UMASS/AMHERST # 
 
 V:^7^ 
 
 31EDbtiDDS177tiD3 
 
 fe"^^^^^^ 
 
 i*a 
 
 
 
 S3) 
 
 SO 
 
 jj'^y .>^4. 
 
 Tm> 
 

 LIBRARY 
 
 OF THE 
 
 MASSACHUSETTS 
 
 AGRICULTURAL 
 
 COLLEGE 
 
 uK 
 
 No.Ll-S__4 
 
 SOURCEvl..]'. ■?4i 
 
 
This book may be kept out 
 
 TWO WEEKS 
 
 only, and is subject to a fine of TWO 
 CENTS a clay thereafter. It will be due on 
 the day indicated below. 
 
 0CT54 1904 
 
 f3 
 
WA^c, 
 
 '^ 
 
 >: 
 
 
 'i' t' 
 
THE NEW YORK MEDICAL JOURNAL, JUNE 20, 1891. 
 
 ILLUSTEATING THE AETICLE BY DE. PEUDDEN AND DE. HODENPYL. 
 
 0^ 
 
 Fig. 1. — Nodule in rabbit's lung, ten days after intravenous injection of dead tubercle bacilli. Steamed three hours 
 
 Fig. 2. — Nodule in rabbit's lung, twenty-eight days after 
 intravenous injection of dead tubercle bacilli. Cult- 
 ure steamed two hours. 
 
 /:/ 
 
 i 
 
 o 
 
 @ 
 
 \ V ej 
 
 
 hC f 
 
 L- t- 
 
 ^ 
 
 *s) r 
 & o 
 
 / 
 
 Fig. 3.— New-formed cells in capillaries ot the rabbit's 
 liver, seventeen days after intravenous injection of 
 dead tubercle bacilli. Culture steamed four hours. 
 
 ■-^,-^-5*i? ^ - C35^?fe' :^;,, 
 
 Fig. 4. — Diffuse cell-proliferation in vessels of rabbit's liver, twenty-five days after intra- 
 venous injection of dead tubercle bacilli. Culture steamed an hour and a half. 
 
Digitized by the Internet Archive 
 
 in 2010 with funding from 
 
 Boston Library Consortium IVIember Libraries 
 
 http://www.archive.org/details/studiesonactionoOOprud 
 
THE NEW YORK MEDICAL JOURNAL, JUNE 20, 1891. 
 
 
 'K''- 
 .".■f** 
 
 Fig. 0. — Small ^circumsoribed epithelioid cell-mass in 
 I'ahbit's livei-, surrouiided by zone of small spberoidal 
 cells, twenty-five days after intravenous injection of 
 dead tubercle bacilli. Culture boiled an hour and 
 a half. 
 
 Fig. 5. — Small circumscribed area of cell-proliferation in 
 vessels of rabbit's liver. Same animal as in Fig. i. 
 
 Fir 8 — Small Iner nodule composed of epithelioid 
 cells. Same animal as m Figs. 6 and 9. 
 
 Fig. 7. — Small nodule in rabbit's liver consis^ting ot a mass of 
 epithelioid cells, forty-eight days after intravenous injec 
 tion of dead tubercle bacilli. Culture steamed thiee hour'. 
 
 
 — ~1 1. 5_ 
 
 t, 4, -- — 
 
 ^ -?r 
 
 ■i V-" 
 
 ./J-. 
 
 
 w 
 
 > 
 
 Fig. 9. — Kodule in liver composed of 
 epithelioid and giant cells. Same 
 animal as in Figs. 7 and 8. 
 
 Fig, 10. — Complex nodule in rabbit's liver, consisting of a congeries of epitheli- 
 oid and giant cell masses, thirty-five days after intravenous injection of dead, 
 tubercle bacilli. Culture steamed two hours. Very few bacilli are remaining 
 
STUDIES OH" THE 
 
 ACTION OF DEAD BACTERIA IN 
 THE LIYING BODY 
 
 BY 
 
 T. MITCHELL PRUDDEX, M. D. 
 
 DIRECTOR 
 
 AND EUGENE HODENPYL, M. D. 
 
 FIRST ASSISTANT IN PATHOLOGY IN THE LABORATORY OF THE AXUMNI 
 
 ASSOCIATION OF THE COLLEGE OF PHYSICIANS AND SURGEONS, 
 
 NEW YORK 
 
 First Papee. — Introductory 
 
 Second Paper. — A Study of the Tubercle Bacillus 
 
 REPEINTED FROM 
 
 THE NEW YORK MEDICAL JOURNAL 
 
 FOR JUNE 6 AND 20, 1891 
 
 NEW YORK 
 
 D. APPLE TON AND COMPANY 
 
 1891 
 

 Copyright, 1891, 
 By D. APPLETON AND COMPATSIY. 
 
STUDIES ON THE ACTION OF 
 DEAD BACTEKIA IN THE LIYING BODY. 
 
 FiEST Article. — Inteodtjctoey. 
 
 The researches of the past decade, bringing to light, one 
 after another, the specidc micro-organisms of some of the 
 most common and fatal diseases, have been so surprising, so 
 definite, so fall of the promise of fruitful outlooks upon 
 hitherto untrodden fields, that we have scarcely yet had 
 time to recover from the glamour of the new light or to 
 realize, in the urgency of fresh practical problems, the exact 
 extent and bearings of the new knowledge. 
 
 For a time it seemed enough, and even more than 
 enough, that montli by month the proof grew stronger that 
 anthrax and tuberculosis and typhoid and erysipelas and 
 pneumonia and tetanus and diphtheria, and a whole group 
 of allied " wound diseases " and others of the so-called " in- 
 fections" class, were always associated with certain germs, 
 each peculiar in its life history, and each standing, as we 
 say, in an setiological relationship to its particular disease. 
 
 But as we have little by little become accustomed to the 
 new light, it has become evident not only that we are still 
 ignorant about the relationship of micro-organisms to sev- 
 eral extremely frequent and important infectious diseases — 
 the exanthemata, for example — but also that when we have 
 learned that a given acute infectious disease is always asso- 
 ciated wiih a particular form of germ, when the life history 
 of that germ is made out, and we can say that it stands in 
 an setiological relationship to the disease, there yet remains 
 a series of accessory problems to be solved in each particu- 
 
4 THE ACTION OF DEAD BACTERIA. 
 
 lar case scarcely less important than the establishinont of 
 the invariable association of the germ with the disease. 
 
 We are just beginning fairly to realize that the disease 
 is not an entity, a thing imparted by the invading germ to 
 the body, but that it is the result of the reaction of the 
 body cells in the presence of the germs ; that the body-cell 
 factor is just as important and just as much in need of 
 studv as is the germ-cell factor. We have been largely for- 
 getful hitherto, as with painful detail the characters and 
 preferences and metabolisms and vulnerabilities of the 
 pathogenic germs have one by one been brought to light, 
 that before our knowledge of the acute infectious diseases 
 can be at all complete, the characters and preferences and 
 metabolisms and vulnerabilities of the body cells must be 
 subjected to an equally careful scrutiny. The germ side of 
 the problem is new and fascinating; the man side is old, 
 and cellular pathology is a phrase familiar to our ears. But 
 these old problems have become fairly new in their new 
 light, and can not too soon be taken up afresh if our knowl- 
 edge of the acute infectious diseases is to be symmetrical 
 and of lasting use. 
 
 Partly by clinical observation and partly by laboratory 
 studies is the new knowledge of the man side of this theme 
 to be acquired, and old clinical observations, which have 
 lain uninterpreted or misinterpreted, and new facts which 
 the new points of view can not fail to elicit, will surely be 
 fast forthcoming. 
 
 But, returning to the bacterial side of the problem, it 
 became evident, very soon after the definite status of patho- 
 genic germs was made out, that something more than their 
 mere presence was necessary to account for the manifesta- 
 tions of the acute infectious diseases. 
 
 The earlj discovery that certain pathogenic germs set 
 free poisonous substances of one kind or another as the 
 result of their life processes, and the evidence that these 
 substances were directly accountable for many of the mani- 
 festations of the acute infectious diseases, drew attention to 
 the complexity of the problems involved, called in the serv- 
 ices of the physiological chemists, and for a time it seemed, 
 and still does to many, th;it, after all, it was the poisons 
 which the bacteria elaborated and sent out into the body 
 
THE ACTION OF DEAD BACTERIA. 5 
 
 on their destructive missions which was the most important 
 thing. "Ptomaines" became a favorite word. When we 
 had said that a given germ produced a given disease or ef- 
 fect by the elaboration of a given ptomaine, it seemed to 
 many fairly utireasonable to ask for any further explanation 
 of the acute infections diseases. The germs were relegated 
 to the more humble function of poison-factories, and the 
 ptomaines were invested with the insignia of malevolent 
 power. The pendulum seems disposed to swing back germ- 
 ward now, and in this paper, which is preliminary to the 
 record of some experimental studies made by Dr. Iloden- 
 pyl and the writer, an(i shortly to follow, on the tubercle 
 bacillus, it is my purpose briefly to review a series of recent 
 studies on the germ-cell bodies which throw a curiously in- 
 teresting new light on some old boilycell problems. 
 
 A very curious vital phenomenon which has long been 
 known in certain unicellular organisms — such as the fresh- 
 water amoiba and in the leucocytes of both the cold and 
 warm-blooded animals — is their response by movement 
 to contact with solid substances. Thus the amoeba floating 
 free in fluids tends to assume a spheroidal form and to re- 
 main immobile. When, however, under suitable conditions, 
 it touches a solid surface, like that of a glass slide, it sends 
 out pseudopodia and performs those curious progressive 
 evolutions known as the amoeboid movement. Essentially, 
 the same series of movements is observed in leucocytes 
 when they, under favorable conditions, come in contact with 
 solid surfaces — such as a glass slide or the walls of the body 
 lymph spaces. This faculty in these primitive forms of 
 life, consisting of a simple lump of protoplasm, is called 
 tactile sensibility, and it is in virtue of this that many of the 
 remarkable and useful evolutions of the leucocytes in the 
 body transpire. 
 
