VM dee wadunue ims a Ce ree ee ere oer ir PR PV a ai ee bye Sisk edule hae aed ee ae Adad wadadsaun day i ‘feria ad AA be p Ben aed addy eo ey er eee Ser ot ew hes 4 Ws) SORA i Nee ee in ive ¥ 4 treed 440d B47 4a Dae pra bere hee MS sak Pear eny sy AAV LA av 4 sa) RA ee dd POR ence viv pec Ree) Peek ea} ee OEE ¥ eee Wy Cece 15 i) Ho 4 A * Saat e raed ae tea dee te o4 4 VANE Y Cede SOUR stat ase me ney Purdy. yee | vay “ Cn Roe dia new 440 is adda baad PAae aa i ik ta ASIN i Oe Sie E Lea i tate ead abe Un} tad 4 ) Vowae a { wud ad ede eka Cay d dade v wo Lian ) vi ‘ PL oe) wa VR red ie Pro Wu ata a ; arava Ate aay Wed Watuldade te pik) edad Py 4 Fad hai ye Xt RA Hee < e aly Pain PNG a, Vay herr i ‘ Wii sy nN Ci} ies haath tet ec) gel yi. hey Yor Vie Re Fu) UA ee AE PRALTGE AS Aa a an relented a ey Paes ite ia a i AEs oly . ‘ 2 het? fh re Wee ee open aa aay Ud at Pah Kite me Data Mi Taryn bes eae AN at eC Bl dado ak vases AT Oe Ca ER SEAN CTR Ve CA RIE as SHUR AS) cer % f} Ve eed ei ae pe Kaew inhi AON) ak Ae Hee AR iS ay “aa ¥ 7 Solberg aie’ é a et Sead eae a ed POI MER daa n BOK) TUNE Coe CORA ai Pritt ee N Hoes) 4 5X fe wre} kpasaée . < spas cst * is ig ‘a ue Mee ’ i es Pian a ; f ae 8 ed WW bt does ie Buia IY Ae, ‘ aoa ” f Moe ote "be 4944 LA ROW gy te ee bide acre ost Rae ay ee RR) noe set ti 4% «yw Thea ay ily iy wt sah eat ea Ret) : ia apts ya hana Gy Pyrite oe Deter t Greece) Paes 7 eyes iad Hed, Hin beg aes EAHA bs AH ave banca ao Are ais é it Ai NG iaatah dod Shey F hhh ay ies mC Pe ea teeta 1 pated ae er ants bg { ny pron ; y pene P me i IAD Bat Bs how be em a aie ae ‘fet ry) Nata Neves ve ase ae a Gaeta! pe jk et ate “ My a, ¥ ‘ PEAT Dereon ey we + ae ; iti) tr: y = iz - ef 28 a ea teers as re b ree PROCEEDINGS OF THE ROYAL SOCIETY OF LONDON Series B CONTAINING PAPERS OF A BIOLOGICAL CHARACTER VOL. LXXXVI. LONDON: Printep For THE ROYAL SOCIETY anp Soup By HARRISON AND SONS, ST. MARTIN’S LANE, PRINTERS IN ORDINARY TO HIS MAJESTY. Avueust, 1913. LONDON : “HARRISON AND SONS, PRINTERS IN ORDINARY TO HIS MAJESTY, ST. MARTIN'S” LANE. ' CONTENTS. SERIES B. VOL, LXXXVI. No. B 584.—December 17, 1912. The Chemical Action of Bacillus cloace (Jordan) on Citric and Malic Acids in the Presence and Absence of Oxygen. By James Thompson. Communicated by PANU NUM ENC ON EIEN Sep urciactsceiinsp eaten westecsccmemsidstnscetitecenaals suwemesbitbensraceedachaosts The Origin and Destiny of Cholesterol in the Animal Organism. Part X.—On the Excretion of Cholesterol by Man, when Fed on Various Diets. By G. W. Ellis and J. A. Gardner. Communicated by Dr. A. D. Waller, F.RS. ............ 1c Some Experiments with Arsenphenylglycin and Trypanosoma gambiense in Glossina palpalis. By H. L. Duke. Communicated by Sir John Rose Bradford, Ee COVE SCC PERO Me an sath acetee sp cedeosetrsactpavrnde-cis-iscehceveeteronaeerdscdedsaccoczaness On a Gregarine—Steinina rotundata, nov. sp.—Present in the Mid-Gut of Bird- Fleas of the Genus Ceratophyllus. By J. H. Ashworth, D.Sc., and Theodore Rettie, D.Se., Zoological Department, University of Edinburgh. Communi- Cated evade Capblyraits whieh. Saal (PLabepll) heewauesrcst=ksaccasearsastevenasceeenertbansaceaces The Size of the Aorta in Warm-Blooded Animals and its Relationship to the Body Weight and to the Surface Area Expressed in a Formula. By Georges Dreyer, William Ray, and E. W. Ainley Walker. Communicated by Francis Gotch, HEE: Semrcetev ne tacccaekrncacievictese csl/essise senna sacha deabjeae se slcuscmeeasermcseaaseaedesctaee casts The Size of the Trachea in Warm-Blooded Animals, and its Relationship to the Weight, the Surface Area, the Blood Volume, and the Size of the Aorta. By xeorges Dreyer, William Ray, and E. W. Ainley Walker. Communicated by Hira CS) GOL CHMMEMEC ORME eee cactsea ec teiaecsccusis suctetalssesocantiecascua deviance vaueusves canes Notes on the Life-History of Trypanosoma gambiense, ete. By Muriel Robertson. Communicated by the Tropical Diseases Committee of the Royal Society. (CATOSUIET OL) armoire casein Scocbob re oaAbiee Cot ad Hee monn acer BaceriocnAcuee Bee aTentEe renee arr aaee No. B 585.—February 7, 1913. On the Comparative Anatomy and Affinities of the Araucarinee. By Prof. Robert Boyd Thomson, University of Toronto. Communicated by Dr. D. H. Scott, BEES Sen CAUDSUACE) wore dothaccstlacen sioeaecnnena suneet ead mem aunM Mele tome ala (Cth catasetes The Relation of the Islets of Langerhans to the Pancreatic Acini under Various Conditions of Secretory Activity. By John Homans, M.D. Boston. Com- municated by Prof. K. H. Starling, F.R.S. (Plates 2 and 3) ..................:00++ PAGE 19 31 39 56 66 1V The Metabolism of Lactating Women. By Edward Mellanby, M.A., M.B. (Cantab.), Beit Memorial Research Fellow. Communicated by Dr. F. G. Ts faye) ate Jal ay Se mcrcopoasocaoscaddsoo.qa0aeddcanda0cbanudcsobbusdsDUEDeUodoOAuoUbEMoDIDOsOdDoONESC27< Colour Adaptation. By F. W. Edridge-Green, M.D., F.R.C.S., Beit Memorial Research Fellow. Communicated by Prof. KE. H. Starling, F.R.S. ............... The Transmission of Environmental Effects from Parent to Offspring in Simo- cephalus vetulus. By W. E. Agar, Glasgow University. Communicated by IDR Ja Coeiaehan ere, IRIS, (CNIOSBIE)) Coocscoanoansacosonacsnoonenosdaadeenodnaooonsce On Negative After-Images with Pure Spectral Colours. By George J. Burch, M.A., D.Se: Oxon, FURS: | ceckececensensse siosew cles osecuceepiemase cteieistiie cas sts dane eee antes eee eee ; Contributions to the Histo-Chemistry of Nerve: On the Nature of Wallerian Degeneration. By Henry O. Feiss and W. Cramer. Communicated by Prof. Ky, Az, Schafer, PR.S.” @PlateyA) rec aceisiiseaticaseaeiieiseemsenensbareesesasteseesectt eee eaeeeee Factors Affecting the Measurement of Absorption Bands. By H. Hartridge, M.A., Fellow of King’s College, Cambridge. Communicated by Prof. J. N. Langley, PERS: «sisunwctiinnaectabaalete teas toms ouaedinbsidsioalss Sennwaloasoaamadeaaiinss salelte senetdeant eee No. B 586.—March 6, 1913. The Phenomenon of “ Narcosis Progression” in Mammals. By T. Graham Brown (Carnegie Fellow). Communicated by Prof. C. 8. Sherrington, F.R.S. ............ Trichromic Vision and Anomalous Trichromatism. By F. W. Edridge-Green, M.D., F.R.C.S., Beit Memorial Research Fellow. Communicated by Prof. E. H. Starling, PRS .cccevsccamesmumenane pict bine eis seeiclssenisne aiclettueicenione dels coasts eee shee Reet A Preliminary Note on a New Bacterial Disease of Piswm sativum. By Dorothy M. Cayley, Research Student, John Innes Horticultural Institution, Merton, Surrey.) Communicated ya Weebabeson Enna sseeeseesesscrtaree stent erseereee ease neeeae On the Manganese Content of Transplanted Tumours. By F. Medigreceanu, M.D. Communicated by Sir J. R. Bradford, K.C.M.G., Sec. B.S. ..........ccsceceeese seen The Influence of the Resilience of the Arterial Wall on Blood-Pressure and on the Pulse Curve. By 8S. Russell Wells and Leonard Hill, F.R.S. ...........0.00s0c0ee0es On the Non-identity of Trypanosoma brucei, Plimmer and Bradford, 1899, with the Trypanosome of the Same Name from the Uganda Ox. By J. W. W. Stephens, M.D., D.P.H. (Cantab.), and B. Blacklock, M.D., D.P.H. Communicated by Sir. Ross, KeCB:,, FAR Si ias.saesesedeesesasacanetetese rete Coe etter Ee ece ee Cee The Action of Adrenin on Veins. (Preliminary Communication.) By J. A. Gunn and F. B. Chavasse. Communicated by Prof. Francis Gotch, F.R.S. ............... An Apparatus for Liquid Measurement by Drops and Applications in Counting Bacteria and other Cells and in Serology, etc. By R. Donald, B.Sc. (N.Z.), D.P.H. (Oxf.). Communicated by Dr. L. Hill, F.R.S. ......... siolehclelsindaieome in eerenss PAGE 88 110 119 128 140 164 187 192 198 Vv A Preliminary Report on the Treatment of Human Trypanosomiasis and Yaws with Metallic Antimony. By H. S. Ranken, M.B. Glas., M.R.C.P. Lond., Captain R.A.M.C., Member Sudan Sleeping Sickness Commission. Communi- eine days Eley Geb lime Ah VS. dersees C,H,O;. 8 Mr. J. Thompson. Action of [Aug. 27, In three further experiments, the carbon dioxide and acetic acid only were determined, the following results being obtained :— C: d. é. Carbon dioxide ......... 2°86 2°36 2°39 grms. Acepiciicideperee erereres ew 0°70 0:32 ae. The molecular ratios CO2/C2H4O2 calculated by the formula a £ = 0:67. (1) 4C.Hs0,+4H20 = 9C2H,O2 + 6CO>. 1912.]_ _B. cloace (Jordan) on Citric and Malic Acids. 11 The effect of the presence of oxygen is to increase the relative proportion of carbon dioxide with respect to the acetic acid. This may be due to the occurrence of reactions such as the following :— (2) 4C.H;0;+ 2H20 + 202 = 8C2H,02+ 8CO2+ 2H20. (3) 4CsHs0;+402 = 7C2H,02+ 10CO2+ 2H,0. These give the ratios shown in the following table, which also includes the numbers actually found :— Molecular ratio, | Respiratory coefficient, ; CO, C,H,0, On | IDG rey el Lee -eopeeeesead 0°67 Bes esrb idcs0:.| 10 40 33. adeonreceseooucer 1:4 2°5 Found by experiment...... 0°97 | 1:0 1°56 } 1:18 — 2°35 — 3°2 It is also possible that the decomposition takes place in the same way in the presence or absence of oxygen, according to equation (1), but that in presence of oxygen a portion of the acetic acid is subsequently oxidised. Summary. (a) Malic acid is not fermented by BS. cloacew in the absence of oxygen. (6) Malic acid is decomposed by B. cloace in the presence of oxygen into _ earbon dioxide, acetic acid, and succinic acid, with traces of alcohol. The decomposition probably goes on in two ways; oxidation by atmospheric oxygen to carbon dioxide and acetic acid, and oxidation at the expense of another portion of the malic acid, which is thereby reduced to succinic acid. With a good supply of air the respiratory coefficient CO2/O2 and the molecular ratio CO2/C2H.O2 found agree well with the values given by a reaction proceeding as follows :— 3C;H,05+502 = 2C2H,02+ 8CO2+5H;0. (c) In contradistinction to malic acid, citric acid is fermented by B. cloace in the absence of free oxygen. In addition to carbon dioxide, acetic, and succinic acids, the products resulting from aérobic fermentation, formic acid is produced, while there is an increased production of acetic acid. The molecular ratio CO2/C2HsO2 found agrees with the value for the following equation :— (1) 4C;H,0;+4H20 = 9C2H,02+ 6CO2. 12 Action of B cloace (Jordan) on Citric and Malic Acids. (d) Citric acid is decomposed by B. cloace in the presence of oxygen into the same products as malic acid. The values found for the respiratory coefficient and for the molecular ratio CO2/C2H4O2 are intermediate between those required for the following equations :— (2) 4CgHsO, + 405 — 7C2HsO2 + 10CO2 + 2H.0. (3) 4C0.H:07+ 2H20 +202 = 8C2Hs02+ 8C02+4 2H20. This is probably due to the difficulty of maintaining complete aération of the medium during the experiment, the decomposition being therefore partly anaérobic and partly aérobic in character. (e) It is possible that the decomposition of citric acid by B. cloacw takes place in the same way in the presence or absence of oxygen, according to equation (1), but that in the presence of oxygen a variable portion of the acetic acid is subsequently oxidised. (7) Methylacetylearbinol, which is produced by the action of B. cloace on mannitol and glucose, is not formed in the fermentation of malic acid or of citric acid by this organism. REFERENCES. O. Emmerling, ‘ Ber.,’ 1899, vol. 32, p. 1915. Bosworth and Prucha, ‘ Journ. Biol. Chem.,’ 1911, vol. 8, p. 479. Thompson, ‘ Roy. Soc. Proc.,’ 1912, B, vol. 84, p. 500. G. Ritter, ‘ Central. fiir Bakt., II,’ 1907, vol. 20, p. 21. oo bo 13 The Origin and Destiny of Cholesterol in the Animal Organism. Part X.—On the Excretion of Cholesterol by Man, when fed on Various Diets, By G. W. ELuis and J. A. GARDNER. (Communicated by Dr. A. D. Waller, F.R.S. Received September 3,—Read November 14, 1912.) (From the Physiological Laboratory of the University of London, South Kensington.) In earlier papers of this series we have shown that cholesterol is never excreted in the normal feces of herbivorous animals such as horses, cattle, sheep, and rabbits. In the case of carnivora such as dogs and cats, provided the body weight remains constant, the cholesterol excreted in the feces can be all accounted for by that naturally ingested with the food. Klein in his experiments also arrived at a similar conclusion. Evidence was also brought forward which rendered probable the view that, in herbivora, at any rate, cholesterol is a substance which is strictly conserved in the animal economy, that when the destruction of the red blood corpuscles and possibly other cells takes place in the liver, their cholesterol is excreted in the bile, and that the cholesterol of the bile is re-absorbed in the intestine along with the bile salts, finding its way into the blood stream to be used in cell anabolism ; further, that any waste of cholesterol might be made up from that taken in with the food. This latter process would be limited in herbivorous animals by the fact that their normal food does not contain cholesterol, but isomeric substances such as phytosterol, which have to be converted into cholesterol before utilisation, and in carnivorous animals by the partial, or even complete, change of cholesterol into coprosterol which takes place under certain dietetic conditions. In man, under normal conditions, cholesterol is never excreted as such in the feces, but always in the form of coprosterol. It seemed therefore desirable to estimate the amounts of coprosterol found in the feces of man under various dietetic conditions. The opportunity of making such investigations was very kindly afforded us by Dr. R. H. A. Plimmer, who handed over to us the dried feeces collected during a series of experiments carried out in the Physiological Institute, University College, London, and published in the ‘Journal of Physiology, August 26, 1909, under the title of “A Metabolism Experiment, with Special Reference to the Origin of Uric Acid,” by R. H. Aders Plimmer, Maxwell Dick, and Charles C. Lieb. The subject of the experiment was a healthy man, aged 39, The three 14 Messrs. Ellis and Gardner. Origin and [Sept. 3, diets selected were chosen so that each yielded 110 erm. protein, 240 grm. carbohydrate, and 100 germ. fat per diem. The carbohydrate and fat constituents consisted of potato and butter, and the protein constituents of (1) beefsteak, (2) egg-white, or (3) herring-roe. The experiment was commenced with an ordinary mixed diet for one week. After this the beefsteak diet was administered, and this was followed by the egg-white. After one week the subject unfortunately suffered from an attack of influenza, so that the experiment had to be discontinued for about 10 days, though analysis of urine was continued except for a period of four days. The egg-white diet was then taken for a period of 35 days. The herring-roe diet concluded the experiment, but this diet was taken only for three days owing to its bad effect on the patient. The feces in this last period were not kept separate, but included with those obtained during the 35 days of egg-white diet. For further details as to the experiment the reader is referred to the original paper. Cholesterol Contents of Various Diets. The cholesterol content of the constituents of the various diets was determined by the digitonin method in the usual manner. Diet I.—Total per diem :— IBSSISIGEIR segs ogdosbades. 500 grm. POGAT OMS Nes acre Santee 800, J BLELUSIOMN ae OSS AA GRE RTS 5 100 _—s,, LUE Ens cosdoaasddosausane SOs. Meat, according to our own observations and those of others, may be taken as containing not less than 0:0685 per cent. of cholesterol—free and combined. Butter, according to Magnus-Levy, contains 0°19 per cent. A sample of “best dairy butter” obtained from Harrods’ Stores was found, by the digitonin method, to contain 0'1744 per cent. of cholesterol—free and combined. The total cholesterol ingested in Diet I would, therefore, be at least 03425 +0:1744 = 0°5169 orm. per diem. Diet I1—Total per diem :— Bigg-white| i. saqssse ee 800 grm. Potato. cc Pause epee 800 ,, Butter tach -e-keneeeee USO 5, SUL deaacauee ne ondancoe st 40) 5 It was found that neither potato nor egg-white contains any trace of cholesterol. The total cholesterol ingested was, therefore, 0°227 grm. per diem. 1912.] Destiny of Cholesterol in the Animal Organism. 15 Diet I11.—Total per diem :— Herring-roe ............ 500 grm. ROUaILO! “eeneechoneeecssce 800 __,, SULIDEIIO rede eicn cee os «a's lis} ee, SU tees onvee ress 80 100 grm. of tinned herring-roe (soft) of the kind taken during the experiment was found to contain 0595 per cent. of cholesterol—tree and combined. The total cholesterol ingested was, therefore, 2°975+0°227 = 3'202 germ. per diem. Treatment of the Feeces. The feces were supplied to us in a dry, finely powdered condition. Those from each series were thoroughly extracted with ether in a Soxhlet apparatus, and the fatty matter in the ethereal solution saponified with sodium ethylate. After separating the soap, the ethereal solution was thoroughly washed with water and evaporated to dryness. From the residues it was found possible to isolate a quantity of pure coprosterol by fractional crystallisation from alcohol. The mother-liquors were evaporated to dryness, and the residual coprosterol benzoated by means of benzoyl chloride in pyridine solution. The feces from Series I, III, and IV, containing an excessive amount of oily impurity, rendered the isolation of pure coprosterol benzoate so difficult that the residues were further treated with digitonin, and in this way an amount of digitonin coprosteride was obtained. In order to be certain that the compound so formed was none other than the coprosteride, we recovered the coprosterol in combination by means of the xylene method. Results of Analysis. Series I.—In this series the subject was fed on an ordinary mixed diet for seven days, Five stools were passed, and yielded 167-7 grm. of dry material. The patient’s weight was practically constant, varying from day to day from 75:8 to 76:2 kgrm.—average 75°99. ' The feeces yielded on extraction 67445 grm. of unsaponifiable matter, from which 41669 erm. of coprosterol were obtained. This would corre- spond to a yield of 0595 grm. per day. This daily output corresponds very closely to that found in the cases of one of us and another colleague on liberal diet from observations extending over a year. Feces from a public latrine, however, yielded a smaller quantity. Series I1—The subject was then fed for seven days on Diet I. His average weight was 70°3 kegrm., and varied on six of the days as follows: 76, 75°3, 75°3, 75, 75, 75°6. Four stools were passed, corresponding to 16 Messrs. Ellis and Gardner. Origin and [Sept. 3, 1181 grm. of dry feces. On extraction, this yielded 4492 grm. of unsaponifiable matter, from which 33306 grm. of coprosterol were obtained. The balance sheet works out as follows :— Total cholesterol (free and com- Total coprosterol excreted ...... 3°3306 bined) ingested with food ... 3°6183 Output peridicmipe:-.seee-seseeee 0°4758 Intake per diem) <2..0......-+-+n<« 05169 Difference, 0:0411 of cholesterol per diem absorbed. Series II1].—Then followed seven days on Diet II. During this period the patient was sickening for an attack of influenza. His weight varied as follows: 75, 746, 74:2, 74:3, 74:3, 74:2—average 744 kgrm. The average loss in weight from the average of the previous period was thus 900 grm., or an actual loss during the week of 800 erm. Three stools were passed, containing 118 grm. of solid matter. This yielded 3°9605 grm. unsaponifiable matter, from which were obtained 2°6505 grm. of coprosterol. The balance sheet works out as follows :— Total cholesterol ingested ...... 1587 Total coprosterol excreted ...... 26505 Tronics) [OEE CHET acoonnocncanceaccoce 02267 Outpubipenidie mitre tenes see 03786 Excess of cholesterol excreted ............ 1:0635 Ss mm joa? CHOWN Zog5dco008ce 071519 If we reckon the loss in weight as due to fat and protein tissues the loss of cholesterol in this way would be from 0°45 to 0°8 grm., which would largely account for the excess excreted over intake. Series 1V.—During the next five days the patient suffered from influenza. Two meals of Diet Il were taken and 35 grm. of dry feeces were obtained. The weights of the patient were as follows :—74:2, 72°6, 73:2, average 73°3. The feces excreted during the first four days yielded 1°7638 grm. of unsaponifiable matter, from which 13109 grm. of coprosterol was obtained. Total cholesterol ingested dur- Total coprosterol excreted ...... 13109 THAVE FHEIMOC! oocanannosnbapo0no029000 0°4534 Excess of cholesterol excreted ............ 08575 5 5 JOG! CHEN pacoeons0059 02144 The patient lost during illness 1°5 kgrm., so that about 1 grm. of cholesterol might be accounted for in this manner, and the excess of output over intake explained. Series V.—During recovery from illness the patient was fed for six days on an ordinary mixed diet, after which the experiment was continued. He was fed for 33 days on Diet II, with addition of various salts, for one day on 100 grm. egg-white and 250 grm. boiled-out meat, for another day on 250 grm. of boiled-out meat, and for three days on a total of 400 grm. ege-white + 1250 grm. herring-roe. 1912.| Destiny of Cholesterol in the Animal Organism. U7 The daily weights of the patient showed a steady decrease during this long period—from 73°8 to 70°7 kerm.—a total loss of 3:1 kgrm. The total weight of dry feces passed during the period was 665 grm. This yielded 20'2195 orm. of unsaponifiable matter, from which 14:°7324 grim. of coprosterol were obtained. Total cholesterol ingested with Total output of coprosterol ... 14°7324 1HO (0! dancobteddonoboseRaEGobaseenqgan 14:920 Output per diem.................. 0°3877 Ditto intake per diem ............ 0393 Excess of intake over output per diem, 0°0053. The loss in weight during this experiment amounts to about 82 grm. per diem. This is considerably less than the loss in Series III and IV, which is about 114 grm. and 250 grm. per diem respectively ; there must, therefore, have been an absorption of cholesterol going on in the intestines during the period. Conclusion. It would appear from these experiments that in man, as in the case of other animals, the excretion of cholesterol in the feces can be accounted for by that taken in with the food, provided that the body weight remains constant. If, however, a rapid loss in weight takes place, as in illness, the output of cholesterol exceeds the intake. Further work will, however, be necessary before this view can be regarded as fully established. Note on the Sterol Contents of Rabbit Feces—In Part VIII of this series of papers we described an experiment in which we succeeded in isolating cholesterol by the digitonin method from the feces of a rabbit which had been fed on extracted bran, but into the peritoneal cavity of which olive oil had been injected. The animal in question had lost nearly a third of its weight during the experiment. In consequence of this result it was thought desirable to submit the feces of rabbits fed on extracted bran, but which were not losing weight, to a more careful examination. For this purpose four rabbits were fed for about ten days on bran which had previously been roughly extracted with ether. The animals remained during the experiment perfectly constant in weight. About 1200 grm. of air- dried faeces were obtained. These feeces were extracted with ether and treated in the manner already described, and yielded about 3 grm. of unsaponifiable matter in the form of a stiff oil. This oil was dissolved in alcohol and mixed with excess of digitonin in alcoholic solution. The precipitate was filtered and washed with ether to get rid of oily matter, and repeatedly boiled out with methyl alcohol, in which it proved very insoluble. The oily matter separated from the compound did not give any sterol colour reaction in VOL, LXXXVI.—B. Cc 18 Origin and Destiny of Cholesterol in the Animal Organsm. chloroform solution with acetic anhydride and sulphuric acid. The digitonin precipitate, which the above treatment had not freed from traces of some fluorescent colouring matter, was finely powdered and decomposed by heating in xylene vapour. The clear xylene solution on evaporation gave a yellow, crystalline, oily solid. This was purified by repeated crystallisation from dilute alcohol, from which it separated when pure in microscopic hexagonal plates. It melted at 136°-137° C. and gave the usual sterol colour tests. The acetate, made in the usual manner with sodium acetate and acetic anhydride, crystallised from alcohol in glistening leaves. It was less soluble in alcohol than the original substance. It melted at 135°-136° C., but if heated very slowly, at about 130° C., the benzoate, made by the action of benzoyl chloride in pyridin solution, crystallised readily from strong alcohol, in which it was sparingly soluble. It melted at 142° C. to a clear liquid, which on cooling gave at the moment of solidification a brilliant green play of colours, gradually changing to brown. This substance was one of the phytosterols of the bran, which had not been eliminated in the rough extrac- tion with ether. The same substance was obtained by Dorée and Gardner from horse dung. ; No trace of cholesterol was discovered. This bears out our previous conclusion that cholesterol isnot found in the normal feces of herbivorous animals. We are indebted to the Government Grant Committee of the Royal Society for help in carrying out this work. 19 Some Experiments with Arsenphenylglycin and Trypanosoma gambiense 77 Glossina palpalis. By H. L. DUKE, (Communicated by Sir John Rose Bradford, K.C.M.G., Sec. R.S. Received September 28,—Read November 14, 1912.) The experiments detailed below were devised with a view to investigating the action of arsenic in the form of arsenphenylglycin upon Trypanosoma gambiense as carried by Glossina palpalis. It was found convenient to deal with the subject under four separate enquiries— I. Does the presence of arsenic in the blood ingested by a positive fly destroy the trypanosomes in that fly ? II. Does preliminary feeding of flies on blood containing arsenic have any effect on the subsequent development of trypanosomes in their interior ? III. If flies are fed on blood containing arsenic shortly after the infecting feeds on a gambiense monkey, are the flagellates still capable of development in the fly? If they can still develop, is the resultant strain arsenic-resistant in the blood ? TV. Has arsenphenylelycin any prophylactic action against the bite of a fly infected with 7. gambiense, and, if so, what is the extent of this protection ? In all these experiments it was deemed advisable to feed each box of flies for two consecutive days on the monkey in order to make certain, if possible, that each fly fed. For the same reason special attention was given to each box of flies handled, with the result that it was found that flies reluctant to feed in the morning would often, if given another opportunity some hours later, bite with greater readiness. In spite, however, of all precautions, it is impossible to be absolutely certain on this point. In the great majority of cases the flies fed readily every day. I. Does the Presence of Arsenic in the Blood imbibed clean an Infected Fly of its Flagellates ? To answer this query boxes of laboratory-bred G. palpalis, known to be infective with TZ. gambiense, were placed upon monkeys which had previously received a subcutaneous dose of arsenphenylelycin, 0-1 grm. per kilogramme body weight. The experiments were commenced 24 to 48 hours after the administration of the drug, and subsequently the period of time was increased. No change in the flagellates was dis- cernible if the interval between the giving of the arsenic and the feeding exceeded 72 hours. The flies were dissected by Miss Robertson at c 2 20 Dr. H. L. Duke. Haperiments with [Sept. 28, varying periods after the feeds upon the arsenic monkey. I have here to express my indebtedness to her for the description of the flagellates found. It was constantly noticed in the case of flies which had fed upon a recently inoculated monkey that the walls of the intestine were unduly brittle and friable, and clean dissection very difficult. This was to a great extent avoided by starving such flies for several days after the ingestion of the arsenic blood before dissecting them. In the following experiments various changes were noted in the flagellates in the fly, due presumably to the action of the drug in the blood imbibed :— Expt. 29.—Positive Box of G. palpalis. Date. Procedure. Remarks. | Oct. 1—2 ...... Fed on Monkey 418 24—48 hours after administration of 0 °1 grm. arsenphenyl- glycin per kilogramme. FEMS Sep ossmodrta Starved Ji sstosteieercncserocssecnsneceeneeeeaces 1 fly dies; nil. pil) Glesemea conn Fed on clean Monkey 502..................... Monkey 502 becomes infected. 2 flies die ; nil. Ay | Ba08 Sot onares Fed on clean Monkey 427 ............:.000008- Monkey 427 becomes infected. 1 fly dies; nil. Wy Gleeesctesres| Starveasandidissectedicacmemtnessesseceeeees 1+fly found. Description of the positive fly :— Gut: Shows flagellates very much modified, with swollen posterior end ; many dead. Proventriculus and thoracic gut: No flagellates seen. Salivary glands: Normal, though somewhat slight infection. It will be noted that for two consecutive days after feeding on the arsenic monkey this fly was infective to monkeys. The condition in this fly affords strong confirmation of the theory that the salivary gland flagellates are the true infecting form. Expt. 74.—Positive Box of G. palpalis. Date. Procedure. Remarks. Oct. 24—25 ...... Fed upon Monkey 488 24—48 hours after the administration of arsenphenylglycin, 0O°1 grm. per kilogramme. EEG Acconci Starved ace ciscc cs scmagueeeemenean een asceeceee 9 flies die, among which is + Fly No. 1. ARCATA Repo ee sete Fed on cock. Pia he Pera aaHage Hed on scock aylaicseeenck ceo eee eee 2 flies die, one negative, the other + Fly No. 2. » 29—Noy. 1...} Fed on cock. Nov. 2—4 ......... Fed on clean Monkey 508 ...............00208: Monkey 503 became infected. 1 fly dies ; nil. Be Da Owaienanees Starved and dissected...............cseesecseeee +FEly No. 3 found. 1912.] Arsenphenylglycin and T. gambiense i G. palpalis. 21 Description of the three positive flies :— Fly 1. Gut: Few altered flagellates, all dead, with swollen posterior ends, and generally altered. Proventriculus : Nil. Salivary glands: Only one gland obtained; this contained living and apparently quite normal flagellates. Fly 2. Gut and Proventriculus : Show no flagellates, living or dead. Salivary glands: Contained flagellates + + +, actively motile. Fly 3. Gut: +++, apparently normal. Proventriculus : Nil. Salivary glands: + + +, normal. Whether the condition of this fly is due to the original gut flagellates having escaped destruction by the arsenic, or whether in the 10 days which elapsed between the arsenic feeds and dissection the gut became reinfected from the salivary glands, cannot be decided. In relation to the first alternative it is improbable that a fly would refuse to feed upon two successive days unless at the point of death. The mortality after the arsenic feeds should be noted in this and other similar experiments. Expt. 357.— Positive Box of G. palpalis. +, EP ea PB ISR E CHOU S oxic tee eet set oe casas ohn oe 1+ fly found. Date. Procedure. Remarks. | ; | | Oct. 10—11 .., Fed on Monkey 452 24—48 hours after administra- | tion of arsenphenylglycin 0:1 grm. per kilogramme. | eepe Lt ecard SS bear VOU te esc eae aay tee eM Me ec casi Seshus css emieponns | 7 flies die; nil. (This box was killed by wood-smoke instead of the usual chloroform.) Description of the positive fly :-— Gut: Flagellates all dead, shape considerably altered, the posterior end being frequently swollen up to a marked degree. Salivary gland : Numerous flagellates, but all dead; shape perfectly normal. Apparently this curious condition of the gland flagellates was due to the smoke. The effect of this treatment is very marked; the glands show a dark granular appearance very different from the clear transparent state to be observed in normal specimens. This darkening of the glands proved by subsequent experiments to be a characteristic of smoke-killed flies. This condition has never been observed in flies killed with chloroform. Note the mortality after the arsenic feeds. Dr. H. L. Duke. Haxperiments with [Sept. 28, Remarks. 26 flies die; nil. Monkey became infected. 1 fly dies; nil. 1 fly dies; nil. 22 Expt. 383.—Positive Box of G. palpalis. Date. Procedure. Oct. 24—25 ...... Fed on Monkey 487 24—48 hours after | the administration of 0:1 grm. arsenic per kilogramme. Peel ammecodscecas Starved. SS =28) reese Fed on Monkey 493 24—48 hours after arsenic 0'1 grm. per kilo. Pe Si penrecocoree: Starved. PRC UME ei looatas Bed ioni(cocky..c.c; uncentar noses ccteeorerneeees » 81—Nov. 2...| Fed on clean Monkey 502 (a) ............ INOW, Basocosscacesece SU aE lopndinggosoncn anacoqasuubedotooanen dadobbads SyntAltoaiacsntenosen He Starved and dissected..................00008 + Flies Nos. 1 and 2 found. Description of the two positive flies :— Fly 1. Gut and proventriculus: Devoid of flagellates, alive or dead. Salivary gland: +-+ +, with normal active flagellates. ++, no alteration in form observable. Foregut: +, no alteration in form observable. Fly 2. Gut: Proventriculus : Nil. Salivary gland: +++, normal. In this experiment the flies were afforded an opportunity of a second feed of arsenic blood. In spite of this the salivary flagellates—the infecting form— were, it appears, unaffected. Apparently fly No. 2 illustrates re-infection of the gut from the salivary gland, as it is very improbable that any of the original gut flagellates would survive the double feeding on arsenic blood. Note the mortality after the arsenic feeds. Expt. 358.—Positive Box of G. palpalis. Remarks. | Date. Procedure. Oct. 10—11 ...| Fed on Monkey 488 24—48 hours after administration of arsenphenylglycin, 0:1 grm. per kilogramme. op JL caos0n000 Starved. T OHLLONE tenets Dissected! 5.suin.3/eac bead eae erat eee eee +Elies Nos. 1, 2, 8, 4 found. ° Description of the four positive flies :— Fly 1. Gut (containing no blood): ++ + active, normal. _ Proventriculus: +, normal. Salivary glands: +++, active and normal. This fly showed no sign of blood in its gut, and therefore presumably did not feed on Monkey 438. 1912.| Arsenphenylglycin and T. gambiense in G. palpalis. 23 Fly 2. Gut (empty forward with a little pale fluid posteriorly): Some tendency towards posterior swelling, but still active. Salivary glands: +-+-+, active and normal. Fly 3. Gut (containing altered blood): +++, active, but some show posterior swelling. Salivary gland: ++, apparently normal. Fly 4. Gut (containing altered blood): +++, few dead, many altered in shape but still slowly motile, others active. The flagellates of these flies show less effect from the arsenic than is usual at this period. It may be remarked, as justifying the conclusion that the above departures from the normal in the flagellates are due to the action of the drug, that such alteration in form and motility have never before been seen in positive flies. The dead flies are removed every morning, and in the normal course of events the flagellates are always seen to be actively motile with no such morphological changes as those described above. Moreover, in cases in which the normal flies have been left until the flagellates are moribund, death occurs without the characteristic swelling of the posterior part always to be observed in the presence of arsenic. ; Expt. 550.—Positive Box of G. palpalis. Date. Procedure. | Remarks. Jan. 31—Feb. 1...) Fed on Monkey 632 48—72 hours after | administration of arsenphenylglycin, 0°'1 grm. per kilogramme. INGOs 71 Saoneondorsobed Starved. | Mea Ore rcecioeriaerae aac MDIESECHEM estat snc sNeticeacestectmenes segatttee | + Flies 1 and 2 found. | Description of the two positive flies :— Fly 1. Gut: +++, normal (no traces of blood seen). Salivary glands: +++, normal. Fly 2. Gut: ++, a few are altered as described above. Salivary gland: +++, normal. A very slight effect appears to have been exerted in Fly No. 2 by the arsenic. Fly No. 1 either fed very slightly or not at all on Monkey 632. 24 Dr. H. L. Duke. Eaperiments with [Sept. 28, Expt. 567.—Positive Box of G. palpalis. Date. | Procedure. ‘ Remarks. Jan. 20—21 ...... Fed on Monkey 620 48—72 hours after administration of arsenphenylglycin, 0-1 grm. per kilogramme. Bp PAD) conn | Starved. Pre die neriio sonue | Dissected ........... sas quscosootencaa sutuaaoobos 1+ fly found. | | Description of the positive fly :— Hindgut : Nil. Posterior part of thoracic gut: Normal, + ++. Anterior part of thoracic gut: Nil. Proventriculus: Nil. Salary gland: +++, normal. From the above experiments it will be seen that the flagellates in the gut of flies fed upon a monkey within 24 to 48 hours of the administration of arsenphenylelycin in doses of Ol grm. per kilogramme are markedly affected. The flagellates in the salivary glands are apparently not injured in any way, nor does the fly lose its power of infecting. This evidence supports the theory that the salivary gland flagellates are the normal infecting agents. In a paper shortly to be published, Miss Robertson brings forward a further mass of evidence to support this conclusion. There is no reason to doubt that the secretion of the salivary gland is poured out into the wound made by the fly’s proboscis at the commencement of the act of feeding. Whether or not this process is repeated during the course of feeding cannot well be determined. These functions of the salivary gland and its contained flagellates are well borne out by the interrupted feeding experiment to be referred to shortly. It is plain that any attempt to clean a positive fly of its flagellates by feeding it upon an animal whose blood contains arsenic will fail, as the gland flagellates will not come into contact with the drug. Il. Has the Preliminary Feeding of Flies on Arsenic-containing Blood any Effect on the Subsequent Development of the Flagellates in their Intervor ? The first pair of experiments, Nos. 336 and 337, devised to elucidate this point proved fruitless, as no positive flies were found either in the arsenic box or its control. 1912.) Arsenphenylglycin and T. gambiense in G. palpalis. 25 Expt. 712. | Date. ‘ | Pee zs Procedure. Remarks. | A\y SJE)" aAscpoonaeoe | _— Fed on Monkey 708 48 hours after the administration of arsenic, 0°1 grm. per | kilogramme. PLO fb fo asi. 1—2 | Fed on Monkey 711 24—48 hours after the administration of arsenphenyl- glycin O'1 grm. per kilogramme. fy WHS) | O-820) 2715 | 20-4 | 1:05 | 176 22-16 =| -1°6 34-4 | : 8 | 9 | 210 | 0-796] 1°99 | 22-4 | 0-95 | 1°79 11-17 | 1°64 | 21-4 Sei ile 2200 | Ori4o |) 1-74 | 26-5 9) 0-79 | - 2-85 5°95 | 1-72 1°16 TO" +2 |) 220° 0-752) 1-77 | 26°0 | 0-8 | 1-85 4°32 | 1-72 2°91 | ess, J280, ).0--749)), 1-76) -| 27-0 | 0:77 |: 1-91 7:85 | 1-79 1°68 Pleo | 230° | Oreo) 1-71 127-8 | 0-74 | 1-91 10-47 elo 4°47 | j H 13) go | 232 | 0-952 | 2°84 | 16:8 | 1:22 | 1-92 47-92 | 1°81 | 56°9 | > Piss) ¢ | 800 | 0-78 |. 1-91 | 30°0.| 0:64 | 2:3 16°96 | 2°34 | 18-4 | ; 15g | 300 | 0-854) 2-28 | 25-2 | 0-76 | 2:93 0°87 | 2°34 2-56 4 }16 | g | 305 | 1°03 | 3:34 |17-4 | 1°09 | 2:33 43°34 | 2°38 | 40:3 ete a) 3820 | 0°98 2-77 “| 21-7" | O87 | 2:41 14°94 | 25 | 10:8 Hel S|) 380.4] 02745 | 1-74. 11.35°3 |, 0°53 | 2-47 29°55 | 2°57 | 6°62 HOmeoa| (340 | LOW | 3:22 | 20°7 | O-87 | 2°67 | 20°6 2°89 | 11-4 20 | 9 | 3870 | 0°933 | 2-78 | 24-4 | 0-74 | 2°67 Revol le i289 |) Sok 121 | ¢ | 400 | 0-971 | 2:96 | 28-8 | 0-74 | 2°83 4°59 | 3-12 5-16 22 | g | 420 | 0-905 | 2-57 | 28-4 | O-61 | 2-93 12°29 || 3:28 | 21°6 Paolo) 4805} 08. | 3-32 22:3 |. 0°77, | 2-98 1141 | 3°35 0-89 124 | g | 490 | 1:02 | 3-24 | 25-1 | 066 | 3:26 | O61 | 3°82 | 152 25° |) 9) 600° | 0-955 | 62-87 | 28-7 | O57. | 3-31 | 13-29 | 3:9 | 25°4 126, | 9 | 550 | 0-939) 2-77 | 31-9 | .0°5 3°54 B75 | 4-29) | 354 eae te |esoo fae lom (aanion P20 ph O75" | S57 16-25 | 4°33 | 4-16 PS Ne) 600, ed 29 01 i522. | 1800.) 0:87, |, 3-27 38-46 | 4°68 | 11°5 | 29 | 9 620 | 1-13 4:03 | 23°8 0 “65 3-86 4-4 4:84 | 16°7 430 | g | 640 |1°3 5°38 | 18-4 | 0°83 | 3°95 34°94 | 4°99 | 6°81 131 | 9 | 640 | 1°03 | 3°36 | 29°3 | 0°53 | 3°95 14:94 | 4°99 | 32-6 Pao Fey a YOSsoR Pe77 4 Obs |. 4-37 10°30 | 5°76 | 22-9 esa ied) 400) 4) 2:29) | 5-22 | 21-9 | 0-66 | 4:58 107 6) oG1Gs.| 15:6 34 | 9 | 880 | 1:21 | 4°37 | 25-9 | 0°55 | 4°75 3-79 | 6-47 | 29-4 Be lage i) Go0n || M23 4} 4-7 | 27-6e),). 0-5 5°22 9°77 | 7:41 | 36°6 | : = : Average ......... 24°9 0-78 — 16 -44 — 19 68 A. Mammals.—In Table I are given the figures and calculations for 35 guinea-pigs ranging in weight from 130 to 950 grm. (ze. increasing more than sevenfold). From this table it is at once evident that, as would be A2 Messrs. Dreyer, Ray, and Walker. [Sept. 30, expected, the radius of the aorta increases much more slowly than the weight of the animal. The area of the aortic cross-section also increases more slowly than the body weight (though of course much more rapidly than the radius), so that the ratio of the sectional area of the aorta to the body weight decreases steadily as the weight of the animal increases. But it appears on calculation that the body weight (W) to the nth power (where x is approxi- mately 0°70-0°72) divided by the sectional area (A) is a constant (4). This gives us the formula W”/A =, which indicates that the sectional area of the aorta is a simple function of the surface of the body since, as was shown in a previous paper, the body surface, which can be calculated from the formula S = kW” is more accurately determined by taking 2 to be approximately 0°71—0°72 than by taking it equal to 2/3 as was done by Meeh (5). Now it has been proved on former occasions that the blood volume is proportional to the body surface, hence it follows that the sectional area of the aorta is proportional to the blood volume of the individual. Table I further shows that the average value of / is 24-9, corresponding to an ” of 0°71, which is by calculation the best 1 for these individuals, and that if the aortic cross-section be calculated from our formula A = W"/k using these values for n and k, the average percentage deviation between calculated and observed values is 16°44. If, on the other hand, the sectional area is expressed as a percentage of the body weight (0°78), the average deviation between the calculated and the observed values is 19°68 per cent. It may be stated further that if the value 0°72 is taken for n, the average value of & becomes 26°5, and, if the sectional area is calculated from A = W°”/26:5, the average percentage deviation between the calculated and the observed figures is found to be 16°53. In order to bring out the various points more clearly, to get rid of irregularities due to individual variations in the animals, and to diminish the influence of experimental error, the animals have been grouped in Table II. In this table the guinea-pigs are arranged in five groups according to weight, and the weights, the aortic radii, and the aortic cross-sections of the animals in each group averaged. The other figures are calculated from these average values. It is found that, under these circumstances, the best is 0°72 (exactly as we found to be the case in calculating the surface from the body weight), and the average value of k is 25°6. Using these values for 1 and &, the average deviation between the calculated and the observed figures is 2°97 per cent., whereas, if the sectional area be calculated as a percentage (0°78) of the body weight, the average percentage deviation becomes 14:2, that is to say, “nearly five times as large. Moreover, it will be observed that, while the & PF Oe PO ee =. . a. oe Lo 1912.] The Size of the Aorta in Warm-Blooded Animals. 43 Table IJ.—Guinea-pigs (grouped). (26 ee ro ae qe | 3 5 | se, |) iS me) |p eae sé we = 3S | org | hecWes | es.) 3 Ae |e ale | 23 fee ea fos || So a ex-| 3 (re ge Pl eon. Bey Ara ReS |e acai | Bee |S Peele hae | 28 Loosead | Seomle es) es | tao | RS | 2 Sei S $28 | | me rs) | So) I oun 5 > = | S = cee | S| £8 ep iewian bk tees || bea ts eq =| 2 223 ope 2) 3a les | 4 = |< tS) 1A i grm. mm. sq.mm per cent per cent | A d-5 | 156 | 0°679 | 1°45 | 26-2 | 0-93 1-48 2°03 1°22 18°85 B 6-13 | 218 | 0°794) 1:98 | 24-4) O-91 1°89 4°76 <7/ 16°47 C | 14-20 | 328 | 0:904) 2°57 | 25-2) 0-78 2°55 0°78 2°57 0:0 | D | 21-27 478 | 0°998| 3°13 | 27:1) 0°65 3°32 5°72 3°73 | 16°09 E | 28-35 | 726 | 1:204) 4°55 | 25:2) 0°63 4-48 1°56 5°66 19°61 | PACT EE Gre sce =e ian 25°6 | 0°78 = 2-97 — 14 °2 exhibits no periodic variation as the weight of the animal increases, the figure representing the sectional area in percentage of body weight decreases with absolute regularity from 0°93 to 0°65. As regards the question of sex, if the males and females be considered separately in Table I, it will be seen that the average i for the 18 males is smaller than the & for the 17 females. Thus, with 2 equal to 0°71, the & for the males is 23°9, and that for the females 26, while with n equal to 0°72, k is 25-4 for the males and 27-6 for the females, indicating in each case that the male animals had somewhat larger aortas than had the females of corresponding weights. But this is a point to which we sball return. In Table III are given the figures and calculations for the aortas of 27 rats, ranging in weight from 301 to 303 grm. (Ze. increasing more than tenfold). The average aortic constant (/) is 21°37, with an n of 0°71, which is the best 7 for these observations, and the average aortic percentage (z.e. sectional area of aorta expressed as a percentage of body weight) is 1°27. It is seen that, as in the case of the guinea-pigs, the variations of the aortic constant show no periodicity, but the aortic percentage decreases markedly and steadily, although not regularly, as the animals increase in weight. If the area of the aortic cross-section is calculated by our formula, the average deviation between the calculated and the observed figures is 11-09 per cent., while it is 19°72 per cent. or nearly twice as large, when the area is calculated in per cent. of body weight. If the value 0:72 be taken for n, k becomes 22°36, and, if these values be used in calculating the sectional aortic area by our formula, the average percentage deviation of the observed from the calculated figures is 11°37 per cent. 44 Messrs. Dreyer, Ray, and Walker. [Sept. 30, Table IJJ.—Rats (individuals). 1 |} fa 6 j 1 am . | a Glee z 8 5 aa a S 5 4g | 2 as = Ea @ | Se Bao =) ore SS A BOF Bo Waa One : | S See Sse owas | = F , | A eo wn Ske ag | So) ge a4 nde 4/28 | §Fa | 25 | Se? | 3c "eo Soon F = 2) i) Ss @d8 | BSp| gag oa) ie) | 4 & 3 Sees celal Se auc) Sig > | 2 | oe | Ee lees) Pe | bee teas 6S ies S 2B I HB | 6 =sS | $s0|855 ai i) pQ es | =< x = iS) A .S) A erm. mm. | sq. mm per cent. per cent, i i @ 30°1 | 0:486 | 0°597 18°78 1-98 0-525 13-72 0 °382 56°28 2 io) 39:1 | 0°417 | 0°546 24 74, 1°4 0 632 13 61 0-497 9 84 3 3 40-3 | 0°468 | 0-681 20-08 i o7/ 0 -646 6°35 0°12 34°18 4 é 47-2 | 0°455 | 0°657 23 71 1°38 0 °722 9°83 0 ‘599 8 68 5 | 29 | 53:0 | 0-493 | 0-764 | 21°94] 1-44 | 0-784 2-56 | 0-673 | 13°52 6 3 56:0 | 0-510 | 0°815 21°38 | 1°46 | 0-816 0°12 Opor/lit 14°63 7 2 57°7 | 0548 | 0941 18:92 | 1-63 0-833 12-97 0°733 28°37 8 | 2 69°7 | 0°566 | 1:01 20°16 1:45 || 0-958 5°98 0-885 14°12 9 g 75°5 | 0°599 | 1°12 19°23 | 1:48 ao 1:09 0-959 16°79 10 g 79-5 | 0:580 | 1-06 21 -09 1°31 1-05 0-95 1°01 4:95 11 rh 84°5 | 0-669 | 1°4: 16°67 | 1-66 1-09 28 44 1:07 30 84 12 | ¢ 96-7 | 0°484 | 0°735 34 93 0-76 1-2 38 8 1°23 40°24 13 | g | 104°0 | 0°643 | 1°3 20°81 | 1°25 i SA7/ 2°36 1°32 1°52 TA | OSE O OL 77; 1 ‘86 14:94 | 1°72 1°3 40 °38 1°37 35-77 15 | fg | 140°0 | 0-669 | 1°4 23 85 | al, (0) 1°56 10-26 1°78 21°35 16 Be | Loy |) @sOr |) ak 557 PAL sts} |) IL cilil 1-57 0-0 1°79 12-29 17 36 | 141°0 | 0-669 | 1-4 23°97 Ooo Tog 10 83 1°79 21°79 18 Q | 141-0 | 0-678 | 1°44 23°31 Lewy | i wy 8°28 1:79 19 -55 19 6 | 157-0 | 0-723 | 1-64 22:09 1:04 | 1-69 2-96 1:99 17 -59 20 6 | 15736) | 0:77 1°86 19 -54 1:18 ii 9°41 2:0 “sO 21 3d | 160°0 | 0-819 | 2-1 17-47 1-31 1°72 22 09 2-03 3°45 22 g | 164°0 | 0-723 | 1-64 23 °32 1-0 17/5) 6°29 2-08 21°15 23 6 | 172-0 | 0-812 | 2-07 | 18 ‘67 1-20 1 eill 14°36 2°18 5°05 24 Oo 50) | (OES2) EGS | 23-29 0-96 1°83 8:2 2°22 24 °32 25 & | 179°0 | 0°755 | 1°79 22-22 1-0 1°86 3°76 2-27 21-15 26 6 | 289°0 , 0-885 | 2°47 22°62 | 0°85 2-61 5 36 3567 32 °7 27 6 | 803-0 | 1:01 3°22 17 95 1-06 27 17°78 3°85 16 °36 | Average! ... ones: 21:37 | 1:27 — 11-09 — | 19°72 In Table IV the rats are arranged in six groups according to weight, and the aortic constant and aortic percentage are calculated from the average figures of these groups. As in the case of the grouped guinea-pigs, the best n is 0°72, giving an average k of 21-9. It will be observed that in these grouped animals the variations in the aortic constants of the groups are very small and are non-periodic, while the aortic percentage falls markedly and very regularly from 1:67 to 0°96. Using the values just stated for m and k, the average deviation between the calculated and the observed figures for the aortic cross-section is only 1:18 per cent., while it is 16-22 per cent. (nearly 14 times as large) if the area be calculated in per cent. of body weight. 45 1912.] The Size of the Aorta in Warm-Blooded Animals. Table [V.—Rats (grouped). “poaatosqo puv pozry OTRO WOTJAS-sso.to Tea 4aq a0UILO A(T ‘yystam Apoq jo (9z. 1) esvquoosed se popepNopVo WoTyoes-sso.t— ‘paadosqo pur poyry “NO[RO TLOTJOOS-SSO.T) W99M4Oq oUALO UT (6.12 = 4) “Ylsr-oM = V *po}e[NoTvo WOT OS-ssoay) “qu sTOM Apoq jo onuguoo1od SB WOIOOS-ssOld 9T4.L0V ‘Vier M = 4 €© NI OD 6 10 =H Tm mk O19 OonooOOo% an ONAN ANNAAA a a “UOTJOAS-SSO.A9 SERS aes OTIOR JO Vole asRIOAW HO Sees n G8eesn "R}.LOV JO SNIPRL adVIOAV HHIDGKRA ooooco REO) RE) SS) 4ys1em Apoq osvrory 5 SR2AeSee 2 aan UC | eee ies Ul TIT %7qRy, woaz t {celles lie : rH HO oD O26 S[BNPIATPUT FO s.toquan py aon ‘dnory 4R0AR& Table V.—Rabbits (individuals). ‘poArosqo puvx pozrT 3 Ao om tom nano st “NO[RD WOTFDAS-SS0.10 SS) SCP pire) Pos MES) nal on) t f : Ae) m00.M4oq aOUOLOIFLCT SSRABAN SR” NOD a ‘qystem Lpoq jo fen Pog F DWH I9HDMOI~ ORM (609. 0) o8uquoo.0d sv DNNGOAOHD HON I6 poyB[MopRo WOTQDaS-ssorgH Wa NG) sil C2) 1C9) (er) teal FOS ONIN Niet te oe) PoAtosqo pu poyrt BHO OM 19 HIG ro —Noywoa — WoOlyDas-sso.t0 Se oe sens GUIS IS eet =! : 5 o woaM4oq dDUOLO TCL pete od Feta) si 1S] Fee) Me ON COS 2) fo¥ ‘GR. ZS = 4 HONMID NADBWM00 Wee) SGP GOED Go S10) si) veel tl) 120 om > NO *poqu[noypeo WoIy9as-ssoa— ae aie SC : ‘ Oo bho aa 00 ror) WSTON OMAR RAATHOR S Apoq jo osvquoosod TOS 5) C2) ECS) Sse) 2 SW UOTPIS-ssold O1LOW moooseeseeeeo°e ) BIO 19 09 OO HID NOM H ror) ; DOD H19 ~ 16 WH B10 H Vierodd = ¥ INS S ft 10 ent : 3 HS DHAOhOMDOHOAN | ANAAANRHAAAAN nN TOI g IO 20 & 1910 d moroer Foe a PAR ASS FAG -SSOL0 91910B JO Voty MNHODODORONAD ci a AeA aR 5p © 3 GANA O Oh OaR aE + 3 24LOV JO sNIpPTyT ERGO igs a alley Ooi sy p COCR FA AA AANA = 2 SSOOrSSOWONHSS “‘qyustom Apog FSRESSASSZ LESS : bho? 19H OD HOO SON10 AAAI NANAN *xeg o+otot | Foo | too | to% ‘ON HANMHAMO-DROHA * * Calitics * The data for the individuals indicated are taken from Keilson (6). Table V gives the figures for the aortas of 12 rabbits, ranging in weight 250 erm. (7.e. Increasing more than eightfold). The average 2 from 310 grm. to aortic constant (k) is 22°49, with an xv (best 2) of 0°71, and the average 46 ) Messrs. Dreyer, Ray, and Walker. [Sept. 30, aortic percentage is 0°609, The variations of the aortic constant show no periodicity, but the aortic percentage decreases very greatly (although not regularly) as the animals increase in weight. When the aortic area (ze. area of the cross-section of the aorta) is calculated by our formula, the average deviation of the observed from the calculated figures is 16°49 per cent., while it is 26°34 per cent. if the area be calculated in per cent. of body weight. If the value 0:72 be taken for n, & becomes 24°15. Table VI.—Rabbits (grouped). a To: g 3 | oo | 3 gs .| 32 | See. 5 35 ei, B [erect mete Saal ee | fs oS "ep B) Fale ic iS) | eh te BSep 8 5 alesnons Ps heath Ss ct Sa | SS) 8 8 | See Ze E z og Eee |) Gare ad | Semi" 22 “AS Pe oe 3.2 ; 4 ® lain @® ga SM) qy 8 | sS ire} o~ =< m oO | oeN oO oye Sw on | Ss z aa |= |e8./S7' | ee | Son emem -|$a%) & | Sil Sen) @ | Sse ele) el. eee QB) ons S S aS e 2980) | a | 2 a |e Son onl =} a © 2 | x = HO 2A |) oon ae, oOo | o | SH bb | 2 ee lee so Il | Bae |.9 Hos SS) oS | oa < < cy (er Es |S A Ss) 6 aha ean ae Tas aan ag —— i in > grm. mm. | sq. mm, per cent. per cent. A | 1-2 340 | 0:961| 2°91 | 21°6| 0-856 | 2:84 | 2:46 | 2:14 | 85-98 1B | ae 649 | 1:29 | 5-27 | 18:7/ 0-812 | 4:45 | 18-43 | 4:08 | 29-17 CG | 5-6 13875 4a | Gea | 2650090473) | 7 -Gau | 14-9 8°64 | 24°65 D | 7-8 | 4763> | 1e72 | 19) Bees Orsie | Torta (o)-83 none eimmee E | 9-12 | 2207 | 1°84 | 10°69 | 22:1 | 0-484 | 10-71 | 0-19 | 138-86 | 22-87 | if | | z : : Sat. Average ......... | oh | OS | = 726 ==.) / 26ers | | In Table VI the rabbits are arranged in five groups according to weight, and the aortic constant and aortic percentage are calculated from the average figures of these groups. In this case the best n is 0°71, giving an average k of 22:1. The variations in the aortic constant are without periodicity, but the aortic percentage falls gradually, though not quite regularly, from 0°856 in the lightest, to 0°484 in the heaviest group. Using the above values for n and k, the average deviation of the observed figures from the calculated is 7°26 per cent., while it is 26°08 per cent. (between three and four times as great) if the area be calculated as a percentage of body weight. We now return to examine the question of sex as regards its bearing on the size of the aorta in these mammals. Since the number of our observations within any one of the species can hardly be regarded as sufficient to justify a general inference, we have taken all three species of animals together and made the figures comparable, inter se, by reducing them in terms of a common standard. When this is done, and the male and female animals are taken separately, it appears that the sectional area of the aorta in the male expressed as a function of the body surface is about 3 per cent. greater than it isin a 1912.] The Size of the Aorta in Warm-Blooded Animals. 47 female of the same weight. This is of special interest in view of the fact that a difference of the same size and character was observed by two of us in the blood volume of male and female rabbits. It may be noted, further, in this connection that, so far as we have yet ascertained from our data, the heart in the male animal is somewhat larger than in the female of the same body weight and species. Table VII.—Ducks (individuals). | eal ae = For Ree TG Z | as E af. | 3S S | | ee ht Bie a lhee eels | eet a |e a ce are ica ea ree er eae | oll) | 8.2 a We 2 aes (eer eee | fa kage] s £ a mM a) fa 1°84 10°66 2 62 1°82 0 82 1:09 | 68:04 | 2 |— (MO | Oc7stes |) ls} 10°87 | 2°57 1-82 16H 1:09 64°84 of — 70 | 0°745 1°74 11°21 2°49 1 -82 44, 1°09 | 59°34 4 /— 520 | 1°56 7 64 10 “42 1°47 7°38 oars: |) eyealil 5°8 3) || == 600 E66 aja Ou 10°23 | 1°43 8°25 4°24 9 36 8:12 6 g | 1180 1:99 | 12°42 11°39 1°05 13 14 5°48 18°41 | 32°45 7 6 | 1330 | 2:14 | 14°3 10°75 1°08 14°3 0-0 20°75 31°09 8 Q | 1420 2°16 | 14:72 10:93 1-04 14 ‘96 1°6 22°15 33 54 9 Q | 2070 | 2-47 Lone | OES 6°93 19 48 1°69 | 32:29 | 40-69 10 Q | 2850 | 2°86 | 25°8 10'16 | 0-91 24°37 587 =| 44-46 41°97 | | | | | | | Afverage vo... | 10-75 | 1°56 = 2-87 — | 38-59 | B. Birds.—Table VII gives the figures for the aortas of 10. ducks, ranging in weight from 70 erm. to 2850 grm. (7c. increasing more than fortyfold). The average aortic constant (£) is 10°75 with an n (best 7) of 0°70, and the average aortic percentage is 1°56. The variations of the aortic constant are small and show no periodicity, but the aortic percentage decreases very greatly and with absolute regularity as the animals increase in weight. When the aortic area (A) is calculated by our formula, the average deviation of the observed from the calculated figures is only 2°87 per cent., while it is 38°59 per cent. (more than 13 times as great) if the area be calculated in per cent. of body weight. If the value 0°71 be taken for n, & becomes 11°54 and the average deviation between calculated and observed values is 3:12 per cent.; with 2 equal to 0°72, k is 12:19. That the value 0°71 for m is very nearly as good as the “best” » (0°70) is shown by the fact that while the average deviations per eent. are 2°87 and 3:12 respectively with » = 0°70 and » = 0°71, the mean 48 Messrs. Dreyer, Ray, and Walker. [Sept. 30, deviations caleulated by the method of least squares are 3°67 and 3°63 per cent. respectively with the same values of x. That is to say that the value 0°71 gives a very slightly smaller mean deviation than the value 0°70. Table VIII.—Fowls (individuals). | i ae =| si, Ze =e | | 2 ee cie iae 23, | 25 ee B 2.8 = ee eo | SU e =O | : a= [ Sea |ee2k eS iam b eee 23 aan | °_ 2 | got | oe an) oe oo) leeeaeee Sea | Sg | See) Meee 2p a | secs OFS) || 38 o3qa |s3°8 esq 2 S | wg | 2 | See) Sil) e228 eee ie Z -s | & | osm | tee | £aR | Pee ieee og ee | ea SCS tien eperecamiene Ee 224/288 } 5 =] 5 2 6 HF S seo 2 eo Boe | ala A < ~s = ‘s) (A Ss) A grm. mm. | sq. mm. per cent. per cent. 1|—| 40°6| 0-462 | 0-669 | 21°51 | 1°648 | 0-679 | 1:47 0-463 | 54°13 2|—| 42-7| 0-462 | 0-669 | 22-31 | 1:567 | 0-704 | 4:97 0-487 | 43-53 | 3|—| 43-9| 0-484 | 0-736 | 20°54] 1°677 | 0:°718 | 2-51 05 | 47-2 | 4/9 | 919 | 145" |) Greles20%61)|/o-718 9) eran 2°96 | 10-48 | 37-02 5 | 9/1580 |1-72 | 9:3 | 21°61] 0-589 | 9:48 1:9 18°01 | 48°31 6 | 9 | 1598 |1:77 | 9:81 | 20°65 | 0-614 | 9°55 2°72 | 18-22 | 46-16 | Average wee... 21-20) 1-14. |) i=.) 2-76) | neon In Table VIII are given the figures for the aortas of six fowls, ranging in weight from 40°6 grm. to 1598 grm. (z.e. increasing nearly fortyfold). The average aortic constant (4) is 21:21 with an nm (best n) of 0°72, and the average aortic percentage is 1:14. The variations of the aortic constant are quite small and show no periodicity, but the aortic percentage decreases very greatly and almost regularly from 1-648 in the lightest animal to 0°614 in the heaviest. When the aortic area (A) is calculated by our formula the average deviation of the observed from the calculated figures is only 2°76 per cent., while it is 46°06 per cent. (nearly 17 times as great) if the area be caleulated in percentage of body weight. With x taken as 0°71, becomes 20°08 and the average percentage deviation between observed and calculated figures is 3°28. Table [X contains the figures for 10 ptarmigan purchased from a game- dealer. These were birds which had been shot, and they show the greatest range of weight we were able to obtain, namely, from 470 grm. to 710 grm. The average aortic constant () is 11-4 with an a (best 7) of 0°71, and the average aortic percentage is 1-4. The variations of the aortic constant show no periodicity, but the aortic percentage decreases from 1°66 in the lightest to 1:23 in the heaviest animal. When the aortic area (A) is calculated by our formula the average deviation of the observed from the calculated figures * 1912.] The Size of the Aorta in Warm-Blooded Animals. 49 is 712 per cent., while if the area be calculated as a percentage of body weight it is 7°22. If the value 0°72 be taken for n, & becomes 12:2. Table IX.—Ptarmigan (individuals). | , 2 > = =| | b | a> - gat Sal 2 Nee g 2 ord | pe Abie 24 Z 3°5 | See) se eae ic ee. Set be a @ Sp 2S aes oe) See ee eee. | ey! lhe a2 Mee 2 ee ea ees ele | iis Pa] = | 290 er n a AES eee) | Sut gay ble 2 = eta meee | eo) lee 5 A A | | | | | | grms. | mm. | sq. mm, | per cent. | per cent oa) 7420) 1-58. p 2-8) \OL =|). 166 | 6-92 12°71 | 6°58 | 18-54 Bo | 470 | 1-4 G19) lass ar t-32, || 6:92 10°53 | 658 | 5-93 8 | 2 | 580 | 1-48 | 6-84 | 126 | 1:29 | 7°54 Qe28. (sess T's 82 4 | 9 | 538 | 139 7-93 |10°9 | 1-47 | 7-62 4:06 | 7°53 | 5°31 eg | 663 | lb) 7-19 | Ieee | 1:3 Vag is NMG Wane TLL Gripes |) "B90 | 1-68 4) 8-35°| 11-1 || -42 | 8°14 25Guie 8°26.) EO, mon GeO.) 1-682) 8-9L |. 10rd |) L-49 | 8-28 8-26 | 8-4 6-07 | 8 | 2 | 630 eee S-SStavELrS wih b8G 6 )4 8-52 ORS Sale 2-72 Sheer) 1600) | 1-4 9°58) | 1074 41°47 ©) «8-72 9°87 | 91 5:27 10 | 2 | 710 | 167 SVingiel2 sl We l-23 99-24 5-73 | 9°94 | 12-87 | ] | PMTENE Oesosec ase a 1-4 — 712 _- 7°22 | Table X.—Ptarmigan (grouped). eae elves lets E ¢ Boi ce [ep N un ML a “ a a= ess|8°s|S~s 18° % le tee et Ne ert S so i eae soso | 2 2 Peau lesn Ss | pe fa be | SEO ie.) See | 8S | on Ea 00h lies as an iS Bled |] petal | Bae oi atonal Ronit ce oF 2 S sis Shenae Sa .| 88 See Wont: bbe 2 Bat ave So) Said eo aecvanlt -ainnl be 25 Eales 2s Sil ais ies 2 Hore 60,1 ger NT Fe 25 I a S 2) 226 |S es 614 | = < < me | 5 oS = | | | m mm. | sq. mm per cent A} 1-2 470 | 1:49 | 6°69 |12-0| 1-49 | 6°94 | 5-43 |B} 34 bet a) Leda eaa cal 12-6. I P38) 266 53 | 1°86 Menmoawe | 5St hil Gel 8-15 while.) acd 8-08 19 | 0-49 D | 10-12 | 663 | 1°69 | 8°96 | 12°0| 1°35 | 8°89 35 | 4:17 i | | Average ......... 12-1) 1-41 — | 1:28 — 2-99 In Table X the ptarmigan are arranged in four groups according to weight, and the aortic constant and aortic percentage are calculated from the average figures of these groups. In this case the best 7 is 0°72, giving an average value for # of 12:1. The variations in the aortic constant are small and non- VOL. LXXXVI.—B. E 50 Messrs. Dreyer, Ray, and Walker. [Sept. 30, periodic, but the aortic percentage falls gradually though not quite regularly from 1:49 to 1°35. Using the above values for n and #, the average deviation of the observed figures from the calculated is 1°28 per cent., while it is 2°99 per cent. (more than twice as great) if the area be calculated as a percentage of the body weight. Table XI.—Sparrows (individuals). | p ze | 3 gs, | zo° ee £ 5.8 2 Si = eas = ier esa | S°R | Slee eee See eee | os Ce, Be a Ss ee ¥ Bo Wie S a 2 aeN 2 = <& “Sp Sian Hens =) 2.) 2 y 282 | Se 8 5 are 2 |. S85) Sle | S22 | Se clas S Se ees 2 I BEE | 6 asS | S$a6 | HSS A ea Ve fe < 2 < S) a ‘S) A grm. | mm. | sq. mm. per cent. per cent 1 | 22-05 | 0°35 0 385 23-1 | 1°748 0 398 3°26 0 °391 1°53 2) 22°55 | 0°366| 0-42 21°8 | 1°782 0 °405 3-7 0°399 5 ‘26 B |) Aayal 0°398 | 0-487 19‘1 | 2°105 0-412 15°77 | 0-409 19 03 4} 24:6 | 0-414) 0537 18-1 2-098 0 °429 25°18 0 436 23 °2 5 26°55 | 0°35 0 *385 26 6 1°45 0 453 15°01 0-47 18 ‘09 6| 26°6 0°35 0 °385 26°7 | 1-448 0 °454 15:2 | O-471 18°3 | Average ......... PRAKSS WW ALSCTAL || | 13°02 = 14°24 Table XI gives the figures for the aortas of six sparrows ranging in weight from 22°05 grm. to 26°6 grm., a very small range indeed, but the best which we were able to obtain at the time. The average aortic constant (£) is 22°6 with an n (best 7) of 0°71, and the average aortic percentage is 1771. The variations of the aortic constant show no periodicity. When the aortic area (A) is calculated by our formula the average percentage deviation between the calculated and the observed figures is 13°02, and it is 1424 if the area be calculated as a percentage of body weight. If 7 be taken as 0°70 the value of & is 21°9, and with an n of 0°72 & is 23°4. C. Thoma’s Observations on Man—Table XII contains the figures for the aortas of 33 human individuals calculated from the grouped observations published by R. Thoma(7). The individuals in question ranged in age from two months to 29 years and the range in weight of the groups is from 8941 grm. to 49,000 grm. The best 7 is 0°70, which gives an average value for k of 5°03. The variations in the aortic constant are without periodicity, but the aortic percentage falls gradually though not quite regularly from 1:309 in the lightest group to 0°804 in the heaviest. Using these values for m and k, the average percentage deviation of the observed from the 1912.] The Size of the Aorta in Warm-Blooded Animals. 51 Table XII.—Man (grouped). Thoma’s Observations.* 2 a 2 2 ay aes al A = = ; + 3 Sirs g Sil in key or . ie ‘S 4 | 3. BS>|/ OCR | Sau | ook S| Bs ° HH. Sis ras? rQ 2 aow Q 2 | : e og a Ss & a | S G50 go 7 Pe Sy 2.2 ; nm © de. CNS. aso ANS a "S Ke] ~~ = a) oF bse} oBF estas} om s ef ee US Lee os See || cy by feb a) Set |} osaq 3 rQ A Coe ot Seqs | ot Sada Sop | one oes © © CR ie eee Gs | © S| ge 25S | gag &.| .o & = 2 a 2 S © BSN eS | eis) es) as) eelesca | oS a Be Bae | 2 CURSE sages as Sr es 2 2 go ll BRE | 9 SsS|/oao/ os oA ay 5 2 5 & 2 2..¢ | 29) 3 los eee S m | = i edae |88F| eos | Eee is. on 4 I] 5 td oo o~ of | oes 3 = 2 = nN = = |ll mano = Saier 2 on mn z mB Se ee | eo) en yee rs < 5 ENGs rap |e Nee es oe) 2 z = = Boos | 382 | sees | eee Bl eS @ » = = x SLE, x eS) BES & | 8 &b 2 2 i j 5 || itl 29 ot 255 ie aa eS z | at Enea. | eee B52 e| sa | 8 g o | Bef |BeS | 55 @ | eee oe “ = i < < e eee | 2 ‘ ; | | erm. | mm | per cent., per cent. Aaa 207, 8,941 | 6-1 | 89°4 | 54:6 616 | 0-97 6:05 ‘88 B/| 9 | 11,950] 6:7 | 89:7 | 56:0 6-78 1°18 6°71 0-15 g 5 | 18,680] 7:25] 85:8 | 51:8 7:09 | 2:26 7:08 3:13 Di} 8) 1,510), Coase orta Sere 7°70 + | 0-65 7°68 0°39 E| 4 | 43,250] 10:2 | 40:8 | 61:9 10°41 2-02 10:58 3:59 F | 5 | 49,000] 11:2 | 84:9 | 53-9 10:86 3:13 11:06 1:27 Average ......