STUDIES ON CULTURAL REQUIREMENTS OFBACTERIA. I

J. HOWARD MUELLERFrom the Department of Bacteriology, College of Physicians and Surgeons,

Columbia University, New York

Received for publication November 1, 1921

INTRODUCTION

With the exception of a few such media as Uschinsky's andFrankel's, which are of interest only from the fact that theydemonstrate the ability of certain species of bacteria to syn-thesize protein and other complex physiological substances fromsimple salts, all the pathogenic bacteria are cultivated for'prac-tical purposes on empirical mixtures containing infusions ofmeat, the digestion products of protein and so forth. Mostpathogens either fail to grow, or produce very scant growth onsimple synthetic media. Many fail to multiply even on meatextract-peptone media, but flourish after the addition of serum,blood, and similar materials. It occasionally happens that alot of stock media, prepared with all due regard to known com-ponents and hydrogen ion concentration will fail to grow suchorganisms as pneumococci, although other lots prepared in thesame way have been successful.Numerons attempts have been made to amplify the use of

synthetic media by the addition of known compounds as a sourceof nitrogen or of some particular chemical grouping which hasbeensuspected of playing a part, but as far as progress in the directionof routine cultivation of bacteria on media of known composi-tion is concerned, the results have been uniformly disappointing.On the other hand, there seems to have been very little effortmade to attack the problem from the angle of an analysis of thebasic factors supplied by the physiological mixtures which are

309

sOuNAL 0P BACTzRIOLOGY, voL. vii, NO. 3

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELIER

known to induce growth. One is struck by the lack of referenceto the identity of the components of meat extract or meat infusionwhich seem to be so universally favorable. Purine bases arementioned, but apparently on insufficient experimental grounds.In fact the only piece of work which the writer has found inwhich an effort was made to determine this point was reported byArmand-Delille (1913), who claims to have substituted argininefor meat infusion in media used for the cultivation of the tuberclebacillus. Other work (Long, 1919) indicates that this organismis able to grow on a great variety of substances, and one is notjustified in applying the conclusions of Armand-Delille to othertypes of bacteria without further experimental evidence. Thepossibility of obtaining some light on the requirements of patho-genic bacteria from this point of view is, however, fairly promising,and it is the purpose of the present paper to outline the methodof approach and to cite a number of preliminary experimentswhich serve to indicate the lines along which work may be done.The results already obtained with two of the factors whichit has been possible to single out and follow up has been made thesubject of two preliminary reports (Mueller, 1920), and willshortly appear in greater detail.A word regarding the purpose of such an investigation ad the

results which can be expected from it may not be out of place.The present empirical methods of media preparation are gen-erally admitted to be uncertain from the standpoint of result,and wasteful in materials used. It may be justly questioned,however, whether a more thorough understanding of require-ments will lead to a radical change in our methods of prepara-tion of media. For example, even if certain amino acids andorganic bases should prove to be the only essential factors be-side salts and perhaps carbohydrates, the difficulty and expenseof obtaining them in pure form might well prevent their ex-tensive practical use, and it is quite possible that peptone andmeat will continue to be their most available source. It should,however, be possible to understand the reasons for the uncer-tainty of results and occasional failures now eisting, and per-haps to guard effectively against them, even though with ex-perienced workers, they occur rather rarely.

310

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA

Perhaps the most important results to which success in sucha piece of work might lead, are the applications of the findingsto problems of more general biological importance, particularlyto those of animal metabolism. For, whatever may prove tobe the nature of these substances which cause growth of bac-teria, they are largely or entirely components of animal tissue,and it is probable that they are either needed also by the animalbody and supplied by plant or other sources, or else are synthe-sized by the animal itself to fill some metabolic requirement.When it is possible to catalogue the substances required bypathogenic bacteria for growth, it will probably be found thatmost of them are either required by, or important in, animalmetabolism, and while many of them will surely be compoundsat present familiar to the physiological chemist, it is equally prob-able that some will be new, or at least of hitherto unrecognizedinportance. This point is sufficiently clear in the light of manyrecent publications in connection with the relation of vitaminesto the growth of bacteria and of yeast.Probably of no less importance will be the results from the