 It was found by Pfeflfer (1), a good while ago, that some 
 of the lowly vegetable organisms endowed with locomotion 
 — the Flagellata, Bacteria, etc. — were capable of moving 
 toward or away from certain substances which exerted a 
 chemical action upon them. This property he designated 
 as chemotaxis, and further postulated &% positive chemotaxis 
 the attracting effect, and as negative chemotaxis the repel- 
 ling effect on such organisms of the chemical substances. 
 
6 THE ACTION OF DEAD BACTERIA. 
 
 Pfeffer has shown that mobile bacteria move toward nu- 
 trient substances, such as beef-tea, and Engelmann and 
 others (2) have demonstrated their movement toward oxy- 
 gen, both effects being apparently due to the positive chemo- 
 tactic action of tliese substances. Stahl (3) showed that 
 similar properties exist in the plasmodia of myxomycetes. 
 This movement has been proved to be due, not to currents 
 in the fluids, not to diffusion, but to the specific action of 
 the particular chemical substances in question on living or- 
 ganisms. 
 
 Tbe chemotactic powers of the juice of raw potatoes, 
 which contains, as Pfeffer showed, potash salts and aspara- 
 gin, has been used in capillary tubes by Ali Cohen (4) to 
 separate mobile from immobile bacteria in mixtures. In 
 this way he found that be could separate cholera and ty- 
 phoid bacilli from the numerous other forms in fjeces, and 
 thus make easier the obtaining of pure cultures for diagnos- 
 tic purposes. 
 
 Now, the same condition of affairs exists in the leucocytes 
 of both the cold-blooded and warm-blooded animals, and 
 the conditions and bearings in them of this positive and 
 negative chemotaxis wore studied in detail by Massart and 
 Bordet(5) and by Gabritchev&ki (6) in 1890. Tbe latter 
 observer has grouped as the result of his experiments cer- 
 tain chemical substances in accordance with their action in 
 this way upon leucocytes. Tlius, inthe group of substances 
 exciting a neixative — repelling — chemotaxis, we have con- 
 centrated salt sohition, 10 per cent. ; lactic acid ; quinine, 
 0*5 percent.; alcohol, 10 percent. ; chloroform; jetpiirity ; 
 glycerin ; bile. Substances having no effect — indifferent 
 chemotaxis — are distilled vpater; dilute salt solution, 0*1 
 to 1 percent.; carbolic acid, 1-per-cent. solution; antipy- 
 rine ; glycogen ; peptone ; beef-tea ; blood ; aqueous hu- 
 mor. Among the most prominent substances exciting a 
 positive chemotaxis are especially sterilized and non-ster- 
 ilized cultures of various pathogenic and non-pathogenic 
 bacteria. 
 
 The general method of testing the powers of these vari- 
 ous substances is to fill small capillary glass tubes, closed 
 at one end, with the substance to be tested, and to thrust 
 these beneath the skin of an animal. After a few hours 
 
THE ACTION OF DEAD BACTERIA. 7 
 
 these tubes are withdrawn and their contents examined. 
 Into tubes filled with substances inciting' positive chemo- 
 taxis the leucocytes crowd in great numbers, while they are 
 held away from tubes having negative chemotactic contents, 
 and when filled with indifferent substances there is no ef- 
 fect at all. While the tactile sensibility of leucocytes may 
 cause them to cluster in small numbers about the surface of 
 the glass tubes, the effect of this property in the leucocytes 
 is altogether insignificant as compared with the chemical 
 substance exciting positive cheraotaxis. 
 
 It appears, then, that there are certain substances asso- 
 ciated with bacteria wliich excite in the leucocytes a move- 
 ment toward the germs. The culture medium itself has no 
 such effect, but the action is developed equally whether 
 the cultures be living or have been killed by boiling. It 
 would thus appear that either the bacteria themselves or 
 some result of their life and growth must be the exciting 
 agency. 
 
 Under the dominant views regarding the significance of 
 the various chemical substances set free by bacteria as the) 
 grow, it has been assumed that it was largely under the ex 
 citing influence of the ptomaines that leucocytes exhibited 
 the phenomena of chemotaxis in the presence of bacteria, 
 A practical bearing was given to the subject, under the in- 
 fluence of this view, by the assumption that in the process 
 of suppuration, as commonly induced by various species of 
 bacteria, the leucocytes gathering at the inflammatory foci 
 were drawn thither in virtue of their chemotactic properties 
 which the metabolic bacterial poisons brought into play. To 
 this view the doctrine of phagocytosis, as held by Metsch- 
 nikoff and his adherents, readily attached itself, and we had 
 a well-rounded hypothesis, in accordance with which the 
 leucocytes, drawn, in virtue of their chemotaxis, into the vi- 
 cinity of invading bacteria, at once set to work to destroy 
 them, and with them the poison sources which were stimu- 
 lating excessive cell inroads. 
 
 But while these observations were going on, an allied 
 but quite independent series of experiments was being car- 
 ried out by Buchner and his associates in Munich, which 
 have thrown a new and apparently most significant light 
 upon both the phenomena of chemotaxis and the nature 
 
8 THE ACTION OF DEAD BACTERIA. 
 
 of snppnratioTi. To these experiments let us then briefly 
 turn. 
 
 While it is fully established that a true suppurative in- 
 flammation may be experimentally induced by a variety of 
 inorganic substances, it is still true that the suppurative 
 processes which occur in the body, either as independent 
 lesions or as complications of a variety of diseases, are prac- 
 tically always due to the action of bacteria. So that in a 
 clinical sense the summary statement, " no suppuration with- 
 out bacteria," is true. While, as above indicated, it has 
 been the general belief of late that the metabolic products 
 of bacterial life, the "ptomaines" or the "toxines," were 
 the active agents in inducing suppuration, this, save in a 
 few instances, has not been proved. 
 
 Biichner (V), in the course of some experiments on the 
 introduction of anthrax spores and anthrax bacilli into the 
 trachea of rabbits and guinea-pigs, had observed some time 
 ago that while the introduction of the bacilli was followed 
 by an intense inflammatory reaction of the lung tissue, with 
 accumulation of leucocytes, fibrin, etc., in the air spaces, the 
 introduction of the spores alone was followed by no such 
 marked inflammatory reaction, but that the spores entered 
 the blood channels and induced, in due time, the usual sys- 
 temic effects of anthrax poisoning. 
 
 There is one factor — so reasoned Buchner — which has 
 not been taten definitely into the account in the causation 
 of suppurative inflammation by bacteria, and that is the 
 possibility that the effect may be produced, not by the 
 ptomaines, not by the toxines already so much studied, but 
 by the albuminoid constituents of the bacterial cells them- 
 selves. If this were true, then the inteuse exudative inflam- 
 mation in the lungs following the introduction of the an- 
 thrax bacilli might be explained by the local disintegratiun 
 of the bacilli and the setting free of their potent proteid 
 constituents, while no such effect would follow the intro- 
 duction of spores. 
 
 It has been repeatedly shown by numerous observers (8) 
 that sterilized cultures of various pyogenic bacteria — such as 
 Staphylococcus pyogenes aureus, Bacillus pyocyaneus, etc. — 
 were as capable of producing suppuration as were the fresh 
 living cultures. But it was believed that this was due to 
 
THE ACTION OF DEAD BACTERIA. 9 
 
 the retention of a toxic substance furnished by the life pro- 
 cesses of the ^erm which had not been destroyed by the 
 sterilization, but clung" about the dead gern3 bodies. Al- 
 though Wyosokowitsch had filtered off the fluid from steril- 
 ized anthrax cultures and found that the filtrate was not 
 pyogenic, while the solid material was, he inferred only 
 that the toxic material assumed to cause suppuration was 
 not soluble in the nutrient fluid. 
 
 Buchner had also shown (9), in the course of some other 
 experiments, that the sterilized emulsion of the so-called 
 pneumo-bacillus of Friedlander, subcutaneously injected, 
 could cause suppuration in rabbits and guinea-pigs. He 
 found further that if such a sterilized emulsion were allowed 
 to stand for some time, so that the solid could be separated 
 from the fluid parts of the mass, the fluid part did not cause 
 suppuration, while the solid part did. That the effect of 
 such sterilized bacterial emulsions was not due to their me- 
 chanical effects in the tissues was shown by such control 
 experiments as the introduction of powdered charcoal, in- 
 fusorial earth, magnesia, potato emulsion, etc., beneath the 
 skin, with negative results. 
 
 By a series of manipulations similar to that practiced 
 with the pneumo-bacillus, Buchner now tested the effect of 
 sterilized emulsions of cultures of seventeen different species 
 of bacteria, among which may be mentioned Staph ijlococcus 
 pyogenes aureus, Staphylococcus cerens flavus, Sarcina au- 
 rantiaca, Bacillus prodigiosus. Bacillus fitzianus. Bacillus 
 cyanogenus, Bacillus megatherium. Bacillus suhtilis, Bacillus 
 coli cominunis, Bacillus acidi lactici, Bacillus anthracis, Pro- 
 teus vulgaris, Finkler's comma bacillus, etc. The injection 
 of one cubic centimetre of the sterilized emulsions of each 
 of these germs resulted within two to three days in an 
 aseptic — that is, bacteria-free — purulent infiltration in the 
 subcutaneous tissue at the seat of injection. On the other 
 hand, the clear fluid obtained by sedimentation from the 
 sterilized emulsions of Bacillus cyanogenus. Bacillus mega- 
 therium, and Bacillus anthracis, induced no suppuration, 
 while the separated sediment invariably did. 
 
