SKELETAL STRUCTURES. 81 
In Plinthosella, Zitt., and Phymaplectia, Hinde, the spicules unite together by the 
interlocking of the tubercles on their lateral surfaces (Fig. 5, e) as well as 
terminally, without forming prominent nodes, thus showing an approximation to 
the mode of union in the Rhizomorina family. 
In the Megamorina family there are two distinct modes in which the skeletal- 
spicules unite together, and both may take place in the same Sponge. In one, the 
spicules and their branches are twisted round each other almost in the same manner 
as the strands of whipcord. The spicules may be merely twisted at their ends, or 
throughout their length; and owing to the lateral and terminal notches (Fig. 4, 4) 
they are very closely fitted together, so as to produce a fibrous meshwork. This 
mode of union is typically shown in Carterella, Zitt., and Isorhaphinia, Zitt. The 
second mode is exemplified in spicules, in which the branches terminate in flattened, 
concave, spoon-shaped expansions (Fig. 4, ¢), which closely and evenly fit, and 
clasp the surfaces of adjoining spicules, and thus form a meshwork of open irre- 
gular interspaces. This kind of union is well shown in Doryderma, Zitt. (Fig. 
4, f), Heterostinia, Zitt., and in the recent Lyidium, Os. Schmidt. 
In the Anomocladina family the union of the skeletal spicules takes place, in 
some respects, in the same manner as in the Megamorina family described above, 
that is to say, the terminal ends of the spicular rays are similarly furnished with 
expanded surfaces (Fig. 5, a, b), which are firmly attached to the central nodes, 
and occasionally to the rays of adjoining spicules, thus forming a mesh apparently 
composed of star-shaped bodies, whose rays are all united together (Fig. 4, /). 
These stellate bodies of the connected skeleton are thus of a compound character, 
for in each there is the node and the rays proper to it, forming the elementary 
spicule, and also other rays belonging to adjoining spicules whose terminal expan- 
sions are firmly attached to the node. This union is usually so close that ib is not 
practicable to determine in the connected skeleton the rays proper to the node 
from those which are merely adpressed to it (Fig. 4, 4). In some instances, in the 
skeleton of Astylospongia, the spicular rays converge to a point in which no central 
node is present for them to clasp, but they terminate in precisely the same way as 
if the node were in the proper position, and are thus grouped round a sub-spherical 
cavity. In some examples of Cylindrophyma, the rays project from the node at 
right angles to each other, and the connected skeleton has quadrate or subquadrate 
interspaces, singularly resembling the mesh of hexactinellid Sponges, for which 
it has sometimes been mistaken. Typical examples of the skeleton of Anomo- 
cladina Sponges are shown in Oylindrophyma, Zitt., Astylospongia, Ferd. Roem. 
(Fig. 4, 1), and the recent Vetulina, Os. Schmidt. 
The skeletal-spicules of the Rhizomorina family are united together by the 
close adpression of the minute facets terminating the numerous branches and 
spinous processes of the spicules, to the main axis and branches of adjoining 
L 
