Report of Dr. Avery (assisted by Drs. Adams, Coburn, Curnen, Dubos, Goebel, Ccodner, Hot&kiss, MacZeod, Mirick and Stillman) Relation of the presence of the C-reactive protein to other evidences of disease activity (Curnen, Mirick and MacZeod). It has been shown previously that the presence of the C-reactive protein in the blood of patients during a variety of infections closely parallels the clinical course of the disease. During the acute phase of illness the precipitation test with C polysaccharide is most marked, and diminishes In intensity as convalescence is established. A specific antiserum prepared in rabbits by immunization with the C-reactive pro- tein of human blood has provided another means of toSting for the presence of this abnormal protein. As previously pointed out the immunological reaction aith immune rabbit serum is far more sensitive than the precipitation test with the C polysaccharide in detecting small amount6 of the reactive protein. For example, if a patient's serumknown to contain the reactive protein 1s diluted 1:4C the precipitation test with the C polysaccharide may not be demonstrable. HOVJeVer, , if the same acute phase serum is diluted 1:600 the presence of the reactive . prOt8in can Still be detected by means Of the Specific Anti68XWII (anti-CP serum). ~ Use has therefore b88n made of the immunological reaction In determining how long the reactive protein pereista in the blood of patients during the course of conval?scence. : , ._, ,. . . An attempt has been made to correlate the presence of the reactivd. protein during the acute disease and in convalescence not only with the state of the lesion itself but also with various clinical tests used as a measure of disease activity, .particuiarly the erythrocyte sedimentation rata and leucocyte count. In a series of patient6 admitted to this hospital suffering from VariOUS acute infection6 of the respiratory tract, frequent determinations of the erythrocyte sedimentation rate have been made, both during the acute phase of 132 disease and in convalescence. Estimations of the amount of C-reactive protein have also been made on the same specimens of blood by the precipitation test with, the C polysaccharide end the immunological reaction with anti-CP serum. The c-reactive protein is present in the blood of pneumonia patients soon after the' onset of the disease. In some cases its presence in considerable amount precede6 the increase in sedimentation rate. Likewise, In certain severe in- fections where the sedimentation rate remains normal, the precipitation reaction with the C polysaccharide msy be strongly positive. In a number of instances late in convalescence when the patient's clinical condition has apparently re- turned to normal, the sedimentation rate has remained rapid, while only small traces of reactive protein were present In the blood as measured by the reaction tith anti42 serum. The reaction with the test carbohydrate becomes negative in patients with pneumonia Then the temperature returns to normal. However, if tested with anti-CP semm the presence of the reactive protein may be demonstrated in the blood for a variable period after the reaction with the C polysaccharide has become negativo. In certain other patients Tvho68 sera have not reacted at any time with the C polysaccharide, the presence of the reactive protein may be readily detected with specific antiserum. The lack of parallelism between the titer of the C-reactive protein and the increased sedimentation rate applies also to the leukocyte count. The reactive protein has invariably been demonstrated in the blood during infections associated with leukocytosis, and in a general may the presence and amount of thi6 protein parallels the degree ef 18UkOCytOSiS, Moreover, in lobar pneumonia, ? ? o ES the leukocyte count falls during convalescence, so also the amount ?? reac- tive protein in the blood diminishes, although it may be detected for a consid- erable period after the leukocyte count has returned to normal. In this con- nection it is worthy of note that in acute infections associated with 18ukOpenia the reactive protein mey be demonstrable in the blood. 133 From the data 'thus far obtained it P.:')Trxrs that the titer of C-reactive protein in the blood during acute infections parallels both the clinical course ad the o,c tivity of tha lesion more closely than does cithcr the crythrocyte -cdimcntatlon rate or the leukocyte count. The findings also suggest thnt the !cttcction of the rcsctivo protein by moens of the C polysacchnrido and anti-CP ::crum may be useful CliniCFilly a6 nn index of disease r!c tivity, pnrticulP.rly during Infections of a polycyclic nature such RS rheumtic fcvcr, tUberCUlOSiS, itc. Development of drug-fastness in ~neUmOCOCCi isolated from pr?ticnts trcatod with sulfonmide drugs (Mlrick nnd McLeod). Strains of pneumococci (Ancountered clinically show considerable variation in sensttivtty to the bac- teriostatic action of tho sulfonamide drugs. In a few instances It has been reported that druefnstnees may develop during the course of