MATERIALS FOR.A MONOGRAPH OF THE ASCONS. 343 
generalisation :—JIn all trivadiate systems of calcareous 
sponges, whatever their form, there is one ray, the morpho- 
logical axis of which lies entirely in a principal plane of 
crystalline symmetry, that is to say, a plane which includes 
the optic axis, and which also bisects the angle between the 
other two rays. Moreover, the ray which defines this plane 
of symmetry is the posterior ray, that is to say, the ray which, 
in the primitive Olynthus, points in the opposite direction to 
the oscular aperture, and therefore occupies, primitively at 
least, a definite position in the sponge-body. ‘Thus, in the 
primitive orientation of the triradiate systems, such as is 
found to persist in the oscular tubes of Ascons, probably also 
in the oscular rim of any calcareous sponge, derived directly 
from the arrangement presented invariably by the embryonic 
Olynthus-form, the plane of crystalline symmetry defined by 
the posterior ray would also halve more or less accurately the 
entire sponge-body, since it would pass through the morpho- 
logical axis of the body. The triradiate systems of the 
Calcarea exhibit a plan of crystalline symmetry 
which is in relation, not only with the morphologi- 
cal symmetry of the spicules themselves, but also 
with that of the sponge-organism. I think it may be 
reasonably inferred that this striking fact must be explained 
by physical peculiarities of the material, since it is a character 
of the spicules which cannot possibly have any biological or 
functional significance. 
Maas (1904 [2], etc.) has advanced, as a proof that the 
form of the spicules is determined entirely by the organism 
in which they develop, the observation that in calcareous 
sponges grown in water deprived of CaCO,, the sclerites, 
though not containing calcite, still show the triradiate form, 
Ido not gather, however, from Maas’s memoirs, that these 
organic triradiate spicules have a regular symmetry, but 
simply that they consist of three rays joined together. As I 
have pointed out above, the operation of a hereditary tendency 
is an adequate explanation up to this point, and, in my 
opinion, no further. 
