53 
^Qdinal coat in retaining the tubes to their proper size. 
1 * le Erst or inner tunic appears to be of a serous character, 
?d to resemble the lining membrane of arteries ; but has this 
j 'sracter in addition, that it is the sole seat of reproduction 
11 the order. 
^he spicula are dispersed irregularly through the sub- 
a Qce of the polype mass ; near the surface far more 
P^ringjy th an towards the centre, where they aro more 
Qsely aggregated. They are very irregular in shape, but 
'f 1 all possess a character in common. In many cases they 
osely resemble the bodies described in the Gorgonia, and 
j? Te a similar formation. Some are K-shaped in various 
^figurations of its parts. By maceration they readily 
odergo decomposition, and are then found to have the same 
c e jmlar composition as the bodies in the Gorgonia. The 
’mles are very closely and intimately connected, yet the 
■j, Section is readily broken by maceration or weak acids, 
plough they are thus diffused through the fleshy gelatine 
l *h but very little organic connection, yet from their com- 
j^'tion they appear to be the result of some vital action. 
. r ‘ Johnston considers them as the lirst appearance of a 
t ^'ypidom or axis, he says “ that it would not be difficult 
0 kace them through all their gradations to the horny 
^ x *hle axis of the Gorgonia.” Such, from the observations 
^ fle on the pith and investing membrane of the axis, and 
t(, Membrane of the crust in the Gorgonia, may possibly be 
ti) e Rase ’ fi ut I ratber hesitate to consider them similar to 
'aphides of plants, or the spicula of sponges, 
tjj .production occurs id this order only in one way, by 
'oternal generation of geinmules or ova. There is no 
ljj e llc ular set of organs appropriated to this function as in 
fi'gher animals; nor is there, as in the Hydroidae, a 
::les. The function 
p 6r . u >gher animals; nor is there, as in the Hydroidae, a 
dp '°fiical dovelopement of ovarian vesicl 
ofi^rs to he of a diffusive kind, and is common to all parts 
tjj . ® lining membrane of the tubes, and according to au- 
Jj'hes, to the walls of the abdominal cavity. 
,nan ner in which the ova are developed is best 
Caiil rve d in a transverse section. In a natural state the 
<le v °. re °I the tube is circular, but when ova are about to bo 
file °P e fi> a segment of the circle is slightly bulged towards 
Cet| tre. As developement progresses, the bulging in- 
( l a itp eS an< ^ becomes more and more circular, till it becomes 
is Slobular and lies on the tube as a tangent. The ovum 
is g R 1 to the lining membrane by an umbilical cord, which 
clear *y to fi e K ot* ce d '• this gets more and more at- 
lef't et f a nd is finally absorbed or ruptured and the ovum is 
d liberty in the tube. From the very earliest period at 