-+- | 383 | 55:8 | = 1°70 a= 1-56 * Thoma’s data are printed in light type. The figures calculated by us are printed in heavy type. In Table XIII we have caleulated K for Thoma’s figures from the formula W/r? =K, first giving the value 3 as in Thoma’s formula, and then giving it the value 2°82, which corresponds to n = 0°71 in our formula, and is the best m for Thoma’s observations. It then appears that if the aortic radius be calculated by means of these values of K the average percentage deviation between 7 calculated and 7 observed is 1°70 when has the value 3, while it is only 1:56 when m has the value 2°82. Moreover, if the appropriate allowance be made for the number of individuals in each group the figures become 1°65 with nm taken as 3, and 1°35 with n taken as 2°82. It will, of course, readily be apprehended that differences of this amount in the radius assume a considerable importance when calculations are made by area (7.¢. 777°), as is necessary in referring the aortic area to the body surface. In Table XIV are tabulated our main results arranged in such a manner as to show at a glance the range of weight, the best , the value of &, the percentage deviation, and so forth for each species of animal. It will be seen from the averages brought out at the foot of the table that taking all 53 ls. Warm-Blooded Anima a ize of the Aorta 1912.] The S e a 3 ee cd j co a Se Be ote see ee U 4s80q SUIYLY WOTYVIAop osvyuoodod oSvIOA Wy a Sa! = sa ee ee | i =e : 19. ¢ Le. LT = 91-€ 80.8 SI-9 | @¢.gG £0.¢ | 04:0 SF. 0006F-LF68 | Sa pcoRGbOSeEOP podnoas : uvy 60. T TZ. PL ia 60: ST = F-& | 9.88 GulZ Nl) Deno T@- 1 9. 96-0. GB" YeNpTATpur : momtedg Fe: Z 66:3 | 8@-T = = [lsat — — | &4.0 IFT €99-04h °°" pednoas : 10: T GG L = SL & = G- Gl P- IT — | TZ.0 TS-T OLZ-OLF | “ 7enplATpUr : WesTULLTy J 1.91 90:9F | 94.8 8z- & = Gus | ewe — | 8&0 %- 68 || S6ST-9-0P | 1BOPLATpUr + [MOT G.I | 69: 8€ a 6I-& 48.6 6 GL ¢. 11 8-01 || 04-0 L. OF 0e8z-02 OR OH CRE [enpraTpur youd 69-€ _ 80-93 = 98. & = L. 83 I. 83 = il 22 G.9 Loze-oFe |" pednoasd 9.1 FE. 9% — | 6P.9T = aad Gg. eS — || t4-.0 Ze. 8 ossz-O1s | Tprarpur + (ourey) 91qq 2 8-81 || 22-9T | S8T-T — = 6-18 | 6-02 i ea) IL-8 962-¢.98 | pednoxs 8Z-T GL. 61 Lé- IT 60. TT = V- GZ P-16 == | Ta&-O T- OF G08-1- 0g | TeMprAlpar + ye 8-7 || Oc tL | 26.8 €g. ¢ — 9. G2 oS — || 2.0 C9. F 9ZL-9¢T |" podnoad 5 @1T =|: 89-6T | -€¢-91 | bb. OT = g. 92 6. bs == lll 3. @ Te. 4 OS6-O8T |" eupratpur + dyd-vomn 4H H miiees) i | 2 | “qYSTO.M || & vw qsoq £q || ood, [24-0 = #| 12.0 = # | 04.0 = «||-24.0 = «| 'TL.0 = « |0L.0 = 4 -gsoq 31] uoTyeTAep || © | JO sultoy A £q popratp | ape, | -w qsogr ur ioe “yRULLy qysiom (aa ea =a S ( = ae Seas Sai 3 eee Re — 8 Ra LY. PPS = | &2-0 ae | 86-4 bv. 4 86.9 Ta. 0 | PP. T OL9-G9P mers padno.tg ea PLT GF = | 6b-S = || €6.2 Gv. 4 0.2 T4-0 | po. T OTL-O09P ie ere ast | | | + UVSTULUY 7 = 2g. 8 16. OT = = 83-T = v. &T G. ST 04, 0 0. € 0982-996 J podnoap Se 41-6 68. €T = =e £P.9 a Vv. EL G. GL 04-0 | OT-€ OL6Z-OPG ||" © FE GH NERC = + (ourey) 4rqquyy = sI-b §©=||_ «G- 2S | «OL. OF. = G. BG ike Tinh Hh 72 AC0) 0-6 089-0, |" meres pad nowy aS 8é- 1 || 1T-S@ &- 8T = => v-&S G. 1% “= 64-0 | PL-6 OV9-OL | [enprATpay > Std-voumy © S | | | “SULLG > be Seine = i S i eae ae e tA wea et ? Sales = 2 2 =! » 2 ‘QUSIOM | Ny “wu qsoq Lpoq yo |, : ; , ; Peete ny Aq woryvtaop || osujzued | 4.0 =% | Th0 = % | 040 = % | GL.0 = 4 | TL-0 = 4 | 02-0 = % FO 9ysTOM P re ea | ‘ Jo sutroy ‘qu sIOM. : Ss See erent aoe. | SESE MP Ser jeune || jo ofuny eT ws augions Aq | | qsoravot, UOT}RIAD, z i) 0 e eee “MOTZVIAOp osuzueo1ed osurOA : | J° 9757 M | D . - V 1} Y | XR | ‘=> jee Bie ts See ee 2 - 5 ue 3. eee ee el me ‘Ojo ‘SJULISUOD ‘SUOIyVIAG 98eyUEDIO0g— arqe 2 A MBNS KG | d— IIA "Ll = : Ce N — a — a Pie, a ee ae Si 64 Messrs. Dreyer, Ray, and Walker. [Sept. 30, The ratio between these figures (3°27 and 17:34) is also almost exactly the same as that which was found between the corresponding figures in the case of the blood volume (2), although in that case the mean deviations themselves were much smaller (namely, 1°39 and 7:82 per cent. respectively), owing to the much greater intrinsic accuracy of the technique. From these figures it is seen that if a series of observations of the tracheal area are made and averaged, it follows that if the difference between this average and the theoretical value given by our formula is as much as 7 per cent. the tracheal area is probably abnormal, and if it amounted to about 10 per cent. it would be almost certain that the trachea was abnormally large or small. But if the measurements were expressed in percentage of body weight it would only be possible to say with the same degree of certainty that the tracheal area of the animal was abnormal when it differed from the calculated value by 50 per cent. or more. The inter-relation of the various constants (for surface, blood, aorta, and trachea) given in the present and in previous papers, together with its significance, will be dealt with in a later communication. But it may here be pointed out that the value of ~ in the expression W”/a = k has now been shown to be 0°70—0°72 for the surface area of three different species of mammals, for the blood volume of siz mammals, for the aortic area of four species of mammals and jour species of birds, and for the tracheal area of two mammals and one bird. Accordingly we regard our formula W"/a = k as a rational formula indicating that the blood volume, the aortic area, and the tracheal area are all proportional to the body surface in warm-blooded animals. From an examination of the large number of data which we have now collected it appears that if one desires rapidly to compare a series of individual observations by means of the formula W”/a =k, the value of k may readily be determined approximately by using the power (= 0°67) or the power ?(= 0°75) instead of the accurate value of n (0°70-0°72). The results thus obtained will be approximately correct over a moderate range of weight. But as the range of weight increases the results deviate from the true values, and those obtained with n = 2 deviate more rapidly than those obtained with n = 3. The difference in the relative accuracy of the results given by these two values of 2 has been ascertained in the following manner :—The percentage deviations for the surface, the blood volume, the aortic area, and the tracheal area for four different species of animal, with “ best n,” with n = 2, and with n = 3, were tabulated, and the figures for these three values of m averaged. The final average deviations thus obtained were 2°52 per cent. with n taken as 0°70-0°72, 3°25 with n taken as ?, and 3°81 with » taken as 3. Thus 1912.) The Size of the Trachea in Warm-Blooded Ammals. 65 it appears that the value ? for m gives a deviation 1:29 times as great as that obtained with the best n, while if the value of m is taken as ? the deviation is 1°51 times as great as that with the best x. Accordingly, it is clear that if an approximate value of 7 is employed for the sake of convenience in rapid calculation it is, as a rule, preferable to use the value } rather than 2. But wherever observations covering a wide range of weight are concerned it is essential to make use of the “best 1” in order to obtain reliable results. In this connection it may be noted incidentally as a point of interest that the ratio between the deviations just quoted for 7 = 2 and n = best n, namely, 3°81/2°52 (ic. 1:51), proves to be practically identical with the corresponding ratio already given elsewhere (2) in the case of the blood volume, namely, 2:08/1°39 (ze. 1°5). Conclusion. Within a wide range of weight in any given species of warm-blooded animal the sectional area of the lumen of the trachea is proportional to the body surface, and can be calculated from the body weight by means of the formula T = W”/k, where 7 has the value 0°70-0°72 and & is the constant to be ascertained for each particular species. REFERENCES. 1, Dreyer, G., and Ray, W., 1910, “The Blood Volume of Mammals, as determined by Experiments upon Rabbits, Guinea-pigs, and Mice, and its Relationship to the Body Weight and to the Surface Area, expressed in a Formula,” ‘ Phil. Trans.,’ B, vol. 201, p. 133. 2. Dreyer, G., and Ray, W., 1911, “ Further Experiments upon the Blood Volume of Mammals and its Relation to the Surface Area of the Body,” ‘ Phil. Trans.,’ B, vol. 202, p. 191. 3. Dreyer, G., Ray, W., and Walker, E. W. Ainley, 1912, “The Size of the Aorta in certain Animals and its Relation to their Body Weight,” ‘Journ. Physiol. vol. 44, p. xvii. 4. Dreyer, G., Ray, W., and Walker, E. W. Ainley, 1912, “The Size of the Aorta in Warm-blooded Animals and its Relationship to the Body Weight and to the Surface Area, expressed in a Formula,” ‘ Roy. Soc. Proc.,’ this vol., p. 39. 5. Dreyer, G., Ray, W., and Walker, E. W. Ainley, 1912, ‘‘The Relation between the Sectional Area of the Trachea and the Body Weight in certain Animals,” ‘ Journ. Physiol., vol. 45, p. vi. 6. Meeh, K., 1879, ‘“‘Oberflachenmessungen des menschlichen Kérpers,” ‘ Zeitschr. f. Biologie,’ vol. 15, p. 425. VOL. LXXXVI.—B. E 66 Notes on the Infe-History of Trypanosoma gambiense, éte. By MuRIgEL ROBERTSON. (Communicated by the Tropical Diseases Committee of the Royal Society. Received September 28,—Read December 5, 1912.) (Abstract. ) The following is a brief account of some of the more salient features in the life-cycle of Trypanosoma gambiense. The results are drawn from a large number of experiments carried out at the Mpumu Laboratory in 1911 and 1912. The present paper is in the nature of a very brief synopsis, and is not a full account of the experiments and conclusions. I. Endogenous Cycle in the Blood. The part of the life-history of 7. gambiense spent in the vertebrate—the experiments were carried out with monkeys—is characterised, as is well known, by a marked fluctuation in the numbers of parasites present in the blood. The individuals show a wide range of variation in length and breadth. During the depressed periods, the few parasites present are of the short, relatively broad type. Periods of increase are characterised by the appear- ance in addition of the intermediate and long slender forms. The latter are the individuals about to divide. The short form may be looked upon as the adult blood-type, and is usually the most numerous form present, except during the periods of rapid multiplication. These periods set in regularly in every revolution of the cycle during the earlier months of the disease ; their recurrence is less marked in the later stages. The short forms appear to be responsible for carrying on the infection in the Glossina, and the blood of a monkey is only infective to fly when these forms are present in sufficient numbers and in a suitable physiological condition, 7.c. not suffering from exhaustion. Intracellular multiplicative phases do not oceur in the lung, liver, or spleen of monkeys. Rounded, non-flagellate individuals are occasionally found in the liver and lung, apparently between the cells, but it is not perfectly clear whether they may not be in rare cases within the cells. They appear at the time when the trypanosomes are being destroyed, before the depressed periods of the endogenous cycle, but have only been found in a teeming infection examined during the earlier months of the disease. These creatures are apparently about to be destroyed, but their survival in very small numbers as latent forms cannot be entirely excluded. Notes on the Infe-History of Trypanosoma gambiense, etc. 67 No sexual differentiation of any kind is to be observed among the blood- types. The differentiation into long and short forms is a phenomenon of growth and division, and is not an expression of sex. A wet fixation of the blood-films in corrosive alcohol, and subsequent staining and mounting without drying in air, gives the nuclear picture shown in figs. 25 and 26. This is in accord with the observations upon live specimens, and is identical with that found in other trypanosomes studied by similar methods. A division-stage is shown in fig. 27. Il. Hxogenous Cycle in the Fly. A great deal of information can be obtained from a close study of the conditions in the early days of the cycle as to what are the factors inhibiting the development of the trypanosomes in the Glossina. I do not, however, propose to discuss these here; suffice it for my present purpose to take those cases in which trypanosomes have established themselves, and to trace the usual development. While the series of changes undergone in the Glossina up to the time when it becomes infective to clean vertebrates is very definite and constant, nevertheless, the duration in time of this cycle varies in different cases within the limits of more than a fortnight. This must naturally be borne in mind when considering the successive stages in the cycle of any given fly. The trypanosomes never attach themselves while in the gut, nor do they ever disappear from this situation at any period; the development occurs free in the lumen of the alimentary canal from the very start. At no period do the parasites enter the body-cells of the host, nor do they penetrate through the gut-wall into the body cavity. The earliest processes that take place in the fly are characterised by a slight and rather indefinite change of form (figs. 1 and 2). Broad, slender and degenerating specimens are all present, but only the broader types are ever found in division at this early stage. These first divisions (figs. 3 and 5) are remarkable in that they show a suppressed crithidial phase in the young individual. This disappears before the separation of the two products (fig. 4). The peculiarity just noticed does not occur in the later divisions, and has never been observed after the 10th day. The gut stages do not show any other crithidial phase. The trypanosomes usually start developing in the middle or posterior intestine (mid-gut), and by the 7th to 10th day there are a large number of trypanosomes present, showing the general features depicted in figs. 6-12. Division goes on rapidly, and the nature of this process is shown in figs. 8-10. It will be observed than the granule at the base of the flagellum 68 Miss M. Robertson. Notes on the [Sept. 28, (blepharoplast of Minchin) plays the réle of centrosome in the division of the kinetonucleus. It must also be noted that the division is not really longi- tudinal but practically transverse, the plane of division being at right angles to the long axis of the parent individual (fig. 10). Division is often unequal. | | ; I 8 i 12 Magnification 2500. Multiplication proceeds until the whole of the middle and hinder and part of the anterior intestine is filled with parasites. Very slender long forms are developed about this period (8th to 18th day or thereabouts) and these gradually pass forward into the proventriculus. This slender type (fig. 13) is essentially the proventricular form and is the culmination of the development 191235] Life-History of Trypanosoma gambiense, etc. 69 in the gut. The trypanosomes may overflow into the sucking-stomach or erop, but are not permanently established there. The flagellates are, moreover, unable to retain their position in the proventriculus if the fly is subjected to a fast of a considerable duration, such as any period exceeding two or three days. x Magnification 2500. Multiple Forms —Up to about the 10th or 15th day of the cycle multiple forms such as those shown in fig. 14 may be seen in certain cases. The evidence is largely in favour of these being degenerative stages but is not sufficient to entirely exclude the possibility that some of them (compare fig. 15) may not be involution-forms or resting phases capable of further development 70 Notes on the Life-History of Trypanosoma gambiense, ete. and activity. Proventricular forms when injected into clean monkeys do not produce infection. Inwasion of the Salivary Glands.—The long slender forms from the proven- triculus come forward into the hypopharynx in small numbers at a time and may be found lying free in this situation im carefully dissected specimens before the glands are infected. From the hypopharynx they pass back along the narrow ducts of the salivary glands and it is not atall a rare occurrence to find trypanosomes in the ducts of the glands in 16- to 30-day flies when the rest of the glands show no flagellates at all. The trypanosomes reach the glands as long slender forms and attach themselves where the duct joins on to the slightly broader part which leads to the glandular portion proper. They become much shortened and very much broader and assume the erithidial condition shown in figs. 16-21. They break free occasionally but seem to attach themselves again. Multiplication (fig. 22) occurs and the trypanosomes gradually invade the whole gland; new specimens keep on arriving from the hypopharynx. The short dumpy crithidial forms develop into trypanosomes almost identical with the blood-forms but often a little below the normal adult length (figs. 23 and 24). These trypanosomes are found swimming free in the lumen of the glands and there is the strongest presumptive evidence for considering that these are the types that produce the infection in the vertebrate. Not only is this second development in the glands necessary to produce an infective fly, but from a number of considerations, amongst others the appearance here of the very clear and definite crithidial stages, it may be held that the development in the glands is the really essential part of the whole cycle. The development in the gut may be considered as a somewhat indifferent multiplication—a mechanical device to enable the trypanosomes to establish themselves in sufficient numbers in contact with the salivary fluid, which alone, in the Glossina, seems able to stimulate the trypanosomes to the apparently essential reversion to the crithidial type. Conjugation—Sexual differentiation has not been observed at any part of the cycle; this is not, however, a characteristic feature of flagellate life- histories. Isogamy seems to be usual among the group. ‘The direct evidence of conjugation is slight and not sufficiently convincing. General theoretical considerations are, however, very strongly in favour of some such process occurring, and from comparative evidence drawn from the consideration of the cycles of 7. nanwm and T. vivax it seems possible that the sexual part of the cycle might take place in the salivary glands. It is obvious that much of the foregoing work has been simply to carry somewhat further the researches of Minchin, Roubaud, Bruce, and Kleine, Notes on the Infe-History of Trypanosoma gambiense, etc. 71 more especially those of the two last-named workers. There are no serious discrepancies between the cycle in the fly sketched by Bruce, Hamerton, and Bateman and that described above, except that I consider the fly history to be in reality a double development. In many points my work is also in agreement with that of Kleine and Taute,* except that I do not consider that the “male” forms described by them play any important part in the eycle. A further discrepancy consists in the view held by the latter authors at the time of writing their paper in regard to the salivary gland phases being a non-essential part of the cycle. My interpretation of the endogenous cycle in the blood of the vertebrate is at present, so far as I am aware, unconfirmed by other workers, largely, I imagine, owing to the fact that the interest has been concentrated for some time past on the appearances in the fly rather than on those in the vertebrate. * “Arbeiten aus dem Kaiserlichen Gesundheitsamte,’ vol. 31, part 2. rie ifs Wate a a awta teh is ‘i BAM Vc aL NW ape Se Rice ‘shel, VER? Geeta. i Comparative Anatomy and Affinities of the Araucarinee. 71 more especially those of the two last-named workers. There are no serious discrepancies between the cycle in the fly sketched by Bruce, Hamerton, and Bateman and that described above, except that I consider the fly history to be in reality a double development. In many points my work is also in agreement with that of Kleine and Taute,* except that I do not consider that the “male” forms described by them play any important part in the eycle. A further discrepancy consists in the view held by the latter authors at the time of writing their paper in regard to the salivary gland phases being a non-essential part of the cycle. My interpretation of the endogenous cycle in the blood of the vertebrate is at present, so far as I am aware, unconfirmed by other workers, largely, I imagine, owing to the fact that the interest has been concentrated for some time past on the appearances in the fly rather than on those in the vertebrate. On the Comparative Anatomy and Affinities of the Araucarinee. By Prof. Ropert Boyp TuHomson, University of Toronto. (Communicated by Dr. D. H. Scott, F.R.S. Received September 14,— Read November 14, 1912.) (Abstract. ) From a study of the anatomy of the different regions of the plant, evidence is found of the relationship of the Araucarinez to the Cordaitales. In the first place, the presence of a leaf gap opposite the outgoing foliar trace, in all forms whether the leaf be large or small, is taken as indicating the Pteropsid ancestry of these forms and is considered of sufficient importance to preclude the possibility of the Lycopsid connection of the Araucarinee, of which view Seward has been the recent exponent. The presence of a gap in the cone and in the seedling seems to put the question beyond doubt, since this indicates the ancestral presence of a leaf gap. One evidence of relationship to the Cordaitales is found in the retention of Cordaitean pitting of the tracheids in the different regions of the plant which are recognised as primitive, in the cone especially, where the pitting may be as much as 5-seriate, the pits, alternate, hexagonal and extending from end to end of the tracheid as in the Pteridosperms and the primitive members of the * ‘Arbeiten aus dem Kaiserlichen Gesundheitsamte,’ vol. 31, part 2. VOL. LXXXVI.—B. G 72 Comparative Anatomy and Affinities of the Araucarinee. Cordaitales. In medullary ray structure, too, all the forms are Cordaitean— the ray cells thin walled and with numerous pits on the tracheids where the cells come into contact with them. In both groups the rays are resinous but devoid of resin canals. In both there are ligneous parenchyma cells and resinous tracheids in the secondary wood. The latter are considered the ancestral form of the resin tissue from which the other types in the secondary wood of the conifer series have been derived. With regard to leaf traces, too, both groups agree—the trace may be single or double even while it is still in the secondary wood. In the leaf there is centripetal primary wood directly opposite the protoxylem. In no case was there found in the primitive regions of the Araucarian forms any indication of Abietinean structure, which would be expected if the view of the Abietinean ancestry of Araucarinee which is advocated by Jeffrey be correct. In contrast to this, in the primitive regions of the Abietinez there are evidences of Araucarian pitting, etc. In addition, evidence is advanced to show that the transitional forms upon which the claim for the greater age of the Abietinee is based, indicate rather the derivation of the Abietinesze from the Araucarinee or Cordaitales. Of special interest in this connection is the evidence that the traumatic resin canals of Araucariopitys are of a primitive type and in the process of acquirement. The determining points are the resemblance of these to both the normal ones of the cone and to the traumatic series of the vegetative parts of the living pine and their difference from those of such a form as Abies, where it has been shown that resin canals are revived by injury. The Abietinean theory of the ancestry of the Araucarinez recognises only traumatic series of the “revival” type, and yet there is no record of authentic Abietinean forms, as has been recently shown, either in or previous to the Triassic, in which the first Araucarian supposed to be derived from the Abietinez (Woodworthia) makes its appearance. Thus geologically as well as structurally the superior antiquity of the Araucarinez rests upon avery firm basis. This basis is made the more secure by the practically unbroken sequence of forms with essentially Araucarian structure right up to the Triassic. In every respect confirmation of the old view has been found, which regards the Araucarinee as anatomically very closely associated with the Cordaitales. The Relation of the Islets of Langerhans to the Pancreatic Acin under Various Conditions of Secretory Activity. By Joun Homans, M.D. Boston. (Communicated by Prof. E. H. Starling, F.R.S. Received October 1, 1912,— Read January 23, 1913.) (From the Institute of Physiology, University College, London.) [PLatEs 2 AND 3.] The object of this research* was to examine the relations of the islets of Langerhans to the pancreatic acini under the following conditions: first, at a time when the pancreatic tissue had been cut down to a very small amount, and second, when the gland had been exhausted by the action of secretin. Under the first of these conditions it was expected that, as the removal of the entire pancreas was inconsistent with life, the removal of all but a little might lead to the hypertrophy of one or both elements in the part remaining, and throw some light on the possibility of a change of acinous to islet tissue or the reverse. As a control to this procedure the pancreatic ducts were ligatured in one animal because the pancreatic acini rather than the islets are generally supposed to degenerate under these circumstances. In the second series of experiments—exhaustion of the gland by secretin—it was intended to test the assumption by Dale, Vincent and Thompson, and others, that this proceeding induces a change of acinous into islet tissue. It is hardly necessary for me to summarise the literature of this subject beyond repeating the most typical of the opposing views as to the functional and anatomical independence of the islets and their relation to acinous tissue. Ranged on the side that the islets are more or less interchangeable with acini are a number of investigators, whose views, however, show important differences of detail, LLewaschew (1886) holds that as a result of exhaustion acinous cells become converted into islet tissue and are capable of re-forming after rest. Laguesse (1893-1911) maintains a somewhat similar view of “balancement” between the tissues, both on embryologic grounds and asa result of a long series of anatomical and experimental observations. He considers, however, that the islets have an independent function. Mankowski (1902) considers that the islet is the highest stage of acinous cell activity into which all such cells must go. Dale (1904) sees, as a result of exhaustion, * The research was carried out with the aid of funds from the Peter Bent Brigham Hospital of Boston, U.S.A. G2 74 Dr. J. Homans. The Relation of the [ Oct. 1, starvation, and duct ligation, the formation of new islets from acini. Karakaschiff (1904-1906), in a study of diabetes in man, finds that islets are without function except to form acini. Herxheimer (1906), in a similar study, concludes that islets are made from acini as a regenerative process. Vincent and Thompson (1907) agree with Dale. Fischer (1912) believes that the islet cells are able rapidly to change in function and appearance. From the point of view of independence of function, the opinions of those who, following the lead of Mering and Minkowski, have endeavoured to discover which part of the pancreas is concerned with the sugar function are more in agreement. Thus Schultze (1900), Ssobolew (1902), Sauerbeck (1904), Tschassownikow (1905), Tiberti (1909), and Laguesse (1911) substantially agree that duct ligation does not produce diabetes. Of these, all but Sauerbeck find that islet tissue is not destroyed in the process, and this conclusion, taken into consideration with the occasional finding by Opie (1901) and others of extensive islet disease in human diabetes, furnishes a strong argument in favour of the functional independence of this tissue. Partial extirpations and pancreatic grafts by Laguesse (1902), Kyrle (1908), Tiberti and Franchetti (1909) have led to similar conclusions. It must be admitted, however, that except for Ssobolew (1902) no one has been able to see as a result of stimulation with carbo- hydrates or phloridzin any changes in the islet cell indicating stages of functional activity [Tiberti (1909), Frugini and Stradiotti (1909) ]. The supporters of the view that islets are unchangeable anatomic structures bring evidence from a number of sources. Embryologic investigations by Pearce (1903), Kiister (1904), Helly (1905), and Weichselbaum and Kyrle (1909) point to the origin of islets in early fcetal life from primitive duct tissue and to their subsequent anatomic independence. None of these investigators have been able to confirm the observations of Laguesse (1895-6) in regard to the disappearance of his “ primary ” islets and the substitution of “secondary ” islets from acinous tissue. Other investigators, by methods similar to those employed by Lewaschew, Minkowski and Dale, have come to conclusions diametrically opposite to theirs, a fact which suggests that the matter is, to a great extent, one of interpreta- tion of the histologic appearances and of microscopic technique. Thus Diamare (1899) finds that the islets contain two kinds of specific granules, and have an important sugar function. Opie*(1900) finds no increase in the number of islets after prolonged stimulation with pilocarpin. Ssobolew (1902) holds the islets to be independent structurally as well as functionally. He sees specific granules in the islet cell. Dewitt (1906) finds a great variation in the normal number of islets, but no increase or change resulting from starvation, feeding, and so on. Lane (1907) shows that there are two 1912. | Islets of Langerhans to the Pancreatic Acint. 75 kinds of granules in the islet cell, differing chemically from each other and from the zymogen of the acini. Finally, Bensley (1911), in a most exhaustive study of the pancreas of the guinea-pig, has cleared up much of the doubt surrounding this subject. By the use of vital stains he shows that the islets can be readily identified and counted, and that their number is not influenced by starvation, or any known method of stimulation. He demonstrates, by the use of refined methods of fixation and by granule stains, that the islet cells have a structure distinct from that of the acinous cells, as well as from the duct and centroacinous eells with which they are likely to be confused, and among which they are frequently found. He points out with justice the error in the common practice of identifying the islets by purely negative means, that is, by fixations and stains which do not bring out their distinguishing characters. I have repeated some of his experiments with vital stains to satisfy myself of the fairness of his basis of identification, and this research would have little justification if it did not take a direction in which Bensley’s work has not carried him. I have borrowed freely from his methods. Technique. The animals used in this research were dogs and guinea-pigs. No especial selection of animals was made. The experiments* include two exhaustions of the dog’s pancreas by secretin, five nearly complete extirpations and one duct ligation in the dog, and three pancreas exhaustions in the guinea-pig by purified secretin. In addition, a number of injections of neutral red, methylene blue, and pyronin have been made in guinea-pigs, according to. the methods of Bensley. Fixing and Staining Methods.—All specimens were removed immediately after the animals had been killed by bleeding under anesthetics. Pieces of pancreas not more than 2 mm. in thickness were placed at once in a number of fixative solutions and stained by various methods. I. Acetic-osmic-bichromate (Bensley). Osmictactd? s225 smears boas 5 dackioates sew tats meets tees 4 ce IPotassiimabichomategete aeeeee cosas eaceeee ee eee ee 15 c.¢ INCObIG ACI teases. cemee bee ae sess -setdes spe nessosesateeeee h e ete Aes 2 drops Tissues fixed from 16 to 24 hours were washed in water and passed through alcohol to paraffin. Ordinary tissues were cleared in benzol before paraffin, but pieces containing considerable scar tissue were passed instead through carbon bi-sulphide for greater ease * All experiments in which the animals have been allowed to recover from the anesthetic—five extirpations of the pancreas and one duct ligation in the dog—were performed by Prof. E. H. Starling. The exhaustion experiments were performed by the writer. 76 Dr. J. Homans. The Relation of the [Oct. 1, of sectioning. Sections cut less than 4 micra in thickness were stained by the acid- fuchsin-methyl-green method advocated and fully described by Bensley. II. Aqueous-chrome-sublimate (Lane’s method for B cells). TELIRISSHIUUATL |OOUROLTEEE osarenocancanasagouosncensnosehoosee 25 orm. Miercunic chllonidemereeneeeeceee ee eneeec eee eeere eect: By oe Distilledswater: ee eedco. sastemaeaaeeeen epee cineeetes 100 ce. Very small pieces of tissue were fixed for 12-24 hours, washed in water and passed through alcohol (including an iodine solution) to paraffin. These sections were stained for 24 hours in a 20-per-cent. solution of neutral gentian. (For the details of preparing this solution see Lane’s or Bensley’s article.) III. Alcohol-chrome-sublimate (Lane’s method for A cells). Saturated alcoholic solution of mercuric chloride. Potassium bichromate, 2°5 per cent. (Equal parts.) Tissues fixed in this solution for three to four hours with one change were washed in 50-per-cent. alcohol and passed rapidly through alcohols to paraffin. Neutral gentian stain. (As this method produces considerable shrinkage, and as in my hands it has proved quite uncertain, especially in the dog, I have made little use of it.) IV. Zenker’s fluid. Tissues treated were stained with methylene blue and eosin or a similar combination. This technique does away with the granules in the acini and islets as well, and may be used to show the “ negative ” of the granule stains. Demonstration of Islets by Vital Staining Methods.—I have repeated a number of Bensley’s experiments to satisfy myself that they are valuable in less skilled hands than his. Only the neutra] red, pyronin and methylene blue methods have been used. A full description will be found in Bensley’s article. Operative Procedure. 1. Partial Removal of the Pancreas.—Under morphine and ether the pancreas, with the exception of about one-tenth at the duodenal end, was removed by blunt dissection (Hédon’s method) without ligation of blood vessels. There was no bleeding. The cut end of the remaining part, generally about 2 cm. in length, was tied off with silk and was either transplanted with its blood supply outside the oblique muscles through a slit in the muscle or left 77 situ. 2. Ingation of the Ducts.——Under morphine and ether the large duct was isolated for 3 an. and divided between two silk ligatures. The pancreas in this region was separated from the duodenum, and omentum was inserted between the gland and the bowel. To divide the smaller duct the tissues about the common bile duct were ligated on both sides and divided. The gland was further separated from the bowel and omentum inserted between. 3. Exhaustion of the Dog's Pancreas by Secretin.—The animals were fed as usual the afternoon before the experiment. They were anesthetised by morphine and ether. A secretin solution prepared from small intestines by the method of Bayliss and Starling was allowed to run into the external jugular vein by means of a cannula. A cannula was tied into the pancreatic duct and the juice measured. A piece of pancreas was isolated at the beginning of the experiment as a control of the exhaustion. The stimulation was kept up until the pancreas failed to respond to the stimulation or until the dog died. 1912. ] Islets of Langerhans to the Pancreatic Acini, 77 4. Exhaustion of the Guinea-Pig’s Pancreas by Purified Secretin.—In order to do away with the fatal depressing effect of secretin on the guinea-pig, purified secretin prepared by the method of Dale and Laidlaw* or Stepp was used. The animal was anzesthetised by morphine and ether. The method of introducing the secretin was the same as in the dog, but no cannula was tied into the pancreatic duct. The degree of exhaustion was estimated by the examination of a piece of pancreas under the microscope. The Normal Resting Pancreas of the Dog. A detailed description of the pancreas is superfluous, but in connection with the accompanying illustration (fig. 1, Plate 2) a brief description of the different kind of cells is necessary as a base with which the experimental conditions produced may be compared. The anilin-acid-fuchsin-methyl-green stain after osmic-chrome-acetic fixation has proved, in my hands, so much more successful than any other for the dog’s pancreas that it is upon this that most of my conclusions are based. By this technique the normal charged acinous cell shows the characteristic zymogen granules stained bright red, the mitochondrial filaments the same colour, while the protoplasm of the cell takes a smooth greenish tint. In association with many acini, the centroacinous cells can be distinguished by their clear uncoloured protoplasm and their oddly shaped red stained granular elements. Some of these cells are indicated in the illustrations, and it is easily seen that they are, apart from the shape of their granules, strikingly like the clear granular cells of the islet. The third cell which it is important to recognise is that of the ducts, and especially the finer ducts, or duct-like rows of cells, which are so often found associated with the islets. These cells are usually oval in shape and often considerably compressed. Their nucleus is central and their clear protoplasm dotted with a variable quantity of mitochondrial elements. They have, like centroacinous cells, a distinct resemblance to islet cells and, as Bensley has pointed out, the latter are often found among them. The islet cells may be easily divided into the two varieties which have been noted by many investigators, and to which Lane has given the name A and B cells accordingly as their granules are fixed respectively by alcoholic or watery solutions. With the acid-fuchsin-methyl-green stain the B cells, which form the vast majority, take a slate blue colour. They are usually smaller than the A cells, which show red granules and a clear uncoloured protoplasm. Both cells possess mitochondrial elements. In the B cells which have not taken up sufficient of the methyl green to give them their character- istic blue colour, the mitochondrial filaments give the impression of red granules, but in the heavily stained cells they hardly show. * T am much indebted to Dr. Dale for a supply of secretin prepared in this way. 78 Dr. J. Homans. The Relation of the (Oct: ae The nuclei of the various cells present less characteristic differences than do the granules. Those of the acinous cells do not always show a bright red nucleolus. On the other hand, the chromatin which they contain is often arranged like that in the islet nuclei. Moreover, the islet nuclei may often contain what appears to be a typical nucleolus. Accordingly, for purposes of differentiation, I have given little attention to the nuclei of the different cells. I shall describe the appearance of islet tissue stained with neutral gentian in connection with pancreas exhaustion experiments in the guinea- pig. In the dog, neutral gentian, though it stains beautifully the zymogen granules, has proved rather unsatisfactory for the islet granules. In the guinea-pig, on the other hand, it gives the most striking pictures. The Exhausted Pancreas of the Dog. Of the two exhaustion experiments performed on the dog only one was completely successful. In this the flow of pancreatic juice was very free, about 210 c.c. being secreted in the course of nine hours. At the end of this time, though the flow had not ceased, the pancreas appeared rosy and translucent in contrast to the opaque appearance of the small segment tied off for comparison. Examination of the discharged organ (fig. 2, Plate 2) shows the exhaustion to be fairly complete. There is no evidence of any transition of acinous to islet tissue. The acinous cells have for the most part lost their zymogen granules. In consequence, the mitochondrial filaments appear more prominent than usual, but the protoplasm of the cells, though somewhat vacuolated, still takes the characteristic smooth green stain. The islet shown in the figure is well charged with granules. Whether the gathering of these granules along the capillaries is an effect of secretin stimulation I do not know, but as most of the islets do not present this appearance, and as it has been noted in the resting gland, I believe that it is accidental. The most noteworthy change in the relation of the islet to the surrounding acinous tissue is that the exhaustion of the zymogen in the surrounding pancreas causes the islet to appear more sharply marked off than usual. This is in strong contrast to preparations of the same organ fixed in Zenker’s fluid, and stained with eosin and toluidine blue, when the islets, being merely negatively stained, can be distinguished with some difficulty from the more exhausted portions of the acinous tissue. 1912. | Islets of Langerhans to the Pancreatic Acim. 79 The Exhausted Pancreas of the Guinea-Pig. Three exhaustion experiments were performed on guinea-pigs, in only one of which could the animal be kept alive long enough to produce any obvious exhaustion of the gland. This animal lived for six hours under stimulation by secretin purified by the method of Stepp. The appearance of the gland treated by the acid-fuchsin-methyl-green method differs in no essential way from that of the dog, but under the neutral gentian stain the islets are marked off with a distinctness which I have not been able to obtain with a similar stain of the dog’s pancreas. Figs. 3 and 4 show an islet with the surrounding tissue in a charged gland and after exhaustion. By this method the zymogen granules of the acinous cells are stained a deep lilac and the protoplasm a pale purple. No mitochondrial elements are seen, and the centroacinous and duct cells likewise fail to stain. In the islet of the charged gland the B cells are stained blue, the fine uniform granules alone taking the colour. No mitochondrial elements are to be seen. It therefore appears probable that the red rods and granules which appear among the slate blue granules of the B cells of the acid-fuchsin preparations are in reality mitochondrial in nature and not granular. The A cells, or those which stain red with acid-fuchsin, fail here to take a positive stain. They appear as faint yellowish to purple homogeneous masses with an indistinguishable outline. In the exhausted gland the zymogen granules are greatly reduced. Occasional small masses are present, especially close to the surface of a lobule, but about the islet shown in fig. 4 (Plate 3) they have almost com- pletely disappeared. The protoplasm of the acinous cell, however, takes the same stain as in the charged organ, though it is often considerably vacuolated. Among the acini a cell appears to which Bensley has called attention. Several of these are shown in fig. 4 about the border of the islet, though they are no less common in other parts of the lobule. This cell is filled with perfectly round, uniform, purple granules, considerably smaller than zymogen and distinctly larger than those of the islet. I have found cells similar to these in specimens of discharged gland stained by acid-fuchsin. Here the granular stain is uniformly red, but in addition the cell protoplasm also takes a diffuse red stain. Such cells have a typical acinous arrange- ment, and rarely one is found with some of the diffuse green colour of the normal cell. I have not been able to find such cells in the acid-fuchsin preparations of resting glands, but have found one or two in neutral gentian stains of similar material. They resemble those which Bensley describes as 80 Dr. J. Homans. The Relation of the [ety a result of a post-mortem change peculiar to the guinea-pig, and which he holds to be the same cells described as a transition form by Mankowski. At first I was inclined to disagree with Bensley, and to consider the cell a result of exhaustion rather than of a post-mortem process, for while I find them abundant in exhausted glands, I have rarely seen one in a normal organ. It seems to me, however, that as the circulation in animals subjected to secretin stimulation becomes in the last half-hour very poor, and as their temperature becomes considerably subnormal, the change may be partly ante-mortem degeneration. Bensley’s evidence that such cells can be pro- duced as a result of post-mortem changes is very complete. That they are produced by exhaustion is suggested by my preparations, but not proven. Moreover, they have not been noted in the dog. In any case they furnish the only suggestion of a possible transition of acinous to islet cells which I have been able to find in the exhausted pancreas of the guinea-pig. Demonstration of Islets and Ducts by Vital Stas. I have made, perhaps, half-a-dozen injections of neutral red, pyronin, and methylene blue, according to Bensley’s method, and I can only say here that the results fully bear out Bensley’s contention that the islets have a distinct staining reaction of their own. In the best preparations, the individual cells can be distinguished, and, when the neutral red injection is combined with pyronin, the relation of the islets to the ducts is quite obvious. Though I have only made one attempt to demonstrate the islets in the exhausted gland, the clearness of outline, size, and number of the islets was quite the same as in the resting gland. For a full account of this technique and its results in exhaustion and starvation, the reader is referred to Bensley’s work. Nearly Complete Removal of the Pancreas in the Dog. The choice of the duodenal end of the pancreas as the part to be studied was made with the full knowledge that the number of islets left would be comparatively small. It was thought, however, that if an amount of this pancreatic tissue could be left, such as would barely support life, a better idea of the relation between acinous and islet tissue could be gained than by the use of the splenic end, in which the islets are known to be much more abundant. Although these experiments were planned primarily for this purpose, the importance of the relation of the histological changes to the carbohydrate metabolism made it necessary to observe the general condition of the animals, and especially the amount of sugar appearing in the urine. Accordingly, a brief protocol of each experiment is given, with a summary of the histological findings at its termination. The observations are arranged 1912. ] Islets of Langerhans to the Pancreatic Acim. | 81 according to the number of days the animals were allowed to survive operation. ExpERIMENT 1.—Young female. Weight, 51 kgrm. Rather fat. Morphine, ether. Removal of the pancreas except for a segment at the duodenal end, measuring 1°5 x 1°5 cm., which is tied off from the intestine and the rest of the organ with silk. With its blood supply intact it is transplanted through a slit in the oblique muscles to a space outside the muscles. Good recovery from operation, but the animal, though it does not look ill, remains quiet for the next three days. Urine for the second 24 hours following opera- tion 250 cc. It contains 3 per cent. sugar. On the morning of the fourth day the animal looks ill, has lost fat, and a fluctuating mass is felt at the site of the graft. Killed by bleeding under ether. Examination shows a thin, yellowish fluid surrounding the graft, which appears white and opaque. No pancreatic remains found in the abdomen. Microscopic Examination.—Superficial infection of the graft. Acinous cells highly charged with zymogen. Apparent slight increase of duct tissue. The islets appear normal in size and shape. The B cells are for the most part small, and very few of the typical slate blue granules are seen. The A cells stand out darkly stained and prominent. EXPERIMENT 2.—Adult female. Weight, 8 kgrm. (Good condition. Morphine, ether. Removal of the pancreas except for a piece 2x2 cm. tied off with silk at the duodenal end. Transplanted with its blood supply outside the oblique muscles. Considerable handling of the transplant. Good recovery from ether. Twelve hours’ specimen of urine on night of third day contained 3°35 per cent. of sugar. 7th day.—Animal thin but seems lively and eats and drinks well. Uvine for 17 hours (550 c.c.) contains 3°95 per cent. of sugar. 8th day.— Wound puffy and graft swollen. 10th day.—Site of graft looks better. Animal lively. Sugar, 1:9 per cent. (single specimen). 13th day.—24 hours’ urine contains 4°%5 per cent. of sugar. The graft looks normal, and the animal seems lively and well. 16th day.—Killed by bleeding under ether. Examination shows that the site of the graft is clean. A moderate amount of scar tissue surrounds the graft, which appears about the same extent as at operation, but considerably thicker ; white, opaque, and very firm on section. No pancreatic remains found in abdomen. Microscopie Examination.—The acinous cells are highly charged with zymogen. They appear slightly shrunken, though there is no evident destruction. There is a great relative, and probably absolute, increase of small ducts and centroacinous cells. No mitotic figures are found. The islets appear considerably altered. The B cells stain poorly, showing few slate blue granules. They contain some red and some uncoloured granular material, and are generally shrunken. EXPERIMENT 3.—Adult female. Weight, 9°6 kgrm. Good condition. Morphine, ether. Removal of the pancreas except for a piece 2x2 cm. tied off with silk at duodenal end and left in place with blood supply intact. Slow recovery from ether, and animal seems very quiet on the day following. 3rd day.—Animal lively and well. 7th day.—Animal eats and drinks well. Urine contains no sugar. 10th day.—Weight, 9 kgrm. Animal seems well. 20th day.— Weight, 8°75 kgrm. Animal seems well. Overnight urine contains 2 per cent. sugar (estimated). 26th day.—Animal in good condition. Killed by bleeding under ether. 82 Dr. J. Homans. The Relation of the [Oct. 1, ELxamination.—There is a little clear fluid about one end of the graft, which is thick, white, and opaque. The other end, or about one-half of the original fragment, is thin, grey, and scar-like, but shows the normal markings. No other pancreatic remains found in the abdomen. Microscopic Hxamination.—Sections from the swollen opaque end differ in no essential particular from those of Experiment 2 (16-day graft) except that the islets are rather better preserved. The B cells are very few in number, which takes away from the islet its characteristic appearance (fig. 6), but it is easy to identify. Sections from the atrophied end present appearances shown in fig. 5. The acinous cells still contain zymogen granules and stain characteristically, but they are shrunken and gathered into small compressed groups. There is an obvious relative increase of duct and centro- acinous tissue. A few islets can be found, but they take almost none of the bluish stain characteristic of B cells, and as they thus resemble duct cells (compare figs. 5 and 6) they are not easy to identify. They are not, however, increased in size, nor is there any evidence of a change of acini into islet tissue. My impression is that the islets are undergoing destruction by the scar tissue. EXxprerRiMenT 4.—Adult female. Weight, 6 kgrm. Rather thin. Morphine, ether. Duodenal end of gland tied off as usual and rest of pancreas removed. Transplantation of fragment 2x2 cm. with its blood supply intact outside the oblique muscles. Good recovery from ether. Although it has lost some weight the animal remains lively and well until just before it is killed. 8th day.—Urine contains no sugar. Graft plainly felt through skin, which moves freely over it. 29th day.—Animal seems well. Overnight urine (250 c.c.) contains 2°7 per cent. sugar. 33rd day.—The animal appears ill and has not eaten during the last 24 hours. Killed by bleeding under ether. Examination.—Peritonitis following perforation from fecal impaction (undigested meat). The graft is shrunken, grey, and scar-like, perhaps two-thirds its original size. A firm elastic cyst is embedded in it. The lobulation is still evident. Microscopic Examination.—The whole graft is invaded by scar tissue. The acini are broken up into small groups. The acinous cells are generally well enough preserved to be identified, and contain many zymogen granules. On the whole the general colour of the protoplasm is less evident than usual, and only the group arrangement and the presence of typical zymogen granules allow the acinous cells to be distinguished. The duct and centroacinous cells seem to be increased in number, and when found in groups resemble islets. A great many cells scattered through the specimen have the appearance of islet cells, but on looking through 30-40 sections I have found no complete typical islets. Duct Ligation. Experiment 5.—Adult female. Weight, 7:3 kgrm. In good condition. Morphine, ether. Ligation of ducts. Good recovery from ether. 4th day.—24 hours’ specimen of urine contains no sugar. 11th day.— Weight, 6°25 kgrm. 18th day.—-Weight, 7 kgrm. 28th day.— Weight, 7:25 kgrm. 35th day.—No sugar in overnight urine. Weight, 7°75 kgrm. Animal seems fat and well. Killed by bleeding under ether. EHxamination.—The pancreas is shrunken to two-thirds its original length. The whole splenic end is much narrowed and a little thickened. It appears pale and scar-like. The duodenal end is much the same. The region of the duct outlets for about 3 cin. is broad, thick, white, and opaque, resembling exactly the 16-day (Experiment 2) and 1912. | Islets of Langerhans to the Pancreatic Acini. 83 part of the 26-day (Experiment 3) grafts. There is a little clear fluid in the tissue about this region, but no evidence of the re-establishment of the ducts or the escape of pancreatic juice into the bowel can be made out. (Probably some slight connection has been re-established, as the animal’s digestion after the first two weeks became so much improved.) Microscopic Exvamination.—The thick, opaque portion resembles that of Experiments 2 and 3 referred to above. The shrunken splenic end resembles in some degree the contracted portion of Experiment 3 and the whole of Experiment 4 (26- and 33-day grafts), but there is less increase of duct tissue and considerably less destruction of the acini. A number of large islets are present. These appear normal in every respect, and though their large size and number might lead to the suggestion that new islets had been formed, they are no more in evidence than would be expected in the splenic end after the acinous tissue had shrunk to perhaps one-half its usual volume. Results of Nearly Complete Pancreas Removal. These results are to be considered from two points of view: first, and most important, according to the altered appearance of the acini and islets, and second, according to the apparent relation of these appearances to the sugar function of the pancreas. It is quite evident from my specimens that, in general, the principal changes which have occurred in the pancreatic remains are the increase of duet and centroacinous tissue, which appears to reach its height in the course of the first few weeks, and an invasion by scar tissue of the acini, and probably of the islets as well. Instead of an interchange of acinous and islet tissue, there is much more a suggestion of a relapsing of islet into duct tissue, and possibly a change of duct to islet. Even in the small fragments which have remained in the animals for several weeks, the acinous cells, though shrunken and considerably altered in their staining reaction, are comparatively easy to recognise. One distinguishes between acinous and islet tissue much more easily than between duct and islet tissue, a fact which might well be expected from a consideration of the embryology of the islets and their constant relation (Laguesse, Bensley) to the ducts in adult life. There are present, then, in these specimens, a number of cells which might conceivably be either duct or islet, their grouping suggesting now one and now the other. In all the grafts, the islet cells, that is, the prevailing cells containing the fine slate-blue granules demonstrated by the acid-fuschin technique, the B cells of Lane and Bensley, tend to lose these granules while retaining their mitochondrial filaments, and so resemble strongly the cells of the finer ducts. In the earlier, less cicatrised grafts, such islets are easily distinguished, and often show only one or two normally stained B cells. In the more scar-like specimens, when the islet cells retain their grouping, and 84 Dr. J. Homans. The Relation of the [Oct. 1, are associated with one or two slate blue cells, they are again easily dis- tinguished, but, when no such granule cells are present, and especially when the islets are disintegrated, a positive identification of any one islet cell, as opposed to a duct cell, seems to me impossible. This alteration is illustrated in figs. 5 and 6. Fig. 6 shows an islet of the 26th day fragment (Experi- ment 3). Several of its cells are obviously B cells, well stained, but the greater part of the islet is not to be distinguished from the masses of duct cells shown in fig. 5, which is drawn from the same specimen. I have been unable to determine whether this peculiarity is a result of degeneration or of over-activity, though I am inclined to attribute it to the latter. For, inasmuch as a very small amount of pancreatic tissue, and that portion containing the fewest islets, is left after operation, the islets, whatever their function, must be working at their maximum capacity. That their characteristic specific granules should therefore, after a time, be absent, is not surprising. Moreover, in the duct ligation specimens, though there is considerable formation of scar tissue, the islets, which are apparently not diminished in numbers, and therefore under no physiological strain, present their normal granule content. Even by a neutral gentian stain, with which I have never succeeded in showing the specific granules in the graft islets of the dog, the granules of the duct ligation islets are easily seen. It is suggested, then, that when only a small portion of the pancreas is left, the characteristic granules of the B cells are exhausted. In a badly cicatrised specimen such cells are hard to distinguish, but there is reason to believe that more are present than are immediately evident, and it is possible that the ducts may even be forming islet tissue. A consideration of the sugar function in connection with these changes shows that none of the animals became severely diabetic. A considerable amount of sugar appeared sooner or later in the urine of nearly all, and in the dog killed four days after operation it was present from the start. Less pancreas was left in this instance than in the other observations, probably too little to support life. Though the islet cells in this specimen were altered in the manner already noted, no conclusions can properly be drawn as to the relative importance of the acinous or islet tissue to carbo- hydrate metabolism. The animal killed 16 days after operation was mildly diabetic from the start. Here the pancreatic acini were highly charged and in good condition, while the islets were composed of feebly stained shrunken cells. It would be reasonable to suppose that the latter were exhausted from carrying too heavy a load. The animal killed at the end of 26 days was not diabetic at first and never appeared at all ill. Im this case, as in the 33-day graft, T9125 Islets of Langerhans to the Pancreatic Aci. 85 both acinous and islet tissue were apparently in process of destruction, and in both the characteristic appearance of normal islets was for the most part absent. It is unfortunate that the duct ligation experiment could not have been extended, but it is possible that some communication between the pancreas and bowel was being established. Though a number of investigators have succeeded in reducing the pancreas to a condition in which no zymogenous tissue remained, and only islets were left, without bringing on diabetes, they have not been able to apply the final test of discovering whether the animal could live without what remained. This is true of the experiments of Laguesse, and others, on duct ligation in the rabbit, in which removal of the pancreas seems to be impossible. I believe also that no such final test has been applied to the dog with a satisfactory proof of the existence of only genuine islet tissue in what remained. Until such final proof is given, the association with the islets rather than the acini of an important sugar function is only implied. Nor does my incomplete investigation more than suggest this conciusion. Conclusions. 1. The islets of Langerhans contain specific granules which allow of their positive identification. 2. There is no alteration in the islets, nor any evidence of conversion of acinous to islet tissue, under prolonged stimulation with secretin. On the contrary, the distinction between the two tissues is, under appropriate staining methods, more clear than usual. 3. There is no evidence of the conversion of acinous to islet tissue or the reverse, when only a small part of the pancreas is left to support life. 4. There is evidence that islet cells are reduced to a condition in which they appear similar to duct cells under these conditions, the first change being a disappearance or discharge of the granules characteristic of the B cells of Bensley. 5. There is evidence of an increase of duct tissue under the same condi- tions, but no evidence that this tissue produces new islets or takes up their function. 6. There is no positive evidence that islets are of vital importance to carbohydrate metabolism, but as between islet and acinous tissue the evidence favours the islet. 86 Dr. J. Homans. The Relation of the [ Oct. 1, BIBLIOGRAPHY. Bayliss and Starling, ‘ Journ. Physiol.,’ 1902, vol. 28, p. 325. Bensley, ‘Amer. Journ. Anat.,’ 1911, vol. 12, p. 297. Dale, ‘ Phil. Trans.,’ 1904, B, vol. 197, p. 25. Dale and Laidlaw, ‘Journ. Physiol.,’ vol. 44; ‘Proc. Physiol. Soc.,’ 1912, p. 11. Dewitt, ‘Journ. Exp. Med.,’ 1906, vol. 8, p. 193. Diamare, ‘Internat. Monatsch. f. Anat. u. Physiol.,’ 1899, vol. 16, p. 155. H. Fischer, ‘ Arch. f. Mikr. Anat.,’ 1912, vol. 79, p. 276. Frugini and Stradiotti, ‘Arch. Ital. de Biol.,’ 1909, vol. 51, p. 186. (Author’s resume from ‘ Lo Sperimentale,’ 1909, vol. 63.) Hédon, ‘ Arch. Internat. de Physiol.,’ 1911, vol. 10, p. 350. Helly, ‘ Arch. f. Mikr. Anat.,’ 1905, vol. 67, p. 124. Herxheimer, ‘ Virchow’s Archiv,’ 1906, vol. 183, p. 228. Karakaschiff, ‘Deutsches Arch. f. klin. Med.” 1904, vol. 82, p. 60; also 1906, vol. 87, p- 291. Kiister, ‘ Arch. f. Mikr. Anat.,’ 1904, vol. 64, p. 158. Kyrle, ‘Arch. f. Mikr. Anat.,’ 1908, vol. 72, p. 141. Laguesse, ‘Journ. de ?Anat. et de Ja Physiol.,’ 1895, vol. 31, p. 475; 1896, vol. 32, pp. 171 and 208. Ibid., ‘ Arch. d’Anat. Micr.,’ 1901, vol. 4, p. 157. Ibid., ‘Compt. Rend. de la Soc. de Biol.,’ 1902, vol. 54, p. 852. Lbid., ‘ Arch. d Anat. Micr.,’ 1910, vol. 11, p. 1. Tbid., ‘ Journ. de Physiol. et de Path. gen.,’ 1911, vol. 13, p. 5. Ibid., ‘ Journ. de Physiol. et de Path. gen.,’ 1911, vol. 13, p. 673. Lane, ‘Amer. Journ. Anat.,’ 1907, vol. 7, p. 409. Lewaschew, ‘ Arch. f. Mikr. Anat.,’ 1886, vol. 26, p. 453. Mankowski, ‘ Arch. f. Mikr. Anat.,’ 1902, vol. 59, p. 286. v. Mering and Minkowski, ‘ Arch. f. Exp. Path. u. Pharm.,’ 1890, vol. 26, p. 371. Opie, ‘Johns Hopkins Hosp. Bull.,’ 1900, vol. 11, p. 205. Tbid., ‘Journ. Exp. Med.,’ 1901, vol. 5, pp. 397 and 527. Pearce, ‘ Amer. Journ. Anat.,’ 1903, vol. 2, p. 445. Sauerbeck, ‘ Verhandl. der Deutsch. path. Gesellsch.,’ 1904, p. 217 ; ‘Centralbl. f. allgem. Path. u. path. Anat.,’ vol. 15, p. 217. W. Schiiltze, ‘Arch. f. Mikr. Anat.,’ 1900, vol. 56, p. 491. Ssobolew, ‘ Virchow’s Archiv,’ 1902, vol. 168, p. 91. Stepp, ‘Journ. Physiol.,’ 1912, vol. 43, p. 441. Tiberti, ‘ Arch. Ital. de Biol.,’ 1909, vol. 51, p. 117. (Author’s resumé from ‘ Lo Speri- mentale,’ 1908, vol. 62, p. 1.) Ibid., ‘Arch. Ital. de Biol.” 1909, vol. 51, p. 123. (Author’s réswmé from ‘Lo Speri- mentale,’ 1908, vol. 62, p. 399.) Tiberti and Franchetti, ‘Arch. Ttal. de Biol.,’ 1909, vol. 51, p. 127. (Authors’ réswme from ‘Lo Sperimentale,’ 1908, vol. 62, p. 81.) Tschassownikow, ‘ Arch. f. Mikr. Anat.,’ 1905, vol. 67, p. 758. Vincent and Thompson, ‘ Internat. Monatsch. f. Anat. u. Physiol.,’ 1907, vol. 24, p. 61 Weichselbaum and Kyrle, ‘ Arch. f. Mikr. Anat.,’ 1909, vol. 74, p. 223. Wright and Joslin, ‘Journ. Med. Research,’ 1901, vol. 6, p. 360. Soc Proc. 8B. v0l.86, Pl. 2. oy Homares. WATERLOW A SONS LIMITED, LONDON WALL, LONDON, Homans. hoy. Soc.Proc. B. vol. 86, PL. 2. WATERLOW & SCNS LIMITED, LONDON WALL, LONGON, 1912. | Islets of Langerhans to the Pancreatic Acin. 87 DESCRIPTION OF PLATES 2 anp 3. Fig. 1.—Normal islet in the charged pancreas of the dog. a, A cells of Lane. 0, B cells of Lane. c, centroacinous cells. 7.b.c., red blood corpuscles. Notice the arrangement of the large round zymogen granules in the surrounding acini, also the distribution of the mitochondrial filaments. The borders of the islet are extremely irregular and no limiting membrane is present. Fig. 2.