treatment but an ndequp.te study of this important phenomenon ho.6 not been made by the more prc- cise techniques rocontly developed in this laboratory. The presence in ordinary culture medium of inhibitors to the sulfon- amide drugs may so affect the outcome of bactcriostatic tests that the actual sensitivity of various strains may be impossible to det8rmine by the usuti cultural methods. To overcome thie difficulty an infusion of froeh liver has been use d in tha present experiments, since this medium is not only free of sulfonamide Inhibitor but is capable of supporting the luxurient growth of small inocula of pneumococci. The bactoriostatic tests have been performed by seeding A standard lnoculum of young, actively growing, pneumococci into tubes of the liver infusion containing varying dilutions of sulfapyridine. Pneumococci have been isolated from twenty-two patients rpith pneumonia before or early in treatment as well as after receiving a full course of a sulfonamide drug, and the rnspective strains have been compared for their sensi- tivity t0 6Ulfapyridine. In ton instances there was PA increase in drug 134 resistance of the organism isolated after treatment, averaging a fourteen-fold increase in the concentration of drug required for complete bacteriostasis. In ei?Jlt cases there was no change. In four cases the culture obtained after treat- ment was more susceptible to sulfapyridine than the original, bacteriostasis being first final effected,by one-third as much drug as was necessary for the strain when isolated, An exception qas a case of pneumococcal endocarditis where the culture was fifty times more susceptible than the original. The greatest increase in drug resistance, was seen in those cases receiving prolonged and con- tinuous treatment. A comparison of the in vitro test and the clinical response of the -- patient showed that eleven strains of pneumococcl from thirteen patients who responded poorly to the drug grew well in a concentration of l/10,000 sulfapyri- dine and five of these in l/5,000 indicating definite drug-fastness of these strains, The strains Isolated after treatment from twonty out of twenty-two patients who showed a good clinical response to the drugs were inhibited by a dilution of sulfapyridine of l/20,000 or higher. The difference In drug sensitivity shown in vitro between three sus- -- ceptible and three resistant strains of Pneumococcus Types I, III, and IV was experimentally confirmed In viva by comparing the therapeutic response of in- -- fectod mice to sulfnpyridlne treatment. Further studies are planned to determine the conditions under which pnaumococci develop roeiatance to tho sulfonamide drugs, and the nature of the alterations brought about In the microorgmisms when' this phenomenon occurs. Studies on sulfonamide inhibitors by the use of a soil bacillus cp?able f destroying parRpe.mknobonzoic acid (Mirick) . The action of the sulfonamide drugs both in vitro and in the animal body appears to be closely related to the -- occurrence of substances which inhibit their bncterlostatic effect, It has boon 135 shown in this lc?bora.tory that inhibitory substances arc present not only in the tissues ,and in certain fluids of the animal body, but also in bacteria. In some tissues, notably liver and kidney, inhibitor is not demonstrable until nutolysis has taken place or the tissue has boen subjected to mild acid hydrolysis. The inhibitor present in urine is likewise not demonstrable unless acid hydrolysis has been used. Yeast and ell the bacterial species studied have bocn shown either to contain inhibitor in the cells or else to produce such a substance in the culture medium in which they have been grown. The amount of inhibitor produced by bacteria varies considerably from one. microbial species to another and between different strains of the same species. A strain of Pneumococcus Type I which had been rendered sulfapyrldine-fast in vitro has been shown to produce approximately -- ten times as much Inhibitor as the susceptible pnront strain from which it was derived. Howover, the increased production of sulfonamide Inhibitor by pneumo- cocci is not the only factor involved In drug-fastness, since some strains which have become relatively fast In viva do not produce any more Inhibitor than do -- susceptible strains. Woods has shown that p-aminobenzoic acid acts ae a powerful sulfonamide inhibitor, and Rubbro and Gillespie have recently isolated thie compound from yeast. It has been suggeatod by ~lldee and Voods that p-aminobenzoic acid is an essential metabolite for various btkt&ia t&d that the sulfonamide druge act by competikvi inhibition of the enzyme systems concerned 4th its metabolism. More recently it hae also been ehomn that p-aminobenzoic acid is an accessory growth factor for Clostrldium acatobutylicum, as well as for chicks, and Its role as an anti-grey hair factor for rodents has been demonstratod by Ansbacher,- ' Although extracts of various organs and of bacterial cells exert a pronounced Inhibitory offoct on the sulfo%!!ido drugs, it by no means follows - 136 