No liquation of silver to the centre.

I ultimately reduced the thickness of the mould to 4 mm., and cast a bar into this, with the results shown in diagrams C and C'.

These results will, I think, be considered sufficiently remarkable by metallurgists who have been accustomed to deal with castings of standard silver. It must not be supposed, however, that liquation has been entirely prevented; it has, however, practically disappeared.

The excellent results now submitted to the Society have been obtained by limiting the possibilities of re-arrangement as much as may be, and by ensuring that the conditions of cooling shall be as uniform as possible. The need of obtaining uniform alloys is met with in other branches of industry than those which involve the use of silvercopper alloys, so that the conclusions to which the present experiments point are somewhat far-reaching.

མ.

"A Contribution to the Study of Descending Degenerations in the Brain and Spinal Cord, and of the Seat of Origin and Paths of Conduction of the Fits in Absinthe Epilepsy." By RUBERT BOYCE, M.B., Assistant Professor of Pathology, University College, London. Communicated by Professor V. HORSLEY, F.R.S. Received February 8,

1894.

(From the Pathological Laboratory of University College, London.)

(Abstract.)

For the purposes of this research, the following are the experiments which have been performed in the cat :

I. Lesions after which Animal was Kept Alive.

1. Removal of one complete cerebral hemisphere in 40 cats. 2. Removal of motor area only in 4 cats.

3. Division of the crus cerebri in 2 cats.

4. Removal of a lobe of the cerebellum in 10 cats.

5. Hemisection of the spinal cord in 4 cats.

6. Complete section of the spinal cord in 2 cats.

II. Lesions after which Animals were not Kept Alive for any Length of Time.

1. The preceding operations.

2. Removal of both cerebral hemispheres.

3. Removal of the cerebellum.

4. Removal of one cerebral hemisphere and opposite lobe of cerebellum and vice versa.

5. Removal of one hemisphere and division of opposite half of the spinal cord.

Where the animals have been kept alive, the symptoms during life and the anatomical changes found after death have been investigated, and in all cases the results of absinthe stimulation have been recorded by the graphic method.

In order to enable a clearer comparison to be made between the various results obtained by the experimental and anatomical methods, the paper is divided into

I. The anatomical changes.

II. Behaviour of the animal during life.

III. Results of absinthe stimulation.

IV. Conclusions.

PART I.-Anatomical Changes.

The investigations of the degenerations have been made in every instance by the Marchi method upon animals which have been kept alive from a few days to three months.

Removal of one Cerebral Hemisphere (Left), or complete Section of the Left Crus Cerebri flush with the Tentorium.

The results of these lesions as regards the descending degenerations are identical.

They show that in the higher bulbo-spinal segments, from the 3rd nerve downwards, there exist long descending internuncial fibres, which are grouped anteriorly and laterally as in the spinal cord; in other words, that the anterior, antero-lateral, and lateral columns of the cord can be traced from the level of the superior corpora quadrigemina.

Anterior Columnar Fibres.-These fibres begin to group at the level of the uppermost part of the 3rd nerve against the anterior and inner aspect of the grey matter of the aqueduct of Sylvius; they constitute the commencement of the posterior-longitudinal bundle. The degenerate fibres are absolutely limited to the side of the lesion; they occupy, for the most part, the inner portion of the posterior-longitudinal bundle, and they can be traced to the end of the cervical spinal cord. The higher the lesion the fewer are the number of degenerate fibres in the post.-longitudinal bundles; if the lesion extends much below the 3rd nerve, the degeneration of the post.-longitudinal bundle is very complete. New fibres are continually being added to the post.-longitudinal bundle in its passage through the bulbar segments, the degenerate fibres in consequence moving more anteriorly. In the spinal cord the degenerate fibres are scattered through the middle and post.-thirds of the anterior column on its inner aspect and closely simulate a direct pyramidal tract.

