EEPOET ON THE TETEACTINELLIDA. 
Ixix 
rhabdus occur as accidental varieties in sponges of wliicli the rhabdus is the characteristic 
spicule [Epallax callocyatlius, Poecillastra schulzii). 
The origin of the trisene is a subject on which opinions differ ; on d priori grounds 
it may vdth equal probability be derived from the calthrops by a lengthening of one of 
the actines or from a rhabdus by terminal cladosis of one of the ends, naturally of the 
centrifugal end ; the former hypothesis has in its favour an appearance of simplicity ; 
the latter is supported by a great deal of direct evidence, both ontogenetic and 
morphological. 
In the simplest trisene-bearing sponges, viz., the Tetillidse, which are not much 
more complex than the Placinidse, the anatrisene at its earliest recognisable appearance 
is an oxytylote, with a very feebly expressed tylus ; the tylus increases in size, and a 
single cladus appears ; and in the young sponge, while still within the body of the 
parent, this single cladus acquires a considerable size before the remaining cladi 
appear. 
The strength of the argument to be drawn from this fact would be greater, were it 
not for the possibility that it may be merely an adaptive character, standing in relation 
to the advanced stage of development reached by the young sponge before it quits the 
body of the parent ; and this seems the more likely as the body of parent sponge is 
invested with a strong fibrous and spicular cortex, through which the young sponge, which 
may be as much as 1‘75 mm. in diameter, must make its way to the exterior, in 
what manner we know not {vide p. 39). The argument may be supposed to be 
further weakened by the fact that in the- adult sponge all three cladi appear to develop 
simultaneously. While the appearance of one cladus earlier than the rest may not 
be a matter of any special significance, the origin of the trisene from an oxytylote 
is a very strong argument in favour of its rhabdal origin. The protrisene also 
develops from an oxytylote, but all three actines appear simultaneously, even in the 
young sponge ; they are at first excessively minute, the merest spines, but subsequently 
attain considerable size {vide p. 13, PI. V. figs. 10, 15). 
Whatever doubt may be felt as to the value to be attached to the appearance of only a 
single cladus in the young Tetillid, the evidence in the case of Thenea, though of a different 
nature, seems to admit of but little dispute ; in very young examples of this sponge, 
which do not develop within the parent, the anatrisene of the radical filaments commences 
as an oxystrongyle, subsequently numerous small spines appear at the strongylate end, 
and as these are absent from the adult spicules we may conclude that three of them by 
over-development become the adult cladi, the rest being suppressed ; in some instances the 
strongyle presents in the young sponge only one or two strong spines, which are evidently 
developing cladi, and thus we have a variability in the number of the cladi of the early 
form of trisene, such as we might expect if it were, as we suppose, derived from a rhabdus 
with an at first spined and subsequently cladose termination. 