that the active sutistancos @-Abiting this property are in all instances identi- c:ril vri th p-aminobenzoic acid, Certain of the chemical properties of the natur- ;tlly occurring inhibitors lend sup)>ort to this view, Consoquently a method vas sought which would diffcrentinto botvoen the inhibitors from various sources Athout recourse to the laborious prccedures of chemical isolation and identi- fication. Although by the very sensitive diazo reaction, p-aminobenzoic acid 1::~ bo detected in concentrations as small as one part in a million, this com- 1 cund is active as a sulfonamide inhibitor in A dilution 100 times greater than thst which can be detected by digzotization, Moreover, the diazo reaction is not L;pecif ic for p-aminobenzoic acid. The possibility suggested itself that there might occur in nature micro- ..rganisms possessing enzymes capable of so altering p-aminobenzoic acid that it rA3 longer is active as a sulfo,namide inhibitor. 9e have isolated from alkaline ::~il a bacterium capable of modifying p-aminobenzoic acid so that this com:>ound no longer gives the diazo reaation and is comj>letely inactive as a sulfonamide inhibi tar, Follc~ing the technique so successfully employed by IYubos in the adaptive production of other bacterial enzymes, the isolation was accomplished by inoculating alkaline soil into a simple basal medium to which p-aminobenzolc acid f7as added as the sole source of carbon and nitrogen. The soil bacterium capable of annulling the action of p-aminobenzoic acid is a small, motile aerobic, Gram negative bacillus. It is not pathogenic for mice, ferments no sugars, and produces an alkaline reaction fn litmus milk. It grows abundantly and exerts'its action on p-aminobenzoic acid in a synthetic mineral medium containing an acid hydrolysate of casoln, as Tell as in ordinary la.boratory media. Its action is not entirely specific, holnover, since it readily. attacks acetylated p-aminobenzoic acid, novocaine, and also acts, although nt a sloaer rate, on o-aminobenzolc and p-aminophenylacetic acids. It is inactive on the meta form as me11 as the methyl and ethvl esters of p-aminobenzoic acid, 13rl arsanildc acid, sulfanilic acid, and the sulfonamide drugs, Sulfapyridine 1:500,000 is bacteriostatic for the soil bacillus, and although it attacks p-amincbenzoic acid it is worthy of note that the latter corn- pound is itself bacteriostatic for this bacillus. This statement at first glance appears paradoxical, However, a similar phencmenon has been observed in this laboratory in the case of the Inhibition of growth of staph,vlococcus by hydrogen peroxide despite the fact that these bacterial cells contain an active cntalase capable of decomposing this toxic compound. The action of the bacillus, on p-aminobenzoic acid is apparently accon+ plished by means of adaptive enzymes. Vashed bacterial cells vllhich have been grown in Inhibitor-free synthetic medium are inactive. However, if cultivated in the presence of p-aminobenzolc acid the cells now contain potent enzymes cap& ble of attacking this compound. The action of the soil bacillus on some of the naturally occurring sulf onamlde inhibitors has been studied. The inhibitory action of yeast extract is destroyed, confirming the evidence of others that yeast cells contain p-amino- benzoic acid. However, the inhibitor found in the cells of Streptococcus hemoly- ticus (Group C) is only slightly affected, and that occurring in peptone broth is not altered. In liver Infusion devoid of free sulfonamide Inhibitor, the growth of the soil bacillus gives rise ta a potent inhibitory substance. In the lrctter instance, the inhibition is apparently not due to the presence of coenzymes, which have been shown to be Inhibitory under certain circumstances, since the inhibitor formod in liver infusion is not inactivated by treatment knom to destroy coenzymes, The nature of the mechanism is unknown whereby the soil bacillus in- activates pkninobenzoic acid pnd certain related compounds. However, the change produced is not due to acetylation or to dcamination alone. 138 The observations thus fn.r mndo indicate thr?t the use of this soil bncillus in which the enzymo systems have been specifically oriented, affords a very useful method for the further study of the nature and action of the sulfona- mide inhibitors and for the detection of minute amounts of p-eminobonzoic RcI~. The effect of sulfapyridino upon the development in rnbbits of specific immunity to pneumococcus (Curnen and MacLeod). It hns been m4ntaincd thr?t tho effectiveness of tho sulfonamide drugs in the thcrnpy of pnewnococcal infections in man is unrelated to tho development of specific immunity to the inva-ding microorgnnism. Tho evidonco for this view is bnsc?d upon observrtions that r?t tho time of defervnsconce circulating r?ntibodies are rnrely domonstrablo in the blood of patients trotted with tho drug. Provious experiments in this laboratory have indicntod that effective sulfapyridino therapy of evarimental pneumococcal infection in mice is dependent upon develoqment of a specific immune