—Normal islet in the exhausted pancreas of the dog. a, A cells of Lane. }, B cells of Lane. Notice the distribution of their granules along the capillaries, which are indicated by the red blood corpuscles (7.b.c.). ¢, centroacinous cells. d, duct cells, which are continuous with those of the islet below and low down on the right. z, exhausted acini. Notice the prevailing absence of zymogen granules, the presence of some mitochondrial filaments, and the vacuolisation of the cells. In the low right-hand corner a small group of acinous cells is cut off from the rest by a small duct. Fig. 3.—Normal islet in the charged pancreas of the guinea-pig. a, A cells of Lane, undifferentially stained. 06, B cells of Lane, containing fine blue granules. 2, acini, containing zymogen granules. Fig. 4.—Normal islet in the exhausted pancreas of the guinea-pig. a, A cells of Lane. b, B cells of Lane. m, acinous cells, containing “ Mankowski” granules. 2, acini without zymogen granules. Fig. 5.—Group of duct cells and atrophied acini (from 26-day fragment). c, centroacinous cells, d, masses of duct cells, which partly retain their tubular arrangement and occupy the central part of the figure. Notice the resemblance of these cells in the arrangement of their mitochondrial filaments and granules to the islet of fig. 6. ig. 6.—Islet containing only a few typical B cells (from 26-day fragment). 0, B cells of Lane. d, duct cells (poorly preserved). z, acini containing zymogen granules. Notice the small number of typical B cells. The rest of the islet cells might, from their appearance, be duct cells or A cells. Compare with figs. 5 and 1. zl VOL. LXXXVI.—B. H 88 Lhe Metabolism of Lactatung Women. By Epwarp MeEtiansy, M.A., M.B. (Cantab.), Beit Memorial Research Fellow. (Communicated by Dr. F. G. Hopkins, F.R.S. Received October 3, 1912,— Read January 23, 1913.) CONTENTS. PAGE 1, Previous Work wie diseescies Reapers ene eas eins ceh aut aas ecees ne ee ae eee ee Cee ' 88 2. The relation of creatin excretion to the involution of the uterus ............... 89 3. ts relation to mammary, gland) activiby....2:...-+---sa0-cesscaeon-soecueersseeseeeers 92 a. Cases illustrating a quantitative relation between creatin excretion and Milk Secretlowss. cidsccnseesenstelshqosbh eateries. Soldceeates aeplen lacteese Merce 93 b. A case of mammary gland activity and creatin excretion developing late, after childbirth (ac8..2A sccctesssae. snes se seeeisnantien cots tock uareeee eee eee 97 ce. A case illustrating simultaneously suppressed creatin excretion and Mammary: gland AChIVADY...ccecss 6 noes cocesn sickness coke sate ceeeeene seared 98 4. The effect of adding casein to the diet of a puerperal woman ................+ 100 5. The independence of the puerperal excretion of creatin and carbohydrate Metabolism .sicecachawasteseeedeecstes de ttkleat’s sotes suactickien bugeiock oats tube AOC REE Eee eee 101 6. General considerations: Gaccon-ctdctaes.seersiat. scout « sot boceetiasseeeee veneers Ieee eae 106 Te SUMMALY ce cass. sud saoareceeations Sabeieebnshee bee al Vetres obtener Meee eee eee eee 108 1. Previous Work. The metabolic changes of pregnancy have been studied by various workers, principally from the point of view of comparing the total output of nitrogenous material with food intake. The following results may be taken as proved :— (1) There is a marked rise in the output of nitrogen following childbirth. Grammatikati(1), Zacharjewsky (2), and Slemons(3), among other workers, have definitely proved this. (2) This increased nitrogen output more than counterbalances the nitrogenous intake, so that women, at this time, lose nitrogen. This is in marked contrast to the storage of nitrogen taking place before delivery. The explanations offered by the various workers on these points vary considerably. Heinrichsen (4), Zacharjewsky (2), and Longridge (13) ascribe the increased nitrogen excretion to regressive changes in the puerperal woman, particularly changes affecting the uterus. Grammatikati(1) thought it was to be explained by mammary gland changes, more especially by the formation of milk fat from protein and the excretion of the nitrogenous residue. One other point of interest observed by Slemons(3) is that the total nitrogen of the urine is less on the day of delivery than any other day and that the drop is greater the more prolonged the labour. The Metabolism of Lactateng Women. 89 As for the individual nitrogenous substances excreted at this period but little is known. Ammonia, which forms a larger percentage of the total nitrogen with the advancement of pregnancy, gradually diminishes to the normal amount, Urea, as might be expected, forms the greater part of the increased nitrogenous excretion following childbirth and, according to Grammatikati, is a maximum when milk appears in the breast and diminishes with the weaning of the child. Other observers have not been able to corroborate this observation or to ascribe the causal connection between urea excretion and mill formation advanced by Grammatikati. The appearance of creatin in the urine of lying-in women was first observed by Shaffer(5) and in dogs by Murlin(6). The present account is more particularly connected with this excretion of creatin by puerperal women. The analysis of such a condition seemed likely not only to furnish important results as to the life history of creatin, but also to shed light on - the strange metabolic changes taking place in the body at this time. 2. The Relation of the Puerperal Creatin Excretion to the Involution of the Uterus. Other well recognised conditions in which creatin is excreted include inanition and cancer of the liver (14), and since a striking feature about these conditions is the rapid wasting of the patient, it has been assumed that when the voluntary muscle breaks down, creatin is liberated into the blood-stream and excreted. This explanation was extended by Shaffer(5) to explain the puerperal excretion of creatin, but, in this case, the muscular tissue supposed to supply the creatin to the blood-stream was not the voluntary muscle but the involuntary muscle of the uterus. A serious difficulty, however, prevents the acceptance of this explanation, in that, while voluntary muscle contains abundant creatin, uterine muscle is quite.devoid of this substance. In a previous paper (14), it was pointed out that creatin has a very limited distribution in nature and can only be found in the cross-striated muscle of vertebrate animals. The cross-striated muscle of invertebrates such as the lobster and the king crab, and, on the other hand, the smooth muscle of vertebrates, represent types of muscle which contain no creatin. Consequently, creatin cannot be found in the smooth, unstriated muscle of the uterus. It has been maintained that creatin and creatinin are present in tissues like the uterus because extracts of such tissues frequently give the Weyl or the Jaffé colour reactions. Both these colour tests are given by so many other substances that they are unreliable* as proofs of the presence of creatinin. * Weyl’s colour test with sodium nitroprusside and caustic soda is also given by aldehyde, acetone, acetophenone, and aceto-acetic acid. Jaffé’s colour test with picric acid and soda is given by any reducing agent. H 2 90 Mr. E. Mellanby. [| Oct. 3; The red colour, which may be developed by mixing extracts of uterine muscle with alkaline picric acid, disappears rapidly on dilution, and cannot, therefore, be due to creatinin. Consequently, Shaffer’s explanation, as it stands, does not adequately explain the puerperal excretion of creatin. However, it is possible that, although uterine muscle contains no creatin, yet some substance is present in the muscle which is converted into creatin when that tissue involutes, and appears ultimately in the urine. In order to test this point, the following observations were made. In a lying-in ward two women were delivered of children by Cesarian section, the one case (A) because of a contracted pelvis, the second case (B) because of a ruptured uterus. At the time of operation the uterus in Case A was stitched up and retained, while in Case B, with the ruptured uterus, it was completely removed. If the involution of the uterus is accountable for the post-partwm excretion of creatin, then it is clear that Case A would excrete much more creatin than Case B, where there was no . uterus to involute. A. Cesarian Section. Uterus B. Cesarian Section. Uterus stitched up and retained. Operation January 13, 1910. removed. Operation January 16, 1910. | | | Total Total Creatin s Total Total Creatin | Vol. | : =e == Vol. : vad eee | creatin. | creatinin. | (Oreatinin creatin. | creatinin, Creatinin mgrm. in| mgrm. in C.c. grm. grm. C.c. 10 c.c. 10 c.c. Jan.14| 815 | 0-987 1-03 0°96 Jan.17| Spec. 9°7 10 0:97 1D |) CKD) |) FeV 0°91 0-95 cp ls) » 16] 450 | 0°58 0°54 1:08 10 Mean ae Tore pa » Le GO | Oc 0-7 1:03 »» 205| ay Ls} || LC®) |) Oz 0:97 0°72 Badal L270 1°76 1:27 1-38 », 19 | 1650 | 0°7 0-86 0-89 | These figures show that the removal of the uterus did not prevent the excretion of creatin, and in fact, Case B, where there was no uterus to Fortunately, creatinin has a much more potent action than other physiological reducing agents and carries on the reaction very quickly with the formation of di-amino- mononitrophenol—a red substance which retains its intensity on strong dilution. The end product of such reducing agents as dextrose, levulose, maltose, aldehyde, is generally mono-aminodinitrophenol, which is also a red substance, but loses all colour intensity on dilution [Chapman (15) ]. This explains why the full dilution of Folin’s method of estimating creatinin is absolutely essential ; for it is clear that on small dilution the colour of the mono-amino- dinitrophenol will interfere with that of diaminomononitrophenol. This point concerning dilution is a common source of error in recent research on creatinin. 1912. | The Metabolism of Lactating Women. 97 involute, excreted much larger quantities of creatin. It would be unfair. to press the interpretation of these experimental figures too far, because I think the excretion of creatin following a Czsarian section may not be completely analogous to that accompanying a normal pregnancy. For instance, the following figures show that, even after an abdominal hysterectomy for uterine fibroids, creatin is excreted :— Abdominal Hysterectomy. Fibroids of Uterus. Time after | Creatin in Creatinin in | Creatin operation. | 10 c.c. 10 c.c. Creatinin™ | mgrm. mgrm. ts} MOWHE a adanese | 9°9 14:0 0°7 CO Ch peceadeoerca | 16-0 12°8 1°25 Bee rates av 4:3 9°6 0°45 It may further be stated that all abdominal operations result in the excretion of some, but very variable amounts of creatin. The significance of this fact will be considered elsewhere.* As regards the Cesarian section figures two points are worthy of mention. (1) Although such large amounts of creatin were excreted, the creatinin excretion was normal or but little diminished. In other words, the large creatin excretion was not produced at the expense of the creatinin. For instance, Case B excreted 2°86 grm. of (creatin + creatinin) per diem on an average over three days, whereas in normal health such a woman would excrete but little more than 1 grm. of creatinin and no creatin. This point is important, because it is commonly said that creatin is converted into creatinin, probably by the liver, and also that an increase of creatin excretion is accompanied by a diminution of creatinin. This relation may possibly hold in conditions like inanition, and in the absence of carbo- hydrate (Cathcart, 7) from the diet, for in such conditions there is usually a diminution of creatinin excreted, together with an increase of creatin. But it does not hold in the above cases of Cesarian section. (2) The creatin excretion following Cesarian section did not depend on inanition or absence of carbohydrate from the diet. Both patients were taking an adequate amount of food throughout the period of examination and the urine did not indicate any condition of acidosis. To sum up, there is no evidence that the puerperal creatin excretion depends, to any extent, on the involution of the uterus. * In the meantime I should like to utter a warning with regard to the interpretation of experiments on creatin metabolism which involve opening up the abdominal cavity. 92 Mr. E. Mellanby. [Oct. 3, 3. Evidence of the Relation between the Puerperal Excretion of Creatin and Mammary Gland Activity in Women. Pregnant rabbits, hke pregnant women, excrete abnormal quantities of ereatin. In fact the estimation of the creatin in the urine of a rabbit is a useful means of diagnosing pregnancy. It may be well to state that in my experience all rabbits, male and female, excrete small quantities of creatin. Oxen also, independently of sex differences, normally excrete creatin in addition to creatinin. This observation, affecting herbivorous animals, is in marked contrast to the total absence of creatin from the urine of normal people. When I started to investigate the puerperal excretion of creatin, rabbits seemed to be the best animals for examination. It was surprising, however, to find that the excretion of creatin by rabbits stops immediately after delivery, and consequently they were useless for this investigation. One possible factor suggested itself as an explanation of this difference between rabbits and human beings, namely, that rabbits eat their placentze after delivery. It might be imagined that the proper performance of the functions of the organism after parturition depended upon the presence of certain chemical groupings, which were supplied by the digestion and assimilation of the placenta; but that, in the human being, since no placenta was available, the substances had to be supplied at the expense of other tissues such as the muscles, with the result that creatin was liberated and excreted at the time of the transference of material. If natural craving for animal food is any indication of physiological needs, then it is certain that female animals require the chemical substances of tissues such as muscle after parturition. The longing which such women have for meat has its analogy in the lower animals in their eating placente and frequently their young. In the case of herbivorous animals this seems to be the only time in their lives that they are carnivorous and may have some special significance such as the urgent requirements of the animal organism for substances such as extractives. From this point of view, therefore, a cow was allowed to eat its placenta after parturition, in order to see whether the creatin excretion would be suppressed, The following figures were obtained :— £912. The Metabolism of Luctating Women. 93 Cow. Creatinin in Creatin in Creatin 10 c.c. 10 c.c. Creatinin’ | mgrm. mgrm. SUNIG 2 25e denne eee 6-0 POH 0°45 aah OhCarcaenree 8-0 2°6 0 *325 | Tea eee 10°1 4°] 0-405 | BE NOW snskaaslan 9°0 3°2 0 °356 eeall bigl Pam eae 8°5 3°5 0-41 15. peek Censor ee 6°6 5-3 0°8 | ROR Bae oe 60. | oe Mer PRO Oia 2 a4 || Sah Osea eeescisee 6 °2 | 3°0 0-48 | a (ieee 6-0 4°3 | 0-71 | * Calf born June 6,4 a.m. Placenta eaten by 10 a.m. It will be seen that a rise of creatin excretion follows the ingestion of the placenta rather than a fall. Consequently it is unlikely that placenta eating explains the absence of creatin in a rabbit’s urine post partum. The investigation was continued on women and this will now be described. (a) A Quantitative Relation between Creatin Excretion and Milk Secretion im Nursing Women.—The women were patients at a lying-in hospital,* and the diet in all cases was ecreatin-free, in that no meat, fish or meat extracts were given. Such a diet} has great drawbacks because of the desire, mentioned above, which these women have for meat and tasty food. The disappoint- ment at the plainness of the diet often makes them depressed, and the mental condition is soon reflected in the poor progress made by the child, either because it is not properly nursed or because the mother’s milk is of poorer quality. This difficulty was especially obvious in cases of multipare who had been in the lying-in hospital before and were therefore accustomed to plenty of meat. Most of the cases studied, therefore, were primipare, in whom this mental depression was not so obvious. Below are the figures obtained in the examination of four normal cases of childbirth. They were all on the same diet and were living in the same ward under precisely the same conditions. * The Lambeth Lying-in Hospital, S.E., to the staff of which, and in particular to Dr. J. 8. Fairbairn, I wish to express my indebtedness. + A typical day’s diet, though it varied to some extent, was as follows :—Milk 25 oz., gruel 5 oz., cocoa (made of milk) 5 oz., an egg, bread and butter 5 oz. Occasionally liver up to 4 oz. was added. [(Oetrss Mr. E. Mellanby. 94 ‘uroIp ted *z0 eG. 0 = 20 G. 9 = potmed Surpuodsarxoo Sammp 4ys1eM s.Aqeq UI esvetoTy “127.0 = (TONeIov] Fo yuaUIYSTTqQVyse 104ye) ORI — UTUTYBETO, uTyBaLD) SIT Te-0 PPE: O SI-T P- PL 8.7 0- FL 60- 0 oos — Fes = Rs 146 ELIT 8g. 0 168: 0 ZO: T 8. ST GF 0: IL G80- O ORS —(°% ™ _ 198 ELIL 62: 0 HE: 0 O1-T 6: FI 6-8 g. ST 60: 0 oss | T ounp “wa 9 09 dn ‘Kep 447, E411 eS. 0 009- 0 €I- 1 &. 9T 9.9 PSL 680- 0 o1i6 | te “ s 199 S911 0s: 0 999. 0 Zé. T 8-41 8-9 G. eT 160: 0 os6 = |0og “ a 19g ) Variations in wave-length determination by different observers on same sample of blood (individual variation). Description of Personal Variation. As a result of wave-length measurements of the bands of the O2 and CO compounds of hemoglobin, I found that those previously published differed considerably from my own; further, in these there was found from time to time a small but quite definite divergence. Both these points are shown in the table below. It will be seen from the table that by my measurements the change in wave-length of the a-band, when O: is replaced by CO in the hemoglobin compound, is 53°3 Angstrom units ;+ a value which is lower than that given by the other observers quoted. This may have been due to dissociation of * ‘Journ. Physiol.’ 1912, vol. 44, p. 1. + Angstrém unit = 1078 cm. Factors Affecting the Measurement of Absorption Bands. 129 Table I—Wave-leneth Measurements on Heemoglobin Bands, | a-band. B-band, Observer. | Date. O,, CO. | Difference. Os. co. Difference. ca z | een shea ae ial, 18 (ap hae 5764 | 5710°5 54085 | 53863 | 5764 5713 5405 5357 °5 | 5765 *5 5713 °5 5400 5361 | 5766 5709 5402 5362°5 | 5764 "5 5712 °5 5401 °5 5363 5763 °5 5710 5402 5364 5763 °5 5711 5402 5361 57/65°5 | oF 5402 °5 5360 57638) | 5 709)5 5404. °5 5361 °5 5764 | 8712 5405 $364 5764. °5 5711-1 53 °3 | 5402 °8 5361 °5 41°3 1912 | | Gamegee...... 5790 5720 70 5438 5380 58 1880 Formanch...| 5781 5710 | 71 5417 5375 42 1901 Dilling ...... 5785 5715 70 5420 5378 42 1909 Heubner* ...| 5769 5402 1912 | * © Bio-chem. Ztsch., 1912, vol. 38, p. 345. the COHb by light, since it was considered inadvisable to employ any screening fluid, in case this influenced the normal positions of the bands. Ammonium sulphide has the property of rendering the CO compound much more stable to light; measurements done in the presence of this body gave a value as high as 59 A.U. Variations were, however, still found in the absolute values, in one case the wave-length varied as much as 5:4 A.U. from the typical position. Alterations in the scale of the instrument were in every case avoided by making the measurements from two standard lines, sodium supplying that for the a-bands, the thaluum line 5351 A.U. for the @-bands. These differences in wave-length must be due to one of two causes, either to actual differences in the wave-length of the various blood solutions measured, or to retinal changes brought about by alterations in the intensity of the incident light, or the previous stimuli received by the eye. Before dealing with the explanation of these phenomena I will proceed with the description of the second, somewhat larger, individual variation, Description of Individual Vuriation. Measurements of wave-length carried out by several observers on the same blood solution showed that considerable divergence of opinion could exist as to the apparent positions of the absorption bands. In this case also the first 130 Mr. H. Hartridge. Factors Affecting the | Nov. 1, question. to require consideration was the cause of the divergence, but the effect this might have on the accuracy of the carbon monoxide estimations had also to be investigated. The differences that may exist between the readings of two observers may be seen from fig. 1. On the left are plotted the observations and calibration curve obtained from them for the observer G. W.. On the right is drawn a portion of my own calibration curve taken from the previous paper.* On comparing corresponding points on the two curves it will be seen that the bands appear 80 40 e e e e e (o) e be Ce e e e 8 « eV ° @ © e a 5 e-M eo e 7 e-M fe} (@) -20 +20 60 100 Bie: 1: @, initial observations ; M, mean of several readings; 0, final observations. everywhere displaced by about 20 scale divisions (10 A.U.). The initial observations of G. W. are given in the table below and are shown plotted as black dots. But to ascertain how far subsequent readings could be relied on to give the actual percentage saturation of a sample of blood, a further series of isolated readings were made. These are shown plotted in fig. 1 as white dots, and are given in the table side by side with the actual percentage saturation present; it will be seen that the latter readings have an average error of 11 per cent. These figures show that although a considerable * ‘Journ. Physiol.,’ 1912, vol. 44, p. 23. 1912. | Measurement of Absorption Bands. 131 Table I1.—Observations for Obtaining and Checking Calibration Curve. 7 Observer G. W. Initial observations by which calibration curve was obtained. co. Reading. | Co. Reading. per cent. | | per cent. 75 °8 56 10°1 M— 4°5 20°5 } — 2 45 °8 19 27°5 — 4 | 16 | -— 5 75°8 62 72 60 62 | 41 39 4 14 45 +20 29 M7 20°5 0 68 -2 47 46 M 20 28 °5 ) TO ai — 5 11 —10 30 2 52 26 | 60 28 58°6 38 and 32 66 °5 45 43 °5 20 50 °6 26 66 °5 41 | 26 °8 | 4 52 28 38 15 23 -9 2 ; 72 66 67 43 | 50 16 53°8 35 M = mean of 4 or 5 observations. Final observations used to check accuracy of calibration curve. | Readin | CO ealeulated from : Actual CO. | of SDECEEOREORE: | a4 Difference. | | Cae pene per cent. per cent. per cent. | 50°6 28 53 “A, 32 °4 +7 } 31 —1°4 75 °8 66 76 +0°2 6°7 —10 | — 3 | -—3 77 43°35 16 4l —2°5 63 °6 46 66 +2°4 58-6 31 55 °*4 —3°2 26°8 2 24 °3 —2°5 12 — 6 10°5 -—1°5 Average error —1°1. difference may exist in the curves of different observers, yet the final results obtained by them need not vary greatly from the true value; the variations depending largely on the degree of practice of the observer. ay Mr. H. Hartridge. Factors Affecting the [Nov. 1, Experimental Demonstration of the Factors that Control the Position of Boundaries and Mean Wave-lengths of Absorption Bands. Theoretical considerations showed that the factors of importance are :— (1) The number of molecules of pigment encountered by light in trans- mission (7.¢. concentration x thickness). (2) The initial intensity of incident light. (3) The activity of the retinal response. (1) The effects of alterations in concentration or thickness of the solution under examination are well known, having been first investigated by Rollett.* These observations have been repeated with the new method, the results being given in the table below and shown graphically in fig. 2. The widths of the bands increase with an increase in concentration or thickness of the solution till a point is reached when fusion of the neigh- bouring bands takes place. The mean wave-lengths of the «- and @-bands, on the other hand, do not appear to be affected by changes in concentration, and this would point to the bands being symmetrical in form. Mention, has Table I1J.—Effect of Change in Concentration on Mean Wave-length and Boundaries of O2Hb Bands. | | Hb ja B | Clan: E Ty he! H Ee ieee | | | i per cent. | | O01 7260 5764 | 5403 4210 0-2 7260 | 5800 | 5765 | 5738 | 5600 | 5485 | 5401 | 5845 4420 0:3 5810 | 5765 | 5783 | 5600 | 5487 | 5403 | 5340 | 4590 0 +4 5829 | 5765 | 5716 | 5602 | 5490 | 5400 | 5310 | 4690 05 5837 | 5762 | 5703 | 5604 | 5505 5397 | 5300 4750 0°6 | 5848 | 5765 | 5685 | 5607 | 5526 5396 | 5290 | 4830 Or | | 5856 | 5759 | 5678 | 5608 | 5540 5395 | 5265 4880 0:8 | 7210 | 5860 | 5678 | 5612 | 5546 | 5250 | 4950 0:9 7160 | 5865 | 5610 | | 5230 | 5130 5000 1-0 7130 | 5863 | | 5226 | 5126 | 5010 | | | iL oil 5864 | | 5210 | 5123 | 5010 1) 5876 | 5180 | 5123 | 5030 1°3 5880 | | 5125 | 1-4 5886 | | | | | 2-0 7000 | 5900 | | | | } | | A, the apparent commencement of spectra in the infra-red. B, the yellow edge; D, the green edge; OC, the mean wave-length of the a-bands. H, the centre of unabsorbed area between a- and 8-bands. F, green edge; H, blue edge ; and G, mean wave-length of 6-band. I, centre of unabsorbed area between §- and y-band. J, green edge of y-band. * Schafer, ‘Text-Book of Physiol.,’ vol. 1, p. 233. 1912.] 0% Concentration of hemoglobin. ae 2% Measurement of Absorption Bands. 133 4000 5000 Wave-length. Fic. 2.—Sheep’s Blood, 1 cm. thick. ©, apparent edges of bands; @, mean wave-length. 6000 been made elsewhere* of the possession of symmetry by these bands, this is of considerable theoretical importance, and is to be more fully investigated at a future date. (2) The effect of alteration in intensity of the incident light on the boundaries of the bands of O2Hb is shown in the following table :— Table IV.—Effect of Changing Intensity of Incident Light on the Bands. Intensity. | A. B. | C. | D. E. F G. | 8 | 5812 5764 ) 5698 | 5598 5492 5398 5316 4 5823 5766 5698 | 5594 5501 5403 53814 2 5833 5764 5692 | 5595 5500 5404 5302 1 5841 5763 5688 5583 5511 5402 Cy , 5840 5765 5684 | 5591 5396 m 5852 5765 | 5588 5402 ry 5856 5766 | 5596 5391 a 5868 5765 | 5592 5393 A. Yellow edge E. Green edge B. Centre io a-band. F. Centre fos B-band. C. Green edge G. Blue edge D. Centre of unabsorbed area. It will be noted that a decrease in the intensity of the light tends to make the bands broader, whereas an increase makes them narrower. In this case * “Journ. Physiol.,’ 1912, vol. 44, p. 8. 134 Mr. H. Hartridge. Factors Affecting the |[Nov. 1, also evidence is afforded of the symmetry of the bands. Limiting the change of intensity to that part of the spectrum occupied by one side of a band still causes the same change as those described above, an increase in intensity making that side of the band narrower, a decrease making it broader, the change being always such as will cause the band, as a whole, to appear shifted away from the side of greater intensity (see Table V). Table V.—Effect of Changing Intensity of Incident Light on One Side of the Bands. Increased intensity. Decreased intensity. Normal band. [a ea Green. Yellow. Green. Yellow. 5765 5767 5760 5759 5773 5768 5762 5760 5771 5760 5774 Gelatin films, stained with methyl orange and malachite green, were used to reduce the intensity of the incident light in the green and orange regions of the spectrum respectively. The increased intensity was obtained by supplementing the light used above by a Nernst lamp, which could be fitted with light filters that transmitted either the orange or green parts of the spectrum. (3) It is well known that by decreasing the intensity of the light incident on the retina, a point is ultimately reached where no sensation of light can be obtained. This point is called the “visual threshold.” The amount by which the light has to be decreased to reach the threshold depends on the activity of retinal response. This not only varies with the wave-length of the light but also with the stimuli received by the retina previous to the threshold measurement. Thus, when the previous stimulus has been intense, a light of considerably greater intensity than the threshold may cause no sensation. Moreover, a stimulus of a certain wave-length not only affects the threshold for that wave-length, but also to a less extent that of wave-lengths on either side; in cases where the previous stimuli have been very intense and prolonged every part of the spectrum becomes affected. The position of the visual threshold at different points of the spectrum is one of the factors which determine the appearance of absorption bands ; this is well shown in the case of oxyhemoglobin in Table VI. Previous stimulation by a bright light causes the threshold to rise and makes the bands become wider; a fall in the threshold conversely makes 1912. | Measurement of Absorption Bands. 135 Table VI.—Effect on Band Widths of Altering Value of Threshold. Ioreaege Cue SONWRMONE — Geocemcadnanergornendeseoda cnn ul cor 04 per cent. Hb. 0°6 per cent. Hb. | see, es ese. Swe | INGVINMIFOV ON ctnanceacteedtone tect desceonroctc wees 5829-5712 5826-5700 ANG SD EGR) ayn FS shal GNA! Gecapn ae 2d0de8 09D =COBSe UND seHooObae 5811-5726 5852-5691 | After stimulation by a bright light ........... boat he 5845-5694. 5859-5680 | The measurements are of the a-band. them narrower. Further, if the bands be measured after alteration in the threshold on one side of the band only, then a movement of the mean wave- length of the band will be found to have occurred; this being in every case towards the side on which the threshold is highest. Table VII.—Showing Change in Wave-length of Bands produced by Alterations in Retinal Adaptation. | | | co Norma! position of | i B : a-band. : j } | per cent. 24, 5757 °5 —1°5 +3 39 °4 5750 —l11 +2 10-1 5762 Sige lal 53°8 5739 -—1 | +2°5 68 °2 5728 5 — il +3°6 The readings in Column A were taken after the eye had been blinded for green rays by a powerful light; in Column B, the light used was orange. Lastly, there is another important factor, viz., contrast, the full effects of which are to receive future consideration. By comparing in several cases the visual appearance of absorption bands with their spectrophotometer curves, the following observations have been made in nearly every case :— (1) A certain density gradient is necessary for an absorption band to be recognised. (2) Any gradient that exceeds this tends to be increased in slope by contrast. (3) Light and dark areas connected by such gradients increase in width. (4) Areas of high intensity become brighter, those of low intensity darker. (5) Small density changes within such areas tend to become obliterated. (6) A line breaking the continuity of a gradient causes a considerable increase in the differences of density on the two sides. The method of observation and data will appear at a future date. Fig. 3 shows these six effects, it also demonstrates how contrast impairs VOL. LXXXVI.—B. L 136 Mr.H. Hartridge. Factors Affecting the [Nov. 1, the accuracy of measurement, either of the mean wave-length or of the apparent boundaries of bands by the cross-wire and micrometer. Where such a method is alone available, the use of a broken line is indicated. Fic. 3.—To show the Effects of Contrast on the Appearance of Absorption Bands. A, normal density curve; B, curve altered by contrast ; C, effect of breaking continuity of density gradient by the line of the micrometer; D, suggested use of “broken ” line in the micrometer. A brief theoretical consideration may now be given of the manner in which threshold and intensity of incident light control the appearance of absorption bands. Any desired value may be given to the density at any point of the spectrum, by causing the incident light, on transmission, to encounter the correct number of pigment molecules. Any part of a spectrum may therefore be made to fall below the value of the threshold of vision, and, when such is the ease, these portions must give to the eye the appearance of uniform black areas having no visible luminosity. This explains why, by starting with a dilute solution of such a pigment as hemoglobin, and gradually increasing either its concentration or thickness, a point is reached at which a narrow black area appears in the centre of the «-band, and, further, why this area increases in width on either side as the concentration or thickness is further increased, until fusion of neighbouring areas ultimately takes place. Similar changes may clearly be brought about by altering either the initial intensity of the incident light or the threshold. In fig. 4 these changes are shown diagrammatically. Such considerations explain the existence of certain conditions upon which the accuracy of the estimation of CO in blood is found to depend. As much of the bands as possible should be below the threshold, without the conditions being such as to impair the luminosity of the area separating the « and @ components. These optimum conditions are always aimed at when making CO estimations; their existence is of great practical importance, and is well shown in fig. 5. 1912. | Measurement of Absorption Bands. 137 7 att! A B 1S T2 atl laC constant TaC consrant Tal constant ai a: als: Fia. 4.—To show the Effect of Changes—A, in the visual threshold ; B, in the initial intensity of the light ; C, in concentration—on the Width of Absorption Bands. +6 +4 +2 i . OPTIMUM 3° © 10% ‘75% 5% 25% Fic. 5.—To show the Existence of an Optimum Concentration at which Best Measurements are to be obtained. The alterations have been considered above as taking place uniformly throughout the spectrum. It remains to consider the case when alteration on one side only of a band is involved. It has been shown above that a rise in the threshold or a fall in the intensity of the incident light both have the same effect, namely, to increase the width of the bands. Where one side only of a band is affected, it is clear that a change in the mean wave-length of the band will be involved, since one side of the band increases in width, the other remaining stationary. This will cause the band as a whole to appear shifted towards that side on which the intensity reaches its lowest, or the threshold its highest value. This is shown diagrammatically in fig. 6. It will be remembered, when describing the personal and individual variations in the wave-length of the bands of hemoglobin, that even under the same conditions differences are still found to occur; these are to be explained as being due to alteration in the threshold. For observers are agreed that the threshold value for light of different wave-lengths is a factor 138 Mr. H. Hartridge. Factors Affecting the [Nov. l, Tz —L- T2 Fic. 6.—To show the Effect of Changes in the Threshold or Intensity of Incident Light, taking place on one side of a band only, on the mean wave-length of the band. At A the value of the threshold on one side has been changed ; at B’ and B the normal intensity of the incident light has been altered, moving with it the threshold. which not only varies greatly with the individual, but is also modified by the quality and time of the stimuli the retina has previously experienced.* In the case of the observer G. W. and myself, a difference of 10 A.U. was found in the wave-length of the bands; this would point to an abnormality in the threshold values of the green or yellow regions of the spectrum for one of us. To test whether this was accompanied by any change in the normal intensity of sensation, tests were carried out by Rayleigh’s method, a yellow being matched with a combination of red and green ; we were unable, however, to demonstrate any difference in our sensation of these colours. The variations demonstrated above in the visual measurements of wave- length of absorption bands are also to be found when a photographie process has been employed; for colour, sensitive plates have a threshold at different wave-lengths which depends on the dyes used to sensitise the emulsion. Further, the actinic value of the light used to illuminate the slit of the spectrograph should be as far as possible -adjusted to be of uniform intensity throughout the spectral region under examination. For this purpose the cadmium are and similar light sources which yield a number of fine lines of very varying intensity are unsuitable. When a suitable source has been chosen the true density gradient may be ascertained by determining experimentally the threshold values at nine or ten evenly distributed positions in the desired spectral region. It thus becomes possible to obtain for the infra-red and ultra-violet portions of the spectrum density curves that are the equivalent of spectro-photometric measurements in the visible spectrum. It is interesting to observe that the factors of contrast, etc., here described may in their relation to the phenomena of absorption bands be conveniently * © Physiology of Special Senses,’ Greenwood, p. 104. + Rayleigh, ‘Collected Papers,’ vol. 1, p. 543. 1912. ] Measurement of Absorption Bands. 139 investigated without the use of spectral apparatus; all the effects being obtained by means of a/rotating disc calculated from the spectrophotometer curve of any chosen band. Such a disc is ‘shown in fig. 7. | oreen {/ ORANGE | ORANGE GREEN Fic. 7.—Diagram of Colour Disc used for imitating Absorption Bands and Effects of Changes in Intensity and Threshold. Summary. (1) When absolute values of the CO saturation of hemoglobin are required the spectroscopic method is complicated by the fact that each observer must obtain a calibration curve for himself, and this should also be checked from time to time against the blood from the particular source under examination. (2) These individual differences are due to the particular threshold values at different wave-lengths. (3) The differences greatly detract from the value of accurate wave-length measurements of absorption bands, both visual and photographic. (4) They do not, however, prevent accurate CO estimations being made, provided that care be taken to work under standard conditions. (5) The effects are considered of variations, on one or both sides of an absorption band, in— (a) The initial intensity of the light. (0) Value of threshold and adaptation of retina. (c) Contrast. VOL. LXXXVI.—B. M 140 Mr. Graham Brown. 3 [Nov. 12, I wish to thank Mr. G. Winfield for the trouble he has taken in making himself familiar with the use of my spectroscopes. I am also most grateful to Prof. Gotch for the kind suggestions he made to me early in the present research as to the best line of attack; these have been of the greatest assistance. The Phenomenon of “ Narcosis Progression” in Mammals.* By T. Granam Brown (Carnegie Fellow). (Communicated by Prof. C.S. Sherrington, F.R.S. Received November 12, 1912,— Read January 23, 1913.) (From the Physiological Laboratory of the University of Liverpool.) CONTENTS. PAGE L.. Introduction: .. ccs cscsccdscusccesssedccsscwer cesses os cdc cisettsaSule cuales niceties samenieneites 140 LT. Methods: Em ployed inc. cccscccctseet» selec sennisecaecsstaaceannactvecsensescnecsenmeeeeetns 142 III. Narcosis Progression in Guinea-pigs ............:ssscseccsceseeeeceeecserersansensenees 142 IV. Narcosis Progression in Cats: Intact Hind Limbs, Bilateral Progression ... 143 VY. Narcosis Progression in Cats: Intact Hind Limbs, Unilateral Progression... 146 VI. Narcosis Progression in Cats: The Effect of some Lesions of the Nervous eh, 2) 1 ROBES Scoricod050 3550050000000 00 ce boo Da HOOD Hao bUcdoHodadeasabcuosoddoodQonsOsqN 148 VII. Narcosis Progression in Cats: In Individual Muscles ...............:ssecsseeeeeees 150 VIII. Narcosis Progression in Cats: The Effect of Asphyxia .............:00.sseceeeeee 151 TX. Conclusions (1.7.2 ..scneesdescee- usseaeruattennecortrrpasciesssatodd seeker erecea tt ne eee 158 B-GM= 1151 Ales oseeerotsd sacene soe oadosiudD 6 Suton usr aboodoneneoostonsonouonapssadesedodd soo sc000054 162 I. Introduction. At present the subject of “ reflex walking ” is one almost entirely neglected by physiologists, yet it is of fundamental importance in the physiology of the nervous system. For this reason it is desired to lay stress upon it by means of the present paper, which is to be regarded as a prelimi- nary account of a phenomenon which will later be described in greater detail. The present author has already described certain movements of progression * The expenses of this research have been defrayed by a grant from the Carnegie Trust. The results here described were embodied in a thesis presented to the University of Edinburgh on March 31, 1912. ; 1912. | “ Narcosis Progression” 7% Mammals. 141 which occur in rabbits while subjected to the state of ether or chloroform narcosis.* These movements are of interest in that they exactly resemble the peculiar form of locomotion in that animal—simultaneous movements of the hind limbs (“hopping”) and alternate movements of the fore limbs. An additional point of interest is that scratching movements may occur also in narcosis, and the phenomena of narcosis movements may slide, as it were, from the one type into the other. In previous papers read before the Society, mention has been made of the fact that similar movements (that is, of progression) may be induced under narcosis in the guinea-pig} and in the cat.{ The movements in the guinea-pig are of interest because they do not occur under normal conditions of ether or chloroform narcosis—at any rate, they have not been observed in a long series of experiments, although some- times after an operation and when the animal is recovering from the narcotic it may make such movements. Under normal conditions the movements which occur in the guinea-pig under narcosis are those of scratching—the “ narcosis scratch.”§ In the cat scratching has not been seen by me to occur in narcosis amongst about 200-300 individuals which I have observed in this state. On the other hand, movements of progression occur with great frequence, probably in about 50-75 per cent. of cats subjected to general chemical narcosis induced by ether or chloroform or by a mixture of the two. In the cat the progression movements are best seen in the hind limbs, where they may appear when the fore limbs exhibit no movement. They exactly resemble the movements of progression in the normal animal. Some- times the two hind limbs move simultaneously (or nearly simultaneously) in the same direction, but more usually the narcosis progression is alternate. It may rarely be confined to one of the hind limbs. The rhythmic phenomenon may be transient, when it appears then lasting but a few seconds. In other cases it may last for long periods of time. Records in which the movements have continued unchanged for several minutes have been obtained. In such cases the narcosis must be kept at a constant level. If the depth be increased the movements decline in extent and finally die out. If the depth of narcosis be decreased the movements cease somewhat suddenly when in full strength. * ‘Quart. Journ. Exper. Physiol.,’ 1911, vol. 4, p. 151. t ‘Roy. Soc. Proc.,’ 1912, B, vol. 84, p. 555. { ibid., 1912, B, vol. 85, p. 278. ; § ‘Journ. Physiol.’ (‘ Proc. Phys. Soc.’), 1909, vol. 38, p. 86; ‘Quart. Journ. Exper. Physiol.,’ 1910, vol. 3, p. 21; zb¢d., 1911, vol. 4, p. 19; é#d., 1911, vol. 4, p. 151. M 2 142 Mr. Graham Brown. [Nov. 12, Of the three chief joints of the hind limbs the ankle joint is that at which the movements of narcosis progression in the cat are best seen. Where the phenomenon is well marked it may appear at all three joints; intermediate grades may be characterised by movement at knee and ankle; where slight, movement only at the ankle may be observed. Even where there is no obvious movement there it is sometimes possible to feel rhythmic “beats ” of tibialis anticus by palpation of its tendon of insertion in front of the ankle. Although .these movements of narcosis progression must have been observed in the cat by many previous investigators, to the best of my knowledge they have not before been minutely examined and described. They are of interest on account of the light which they throw upon the phenomenon of progression in comparison with the various other rhythmic phenomena which may be observed. Such rhythmic phenomena may be seen in the scratch reflex, in simple reflexes, in compounded reflexes, as reflex’ rebound, in response to central stimulation of the cut surface of the spinal cord, in the progression phenomenon which may follow a rapid division of the spinal cord, and in the narcosis movements here described. In some experiments several of these forms of rhythmic phenomena have been obtained. Il. Methods Employed. In the guinea-pig the movements of the intact hind limbs have been registered upon the moving surface of the kymograph through the mediation of a pair of levers, which were connected to the hind limbs by means of threads. During the taking of a record the animal was laid upon its back, being sustained with its long axis parallel to the table upon which it rested. When the movements of narcosis progression in the intact hind limbs of the cat were registered, the animal was placed prone upon the table. A hot-water bottle was placed under the lower part of the abdomen and had the effect of raising the pelvis.. A steel bar (parallel to the table at a distance of about 5 cm. and at right angles to the threads connecting the toes to the recording levers) was placed under the ankles. The movements then registered were almost only those at the two ankle joints. The.movements at the ankles have been observed after various operative procedures. Thus they have been examined after decerebration of the animal by rapid division of the brain stem through the anterior colliculi; after rapid division of the spinal cord in the region of the lower thoracic segments; after de-afferentation of one hind limb by the division of the posterior spinal roots proper to it ; after motor paralysis of groups of muscles in the two hind limbs ; and, finally, in two individual muscles (gastrocnemius and tbialis anticus—antagonists at the ankle joint) after motor paralysis of all the other muscles of both hind limbs. III. Narcosis Progression in Guinea-Prgs. In the normal condition under narcosis the rhythmic movements which occur are those of the scratch. On an attempt being made to induce these after local anzsthesia of the receptive skin fields for the scratch-reflex 1912.] ‘““ Narcosis Progression” in Mammals. 143 by means of the subcutaneous injection of “novocain” (0°1-0'2 grm.), it was found that movements of progression occurred, and that scratching could not be obtained. This accidental discovery was confirmed in other cases, and it must appear that the application of this drug causes the narcosis movements of the guinea-pig to change from those of the scratch to those of progression. This result is not conditioned by the site of the application of the drug, for it is obtained after intra-peritoneal injection. In cats in which narcosis progression has not occurred, an injection of “novocain” may sometimes be followed by their appearance. When these movements occur in the guinea-pig they are bilaterally alternate in the two hind limbs. Their amplitude may be greatly increased during the application of a peripheral stimulus. Such an augmentation, for instance, occurs if mechanical pressure be applied to the fold of skin which passes from the abdomen to the front of the thigh. In this case the amplitude of the “beats” is increased in the limb of the same side and diminished in that of the opposite side. The usual rate of rhythm of the movements is about 2:0 beats per second in each hind limb (two complete cycles of progression per second). This has been remarkably regular. In one and the same individual variations of rate of rhythm between the extremes of 1°75 and 3:0 beats per second have, however, been observed. In this same individual the movements of the narcosis scratch were recorded on another occasion. The rate of rhythm of the beats of the scratch varied between 7:0 and 8-0 beats per second. This rate corresponds very closely with the average rate of 7°5 beats per second obtained from a large number of different individuals.* It seems to bear the simple relationship of 4:1 to the rate of the movements of progression in the same individual. If this is the case it is of interest that the relationship should be 2:1 in the case of the rabbit+ (where the progression is that of hopping), and of 4:1 in the guinea-pig (where the progression is that of running). IV. Narcosis Progression in Cats: Intact Hind Limbs, Bilateral Progression. The movements in the intact hind limbs, as studied at the ankle, are of flexion followed by extension—rhythmically repeated. Perhaps it would be more correct to say that the movements are of flexion followed by relaxation of flexion—because palpation of the extensors usually fails to detect any movement. There is probably a reciprocal relaxation during * ‘Quart. Journ. Exper. Physiol.,’ 1910, vol. 3, p. 21. + Ibed., 1911, vol. 2, p. 151. 144 Mr. Graham Brown. [Nov. 12, she flexor contraction, and a restitution of maintained tonus during flexor relaxation, but there is no evident contraction in this phase. The movements occur in the two limbs. They are then related in such a manner that flexion at one ankle occurs during relaxation of flexion at the other ; and when the second limb exhibits flexion the first relaxes. It happens (although rarely) that sometimes the movements in the two hind limbs are nearly synchronous, in the sense of being in the same direction at the same time. Graphic records of the movements of the two ankles during this narcosis progression demonstrate the relations of the movements in the two hind limbs, and also the variations of their rates of rhythm. In a typical instance the curve traced by each foot is rhythmically discontinuous (fig. 1). Each remains for intervals parallel with the abscissa — corresponding with a posture of extension. Between these intervals each curve describes a sharp rise and fall(+ and — flexion). This movement may be termed a “beat”; and there is usually no pause at the top of a beat. The movement of relaxation of flexion (or of extension) immediately succeeds that of flexion contraction. Occasionally there is a slight pause at the top, and the curve then remains parallel with the abscissa in a posture of maintained flexion. The fall of a beat is usually a more rapid movement than the rise. When the tracings of the two ankle movements are compared it is found that the beat of one falls within the pause of the other. The exact relation- ship depends upon the rate of rhythm of the movements. Where the movements are slow the intervals may be of greater duration than the beats. In such a case the top of a beat in one foot corresponds in time to the mid-point of an interval in the other. The end of an interval in one foot then overlaps the beginning of one in the other. Immediately after each beat there is a short period during which both feet are in a state of maintained extension. This is then terminated by the appearance of the following beat at the other ankle. If the rate of rhythm is faster and the duration of the pauses exactly equal to the duration of the beats then the termination of a beat at one ankle is immediately followed by the commencement of a beat at the other; the termination of that by the commencement of a beat at the first—and so on. With still faster rates the interval between beats at either ankle becomes shortened. The apex of a beat at one ankle then falls mid-way in this pause, but the commencement of the beat occurs at a point during the relaxation phase of a beat at the other ankle. The relaxation phase of the 1912.] “ Narcosis Progression” in Mammals. 145 first beat then synchronises at a certain point with the point of commence- ment of a beat at the second—and so on. Increase in the rate of rhythm is then accompanied by a disappearance of Fie. 1.—Experiment, C, XXXIV, record 60; 11.4.11.—Normal cat, record of narcosis progression obtained by registration of the movements at the ankles of the intact hind limbs. The movements are here slow. It will be observed that there are distinct pauses in flexion relaxation between the beats, and that there are slight pauses at the top of the beats. The pauses in relaxation are of greater duration than are the beats. The beats are alternate in the two hind limbs. The corresponding ordinates marked 1, 2, 3, 4 demonstrate that the commencement of a beat at the left ankle occurs after the commencement of a pause between beats at the right, and that the termination of that beat occurs before the termination of the same pause. Thus, for instance, immediately after ordinates 4 both limbs for a short time are in flexion relaxation. In this and in all the other figures, except figs. 7 and 8, the upper tracing is that obtained from the movements of the right foot and the lower that from the left. The rise of the curve denotes flexion, and the fall extension at the ankle. Corre- sponding ordinates on the two tracings (usually numbered 1, 2, 3, etc.) demonstrate the time relations of different points. A millimetre scale has been drawn before the record was varnished, and is thus reduced in proportion with the rest of the record. The lowest line registers time in seconds. the pause between the beats of either foot. At each ankle the commence- ment of a beat follows immediately upon the termination of the preceding 146 Mr. Graham Brown. [Nov. 12, beat. The apex of a beat at one ankle is then found to correspond in time with the point of transition from beat to beat at the other. Where the movements of narcosis progression are fast 1t may hase that they are no longer bilaterally alternate at the two ankles, but are more or less exactly synchronous. The progression is then that of the gallop. The rate of rhythm in different experiments has been found to vary between comparatively wide limits. Thus a rate as slow as 0°6 cycle per second has been registered. (A cycle may be measured by taking the average duration of time between successive apices of the beats at one ankle.) Rates as fast as 25 cycles per second have been registered under normal conditions; while under asphyxia in addition to the narcosis the rate has been found to rise as high as 3°3 beats per second—or even higher. Even in one and the same individual the rate may vary considerably. Thus on two successive days in one individual rates of 0:6 and 2 cycles per second have been registered. Nine days later the rhythm was 1 cycle per second. . The movements are not always regular. Sometimes “grouping” may occur. The beats may then occur in pairs with shorter pauses between the elements of a pair than between successive pairs. Other variations may occur, and sometimes there may be a dropping out of the movements of a limb for short intervals of time during the registration of a long record. V. Narcosis Progression in Cats: Intact Hind Limbs, Unilateral Progression. Unilateral progression as regards the pair of hind (or of fore) limbs is of course a phenomenon often seen in the case of the mammal which uses -quadrupedal progression. Three-leg progression in the dog after injury to one limb is a common sight. It occasionally happens that the phenomena of narcosis progression are entirely confined to one hind limb; while it more often happens that for short periods of time a bilateral progression in the hind limbs becomes unilateral. In the forms of unilateral progression in which the phenomenon is entirely confined to one of the pair of hind limbs there is little to describe. The rate of the movements is almost always very fast, and corresponds to that which obtains in bilateral progression when the movements are synchronous in the two limbs (galloping). Temporary abolition of the movements of progression in one limb of the pair may take place during long records of bilaterally alternate movements (fig. 2). | This abolition occurs suddenly. The beats in one limb fail, and there ensues a shorter or a longer pause during which the beats occur alone in the 147 1 Mammals. vi ” JV essvoi Pro Ss « Narecosi 1912.] ‘(uoustmousyd [etaze[LUN ey Sutpeoedd AToyvrpeumurt § sepAd Oy OY ‘[e109qRIIq) CE ‘OD “Gf SeyvUTPAO YALA poaeduioo se (sajoAo Z = ‘[erogeyiun) Ff “a “q seqeurpaz0 4e Quy pully 9J9] 94} UL poyteUL sou st UOUetUOUeYd ous oy, “(SuLAOM oe Sqm] PULY YIOq etayA) LE “cg SoyVUIpIO uBEAjeq yey} ULYy 10zLOYS st (seppAo G =) Th pur GE SoyvuIpao UaEAMG0q eoUvISIp eY2 * Lh ‘OP “6g SeqvUTp1O ye quUITT FYSTA oY} Jo osko ay} UL UdEs oq AvU SIYT, ‘etojoq wey reyomb ore oaowt 0} sonuiU0Dd ToryAs QUAI] Oq9 Ul syvoq O44 “etoy SIN000 Sig] UEy AA “[RAA}LTIUN oUMODEq 09 STeAde}UT 4104S Of pu} syreq ayy Surova, oy} JO PUS OY} SPAVAOF 4VYF POATOSYO oq OSTL TIM 4] ‘SooURZStINOIID osoy} 1opuN uOUsUIOUEY a1} JO WOTY"VSSeD UOppns O19 Jo yeoid 44 st Surovay sty, ‘poddoys Ayueppns Loy} [UN zoSael acurede oq AT[enpras Aoyy posvedoep sea sisooreU jo yydop oy} sw epee PTOY SOUO ISL] OI} JO 4U9}XO oY} JO G.0 JO JUOMOOUETIIUOD 944 4% a10M SJUOTMOAOUT oT, "Spuodes QOL Jo porsed BLOF poplooad BOA SYMOMPAOTI Of} YOIYA UL Su1oeV1y wv Jo qaved yse_ oy} soars AMSY SIT, “WoIssersord stsooavu jo paooea ‘(arogaq SAv P OT peareyqo svar | ‘DY OYA WLO.TF OMLO oULeS 9173) yo [PUMLON—"TL'F'1G ‘69 plover ‘ATX XX ‘O Wuouts0edxy-—"z ‘org lp Or 6CH Dec 4 3 qecogcase vibe zee ze x See RrES rt SNS RN an oo " WHE ZEC,4A 2c |x MICO 6292 Hil 148 Mr. Graham Brown. [Nov. 12, other limb. Reappearance of the beats in the first limb then terminates this pause; but these beats are at first smaller than usual. The common extent of the beats is soon attained. The duration of the long pause is approximately equivalent to that of so many cycles. That is, it is approximately equal in duration to a simple multiple of the duration of a single cycle when the two limbs are acting together alternately. This correspondence is, however, only approximate ; for minuter examination of the records reveals the fact that the pause is in duration smaller than would be this simple multiple. The beats which continue in being in the other limb become more rapid in rate of rhythm than before. The change of rate of rhythm in such temporary unilateral progression may be slight. It may change from 0°75 to 0-9 cycle per second only. In other cases it may change from 1°5 to 2°5 cycles per second. The beats of a limb in which the movements occur during the temporary abolition of the beats in the other are often increased in extent as well as in rate of rhythm. This increased extent may persist for a short time after the resumption of movement by the other limb. It then gradually disappears synchronously with the attainment by these beats of their normal extent. VI. Narcosis Progression in Cats: the Effect of some Lesions of the Nervous System. The movements of progression in the intact hind limbs may be recorded —as they occur at the ankles—after certain lesions of the central and peripheral nervous systems. Of these, the first which we may mention is that of decerebration by rapid division of the brain stem through the anterior colliculi. In one experiment the movements of narcosis progression were recorded before decerebration, and also 15 seconds after the infliction of the lesion. Before decerebration the movements were very well marked, and were alternate in the two hind limbs. After decerebration they were very much reduced in extent, the reduction being greater in one limb than in the other. The movements were synchronous in the two limbs at the beginning of the record, but later they became alternate. The rate of rhythm also changed after decerebration. Before it had been about 1-2 cycles per second. After- wards it became about 2°6 cycles per second. The movements entirely disappeared 40 seconds after decerebration. The movements of narcosis progression in the hind limbs may persist after local injury to the lumbar region of the spinal cord. Thus, after the movements of normal narcosis progression have Been recorded, the lumbar spinal cord may be transected at the level of the entry 1912. ] “ Narcosis Progression” in Mammals. 149 of the most caudal fibres of the VIth post-thoracic posterior spinal root without abolishing the narcosis progression. This lesion, if accompanied by the destruction of the lower part of the cut cord, removes from the body the centres for the extensors of the ankle joint, and leaves that for the flexors (or part of it). Yet the movements at the two ankles may still persist. Again, in the same experiment, the remaining portion of the lumbar spinal cord has been split in the middle line of the body in the VIth, Vth, and lower part of the [Vth post-thoracic segments, and the left part of the cord in these segments has been removed. Movements of narcosis progression persisted in the right hind limb at the right ankle. They were very slight, and it was almost impossible to register them, but palpation of the tendon of tibialis anticus revealed the fact that that muscle continued rhythmically to contract, and it appeared that the rate of the movements was about twice as fast as before. The movements of progression in this experiment before the first lesion (in the normal condition) had been of good extent and of a rate of about 0°95 cycle per second. After the first lesion the extent of the beats was reduced, but their rate remained at about 1 cycle per second. After the second lesion the extent of the beats was very markedly reduced, and the rate of rhythm appeared to be about 2 cycles per second. At the time of registration the beats appeared to be of a rate of about 1 cycle per second— but possibly of about 1°3 cycles per second. Division of the posterior spinal roots of one of the two limbs in another experiment was followed by abolition of the narcosis movements as they occurred in that hind limb. The movements in the other limb continued, but were slower than before. Only one experiment of this sort has been successful. It is not usual for the movements of narcosis pro- gression to last as long after the commencement of narcosis as is necessary for the preparation of the spinal cord and roots before they are cut. It is, perhaps, remarkable that the movements should survive the procedure in any case. It is but fair to add that in this successful experiment the narcosis progression had at the commencement of narcosis showed a marked tendency to be unilateral in either hind limb, and especially in that the posterior roots of which were not cut. ' In another case motor paralysis of one hind limb was produced by division of all its spinal roots—both motor and afferent. In this instance the movements of narcosis progression persisted in the other hind limb, but were then more slow than before. In yet another experiment the movements of narcosis progression at one ankle survived not only the motor paralysis of all the muscles save the knee 150 Mr. Graham Brown. [Nov. 12, extensors of the other hind limb, but also the motor paralysis of all the muscles (save the knee extensors) acting upon its own hip and knee joints. The movements at this ankle still survived after transection of the lumbar spinal cord at the level of the lower border of the VIIth post-thoracic segment, and also after division at the lower border of the VIth segment. Finally, I have lately divided the spinal cord in the lower thoracic region very rapidly while the narcosis progression (in deep anesthesia) was in being. In this instance the movements were not recorded, but there appeared to be little change in them. They ceased about 30 seconds after the lesion.* VIL. Narcosis Progression in Cats: in Individual Muscles, When the individual antagonists at the ankle (tibialis anticus and gastrocnemius) are prepared for the registration of their movements all the muscles of the other hind limb are put out of action by motor paralysis, while all the other muscles of the same hind limb are similarly paralysed. After this drastic procedure it is not strange that the movements of progres- sion narcosis are difficult to obtain—especially as they soon tend to disappear in the normal cat as the narcosis is continued for‘any great length of time. The movements of progression narcosis have, however, been recorded in the individual muscles, and in an experiment in which they were also recorded in the intact hind limbs. In the intact hind limbs the movements were of good extent, regular, and of a rate of about 1:6 cycles per second. They were recorded for a period of about 90 seconds. After preparation of the individual muscles at the ankle the movements persisted and were very well marked. The extensor—gastrocnemius— throughout exhibited no trace of movement; but the flexor—tibialis anticus— presented a record composed of well marked beats, regular in extent and rate of rhythm. These were composed of contraction phases immediately succeeded at the summit of contraction by relaxation. The termination of the phase of relaxation was succeeded by a pause in which the muscle remained in relaxation, and then contraction again appeared (fig. 7). The rate of rhythm of these beats was about 1°5 cycles per second, so that here the motor paralysis of one hind limb and the motor paralysis of all the * Note (added January 23, 1913).—I have lately repeated this result, with graphic registration. The narcosis progression was occurring at a depth of anesthesia so great that rapid division of the lower thoracic spinal cord evoked scarce a movement. Again, I have found narcosis progression to occur at a depth of anzesthesia at which reflex movement at the ankle could hardly be elicited. To these recent observations I would like to add a third : narcosis progression may persist through the procedure of decapita- tion, and thereafter may suddenly merge, in one hind limb, into the scratch-reflex. 