from the Miocene of Australia. 101 
It will be noticed that A is of a slightly different form from 
B, being smaller, but of greater proportional length and height. 
Specimen B is, in fact, rather broader and flatter than is usual for 
the species. On the evidence of size and tuberculation, it must be 
inferred that A (Text-fig. 1) is a somewhat younger individual than 
B (Text-fig. 2), although both specimens possess approximately the 
same number of plates in their tests. In both examples the ambital 
angle is very acute, and the adoral surface almost flat. 
In each of the four paired interambulacra (areas 1, 2, 3, and 4),' 
the adoral surface is entirely occupied by three plates—the unpaired 
plate at the peristome border, and two large multitubercular plates 
which make a kind of plastron. Of these large paired plates, those 
-in columns la and 46 are crossed by a faint curved suture, seeming 
to indicate that they are of dual origin. Lovén (Htudes) has inferred 
that one or both of these plates are often double in Spatangids, but 
seems not to have detectedasuture in them. Thereis the appearance 
of a similar suture across the corresponding plates in columns 2a 
and 35 in specimen B, but as this occurs in that individual only it 
requires confirmation, and is ignored in the numbering of the plates 
in the succeeding description and figures. 
In column la (Text-fig. 1) the plate next above the ambitus 
(plate 4) is narrow but complete. In column 14 there are two 
imperfect, lath-like plates intervening between the large ‘ plastronal’ 
plate (2) and the first tubercle-bearing plate of the adapical surface (5). 
Both of these imperfect plates take part in the interradial suture, 
but taper away towards the adradial suture, the adorally situated one 
being the shorter. As a result, plate 5 is in contact with plate 2 at 
the margin of ambulacrum II. In specimen B (Text-fig. 2) the 
condition in this column is precisely similar, but plate 3 is represented 
by a minute fragment extending for only a millimetre away from the 
interradial suture. 
In column 2a, in specimen A (Text-fig. 1), plate 3 is complete, 
but tapers towards the ambulacrum, and is reduced to a very narrow 
strip at the adradial suture. In specimen B (Text-fig. 2) the 
corresponding plate is imperfect, disappearing at a distance of about 
two millimetres from the ambulacral margin. In column 26 plate 3 
is complete, its height being about half that of the succeeding plates 
in specimen A, and about a quarter that amount in specimen B. 
On: the other side of the test precisely symmetrical and similar 
conditions obtain. In area 5 there is nothing comparable in degree 
with this compression and reduction of the plates. 
From the foregoing description, and particularly from the figures, 
it will be seen that there is developed in the interambulacra of 
Lovenia forbest a structure closely analogous to plate-crushing as 
found in the ambulacra of most Regular and many Irregular Echinoids. 
The only difference in character is that, whereas in ambulacral crushing 
the plates are usually detached from the perradial suture, here they 
survive at the interradial suture and fail to reach the adradial. 
The plate-crushing, in so far as it produces plates homologous with 
demi-plates, is restricted to the two sets of interambulacral columns 
1 Lovén’s notation is used throughout. 