standpoint of the classification of 'bacteria. Doryland (1920)has discussed this question at some length, and it is quite possi-ble that unexpected relationships or dissimilarities in bacterialspecies may develop on the basis of food requirements. Thatrelated species among pathogenic bacteria do have similar needscannot be questioned. The colon-typhoid group grow easilyon simple meat extract broth. The streptococci and pneumo-cocci require the presence of an infusion of meat, the meningo-cocci and gonococci usually grow poorly without the additionalpresence of "hormones," blood serum, etc., while the influenzabacillus needs a substance associated with hemoglobin. It isquite possible that the more fastidious types of bacteria mayrequire some of the factors necessary for the more easily growingforms, plus one or more additional substances. For example,if growth of the typhoid bacillus depends on the presence ofthree compounds in meat extract broth, A, B, and C, then thepneumococcus will perhaps fail to grow unless A, B, and C are sup-plied together with D and E, and so on to such organisms as the

311

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELLER

gonococcus and others. Such a conception will at least serveas a working hypothesis upon which to begin investigation.The choice of organisms with which to work must be governed

by several considerations. In the first place, since several fac-tors are probably involved in the growth of all parasitic bacteria.it would be well to select a species having a somewhat limitednumber of requirements. In this way one might hope to singleout one or two factors at a time for identification and, usingthese as a basis, proceed to the study of others. On the otherhand, bacteria with too simple requirements should be avoided,because of the probability that such organisms possess the powerof using a variety of different materials having no immediaterelationship to the components of meat extract and peptone.The colon bacillus, for example, can grow on Uschinsky's orother simple media, about as well as on ordinary extract broth,and for that reason would not be suitable for the purpose athand. A further point to be considered is that since the workwill probably extend over a long period of time, the type of or-ganism selected must be such that its cultural requirements willnot change materially during preservation on culture media.For these reasons, the pneumococcus-streptococcus group was

selected as being probably the most satisfactory. These organ-isms grow well on meat infusion broth, but poorly or not at allon extract broth. While blood or serum improves the growth,neither is essential. In addition, these organisms can be pre-served almost indefinitely and kept at what should be quiteuniform food requirements, The method by which this is at-tempted will be described fully below.The original plan of the work was to start with meat infu-

sion-peptone broth, to eliminate such factors in its compositionas could be managed expeentally, and to substitute knowncompounds, such as amino-abids, purine bases, etc., or failingin this, to determine, if possible, the chemical nature of thematerial removed from the media. As will appear, the workhas been little more than begun, since the difficulties in substi-tuting known compounds for the meat infusion have provedunexpectedly great.

312

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA

METHODS

Preparation of cultures

Three types of pneumococci, I, II and IIA, together with astrain of Streptococcus hemolyticwus, were used. The pneumo-coccus strains were each passed through two mice, and thestreptococcus through one mouse and stock cultures were madefrom the heart blood of the second mouse into small tubes con-taining about 1 cc. of sterile human blood. The latter was ob-tained in the usual way from the median basilic vein and trans-ferred from a syringe to small sterile tubes, each tube containingtwo or three glass beads. The tubes were than shaken untildefibrination was complete, and incubated twenty-four hours toinsure sterility before being inoculated from the heart blood ofthe mice. After inoculation the tubes were incubated eight toten hours until smears showed that the organisms were multi-plying, and they were then stored in the ice box. For trans-plants, ordinary meat infusion peptone broth containiing 0.1per cent glucose and brought to pH 7.4 to 7.8, were used. Asmall loopfull of the blood culture was transferred to the broth,and after incubating eighteen to twenty-four hours, the culturewas used, in the case of the pneumococci for the inoculation ofa second similar tube of meat infusion broth. These two cul-tures are called respectively the A and B cultures. Experi-mental media were inoculated from the B cultures in the caseof pneumococci, and from the A tubes with streptococci. Inthis way it is believed that the food requirements of the bac-teria remain reasonably constant. The stock culture remainsalive for long periods of time, up to six months or more. Afterbeing opened repeatedly, old tubes gradually dry up, and suchcultures have been transferred to fresh blood tubes, using thesame technic, and always passing the strain through one mouseto guard against a gradual change in cultural requirements inspite of the blood media. It is not unlikely that significantquantities of blood may be carried over into the test media withonly a single passage intervening, but it certainly is not enoughto produce growth on unsuitable media, although it cannot beoverlooked in estimating the value of results obtained.

