1895 - 96 .] Mr Masterman on the Structure of Actinotrocha. 133 
not found. The coeca remain in continuity with the wall of the 
gut, and retain their lumen. 
An inspection of fig. 4 will show the close resemblance of these 
coeca to the typical notochordal tissue of the Cliorclata. They 
evidently lend support to the anterior region of the tentacular ring, 
and being thus sustentative organs, developed in response to the 
stimuli from outside themselves, the direction of the vacuolising 
process corresponds to that which would be expected on theoretical 
grounds. 
I think there can be no reasonable doubt that these two hypo- 
blastic vacuolated diverticula have the morphological value of a 
notochord. 
Their origin in the collar region corresponds to that in Balano- 
glossus, the only hemichordate form the development of which is 
known, and the fact that they are paired and lateral instead of 
medium and dorsal need not be a serious impediment to the com- 
parison here instituted; the difference between “paired” and 
unpaired is obviously one of degree rather than of kind in bilater- 
ally symmetrical animals. On theoretical grounds one would 
expect a skeletal organ arising in the collar region to be paired. 
It may be well not to press the immediate homology too far, 
especially as it is questionable if the precise homology of the 
notochord in such forms as Balanoglossus, Tunicata, Amphioxus, 
and Vertebrata can be maintained. It would be preferable to 
make a general definition of the notochordal feature, thus : — A 
skeletal structure formed by the vacuolisation of certain of the 
hypoblastic cells in definite areas determined by the special need 
in each group (in head, collar, or tail) which may ( Vertebrata ) 
or may not ( Hemicliorda , Plioronis ) be separated completely from 
the gut wall, and may (higher Vertebrata , Plioronis) or may not 
(Hemicliorda, Ceplialochorda) be completely replaced in the adult 
by a mesoblastic chondroid skeleton. 
The condition of the notochordal tissue in Actinotrocha is 
probably the most primitive yet described, for two reasons in 
particular ; firstly, because the diverticulum still remains in 
organic continuity with the rest of the hypoblast, and its lumen 
in continuity with that of the gut ; and secondly, because a certain 
area of the ventral wall of the gut just opposite the heavy sac- 
