LVL-The Chemical Actions of some iWicro-organisms. By R. WARTNGTON. THE present research was commenced with the intention of isolating, if possible the nitrifying organism contained in soil. I n the autumn of 1886 I spent through Dr. Klein’s kindness some time in the laborat,ory of the Brown Institntion learning under his instruction, the methods usually employed for the separation of different species of bacteria. On returning to the Rothamsted laboratory I com-menced the study of five organisms-two isolated from visible growths in some of my nitrifying solutions; two obtained from soil at the Brown Institution ; and one the well-known Bacillus subtilis obtained from hay. A short time afterwards I became acquainted with the paper by W. Heraens (Zeit. f. Hygiene 1886 193).In this communication he names seven well-known organisms :-LW. prodigiosus Staphylococcus citreus Finkler’s bacterium the spirillum of cheem the bacteria of anthrax and typhoid fever and the “ Wurzelformige Bacterien,” as capable of nitrifying urine and concludes that the power of producing nitrification is by no means unusual among bacteria. In April 1887, D c. Klein kindly supplied me with pure cultures of 12 bacteria includ-ing several of tliose specifically named by Heraeus. The study of these organisms afforded results of considerable interest and in consequence others were from time to time supplied by Dr. Klein. The investiga-tion thus deviated from its original intention and became to a con-siderable extent a stmdy of some of the chemical propertiea of a variety of oipnisms many of them well known as pathogenic.The action of these organisms has been tested in four particulars :-1. The hydrolysis of urea. 2. Action on milk 3. Capacity for reducing nitrates. 4. Power of producing nitrification. It is proposed after publishing the results at present obtained to return to the investiga-tion originally intended and recommence the study of the micro-organisms in soil. THE MICRO-ORGANISMS STUDIED. The following were received in the state of pure cultures from Dr. Klein :-1. Bacillus of swine fever. 2. , of typhoid fever. 3. , of infantile diarrhoea 7%8 WARIKGTON THE CHENICAL ACTIOXS 4. Bacillus of anthrax. 5. , of septicaemia (mouse). 7. , subtilis (jequirity extract). 8. 7 ) , (scurf of scarlatina).9. , JEz~orescens. 6. 99 7’ (guinea-pig). 10. , $uorescens liguescens. 11. ,) intesthi.” 12. Spirillum Koch’s cholera asiatica. 14. , Deneke’s cheese. 16. Micrococci~s aureus. 17. Staphylococcus luteus. 13. 7 Finkler’s cholera nostra. 15. ,> Liagard’s noma. 18. 7 7 candidus. 19. 9 candidus liguescens. 2Q. Streptococcus s c a r l a t i n a h o r n Dr. W. R. Smith was received :-21. Micrococcus urece. The following five were isolated under Dr. Klein’s superin-tendence :-22. Bacillus su&ilis (hay). 24. , torulqormis. 25. , sulphureus. 26. , tardecrescens. 23 1 $occus. There mas also separated from specimens of M. prodigiosus-27. Nicrococcus gelatinrjsus. A considerable number of the organisms here enumerated are well In some other cases a reference to See Reports of Ned.Oflc. LOG. See same Reports 1886-7, It will be more fully described in the following known and need no description. the original description will suffice. Gov. Board 1877-8,169 ; 1686-7,446 P1. XXV. 447 P1. XXVII. Report. 1. Bacillus of swine fever (Klein). 3. Bacillus of i i f a n t i l e diarrhea (Klein). 5. Bacillus of septicmnia mouse (Klein). 6 . Bacillus of septicamia guinea-pig (Lingard). Ibid 1885-6 17@, Ibid. 447 P1. XXVI. Plate XIX. * When this paper was communicated to t.be Chemical Society the organism now called B. intestini was spoken of provisionally a8 Bacterium fepmo OF SOME MICRO-ORGIANISMS. 729 9 and 10. BaciZlus$uorescens. 11. BaciZEus iniestini. These two fluorescent bacilli will be described by Dr.Klein in.Med. O$ic. Report 1887-8. Obtained by Dr. Klein from the large intes-tine in a case of diarrhoea i n a rabbit. Grown in broth it appears as a short bacillus with rounded ends. The longer forms are 1.0-1.5 p in length and about 0.4 p in thickness, I n a plate culture it appears the first day as small translucent dots, which afterwards develop in the depth to opalescent spheres and on the surface produce a shining expanse of pearl-like lustre. I n a stabbed culture the growt,h at first is chiefly in the depth. The track of the needle is bounded by rows of opaque spheres. There is no liquefaction. In broth at 22" or 33" a dense white turbidity is produced with no film. 25. XpiriZZum of noma (Lingard). This non-liquefying spirillum is described in the Practitioizer 38 121 Figs.32-35. 18. StaphpZococcus candidiu (Klein). This non-liquefying organism was obtained from condensed milk. Hed. Ofic. Report 1886-7 386, No. 3. Obtained from the blood of a scarlatina patient. Ibid. 1886-1 374 Pls. I-XII. The organism used in the experiments described in this paper was obtained from the blood of a scarlatina patient. 21. d/li'crococcus urece (Smith). Quart. Jour. Micro. Science 1887, 371. 23. Bacillus ;Roccus.-Obtained from garden soil at the Brown Institution. A straight bacillus 3.0-7.5 p i n length and 1 . 0 ~ in thickness ; motile in broth forms long interlacing threads. In gelatin plate cultures at 22" it forms the first day faint colonies which under a low magnifying power resemble balls of cobweb; from these fine interlacing threads resembling the mycelium of a fungus spread in every direction.Liquefaction commences the first day and on the second extends over most of the surface. In stabbed gelatin cultures at 22" it forms a liquefied channel in the course of the needle the first day. From this channel numerous fine branches resembling knotted threads pass into the gelatin. The second day liquefaction has extended over the surface ; the channel has disappeared and the gelatin is filled with finger-like branches starting from the old tube. A soft, thick film forms on the surface ; below this are two layers of growth with clear spaces between. These two layers finally become one and sink to the bottom. The liquefied gelatin becomes brown near the surface.19. Staphy lococczrs caiididus liguescens (Klein). Ibid. 370 P1. VII. 20. Mreptococcus scarlatince (Klein). The fifth day liquefaction is completed 730 WARINGTON THE CHEMICAL ACTIONS On agar agar at 22" a grey finely crinkled film quickly spreads over the whole surface. I n diluted urine or in very Teak broth a t 22" it forms dense white flocks resembling those produced by anthrax the rest of the liquid remaining clear. In normal broth at 33" it forms after some days a thick soft film which falls as ribbons. 24. Bacillus toru1iformis.-Obtained from garden soil a t the Brown Icstitution. Maximum length 5.3 p a few 6.4 p ; thickness 1-0-1.6 p. The ends thickened and much ronnded. Many short, thick forms joined together.The general arrangement very crooked and irregular. In gelatin plate cultures i t forms the first day white dots, having a granular margin. The second day there are large white colonies many 2x or & inch in diameter. Under a low magnifying power they appear of granular structure. Liquefaction hw com-menced. I n stabbed gelatin the first day's growth is chiefly confined to the needle channel and consists of moderately opaque dots. The second day liquefaction commsnces. I t occurs principally a t the surface. A funnel of liquefied gelatin is formed with a short conical pipe and white organism in the pipe the upper liquid being clear without film. The gelatin at the bottom of the tube remains long un-liquefied. 011 agar agar a t 22" it grows rapidly forming a pasty moderately opaque white film.In broth at 33" it grows luxuriantly producing great turbidity an abundant fine deposit b u t no distinct film. 25. Bacillus su1phzweus.-This was isolated from a surface growth which had appeared on some solutions which had nitrified. In a deposib from dilute broth it had a leiigth of 1-1.5 p and a thickness of 0.3 p. Taken from the surface of the broth it appeared as a net-work of bacilli the sheaths faintly stained with well-stained oval dots within. The colonies were a t first milky points becoming afterwards translucent spherical masses. The colour of the colonies is at first yellowish-white. By exposure to light the colonies become opaque and of a bright sulphur colour. There is no liquefaction. In a stabbed culture there is scarcely any growth in the depth of the gelatin.In broth a t 2 2 O it produces turbidity. Afker some days small clots are deposited some strung together in ropes together with some firrer threads. Festoons of these ropes hang from the surface but, The bacillus is motile. In a plate culture no growth appeared for several days. Growth soon appears upon an inoculated surface OF SOXE XICRO-OXGAKISMS. 731 there is no film. light. The deposit in broth turns yellow on exposure to The bacillus does not grow at 35". 26. Bacillus tardecrescens.