Semmelweis spoke in public about his discovery three times, on 15 May, 18 June and 15 July 1850, at the general meetings of the Imperial and Royal Society of Physicians of Vienna, with Rokitansky presiding. He wrote none of it down. The minutes, kept by the Society’s secretary Dr Herzfelder and reprinted in Győry’s edition of the collected works, are all that survives, and this piece follows them. The entry gives the substance of the first evening and one exchange from the third; the rest is here. All quotations are translated from the German by the writer of the entry.
15 May: the case
The minutes say that Semmelweis, “emeritus assistant of the First Obstetrical Clinic, developed his views on the genesis of puerperal fever.” He “showed numerically from the records that since the year 1839, since which a special clinic for the instruction of obstetricians and another for midwives have existed, on the first taken together more than four times as many of the lying-in women have died, for the most part of puerperal fever, as on the second, notwithstanding that the admission of the pregnant has alternated between the two every twenty-four hours, from which he drew the simple conclusion that the stronger appearance of childbed fever from time to time on the First Obstetrical Clinic can have been caused not by more general epidemic influences but solely and alone by local, that is endemic, conditions of that clinic.”
He then went through “the usual endemic causes as others have adduced them”: overcrowding of the wards, “the impregnation of the premises over many years with puerperal miasma,” the frequent “and supposedly rougher” examinations by the apprentice obstetricians, the fear of the pregnant when they were brought to that division, “and so on, one after the other, and showed that none of them makes the often great ravages of puerperal fever on the First Obstetrical Clinic explicable, and that on the contrary many of these supposed causes are to be found in a higher degree on the Second Obstetrical Clinic.” This, and the post-mortem findings in childbed fever, “which show the greatest similarity with that pyaemia which tends to develop in anatomists and surgeons after wounds at corpses and the impregnation of the freshly injured places with organic matter in putrefaction,” had led him to the conviction “that puerperal fever is likewise a pyaemic process produced by the uptake of putrefying organic matter into the blood of the mother from the inner lining of the uterus, and that the continual fresh importation of such matter precisely on the First Obstetrical School is grounded chiefly in the dissection of corpses, so frequent today, by the assistants and students of that clinic, without however another mode of transfer of putrefying organic constituents to the maternal organism being excluded, as must be assumed with decomposed remains of the placenta, with the uninterrupted examining of sick and healthy pregnant and lying-in women, and then with other patients who lie ill of a dissolution of the humours.”
Following this idea he had “introduced the rule that each of the students or other examiners should wash his hands carefully in a chlorinated lime solution before every exploration of a pregnant, parturient or lying-in woman, so as to extinguish completely every putrefying organic atom that might possibly cling to the fingers, even to its smell, and behold, the most brilliant success crowned this procedure, and now for full three years; the mortality, namely, which was otherwise 8.3 percent of the lying-in women on the First Obstetrical Clinic, has now become that which is also observed in private practice and on other lying-in clinics, namely 2.3 percent.” Other things supported it: “that whenever there were assistants on the clinic for midwives who made many examinations of corpses, the disease also claimed more numerous victims there”; that from the founding of the General Hospital in 1784 “to the creation of an independent chair of pathological anatomy no so-called epidemic of puerperal fever prevailed, and the mortality of the lying-in women did not even amount to one percent”; that from then “to 1 May 1847, namely to the beginning of the chlorine washings, as the period of the flowering of pathological anatomy, 5.7 percent died on the two clinics taken together, and from then to the last day of April of this year only 2.2 percent.” The fever did not spread outside lying-in houses, the seasons had no influence, it arose after injuries “as no epidemic disease otherwise does,” it appeared in animals, “and it can even be produced artificially in the latter.”
So puerperal fever, “according to the nature of it given above, is therefore just as little a contagious as a disease specific in itself, but develops in that an animal-organic matter that has passed into putrefaction, no matter from which patient and no matter whether from the living organism or from the corpse, taken up into the mass of the blood of the lying-in woman, produces the puerperal (pyaemic) decomposition of the blood, thereupon the known exudation, and as a third thing the metastases.” The matter reached the woman “by means of the examining finger, or through the use of utensils impregnated with it, or also through the air, laden with putrefying matter, that enters the cavity of the uterus after the birth, for which last mode of communication two striking examples are adduced.” The place of absorption was the part of the womb just above the inner mouth, reachable by the finger, which during pregnancy had lost its mucous membrane; the vagina was too thickly coated to absorb. The womb was most accessible in the first and second stages of labour, “and in this time too the examination is most frequently made; hence it is also to be explained that not only the mother but mostly the child too perishes by the same great process of exudation when the disease comes about, and why lying-in women who because of a delayed first and second stage of labour remained two or three days in the labour room mostly succumbed to the most malignant puerperal fever.”