response, If drug therapy is discontinued beforo sufficient time has elnpsed for active immnity to dovelop the animals do not survive. The apparent discrepancy between the clinical and experimental obscr- vations was thought to be attributable to differences in the methods used for detecting the immune response. In patients recovering from pneumonia tho de- velopment of immunity is recognized by omsloying serological techniques to demonstrate antibodies in the circulating blood. In animals, however, the presence of active innnunity may be detorminod experimentally by infection e.t an appropriate interval following previous infection or prophyle,ctic vaccination. Numorous studies have shown that under such circumstances active immunity to pneumococcal infection appears in both mice and rabbits before type specific antibody is demonstrable in the blood. The present study was undertaken in order to establish whether or not 139 the admfnistration of sulfapyridfne alters the development of immunity to pneumococcus in rabbits and to compare the relative sensitivity of various methods for the detection of the early immune response, The ezqeriments were carried out In rpbbits because of their suitability for the study of both active resistance and the presence of humoral antibodies. The animals were first vaccinated and after Intervals of 48, 72, and 96 hours thereafter comparable groups pere infected to determine the degree of active resistance that had developed. Prior to vaccination and at Intervals of 24 hours until just before Infection, blood pas drawn for serological studies which included tests for agglutinins, precipltins , and mouse protective antibodies. Each rabbit was vaccinated intravenously. by a single injection of heat-killed Type I pneumococcl equivalent to 10 cc. of broth culture. Half of the vaccinated rabbits In each of the groups previously designated received sulfapyrldlne during the period when immunity might be expected to develop, that is, cainoiding with and following vaccination. The drug was given by stomach tuba in divided doses totalling 4.5 grams for each animal. Admi ni s- tration of the drug was scheduled to terminate 80 that at the time of infec- tion no remaining lffreetl sulfapyridine could be detected in the circulating blood. All vaccinated animals including those which had and those which had not been given sulfapyridine as well as an appropriate number of un- vaccinated and untreated controls were infected by, the intradermal injection of a rabbit virulent strain of Type I Pnomococcus. The infecting dose was invariably fatal for the control animals. The course of infection including the occurrence and degree of bacteremia YEB observed and recorded for each rabbit. Regardless of the interval botwecn vaccination and infection, there was nr, significant difference in the immune response of the rabbits which 140 had received sulfapyridine as compared with those which had not. Rabbits infected as early as 46 hours after vaccination irrespective of whether or not they had received sulfapyridine showed that pctive immunity of a type specific nature was already present at this time. Although the course of experimental infection was usually severe and accompanied by bactcromia 80 per cent of those animals survived, Further dovelopmcnt of active It+ munity was manifested in the rabbits infected 72 hours after vaccination by a mllder.clinical course and a lomr incidence of bnctoromin, although the survival rate was tho some as in the procoding group. Of the rabbits In- fected 96 hours after vaccination few showod bactoremia and all survived. Although demonstrable antibodies had not appeared in tho sora of rabbits earlier, practically all of tho sora obtained 96 hours after vaccination CON ferrod protection on mice, and about half of these scra contained agglutinins and a fow contained precipitin in low titer for the type specific polysac- charlde, From those observp.tions it is ovidont that in rabbits the immune response to pneumococci is not lnfluonocd by the Rdministretion of sulfa pyridino. Moreover, the development of active immunity considerably pre- cedes the earliest appearance of antibody in ths circulating blood. Tho oxperiinental findings lend further supgort to the view that in man effective sulfapyridine therapy Is intimately nesociated with the development of active immunity, Epidemiologic4 studies ( Stillman), In continuing an apidemio- logical study of the disseminp,tion of disease producing types of pnoumococcl, experiments have been carried on tc dctormine how long these organisms will survive after being rapidly dried in the frozen state. It was found that the desiccated cells may remain viable for at least three ycnrs provided moisture is prevented from coming in contact with the dried material. Marc recently, 141 living organisms have bcon recovr?rmd from a specimen of.driod cells preserved in hermetically sealed vials for a period of f fve years. In all instances the recovered strains have exhibited all the distinguishing characteristics of the ortginal culture including