-This organism was isolated from floating gelatinous masses occurring in some solutions of ammonium carbonate which had nitrified. Grown in dilute broth it appears as a small oval bacillus 1.0 p in length and about 0.5 p in thickness.Grown on gelatin it is somewhat smaller and appears more like an oval coccus. It is stained by gentian-violet very slowly. The bacillus is under all circumstances B ~ O W in growth. In a plate culture the colonies became visible after a week or more ; they grow into small translucent droplets of a somewhat smoky tint. In a stabbed cultwe the growth was chiefly in the depths it con-sisted of strings of spherical coloniea. In broth growth is slow and only produces a very moderahe turbidity. 27. Microcoecus gelatinosus.-T wo specimens of M. prodigiosu.9 from different sources produeed a mixture of red and white growths on the surlace of agar agar. The red I did not succeed in obtaining pure; the white was easily isolated by cultivation on nsar agar at 35" when the white only was developed.The white organism resembles the red in microscopical characters but grows much better than the red at high temperatures and preserves its vitality for a longer period. Grown in broth it appears as an oval COCC';LS or short bacillus, 0-5-1.3 p in length and 0-3-1*0 p in thickness. I n plate cultures of gelatin it appears the first day as milk-white dots which the next day are increased to Q or 4 inch diameter and liquefied. In stabbed gelatin liquefaction in the channel commences the first day and has extended over the surface by the second day. A funnel is then formed with a wide pipe. On agar agar it grows rapidly at 22" less rapidly at' 35' forming a thick pasty whitish mass moderately opaque covering the whole surface .I n broth it produces an abundmt turbidity and a considerable amorphous deposit but no film. There was no liquefaction. There is no film. GENERAL METHOD. The methods of culture on gelatin and agar agar were those originated by Koch and now generally employed. The nutritive gelatin and agar agar were of the composition employed by Dr, Klein the former containing 19. per cent. of gelatin. The prepara ‘732 WARIXGTON THE CHEMICAL ACTIOSS tion of the various liquid mediums will be described under the head of each investigation. The micyoscope was one kindly placed a t my disposal by the Chemical Society for the purpose of this investigation. The powers used were Zeiss’ D and Leitz’s & oil immersion lens. The stained preparations were made with gen tian-vio le t.All inoculations of solutions were made by means of Klein’s capillary glass pipette or hollow needle the cotton- wool stopper not beiiig removed from the tube o r bottle containing the sterilised liquid. By means of the same pipette the acidity or alkalinity of a solution could be a t any time ascertained without removing the st’opper. I n examining a liquid for nitrites or nitrates a much larger pipetbe was employed the capillary tube of which was passed between the stopper and the glass ; sterilised wool was held against the pipette at the point where i t entered the bottle so that oiily filtered air might enter when suction commenced. The large pipettes were kept in boiling water till the moment when required.By proceeding in this way the cultures were commenced and brought to a close without once removing the cotton-wool stoppers. As a check on the results a culture was generally made on gelatin or agar agar a t the end of each experiment or series of experiments, made in a fluid medium the object being tlo prove that the action had proceeded without the entrance of any foreign organism. This check was at first employed in every case but after confidence in the mode of work had been attained it was made use OE only to check striking o r irregular results or to ascertain the growth of an organism in milk. The method was especially necessary for the latter purpose as successful inoculation with an organism which produced no striking change in the milk could in no other way be ascertained.THE HYDROLYSIS OF UREA. The hydrolysis of urea and its conversion into ammonium carbonate which takes place in putrefying urine is now universally acknowledged to be due to the action of micro-organisms. Pasteur has described a micrococcus which acts i n this manner. More recently (1F85) Leube and Graser (Yi~chow’s A ~ c h i v 100 555) have separated the organisms occurring in samples of ammoniacal urine, and have found four which were capable of converting urea into ammonium carbonate. The most active of these was a short bacillus, which they name Bacterium ureoe. The next in activity and the most abundant was a micrococcus. There were besides two bacilli, which exerted a much feebler action. The authors also name a fift OF SONE MICRO-ORGANISMS.7 33 organism " Lungen Sarcin," as capable of converting urea into ammo-nia. Heraeus (Zeif. f. Hygien,e 1886 215 221) isolated four bacilli which effected an ammoniacal fermentation in urine ; three of these liquefied gelatin and were therefore distinct from the organisms of Leube and Graser. During last year Dr. TV. R. Smith (Quart. Jour. Xicmscopicab Science 1887 371) has separated about 20 organisms from nmmoniacaJ urine and found among them only one which was capable of producing ammonia from urea. The active organism separated by Dr. Smith was a micrococcus ; it was apparently not identical with the micrococcus of Leube and Graser as it slowly liquefied gelatin whilst their micrococcus did not. I t appears there-fore that at least nine organisms are known which are capable of hydrolysing urea.Cultivations in diluted urine have been attempted with all the organisms studied during the present investigation. Nos. 13 14, 16 17 and 26 have not occasioned a distinct turbidity in the solu-tions; Nos. 19 and 20 have given only a slight turbidity; the remainder have afforded plain evidence of growth. The urine soh-tions were of two kinds ; the first was 8 1 per cent. solution contain-i n g gypsum used for the experiments on nitrification. The cultures in this series of solutions were tested with Nessler's reagent after being kept for two or three weeks a t 32" ; the presence of gypsum in this case preventing volntilisation of the ammonia. The tint given by Nessler's reagent in the various cultures was compared with that produced by the same test in similar solutions unseeded with organisms.The organisms Nos. 1 2 4 5 11 13 15 17 18 19 20, 22 23 24 25 and 26 were tried in this manner. I n one case only (No. 18) was any distinct increase of ammonia perceived over that present in the unseeded solution. The second urine solution contained 25 per cent. of urine and was prepared especially to determine the action of the organisms on urea. The test-tubes containing the seeded urine were kept a t the temperature of 22" for 7-10 days the alkaliiiity of each cultivation was thexi determined by means of standard acid and alkali and the result compared with that given by unseeded urine placed under the same conditions. The method did not admit of extreme accuracy as the indicators employed failed to determine the point of neutrality with great exactness.All of the organisms under investigation with the exception of Nos. 7 8 16 and 19 were tried by this mehhod; in the case of 13 14 and 17 however no quantitative result was attempted as there was no evidence of any growth in the solution. With the exception of two organisms to be presently mentioned, none proved capable of distinctly affecting the alkalinity of the urine, In the case of the Jficrococcus weCe (Smith) a very distinct effec 731 WARINGTON THE CHENICAL ACTIONS was produced on the urine. I n two experiments made with different solutions the urine acquired a distinct ammoniacal odour. In one case the alkalinityat the end of seven days was increased to an extent equivalent to 0.0066 gram of ammonia per 10 C.C.of solution while in the second experiment the increase of alkalinity in eight days was equal to 0.0043 gram of ammonia. A very distinct increase in alkalinity also occurred in the case of B. Jluorescem. In the first experiment the increase of alkalinity in eight days was equivalent to 0.0039 gram of ammonia in 10 C.C. of the solution I n the second experiment the increase in seven days was equal t o 0.0037 gram. It seems probable from this result that the B. jluorescens is identical with the bacillus giving a green fluorescence obtained by Heraeus from soil as this organism was found by him to convert urea into ammonia. The property of effecting the hydrolysis of urea is apparently but rarely met with among micro-organisms ; in the present case out of 24 organisms tried only two could certainly be shown to possess it.I am doubtful however if either of these organisms is the cause of strong ammoniacnl fermentation. At all events in two comparisons of the result produced by the M. urea? with that yielded under the same circumstances by the mixed organisms of arable soil the latter proved far more effective. The soil mas from heavy nnmanured, arable land. The fragment used was about the size of a large pin's head; it wa8 taken half an inch below the surface. The two com-parisons were made in different solutions. The amounts of ammonia found per 10 C.C. of diluted urine were as follows :-Soil. M. uvea. I. 0.0252 gram 0.0066 gram. 11. 