor- 

 hovis and 31. LesJcei with respect to this area is brought out clearly 

 in PL XXXIX, figs. 4 & 5, and in PL XXXVIII, fig. 1, and the 

 differences are formulated in the description given of 31. cor-hovis 

 on p. 521. 



In all low-zonal forms the granulation is fine, becoming coarser 

 as we reach the top of the 31. cor-testudinaritmi-zone ; while at the 

 base of the 31. cor-angidnum-zowQ it is notably stronger, and in 

 the high-zonal forms it reaches its maximum, and is there invariably 

 mammillated (PL XXXVII, figs. 6-9). 



Depth of ambulacra. — The progression from the deep low- 

 zonal grooves to the shallow high-zonal grooves is plainly brought out 

 in the Tabular Summary (pp. 512-513), and is diagram matically 

 depicted in the vertical sections of the ambulacra given below each 

 photomicrograph on PL XXXVI. The high-zonal exceptions to this 

 rule are found only in the broad forms, such as the 31. cor-anguinum 

 var. latior of this paper, and 31. glyphus^ Schliiter, of the zone of 

 Belemnitella quadrata. This merely illustrates the tendency of broad 

 forms, in all zones, to have deeper ambulacra. 



