60 PART I. GENERAL ACCOUNT 



The coelom of the protosoma is considerably larger in Oligobrachia. The 

 tentacular canals spring from the frontal and latero-ventral surfaces from 

 points arranged in a horse-shoe, in accordance with the shape of the base of 

 the tentacular crown. The internal apertures of the coelomoducts open at the 

 postero-lateral corners of the protocoele (PL II, facing p. 68). 



The horseshoe-shaped protocoele of the Thecanephria is arranged 

 parallel to the bases of the tentacles, so that it reflects the shape of the ten- 

 tacular crown. Thus in Lamellisabella, with its tentacular crown forming 

 almost a complete circle, the ventral ends of the coelomic horse-shoe almost 

 meet (PL III, facing p. 92). In Spirobrachia the right wing of the crown is 

 hypertrophied and a considerable fraction of the tentacles are situated on a 

 large corkscrew-shaped lophophore ; the right half of the coelomic horse- 

 shoe is correspondingly elongated inside this outgrowth of the protosoma, 

 following its shape and reaching right to the tip (Fig. 41). 



It is possible that the pericardial sac of the Athecanephria is a derivative of 

 the protocoele, but since the nature of the pericardium is not at all clear we 

 may more usefully discuss it in the chapter on the circulatory system (p. 75) 

 with which it seems to be functionally connected. 



At this point we may conveniently consider the various interpretations of 

 the coelom of the protosoma which have been put forward by earlier authors 

 (Johansson, 1939; Jagersten, 1956). My own interpretation, based on a study 

 of a number of genera of Pogonophora does not agree with either of them 

 (Ivanov, 1955a, 1955b). 



On the basis of his study of Lamellisabella zachsi Johansson suggested 

 that the horseshoe-shaped cavity at the base of the tentacles (he called it the 

 tentacular coelom) communicated with the paired coelomic sacs of the 

 mesosoma. If this were so then the same condition would exist in the Pogono- 

 phora as in the Pterobranchia, in which the tentacles belong to the second or 

 collar segment. My investigations on abundant and varied material emphati- 

 cally repudiate this interpretation. 



Then behind the first coelom Johansson found an unpaired cavity which, 

 he supposed, communicated with the first coelomoducts and lay medially 

 between the front ends of the coelomic sacs of the mesosoma. This thin- 

 walled sac, however, is present in only a proportion of the Pogonophora and 

 is characteristic, though Johansson could not know it, of the Thecanephria 

 alone (Ivanov, 1955c). Histologically its walls clearly show that it is a deep 

 invagination of the ventral wall of the dorsal blood vessel surrounding the 

 excretory portions of the coelomoducts, and it cannot be considered as an 

 independent coelomic cavity. 



