CALMA GLAUCOIDES 451 



opaque in life and more slender than the rest. In it single eggs 

 or small groups of eggs receive a separate investment of 

 a substance giving the chemical tests for mucin. This is at first 

 laid on in a fluid condition, while the later layers are dense and 

 firm. Shell-gland would therefore perhaps be the appropriate 

 name for this portion. In the rest of the oviducal passage 

 (n.g.) which is pellucid in the living animal, the eggs in their 

 shells are enclosed in the substance of the nidamental ribbon, 

 also a mucin. The outer layer of this, like that of the shells, 

 is firmer and denser than the rest. Hitherto we have considered 

 only those changes that affect the dorsal wall of the original 

 oviducal sac, and result in the formation of a twisted egg- 

 passage (see also Text-fig. 3, e.jp.). The ventral wall (Text-fig. 3, 

 v.w.) remains flat, thin, and non-glandular. Distally, near the 

 atrium, the flask-shaped receptaculum seminis (Text-fig. 2, 

 r.s.) is formed as an evagination of it. From the atrium it 

 extends back as the floor of a broad, shallow chamber (sp.p.) 

 which narrows as it becomes continuous behind with the initial 

 part of the female duct at the point of departure of the func- 

 tional oviduct {y). The impression so far conveyed is that the 

 original sac-like female duct has been divided into two passages 

 by a process resembling the pinching off of the vertebrate 

 semicircular canals, namely, a long coiled dorsal one, ciliated 

 and glandular for outgoing eggs, and a short thin-walled ventral 

 one for incoming sperms from the receptaculum, which is 

 neither ciliated nor glandular. Such a complete female diauly 

 is, however, not strictly true. Text-fig. 3 of a section through 

 the nidamental region in the plane lr in Text-fig. 2 shows that 

 three (1.2.3) out of the four oviducal loops thus cut across 

 are incompletely separated from the vaginal chamber below, 

 while the fourth or proximal loop (4) is a discrete tube. Thus 

 for a considerable length of ribbon-forming oviduct a facultative 

 but not a structural diauly is present. The tube of the shell- 

 gland is, however, completely separated except at its commence- 

 ment, as mentioned above. This is essential since it deals 

 with loose eggs, or with eggs receiving a fluid envelope. The 

 continuity at /y of the undivided oviduct with the vaginal 



