IOO 
3 * 
<x 
r\ 
branches is ± i cm., except the lowest ones; they are inserted at an angle of 45 0 (fig. 96, /), 
distally inclined and straight. At the side of _ 
both these rows there is a tertiary row on the 
middle of the convex side of the colony (figs. 
95 and 96, 2), which row is at equal angles 
with the other rows, so that in the top-part 
of the colony, the three rows are equidistant. 
The branches of this third row are inserted 
Fig. 96. Parantipathes (?) iris- 
tic ha sp. n. Lateral view of the 
branches (schema). 
O' 
ii 
o 
o © OO 
^ r 1 o 
. <5 
Fig. 97. ParantipalhesQ') tristicha sp. n. 
a Spines on the base of the colony; 
b spines on the top of the stem ; c spines 
on an ultimate branch; a, b, c 52 X- 
at right angles with the axis, except the lower ones, which are 
distally inclined at an angle of nearly 6o°. — These branches are 
inserted at the level of the branches of both the other rows, so that 
the number of these branches is double the number of the branches 
in the other rows. The branches are branched themselves (figs. 95 
and 96), so that all the secondary branches of one and the same 
branch lie in a plane, which intersects the plane of the simple 
branches in a line at right angles with the stem. Since nearly all the 
branches of this row are subequal in length (0.5 cm.) and straight, 
and since the secondary branches are very regular, the whole colony 
has a very regular appearance. The secondary branches are grouped 
in two pairs. Ihe lowest pair (figs. 95 and 96, 3) is 0.5 cm. long and these branches are 
straight; the highest pair, inserted at, a 0.75 mm. higher level, is 3 mm. long and curved 
towards the outside of the colony (fig. 95, 4). Often the branches of one pair do not lie in 
the same plane; in this case there is an obtuse angle between them on the distal side. — 
1 here are no polyps. The spines are blunt knobs, 80 p. long, with a mutual distance of 150 p,, 
on the base of the colony (fig. 9 j a), where they leave one side of the axis (the concave side 
of the colony) entirely smooth ; there are three longitudinal rows, with some sparingly distributed 
spines on the transition zone towards the smooth half of the axis-circumference. Where the branched 
part of the axis begins, the spines are more acute, but they are small (fig. 9 y b) and placed on a 
kind of longitudinal crests, which wind in a very steep spiral around the stem; on the top of 
the stem there are 6 longitudinal rows, with a distance of 165 — 300 p between the spines. 
On the branches (fig. 97c) the length of the spines is 225 p. and 130 p. ; the longest ones are 
placed on the distal side of the branches. The spines are more acute and distally inclined, 
while the base of the spines is very elongated. There are 4 longitudinal rows, with a distance 
of 600—750 p. between the spines. The surface of the spines is entirely smooth. The absence 
of the polyps makes it difficult to decide to which genus this species appertains, but in view 
of the mode of branching, I put it in the genus Parantipathes, in expectation of a better 
determination. It is not to be denied that the two rows of simple branches, nearly in the same 
plane, the double-numbered row of branched branches, which are to be compared with the two 
rows of short branches, and the presence of longitudinal crests at the base of the spines, are 
points which this specimen has in common with Eubathypathes qttadribrachiata Brook. The spines 
are entirely unlike the very widely distributed type of spine of Ettbathypathes , but without doubt 