0.0116 , 0.0043 ,, It would appear probable therefore that the soil contained a more powerfully hydrating organism than the micrococcus of Dr.Smith ; it possibly contained the active bacillus of Leube and Graser. ACTION ON MILK. Milk is an excellent medium for studying the chemical action of micro-organisms being a liquid susceptible of many changes. I n all the experiments I have made skim-milk has been employed; the inconvenience of a surface layer of cream was thus avoided. The first trials were made wit,h skim-milk kindly supplied by the Ayles-bury Dairy Company. This milk had been separated in the centri-fugal machine it was thus perfectly sweet and as nearly as possible free from fat. The milk had however been sterilised by heat before it reached me and had of course again to be heated after it had bee OF SOME JIICRO-ORGLIKISIW.735 transferred to test-tubes. Owing probably to this double heating of the milk it required a loriger time to curdle than the milk subse-quently employed. The milk afterwards used in the experiments was taken directly from the cow into a sterilised stoppered bottle; the portion collected was never the first or the last milk from the udder. The bottle of milk was stood in a cool place for 36 hours. The milk below the cream layer was then drawn off by a syphon and 10 C.C. measured into each test-tube The test-tubes closed of course with sterilised cotton-wool were then heated a t 100" for 20 minutes in a steamer. Only three imperfectly sterililred tubes were observed during the whole investigation. The precipitation of the caseiin which determines the curdling of milk may be brought about either by an acid or by a soluble ferment, as rennet ; the latter is capable of solidifying milk even when it has been made alkaline with sodium carbonate.The proportion of acid required to curdle milk is smaller the higher is the temperature; thus milk that is only slightly sour may be completely curdled if heated to boiling. Attempts were made to ascertain what was the smallest proportion of lactic acid that would suffice to curdle milk. Very dilute lactic acid was slowly dropped into 10 C.C. of sterilised skim-milk ; milk acidified with different pro-portions of acid w~ then heated and the effect observed. With 0.15 gram of' lactic acid per 100 C.C. of milk a tolerably complete solidification of the milk was obtained when the test- tube was plunged into boiling water.A considerably smaller proportion of lactic acid sufficed to produce R partial curdling. With 0.32 gram of lactic acid per 100 C.C. of milk the whole became nearly solid in two hours at 33". The proportion of lactic acid required to curdle at lower tem-peratures could not be ascertained by this method as it was found impossible to mix the necessary proportion of acid with the milk without a t once curdling it the acid being unavoidably present in great excess at the spot on which the drop of acid fell. It is difficult to determine the quantity of acid in milk with much accuracy the reaction with litmus-paper being far from sharp. Neutral litmus-paper was used as the indicator in all the experiments.To Hueppe ( M i t t h e i l . a. d . k. Reiclzsgesundheitsamt 2 309) we are indebted for the fullest study of the organisms affecting milk. He names five as capable of souring milk. One st bacterium or bacillus, which curdles with the evolution of carbonic acid gas; two cocci obtained from the mouth; M. prodigiosus; and the coccus of ostao-myelitis." * Marpmann (Arch. Pharm. 1886 243) states that he has obtained five organisms producing lactic acid from the milk supplied to Gottengen j his desirip-tions do not seem to be publiehed yet 736 WARIKGTON THE CHEMICAL ACTIONS Among the organisms which I worked with five have distinctly acidified milk ; most of these organisms are certainly different species from those enumerated by Hueppe.The acidification by these or-ganisms has been in very different degrees. The three first named act apparently simply by the production of acid ; in the case of the fourth and fifth the action is more complicated. StapA ylococcz~s candidus.-At 22" this organism distinctly acidified the milk. The cultnre was continued for 43 days without visible change in the milk save the production of a slight sediment. At 32" after eight days the acidity of the milk was equal to 0.13 gram of lactic acid per 100 C.C. The milk was not curdled and did not solidify when placed in boiling water. At 33" with another batch of milk the acidity after 17 days was 0.15 per cent. Bacillus i.lztcstini.-Grown in milk a t 2 3 O it produced in 11 days an acidity equal to 0.33 per cent.of lactic acid; in 21 days an acidity of 0.33 per cent. ; in 31 days an acidity of 0.33 per cent. The milk did not curdle at 23"; when the cultures were placed in boiling water they immediately solidified. At 34" the milk became solid in two days ; the acidity was then equal to about 0.35 per cent. of lactic acid. Rncillus of Infantile Diawhcea.-A culture at 22" has been quite fluid after 43 days but examined at 83 days was found to be solid. The following amounts of acidity were found in experiments made at different times:-At 10 days 0.33 per cent.; a t 11 days 0.34 psr cent. ; at 7 days 0.31 per cent. ; a t 14 days 0.35 per cent. ; at 21 days, 0-38 per cent. ; a t 31 days 0.44 per cent. of lactic acid. At 33" the milk became solid in two days; the acidity was then 0.32 per cent.A t 36" the milk curdled in one day and gas was distinctly produced. No peptone was found in old cultures either of the curdled or un-curdled milk. 2Cli'crococcus zcrece.-At 22" the milk begins to thicken in 12 days, and some days later becomes solid ; the curd is however always soft. The acidities determined were-at 10 days 0.17 per cent. ; a t 7 days, 0.16 per cent. ; a t 14 days 0.24 per cent. ; a t 21 days 0.22 per cent. The last-named culture was tested for peptone but only a slight re-action was obtained. The amounts of acidity in different experiments were-at 7 days 0.19 per cent. ; a t 8 days 0.24 per cent. Micrococcus gelatinosus.-At 10" the milk becomes solid in about 15 days. The curd afterwards very slowly shrinks or partly dissolves.Peptone was distinctly present in the whey of an old culture. At 33" a soft curdling takes place in 5 days OF SOXE MICRO-ORGANISMS. 737 A t 23" the milk solidifies in 2 days; the amount of acid a t that time is about 0.15 per cent. The curd very slowly becomes smaller ; the whey then contains peptone very distinctly. At 32" the milk curdles in 1 day ; the acidity is then about 0.14 per cent. On reviewing the results given by these five organisms some points of interest appear. The first three organisms apparently do little more than produce lactic acid. The amount of acid produced by the second and third is considerably larger than that yielded by the first, but with each organism the proportion of acid which can be formed seems to be nearly a fixed quantity ; a t least after a certain acidity is reached further increase takes place with extreme slowness so that apparently a culture of six weeks will not enable Staph.candidus to curdle milk at all nor will it enable Bacillus intestini and the Bacillus of infantile diarrhma to curdle milk a t 22". We may probably assume that when a certain proportion of lactic acid has been produced the further growth of the organism is checked and with this the pro-gress of chemical action. That the curdling which does take place is determined solely by the lactic acid produced appears very pro-bable €or the amount of acidity in cultures freshly curdled a t 33" of B. intestini and the B a c i l l ~ ~ of infantile diarrhea is practically the same as the amount found requisite to produce curdling in the pre-vious experiments with lactic acid.When we turn to the last two organisms M. urece and 174. gelatino-sus the circumstances are very different. 21.2 gelatinosus is far more active in curdling milk than either Bacillus intestini or the Bacillus of infa7atCle diarrhma for these fail entirely to curdle at 22" while M. gelatinosusnot only curdles in 2 days a t 22" but also after 15 days a t lo" and yet M. gelatinosus produces far less acidity than these two organisms. They fail to curdle milk a t 22" with an acidity of 0.38 per cent. ; J!. gelatinosus succeeds in curdling milk a t this temperature wit8h an acidity of only 0.15 per cent. ! It is clear then that in the case of Jf. cyelafiizosus some curdling agent other than lactic acid takes a considerable share in the reaction.The same may be said of the far less perfect curdling effected by 31. ureE as here too the amount of acidity produced is by itself insufficient to effect the curdling which takes place. Further light is thrown on this subject by the results afforded by the next) three organisms. BacillzLs.~zLorescens 1iquescens.-At 2 3 O milk seeded w i t h this organism thickens in 3 days and becomes solid in 5 days. The milk is neutral a t the time of curdling. The small quantity of whey a t the surface exhibits a slight bluish-green colour. I n a culture a month old the curd remained undissolved ; it had become distinctly acid. Tests for peptone in cultures a fortnight old showed none or a trace only.VOTI. LTII. 31 738 IVARISGTOS THE CHEMICAL ACTIOSS Spirilli~m of Asiatic Cholera (Koch).