STRUCTURE AND DEVELOPMENT OF REISSNER’s FIBRE. 85 
attributed rather to the unequal growth of dorsal and 
ventral surfaces of the filutti term inale. The terminal 
neurenteric mass of cells disappears, and growth appears to go 
on much more rapidly upon the dorsal surface of the filum 
terminale. In this way the sinus terminalis becomes 
carried downward behind the notochord, and as a consequence 
of this unequal growth the terminal neural pore undergoes 
a complete cliange of position. Primarily directed dorsally,^ 
it has in turn a postero-dorsal, a posterior and a postero- 
ventral presentation, and finally takes up an actual 
ventral presentation, 
The sinus terminalis is present in those adult specimens 
of which I have been able to examine undamaged material, 
and in every case the terminal neural pore has the ventral 
position. 
(e) Reiss ner’s Fibre in the Sinus Terminalis. 
In the manner of the ending of the fibre posteriorly, too, 
the condition in Petromyzon fluviatilis and laival 
Ichthyomyzon tridentatus is markedly unlike that 
which Sargent has described for larval P. marinus (’01), 
adult P. marinus and larval P. planeri (’04). 
According to that author (’04, p. 149) the posterior part of 
Heissner’s fibre arises, in early larval development, by the 
coalescence of forwardly growing axons from a group of 
“ posterior canal cells ” situated wholly within the terminal 
sinus. This statement was made in the first instance of larval 
P. marinus (’01, p. 448), probably in mistake, for sub- 
sequently Sargent states that his investigations of the de- 
velopment of the optic refle.x apparatus in Cyclostomes were 
“confined to the larval stages of Petromyzon planeri” 
(’04, p. 149). 
‘ It is of interest to note that in Ann)liioxus (in which I have been 
unable to determine with certainty the occurrence of Reissner’s fibre, 
although I believe it to be present) there is a sinus terminalis dorsal 
to the notochord, and with an apparent terminal neural pore, directed, 
as in young ammocoetes, dorsally. 
