304 Transactions. — Zoology. 



covered by cuticle, which is thicker on the external convex surface 

 than on the apex or inner face ; below this are several " bicellu- 

 lar glands," which project downwards from the epidermis into 

 the dermis or fibrillated connective tissue (Plate XVI., fig. 12). 

 The latter is excavated by a great space which at the base of 

 the tubercle is in free communication, by means of a small 

 aperture, with the general body cavity. In this space is a mass 

 oi spherical cells, each with a refringent body and a deeply 

 staining nucleus. At first I took this for a specialised group of 

 connective tissue-cells, serving as a skeletal axis to the tubercle ; 

 but examination of serial sections shows that this mass of cells 

 is continuous with masses of granular ccelomic cells adhering to 

 the body- wall, and in some cases the transition between the two 

 appearances — refringent and granular — can be made out. Unfor- 

 tunately, the material is not sufficiently well preserved to enable 

 me to describe in greater detail this structure ; but it appears to 

 serve as a means of erecting these tubercles, for when the strong 

 muscles of the body-wall contract this ccelomic fluid will be 

 driven into these subdermal spaces. 



On the body-wall itself, as opposed to the wall of the intro- 

 vert, there are no tubercles : the usual lymph-spaces alternate 

 with groups of bicellular glands, as has been described by previous 

 writers on the histology of Sipunculus. 



As the tentacular crown was invaginated, it is not altogether 

 easy to reconstruct the appearance which it has when fully 

 extended. On slitting open the oesophagus, five ridges, covered 

 with tentacles, are seen (Plate XV., fig. 5) : of these one is dorsal, 

 two are lateral, and two late ro- ventral ; the dorsal ridge is the 

 largest, and bears more numerous tentacles. The tentacles them- 

 selves are flattened, membranous, and truncated distally — quite 

 unlike the more or less cylindrical tentacles of Phascolosoma and 

 other genera ; and it has been shown by various authors that 

 in the genus Sipunculus the tentacles are really the jagged edge 

 of a membrane : various species show different degrees in which 

 the margin of the circumoral membrane is cut into, so that in 

 some the " tentacles " are short and in others longer. In the 

 present species the membrane is almost entirely " frayed " out 

 into these tentacles. By artificially pushing the oesophagus and 

 crown upwards we may cause the crown to assume, to some 

 degree, the condition of eversion ; and from a study of such a 

 preparation, and of the ridges themselves, I venture to " recon- 

 struct " this tentacular crown (Plate XV., fig. 4), which is very 

 unlike that of the species of Sipunculus usually figured ; but 

 from the brief diagnosis of " Phallosoma," Levinsen, given by 

 Yves Delage and Herouard,* there seems to be some resemblance 



* Traite de Zool. Concrete, v.. p. 21. 