.313

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELLER

Determination of limiting hydrogen-ion concentration for 8trainsemployed

It was found, by inoculating a series of plain broth media withpH ranging from 6.8 to 8.6, that the pneumococci foundconditions suitable for growth between pH 7.4 and 8.0, whilethe hemolytic streptococcus grew well throughout the samerange, and even as far down as pH 6.8. For the adjustment ofexperimental media, therefore, a reaction of pH 7.4 to 7.8 hasbeen used for all the strains of test organisms. In cases whereseveral experimental lots of media are prepared at one time, theadjustment of the reaction is facilitated by the direct additionof phenol red to the entire lot. It is a simple matter to reachthe particular pH desired by using a solution of phenol red of0.02 per cent concentration, in the proportion of 1 cc. to 25 cc.of media. Normal or semi-normal NaOH (moderately free fromcarbonate) is then added drop by drop until the color is distinctlyred, but not purple. One can quickly learn to recogalize thecorrect color without the need of standard color tubes. In thecase of unusually dark colored media the reaction can be ad-justed approximately by this means and finished by the usualmethod of dilution and addition of more indicator followed bycomparison with standard tubes. Experiment showed thatseveral times this amount of phenol red could be added to mediawithout any influence on the growth of the organisms in question.

Limiting range of osmotic pres88ureAlthough moderate variations in osmotic pressure are prob-

ably of little importance in culture media, an experiment wascarried out in broth made up with varying concentrations ofNaCl. The test orgaisms grew equally well in salt-free brothand through several intermediate concentrations up to an equiva-lent of 2 per cent NaCl, so that with ordinary care the questionof osmotic preure apparently need not enter into consideration.

Presence of glucoseIn preliminary work with the pneumococci it was observed

that occasionally a lot of meat infusion was met with which

314

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTB OF BACTERIA

failed to give growth, although prepared with all due attentionto hydrogen ion concentration, etc. The addition of as littleas 0.025 per cent glucose to such media at once improved them tosuch a degree that marked growth occurred, and with twicethis quantity, growth was approximately as heavy as in theaverage broth prepared without glucose. While the point hasnot been verified experimentally, it is not improbable that mostmeat contains small quantities of glucose or other carbohydrate,and that occasionally this may be much diminished or absent.For routine purposes, therefore, and in all experimental media,0.1 per cent glucose is added. This quantity is insufficient toproduce acid in amounts great enough to kill the cultures intwenty-four hours, or to interfere with agglutination withspecific sera.

Inorganic constituents

Little is known definitely of the salt requirements of bacteria.There is some evidence in the literature to indicate that theyare moderately elastic, and in any case such minute traces as maybe required, must occur in the peptone and meat infusion usedas routine media. Where these or similar substances are omitted,small quantities of inorganic salts may well be required. Thenature of these can only be determined when purified organiccompounds can replace the meat infusion and peptone.To guard as far as possible against failure of experimental

media to support growth through lack of inorganic material,the same salt mixture which is used in Uschinsky's and othersimilar media has been employed instead of simple NaCl, inthe preparation of all experimental media in which the ordinarymeat infusion is not used.To simplify the preparation of media, the salts are all dissolved

in twice the concentration required, together with glucose, andphenol red is added to this solution. Media are prepared byadding an equal volume of meat infusion or other solution tothis preparation, thereby reducing the concentrations of theconstituents to that desired. The composition of the solutionis shown below, and it will be referred to subsequently as"glucose-salt solution."

315

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

316 J. HOWARD MUELLIER

per cent

NaCi...... 1.0MgO4..0.04CaC..0.02HK2P4.0.2Glucose.0.2Phenol red.80 cc. of 0.02 per cent solution per liter

If all the components except- the potassium phosphate are dis-solved, and the solution diluted almost to the final volumebeforethis substance is added, calcium phosphate is not precipitated,although a good deal is probably lost in the precipitate whichusually forms on adjusting the reaction and boiling.

Sterilization of media

Since it has been shown that ten minutes autoclaving at 10pounds pressure is even less destructive to sugars then the Arnoldtemperature for three quarters of an hour, this method has beenused in most of the work. Practically no change in pH occurs,and contaminations are exceedingly rare.

