22 THE VOYAGE OF H.M.S. CHALLENGER. 



tissue still supports its margin all round, and especially posteriorly toward its termina- 

 tion. The secerning tissue is thus largely developed at its outer and inner regions. 



The foregoing canals were noticed by Kowalewsky, and their function correctly 

 interpreted. Nothing, however, has been seen resembling the long moniliform canal 

 described by KoUiker,' which ends blindly at the anterior region. Dyster states that 

 the ova were discharged alternately by the two ducts. 



Caldwell ' describes the posterior pair of mesoderm ic diverticula opening in the 

 larva in the middle line and communicating with the exterior in each case by a 

 small pore, which he thinks probably persists as the nephridial opening of its side. 

 The excretory cells lie in the blood-space of the splanchnopleure, and not in the body- 

 cavity. He also states that the external openings of the nephridia are parts of the 

 blastopore. The nephridia attach themselves entirely to the posterior mesoderm (for 

 the latter is divided into an anterior and a posterior part). " The division of the 

 blastopore caused the division of the mesoderm, and results, amongst other things, in 

 metameric segmentation." 



"O 



Reproductive Organs and Development. 



The ova are developed in the posterior region of the body in racemose masses 

 attached to the vessels round the main division of the alimentary canal. While one 

 side of the body is thus occupied, the other has a bulky lobulated (in section somewhat 

 areolated) granular tissue, which represents the male elements (PL II. fig. 2, ovr and t). 

 Kowalewsky mentions that the latter occur in a different region from the former, but 

 such was not indicated in the present specimens. Dyster correctly observed both ova 

 and spermatozoa " beneath " the stomach. The ova are thus shed into the body-cavity, 

 fertilized therein, undergoing, according to Kowalewsky, the earlier stages of their 

 development there, and then pass outward by the nephridia to lodge in the branchial 

 whorls. 



The ova lie freely amongst the tentacles ; and there was no indication in the 

 preparations that they adhere together. Certain specimens would indicate that the 

 older larvae occupy .the outer whorls of the tentacles, the earlier forms mainly occurring 

 in the centre (PI. III. fig. 1). This condition, however, may simply indicate that the 

 action of the cilia in the free larvas had caused a change of position. 



The history of the development of the young Phoronis is most interesting. The 

 larval form was first found in the sea off Heligoland by Johannes Miiller,^ in one of his 

 memorable excursions ; and he gave it the name of Actinotrocha, from its ciliated arms ; 

 but he was unaware of its relationships — indeed, he thought it might be a larval 

 Mollusk. A more detailed account of its structure was published the following year by 



' Op. ctt., p. 12. - Quart. Jonrv. Micr. ScL, vol. xxv. p. 19, 1885. 



3 Mulkrs Archivf. Anat. u. Phijswl., p. 101, Taf. v. figs. 1, 2, 1840. 



