1897-98.] Dr Masterman on Ar chimeric Segmentation. 297 
metameric (posterior enterocoeles), then the left mesocoele (hydro- 
coele), and, lastly, the protocoele (anterior enterocoele). 
In the other Echinodermata the open part of the blastopore 
becomes the anus or opens at the posterior end of the body till after 
the mesoderm has commenced to arise (anus of bilateral larva). 
Thus in these, in accordance with the law stated above, the archi- 
coeles arise from the distal (front) end of the archenteron, and give 
rise backwards to metacoeles, mesocoeles, and protocoele. The 
main irregularity consists in the fact that the right mesocoele (right 
hydrocoele) may be absent or vestigial, whilst there are in the 
various groups variations with respect to' the relative time of division 
of, e.g ., the metacoele into its two portions, and of the metacoele 
from the other archicoeles, and, in addition, the formation of the 
protocoele has not been so clearly followed in some as in others. 
It is usual for a pair of vaso-peritoneal vesicles to be formed, 
which each divides up into mesocoele (hydrocoele) and metacoele 
(enterocoele).* 
In Synapta f the metacoele is at first unpaired, and only later 
divides into two, but, in addition, the assumption by the archenteron 
of the secondary axis on the one hand, and the formation of the 
protocoele pore on the other hand, are both so accelerated that the 
archenteron at one time appears to have two apertures to the exterior, 
the future anus, and the water-pore. 
It can scarcely be maintained that the hydrocoeles and entero- 
coeles of the Echinodermata are not respectively homologous 
throughout the group, and their variations in development appear 
to be inexplicable except as modifications of a primitive type of 
archicoele development, such as suggested here. 
The archimeric segmentation may, therefore, be described in its 
ontogenetic characters as consisting primarily of four archicoeles, 
sometimes secondarily divided into five or six, which arise prim- 
itively as four archenteric diverticula, but which may by a simple 
modification arise as a pair of diverticula, which later divide up 
into the archicoeles themselves. 
* In certain pelagic larvae, e.g. Bipinnaria , the functional oesophagus is 
early formed although not actually part of the blastopore. In these the two 
lateral coelomes move forwards later to form the protocoele. 
t Salenka, Jena. Zeits ., xxii., 1888. 
YOL. XXII. 26/9/98. 
U 