He ended by announcing that he would answer, at a future meeting, the objection that Doctors Scanzoni and Seyfert had raised in the Prager Vierteljahrschrift against his views “as brought up by Professor Skoda in the Imperial Academy of Sciences.” Rokitansky asked the Society “whether it would not be generally inclined, in view of the high importance of the subject, to enter on a discussion of it next time, which was also agreed.”
18 June: Prague
Scanzoni had drawn up a table of women delivered and dead of puerperal fever at Prague which showed a better mortality than Vienna had obtained with the chlorine washings, and had argued that at Prague, from other local circumstances, the occasion for the assistants and students to be impregnated with putrefying matter from corpses was “only extremely rare.” Semmelweis “finds in this by no means a proof against the usefulness of the chlorine washings in themselves, but solely and alone the logically justified conclusion that the chlorine washings there, and indeed for the cases of possible introduction of putrid matter from the corpse, are superfluous, but by no means for those illnesses that can develop through the transfer of such matter from other patients, or from a parturient (perhaps) afflicted with discharging ulcers, to a healthy parturient; for the physician the occasion for impregnating his hands with putrefying matter is after all not only the corpse.”
On the transfers of sick women out of the First Clinic, and the present better mortality, he “gives the explanation that these were frequently undertaken precisely in the time before the chlorine washings, but that at present, since June 1847, no lying-in women (except for single cases because of syphilis, smallpox and so on, as the regulations require) are moved to other divisions of the General Hospital, so that the mortality ratio now given is a true and correct one.”
To Scanzoni’s doubt of the rabbit experiments, “on account of the larger quantity of putrid matter that was required to produce pyaemia in them,” he answered that Scanzoni “has surely not taken into proper account the wholly different structure of the genital apparatus of these animals and the lesser absorption caused by it, any more than the circumstance that in these animals the injected fluid is for the most part at once pressed out again by the abdominal press; besides, Dr Scanzoni does not seem to be very serious in his opposition to experiments on animals, since he himself proposes to the government that such be made.”
To Seyfert, “who puts the question what the usual number of puerperal illnesses may be, down to which the chlorine washings prove their efficacy,” he answered “that this is that number of lying-in women in whom the putrefying matter is taken up not from outside but from the sick individual herself, and whose absorption forms precisely that potency of which Dr Seyfert thinks ‘that it must first be invented’.” Seyfert had also said that in February 1849 an epidemic of puerperal fever had prevailed in the city of Prague without one in the lying-in house, which Semmelweis found “almost inexplicable, since the building of this institution lies within the city, and certainly inhabitants of the latter sought help there as pregnant and parturient women.” As for Seyfert’s belief in the weather, the lecture of 15 May had shown the contrary; and that puerperal fever was really a pyaemic process “should be found sufficiently set out in all the pathological-anatomical works of modern times.” On gynaecological wards, where the parts were excoriated, pyaemia could be caused in the same way, “for which reason the chlorine washings have been introduced on those here too.”
Then the minutes record the intervention that the entry quotes. Semmelweis’s remark “that on the Second Obstetrical Clinic too, in the time of such assistants as had occupied themselves much with the opening of corpses, the puerperal fevers had appeared more frequently, in which Dr Zipfl’s period of service in particular was named,” had “caused the latter to make closer investigations into this subject, the results of which he communicates to the Society in the following.”