 While it thus seemed probable that the albuminous ma- 
 terial of the bacterial cell was at least chiefly concerned in 
 inducing suppuration on the injection of sterilized emul- 
 
10 THE ACTIOK OF DEAD BACTEHIA. 
 
 sions, this was not yet fairly proved. Buchner now sought 
 to strengthen the evidence by a very ingenious experiment. 
 The modern technique of staining bacteria with the aniline 
 dyes depends, as is well known, upon the power of these 
 dyes to enter into chemical combination with the bacterial 
 cell plasma. Acting upon this principle, Buchner found 
 that if he added to a sterilized emulsion of the pneumo- 
 bacillus, which was strongly pyogenic, an aqueous solution 
 of methvl violet, the emulsion was wholly bereft of its pyo- 
 genic powers. Anent of this interesting bit of evidence of 
 the importance of the bacterial cell proteids, Buchner calls 
 attention to its bearing upon the theory of the antiseptic 
 and anlipyoDfenic action of the so-called pyoktanin of Stil- 
 ling, the usefulness of which in practice is still sub judice. 
 
 But more definite proof of the importance of the bac- 
 terid-protein in inducing suppuration was still needed, and 
 Buchner proceeded to separate it from cultures of the 
 pneumo-bacillus after the method of Necki, by digestion of 
 masses of culture in dilute potash and precipitation with 
 acetic or hydrochloric acid. The precipitate separated by 
 filtration was again dissolved in dilute potash sohition and 
 reprecipitated. This was done the third time, and at last 
 the purified product was brought into solution. This ma- 
 terial gave the chemical reaction of an albuminoid body. 
 Subcutaneous injection of this material in rabbits in some 
 cases was followed by a gathering of leucocytes, in others 
 not. As it seemed likely that on simple subcutaneous in- 
 jection the material was readily and rapidly absorbed be- 
 fore it produced local effects, recourse was had to a method 
 of experiment used by Councilman in his well-known studies 
 on suppuration (10). Small glass tubes, drawn out at the 
 ends, were filled with the pneumo- bacillus protein, sealed up, 
 and steiilized by steam for an hour. These were then in- 
 troduced, with strict antiseptic precautions, beneath the skin 
 of rabbits, shoved away from the opening, and, after they 
 were healed in, their tips were broken off. After five days 
 the tubes were exposed. Around the openings of these, as 
 well as extending deep into their interior, were masses and 
 plugs of leucocytes. Cultures showed no living bacteria. 
 Control experiments with tubes filled with salt solution 
 showed no collection of leucocytes. 
 
THE ACTION OF DEAD BACTERIA. H 
 
 It tlius seemed to be proved that the pyogenic action of 
 sterilized cultures of Friedlander's pneiimo-bacillus is due to 
 the freed albuminoid constituents of the bacterial cell. 
 That such a freeing of tlie contents of the bacteiial cells oc- 
 curs in the tissues of the living; body seems evident from 
 their well-known proneness to disintegration and the devel- 
 opment of involution forms in suppurative foci. 
 
 The next thing to be done was to carry on a similar 
 series of experiments with other well-known pathogenic 
 bacteria. To this task Buchner and his associates addressed 
 themselves in a series of studies as yet not fidly published 
 (11). But, even so far as their results are known, some 
 most significant facts have been elicited. Buchner endeav- 
 ored to separate by the method of Necki (see above) the 
 bacterio-protein from about fifteen species of bacteria, but 
 in many of these the attempt was unsuccessful, because 
 sufficient solution and extraction of the proteid ingredients 
 of the germs did not occur. The Bacillus i^yocyaneus gave 
 the most abundant albuminous extract, but a sufficient 
 amount was obtained from Staphylococcus pyogenes aureus, 
 Bacillus typhosus, Bacillus suhtilis, Bacillus acidi lactici, 
 and from the red potato ba':'illus for animal expeiiment. It 
 was, in fact, found that capillary tubes filled with the puri- 
 fied proteids from all these species of bacteria and placed 
 beneath the skin of the rabbit showed after two or three 
 days, extending into the open end, a plug of fibrinous pus 
 several millimetres in length. This plug was found, on mi- 
 croscopical examination, to consist largely of leucocytes. 
 
 That the ordinary chemical decomposition products of 
 bacterial cell life are not concerned in inducing this posi- 
 tive chemotaxis in the leucocytes was shown by introducing 
 beneath the skin of rabbits tubes filled with such substances 
 as butyrate and valerianate of ammonia, trimethylamin, am- 
 monia, glycocoll, leucin, tyrosin, urea, etc. These were, for 
 th"e most part, wholly without effect upon the leucocytes, 
 only glycocoll and leucin exciting in some cases a moderate 
 chemotaxis, not at all to be compared, however, with that 
 of the bacterio-proteins. 
 
 It would thus seem to be highly probable, if not abso- 
 lutely proved, that the power of exciting positive chemo- 
 taxis, which at least many species of bacteria display, is 
 
12 THE ACTION OF DEAD BACTERIA. 
 
 due to the proteid iuji-redients of their bodies when these 
 are set free, as they may be naturally when the germs disin- 
 tegrate in the tissues, or artificially by chemical extraction. 
 
 With that keenness and fertility of thought which char- 
 acterizes Buchner's work, he now gave wider range to his 
 investigation. He recognized the fact that, though of late 
 the phagocytic action of the leucocytes has been especially 
 emphasized in relation to bacteria, this is by no means their 
 chiefest or most constant role. Dispose of bacteria the 
 leucocytes undoubtedly do; whether after themselves killing 
 them, or after they are destroyed by other ag;oncies, ])as not 
 yet been fully determined. But by far the most constant 
 phagocytic role of the leucocytes is in carrying on the 
 process of resorption and disposal of useless particles and 
 dead material in the living body. About such material 
 they gather much as the}- do in the vicinity of bacteria, 
 though not in such marked degree or under such dramatic 
 conditions. 
 
 Now, what attracts the leucocytes into the vicinity of a 
 .particle of dead and useless muscle, or cartilage, or connect- 
 ive tissue which they are to absorb and remove ? Certainly 
 not bacterial poison, certainly not bacterial proteids, for 
 with what may be called the normal phagocytic functions 
 of the leucocytes bacteria have nothing to do. Having 
 shown that a proteid substance derived from the bacterial 
 cells was capable through chemotaxis of attracting leuco- 
 cytes, Buchner now studied in a similar way the effects of 
 closely allied substances — namely, the so-called vegetable 
 caseins, gluten casein from wheat and legumin from peas, 
 both separated by precipitation from alkaline solutions. 
 Both of these substances were capable of exciting the most 
 marked chemotaxis in the leucocytes of rabbits. More- 
 over, as it has been shown that vegetable casein exists as 
 such in the grain of cereals and of the leguminosse, he in- 
 troduced beneath the skin of rabbits or guinea pigs, under 
 strict antiseptic precautions, masses of wheat and pea meal, 
 ami found that within two days these masses were sur- 
 rounded and penetrated by enormous masses of leucocytes. 
 Cultures from these masses proved the entire absence of 
 bacteria. Starch introduced subcutaneously under the same 
 conditions induced no gathering of leucocytes. 
 
THE ACTION OF DEAD BACTERIA. / I3 "Mf>^v 
 
 That this gathering of leucocytes was due to cheraotaxis 
 and not simply mechanical, owing to the tactile sensibility 
 of the leucocytes, was shown by introducing subcutaneouslv 
 in a rabbit in one place an emulsion of infusorial earth with 
 0*7-per-cent. salt solution, and in another place an emulsion 
 of the earth with glutin casein. In the first, after three days, 
 but few leucocytes had gathered about the foreign material, 
 while the second was surrounded and partially penetrated 
 by an enormous number of leucocytes. 
 
 But still another step remained to be taken. As the 
 gathering of leucocytes about dead organic fragments in the 
 tissues which are to be removed, as so often happens, can 
 not be ordinarily due to bacteria or bacterio-protein, so, 
 also, interesting as the observation may be, can vegetable 
 proteins have no part in the matter. So alkali albuminates 
 were prepared and purified, in a manner similar to that em- 
 ployed with the bacterial and other vegetable proteins from 
 muscle, liver, lungs, and kidney of rabbits. These tested 
 in the same way were all found to strongly attract leuco- 
 cytes when introduced beneath the skin in tubes. Of the 
 alkali albuminates prepared from blood, fibrin, yolk and 
 white of egg, only the blood and yolk of egg showed mod- 
 erate power of exciting positive chemotaxis. 
 
 These experiments show that it is only ceitain of 
 the decomposition products of animal tissue which pos- 
 sess chemotactic powers, and that these, as a rule, are 
 the earlier and not the ultimate products of the decom- 
 position. 
 
 Finally, as it has been shown that a general leucocytosis 
 is apt to be associated with febrile inflammatory processes, 
 Buchner and Roemer studied the effects of intravenous in- 
 jections in rabbits of these various chemotactic proteids. 
 They found that within eight hours of their introduction 
 into the blood there was a marked leucocytosis lasting for 
 several hours, and that this might he heightened by repeat- 
 ed injections. Thus they found, by a daily injection of 2 c. c. 
 of an eight-per-cent. solution of the bacterio-jiroteins of Ba- 
 cillus pyocyaneus, the relation of white to red blood-cells, 
 which at first was 1 to 318, was on the second day 1 to 126 ; 
 on the third day, 1 to 102 ; on the fourth morning, 1 to 73 ; 
 and on the same evening, 1 to 38. From this time on no 
 
14 THE ACTION OF DEAD BACTERIA. 
 
 increase was noted. The absolute number of the red 
 blood-cells remained unchanpted, while there was an ab- 
 solute sevenfold increase in the number of leucocytes. 
 Gluten casein, as well as alkali albuminate from muscle, 
 injected into the blood, showed similar but less pronounced 
 effects. 
 