Antero-lateral Columnar Fibres.-Meynert's fibres (fountain decussation) pass from the side of the lesion across the raphe immediately ventral to the post.-longitudinal bundles, and turn vertically down, lying in the raphe in front of the post.-longitudinal bundles. The fibres can be traced into the antero-lateral columu of the cord as far as the lower cervical region. In their descent they move slightly forwards, but they are always more anterior than the fibres of the post.-longitudinal bundles. The degeneration method shows that Meynert's fibres arise in a focus close to the point of commencement of the descending root of the 5th, on a level with the roots of the 3rd nerves. The vast majority of the fibres decussate completely; the decussation is a very horizontal one.

In

Lateral Columnar Fibres.-In the highest segments these are derived from fibres, which decussate in front of the preceding group, and originate from a slightly higher level in the tegmentum; they probably correspond to Forel's decussating fountain fibres. the mesencephalon they occupy a position in front of and slightly external to the antero-lateral columnar fibres; lower down they are more laterally situated, and in the pons and medulla form a well defined group, immediately in front of the ascending root of the 5th and the substantia gelatinosa, and dorsal to the nucleus lateralis; they are traversed by the root of the 7th, and are bounded externally by trapezoid, and ascending cerebellar fibres. They are internal to the lemniscus and quite distinct from it; they can be traced to the upper lumbar spinal cord. In the spinal cord they occupy a position immediately in front and to the outer side of the pyramidal tract; they are readily distinguished from the latter by their large size.

Degeneration of Descending Root of 5th.-The Marchi method shows that this root is invariably degenerate upon the side of the lesion. It arises in the lateral aspect of the Sylvian grey matter at about the level of the upper oculo-motor nucleus, descends to meet the ascending root, and passes out without interruption.

Pyramidal System.-The degeneration is confined exclusively to the pyramid on the side of the lesion, and the degeneration appears complete.

There is a slight bifurcation at the decussation, a small group of fibres passing back to the lateral tract on the same side. There is thus a direct lateral pyramidal tract, but there is no anterior direct pyramidal tract.

The decussation of the pyramid is not confined to the cervical region; degenerate fibres leave the pyramidal system from the internal capsule, the crusta, and from the pyramid in the medulla. A large group passes through the thalamus and beneath the corpora quadrigemina, many of the fibres crossing over in the roof of the aqueduct. A considerable number of degenerate fibres pass back to

VOL. LV.

U

the quadrigeminal region from the outer and dorsal part of the

crusta.

In the posterior commissure, pineal commissure, and corpus callosum there are numerous degenerate fibres.

When the motor area or anterior third of the brain is removed, the degeneration is confined to the pyramidal system, the contrast with the results of hemisection of the mesencephalon being, therefore, very striking.

Comparing hemisection of the mesencephalon with hemisection of the cord, the difference, as regards the internuncial fibres, is the greater amount of decussation in the mesencephalon than in the cord. The internuncial fibres are conspicuous by their large size.

Comparison with ascending degenerations at the various levels shows that the descending tracts described above are distinct from the ascending.

Lesion of the Cerebellum.-When one lobe of the cerebellum is removed there is degeneration of the superior cerebellar peduncle. I have found no evidence of the descending columnar degeneration described by Marchi.

PART II.-Symptoms of Animals during Life.

Comparing the relationship between the extent of the degenerations and the symptoms, it is found that

Removal of the motor area entails degeneration of the pyramid alone, and a temporary paresis; the animal appearing, after a short time, quite like the normal.

Removal of a cerebral hemisphere or hemisection in the quadrigeminal region gives rise to degeneration of the pyramidal and internuncial systems and fibres. The symptoms are more pronounced; sensation is much altered; there is great difficulty of feeding.

Hemisection of the cervical spinal cord produces extensive degeneration of the anterior and lateral columns of the cord. There is at first hemiplegia, the vasomotor disturbance is much greater, and the sensory and motor paresis lasts longer. Removal of a lobe of the cerebellum does not appear to be accompanied by descending columnar degeneration; the symptoms may be very slight, or there may be incoordination and sensory and motor weakness upon the side of the lesion. The manifestation of the motor weakness differs from that seen in the cats, in which the pyramid is degenerate. There is no wrist drop in the cerebellar cat. The weakness in the case of the pyramidal cats is uncomplicated; on the contrary, that in the cerebellar cat is intimately bound up with complex phenomena of incoordination.