capsule formation, type specificity, and virulence. In this conllectlon it is intcrosting to recall that although pneumococcl are not generally rognrded as capable of surviving long under the naturally more rigorous conditions of the outside aorld, they have been isolated directly from dried pneumonic sputum which has been exposed to air and diffuse light at room temperature for from 4 to 8 weeks, pncumococc i preserved in dried rabbit blood and similarly exposed hsvo been recovered a month later without suffering any loss in virulence or type specificity. Duplicate epeclmens similarly dried but stored in tho cold have yielded living pneumococci when cul tfvEtod a year later. The capacity of individur.1 strains of the same type to survive undor these conditions varies, and there are distinct dlfforences in this regard botmeen different types of pneumo- coccus. Studies on capsular synthesis by pneumococci (E4acLood and Avery). The capacity of pneumococci to grow and produce disease in the animal body is conditioned by, if not rdnslly dependent upon, the activity of the particular group of enzymes concerned in the synthesis of the cell capsule, Capsular synthesis is most highly developed crud the product of its titlvity most pronounced in cells best adaatod to growth in the animal body. The presence In the capsule of a chemically unique end sorologically reactive polysac- charide confers upon the cell a highly selective specificity which m&es possible the differentiation of sharply defined and specific types nithin the species. The enzymes responsible for capsular synthesis can be reversi- bly inactivated by known changes in environmental conditions rrithout impair- 142 ing the viability of the microorganisms. The selective inactivation of this particular function results in the loss of capsules, together with the con- sequent loss of type specificity and invasive properties. Under these con- ditions hi&ly pathogenic microorganisms are reduced to a state in which they are no longer capable of inducing disease in animals highly susceptible to fatal infection with the originally encapsulated parent strain. Important and essential as capsular synthesis is to the pathogen- icity of pneumococci in the living host, this function is not vital to the life and growth of the microorganism outside the animal body, since cells in which the formation of capsules has been inhibited are still capable of carrying on the vegetative processes of metabolism and multiplication in artificial media. Capsule formation may be regarded therefore as an adaptive mechanism whereby the bacterial cell seeks to protect Itself against the defense reactions of the host. It has long been recognized that pneumococci can exist as in en- capsulated or non-encapsulated forms which in terms of bacterial dissocle,tion are referred to respectively as S and R variants. Under suitable conditions this fo'rm of d$ssociation ( S&&Z) is reversible and may occur spontaneously or be experimentally induced in response to environmental influences which favor either activation or inhibition of the capsule-producing enzymes, By appropriate methods these changes can be brought about In vitro. For example, -- it is possible to inactivate capsular synthesie in pneumococci of any specific type and to derive thereby the non-encapsulated R variant; conversely, it is also possible to reactivate these sane R cells under conditions such that they invaric,bly revert to the original type `and regain all the specific character- istics of the parent strain from which they were derived. In the study of the mechanism of capsular synthesis the phenomenon 143 of the transformtion of specific types .of pnoumococcus lends Itself more readily than does the direct form-of revorsiblo dissociation. Griffith first showed in mice that encapsulntcd ~011s of one specific type mny be transformed into other specific typos through the intermcdi?.te stage of the R forms. Dawson found subsequently that the use of animal inoculation, by growing R cells in a medium tract prepared by dissolving S trs.nsform~.tion may be effected in vitro without -- and ~lloway succeeded in c,?using trsnsformPtion containing small amounts of a filtered ex- pneumococci by means of desoxycholate. It is possible therefore to study the properties of the specific activator of cap- sular synthesis in a soluble form entirely free of bacterial cells, and to define the conditions under which it exerts its greatest activity. In tho preparation of potent extracts it is essential to work under conditions which limit as far as possible the action of the autolytic onzymos of the pneumococcal cells. Autolysis, whether occurring spontaneously or brought about by the action of tho extracted enzymes, is kno*?l to destroy the transforming principle. Since the autolytic system is a mixture of various enzymes an attempt was made to determine if possible what particular enzyma is involved in breaking down the activity of extracts. For this purpose selective substances known `to inhibit certain enzymes were tasted. MacLeod first showed that