3. "When the ovum is manifestly dead, yet unexpelled." (Jellett, Dublin Jour. Med. Sci., May, 1897).

No amount of hemorrhage of itself signifies inevitable abortion, though this might demand-its induction. "Hemorrhage usually precedes pain when abortion results."

Gould and Pyle give the following as denoting "inevitable abortion:"

1. "Hemorrhage profuse and continuous, clotted and dark colored.

2. "Pain, cramp-like and severe.

3. "Cervical canal dilated.

4. "Uterus soft, enlarged; angle between upper and lower uterine segments effaced.

5. "Discharge consists of dark blood, clots and portions of

ovum.

6. "Signs of pregnancy present except amenorrhea."

A very sure symptom of impending abortion is a very dark discharge of blood from the uterus, usually persistent or recurring; this is usually very scant for several days or longer. Sooner or later following this abortion either occurs spontaneously or other symptoms arise which render active interference imperative. Such cases are prone to leave behind some unpleasant sequel, such as a damaged tube or chronic congestion of the pelvic viscera, which invites inflammation if one delays too long.

Having decided that abortion is inevitable, what course shall we pursue? No hard or fast rules can be given here, as abortion occurs in a variety of ways. Of the different ways in which abortion may occur I take the following from Thomas: 1. "There may be a complete emptying of the uterine cavity.

2. "The fetus may be expelled with the amnion and chorion while the decidua vera is left in the uterus, but later coming away in the lochial discharge.

3. "The fetus (and liquor amnii) alone may be expelled, leaving behind (up to the third month) the amnion and chorion.

4. "When utero-gestation is a little further advanced (beginning of the third month on) the fetus and membranes may be expelled and the placenta is left behind.

In investigating the subject of treatment I find the weight of evidence is against active interference so long as abortion proceeds in a normal way, i.e., hemorrhage, cramp-like pains, and followed in the course of a few minutes to two days by the entire contents of the uterus. Until two and a half months the placenta is unformed and we can usually leave the case to nature; after this time the placenta is the all-important element. Many cases no further advanced than a month or six weeks occur without knowledge of the patient. Yet when we have reason to know or suspect that abortion has occurred, even thus early, we should counsel rest in bed for some days.

When called to see a case of inevitable abortion, if the trouble is just beginning, put the patient to bed, and, if the hemorrhage is at all free, sterilize the parts and tampon with properly prepared gauze or pledgets of cotton. This is the safer and more expedient plan, anyway, if conception is more than two months advanced. Under such circumstances it is safe to leave a patient, removing the tampon in from six to twelve hours. Frequently the uterine contents will come away with the tampon, but, if not, another should be applied after careful preparation of the patient (administering an antiseptic vaginal douche), unless there should be some contraindication, as a rigor, fever or an offensive discharge. After twenty-four to thirty-six hours of expectant treatment, if the uterus has not emptied itself, its contents should be removed with the finger or curette. Of course this should be done under all antiseptic precautions. If the cervix is not already dilated this should always constitute the first step in the operation, because in free drainage is largely dependent the safety of the patient. If the cervix is already dilated or very dilatable, an anesthetic is not usually required.

In using the curette the speculum should always be used and the uterus drawn well down by means of the double tenaculum in the anterior lip. Then introduce a sound to learn the depth of the canal. My preference is for the sharp curette, and the whole surface of the uterus (excepting the cervical canal) should be carefully and thoroughly curetted and the cavity washed out with a hot 2 per cent. carbolic acid solution. If the uterus has been emptied, hemorrhage almost

certainly stops at once, and no packing should be employed. However, should hemorrhage not cease or the uterus be acutely flexed-thus hindering drainage—an intra-uterine tampon, or a strip of gauze for drainage, should be applied, according to indications. Keep the patient in bed for at least one week.

In cases of abortion which do not come on so acutely, but linger-hemorrhage (usually of very small amount) continuing or continually recurring-the curette (or the finger) should be used to empty the uterus. So should any case beginning with a rigor or fever, or whenever such symptoms arise during the course of an abortion, be treated. The presence of a foul discharge demands the use of the curette, and in such cases the irrigating curette should be used or the intra-uterine douche used later. The douche should be used twice each day till the temperature remains normal for a day or two and until there is no odor from the discharge. Should the temperature not be due to the absorption of putrid material from the uterine cavity, but to other complicating conditions, the douche might prove harmful and should not be given.