-At 22" the milk exhibited ft loose curdling in 4 days ; i t was then neutral. Milk made distinctly alkaline with sodium carbonate became solid in 3 days. After 25 days the milk was still solid ; it had then become distinctly acid. At 33" milk curdled firnily in 4 days ; in 11 days it was slightly acid. Alkaline milk became solid in 1 day and was then still alka-line. The curd in the last-named experiment shrank a good deal. The whey tested after 15 days was distinclly acid and p v e a strong reaction of peptonc. Peptone was also found i n smaller quantity i n cultures in the ordinary milk 11 days old." SpiriZZ.um of Cheese (Deneke).-This organism acts but slowly on milk.At 22" a bright yellow ring forms a t the surface of the milk in 4-6 days. A tough crust is then developed giving the milk the appearance of solidification. Later a t somewhere about 11 days the milk gelatinises. It is still neutral and has a butyric ociour. Still Iatci. the opaque jelly par ti all^ redissolves ; the whey becomes slightly acid and contains a distinct amount of peptonc. I n alkaline milk the gelatinisation begins in about 12 days and gradually increases in solidity. At 33" a similar course of change takes place but apparently more slowly. The milk after 21 days is found to be alkaline and little or no peptone is present. I n the case of the TI. j7.riorescens liquescens and the Spirillum of Asiatic cholera we have clear instances of the complete curdling of the casein without the formation of any acid ; the action is in fact similar to that produced by the ferment contained in reunet.It has been frequently assumed that when an organism does work exactly similar to that performed by a ferment that it does SO by the production of a ferment. There is a t least one undoubted instance of the production of a ferment by a micro-organism namely the formation of invertin by yeast. According to S. Lea ( J . l'hys. 6 136) the bacteria which concert urea into ammonia also act by the production of a ferment. Riicrophytes would t h u s appear to effect certain hydrolytic actions by the same agency as similar actions are accomplidred in the cell sap of plants of more complex organisstion. Facts and analogy thus seem to support the hypothesis that ferments are produced by bacteria; and until more certain light is thrown on the subject we may well assume with Hueppe and others that the rennet-like curdling of milk is due to the production of a ferment.With the information gained by the study of the last group of * Since writing tlic above Dr. Klcin has informed me that the f w u l t j of curdling milk lias not been hitlicrto rccogniscd RS belonging to Kocli's spirilluni. The cspcrimcnts liavc therefore been repeated but with thc same results. The gelatin cultures prcpared from the curciled milk BL] a cliecli gave normal growtlie OF SOME MICRO-ORGANISMS. 739 organisms we can now understand how M. gelatinosus and M. urem are able to curdle milk with the production of only a part of the acidity necessary to effect this purpose.They act in fact as ferment producers as well a8 acidifiers. The cheese spirillurn possesses only a feeble gelatinising power ; it appears to occupy an intermediate position between the group of rennet-like organisms and the group of peptonising organisms which we have next to consider. Hueppe names four bacilli and two cocci as capable from his own observation of gelatinising the casein and then redissolving it with conversion into peptone. According to him this is by no means an uncommon property of bacteria; he ascribes it to the presence of two ferments a rennet ferment and a trypsin ferment. Peptone has been looked for in the present experiments by the application of the usual test of sodium hydrate and copper sulphate ; and also by means of phosphotungstic acid a reagent considerably more delicate.The milk t o be tested was in every case placed in a small dialyser of parchment-paper ; after 24 hours the diffusate was examined for peptone. According to Vines and Green (Phil. Trans., 178 43) dialysis is needed to avoid a confusion between peptone and hemialbumose both of which give the same reaction with chemical tests whilst only the first will pass through a membrane. In the case of a large number of organisms small quantities of peptone have been found in the milk cultures especially Then they became old. The peptone is usually most abundant in cultures made at a low temperature (23") ; a.t a higher temperature (33") less pep-tone and more ammonia is found.Milk cultures have for the same reason more tendency to become alkaline at the higher temperature. The following five organisms do not produce a solidification of the milk which corresponds to typical curdling but the casein is more or less gelatinised and then redissolved. Bacillus subtiZis (hay ).-Milk cultures of this organism at 22" begin in two or three days to show a translucent space immediately below the surface. Each day this space increases and the milk is resolved into a nearly clear liquid above and a soft jelly below the latter steadily diminishing by re-solution. The clear liquid afterwards becomes turbid from bacterial growth. The reaction of the milk is neutral during the early stages of the reaction but becomes strongly acid later ; tho odour is then pungent and acids of the fatty series are clearly present.As soon as the separation into fluid and jelly is clearly established, peptone can be distinctly found and it becomes abundant when the re-solution of the casei'n has made some progress, At 33" the action commences earlier but the re-solution of the casein seems to proceed more slowly than at a lower temperature. 3 ~ 740 WVARINGTON THE CHEMICAL ACTIONS The bacilli from scarlatina scurf and from jeyuirity extract, peptonised as vigorously as the bacillus from hay. Bacillus anthracite.-The experiments were made a t 22". The action goes through the same course as that described in the case of B. subtilis the production of acids in the latter stage of the reaction did not apparently occur.B. Jlocczhs.-The peptonising action is apparently as vigorous as with the two preceding organisms. No production of acid was observed. SpiriZlum of Cholera nostra (Finkler).-The culture was made both a t 23" and a t 37". A t the higher temperature the entire gelatinisation of the milk preceded the formation of a clear solution. Peptonisatiou was more vigorous a t the lower temperature. This organism did not apparently act quite as rapidly as the three previously named. Bacillus tomZi)?ormis.-The experiment was made a t 22". The action was considerably slower than in any of the preceding cases. The milk did not gelatinise for rather more than a week and after 26 days one half of the jelly was still undissolved. The culture remained neutral.The remaining organisms produced little effect on milk although nearly all grew freely in it as was proved by making gelatin cultures from the milk many days after it had been seeded. Streptococcus scadatin@ is stated by Klein to curdle milk a t 37" in three days or a little later. I obtained no curdling with this organism. At 22" the milk became distinctly acid but at the end of 43 days was quite fluid. At 32" the acidity after eight days was equal to 0.07 gram of lactic acid per 100 C.C. ; the milk was quite fluid and did not curdle when placed in boiling water. A culture a t 37" continued for several weeks. was equally unsuccessful in producing curdling. Fresh cultures on gelatin were used for inoculation in all these experi rn ent s. On reporting these results to Dr.Klein he informed me that the cultures of Xtr. scarzatinae in his possession had also lost the power of curdling milk. Since then he has obtained a fresh supply of the organism and finds that as before it curdles milk completely a t 37', rendering it very acid. This loss of the power of producing lactic acid during a long series of cultures on gelatin is a fact of great interest. The organism I experimented with had been grown on gelatin €or a year and four months before the experiments in milk were made. The two BaciZZi of septicmnia make the milk distinctly alkaline a change which is visible to the eye as the milk loses much of it OF SOME MICRO-ORGANISMS. 