The brain of Acipenser. 191 
question run in the molecular layer (cf. page 93). This is contrary to 
what we should expect from the structure of the cerebellum in higher Ver- 
tebrates and from SCHAPER’s descriptions of the cerebellum of fishes: 
However, there is one other fact which tends to support my obser- 
vations. I have mentioned above (page 83, 99) two bundles of fine 
fibres which leave the molecular layer at the caudo-lateral limit of 
the cerebellum, or beginning of the cerebellar crest. One of these 
bundles runs into the internal part of the acusticum, following the 
course of the internal arcuate fibres from the lobus lineae lateralis. 
The other bundle leaves the outer border of the cerebellar crest and 
runs cephalo-ventrad over the external surface of the medulla at its 
cephalic end. Although these fibres are fine and non-medullated it 
can not be thought that they are the neurites of granule cells, and there 
are no other cells in the cerebellum from which they could come 
except the PURKINJE cells. According to the theory of the phylo- 
genetic origin and dovelopment of PURKINJE cells given in the 
next paragraph, it is to be expected that the PuRKINJE neurites 
would take the same course as the neurites of the large cells in 
the acusticum. This is probably the case for most of the PURKINJE 
cells in the acusticum and lobus lineae lateralis, and possibly also 
for some of those in the cerebellum. The bundle of fibres mentioned 
above which runs through the internal part of the acusticum may 
be such fibres. But the differentiation of the PUKINJE cells even 
in so primitive a brain as that of Acipenser is sufficient to lead us 
to expect a new course for the neurites of at least a part of these 
cells. In the acusticum I have described fibres from the PuRKINJE 
(and other large?) cells running for some distance with the spinal 
V. I have suggested (page 73) that these may be the fibres which 
go to the lower olive. If.so, they may represent the early connection 
of the PuRKINJE cells with the olive. The bundle of fine fibres 
which run over the lateral surface of the cephalic end of the 
medulla, a little caudal to the commissura ansulata, is about in the 
position of the pons of higher Vertebrates. It is possible that this 
is the earliest representative of the pons. I am unable to offer 
any explanation of the PURKINJE neurites running in the mole- 
cular layer. 
It is perhaps possible to indicate in a general way how the 
cerebellum has developed from the anterior end of the acusticum, 
and how it has come to have its characteristic structure and great 
importance. I have shown that in the most general and fundamental 