1912. | “ Narcosis Progression” in Mammals. 151 muscles of the other save only the two retained had no apparent effect upon the rate of rhythm of the movements. The general appearance of this tracing was such that to inspection it might easily be mistaken for one in which the movements of the intact hind limb were recorded. \ aa hil Sear ey \ \ AHI ‘ie UY 423 C.£. RATHER VEEP MARKED ASPHYXTIA Fic. 3.—Experiment C, LXII, record 110; 19.7.11.—Normal cat, narcosis progression under the influence of chloroform-ether narcosis reinforced by the application of 0°2 grm. “ novocain.” Effect of asphyxia—trachea closed 1-2 seconds before ordinates 1. In the left limb the beats are seen for a time to continue to relax to the same level, but to become of increased extent and rate of rhythm. Then the relaxation becomes less complete, and as this becomes still less the beats decrease in extent, although their height does not fall. In the left limb the beats do not disappear, but in the right they disappear and a state of maintained flexion is left. In the left limb the pauses between the beats are seen to disappear. The beats in the two limbs are at first alternate (as at ordinates 1, 2, 3), but when they are of maximum extent in the left limb they are very nearly synchronous. Later they are again alternate. VIII. Narcosis Progression in Cats: the Effect of Asphyxia. In the cat under narcosis the effect of asphyxia may easily be studied by completely closing the elastic trachea (or by closing a tracheal cannula when that has been inserted). In these experiments this complete closure was applied for short periods—about 14 seconds to about 50 seconds (maximum). An asphyxia of about 30 seconds is sufficient to give the effects to be described. These have been examined both in the intact hind limbs and in the individual antagonists at the ankle. The records demonstrate many minor variations in the phenomenon. The description here is that of the typical effect. 152 Mr. Graham Brown. [Nov. 12, A short period of complete asphyxia produces three kinds of effect upon movements of narcosis progression then in being. These are: Change in the rate of rhythm of the beats; change in the extent of the beats: and change in the mutual relations of the beats in the two limbs. Change in Rate of Rhythm—Two main phases occur between the com- mencement of asphyxia and the attainment of its complete effect. The first of these is a slowing of the rate of rhythm. This may be absent, or it may be so great as to cause complete cessation of the beats. It seems to be conditioned by an increase in the pauses between the beats. In extent of duration it is usually slight; a rate of rhythm before closure of the trachea of about 1:3 cycles per second may become one of about 1 cycle. This phase lasts for about 9-12 seconds, at the end of which time the rate of rhythm again increases (or the beats reappear if they have been suppressed). The phase of the increase of rate of rhythm is the second. It is never absent unless the beats at the commencement of asphyxia are already very fast. The rate of rhythm becomes progressively more fast, at first by a reduction of the pauses between beats, but later (when the pauses have disappeared) by a reduction of the duration of the beats. This phase may last for as long as 20 seconds, and the rate of rhythm may become thrice that which obtained before the application of asphyxia. Thus a rate before asphyxia of 1:3 cycles per second has been observed to change to one of 3 cycles per second 25 seconds after closure of the trachea; and one of 1 cycle has changed to 2°75. The beats themselves may shrink in duration to 0°75 of the duration before asphyxia. Change in Extent—Synchronously with these changes in rate of rhythm, changes in the extent of the beats may appear. This change may be measured either by the examination of the heights of the beats (that is, of the heights of the apices of maximum flexion); or by an examination of their lengths (that is, of the distances between the apices and the lowest points in the beats). If the apices of the beats in an asphyxia record be joined by an imaginary line, and if the lowest points be joined all together by another, the beats will appear as conditioned in height at any one point by the distance between the two curves there. The first curve—which is that of maximum flexion— is parallel with the abscissa at the time of the application of the asphyxia. It then sometimes slightly falls (the fall may, of course, be great—as when the beats disappear in the first phase). This fall is soon succeeded by a gradual rise of the curve, and it is most common for this gradual rise to commence at once and without preceding fall. Throughout the phase in which the rhythm slows the curve of maximum flexion gradually rises, and it continues to rise throughout the whole phenomenon until the complete 153 Mammals. “wn 27 TESSLON s Proc PCOSVUS “ Na 1912. ] “Bsslosqe oy} Suoye yISus] ul (a[s peonped) ‘wu g = puooes | : polley TayxIeU stUa1y ayy 19 FT ‘Quy pury 4ys11 oy ur ATor14U0 aeeddesip jou Op szeeq oya pur ‘pexreor ATpeq si UoIxey poureyureM jo o9R48 oy) ‘snouoryouds AT[B10}e/I1q eWODeq Jou op sxveq oY} ‘1eaddestp jou op sasned 049 ynq ‘oqe1 pue 4ue7xe ur osverour ATpeyreur syvoq 04g ‘ Suoy Ajaayereduros st yooya erxAydse 044 Jo Aouayzey 04 eo] EGF P2oigou oq [IIA 4] “szeeq JYSt1 oy Jo quaMedUeMTIMOD Jo quod ayy sFuIORIy qyOq wo YALE “ZP—-] porequinu ‘seyeuIpso 10440 aut (4 %) & ye paddoys pue (a ‘wv SeqyeUIpdo) oul] [eUsIs 943 JO v 4e peoueUIUIOD sea (eayort4 ey Jo eansojo o40[dm00 fq) uxdydsy “Risoyseue 19Yy49-WNA0JOI1O[YD Japun uorsseaZoad SISOoLeU “9VO TRUMON— TILE {eel ploet TITAX XT ‘9 quewmltedxy—f “oly WIHIVYL ONISOTD AG -A-X —~WIYAHASY - OLLI be 9€ tf Ok KE AUI6e L 9 §F_ #& i 154 Mr. Graham Brown. [Nov. 12, asphyxia effect is obtained. Towards the end of the second phase the rise becomes much less rapid than at its commencement, and the curve finally becomes again parallel to the abscissa as the maximum effect is reached. The curve of minimum flexion—that joining the lowest points in the beats— may also fall in the first phase. If it does so its fall is less than that of the curve of maximum flexion, and more usually it remains parallel with the abscissa throughout the first phase. At the end of that phase, and when the beats are again increasing in rate of rhythm, this curve commences to rise more rapidly than that of maximum flexion. This rise continues to be more rapid until the maximum asphyxia effect is attained. In fact, both curves continue to rise throughout the phenomenon. That of maximum flexion has a shorter latency and a more gradual rise than the curve of minimum flexion, which has a longer latency and a more rapid rise. In consequence of this the beats are of greatest relative extent (that is, from apex to their lowest point) at the end of the first phase when the curve of minimum flexion is just about to rise, and thenceforward they progressively diminish in extent—at the same time becoming increasingly more rapid. The maximum effect is attained when the curve of minimum flexion coincides with that of maximum flexion. Just before this is attained the beats are very small in extent and very fast. When it is attained they disappear and there is left in their place a state of maintained flexion. This maintained flexion may continue to increase for a short time after this. Change in Bilateral Relations—The temporal relations of the movements in the two hind limbs are of interest during this phenomenon of asphyxia. At the commencement of the condition it may be supposed that the beats in the two hind limbs are accurately alternate. In the first phase of the asphyxia phenomenon, when the beats slow in rate and increase in extent, this relationship persists. Thereafter a change makes its appearance in the relationship of the beats on the two sides of the body. The apex of the beat on one side at this point falls midway between the apices of two beats on the other. In successive beats the apex then leaves the mid-point and either advances or retires towards the first of the two apices of the other limb between which it falls (fig. 5).° This process is continued until the apex of a beat of one limb actually coincides with that of a beat in the other. The movement is then that of the gallop—synchronous beats in the two hind limbs. This state often persists for the remainder of the period of asphyxia, but sometimes the beats become again less completely synchronous. This phenomenon of progression in asphyxia, if it is induced in one of the rare cases in which the progression is already synchronous, may not interfere with that synchronism. The beats may become slightly more fast. 4 155 s Mammal an L ? gression 1s Pro COSUS “ Nar 1912.] TOHCUTUNIOZ PUB FUSMIEdUEUIMTOD snouoryoUAs r1043 Lq MMOYs sI se ‘SqUIT] puty omg 049 ut snouoryouds ome s7¥9q 94} EF SeyeUIPIO ynoge ye [MWUN seSueyo Ayyonpoi6 diysuoyejer e134 TojfeotoG], “oyeutezy ore syveq 049 (6T SeqzeUIpso *119) Qual] 44811 04} Jo szeeq oy} usemgeq woz sesned jo souervaddesip 243 Jo emt} OU} FW “quuIT 2624 949 UL sy%eq Jo yuemMeoUETUUOD eu9 YIM puodserioo [Te yoy ‘seyeUtpso peroqrint aq} Jo suolgejar ayy Aq pozerqsuouep sit pue ‘poyreu A[1ee]9 4ysour st uouetmoueyd yse] ony, “‘UustuoIgouds [ele9ze][1q 07 SqUIT[ Omg ayy Ut Syeoq OY} Jo UOIEUIE;[B [e1eqeIIq WoIZ asueYO (9) pue { (sosto autos UI SB poyreUr [fom OS 10 4018 OS JOU) UOTxay pourezureut Jo 94e48 e (Gg) Ul SuIsr0U ‘syveq oY4 JO JU9}X9 OF UI osvotoep ayer (F) ‘$j¥eq U9eM40q sosned 94} Jo sourreeddestp (¢) { wy4tyat Jo aye1 Tray} Ul eseedour (Z) ¢ squuiT YI0q UT syveq B44 Jo 4UE}xe 9U} UL aSvaIOUT 944 (T) soyeAysuOTEp prodded SIU, “erxXydse oy9 Jo quatteouemtuOD ayy STOFOq US ST QUIT PUIY 4J9T @Y} UT syeaq oY Jo ao1gvorjdnpoa Snonms W "A-X weemjeq peardde sea eixdydsy [49430309 peqyy A[[NJorvo Weeq Sel{ JI : eAoqe G-g pu Moraq 1-9 SOJ€UIpIO U9eeM40q euIT Teuoserp & url Surysturea ut utoq sem aaded eyL] ‘woisserSo1d stsooreu “R09 (UMON—TEIL9T | OFLNpt0004 “ATX KX XT ‘9 quewttedxay—¢ ‘org VIHIVYL JO ONISOID FLFIAWOD VixXAHASV Ex Ze 9f S€ EC IE 62 42 Se E2 le 6/ dz) SI &/ a 4 oO eel ee 9 \ wh Fryer, o .o ° ZE 9€ S€ € 1€ 62 £252 €2 12 «6 LIN Srey Giared, / Se HRUALLO Waa B. VOL. LXXXVI. [Nov. 12, Mr. Graham Brown. 156 ‘sisoozeu Jo Yydep ey} Jo Suruesse] @ 07 onp oq Avur yavd ur SIG, -erxhydse o1ojoq Suiuulseq ey} 48 UBYy 1ed1e] PUL TAAO[S TORU av 8yvoq ot} PLoder ey} JO PUES OY} FW “SenUTATOD ss9001d 9} 88 IAMOTS PUB JedIB[ GUIODEG §}¥eq 44 PUY SedTeTIUIOD Araooea [RoI Vay], ‘S7eq GATJLoGe yswy PUL [TBS Jo SolTes JLOYS ysiy oT OLY, -eixAydse tosy AAA0IeI OY} SO}BIYSUOMIEp Sulowsy sty, ‘pider pue [[eus cae syvoq oy} pouTeyge SI Jooyo [[NF OY e1oFor qsn{ pue ‘sopsnur 430q UL pexTVUL St UOIXeY peurequreUt jo 04e48 oy “erxdydse yy Jo uonrorddy oy} e1oyoq ueae snouoryouds ATAvou pue wmygAyx Jo ayer ur pider ore szvoq OYJ, “peonput st vrxdqdse uegM 4U9}x9 UI SUISva1OAp 91B SJUSUIOAOW OY} OOF] “WOISsetcoId sisooreU ‘4e0 [VUION-—"1T SST £ ge] paooed “TT AOX ‘O quewliedxy—9 ‘O14 LHINY til 21 2! gor 901 vo! 2204999949 290985 ANN ez: OZ! an on, on en 91! Bor 901, wor 20100! G6 96 bE 2696 99 op pp 2208 By seve ee Of 1912. | « Narcosis Progression” 1% Mammals. av It sometimes occurs that, if the movements of narcosis progression have occurred in a cat but have then ceased, the production of a state of asphyxia may induce them again. They then appear some time after the commence- ment of asphyxia at that point at whieh it might be expected that the beats would be reaching their maximum extent. Recovery.—The asphyxia effect ends in the production of a state of maintained flexion. If the’ asphyxia be continued there may then be no recovery of the movements of progression. But if the asphyxia be terminated whenever the complete effect is attained—or, better still, a few seconds before it is attained—the movements of progression again make their appearance (fig. 6). ven when the asphyxia is stopped before the attainment of the complete effect (that is, when the beats are still present) the full effect is attained. The state of maintained flexion then persists for a few seconds; the beats reappear—being then of small extent and rapid rate of rhythm; become of greater extent and of slower rate; and finally again attain their normal appearance. The maintained flexion PROGRESSION IN IND/K MUSCLES Fie. 7.—Experiment C, LX, record 107; 14.7.11.—Normal cat, narcosis progression registered in the isolated tibialis anticus and gastrocnemius muscles after motor paralysis of all the other muscles of both hind limbs. The upper tracing is that of the flexor (tibialis anticus), while the lower is that of the extensor (gastrocnemius). Rise of the curve denotes contraction and fall denotes relaxation of a muscle. It will be observed that here the flexor beats are very well marked, but at this period of the experiment are somewhat irregular in extent and in rate of rhythm. No movement of gastrocnemius is registered. N 2 158 Mr. Graham Brown. [Nov. 12, may, however, persist for long periods, as long as 20 seconds. The beats may then reappear and again become suppressed, again to appear and persist. The beats at first may be more rapid than before the attainment of main- tained flexion. At first they are synchronous in the two hind limbs, but as they slow in rate of rhythm they again become alternate. In records obtained from the individual muscles at the ankle the effects of asphyxia have also been studied (fig. 8). Here the synchronism between the two hind limbs was not investigated, but with this exception the flexor showed the phenomena of asphyxia seen in the case of the intact hind limb. After the commencement of asphyxia there was little or no slowing of the rate of rhythm. The beats at once commenced to increase in height. The maximum was attained just before they began to increase in rate of rhythm. There- after the curve of minimum contraction of the flexor muscle began to rise (that is, the point of minimum contraction occurred at an ever greater level of maintained contraction). At the same time pauses disappeared from between the beats, their rate of rhythm progressively increased, and their extent progressively diminished. Their rate increased from about 1 cycle: per second just before asphyxia, to about 3:4 cycles per second just before the complete effect appeared. The beats absolutely disappeared and left behind a state of maintained contraction of the flexor, which gradually increased. Recovery from the effects of asphyxia has also been observed in the individual flexor muscle (fig. 8). ‘The state of maintained contraction was then broken by groups of abortive beats which were of very small extent. Soon undoubted beats appeared. These were irregular, of slower rate than before; and of smaller extent. The extensor played no part in the phenomena of asphyxia, or of recovery therefrom. Throughout it remained inactive. . IX. Conclusions. That these movements of narcosis progression are strictly equivalent to the normal act of progression there can be no doubt. In the rabbit the progression is almost invariably that of hopping—where the two hind limbs move synchronously in the same directions. In the phenomenon of narcosis progression of the rabbit the movements of the two hind limbs are also synchronous. In the guinea-pig the progression is almost invariably that of bilateral alternation of movement in the two hind limbs. In the narcosis progression of that animal the movements of the two hind limbs have always been alternate. In the cat the movement may either be that of bilateral alternation—as 159 Mammals. pp) he Ww ‘ogression 1s Py 'ONCOSUS 73 Ni 1912.] ‘queqxe jo AquemnBarat euteayxe jo ese Aoyy paaveddvat AT]vat syvoq oy} UAYA yNq “eyVEq GATIAOGL OTIOS JST a1OM TOUT, ‘pervadde Ayyenpeas Aa@aooad pur paysisied sty, “WorjovsqUOD Loxey pouTeyUIeM jo a4yvys Vv ponsue sto} pue perveddesip sjraq oy} Ue pourezqo SM Joaya ojeTdwoo oy} pur ‘asot UOIQeXEIAL Jo [PA9T 94} ‘porvaddestp syveq ayy U9eMJ0q WOVXE[EI OXY UT sesned oyy, “pide a10ur outa £19A PuL TE[NBea ecom auredeq uayy Aeyy, “Ae[nSerat ATOA oO 8}vEq 1OXE] 243 eixAydse Jo FUeMIADUSMAUIOD ET} JO OUIIG 949 FV ‘(oqvurxoidde . xX ,,-,, X ,) erxdydse Jo yooyo oy, “op yur oy} 4B S4SIUOSIU [VNPIAIPUl olf} UL MoIsseaSoad stsoa.1wu Jo plooII—J “Sy se quowiiiadxe oures oy} WoAy YVO [RULION—TL'L'FT * LOT proved “XT “) quewtiedxq—'g oI] a _ (NO/LWe4IdS7Y 4O INIddOLS FLITANOI) Ben NoIsciDCdd 160 Mr. Graham Brown. [Nov. 12, in walking; or of very nearly perfect bilateral synchronism—as in the gallop. Its narcosis progression may take either form. The great part played by the flexor muscle is a point of interest. It must be remembered that the movements of extension are much more affected by chemical narcosis than are those of flexion. It is perhaps because of this that when the individual muscles are examined the flexor alone appears to take part in the movements. It is probable that the extensor does take part in the better marked types of narcosis progression, although it is difficult to shew that it does. That muscle certainly takes part in the phenomena of progression which follow mechanical stimulation of the spinal cord and are extremely like the movements of narcosis progression. It also takes part in many instances of the rhythmic rebound phenomenon, which again is very similar to these two types of movement. But it is equally clear that the phenomenon of narcosis progression may appear when there is no evident contraction of the extensor at the ankle. , Again, the independence of the flexor centre is a point of great interest. The movements of narcosis progression may be present after decerebration. They may persist even after the isolation of the aboral part of the spinal cord by division in the lower thoracic region. The phenomenon of narcosis progression may therefore be conditioned by the lumbar centres alone, although it is probable that the higher centres in the cord and upper parts of the cerebro-spinal axis play an important secondary part in its conditioning. The movements may also occur after the destruction of the lower part of the spinal cord which contains the centres for the extensors of the ankle. But this does not necessarily mean that they would continue if all the extensor centres could be removed from the lumbar spinal cord. In this experiment there yet remained the extensor centres for the thigh. The movements may finally occur after the removal of a great part of one lateral half of the lumbar spinal cord. This all speaks for the independence of the lumbar centres. It is true that in these experiments the movements were not observed after the de-afferentation of a hind limb in that limb. It must, however, be remembered that the movements of progression after division of the spinal cord have appeared in these circumstances, and so have the similar move- ments of rhythmic rebound. Perhaps the afferent proprioceptive impulses, which are almost certainly reinforcing impulses and not essential to the act in the phenomenon of narcosis progression, play a greater part in the depressed state of the lumbar centres which is conditioned by the chemical narcotic. If so, it is of interest that the motor paralysis of a limb (accompanied by its de-afferentation, either actual or virtual) has little effect upon the movements of progression narcosis which occur in the other. 1912. | “ Narcosis Progression” in Manmals. 161 There is no space here to discuss all the points of interest raised by these phenomena, but I should like to mention one. The rhythmic act of pro- gression resembles that of respiration in so complete a degree that it is difficult, if not impossible, to resist the idea that in all essentials they are the same, and conditioned in similar manners by similar mechanisms and by stimuli of similar sources. There is a phenomenon of “voluntary breathing” just as there is one of “voluntary progression.” But the former act tends soon to become ? involuntary; and although progression seems to be more under the influence of the higher centres, yet it, too, tends to become an involuntary act once started. The rhythmic movements of respiration seem essentially to be central; they appear to continue after the abolition of self- engendered impulses. In a similar manner it appears that the movements of progression in the lumbar centres may appear when self-generated proprioceptive impulses are excluded. Yet both respiration and progression are reinforced by a peripheral self-regulative mechanism. The effects of asphyxia upon the two centres are again very similar; for in its effects upon the respiratory centre it first produces an increase in the amplitude of the respiratory movements. A state of maintained inspiration underlies this, the diaphragm may continue contracted to a certain extent even at the end of expiration. With an increase of the expiratory movements the inspiratory movements become small. It looks as if there were here, and up to this stage in the asphyxia phenomena of respiration, a resemblance between the behaviour of the inspiratory centre and the flexor centre in progression. Is it possible that the flexor centre is strictly comparable to the inspiratory centre? That flexion = inspiration, and extension = expiration ? In the case of progression it is again of interest that asphyxia should produce an increase in extent and rate of rhythm of the movements. The act of progression if of sufficient speed itself may cause a certain degree of asphyxia. If this be not too great it will assist rapid movement through the environment. When carried to too great an extent the movements will be retarded, and the state of asphyxia will therefore be lessened. There is thus here possible a nice internal regulation of the speed of progression, and an optimum speed may be set for each resultant of the balance between the local peripheral and central factors, and the higher central and peripheral factors, which all influence the final centres which condition the act. 162 Mr. Graham Brown. [Nov. 12, X. Summary. 1. Movements which seem exactly to resemble those of progression occur in some animals when subjected to general chemical narcosis. 2. In the rabbit these movements are synchronous in sense of direction in the two hind limbs. This corresponds with the normal hopping type of, progression in these animals. 3. In the guinea-pig the movements which normally occur in narcosis are those of the scratch. If, however, the chemical narcosis be combined with novocain the resultant movements are those of progression. They are alternate in the two hind limbs, and thus resemble the movements of ordinary progression in the guinea-pig. 4. In the cat the movements which occur in narcosis are those of progression. 5. If the movements be examined in the two hind limbs they are found usually to be alternate. Rarely they are synchronous under normal cireum- stances—narcosis gallop. 6. If the depth of narcosis be gradually increased the movements of narcosis progression gradually fade out. If the depth of narcosis be decreased the movements progressively increase and then suddenly cease. 7. In either hind limb the movements—as examined at the ankle joint— consist of flexion succeeded by relaxation of flexion (extension), and with a pause in the posture of minimum flexion. Palpation of the tendons at the ankle shows that the flexors are active in this movement. It fails to demonstrate any extensor movement in the intervals of flexor contraction. This may, however, perhaps be present in the phenomenon. 8. The pauses may be long—of greater duration than the flexor beats— or they may be absent. In the latter case beat succeeds beat without intermission. 9. Occasionally in a record the movements may fail in one hind limb. It is then found that there is usually an exaggeration of the movements in the other. The beats become of greater extent and quicker than before. This augmentation gradually disappears if the beats in the other hind limb reappear. These, then, are at first smaller than usual, but soon attain their normal extent. 10, The phenomenon when present has been observed to continue after decerebration. In the lumbar centres it may also outlast rapid division of the spinal cord in the lower thoracic region. In the lumbar flexor centres it may also outlast a removal of the spinal cord aboral from them—in which the ankle extensor centres are cut off. The progression at one ankle may 1912. ] “ Narcosis Progression” in Mammals. 163 also outlast removal of the lumbar cord of the opposite side of the body. The movements of narcosis progression have also been observed in a pair of antagonistic muscles at the ankle-joint after the motor paralysis of all the other muscles of both hind limbs. Immediately after the de-afferentation of one hind limb (by rapid division of the posterior spinal roots) the move- ments of narcosis progression have been found to be present in the normal hind limb, but have not then been present in the de-afferented limb. 11. When examined in a pair of individual antagonists at the ankle joint the movements of narcosis progression are found to be confined to the flexor, and then exactly to resemble the movements at the ankle examined in the intact limb. 12. Asphyxia produced by the complete closure of the trachea for a short period of time (15-40 seconds) produces a change in the movements. This is the same when examined either in the intact hind limb or in the individual flexor at the ankle. The flexor beats at first increase in extent and at the same time slow in rate of rhythm. They may sometimes be decreased in extent for a short time, and may even completely disappear. This phenomenon forms, the first phase. When the beats have attained a maximum (after reappearance if they have previously disappeared) they commence to become quicker, and although they still continue to increase in height their relaxation is less complete than before, and their extent decreases. In this second phase of the asphyxia phenomenon there appears to be an increasing factor of maintained flexion. The beats become still more rapid and still smaller until they finally disappear. There is then left a state of marked maintained flexion (flexor contraction as seen in the flexor muscle). 13. If the movement in the two hind limbs is alternate at the point of commencement of asphyxia this alternation may change to synchronism as the beats become more rapid. This change is a gradual one. The apices of the beats in one hind limb gradually advance in temporal relationship to the apices of the beats in the other from the mid-point between these apices. This advance proceeds gradually until the apex of a beat in one limb synchronises with that of a beat in the other. 14. If the asphyxia be terminated at the point at which the beats disappear and a state of maintained flexion is left that state of maintained flexion may persist for a few or for many seconds, and may then be broken by the reappearance of beats. These then are fast and small, but become slower and larger in the reverse order to that which obtained during the establishment of the full asphyxia effect. 15. In narcosis progression the rate of rhythm is usually one of between 164 Dr. F. W. Edridge-Green. [Nov. 15, 1 and 2 cycles per second. It may be as slow as 0°6 cycle per second; or as fast as 2°5. In asphyxia the rate which obtained at the point of commence- ment may be triplicated before the attainment of the complete effect. Thus a rate of 1 cycle per second may become one of 3:4. In normal narcosis progression the rate of rhythm may vary considerably in the same individual on different occasions. Trichronuc Vision and Anomalous Trichromatism. By F. W. EpripGe-Green, M.D., F.R.C.S., Beit Memorial Research Fellow. (Communicated by Prof. E. H. Starling, F.R.S. Received November 15, 1912,— Read January 23, 1913.) (From the Institute of Physiology, University College, London.) DEFINITIONS. A. Trichromie Vision. The trichromic in my classification of degrees of colour-perception are those who have only three colour sensations—red, green, and violet. They see only three colours in the bright spectrum and describe it as consisting of red, red-green, green, green-violet and violet. They apply the designation red- green to the orange and yellow regions of the spectrum and green-violet to the blue region. There are many degrees and varieties of trichromic vision (1, 2, 3, 4,5). I have classified the colour-perception of individuals as dichromic, trichromic, tetrachromic, pentachromic, hexachromic, and heptachromic. This classifica- tion is made by estimating the number of definite colours seen in a bright. spectrum, and the persons belonging to each class behave in every way as if they possessed the number of colour sensations indicated. On my theory of colour-vision each colour sensation is separate and distinct and not compounded of two or more fundamental colour sensations. For instance, there is the strongest evidence that yellow is a simple sensation (18, 24, 25, 26) and that. spectral yellow light does not excite the red and green sensations. B. Anomalous Trichromatisne. The term anomalous trichromatism is used in the sense of the Young- Helmholtz theory in which all colour sensations are supposed to be made up of different proportions of three fundamental sensations. A trichromat on this theory is therefore a person with normal colour-perception. An 1912.] Trichromic Vision and Anomalous Trichromatism. 165 anomalous trichromat is a person who is supposed to have three fundamental sensations but they have not the same proportions as in the normal-sighted. Those are designated anomalous trichromats who, when making the equation » 670+2535 = 1589, use proportions of red and green different from the normal. At the same time the subjects of this abnormality object to the normal equation. Those who put too much red in the mixed colour are called red-anomalies and those who put too much green in the mixed colour green-anomalies. » Anomalous trichromatism was discovered by Lord Rayleigh (6), who stated that the colour vision is defective only in the sense that it differs from that of the majority. In 1904 Guttmann (8) stated that the anomalous trichromats were colour weak and described a number of symptoms similar to those given by me as associated with trichromic vision. I then examined a number of persons with Rayleigh’s apparatus (14) and found that many colour-blind persons, both dichromic and trichromic, can make a match which agrees in every particular with that of a normal sighted person. T could find no evidence that colour weakness was necessarily associated with anomalous trichromatism. I examined 15 students from Newnham College on the same afternoon, the conditions being precisely the same for each. There was considerable variation in the observations and those who made an anomalous equation in every case strongly objected to the normal match. I could find no evidence of colour-blindness in any of those examined. All saw yellow in the spectrum. Of the 15 examined, five made the normal match exactly and one required slightly more green, the others more red in proportions varying in different cases; there was considerable difference between the two extremes, one requiring nearly twice as much red as the other in the mixed colour. This year Lord Rayleigh kindly lent me his colour-mixing apparatus and I examined 100 women students, 25 belonging to the London County Council training college and 75 to University College. The last 75 were examined in precisely similar conditions. The illumination was incandescent electric light and the equation did not vary from day to day. All were examined with Lord Rayleigh’s colour-mixing apparatus; 51 were examined by some kind of test for colour-blindness, and 36 of these were examined by my lantern. I have designated as “anomalies ” those who, on an average of a number of observations, had a deviation of more than one whole division from the normal and did not agree with the normal equation. The colour-mixing instrument of Rayleigh was arranged so that 0 corresponded to full red and 25 to full green. Then by the laws of double refraction the exact proportions of red 166 Dr. F, W. Edridge-Green. [Nov. 15, and green in any mixture can be ascertained. For instance, 12°73 corresponds to a ratio of intensity 1:061 green/red, and 10°371 to 05829 green/red. The other figures can be easily understood by remembering that a difference of one-tenth of a division corresponds to a difference of about 24 per cent. in the ratio of intensities of red to green when the figures are in the neighbourhood of normal vision. Out of the hundred examined, 86 made the normal equation or within one division on either side of it, 12 were anomalous trichromats, 10 being red- anomalies and 2 being green-anomalies. ‘ Red-anomalies. Green-anomalies. az LS aN Ve 11955 6. 1:2 are! g. 14 re 8 4, 1153} Bt ie Sh 95) “he 195) Oy 1133 5 1158) WOR 183 Two others on an average of five observations appeared as anomalies (one 1°3 red, the other 2°0 green), but, as they both agreed with the normal equation, they do not come under the definition. Excluding the last mentioned, who were to a certain extent colour blind, none of the anomalies were found to be colour defective. Of those who made the normal match 9 were found to be colour defective. No. 1 of the green-anomalies was examined very carefully on three occasions ; there was no evidence of colour-blindness ; she passed my ordinary lantern test and also my triple lantern with ease and accuracy, and saw red and green through small apertures as far as I did. She also passed my bead test. Examination with Spectrometer—Pure yellow was isolated at 5770 to 25882. This is quite normal. The area of greatest luminosity was 5697 to 15795; this is considerably to the green side of the maximum of the normal luminosity curve. She marked out 18 monochromatic divisions in the spectrum. This is the normal number; she also named all the colours red, orange, yellow, green, blue, and violet correctly. I have also examined a large number of men and find that when there is a large mean deviation there is colour weakness. The following case is instructive as an example of a high grade green-anomaly without any trace of colour weakness. The observer was an assistant in the Chemical Laboratory of the Physiological Institute, University College. Rayleigh Apparatus —Shown red and yellow, named them correctly as red and yellow. The mean of seven equations was 17°3, the mean deviation 0:1. The normal equation was 14°5. The mean deviation is very small. 1912.| Trichromic Vision and Anomalous Trichromatism. 167 Strongly objected to the normal equation ; said that the mixed colour was orange, and the simple, yellow. Nagel’s Test and Stilling’s Test.—Passed both these tests with much greater ease’and more rapidly than most normal-sighted persons. My Lantern Test.—Passed easily. The above tests were made in the presence of Prof. Starling and Dr. Homans. Spectrometer.— Region of greatest luminosity......... A 589-A 605 TM PUT eH YVellOW wna scscseveces.< « A 591-A 596°5 _ My yellow region \583-A 590 appeared greenish-yellow to him. This region inclines to orange-yellow to me. Pure blue was A 472-A 476. Pure green, A 510-A 514. Simultaneous contrast was not more marked than normal. Saw red below A 780. The following are the monochromatic regions marked out by him:— Hye Bp. 1 a mee } Called by him Red. 14 m tg } called by him Green-blue. A 62 A 2, a } ” CHEE 15. = } ” » 3. a } ” Orange-yellow. 16. A a } # Blue. A 605 r : Ase } * Yellow. 17. ei a a Deep blue. git Say ¢ Greenish yellow. ,, *47° } t Violetbiue: A590 A 466°5 = ; 6. 1 } ” ” ” 19. ue } ” Blue-violet. 7. x ee) § Yellow-green. 20. us ai } +f) ° A 56 447 , 8. 4 } ue ” 21. i } ” Violet. A558 } dA 435 } 9. 1 23 ” 22. 1 ” ” 10. ‘ hit } ” Sueen. 23. : ir } ” ” ile a ee } $5 Blue-green. 24. , a } A ba A516 A411 12. 1k ” ” 95. 1 } 2 ” 13. : ae } ‘ Green-blue. A407 It will be noticed that the region regarded by the normal-sighted as orange-yellow is named and seen by him as greenish yellow. This gives an explanation of the anomalous trichromatism. If the region to be matched appears greener than usual, it will obviously require more green and less red in the mixed colour. These were the results of single observations; the available time would not admit of more and they clearly confirm the other tests. 168 Dr. F. W. Edridge-Green. [Nov. 15, THE RELATION BETWEEN TRICHROMIC VISION AND ANOMALOUS TRICHROMATISM. Anomalous trichromatism should be clearly defined as the condition in which anomalous matches are made by a person who refuses to accept the normal match. Much confusion exists on this point; a person who agrees with the normal equation cannot be regarded as an anomalous trichromat even though he agrees at the same time with the anomalous matches. This is only evidence of colour weakness, inasmuch as both equations are regarded as satisfactory. There are many anomalous trichomats who are not colour weak and there are many trichromics who make absolutely normal equations. Trichromic vision in my classification is therefore not synonymous with anomalous trichromatism. There are also persons who will make the normal equation in one set of circumstances and anomalous equations in another (14). There are also those who will make normal equations when the red employed is } 670 but will make an anomalous equation with a red of larger wave-length, as for instance » 690, putting twice as much red in the mixture compared with the normal equation in similar circumstances (16). Anomalous trichromatism when too much red is put in the mixed colour may correspond to defect in the perception of certain red rays, namely those employed in the mixed colour. JI have shown(5) that when there is shortening or much defect in the perception of red the junctions of the other colours are shifted towards the violet end of the spectrum. The yellow, therefore, corresponding to the D line, is seen as a much redder colour than the normal, and if we consider that the green is similar to the normal it is obvious that more red will be put in the mixture than by the normal-sighted. This shortening of the spectrum may be associated with normal vision in other respects or with any degree of defective colour differentiation, that is to say, it may be associated with dichromiec, trichromic, tetrachromic, penta- chromic, hexachromic or heptachromic vision. A similar condition is also found for the violet end of the spectrum. It is obvious that a man, who has shortening of the red end of the spectrum or defect in the perception of red, is colour weak as far as red is concerned. Unless, however, he has defective hue perception he may make no other error than that directly connected with the defective perception of certain red rays. It is different with those who make an anomalous match in which too much green is put in the mixed colour. As found by Rayleigh (6), Kollner (20), v. Kries (10), Nagel (13), and myself (14), a man may make an anomalous match without presenting any other colour defect. I have found 25 per cent. of men to be more or less colour weak, and it is, therefore not surprising that anomalous trichromatism 1912.] Trichromic Vision and Anomalous Trichromatism. 169 is frequently associated with colour weakness. The colour weak are also particularly liable to fail in making an equation, but in addition to making the anomalous equation they are in most cases satisfied with that of the normal, Anomalous trichromatism cannot be due to the diminution of a ereen sensation in the sense of the Young-Helmholtz theory. Apart from the fact that. I have shown that yellow is a simple and not a compound sensation (18, 24, 25, 26), there would be no reason why more green should be required in making the compound yellow, since the simple yellow would also contain less of the hypothetical green sensation. If whilst the yellow remains as in the normal the sensitiveness to green light were diminished or to red increased we should have an explanation of the facts. Schuster(19) found that the position selected as pure yellow was the same with the green- anomaly as with the normal-sighted. Whilst there are red-anomalies who show weakness for red, there are others who do not, and this may be explained by an increased sensitiveness to green whilst the red and yellow remain as in the normal. SUMMARY. 1, Trichromic vision is not synonymous with anomalous trichromatism. 2. Many persons with otherwise normal colour-perception make an anomalous equation. 3. Many colour-blind persons (dichromics and trichromics) make an absolutely normal match with no greater mean deviation than the normal. 4. Colour weakness is not characteristic of anomalous trichromatism but of trichromic vision. 5. Anomalous trichromatism and colour weakness are not synonymous. 6. A large mean deviation indicates colour weakness. 7. Anomalous trichromatism appears to be due to an alteration in the normal relations of the response to the three colours (lights) used in the equation. If the eye be more or less sensitive to one of the components of the mixed colour whilst the other has its normal effect, an anomalous equation will result. An anomalous equation will also result when the yellow is more allied to green or red than is normal. BIBLIOGRAPHY. 1, Edridge-Green. “Colour Blindness and Colour Perception,” ‘Int. Scient. Series,’ 1891 and 1909. 2. 5 “‘A Trichromic Case of Colour-blindness,’ ‘Ophth. Soc. Trans.,’ 1901. 3. * “The Evolution of the Colour Sense,” ‘ Ophth. Soc. Trans.,’ 1901. 4, a “ Two Cases of Trichromic Vision,” ‘ Roy. Soc. Proc.,’ 1905. 170 25. 26. Trichromic Vision and Anomalous Trichromatism. Edridge-Green. ‘Hunterian Lectures on Colour Vision and Colour Blindness,’ London, 1911. Rayleigh. “Experiments on Colour,” ‘Nature,’ 1881, vol. 25, p. 64. Donders. ‘“ Farbengleichungen,” ‘Du Bois-Reymond’s Archiv,’ Jahrg. 1884. Guttmann. “ Untersuchungen an sogenannten Farbenschwachen,” ‘ Kongress f. Experimentelle Psychologie in Giessen,’ 1904. Konig, A., and Dieterici. ‘Zeitschrift fiir Psychologie und Physiolggie der Sinnes- organe,’ 1893, vol. 4. . v. Kries. ‘Anomalen trichromatischen Farbensysteme. Physiologie der Gesicht- sempfindungen,’ Leipzig, 1902. Levy, Max. Inaugural Dissertation, Freiburg, 1903. Edridge-Green. ‘Colour Systems,” ‘Ophth. Soc. Trans.,’ 1905. Nagel. “Fortgesetzte Untersuchungen zur Symptomatologie und Diagnostic der angeborenen Stérungen des Farbensinnes,” ‘ Zeitschrift fiir Sinnesphysiologie,’ Leipzig, 1906. Edridge-Green. ‘Observations with Lord Rayleigh’s Colour-Mixing Apparatus,” ‘Ophth. Soc. Trans.,’ 1907. Lotze, A. ‘Untersuchung eines anom. trichr. Farbensystems,’ Diss., Freiburg, 1898. Edridge-Green. “The Relation of Light Perception to Colour Perception,” ‘ Roy. Soc. Proc.,’ 1910. Kéllner. “Uber das Grenzgebiet zwischen normalem Farbensinn und Farben- schwiche,” ‘Ophthalmologische Gesellschaft,’ Heidelberg, 1911. Edridge-Green. “The Simple Character of the Yellow Sensation,” ‘Journ. Physiol.,’ 1912. Schuster, A. “ Experiments with Lord Rayleigh’s Colour Box,” ‘Roy. Sec. Proc.,’ 1890. Koéllner. ‘Die Stérungen des Farbensinnes,’ Berlin, 1912. Guttmann. “ Untersuchungen tiber Farbenschwiche,” ‘ Zeitschr. f. Sinnesphysiol., 1907, vol. 42, pp. 24, 250; vol. 43, p. 146, 199, 255. Edridge-Green. “Die Wahrnehmung des Lichtes und der Farben,” ‘ Berliner Klin. Wochenschr.,’ 1909, vol. 46, p. 12. “‘Dichrormatisches Sehen,” ‘Archiv f. d. ges. Physiol.,’ 1912, vol. 145, p- 298. 5) “Simultaneous Colour Contrast,” ‘Roy. Soc. Proe., 1912, B, vol. 84. Porter, A. W., and Edridge-Green. “Negative After-images and Successive Con- trast with Pure Spectral Colours,” ‘ Roy. Soc. Proc.,’ 1912, B, vol. 85. Edridge-Green and Marshall, C. D. “Some Observations on so-called Artificially Produced Temporary Colour-blindness,” ‘Ophth. Soc. Trans.,’ 1909, vol. 29, p. 211. NL A Preliminary Note on a New Bacterial Disease of Pisum sativum, By Dororny M, Cayty, Research Student, John Innes Horticultural Institution, Merton, Surrey. (Communicated by W. Bateson, F.R.S. Received November 19, 1912,—Read January 23, 1913.) Investigations have been carried out this year at the John Innes Horticultural Institution to elucidate the nature of a disease which affects culinary peas (Piswm sativum), The disease, in this district at all events, is a serious one, killing a large proportion 6f the crop, but I have no information as to its prevalence in other parts of the country. I have succeeded in proving that the disease in culinary peas is caused by a large bacillus which exhibits a peculiar feature, inasmuch as it is transmitted in the interior of the seeds of the plant. As far as I am aware no analogous instances are known. Owing to the work of Chamberland (1879), Lehmann (1889), Laurent (1891), and others, it has been definitely proved that not only is the interior of a normal seed sterile, but also beans and peas taken under sterile conditions from healthy pods are free from bacteria. A very large rod-shaped bacillus has been isolated from the stem of the living pea plant and from the centre of the cotyledons of the pea. The life history of the organism is complicated by involution-forms and a zoogleeal stage. In the rod stage the bacillus is Gram-positive, non-acid-fast, very motile when young but enveloped in a capsule when at rest. It varies considerably in size according to the amount of water, food material, and other conditions. It grows well on acid (1 per cent. normal HCl), alkaline (1 per cent. normal NaOH), and neutral pea agar agar, forming small circular, pale buff, translucent, watery colonies on the surface of the medium, and when submerged the colonies are deeper in colour, opaque and lens-shaped. The colour of the colonies varies according to the medium used. Under certain conditions the colonies may have a decided orange tint. This was especially noticeable in an impure culture into which a spore of Penicillium had been introduced. The bacterial colonies immediately round the Penicillium were both larger and of a deeper tint than in other parts of the Petri dish where no fungoid growth occurred. This orange tint has also been observed in badly diseased cotyledons after VOL. LXXXVI.—B. ) 172 Miss D. M. Cayley. Preluminary Note ona [Nov. 19, germination, but does not necessarily occur in all cases. Further elucidation of this point is necessary. No growth has so far been observed on lactose pea agar agar. In liquid peptone beef broth the rods grow to a great length and are strung together in chains. The organism occurs in the phloem, cambium, medullary rays, and occasionally in the pith of the stem, also in the parenchyma of the vascular bundles which run along the mid-rib of the pod, in the tissue of the funicle and cotyledons. In the very young plant grown in sterile sand the bacillus has been found in the primary ground-tissue of the radicle inside the pericycle, and in the young phloem and cortical tissues of the shoot. The general symptoms are. as follows:—In mild cases after germination the shoot can develop normally, but in bad cases it is frequently abortive, brown and dead at the tip, and laterals grow out prematurely to take the place of the main shoot. Quite early in the development of the plant, when the plumule is from half an inch and upwards in length, light brown longitudinal streaks can be seen on the stem and root, and the first leaves are often brown at the tip. These streaks develop later into slits. Im very. bad cases little or no germination takes place. After this stage no further definite signs are noticeable till about the flowering period. Then the development of the disease depends a good deal on external conditions. If the weather is warm and dry, and the plants are growing vigorously, the disease develops rapidly, and in a few days the plants become unhealthy and change colour. The stem turns slightly brown, and looks somewhat water-soaked. Brown longitudinal streaks appear at the base of the petioles on either side of the rib of the stem, which is continuous with the mid-rib of the leaf. The streaks split open and dry out. ‘The collar may be badly disorganised. The leaves become spotted, streaked and yellowish in colour, and if the disease is progressing rapidly the younger portions of the plant show discoloration, and fail to develop properly. Except in bad cases the plants grow to full height, and can flower and set a certain amount of seed, but on examination the cotyledons of the seeds of a diseased plant show brown discoloration, which may be limited to a mere spot in the centre of each cotyledon, or, on the other hand, nearly the whole of the cotyledon may be involved. In the latter case there is often a cavity in the centre of the cotyledon. Sections of the diseased cotyledon show large numbers of bacilli in various stages of development in the cells and intercellular spaces. The bacillus works its way into the intercellular spaces and then breaks 1912. | New Bacterial Disease of Pisum sativum. 173 into the cells. There the nucleus is often attacked, the cytoplasm destroyed, and the cells collapse, thus forming rents in the tissues. There is considerable evidence to show that the bacillus passes up the plant through the tissues above mentioned, through the funicle, and probably the micropyle into the young developing seed. If one pea is diseased all the other peas in the same pod are diseased to an equal extent. The disease is chiefly spread by the seed, but fresh infection may take place through the soil. Inoculation experiments were carried out in the open, but little stress can be laid on the results, as the disease was so prevalent throughout the experimental plot. Pea plants grown in heated soil in boxes, and inoculated just above the ground, when the plants were about 1 foot in height, showed no disease, whereas, in the open, seven out of ten inoculations on the stem just below the youngest unfolding leaf were successful. Further inoculation experiments are necessary, but the above results tend to show that the bacillus can only penetrate very young tissue. This is supported by the fact that large numbers of the bacilli have been found in the inner tissues of the radicle when only about half an inch long. Further investigations are in progress. In many respects the symptoms resemble those of the formidable disease of sweet pea (Lathyrus odoratus) known as “streak.” This disease has been held to be due to Thielavia busicola, but, in view of these observations, that conclusion seems very doubtful, and I may add that, in the stem of diseased sweet peas, I have already found bacteria like those here described. 174 On the Manganese Content of Transplanted Tumours. By F. Mepicreceanv, M.D. (Communicated by Sir J. R. Bradford, K.C.M.G., Sec. R.S. Received November 21, 1912,—Read January 23, 1913.) (From the Laboratory of the Imperial Cancer Research Fund.) The occurrence of manganese in plants is well known. Its quantitative distribution and biological significance have been carefully studied from many points of view. A few examples of the more recent, especially experi- mental work, illustrating some of the biological properties of this metal, may be mentioned. Bertrand* showed that there exists a close relationship between the activity of vegetable oxydases and the amount of manganese present. In a series of very exact experiments with Aspergillus niger, the same author demonstrated that the presence of manganese is necessary to the formation of conidia of this mould,t and also that the rapidity of its growth may be largely influenced by the quantity of manganese added to the culture medium.t The study of manganese in animals is far less advanced than in plants. Since the food-stuffs contain manganese, it is obvious that this element is continuously introduced into the animal body. The detection of manganese in animal tissues has been the subject of repeated investigation during the last 70 years. The conclusions, however, which the earlier authors have drawn are very contradictory, undoubtedly attributable mainly to the insufficiency and the defects of the methods and the technic used for the detection and estimation of this element.§ Recently Bertrand and Medigreceanu applied Bertrand’s colorimetric method for estimating the manganese in organic substances to an extensive analytical study of this metal in normal animals. By means of this method manganese can be estimated even when present in very small quantities, 2/1000 mgrm., and with an error not exceeding 10 per cent. Manganese was thus found to be a normal constituent of the organism throughout the animal kingdom.{1 The invertebrates usually show relatively * G. Bertrand, ‘Comptes Rendus,’ 1897, vol. 124, p. 1032. + ‘Bull. Soc. Chim. France,’ 1912, Ser. 4, vol. 11-12, p. 494. { Lbid., p. 400. § See Bertrand and Medigreceanu’s article, ‘Bull. Soc. Chim. France,’ 1912, Ser. 4, vol. 11-12, p. 656. || ‘Bull. Soc. Chim. France,’ 1911, Ser. 4, vol. 9, p. 361. J See Bertrand and Medigreceanu, ‘Comptes Rendus,’ 1912, vol. 154, pp. 941, 1450 ; 1912, vol. 155, p. 82. On the Manganese Content of Transplanted Tumours. 175 much larger quantities of manganese than the vertebrates, and of the verte- brates the mammals contain the smallest amounts—a few hundredths of a milligramme per 100 grm. of the total weight of the organism—while the birds, reptiles, batrachians, and fishes show 5-10 times as much. The quantitative distribution in the different organs, tissues, and animal products, especially of the higher classes that have been studied, is very interesting. The blood, for example, contrary to the claims of most previous investigators, has been found to contain much smaller amounts of manganese than sometimes admitted, usually only a few hundredths of a milligramme per litre. The hemoglobin of horse blood contains no manganese. Of the organs and tissues of principal functional importance higher manganese values have been met with in the liver (0°265-0:416 mgrm. per 100 grm.) and in the kidneys (0:063—0°238). Lower values are found in the muscular tissue (< 0:005-0:018), the nervous tissue (< 0:005-0:036), and the lungs (0°006-0:023). The organs of the birds are generally richer in manganese than those of mammals, and the highest value obtained has been for the oviduct of birds (0°786-2:201). It may be mentioned that the grey matter of the ox brain is much richer (0022) in manganese than the white (< 0:005), and also that, in general, the heart and muscles of the tongue contain larger amounts of this metal than the trunk muscles and the muscles of the extremities. Among the organs or tissues of minor functional importance, the hair, plumage, and nails contain relatively large amounts of manganese (0°111-3:214), The milk is very poor in manganese, although apparently richer than the blood. In the egg-white (fowl and duck) they were unable to detect this metal, even when analysing 100 grm. of the fresh substance. The yolk seems to contain all the manganese present in the egg. Considering the ubiquity of manganese throughout the animal kingdom, and its remarkable distribution in the various tissues, these authors emphasise the importance that it probably has as a catalytic agent of living matter. Again, the wide differences shown to obtain between the amount of manganese found in plants and in animals is an observation which may have considerable importance. The quantities of manganese present in the various organisms and tissues may very well be taken into consideration in studying the problems of the origin of species, as well as those of bio- chemical adaptation to the medium, in interpreting the influence of vegetarian and flesh diets, and finally in drawing deductions as to the nature of the physiological soil. In connection with Bertrand and Medigreceanu’s work on manganese, it 176 Dr. F. Medigreceanu. On the [Nov. 21, seemed of interest to study its occurrence and quantitative distribution in tumours. The transplantable tumours (mouse, rat, dog) were chosen for the purpose. These kinds of tumours are at present the best known as regards their biological and morphological properties, and the most suitable for an exact and rapid orientation. The tumour strains analysed belong to both the principal morphological groups—carcinoma and sarcoma. Lach strain shows different morphological and biological properties.* Technie. As already mentioned in the introductory part, the estimation of manganese was made by Bertrand’s colorimetric method. It consists essentially in converting the manganese present in the sulphate ash of the organic substance into permanganic acid, in oxidising the ash dissolved in concentrated nitric acid with potassium persulphate in the presence of silver nitrate, and in comparing the intensity of its rose-pink to violet colour with the colour of standard solutions of the same acid prepared in a similar way. The details of the method were followed exactly as given by Bertrand and Medigreceanu.t Of the tumour tissue to be analysed quantities not exceeding 100 grm. were first dried at 100° C. and then incinerated at the lowest possible temperature, using sulphuric acid for the destruction of the final traces of carbon. The sulphate ash was then dissolved in concentrated hydrochloric acid, again treated with a little sulphuric acid, and finally heated until the appearance of white fumes of sulphuric acid denoted the absence of hydrochloric acid. The residue was then dissolved in 10 c.c. of 25-per-cent. nitric acid, and if necessary the undissolved part of the ash allowed to precipitate. A few drops of 10-per-cent. silver nitrate was then added to the clear solution, the tube warmed, and its contents oxidised with a few decigrams of potassium persulphate. The greatest care was always taken to avoid introducing impurities containing manganese into the samples for analysis, and pure reagents were used throughout. It may also be mentioned that the small quantities of blood contained in the tumours do not influence the analytical results, for it was found that 25 grm. of mouse blood treated in the same way did not show any visible trace of manganese. This fact fully agrees with the previous findings of Bertrand and Medigreceanu, who observed only traces of manganese in larger * Full details on the tumours analysed may be found in the ‘Fourth Scientific ‘Report on the Investigations of the Imperial Cancer Research Fund,’ London, 1911, Taylor and Francis. + ‘Bull. Soc. Chim. France,’ 1912, Ser. 4, vol. 11-12, p. 656. ibraze Manganese Content of Transplanted Tumours. 1912. | | ¥ 09 ‘mad g srnoumy “MIs ZZ corr :4Y Fem oAeIOAY “eorut ET 400-0> Z00. 0> 9% O8T-s2 Aas Lyeo 26 WI J. Z SMMOMINY “UMS GT GOTU :4YSTOM OsvIOAW “ooTUL FZ F00- O £00. 0 99 &I agoL oot “eYRULOIILG | V re ‘US T. [ smmourng “us JT OOIUt : 4YSIeM osvIoAW “9OIUl ZT FIO. 0> Z200: 0> Gal 69 -2z V 8g | @ 3& 982 ‘pmbry queoseyedo Sururequoo ‘sqsho ofrerT a 1s ‘WIS G. T SINOTINY ‘TAS OT GTM :4YSIOM oHVIOEAW ‘OOIUL ET TO. 0> 200: 0> 0Z 89 -Fr 1g aH 0g OOT d 69 ; ; j d $9 ma ¢. ¢ sMMouINg “UI JT GOTT :4YSTaM oFICAW “OOTUL ET 210-0 900. 0 9F O&I-8F 19 : Wos LHS 18 ; Vv bY “WAS Z SAMOUINy “UIT g. GT COTM :4ySIom osvieAW “OOTUL TT €10- 0 £00: 0 63 69 —62 V & Vor 66I “WHS Z SIMOUING “WIS QT SOTUL : 4ySI0M ODeIEAW -odTUI EZ 800- 0 F00- 0 Lt 61 V $F (2) ‘orysho puB 9140.100 NT oF ‘WAS g. J SINOTING ‘UAT QT SOTUL :4FYSIOM OSBIOAW “dT OT 10: 0 €00- 0 82 8-82 M og (1) 16 ; : Ves WAS T. Z sNOUINy “UIIS F. CT SoTUL :4ySIEM OFBIOAW “OO1UI OT 600: 0 G00: 0 1Z os -8T d 99 -oysfo AT[eroues smmoumyy, aso (2) ‘WA g. [ smMoung “UIs JT GIT :4YSIOM OSBIOAW “OOTUI GE 10-0 900: 0 09 8z as9 (1) 89 “eyeUIOULIeD PUe vyeUMOUIIIeI-OUDpy | | ‘eome4sqns . Yseaz O49 JO Cigutee ‘SOULULBIO UT f ‘MIs OOT 19g bs eu oE ‘peuruexe aur syieueyy yeur9 your Ysoay utr aru ‘UIBI48 NOUN, ay fo qusiom | #9 °° ‘eSOURSUBUL JO SOMMBISTTIIPL ‘SINOWUN]T, Isnoyy On the [Nov. 21, Dr. F. Medigreceanu. 178 “UNIS J SINOUINY :44SI0M OSBIEAW “ssOp ge €00- 0> Z00: O> 2G 1% V OF woys10ue4 4, eulooreg-oydurAT : ‘Moun y, Soq V6L é. ' 800: 0 100: 0 98 (7) z WIS OF sinouINy “UIs BE syBI :44SIOM oSvacAy ‘s}vI Cc { 10.0 | <00. 0 0g (») 21-68 VLL ( VSL () “ULIS F. & SIMOUINY “UAT PH) 8¥BI : 4YSIOM OSvIOAY *s4BI QT 600: 0 S00: O cs 61 V 08 (8) ‘US [p SINOUING “WIS Eg SyeI :qYSIOM ODBIOAW “84vI Z 900. O coo: O 08 06-2S vee (2) “WAS ZG AMON “UIs JOT FV > 9USIOAA “9BA T 900- 0 £00- 0 0g &L V LL (1) ‘s'ae “euLOdIeG ‘oigoaoou Ar0 A. s | { V8s ‘ULIS 9. SANOUING “WIS FO syed :4YSIOM OBRIOAY ‘szeI C |} 10.07 B00. 07 BE Sulsey l\lL¥SL OU “*BULOULDIEA t . | | eounjeqns -ojdures I. | -‘syaBuray | Wis OOT 2d | enaarcies acess | ur ‘smmoumy ‘UIBI4s Moun, (Eee 2 a ae he ee Bee! Kt jo os ‘Qs0UBSUBUL JO SAULUBISIT[L aq} FO 4USTO AA ‘SINOWNT, Jey 1912.| Manganese Content of Transplanted Tumours. 179 quantities of the blood of several mammals and birds. It serves also as a control to the purity of the reagents used. The analytical results in the several tumour strains examined are shown in the adjoining table. Summary and Conclusions. As a general conclusion it may be stated that the quantities of manganese found in transplanted mouse and rat tumours, whether carcinomata or sarcomata, and also in the so-called lymphosarcoma of the dog, are very small —they vary between 0:004 and 0:012 mgrm. per 100 germ. of the fresh material. In order to obtain an idea of the comparative amounts of manganese con- tained in the normal mammary gland and the epithelial tumours derived from it, two manganese estimations were made of normal mouse mamma. In the first, 15 grm. of lactating mamma were analysed. Only 0:004 mgrm. was found, ze. 0026 mgrm. manganese per 100 grm. For the second determina- tion, 14 grm. of resting mamma were incinerated. The amount of manganese present was 0:002, 22. 0:014 mgrm. per 100 grm. Though the comparison be not strictly exact, nevertheless the figures obtained allow this general conclusion to be drawn, that the epithelial transplantable mouse tumours developing in the mammary gland do not contain a larger amount of manganese than their normal mother tissue. Furthermore, there are not very marked differences in the percentage distribution of manganese between carcinomata and sarcomata. In connec- tion with this observation it may be also mentioned that the carcinoma and sarcoma strains of the mouse tumour “100” do not exhibit appreciable differences in their manganese content. 180 The Influence of the Resilience of the Arterial Wall on Blood- Pressure and on the Pulse Curve. By 8. RussELL WELLS and LeonarD HI, F.R.S. (Received November 29, 1912,—Read February 6, 1913.) This communication is the result of two independent but converging lines of research. It is well known that when a fluid is driven with a rhythmically varying pressure through a sufficient length of a distensile elastic tube, the pressure at the exit loses its rhythm and becomes constant and the flow continuous, whereas if the tube is rigid, the pressure at the outlet varies as that at the inlet (less the change due to friction) and the outflow is intermittent. Since the arteries are distensile elastic tubes and the blood is rhythmically forced into them by the heart, it follows that the curve of blood-pressure must be altered to a greater or lesser degree by the distensibility and elasticity of the arterial wall. We use the term resilience in this paper to express the ease with which an elastic tube distends with a rise and recoils with a fall in pressure of the contained fluid; thus, a rubber tube with a wall 0:2 mm. thick is more resilient than one with a wall 0-4 mm. thick, the thinner, more resilient tube yields with the rise and recoils with the fall of pressure more than the “harder,” thicker walled, less resilient tube. A glass tube in this sense has no resilience, and the same may be said of rubber pressure tubing. As the arterial wall contains muscle its resilience will be altered by a more or less contracted state ; as the degree of contraction and resilience may vary locally it is to be expected that the curve of blood-pressure may also vary, e.g. in the brachial and femoral arteries. Further, as the peripheral resistance in any area may alter the tension of the arterial wall, its resilience may vary without any change in the muscular state of the arterial wall. Observations made by one of us (L. H.) with W. Holtzman and Martin Flack, and later with R. A. Rowlands,* on cases of aortic regurgitation placed in the horizontal position, have shown that the systolic pressure is much higher in the leg than in the arm, eg. 100-150 mm. of mercury higher, and so characteristic is this difference that it is a diagnostic sign of the condition. Thomas Lewis found that the same held good in the case of a dog in which he had experimentally rendered the aortic valves incompetent one month previous to taking the observations.+ His measurements were * ‘Heart,’ 1909, vol. 1, p. 73; and 1912, vol. 3, p. 219. + ‘Heart,’ 1912, vol. 3, p. 222. Influence of Resilience of Arterial Wall on Blood-Pressure. 