74 1 opacity. The bacillus from the mouse acted more energetically than that from the guinea-pig.A trace of peptone was found in both cases in cultures three weeks old. The Bacillus of swine fever, B. fiuorescens and B. tardecrescens also slowly produce an alkalinity in milk. The Typhoid bacillus and Staphylococcus luteus slowly render the milk slightly acid without producing further change. B. suZphureus grew without producing any appreciable change. The SpiriZZum of noma did not grow easily in milk. I obtained a culture in alkaline milk only. My stock of Staphylo-coccus candidus liquescens and Jli'crococcus aureus was unfortunately dead before the milk experiments commenced. It is of great interest when one property of an organism can be correlated with another. I venture to think that these experiments with milk enable us to do this.The whole of the organisms which fail to gelatinise milk are organisms that do not liquefy gelatin. The three organisms first mentioned in this section which apparently simply attack the milk-sugar and produce lactic acid are also non-liquefying. On the other hand the whole of the organisms which act on milk as ferments liquefy gelatin. This conclusion is quite in accordance with the view already taken by some investiga-tors that the liquefaction of gelatin is itself an action produced by a ferment. We may venture therefore to predict that every liquefying organism will be found capable of gelatinising the casein of milk. This precipitation of the casejin may take place with or without the formation of lactic acid and with or without a subsequent re-solution of the case'in as peptone.Some experiments were made to see how the mixed organisms of soil would attack milk ; the soil employed was a heavy loam from an arable field long unmanured. At lo" milk seeded with a small fragment of soil became gelatinised in nine days ; the milk was then slightly acid. I n 11 days gas began t o appear. Half the volume of the milk was then an opaque jelly. The odour was very bad a circumstance that had not occurred with any culture made with pure organisms. Peptone was distinctly present. At 22" the milk curdled in 2-3 days and became slightly acid. In 4 days the evolution of gas commenced and continued active for some days. The curd slowly redissolved and the solution contained peptone very distinctly. The reaction of the milk was for some time only slightly acid but at the end of 5 weeks the acidity had become very large.The odour passed through various unpleasant stages, which it is difficult to describe. Soil thus showed itself possessed of organisms acting like rennet It produced no apparent change 732 WARlNGTON THE CHEMICAL ACTIONS and trypsin ferments. It differed in action from the pure organisms tried first by the production of much gas and secondly by the formation of putrefactive products having a powerful odour. TEE REDUCTION OF NITRATES. The reduction of nitrates to nitrogen in sewage and in waters containing sewage seems to have been first noticed by Angus Smith (Mem. Lit. and Phil. Xoc. Manchester 1867 [3] 4 56). He sub-sequently made many experiments on the snbject which are described in his Reports to the Local Government Board on the Pollution of Rivers 1882 and 1884.Schloesing in 1868 (Compt. rend. 66 237) showed that during the fermentation of tobacco juice or of putrefying urine or during the lactic fermentation of sugar any nitrate that was originally present disappeared nitrous oxide nitric oxide and nitrogen gas being produced. Schloesing further showed in 1873 (Cornpt. r e d . 77 353) that a rigorous reduction of nitrates to nitrogen gas occurs in moist vege-table soil when a change of atmosphere is prevented. Muntz has shown (Ann. Chinz. Phys. 1887 11 125) that under the same cir-cumstances chlorates are reduced to chlorides bromates to bromides, and iodates to iodides. Some experiments were made by myself in the Rothamsted laboratory in 1880 (Jour.Roy. Agri. Soc. 1881 332) on the reduction of nitrates in soil. 7 lbs. of arable soil were placed in a percolator, forming a column 8 inches deep. The soil was saturated with water, and sodium nitrate in quantity equal to a large agricultural dressing, placed on the surface. After a week water was applied daily for nine days in quantity sufficient to keep the surface covered and the drainage was collected and analysed. At the end of this time nitrates and nitrites ceased to appear in the drainage water Only 21 per cent. of the nitrates applied were recovered as nitrates and nitrites in the drainage. Large transverse fissures were formed in the soil by the production of gas. The power of soil to reduce nitrates to nitrites and finally to destroy the latter nitrogen being probably evolved was further shown by later experiments not hitherto published.To 125 C.C. of a sterilised 40 per cent. solution of urine contained in a bottle closed by a cotton-wool stopper about 0.2 gram of arable soil was added and the bottle placed in a cupboard having a tempe-rature of about 10". In five days nitrites were distinctly present, resulting from the reduction of the nitrates naturally occurring in the urine. I n L2 days neither nitrites nor nitrates could be found, the solntion giving no reaction with diphenylamine OF SOME MICRO-ORGANISMS. 743 To 100 C.C. of a 213 per cent. solution of urine containing 1 gram of nitre per litre about 0.5 gram of arable soil was added the surface of the solution was covered with a thin layer of paraffin oil, and the bottle then placed in an incubator at 20".In two days a slight evolution of gas commenced and nitrites were distinctly present. Nitrites have per-manently remained in this solution. A similar experiment was made at the same time the only differ-ence being that 5 grams of nitre and 5 grams of glucose per litre were present in the solution. Gas was evolved as before but in much larger quantity. In 11 days all nitrites and nitrates had disappeared. In a duplicate. experiment to the preceding but conducted at a temperature of 35" gas was evolved in one day and ceased in four days by which time the nitrites first formed had disappeared and no nitrates could be found. The oxidisable organic matter contained in the soil aud urine is seen by the first twa experiments to be capable of reducing only a small amount of nitrate ; the addition of glucose in the last' two experiments determined the reduction af a much greater quantity of nitratre and the final disappearance of nitrite from the solution.That the reduction of nitrates is due to a reaction between the nitrates and the organic matter present has been recognised from the first by every investigator. Meusel in 1875,(Jozlr. pharm. [4],.22 430) was the first to prove that the reduction of nitrates to nitrites in natural waters is brought about by the agency of living organisms which he pronoiinced to be bacteria. Deh6rain and Maquenne in 1882 (Compt rend. 95 732) were the first to establish the same agency in the case of the reduction of nitrates in soil.It has in fact been abundantly shown that if sewage or soil is sterilised by the action of heat or antiseptics no reduction of nitrates will take place. The conditions necessary for the reduction of nitrates include a nourishing medium and temperature suitable for the growth of micro-organisms ; also the presence of organic matter capable af oxidation. The exclusion of air is favourable to reduction but not essential to it ; thorough asration is however fatal to the process. When the organism is capable of growing at a high temperature such a temperature is most favourable to reduction. The organic matter suitable for effecting reduction is very varied. According to different observers albuminoids sugar propyl alcohol ethyl alcohol fats, glycerol glycol acetates and tartrates are all active in this respect.The amount of reduction if other conditions are equal depends entirely on the quantity of oxidisable organic matter present. Thus Gayon and Dupetit found that in sewage seeded with putrid urine, In nine days evolution of gas ceased 744 WARINGTON THE CHEMICAL ACTIOXS not more than 01-0.2 gram of nitre per litre was reduced ; while in chicken broth seeded from the same source 50 grams of nitre per litre could be reduced. The reduction may be to nitrites to nitric oxide to nitrous oxide or to nif rogen. The formation of ammonia has been sometimes observed, but the evidence generally points t o its origin in the decomposition of nitrogenous organic matter rather than from the reduction of nitrates.The formation of nitric oxide has been frequently noticed during the fermentation of sugar-beet molasses when the solution is not kept sufficiently acid (Compt. rend. 66 171 237). The pro-duction of nitrous oxide has been observed by Deherain and Maquenne during the reduction of nitrates by soil (Cowpt. r e i d . 95, 691 854). The differences in the products of reduction are determined :-1. By the conditions of the experiment; 2. By the specific nature of the acting organism Till within the last few years all experiments have been made with mixtures of various organisms such as naturally occur in the air of the laboratory in sewage or in soil. Working with such natural mixtures one is easilyled to conclude that the character of the reaction depends entirely on the composition of the solution or on the other conditions of the experiment.Thus Munro (Trans. 