 Considering now the bearing- of all these experiments 
 on suppuration and on physiological resorption of dead or- 
 ganic materials in the tissues, it would appear that in sim- 
 ple resorption, as in l>acterial suppuration, the leucocAtes 
 are drawn to the seat of operation by a proteid material. 
 This in resorption seems to be furnished by the dead and 
 disintegrating ti^sues themselves, and when the leucocytes 
 have gathered up a certain amount of refuse in their bodies 
 they may carry it away. In bacterial suppuration, on the 
 other hand, the attracting material may be furnished by the 
 protein of the disintegrating bodies of the bacteria them- 
 selves, but poisonous ptomaines fumished by the live bac- 
 teria may cause the destruction and degeneration of the at- 
 tracted leucocytes, which thus collect as pus. 
 
 Whether the ptomaines themselves may not indirectly 
 furnish chemotactic material by causing the destruction of 
 the tissue elements and the setting free of their albuAiinous 
 constituents, is a matter requiring further study. 
 
 It is also not improbable that the limited suppuration 
 induced by bacteria-free chemical substances — such as tur- 
 pentine, calomel, etc. — may be due to the chemotactic 
 tissue-proteids set free by the action of the chemicals on 
 these tissues. 
 
 It seems probable that not only are the leucocytes drawn 
 toward the chemotactic proteids thus produced in or intro- 
 duced into the body, but that the fixed connective-tissue 
 cells are stimulated to proliferation. In fact, Buchner 
 found that by the introduction of a sterilized emulsion con- 
 taining 3'5 milligrammes of pyocyaneus protein into the 
 forearm of one of his associates, a severe inflammation was 
 induced with all the symptoms of an acute typical erysipe- 
 las, with lymphangcitis, such as must have involved the fixed 
 connective-tissue cells. On the fourth day the inflamma- 
 tory process underwent resolution. Gluten casein induced 
 similar but less acute effects. 
 
THE ACTION OF DEAD BACTERIA. 15 
 
 These most clever and striking researches of Bnchner 
 would seem to throw niucli light on the whole subject of 
 the theory of suppuration, and to promise large accessions 
 to ourknowleHge of inflammation when the many lines of 
 thought and study which they suggest shall have been fol- 
 lowed out. 
 
 It is now evident that an aseptic suppuration is possible 
 under a variety of conditions. 
 
 It still remains true, however, for the purposes cf sur<>i- 
 cal practice, that the suppurative processes as we see them 
 in the clinic and at the bedside are due to the presence of 
 bacteria of one form or another. It is true, also, that the 
 suppurations which we can induce experimentally with ster- 
 ilized — that is, dead — bacterial cultures, or with certain dead 
 proteid substances — aseptic suppurations — are limited in 
 their duration, extent, and destructive power, as compared 
 with those occurring under the influence of living germs. 
 This is because in the latter case the growing and new 
 forming germs may keep up the inflammation once alight 
 to an almost indefitiite extent. 
 
 We purpose, in the paper which is to follow, to detail 
 some results of a series of experiments on the action of dead 
 tubercle bacilli on the living tissue not less striking than are 
 those which show the power of other sterilized bacteria to 
 induce suppuration. 
 
 Bibliography. 
 
 1. Pfeffer. Unters. a. d. lotanischen Institut zu Tubingen, 
 1886-1888. 
 
 2. Engelmann. Bot. Zeitg., 1881. Prudden. Medical Rec- 
 ord, March 26, April 2, 1887. 
 
 3. Stall). Zur Biologie der Myxomjceten. Bot. Zeitg., 1884:. 
 
 4. AH Cohen. Ctrlbl. far BaMeriologie, Bd. viii, 1890, No. 
 6, p. 165. 
 
 5. Massart and Bordet. Journal public par la soc. royale 
 des sciences medicates et natureltes de Bruxelles, 1890, v. 
 
 6. Gabritchevski. Annates de Vinst. Pasteur., June 25, 1890. 
 
 7. Buehner. Berl. Min. Woch., 1890, No. 30, p. 673. 
 
 8. Bachner. Op. eit., p. 674. 
 
 9. Bachner. . Berl. Uiii. Woch., 1890, No. 10. 
 
 10. Councilman. Virchow's Archiv, Bd. 92, 1883. 
 
 11. Buehner. Berl. Min. Woch., 1890, No. 47. 
 
16 THE ACTION OF DEAD BACTERIA. 
 
 Second Aetiole. — The AcTioisr of Steeilized Cultures of 
 
 THE TUBEECLE BaCILLUS. 
 
 We have seen in the first article (1) that the action of 
 bacteria in the body is by no means limited to the chanyjes 
 induced by those substaoces called ptomaines and albu- 
 moses, or toxaibarains, which are set free by living germs. 
 Tt app?ars that the proteid constituents of the bacterial 
 cells themselves, when these are set free, either by a natural 
 disintei^ration of the germs or by an artificial extraction, 
 are capable not only of stimulating the fixed body cells to 
 proliferative changes, but may, by calling into play the 
 forces involved in chemotaxis, cau-^e either a moderate col- 
 lection of leucocytes or induce marked suppuration. 
 
 It is probable that these two fairly distinct factors, the 
 eliminated ptomaines and albumoses on the one hand, and 
 the bacterio-proteins on the other, arc both involved in 
 many, if not most, of the acute infectious diseases. But 
 one or other of these factors is apt to be so preponderant 
 in many cases as to fairly dominate both the symptoms and 
 lesions. 
 
 For example, in tetanus, typhoid fever, and Asiatic 
 cholera, the toxic substances absorbed into the body at 
 large from the seat of growth of the pathogenic germs may 
 determine profound and even fatal symptoms without much 
 local change at the proliferating germ centers. 
 
 On the other hand, in many of the local suppurations — 
 in pneumania and in tuberculosis, for example — the most 
 marked effects are apt to be induced in the immediate 
 vicinity of the invading germs. 
 
 In still other diseases — diphtheria, for example — we may 
 have extreme local tissue changes together with profound 
 systemic effects. 
 
 Finally, under a variety of unusual conditions, patho- 
 genic germs which commonly induce one particular order 
 of lesions may bring about most striking alterations of 
 another class. 
 
 As an example of this variability of effect, we may cite 
 
THE ACTION OF DEAD BACTERIA. 17 
 
 the suppurative inflammations which not infrequently occur 
 as complications of typhoid fever, when these germs lodge 
 or grow in unusual situations. 
 
 Perhaps the most striking example of a pathogenic 
 germ whose effects are closely limited to the immediate 
 neighborhood of its seat of growth is the Bacillus tuber- 
 culosis. That there may be systemic effects induced by 
 the absorption into the body at large of poisons eliminated 
 by the tubercle bacillus, we, of course, can not deny. But 
 if such there be, we certainly know very little about them, 
 and they must be insignificant in comparison with the 
 dominant lesion of tuberculosis — namely, the production 
 immediately about the germs of a new short-lived tissue 
 having a moderately characteristic morphology and prone 
 to undergo degenerative changes of great significance, both 
 to the germs which cause it and to the integrity of the in- 
 vaded organ. 
 
 It is because of the peculiarly direct and constant re- 
 lationship between the germ and the lesions which it in- 
 duces that we have selected the tubercle bacillus for an 
 experimental stady along the lines which our preliminary 
 paper has suggested. 
 
 But, before detailing our own experiments, it seems de- 
 sirable to briefly notice some observations already made in 
 this field, which will have a bearing in establishing our 
 point of view. 
 
 A number of studies have been made on the metabolic 
 products set free during the growth of the tubercle bacillus 
 both in cultures and in the body. But these studies have 
 been made from such a variety of different standpoints and 
 with such different technic^ue that it is difficult to glean 
 any very positive and definite data from the results. 
 
 Hammerschlag (2) found in the alcohol and ether extract 
 of the t'lbercle bacilli, fat, lecithin, and a poison which in- 
 duced spasms and death in rabbits and guinea-pigs. The 
 residual material of the bacilli contained an albuminoid and 
 cellulose. He could not separate a ptomaine by Brieger's 
 method, but did find evidence of a fever-causing toxal- 
 bumin. 
 
 The recent publications of Koch on the nature and 
 effects of substances in or derived from cultures of the 
 2 
 
.18 THE ACTION OF DEAD BACTERIA. 
 
 tubercle bai^illas we may assume to be familiar to all our 
 rea<lers. But the manner in which they have been made 
 known and the lack of detail in the description of the mode 
 of preparation of his " lymph " leave his results in a most 
 unsatisfactory state of indefiniteness, so far as the scientific 
 bearing of the subject is concerned. Some substance or 
 substances, it would appear, either set free during the 
 growth of the tubercle bacilli or, as it is said, extracted 
 from them by glycerin, are capable of exerting rather ill- 
 understood effects upon tubercle tissue in the body and are 
 not without extreme effects in many cases upon the body 
 at large. With the therapeutic bearing of the substances 
 contained in Koch's tuberculin we have here nothing 
 to do. 
 
 One point, however, in Koch's last publication on this 
 subject has an important bearing upon the subject in hand 
 — namely, the statement that pure cultures of the tubercle 
 bacillus killed by boiling, or in other -ways, are capable of 
 inducing local suppuration witb no other effects, if intro- 
 duced subcutaneotisly into the guinea-pig. 
 
 Hueppe and Scholl (3) cultivated the tubercle bacilli in 
 a glycerin-peptone bouillon, and found that the fluid con- 
 tained substances having essentially the same eff"ects upon 
 animals as those noted by Koch. As the result of an inter- 
 esting series of experiments, these observers came to the 
 conclusion that the specific poison of Koch's " lymph" did 
 not belong to the proteins, but to some substance or sub- 
 stances eliminated by the bacilli during their growth and 
 contained in the culture fluids, and which were not, as Kocb 
 asserted, extracted from the culture by glycerin. Ilucppe 
 and Scholl separated from the " lymph '' which they had 
 prepared, and which seemed to be practically identical with 
 the ''lymph" of Koch, by precipitation with alcohol, a sub- 
 stance which appears to contain the active ingredient of the 
 " lymph." They separated in the same way the same sub- 
 stances from Koch's "lymph," and, on testing these in tubes 
 beneath the skin of animals, they foinid that in both cases 
 this substance was markedly chemotactic. 
 