181 recorded by means of cannule placed in the arteries and connected with Hiirthle manometers. Hiirthle* and others have recorded previously higher readings of pressure in the femoral than in the carotid artery of the dog. Tigerstedt ascribed these to reflection of the primary pulse wave without change of sign and addition of the reflected to the primary wave in the femoral artery.t The difference of the systolic pressures in the arm and leg in aortic cases was ascribed by L. H. and his co-workers to the better conduction of the systolic wave crest in the leg arteries, which were assumed to be in a more contracted and harder state. This view was confirmed by experiment, for it was found, on placing the legs and buttocks of the patients in a hot bath, the difference between the readings of arm and leg arteries was abolished, and this was ascribed to the expanding and softening of the contracted walls of the latter. Also, in the case of healthy young men placed in the horizontal posture, while it was found that the leg and arm readings of systolic pressure were normally the same, these were rendered temporarily unequal after the subjects had run twice up and down a long flight of stairs (particularly if the arm were placed in hot water beforehand); the heart was thereby made to beat forcibly, while the leg arteries became more contracted, so the crest of the wave was better conducted in them than in the arm arteries. By placing one arm in hot water, it was found possible to render the reading different in the two arms, even in the resting subject, much more so after a short period of violent exercise. If the wrist alone were placed in hot water, the radial gave a lower reading than the brachial, but if the elbow were placed in hot water, readings of brachial and radial were equal, both being lower than in the other and cooler arm; bandaging the hand tightly made no difference to the reading. The conclusion arrived at was that the inequality was due to an altered condition of the arterial wall and not to diminished peripheral resistance, and these experiments led to the conception that the nature of the arterial wall affects the conduction of the systolic wave, and that the blood-pressure, as ordinarily measured by a sphygmometer, by the method of obliteration of the pulse, depends not only on the pressure wave produced by the heart, but also on the effect on this wave of the arterial wall, a new factor which has not hitherto been taken into account. A difference of pressure between the arm and leg readings has been noted by several observers in cases where the arteries are thickened and hardened as in old people. This difference has been ascribed to an error in the * “Arch. f. d. ges. Physiol.,’ vol. 47, p. 32. + ‘Lehrb. d. physiol. des Kreislaufes,’ 1893, p. 352. 182 Messrs. 8. R. Wells and L. Hill. Influence of [Nov. 29, readings due to the thickened artery resisting compression, just as an empty rubber tube does. One of us (L. H.) and Martin Flack have found that such differences of pressure are lessened by keeping on the pressure of the armlet, and lowering and raising it so as to take several readings of : (1) the reappearance ; (2) the disappearance of the pulse. Our explanation is, that the artery cut off from the blood relaxes and softens, and therefore the crest of the systolic wave is diminished. It has been shown by Bayliss that compression of an artery is followed by vascular dilatation in the area cut off from the blood. In many of these cases the force of the pulse is irregular; now and again an extra large systolic crest forces its way beneath the armlet, and such large waves are better conducted by the leg arteries, just as happens in the case of aortic regurgitation. By means of a circulatory schema, in which two lengths of artery are inserted, one to be compressed, the other to be palpated (the latter gave the . index, the disappearance of the pulse), it was easy to demonstrate that the systolic pressure is read more accurately when the palpated artery is made tense (produced in this schema by obstructing the outflow by means of a mercury valve) than when it is soft. In the first case the readings of systolic pressure taken in the pump and in the artery are the same, in the second case the reading taken in the artery is lower. In the living animal with its vasomotor nerves, and pressure changes of rapid rate, and output of the heart varying from second to second, it is extremely difficult to study exactly the effect of the various factors on the character of the pulse curve, for one cannot vary at will one of these factors without affecting the others. From these considerations it appeared desirable to one of us (S. R. W.) to investigate the subject by means of non-living elastic tubes. Halls Dally and K. Eckenstein have assisted in this research, which will be published in full later. After considerable experiment an apparatus was devised, by means of which fluid at a known rhythmically changing pressure could be passed (a) through elastic tubes of the same calibre, but with walls of various known thicknesses; (6) through various lengths of the same tube, and (c) keeping to the same tube, the absolute pressure could be varied, while maintaining the same difference between the systolic maximum and the diastolic minimum, or this difference could also be varied at will. The tubes used in the experiments were various lengths of rubber tube all of the same internal calibre, but with walls of 0°8, 0°6, 0-4, and 0:2 mm. thickness. The pressure variations of the fluid before flowing through the resilient tube and at the end of it were recorded by Hiirthle’s manometer. 1912.] Resilience of Arterial Wall on Blood-Pressure. 183 It was found when the same resilient tube was used, but the diastolic pressure of the entering fluid varied, keeping the interval between the systolic and diastolic pressures as far as possible the same, that the higher the pressure and consequently the more the resilience of the tube was brought into action by stretching, the nearer together were the diastolic and systolic pressures at the end of the resilient tube. In other words the smaller was the amplitude of the pressure waves, and the more closely did the pressure approach to a continuous one. As an instance, the following experimental results may be cited, working with 30 cm. of a rubber tube, the walls of which were 0°8 mm. thick and recording the pressures in milli- metres of Hg. Entering pressure. | Pressure at end of rubber Difference between tube. initial and end pressure. | Systolic. | Diastolic.| Difference. Systolic. | Diastolic. Difference. | Systolic. | Diastolic. lel | | | a 145 50 95 | 120 GoM 60 Pige || og zai 184. 86 Ogi edno) yal remo |)" 48 —32 +18 220 125 95 | 188 14s. 40 —32 +23 The same general results followed, no matter what the thickness of the wall of the tube experimented on might be, and no matter what its length, but the difference in the case of the thinner walled tubes was even more striking. Working with a rubber tube 30 em. long, and with walls 0:2 mm. thick, with an entering pressure of 78 mm. of Hg diastolic and a 148 mm. systolic, an almost continnous pressure of 104 mm. diastolic and 107 mm. systolic was obtained at the end of the resilient tube. With raised pressure not only was the curve of less amplitude, but its form also was altered, the top becoming flattened and the dicrotic wave less marked, indeed it took on the characters which have frequently been described as occurring in the sphygmograms of cases of high blood pressure. In order to test the correctness of the supposition that as the general level of pressure was raised the resilience of the wall was increasingly brought into play, a series of experiments was carried out, using the same initial pressures and the same thickness of tube wall, but varying lengths of tube. It was found that lengthening the tube had the same effect of approximating the systolic and the diastolic pressures and making the curve take on the characters of a “high pressure” sphygmogram. For instance, using a tube with walls 0°8 mm. thick, the following results were obtained :— 184 Messrs. S. R. Wells and L. Hill. Influence of [Nov. 29, ous Pressure at end of rubber Difference between Length Entering Pres: tube. initial and end pressure. of tube in | | : ee ; em. (Systolic. | Diastolic. | pies Systolic. | Diastolic. | Dies; Systolic. Diastolic. | | | | | 15 160°" | “40 S'S 2am atz6 ee 66 —34 +20 30 160. |. e420) | ete ienTaS | 74 44 —42 +32 60 ily | 40 he ails 108 | 78 33 —49 +38 | | The same initial pressure differences were then tried on tubes of the same length and calibre, but with walls of different thicknesses, namely, 0°8, 0-6, 0:4, and 0-2 mm., when the same sort of results were obtained, viz., the thinner and consequently the more resilient the tube, the more was the systolic pressure lowered and the diastolic raised by passing through the tube, that is, the nearer the resultant curve approached a straight line. It was quite remarkable to observe how with an entering pressure such as 160 mm. systolic and 40 mm. diastolic, a curve in the 0°8 mm. tube would have all the characters of a low pressure sphygmogram, great amplitude, sharp rise and fall and very well marked dicrotic wave, while with exactly the same entering pressure the curve in the 04 mm.,and more so in the 0-2 mm., took on all the characters of a high pressure sphygmogram, slow rise, flat top, slow fall, and shghtly marked dicrotism. L. H. and Martin Flack have since found that the introduction of, say, 6 cm. of cat’s carotid artery in place of an equal length of pressure tubing alters the characters of the pulse curve from a low to a high pressure curve. The experiments demonstrating this will be published in full later. From these experiments conducted by S. R. W. and those of L. H. it seems legitimate to draw these conclusions: the form of curve obtained by a sphygmograph or other instrument recording the pulse is the resultant of two factors, the blood-pressure variations produced by the heart and the resilience of the arterial wall, using the term resilience in the sense defined above. Much at times has been made of the supposed influence of reflected waves on the pulse curve. It is the resilience of the wall which we believe to be the important factor in modifying the curve, and not the reflection of waves from the periphery. The blood-pressure measured in any artery by the sphygmometer is likewise the resultant of these two factors, and the measurement does not necessarily give us the full systolic pressure produced by the heart; much of the force is spent in dilating a soft distensile artery. Further, since the character of the flow in an artery largely depends on the resilience of its 1912.| Resihence of Arterrval Wall on Blood-Pressure. 185 walls, it is obvious that, the more resilient or yielding are those supplying any part, the more closely will the blood stream at the threshold of the capillary area supplied approach a uniform pressure (roughly the mean between the systolic and diastolic pressures, less, of course, what has been lost by friction), while the harder or less resilient the arterial wall, the more closely will the variations approach those in the aorta, Now all the arteries, and to a greater extent the arterioles, are contractile, and under the influence of the nervous system, and with an increased tone or contraction, they become not only narrower as to lumen, but also thicker as to wall, that is less resilient, so it may happen that the organism can with the same heart force vary the pressure at the threshold of a particular capillary area between an intermittent pressure with a high systolic beat and an almost continuous one, with a lower systolic pressure. In the one case there would be a hammer-like percussive wave beating open the capillaries, the blood would be hammered in; in the other there would be a more continuous pressing in of the blood at a lower tension. It may be that narrowing of lumen and lessened total flow goes with the more percussive wave due to hardening, but this does not necessarily follow, for it is possible that a tightening of the muscular coat, and a lessening of the resilience, may take place before actual narrowing occurs. It is further possible that the great arteries and the arterioles act differently, or independently in some cases. We advance the view that the throbbing and capillary pulse observed in an acutely inflamed area is due to an increased tone of the arterial walls, a lessening of their resilience ; this throbbing is often relieved by hot fomenta- tions which act by relaxing the contraction of the vessel walls. In cases of aortic regurgitation the hammer-like pulse propelled through the harder leg arteries secures to the legs an adequate supply of blood, compensating as it does for the diastolic fall due to the regurgitation ; there, again, hot water baths relax the arterial wall. One of us, L. H., with Martin Flack,* has shown that in the case of the salivary gland each alveolus is surrounded by a tough membrana propria which resists expansion and allows the secreting cells to draw fluid from the capillaries and raise the secretory pressure to almost double the height of the arterial blood-pressure, without obliterating the surrounding capillaries or interrupting the venous outflow. Under such conditions the veins are narrowed, by the expansion of the alveoli up to their limiting membranes, and the blood vessels, arteries, capillaries, and veins form a system of rigid vessels with a rapid rate of flow, the pulse even coming through into the * ©Roy. Soc. Proc.,’ 1912, B, vol. 85, p. 312. 186 Influence of Resilience of Arterial Wall on Blood-Pressure. veins, the whole gland feeling tense to the touch. Increased hardness of the arteries, supplying such an active organ, permits the full force of the systolic wave to come into play, and insures a flow of blood in the face of the increased osmotic pressure and swelling of the tissues. In the condition of local inflammation the heart beats forcibly, and the arteries conduct the full force of the wave to the swollen, tense, inflamed part; the part throbs with pain. A fomentation, by softening the arteries and the confining frameworks which surround the tissue cells, or the surgeon’s knife by relieving the tension, permits an ampler flow of blood with its curative properties. It is the altered osmotic condition of the infected inflamed tissue which causes the swelling, and this may advance to such a degree that the circula- tion is strangled and the part necrosed ; before this happens, however, the full stroke of the heart’s systole is conveyed by the hard contracted arteries with hammer-like strokes to the part and forces the blood through the vessels, maintaining the circulation and thus allowing the bacterial poisons to be neutralised up to the utmost possible limit. Probably the hypertrophy of the muscular coat of the arteries in certain pathological conditions is correlated with a need for the hammer-like stroke. 187 On the Non-identity of Trypanosoma brucei, Plummer and Bradford, 1899, with the Trypanosome of the Same Name from the Uganda Ox. By J. W. W. STEPHENS, M.D., D.P.H. (Cantab.), and B. BLACKLOCK, M.D., D.P.H. (Communicated by Sir R. Ross, K.C.B., F.R.S. Received December 7, 1912,— Read January 23, 1913.) Introduction. Before considering our own observations it will be necessary to review briefly previous statements regarding the morphology of Z’rypanosoma brucei. (1) Bruce (1) states that the hematozoa vary among themselves a good deal in shape and size and seem to take on slightly different forms in different species of animals. He publishes four figures depicting nine trypanosomes. Possibly one or two of the five figured from the dog might be considered to be “ stumpy ” forms. (2) Kanthack, Durham, and Blandford (2) state that the Nagana parasites vary considerably both in size and form. They may be long and pointed and sometimes stouter, some individuals are short and thick with a short flagellum, their protoplasm being crowded with granules. This description suggests dimorphism, but it should be noted that forms without a free flagellum are not mentioned. No slides were available belonging to these observers, but Dr. Durham kindly lent us a large series of photographs. On examining these, one or, perhaps, two show a “stumpy” form, but it is difficult to be certain, and the uniformity of the remainder is striking. They state that “the material for our observation was obtained in the first instance from the blood of a dog infected by the disease on the voyage from Africa, and brought to England in November, 1896, by Dr. Waghorn.” This animal we believe to be the origin of the strain of 7. brucei, Plimmer and Bradford, 1899, described by these authors, and at present maintained in England, so far as we can gather, solely at Liverpool. It is not stated above from what animal the dog was infected on the voyage, nor is it stated what the exact original source of the strain derived from Zululand was. (3) Plimmer and Bradford (3 and 4) describe four forms in the blood, but neither their description nor figures suggest that they have seen stumpy forms. They describe “a large hyaline form.” “This organism is much larger than the ordinary adult form, and is much wider, often more than VOL, LXXXVL.—B. FP 188 Drs. Stephens and Blacklock. T. brucei, Plimmer and [Dec. 7, double the width, and is more irregular in shape. The protoplasm is quite homogeneous and much more delicate, and it stains very faintly with the methylene blue.” They are still to be found in films, and we easily found them in Dr. Plimmer’s old films, but we found only extremely rarely forms that could be called “stumpy.” We think that if they had been present these observers would hardly have failed to have noticed and drawn attention to them, as they have a striking appearance. (4) Bruce and others(5) make a comparison between 7. brucei, Uganda, 1909, and 7. brucei, Zululand, 1894. They state that many of the old Zululand preparations are still extant, so that it has been possible to do this. The preparations were, however, 15 years old, and had been stained with carbol fuchsin. ‘The slides were got from horse, donkey, ox, monkey, dog; 200 trypanosomes were measured from the Zululand strain and 172 from the Uganda strain. The curves obtained in this way certainly resemble one another, though in one case the peak is at 18 yw, in the other case at 20 wu. Also trypanosomes are figured from each strain, and there can be little doubt that there is a close resemblance, if not identity, viz., in the fact that both possess both long and stumpy forms. “ With the evidence available the Commission consider themselves justified in considering the trypanosome recovered from the Uganda ox to be identical with 7’. brucei, the cause of Nagana in Zululand and other parts of South Africa.” Further, Bruce states in another paper (6) that 7. brucei (Uganda strain) has actually 26 per cent. of non-flagellated forms. (5) In 1911 Laveran(7) published an article, entitled “Identification et essai de classification des trypanosomes des mammiferes.” In this article he places 7. brucei in his Group I, “ Trypanosomes chez lesquels le flagelle présente toujours une partie libre,” whereas he places T. gambiense in his Group III, “ Trypanosomes ayant des formes a flagelle libre et des formes sans flagelle libre.” Or, in other words, 7’. brucei is classed among the monomorphic trypano- somes while 7. gambiense is among the dimorphic. We have then two opposite statements as to the morphology of 7. brucei, (1) that it is mono- morphic and (2) that it is dimorphic. We possess two strains of (so-called) 7. brucei in the laboratory, viz., the Zululand strain, of which we have given the origin above, and the Uganda strain from Surgeon-General Bruce, obtained originally from the ox in Uganda in 1909. These strains have been maintained continuously at Runcorn in a variety of animals, the Zululand strain for 44 years, and the Uganda strain for 24 years. In the case of the Uganda strain it was lost in 1912.] Bradford, 1899, and Trypanosome from Uganda Ox, 189 1912 fora short period but was returned to us again by Prof. Mesnil, who had previously received it from us. We made then a preliminary examination of these two strains, and found to our surprise that they could easily be distinguished morphologically. We next proceeded to make a detailed examination of the two strains in a series of slides throughout the entire period of infection in various animals, viz., rats, guinea-pigs, and rabbits. As the result, we believe we have established the following facts :— (a) The Zululand strain is typically monomorphic. The trypanosomes are long, with a long free flagellum. We must admit, however, that it is possible (as we believe is the case also in another typically monomorphic trypanosome, viz., 7. evansi) to find by long search short forms which somewhat resemble true stumpy forms, but we must emphasise the fact that, in all the slides we have examined, prolonged search is necessary to find them. (b) We have also verified the fact that Laveran’s 7. brucei strain also is, as he says, monomorphic. The origin of this strain seems uncertain. Laveran probably received it from Ehrlich, but where the latter got it from cannot now be ascertained. Unless it came from England, there must be two monomorphic 7’. brucei strains in existence, not to mention the possibility of other 7’. brucez strains of uncertain parentage in various laboratories. We have examined also old slides from the Zululand strain lent us by Prof. Nuttall, Colonel Skinner, R.A.M.C., and Dr. Plimmer.* All these were monomorphic. We repeat here that in these films, or at least some of them, it was possible by long search to find a short form somewhat resembling a stumpy form, but not having the somewhat indefinable characteristic appear- ance of the latter. (c) The Uganda strain, on the contrary, is typically dimorphic, i.c., besides the usual long forms of trypanosomes, stumpy forms are readily found, even in abundance occasionally, when the infection is well marked. Bruce (6), as we have noted above, states that this trypanosome has 26 per cent. of non- flagellated forms. The typical stumpy form we may define as a short, thick trypanosome, 12-14 yw, almost straight or slightly curved along one edge, while along the other the membrane is thrown into bold folds, there being no free flagellum, or at times a very short or doubtfully free one. It is thus easy to distinguish a typical Uganda specimen from a Zululand specimen, and, in fact, we may express the difference in this way, that it is * We desire here to express our thanks to these gentlemen for their kindness in sending us slides. P 2 190 Drs. Stephens and Blacklock. T. brucei, Plimmer and [Dee. 7, impossible to match a typical Uganda slide by any slide from the Zululand strain. We have stated in the above history of the Uganda strain that it was recently returned to us by Prof. Mesnil, who remarked in his letter that he had maintained it in mice (for nearly a year), and that it showed now very few “trapues” forms. This we have been able to verify in the films made from the infected mouse sent to us. But, as soon as we had re-inoculated it into guinea-pigs, it again showed numerous stumpy forms. But the same does not hold good for the Zululand strain; in guinea-pigs, as in rats and rabbits, the strain is typically monomorphic, 2.e. it does not show stumpy forms. We therefore conclude that the two strains, as we now possess them in the laboratory, are different. How, then, are we to explain these facts? There seem to us three possibilities :-— 1. That the strain we now possess, which we have been designating T. brucei, Zululand, is not this strain at all, but some other trypanosome inoculated erroneously during the course of imoculations extending over years. We think this view is untenable, for it would not explain the monomorphic character of the old slides we have examined, nor would it explain Laveran’s monomorphic trypanosome. 2. While Bruce may have been working with a dimorphic trypanosome in Zululand, and still has slides showing these characters, it is quite possible that the strain sent by him to England was something quite different. This is all the more likely, as Bruce successfully infected dogs from a variety of wild game, viz., wildebeeste, kudu, bushbuck, and buffalo, and, as Bruce himself states, “when Z. brucei was discovered in Zululand in 1894, it was naturally thought to be the one and only trypanosome in Africa,” and no suspicion arose at that time of a multiplicity of trypanosomes in native game. This is the simplest explanation, and the fact that Plimmer and Bradford do not describe or figure stumpy forms, and our examination of Dr. Plimmer’s slides had the same result, makes it probable that this is the true one. 3. That the strain originally sent to England was dimorphic, but that it has now become monomorphic. This may have come about in two ways :— (a) The strain originally was a mixture of a long trypanosome and a stumpy trypanosome, and the stumpy has now died out. If this explanation were valid, it would probably imply that 7. gambiense and other dimorphic trypanosomes were also mixtures. This we regard as a not impossible view, but one we cannot at present prove or disprove. (b) The strain was originally dimorphic (but not a mixture), and that it has now become monomorphic. If this were so, it would modify materially 1912.| Bradford, 1899, and Trypanosome from Uganda On Wit: our notions of specificity of trypanosomes, at least in laboratories. Of such a change we have at present not much evidence. We have noted, however, above that the Uganda strain kept in mice for a year was almost (but not entirely) monomorphic, but that in guinea-pigs it at once showed its normal characters. It is impossible at present to decide between these explanations. We come back, therefore, to the fact of which we ourselves have no doubt, viz., that the trypanosome that Plimmer and Bradford worked with, and which they named 7. brucei in 1899, is certainly now a monomorphic trypanosome, and is not the same as the trypanosome from the ox described under the same name by Bruce and others in Uganda. We believe, then, that the facts we have brought forward prove the non-identity of the Zululand and Uganda strains. In order to avoid confusion, we think it advisable that this Uganda trypanosome should be re-named. We therefore propose for it the name T. ugande. REFERENCES. (1) Bruce, ‘Further Report on the Tsetse-fly Disease or Nagana in Zululand,’ by Surgeon-Major David Bruce, A.M.S. (1896 ?), Harrison and Sons, London. (2) Kanthack, Durham, and Blandford, ‘ Roy. Soc. Proc.,’ 1898, vol: 64, p. 100. 3) Plimmer and Bradford, “ A Preliminary Note on the Morphology and Distribution of the Organism Found in the Tsetse-fly Disease,” ‘Roy. Soc. Proe.,’ 1900, vol. 65, p. 274. (4) Bradford, J. R., and Plimmer, H. G., “The Trypanosoma brucei, the Organism Found in Nagana or Tsetse-fly Disease,” ‘Quart. Journ. Micro. Sci., 1902, vol. 45, p. 449. (5) Bruce and others, ‘Reports of the Sleeping Sickness Commission of the Royal Society,’ 1911, No. XI, p. 147. (6) Bruce, zbid., 1912, No. XII, p. 24. (7) Laveran, “Identification et essai de classification des trypanosomes des mammi- féres,” ‘Ann. de l'Institut Pasteur,’ July, 1911, No. 7, p. 497. 192 The Action of Adrenin on Veins. (Preliminary Communication.) By J. A. Gunn and F. B. CHAVASSE. (Communicated by Prof. Francis Gotch, F.R.S. Received December 13, 1912,— Read February 6, 1913.) (From the Pharmacological Laboratory, Oxford.) It would be remarkable if the vein wall were the only tissue in the body to possess contractile fibres without a functionally important duty of con- tracting. Very little attention, however, has been paid to physiological alterations in the calibre of the veins, though such alterations may be of high importance in modifying physiological and pathological conditions of the circulation, and in explaining certain actions of drugs. The following investigation was undertaken in the hope of adding something to the knowledge of the contractile power of the veins; and, though the intended scope of the inquiry has not yet been completed, results have already been obtained which appear to be of sufficient importance to justify their being placed on record. Method. The method employed for recording the contractions of veins was similar, in essential respects, to that used by Cow* for determining the reactions of surviving arteries. In our experiments the veins were obtained from freshly killed sheep, and put, as soon as they could be obtained, into a Dewar flask containing oxygenated Ringer’s solution at 37° C., and so conveyed to the laboratory. A large water-bath, kept, unless otherwise stated, with a variation of half a degree on either side, at 36°C., held two beakers containing oxygenated Ringer’s solution at the same temperature. In one of these beakers the veins were put until required; in the other was put the part of the vein used for each experiment. For these experiments ring preparations were made. It is difficult to cut quickly and without undue manipulation of the vein a ring of absolutely uniform cylindrical length; but the rings used had a length averaging 15 mm., which varied not more than 4 mm. on either side at different parts of the ring. The ring was suspended between platinum hooks, the lower hook being fixed, the upper attached by a silk thread to a lever, which recorded variations in the calibre of the ring upon a slowly revolving drum. * Cow, ‘Journ. Physiol.,’ 1911, vol. 42, p. 125. The Action of Adrenin on Veins. 193 (a) Veins Remote from the Heart. All the ring preparations of (large) veins which we have so far subjected to the action of adrenin have responded by contraction. In the experiments illustrated in the accompanying figures the magnification of movement was the same in each case. Fig. 1 shows the contraction of an external jugular, External Jugutlar Vern. LN \ Adrenin 7 in 700,000 Fria. 1. fig. 2 of a mesenteric, vein. In regard to the amount of contraction produced they are, however, not comparable, because the temperature in the case of the mesenteric vein was higher (41° C.) than in the other experiments, in which it was 36° C. Mesenteric Vein T-41°C. Say , ge Adrenin 71 in 100,000. Fie. 2. The fact that veins contract under the action of adrenin renders it highly probable that veins possess venoconstrictor fibres supplied from the thoracico- lumbar sympathetic system. Several observers have concluded that the portal veins contain venomotor nerves ; but the presence of such nerves in other veins rests mainly on the evidence of Thompson* and Bancroft,+ who * Thompson, ‘ Archiv f. Physiol.,’ 1893, p. 102. + Bancroft, ‘Amer. Journ. Physiol.,’ 1898, vol. 1, p. 477 194 Messrs. J. A. Gunn and F. B. Chavasse. _—[Dec. 13, found that stimulation of the sciatic nerve in the cat and rabbit produced visible contraction of the superficial veins of the hind limbs. (b) Veins Near the Heart. Considerable interest attaches to the action of adrenin on the great veins near the heart. It has long been known that in the mammal the great veins near the heart, which correspond in some ways at least to the sinus venosus in the frog, beat rhythmically with the heart proper. One of us has on previous occasions made unsuccessful attempts to obtain a spon- taneously contracting ring preparation of the superior vena cava of the cat and rabbit. A renewal of this endeavour with the larger rings from the sheep and bullock has been equally unsuccessful, though it is not suggested that this may not yet be accomplished. What we believe to be a phenomenon of considerable interest and import- ance is that such a quiescent ring preparation of the superior vena cava near the heart can be made to beat vigorously and rhythmically by the action of adrenin. For these experiments the venz cave were removed along with part of the auricle, so that the distance of the ring from the auricle could be measured accurately. Fig. 3 shows the effect of adrenin, 1 in 20,000, on a ring preparation of the superior vena cava of a sheep, 6 mm. distant from the angle of its junction with the auricle. The kind of effect produced by adrenin on this venous ring is exactly like that produced by it on the whole heart. Though it cannot be postulated with absolute certainty in the case of the quiescent tissue, inspection of the tracing leaves little room for doubt that adrenin augments and accelerates the contractions of the ring. In the first place, therefore, this method of experiment affords evidence in favour of the view that the augmentor-accelerator nerve supply of the heart extends for some distance up the superior vena cava. Secondly, it throws some light on the origin of the rhythmicity of the heart. The ring of the superior vena cava is quiescent. (This is established not only by the absence of movements of the lever, but also by observation of the ring in a strong light.) Adrenin almost immediately induces powerful rhythmic contractions. The action of adrenin is a continuous stimulus to which the muscle responds by a discontinuous (rhythmic) contraction. Now the researches of Lewandowski, Langley, Elliott, Dale, and others have established with an unusual degree of certainty that adrenin acts on the myo-neural junctions of the thoracico-lumbar sympathetic system, and that its action is confined to these. Unless, therefore, the action of adrenin 195 S. un . Adrenin on Ve of 1on O The Acti 1912] TTT TY SPUu0Izas 4 Kn 000‘0e ULE UIMaIarp EF WL A 9 (La auyg ) DAD) DUA, s01.1L3adn re) . SY 196 Messrs. J. A. Gunn and F. B. Chavasse. _—‘[ Dee. 18, on the superior vena cava is unique, there are apparently only two ways in which its action on the quiescent superior vena cava can be explained. On the neurogenic hypothesis of the rhythmicity of the heart, it is possible to hold that the continuous stimulation by adrenin of the sympathetic myo-neural junctions so raises the excitability of the muscle that previously subminimal rhythmic impulses from intrinsic motor ganglia (hypothetically present in the ring preparation) are now adequate to elicit rhythmic contractions. On the myogenic hypothesis, on the other hand, stimulation of the myo- neural junctions of the sympathetic nerve causes the muscle to enter into the rhythmic activity which is inherent in it. All that can be said at present is that the latter explanation seems some- what more probable. If, however, we should be able to elicit, by adrenin, rhythmic contractions in a ring of the superior vena cava in which subsequent histological investigation can reveal no ganglia, then it would furnish a cogent argument in favour of the myogenic hypothesis of the rhythmicity of the heart. ; Further, this kind of investigation has afforded, and with further experi- ment, it is hoped will afford with greater accuracy, a physiological method of determining how far the rhythmically contractile tissue extends up the great veins, and where it merges into non-rhythmic contractile tissue. The difference in physiological reaction can be controlled by subsequent histological investigation. In the meantime it can be said that the rhythmically’ contractile tissue extends up the superior vena cava of the sheep for at least Superror Vena Cava (Sheep) 75mm. Fig, 4. 6—8 mm. from the veno-auricular junction. Fig. 4 shows that 16 mm. from this junction, a ring of the same superior vena cava which gave the rhythmic contractions shown in fig. 3 responds to adrenin by simple contraction. It must be emphasised that a negative result in inducing rhythmic responses is not in itself conclusive, because, even at 6 mm. distance from the heart, adrenin does not always induce rhythmic contractions. Under the conditions of the present experiments it is not possible for the veins to reach the laboratory always in the same condition of excitability ; and we have found 1912. ] The Action of Adrenn on Veins. 197 that a negative result is likely to arise when the muscle of the vein is subnormally excitable to electrical stimulation. In the inferior vena cava we have on no occasion been able to induce rhythmic contractions. The rings of the inferior vena cava have always Tesponded by simple non-rhythmic contraction, as shown in fig. 5. These experiments are being continued, and extended to the action of other drugs. Inferior Vena Cava OSheep) f Aa CLI 7 74 1706,000.. Fie. 5. Sunumary. 1. The action of adrenin upon ring preparations of veins remote from the heart is to diminish their calibre, as in the case of arteries. They, there- fore, probably contain venoconstrictor nerve fibres from the thoracico-lumbar sympathetic system. 2. The action of adrenin on quiescent rings from the superior vena cava near the heart is to cause them to beat rhythmically and powerfully. _3. (a) The accelerator-augmentor nerve supply of the heart, and (0) the rhythmically contractile tissue, extend up the superior vena cava for at least 6—8 mm. from the veno-auricular junction in the heart of the sheep. 4. The induction by adrenin of rhythmic contraction in the quiescent superior vena cava seems, on the whole, in accordance with the myogenic theory of mammalian heart rhythmicity. 198 An Apparatus for Iiqud Measurement by Drops and Applications m Counting Bacteria and other Cells and in Serology, ete. By R. Donatp B.Sc. (N.Z.), D.P.H. (Oxf.). ‘ (Communicated by Dr. L. Hill, F.R.S. Received November 21, 1912,—Read January 16, 1913.) (From the London Hospita Bacteriological Laboratory. Dr. William Bulloch, Director.) To promote drop-measuring in serological and bacteriological work, etc., the writer has devised a simple system of producing uniform pipettes, clean and sterile, which deliver uniform drops of any required size from 3 c.c, down to 1/200 c.c. or less, and has devised also simple forms of constant-pressure apparatus for use with the pipettes. The fundamental principle of his method rests on the fact that the size of a drop of a given liquid yielded by a clean pipette is determined by the outer circumference of the pipette at the level where the contact-edge of the drop clings round the glass—due allowance being made for the rate at which the drop is detached and the temperature. The pipettes, freshly drawn out from glass tubing in a Bunsen flame to a nearly cylindrical capillary form, are gauged in a wire gauge and cut off at the required sizes. The gauges used are such as the Starrett Morse Drill and Wire Gauge, which has holes ranging in diameter from 5°79 mm. down to 0°34 mm. Tubes larger than these sizes may conveniently be gauged by the Columbia vernier slide gauge. Capillary tubes less than 0°34 mm. may be gauged in a wire drawplate. In gauging, the capillary tube is pushed gently down into the particular — gauge hole required and is then cut off—preferably at the upper surface of the plate, so that the dropping-point shall not come into contact with any trace of greasy matter which may remain on the cleaned gauge plate. To ascertain the size of drop yielded by such a dropping-point an adjustable constant-pressure apparatus was devised. This (fig. 1) consists of a straight tube of 3 to 4 mm. internal diameter and of such a length, e.g. 50 or 60 cm., that the free air space within shall be amply greater than the capacity of the pipette employed. The tube is carefully cleaned, washed finally with distilled water, and dried with grease-free cotton wool drawn through on a thread. The ends are opened out slightly funnel-shaped to facilitate the ramming in of an inch or so of pure cotton wool, which is required to retard An Apparatus for Liquid Measurement by Drops. 199 the passage of air and to prevent effectively the escape of any of a column of pure dry mercury, 20 cm. long, which is introduced to act as a plunger. The tube is arranged as shown in the figure. The right-hand end is joined by a short rubber tube to the upper end of a pipette, which is supported by a clamp, and the left-hand end, attached to a screw-clamp, can be moved up or down the tall stem of a retort stand. The stem of the pipette has an internal diameter of 1 mm. or so, and the lower of its two calibration marks is 1 or 2 cm. below the bulb. Then SEUELEDD /i I | | i 1 or 2 cm. below this the pipette is jomed by a short piece of cleaned bicycle valve-tubing to the upper end of the capillary dropping nozzle. For special work the junction may be made by means of a sleeve of good cork or by grinding or fusing the nozzle on to the pipette point. The liquid friction in a small dropping nozzle ought to be such that the head of mercury employed may be great enough to render negligible the loss of 2 or 3 em. head of water as the pipette empties. The acceleration of the falling mercury down the well-throttled sloped tube is negligible. To carry out a dropping experiment the mercury is first brought to the 200 Mr. R. Donald. [Nov. 21, right-hand end of the tube, or, if the dropping nozzle has great friction, to within 1 or 2 cm. of the end. The vessel of liquid is raised to cover the end of the nozzle. The left-hand end of the mercury tube is then depressed till the pipette is nearly full) Next it is gently raised until the upper meniscus of the liquid rests at the upper calibration mark. Finally the vessel of liquid is lowered. Now the end of the mercury tube is raised till the drops fall at the required standard rate, say one per second. The height required for the dropping- point in use is marked by a sliding ring. Then the pipette is refilled, the mercury tube is at once elevated to the required point, and the drop-count at the uniform standard rate is observed. If necessary, the mercury tube may be slightly lowered at the last drop to allow estimation of the fraction of a drop. The drop-counts for distilled water thus found are given in the following table. The fourth column contains the quotient drop-weight in mgrm./diam. in mm., eg. for Morse gauge 80, 1000/131/0°340 = 22-45. These quotients . are seen to form a fairly uniformly falling curve as the dropping-point increases in size. . Drop-count. Distilled Water. | Morse gauge No. | Diameter. Broppconan (UG: We. in mgm | per 1 c.c. Diam. in mm. | | mm. 79 0 :366 122 0 22-4 78 0-406 112°9 21°9 79 0-633 73°5 21 “4 68 0-787 58-5 21°4 62 0-965 | 51-6 20-1 60 1°016 | 49°8 19'-9 57 1-092 47-9 19°1 Bd, 1 ‘397 38 9 oo 43 2 -261 25 °5 17 *4 33 2-870 20°3 17-2 28 3-569 16 °7 16°8 At different rates of dropping the drop-count differs as shown by the following observations :— 1912.] An Apparatus for Inquid Measurement by Drops. 201 | Diameter. Mae No. of drops obtained. mm. Dropping-tube (throttled on separator (14°4 cylinder with Mariotte’s tube) from 10 c.e. ASH cr a a Dropping-tube similarly fitted ............ 8-4 % Or or e Mercury tube pressure, Morse 33......... 2°87 from 1°125 c.c. NPRWONHO SwoNIS to wo a Pa Morse 56......... 1-181 ~I rs Or 0 5 0 0 0 0 2 0 23 °2 a3) 3 6 0 0 0 3 5 CAOEN Or e) 5 on Morse 80......... 0°34 from 0°5 c.c. Or for} ~J 9 IDE MLO scece 0°29 LN) ve} Or SAWSOGDH SSH w oo : Je) S wo ee QD Hand-teat pressure, D.P. X ............... 0°25 io) ror) SSO BNFKESSO SSSS lores ms = fo) ior) From the above observations it may be seen that just in the cases where fine accuracy is most desirable—namely, in measuring by hand one or two drops—these small drops have in their favour the peculiarity (pointed out by Ollivier) of being practically constant in weight at all expulsion-rates slower than one drop per second. Such rates can be easily and reliably secured by the use of a hand mercury-plunger tube—a miniature form of the mercury tube described above. The narrow limits of drop-rate required for large drops, eg. of 1/5 to 1/4 c.c. in the Wassermann test, may be secured by the use of the arrange- ment shown in fig. 2. A separate cylinder of convenient size has a small Mariotte’s tube fitted by a rubber cork through the upper aperture. Various dropping-tubes, smaller to telescope inside, or larger to telescope outside the stopcocked 202 An Apparatus for Liquid Measurement by Drops. tube, may be fitted on by a ferrule of washed iubber tubing or of good cork. A dropping-point to give watery drops of + cc. may be formed of tubing 144 mm. external diameter. The dropper is held in a stand at a height convenient to allow a rack of test-tubes to be slid under the dropping- point. The drop-rate may be regulated as desired by fitting inside the dropping- tube, and within forceps-reach of the orifice, a throttle made of suitably drawn out capillary tube. The author has found the apparatus efficient in the following applications :— (1) Comparison of surface-tension of liquids, in quantity as little as 4 c.c., eg. in testing the value of the meiostagmin reaction in syphilis. (2) As the drop yielded by one pipette differs by not more than a fraction of 1 per cent. from the drop similarly yielded by another pipette similarly gauged in the same hole, the method may be used, with successive fresh pipettes, for rapidly and accurately measuring off small quantities of any liquid, from 0°25 cc. to, say, 0:004 cc. of watery liquids of high surface tension, or even smaller quantities of liquids of low surface tension, for instance, 1n :— (a) Measurements of various small quantities, in e.g. Wassermann reaction, micro-Wassermann re- action, and other complement-fixation tests; Widal reaction; alkalimetry, acidimetry, and other volu- metric tests ; testing cultures and disinfectants ; and in pharmacy. (6) Direct estimation of cells, numerically, and roughly qualitatively, in small drops of cerebro-spinal fluid, dried on a slide. (c) Blood-count, red cells and white cells, from drops of diluted blood, similarly dried. (d) Direct counting of bacteria in small dried drops of diluted vaccines, distilled water, domestic water, sewage effluent, diluted milk. (6), (c), and (d) yield permanent preparations. : : ; 203 A Preliminary Report on the Treatment of Human Trypano- somasis and Yaws with Metallic Antimony. By H. S. Ranken, M.B. Glasg., M.R.C.P. Lond., Captain R.A.M.C., Member Sudan Sleeping Sickness Commission. (Communicated by H. G. Plimmer, F.R.S. Received December 23, 1912,— Read February 6, 1913.) (From the Yei Camp, Sudan Sleeping Sickness Commission.) The use of precipitated metallic antimony, in a state of finest division, was devised by Plimmer for the treatment of trypanosomiasis, and the results of his and other experiments with this substance have been published.* After a long series of experiments on subcutaneous and intramuscular injections of this form of antimony suspended in water, oily media, egg yolk, etc., all of which caused great irritation, he found that it was possible to inject it intravenously with safety, and without causing any irritation of the tissues. A large number of ‘animals were cured of trypanosomiasis by this means, and in May, 1910, Major W. B. Fry gave a dose intravenously to a late case of Kala Azar, thus demonstrating that it could also be used safely in this manner on human beings. Captain R. J. C. Thompson, R.A.M.C., gave this preparation by intravenous injection to the cases first admitted to the Yei Sleeping Sickness Camp between January and March, 1911. LHighty-one injections were given to 38 cases, but pressure of administrative work prevented these cases being fully treated and investigated, and they were all trausferred to atoxyl treatment. Captain Thompson states that from a clinical standpoint some of these cases showed great improvement. Later I had the honour of being appointed to the Sudan Sleeping Sickness Commission for work at Yei in the Lado Enclave, and since October, 1911, I have collected a series of cases treated by metallic antimony alone and in conjunction with other drugs. This report is consequently only preliminary ; but it demonstrates that antimony in this form is a safe drug to employ in the treatment of sleeping sickness, if used with reasonable precautions, and that the results obtained so far certainly call for extended use and further investigation. During the past year 76 newly admitted cases have been treated with antimony, alone and combined with salvarsan and atoxyl, and shorter courses of antimony have been given to 143 old cases previously treated with atoxyl, * “Roy. Soc. Proc.,’ B, vol. 80, p. 483 ; B, vol. 81, p. 354 ; and B, vol. 83, p. 140. VOL. LXXXVI.—B. Q 204 Captain H.S. Ranken. Treatment of Human [Dec. 23, atoxylate of mercury, etc. Over 1400 intravenous injections have been given, and three deaths have occurred which can be attributed to the treatment. All these fatal cases were amongst the first 150 injections. The method of administration is the same as that of an ordinary intra- venous injection of saline solution given hydrostatically through a needle of somewhat large bore.* The dose of antimony is stirred with about half an ounce of normal saline in a small glass mortar and becomes a temporary suspension. Two ounces of saline are then poured into the funnel and tubing, the needle is inserted into any vein in the forearm and the clip opened. As soon as it is seen that the saline is flowing freely into the vein, and there is no swelling round the site of the puncture, the suspension of antimony is poured into the funnel and the mortar is washed out, with a little more saline, into the funnel in order to leave no residue. The antimony is allowed to run into the vein, the funnel being gently shaken from time to time, and, when it is on the point of becoming empty, more saline is poured in. The window in the rubber tubing should be watched, and after all traces of antimony have passed it some more saline is allowed to run in to clear it out of the part of the tubing below the window; the clip is then closed and the needle withdrawn. About six ounces of saline seems to be a sufficient quantity, and the time occupied in giving an injection varies from three to seven minutes— depending on the calibre of the vein and the bore of the needle employed. I have found it possible, without any hurry, to give as many as 10 injec- tions in an hour; I mention this to show that this method of treatment is feasible on a large scale. The question of dosage is one of considerable importance. The dose usually given was one grain. Large doses up to three grains have been given, but these seem to be attended with risk and have been given up for the present; it is probable, however, that in good, selected cases one and a half, or even two, grains may be safely given. Several cases of extreme susceptibility have been met with, however, so that the initial dose should be one grain. The interval between the doses seems to be of the greatest importance. Four days would seem to be most suitable, but injections have frequently been repeated on the third day after a dose ; they should certainly not be postponed any later than the fifth day. A few of the earlier cases were treated with weekly doses; two of these were very susceptible and were unable to take doses at shorter intervals ; both cases have relapsed. The usual course of treatment has been five doses at intervals of four days, * The size of the one I have used is No. 19 of Imperial Standard Wire Gauge internal diameter and No. 15 external. 1912.] Trypanosonuasis and Yaws with Metallic Antimony. 205 covering a period of from 17 to 20 days; after an interval of six weeks a short course of three doses extending over nine days has been given. The table on pp. 206-7 gives in shortest form the results of the treatment with antimony alone. Thirty-five cases have been so treated: in 15 of these there were no symptoms other than enlargement of the cervical glands, except varying degrees of debility ; 17 had also fine tremor of the tongue and five cedema of eyelids. The cases were not especially selected for treatment, but were taken just as they came. Up to the present it has not been possible to procure susceptible animals for inoculation from the cases treated, so microscopical examination has been the only available method of controlling the results. Blood and gland juice have been examined in all cases. Three observers have undertaken each examination, and the total time of each examination has been one and a half hours. Many of the cases have been repeatedly examined. Of these 35 cases four are dead and two have deserted. Try- panosomes have reappeared in the blood of four cases and another has relapsed clinically. The remaining 24 are all improved and all microscopical examinations are negative up to date of writing. (a) Deaths.—Two only of the four deaths can be connected directly with the treatment. Both were advanced cases and had been given larger doses than usual. No. 2,a boy of 10 years, had 2 gr. on one occasion and 13 gr. twice. He showed no symptoms till 48 hours after the last dose, when meteorism appeared accompanied by acute abdominal pain, and he died seven hours later. I have seen a somewhat similar death here in a case treated only with atoxy]l. No. 24 was also an advanced case; he took two injections of 1 gr. without discomfort and then had a dose of 2 gr. Twelve hours later epileptiform symptoms set in quite suddenly; the patient became unconscious and died six hours later. Of the other two deaths—No. 7 was due to broncho-pneumonia; the patient was debilitated and broncho-pneumonia was very prevalent at that time. No. 47 died from an epileptiform attack, but had not had any form of treatment for a week. (b) Deserters.—One, No. 38, had recently completed his course of treat- ment and had shown great improvement. The other, No. 3, belonged to a district 100 miles from Yei. He had been without treatment for five months and had done very well—taking large doses without untoward symptoms. All examinations had been negative, though repeated very frequently. (c) Relapses—No. 27 is a clinical relapse. He was a soldier and his Q 2 Treatment of Human [Dee. 23, Captain H. 8. Ranken. Le OATZEHOU 9.18 206 pimp jeurds-orqoreo pur ‘spue[s ‘poog «¢ [OFUSUL,, SI pus “emIepa MoU sey FT ‘QUBISISSR SB YDB PTROd oY 4BIy Os ‘yuouL -SAOAdUUT BAIS SLM aLOTy sYyMOUL $e Oy ao feduqs pxlyy oy} UO SUISIOA SBA osteo STUY, “* ssay PONT, - Pesce Gooey E. Ghee +++ |910py] ‘We | “ 4 ‘poaoadury | aeqyst1q yonzy | to; yeus yonTy = asco. (‘ad }) sesop ¢ ++ Stacie OL ‘a 4J9] EO pavy BS ‘porwoddusip comtery onsuoy, “pood {aq 8 poor | [pects au0 <{uG =) iieosbemi5es aE a Oy ati i a. WT ‘esOp “I9-gZ LoyyR | Ieee: | sMMOY ZT Jnoge aanzias utozydopido ygra | ped ‘uwoxyrpuos aood ut ‘esvo poouvapy | (a8 Z09T) “ @| +++ pects 3 W “paavaddustp “ ¢ yor ‘TOULOL} anso} pry *poos Kao A tee ene eee poo peonpox ou = _flecana ween eee ec G 46 4b oh + “c “WH piry as *paavoddestp Yor a ‘eua pa pry wee ae poos LECH pus [reug == | fleSaaon eee ee eee ce G + + + + + ce “WH a1qe ae “UOIyIpuod rood ‘PIO eee ee ee “ec -so1jou ATao.uvag ee | OocoCG Ca 6 G + + + + + + “ a I ce e | ‘poos A190 A See eeenerees «“ Ts 2 Pee ee seen es ee se G fy + 79 W | pee ‘poos £10 A. Seen ernaee poon powvoddustcy — | |bonbonbenccccs 5 “ G fot + + 73 oth ‘quoyvaord sum yor “etuommend | -oyouorq Fo perp * peqvatiqep pue PIO _ Sie CET Ye tee | gta ae ae Beep ait ‘osdvyor B pey oy |asdvjoraeqye “* @ woT_M ‘stQUOUE 4G LOF UOTYIPUOD poos ut ac poureureayy ~“Auourgue 07 o[qiydeosns A100 A ronoon (sty py (FG) par + sf ‘W "[Joau oginb | . SI pus “plryo SULA, @ Jo pauyuos meaq Ajqueoer seAY “MOyTpuod pood TOA UP of poavoddustqg, = See ae we 8 + + af aqqe | “c g *poaoaduit yonur £10 A bee wenene tae (73 -amjound qsnp aa Jaseees (Ges) ¢) (a3 P ae + + ce ‘WH seteee (as T) “ g ‘ABMB SOTIUL OOT peAaty oy sv ‘paquesacy jus poo | porvoddesiq = | (‘a3 ¢ 09 1) Sesop PF + +r qUpy| ‘ ‘esto poousape AraA @ £ quouIyBery (48 % Jo T pur jo quemeouemmos «047n skup eT pal = (crs. Sy Jom) “Se ak rick ar oP ae Or ‘A oAT}eS | -ou ommyoung yee ‘poystanou teqygoq S peaordoat yonyy | poaoadu yong | ‘xoypeus you Tt — hae ae eae sssopg| ++ + ++ (Spy | oom | "[87 9 TN. “‘SpuBLO pucys pue ‘UOIyTpuoD [Bteme4 | poole, “quoulyvory, < eee ‘spuBly | ‘ody | “xeg | ‘MOTYIPMOD yuosetg ,Ajuo AuommMUy YIM poyeory, seseg Jo Jourysqy 207 umony. nd Yaws with Metallic Ant SOMILASUS CL 1912.] Trypano QJBIMOIB UR OATS JOU OP sX{O}JOVF osoyy sv ‘sautosouvd{.y Jo coquinu to spue *poaordear youu { poweedde -sIp ower}, onsuoy, “mojyyeg aq | jo WoYsessns ye sosop yyy pur paryy Q8cy oL0Jog UAAIS sv “As T fonTq UIpMpOT ye Say ‘poo te “noon “quouteaord at qystg ‘porvoddestp yorpm “eumopa pep ot 10990g7 *qUOTI4BEL} O1[9 FIM pozoouU0d Ayqueredde gou ‘youqqge utcozrydozide UB UL OSOP ySBT 1O¥FR Yoom wv policy ‘ens PoouUBApPe UT ‘URI poyLIonute plo UW ‘uoT}IpuoD ur posordur yonur mou furyg Atoa ourng jo ss “poo te poo ‘syquow $z 109; porvoddvor sourosourd (xy, *poaordury Heenan et eee i3 “poos £10 A Jenene ree eee pooxy “exe e) -u0d pedaordur Youu Ul quoTI4ReL} Jo OSINOD 1O9JB SYOOM XIS poytosop quoyegy ‘syguour omy «0978 = patvoddrox soutosound day, *posorduay fefeevee cveen “ “poos 4.10 4 ss “poaordunt pony «8 th *patvoddesrp come.19 ensu0y pus evutepapD “quoutesord unr jvats paMoys pu “UOTIpuod pq UL oUED poo ‘AuOUITZUB Of OATYISMOS L194 “pooyq ur urese porvodde somosourdday, ‘esOp ysB] 109} syquow omg osdepoa puzf | *poos AiO A beeeeeenees on “ec *osm00 qsiy aoyye sigryouosq peyT “pood faq ve poor “BuLepa YyLM ‘aoripuod aood ut su AQ ——— eee a a el 7 poos A104 *asBo 09 Jo AGILOAOS Ol[4 FO WOTZBOTPUT [3 oy4 Jo ozIs O44 07 patyouyqu oq you pjmoys oourgtoduat yonut OOg, » sete eeees a3 Teug TOT[BUIS TON FAL iis “ ToT [VULs YON IAL TOT[VMUs TONAL “ qyeus £10 4 LOTUS YONTA arqu -901j0uU = JON 4jos agin’ atquydeo -rod oo £yuQ « « “guos A].09 yy pley pue [[euig ce g eee eoeresccaces g i g Secon) « G “e g seer e reese ii g i) _ g € . . e g “ g dee eeenes beens “ G “ eg eeveee tet eeenes “ G “ g see ee wrens « @ ee seeeee wee g “ g raepesaneerenes “ Gg sc g eee eee eeneeseee “ g cc g tee eee wee on be G “ g Goong pee eeenes «“ g « g ween ee eres oe “ G “ e pete ween sosop G 77 e Eee eee e ne nee 4 4g € “ ro see eee eee eee ees “ ry “ @ ‘ @ Sree e eee ensene ‘ € “ g . soe (cad ¥) “ e ++ +++ *YOOM OO JO S[VALojUT FY 4+ ++ + + +++ ++ | |W tas “W « W ee . 6c . “ . “ . “ . “e ‘WH “cc (73 “ Ww its “A WEPV | “IW — . ~-— oe Tg (0) ] 8P i oF oP iz IP OP 8& LE 98 gE PE €& Té 08 66 82 208 Captain H. S. Ranken. Treatment of Human (Dee. 23, condition on arrival at the Camp was alarming, and he was quite unable to walk. A full course of antimony, in doses of 1 gr., was successfully given and there was a great improvement in his condition. The large masses of glands were very greatly reduced in size. The tremor was unaffected, but he became much stronger. He was able to act as a “native assistant” in the laboratory and a “ headman ” in the camp. His improvement was maintained for three and a-half months, when suddenly he relapsed and became a third stage case. Blood examinations have been negative. The cerebro-spinal fluid was examined also and the cellular elements were increased, but’ trypanosomes were not found. In four other cases trypanosomes have reappeared in the blood. Two of these were very susceptible to antimony and were unable to take doses at shorter intervals than a week. After the relapse attempts were made to give them another course of antimony, but it had to be given up owing to their unusual susceptibility to the drug. The other two had heavy infec- tions, but did not seem to improve. They have all been transferred to atoxyl. (d) The other 24 cases have shown improvement. In three it is slight and these are being kept under close observation. In some cases the improvement has been very striking; in others who were admitted in fairly good condition, it has not been so evident, but the whole appearance of the patients is different. After a course of antimony there is not uncommonly depression and debility. This, however, passes off in a few days; the patient becomes more active, feels and looks better, and loses the languor that is so often seen in patients just admitted. They put on weight and the skin becomes healthier; this is no doubt due in part to the regular feeding in the sleeping sickness camp, but the improvement would not occur were it not initiated by the treatment. In some cases the tongue tremor has disappeared under treatment. Probably some other cases will relapse, so these cases are being kept without further treatment at present, as it is desirable to test the effect of this treatment with antimony alone. Il. Antimony and Salvarsan. Ten cases have been treated with these two drugs, but a regular course has not been carried out; the salvarsan was given, in the majority of cases, when a patient was unable to take a complete course of antimony. No information has been obtained as to the best line of treatment to be followed with this combination of drugs, but a further series has been com- menced. The results up to the present are: Eight out of the ten cases are very well and have shown definite improvement ; one deserted, but was in very good 1912.| Trypanosomiasis and Yaws with Metallic Antimony. 209 health and had been without treatment for four months. She was one of the few who had a complete course. The tenth case had one injection of salvarsan and was unable to take antimony on account of extreme sus- ceptibility. Atoxyl was substituted, but the patient developed toxic symptoms. She is now having small doses of atoxyl, but is getting steadily worse. The salvarsan has been given both as an intramuscular injection in olive oil, and by intravenous injection in alkaline solution. III. Antimony and Atoxy!. The patients in this series have been treated as follows :— (1) Five doses of antimony 1 gr. at intervals of four days. (2) Atoxyl 5 gr. every three days for 40 days. (3) Three doses of antimony 1 gr. at intervals of four days. (4) Atoxyl 5 gr. every three days for one month. There is no interval between these courses, so the patients have continuous treatment over a period of rather more than three months. An interval of a month is then given and atoxyl treatment is then continued, as a tonic and precautionary treatment. Thirty-one patients have been treated; but three were unable to take the complete course of antimony. This is the most recent series, and the three months’ course has just been completed; it is too early to speak of results. At the time of writing all have done well; the improvement, in many instances, has been striking; some, however, were advanced cases, and it is not expected that the improvement can be maintained permanently. IV. Old Atoxyl Cases, Treated with Short Courses of Antimony. After we had seen that intravenous injection of antimony was bringing about improvement in the newly admitted cases and could easily be carried out on a large scale, it was decided to give short courses of antimony to the old cases that had been under treatment with atoxyl, atoxylate of mercury, etc. The majority of the patients so treated have had two separate courses of three doses of antimony, 1 gr. In some instances, patients have not had all these six doses on account of susceptibility, intercurrent affections, bronchitis, etc., but 813 injections have been given to 143 patients. In this series there has been one death. The two fatal cases in the first series had doses which have been found to be inadvisable in advanced 210 Captam H. 8. Ranken. Treatment of Human [Dee. 23, cases. In this case only two injections of 1 gr. had been given. The patient was a man in good health; he showed slight depression after the first dose, but had quite recovered and was ready for a second dose four days later. He was sick after this and soon developed alarming symptoms, and died. There were no late effects in any of the other patients. SYMPTOMS FOLLOWING THE INTRAVENOUS INJECTION OF ANTIMONY. Antimony has a powerful depressant action and, as the preparation employed is very active, it is only to be expected that a course of treat- ment should cause the appearance of some symptoms. It has been found that some patients have a high degree of susceptibility to antimony, and they have suffered somewhat severely ; also, at the beginning of this work, when attempts were bemg made to obtain some information as to the dosage, larger doses were given and the effects were, in some cases, more marked. With further experience symptoms of any degree of severity have become very rare. The following are the symptoms which may be seen in patients who are not unusually susceptible :— 1. Fever: The “ réaction thermique.” 2. Pulse: Is not much accelerated; but, so far as can be ascertained by digital examination, there is a fall in blood pressure. 3. Diuresis: There is variable diuresis. 4. Cough: This occurs in the majority of cases. It begins a few minutes after administration and lasts for 5 or 10 minutes. Very rarely it has persisted for 12 or 24 hours, but only in cases which have had slight bronchitis. 5. Pain in the xiphisternal region: This is a common occurrence and is most severe the day following an injection. It is probably of gastric origin. 6. Headache is not uncommon, but is not severe. The following symptoms have only occurred in cases of great susceptibility to the drug :— 7. Sickness and vomiting: This has occurred six times ; once it occurred in a fatal case (No. 32a). 8. Fainting : Only one case. 9. Meteorism: Was seen in one case (No. 2), which was fatal. Captain Thompson saw a similar case 18 months ago. 10. Herpes: There have been seven cases. In six it was seen on the face and lips, and in the seventh along the ribs. This condition is probably similar to the herpes that occurs in certain forms of arsenical poisoning. LO 2) Trypanosomiasis and Yaws with Metallic Antimony. 