In the cases of so-called "missed abortion" nature should be assisted after we have assured ourselves the fetus is dead. Under such circumstances the woman is in constant danger of hemorrhage or of septicemia. I believe the plan of treating incomplete abortion advocated by Anna M. Stuart (N. Y. Med. Jour., Sept. 6, 1896) is well worthy of trial, and I believe it would serve a good purpose here. It consists in packing gauze in the uterine cavity, thus stimulating that organ to contract.

It will be noticed the expectant plan of treatment is here given the preference in the treatment of suitable cases of abortion, as the curette, even in the most skilled hands, is not without danger-numbers of cases of perforation of the uterus having been reported as resulting from its use. It will also be noticed that I have made no mention of any medicine to be given because none is indicated, unless it be an occasional dose of quinin for its oxytoxic effect, except when the cervix is well dilated and the fetal ball (or the placenta) is well engaged in the canal, when ergot may prove of service.

CORRESPONDENCE.

CLASSIFICATION

OF THE CONTINUED FEVERS OF THE SOUTH.

Editor Memphis Medical Monthly :

Since the discovery and elaboration of the life history of the malarial hematozoon, and the method of its transmission to man, there can be no more interesting or important subject connected with malaria discussed in this special issue than the classification of the continued fevers of the South.

In the Phila. Med. Journal of August 17, 1901, I offered a solution, based on purely clinical grounds, of the vexed question of our Southern fevers. The paper was directed especially to laboratory men, in the hope that they might confirm by chemical and microscopical methods the deduction drawn from bedside study of these cases.

The position was taken that the continued fevers under discussion were due "to lesions of the mucous membranes or of the nervous system produced by the malarial plasmodium, or more likely, to an irritant effect of plasmodial toxins on the central nervous system, or yet possibly to both in varying degree in different cases, corresponding to the characteristic regular irregularity of the disease in question," and the term " postmalarial fever" was suggested, as better indicating the etiology and pathology of these cases than any other. That the toxemic (malarial) etiology is much more probable, as indicated above, if not alone responsible, seems proven from the fact that so many of these cases have no nausea nor abdominal tenderness, and have some appetite throughout the course, a thing unlikely were they produced by lesions of mucous membranes.

In the November issue of the MONTHLY Dr. Krauss contributed a valuable paper on the "Occurrence of Leukocytosis and Socalled Dust Granules," in which he brings the strongest confirmatory microscopic proof of this position, without himself accepting it.

The mere fact of a constant leukocytosis in these particular continued fevers mentioned by Krauss, and the known absence of that condition in typhoid, is in itself a sufficient proof that they do not belong to the latter class, and it is almost incredible that Osler and others could in the face of such testimony as is adduced by Southern writers, so classify them.

May it not be possible that Dr. Krauss's dust granules, which he once thought embryones, are at the other end of their biology, and are fragments of plasmodia, whose ptomains are to blame for the septic fever? Dr. Krauss's statement that "some spreads look like a bursted red corpuscle from which eight to twelve of these bodies seem to be escaping, the similarity of their staining reaction to malarial

plasmodia,

are present in such great numbers in these cases and are almost totally absent in other blood," would at least point in that direction.

Again quoting from my paper: "This view readily accounts for the great irregularity and mild type of these cases, varying of course with the degree of resistance of each individual, and the amount of original toxic matter absorbed; for the absence of malarial hematozoa in the blood late in the cases, and hence with the failure of quinin to cure. It also accords with the fact that these cases come in at the height of the malarial season, after the system has been for some time under malarial influences. Furthermore, not only the blood tests, but the remittent character of these cases in the beginning, marks them surely paludal; but when we would expect our patient's convalescence, we are confronted with a continuing fever, with no hitherto well defined cause, lasting as long as two or three weeks, and in many cases much longer, and which is readily understood when the toxic nature of the cases is remembered, behaving much as any septic fever does."

Dr. Krauss also points out the late seasonal appearance and septic cause of these fevers, which, taken with the leukocytosis, and other microscopic findings of his enumerated above, would seem to admit of but one logical interpretation, viz.: that the fever is due to a poison, possibly alkaloidal, generated by the Laveran plasmodium. And Dr. Krauss himself says, "That it depends upon a malarial infection is a conclusion hard to get VOL. XXII-16