her friends but at the time of her entrance into the hospital she was completely unconscious. During these convulsions the body became rigid, with quick, short, clonic movements of the various muscles. The face became cyanotic, pupils dilated, and she frothed at the mouth. She was

Examination. This showed the patient to be unconscious.

a blonde of medium size. The skin was pale. Pupils were dilated. There was very slight edema of the ankles. The tongue was swollen and bleeding from recent injury. Heart and lungs were negative. Blood pressure was 160 (Faught's sphygmomanometer, four-inch arm band). The abdomen was distended above the pubes, the fundus reaching as high as the umbilicus. The fetus was not distinctly palpable. The fetal heart was heard in right lower quadrant; rate, forty-two per quarter minute.

A catheterized sample of urine when boiled became almost solid and addition of nitric acid intensified the precipitate. A drop of uncentrifuged urine showed numerous hyaline and granular casts. The sample obtained was too small for a complete analysis.

Diagnosis. The diagnosis of eclampsia was made and because of the frequency of the convulsions it was decided to empty the uterus at once. No labor pains were present.

Operation-Ether anesthesia; vaginal Cesarean section. The vagina was dilated manually, after which the cervix was seized and pulled down with a vulsellum forceps, and a transverse incision made through the mucous membrane just below the vesicovaginal junction. The bladder was easily pushed up with a sponge. A transverse incision was also made in the mucous membrane just behind the cervix, and the mucous membrane separated from the cervix. The anterior cervix and uterine wall was then incised in the median line for a distance of about eight centimeters and the posterior wall was likewise incised for a distance of about six centimeters. The opening thus made was large enough to admit the hand. The membranes were ruptured and with the hand in the uterus a foot was seized and an internal podalic version performed and the child delivered. The child gasped several times but was not viable because of its prematurity. It was, however, well formed and measured thirty-one centimeters in length. The placenta was delivered. manually. The incision in the uterus and vagina were sutured with catgut after the delivery of the placenta. Before leaving the operating table ten ounces of salt solution were injected under each breast.

Subsequent History. The patient was put to bed in good condition, and there were no convulsions after the operation. She soon recovered from the anesthetic, and became much excited, being very hard to manage. During the following night she got up out of bed on three different occasions. Eliminative treatment was begun as early as possible after the operation. Two drops of croton oil were given as soon as she began to show signs of consciousness. Elimination from the skin was obtained by the application of heat. The day after the operation her mind. began to clear, and in the second day she was perfectly rational. After this

the recovery was uneventful and the patient was able to leave the hospital in good condition on the fourteenth day after the operation, the cervix and vagina being well healed. Three days after the operation the albumin had entirely disappeared from the urine, and in four days the blood pressure had dropped to 122. The highest temperature after the operation was 100 on the fifth day.

After the patient became rational it was interesting to learn how little she recalled events just prior to her convulsions. The patient came to Ann Arbor in the afternoon, and her convulsions began suddenly that evening at about 6 o'clock. She remembers packing her suit case preparatory to coming to the city, but could not recall the fact that she had come here, and on regaining consciousness was surprised when told where she was. She did not know whether she had come here via the Michigan Central or the electric car. Neither could she recall where she had visited after reaching Ann Arbor.

The diagnosis is quite plain in this case. The only conditions to be seriously considered in a differential diagnosis are eclampsia and nephritis. No doubt many cases of nephritic convulsions occurring during pregnancy have been called eclampsia. The two conditions are often difficult to differentiate, and often some observation is necessary before a positive diagnosis can be made.

In the case reported there was no previous history suggestive of a nephritis, and her childhood had been usually free from infectious diseases. The almost complete absence of edema after having been in convulsions for eighteen hours would also argue against a nephritis. The blood pressure of 160, although raised, is not as high as in the average patient with uremic convulsions. The urea content of the urine passed during the first twenty-four hours in the hospital was low-only seven and eight-tenths grams as estimated with the Doremus ureometer. The rapid increase in urea as the condition of the patient improved was also significant. If the convulsions had been due to a nephritic condition one would hardly expect the urine to be absolutely free from albumin in three days. From a consideration of these observations it would seem that a nephritis could be well ruled out, and a diagnosis of true eclampsia could be established.

