H. L. Hawkins—Ambulacral Structures in the Holectypoida, 349 
I11.—Somr AmBpunacRAL STRUCTURES IN THE HOLECTYPOIDA. 
By Herzsert L. Hawxins, B.&c., F.G.S. 
FJ\HE phenomena of ‘plate-crushing’ in the course of the growth 
of the ambulacra of the regular Echinoids have for many years 
been utilized for purposes of classification of those forms, but the 
similar features which exist among some of the irregular types have 
not been studied in such detail as to render them systematically 
useful. It is with a view to showing that the structure of the 
ambulacra in one group of the Echinoidea Jrregularia can be relied 
upon as an index ofthe relationships of the various genera, that this 
note has been written. It makes no pretence at completeness, as the 
material and time available have not been sufficient to allow of an 
exhaustive study. But the results of those observations that have 
been made seem to indicate that among the Holectypoida it should 
be possible, with the aid of zonal collecting, to trace genetic series 
among the various species, and perhaps to bridge the gulfs existing 
between the genera, by a study of their ambulacral plating. 
That this method could be applied to the Holectypoida only 
seems certain. Among the Spatangoida the ambulacra are composed 
throughout of simple primary plates, varying greatly in shape and 
size, but not complicated by the development of demi-plates, except 
in the immediate vicinity of the peristome. The petaloid portions of 
the ambulacra in the Clypeastroida may exhibit the phenomena of 
‘ plate-crushing ’, but I have only observed demi-plates in one 
specimen of a Clypeaster from the Miocene of Southern Europe. In 
any case, when the ambulacrum is traced beyond the limit of the 
petal, it is found to be composed of polygonal primary plates 
throughout its length. 
In the Holectypoida, a group which is regarded as being the order 
of the Irregularia least removed from the regular type, the complexity 
of the ambulacral plating is often comparable with that in the simpler 
forms of the Diademoida. Among the latter group, in the more 
primitive sub-order of the Diademina, the ambulacrals are united in 
sets of three to form one compound plate, and a triple arrangement 
also obtains in the plates of the Holectypoida. But the arrangement 
of the component parts of a complete ambulacral in such a genus as 
Conulus, although in parts similar in appearance to that which is 
found in forms allied to Ardacia, is arrived at in a strikingly different 
manner. So regular is the order in which the plates become crushed 
into demi-plates that it is possible to regard the series of three which 
ultimately form a compound plate as being made up of three entirely 
different kinds of plate. 
Before the region of crushing is reached there is no visible difference 
between the members of the simple row of primaries (see Fig. 2, 
Pygaster), but when the modification sets in, these plates succumb 
with surprising regularity, always in a definite order. It will be 
convenient for the purpose of this note to designate the individuals 
of the series of three plates by the letters a, 4, ¢ (see Fig. 1). These 
individual plates bear the following relation to one another: plate a 
will remain a primary throughout the entire ambulacrum, and will, 
