Echinoid Tests, Diademoida. 111 
case is merely a result, in the case of Diademopsis, of a greater development 
of the interradiad secondary series, and in the case of Hemipedina of a greater 
development of the main series. Point 5 is also rather vague and is so qualified 
as to lose its strict diagnostic value; here again D. acquituberculata con- 
travenes the diagnosis. 
We fall back, then, on point 4, which may be expressed thus: below 
the ambitus each interambulacral column has one or more distinct series of 
secondary tubercles in Diademopsis, but has no such distinct series in Hemz- 
pedina. In a word one may describe Hemipedina as pauci-tuberculate, and 
Diademopsis as multi-tuberculate. Let us consider how far this can be taken 
as a criterion. It is the case that no species of Hemipedina, however pauci- 
tuberculate it may appear, is really devoid of all trace of both an adradial 
and an interradial secondary series; and this appears to be admitted by the 
wording of Lampert’s diagnosis. It is true that the secondary tubercles are 
hard to detect in such species as H. Etheridgei, H. Jardinei, and H. per- 
forata; but they are there, and always originate in the same position with 
regard to the peristome. In H. Waterhousei they are slightly more marked, 
and in 4. Woodwardi the interradiad series, which reaches just above the 
ambitus, was large enough to have been mentioned by Wricnt. In H. tetra- 
gramma the interradiad series is, at the ambitus, not much smaller than the 
median series, but higher up becomes quite inconspicuous; the adradial series 
begins at the peristome as two parallel series of quite small tubercles, of which 
some become larger towards the ambitus while the others are reduced, and 
the series thus formed dies out halfway between the ambitus and the apex. 
This external series was not noticed by Drsor (1858, Synops. p. 430), who 
however said «Par ses rangees de tubercules accessoires dans les aires _inter- 
ambulacraires, cette espece forme en quelque sorte le passage des Hemipedina 
aux Diademopsis». The only reason that I can find why Desor did not put 
Hi. tetragramma into Diademopsis, is that, as may be gathered from his 
contemporary statement under H. microgramma (p. 433), he placed in Diaa- 
emopsis only species with ambulacral tubercles equal in size to interambulacral, 
limiting Heniipedina to species with the tubercles unequal. It is unnecessary 
here to trace the introduction of another interradiad series, as in H. (Pseudo- 
pedina) Smithi, or of yet another one, as in H. (Phymopedina) Bouchardi 
and marchamensis. Enough has been said to render it clear that there is no 
obvious reason why a line of generic division based on the number of tubercle- 
series should be drawn anywhere between H. Etheridgei and H. marcham- 
ensis ; or, if one must be drawn, why it should be between H. Waterhouset 
and H. Woodwardi rather than between H. Woodwardi and H. tetragramma 
— or anywhere else. 
There is, however, a reason for drawing a line, although the reason is 
not obvious and the line must, with our present knowledge, prove doubtful in 
practice. The Diademina in question are descended from Triassic ancestors, 
perhaps from Cidaridae, such as Triadocidaris. The median tubercle-series 
certainly corresponds to the unitubercular series of Cidaridae, and the secondary 
series have arisen by modification of miliaries or of scrobicular tubercles. One 
would therefore expect to find the pauci-tuberculate species preceding the multi- 
