which the double-bladed saw (Luer's rhachiotome) is perhaps the safest, at least for beginners. The single-bladed saw with rounded end is also very useful and can be thoroughly recommended. The operation can be done most quickly by biting off the spinous processes with the heavy bone-forceps and cutting through the lamina with chisel and hammer, but there is greater danger of injuring the cord.

The numerous artifacts in the cord, reported as neuromata and heteroplasia even within very recent times by competent pathologists, would seem to indicate that the need of careful and delicate technique in the removal of the spinal cord is not yet fully appreciated.

The laminæ should be sawn nearly or entirely through in a line with the roots of the transverse processes from the third or fourth lumbar vertebra to the cervical region. The arches of the cervical vertebræ are best divided with a heavy bone-cutter, because they cannot be easily sawn, and there is sufficient room here for the points of the bone-cutter without danger of their pressing on the cord.

It is important to strike the outside limits of the spinal canal, so as to get as much room as possible for the removal of the cord. Test if the sawing be deep enough by the mobility of the spinous processes. If necessary, they can be freed by means of the hatchet-chisel and a hammer in the same way that the calvaria is loosened.

As the cord reaches only to the second lumbar vertebra, cut through between the third and fourth, free with the heavy bone-cutter the lower end of the row of the spinous processes, which are held together by their ligaments, and strip them up to the neck; then cut through the cervical arches with the bone-cutter, taking care that the point within the canal does not come in contact with the cord.

The nerve-roots are to be divided with a sharp scalpel by means of a long cut on each side of the cord. Then cut across the dura and the nerve-roots at the lower end of the exposed canal, and, while holding the dura with forceps, carefully free the cord from below up with scissors or scalpel, taking care all the time not to pull or bend the cord, be

cause in either way artifacts may be produced. Cut the cord squarely across as high in the cervical canal as possible, so that the remaining portion may be easily removed with the brain.

Lay the the cord after removal on a flat surface and incise the dura longitudinally, first posteriorly and then in front. A series of cross-sections, usually 1 to 2 cm. apart, is made through the cord while supported on the fingers during the

cutting, so that the cut surfaces shall fall apart. The different segments should ordinarily be left attached to the dura, so that their position in the cord can easily be determined.

A diagnosis from the fresh, macroscopic appearances of the cord is often very difficult to make, according to the best authorities.

The Eye. The contents of the orbit, including the posterior part of the eye, can be readily examined by chiselling

off the roof of the orbit. The posterior half of the eye can be removed by cutting around the eyeball with sharp scissors without changing the hold of the forceps on the sclera. If done quickly, the retina remains quite well spread out. The anterior half of the eyeball is to be propped in place by a plug of cotton dipped in ink or in a solution of permanganate of potassium.

The Ear.-The middle ear can be exposed by chipping off with a chisel its roof, which lies in the middle of the petrous portion of the temporal bone. The roof can also. be very easily bitten off with the heavy bone-cutters. If, however, it be desired to examine the ear more carefully by means of a section through the external meatus and the middle ear, it will be necessary to remove the whole of the petrous bone. For this purpose the incision behind the ear must be carried back along the anterior edge of the trapezius muscle halfway down the neck. Then the skin-flaps, including the external ear and the underlying tissues, must be dissected back for some distance on each side of the incision. Two converging incisions are then to be sawn, the anterior passing through the root of the zygomatic arch, the posterior just back of the sigmoid sinus, so as to come together at the apex of the pyramid of the petrous bone, or, better still, to meet in the foramen magnum. An ordinary chisel and a hammer or mallet will be found very convenient for freeing the petrous bone after the incisions have been

sawn.

In the examination of the petrous bone after it has been removed the first step is to chisel off the tegmen tympani so as to get a view of the middle ear. Next remove the lower wall of the external meatus, so as to expose the outer surface of the membrana tympani. Finally divide the petrous bone with a fine hair-saw by an incision starting in at the styloid process and coming out at the carotid canal, parallel to the crest of the pyramid of the petrous bone.

This incision divides the cavum tympani into halves. In the lateral half can be seen the membrana tympani with the hammer and the anterior half of the mastoid cells. In the

median half are the labyrinthine wall of the cavum tympani with the stapes and the posterior half of the mastoid cells. It is best to remove the anvil before sawing through the bone. The Eustachian tube can be easily exposed by starting from its termination in the middle ear.

The Naso-pharynx.-Although a fair view of the nares and pharynx can be obtained by chiselling off the portion of the base of the skull lying over them, the method does not begin to offer the satisfactory view that can be obtained by the method of Harke,1 a method which is not so difficult as might at first sight seem, and which consists in halving the base of the skull by a longitudinal incision. To do this the original incision in the scalp must be extended on each side over the mastoid processes and along the anterior edge of the trapezius muscle to a point below the middle of the neck. Then the posterior flap and the underlying muscles must be freed from the occipital bone and the upper portion of the occipital vertebræ. In like manner, the anterior flap must be dissected from over the root of the nose and the upper edge of the orbits, and be drawn down over the face. Then flex the head strongly forward and saw through the occipital bone and the base of the skull, dividing the occipital and frontal bones, the sella turcica, the cribriform plate, and the basilar process into equal halves. Anteriorly, it is well to go a little to the left or right, so as not to injure the nasal septum.

The next step is to cut the pachymeninx and the apparatus ligamentosis between the anterior edge of the foramen occipitale magnum and the processus odontoideus, as well as the inner side of the atlanto-occipital joint from within. Then the two halves of the skull are to be drawn forcibly apart. The nasal bones, the hard palate, and the alveolar process of the upper jaw break, and the two halves of the base of the skull open like a book, revolving around an axis which passes through the joint of the lower jaw and the atlanto-occipital joint.

If the foramen occipitale magnum offer too much resist1 Berliner klin. Wochenschrift, 1892, No. 30.

ance, break through it with a chisel, and also if necessary through the anterior and posterior arches of the atlas.

It is now easily possible to inspect the sinus sphenoidales, the nasal septum, the frontal sinuses, and the nasal passages. The antrum of Highmore is easily opened with forceps and a pair of bone-shears.

After the operation the two halves of the base of the skull are brought together, and wired if necessary. When the skin-flaps have been replaced all evidence of the operation is covered up.

Examination of New-born and Very Young Children.-I. The head is preferably opened by the method given on page 52.

2. According to Nauwerck, the spinal canal can be opened by dividing the vertebral arches with strong scissors.

3. The umbilical cord, if present, and the umbilical arteries demand close attention in children who have lived a few days or weeks, for the purpose of determining if infection has taken place at that point. Nauwerck advises a modification of the primary long incision. A little above the umbilicus it should divide into two diverging incisions running to the pubes. In this way a triangular flap is left containing the umbilical arteries, while from the upper end is given off the umbilical vein. The vessels may be ligated or opened at any point that seems advisable.

4. Anomalies of circulation should be looked for in all "blue babies." The closure or non-closure of the ductus Botalli (arteriosus) is best determined in situ by dissecting off the thymus and opening up the pulmonary vein in the middle of its anterior surface. The cut may be extended downward, if desired, through the pulmonary valve and the wall of the right ventricle. The duct lies in the median line of the pulmonary artery, a little above its division into its two main branches. A small probe can be passed through it into the aorta. The condition of the foramen ovale between the auricles is easily examined.

For other anomalies of the circulation it will usually be found most satisfactory to remove the thoracic organs in