1886 667) found that river-water readily oxidised ammonia to nitric acid but after the addition of a tartrate reduction set in and all nitrades disappeared from the solution. From such facts some have concluded that not only the extent of reduction but the whole difference between oxidation and reduction is simply deter-mined by the composition of the solution. These ideas become, however greatly modified when we become acquainted with the specific properties of the different organisms which together produce the actions in question it then becomes evident that in many cases one stage of the work accomplished is performed by one organism, while a second stage is effect'ed by another.Within the last few years considerable progress has been made towards ascertaining the reducing faculty of individual species of bacteria. I n 1882 Gayon and Dupetit (Compt. rend. 95 1365) pub-lished quantitative results showing the amount of nitrate reduced in the same time by seven distinct organisms. The most active was a small, mobile bacillus producing few spores anaerobic. This organism, cultivated in chicken broth at 35" with exclusion of air reduced 9.6 grams of nitre per litre to nitrite in one day. Several other organisms only reduced a t the rate of about 0.5 gram per day. I n 1886 the same authors published a splendid research upon the reduction of nitrates by bacteria (Ann.de la science agronornique 1885 1 226). The products of the reduction of nitrates are very varied OF SOME lIICRO-ORGAISIS?rlS. 745 A good abstract of this paper will be found in this Journal (Abstr., 1886 823). The authors found that the reduction of nitrates to nitrites was a very usual property of the bacteria examined ; they met with only one species (they notice 11) which could be cultivated in broth containing nit're without occasioning reduction. They state that a large class of bacteria reduce only t o nitrites no nitrogen gas being produced. They experimented with but did not succeed in isolating in a pure state the organism producing nitric oxide. The-y isolated two bacilli from sewage which they named Racteriuiiz denitri-ficans oc and /i? ; these reduced nitrates to nitrogen gas nitrites being formed under most circumstances as a stage in the reaction.The action of the a-organism was under favourable circumstances very energetic the liquid in which it is cultivated becoming covered with foam and evolving in one day its own volume of nitrogen the temperature at the same time considerably rising. The same orgmism when grown in an artificial solution containing nitre and asparagine, produced a considerable amount of nitrous oxide ; when the aspara-gine was omitted no nitrous oxide was formed. Heraeus (Zeit. f. Hygenie 1886 215) grew 10 bacteria isolated from water and soil in solutions containing a nitrate and sugar. Six grew well in this medium and of these t w o bacilli reduced the nitrate t o nitrite and ammonia.In the autumn of 1887 I communicated to the British Association (Report 1887 653) a preliminary account of the results obtained with 20 organisms part of those forming the subject of the present paper. In March of the present year a paper was read before our Society by Dr. Percy Frankland (Trans. 1888 373) in whicb the author describes the results he has obtained respecting the reduction of nitrates to nitrites with 32 species of bacteria separated by himself from the atmosphere or from natural waters. The main series of experiments were made in a weak solution containing as organic matter 0.25 gram of peptone and 0.3 gram of sugar per litre at the tempera-ture of 30". Of the organisms tried 15 or 16 were found incapable of reducing nitrates to nitrit,es; it is to be remarked however that seven of these organisms produced either no visible growth or a very slight turbidity.It would seem possible therefore that under condi-tions more favourable to growth the proportion of active organisms might be increased. In the experiments with pure organisms which I have now to describe the power of each organism to reduce nitrates has been tested in nearly every case in two solutions :-1. Beef broth contain-ing 5 grams of potassium nitrate per litre ; 2. A 20 per cent. urine solution containing 1 gram of nitre per litre. In some case 746 WARINGTON THE CHEMICAL ACTIONS peptone has beeu added to the broth and glucose to the urine solution ; but these were for additional trials outside the main series.For the broth 2 Ibs. of lean beef were taken cut small treated with cold water slowly heated with stirring and finally boiled for half an hour. The filtered broth was made slightly alkaline to neutral litmus-paper by the addition of sodium carbonate brought to boiling and again filtered. The perfectly clear filtrate then received the necessary amount of nitre and was diluted t o 1200 C.C. The urine solution was in some cases clarified by making it dis-tinctly alkaline with potassium or sodium carbonate boiling and filtering ; the filtrate was then neutralised with. phosphoric acid. Urine thus prepared remains perfectly clear when heated for sterili-sation and also when it becomes alkaline a fact of considerable advantage if the growth of an organism is to be determined by turbidity or by the formation of a deposit in the solution.Urine so clarified is not however so nutritive as urine which has not been filtered after boiling. The nitrated broth or urine was placed in small wide-mouthed bottles previously baked at 140" ; each bottle received 100 c.c. which nearly half filled it. The mouths were closed with plugs of sterilised cotton-wool a paper cap tied on and the contents of the bottles sterilised by heating for several hours at a temperature near 100". The trials were as a rule made a t two temperatures ; the lower one varied at different times from 20-23" the higher from 32-35'. The formation of nitrite was ascertained by two reag,entu 1. The zinc iodide and starch solution of Trommsdorf ; 2. Metaphenylene-diamine.The first reagent is so extremely delicate that one is apt to conclude from its indications that a large amount of nitrate has been reduced when in fact the amount of reduction has been extremely small ; the extent of reduction has therefore always been judged from the results of the second test. No strictly quantitative determina-tions of the nitrous acid formed have been made. The reducing power of each organism must be concluded mainly from its behaviour in the experiments with broth. The reduction of nitrate in the urine solutions was never large the proportion of nitrous nitrogen very seldom exceeding 1 per million of the solution. The results with urine also varied a good deal at different times, according probably to the varying nutritive character of the solution.With freely growing organisms the reduction of nitrate generally takes place speedily (if it occurs at all) both in broth and urine solutions and most speedily at the higher temperature if the organism will bear it. To avoid however repeated testing of the solations the cultures in broth at 32" to 35" were seldom examine OF SOME MICRO-ORGAh'ISMS. 747 till at least three days old and the cultures at 20-23" not before five days. The cultures of slowly growing organisms and many of the cultures in urine were allowed a longer time before examination. Looking first at the bacilli which form the largest class examined, I am disposed to place the following three first as possessing the greatest power of reducing nitrat'es to nitrites. Bacillus JcEoccus., jluorescens non- lipuescens. ? of swine fever. In the case of these organisms the culture in broth gave an exceed-ingly strong reaction with metaphenylenediamine a large precipita-tion of the colouring matter taking place immediately. Next in order stand the following eight bacilli the cultures of which in nitrated broth gave a very strong reaction with meta-phenylenediamine but with no immediate precipitate of colouring matter. Bacillus intestini. , of typhoid fever. , of infantile diarrhcea. , of septictemia (mouse). , anthracis. , 7 ) (guinea-pig). Spirillum of Asiatic cholera. , of cheese. Far removed from these in reducing power stands-Bacillus subtilis. The hay bacillus does not reduce nitrates in urine even when glucose is added.In broth ah 22" there is frequently no reduction for a week or more but generally a slight reduction afterwards occurs. l n broth at 35" a small reduction can be perceived after a few days. Reduction does not apparently occur till the broth becomes alkaline. The addition of peptone favours the reduction. The largest amount of nitrite noticed has probably not exceeded 5 parts of nitrous nitrogen per million of solution. B. subtilis from the scurf of a scarlatina patient gave the same amount of reduction as the bacillus prepared from hay. The bacillus from jequirity extract gave a somewhat greater reduction. Frank-land speaks of B. subtilis as giving no reduction; this is probably owing to the small amount of reduction it occasions when it feebly nutritive solution is employed.The following six bacilli have given no reduction of nitrates to nitrites : 748 WARINGTOW THE CHEMICAL ACTIOSS Bacillus Juorescens lipuescens. , toruliforinis. , sulyhureus. , tarclecrescens. Spirillum of cholera nostra. , of noina. 