262 Miscellaneous. 
glands are still rudimentary. From this we must conclude that it 
represents simply an apparatus of protection for the genital glands; 
and this conclusion of course applies also to the other Asteriz, of 
which the dorsal skeleton, which is, moreover, so variable, thus loses 
all typical significance. 
The arms in the Brisinge do not complete themselves until the 
appearance of the genital glands. The disk, on the contrary, is 
produced early, and forms around the digestive sac, the prolonga- 
tions of which towards the arms do not appear until later. These 
various facts are absolutely in conformity with the theory of the 
Echinodermata that we proposed in our memoir on “ Animal Colo- 
nies,”—a theory which leads us to see in these creatures, as in the 
Medusez and Ooralliaria, the result of the amalgamation of repro- 
ductive individuals, usually to the number of five, around a central 
nutritive individual. But with age the disk itself undergoes consi- 
derable changes. 
_ In one of the young individuals that we have been able to study 
it is formed by a central piece and nine large, contiguous, triangular 
pieces, nearly coming into contact with the central piece, and with 
it covering the whole surface of the disk, or even extending beyond 
it, and projecting in the intervals of the arms, with which they 
alternate. All these plates, and especially the interbrachial ones, 
bear large movable spines; some smaller plates, alternating with 
them, and evidently of new formation, already exist between them 
and the central piece. 
This constitution of the disk, resembling that of the calyx of a 
Crinoid which possessed nine arms, contrasts singularly with the 
structure of the disk of the adult Brisienge, reduced to an integu- 
ment supported by a ring of calcareous pieces. 
This is how the passage takes place from the one form to the 
other. The growth of the disk is effected by the separation and 
dissociation of the pieces which constituted its central part; the 
interbrachial pieces are thus constantly pushed towards the margin 
of the disk ; at the same time they become reduced more and more, 
place themselves exactly in the angles of the arms, and thus by 
degrees cease to form part of the skeleton of the disk, and finally 
constitute the odontophores. 
Thus the odontophores are the remains of the pieces of the first 
order in the original disk of the Brisinga. The evident identity of 
the plan of organization of the Brisinge and the true Asteriz renders 
the same conclusion probable for the other starfishes. Upon one of 
these plates of the disk of the first order, the madreporic plate is 
always formed; this leads to an unexpected comparison. In the 
Brisinge the displacement of the pieces forming the primitive disk 
stops when these pieces arrive at the outer margin of the buccal 
ring; but if the phenomena of growth have been sufficiently active 
to displace all these pieces to the ventral surface, the madreporic 
plate will pass with them, and we shall have an Asterias, of which 
the ventral surface of the disk will be identical with that presented 
by all the Ophiuri. The development of the Brisinge, which, on 
the one hand, borders on that of the Crinoids, on the other hand, 