211 11. Stomatitis : Two cases; one was vesicular, but the other was more severe and ulcerated. In both cases the lesions were limited to the anterior portion of the hard palate, and the condition may perhaps have been herpetic. It is of interest to note that very similar symptoms were observed in experimental dogs treated in this manner. Short mention must be made of the effect of antimony on the temperature and upon the leucocytes. I. Temperature. A “réaction thermique” has been described, occurring about 20 minutes after injection of salts of antimony. This subject may be of greater interest owing to the large amount of work done on the subject of salvarsan fever. Temperatures have been taken in over 100 cases: (1) at short intervals on the day of an injection ; and (2) morning and evening temperatures for three days after treatment. There is no effect on the temperature till two or three hours after treat- ment, when it rises 1° F., and at four hours after treatment the average rise is 14°F. The same evening, 10 to 12 hours after treatment, some cases have fallen to normal, while others have gone up to over 101°. For the next two days there is an average evening temperature of 100-2° and 99°7°, and the temperature returns to normal. The results are identical in treated and untreated cases ; trypanolysis and the disposal of dead trypanosomes do not, therefore, cause this rise of temperature. The initial rise may be due to antimony, but the temperature for the two following days is probably due to the saline solution. With the apparatus available here it is impossible to) get pure distilled water; it would be of interest if some injections could be given with water twice distilled in a good apparatus. The temperature does not seem to have any effect on the general condition of the patients. Il. Leucocytes. The action of leucocytes on this preparation has been described.* Blood films, taken from animals after treatment, were stained and the leucocytes found to be crammed with the minute particles of antimony. They show a great avidity for this preparation, which they discharge, presumably in soluble form, into the blood-stream, the action being thus spread over a longer period. A series of 200 leucocyte counts has been made to determine the effect of * “Roy. Soc. Proc.,’ B, vol. 83, p. 140. 212 Captain H. 8. Ranken. Treatment of Human (Dee. 23, treatment on the numbers of leucocytes. The cases selected were all under treatment, but some enumerations were made on untreated cases and the same changes observed. Immediately after injection of antimony there is a considerable reduction of the leucocytes in the peripheral blood and in half an hour they have fallen to 60 per cent. of the number obtained on enumeration just before adminis- tration. In some cases the count is still lower an hour after the injection. From this point the leucocytes began to rise in number and in the majority of cases have returned in four or six hours to the level of the first count. This increase continues and 24 hours after treatment there is an average count of over 16,000; this is maintained for another 24 hours, but four days after treatment the numbers have fallen to approximately the original count. The estimations were continued over a series of three doses, and showed the same changes after each. III. Action of Antimony on Trypanosomes. The exceedingly rapid action of the salts of antimony in man has been reported (4), but it was not anticipated that the metal would have an almost immediate trypanocidal action. Trypanosomes were never found in eases examined a few hours after treatment, so observations were made to determine the time required for an intravenous injection of antimony to clear the glands of trypanosomes. Only heavily infected cases were selected, a case being considered suitable if 10 trypanosomes were found in five minutes in the gland juice, but in many instances they were much more numerous, two and three trypanosomes being often seen in one field. The time was taken from the moment the antimony suspended in saline solution entered a vein in the arm, and at periods from 3 to 30 minutes after this time glands were punctured and wet preparations made. These were examined both by dark ground and ordinary illumination. In all cases 15 minutes was allowed for each film, and in many cases the search was prolonged up to 30 or 45 minutes. After 3 minutes (slide made before the injection is completed).—Trypano- somes are still abundant, but some are already very obviously affected by the treatment. They show greatly exaggerated activity and dash rapidly across the field, so that they are difticult to keep in view. Frequently a trypano- some becomes anchored by the blunt end and lashes about most vigorously. This exaggerated motility soon passes into a state of fatigue, when the trypano- some wriggles more slowly and more and more feebly till movement ceases. Some retain their normal form, while others become swollen and bloated. 1912.) Trypanosomasis and Yaws with Metallic Antimony. 213 Occasionally a trypanosome comes rapidly to a standstill and dissolves away so that only a haze of protoplasm can be seen left behind. After 5 minutes.—The exaggerated motility is not so frequently observed, but the trypanosomes are more often anchored and may be seen lashing about in an extreme state of activity. They appear to be somewhat reduced in number. After 7 minutes—Trypanosomes are more scanty. Exaggerated motility is not a feature of these preparations, but all the other changes occur as in the earlier preparations. After 10 and 15 minutes—In all preparations taken at these times a considerable search was required to find a trypanosome, but in the majority of cases one could be found in a search of 10 to 20 minutes. After 20 minutes.—Very many preparations have been examined and have invariably proved negative. An intravenous injection of antimony therefore kills the trypanosomes in the circulating blood in 20 minutes. ON THE TREATMENT OF YAWS BY THE SAME METHOD. Some cases of yaws have been met with in the course of this work. In view of the successful results published by Strong, Castellani, and Alston, the first three cases were treated with salvarsan. One was given an intramuscular injection of 0°6 grm. in olive oil; the lesions—plantar ulcers—were healed in three weeks. The two others had an intravenous injection of 0°45 grm., they showed several small patches on the face and trunk and responded much more rapidly to intravenous treatment. I had been much impressed with the rapid trypanocidal action of this metallic antimony, and decided to treat a case of yaws with this drug. At first doses of 1 er. were given and the condition improved, but not very rapidly. With a larger dose the effect was much more striking; 14 gr. seems to be quite efficient, but 2 gr. have been given to the last four cases. None of the 10 cases manifested the hyper-susceptibility to antimony that has been seen in sleeping sickness, and there have been no after effects. In an adult of fair condition 1 gr. should be given as a first dose, and doses of 14 er. or 2 er. repeated twice with intervals of four days. I have sometimes shortened the interval by one day. Three doses, I believe, is a quite sufficient course of treatment, but in the majority of our cases a fourth dose has been given to ensure, as far as possible, a permanent result, as most of these cases have been collected from different villages and pass out of observation when they are discharged from hospital. The antimony is administered as described above (p. 204) for trypanosomiasis. 214. Treatment of Human Trypanosomasis and Yaws. The following short notes give an abstract of the cases :— 1. Lesions small, but generally distributed with septic or impetiginous patches in the beard. Four doses of antimony were given—two of 1 gr. and two of 1% gr. Lesions were all healed in 12 days. 2. Lesions general and larger than in preceding case. Treatment, four doses of 1% gr. All lesions healed in 11 days. 3. Extensive ulceration of scrotum. The ulcers had moist surfaces and an exceedingly offensive discharge. Treatment, four doses of 14 gr. All healed in 11 days. 4. A large crusted patch on perineum and several small lesions on face. Treatment in four doses of 14 gr. Healed in 10 days. 5. Extensive ulceration on soles of feet. Treatment four doses of 14 gr. ; local treat- ment, perchloride of mercury 1/1000 as a lotion. Quite healed in 14 days. 6. Discharging ulcers between toes. Treatment, four doses of 14 gr. MHealed in 14 days. 7. Generalised eruption. There were large confluent patches all over the face, trunk, perineum, and limbs. There was also a large primary ulcer on the scrotum 3 inches in diameter, with very foul discharge. The patient was debilitated, and three doses of 1 gr. were given. There was some improvement, but it was very slow, so a dose of 2 gr. was given, and all the lesions were completely healed in six days, except the ulcer, which was reduced to less than 1 inch in diameter. 8. Some patches on the face with thick raised limpet crusts, and many small lesions on the scrotum, perineum, and buttocks, with moist discharging surfaces. Treatment, one dose of 14 gr., followed by two doses of 2 gr. and a fourth of 14 gr. All healed in 10 days. 9. Small patches on scrotum, penis, and perineum. Treatment, one dose of 1} gr. and two of 2 gr. All healed in 10 days. 10. Small patches on scrotum, penis, perineum, and axille. Treatment, one dose of 14 gr. and two of 2 gr. All lesions healed in 11 days. In all cases the lesions were characteristic of the various stages of yaws, from minute vesico-pustules up to confluent, crusted lesions, rupioid patches with limpet-shell crusts, or late plantar-ulcers. The diagnosis was confirmed in the earlier cases by examination of scrapings from the deeper parts of the lesions. Films were examined by dark ground illumination and spirochetes were found, sometimes in large numbers. After the first dose the discharging ulcers showed signs of drying up and they skinned over rapidly. In 48 hours a distinct improvement was seen ; the crusted lesions had shrunk somewhat and no longer contained fluid. The yellow colour disappeared and was replaced by a pearly grey; the underlying raw surface healed very quickly, and soon there was only a desquamating flake representing the site of the lesion. Accompanying the local changes there was improvement in the patient’s general condition. The number of cases treated is small, but they have been uniformly successful. The treatment can easily be carried out on a large scale, and it may be possible to cure large numbers of persons affected with this most unsightly, and hitherto long-enduring disease, without causing too Liberation of Ions and Oxygen Tension of Tissues. 215 great a drain on the funds allotted to medical work in these colonies, etc., where this condition is prevalent. This preparation of antimony was tried successfully in a few cases of syphilis* by intramuscular injection, but the pain was so severe that the method was not continued. I venture to suggest that, as intravenous injection of the metal has proved feasible, a further trial may be warranted in this other spirochetal disease so closely allied to yaws. The Inberation of Ions and the Oxygen Tension of Tissues during Activity. (Preliminary Communication.) By H. E. Roar, M.D., D.Sc. (Communicated by Prof. C. S. Sherrington, F.R.S. Received January 10,— Read February 20, 1913.) (From the Physiology Laboratory, St. Mary’s Hospital Medical School.) The hypothesis, that when cells become active ions are liberated, is frequently quoted in physiological writings, but measurements of the ionic changes are not given. The present research is an attempt to measure the ionic changes during tissue activity. A knowledge of these changes is desirable in order to compare the action of ions on colloids outside the cells with the changes that occur inside the cells. On turning to the physiological literature one finds that little or nothing is given in the way of measurement of the ionic concentrations with which biology has to deal. Thus, neither the recent work on electrobiology by Bernstein, nor that on the hydrogen ion in biological processes by Sorensent contains any reference to the measurements of ions in tissues during rest or activity. Galeotti has, however, measured the hydrogen ion in heart muscle.§ Method. The method used in the present series of experiments was to prepare a frog’s sartorius muscle and arrange it for direct stimulation from an induction coil. A high resistance reflecting galvanometer was used and the muscle was tetanised. The muscle was tested with non-polarisable (Ringer solution, calomel) electrodes. Two of these were placed to touch the muscle directly * “Roy. Soc. Proc.,’ B, vol. 80, p. 481. + J. Bernstein, ‘ Electrobiologie’ (Braunschweig, Vieweg und Sohn), 1912. { S. P. L. Sorensen, ‘Ergebnisse der Physiologie,’ 1912, vol. 12, pp. 393-532. § G. Galeotti, ‘ Archivio di Fisiologia, 1904, vol. 1, p. 512; ‘ Zeits. f. Allge. Physiol.,’ 1996, vol. 6, p. 99. 216 Dr. H. E. Roaf. The Inberation of Ions and the [Jan. 10, opposite each other and the muscle was stimulated. If there was no definite deflection of the galvanometer, the cross-section of the muscle was considered to be iso-electric. Some other form of electrode was then substituted for one of the calomel electrodes and the muscle was stimulated. In order to avoid any influence of the stimulating current, the direction of the current in the primary coil was reversed and the stimulation was repeated. Chlorine Ions. The electrode used to replace one of the calomel electrodes was a silver wire coated with silver chloride. The arrangement was Ag | AgCl | muscle | Ringer solution | HgCl | Hg. This arrangement gave an electrical potential which was compensated in the usual way, and from the amount of compensation it was possible to calculate the concentration of chlorine ions in contact with the silver chloride—silver wire. After the compensation was accomplished the muscle was stimulated, and the direction of the galvanometer deflection was noted. The results showed that the silver electrode became more negative than it was with the resting muscle. There are two contacts where the potential may be produced; the contact Ringer solution-HgCl-Hg not being affected by the contraction of the muscle. The first is the contact Ag-AgCl-muscle, and an increase in chlorine ions would produce the increase in negativity when the muscle contracts. The second contact is that between muscle and Ringer solution, and the potential would depend on the relative ionic mobilities of the positive and negative ions set free. For a binary electrolyte the formula would contain the ratio (w — v)/(w + v), where vu = the rate of migration of the positive ion, and v = that of the negative ion. A positive ion diffusing away from the muscle more rapidly than the negative ion would cause the mercury electrode to become positive, that is, the silver would become negative. The only positive ion that need be considered is the hydrogen ion, as that is the only one that would give an appreciable value for (w — v)/(w + v), and it will be shown in the next section that hydrogen ions are liberated when muscle contracts. Against the possibility that when the muscle contracts the negative charge on the silver electrode is due to hydrogen ions diffusing into the Ringer solution, it may be pointed out that a saline electrode in contact with the active portion of a muscle becomes negative. Hence there may be a potential opposed to that due to the chlorine ions, and perhaps the action current may be added to the observed potential to give the true potential 1913. | Oxygen Tension of Tissues during Actiwnty. 217 at the silver electrode. If, however, the usual electrical change is due to an increased permeability of a membrane to negative ions both silver and calomel electrodes will be equally affected, and hence the Ringer solution may become positive from diffusion of hydrogen ions in my experiments, whilst with the usual arrangement of electrodes the negative charge at the membrane overbalances that due to the diffusion of the hydrogen ion. The potential due to diffusion of the hydrogen ion should be reduced«to a minimum by excess of indifferent electrolyte in the Ringer solution and muscle lymph. Hydrogen Ions. Galeotti (/oc. cit.) used a hydrogen electrode to measure the hydrogen ion, but there is one objection to this electrode, namely, that if the electrode and the tissue in contact with it are saturated with hydrogen the behaviour of the tissue may be abnormal. To avoid this difficulty I have used a different form of electrode. Galeotti added the action potential to his observed potential, but until we know the true cause of the action current we cannot decide whether or no this addition is legitimate. The oxygen electrode was not applicable and the reason for this will be found in the third section of this paper. The electrode used to replace one of the calomel electrodes was a platinum wire covered with manganese dioxide.* The arrangement of the electrodes was Pt | MnO, | muscle | Ringer solution | HgCl | Hg. Using this arrangement and compensating as before, on stimulating the muscle the platinum electrode showed a very strong positive potential; so marked was the effect that a deflection was obtained by a single break shock. This cannot be due to negative ions diffusing into the Ringer solution, because in the ratio (w—v)/(w+7) the hydroxy] ion is the only one that would give any appreciable negative value and the liberation of hydroxyl ions would be more effective in making the manganese dioxide electrode negative. Therefore, it is evident that hydrogen ions are liberated during muscular contraction. Oxygen Tension. Many text-books contain the statement that there is no oxygen tension in tissues. The author has pointed out that there must be some oxygen tension+ and Verzir has shown, by an indirect method, the limiting values for sub- maxillary gland and muscle.t * O. F. Tower, ‘ Zeitschr. f. physik. Chem.,’ 1900, vol. 32, p. 566. + H. E. Roaf, ‘ Brit. Med. Journ.,’ September 28, 1912. t F. Verzar, ‘ Journ. Physiol.,’ 1912, vol. 45, p. 39. 218 Inberation of Ions and Oxygen Tension of Tissues. The electrode used to replace one of the calomel electrodes was a piece of platinum wire covered with platinum black. The arrangement was Pt | muscle | Ringer solution | HgCl | Hg. The platinum electrode in oxygen or air gave results which when the muscle contracted showed sometimes a positive potential, sometimes after a slight positive potential a negative potential, and sometimes only a negative potential. The formula for the potential at a gas electrode contains the ratio \/ P/p, where P = the partial pressure of the gas and p= the osmotic pressure of the corresponding ion. From this ratio it can be seen that a fall in oxygen tension would produce the same effect as a rise in hydroxyl* (fall in hydrogen) ion concentration. Therefore, there are two factors to consider, an increase in hydrogen ions, which tends to make the platinum positive, and a fall in oxygen tension, which tends to make it negative. The deflections of the galvanometer would, therefore, be explained as a rise in hydrogen ion concentration which is frequently overbalanced by a fall in oxygen tension. The proof of this is very easy. When the access of air is prevented by a piece of rubber sheeting placed over the platinum wire as it lies against the muscle, the galvanometer may not show any deflection, but on stimulating the muscle the platinum becomes negative, thus indicating a fall in oxygen tension. This experiment has been frequently repeated and it is possible to convert one form of reaction into another. When the electrode is exposed to air the fall in oxygen tension is less, and hence, especially at the beginning of stimulation, the increase of hydrogen ions causes the platinum to become positive, but when the electrode is covered the fall in oxygen tension is greater, so that the rise of hydrogen ions is masked and the platinum always becomes negative. It is evident that this result may form the basis of a method for the direct measurement of oxygen tensions in tissues. The results so far obtained are purely preliminary. Measurements of the time relations and the actual potential produced will show whether the liberation precedes, accompanies or succeeds the muscular contraction and will give some approximation to the relative increase in ionic concentrations. Finally, it must be pointed out that the concentrations of ions measured are those in the lymph on the surface of the muscle. An increase may be due to an actual increase in ionic concentration inside the muscle or to an increase in permeability of a membrane previously impermeable to the ion measured. * Since the concentration of the doubly charged oxygen ion is proportional to the square of the hydroxy] ion concentration. 219 Reciprocal Innervation and Symmetrical Muscles. By C. 8. SHERRINGTON, F.R.S., Professor of Physiology, University of Liverpool. (Received November 13, 1912,—Read January 23, 1913.) (From the Physiology Laboratory, University of Liverpool.) I. Introduction. If we attempt to decipher the biological meaning of reciprocal innervation its various instances when marshalled together say plainly that one of the functional problems which it meets and solves is mechanical antagonism. Where two muscles have directly opposed effect on the same lever, “ reciprocal innervation” is the general rule observed by the nervous system in dealing with them, and this holds whether the reciprocal innervation is peripheral as with the antagonists of the arthropod claw, or is central as with vertebrate skeletal muscles. Also where one and the same muscle is governed by two nerves influencing it oppositely, reciprocal innervation seems again the principle followed in the co-ordination of the two opponent centres, as has been shown by Bayliss* in his observations on vasomotor reflexes. But the distribution and occurrence of reciprocal innervation extend beyond cases of mere mechanical antagonism. The reflex influence exerted by the limb-afferents on symmetrical muscle-pairs such as right knee-extensor and left is reciprocal.t Thus right peroneal nerve excites the motoneurones of left vastocrureus, and concomitantly inhibits those of the right. The reflex inhibition of the one is concurrent with, increases with increase, and decreases with decrease of, the excitatory effect on the other. Here the muscles are not in any ordinary sense antagonistic; not only do they not operate on the same lever, but they are not even members of the same limb, nor do they belong even to the same half of the body. They are, however, actuated conversely in the most usual modes of progression—the walking and the running step—though not always in galloping. Similarly with other symmetrically paired limb muscles,} the limb afferents when tested for their reflex effect on such twin muscles commonly exert an opposite and reciprocal effect on the members of the pair. Here the bifurcation of the afferent path which leads to the reciprocal effect sits, so * ‘Roy. Soc. Proc.,’ 1908, B, vol. 80, p. 339. + ‘Journ. Physiol.,’ 1898, vol. 22, p. 398. t ‘Roy. Soc. Proc.,’ 1905, B, vol. 76, p. 286. VOL. LXXXVI.—B. R 220 Prof. C. 8. Sherrington. Reciprocal [Nov. 13, to say, astride of the median longitudinal plane of the body, and a somewhat analogous case is that of the reciprocal innervation exerted by cortex cerebri on certain symmetrical muscle pairs in, for instance, the musculature of the eyeballs.* Stimulation of a point in the right cortex, while causing left external rectus to contract, causes right external rectus to relax, and con- versely inhibits left internal rectus while exciting contraction of right internal rectus. But in the case of such symmetrical muscle-pairs, though some circum- stances and some afferents deal with the two members of the pair by “reciprocal innervation,” it is equally clear that some deal with them by “identical innervation.” Thus with the two vastocrurei, right and left, though most of the limb-afferents influence the two reciprocally, the genito- crural nerve influences them identically,t namely, excites concurrent contraction of both muscles, and it is clear that in their natural use both the muscles are sometimes thrown into contraction together and thrown out of contraction together, as happens in the gallop and in standing and sitting. Similarly with the lateral recti of the eyeballs, stimulation of certain brain points and certain voluntary acts which cause ocular con- vergence exert an identical influence on the two internal recti exciting both together, and similarly an identical influence on the two external recti inhibiting both. The relation of reciprocal innervation to the symmetrical ‘muscle-pairs differs, therefore, from its relation to antagonistic muscles, in so far that in the former case it is only one of the ordinary modes of innervation obtaining for the muscle-pair, whereas in the latter case there is little evidence at present that reciprocal innervation of the muscle-pair is at all commonly under normal circumstances replaced by identical innervation. Such symmetrical muscle-pairs present, therefore, the problem that some- times they are co-ordinated by reciprocal innervation and sometimes by identical innervation. In the present observations it has been sought to see whether by experimental means in the purely reflex preparation they can be made sometimes to contract together or to relax together and at other times to behave reciprocally, the one member of the pair contracting con- comitantly as the other relaxes. II. Change from Reciprocal Innervation to Identical in Symmetrical Extensors. When an afferent nerve of one hind limb is stimulated in the double (right and left) vastocrureus preparation (decerebrate cat), reciprocal innerva- * “Roy. Soc. Proc.,’ 1893, vol. 52, p. 333. + ‘Journ. Physiol.,’ 1910, vol. 40, p. 53. 1912.] Innervation and Symmetrical Muscles. 221 tion is seen to obtain in the reflex effect on the two muscles. The muscle contralateral to the nerve exhibits excitatory contraction; the ipsilateral muscle inhibitory relaxation (fig. 1). This result holds good over a wide range of intensities of stimulation of the afferent nerve. The contralateral le Tune 5 Sees AVaslocriua (4 S44 Sp Hf a OL me eS eI cI me — Right lestocrureus Lerl Vastocrures LVaslocrus } I R.perons. |! Of 3 : ==. SS EE EEE L peron.m. 4 eS Tt DET TLE VE L per tel Ve Fig. 1. Fig. 2. Fic. 1.—Reciprocal innervation of the extensor muscles, vastocrwrei, of right and left knee, first from left peroneal nerve, then from right peroneal. Decerebrate cat. The inhibitory relaxations are in each case followed by post-inhibitory rebound. Fic. 2.—Extensor muscles, vastocrurei, of right and left knee. The right peroneal nerve is first stimulated; then the same nerve again, and during its stimulation left _ peroneal is stimulated with same intensity approximately as right, and the left stimulus withdrawn, and finally the right. During the double stimulus both muscles exhibit inhibitory relaxation. Decerebrate cat. excitation and the ipsilateral inhibitory relaxation increase pari passu as the intensity of the stimulation is increased. But it is also possible experimentally to obtain simultaneous inhibitory R 2 222 Prof. C. 8. Sherrington. Reciprocal [Nov. 13, relaxation of both muscles followed by simultaneous contraction of them. Ii the preparation be carefully made, the condition of the twin isolated muscles remaining sensibly similar, and if similar right and left afferent nerves, ¢.g. right and left peroneal, be fitted with electrodes from similar induction coils similarly supplied and interrupted for faradisation and the intensity of the stimulation of the two nerves be kept as far as practicable equal, observations can be obtained as follows :—Suppose, as in fig. 2, right peroneal stimulated; right vastocrureus relaxes and left contracts; if then, while right nerve continues to be stimulated, left peroneal be stimulated in addition, right vastocrureus still remains relaxed and indeea may relax further, but left vastocrureus relaxes also. On then withdrawing the stimulation of left nerve left muscle contracts and right remains relaxed still, and finally on withdrawal of stimulation of right nerve right muscle contracts by rebound. This shows that each of the afferent nerves employed taken by itself unfolds in response to stimulation an inhibitory effect on the ipsilateral muscle stronger than is its excitatory effect on the twin contralateral muscle. Stimulation of the right and left nerves concurrently if the stimulations be fairly equal in intensity causes therefore concurrent relaxation of both muscles. Thus when the stimulations of the two nerves are repeated synchronously, as in fig. 3, both muscles relax together at each repetition of the stimulation. Further at each discontinuance of the double stimulation both muscles exhibit synchronous rebound contraction. So that in response to the synchronously repeated and synchronously remitted stimulations both muscles relax and contract synchronously. If, however, the intensities of the two stimuli be markedly unequal the muscular reactions, right and left, though synchronous are, of course, reciprocal, not identical. The identical form of reaction holds true over a wide range of intensities of stimulation, so long as the two stimulations, right and left, are kept of approximately equal intensity. The preponderance of ipsilateral inhibition over contralateral excitation obtains therefore both with moderate stimuli and with strong, the ratio between the intensities of the reflex inhibition and reflex excitation remaining apparently about the same for a wide range of stimulus intensities. The difference between the effect of synchronous double stimulation of strong intensity and of weak is in the main merely that with weaker stimuli the synchronous relaxations of the two muscles are weaker and are followed by less powerful rebounds than are strong,. though the rebounds are still synchronous. ‘geo aqeaqereoeq )4=‘yenba Ajeyemrxoadde 4jo] puv yyst1 SUOIFR[NMYS JO SalzIsusqUI ey, ‘“euoTe Ja, Jo Aj[vuy pur ‘ouoyR 4YSII Jo UST) ‘1eY}050} seAtoU YI0q Jo SUOTyRTNMTS UBAIS UOT 4Jo, Jo ust ‘rouored 4ySta Jo UOR[NMIG ‘9euy Jo] pur AYSII Jo “amuoozspa ‘sofosuuI JOSUEyXH—'e ‘DI\| 223 ub qnauoiad 7 21 joaunsad oy 1IO)S VAT Innervation and Symmetrical Muscles. LI04S DALY ete 1912.] o eee AFRO L_OLIlOIlOlIOIIl™OI™PIIIIL_II—I_9@q UII eee ee eee > TG fr ELT fa -~ POW 7 u IIL. Change from Reciprocal Innervation to Identical in Symmetrical Flexors, If instead of extensor muscles we take a pair of symmetrical flexors, upon them again reciprocal innervation is found to be the reflex result of stimula- tion of an afferent nerve of either limb. Thus with the two tensor fascize 224 Prof. C. 8. Sherrington. Reciprocal [Nov. 13, femoris muscles, hip-flexors, the effect of stimulation of the central end of the peroneal nerve is contraction of the ipsilateral muscle and inhibitory relaxation of the contralateral (fig. 4). So also with psoas, and sartorius (fig. 5), and semitendinosus, the latter a flexor of knee. Yet with these flexors, although thus reciprocally innervated, it has been shown that, under certain circumstances, they exhibit identical reflex innervation, for instance when nociceptive stimulation is applied concurrently to both feet.* And Sertorits : 0 Time£ SRR ERAS AR ALAA TAA CALA TRIAL AS fitse fen. BR peron. * /, peron. Fig. 4. Fie. 5. Fie. 4.—Flexor muscles, tensor yascie femoris, of right and left hips. Stimulation of right peroneal nerve, and then of left peroneal nerve. Decerebrate cat. Fie. 5.—Flexor muscles, psoas and sartorius, of right and left hips. Stimulation of right peroneal causes contraction of the right muscles and inhibitory relaxation of the left. Stimulation of the left peroneal causes inhibitory relaxation of the right muscles and contraction of the left. Decerebrate cat. when the two peroneal nerves right and left are concurrently stimulated with faradisation of approximately equal intensity, the two flexor muscles right and left both contract together (fig. 6) and exhibit fully identical reflex innervation. Here, however, the identical innervation presents itself in the form of concurrent contraction not of concurrent relaxation as was the case with the extensors. With the flexors, therefore, the excitatory effect of each * Sherrington, in ‘ Schifer’s Handbook of Physiol.,’ 1900, vol. 2, p. 840 ; ‘ Integrative Action of Nervous System,’ p. 225, 1906. 1912. ] Innervation and Synmetrical Muscles. 225 afferent is stronger than the inhibitory effect, and the excitatory effect is ipsilateral. So that with both flexors and extensors the ipsilateral effect is Sartorris R. pie OEE eee EE eee Lyperon Fic, 6.—Flexor muscles, psoas and sartorius, of right and left hips. Stimulation of the right and left peroneal nerves together causes contraction of both the right and left muscles synchronously. The stimulation intensities right and left were approxi- mately equal. Decerebrate cat. the stronger, but in the case of the flexors it is excitatory while with the extensors it is inhibitory. IV. Algebraic Summution as a Factor in Producing the Change. In the case of these symmetrical muscles therefore the change from reciprocal reflex effect to identical is clearly explicable by algebraic summa- tion of excitation and inhibition.* The result may be stated numerically. Right peroneal nerve under a stimulus whose intensity may be figured as 10, causes an excitation of the motoneurones of right flexor muscle of an intensity expressible as 10, and a weaker inhibition of the motoneurones of left flexor whose intensity may be called 6. It at the same time causes also an inhibition of the motoneurones of the right extensor of intensity 10, and a weaker excitation of the motoneurones of left extensor of intensity 6. Similarly, the reflex effect of the left peroneal nerve under a stimulus of like intensity is on the left flexor’s motoneurones an excitation of value 10, and on right extensor’s motoneurones of value 6, while left extensor’s motoneurones it inhibits with value 10 and right flexor’s motoneurones with value 6. Denoting (fig. 7) excitation by the prefix + and inhibition by the prefix —, the resultant reflex effect on the motoneurones of the muscles severally is, when both nerves are stimulated concurrently, + 4 for the flexor in each limb, and — 4 for the extensor in each limb. So that, if we suppose all the muscles to be previously in a condition of medium tonus the two symmetrical flexors then contract and * “Roy. Soc. Proc.,’ 1908, B, vol. 80, p. 565. 226 Prof. C. 8. Sherrington. Recrprocal [Nov. 13, the two symmetrical extensors exhibit inhibitory relaxation. It is to be noted that if the initial condition of the muscles be full repose, e.g. —10 in the above notation, then in the above instance the symmetrical extensors (as well as the flexors) enter upon a certain degree of contraction, namely —4, full relaxation being —10. In any case the double stimulus gives an identical reflex effect on the symmetrical muscles, though the reflex effect <7 +6 Fic. 7.—Explanation in text. still remains reciprocal as regards the pairs of antagonistic muscles. Under the double stimulus the reflex effect becomes symmetrical and identical in the two limbs, although under either of the two components singly the reflex effects in the two limbs are diametrically opposed. These results hold over a wide range of intensity of stimuli so long as the intensities of the stimuli right and left are approximately equal. This indicates that the intensity ratio between the ipsilateral and contralateral effects, both of inhibition and excitation, remains but little changed over a wide range of different intensities of stimulation. The absolute values rise and fall with increase and decrease of stimulation, but the relative values remain about the same, subject to two alterations which will be mentioned later. The above observations therefore come under the rubric of algebraic summation in double reciprocal innervation. They show that double reciprocal innervation can change reciprocal innervation of symmetrical muscles into identical innervation of them. Double reciprocal innervation* applied to antagonistic muscles does not result in identical innervation of * “Roy. Soc. Proc.,’ 1909, B, vol. 81, p. 249. 1912. | Innervation and Symmetrical Muscles. 227 them, although it can bring about, it is true, the exhibition of some degree of contraction by both the antagonists at the same time. The difference in intensity of reflex influence exerted on the ipsilateral and contralateral limbs respectively may have its biological meaning in the opportunity thus given for the limbs to exhibit either symmetrical reflex movements, or movements of opposed direction right and left, according as there is equality or inequality between their right and left stimuli. V. Symmetrical Rebounds. It will be noted that in the above attempt to change by experimental means the reciprocal innervation of the symmetrical muscle-pair consisting of right and left knee-extensor into identical innervation, success 1s reached as regards “immediate ”* reflex effect only in so far as identical inhibitory relaxation of the two. Simultaneous contraction of the two in response to concurrent stimulation occurs occasionally with weak stimuli, especially where one of the nerves tends to give ipsilateral contraction. The identical contraction of the two which ensues, however, after their concomitant relaxation is, so to say, not an immediate but a “ terminal ”f reflex result, for it is due to post-inhibitory rebound. Rebound contraction in one muscle of an antagonistic or of a symmetrical pair 1s so commonly associated with concomitant relaxation of the fellow muscle (fig. 8) of the pair that large identical rebound contractions vastoer Li. va stocr. Fic. 8.—Extensor muscles, vastocrure’, of right and left knees. Stimulation of right peroneal nerve. The post-inhibitory rebound of the ipsilateral extensor is synchronous and reciprocal with the post-excitatory relaxation of the contralateral extensor. Decerebrate cat. * Of. T. Graham Brown, ‘Quart. Journ. Exp. Physiol.,’ 1912, vol. 5, p. 237. + Cf. T. Graham Brown, ‘Quart. Journ. Exp. Physiol.,’ 1911, vol. 4, p. 331. 228 Prof. C. S. Sherrington. Reciprocal [Nov. 13, occurring concomitantly in the symmetrical muscles is not without interest. T. Graham Brown* has recently pointed out that in antagonistic muscles in many cases the terminal relaxation following an excitatory reflex may be regarded as of the nature of an inhibitory rebound, the converse of rebound contraction. It would seem, therefore, that the terminal rebound following a reflex of reciprocal effect on a muscle-pair is often itself of reciprocal character in the two muscles. In the symmetrical muscles dealt with in this paper the terminal effects when the reflex itself has been of reciprocal influence on the two muscles are quite usually of reciprocal character in the two muscles (see fig. 8). When, however, the character of the reflex itself has been changed, by the procedure described, from reciprocal into identical the terminal rebound also is changed from reciprocal into identical (fig. 3). VI. Factors Outside Algebraic Summation Involved in the Change. The above seems to me what the experiments clearly indicate as the main principle involved in the change from reciprocal innervation of symmetrical muscles to identical innervation of them when the stimuli are appropriately duplicated. This principle rests on the inequality of the excitation-poteney and inhibition-potency respectively inherent in the components of the summed duplicate reflex. But the experiments have shown certain further features outside this principle. In the observations on the vastocrureus muscles, when the reflex inhibition due to the ipsilateral nerve is in progress, and stimulation of the contralateral nerve is then added, the effect of the latter is very occasionally not a mitigation but a distinct increase of the inhibitory relaxation (fig.:2, abscisse 4,5). A similar result is sometimes met with in the flexor, tensor fasciz femoris, though there conversely in regard to the nerves used. A phenomenon comparable with these, although in the opposite direction, was reported previously by Miss Sowton and myself working with the knee-flexor, semitendinosus.t We noted that if the stimulation of contralateral nerve is relatively weak in comparison with that of the ipsilateral nerve, the former, if added when the latter is in progress, may, instead of lessening the reflex contraction due to the latter, actually increase it. In the observations of this paper a strong reflex inhibition of the extensor centre already in progress seems to convert a weak excitatory influence into an inhibitory one. In the previous obser- vations a strong reflex excitation of the flexor centre seemed to convert a weak inhibitory influence into an excitatory one. And it has been the case in some of the present observations on the hip-flexors that the addition of a * Ibid. + ‘Roy. Soc. Proc.,’ 1911, B, vol. 84, p. 204. 1912.] Innervation and Symmetrical Muscles. 229 strong contralateral stimulus to an already existent ipsilateral has increased instead of lessened the contraction (fig. 9), and this has been so not only during the application of the contralateral stimulus but still further on its withdrawal. A rebound contraction of the flexor muscle has then occurred, RR ipsoas | a Sortortus 2 3 s iL psvuas Sartorius Fie. 9.—Psoas and sartorius muscles, flexors, of right and left limbs. Stimulation of right peroneal with a weak stimulus, coil at 24 cm. from primary ; during this stimulation the left peroneal is stimulated for about one second with a strong stimulus, coil at 14 cm. from primary. The contraction of right flexors is increased during stimulation of left nerve, and on cessation of this stimulation a still further rebound augmentation of the right muscle’s contraction occurs. Decerebrate cat. adding itself to the contraction due to ipsilateral nerve, and causing a combined contraction of large amplitude. It is obvious that such reactions though outside the main principle above stated work in the direction of assisting a double reciprocal innervation to change the reflex effect on the symmetrical muscles from reciprocal to identical character. The contraction of both fiexors will be increased and so also the inhibitory relaxation of both extensors. 230 Prof. C. 8. Sherrington. Reciprocal [Nov. 13, It seems, therefore, that in the combination of two reciprocal reflexes of opposite effect on the symmetrical muscles of the limbs there is, besides - simple algebraic summation of the respective excitation and inhibition of the components, a further factor sometimes present. An actual reversal of the weaker element of one of the components appears to take place. Thus, the excitatory effect on the contralateral extensor tends occasionally to be reversed to inhibitory effect, and the inhibitory effect on contralateral flexor tends to be reversed to excitatory effect. There remains the question whether identical concurrent contraction of ‘the symmetrical extensor pair can be obtained as a direct reflex. With concurrent stimulation of both nerves, right and left, it can occasionally, when the stimulation is weak, especially if one or both nerves be then giving ipsilateral extensor contraction. Experiment shows that it can also be obtained by a procedure quite other than that followed for producing the identical innervation which has symmetrical relaxation for its result, namely by weak stimulation of the afferent nerve of one side alone, right or left. The range of intensity of stimulus over which reciprocal innervation of the muscle pair results from excitation of a limb afferent is wide and ranges from weak intensities without break right up to the very strongest. But with stimuli of little above threshold value the result changes in the decerebrate preparation. With these stimuli although the response in the contralateral muscle remains reflex contraction, in the ipsilateral muscle the response becomes reflex contraction instead of inhibitory relaxation. This alteration is that noted previously by Miss Sowton and myself.* That it is accompanied by concomitant contraction of the symmetrical extensor of the fellow knee shows that with change merely of intensity of the stimulus the reflex innervation of the fellow muscles alters from reciprocal to identical. The reflex contraction so obtained is weak but quite indubitable and clearly concomitant (fig. 10). Over a range of stimulus-intensities running from threshold value up to weak or moderate, the exact limit upward varying with the condition of the preparation, the reflex result in the decerebrate animal passes from identical innervation (fig. 10, A) to an admixture of identical and reciprocal innervation (fig. 11), until with stronger stimuli reciprocal innervation is fully established (fig. 10, B). In the admixed form of result the reflex opens with bilateral contraction, 7.e. contraction of both right and left extensor muscles, and then passes over into ipsilateral inhibitory relaxation, with contralateral pure contraction (fig. 11). This transition is speedier the less weak the stimula- tion. It is brought about by a change occurring, as noted by Miss Sowton rob) * “Roy. Soc. Proc.,’ 1911, B, vol. 83, p. 435. 1912. | Innervation and Symmetrical Muscles. 231 and myself,* in the reaction of the ipsilateral muscle during the progress of the stimulation. The statement was made above that the relative values of ipsilateral Tape ea eS Fah pad cal pa sad call ca ts ea) (a ae a ad fa) eo od) oe BR vastocr vastocr ae Ly peroneal | covl, 29cent | , Z | fil L.peroneat -————— covl,77cenk. Fie. 10.—Extensor muscles, vastocrwret, of right and left knees. In the left-hand record the stimulation of left peroneal is weak and little above threshold value ; in the right-hand record the stimulation of the same nerve is repeated, but with increased though still moderate intensity. Decerebrate cat. Tn © ADADADADADARADDADAAARD LAAN ADADAN Ht sec ; / R vastocn peron. Fie. 11.—Extensor muscles, vastocrurei, of right and left knees. Stimulation of left peroneal with moderately weak faradisation. The reflex effect opens by being “identical” in the two muscles, but later changes to “reciprocal” as the stimulation continues. Decerebrate cat. inhibition and contralateral excitation in the extensors remain about the same over a wide range of intensities of reflex stimulation, although the * Ibid. 232 Reciprocal Innervation and Symmetrical Muscles. absolute values rise and fall part passu with the stimulus intensity. What has just been said has to be remembered in relation to that statement. With decrease of stimulus intensity, as just said, a value of stimulus is ultimately reached, at which in the decerebrate preparation contralateral effect still remains excitatory, but ipsilateral becomes excitatory instead of inhibitory. The biological meaning of this may be that with these weak stimuli the reflex produced is that of standing, 7.e. a local reflex contributory to the great compound reflex of standing, whereas, with stronger stimuli, the reflex produced is the nociceptive flexion reflex, or the flexion phase of a locomotor step-reflex. With the flexor muscles of the hip, psoas, tensor fascize femoris, and sartorius, strong stimulation of one peroneal nerve sometimes excites contraction in these muscles in the contralateral, as well as in the ipsilateral limb. The contraction of the contralateral muscles is less strong than that of the ipsilateral. It tends to be followed on withdrawal of the stimulus by marked rebound contraction. The contralateral contraction of the hip- flexors recalls the contralateral contraction of the ankle flexor, tibialis anticus, noted by T. Graham Brown,* as sometimes occurring in both decerebrate and spinal preparations. VIL. Summary of Conclusions. 1. The oceurrence and distribution of reciprocal innervation extends to cases of muscular co-ordination beyond those-involving simple mechanical antagonism. Thus it is exemplified also in reflexes actuating symmetrical muscles, for instance, muscles symmetrically placed in the right and left limbs. 2. These muscles present the problem that, in reflexes, though often worked reciprocally, they are also often worked identically. 3. Experiments cited show certain ways in which the stimulations can be experimentally arranged to give either reciprocal or identical innervation of symmetrical muscles of right and left limb. 4, Itis shown that algebraic summation of excitation and inhibition can explain this result. 5. It is further’shown that there is evidence that other factors besides simple algebraic summation of the individual component reflexes have a share in changing the reciprocal innervation into an identical. A reversal of the weaker element of one of the components appears to occur. Thus, the excitatory effect on the contralateral extensor tends occasionally to be reversed to inhibitory effect, and the inhibitory effect on the contralateral flexor tends to be reversed to excitatory effect. * © Journ. Physiol.,’ 1912, vol. 44, p. 125. 233 Nervous Rhythm arising from Rivalry of Antagonistic Reflexes : Reflex Stepping as Outcome of Double Reciprocal Innervation. By C. 8. SHERRINGTON, F.R.S. (Received February 3,—Read February 20, 1913.) (From the Physiological Laboratory, University of Liverpool.) EL The observations with which the present communication deals were met with in experiments continuing those on reciprocal innervation of symmetrical muscles. In my previous paper on that subject* it had been reported that in regard to symmetrical extensors of the knee the ratio borne by intensity of the ipsilateral inhibition to the contralateral excitation is such that with equal stimuli to right and left symmetrical afferent nerves there is inhibitory suppression of contraction in both the muscles. In other words, under double reciprocal innervation the ipsilateral inhibition by each nerve completely overcomes the contralateral excitation of the other. It was shown that this mutual suppression holds over a wide range of the scale of intensities of stimulation. It was also shown that with quite weak stimuli a simultaneous stimulation of both nerves, stimuli being equal in intensity, often results in concurrent contraction of both muscles. Indeed, with quite weak stimuli, the effect of stimulation of each afferent nerve by itself is, in the decerebrate preparation, usually contraction of the ipsilateral as well as of the contralateral muscle. This being so, it is evident that at some point in the scale of intensities of stimulation there should be a place below which contralateral excitation is stronger than ipsilateral inhibition, whereas above it ipsilateral inhibition is stronger than contralateral excitation. My further experiments were directed to finding where on the intensities- scale this point actually lies. In prosecuting this search there began to be met instances of rhythmic contraction of exceedingly pronounced character. Method employed. The mode of preparation when the knee extensor was observed was as follows:—The animal (cat) was decerebrated under deep chloroform and ether narcosis. In both limbs all muscles were then detached from the great and lesser trochanters and intertrochanteric line. The insertions of right and left iliopsoas, psoas parvus, and tensor fascisze femoris were then carefully resected up, and the origin of rectus femoris right and left. In both limbs the following nerves were severed: popliteal, small sciatic, hamstring, external * Supra, p. 219. 234 Prof. C. 8. Sherrington. Nervous Rhythm [Feb. 8, cutaneous, obturator, internal saphenous and branch to the sartorius and pectineus. The peroneal of each limb was ligated tightly at its entrance into tibialis anticus just below knee. This procedure leaves the vastocrureus muscle of each limb the muscle operative in or on the limb. A steel drill in the condylar end of each femur is clamped to a heavy upright in such a position that, the preparation being supine, the two femora are nearly vertical and are parallel, and both hips are flexed to about a right angle. The root of the tail is fixed to the table by a steel pin. The leg is removed one-third down the tibia, and into the free end of that bone a fine steel pin is set ; to this the thread attached to the myograph lever is fastened. The angle made by the tibia with the thigh varies with the degree of contraction of vastocrureus, and, the preparation being decerebrate, that muscle exhibits tonus, z.c. maintains the standing posture. The resistance to the contraction of each vastocrureus was provided by the weight of the remaining portion of the limb below the knee and the adjustable tension of a light coiled wire spring attached to the lever of the myograph near its axis. The resistance was made as nearly similar as possible for both the right and the left muscle. The peroneal nerve on each side was stimulated with platinum electrodes, 4 mm. apart and placed with the nerve-trunk obliquely between them. The pair of electrodes was inserted sidewise through a glass tubulure in the wall of a glass tube, in which lay the ligated nerve itself. The muscles and skin around and over the electrode and tube were brought together and stitched. For flexor observations the psoas and tensor fascize femoris were employed and isolated, the animal being prone. For stimulation a pair of similar Leeds induction coils were used, one for the right nerve, the other for the left. In the secondary circuit of each a resistance box of 100,000 ohms was introduced. A double switch connected to the circuits for both nerves allowed the coils to be interchanged for the two nerves at will, furnishing some oppor- tunity of testing the approximate equality of the two circuits and of the relative excitability of the two nerves. The right and left vastocrurei, thus completely isolated in each limb, but retaining blood and nerve supply and natural attachments quite intact, form a symmetrical pair of extensor muscles. Under the reflex action of, for instance, an afferent nerve of either limb, reciprocal, innervation obtains for them as it does for an antagonistic muscle-pair. As shown, the afferent of each limb relaxes the ipsilateral vastocrureus by reflex inhibition, and causes the contralateral to contract by reflex excitation. The motor centre of the contralateral extensor reacts therefore to the afferent in the same direction as does that of the flexor of an ipsilateral antagonistic pair. For the observations on double reciprocal innervation required I employed symmetrical afferent nerves, usually the peroneals, sometimes the popliteals. With simultaneous stimulation of both right and left nerves algebraic summation of the excitatory and inhibitory effects is of course the result.* Over a certain range of combinations this summation is readily observable because it results in contractions not wholly suppressed but of various grades of submaximal intensity. Somewhere within this restricted range of combinations of the opposed stimuli should lie the neutral point sought * Vide supra, p. 226. 1913.] arising from Rivalry of Antagonistic Reflexes. 235 for in my experiments. When pursuing adjustments of the stimuli in order to arrive at determination of this it was found that rhythmic alter- nating contraction and relaxation, reciprocal in direction in the two muscles, appeared directly both stimuli were employed concurrently, though quite absent under either stimulus alone. During these rhythmic reactions a glance at the preparation was enough to show that, although only the one thigh muscle remained in each limb, the animal was stepping with both limbs, usually at a quick walking pace, and whether the muscle remaining were an extensor or a flexor. It began to step as soon as ever the two antagonistic stimuli were concurrently com- bined, and it ceased to step directly their union was dissolved. Fig. 1 exemplifies this. There it is seen that it did not matter whether the combination were made by adding the inhibitory stimulus to the excitatory already in action, or the excitatory to the previously-acting inhibitory; in either case the stepping did not occur under either stimulus acting alone, yet ensued directly the two opposed stimuli were in action together. Conversely, on withdrawing either of the stimuli and leaving the other one in operation the stepping immediately ceased, each limb then passing into either steady contraction or steady relaxation, according as the remaining stimulus was contralateral to it or ipsilateral. II. Nervous inhibition seems in many ways the exact opposite and counter- part of nervous excitation. And since the process of nervous excitation is certainly rhythmic, its natural rhythm in mammalian motoneurones being probably about 50 impulses per second (Piper),* it may be argued that the process of nervous inhibition is probably similarly rhythmic. Yet in com- parison with the slow periods of many rhythmic muscular acts, for instance those of breathing and stepping, a rhythmic frequency of 50 per second can be considered as tantamount to continuous and steady operation. It is a question how there are developed from such minutely oscillatory nervous discharge those coarser rhythmic actions with periods recurring once in 3 secs. as in breathing, or once a second as in the step, or four times a second as in the scratch reflex. The suggestion has often been made that such rhythmic nervous actions are the result of the concurrent action of two opposed nervous forces, the outcome of a constant opposition or resistance acting against a constantly discharging nervous activity. e“How can the constant motion of the nervous fluid be changed to a periodic motion ? When a conductor of electricity is held at some distance from the electric * ‘Hlektrophysiologie menschlicher Muskeln,’ Berlin, 1912. VOL, LXXXVI.—B. SS) 173d 4x0} UI UAAIS UOTyVURfdxe aoTjAN ‘UOexeper ALOFIqIYUT OAT}OW sWeouT OUT] weigofur Jo yudosop geyQ os ‘ouo o1N0} & sx UOTyeaedoad oy, ‘NOEXe[AL GuEdSep “UOIZORIZUOO SojoUap our, Yderdokw yo yUoOSW “9Y}O OY} JO GUO Jo [AveIPYyTA Aq weyxorq st uorzenMtys 410749 10xe pure A107IqIYUT Jo VOIVEUIGUIOD oY} YOITA ye SWOMOUL 94} YAVUT Sossodo [[BUIS OTT, “CAO SPU jo syqjy UI OWT, ‘eposnuT snemn.10098vA qo, CATT { OPOSNUA sneanro0yseA FYSIT “AY “Yeo ozetqotooa(T “‘s}USJEHe se posn (‘d'7) jeouosed 4Joy pue (dw) jeouored yy St ‘a1ed-aactou oystuoSeque yy Jo UOWLNMMS JUeInoU0D Aq paxorg “4Jo] pue 4ysi1 ‘seouy JO e[OSNUT 10sU94xe pezeTOST JO Suiddoys yeoordioer xepey— | “1 [Feb. 3, papa Sis Se ae en ie + Fe £0) Ar ee | eee RAE ES a uN eT wean ecg 8 oa : F : x i Nervous Rhythm Prof. C. S. Sherrington. : Wo” WAILALS LISS HLA IID IAI LASS MAILS IS AIA IIS IA SASSI AL IIS ISLS MISS AD IAAI SS AAD IAAI AIA ASSAD ASI SSSI S LASS ISI IS ASD S LISS BAIA S KILL, NONTTTATNDTTIITIITIN TTT TTTTTITTTS 236 1918.| arising from Rivalry of Antagomstic Reflexes. 237 machine, kept in a state of constant excitement, the electricity is given off periodically. The dry air between the machine and the conductor held near it impedes progress of the electricity to the conducting body until the electricity accumulates in quantity to overcome the impediment offered in its course, and the electricity escapes in a succession of sparks. This is’ an illustration of how the current of the nervous principle may be rendered periodic.”* A well-known application of this explanation to a particular case of nervous rhythm was that given by Rosenthalf for the inhibitory action of the superior laryngeal on the respiratory centre (1862). He concluded that the stream of nervous impulses from the constantly discharging and blood-: stimulated respiratory centre meets a resistance to its outflow and thereby becomes rhythmic, much as does a stream of air passing out from a tube under water (op. cit., p. 242). He described the function of the superior laryngeal nerve as being to increase this resistance and thus slow and deepen the respiratory movements ; and the function of the rest of the vagus as being to: lessen the resistance and thus render the respiratory rhythm less deep and more frequent. He argued that other inhibitory nerves act on their respective mechanisms in the same manner as does the superior lyneeal on the respiratory centre. nes =. ‘ When a skeletal muscle’s reflex contraction is partly ainitdronieed by a reflex inhibition insufficient to suppress it entirely, the contraction thus moderated by inhibition often shows oscillations. A. Forbes{ has put the question whether these oscillations may not mean that nerve-impulses constantly generated by the reflex excitatory nerve break through the inhibitory resistance periodically, just as, on Rosenthal’s view, do the inspiratory impulses from the respiratory centre. Undulations tend commonly to appear in reflex contractions obtained under concurrent reflex excitation and inhibition,§ and both T. Graham Brown|| and Forbes’ have independently called attention to features in them: Of such undulations two main forms may be distinguished, and it seems at: present well to regard the two separately, although the same principle of production may underlie both. In one form, the undulations are smaller, more rapid, less regular, and * John Miiller, ‘Handb. d. Physiol.,’ 1835, vol. 11, p. 77. + ‘Die Athembewegungen u. ihre Beziehungen zum Nervus Vagus,’ Berlin, 1862. t ‘Roy. Soc. Proc.,’ July, 1912, B, vol. 85, p. 289. § ‘Roy. Soc. Proc.,’ 1909, B, vol. 81, p. 268; ef. also ‘Quart. Journ. Exper. Physiol.,’ 1908, vol. 1, p. 7. || ‘Roy. Soc. Proc.,’ 1912, B, vol. 85, p. 278. 3 Lbid. 238 Prof. C. 8. Sherrington. Nervous Rhythm [Feb. 3, have a frequency of about 8-10* or 7-12 per second.t Figures showing this form have been furnished in previous papers in these ‘ Proceedings.’t The oscillations are often compounded into or with slower ones so that the graphic records show compound waves. In the other form the undulations are slower and often of much more regular rhythm and amplitude than the above. As seen in figures furnished in these ‘ Proceedings, by T. Graham Brown,§ they show on the muscle contraction as notches or teeth in series from three to seven in number, each undulation lasting not far short of a second. Forbes figures|| a remarkable example with oscillations more ample, less regular, proceeding through a long series with a rate often of about 1 per second. In a paper of my own an example of this slower form of undulations was figured, recorded simultaneously in flexor and extensor muscles of the knee ; it was there (p. 268) pointed out that these undulations are reciprocal in sense in the antagonistic muscles, puffs, so to say, of inhibition in the one muscle’s centre corresponding with puffs of excitation in the other’s. Grahain Brown,** observing the antagonist muscles of the ankle, notes also there that where (fig. 3 of his paper) the undulations were visible in both the antagonists they were reciprocal in them. An important suggestion is made by Graham Brown, namely, that these slower undulations are “akin to the rhythmic act of progression.”++ This suggestion the results given in the present pages leave no room to doubt is correct. Graham Brown believes “that the rhythmic phenomenon is conditioned by a balance of” “activities which produce in the same centre equal and opposite effects (excitation and inhibition).’{{ Forbes,§§ referring to the less regular and less orderly oscillations met in his experiments, writes: “ Perhaps all” these oscillations “are manifestations of a general tendency of opposed influences in reflex centres, although themselves con- tinuous, to produce rhythmic activity. If so, it is conceivable that some conditions of intensity or time relations produce the regular movements of progression by enabling the centres to fall into a rhythm natural to them, * Forbes, zbzd., p. 293. + Of. ‘Quart. Journ. Exper. Physiol.,’ loc. ctt., p. 74, fig. 6. { Forbes, ‘Roy. Soc. Proc.,’ B, vol. 85, pp. 293, 297, figs. 3, 4; Sherrington, zbzd., B, vol. 80, pp. 569-574, figs. 1-4; zbzd., B, vol. 81, pp. 250, 262, figs. 1, 10. § ‘Roy. Soc. Proc.,’ 1912, B, vol. 85, pp. 281-283, figs. 1, 2, 3. || ‘Roy. Soc. Proc., 1912, B, vol. 85, p. 292, fig. 1. 4 ‘Roy. Soc. Proc.,’ 1909, B, vol. 81, p. 261, fig. 9, B. ** ©Roy. Soc. Proc.,’ loc. cit. +t Ibid. t{ Zbrd., p. 287. §§ ‘Roy. Soc. Proc.,’ op. cit, p. 297. 1913.] arising from Rivalry of Antagonistic Reflexes. 239 whereas other conditions . . . just miss the natural rhythm of the centres and produce a confused rhythm instead.” In conformity with this opinion the discrete and regular contractions shown in fig. 1, presenting the appearance of a rhythmic series of wholly separate contractions, which are reciprocals of similar complete relaxations in the fellow muscle, are in reality examples of the above discussed slow form of undulation incident to reflex contraction under concurrent inhibition and EA. SERRE LE Fig. 2.—Reciprocal stepping of isolated extensor muscle of knees, right and left, reflexly evoked by concurrent stimulation of the antagonistic afferent nerves /.p., left peroneal, and 7.p., right peroneal. Decerebrate cat. R.V., right vastocrureus ; L.V., left vastocrureus. Correspondences between the stimulation signals and the myograph events indicated by numerals. The right peroneal stimulation following the left produces the rhythmic though somewhat imperfect stepping, but the left following the right, although the same physical stimuli are employed, fails to do so. Time above in fifths of seconds. 