The diagnosis is of prognostic importance. In a nephritic patient it is reasonable to expect a recurrence of trouble with each succeeding pregnancy, while an eclamptic patient is less likely to have a recurrence. Williams contends that a patient can have a true eclampsia but once. This view has not as yet been generally accepted. However, there is no doubt that recurrences of convulsions during future pregnancies are less common after eclampsia than after nephritis.

The object of this report was not simply to report a case of eclampsia, but to call attention to vaginal Cesarean section as a treatment for this condition. The merits of this operation for eclampsia lie in the rapidity with which the uterus can be emptied, and in its freedom from shock and trauma.

CLINICAL PROFESSOR OF DISEASES OF THE NERVOUS SYSTEM IN THE UNIVERSITY OF MICHIGAN.

INVESTIGATIONS of the cerebrospinal fluid may be classified under four headings-physical properties; cytology, bacteriology, and chemistry. Of these four it would be difficult to say which is the more important nor is it necessary for clinical purposes to settle this question since in most cases where the findings in the cerebrospinal fluid are of value in the diagnosis of disease of the central nervous system the conclusion reached will be based upon the condition present in all four.

The following abstract, based largely on observations from a large number of examinations made at the University Hospital, deals only with the chemical findings, the cytology and bacteriology being treated elsewhere.

The cerebrospinal fluid is obtained by lumbar puncture, the needle. being introduced between the third and fourth or fourth and fifth lumbar vertebrææ, a point well below the lowest portion of the spinal cord. As much as twenty cubic centimeters may be drawn from normal individuals without evil effects. The average amount usually withdrawn for clinical purposes is somewhat less. It is important to note the pressure of the spinal fluid; this may be determined accurately by means of a manometer or judged by the rate of flow from the needle. The pressure is high in certain conditions such as brain tumors, abscesses, et cetera. It is also high following traumatism to the brain, even though there is no hemorrhage or fracture of the skull. It is also important to note the *Read before the Clinical Society of the University of Michigan, February 16.

1911.

color of the spinal fluid. Normally it is clear and colorless. In conditions of cerebral or spinal hemorrhage it would be stained a yellowish, or yellowish brown. In acute inflammations of the meninges, cerebrospinal meningitis, et cetera, the fluid is turbid, owing to the presence of pus cells. The specific gravity of the spinal fluid is about 1005; it varies slightly in different individuals but these variations probably have little clinical value. The freezing point is fairly constant at .53° centigrade. This lowered freezing point is probably due to the presence of sodium chlorid. The cerebrospinal fluid is slightly alkaline.