Recording of results

When growth has continued for twenty-four hours, the degreeof turbidity is recorded by comparing the tube with a set of seventubes containing suspended BaSO4 (Koser and Rettger, '19)ranging from the faintest trace of a cloud to a suspension as heavyas the best growth obtained with these organisms on good media.Negative growth is shown by "0," and any degree of growth by anumber corresponding to the series number'of theBaS04standards.This affords a means of correlating experiments done at differenttimes, although of course it is not quantitative. It usuallyhappens, however, with the pneumococcus cultures, that afteroptimum growth has been reached, autolysis follows very quickly.Within a few hours, a tube which has shown growth equal tostandard tube nos. 4 and 5 will clear up and become simplyopalescent. With Type I this happens quite regularly in lessthan forty-eight hours on good media, since maximum growthis reached by this strain in fifteen to eighteen hours. Types IIand IIA grow more slowly, but are often autolyzed in forty-eight

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA 317

hours. In recording growth, such tubes are marked "A." Thisautolysis is of course not a criterion of good growth, for in experi-mental media, scanty growth is also followed by clearing up.The streptococcus appears not to show this phenomenon.

Preliminary experiments on infusion broth

The growth of the test organisms on peptone-free infusion andon peptone water alone was first investigated.

[Meat infusion ..................... 25 cc.Lot 1. Glucose-salt solution..................... 25 cc.

(Peptone1.....................0O.5 gm.

Lot2 .Meatinfusion..................... 25 cc.Glucose-salt solution..................... 25 cc.

[Peptone.....................0O.5 gm.Lot 3. tWater..................... 25 cc.

(Glucose-salt solution..................... 25 cc.

LOT NUMBER pH TYPE I TYPE II TYPE IIA STOCOOC

1 7.8 A A 4 52 7.6 A 5 2 63 7.8 4 0 0 0

It appears from this experiment that peptone-free infusion ispractically as satisfactory a medium for the growth of the pneu-mococci and streptococci as the usual meat infusion broth con-taining 1 per cent of peptone. On the other hand, with theexception of the strain of Type I pneumococcus, peptone wateralone will not support growth. As a working hypothesis, ithas been assumed that these organisms have a nitrogen require-ment supplied by peptones or amino acids, and also a need forcertain accessory substances supplied by meat infusion. There-fore, on the basis of such a supposition, meat infusion seems tocontain.not only the accessory substances, but also peptones,amino acids or other sources of available nitrogen. The problem

I Throughout the work, "Difco" Peptone has been used for the sake of uni-formity.

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELLER

is complicated by this fact, since it is necessary to separate bysome means the accessory substances from the nitrogen supplybefore either one can be studied separately.

The need for an accessory factor

Before describing the various methods by which a separationof the accessory factors of meat infusion was attempted, anexperiment will be described which strengthens the probabilitythat such substances, other than protein degradation products,are necessary. It is possible that only amino acids or peptonesmight be required for growth, but that in the preparation of com-mercial peptone some essential amino acid, as, for example,tryptophane, is wholly, or for the most part, altered or destroyed.It is quite conceivable that unstable groupings other than tryp-tophane may be present in the original protein molecule, whichmay withstand moderate heating in neutral solution and thusbe present in meat infusion among the amino acids or peptonesin a soluble form, and yet be almost or quite lacking in commer-cial peptone. If such were the case, a whole protein hydrolyzedby trypsin or erepsin, together with salts and glucose, wouldprobably serve as a complete culture medium. A specimen ofcommercial casein after several day's digestion with trypsin was,as a matter of fact, found to be quite satisfactory without thepresence of infusion, for the pneumococcus and streptococcus.However, when the casein was purified by three precipitationsfrom Na2CO3 solution by acetic acid, washed with alcohol andether and then digested as before, growth was negative. Thefollowing protocol shows the results of such an experiment, inwhich casein was prepared directly from milk. The "crude"casein is the first precipitate obtained by acetic acid, the "pure"casein has been three times reprecipitated and finally washed inalcohol and ether.The two preparations of casein were dissolved in 0.5 per cent

Na2CO and digested with a small quantity of trypsin (Fair-child) at 370 under toluol for two weeks. At the end of this time,the two solutions were boiled and filtered. Twelve cubic centi-

318

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA

meters of each ifitrate represented about 0.5 gm. casein. Thefollowing media were prepared:

Digest of impure casein....................... 12 cc.Lot 1. Water....................... 13 cc.

Glucose-salt solution....................... 25 cc.

Digest of pure casein....................... 12 cc.Lot 2. sWater....13 cc.

(Glucose-salt solution....................... 25 cc.

