134 Mr. G. Bidder. On the Flask-shaped Ectoderm [June 16, 



of the ccelom, divided into right and left halves by median septa, 

 each half communicating with the exterior by means of its own 

 collar canal; on the posterior face of this cavity is an ectodermal 

 thickening, which corresponds in position with the nerve-plate of 

 Cephalodiscus and the nerve-tube of Balanoglossus. The collar cavi- 

 ties are continued upwards into the tentacles, and surround the 

 mouth. From the pharynx a short divert! culum is given off upwards, 

 which is continuous with a rod-like structure, apparently half 

 cellular, half gelatinoid, which lies in the line along which the 

 median posterior septum of the collar meets the nearly vertical 

 septum between proboscis and collar cavities. It thus corresponds in 

 origin, structure, and position with the notochord of Cephalodiscus. 



The trunk contains the greater part of the alimentary canal. Its 

 body cavity, as in the other Hemichordata, appears to be completely 

 shut off from the paired cavity of the collar. The only part of the 

 intestine calling for remark is a short semicircular diverticulum, 

 which occurs also in Cephalodiscus. 



The points in which Rhabdopleura differs from both the other 

 Hemichordata are purely negative, viz., the absence of a proboscis 

 pore or pores, and the absence of gill-slits ; the points of agreement 

 are so striking that it is impossible to separate the three organisms. 



A more fully illustrated paper on the subject will shortly be 

 published. 



DESCRIPTION OF FIGURES. 



FIG. 1. Longitudinal section, taken slightly to one side of the middle line, so- as to 

 avoid the median septa of the collar region. The dotted line e marks the 

 ventral limit of the collar. 

 FIG. 2. Transverse section along the line c d in fig. 1, through proboscis stalk 



and upper part of collar, and cutting the base of a tentacle of one side. 

 FIG. 3. Transverse section along the line a b in fig. 1, showing the external 



opening of one, the internal opening of the other, collar canal. 

 In all three figures the mesoderm is represented by a thick line, the ectoderm 

 and endoderm by thin lines. The gelatinoid part of the notochord is marked by 

 Cross-hatching ; its cellular part has not been distinguished from ectoderm or 

 endoderm; b.c. 1 , prse-oral ccelom, or proboscis cavity; J.c. 2 , ccelom of the collar divided 

 by the median septa into right and left halves ; I.e. 3 , ccelom of the trunk ; c.c., canals 

 leading from the collar cavities to the exterior. 



XII. " On the Flask-shaped Ectoderm and Spongoblasts in one 

 of the Keratosa." By GEORGE BIDDER. Communicated 

 by ADAM SEDGWICK, M.A., F.R.S. Received June 15, 1892. 



In my " Note on Excretion in Sponges," published by the Society 

 in the ' Proceedings ' of this year, I said : " Both from my own ob- 

 servations on an Aplyntta (?) V V V '. and from a study of 