13. fluorescens liquescew and B. su@hureus have not been tested a t a temperature above 23" as they refused to grow in the hot incubator. H. toruliformis Finkler's spirillum and the spirillum of noma, have each been grown with an attempt at complete exclusion of air, the surface bf the culture-liquid being covered after seeding with a layer of paraffin oil ; this addition did not a t all prevent the growth of the organisms. The use of paraffin did not in any case determine a reduction of the nitrate.It was thought a t first that the layer of oil would effectually protect the liquid from contact with oxygen but a subsequent experiment in which a solution of cuprous oxide in ammonia was substituted for the culture-fluid showed that oxygen did in fact pass beneath the layer of oil. I hope a t some time to try the effect of an absolute exclusion of oxygen. Proceeding next to arrange the micrococci examined ascording to their reducing effect and classifying them as far as possible on the same principles as the bacilli we place in the first rank-Jficrococcus w e e . Xtaphylococcus candidus. , gel at irtosus. 7 luteus. These are apparently practically equal in efficacy to the three bacilli In the second rank stands one coccus-previously named as possessing the highest reducing power.Xtaphylocozcus candidus 1iquesceBs. Far removed from these comes-Streptococcu.s scarlatiim. This must for practical purposes be classed as without a reducing power for nitrates but in fact a trace of nitrous acid is shown 1.y the delicate iodide test when it is cultivated for some time in nitrated broth. Strept. scarlatince has been grown covered with a layer of paraffin oil without developing further reducing powers. One other micrococcus, Micrococcus aureus, has been grown in nitrated broth with and without a layer of paraffin on the surface without any production of nitrite OF SONE MICRO-ORG~AXISMS. 749 Among the organisms examined by Prankland it was the bacilli only which exhibited reducing properties ; among those now reported on the micrococci have proved to be in many cases as powerful as the bacilli.The table (p. 750) summarises the whole of the experiments on reduction. The letters R R r tr. 0 indicate the extent of reduction observed. The letters n. gr. signify “ no growth.” As already ment,ioned the results in urine varied a good deal a t different times the whole period of experiment being more than a year I n some cases single experiments showed a distinct amount of nitrite by the iodide test though none is mentioned in the table a majority of experiments made a t other times showing no nitrous acid. In several cases in which a large reduction occurred in broth a t R low temperature the experiment was not repeated a t a high one. It may be safely concluded that in each of these cases the reduction would have been equally great at the high temperature there being in none of these instances any ditficulty of growth at a high tempera-ture.Whenever an organism showed no or very little reduction a trial was always made at both temperatures. In several cases of small or no reduction the experiment was repeated with an addition t o the broth of 5 grams of peptone per litre; in no instance was the amount of nitrite increased by this addition the broth being in itself sufficiently rich in organic matter. The addition of glucose to the urine was chiefly tried with organisms which afterwards proved to be destitute of reducing power ; in one case in which the organism was able to reduce it,s action was dis-tinctly increased by the glucose.The whole number of organisms reported on is 25 ; of these 7 were entirely witlhout reducing power 1 only produced a mere trace of nitrite and 1 only a very small quantity. The remaining 16 reduced nitrates in broth with considerable vigour. Of those which failed to reduce one was of feeble growth produc-ing only a very limited turbidity in the broth but others in the same class were among the most vigorously growing organisms examined. The power of reducing nitrates is thus in no way determined by rapidity and vigour of growth. As far as I am able to judge the reduction which has occurred has been simply from nitrates to nitrites but this point can only be decided by strictly quantitative experiments. No production of gas has been observed in any of the broth cultures although the amount of nitrate present was quite sufficient to occasion visible gas if a reduction to nitrogen had taken place.B.$occzcs was grown in broth under a layer of paraffin with the especial object of observing any bubbles of gas that might be formed but without any such result Experiments on the Reduction of Nitrates to Nitrites by various i I n broth. I I 20-23". No. 32-3'7". -23 9 1 11 2 3 12 5 6 4 14 22 10 24 25 26 13 15 21 27 17 18 19 20 16 I Number of trials. --B.JEoccus 1 of swine fever 1 intestini . 1 of typhoid fever 1 of infantile diarrhea,. . 1 Sp. of Asiatic cholera 1 B. of septicaemia (mouse) 1 9 , (guinea-pig) . . .1 of anthrax 3 Sp. Deneke's 2 B. subtilzs 6 JEuorescens liiquescens 4 torulzyorrnis . 1 sulphureus 1 tardecrescens. 1 Sp. Finkler's . 1 of noma . 1 M. urea 1 gelatinosus. . 1 Htaph. EzLtezLs 2 candidus . 1 , , liquescens 1 Streptococcus scarlatina . 2 Juorescens non-lipuescens 1 M . aureus . 1 Reduo-tion. Reduc-tion. Number of trials. R R --I__-R R R n R R R R n It R tr. I) 0 0 0 0 0 R R R 0 0 R R -B -n R r 0 0 n. gr. 0 0 0 R R -R 0 0 2 1 -A - -1 I -2 2 2 3 2 1 1 4 2 1 2 1 5 3 - OF SOME MICHO-ORGANISMS. 751 this bacillus which was obtained from soil is thus not one of the clenitrifying bacteria of Gayon and Dupetit.Another means of jadgiug whet her reduction has proceeded further than the stage of nitrite is by studying the alkalinity of the solution. If the reduction of potassinm nitrate yields gaseous oxides of nitrogen, nitrogen or ammonia the solution must become alkaline from the formation of potassium carbonate. With the exception of the broth cultmes of Staph. candidus Ziquescens I have observations of the alka-linity of the broth cultures of each organism at the time when trial was made for nitrites; at this time 3-10 days from starting the cultures the reaction of the broth was slightly alkaline in cultivations of M. u r e a and B.$uorescens Ziquescens and decidedly alkaline in the case of B.$uorescens non-liquescens; in all other cases the broth was neutral or in a very few instances feebly acid.With the exception, possibly of the culture of B. fluor. non-Ziq. the reduction to nitrites would therefore appear to have occurred without the production of nitrogen oxides of nitrogen or ammonia. After a considerable time, most of the broth cultures became distinctly alkaline; but this in itself is no proof of the destruction of nitrites for many organisms render simple broth alkaline after growing in it fop some time the production of alkali being due to the destruction of the organic salts which the juice of meat contains. As far then as this imperfect evidence goes none of the organisms examined (with the possible exception of B. Jluor. non-Ziq.) possessed the power of reducing nitrates to nitrogen or to its gaseous oxides.That soil does actually contain organisms which are capable under certain circumstances of so reducing nitrates has been confirmed by my own experiments already described (pp. 742-3). Nitrites are not normally present in well aerated fertile soils or at least only to that minute extent in which they can be shown by delicate tests to occur everywhere. The drainage watera collected at Rothamsted only rarely contain sufficient nitrite to give a distinct reaction with the delicate iodide test. Misapprehension has some-times occurred on this point as some analysts are in the habit of reporting the quantity of nitrogen present as " nitrates and nitrites," without apparently ascertaining in each case the presence of nitrites in the water.NITRIFICATION. A good many investigations have been published during the last two years having for their object the discovery of nitrifying organisms or the farther elucidation of the process of nitrification : little success seems however to have attended these endeavours 752 WARI?;GTO?J THE CHXMICAL ACTIOSS Celli and Zuco (Gnzz. chim. Italians 17 99) found that a solution of ammonium chloride containing i+GG its weight of mercuric chloride passed and repassed two or three times a day through a column of sterilised sand and calcium carbonate gave after a few days a distinct reaction with diphenylamine. The amount of nitric or nitrous acid present was iiicreased when spongy platinum was sub-stituted for the sand. They conclude that the nitrification of ammonia in soil may take place without the intervention of an organism although organisms may assist in nitrification.Frank (Forschtingen auf clem Gebiete der Agdxlturphysik 10 56) expresses the opinion that nitrification is in greatest part an inorganic process. I n his hands soil which had been ignited was still capable of nitrifying ammonium chloride. Frank's experiments have been since repeated by Plath (Landw. Jahbiicker 16 89l) who comes to the opposite conclusion that soil when sterilised has no power of oxidising ammonia. We surely need not controvert a t length these conclusions of Celli and Frank contrary as they are to the accumulated evidence of many well-ascertained facts. I may perhaps refer to one of my own early experiments (Trans. 