_ II IIA STREPTOCOCCI

NUMBER pH24 hours 48 hours 24 hours 48 hours 24 hours 48 hours 24 hours 48 hours

1 8.0 6 5 0 6 6 6 5 62 8.0 0 0 0 1 0 0 0 0

Control experiments showed that the digest of pure casein wasnot inhibitory, for the addition of meat infusion produced heavygrowth.The above experiment was confirmed several times, using

different preparations of casein, always with the same result.The conclusion seems warranted that while crude casein containssome accessory substance, this is not a part of the protein mole-cule (and hence probably not an amino acid or polypeptidegrouping), and can be easily removed by purification of theprotein by standard means.

Possibility of another source of growth accessory substance freefrom protein nitrogen

By the use of the term "accessory substance" in connectionwith these studies, it is desired to avoid, as far as possible, theconception of vitamines. There is now abundant evidence inthe literature that vitamines, particularly the water-solublevitamine, may be as essential for certain micro6rganisms as foranimals, but there is little probability that the accessory factoror factors of meat infusion is in any way connected with thewater soluble vitamine. -In the first place, muscle tissue isbelieved to be low in vitamine. In the second place, experi-

319

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELLER

ments with the pneumococcus and streptococcus have shown that"protein-free milk," which does contain the water soluble vita-mine, is almost without activity when substituted for meat in-fusion in media. To economize space, protocols of these ex-periments are omitted.

It is not impossible that physiological extracts other than meatinfusion might supply the accessory factors in a form more freefrom other nitrogenous compounds such as amino-acids, thanthe latter, and in a few experiments it has been possible to showthat some other preparations, notably one from blood, and onefrom spinach leaves, gave growth when mixed with peptone andonly scant growth in its absence. The blood was diluted withwater, acidified, boiled and filtered. The spinach leaves weredried, ground fine and extracted with water. In the case of thelatter, initial extraction of the dried powder with ether did notremove the accessory substances, and boiling with repeatedchanges of alcohol for several hours extracted only a small part.It would perhaps be possible to develop a technic along eitherline for the preparation of a solution of the accessory factorssufficiently free from protein nitrogen to investigate the natureof the requirements of the test organisms by the addition of purea o acids, but is has seemed more satisfactory first to exhaustas fax as possible the more obvious methods for a separation ofthe meat infusion.

Attempts to separate the growth accessory factors from the aminoacids of meat infusion

1. Repeated extraction of meat. While carrying out some ex-periments along another line, it was observed that the test or-ganisms grew as well on a trypsin digest of the insoluble meatresidue remaini g after the preparation of meat infusion whichhad first been thoroughly boiled out in three changes of water,without the addition of any meat infusion, as upon the usualpeptone broth. This suggested the possibility that the growthaccessory material might be extracted from the coagulatedprotein-with some difficulty, and might be partially separatedin this way from amino acids, etc. Chopped beef was, therefore,

320

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA

soaked in cold water in the proportion of 1 pound of meat to 500cc. of water, and heated to 550 for a few moments and strained.The residue was boiled for five minutes with a second 500 cc. ofwater, and strained, and the extraction repeated a third time.The first extract was boiled to remove coagulable material, andall three extracts were filtered. Media were prepared in thefollowing way, and after inoculation, gave the results indicated:

GROWTH 86 HOURSEXTRACT EXTRACT EXTRACT PEPToNE H:O GL"ucosE _____

EI Ct II E=ITII^CT PEPq!ON H20 I II IIA_SALT~~ u hIA Sltrep-

tooocci

CC. CC. CC. grams CC. CC.

1 25 1.0 25 5 7 7 62 25 25 1 1 2 53 25 1.0 25 3 3 4 44 25 25 0 1 1 35 25 1.0 25 5 3 5 66 25 25 0 0 0 07 1.0 25 25 3 0 1 0

Other experiments of the same kind have given like results,but there is enough irregularity so that the method is not ideal.It does, however, bring further evidence as to the presence oftwo classes of compounds in the meat infusion. Further workwith the method, substituting known amino acids instead ofpeptone, may lead to more positive results.

2. Chemical fractionation of meat infu8ion. From the factthat commercial meat extract cannot replace meat infusion forthe growth of the pneumococcus and streptococcus, one maysuppose that heating, oxidation or simply long preservation maydestroy the active material. If it can be shown experimentallythat these or other comparatively simple chemical processesare destructive, plans for chemical separation can then be madein such a way as to avoid or minimize this difficulty. Accord-ingly, a few simple tests were made on the influence of suchobvious procedures as suggested themselves. It was found thatdrying on the water bath, long boiling (two to three hours) inneutral or moderately acid solution (2.5 per cent HCl) and

321

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

J. HOWARD MUELLER

exposure to mild oxidizing or reducing agents in no way dimin-ished the power of infusion or destroyed its activity. It wouldseem that if hot alkali and strong acids are avoided, it should bepossible to preclude loss of activity while carrying out simpleprecipitations on meat infusion.