the `transforming substance was not ix+ paired khen cell auiol$ds takes' place `in ' the .presence of fluoride. Although under these conditions celi dl*sintegratidn occurs, such `autolysates are still potent `in inducing transformation. Arvparently fluoride'selectively Inhibits the enzymes responsible for the destruction of the transforming principle without impairing the activity of the enzymes which bring about cell lysis. Cell-free extracts of encapsulP,ted pneumococci prepared by dis- solving the bacteria in sodium desoxycholate at low tem?ernture are immcdiate- 144 ly heated at'65cC. for 30 minutes in order to destroy the bacteriolytic enzymes. Sodium desoxycholate is removed by precipitating the extract in alcohol , in which the desoxycholate is soluble. The alcohol precipitate is then dissolved in saline at a slightly alkaline reaction and the bacterial protein removed by shaking repeatedly with chloroform according to the method of Sevag. Extracts prepared in this manner contain considerable amounts of nucleic acid. The latter substance may be almost completely removed by digestion with crystalline ribonuclease without affecting the transforming potency. Provided the various purification procedures are carried out with- in a fairly narrow pH range, little of the specific transforming activity is lost, and protein-free and lipid-free extracts have now been obtained which will effect the complete trangformqtion of R cells to the S form when used in amounts corresponding to 1.0 cc. or less of the original culture from which the extract was prepared. The transforming activity of purified extracts has been found to be resistant to the action of the crystalline proteolytic enzymes, trypsin and chymotxypsin. A purified phosphatase preparation from swine kidney `did not sffect the transforming principle. Pho sphatase preparations from rabbit bone and dog intestine, on the other hand, quickly destroyed the transforming activity, In addition to their phosphatase activity both of the latter ~ enzyme solutions showsd considerable esterass activity, whereas the kidney phosphatase preparation was free of this enzyme, In this regard it is of interest that sodium fluoride, which is known to inhibit the action of ester- ases, prevents the destruction of the transforming principle by tho bacter- iolytic enzymes of pneumococcal cells. As noted above the transforming activity is not affected by digestion with crystalline ribonuclease. 145 Normal human serum contains a.n enzyme which is capable of quickly inactivating the tr`ansforming principle. Cn the other h;ind, serum obtained from patients during the acute phase of vgrious infections doos not destroy the activity of potent bacterial extracts, The lack of activity of acute phase serum is probably attributable to the presence of an antiformcnt which inhibits the activity of the serum enzyme, From the cvidencc obtained by the use of various enzyme prepara- tions it is not held that the spcciflc transforming principle is neccssari$y of the nature of an ostor, inasmuch as other enzymes may hnvo been present. However, the evidence obtained by the use of sodium fluoride as an enzyme inhibitor when taken in conjunction with the other studies, suggests that the transforming principle mny be an csterified compound. Vhatover may ultimately prove to be the nature of the transforming principle, ono of its most striking chnractcristics is the type-specificity of its action on Ii cells. The mechanisms involved in the phenomenon of trans- formation and the nature of the activating principle are still undetermined. However, the results indicate that all X cells in nfiich retrogression has not advanced too far, possess a potential but inactive system of enzymes capable of synthesizing any of the type specific polysnccharides. Once the capsular function has been specifically activated, the newly transformed cells con- tinue to synthesize the same capsular material and retain their type spocif- lcity through innumerable transfers on artificial media. The study is being continued with the hope that knowledge of this important cellular mechanism may lead to a better understanding of the prin- ciples Involved in certain induced variations of living cells, not only of Pncumococcus, but also those of other biological systems. Furthermora, it is possible that knowledge portnining to the nature of the substances which serve as activators and inhibitors of the capsule-producing enzymes might afford a specific apprcach to the suppression of the capsular function, upon the activity of which the pathogenicity of pneumococcus depends. . The reduction of Fe+* iron as a measure of bacterial metabolism (Cubes). Bacterial metabolism has often been studied by fcllowing the rate of reduction of methylene blue or other reducible dye; this method does not lend itself to quantitative analysis and also suffers from the fact that dyes are toxic for several type6 of enzymes. It is known that the Fe+*& Fe* system possesses a very high redox potential #and ten therefore respcnd tc most of the oxidation-reduction reactions associated with metabolism. Inorganic Fecy'orms ni th .