108 
Triassic Echinoderms of Bakony. 
weight can be attached to the more median position of the main _tubercle- 
series on the interambulacral column in Orthopsis, especially when one notices 
that in the genotype it becomes nearer the adradius above the ambitus, and 
when one remembers that it is no less median in some admitted Diademopsis, 
e. g. D. Bowerbanki (Wr.) and D. aequituberculata Lampert. It appears, 
however, that all recognised species of Diademopsis show more arcuation of 
the pore-pairs around the main ambulacral tubercles, and although we must 
suppose the existence of a Diademopsis ancestor with its pore-pairs in a 
straight line, there is at present no evidence of any Liassic or Bajocian species 
reverting to that structure. 
Let us then consider the hypothesis that Orthopsis is a later reversion, 
and let us look for any characters forbidding us to regard it as primitive. 
Such characters are, I believe, to be found in the genotype. Both Cenomanian 
and Senonian specimens in the British Museum show the following: — In the 
interambulacrum the convergence of the main tubercle-series as they near 
the peristome is not nearly so great as in the undoubted species of Diademopsis 
and Hemipedina that I have examined. In those genera the tubercles seem 
almost to meet on the interradius, but in O. granularis they remain side by 
side, with no attempt at inosculation. This seems to indicate a greater extension 
of the peristome into the interambulacrum, or in other words the resorption 
of more interambulacral plates: it is a sign of accelerated development. 
Connected with that character is the wide and straight interambulacral lip, 
and the rapidity with which the tubercles assume prominence. Below the 
ambitus the scrobicules of these tubercles are subquadrate, separated only by a 
line of granules or a ridge, as in Pygaster: this also is an advanced character. 
A striking feature in this species is the adoral position of the main tubercle on 
each plate and the adapical position of the external and internal tubercles. Thus, 
the scrobicular circle of the main tubercle invades the plate below, while those 
of the secondary tubercles invade the plate above. In this way all the tubercles 
alternate, and so in a column of less relative width than in Diademopsis the 
tubercles are no fewer in number and no smaller in size, for in Diademopsis 
the tubercles are all on the same level. Obviously the alternation is a later 
stage of development. A slight tendency towards it is seen in Hemipedina 
(Phymopedina) Bouchardi. In the ambulacrum similar scrobicules and a similar 
alternation of an internal series at the ambitus are observed. At and below 
the ambitus, each main tubercle, starting apparently from the middle primary 
of a triad, overlaps the two other primaries equally; but above the ambitus, 
it is smaller, shifted adorally, and occupies only two primaries, while the 
adapical primary of the triad bears one or two of the scrobicular tubercles 
enlarged. This then seems not a simple reversion to a primitive condition, but 
a step in the direction of increasing the number of tubercles in the main 
series. Below the ambitus, where the ordinary Diademine relation of the tub- 
ercles is maintained, there is still some faint trace of arcuation of the pore- 
pairs; but above the ambitus this has disappeared with the change in tuberc- 
ulation. At the peristome the pore-pairs are much crowded and pressed out 
to both sides in a manner which I have not observed in ordinary species of 
Diademopsis and Hemipedina. Hemipedina Saemanni Wr. of Rauracian age 