This has been found to be the case, and several different re-agents have been employed, without, however, any markedsuccess in the separation of the constituents of the infusion.Precipitation with alcohol, up to 85 per cent concentration,carried out by preliminary evaporation of the infusion to a small,measured volume, followed by addition of the required amountof 95 per cent alcohol, has in several experiments yielded a pre-cipitate which contains the growth accessory substances, sincethe test organisms grew on media prepared from the precipitateplus peptone, but not in the absence of peptone. The separa-tion is somewhat more complete if the first precipitate is dis-solved in a little water and reprecipitated. However, in eithercase, a part both of the growth accessory factors and of theamino acids remains in the alcohol filtrate, and it has not so farbeen possible to separate them quantitatively in this way.Lead hydroxide, lead acetate, mercuric chloride and silver

nitrate and baryta have all been tried, none of them with resultssatisfactory enough to follow up extensively. In general, thefiltrates from these reagents will permit growth to some extentwithout the addition of peptone, and better growth in its pres-ence, which is taken to indicate t-hat the growth accessory frac-tion is not precipitated readily by these metals, while the aminoacid fraction, perhaps in the form of peptone, is partially throwndown. By precipitation with tannic acid, followed by removalof the latter with Ba (OH)2 + Pb (OH)2 in the usual way, verysimilar results are obtained.While none of the methods of chemical separation have given

successful results, they have at least shown that it is possible tosubmit meat infusion to such processes without loss of activity.It has, indeed, been observed that where a single preparation wasput through several consecutive precipitations, the activitywas gradually diminished or lost, but it is at least as possible that

3022

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

CULTURAL REQUIREMENTS OF BACTERIA

this may have been due to the loss of some essential factor in aform which could not be recovered, as for example a tanninprecipitate or an extremely insoluble silver combination, as toactual chemical decomposition of the substance. It is, in fact,quite probable that there are many essential factors present, andthat progress can be made only as methods are developed whichwill enable one to single these out for identification. Such amethod seems to have been found in the treatment of meat infu-sion with charcoal. By this means, one or more factors are re-moved from the infusion which may be again supplied by theaddition of a small amount of peptone, or of a sulphuric acidhydrolysate of casein. In following up this lead, a considerableamount of work has been done, and a number of interestingobservations made which have already been briefly reported(Mueller, 1920) and which will be dealt with in the next paperof this series.

SUMMARY

The purpose of this paper has been merely to outline the planof work and describe the method followed in our studies upon theproblem of the nutritional requirements of certain bacteria.The procedures as outlined in the section on methods will beused in such work unless modifications prove desirable. Suffi-cient evidence has been obtained from the experiments herereported to warrant belief that two classes of organic compounds(in addition to carbohydrates), are required for the growth ofpneumococci and streptococci, the first supplied by proteindegradation products, the second by extractives of meat. Bothoccur together in ordinary meat infusion, but they may be separ-ated more or less completely in several ways. The necessityfor a non-protein substance is shown most clearly by the failureof a trypsin digest of purified casein to support growth, whilethat of impure casein is satisfactory. The possibility of a sep-aration of the two classes of compounds, as they occur togetherin meat is evident from experiments of several types, particu-larly by repeated extraction of meat, alcoholic precipitation of

3233

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

324 J. HOWARD MUflJER

meat infusion, and charcoal decolorization of heart infusion.There is every reason to believe that there may be several indi-vidual factors falling into each of these two groups.

REFERENCES

AEMAND-DELILLB, MAYER, SCHAFFER, AND TERROwiN 1913 Jour. de phyuiol.et pathol. gen., 15, 797.

Doum.&ND 1916 Jour. Bact., 1, 148.KOSER AND REITTGER 1919 Jour. Infec. Dis., 24, 301.LONG 1919 Amer. Rev. Tuberculosis, 3, 86.MuELLER 1920 Proc. Soc. Exper. Biol. and Med., 18,14 and 225.

on July 8, 2020 by guesthttp://jb.asm

.org/D

ownloaded from