: 
352 H. L. Hawkins—Ambulacral Structures in the Holectypoida. 
The four species I have chosen for the purpose of the diagram are, 
as far as I could obtain specimens, representative of the later specific 
developments in all four genera. The specimens utilized are all 
approximately adult. 
Pygaster macrocyphus, Wright (the only traceable co-type of which 
is in the South Kensington Museum), is from the Kimmeridge Clay 
near Boulogne. Although the adapical regions of the test are con- 
siderably broken, enough of the ambulacra can be examined to show 
that considerably more than sixty consecutive ambulacral plates, 
counting from the apex, in each ray are simple primaries. Apparently 
the development of demi-plates does not commence until a region 
about half-way between the ambitus and the peristome. Examples of 
P. semisuleatus and P. umbrella which I have examined show at least 
an equal postponement of the ‘crushing-point’, but it is not often 
that specimens suitably preserved to show the plate sutures are met 
with among the Jurassic forms. 
Holectypus serialis, Deshayes, from the Turonian of Algeria, is 
perhaps the latest member of the genus known. Two specimens in 
my collection show the plate sutures very clearly, although the 
diameter of the larger example is only 15mm. The first indication of 
‘plate-crushing’ occurs (of course, in set ¢) at plate 39, where this 
primary becomes wedge-shaped, reaching the median suture by a very 
narrow strip. Plate 42, the next of the ¢ set, is definitely a demi- 
plate, and plates 41 and 43 meet round it along the median line. 
Whenever the plates of the ¢ set are seen as the ambulacrum is traced 
towards the peristome, they are found to be demi-plates, whose 
extension towards the median suture becomes reduced the nearer they 
are to the mouth. Plate 47 is the first member of set 6 to show 
a cuneiform shape, and by the time plate 56 is reached this set also is 
represented by demi-plates, and the median ambulacral suture is entirely 
formed by the large primaries of set a. In H. depressus from the 
Lower Oolites, the crushing of the plates of set ¢ is postponed beyond 
plate 50, and that of the plates of set 4 yet further from the apical 
system. ‘hus in earlier Jurassic times this genus would appear to 
have been similar to Pygaster in the structure of its ambulacra, but to 
have developed in the course of time a far more complex condition 
than was ever reached by the latter genus. 
Discoidea Forgemolli, Coquand, from the ? Senonian of Algeria, 
exhibits the phenomena of ‘ plate-crushing’ in a degree advanced from 
that in Holectypus. Plates of set ¢ begin to appear compressed and 
wedge-shaped at number 33, and the reduction in size of the plates of 
set b is accelerated to such an extent that while plate 36 (set ¢) is but 
hardly separated from the median suture, plate 35 is already cuneiform. 
In a D. cylindrica from the Lower Chalk the first indication of crushing 
in set ¢ appears at plate 39, and in set 4 at plate 41. So that im this 
earlier form the proportionate rate of crushing of the two sets of plates 
is similar to that in D. Vorgemolli, but the point at which it commences 
is postponed by six plates. It will be noticed that in the case of 
D. cylindrica the ‘ crushing-point’ is identical with that in Holectypus 
serialis from a newer horizon. ‘There is thus some overlapping in this 
feature among the various genera, but in the fully specialized members 