 That Koch's '" lymph " is not, as it is furnished in the 
 crude condition, chemotactic and does not tend to cause ab- 
 scesses, may be due to the fact that it contains glycerin, 
 
THE ACTION OF DEAD BACTERIA. 19 
 
 which exerts an influence on the leucocytes opposite to that 
 which positive cbem-itactic substances display. 
 
 Trudeau (4), working also with fluid cultures, obtained 
 materials apparently the result of metabolic processes of 
 the tubercle bacilli which induced toxic effects upon ani- 
 mals. But these were not marked, since he did not succeed 
 in sufficiently concentrating the fluid containing the mate- 
 rials sought for. 
 
 Zulzer (5) extracted from agar cultures a substance which, 
 on purification and injection subcutaneously in small quan- 
 tities into rabbits and guinea-pigs, caused temporary in- 
 creased rapidity of respiration and an elevation of tempera- 
 ture. Larger does were fatal. 
 
 Crookshank (6) separated the bacilli from glycerin-broth 
 cultures by filtration, and concentrated the filtrate by evapo- 
 ration. This concentrated filtrate, injected subcutaneously 
 into guinea-pigs, induced muscular spasms, a lowering of 
 the temperature, and death. 
 
 Weyl (7) extracted tubercle bacilli with dilute caustic 
 soda, and found in this extract a substance which seemed 
 to him to belong in the mucin group. Subcutaneous in- 
 jection of a solution of this substance produced in animals 
 a local necrosis at the seat of injection. 
 
 Maffucci (8) found that sterilized cultures of tubercle 
 bacilli introduced beneath the skin of guinea-pigs in con- 
 siderable amounts were capable of inducing a chronic poi- 
 soning of the organism, ending in death from marasmus and 
 great destruction of the red blood-cells in the spleen, but 
 does not note any marked local changes beyond the pro- 
 ducing of an abscess. 
 
 Wyssokowicz (9) killed the tubercle bacilli in cultures 
 by boiling, and these on subcutaneous injection in a rat 
 caused a small local abscess, containing well-formed and 
 readily stained though dead tubercle bacilli. 
 
 An emulsion of a boiled culture of the tubercle bacilli 
 was injected into the abdominal cavity of a rat. The ani- 
 mal was killed on the thirtieth day. A few small nodules 
 were found beneath the liver which consisted of a central 
 pus-like mass surrounded by a thin new-formed tissue 
 wall. The central portion of these nodules consisted of tu- 
 bercle bacilli and leucocytes, while their walls were com- 
 
20 THE ACTION OF DEAD BACTERIA. 
 
 posed largely of endothelial cells, with here and there giant 
 'cells among them. Readily stained tubercle bacilli were 
 scattered through the walls of these nodules, and the epi- 
 thelioid and giant cells were often grouped about them. 
 The microscopical examination of the liver revealed numer- 
 ous small spheroidal nodules composed of epithelioid cells 
 and leucocytes, among which were scattered a few tubercle 
 bacilli. 
 
 We know that various kinds of substances, organic and 
 inorganic, when introduced into the body, can become sur- 
 rounded by a tissue in many respects resembling tubercle 
 tissue. This was abundantly shown in the years just pre- 
 ceding the discovery of the tubercle bacillus and the final 
 establishment of tuberculosis as an acute infectious disease. 
 Whether the dead tubercle bacilli in these experiments of 
 Wyssokowicz's act in some special way in stimulating the 
 tissues to these tubercle-like growths, or whether they do 
 this simply as foreign bodies, must be shown by further 
 studies. 
 
 The action of the living tubercle bacillus in inducing 
 the local lesions of tuberculosis seems to be quite complex. 
 But there appear to be several tolerably distinct classes of 
 influences at work. 
 
 In the first place, there seems to be an irritation or 
 stimulation of the tissues near the bacilli which results in 
 the proliferation of cells, primarily and notably of the con- 
 nective tissue cells. The eJBFect of this is to produce what 
 is called the epithelioid cell tissue. Following this, or asso- 
 ciated with it, there is a local action either on the leuco- 
 cytes or the blood-vessels, or both, which leads to the ac. 
 cumulation of small spheroidal cells either in the periphery 
 of the tubercle or distributed through it. Generally there 
 is a necrotizing action which results in the coagulation 
 necrosis or cheesy degeneration of the new-formed tissue 
 or other tissue near the tubercle. 
 
 The giant cells which form such striking features in 
 many phases of tuberculosis appear to be formed under the 
 influence, on the one hand, of those factors which induce 
 cell proliferation, and, on the other, of those which retard 
 the completion of the cell proliferation when once it is 
 under way. The impulse to cell division under these cir- 
 
THE ACTION OF DEAD BACTERIA. 21 
 
 cumstances is in so far successful that a protoplasmic 
 mass is formed with many nuclei, but the cousumraa- 
 tion of this attempt, the division of the cell bodies, is in 
 abeyance. 
 
 In view of the new facts regarding the singular power 
 of the bacterio-protcin and chemotaxis which the researches 
 briefly summarized in our first article have revealed, and 
 the promising but, it must be confessed, rather obscure 
 light which the recent studies on the poisons of the tu- 
 bercle bacillus have elicited, it has seemed to us that some 
 further studies on the tubercle bacillus should be made 
 with the view of learning, if possible, in more detail some- 
 thing of the factors which are concerned in developing the 
 complex structures which we call tubercles and tubercle 
 tissue. 
 
 Our first series of studies was made with the view of 
 determining what effects, if any, are produced in the body 
 of the rabbit by the introduction of cultures of the tubercle 
 bacilli killed by prolonged boiling and freed from any ma- 
 terial elaborated by them during their growth on the artifi- 
 cial culture media, so far as such substances are soluble in 
 the culture media, or in water, or in fifty-per-cent. glycerin. 
 That is to say, we wished in these experiments to learn 
 whether the dead bodies of the tubercle bacilli alone, apart 
 from any of their elimination or soluble metabolic products, 
 produced changes in the body cells of living rabbits, and, 
 if so, of what kind. 
 
 Technique. — We cultivated the tubercle bacillus for this 
 purpose on peptone agar with six-per-cent. glycerin, and in 
 glycerin-peptone bouillon. In all cases we continued the 
 growth until it became voluminous, and we obtained iden- 
 tical results whether the cultures were two months or six 
 months old, and whether they were made in bouillon or on 
 agar. We have used cultures of tubercle bacilli derived 
 from three different sources. 
 
 The solid masses of culture were scraped carefully off 
 from the agar, or filtered from the culture-bouillon. In 
 some cases we were not particular to remove from the cult- 
 ures what culture-bouillon clung to the germ masses. But 
 in most of the experiments the masses of bacilli were 
 washed with sterilized distilled water, and then boiled in a 
 
22 THE ACTION OF DEAD BACTERIA. 
 
 small amount of sterilized water for from an hour and a 
 half to four hours. In still another set of experiments we 
 first carefully washed the culture masses, and then boiled 
 them for from two to four hours in tifty-per-cent. glycerin 
 with water, and then filtered off the fluid, washing the 
 germ masses before making an emulsion with pure water 
 for injection. 
 
 We have found that the condensation fluid in glycerin- 
 agar cultures and also the glycerin bouillon in which the 
 tubercle bacilli had grown, when dropped into strong alco- 
 hol, gives a white precipitate, which, according to the studies 
 of Koch (10) and Hueppe and SchoU (11), contains the act- 
 ive principles of Koch's tuberculin. We furthermore found 
 that after the glycerin-agar condensation water or the gly- 
 cerin bouillon in which the tubercle bacilli are grown has 
 been carefully washed out of the culture masses of bacilli, 
 no more of this tuberculin-containing material is extracted 
 from the bacilli by either water or fifty-per-cent, glycerin, 
 even after boiling for four hours. It would thus appear 
 that the active substance in Koch's "lymph "is a mate- 
 rial elaborated by the life processes of these germs and 
 set free in the culture media, as shown by Hueppe and 
 Scholl (3). 
 
 After steaming the bacilli for from an hour and a half 
 to four hours in the steam sterilizer, and separating them 
 from their metabolic products in the manner described, we 
 have made an emulsion of the bacilli in sterilized water 
 which we have used in our experiments. In a few cases we 
 have boiled the bacilli together with a small amount of the 
 culture fluids containing their metabolic products, but the 
 results of these experiments have been the same as when 
 we have carefully separated them. 
 
 We find that after prolonged boiling, either in water or 
 in fifty-per-cent. glycerin, while many of the bacilli present 
 the usual bizarre broken granular involution and degenera- 
 tion forms so often described, most of them preserve their 
 form intact, and are stained in the usual way by the Koch- 
 Ehrlich method. The broken granular forms are readily 
 stained. 
 
 We have, by a series of separate experiments, proved 
 that the tubercle bacilli are certainly killed by the pro- 
 
THE ACTION OF DEAD BACTERIA. 23 
 
 longed steaming to which they have been subjected, of from 
 an hour and a half to four hours. 
 
 As this is, of course, a point of vital importance in our 
 experiments, we have been very careful about this prelimi- 
 nary sterilizing. We have used the Arnold steam sterilizer, 
 which is one of the most efficient forms of laboratory ster- 
 ilizers known to us. The material to be killed has been 
 invariably placed in small, thin-walled flasks, exposed to 
 the full action of live steam. The duration of the steam- 
 ing was, of course, carefully reckoned from the time when 
 the chamber of the steamer and its contents was at the full 
 boiling temperature. While we have found in a series of 
 preliminary experiments that an hour and a half's steaming 
 sufficed to kill the cultures, as, from what we know of the 
 vulnerability of these germs, it should, in all except one 
 set of experiments, we continued the steaming from two 
 to four hours, the average time being about two hours and 
 a half. 
 
