XVII. SPONGIiE. 
6 Spang. 
the edges of an octahedron; and the terminal branches (spines) of the 
hexactin-rays of Lophocalyx philippensis and the terminal hooks of the 
oxyhexactin-rays in Hyalonema apertum and Bathydorus uncifer (n. sp.) 
appear to be situated in the secondary symmetry-planes of the regular 
crystal-system. 
Ijima (14) corrects some statements concerning Hexactinellid spicules. 
The microscleres of Tetractincllids and Clavulina usually form — when 
present at all — an armour on the external surface and are here generally 
much more numerous than in the interior. Only in species of Vioa 
Lendenfeld (21) finds the microscleres confined to the interior and 
absent in the cortex. 
All the megascleres of the Clavulina have, according to Lendenfeld (21) 
an axial thread, which in the tylostyls widens out in the centre of the 
“head” to form a head- nucleus. This nucleus is composed of the same 
substance as the thread and contains granules of various sizes. If the 
“head” is not situated terminally, the axial thread is continued beyond 
the head-nucleus. In the monactine spicules the axial thread terminates 
abruptly some distance from the blunt end. The relative thickness of the 
spicules and the position of the head in the tylostyles (whether terminal 
or not terminal) are subject to considerable variation even within one and 
the same specimen, so that these characters cannot be used for systematic 
purposes to such an extent as was formerly supposed. The Clavuline 
megascleres are on an average 60 times as long as thick. The stoutest 
spicules (1 : 34) were met with in Suberites aaptus\ the most slender 
(1 : 200) in S ' tell ig era nux. 
It is according to Lendenfeld (21) a remarkable fact that the pointed 
end of the radial monactines is in the Clavulina always situated distally, 
whilst the point of tho shaft of the radial triaens of Tetractinellids, from 
which the Clavulina-monactines are supposed phylogenetically to have 
been developed, is always directed inwards. In Clavulina the monactines 
very often project beyond the surface and form a sort of fur. Lendenfeld 
(21) also considers the large spicules forming the halo in Tnchostemma as 
such radial “fur-spicules” (Pelznadeln) and not as tangential spicules. 
The same applies to Tentorium. Really tangential megascleres are found 
in the dermal layer of Suberanthus , Polymastia and other genera. In 
the boring forms of Vioa and Papillella the megascleres are very much 
more numerous in the exposed portions of the sponge, than in the interior. 
Also in Polymastia the spicules are very much more numerous in the cortex 
than in the pulpa. 
The following new spicule-names have been introduced : 
Pseudosterraster (Lendenfeld, 21): thick sausage- shaped, Sterraster- 
like Spicule developed from a Spiraster. — Strongylaster (Lendenfeld, 21): 
Euaster with cylindrical, blunt rays. — Centrotyl (Lendenfeld, 21): a 
small rod with a swelling at or near the centre. — Microrhabd (Lenden- 
feld, 21): a small cylindrical rod blunt at each end. — Discospiraster 
(Schulze, 34): Discohexaster with spirally wound terminal rays. — Aspido- 
plumicom (Schulze, 34) : Plumicom with shield-like basal plates. — 
Rhopalaster (Schulze, 34): large Hexaster with club-liko terminal rays. — 
Subtylostyl (Lendenfeld, 21): Tylostyl, the head of which is barely 
thicker than the shaft. 
d. Epithelia, Cuticula [cf. Titles 20, 21]. 
Lendenfeld (21) finds a well-developed cuticulo on those surfaces of 
boring Vioa- and Papillella- specimens that are in contact with the stonos, 
shells, etc. in which they dwell. This cuticulo is comparable to the basal 
sponginplate of Horny Sponges. In preserved specimens of Chondrilla 
a continuous cuticule was found on tho outer surface covering all tho 