240 Prof. C. 8. Sherrington. Nervous Rhythm — [Feb. 3, excitation ; examples, it is true, in which the regularity and amplitude of rhythmic contraction and relaxation have reached their full. That these rhythmic contractions are really analogous to the simple undulatory reflex contractions cited and discussed above seems clear from several considerations. (1) Both occur under the same circumstance, namely, concurrence of inhibition and excitation. (2) Both exhibit the same rate of rhythm. (3) It has not been difficult in my experiments to find intermediate examples connecting the more developed forms with the less developed. Thus, the example fig. 2 is intermediate between the form exhibited by fig. 1, and the example figured in observation A, fig. 6; and the example fig. 3 seems to me intermediate between that of fig. 1 and the example figured by Forbes (URES SE wecre Us Sulu cLe SS SESE SS SUveCSUeCCUCURUUEE TEL SUECEBE LED EELEULUUULUCULEULUDUUUSUL NUD DEE EEEUULSUET Fic. 3.—Reflex reciprocal stepping of isolated extensor muscle of knees, right and left, evoked by concurrent stimulation of the antagonistic afferents r.p., right peroneal, and i.p., left peroneal. Decerebrate cat. R.V., right vastocrureus muscle ; L.V., left fellow muscle. Time above in fifths of seconds. In addition to the rhythmic stepping, which is here somewhat rapid, there is a minute tremor of much more rapid rate (see text) engrafted on and disturbing the stepping retlex’s rhythm. 1913.] arsing from Rialry of Antagonistic Reflexes. 241 (op. cit., fig. 1). By varying the conditions of experiment various transitional forms can be obtained between imperfect irregular rhythmic movement and the complete regular rhythmic series of movements constituting stepping. Figs, 6, 7, and 15 furnish some illustration of this. UN Turning now to analyse some of the features of the rhythmic stepping the example in fig. 1 may be used. The observation begins with stimulation (faradic) of the right peroneal nerve, causing immediate inhibitory reflex relaxation of the right knee extensor with synchronous reflex contraction of the left fellow muscle ; both relaxation and contraction are perfectly steady and arhythmic. After 2°5 secs., during which time the left muscle has remained perfectly steadily relaxed and the right perfectly steadily con- tracted, stimulation (faradic) of the left peroneal nerve is commenced, that of the right still continuing unchanged. The steady relaxed condition of the right muscle is at once broken by a contraction and at the same moment the steady contracted condition of the left muscle is broken by an inhibitory relaxation. The r. contraction and the 1. relaxation culminate synchronously in about 0-4 sec., and then die out about as rapidly as they appeared, to reappear and similarly culminate synchronously again. In this way they rhythmically and reciprocally appear and reappear in series so long as the concurrent stimulation of the two nerves right and left is kept up, namely, for 6°5 secs., seven complete steps being taken in that time by the extensor muscle of each knee. The r. stimulus, the one commenced with, was then with- drawn. The stepping immediately ceases in each muscle, except that the r. muscle, which, having begun its stepping by contraction, is at the end of the seventh step, relaxes, carries out a half-step more and passes into steady contraction, thus assuming an attitude of full extension of knee; and that similarly the 1. muscle, which, having begun with relaxation, is at the end of seventh step, contracts, executes a half-step more, passing into relaxation and assuming a posture of flexion of knee. These final half-steps and assumptions of reciprocal states at the two knees are, of course, due to the action of the stimulation of the left nerve now remaining unopposed by any concurrent stimulation of the antagonist right nerve. The 1. nerve stimulation remains in operation until withdrawn 3°5 secs. later. During its sole action no trace of rhythm appears in either muscle. On its withdrawal the contraction of right muscle at once begins to decline, although, the preparation being the tonic one, the “shortening reaction” of “ plastic tonus ” has taken place, and the tonic shortness of the muscle still persists after withdrawal of the stimulus. In the left muscle on withdrawal of left nerve 242 Prof. C. 8. Sherrington. Nervous Rhythm (Feb. 3, Hai stimulus no obvious change occurs, because the stimulus is inhibitory and nh ( under it the muscle is already relaxed and relaxes no further and no less on i cessation of the excitation, a “lengthening reaction” having occurred at end i ka! of the r. nerve stimulation. 1 3) | Three seconds later the left nerve stimulation is recommenced ; steady con- | traction of r. muscle is reassumed; 1. muscle being already relaxed the inhibitory effect there is not apparent, although really fully existent, as subsequent events show. The 1. nerve stimulus is here the pre-current stimulus, and under it no rhythm of any kind appears in either muscle | any more than did under r. nerve stimulus in the preceding observation. | But on then applying stimulation of r. nerve, that of 1. still continuing as i‘) before, rhythmic stepping at once appears. It begins synchronously in Ba | right and left muscles, the opening phase being contraction, 7.c. extension i Hi of limb, in the left muscle, and relaxation, ie. flexion of limb, in the | right muscle. On withdrawing the r. nerve stimulus, about 6 secs. later, when the last step of a sequence of six has been nearly completed, the stepping ceases abruptly with completion of that step. The right muscle then reverts to steady contraction, the left to full relaxation. This resumption is, of course, the effect of the still-continuing 1. nerve stimula- tion, and guarantees that the stimulation of that nerve has remained effective throughout. Rather more than a couple of seconds later the l. nerve stimulation is withdrawn and the contraction of right muscle, in consequence, shows decline, modified, however, by the shortening reaction of the plastic tonus. Two seconds later r. nerve’s stimulation is once more commenced. It evokes steady contraction of 1. muscle and, synchronously with that, full inhibitory relaxation of r. muscle. During the 3°5 secs. for which this stimulus remains thus in operation by itself, the steady reciprocal reaction of the two muscles is maintained unchanged. The stimulation of 1. nerve is then commenced, that of r. nerve continuing unaltered. Synchronous stepping, reciprocal in direction in the two muscles, sets in at once. Here the opening phase in each muscle is the reverse of that in the previous observation. The 1. nerve stimulus is maintained in concurrence with that of r. nerve for 7 secs., and during that period a sequence of six complete and regular steps is performed by each muscle. Stimulation of 1. nerve is then withdrawn, and both muscles at once revert to the non-rhythmic steady state they exhibited previously under stimulation of r. nerve alone, r. muscle being in steady relaxed condition, and 1. muscle held steadily con- tracted. Later, the stimulation of r. nerve is itself finally withdrawn ; the contraction of 1. muscle then at once begins to decline into pure stato- 1913.| arising from Rivalry of Antagonistic Reflexes. 243 tonus, and r. muscle shows a slight return to greater tone than it had during the influence of r. nerve’s stimulation. Although the results, therefore, appear curiously simple, the conditions underlying them are a complex of various factors. Without pretending to offer any full analysis of these, some of them may be here briefly adverted to. 1. Influence of the Laterality of the Stimulus. Decisive points here are (1) that reciprocal innervation holds for the symmetrical muscles, (2) that with extensors the direction of reciprocity is excitation contralateral and inhibition ipsilateral, with flexors conversely, and (3) that the rhythmic stepping here concerned arises, proceeds, and ceases with commencement, continuance, and withdrawal of the exercise of double reciprocal innervation. In such observations as shown in figs. 1, 4, and 9, we may term the stimuli according to their sequence “ pre-current,” “added,” and “remaining.” It will be seen from the figures that the added stimulus when extensor muscles are used always causes, as its first effect, the contraction phase of the contra- lateral muscle’s reaction, and the relaxation phase of the ipsilateral muscle’s reaction. That is, it causes, as its first effect, the extension phase of the step at the contralateral knee and the flexion phase of the step at the ipsilateral knee. This is in agreement with the rule observed in the case of stepping provoked by direct faradisation of the spinal cord. As noted previously,* I found that regular stepping of the hind limbs is readily provoked in the spinal animal by unipolar faradisation applied with a stigmatic electrode directly to a certain area of the distal cut face of the cord in the cervical region. If the electrode be applied in the right lateral half of the cord the stepping begins with flexion phase of right hind limb, and extension phase of step in left hind limb. It might have been supposed that the reflex stepping produced by the double reciprocal action of symmetrical r. and 1. afferents upon the knees r. and |. symmetrical extensor muscles must of necessity be bilateral. Experiment shows, however, that that is, in fact, not the case. The reflex stepping so obtained, although usually bilateral, is sometimes unilateral. Thus, figs. 4 and 5 are from the same experiment at short interval. In the former the stepping is bilateral, in the latter it is confined to the left muscle. The observation in fig. 4 was obtained with values of stimuli right nerve 15 cm. left nerve 17 cm. as measured on scale distances of the two induction apparatus. * ‘Journ. Physiol.,’ 1910, vol. 40, p. 86; ‘Brain,’ 1910, vol. 33, p. 13; also Roaf and Sherrington, ‘Quart. Journ. Exper. Physiol.,’ vol. 3, p. 210; and G. Brown and Sherring- ton, zbzd., vol. 4, p. 202. [Feb. 3, S Ss ~~ > RS 4 S S S i— =; Prof. C. S. Sherrington. 244 ‘g “By sv yueulsedxe owes wo1g “gp ‘d 4x0} UL UoLydi1osep Site) LI ‘d] “tuo @{ “da BUTT, ‘afPOSNUT sneIN10098vA 4JOT “ATT { G[osutu suadnas0j9svA AYSIU “AY ‘jeeuoaed 4yo] “d-7 puv ‘Teouosed 4y Sta “dw syuoIEye ostuoseqzue RELICB CIN | SONIA SN|NUIYG ‘shNuys JUeTAINOUCD eYyy JO [VARAIPYIIA jo QUOLUOUL YALL Sess0I, ‘SpU0dES JO SyIJY Ul oAoge ey} JO (WOIFRSIPLART) TOLAL[NWYS JUaTIMOUOD Aq PaxOAd “QJoT PUL FYSIA ‘SeaUy Jo o[OSNUL Aosta3xe pezEjost Jo Sutddays jeoosdtoa1 xoyoy—yP “DIT os | mars me oT S Ath kh 1913.] arising from Rivalry of Antagonistic Reflexes. 245 The threshold stimulus for r. nerve had been found slightly higher than that for left in this preparation ; the two stimuli although not equal (as shown later v. infra) were certainly not far from equal. Then, for obtaining observation, fig. 5, the r. nerve stimulus was increased to 13 cm. and 1. nerve stimulus to 16 em. Both stimuli were therefore increased, but that of r. nerve much more than that of |. nerve, the steepness of ascending intensity of the physical stimulus being much greater between 15 cm. and 13 cm. than between 17 cm and 16 ecm. of the scale. This unequal increase of VEIL YYYIT YY LIVI TI III III LILI IIT ITI ITI LIVI LONI ILOILO IVT IIIT | EV: LV. Ap. rp.t Fia. 5.—Reflex unilateral stepping as exhibited in an isolated extensor muscle of left knee, the similarly isolated extensor of right knee being relaxed by inhibition while the fellow muscle steps. Decerebrate cat. From same experiment as fig. 4, but with different combined stimulus value, namely peg me, rp. and l.p., right and left ip: cm. peroneal nerves. Time above in fifths of seconds. Further explanation in text, p. 245. intensity of stimulation of the two nerves r. and |. is answerable for the change from bilateral stepping to unilateral. It suppresses the stepping in the muscle on the same side as the stronger stimulus, and at the same time makes the stepping of the other, the left, muscle faster than it was in fig. 4. It brings this about in the following way. With increase of the strength of the stimulus its ipsilateral inhibition increases more than does its contralateral excitatory effect. When (fig. 4) r. stimulus has a value of 15 em. it does not suppress the contralateral excitatory effects of 1. stimulus at value 17 cm., but pe pre rere rer ance arree RESSIOSSOEE a : 246 Prof. C. 8. Sherrington. Nervous Rhythm [Feb. 3, r. at value 13 does suppress all contralateral excitatory effects from |. at value 16. The r. muscle therefore in fig. 5 does not step. But in the 1. muscle i.’s stimulus value 16 restrains the contralateral excitatory effects of r.s stimulus value 15 even less than was the case when l.’s stimulus was value 17 and r’s stimulus value was 16. Hence the 1. muscle steps faster in fig. 5 than in fig. 4, r.’s excitatory effect breaking through 1.’s inhibition more rapidly and frequently. This quickening of the step under the stronger stimulus harmonises with the observation that though the direct faradisation required to excite stepping from a point in the cross-section of the cervical spinal cord is of quite weak intensity, the rate of the stepping of hind limb so produced increases, ceteris paribus, with the intensity of the faradic stimulus. 2. Strength of Stimulus. As to the strength of the antagonistic stimuli which by their concurrence evoke the stepping, the phenomenon, in my experience, is not obtainable with strong stimuli. With stimuli just above threshold intensity I have at times seen traces of the rhythmic undulation; but attempts to develop it with such very weak stimuli have not so far attained much result. Stimuli rather stronger than such but on the weak side of moderate have yielded the best results. Weakening the stimuli beyond that point gives a rhythm not only slower but more irregular with waves of varying amplitude (fig. 6A), and a tendency to pauses between some of the beats. On the other hand, with too strong stimuli there is a tendency for the rhythmic reaction to be suppressed in one or other member of the muscle-pair, and for it to be represented in the other member by a few sharp somewhat irregularly explosive beats separated by longish unequal intervals (fig. 6B). An idea of the range of intensities suitable may be afforded by the data of an experiment. In the experiment from which figs. 64 and 6B, and also fig. 4, have been taken, the threshold value for reflex effect both ipsilateral and contralateral lay for r. nerve at 17°8 cm. and for 1. nerve at 19°2cm. This seems high, but it must be remembered that a resistance box of 100,000 ohms was included in each circuit. Slight but distinct rhythmic stepping was obtained by the combination yr. at 16 cm. and 1. at 18 cm. | =): fig. 64. More regular and stronger rhythm was obtained with ie ILS) we, 144 ie, 113} various stronger combinations, ¢.g., 7 (fig. 4), Tay’ Cis Results with RS eal; Sete dane ee JEL Liss were less good. A result with Liss no stepping with right muscle, and merely two short unequal steps with is given in fig. 6B; it exhibits 1913.] arising from Rivalry of Antagomstic Reflexes. 247 left muscle in a period of concurrence of the stimuli covering more than 3 secs. With ee LU and all stronger combinations tried, there was obtained 2 no rhythmic stepping movement at all, although excellent steady contractions and inhibitions. Weaker stimuli tend, in my experience, to give slower rhythm. Other factors in the production of the rhythm, for instance, the ratio between the LIAO DATO IIITTT TILL TT IVD SOIT LIN LDN TITY OI INTII IY A a" Fic. 6.—Reflex rhythm (imperfect stepping) evoked in the isolated extensor of knees, right and left, by concurrent stimulation of the antagonistic afferents, right and left peroneals, 7. and Up. Decerebrate cat. In Observation A the stimulus values are too weak T:P: 16 cm; in Observation B they are too strong tet LU ae both lp. 18 cm. I.p. 12°5 em. observations are from the same experiment as fig. 4, where good and regular reflex r.p. 15 cm. l.p. 17 em. worthy that the stimulation of U.p. is so weak that it of itself produces no obvious contraction of the contralateral muscle and only the merest trace of relaxation of the ipsilateral muscle. Yet its presence is documented at once when the concurrent stimulus is added in the response being then not steady but rhythmic. Further explanation in text, p. 246. stepping is obtained with stimulus values In Observation A it is note- strengths of the antagonistic stimuli, change, however, of necessity also in such comparisons. Still my facts, as far as they at present go, clearly indicate the above tendency. Thus figs. 4 and 6 are from the same experiment. In fig. 6A the stimuli were a and the beats are 9 during 8:5 secs., while in fig. 4, observation 3, where the stimuli were i the beats are 7 during 5:2 secs. Again fig. 7, observations A, B, and C are all [Feb. 3, Nervous Rhythm Prof. C. 8. Sherrington. 248 give five steps re, ib4e NG In fig. 7A the combined stimuli from one experiment. but strengthening stimulus r. to 13, so that the combined 5) in 14 sees. secs., and x oO 7B) in 13 steps during 17 o to) results (fig. re, 116} 16’ iL. both stimuli further still to stimuli become produces (fig. 7c) five steps me, UY 13°5 1 og to) ethenin tren s during 3°8 secs. "1¥G ‘d “4x04 Ul uoyeuLdxe JoyyIny ‘spuoves Ronin Gc tang 7 ee ee eens Uo eT “a gt ‘dy qyg a q, tL “a gl da ° Q toy pue “5 er da Ot? & 103 SES FL da O28 V UoHeAIaBGO 40y SONJA SNINWUYG ‘spueteye se soatou ‘eauoted 4yo] “d'? pure ‘jeouosed 4ysta “dy yo uOYeNUIYS yUeIINOUOD : kq pexoao (q%0 9q'e1q91900p) 99Uy 4J9] JO “AT GjOSNU 1OSsuUs4xe psyze[OsI Jo Suiddo4s orang Aga xopoy—'), “Oly re) PITTA MM ITAA aa TTL TYTN eT j y PITT TT ne 1913.] arising from Rivalry of Antagonistic Reflexes. 249 More satisfactory for comparison is fig. 8 with fig. 5. Both are observa- tions from the same experiment, with no long interval between them. The ], stimulus was of the same value in both, namely, 16 cm. But the r. stimulus had value 15°5 cm. in the observation fig. 8, whereas in that of fig. 5 its value was 13 cm. The results contrast in three striking features. Where r. stimulus is stronger (fig. 5) the rate of stepping is six MOY YIN NINN NININTY Fie. 8.—Reflex reciprocal stepping as exhibited by isolated extensor muscle of knees, right and left, evoked by concurrent stimulation of antagonistic afferents, right and left peroneals, 7p. and /.p. Decerebrate cat. Time above in fifths of seconds. Compare fig. 5. Explanation in text, p. 249. Stimulus value a double phases in 4°8 secs., as against four double phases in 4 secs., where r. stimulus is weaker. Where r. stimulus is weaker (fig. 8) the stepping is not so well maintained as where r. stimulus is stronger. Where r. stimulus is weaker (fig. 8) the stepping occurs in the fellow muscle of the right side as well as in the left, and where r. stimulus is stronger (fig. 5) the stepping is confined to the left-side muscle. 3. Influence of Intensity-ratio between the two Antagonistic Stimuli. From the’ preceding section it is clear that as might be expected the proportion between the intensities of the two antagonist stimuli is an 250 Prof. C. 8. Sherrington. Nervous Rhythm [Feb. 3, important factor in determining the existence and characters of the rhythmic reflex. Evidence of this was constantly met in the experiments. In fig. 7 the two observations A and B illustrate an aspect of it. Stepping is more marked in B than in A although the intensity of one only of the rival stimuli was altered. Conversely, the change of intensity by even a little of one of the antagonist stimuli may make all the difference to whether the combination be effective or not for rhythmic stepping. My experience put generally is , of hips, right and left, The direction of the D pe xation contralateral. Time above in rp. and lL, y isolated flexor muscles, R.F., L.F. n ipsilateral and inhibitory rela The muscles, both right and left, were psoas and tensor fascize femoris. ght.and left peroneal nerves, ibited b ed by concurrent stimulation of ri reciprocal innervation is here contractio Decerebrate cat. k fifths of seconds. Fig, 9.—Reflex reciprocal stepping as exh evo =“ von) N Les. stic Refle: gon y of Anta valy d from R “ arising 1913 ] "ggg ‘d 4x0} ul uoIeUR[dxoe Joyjang ‘oandy oyg Jo suoTeAsosqo 4seT Omg 04) Jo Burddoys yeroywpIUNn ay) aald Quodanouo LOY “TIM TNMs opeaedos OMY OY} JO SuOyoveL oyvAedes oY} MOYS OINSY oY} UL SMOIYBAIOSqo OA4 ae) a : ‘ ; ial epee t : WI UL “A ep da SONTEA SNINUNGS ‘Spuooos JO SIFY UL OAOGL OWL], ‘ePOSNU sheIns009seA 4Jor “AT { oposnut snoansd -OFSVA JUSTIA “AT "gvo oyviqeaatede(T ‘yeouorad gyo, “a7 pur ‘eouo10d 4yoit “ds ‘SoAdoU JUOAOYR OYSIUOSRIUR OY JO TOIQL[NUITS quormouos Aq Quix oy} JO 4vyy UL you ynq ‘oouy 4Joy ayy JO O[OSNUT OSUE}xe pozETOST ey} Ul poonpul Sutddoys xoyoyy—‘OT ‘Ol d PO a Tal oe Nae Te. a. >}! (ee ee ss Se -—\- a eee a Ay XG Whe, i Et oe rl Ta Ree 6 ee AT JS VOL. LXXXVIL—B. 252 Prof. C. 8. Sherrington. Nervous Rhythm | [Feb. 3, that the antagonist stimuli must not be very unequal in intensity if they are to provoke the rhythmic reflex (fig. 9 shows this with flexor muscles), but that on the otherhand the stimuli may yet succeed in producing the rhythmic reflex when quite indubitably and markedly unequal in intensity, if that inequality does not go beyond certain limits. Fig. 10 will serve as illustration. In this experiment the thresholds of r. and 1. nerves lay at 19:5 and 20°d respectively. The physical cireuits were hardly appreciably unequal, for the thresholds changed only to 19:2 and 20°8 when the coils were interchanged for the two nerves by the double YDIVVSITIVIYIVIIVVL LIVI IY IVI LYYIVIVIVIVIVIIVY IY NIV IVI VIII IVI VIVIV IV III ITV PVT IIT INI “ 2 Fie. 11.—A series of reflexes, phasic and tonic, provoked in the isolated right and left vastocrureus muscles by short non-concurrent faradisations of right and left peroneal nerves. The numerals mark some salient corresponding events in the signal lines and myograms. The inhibitory relaxations produced in right vastocrureus, R.V., by stimulation of 7.p. are followed by rebound contractions ; similarly the inhibitions of L.V. are followed by post-inhibitory rebounds on withdrawal of each stimulation of i.p., left peroneal. Besides the reciprocal reflex movements induced by each stimulus, the stimuli cause assumption of reflex postures, e.g. under stimulation of r.p. the two myogram lines approach each other, z.e. r. muscle is relaxed (indicating that flexion of right leg would be going on, were the muscles not paralysed) and |. muscle is contracted (indicating that left limb would be extended). Under stimulation of l.p. the two myograph levers diverge, indicating that left limb assumes posture of flexion and right limb that of extension. This figure gives the separate effects of the two stimuli which, when concurrent, give the stepping reflex recorded in fig. 1. Decerebrate cat. Time in fifths of seconds. 1913.] arising from Rwalry of Antagonstic Reflexes. 253 switch. This being so we must consider the stimuli r. 13 and 1. 16 distinctly unequal, Their separate effects on the preparation are shown at the beginning of fig. 10. It is noticeable there that the rebound contraction after the inhibition produced by r. 13 is greater than after that produced by 1.16; also that r.’s contractive effect on L muscle is better maintained than is 1.’s on R muscle. Yet this combination produces, as the figure shows, good stepping in one of the muscles. Now this fig. 10 is from the same experiment as furnished r. 14 And =: 13 te 16 must be con- fig. 1 and the combined stimuli in fig. 1 are sidered less equal than re. yet the rhythmic stepping of the left muscle in fig. 10 is as good or better than that in fig. 1. Now the stimuli used in fig. 1 were themselves unequal, not merely on the face of their scale values but as tested on the preparation at the time. Fig. 11 shows the effects of these two stimuli when employed separately ; the tonic effect is obviously not equal but also not far from equal; but (fig. 12) the rebound is greater after inhibition > IDV IIL MMIII NII OI NII TY TON INITOOOYITOYNIIY I OIIYYTOOIIII yy AN a — Moet e MIE Lp Fia. 12.—Right knee extensor (R.V.) and left (L.V.) isolated and reacting reciprocally to reflex stimulation of each peroneal afferent ; 7.y., right peroneal, /.p., left peroneal. Time above in fifths of seconds, Decerebrate cat. ma2 254 Prof. C. S. Sherrington. Nervous Rhythm [Feb. 3, by r. 14 than after inhibition by 1. 17. And when, in fig. 13, the r. and 1. stimuli are applied synchronously (observation 50) and begin together, r. stimulus obviously has the upper hand, for L muscle at once contracts, and though not seen in the figure R muscle at once relaxed. Moreover on with- drawing the two stimuli together L muscle at once relaxes instead of showing rebound contraction. Clearly r. stimulus is stronger than 1. stimulus in its effect on the preparation. Observations 50, 51, 52, 54, fig. 12, in which r. and 1. stimuli are applied together, do not differ in result on the muscle from observations 48 and 49, fig. 12, in which r. stimulus was applied alone, except in the feature that some little while after commencement nl eavaun PU ey LULL | | u thi TM Fie. 13.—Extensor muscle of left knee L.V. Observations 49 and 50 show reflex contractions evoked by stimulation of right peroneal nerve, 7.p. Then 50-54, a series of synchronous stimulations of 7.p. and l.p.; during these there ensues stepping ; the inhibitory phase (flexion phase) of the step appears as a gradually increasing inhibitory notch on the contraction caused by 7p. Time above in fifths of seconds. pf each of the double stimulations (observations 50-54) an inhibitory notch ppears in the reflex contraction, whereas in observations 48 and 49 there is othing of the kind. Reserving this point of difference for the present, the result here is that of the two stimuli r. 14 and 1. 17 the former is the stronger not only in its numerical scale-value but also as actually tested on the preparation. Yet this pair of unequal stimuli give a good rhythmic reflex during their concurrent application (figs. 1 and 13). If r. 14 and 1. 17 form an unequal pair in which r. is the stronger, obviously r. 13 and 1. 16 must form a pair more unequal still. Yet ae gives a good rhythmic reflex in one of the muscles (fig. 13). Certainly, therefore, in order to produce the rhythmic reflex, the two antagonistic stimuli, right and left, need not be exactly equally balanced 1913.] arising from Riwalry of Antagonistic Reflexes. 255 but experience points to the necessity of their being not widely unequal, also to the need for closer balance of r. and 1. stimuli for evoking bilateral rhythmic reflex than for evoking a unilateral one. Results referred to at the outset of this communication throw some light on this. It was there said that as the intensity of stimulus used for evoking the reciprocal reflex on the muscle pair is progressively increased, the intensity of both ipsilateral inhibition and contralateral contraction increase, but that of the former more rapidly than that of the latter. Synchronous application of r. and 1. stimuli of equal strength produces when the stimuli are strong suppression of contraction both r. and 1.; but with weak stimuli bilateral contraction, contraction of both muscles, results. The same result is shown by determining the strength of contralateral stimulus required to force its reflex contraction through an already established ipsilateral inhibition. The kind of result then met with is as follows: A preparation where threshold for r. peroneal was 164 cm., and that for 1. peroneal 15:2 cm., yielded the figures— cm. cm. Inhibition of R. muscle by r. stim. 13 was broken by reflex contract. due to 1. stim. 14 9 ” 12 ” ” 79 125 x - 11 required 1, 10 cm. to break it through. ” ” 105 ” 1. 9 ” ” Such results indicate a relation between intensity of stimulus and intensities of ipsilateral and contralateral reflexes such as is sketched in the diagram. The diagram accounts for the relation between the observations of ponse. R. extensor, Degree of reflex res Degree of reflex response. L. extensor TOmre te On 15) leo WL i ee EO, 9) Distances of secondary coil. 256 Prof. C. 8. Sherrington. Nervous Rhythm [Feb. 3, fig. 4 and fig. 5, both from the same experiment; also for that between the observations of fig. 1 and fig. 10, both from the same experiment. As drawn, the diagram applies to extensors (of knee), but by reversing inhibition to excitation and conversely it is applicable to flexors (of hip and knee). 4. Influence of the Sequence. It is noticeable that the effect of concurrence of the stimuli opens with a result in the direction of that of the added stimulus practically at once on addition of that stimulus. Under the concurrence the newer stimulus tends to dominate at once. This is in harmony with experience on visual rivalry, struggle between rival contours, etc. The very newness of the new stimulus lends it force as against the pre-existent, especially where the pre-existent is not very much the more potent and has been in operation for some time. Probably a similar relation explains why although well-acting combinations of the antagonist stimuli produce the rhythmic reflex whichever of the two stimuli precedes, with less effective pairs of stimuli that is not the case. It appears with the latter a point material for the result which of the two. precedes and which follows (fig. 2). The rhythmic reflex may be much better if x precede y than vice versd, or it may not appear at all in one sequence LIVI INI IIIT NINN Bu a Fie. 14.—Isolated extensor muscle of right R.V. and Jeft L.V. knees. ynchronous stimulation of both right and left peroneal afferents (2. per.), (Z. per.) The reflex opens with identical contraction of both the muscles, but almost immediately the reaction of the left muscle is changed to inhibitory relaxation, with the result that the symmetrical muscles fall into step with reciprocal harmony. Decerebrate cat. Time above in fifths of seconds. 1913.] arising from Rivalry of Antagonistic Reflexes. AES though appearing with the other. A particular combination which works need not necessarily work reversed, though it generally does so. As to what happens when the two rival stimuli are started together, the observations have shown that sometimes then the reflex step starts at once with full reciprocity of phase in the two muscles, and so proceeds co- ordinately in that manner. Sometimes, however (fig. 14), the muscular effect begins identically with contraction in the two muscles, and this is almost immediately checked and reversed in one of the muscles so that that muscle then begins, although a little late, to behave reciprocally in regard to its fellow. In fig. 14 this correction takes place in time for L. muscle, although starting wrongly, to have got into harmonious reciprocal step with its fellow before even the completion of the first phase of the first step is reached Harmonious reciprocity then continues. When the two rival stimuli are started together there seems a tendency for the stronger to overcome the weaker altogether at first, and then to give way to the latter a little later, and then later still to re-establish ascendancy, again soon losing it, and so on. Thus a see-saw alternation of dominance is arrived at. Observation 54 in fig. 13 exemplifies this. The weaker stimulus (v. supra) 1. 17 there makes its effect felt as a deep inhibition of a contraction already initiated by r.14. And in the series of immediately precedent observations 50—53, this effect of 1. 17 is seen to have each time occurred with the same time-relations, although with successively increasing effect. Evidence in the same direction is illustrated in fig. 7a. 5. Influence of the Afferent Nerve of the Reacting Muscle. The rhythmic effect of the antagonistic concurrent stimuli takes place when the reacting muscle has been de-afferented. The presence of the proprioceptive afferents of the muscle is therefore not necessary to the reaction. Fig. 15 shows the rhythmic stepping obtained from the right vastocrureus by con- current stimulation of the antagonistic r. and 1. popliteal nerves. (The signals in this experiment were set to mark upwards, not downwards as in the other figures of this paper.) The observation begins with faradisa- tion of r. nerve producing, after the initial spike (Anfangstetanus), inhibition; but the muscle being already relaxed, the preparation being decapitate, the inhibition causes no visible further elongation of its muscle, except the suppression of the initial contraction which it itself had provoked. After 2°5 secs. the stimulation of the contralateral popliteal js commenced, the stimulation of ipsilateral proceeding unaltered. The result of concurrence of the stimuli is immediate rhythmic stepping of the muscle. Four steps are taken during the continuance of the concurrent 258 Prof. C. 8. Sherrington. Nervous Rhythm (Feb. 3, Fie. 15.—Isolated de-afferented extensor muscle of right knee. Decapitate cat. Faradisation of central end of right popliteal nerve for 6'5 secs., joined 2°2 secs. after its commencement by faradisation of left popliteal nerve ; the latter stimulation is continued until 2 secs. after cessation of right popliteal stimulation. The signal marks are directed upward instead of downward as in the other records. During the concurrence of the stimuli, but not when either of the two stimuli is in operation alone, the muscle gives rhythmic reflex stepping ; it completes four steps in about 4 secs. The muscle had been de-afferented (118 days). The proprioceptive reflex—“ shortening reaction ”—is therefore absent on withdrawal of the contra- lateral stimulus, and the muscle relaxes at once, being toneless. Time above in seconds. 1913.] arising from Riwalry of Antagonistic Reflexes. 259 stimulation, #.c., in 4°5 secs.; the ipsilateral stimulus is then withdrawn, and under the influence of the remaining contralateral stimulus the muscle enters at once into steady maintained contraction, and continues so contracted until the stimulus is withdrawn. The muscle had been de-afferented October 7, 1908, and was used for experiment nearly four months later, February 2, 1909. In the experiments of Graham Brown* it has been shown that stepping occurs after de-afferenting the muscles involved. 6. Influence of a Component Stimulus subsequent to its Withdrawal. In the rhythmic reflex a contraction-phase (extension-phase of step) which is in course of execution does not cease immediately on withdrawal of the contralateral stimulus. On the contrary it continues its course for a brief time and ends in smooth transition and reversal into relaxation. This want of abruptness is strikingly different from the abruptness with which a reflex inhibition excited by an ordinary stimulus against tonus often commences. The excitatory phase thus continued after withdrawal of the contralateral stimulus is frequently less ample than its predecessors where the stimulus was not withdrawn (fig. 16, observation 2). A small final step to the series is thus produced (fig. 1, observation 1). If the contralateral stimulus be withdrawn during the course of the relaxation phase of the rhythmic reflex, i.e. during the flexion phase of the step, often no subsequent contraction- phase ensues (fig. 4, observations 3 and 4), even although the withdrawal occur late in the course of the relaxation phase (fig. 4, observation 3). But if the contralateral stimulus be relatively strong and the ipsilateral relatively weak, and the former be withdrawn just before the end of the relaxation phase, a small and somewhat delayed subsequent contraction phase may ensue (fig. 10, observation 3), and even be followed by one still smaller and still more delayed (fig. 1, observation 3). Conversely with the withdrawal of the ipsilateral stimulus. If this cease during a relaxation phase that phase is completed more or less amply after the stimulus withdrawal (fig. 4, observation 1). If it be withdrawn during a contraction-phase no relaxation phase usually follows, but there may be just the commencement of one if the ipsilateral stimulus is withdrawn at end of a contraction phase (fig. 16, observation 1), and if the ipsilateral stimulus as compared with contralateral be not too weak. * ‘Roy. Soc. Proc.,’ 1911, B, vol. 84, p. 308. [Feb. 3, Nervous Rhythm Prof. C. 8S. Sherrington. 260 “spuooes JO SIFY Ul Ao" OUITT, “7B o7BIGQe1000(T ‘ssoro @ AQ poYACUL SI SN[NUITAS oY} JO [BMLIPYIIM JO FUSUIOU OYJ, “AAOYS SI Ajsnowosd ysnl snpnuys A10zRz1Ioxe eyy Jo [eMBIPYFIA qoqye esvyd (10su9e}xe) UOTJoRI}UOD OY} JO SuUBATVAp 944 J UOIZBATESO UT “SSod0 OY} TOF Fe qsn{ Sur1ain090 Yoyo [[VUIs 94} SUIEq Joao S41 TeMBIPYIIA'S}t 0} quonbesqns snpnunys AroqiqryUr oy} Jo Jooyo-1oxzZe oy} SoPBAASNITT eINSY oY} [ UOIWVAIESqQ UT ‘SUOTSBDD0 SATSSeDONS OM UO TJeM ATenbo YOM you seop VOTYBUIGUIOD sN{NMIys eUES oY} PUB BuO Fey} SMOTS emsy eyy, ‘a7 pue ‘ds ‘syue1oye jeouosed qJo] pue JUSTE oy} Jo UorZR[NUNTYs yuettnoucd Aq peonpoid Suiddoqs xeyet Surmoys “AT eoUH 4Jo[ JO efsNUL JOsUe}xe Peze[OST—9L “Ol ee ak ad fees Te ae Na Ne re Ay AT ARUSSSRR ERMA SUDARARORORASDADDDARDANAI BEADED REAAO OLD OED URE RDERASRREGANINOR RADAR DRI A ARE PEEP REP EEE EPEC E ELECT EAPC ELe TET EET PEEETTAT ET PEPER MTT POCPEPT ET EEEET Mere Tee PEPE E Eee T 1913.] arising from Rivalry of Antagonistic Reflexes. 261 ITV. SUMMARY OF CONCLUSIONS. It is shown that taking an afferent nerve which produces steady reflex excitation of the muscle, and another which produces steady reflex inhibition of the muscle, it is possible by stimulating both nerves concurrently to obtain regularly rhythmic contractions and relaxations of each member of a pair of symmetrical muscles, the phases being reciprocal in the two. To do this requires certain somewhat narrowly adjusted proportions of strength of the two paired stimuli. The stimuli are both of them continuous, in the sense that they are faradic and of a frequency (about 40 per second) much above and bearing no causal relation to the rhythmic reflex produced. In the rhythmic reflex the right and left muscles each contract and relax alternately and move reciprocally, the contracting phase of right muscle being syn- chronous with the relaxing phase of left, and conversely. This rhythmic reflex is shown clearly to be reflex stepping. In short, under the rivalry of the two opposed and so to say equipoised continuous stimulations th limbs exhibit reflex walking. With certain other paired intensities of stimulation of the two antagonistic afferent nerves, it can be arranged that only one muscle of the pair shall step—the left muscle if the right nerve stimulus be the stronger, and conversely. During this unilateral walking or running the other leg is kept steadily flexed by the reflex, i.e. the extensor muscles are kept steadily inhibited. 262 Herbage Studies. I1.—Variation im Lotus corniculatus and Trifolium repens (Cyanophoric Plants). By H. E. Armstrone, F.R.S., E. FRANKLAND ARMSTRONG, and EDWARD Horton. i (Received January 1,—Read January 23, 1913.) Lotus corniculatus. In Part I of these studies, it is shown that Lotus corniculatus is a plant in which a glucoside containing cyanogen is frequently present together with the corresponding enzyme. During 1911 we were able to make observa- tions practically over the whole of Europe, owing to the assistance we received from Dr. Eyre, which led us to the conclusion that the glucoside and enzyme were normal constituents of the plant in almost all districts, though occasionally, in close proximity to plants which were cyanophoric, others were met with in which little if any cyanide could be detected. In Scotland, in South Ayrshire, the plant was uniformly acyanophoric, except on the coast ; nor could cyanide be detected in plants collected in Norway. Lotus major, which is a sufficiently distinct variety to have been recognised by botanists as a separate species, was uniformly free from cyanide and also apparently from enzyme; but no regular distinction could be made between the various other forms which botanists look upon as merely varieties of the plant. Lotus major, it should be added, always affects damp situations and is a rank grower; it can be distinguished by the manner in which the large number of flowers in the umbel spread out from a common centre instead of at intervals from the flower stalk; the calyx teeth also tend to spread outwards, whereas in other forms they are almost uniformly strongly incurved. We have found the double form of JZ. corniculatus (var. pleno) to be strongly cyanophoric. During 1912, we have again examined specimens of L. corniculatus from many localities in England and Wales and, as a rule, have found them to be strongly cyanophoric.* Out * During 1912, in testing for cyanide, the Guignard picrate paper used was always prepared as required and we have substituted toluene for chloroform in order to avoid the possibility of any trace of acid being introduced. In the morning, before going into the field, strips of paper were impregnated with the alkaline picrate solution and the moist strips were at once placed in the tubes which were to be used in testing the plants. The tubes were then incubated in the pocket, in order to ascertain whether any hydrogen cyanide was retained in the cork. Recently prepared undried paper is usually more sensitive than paper which has been dried and then moistened just before it is to be used ; the once dried paper rarely has the bright yellow appearance of freshly stained paper. Herbage Studies. 263 of about 30 specimens collected day by day by a cyclist during a fortnight’s tour from London to Wales and back, in the earlier part of July, only two were found to be almost free from cyanide—one of these was obtained at Church Stretton, Somerset, the other from the foot of Cader Idris, Wales. Specimens sent to us by Mr. Pickering from Woolacombe, South Devon, and by Mr. Stapledon (LZ. incanus, Gray, L. villosus, B. and H.) from Westward Ho, North Devon, were strongly cyanophoric ; on the other hand, material sent to us from near the Lizard, Cornwall, gave no response to the test. At the end of June, one of us found the plant growing very freely, in full bloom, on the retaining wall at the foot of the hill slope bordering the whole Jength of Rydal Water, Westmorland; of seven specimens, presenting no difference in appearance, picked from this wall at fairly regular intervals, all but one were more or less strongly cyanophoric ; the exceptional specimen contained the faintest trace, if any, of cyanide. During August, a very thorough study of the plant was undertaken by one of us, chiefly in the valley of the River Cree on the borders of Ayrshire, Kircudbrightshire and Wigtownshire, in the district where the previous year, at Whitsuntide, the specimens examined were all acyanophoric. Again, in the places visited in the previous year, many specimens were found in which cyanide could not be detected ; here and there, however, along the river bank and in the adjoining fields, patches of the plant were met with now and then which were faintly cyanophoric. On walking along the high road, across the moor, from Drumlamford, about five miles from Barrhill Station, to Newton Stuart, a distance of about 12 miles, a delicate stunted form of Lotus corniculatus was frequently found growing at the roadside among grass, Lotus major being plentiful in damp situations near Newton Stuart. Five specimens of the plant were secured in the course of 10 miles; the first of these, obtained about three miles out, was strongly cyanophoric, the picrate paper being coloured brick red by the evening ; not a trace of hydrogen cyanide was observed in the case of the other four specimens. About two miles from Newton Stuart, where the River Cree comes into view, at the foot of a fairly steep hill, there was a profuse growth of ZL. corniculatus on the top of the retaining wall at the left hand side of the road ; several specimens were tested ; strange to say, none of them appeared to be cyanophoric. Very faint indications of hydrogen cyanide were observed in some but not in all plants collected between Drumlamford and Barrhill Station. On walking from Pinwherry, the next station north of Barrhill, to the coast at Ballantrae, about eight miles, along the road passing through the valley of the Stincher, Zotws major was found to be abundant everywhere in the moist bank at the foot of the hill slope on the right. JZ. corniculatus was also met with here and there at the edge of the road ; specimens collected within the first, within the second and within the next two miles, all gave fair to strong indications of cyanide. At Ballantrae, as in 1911, Z. corniculatus was growing freely in very coarse sand and stones, on the upper level of the beach, in large clumps or compact tufts consisting of long straggling stems bearing small, delicate leaves but large seed-pods, the underground root growth being very strong ; cyanide was present in moderate amount in this plant. A short distance away, across the main road, at the foot of the hill, delicate plants with small seed pods were found growing in the grass beneath a wire fence bordering the rough roadway leading up hill. In 1911, nothing was detected in the plant in this situation but last year distinct indication of the presence of cyanide was obtained— perhaps only because the test applied was a more delicate one. In any case, the very different conditions prevailing on the shore and near at hand on the hillside had favoured the development of very different types of plant. The contrast is equally striking when the plant growington the beach at Ballantrae is 264 Prof. H. E. Armstrong and others. [Jan. 1, compared with that found on the East coast near Dundee and St. Andrews on either side of the Tay estuary. J. corniculatus is abundant in these localities on the outer margin of the sand dunes, where it grows in fine blown sand. The growth is chiefly underground and is much less coarse in character than that at Ballantrae ; usually, the leaves alone, which are very small and delicate, appear above ground. Cyanide was found to be present in this plant, though in very much smaller amount than in the coarse growing Ballantrae form. Taking the observations made zn the field last year and the previous year into account, it appeared to be little short of established that, in addition to the common widely distributed cyanophoric form of JZ. corniculatus, a botanically indistinguishable form exists, different from JZ. major, which, like this latter, is acyanophoric. Considerable quantities of plant collected in the Cree valley were brought to London for the purpose of determining the enzymic activity. Among these were various samples in which cyanide had not been detected ; when these were incubated at 37°, with a few drops of toluene, the presence of cyanide in the plant became obvious within 24 hours, though the amount detected was very small in all cases. On testing Z. major in the same way, no trace of cyanide was detected. It may be added that traces of cyanide have been found in plants grown during 1912 from seed gathered the previous year in Norway from plants in which cyanide could not then be detected. Whilst therefore it appears that Z. major is uniformly acyanophorie and that the common forms of JL. cormiculatus are, more or less strongly cyanophoric, a form of this latter species undoubtedly exists in which the power of producing the cyanophoric glucoside is all but suppressed. Enzymic Activity of L. corniculatus from Various Localities. The determination of the enzymic activity of plants differing in habit and from various localities is obviously of importance in view of the probability that the cyanophoric glucoside and the enzyme are in close correlation. It should be noted that the activity of Z. major towards linamarin is 80 slight as to be negligible (see Part I) and it may almost be assumed that this species is not only acyanophorie but also free from the specific enzyme met with in L. corniculatus. The enzymic activity of all the specimens of L. corniculatus examined during 1911 in which the cyanophoric glucoside was present was high. Of the four specimens from Norway, in which cyanide was not detected, however, one was moderately active, one but slightly active and two active. The results obtained on examining specimens collected last year are as follows :— 1913. ] Herbage Studies. 265 Enzymie Activity of Lotus corniculatus. Percentage hydrolysed. Source of specimen. Linamarin. Salicin. Moscow,* from American clover seed .........020.cs0ee eee eee 84 °2 — nS », West Russian clover seed ...............065 80°5 — Ballantrae (August 27, 1912). .....2 |S = | = Ve pan ote 692 | Reedbuck ...... 8% = = =| = ote ey G2 695 | Sable............ 5 = | = — = = ye 742 | Warthog ...... 9 = | = - _ = hea og 7420 ye 9 - = = = = 28): 743° | Oribi ............ 9 a _ = = 0 283 7434 ail laobood seeane 9 = | = = | = bz py ey 744 | Duiker ......... 7k = = = | = = ent. 755 | Bushbuck ...... 11 — — = | = = ane 768 | Hartebeeste ... 6s = eS = | = = a 29 769 | Reedbuck ...... 5s ~ - | — = ope 780 Fla eens 2 a | = = of 7 Wi 783 Be liatiad aa 4 dp a ar ar July 1.. 777 | Hartebeeste 7 el = = m asia 778 "7 = = = 5 aL 779 rs 7 a + + Re ee SZOMOribivcne ence as | = = ay oust. SO ati te Nee 73 =) os _ - — ees SL Sia Wvemthorre nc: 113 =e = eo Ae 813 | Hartebeeste 8 - | = = = = ” AV 60 814 ” 8 c ai ye ns aR er ec WSIS a 8 = = = | = = Sp be SCRE Neel} ff 8 - | = — = — ie ene f ne 8 Soret = — — My Ole 828) ||| Reedbuck <..... oe eo | ee _ - 5 (3 Goo 825 Hartebeeste ... 8 _ = na Gis 826 | Warthog ...... 42 + | + — - — 6 Covoll B7. |) ORM .cosneccadse 74 | a Tegal = = 5 Sie 844. Hartebeeste ... 3 _ = = = = af ee S635 5 eOribiesceeweere: 1 - — + _ ae ko) 859 | Hartebeeste ... 9? = = = = a 0) 860 Oxibigeeceee ee = + = - — xy. dO) 861 | Warthog ...... = = = = - = ae Ope. SG2meleOxibiercrerees Se oe er Ne is = — = Fae ge Hl bee 866 pb Sidencwoueogs $ = = = = = sep IGS, 869 | Warthog ...... 83 — - — | _ ; = Senet 872 Ei panes 3 — = = | = ile See aa 875 | Hartebeeste = _ - = = , = » 20 912 | Reedbuck ...... ie — _ + _ _ ae 918 | Hartebeeste ... 8 - — _ _ | = ny eal Qe) |) Oral cosoaoobone a _ _ = = = eal 920 | Warthog ...... 5 _ _ = = - eo 920a WS a 5 a 2 = SN eS ah 9205 St Wie 5 = - = Sole peo O21) | Oribie. ee 8 = = = = = op eal 923 | Warthog ...... 10 — — - — - aoe pal 925 | Bushbuck...... 10 - - - - — err ied 927 | Hartebeeste ... 72 } = - — — _ er 929 | Warthog ...... 52 Vena - - - — A ee) 931 a ay le haere ra = = — = — 272 Sir D. Bruce and others. Trypanosomes in (Jan. 18, Table I—continued. | By microscopical By inoculation. Expt , Age of examination. Date. NE, : Animal, blood, in Z O. 7 hours. Nl Thick. | Thin. tea Monkey. | Dog. | | l 1 | 1912, | | | | July 22.. 983 | Warthog ......| 4 + =a a mi bat AOD AN UGESS «| OES ee 4k a bse coal ea hes |) Sunsswlecome 955 | Hyena ......... 8 = = + | = = | 50 28 2c O56 een an eee 8 + = So = = | Bee Ry a 957 | Hartebeeste 5 — — + + + lines, We28:/c(e nObGem fe Oy gz Lo eee zs is = | » 25...) 983 | Reedbuck...... | 6% + + + — - |. )-26 S| SOB all Duke eee I aes ul ae = as aan | ,, 27...) 1000 | Hartebeeste ... 3h + = + | + vit 5 BD 1004 | i ilies Be 2 BS = 3 = | 208.) 1007" | Markert ee nmn fu a ae ee = | ,, 80...) 1010 | Hartebeeste ...| 4 Sie ns = = = 1 gg 80 UI} || WikyrCl oe .snone 4 + + + + + | Aug. 1 WON || OHI srcocncos000 84 = = = = = 5 2 O24 i Salblepeeeesees Ta — — — - —. 5 2 LOZ | Duikerteewerre: 5+ + SS — - 5 4...) 1044 | Bland .........| 63 — — + —- = a 5...| 1045 | Duiker ......... $ = = = ss = || S| 048 as ivyalatcat eres | = sul eae Se a x 7 ...| 1052, | Warthog ...... a - - | = _ — > ce |) SLOSBian aivaldleat mene 23 —-— | - | =] .= = 3) Dera L058! el iicodoo eae | Bs $0 all ae - - | 11..., 1061 | Waterbuck sao} 23 qi tilde ilbetse. | | = et JU O64) | AWiarthoo seers i ele ck ao al = - | WA onal MOBY || TEIGESDA, o...0006 | 4 = | = -— = = » 18...| 1075 | Waterbuck ...| 5 Ai |i — — = » 18...} 1078 | Bushbuck ...... 3 + + + = - se el See Osi Nine eee 3 + = — - - » 18...) 1084 | 30 a0 3 + + + + ot 5 1S cele LOSTAy paar ean oe Naess + + + a = m 1D cca UOMO |) OMI sacsoccnoocc "i _ = — — _ ee Tomes aroo8 Pe Nient ilh .3 Lesh = ay = = {7 29 | L096) | Ga meee 7 = She = - i gg =A 1099 Peston doe 7 — - | = - — | pe Oc eeeeLO2 egal ee 7 = — = = = ay a aeall . alee} Warthog ...... | 53 - — — — — A Diligadl’ UGH) RR Sh IP fe eS — | Rede son 1142 | Hartebeeste 7h = — + — ee | 5 a al MS of 62 as = = = a eno>) |e lGOweReedbuck mem 83 + + + ES a Psy) PPV ecel TUIg Oe 8h a is # ey a . Poa. 24 1156 hn wont ase 7 + = nF = z= 4 PPL ccoll THEE) iN set, Hf = = - — = ye 1162 ci tA 6 + - + — = Wiitgy. 322 1165 EC 6 aa ps = = =| Meg h By osoll TES} Warthog ...... 73 = — = _ — | 1 gy 2Bsoal WG) Wivellel Ga cece 63 = = = - Sek » 28...| 1174 | Waterbuck ... a — - — = = | wy sell aly 5 at a i = = | » anil UI1g0 i nn 5 # — + + ; + | » 24...) 1183 | Warthog ...... 7 = = = = ee BH Elena) MINsYo} cuit Beacon 6 + — + — — | » 24...) 1189 Sahin 6 epee Nr - — = 24) en OD MOribieens tenn ie ape hee = = — | bh) aren, ILI dtc sab aes 74 = # = = | 1913.| Blood of Wild Animals LIaving in Nyasaland. 273 Table I—continued. By tatercseopieal By inoculation. Expt. : Age of examination. Date. No Animal. blood, in Ne ie aaa ; hours. Thick. | Thin. | Goat.| Monkey. | Dog. 1912 . Aug. 24 1198 | Poreupine...... 8} = = = — — oy cash 1202 Eland ......... 4 + + + + Hyehane-te) 1203 | Bushbuck...... 5 + + — - — or Pelee 1205 Eland ......... 4 = _ _ _ » 28...| 1210 | Waterbuck ... 4, pp oe + + + » 980...| 1216 | Bushbuck...... 8 + - - — Sept. 6...) 1250 | Koodoo......... 2 — - — - — Ge alo 5 Aen Oribigaensn eee | 6% = |S = = — ; 7 ...| 1261 | Bushbuck ...... 44 = ofS + — = 56 7 ...| 1264 | Waterbuck ... 33 te eg ee - + + 0 7...| 1268 | Buffalo ......... 64 - | = — — _ . 7 ...| 1272 | Hartebeeste ...| 54 — = lo] _ — 3 7 ...| 1276 Warthog ...... A — —- lac emt — - 3 7...) 1281 || Buffalo %........ 9 _ Seed a — _ » 10...) 1285 ITE emcees 5 ee — = -— | On| 1960), | Blandi) ih), | es = Bate ce ui ark Oi) L293) | Warthor sn. 124 — -— = — - » 10...) 1298 | Buffalo .........| 5 — =e || — _ » 1...) 13804 ) eaoaddunal 34 = aa eee - - Syl 1308 | Warthog ...... 6 | ot - — poten} 1339 Waterbuck ... 63 _ _ + — — «3 US} 1343 | Bushbuck...... 7 — — — | = = re Tals 1347 | Reedbuck...... 5 + en — + + mee le 1351 ne Gore 5S = = a = een kal 13855 | Hartebeeste 74 - - - | - - aati 1359 i vel ee = = a a LA 1863 | Reedbuck ...... 6 = — + — — RIG TSESiuMOriby Ay ween 4 - fo o= — _ = Ha al 1372 Sih) Soonomeneeed 7k = = = — - pe 6 1376 | Elephant ...... 20 — — — = = earls 1380 | Koodoo......... $ _ - + — — Ppaataly/ 1384 | Warthog ...... 64 — a leg — — eS 1388 Waterbuck ... 8 + + + — — es) 1392 | Hartebeeste ... 5 - — - — = vacate 1396 : anes be a 2 ae By LS 1400 | Oribi............ 4% - — — | _ = | 5 20...) 1406 | Waterbuck ... 9 — — + | — = FB ec TO e 9 2s ie ee = 2 » 20...| 1414 | Warthog ...... 64 _ — = a = » 20...) 1418 | Hartebeeste ..: = = _ — — » 20 1422 if 9 a sai e Pa x er) 1426 is nag 4 EGS = bs «6 AD 1435 | Reedbuck ...... 9 — - _ + fh ca) 1439 Be ne 8} ss a ou bas cs 20 1443 Oribinee ees 73 x = af A 2 We espe eeee 1447 | Waterbuck ... 14, = — = = = Zo 1453 | Hartebeeste ... 103 + — — — Et Oct. 6 1471 | Eland ......... PR + — + = = Noy. 10 1577 | Warthog ...... 3h — - = = = | Total 180. Infected with pathogenic trypanosomes 57 = 31°7 per cent. Tn the above table an account is given of the examination of 180 wild animals shot in the fly-area adjoining the Commission’s camp at Kasu. 274 Sir D. Bruce and others. Trypanosomesin ([Jan. 13, This part of the country is situated in the proclaimed Sleeping-Sickness Area of Nyasaland, which extends from the Chirua river (lat. 13° 20’S., long. 34° E.) in the north to the Lintipe river (lat. 13° 50’ S., long. 34° 30’ E.) in the south. It is bounded on the east by the Lake and on the west by the foot-hills. The area is about 50 miles from north to south and 25 from east to west. These figures are only approximate, as the available maps are far from correct. This is the only part of this country in which cases of the human trypanosome disease of Nyasaland, up to the present, have been found. It will be seen, then, that these animals were procured from the very heart of the Sleeping-Sickness Area. Among the 180 animals, 57 were found to harbour pathogenic try- panosomes—31°7 per cent. Table II gives the species of trypanosomes found in the 180 animals. Here a difficulty is encountered—the classification. The tendency in this branch of natural history, as in all others, is to multiply species. In a previous paper* the trypanosome causing human trypanosome disease in Nyasaland was called Trypanosoma rhodesiense, on account of the presence of posterior-nuclear forms. This trypanosome agreed in all other respects with 7'rypanosoma brucei, the common trypanosome of wild animals in South Africa, and the cause of the tsetse-fly disease, or Nagana. In order to compare the two species of trypanosomes more closely, the Com- mission procured, by the kindness of Dr. A. Theiler, C.M.G., Pretoria, a strain of Nagana from the same spot in Zululand where it was first discovered in 1894. Much to the surprise of the Commission it was found that 7. bruces has quite as large a proportion of posterior-nuclear forms as 7. rhodesiense, and that the blunt-ended characteris common to both species. The Commission is therefore driven to the conclusion that 7. rhodesiense is neither more nor less than 7. brucei, and that the human trypanosome disease of Nyasaland is Nagana. To this it may be objected that Nagana has never been known to attack human beings. This has probably been due to faulty diagnosis, cases in man being returned as malaria. The pathogenic trypanosomes then, found in the blood of wild animals in Nyasaland, up to the present, by the Commission are 7. brucei (Plimmer and Bradford) vel rhodesiense (Stephens and Fantham), 7. pecorum, T. sume, and 7. capre (Kleine). 7. ingens is also found, but this trypanosome cannot, with our present knowledge, be considered a pathogenic species to man or domestic animals. In Table II the plus sign means that the trypanosome named at the top of the column was present in the blood. The other plus signs signify that * ‘Roy. Soc. Proc.,’ 1912, B, vol. 85, p. 423. 1913.| Blood of Wild Animals Inning in Nyasaland. 275 Table I1—Species of Trypanosomes found in the Blood of Wild Animals living in the Sleeping-Sickness Area, Nyasaland. Date. |=EzPt:| Aanimat, | Z,2rucetvel| 7. T. | P. | ff. |Thick| Thin } Tnocu- | rant pO: || ; rhodesiense. | pecorum.| simie. | capre. | ingens. | film. | film. | lation. | J ] 1912. | | | | | | Jan. 22 4ieiMland).oes.-. | + [eae E 22 ie + is (age May 19 | 616 | Waterbuck Tah hate | | » 26 591 | Reedbuck .., | + beth June 2] 613 | Buffalo... | + | am ees) 748:| Oribi ..:...... | + sey ol » 80] 783 | Reedbuck ... + ti ae sts ati Fua) | | July 1] 779 | Hartebeeste + es ia | 4, 5 | 828 | Reedbuck ... | cee ce malig lacks S| ' 5, 6 | 826 | Warthog iy ck Soe de seeorn 800 |'Oribi......... + | sty eee | + + |» 20| 912 | Reedbuck ... + | | cot | 5 22 {| 933} Warthog .. + fi + | » 28] 955 | Hyena ...... |} + Niles eA eI nG his gui) ceva jas + | = , 23] 957 | Hartebeeste 4 Re » 23] 958 e + + + | » 25 | 988 | Reedbuck ... ee ae =| + Taek uae » 27 | 1000 | Hartebeeste ee | | + Migtehs | 2 29)|'1007,| Daiker ...... + + 50) {1013 !Wland......... | + |} + a + cea} Aug. 2 | 1027 | Duiker ...... a ae » | 1044 | Eland......... + sf 5 11 | 1058 | Koodoo .. + a it: | » 11 | 1061 | Waterbuck - ewe + vel »» 11 | 1064 | Warthog ... + | | es » 18 | 1078 | Bushbuek ... + | HW + Pees toer + | | + | 5, is fr1084 se Pe Gs eG | | +» 18 | 1087 x + | & ae | aa + eee 9) 1096) | Oriby ©. .2... + | +: » 21 | 1139 | Wartho + + + + » 21 | 1142 | Hartebeeste + | ty » 22 1150 | Reedbuck ... - fat + a ye 22)| 11538 t i; + + + ) £22 be ee + rab os oon eng? i ae + al he » 24 | 1180 | Waterbuck ft + + ic see » 24 1186 | Warthog ... = arnt i oe 24 | 1189 af fees aE K. 28)| 1202 | Winndewh is Lai | es + a) | » 28 | 1203 | Bushbuck ...| + + Tat p28 | 1210 | Waterbuck fe | + | + + » 80 | 1216 | Bushbuck ... + + Sept. 7 | 1261 | 5 Sae| + + 5 7 | 1264 | Waterbuck we ze ae ie » 11 | 1804 | Buffalo ...... fe.) | | » 12 | 1308 | Warthog ... 2 =e, » 13 | 1839 | Waterbuck | + Somer » 18 | 1347 | Reedbuck ... + | + | \e eee 5) a4) 1368 in a hoe | ce alle Sa » 17 | 1380 | Koodoo _...| + | + i » 18 | 1388 | Waterbuck the pe gata Mas aes oe a » 20 | 1406 * hank + » 20 | 1410 rs fc | fee | » 28 | 1435 | Reedbuck ...| + | | Her cd 25 | 1453 | Hartebeeste + ! fe + Oct. 6 | 1471 | Eland......... + Etat + 276 Sir D. Bruce and others. Zrypanosomes in [Jan. 13, the trypanosome was found in a thick or thin film or by inoculation of a quantity of blood from the wild animal into healthyexperimental animals. Table I1I.—Species of Trypanosomes found in the Blood of Wild Animals in the Sleeping-Sickness Area, Nyasaland, and the Number of Times each was found. | Number of T. brucei vel cae ; errant ahodeenenset T. pecorum. T. simie. T. capre. T. ingens. 180 | 14 | 26 | 3 | 20 | 3 In every 100 wild animals living in the Sleeping-Sickness Area, Nyasaland, taken at.random, the following numbers may be expected to be found infected with these species of trypanosomes. Table [V.—Percentage of Animals infected by the different Species of Trypanosomes. T. brucei vel | ih | 4 : Rtas | T.pecorum. T. simie. | T. capre. | T. ingens. | J | (a8 | 14°4 | LY 111 | 17 Table V.—The Species of Animals dealt with, the Total Number examined, the Number found Infected, and the Species of Trypanosomes by which they were Infected. | ; | T. brucei | Animal, UhoeN So, | elo. Kea vel T. pecorum.| T. simie.| T. capre. | T. ingens. examined. | infected. aie sieaise | 27S E. | | | | ‘Blandite 10 6 6 1 Sable ......... 5 | 0 Waterbuck 13 | 9 3 1 | 8 | Koodoo ...... 3 | 2 2 Bushbuck ... 10 "/ 7 1 Hartebeeste 35 6 55 1 Reedbuck ... 19 | 12 3 1 9 1 | Oribie 26 4, | i 1 1 a Duiker ...... i 2 | 1 1 Buffalo ...... 9 2 | 2 | ION dognssoco il (0) | | Hyena ...... 3 2 | | 2 | Elephant ...! 2 (0) | Warthog ... 33 7 | 1 3 | 3 Wild cat ... 3 (0) | Porcupine... ul () | H if | | Total...) 180 59 | 14 | +26 3 20 3 | | 1913.] Blood of Wild Animals Lwwing in Nyasaland. 277 The next table gives the percentages of the different trypanosomes occurring in the wild animals. The numbers are too small to be taken literally, but it is interesting to learn that in this fly-district the waterbuck, hartebeeste, reedbuck and duiker are dangerous neighbours to man; the eland, koodoo, bushbuck and buffalo to cattle, goats and sheep; and that the warthog is the only animal which harbours 7. simie, the lightning destroyer of the domestic pig. Table VI.—Percentages of Different Species of Trypanosomes harboured by Wild Animals in the Fly-area. Animal. ee Paar Nie | T. pecorum, T. simie. | T. capre. | T. ingens. . | percent. | percent. | percent. percent. | per cent. 1D Ehnal gendopannoc 10 10 Gall elmetn irene | 5 Waterbuck ...... 13 23 8 | | 61 | Koodoo)/..:..:5-. 3 66 Bushbucek......... | 10 | 70 10 | Hartebeeste ......, 35 | 14 | 3 Reedbuck......... 19 16 5 47 5 (Oe-sT Tae Aanpasataane 26 4 4, 4 4 UK Ore wes: 7 14 14 (Buttalovres.. cess 9 22 fpmlbton@ hed sto | 1 | | de bye See onocbooce 3 66 Elephant ......... 2 | Warthog ......... fe 3 9 9 Wald’ cats. .: ss.) 3 | Porcupine.........| 1 | CoNCLUSIONS. 1. 31°7 per cent. of the wild game in the fly-country below Kasu Hill harbour pathogenic trypanosomes. 2. The species of trypanosomes found are JZ. brucei vel rhodesiense 78 per cent., 7. pecorum 14:4, 7. simice 1:7, T. capre 11:1, and Z. ingens 1°7. 3. It is self-evident that these wild animals should not be allowed to live in “ fly-country,” where they constitute a standing danger to the native inhabitants and the domestic animals. It would be as reasonable to allow mad dogs to live and be protected by law in our English towns and. villages. Not only should all game laws restricting their destruction in “ fly-country ” be removed, but active measures should be taken for their early and complete blotting out. 4. It must be strictly borne in mind that this only refers to wild animals living in fly-areas. No pathogenic trypanosomes have, up to the present, been found by the Commission in the blood of animals living in fiy-free areas. 278 Trypanosome Diseases of Domestic Animals in Nyasaland. II.—Trypanosoma capre (Klezne). By Surgeon-General Sir Davip Brucz, C.B., F.R.S., A.M.S.; Majors Davip Harvey and A. E. Hamerton, D.S.0., R.A.M.C.; Dr. J. B. Davey, Nyasaland Medical Staff;* and Lady Bruce, R.R.C. (Received January 13,—Read March 6, 1913.) (Scientific Commission of the Royal Society, Nyasaland, 1912.) [Puate 5.] INTRODUCTION. This species belongs to the vivax group, which consists of three species :— Trypanosoma uniforme, T. vivax, and T. capre. They are all characterised by their extreme motility ; clear cell contents; large, round, terminal micro- nucleus; and lastly, by the fact that the wvax group only infects cattle, goats, and sheep, and is harmless to the smaller laboratory animals. All three develop in the proboscis of the tsetse flies and not in the alimentary tract, as do other pathogenic trypanosomes. T. vivax is stated to be pathogenic to horses, mules, and donkeys, but there has been no opportunity of testing these animals at Kasu with 7. capre. It is curious that 7. uniforme and T. vivax have not been met with by the Commission in Nyasaland. This may be due to the absence of Glossina palpalis, which is their carrier, while 7. capre is carried by G. morsitans. MoRPHOLOGY OF T. CAPRA. A. Lnving, Unstained. The description given of 7. vivax can be equally applied to this species. It is just as active in its movements and dashes across the field of the microscope with the same impetuosity. B. Fixed and Stained, The blood films were fixed, stained, and measured as previously described in the ‘ Proceedings.’f * Dr. Davey resigned his membership of the Commission in October, before the completion of the work here recorded. t ‘Roy. Soe. Proc.,’ 1909, B, vol. 81, pp. 16 and 17. Trypanosome Diseases of Domestic Ammals in Nyasaland. 