~a ' dipyridyl, a red complex which per- mits very sensitivo and accurate determination of the former substance. V-e have found that, in the concentrations required to titrate the iron reduced by bacteria, PWU `dipyridyl does not exhibit any toxicity for these cells. It is possible, thereforo, by addlng Fe*' brig ra ~dipyrldl to a cell suspension, to obtain a ccntinuous record of the reduction reactions going on in the system, In fact, it has been found that the production of the red Fe* -m-c * `dipyridyl complex in a metabolizing cell suspension is proportional to time and to the number of cells and can serve as a quantitative measure of metabolism. Bactericidal substances produced by aerobic sporulnting bacilli (Dubos, Hot&kiss, and Little). The culture of aerobic sporulating bacillus which produces the bactericidal subst,ances described in previous reports has now been identified as a strain of B. brcvis. In fact, several different strains of this species have been isolbted from soil, sewp.ge( cheese, etc., ?r obtained from culture collections and all have been found to exhibit antagonistic activities toward other microorganisms, Cultures of 13. brcvi s 147 have yielded two crystalline products, gramicidin rind tyrocidinc,, both of which are endoTed with bacteriostatic and bactericid.al propprtins. Tyrocidine is R polypeptido with n free amino group; it is bat- terisidal for all microorganisms so far tested, sith the possible exception of the tubprclc bacillus. In fact, it behaves like a gcnerel protopksmic poison, analogous to the c?tionic detergents which it resembles by its basic character and ability to depress surfnco tension. Although very m.ctive in vitro against Gram positive and Gram negritive organisms, tyrocidinc is -- completely inactive in vivo p.gninst Gram negative infections, pnd affords only -- very limited protection against infection of mica with pnoumococci. As stated in earlier reports, grnmicidin exhibits a remarkable selective activity against Gram positive bacteria and, mhen used under the proper condl tions, retains much of its activity in viva. Gramicidin in jcc ted -_I_ intraperitoneally into mice exerts n protective and curative action against streptococcus and pneumococcus peritonitis; its efficacy has also been demonstrated in the case of a naturally occurring localized Infection, namely bovine mastitis. This disease, caused by group B streptococci, remains localized in the udder of the infected animal. It has been found that the Injection of grwnicidin directly into the infected qurrter through the tept canal often results in the rapid, complete and permanent disappearance of the streF,tococci, Particularly successful results were obtained when gramicidin was injected in emulsion with minorti oil, thus minimizing the Irritation resulting from the trostmcnt, Although gramicidin is therefore effective in vivo for the local -- treatment of localized Gram positive infections, it is practically inoffec tivo rvhen administered by the intravenous, subcut,aneous, or intrqmuscular routes; as will be shown later, the lack of effectiveness by systemic administra.tion 148 may be due to the presence in the blood of a substance - cephalin - which is a specific inhibitor of gramicidin. Studies on the chemical nature of bactericidal substances prepared from cultures,of B. brevis (Hotchkiss and Dubos). The bactericidal material produced by B, brevis contains the neutral Substance'gra-Iicidin and a weakly - basic substance designated as tyrocidine. Both gramicidin and the hydrochlo- ride of tyrocidlne have been obtained crystalline and in highly purified form. Substances previously described as graminic and gramidinic acids have been recognized as tyrocidine, crystallizing without the full complement of hy- drochloric acid. Tyrocidine is apparently a polypeptide containing a free amino group and having a carbolryl group combined as an amide (40-m). The chlo- ride content of the hydrochloride indicates a molecular weight of about 1300. Of approximately eleven amino acid residues, one is tryptophane and one tyrosine. York has continued on the chemical nature of grpmicidin. A molecw lar weight of approximately 1400 Gas found by use of a physico-chemical method which may have entailed an error of considerable proportion. Con- sequently, in tho chemical degradation of gramicidin by hydrolysis, an at- tempt has been made to obtain more accurate indications of the moleculsr weight. The tryptophane analysis together with other results allow a choice of two values of the order of 1000 and 1600. A number of statements can be made which will be true of either unit: Cyclic tryptophane nitrogen atoms and aliphatic amino nitrogen bound in peptide linkages account for the tote1 nitrogen present in the molecule. One amino group per molecule is derived from a substance which is not a. typical x-amino acid; the remainder bolongs to m-amino acids (present in psrt as &- and in pp.rt as $F or "unnature,l* amino acids as previously reported), One half of the