40 



CHIM-^ROID FISHES AND fHElR DEVELOPMENT. 



terminating near the cardiac region. To add, therefore, a couple of inches to the 

 length of the capsule involves a serious problem in the matter of space. This has 

 been solved as follows : The capsule is gradually released, so that it comes to project 

 from the fish's body ; at first the apex appears at the mouth of the oviduct, then 

 the trunk-sheath, then the tail-sheath (fig. 1 2). If at this time the egg is dislodged, 

 an abortive terminal results, as in plate 11, fig. 10. If it remains, the process in the 

 formation of the capsular filament and terminal organ progresses as seen in plate i, 

 hg. 4. The capsule now protrudes as far as the base of the filament, and with this 



Figs. 24-26. — Egg-capsules of Chimaera colliei, partly opened, so as lo show egg and young. Natural size. 

 24. Late blaslula (about 9 days). 25. Early gasuula (about 19 days). 26. Late gaslrula (about 24 days). 



protrusion occur many changes in the oviduct {cf. plate ii, fig 4, and plate 11, fig. 6), 

 e. g. , its diameter becomes greatly constricted and its dorsoventral characters and 

 web creases obsolescent ; it also loses its earlier differentiation into trunk-sheath 

 and tail-sheath forming portions, and its vaginal region is extended headward, the 

 remaining part of the oviduct becoming correspondingly reduced. All these changes 

 are to the obvious end of molding the long capsular filament and the bulbous terminal 

 organ (plate i, fig. 4, f. /., and c. 0.). For the formation of the latter serves a special 

 region of the capsular gland, /. r., its hindmost zone, a portion pinkish in color, 

 provided with the recesses into which the terminal organ has been traced.* 



* Finally, a note may be given as to the probable mode of attachment of the capsule (cf. pp. 26, 27.) It is evident 

 that the oviduct can be greatly evaginated at the time the egg is deposited (plate i, fig. 2\ and with this phenomenon 

 is evidently connected the forward extension of the cervix [cf. plater, fig. 4, and plate 11, fig. 6). We may thus 

 conclude that at the cervix, then, the bulbous organ of attachment can be held by the fish even when the oviducts have 

 been greatly everted ; and it would naturally be by such elongated, even finger-like, processes that the fish could press 

 the filamentous ends of the capsules against the object of attachment, c. g., a rock fragment, and thus secure their 

 adhesion. 



