16 
THE VOYAGE OE H.M.S. CHALLENGEE. 
slices fairly thick. I then found that the cladi diverge from the rhabdome at about a 
right angle, and that the spicule is usually orientated, with the rhabdome radial and the 
cladi tangential in position with respect to the sponge ; so that the spicule appears to be 
an orthrotrisene with a remarkably short rhabdome, situated more deeply within the 
ectosome than is usually the case with this spicule. Fortunately I observed in several 
instances, lying close by the side of the fully formed spicule, smaller examples 
representing it at a very early stage. These are always plagiotrisenes in which the 
relative length of the cladi and rhabdome is normal, ^.e., the rhabdome is considerably 
longer than the cladi, e.g., in one instance, the rhabdome measured 0'065 mm., and the 
cladi 0’026 mm. in length. These pass into the adult by a general growth, involving 
both rhabdome and cladi, but the latter increase at a far greater rate than the former ; so 
that in the fully formed spicule they attain a considerably greater length. The young 
spicules probably arise from a scleroblast budded off by that of the adult, with which they 
are associated, towards the completion of its growth. 
The supposed calthrops which makes such a startling appearance in Tetilla is thus 
nothing more than an interesting modification of an ordinary trisene ; and not a 
persistent tetraxon. In other sponges, such as Pcecillastra {Normania), in which an 
apparent calthrops occurs in the choanosome, it is equally possible that it is not derived 
from a microcalthrops, but from a trisene ; and furthermore, as the calthrops of the 
Pachastrellidse is almost certainly descended from that of Pcecillastra, a trisene origin 
may likewise possibly be ascribed to it. 
In some cases it appeared to me that when oxeas occurred lying tangentially in the 
ectosome, the young trisenes were displaced from their normal position, the rhabdome 
having a tangential position ; if so we have a still further approach to the conditions 
under which the calthrops occurs in Pcecillastra, and further, the apparent short rhabdome 
of the adult orthotrisenes in the Tetilla may in some cases be really a cladus turned out 
of the tangential into the radial position by a rotation of the whole spicule. These 
cases may, however, possibly be explicable as artificially produced in the process of 
cutting. 
The microscleres, both microxeas and sigmaspires, are scattered generally throughout 
the sponge. 
Tetilla sp. 
Sponge small, ovate, free, surface pilose, spicules projecting obliquely from the 
surface, and chiefly towards the base ; ectosome a fibro- vesicular collenchyma about 0’02 
to 0’047 mm. thick. Choanosome with a richly-developed coUenchymatous mesoderm, 
which forms a somewhat thick layer about the walls of the chief canals, and is 