Zipfl’s table
Zipfl, who had been assistant on the Second Clinic in 1842 and 1843, “took from the records all the post-mortem protocols that were found entered under his name during the first year of his assistantship (1842), forty-one in number; took from the admission sheets and the daily birth lists the women delivered on each post-mortem day and those of them who later died of puerperal fever; and divided them according to the time of their admission into five rows, according to whether they were delivered before the hour usually set for post-mortems or immediately after it, or only the day after.” The table “makes visible that precisely of those delivered before the post-mortem hour and of those delivered latest after it (twenty-four to thirty-two hours), who were thus not at all or least exposed through the examination to impregnation with cadaveric matter, the very most died, namely ten and a half to twelve percent, whereas of those who were delivered in the hours immediately after the openings of corpses fewer, and indeed only five percent, were carried off.”
“This is certainly only chance, as Dr Zipfl thinks; but had the opposite relation been found just as much by chance, everyone would have regarded it as a confirmation of the view put forward by Dr Semmelweis: a proof of how careful one must be when one wants to draw conclusions from statistical facts.” Comparing the months of the year gave the same: “precisely many lying-in women died in such months in which hardly a few openings of corpses were undertaken, whereas with a larger number of post-mortems in the month only a few deaths occurred among the delivered.” In all, on thirteen of the forty-one post-mortem days, among the women delivered within thirty-two hours of the post-mortem, nobody died; of 311 so delivered, 20 died; “whereas of 114 delivered within eight hours before the post-mortem time, 12 departed by death.” And of 147 street births that year, 5 had died, “which thus for lying-in women who could not have been examined at all likewise shows an unfavourable ratio in this year.” Zipfl “therefore cannot, from what has been factually adduced, agree with the opinion that it was chiefly the examination of parturient women with fingers impregnated with cadaveric matter that caused the frequent childbed fevers on the Second Lying-in Clinic during his years as assistant.” Semmelweis’s reply, and Dr Lumpe’s communications on the same subject, “had to be left over to the next meeting because of the lateness of the hour.”
15 July: the reply
Semmelweis began with Zipfl. The table was correctly compiled and its conclusions might be correct, “but according to Dr Semmelweis the number of post-mortems entered in the protocols is by no means the number of the openings of corpses of lying-in women made in all, since of the latter the greater number, because of the almost always identical scientific findings, were not entered at all, and so the whole compilation lacks its correct basis, namely the number.” As for the street births, “other external harms are mostly sufficient to explain them, to which is added that women so delivered are nevertheless often examined, to make sure whether the afterbirth has already come away or not.” The transfer of putrid matter to the parturient was possible “not only from the corpse but just as well from sick lying-in and parturient women.” “But it is always striking that precisely in the two years in which Dr Zipfl was assistant, and to his credit occupied himself frequently with post-mortems, the mortality on the Second Lying-in Clinic was greater than otherwise, so that in the year 1842 it amounted to 7.5 percent and in the year 1843 to 5.9 percent, whereas in the year before it was only 2.3 percent, and in the year after 3.5 percent.” His own Table I of 1861 has the two outer figures the other way round, 3.5 for 1841 and 2.3 for 1844. “Stranger still” was that Zipfl, “despite his objection, nevertheless claims the priority of the views on the origin of puerperal fever for himself and Ferguson,” since until then, as regards childbed fever, “only the absorption of purulent or putrid matter within the uterus” had been spoken of, and the examining finger never.
The minutes go on to Lumpe, who had served as assistant on the First Clinic in the early 1840s and now, from a compilation of the monthly reports from September 1840, argued on the other side; Schürer von Waldheim prints his table and Semmelweis’s book answers him at length. Nothing was decided that evening, and the Society did not take the matter up again.
What came of it
The Society elected Semmelweis an ordinary member that year, with twenty-one of thirty-nine votes. He did not write up what he had said; the three sets of minutes stood as his only published statement of the general doctrine, that the matter might come from the living as well as the dead, until the book of 1861. In February 1850 he had applied a second time to be allowed to lecture, and on 10 October he was made Privatdozent of theoretical midwifery with demonstrations on the phantom; five days later he left Vienna. Zipfl’s table, which he had answered by saying that the forty-one were a fraction of the post-mortems made, is the only count anyone ever made of the thing Skoda’s forbidden commission had been meant to count: whether the women died with the dissecting, day by day. It was made by a man with a motive, from a sample whose size nobody could fix, and it came out against the doctrine. That is not a reason to think it right. It is a reason to notice that the records which could have settled the question were in the hands of men who had no wish to open them, and that the one man who did open them was arguing for himself.