 The masses of tubercle bacilli thus steamed have been 
 proved to be dead, first, by their failure to grow on favorable 
 culture media; and, second, by the much more delicate test 
 of repeated inoculations in guinea-pigs, which invariably 
 gave negative results so far as the development of tuber- 
 culosis is concerned. Finally, we have inoculated guinea- 
 pigs with materials from the lesions about to be described 
 as induced in rabbits by the dead cultures, and obtained in 
 the guinea-pig only negative results. 
 
 In these ways we have satisfied ourselves that the ma- 
 terial which we were using contained in every case no liv- 
 ing tubercle bacilli, but only their dead bodies. 
 
 Experiments. — Subcutaneous Injections. — We first 
 made a series of injections of the emulsion of the sterilized 
 tubercle bacilli in rabbits beneath the skin and found that 
 in a considerable proportion of cases a small local abscess 
 is developed in from two to six weeks. This abscess, some- 
 times firmly encapsulated with fibrous tissue, contains pus 
 cells, granular detritus, and many readily stained and well- 
 formed tubercle bacilli, together with involution forms. 
 This pus, as above indicated, on inoculation into that most 
 susceptible animal, the guinea-pig, gave negative results ; 
 no tuberculosis developed, which was to be expected if the 
 
2i THE ACTION OF DEAD BACTERIA. 
 
 cultures of the tubercle bacilli used for inoculation had been 
 readily sterilized. We have further made cultures of this 
 pus with a view of determining whether this suppuration 
 might not be due to some of the ordinary pyogenic bac- 
 teria, but the results were negative. We thus find, in con- 
 sonance with the results of Koch, and of Hueppe and 
 SchoU, that dead tubercle bacilli are, under certain condi- 
 tions, markedly pyogenic. 
 
 We have tested the chemotactic powers of sterilized 
 cultures of the tubercle bacillus introduced in tubes beneath 
 the skin of rabbits, in the manner described in our first 
 article, and found in every case after six days a plug of 
 leucocytes extending deep into the broken end of the 
 tube. We have introduced the dead cultures beneath 
 the skin of rabbits inclosed between the laminae of Zieg- 
 ler's plates. Tn from three to five days the capillary 
 space was found crammed with leucocytes. The dead 
 tubercle bacillus thus possesses marked positive chemotac- 
 tic powers. 
 
 Peritoneal and Pleural Injections. — We next made 
 injections of the emulsion of dead tubercle bacilli in con- 
 siderable quantities (2 to 3 c. c. of a milky emulsion) 
 into the peritoneal and pleural cavities — in four animals 
 into the peritoneal cavity ; in two, into the pleural cavity. 
 We obtained one positive result in each set of injections. 
 The others were all negative. 
 
 In the positive cases, whose results were similar in both 
 the pleura and peritonaeum, we found several larger and 
 smaller white nodules adherent to the serous surfaces. 
 These consisted of a central soft creamy mass, made up 
 largely of pus cells with cell fragments and granular detri- 
 tus intermingled with large and smaller clusters of readily 
 stained and well-formed tubercle bacilli. While the nuclei 
 of some of the pus cells were not readily stained by hema- 
 toxylin, there seemed to be no evidence of well-marked co- 
 agulation necrosis in the cells except when very large 
 masses of the dead bacilli were introduced. This central 
 pus mass was in all cases closely surrounded and inclosed 
 by a wall of fibrous tissue, in the inner layer of which were 
 numerous larger and smaller masses of epithelial cells in- 
 termingled with giant cells. Some of these epithelial and 
 
THE ACTION OF DEAD BACTERIA. 25 
 
 giant cell islets were irregular in shape and diffuse in out- 
 line. Others were sharply circumscribed and rounded, and 
 presented the general appearances of miliary tubercles. 
 Tubercle bacilli were everywhere abundant in these epi- 
 thelial cell masses, and nowhere else in the fibrous wall of 
 the nodules. Well-marked cheesy degeneration was no- 
 where seen in these cell masses which resemble tubercle tis- 
 sue, even when the animal remained alive for four weeks 
 after the injection. 
 
 We thus see that the injection of dead cultures of the 
 tubercle bacilli is capable of occasionally inducing in the 
 pleural and peritoneal cavity of the rabbit a suppurative fo« 
 cus and the formation about it of a tissue morphologically 
 similar to tubercle tissue. 
 
 While these results are interesting, we should not for- 
 get that a great variety of substances, organized and unorgan- 
 ized, on introduction into the serous cavities, as well as be- 
 neath the skin, are capable of stimulating the cells of the 
 part to the production of a new tissue which in many re- 
 spects closely resembles in its morphological characters the 
 tissue induced by living tubercle bacilli. The new tissue 
 growths formed under these conditions have been the sub- 
 ject of the most extended research in times gone by. And 
 every pathologist who lived through the epoch immediately 
 preceding the discovery of the tubercle bacillus knows to 
 what fallacious conclusions experimental studies on the in- 
 troduction of foreign bodies into animals led many an ob- 
 server, owing to the unwarrantable stress which was laid 
 upon epithelioid cell tissue and giant cells, in determining 
 the morphology of tubercle. 
 
 We have therefore left this line of experiment here and 
 turned our attention to a form of experiment which seemed 
 to promise a more direct insight into the detailed effects of 
 dead tubercle bacilli upon living animal cells. This was the 
 injection of the dead germs directly into the blood-vessels 
 of the rabbit through the ear veins. 
 
 The leading idea in these experiments was to introduce 
 into the body the dead bacilli under such conditions that 
 their effects upon individual cells could, if possible, be de- 
 termined. 
 
 Intravenous Injections. — We have used in these experi- 
 
26 THE ACTIOX OF DEAD BACTERIA. 
 
 ments the sterile emulsions of tubercle bacilli described 
 above, and have injected them in moderate quantity, aim- 
 ing to introduce into each animal an amount of the material 
 which would correspond to a spheroidal mass of the solid 
 culture of from about one to two millimetres in diameter, 
 diffused in about two cubic centimetres of water. In order 
 to avoid gross vascular disturbance, it is important to break 
 up as much as is practicable the flocculent massesof the tu- 
 bercle bacilli with a platinum needle before injecting them. 
 The animals bear this injection into the blood perfectly Veil, 
 and we have in our series of experiments in this way rarely 
 observed that their general health was in any way impaired 
 up to the third week, when a small proportion of the ani- 
 mals have died emaciated (five out of twenty-four). The 
 animals were killed at intervals from the first day to the 
 second month, as follows: 1st, 2d, 5th, 8th, 10th, l7th, 
 18th, 21st, 22d, 25th, 26th, 2'7th, 28th, 32d, 34th, 35th, 
 48th, 60th day. The organs carefully hardened in strong 
 alcohol, the lungs being filled through the trachea. Some 
 of the sections were stained in the usual way with hema- 
 toxylin and eosin, while others were stained for tubercle 
 bacilli. We have experimented in this way upon twenty- 
 four rabbits. 
 
 We deem it best to give a summarized statement of our 
 results, rather than a detailed account of the individual ex- 
 periments, because the effects of the dead bacilli vary con- 
 siderably, depending upon the amount introduced and the 
 size and vigor of the animal, and in many other less well- 
 defined ways. 
 
 If the animal is killed within twenty-four hours after 
 the injection of a moderate amount of dead tubercle bacilli, 
 the organs hardened in alcohol, and sections finally stained 
 for tubercle bacilli, it will be found that the bacilli are 
 present in by far the largest numbers, and in the largest 
 masses in the capillaries of the lungs. They are next most 
 abundant in the capillaries of the liver, and may be found 
 in small numbers in the spleen. 
 
 After the first three or four days we have been unable 
 to find bacilli in the spleen, but they appear to continue in 
 approximately the same numbers in the lungs and liver as 
 at first. We have made no examinations of the other 
 
THE ACTION OF DEAD BACTERIA. 27 
 
 viscera with a view of determining the presence or distri- 
 bution of the bacilli in them. 
 
 We need only refer here to the experiments of Wysso- 
 kowicz (12) and others on the rapid disappearance of 
 micro-organisms injected into the blood.* 
 
 Examinations of sections of the organs of animal$ killed 
 during the first and second week after the injection usually 
 show that while the bacilli have largely disappeared from 
 all the viscera, except the lungs and liver, they may still be 
 found scattered singly or in small clusters in the smaller 
 blood-vessels of these organs. Their situation can usually 
 be best made out in the capillaries of the liver. Here they 
 may be lying apparently free against the wall of the vessel, 
 or they seem often to be surrounded by a small quantity of 
 a nearly homogeneous or finely granular material which may 
 be fibrin or blood plates, or, finally, they may lie singly or 
 in small masses inclosed in variously shaped cells. 
 
 The earliest, and throughout the experiments the most 
 constant, lesion which we have found as the result of the 
 venous injection of dead tubercle bacilli is the development 
 in the lungs of from one or two to innumerable small white 
 nodules, some invisible to the naked eye, some as large as 
 two or three millimetres in diameter. These small nodules 
 may appear as early as the fifth day after the injection, and 
 persist up to the end of the second month. In the earliest 
 days they consist of a central denser mass made up of epi- 
 thelioid and giant cells, interspersed with and often sur- 
 rounded by masses of small spheroidal cells resembling 
 leucocytes. 
 
 Readily stained tubercle bacilli are present often in large 
 numbers in these lung nodules, especially in and among the 
 epithelioid and giant cells. Such a nodule with its bacilli 
 stained is represented in Fig. 1. 
 