The constituent of the spinal fluid which has excited the most interest and which at present appears to have the greatest diagnostic importance is protein. This exists in the form of albumins, normally, and a slight proportion of globulin. The proportion of albumin in the normal individual is probably never over .5 parts per thousand. An increase in the amount of albumin is found in inflammatory conditions of all sorts and is also occasionally found in brain tumor. Various tests have been suggested for determining the relative amount of globulin. The one suggested by Nonne and Apelt, known as the Nonne-Apelt reaction, consists in adding a saturated solution of ammonium sulphate to an equal part of cerebrospinal fluid. An abnormal amount of globulin is present if the mixture becomes opalescent. This is described as "phase I" and, according to Nonne and Apelt, is present in paresis and tabes in a very high proportion of cases. It is also, however, present in many inflammatory conditions of the central nervous system. The other reaction, “phase II", is obtained by adding a few drops of acetic acid and boiling; this gives the albumin content. More recently, Noguchi has suggested a test which consists of adding two cubic centimeters of cerebrospinal fluid, five cubic centimeters of a ten per cent solution of butyric acid, boiling for one minute, then adding one cubic centimeter of normal sodium hydroxid solution and again boiling for one minute. The flocculent precipitate indicates an increase in euglobin which, according to Noguchi, is a specific test for tabes and paresis. More recent investigations have shown that a positive Noguchi reaction is also obtained from other, nonspecific diseases of the central nervous system such as brain tumor, et cetera. A rapid and delicate clinical test for excessive albumin has been described by Pandy. It consists of adding a few drops of the spinal fluid tested to an aqueous solution of carbolic acid, one to fifteen, a distinctly bluish white cloud indicating an increase in protein content, probably globulin. These tests are of great value in the differential diagnosis of inflammatory conditions of the central nervous system from functional disorders, with which they may be confounded. However, all signs of meningeal irritation may be absent in some cases of tabes. In trypanosomiasis there is found an increased globulin and nucleoproteid content in the spinal fluid and some increase has also been reported in diphtheritic paralysis, in uremic coma, and in brain tumor. In epilepsy, neurasthenia, pseudotabes, multiple sclerosis, et cetera, there is no increase in the globulin.

Next in importance to the protein in cerebrospinal fluid is the amount

of copper reducing substance. This substance is probably sugar, although various men have suggested that there may be some other reducing substance besides sugar. There is a certain normal relation between the quantity of sugar in the spinal fluid and in the blood, but the amount of sugar in both is increased in diabetes. In meningitis the amount of sugar is liable to be decidedly reduced and this applies also to syphilitic meningitis. An attempt has been made to diagnose between a tuberculous and meningococcic meningitis by the amount of sugar found in the cerebrospinal fluid but the findings of different observers are somewhat contradictory on this point. Beta-oxybutyric acid, diacetic acid and acetone have been found in the cerebrospinal fluid in cases of diabetes, being present even when these substances are not present in the urine.

Urea is a normal constituent of the cerebrospinal fluid but normally is in very small amount. It is increased in nephritis especially where there is uremic convulsions or coma. At it is not increased in cerebral arteriosclerosis, it affords a very valuable means of differential diagnosis. Mallard and Froment conclude from their studies that the presence of urea in notable quantities in the cerebrospinal fluid is always indicative. of renal insufficiency and is very valuable in the diagnosis of uremia from other states of coma or convulsions.

In eclampsia, Lockman found sarcolactic acid present in the cerebrospinal fluid after the patient had had two attacks. Its presence may have something to do with the causation of the convulsions.

Bile pigments are occasionally found in the cerebrospinal fluid. They are not, as a rule, found there in ordinary icterus unless the choroid plexus is in some way injured. In cases of meningeal hemorrhage bile pigment may be formed from the blood without the intervention of the liver and be present in the spinal fluid.

Halliburton found in the normal cerebrospinal fluid a trace of cholin. Most observers, however, have held it to be of pathologic import. Older studies in this direction were made by the platinum chlorid method but more recent tests for cholin as, for instance, the periodid tests, as described by Rosenheim, show that cholin does not appear in the cerebrospinal fluid, even in pathologic conditions. As Rosenheim's test is very delicate and much more accurate than the former methods, it must be concluded that the presence of cholin described by some authors was due to an error in technic.

Recently Cushing has attempted to show that the cerebrospinal fluid contains a substance that is secreted by the pituitary glands. His method was to concentrate this fluid and inject it into animals. The symptoms resulting were supposed to be due to pituitin. It seems more likely, however, that they were due to a concentrated salt solution. Normal cerebrospinal fluid is nontoxic to animals.

Although the choroid plexus is impermeable for many substances in health, it is not so for all. Potassium iodid and potassium ferrocyanid may be found in the cerebrospinal fluid after ingestion, and also the lithium salts. Crowe has shown that urotropin administered internally appar