310 
THE VOYAGE OF H.M.S. CHALLENGER. 
the sides of the axis, spines slender, conical, terminally tylote ; axis about 0’006 mui, 
long, spines about O’OOS mm. long, total length of amphiaster 0’02 to 0‘024 mm. 
Colour. — Yellowish- white in the dried state. 
Habitat. — Station 192, off the Ki Islands, south of Papua, September 26, 1874; 
lat. 5° 49' 15" S., long. 132° 14' 15" E. ; depth, 140 fathoms ; bottom, blue mud. 
Remarks. — This little sponge, only about 2 mm. in diameter, was found incrustiug 
the fragment of limestone to which Corallistes thomasi is attached. It might easily 
have been mistaken for a Foraminifera, but on mounting in balsam its true character 
at once appeared. 
The monocrepidial character of the desma is obvious enough at all stages of its 
development from the earliest to the latest, but the nature of the disc does not at first 
appear so clear. After a searching examination I can find no trace of a division of the 
rhabdal axis on or after entering the disc, there is no trifid nor any other subdivision, 
simply an abrupt and sudden termination. No trace of subdivision was revealed by 
treatment with hydrofluoric acid. The disc therefore is uniaxial, like that of Neovelta, 
from which, however, it difiers not only by the tuberculation of its surface and the denti- 
culation of its margin, but also by the almost constant presence of a stalk. Only in a 
single instance was the rhabdome not observed, and even then its place was indicated 
by a slight irregularity of the surface, and no tangentially lying crepidial axis was 
observed. In Neopelta the stalk on the contrary is more often absent than not. In 
addition to this the stalk is usually at right angles to the disc in Callipelta, while in 
Neopelta when present it is usually obliquely inclined. The discostrongyle of Callipelta 
appears therefore to form a connecting link between the discotrisene and the disc, it difiers 
from the latter and resembles the former in the almost constant presence of a stalk set at 
right angles to the disc ; it differs from the former and resembles the latter in the 
constant absence of cladal axes. The question then arises, so difficult to answer in all 
similar cases, not only in the Lithistidae, as to whether the disc is descended from the 
discotriaene, or vice versd, or whether both have a separate or a common third origin. 
The last alternative may I think be safely dismissed. With regard to a separate origin 
there is more to be said. The discotriaene can be traced into the dichotriaene, and we 
may safely assert that one has arisen from the other. The dichotriaene is the most 
widely diffused, and departs the least from the usual spicular type, so that the order 
of development has in general been tacitly assumed as from the dichotriaene to the 
discotriaene, and not vice versd. This conclusion becomes strengthened when the 
identity in character of the dichotriaene in the Lithistidae with that of the Choristidae 
is considered. Zittel, indeed, speaks of the resemblance between these spicules in the 
two groups as a deceptive or false resemblance ; this is mere transcendentalism, false 
conclusions may be drawn from the resemblance no doubt, but the resemblance itself 
