212 J. B. JOHNSTON, 
of certain indirect paths or coördinating mechanisms, and perhaps as 
an optic center. This accounts for the presence of the ganglia and 
the bundles of MEYNERT even in those Mammals which have poorly 
developed olfactory organs. 
In its internal structure the ganglion habenulae shows greater 
complexity in the Mammals than in lower forms. Its structure is 
identical in the trout, sturgeon, and frog, as I have noted above. It 
has not been investigated in Reptiles and birds, but has been studied 
by the GoLGr method in Mammals by D. S. Ramon and by KO L- 
LIKER (96). Ramon (as cited by KÖLLIKER) described a small- 
celled median part and a large-celled lateral part, and found that 
the fine fibres of MEyNERT’s bundles came from the small cells and 
the coarse fibres from the large cells. The existence of the two 
kinds of cells is confirmed by KÔLLIKER. In many lower Verte- 
brates the right ganglion is larger than the left. In Acipenser the 
difference is great and is due (page 112) to a larger part of the 
tractus olfacto-habenularis entering the right ganglion. With this is 
correlated a larger number of cells in the right ganglion and a much 
larger number of fine fibres in the right bundle of MEYNERT. 
The termination of the bundles of MEYNERT requires further 
investigation. The failure of VAN GEHUCHTEN to trace the fibres 
through the corpus interpedunculare in the trout to a nucleus similar 
to that in which it ends in Acipenser may have been due to the fact 
that his material was embryological, so that the fibres had not 
reached their full growth. This explanation is insufficient, however, 
for the young rabbits, rats, mice, cats, and dogs investigated by 
S. Ramon y CAJAL, EDINGER and KÖLLIKER. Although CaJAL’s 
material may have presented some embryological features, the struc- 
ture of the ganglion interpedunculare itself suggests that the re- 
lations in Mammals differ materially from those in fishes and per- 
haps also in Reptiles. The ganglion in Mammals is larger and more 
complex, containing well developed cells of both the I and II types, 
and it also receives fibres from other sources than the bundles of 
MEYNERT. The corpus interpedunculare in Acipenser is an almost 
insignificant body. In Teleosts it seems to be better developed. Its 
much greater size and complexity in Mammals suggests that it may 
have gradually grown more important until in Mammals it has be- , 
come the chief, if not the only, nucleus of MEYNERT’s bundles. This 
is rendered very doubtful, however, by the very careful and accurate 
work of EDINGER & WALLENBERG (99a) on birds, where the bundles 
