84 
GEOEGE E. NICHOLES. 
backward into the foui’th ventricle (Petr om yzon). Nothing 
of this sort could happen if the fibre were buried in the brain- 
substance of the rhombo-inesencephalic fold. 
(d) The Sinus Terminalis. 
As to the universal occurrence of a sinus terminalis there 
appears to be some diversity of opinion. Both Retzius (’95) 
and Sterzi (’07) seem to have found this terminal dilatation in 
some specimens of Petroniyzon marinus and P. fluvia- 
tilis; Studnicka (’95) found it invariably present in P. 
f luviatilis and P. planeri, but he denied thatit was present 
in ammocoetes less than 10 cm. long. Schaffer (’01, fide 
Sterzi) apparently found this terminal dilatation to be constant 
in all ammocoetes. Sargent refers to the sinus terminalis 
of the ammocoete (6-10 mm. in length), but gives no figure, 
and his figure of the adult terminal sinus is incomplete. I 
have been able to recognise the sinus terminalis as a 
dilatation of the terminal portion of the central 
canal in all of the ammocoetes of Ichthyomyzon which 
I have examined, but it is only in specimens of about 40 
mm. and upwards that there is an actual bulbous enlargement 
of the end of the filum terininale. The change of 
position which this terminal sinus undergoes during larval 
life is noteworthy. In the young specimens the neural 
tube stretches wholly dorsal to the notochord and the sinus 
terminalis lies immediately dorsal to its extremity. The 
actual terminal neural pore is dorsally directed, and the 
hinder wall of the sinus terminalis is continuous with 
the terminal mass of neurenteric cells which caps the 
end of the notochord. As growth proceeds the sinus 
terminalis enlarges and begins to turn downward behind 
the notochord, ultimately coming to occupy the place of the 
terminal mass. This is not due, however, to an extension 
of the lumen of the neural tube into the midst of this mass 
of neurenteiic cells as a continuation of the process by 
which the central canal is said to have arisen. It is to be 