 A limited area of lung tissue is replaced by these 
 nodules, and the air-vesicles about them show an increase in 
 the epitheliod cells, which sometimes completely fill them. 
 
 * The use of masses of dead tubercle bacilli in the study of multiple 
 embolism would suggest itself as a mode of experimentation on this 
 subject, since the differential staining enables us to locate the seat of 
 lodgment of the emboli with much greater accuracy than is possible 
 when small quantities of ordinary insoluble pigments are used. 
 
28 THE ACTION OF DEAD BACTERIA. 
 
 In animals killed at a later period these nodules are, as 
 a rule, denser in texture, and consist largely of epithelioid 
 cells and loose fibrous tissue (Fig. 2). Tubercle bacilli are 
 now, as a rule, present in very small numbers. 
 
 These scattered nodules in the lungs are the only lesions 
 which ,we have found up to the third week ; no changes 
 whatsoever having been found in any of the other viscera 
 before this time. 
 
 Animals killed in from three to five weeks almost in- 
 variably show a considerable number, often a very large 
 number, of the above-described white nodules in the lungs, 
 so that they look like lungs in acute miliary tuberculosis- 
 These nodules of new-formed tissue are usually much more 
 easily seen after the lungs have been distended with alco- 
 hol and hardened than before. Tn many cases animals 
 show no other gross lesions whatsoever, either in the kid- 
 ney, spleen, or liver. 
 
 But if a microscopical examination of the liver is made, 
 it will be found that in a considerable proportion of cases, 
 as early as the third week after the injection, swollen en- 
 dothelial cells are found in many of the capillaries, and 
 scattered, often very numerous, islets of new-formed cells 
 ramify in spreading clusters in the capillaries or form very 
 minute circumscribed masses, often consisting of but three 
 or four cells with lai'ge nuclei and faintly granular bodies 
 (Fig. 3). Sections showing these lesions and stained for 
 tubercle bacilli show almost invariably one or more bacilli 
 in these tiny new-formed cell masses. But free bacilli or 
 bacilli inclosed in single cells are rarely to be found in the 
 vessels, as they are in the first fortnight after the in- 
 jections. 
 
 In most of the animals killed in from four to six weeks, 
 in addition to the above described nodules in the lungs, the 
 liver was more or less thickly bestrewn with minute whit- 
 ish spots having the gross appearance of miliary tubercles. 
 Microscopical examination of these livers shows that, in ad- 
 dition to these macroscopical lesions, there are microscopical 
 lesions of similar character. 
 
 The simplest alteration that is found at these later pe- 
 riods consists in a proliferation of cells within the capillaries 
 of the liver, so that these are moderately and diffusely di- 
 
THE ACTION OF DEAD BACTERIA. £9 
 
 lated in places, or distended, so as to contain branching 
 nodules of epithelioid cells (see Fig, 4). It seems to us that 
 these new cells in the capillaries are in part at least derived 
 from the vascular endothelium (see Fig. 5). 
 
 Sometimes associated with this diffuse growth of new 
 cells in the capillaries, sometimes without it, the whole liver 
 is thickly sprinkled with minute, sharply circumscribed, 
 rounded nodules. Some of these nodules consist of epithe- 
 lioid cells in part sharply outlined, but in pait merging into 
 one another and intermingled with or surrounded by a zone 
 of small spheroidal cells (Fig. 6). 
 
 Many of the small nodules, on the other hand, consist 
 almost exclusively of small masses of epithelioid cells, either 
 merged together as in Fig. 7, or loosely packed and distinct 
 as in Fig. 8.- Again, the nodules sometimes consist of 
 a single small mass of epithelioid and giant cells, as in 
 Fig. 9. 
 
 Finally, the liver nodules may be more complex in struct- 
 ure and consist of a cluster of larger and sinaller epithelioid 
 and giant-cell masses, as in Fig, 10. We have seen no evi- 
 dence of proliferative changes in the liver cells- ; but these 
 are sometimes squeezed and flattened by the growing nod- 
 ules. So numerous are these small foci of cell proliferation 
 that not infrequently from two to ten of them may be found 
 within the area of a section of a single liver acinus. They 
 are usually situated within the acinus, more rarely in the 
 interlobular connective tissue. 
 
 In a large proportion of these liver nodules a varying 
 number of tubercle bacilli can be found (Figs. 3, 7, 9, 10). 
 But these become less and less abundant as time goes on 
 after the injection, so that after from a month to six weeks 
 we have found but a few scattered bacilli in most of the 
 nodules, and in a considerable proportion none at all. The 
 bacilli do not appear to bear any special relation to the 
 giant cells, as they so often do in genuine miliary tubercles. 
 They are often considerably broken and granular. 
 
 It is most noteworthy in this connection that the tuber- 
 cle bacilli are much more abundant, as a rule, before and 
 during the early stages of the growth of these cell masses 
 than they are when these have assumed the more distinctly 
 tubercular type. A¥e have found no lesions whatsoever in 
 
30 THE ACTION OF DEAD BACTERIA. 
 
 any of the animals, in the kidneys, and the tubercle-like 
 masses but rarely in the spleen. 
 
 Oar observations on the later stages of these new tissue 
 growths are not yet complete enough to enable us to state 
 what becomes of them when the animals are allowed to live. 
 Up to the end of the seventh week we have seen no evi- 
 dence of cheesy degeneration even in the largest and most 
 complex of the nodules. It is perhaps owing to the lack of 
 cheesy degeneration that the tubercle-like bodies are often 
 first revealed by the microscopical examination. 
 
 We do not mean to draw fast lines in assigning definite 
 periods at which these various lesions occur after injection 
 of the dead bacilli, bnt only to indicate in a general way 
 the averages of our experiments. 
 
 We have never seen the slightest indication of the pro- 
 liferation of the tubercle bacilli in the tissues of any of the 
 injected animals. 
 
 The introduction of the dead tubercle bacilli in larger 
 quantity and in as large masses as can be used without 
 killing the animals by multiple embolism leads to some 
 strikingly different results from those which we have de- 
 scribed ; but this series of experiments is not yet com- 
 pleted and need not be further considered here. 
 
 Summary. — We have found that prolonged boiling, 
 while often causing a considerable breaking up of the tu- 
 bercle bacillus, does not interfere with its characteristic 
 staining and does not alter the morpliology of many of the 
 individuals of a culture ; so that bacilli killed in this way 
 and introduced into the body can be readily recognized 
 at their seat of lodgment, even after the lapse of many 
 weeks. 
 
 Oiir experiments have shown that dead tubercle bacilli 
 separated from such of their metabolic products as are set 
 free in the culture media or are extracted by prolonged 
 bailing in water or fifty-per-cent. glycerin are capable of 
 inducing marked effects upon the body cells of the rabbit 
 with which they are brought in contact. 
 
 These dead tubercle bacilli are markedly chemotactic. 
 When introduced in considerable amount into the sub- 
 cutaneous tissue or into the pleural or abdominal cavities, 
 they are distinctly pyogenic, causing aseptic localized sup- 
 
THE ACTION OF DEAD BACTERIA. 31 
 
 puration. Under these conditions they are capable, more- 
 over, of stimulating the tissues about the suppurative foci 
 to the development of a new tissue closely resembling the 
 diffuse tubercle tissue induced by the living germs. 
 
 We have found that dead tubercle bacilli introduced in 
 small numbers into the blood-vessels of the rabbit largely 
 disappear within a few hours or days, but that scattering 
 individuals and clusters may remain here and there in the 
 lungs and liver, clinging to the vessel walls for many days 
 without inducing any marked changes in the latter. After 
 a time, however, earliest in the lung ; later, as a rule, in the 
 liver; a cell proliferation occurs in the vicinity of these 
 dead germs which leads to the formation of new multiple 
 nodular structures bearing a striking morphological resem- 
 blance to miliary tubercles. There is in them, however, no 
 tendency to cheesy degeneration and no evidence of pro- 
 liferation of the bacilli, but rather a steady diminution in 
 their number. It seems to us that the new structures origi- 
 nate in a proliferation of the vascular endothelium under 
 the stimulus of the dead and disintegrating germs. 
 
 Control Experiments. — It will no doubt appear to 
 others, as it did at first to ourselves, as if there must have 
 been some experimental error in the attempts to kill the 
 bacilli, so singular are the results that we have obtained. 
 But we think that the treatment to which the bacilli used in 
 these experiments have been subjected, together with the 
 tests of death to which we have exposed them, make it evi- 
 dent that we have actually used dead germs for our injec- 
 tions. , 
 
 This would appear certain, not only from the tests which 
 we have detailed above, but also from the fact that we have 
 made fresh boilings for each set of experiments (twelve in 
 number), and in most cases have prolonged the boiling for 
 from one to three hours beyond the period which has been 
 established as fatal to these germs. It should be remem- 
 bered that, according to the researches of Schill and 
 Fischer (13), the action of moist heat at 100° C. for fifteen 
 minutes suffices to kill tubercle bacilli. 
 
 The question now arises whether the?e curious effects 
 of dead tubercle bacilli upon the living tissues of the rabbit 
 are due to some special peculiarity of the bodies of these 
 
32 THE ACTIOX OF DEAD BACTERIA. 
 
 particular germs, or may be induced by other dead germs 
 or other substances introduced in the same way. 
 
 To satisfy ourselves on this point, we have made a series 
 of intravenous injections in approximately similar amounts 
 of boiled cultures of the Bacillus diphtherice of Loeffler (six 
 animals) ; of the Bacillus coli communis (five animals) ; of 
 tbe Staphylococcus pyogenes aureus (six animals) ; of wheat 
 flour (two animals) ; and of red pepper (two animals). 
 These substances have been introduced aseptically. The 
 details of these control experiments need not be given here, 
 since in none of them were there any alterations in any 
 way comparable to those so nearly uniformly induced by 
 the dead tubercle bacilli, and in most of them there was no 
 reaction or change at all. 
 