1878 44) which seems to give a distinct answer to the question at issue.Similar soil was placed in three tubes, through which air was daily drawn the air in one case containing a little chloroform vapour in another the vapour of carbon bisulphide. Where air alone was aspirated the nitric nitrogen in the soil increased from 8 to 50 parts per million ; where the air contained the vapour of chloroform o r carbon bisulphide no appreciable increase of nitric nitrogen took place. Although however we may strongly hold that the nitrifying power of soil is due to the action of a living organism we have still to explain the appearance of small quantities of nitric or nitrous acid in the experiments above mentioned. It is clear that in all such experi-ments the presence of traces of nitrites in the atmosphere must always be taken into account.I have already shown (Trans. 1881, 229) that distilled water cannot be freely exposed to air for any length of time without containing traces of nitrous acid and the presence of this acid is soon manifested if the exposure takes place in a room in TT-hich coal-gas is burnt. When therefore an experimenter on nitri-fication makes use of the diphcnylamine test which gives a blue coloration with 1 part of nitric or nitrous nitrogen in 10 millioiis of water or the far more delicate iodide and starch test which will indicate 1 part of nitrous nitrogen in 100 millions of watei' he will not improbably obtain a distinct reaction sooner or later without a living organism having had any share in the operation or indeed without any oxidation of ammonia having occurred in his solutions OF SOME MICRO-ORGANISMS.753 We turn next to the investigations with living organisms. Celli and Zuco isolated various organisms from the subsoil waters of Rome ; these organisms were then grown in a solution containing an ammonium salt. In every case the solution after a time gave a small reaction with diphenylamine. The amount of nitrous or nitric acid did not increase during several months even when a current of oxygen was passed through the solution. When the cultures were passed and repassed through a column of sand the reaction with diphenylamine was distinctly increased. All the organisms tried seem to have behaved in a similar manner. Adametz (Fomch. Agr.-P72ys. 1886,381) isolated 22 organisms from soil.At the end of five weeks an extremely small quantity of nitric acid was present. Heraeus (Zeit. f. Hygiene 1886 193) has conducted an investiga-tion on nitrification in Koch’s laboratory ; about 29 organisms were examined. He obtained a powerful nitrification when a large quantity of soil was placed in a solution of ammonium carbonate. The bactei-ial skin which formed on the surface of the solution in this experiment was also very active in setting up nitrification. I n 6 days the solution seeded with this skin oxidised 1.6 C.C. of indigo, and in 10 days 5.5 C.C. From this bacterial skin he isolated a bacillus, and a streptococcus ; these together with a bacillus and yeast isolated from fermenting urine were grown in an ammoniacal solution similar to that already employed.All the solutions gave a distinct reaction for nitrous acid with the iodide and starch test but the amount was tloo small to determine with indigo. About 13 organisms of well-known kinds were also grown in a 20 per cent. urine solution. Seven of these organisms produced in one day’s growth in the incubator suEcient nitrous acid to give a strong reaction with the iodide and starch test. Heraeus concludes that these seven organisms and the four previously mentioned possess oxidising pyoperties and me capable of forming nitric acid. Frank (loc. cit.) separated a number of organisms from soil but failed on trial to obtain nitrification with any of them. Leone (Atti d. R. Accademia (1. L i n c e i 1887 37) concludes from his experiments that all micro-organisms are more or less capable, under favourable conditions of producing nitric acid and that the same organisms in the presence of organic matter are capable of reducing nitrates.Muntz (Ann. Chim. Phys. 1887 11 128) also cautiously observes that organisms which appear identical with those which produce nitrification are capable of reducing nitrates when air is excluded. Manly Miles ( A g r i c u l t u r a l Science 1887 10‘2) writes as if he had FGur of these were grown in an ammoniacal solution. TOL. LIII. 3 754 THE CHEMICAL ACTIONS OF SOME MICRO-ORGANISMS. pure cultures of the nitrifying organism under experiment but he gives no description of the organism. P. Frankland (Trans. 1888 389) grew 32 orga,nisms obtained from air and water in a nutritive solution containing an ammonium salt for 40 days ; he obtained in no case more than a faint indication of nitrous acid.It seems very clea,r that not one of the investigators who have ex-perimented with isolated species of bacteria has obtained in his solu-tions more than a trace of nitrous or nitric acid; no one 72as obtained an amount that could be determined quantitatively. Another point which generally appears is that every organism tried gives nearly the same result. The failure to obtain the nitrifying organism in a separate form is most conspicuous in the case of Heraeus. He had in his hands a bacterial skin which possessed energetic nitrifying properties ; but the two organisms which he separated from this skin gave in a solu-tion of the same composition as that previously nitrified nothing more than the usual trace of nitrous acid.The statement of Heraens that seven of the organisms examined commenced the nitrification of a 20 per cent. urine solution in one day is. apparently due t o a mistake. My own experiments show that a urine solution of that strength cannot be nitrified by soil without the addi-tion of gypsum the commencement of nitrification in a strong solu-tion is also extremely slow (Trans. 1884 661). The nitrous acid which so speedily appeared in his solutions was doubtless due t o the reduction by the organisms of the nitrates naturally present in the urine (Trans. 1884 669). Celli and Zuco Leone and others apparently believe that the oxidising or reducing action of an organism is determined by the con-ditions in which it is placed and that the same organism is capable of discharging both functions.The capacity for exercising the entirely opposite functions of nitrification and the reduction of nitrates is indeed possessed by soil and by river water mediums which contain a multitude of different organisms but it cannot as yet be said to be proved of any single organism growing in a pure culture. The investigations of P. Frankland and myself have shown that all bacteria do not reduce nitrates even under specially favourable conditions and one that is capable of unmistakably oxidising ammonia has not apparently been at present isolated. My own results respecting the nitrifying power of isolated organisms admit of a very brief description.It has been already stated that two of the organisms B. sdphureus and B. tardecrescens had been separated from visible growths occurring in solutions which had nitrified. Two other organisms B. floccz~s and B. torulifornzis had been obtained from soil. The remaining organisms had not an origi SOME REACTIONS OF THE HALOGEN ACIDS. 755 that would suggest their possession of the nitrifying properties belonging to soil. The four organisms which appeared most likely to include one possessing nitrifying properties were submitted to several trials ; 20 other organisms were mostly tried but once. Four different solutions were employed the composition being varied with the view of obtaining growth with a minimum amount of carbon-aceous matter present. The examination of the solutions extended in every case over many weeks. When a fragment of arable soil was added to any of these solutions nitrification distinctly commenced within a fortnight, and in another fortnight all the ammonia (equal to about 25 of nitrogen per million of solution) had disappeared. None of the pure organisms experimented with gave any such result. A distinct reaction with diphenylamine was in some cases obtained but this did not appear to grow in amount although in such cases the examination was specially prolonged. The amount of nitric or nitrous nitrogen in the solutions did not apparently in any case exceed 1 per million and all of this could not be attributed to the action of the organism as the unseeded solutions in the incubator also gave some reaction with diphenylamine. When we have discounted the trace of nitrites probably obtained from the atmosphere there is clearly very little left that can be attributed to the action of the organism. The question whether any part of the nitrate or nitrite present was produced by the organism I am unable to decide; but it is quite clear that none of the organisms examined possessed any nitrifying power in any way comparable with that possessed by soil. An organism which nitrifies as soil nitrifies has yet to be isolated