SCIENCE. 
253 
mitted to absolute alcohol for a day before entire 
removal from the skull, then put in a mixture of 
methyl-alcohol and bichromate of potash, of a 
muddy beer color (thirty grains of the salt to the 
ounce of alcohol) for a week, and subsequently, for 
a variable time according as the specimen will 
harden, to simple Muller’s fluid. The staining, 
cutting and mounting can be done exactly’ as in the 
former case. 
Specimens prepared by the first method of har- 
dening will furnish better results for the medulla, 
those hardened with the second will yield more 
complete specimens of the higher ganglia. It is a 
well known fact that fluids that will harden the 
medulla oblongata well will sometimes fail to 
render the cerebrum and mesencephalon fit for 
cutting. 
Of course the most important series of sections 
will be one taken transversely to the peduncular 
axis. This should be made first, therefore, and 
studied in conjunction with the delineations made 
from the coarse specimens. Now the student hav- 
ing familiarized himself with the precise topo- 
graphy and extent of every ganglion, cortical ex- 
panse and fibre mass, is ready to proceed to more 
complicated inquiries, that is to study the relations 
of fibre masses. How he may proceed where a 
fasciculus does not run in a straight plane, I have 
indicated in a previous contribution to this journal.* 
It is needless to say that in addition to these 
methods, which may be called systemic ones, inas- 
much as they are calculated to reveal homologies 
and relations, that all other methods of hardening 
and staining may be used to study the finer and 
finest histology. They are of less importance, 
however, both to the zootomist and neurologist, 
than is generally supposed. 
Now a word as to the objects of such an inquiry, 
for unless the investigator has a definite point in 
view, and a provisional notion of the subject he 
intends to develop, his work will be barren of result, 
save he stumble on some revelation accidentally. 
a. The close relation between the cerebral lobes 
and the olfactory lobes of fishes may, if studied in 
all the groups, particularly the lampreys, lead to 
the establishment of a homology with the so-called 
cerebral lobes of the higher invertebrates. 
b. The fact, which we have every reason to sus- 
pect to be a fact, that the cerebral lobes of fishes 
are the true homologues of the cerebral hemis- 
pheres of the mammalia, sauropsida and amphibia, 
requires to be definitely established. Prof. Burt G. 
Wilder questions this homology, on the ground that 
the cerebral lobes of bony fishes are solid, and con- 
tain no ventricles. That so acute an observer, one 
to whom we owe so much in the line of correction 
of gross errors which have found their way into 
standard text books, could lean his objection on 
such a doubtful basis, shows how catholic must 
become the principles, if I may so term them, of 
cerebral anatomy. The embryological develop- 
ment of the fish’s brain presents features which no 
other vertebrate brain exhibits in the course of de- 
r Part I. of this series, Journal 0/ Nervous and Mental Disease , 
1887, p, 668. 
velopment, namely, the entire central nervous axis 
is apparently solid. In truth it is hollow, but the 
cavity is a mere slit, the walls of which are in con- 
tact, and when the cerebral lobes become solid they 
do so by the fusion of these walls and the oblitera- 
tion of the slit. The ventricle is therefore not an 
essential feature of the cerebral hemisphere, and as 
if to prove this fact beyond a doubt, we find that 
among animals as nearly related as sharks, some 
have true ventricles in these lobes communicating 
witn the third ventricle, while others have them as 
solid as the bony fish. 
c. The derivation of the olfactory bulb, a struc- 
ture often and unwarrantably confounded with the 
olfactory lobe, can be best studied in fishes. 
d. The same applies to the cerebral epiphysis 
and hypophysis, still known by the improper titles 
of pineal and pituitary glands. 
e. The relations of the peculiar lobi inferiores 
to the optic nerve, and the asserted homology of 
the corpora candicantea require confirmation. 
f. The question of the homology of the cerebel- 
lum and optic lobes, which is in a very unsettled 
state to-day, is yet unanswered. Wilder, in his 
paper on the brain of the C/iimcera, has exposed 
the fallacious interpretations which most authors 
have made in this regard. His essay will prove 
valuable to those engaged in this inquiry. Possibly 
the discovery by myself of the entire distinctness of 
the post-optic and the hitherto unknown inter-optic 
lobes in reptiles, from the optic lobes proper, may 
assist in unraveling the true relations. 
g. Since among fishes we find many examples 
of remarkable development of the periphery, I need 
but instance the rostrum of Spatularia, the great 
lateral expansions of the skate, the asymmetry of 
the Flounder, the rudimentary eyes of Amblyopsis, 
the marsupium of the Hippocampus, and the im- 
mense jaw of the Angler, an inquiry dealing with 
the relations of nerve centres to the projected peri- 
pheries may be expected to furnish many suggest- 
ive facts bearing on the projection doctrine. 
All through these lines it will be seen that as in 
every other branch of morphology a study of em- 
bryonic development is an essential to a proper 
knowledge of the fish’s brain. A brief considera- 
tion of the methods to be employed in this field of 
the study will not be out of place. 
Spawn can be obtained living from our fish 
hatching depots, whose superintendents will be 
found very obliging towards those requiring mate- 
rial for scientific study. The different stages of 
development, extending to beyond the period when 
the young fry escapes, can- be obtained by permit- 
ting the ova to develop under the eye of the ob- 
server in a hatching trough. 
The ova of bony fishes are dropped into a solu- 
tion of chromic acid, or Muller’s fluid; better, a 
few specimens are taken out each day and dropped 
each into differently strong solutions of the former 
and into the latter. I know of no standard strength 
that will yield uniform results, and have while 
working in this field in Vienna lost thousands of 
ova by following the routine directions. 
From the chromic acid and Muller’s solutions the 
1 spawn is transferred to alcohol in from two to 