 The animals under experiment were strictly isolated and 
 kept under favorable hygienic conditions, so that the possi- 
 bility of a chance tuberculosis is largely barred out. 
 
 We must therefore conclude that the lesions caused by 
 the introduction of dead tubercle bacilli into the body of 
 rabbits are due to some peculiar property of the bodies 
 of this species of germ. 
 
 Remarks. — Two possibilities, at least, exist as to this 
 peculiar power of these dead tubercle bacilli. It may, of 
 course, possibly be due to some trace of poison elaborated 
 in the germ cell during its life and not given up in the 
 culture medium, or extracted by boiling in water or dilute 
 glycerin. Or, wbich seems more probable to us, in view of 
 the peculiar properties which the bacterio-protein of other 
 species of germs has been recently shown to possess, the 
 remarkable power of these dead germs may be due to the 
 specific proteid of the germ cells, which is gradually set 
 free as the germs disintegrate at their seat of lodgment in 
 the tissues. 
 
 Though we have made repeated attempts to isolate the 
 bacterio-protein from the tubercle bacilli by the method of 
 Necki, neither with this nor with any other method have 
 we succeeded. 
 
 It seems to us highly probable that the bacterio- 
 protein of the tubercle bacillus, either living or dead, when 
 set free by disintegration in the tissues, is capable of fur- 
 nishing that local cell stimulus which results in the forma- 
 
THE ACTION OF DEAD BACTERIA. 33 
 
 tion of the various phases of those structures whichi we call 
 tubercles aud tubercle tissue. The slow growth of these 
 structures would correspond to the exceptional structural 
 invulnerability of the tubercle bacillus. 
 
 These structures, which, as we have shown, we can in- 
 duce at will in the rabbit by the injection of dead tubercle 
 bacilli into the blood-vessels, are morphologically absolutely 
 typical of certain phases of miliary tubercles, even to the 
 presence of stainable tubercle bacilli in them. Miliary tu- 
 bercles these structures are, but — and this is of vital impor- 
 tance — the animals do not acquire tuberculosis. It is not 
 an acute infectious disease which we thus impart ; there are 
 no living germs to grow. The disease, if it is a disease, 
 is not indefinitely progressive. The dead bacilli seem to 
 act as foreign bodies simply, curiously stimulating, it is 
 true, but only dead foreign bodies after all. 
 
 There is no tendency that we can observe in these tiny 
 nodules to cheesy degeneration. 
 
 So far as we can see at present, when the limited 
 amount of stimulant or poison, proteid or other, which 
 they contained when we put them into the animal, is used 
 up, the active process is over, and it rests with the recupera- 
 tive powers of the animal to make what disposal they can 
 of the local ravages which have been caused. 
 
 The drawings of the liver of rabbits injected through 
 the veins with living tubercle bacilli by Yersin (14) show 
 at a glance the essential difference between the genuine 
 miliary tubercle of tuberculous rabbits and the non-infec- 
 tious aseptic tubercles of our experiments — namely, the 
 actively proliferating tubercle bacilli which make the lesion 
 in the first case a progvessive one, and which, sooner or 
 later, cause the death of the animal. 
 
 These studies would suggest the possibility that the 
 formative tissue changes which are characteristic of tuber- 
 culosis in man may be largely due to the action of the bac- 
 terio-protein of the tubercle bacillus set free as the germs 
 die and disintegrate in the tissues, while, on the other 
 hand, the equally important and in some ways more char- 
 acteristic action of the germ — namely, the coagulation ne- 
 crosis, and possibly some of the vascular changes and other 
 poisonous effects — may be chiefly induced by those freed 
 
34 THE ACTION OF DEAD BACTERIA. 
 
 metabolic products of the life processes of the germ which 
 the experiments of Koch have recently brought to light, or 
 to others as yet unknown. 
 
 We resist the temptation to elaborate here the many 
 fascinating possibilities which these studies suggest as to 
 the rationale of tubercular inflammation in man, and the 
 closer analysis of its lesions which may be possible in the 
 future. 
 
 If the results of our experiments are confirmed by 
 others, it is evident that our conception of the significance 
 of the various lesions of tuberculosis may be considerably 
 modified and the therapeutic possibilities made essentially 
 clearer. It will be quite comprehensible why so many well- 
 defined miliary tubercles, so many masses of tubercle tis- 
 sue, so many destructive lesions of chronic phthisis, and so 
 many so-called scrofulous and tubercular lymph nodes fail 
 to show the presence of tubercle bacilli if it is established 
 that the morphological characteristics of these struct- 
 ures are dependent upon the disintegration of the bacilli 
 which once found lodgment where they form. 
 
 We are aware of the danger of drawing far-reaching 
 inferences from a single series of experiments. But we 
 think that we are fairly justified, from the experimental 
 data now set forth, in the conjecture that we may have here 
 the explanation of the comparatively slow growth of tuber- 
 cle and tubercle tissue which is observed under a variety 
 of conditions. 
 
 One may conjecture, too, whether a certain number of 
 the miliary tubercles which we find in the body after a gen- 
 eralization of the tubercular process from an old tubercu- 
 lar focus may not actually be innocuous growths, or at least 
 harmful only as foreign bodies in the tissues where they 
 develop. 
 
 The possibility would, furthermore, suggest itself that 
 those dense masses of fibrous tissue in the lungs which we 
 usually regard as evidence of healed tuberculosis may actu- 
 ally be innocuous, even though they harbor stainable 
 tubercle bacilli, and no longer capable of lighting up a 
 fresh infection, as is commonly supposed. 
 
 One may legitimately qiiery, too, whether in the closel" 
 analysis of the tubercular processes in the future we may 
 
THE ACTION OF DEAD BACTERIA. 35 
 
 not find that the formation of the new tissue, the develop- 
 ment of the tubercles themselves, is, after all, an attempt 
 at a conservative process on the part of the organism in- 
 cited by the stimulus of the disintegrating- bodies of some 
 of the germs. It may be, too, that, as Hueppe suggests, 
 the conservative attempt is often rendered futile by a de- 
 structive tendency — cheesy degeneration — developed under 
 the influence of a metabolic product of the living and grow- 
 ing germs which the tubercular foci still harbor. The 
 bearing of this possibility upon the administration of large 
 doses of Koch's tuberculin should not, perhaps, be wholly 
 ignored even at this stage of our knowledge. 
 
 It is somewhat curious to notice how closely in accord 
 with some of our experimental results are many of the feat- 
 ures of Hueppe's (15) recent essay to develop a rational 
 explanation on theoretical grounds of the varied phases of 
 the development of tubercles and tuberculosis. 
 
 We wish, in conclusion, to call renewed attention to 
 the fact, too often overlooked, that the acute infectious dis- 
 eases are much more complex, both in their lesions and 
 their aetiology, than we are wont to realize. We were for- 
 merly perforce content when we had done what we could 
 to make plain the morphology of their obvious lesions. 
 Light dawned at last on the relationship of bacteria to some 
 of these diseases, and the definite establishment of the con- 
 stant and inevitable relationship between some specific germ 
 and its special disease was at first so satisfying that this 
 relationship alone was regarded as a solution of its fetiology. 
 
 But the restless seeker after detail was not long content 
 with this, and new light was speedily forthcoming as to 
 the important role which chemical poisons, elaborated by 
 pathogenic germs, had to play in the induction of the varied 
 phenomena of this class of diseases. We have seen that a 
 new phase of our knowledge is being now developed in 
 the study of the proteid ingredients of the germ bodies. 
 
 But beyond all these lines of study and essential to 
 their full fruition is the development of a fuller comprehen- 
 sion of the powers and reactions and idiosyncrasies of the 
 body-cells themselves, upon whose comportment in the last 
 analysis depend the distinctive features and the ultimate 
 siacnificance of the acute infectious diseases. 
 
36 THE ACTION OF DEAD BACTERIA. 
 
 Thus the harvest garnered in this newly opened field of 
 thought and research must ultimately be brought home to 
 the old storehouse of pathological anatomy and physiology 
 before it can be made in the fullest degree available for the 
 understanding and curtaihnent of bacterial disease. 
 
 1. Prudden. New York MefUeal Journal^ Jiine 6, 1891. 
 
 2. Haramerschlag. Rev^Ctrlhl.f. Bacteriologie, etc., March 
 2, 1891, p. 272. 
 
 3. Hueppe and Scholl. Berlin, klin. TF(?cAe?^^c7i?•., 1891, Jan. 
 26, Feb. 23, March 16, March 23. 
 
 4. Trudeau. Medical Record, Nov. 22, 1890. 
 
 5. Zulzer. Berlin. Tclin. Wochenschr., 1890, No. 4. 
 
 6. Crookshank. Lancet, Feb. 27, 1891, p. 300. 
 
 7. Weyl. Deutsch. med. Wochenschr., Feb. 12, 1891, p. 256. 
 
 8. Maffacci. CtrlU.f. allg. Path., etc., Dec. 15, 1890. 
 
 9. Wyssokowicz. Mitth.a. Dr. Bremer''sHeilanstalt,u.s.w., 
 neue Folge, 1890. 
 
 10. Koch. Reprint, Medical Record, Jan. 17, 1891, p. 85. 
 
 11. Hneppe and Scholl. Berlin, klin. Woehenschrift, Jan. 
 26, 1891, p. 89. 
 
 12. Wyssokowicz. Zeitschrift fur Hygiene, Bd. i, p. 3. 
 
 13. Schill and Fischer. Mitth. a. d. kaiserl. Oesund- 
 heitsamte, Bd. ii. 
 
 14. Yersin. Annales de VlnstitUt Pasteur, 1888. p. 140. 
 
 15. Hiieppe. Berlin. Min. Wochenschr., M;trcli 23, 1891. 
 
^-^i^ v,^; "'^^ 
 
 
 CM 
 
 r--