Echinoid Tests, Diademoida, 119 
diameter of scrobicule averages 33.7 (extremes 25 and 42), diameter of boss averages 
16 (extremes 11 and 19), diameter of mamelon averages 7 (extremes 5 and 9). 
Extra-scrobicular surface covered with distinct regular miliaries, about 4, or fewer, 
to the millimetre (linear). Adradial margin strongly bevelled, with about 4 denticles 
to each plate; transverse margins each with a strong rebate. 
Material. — This consists of 23 interambulacrals, namely nine from Jeru- 
zsalemhegy (lettered a—j), two from Cutting I on the Veszprém-Jutas Railway 
(lettered k, 7, the latter labelled as from bed e), ten from the quarry near that 
cutting (lettered m—v), one fragment from beds a—b of Cutting IV (lettered w), 
and one fragment from Cserhat (lettered 7). The last two, however, are very doubtful, 
and will receive a separate description. The rest are all of Raiblian age. 
The holotype is specimen m. (Figs. 205—207). 
Description of Specimens a—v, — These plates are markedly distin- 
guished from all other isolated plates found in Bakony by their great relative width 
(see table of measurements), which may exceed five times their height. The greatest 
absolute width measured is 9 mm. (in a); but in b (fig. 201), where the missing 
adradial tract cannot well have been less than 3 mm. wide, the total width must have 
been over 10 mm. Specimen k was probably wider still. 
The next conspicuous feature is the irregular curvature of the plates, which is of 
two kinds. The coronal plates of all Echinoids necessarily share in the normal curvature 
of the test. Thus ina perfectly spherical test, did such exist, the transverse and merid- 
ional sections of all plates would both be an arc of a circle. When either diameter is 
short, as is the case with the meridional diameter of the plates before us, this 
curvature may be neglected. In a long diameter, like the transverse diameter of these 
plates, the curvature is often obvious, so that the width of the plate measured along 
its outer surface is appreciably greater than the width measured directly along the 
chord of the arc. It is the latter measurement that is given in the table. Now in many 
of these plates, especially the wider ones, the curve is not an arc but is greater on the 
adambulacral side of the tubercle (figs. 205, 209). As a consequence of this, several of 
the plates have been broken across by pressure while in the rock, and the adradial portion 
has either been lost, as in b and /, or has, though very rarely, been recemented by calcite 
to the remaining portion, as ind. It follows that the interambulacrum as a whole was, at 
least in its ambital region, flat in the interradial tract and then sharply sloping towards the 
adradial margin. If one attempts the reconstruction of specimen b, one arrives at an 
equatorial diameter of not less than 32 mm,; how much greater depends on the width 
of the ambulacra, but 40 mm. would not be unreasonable. One may also infer that the 
ambitus was pentagonal with rounded angles, i. e. sub-decagonal. Assuming that the 
interambulacrals of such a test were not higher than specimen b, there must have been 
about 24 in a column. 
The second curvature to which reference has been made is seen in specimens 
a, c, d, e, f, g,h, k, l, m,n, 0, y, that is in thirteen, while it probably existed in some 
of the incomplete plates as well, so that about three-quarters of the plates may be 
supposed to have had it. This curvature is manifested in many ways. In its simplest 
and commonest form, as in a, c, h, k, m (fig. 207), the plate is so bent that its adapical 
margin is concave, and its adoral margin convex; the greatest concavity is near the 
tubercle. Another simple form is presented by d and / (fig. 203), in which the adapical 
margin is convex and the adoral concave, the concavity being as before near the 
