60 
column, an auxiliary side arm proceeds, formed of numerous minute round 
joints, depressed above and below. These side arms haye, probably, a greater 
number of columnar joints intervening, the further they are distant from the 
pelvis. ‘Their points of insertion are not disposed as in the preceding species, 
directly over each other, but alternately approach the opposite sides of the 
space between the angles, so as to form a double series. One or two cotum= 
nar joints above each point of insertion is grooved, for the purpose of admit- 
ting a closer adhesion of the side arms to the column, when the animal is in 
a state of contraction or rest. 
In the P. Caput Meduse, the round side arms near the pelvis are short, 
and as they lengthen, the distance between the series increases, thereby allow- 
ing free motion without any other provision. Ip P. Briareus, they are very 
numerous and long, but in order that they may divide the space correctly with- 
out infringing on the column or creating confusion, they are lozenge-shaped ; 
therefore when folding around it by packing together with the flattened sides 
towards each other, many will go in a small circumference, and when extended, 
the longest diameter of the lozenge being horizontal, a greater surface for the 
purpose of detention is produced. In the present species they are again round, 
but in more frequent series than in P. Capnt Meduse ; hence their arrangement 
in double rows, becomes necessary, and as they occupy more room, and intrude 
on the angles of the columnar joints, the provision of a groove in those over them, 
as noticed above, becomes necessary to enable them to adhere closely to the 
column when contracted. 
Tn afine specimen of the superior portion of this animal, of an apparently 
nearly full grown size, which I have had opportunity to examine, the First 
COSTAL JoInTS (PL. |.) are acutely angular, and do not overlay so many of the 
subjacent columnar joints. 
The arms and hands resemble those of the former species, but the tentacula 
set off rather differently. The first joint of each tentaculum is not only (as in 
the former species) received into a depression of the finger joint above that 
from which it proceeds, but also occasions a depression in the base of the third 
finger joint above it, and from the upper part of which again another tentaculum 
proceeds, thus producing a scale-like overlaying appearance. As stated be- 
fore, the specimen from which my observation is taken, demonstrates consider- 
able maturity of growth, and hence shows throughout a close structure, and 