279 Length.—The following table gives the length of this trypanosome as found in the waterbuck, ox, goat, and sheep—500 trypanosomes in all. Table I—Measurements of the Length of Zrypanosoma capre, Nyasaland. In microns. Date me Animal Method of Method of | expt. dixing: staining. | Average’ Maximum) Minimum length. | length. | length. | 1912. | | Sept. 18...) 1888 | Waterbuck...| Osmic acid | Giemsa 26 °8 29 -0 25:0 | eee 4b 849 1 cesses: if 25-9 320 220 | We eal CSOOl a.) Marche aeeesis ‘ A 25°5 30°0 18 ‘0 Mareh 4...) 175 | Goat ......... 3 % 233 28 -0 20:0 | |. | TITER aS |e a ‘ +) 26°1 29-0 21-0 | ve | ERRAN TOXSE SOY Sa cea | i % 27-5 30-0 24:0 ETRE y2OOr 4 e 26 2 30-0 220 ee 20 263 Sy Ohcan nore op Fe 26 ‘9 30 0 22:0 Ss Si Ay a i Sa i ‘ 23-7 26-0 20-0 Aymeede 263 A un pesteebeices _ 3 24° 4 29 :0 21-0 April 4 339 Aon Mheahoe eee Bi ii 251 28-0 23 -0 MOO 279.10 ibs vgs oe | A; i 23-4 27-0 21-0 Moe) 989 |a ee, ¥: 8 246 27-0 21-0 » 29... 339 wafer spiccaeaes % FA 27 °6 31 °0 25:0 WNL | GIES |e ee cya “A if 25 °5 29-0 23-0 Bato sso in | koe ‘ a 24° 27-0 23-0 April 8 348 | Sheep ......... 3 Ps 26 6 30 °0 24-0 PIP RAG cay, Sas fi < 24, °6 28-0 21-0 _ycaaie OuNel RERY Sal Mink mn Hea " sf 26 °7 29-0 24-0 Mees Genk GAB I int, a vos ee rss a a‘: 23 °9 27-0 22-0 My gel S48 hae fi a‘ 26-2 29 -0 24-0 Nise sae yh. oe ee if fi 24-4 27-0 22 0 5 SI ASG a a et Sea 3 2 24-7 27-0 22-0 | June 18...) 543 pris eourerconees ¥ : 25 6 29:0 23 0 Peuly 25. OGG ik Oe a 4 27 °5 32-0 24.0 j | 25 °5 32 °0 18 °0 | i The average length of 7. caprew, Nyasaland, in different species of animals, taken from Table I, is as follows :— Table II. In microns. Number of Species of animal. trypanosomes measured, Maximum | Minimum Average length. length. length. Waterbuck ......... 20 268 29-0 25-0 Oma ere . 40 PAS) F/ 32-0 28-0 Goatees ccuescs.) 260 25°83 31-0 | 20-0 INGEN eccossdnocesaoy 180 25 °6 32-0 21 °0 280 Sir D. Bruce and others. Trypanosome [Jan. 13, Table IIJ.—Distribution in respect to Length of 500 Individuals of T: capraw, Nyasaland. In microns. Average length. Animal. bo S2 bo a) 18.| 19.| 20.| 21.| 22.| 23.) 24. | 25. | 26. bo bo (2) er) @ 30. a 32. ] (Os dgnuaneasodageoses oa —|—}]— | | | WAORH Bh | | bo ies) SAG AKH EATS UHH dAoE | | | | bo bo bo RSSfonyrns ADE | | | Le tel el eel | | el wrote! | | [ ere fe lee | ote wwe ell bm OL CO Od Ol & GO 6 bb | STO OsN|| wHeeimes | ey | CICS CHOPS NERS TOS ININGICIIN ES | a. ae HS C169 IS Gueihe Gus Gu | NrFPWNWNHeE we o bo ee (ee | otal ee eeeeeeeee ce 3| 8 | 23| 49] 79 95 | 80 | 68 | 57 | 24 9| 2) 2 l | | | | | \ Percentages AC — |0-6\1°6 4°6 9-8) 15°8|19-0 fee ant 4°8/1'8|0-4|0-4 Cuarr giving Curve representing the Distribution by Percentages in respect to Length of 500 Individuals of Zrypanosoma capre, Nyasaland. Mucrons & 18 {19} 20) 2) |22|23 | 24) 25)26) 27) 23/29 |30/3! |32 S) Percentages —nNakaADAAD © 1913.|| Diseases of Domestic Animals in Nyasaland. 281 This curve is made up of measurements from 20 specimens of trypano- somes taken from the waterbuck, 40 from the ox, 260 from the goat, and 180 from the sheep. From it will be seen that 7. capre is a monomorphic species, varying from 18 to 32 microns in length, the greatest number of individuals (19 per cent.) being 25 microns long. Breadth.—Measured across the broadest part 7. capre averages 3 microns in breadth (maximum 4°25, minimum 1:75). Shape-—T. capre differs from 7. vivax in that it is heavier built and altogether has a larger and clumsier appearance. The posterior half is swollen, and its end is bluntly angular or rounded. The anterior extremity is narrower and pointed (Plate 5). Contents of Cell.—Clear, with a delicate alveolar structure, and free from vacuoles or granules. Nucleus.—Oval, compact, lying about the middle of the body. Mieronucleus.—Large and round, situated, as a rule, close to the posterior extremity, but sometimes removed to a short distance. Undulating membrane-—Much more developed than in 7. vivax, and thrown into bolder folds and undulations. Flagellum.—tThere is a well-marked free flagellum which averages 6°5 microns in leneth (maximum 9°5, minimum 4). No specimens have been seen without a free flagellum as stated by Kleine. Disease set up in Catile by T. capre.—Only two oxen were inoculated from goats suffering from this disease. These animals showed the trypanosomes in their blood in small numbers for two months after imoculation. The trypanosomes then disappeared and have never reappeared. The two oxen at the present time are in good health and have evidently recovered. This strain of 7. capre cannot, therefore, be considered of much pathological importance as far as oxen are concerned, but more cases are wanted. Kleine states that cattle are immune. Disease set up in Goats and Sheep by T. capre—tIn goats and sheep, on the other hand, 7. capre runs a fairly fatal course. In the list of animal experiments 36 goats and 4 sheep are given. Of the 36 goats, 15 had been infected by wild G. morsitans and died, on an average of, from 53 to 59 days. As the flies were fed on a goat, a monkey, and a dog, and, as a rule, three times on each animal, to ensure that all the flies fed, it is not possible to tell the exact day of infection. Four others were inoculated with blood from infected goats or antelope, and these died, on an average, in 57 days. The remaining 17 goats are still alive after intervals of from 61 to 262 days. Of the four sheep, one died in 282 Sir D. Bruce and others. Trypanosome [Jan. 13, 36 days, another lived 89 days, the third 221 days, while the fourth is still alive after 245 days. Table IV.—Animals Susceptible to 7. capre. | No. of Period of Date. | | Source of virus. incubation, oie BER Remarks. expt. ree in days. | “a Cattle. 1912. | | | Mar. 20...| 349 | From Goat 268 ......... | 20 | —_— Alive and well after 245 days. » 20...| 350 5 263 ......... 10 | — These animals appear to | | | | have recovered. Goat Heb (Gie.|)) 2¢3) || Wall ititesiaie ase memes ee | 0-12 |; 41-53 Died of 7. capre. at e200 kal eee ate ee 20-26 128-134 ¥ 5p Ope mesy te MER Mis Bas 2-17 51-66 is K Mar. 3... 272 sill cae eRe en pei | 21-29 — Alive after 262 days. » 11...| 263 | Natural infection ...... ? P Died March 22. 3 LS SBS Wald ilies se eeeers. tee ae — Alive after 251 days. » 16...| 3835 | From Goat 268 ......... 5 — % 249, SOG) ERD | ts DESI Ane. | 9 63 Died of T. capre. Bl BO eS DOR: eke 16 _ Alive after 249 days. » 20... 346 ” 208) ieee ke | 5 it ” 245 ” DOr gana, a DBS hy ie 12 _ > 245, =p Dron 348 a PAB) dabosos-n 5 89 Died of ZT. capre. ANW, — Thsnc ie} | ARYAN TW) 5 nonoaonodesoe 10-21 101-112 7 5 » 11 410 99 te ee tee ee tee 6-11 46-51 ” » i by 412 bahia Raniety occa 0-13 0-20 ; , 16ah Aa Ee | ised Reid 7-13 33-39 s "3 Sein ine 422 EME BA Uno ae Cont 4-10 — Alive after 215 days. » 24...) 415 PNM AANA oA otidoCoc 13-19 46-52 Died of 7. capre. ” 24 420 Moe Soames sdomsnaze 13-19 43-49 | ” ” May 2 433 Pl “kdbaoonsecHconse 10 63 ” ” Shae 435 work fcurpbsdeee antes 18-19 55-61 3 5 Gee) 266 Ph ime ahen asecndonscee 6-11 69-74 3 » 9 9 269 Ajo we sdohcopodtoocdace 10-18 18-26 ” 2 Shale 853 pe SC 6-16 — Alive after 186 days. Semone 565 gi Tero eee eae 3-10 52-59 Died of 7. capre. June Boon 622 Ho | andedesoasan0n0 15 34, ” ” July 25...) 979 | From Reedbuck 988 ... 11 = Alive after 118 days. Aug. 18...| 1039 | From Bushbuck 1087 8 33 Died of 7. capre. » 22...) 1111 | From Reedbuck 1153 iil 43 » » » 22...) 1114 Be 1156 11 — Alive after 90 days. » 22...) 1118 5 1150 11 — 0 90 ,, 225) 91120 1162 7 = » 30 ;, Sept. 13...| 1842 | From Waterbuck 1339 20 — » 68 ,, » 14...) 1866 | From Reedbuck 1363 9 _ 7 67 ,, » 18...) 1391 | From Waterbuck 1388 8 — » 63, » 20...| 1409 ss 1406 17 — 3 61 ,, Sheep July 17... 907 | From Goat 653 ......... 5 36 Died of 7. capre. Mar. 20... 346 * PASS. coovossae 5 — Alive after 245 days. reared. sad 347 i PASS scodeo ono 12 221 Died of 7. capre. » 20..., 348 5 Boars | 5 89 » y 1913. | Diseases of Domestic Animals in Nyasaland. 283 Table 1V—continued. 5 | No.of Period of | ; : é Duration, : Date. expt. Source of virus. nee in days, Remarks. Monkey. 1912. Feb. 2... 2NS) © || AWAITS) | esonoqpanpeanee — — Only showed 7. simie. Mar. 9... 326 | From Goat 175 ......... ost — Never showed trypanosomes. Anruelons|) 405) || Wald flies)... s..0e-n.. aa — Only showed 7. simic. meron) 465 EAD ai ee Ai Nai = c a > 28...| 467 doeh Lhd dansodensenere|| — — Only showed 7’. pecorum. July 25...| 989 | From Reedbuck 988 ...| = = Never showed trypanosomes. Aug. 22...) 1154 PA 1158...| —_— — a . PPL alnGye a 1156... = ss . i Sept. 18...| 13840 | From Waterbuck 1339 — — S ra » 14...) 1864 | From Reedbuck 1363... — = 9 ) » 18...) 13889 | From Waterbuck 1388 | -- — np . » 20...) 1407 My 1406 a — 33 3 Dog. Mar. 9...) 319 | From Goat 175 ......... — — Never showed trypanosomes. on ne) 320 s IL/ES Re ode re _— -— | Only showed 7, brucei. so at) 321 % IBY aon conene = = Only showed 7. pecorum. ‘et 322 at aT eek ait a | a . Cy 344, 5 PASS) cravooson —_— _ Never showed trypanosomes. 4 Odes ck Deane = _ 4 a July 25... 990 | From Reedbuck 988... — — a a Aug. 22... 1155 i 1153... aus = as A pe ooh 1 1s8 if 1156... ee = is os Sept. 13...) 13841 | From Waterbuck 1339 — — ; i » 14...) 1865 | From Reedbuck 1368... — _- = i » 18...) 1890 | From Waterbuck 13888 — — 3 5 Guinea-pig. Mar. 20...| 351 | From Goat 263 ......... = = | Never showed trypanosomes. 45° ADhocll = BER | 4 DESecan | — ea || BE Bs Rat Mar. 20...| 351 | From Goat 268 ......... | _— = | Never showed trypanosomes. F 2. 352 | _ OSE) “aed | —- | = | eh) ” i THE CARRIER OF T. CAPRA. The carrier of 7. caprw in Nyasaland is G. morsitans. These tsetse flies in the neighbourhood of Kasu are heavily infected with this trypanosome. In the experiments made to ascertain with what trypanosomes the wild flies are naturally infected, 7. caprw was found in 61 per cent. The development of this trypanosome in G. morsitans will be dealt with in a future paper; suffice it to say here that it is restricted to the proboscis and runs a course of from 16 to 20 days. VOL. LXXXVI.—B. x 284 Trypanosome Diseases of Domestic Animals in Nyasaland. THE Host orn RESERVOIR OF T. CAPRA. Up to the present 180 specimens of wild game living in the Nyasaland Sleeping-Sickness Area have been examined. Of these 19, or 10°5 per cent., harboured 7. capre. The animals were reedbuck, waterbuck, eland, and bushbuck. CONCLUSIONS. 1. 7. capre belongs to the same group as 7. vivax and T. uniforme, and affects the same animals—cattle, goats, and sheep. Monkeys, dogs, and the smaller laboratory animals are immune. 2. The carrier is G. morsitans. | 3, The reservoir of the virus is the wild game living in the “ fly-country.” DESCRIPTION OF PLATE. Trypanosoma capre (Kleine).—Large, heavily built body ; posterior extremity swollen, bluntly angular, or rounded ; anterior extremity pointed ; nucleus oval, compact ; micronucleus large, round, situated, as a rule, close to posterior extremity ; undulating membrane marked, thrown into bold folds; flagellum well marked, free, average ‘§°5 microns in length. x 2000. Bruce & others. Roy. Soc. Proc. B vol. 86 PLS. Ox. ME. Bruce, det. Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland. 1.—The Human Strain. By Surgeon-General Sir Davin Brucg, C.B., F.RS., A.M.S.; Majors Davip Harvey and A. E. Hamerton, D.S.O., R.A.M.C.; and Lady Brucg, R.R.C. (Received February 8,—Read March 6, 1913.) (Scientific Commission of the Royal Society, Nyasaland, 1912.) Introduction. In order to gain a general idea of this important species of trypanosome, it will be necessary to study as many individual strains as possible. It may be thought unnecessary to describe each strain so much in detail, but without this it will be impossible to get any order out of the chaos which rules at present in the classification of the African species of trypanosomes pathogenic to man and the domestic animals. Up to the present the Commission have only had an opportunity of work- ing with five human strains. Four of these are from natives infected in the Sleeping-Sickness Area, Nyasaland, the fifth from an European who contracted the disease in Portuguese East Africa. It is intended, in later papers, to describe five strains from wild game and the same number from the tsetse fly, Glossina morsitans. The human strains are named : I, Mkanyanga; IIT, E——; III, Chituluka; IV, Chipochola; and V, Chibibi. I. Morphology of Strain I, Mkanyanga. This has already been dealt with in a previous paper.* Il. Morphology of Strain II, E——. The following table gives the average length of this trypanosome as found in goats, sheep, monkeys, dogs and rats, 1500 trypanosomes in all, and also the length of the longest and shortest :— * * Roy. Soc. Proc.,’ 1912, B, vol. 85, p. 423. 286 Sir D. Bruce and others. Trypanosome [Feb. 8, Table I.—Measurements of the Length of the Trypanosome of Strain IT, E In microns. Date. Method of fixing. Methed of | g- Average Maximum | Minimum length. length. length. 1912 Osmie acid Giemsa 22-2 360 | 15:0 | | The average length of the trypanosome of Strain II, in different species of animals, is as follows :— Table ITI. | In microns. | Number of | Species of animal. | trypanosomes | measured. Average Maximum Minimum | length. | length. length. Q ° & D (=) ho (=) ~ ww ns j=) Shee paeessesseeeeeee 20 21°3 | 28:0 IMionkeyarenecsereerce 160 » 22-9 | 36-0 Dogtra. eee 260 21°8 | 31-0 Rat. eee 1000 23-1 | 32-0 Table III.—Distribution in respect to Length of 1500 Individuals of the Trypanosome of Strain I], E——. In microns. 15. | te | 17 | 18. 1 ouneeo: | 20. |siooN le tces| Owens | | Total ......... 2 2 | 12 | 55 | 108 | 159 | 210 | 188 | 215 | 177 | 138 | | Percentage...| 0-2] 0-2] 0-9 a) 72/106/14:0/126|/144|/118| 9-2 In microns. 26. | 27e| .28. |295 \yus0. |, Ble, ||) 325 saeelieo4 wales peae: a i} Motaluenet g3 | 6o | 34 | 26 | 18 8 Baines | rigs (Gees 1 Percentage...) 5°6| 4:0] 2:3] 1°38} 1:2} 06) 0:2) — | © | = O'l 1913. ] causing Disease in Man in Nyasaland. 287 Cuarr 1.—Curve representing the Distribution, by Percentages, in respect to Length, of 1500 Individuals of the Trypanosome of Strain II, E——. MinCiO nes 113 [14] is]ie]17 [18 | 19] 20] 21 |22 25 26], eS aE oO @ P=) = Co) 3) L 7) a This curve is made up of measurements from 60 specimens of trypanosomes taken from the goat, 20 from the sheep, 160 from the monkey, 260 from the dog, and 1000 from the rat. In a previous paper it was suggested that 1000 trypanosomes taken at random would be a suitable number to plot a curve from, for purposes of comparison. This is done in Chart 2. The taking away of 500 rat trypanosomes has changed, to a great extent, the character of the curve. There is no resemblance between this curve and that given on Chart 1 of Strain 1, Mkanyanga. If the two strains, I and ITI, belong to the same species, then little help can be expected from this system of measurement in classifying trypanosomes. It has been suggested by Dr. J. W. W. Stephens that the measurements should be made from one animal, and he proposed the tame rat as a suitable species. There seems much to be said in favour of this. Practically, his proposal is that a series of slides should be made with blood taken on 10 consecutive days from a single rat, and that 100 trypanosomes should be drawn each day. But it is no light task to draw 1000 trypanosomes at a magnification of 2000, and afterwards to measure them. We have therefore made a compromise and measure 60 trypanosomes on nine con- secutive days, beginning from the day the parasites first appear in the blood. In order to deal with a round number (500) only 20 are measured on the ninth day. 288 Sir D. Bruce and others. Trypanosome [Feb. 8, Cuart 2,—Curve representing the Distribution, by Percentages, in respect to Length, of 1000 Individuals of the Trypanosome of Strain IT, E : es im ac Percent - »’® BF OM aw @ WO oO oo) o = No Mitkeiaolnnns 13 [14 16 17 {18 FS) 23) 24)25 26 27 | 28) 29|30 31 }32|33|34| 35] 36| 37/38 Cuarr 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain II, E days from Rat 728. , taken on nine consecutive Percen =—rn&e kf 4D AN & © tages eo = ® — 35|36|37 o 1913. | causing Disease in Man in Nyasaland. 289 This makes a symmetrical curve, which ascends and descends by fairly regular steps, but with little likeness to Charts 1 and 2. In an organism low in the scale of nature, such as this, subject to great variation in form, it might be thought that it would not be likely to behave in any two rats in the same way. The following chart shows that this is not so, but that, on the contrary, the same strain of trypanosome planted in two different animals of the same species grows in a remarkably similar way. Carr 4.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain IJ, E——, taken on nine consecutive days from Rat 726. T T T 1 Si |32}33 134] 3 | | | | | 19 ——a—t rea IE + ‘a ; ] | | | | } e aaa eel | Bane: | i I7 lasthocaia | | —- ahah 16 + ‘ Hae | | | | in| i 15 — —. | =p Wo WY 14 ee % 13 = ee ee Led ) Ve ' ++} 1 jt of Fale ace Ji a -—— - 2» ive all lem Plo —- — Spm ey Ma GN ala Soa < o ° a iecail ch asa Te LST Fe os | ! pas oe es is leit | o 7 T | | rf | | | | a 6é Ss Ss | i | | | 5 = i == + = - al Fis via sip fe aires ian ia 2 ab j | aa | ea a) | It is remarkable how much alike these last two curves are. If curves made in this way from different strains of one species of trypanosome showed the same degree of similarity, this method would certainly be useful for purposes of classification. But, as we have seen, the curve of Strain II has no resemblance to that of Strain I, and it will be found that each human strain of this species of trypanosome differs, more or less, when subjected to this method of measurement. As the occurrence of posterior-nuclear forms has been made the distinguishing character between Trypanosoma brucei, gambiense, and rhodesiense, it will be of interest to note the percentage of these forms in the various strains. The method used is to count the number of posterior ‘ 290 Sir D. Bruce and others. Jrypanosome [Feb. 8, nuclears in 1000 short and stumpy forms in 10 specimens of a single rat’s blood taken, as near as possible, on 10 consecutive days. Table 1V.—Percentage of Posterior-Nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain II, E : Experiment F Percentage among short Dae. ENG. SELullh and eet oes Forni: 1912 | Uibine PB Sa5ec0ceu 728 Rat | 10 BS 26 seueen 728 3 ily spn UT A eee 728 es 3 5B ycun 2 aeeeee 728 a 9 duly» ab osanceco 728 5 5 5 Ditlesneen 728 5 5 gfe iteratey een 728 ai 9 Syl, eatoreaeeeeee 728 3 Ppeiciy ana natios 728 a 18 3 ORL eee 728 5 14 IMMOBRD sero50080500 9°3 In regard to breadth, shape, contents of cell, nucleus, micronucleus, undulating membrane and flagellum, it is not proposed to describe these characters separately for each strain, as was done in Strain I. Suffice it to say that no difference can be made out in regard to these particulars on comparing the five strains. The same posterior-nuclear and blunt-ended forms are present in all. III. Morphology of Strain ILI, Chituluka. The following table gives the average length of this trypanosome as found in the goat, monkey, dog and rat, 1500 trypanosomes in all, and also the length of the longest and shortest :— Table V.—Measurements of the Length of the Trypanosome of Strain III, Chituluka. | In microns. Date. Method of fixing. Me bla od ee Sete Average Maximum ;, Minimum length. length. | length. 1912 | Osmic acid Giemsa | 26-1 | 38 -0 | 15 ‘0 The average length of the trypanosome of Strain III, in different species of animals, is as follows :— 1913. ] causing Disease in Man in Nyasaland. 291 Table VI. | In microns. Number of it 4 anevestiews! Species of animal. trypanosomes measured. Average Maximum Minimum length. length. length. | Goats tov. cescee 80 26 °9 32-0 16°0 Monkeys. Jct es: 160 27°7 36 °0 16°0 OP seit eBesk 260 241 35-0 16-0 15°0 Rati cesses sch: -6s | 1000 | 26 °4 38 0 Table VII.—Distribution in respect to Length of 1500 Individuals of the Trypanosome of Strain III, Chituluka. In microns. | pe a ee | 15. | 16. | 17. | 18. | 19. | 20. | 21. | 22. | | | Lil ae as 1) 8 | a) a1 78| 71| 44 eal | | | | | | Percentages .........| 0-1 La) lua aa 5-2 | 4°8 | 3°0 | 3:1 | 3°8 | 3°6 | 66 | 8:0 In microns. 33. 34. 35. | 36. 29. | 30. | 31 | 32. | 128 |188| 99|117| 91] 63| 27| 11) Byes Percentages ......... 7:4 | | | ] | | | 8°6| 9-2/1 66| 7:8 | 6-2| 42) Ea Ore oisaL On 0-1 Cuart 5.—Curve representing the Distribution, by Percentages, in respect to Length, of 1500 Individuals of the Trypanosome of Strain ITI, Chituluka. ” () ce.) B » = Oo oO — 1) a. 292 Sir D. Bruce and others. Trypanosome [Feb. 8, This curve is made up of measurements from 80 specimens of trypano- somes taken from the goat, 160 from the monkey, 260 from the dog, and 1000 from the rat. It resembles that of Strain I, and differs absolutely from Strain IT. As in the case of Strain II, E individuals of this strain. , @ curve is also given of 1000 Cuarr 6.—Curve representing the Distribution, by Percentages, in respect to Length, of 1000 Individuals of the Trypanosome of Strain III, Chituluka. Microns 1A] 15]16117118]19]20]21 |2zl23|2al25|26 30|31|32|33] 34] 45]36|37 [2e| 7 ae This curve, made up of 1000 individuals, is very similar to the previous one of 1500. It is made up of 80 specimens of trypanosomes taken from the goat, 160 from the monkey, 260 from the dog, and 500 from the rat. The two following curves represent measurements of 500 trypanosomes taken from each of two rats. Cuart 7.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain III, Chituluka, taken on nine consecu- tive days from Rat 952. |_| a ip} 2 ae 3 eel = 7 ee I] ene eAe fal sf g : Eide fo Paaeteraa 3 = 3 Pb fseaC eee Gee oo ONG 2 (Hen? if Sic RE Coase HEARERS Rae eels 1913. | causing Disease in Man in Nyasaland. 293 ' Cuarr 8.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain III, Chituluka, taken on nine consecu- tive days from Rat 953. Ae ee ae Ai el ae ATES 51S a a a JSS SSS Tee ee eee Jae eee a a oe ee ee (Ss a Od eRe eie emit TT Peer SR An | | ss) ss Le AS ASR age 87 A a 2a a a — lf Cats Sg yl a Canes 7) S) Of 3 » s ® o ~ ov a ies LE FE aaa Kiel Ei il [elt OG EL Jie See See ee These last two curves from different rats also closely resemble each other. It is curious and striking that the same strain of trypanosome growing in two different animals should show this remarkable similarity. Table VIII.—Percentage of Posterior-Nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain III, Chituluka. | Experiment . Percentage among short | Pate, | No. aan | and stumpy forms. | 1912. | August z btniee 953 | Rat 4 Aca 958 | 5s 6 is a cece 953 i 3 5 Uharebie 953 a 8 eC 953 | if 6 | tenis i 953 | i 13 Syn O. weeks 953 . 382 \ = 2 AVETALC) Soc sseeesieee 10°3 IV. Morphology of Strain IV, Chipochola. The following table gives the average length of this trypanosome as found in goats, monkeys, dogs and rats, 1000 trypanosomes in all, and also the length of the longest and shortest :— 294 Sir D. Bruce and others. Zrypanosome [Feb. 8, Table [X.—Measurements of the Length of the Trypanosome of Strain IV, Chipochola. = os In microns. Date. Method of fixing. Moet of | Sana ne Average Maximum | Minimum | length. length. length. 1912 Osmic acid Giemsa 22:5 | 34:0 | 15 ‘0 The average length of the trypanosome of Strain LV, in different species of animals, is as follows :— Table X. In microns. Number of Species of animal. trypanosomes measured. Average Maximum Minimum length. length. length. Goat’ incdeemesece 80 20-4 29-0 150 Monikeygaeereeeeaees 160 22-0 34°0 16 °0 1 D6) tee onsbenoeasddiee 260 20°9 31-0 15 °0 Raitiesncetceseemes 500 22°5 34°0 15°0 Table XI.—Distribution in respect to Length of 1000 Individuals of the Trypanosome of Strain IV, Chipochola. In microns. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24, ANI ae ancanciasasnasaes 2 4 32 CS LOM SLOT POS SLOG 95 95 Percentages ......... O25 OLA Se28 GES EON TOs | 101-99) StOLGs Oommen | In microns. 25. 26. 27. 28. 29. 30. 31. 32. | 33. 34. Total se eee sswesnes 74 64 50 38 26 16 5 3 i 1 Percentages ......... C2 CAN BO BS) B45) Ue) OF | OB OU) O2 1913.] causing Disease in Man in Nyasaland. 295 Cuart 9.—Curve representing the Distribution, by Percentages, in respect to Length, of 1000 Individuals of the Trypanosome of Strain II, Chipochola. Mire roan 's TEC oes ee See senor DLL coa Santee 5 na Se le ize NSE | 2 ESY See NG ea aa / y 4. Sage eaecese oc So eee ee Se See Ei This curve is made up of 80 specimens of trypanosomes taken from the goat, 160 from the monkey, 260 from the dog, and 500 from the rat. Percentages an @ ei LE Pa a | “rm es cI aft : EEEEEEEEEEE Cuart 10.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain IV, Chipochola, taken on nine consecu- tive days from Rat 1337. Microns g) 8 7 & S) 4 3 fe 1 296 Sir D. Bruce and others. Trypanosome [Feb. 8, Table XII.—Percentage of Posterior-Nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain IV, Chipochola. | Dees, EY eee Percentage among short 0. and stumpy forms. 1912. Sept. 20 ......... 1337 Rat 1 Bega psa 1337 = 1 MH ail unoon0c 1337 3 2 Ayer) sodauconc 1337 » | 4 FF 40) eon acdioce 1337 > | 0 Ailes) cogoaecon 1337 " (0) py 249) soaucao00 1337 9 1 330 BOM. wnaeaae 1337 55 14 Oct: GOs 1337 i 5 ie ucoosnue 1337 5 AV CLAS voce scerenne 3°3 V. Morphology of Strain V, Chibibi. The following table gives the average length of this trypanosome as found in goats, monkeys, dogs and rats, 1000 in all, and also the length of the longest and shortest :— Table XILI—Measurements of the Length of the Trypanosome of Strain V, Chibibi. In microns. Date. Method of fixing. eee oF g: Average Maximum | Minimum length. length. length. | 1912 Osmie acid Giemsa 22 °4 | 37 ‘0 | 15°0 The average length of the trypanosome of Strain V, in different species of animals, is as follows :— Table XIV. In microns. Number of Species of animal. trypanosomes measured. Average Maximum Minimum length. length. length. Groathe enseeaeceessees 80 19-9 31°0 16:0 Monkey ...........- 160 21°8 320 15:0 Do geese iene sene 260 20-6 370 16:0 IRENE ganacaoasocdacoo0n 500 24:0 320 18-0 1913.] causing Disease in Man in Nyasaland. 297 Table X V.—Distribution in respect to Length of 1000 Individuals of the Trypanosome of Strain V, Chibibi. | In microns. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24, 25. | | | | Motallenenucs. ss 1 8 20 58 117 | 122 | 123 | 107 93 93 | 638 | | Percentages 0'1 0°8 2°0 6°8 | 11°7 | 12-2 | 12°3 | 10:7 | 9:3 | 9°3 6°3 | In microns. ; | ] 26. 27. 28. 29. 30. 31. 32, | 338. | 34.| 35.| 86. | 37. | 2 = ‘ = * Motalienwecars: 51 41 43 30 16 10 83 5-—- |— || 1 Percentages | 5°1 | 4:1] 4°3/) 3:0} 176] 1:0} 0°3 | — | =| — |) = Oar CuaArtT 11.—Curve representing the Distribution, by Percentages, in respect to Length, of 1000 Individuals of the Trypanosome of Strain V, Chibibi. Fa ise [iiss eo] Mucyons 38 ales [ea] oz a7 [es 2a[eo [a [ae|as|salss ss SEG EE a ee ce | ot \ : eee ea ax | ox Queer) This curve is made up of 80 specimens of trypanosomes taken from the goat, 160 from the monkey, 260 from the dog, and 500 from the rat. 298 Sir D. Bruce and others. Trypanosome [Feb. 8, Cuarr 12.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain V, Chibibi, taken on nine consecutive days from Rat 1660. Microns 13 14}15 |16 17 118 |19 }20| 21 |221235|24|25|26| 27) 26|29)350/3) |52)/33|54|55|56|37|58 —e§e MO wWOHKRODND OO Table X VI.—Percentage of Posterior-Nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain V, Chibibi. | | | Dater Papen Sane | Percentage among short 0. | and stumpy forms. | | | | 1912. | IDYt85 863) Goonccase | 1660 Rat | 0 Pon Woogpasudo 1660 5 | 34 | Br Oise bap addcad 1660 % | 2 | 539 1h eens 1660 . | 6 | pial an nee | 1660 is | 8 | pmell Oleeneeeree 1660 49 23 py dil Ssasosene 1660 os 31 oy. I) pbbaccose 1660 ents 28 ng) LB} gan aanoe 1660 6 27 Fe Wet carenooees 1660 5 32 | AMBER coronhacdaas 21:1 Comparison of the Human Strains. The following table gives the average length of this trypanosome in the five human strains under consideration, as found in goats, sheep, monkeys, dogs and rats, 6200 trypanosomes in all, and also the length of the longest and shortest :— 1913.] causing Disease in Man in Nyasaland. 299 Table XVII.—Measurements of the Length of the Trypanosomes of the five Human Strains. The trypanosomes have been taken from various animals. Nuanbor In microns. | Date. | Strain. Name. t om Animals. | an | a Average | Maximum| Minimum | Sea length. length. length. AJ ab es al if ical 1912 I Mkanyanga 1220 Various 24:1 36 ‘0 14:0 | 1912 II E——_ 1500 - 22 °2 360 150 | 1912 | III Chituluka 1500 is 26 ‘1 38 0 15 ‘0 1912 IV Chipochola 1000 $0 22 °d 340 ON 1912 Vv Chibibi 1000 a 22 °4, 37°0 5 0) | ; | | 6220 235 38 ‘0 140 | It must be acknowledged that, in spite of the fairly large number of trypanosomes measured, there is a marked difference in the average length of the five human strains—from 22°2 to 26°1 microns. are similar, varying only from 22°2 to 22:5. Strains II, IV and V This difference in average length may be due to slight variations having taken place in the different strains, resulting from the passage through more or less resistant man. treatment by atoxyl or other drugs. There is no evidence that this variation is due to It has been shown that the same strain grown in two animals of the same species gives like results. Table XVIII.—Measurements of the Length of the Trypanosomes of the five Human Strains. The trypanosomes have been taken from the rat alone. Number In microns. Date. | Strain. Name. t of J Animal. a se, Average | Maximum | Minimum ; length. length. length. 1912 I Mkanyanga 600 Rat 25 -1 35 ‘0 16°0 1912 II E 1000 5 23 *1 32 :°0 17:0 1912 IIt Chituluka 1000 A 26 °4 38 °0 15:0 1912 IV Chipochola 500 i 22°5 34-0 15-0 1912 W Chibibi 500 5 24°0 32 °0 18:0 | 3600 24 °2 | 38 0 15:0 VOL. LXXXVI.—B. Y 300 Sir D. Bruce and others. Trypanosome [Feb. 8, Comparison of the Curves from the Five Human Strains. It must also be confessed that, on comparing the five curves one with another, they do not give as much assistance in classifying this species of trypanosome as was hoped. Curves I and III are alike, and coincide with that prepared by Dr. Stephens from the case of Armstrong in Liverpool, whereas Curves II, IV and V approach more to the type described by Kinghorn and Yorke from the Luangwa Valley. Table XIX.—Distribution in respect to Length of 6220 Individuals of the five Human Strains. The trypanosomes have been taken at random from various animals. In microns. 14. 15. 16. fe 18. USS |) 2205 al) PA, |) Bh || Bel / 25. ] | | | Strains I-V | 1 | 10 | 41 | 154 | 325 | 494 | 528 | 577 | 512 | 525 | = 464 Percentages | — O);2))| OTe) 92E5 53h) 18:0) 8-4 Ors Seshiesed 26. | 27. | 28. 29. | 30. | 31. 8 "83, 34. | 35. | 36. | 37. Strains I-V | 425 347 | 807 | 198 | 167 | | | l | 372 Percentages 68160\56/49/34 27/20/10 0-6 0:2) 0:2 | — |- | | | Cuart 13.—Curve representing the Distribution, by Percentages, in respect to Length, of 6220 Individuals of the Human Strain of the Trypanosome causing Disease in Man in Nyasaland. The trypanosomes have been taken from various animals. Microns 13} 14] 15] 16]17418]19 }20|21 | 22| 23} 24) 25}26|27| 28 |29 | 30} 3) |32}|33 | 34} 35 | 36) 37 |38 a} 19 n 9 iBbi vo Of 8 a i) | el » £6 ae (9) wy & ———+- + | - o 4 a 3 2 t r | 1913. ] causing Disease in Man in Nyasaland. 301 Cuart 14,—Curve representing the Distribution, by Percentages, in respect to Length, of 3600 Individuals of the Human Strain of the Trypanosome causing Disease in Man in Nyasaland, taken from the rat alone. Microns 24|25|26|27|28|29|30\1 |32|33|¢4|a5|s6|37|38 Bement eal aaa Beep See Se sees eee Secee Curves 13 and 14 will be found of use when the human strain of this species of trypanosome is compared with the Wild Game and the Wild Glossina morsitans strains. Table XX.—Comparison of the Percentages of Posterior-Nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Human Strain. Experiment 3 : Percentage among short and No. Strain. Name. Animal. stumpy cosine. — Il Mkanyanga Rat 34°1 728 101 HK “ 9°3 953 Til Chituluka b 10°3 1337 IV Chipochola 5 3°3 1660 V Chibibi 5 32-0 ARETENH® eon noonnnccc0oe 17°8 It is to be noted that in the human strain the percentage of posterior- nuclear forms varies greatly, although the method of enumeration is the same in each case. This presence of posterior-nuclear forms would have been accepted a few months ago as sufficient proof that the species dealt with was 7. rhodesiense. Since then posterior-nuclear forms have been reported as occurring in 7. brucei from Egypt, Uganda and Zululand. Ina strain lately obtained by Theiler from the same spot in Zululand where this species was originally discovered in 1894, this percentage rose to the highest yet recorded. VOL. LXXXVI.—B. Z 302 Messrs. W. Cramer and J. Lochhead. [Jan. 16, Conclusions. 1. The five human strains of this trypanosome, isolated from four natives in Nyasaland and one European in Portuguese East Africa, belong to the same species. 2. This species is 7’. rhodesiense (Stephens and Fantham). 3. Evidence is accumulating that 7. rhodesiense and T. brucei (Plimmer and Bradford) are identical. Contributions to the Biochemistry of Growth.*—The Gilycogen- content of the Liver of Rats bearing Malignant New Growths. By W. Cramer and Jas. LOCHHEAD. (Communicated by Prof. E. A. Schiffer, F.R.S. Received January 16,— Read February 20, 1913.) (From the Physiology Department, University of Edinburgh, and the Imperial Cancer Research Fund, London.) In previous papers by one of us (1, 2) observations on the gaseous metabolism and on the nitrogen metabolism of tumour-bearing rats have been recorded. The present paper contains observations on the carbo- hydrate metabolism of tumour-bearing rats. The tumour-strain employed in our previous work was also used for these experiments; it is a spindle- celled sarcoma (J.R.S. of the Imperial Cancer Research Fund) which has a rapid growth. This tumour does not contain any glycogen. A large number of glycogen estimations were carried out with the liver of normal and of tumour-bearing animals. The glycogen estimations were made by Pfliiger’s method; the glucose obtained by the hydrolysis of glycogen was estimated in the first part of the work gravimetrically according to the technique used by Pfliiger; later on, Bertrand’s method of titration was employed. It is necessary to bear in mind that only 4 to 6 grm. of liver tissue are available for analysis in these estimations, so that with a glycogen percentage below 0:1 to 0:2 per cent. the absolute amount of glycogen present in the whole liver of a rat is so small that it cannot be determined * This research is in continuation of papers in ‘Roy. Soc. Proc.,’ 1908, B, vol. 80, p. 263 ; 1910, vol. 82, p. 307 ; zbzd., p. 316. ae 1913. | Contributions to the Biochemistry of Growth. 303 accurately by these methods. Quantities of glycogen below this limit are therefore indicated in the following tables by the expression “ < 0:2 per cent.” Since the glycogen store of the liver is known to be dependent upon and influenced by external conditions, an attempt was made to equalise these as far as possible. All the animals were of about the same age and weight ; they were kept on a constant diet of bread and milk in the special metabolism cages devised by Professor Schiifer.* In some series the supply of food was limited to a definite quantity, which was sufficient to cover the require- ments of the animal with regard to the growth of both the host and the tumour ; in other series the supply was ad libitum. The results of these first observations, which have been grouped together in Table I, show that the glycogen-content of the liver of both normal and tumour- bearing animals varies within very wide limits. The following data were also known, but since they were found to have no relation to the variations observed in this series, they have been omitted from the table :—-sex of host, weight of host. The fact that the observations extended over several years also excludes the possibility of seasonal changes on the part of the host and of cyclical changes on the part of the tumour-cells as being responsible for these wide fluctuations. These must have been caused, therefore, by factors which had not been controlled by the conditions under which the observations had been carried out. A closer analysis of the results appeared to give a clue as to the nature of these factors. It will be seen that the variations in the glycogen-content of the liver are, as a rule, not so marked in animals killed on the same day. Now when several estimations were carried out on the same day, the animals were killed at practically the same time. Since all the animals were always fed at the same time, it seemed possible that the influence of the time which elapsed between the last meal of the animal and the moment when the liver was subjected to analysis was operative within narrower limits—in such small animals as rats at any rate—than the data given in the literature for larger animals would lead one to suppose. Some estimations carried out with normal rats weighing about 100 grm. showed that the glycogen-content of the liver was always relatively high three to five hours aiter a meal, and that 15 hours after a meal it fell so low (below 0:2 per cent.) that with the small amount of liver available it could not be determined. In order to be able to compare the conditions in tumour-bearing animals with those in normal animals it was therefore necessary to compare animals in the same state of digestion and assimilation. This object was attained in * For a detailed description, see ‘Quart. Journ. Exper. Physiol.,’ 1912, vol. 5, p. 204. Z 2 304 Messrs. W. Cramer and J. Lochhead. [Jan. 16, Table I. Normal rats. Tumour rats. Date. Weight of Liver Days Weight of | Weight of Liver liver. glycogen. of growth. tumour. liver. glycogen. grm, per cent. germ, grm per cent. 20.7.10 ...| 9 1°3 4°5 3°8 9 1°3 4°5 5° 26.7.10 ...| 15 3°5 5¢1 0°4 15 9°5 5¢L 1 °4 28.7.10 ...| 46 2°7 4°8 374 ; 4-7 3°0 AD) MO 3 18 6-5 5°1 0°5 | 18 3-7 4 °2 0:97 31.7.10 ...! 20 8-0 6-7 0°44 | 20 5-7 6-4 0°36 8.11.10 ...) 9 0°35 43 O7 9 0°5 5:2 5°7 12.11.10 ..,| 4:7 1:0 18.11.10 ...| 19 14 *4 5-2 2°1 19 10-4 6-4 1°6 22.11.10 ...| 23 19°3 57 il <8) | 23 19 °8 56 0°8 28.11.10 ... 4-9 2°2 5°5 1:0 1.12.10 ...| 5°3 1-1 32 6°3 5-7 iL 27/ 7.12.10 ...| 15 7-5 4,°8 0°5 | 15 7-2 Il Ol S210 5s 3°8 0°3 3°8 0°4 12.12.10 ... 20 165 4,°5 1°5 20 16°1 5°0 4:1 14.12.10 ... 4°7 1:14 22 3°5 6°3 i 3) 4-9 2°19 22 9-1 5°1 1°5 lez On@elelocen| 4°5 1:0 10 1:0 4-2 0°34, 25.7.11 ... 4°7 1‘4 15 1 *4 5-1 2°65 PAM Il a5 5:1 1°0 1.8.11... 5-1 14 22 1°6 5:0 1°6 | 4.8.11... 4°7 2-6 25 8 4:5 4°9 8 9% 2°15 * Pregnant animal. the following manner: Animals of approximately the same weight (100 grm.) were kept for at least two weeks on a diet of bread and milk. During the last five days the amount of food eaten was determined and, unless otherwise stated, the food was supplied ad libitum and was kept in the cages all the time, so that the animals could feed whenever they were hungry. As a rule the amount of food consumed in 24 hours did not vary very much. In the mornings when only a little food was left at the bottom of the beaker which the animal could not reach without difficulty, fresh food was placed in the cages, so that asa rule the animals would begin to feed at once. Having determined the amount eaten, animals were selected which had consumed 1913.] Contributions to the Biochemistry of Growth. 305 about equal quantities, the food was removed from the cage and a given time afterwards the animals were killed by breaking the neck. In every case a normal and a tumour-bearing animal were killed at the same time. The results obtained are given in Table II. Table II. Tumour rats. Normal rats. Hours | | after | | Diet last | Weight | Weight | 7... | Weight | 7.70. ae meal. | of of 1 of 1 | tumour. liver, | 898°" | iver, | SYCBED: | | : grm. erm. per cent. grm. | per cent. | 3 675 6°3 0 62 3°7 0°66 | Meat. 5d 7°4 1 ‘46 5°8 1°68 Bread and milk. 9-2 4°7 1°81 4°4 3°58 | Bread and milk, fasting 24 hrs. previous to last meal. | 7 9°5 5°0 <0°2 4°5 0°51 Bread and milk. 5:0 4-2 <0°2 An a Pe 9°5 5°0 <0°2 4:8 2°24 . i 45 1°52 » ” . 17 4:0 6°7 <0°2 4°8 <0°2 Meat. | 24 12°0 4:7 <0°2 3°8 <0°2 Bread and milk. 42 4:0 | <0-2 26 |<0-2 i s | The results show that a definite difference exists between the glycogen metabolism of a normal and a tumour-bearing animal : the glycogen disappears more rapidly from the liver of a tumour-bearing rat than from the liver of a normal rat. Se E—————— ee This result is in complete agreement with some observations on the gaseous metabolism, which were carried out by one of us in 1908 (1), and which have been confirmed since by Chisholm (3). It was found then that after a meal a tumour-bearing animal returns to the fasting state more rapidly than a ih eT A a normal animal. These observations, taken together with those made four years ago by Cramer and Pringle on the nitrogen metabolism of tumour-bearing rats. throw some light on the metabolic conditions of an animal in which processes of active growth are taking place. Cramer and Pringle found that a tumour transplanted into young growing rats grows to a considerable size without interfering with the growth of the animal. As the tumour grew the ; nitrogen excretion of the host fell, so that there was a “sparing” of the protein metabolism. The present observations show that this “sparing” is associated with an increased carbohydrate metabolism. In these experiments on the nitrogen metabolism it was necessary to maintain the animals ona constant nitrogen intake. It might be argued that 306 Messrs. W. Cramer and J. Lochhead. [Jan. 16, the “sparing” of the protein metabolism takes place only with a restricted food supply, but that with an unrestricted supply of food the animals simply take in more food and thus cover the requirements of the tumour. But the experiments of Medigreceanu (4), which we are able to confirm from our own observations, have shown that, even with an unrestricted food supply, tumour- bearing animals do not eat more than normal animals. He also found that the growth of the tumour, while it does not stimulate the host to an increased intake of food, leads to an increase in the weight of the liver (5). It is interesting to note that the same holds good also for pregnant animals. The sparing action on the protein metabolism referred to above might be explained simply by assuming that in tumour-bearing rats more carbohydrate material is burnt up instead of protein material, and that the glycogen which disappears so rapidly from the liver of tumour-bearing animals, by being used as a source of energy, protects an isodynamic equivalent of protein material, which then becomes available for the formation of new protoplasm. In that case the part taken by carbohydrates would not be specific and could be taken equally well by fats. On the other hand there is the possibility that the more rapid disappearance of glycogen from the liver is due to the fact that, in addition to the carbohydrate material which is burnt up and used as a source of energy, some glycogen is used together with nitrogenous substances as material for the synthesis of new protoplasm. The results of the observations on gaseous metabolism by Cramer and by Chisholm indicate that the latter alternative is the correct one. For if in tumour-bearing animals more carbohydrate were oxidised in place of proteins than in normal animals, one would expect to find that the rise of the respiratory quotient from about 0°7 to about 1 which takes place after a meal rich in carbohydrates should be more persistent in tumour-bearing animals than in normal animals. But the observations which we have quoted show that this is not the case; on the contrary, if there is any difference between tumour-bearing animals and normal animals it tends, in the case of rats, to be in the direction of a less persistent rise of the respiratory quotient in the tumour-bearing animals. We have repeatedly pointed out the close analogy which exists between the metabolic conditions of pregnancy and those of an animal bearing a trans- planted new growth. In the former, where a definite organ—the placenta— presides over the nutrition of the foetus, we have been able to show that carbohydrates are used as a material for the building up of the fetal protoplasm. The relation of the liver to the nutrition of the tumour is not so intimate as that of the placenta to the nutrition of the fcetus and does not afford so direct an insight into this aspect of the problem. But the evidence 1913. ] Contributions to the Biochemistry of Growth. 307 which we have brought forward in the present paper points to the same conclusions, Conclusions. Glycogen disappears more rapidly from the liver of tumour-bearing rats than from the liver of a normal rat. Since observations on the gaseous metabolism showed that there is no increased oxidation of carbohydrate material in tumour-bearing animals, the results confirm the conclusion arrived at previously from observations on pregnant animals, that in growth carbo- hydrate material is used for the synthesis of protoplasm. The expenses of this research were defrayed by grants from the Moray Fund of the University of Edinburgh. LITERATURE. 1, W. Cramer, “The Gaseous Metabolism in Rats Inoculated with Malignant New Growths,” ‘Third Scientific Report of the Imperial Cancer Research Fund,’ 1908, p. 427. 2. W. Cramer and H. Pringle, ‘The Nitrogen Metabolism of Rats bearing Malignant New Growths,” ‘ Roy. Soc. Proc.,’ 1910, B, vol. 82, p. 307. 3. R. A. Chisholm, “The Respiratory Exchange of Mice bearing Transplanted Carcinoma,” ‘Journ. of Path. and Bact.,’ 1911, vol. 15, p. 192. 4, FE. Medigreceanu, “ Ergebnisse eines Fiitterungsversuches bei Ratten, die iiberimpfte Tumoren trugen,” ‘ Berlin. klin. Wochenschft.,’ 1910, No. 17. 5. F. Medigreceanu, “On the Relative Sizes of the Organs of Rats and Mice bearing Malignant New Growths,” ‘ Roy. Soc. Proc.,’ 1910, B, vol. 82, p. 286. 6. J. Lochhead and W. Cramer, “The Glycogenic Changes in the Placenta and the Foetus of the Pregnant Rabbit: a Contribution to the Chemistry of Growth,” ‘Roy. Soc. Proc.,’ 1908, B, vol. 80, p. 263. 308 The Formation o, the Anthocyan Pigments of Plants. Part [V.*—The Chromogens. By FRreperick KEEBLE, Sc.D., Professor of Botany, University College, Reading; E. FRANKLAND ArmsTRONG, D.Sc, PhD.; and W. N. JonES, M.A., Lecturer in Botany, University College, Reading. (Communicated by W. Bateson, F.R.S. Received February 3,— Read February 27, 1913.) The object of the series of communications of which the present paper forms a part is the elucidation of the biochemistry and genetics of flower- pigmentation. In order to achieve this object it is necessary, in the first place, to ascertain the nature of the chemical processes which determine the formation of the anthocyan pigments, and, in the second place, to discover the chemical nature of the Mendelian characters to which the several varieties of a given species owe their power of forming and breeding true to definite types of flower colour. The history of the working hypothesis which we use in these investigations has been summarised in an earlier paper.t This hypothesis may be expressed in the form of the equations A and B :— A. Prochromogen+ enzyme = chromogen. ?a glucoside. ? emulsin. B. Chromogen+oxydase = anthocyan pigment. N Perdis + organic peroxide. Our previous and present communications are concerned with the latter, consideration of equation A being reserved for a subsequent occasion. The position to which our previous work has led us may be summarised thus :— The presence of oxydase in flowers may be demonstrated by means of benzidine, «-naphthol, or similar “artificial chromogens,” which, when acted * Previous papers of the series, which did not bear the general title, are :— [Part I.]|—“ The Distribution of Oxydases in Plants and their dle in the Formation of Pigments,” ‘Roy. Soc. Proc.,’ 1912, B, vol. 85; [Part II.]—“The Oxydases of Cytisus Adami,” ‘Roy. Soc. Proc., 1912, B, vol. 85; [Part III.]—“The dle of Oxydases in the Formation of Anthocyan Pigments of Plants,” ‘Journ. Genetics,’ Nov. 1912, vol. 2, No. 3. + “The Réle of Oxydases in the Formation of Anthocyan Pigments of Plants,” ‘ Journ. Genetics,’ 1912, vol. 2, No. 3. The Formation of the Anthocyan Pigments of Plants. 309 on by oxydase, yield pigments. These reagents serve not only to demon- strate the occurrence but also to determine the distribution of the oxydases of the flower. By the application of the method it is found that the distribution of oxydase coincides with that of anthocyan pigment. In white flowers oxydase may be present in an active or an inhibited state. In the former case some other part of the pigment-forming mechanism is absent from the flower; in the latter, the whole of that mechanism is present, but its action is prevented by the inhibition of the oxydase. The present communication and that which follows (Part V) deal primarily with the chromogens of the flower. Our first definite results demonstrating the existence of chromogens in the flower and the relation of these colourless substances with the anthocyan pigments were obtained with Brompton Stocks (Matihiola incana). These plants occur in numerous colour-varieties, the chief of which are pink, red, purple, and purple flaked with white. Flowers of any of these varieties, when treated with alcohol, lose their colour rapidly. It is therefore easy to obtain a series of colourless petals derived severally from each of the colour-varieties. If the decolorised petals of such a series be placed in water at room temperature they begin almost at once to regain their colour, and, after a quarter of an hour, each petal is found to be possessed of the identical colour of the variety to which it belongs. The petal originally pink recovers its pink colour, that from a red or purple variety becomes again red or purple, and that from a white-striped purple variety reproduces with faithful accuracy the purple and white pattern of the original. Despite the fact that our experiments have been carried out during the winter months, when suitable material is somewhat scanty, we have been able to prove that a similar recovery of natural colour is exhibited by many other flowers, for example :— Aubretia, wallflower (Cruciferee). Viola (Violacez). Pelargonium (Geraniacez). Cyclamen, polyanthus (Primulacez). Begonia (Begoniacez). Azalea (Ericacez). Bilbergia (Bromeliacez). Dendrobium (Orchidacez). Recovery of colour is shown also by the vegetative parts of plants which contain anthocyan, for instance, the leaves of Fuchsia and Tradescantia, and 310 Prof. Keeble, Dr. Armstrong, and Mr. Jones. _ [Feb. 3, the fronds of the Royal fern (Osmunda regalis). Loss and recovery of colour are therefore phenomena of very general occurrence, and may be regarded as characteristic of many, if not of all, kinds of anthocyan pigments. The wallflower is of special interest in this connection, in that the brown varieties with which we have worked contain representatives of the two types of pigment—anthocyanic and plastid-derived pigments—to either or both of which the colour of flowers may be due. The brown colour of the wallflower is produced by a purple anthocyan pigment and a yellow plast pigment. When acted on by alcohol a brown petal becomes decolorised ; but 1t recovers to a purple colour when treated with water. The recovery to purple instead of brown is due to the fact that the yellow plastid-pigment which contributes to the original brown colour is soluble in alcohol and is therefore extracted from the tissues by this reagent. Thus only the colour- less antecedent of the purple anthocyan pigment is left in the cells. Treated with water that antecedent gives rise to a purple pigment which, since it is no longer mixed with yellow, produces its proper optical effect. The yellow pigment may be obtained free from the anthocyan pigment by evaporating the alcoholic solution and washing the residue with water, in which the plast pigment is insoluble. The power of recovering to the original colour serves as a means of distinguishing the pigments of the anthocyan class from those which are derived from the plastids. The reproduction of the original colour in the petals of stocks and other plants is open to two alternative interpretations. On the one hand, it may be regarded as a phenomenon of like nature to that exhibited by indicators ; on the other hand it may be attributed to the oxidation of a chromogen. Immediate choice between the two interpretations is rendered difficult by reason of the fact that acids and alkalis exercise marked and definite effects on the colours of the anthocyan pigments contained in the flowers. Thus, in the presence of alkalis the pigment in the petals of stocks assumes a green-blue colour and in the presence of acids it becomes pink. Moreover the chromogen extracted by means of 50-per-cent. alcohol from the petals of stocks behaves as a very sensitive indicator. Dilution of the alcoholic extract with ordinary distilled water—which contains carbon dioxide— suffices to produce a pale pink colour. With mineral acids the colour becomes intense and with alkalis it passes through blue and blue-green to green. Nevertheless, and in spite of the complication introduced by these indicator effects, the evidence of the experiments now to be described points very definitely to the conclusion that the indicator hypothesis must be discarded in favour of its alternative. 1913.] Formation of the Anthocyan Pigments of Plants. 311 The petals of stocks decolorised by strong alcohol contain oxydase. The reproduction of the original colour by the immersion of decolorised petals in water is hastened by the addition of hydrogen peroxide. When a pink and a purple petal decolorised in the same alcohol are transferred to water to which a drop or two of hydrogen peroxide is added, the petals recover pink and purple respectively. Therefore it follows that the activating action of the peroxide is due to its provision of oxygen and not to its acidity. In water which has been boiled for a long time in order to remove the oxygen, the recovery of colour, if it take place at all, occurs more slowly than in unboiled water. The addition of dilute hydrogen cyanide—a substance known to inhibit oxydase-action—prevents the recovery of colour. The results of other experiments designed to decide between the indicator and oxidation hypotheses lend further support to the latter. Thus decolorised petals in which the pigment has been caused to reform may be again decolorised either by leaving them in water till the pigment has diffused away or by transferring them to alcohol. Petals treated in this manner, if placed in hot water, produce once again their natural pigments. Again, the restoration of pink colour to decolorised petals of a pink variety may be brought about in an alkaline medium; for example by transferring the petals from alcohol to water containing a small quantity of hydrogen peroxide which has been rendered faintly alkaline. The pink colour which is thus induced changes subsequently to purple. Conversely the purple colour returns to a petal of a purple variety even though the * medium in which the change is effected be acid. In this case the recovered purple soon becomes pink owing to the action of the acid. The last experiment is rendered still more conclusive if the procedure be modified in the following manner. Petals of purple stocks are incubated with 99-per-cent. alcohol to which enough citric acid has been added to render the alcohol distinctly acid to litmus. The petals became almost decolorised, retaining only a faint pink colour. When transferred to distilled water—which is not acid to litmus—pigment is produced in con- siderable quantities and the colour of the pigment is at first red but soon becomes purple. If the colour were of the type of an indicator reaction the effect of the water would be not to intensify but to dilute the tint. The conclusion which we have reached as the result of these observations is that, although indicator changes run parallel with the changes involved in the formation of anthocyanic pigments, the latter arise as the result of the oxidation of chromogens. It remains to mention the remarkable acceleratory effect which high temperature has on the reproduction of the natural pigments of stocks. As 312 Prof. Keeble, Dr. Armstrong, and Mr. Jones. __[Feb. 3, stated already, a petal recovers its colour in water at room temperature in the space of a quarter of an hour. At higher temperatures the recovery is more rapid and if a petal be dropped into water which has been heated to near the boiling point the recovery of colour is almost instantaneous. We turn now to the detailed interpretation of the facts of loss and recovery of colour; and we deal first with the loss of colour which takes place when petals are dehydrated. The evidence about to be given supports the conclusion that the loss of colour is due to the action of a reducing agent. In the present state of our knowledge of the reducing processes which occur in the cells of plants it is not possible to affirm that the agents of these processes are of the nature of specific catalysts. We propose therefore to avoid using the word reductase and to employ the indifferent term “reducing agent” in the description of the phenomena of decolorisation. A careful examination of the petals of stocks subjected to the action of alcohol makes it difficult to escape from the conclusion that decolorisation is due to the activity of a reducing agent. It is easy to demonstrate that the loss of colour is not due merely to a dissolution of the pigment and its diffusion throughout a large bulk of fluid. As evidence that the loss of colour is due to the action of a reducing agent we may cite the following facts :— The immersion of the petals in alcohol produces three immediate effects,— a rapid evolution of gas, a reduction in the amount of colour, and a discharge from the petals of a certain amount of pigment which dissolving in the ~ alcohol gives rise to a marked coloration of that reagent. Similar effects are produced, but more rapidly, if previously to their immersion in alcohol the petals are treated for about half a minute with chloroform. As a consequence of the discharge of the pigment the alcohol becomes deeply coloured—purple or red according to the variety of stock used in the experiment. If the alcohol be decanted at once its colour disappears with remarkable rapidity and in less than 5 minutes the liquid becomes colourless or at most faintly coloured. The partly decolorised petals, from which the first lot of alcohol was removed, if treated with more of this reagent, undergo further decolorisation, but at a much slower rate. The simultaneous evolution of gas and the discoloration suggest that the effect of the alcohol is to liberate a reducing agent which brings about the deoxidation of pigment and an evolution of oxygen. Further evidence of the presence of such a reducing agent is provided by extracts prepared by pounding fresh petals with alcohol. The colour of the extract is at first 1913.] Formation of the Anthocyan Pigments of Plants. 213 identical with that of the petals from which it was made; but sooner or later the colour fades and the solution becomes colourless. The fading is rapid in concentrated alcohol and slow in alcohols of somewhat weaker grades. The agent responsible for the fading is resistant to high tem- peratures. Thus if alcoholic extracts be evaporated to dryness and the residues be taken up with water, the fading of the solutions still takes place Further evidence in favour of the view that decolorisation is due to reduction is offered by the results of experiments on the effect of extracts in inhibiting and in reversing oxydase-action. The experiments were made in the following ways :— 1. Extracts made from Stocks by Grinding the Petals with Alcohol—A solution of the peroxydase of bran is rendered of such a strength that it just gives the characteristic blue reaction with benzidine and hydrogen peroxide. Petals of a coloured variety of stock are ground with alcohol, the extract is evaporated to dryness, and the residue dissolved in water. If a few drops of the latter solution be added to the solution of peroxydase, and if the benzidine-hydrogen peroxide test be applied, no colour-reaction ensues. The oxydase is prevented by the reducing agent from bringing about the oxidation of benzidine. Only if it be increased very considerably in amount is the oxydase able to overcome the opposing influence of the reducing agent, and to bring about the oxidation of the benzidine. 2. Extracts obtained by Immersing Intact Petals im Strong Alcohol—The use of extracts made by grinding petals with strong alcohol is open to obvious objections. We have, therefore, used extracts obtained by the immersion of intact petals in strong alcohol. For this purpose petals of purple stocks are immersed in alcohol of 99 per cent. When the alcohol is decanted from the tube containing the petals, its colour (pale purple) disappears in the course of a few minutes. On evaporation over a water-bath it yields a purple residue. For the purposes of control an equal volume of alcohol of the same strength as that used for the extraction of the petals is also evaporated to dryness. A bran peroxydase is prepared of such a strength that when a given volume of it is added to a given volume of a weak solution of benzidine containing one drop of hydrogen peroxide a definite but pale blue colour is produced. The addition of similar volumes of peroxydase, hydrogen peroxide and benzidine to the purple residue results in the production of no blue colour, whereas the colour develops normally when the reagents are added to the vessel .in which the alcohol alone has been evaporated to dryness. The alcohol which has been in contact with the petals, hke the alcoholic extract obtained by maceration, prevents the action of oxydase. 314 Prof. Keeble, Dr. Armstrong, and Mr. Jones. __[Feb. 3, Yet more conclusive is the result when the blue solution, produced by the action of bran peroxydase and hydrogen peroxide on benzidine, is added to the residue obtained by evaporating alcohol which has been in contact with intact petals. The blue colour of the former is discharged immediately, that is to say, the action of the oxydase is reversed, and the blue product of the oxidation of benzidine is reduced to its original colourless state. That this effect is not due to reducing agents present in the alcohol is shown by the fact that no discharge of colour is brought about by the addition of the blue oxydase-benzidine solution to the residue left after the evaporation of alcohol which has not been in contact with petals. This method of demonstrating the presence of a reducing agent is the more conclusive in that whereas alcohol alone reduces oxydase-activity, it does not bring about a reversal of the action. The only effect of alcohol on the blue colour is to precipitate the blue pigment. We have shown in a previous communication (Part III) that the oxydases of the flower not only act on the forerunners of pigment contained in the petals but also on the artificial chromogen benzidine and give rise to pigments ; we now show that flowers contain reducing agents which are not only capable of inhibiting the action of oxydase, but are able also to reduce both the natural pigments of the flower and the “artificial” benzidine pigments to the colourless state of chromogens. Two facts stand out prominently in the foregoing investigation of decolorisation. These facts are that the reducing agent is very resistant to high temperatures and that it is active in strong alechol. The former we have studied in sufficient detail only to be able to state that the reducing agent is not destroyed by exposure to a temperature of 100° C., the latter fact has been investigated more fully and with the following results :— Both evolution of gas and fading of the flower take place rapidly in aleohol of 95 per cent. These processes go on, albeit more slowly, in yet stronger alcohol. Thus with the ordinary absolute alcohol of the laboratory (99 per cent.) a certain amount of gas is evolved and colour begins to disappear; but when petals are placed in alcohol of approximately 100 per cent. both processes, although they take place, come to an end much sooner than in the alcohol of slightly lower grade. We conclude, therefore, that the reducing agent which brings about decolorisation of the petals of stocks is able to exhibit its specific action in tissues which are almost completely dehydrated. We have, moreover, evidence that loss of colour occurs naturally in the plant. We know, for example, that in many plants light shades of colour are dominant to dark shades; we know also that the flowers of such plant of stocks may assume as 1913.| Formation of the Anthocyan Pigments of Plants. 315 they fade a new colour, and we know that the colour of some flowers under- goes a marked change during the course of the day. Such changes are to be ascribed to the simultaneous presence in the petals of pigments, chromogens, oxidizing and reducing agents. We have now to consider the conditions under which recovery of colour occurs. Evidence has been given already in favour of the interpretation that recovery is brought about by the oxidation of chromogen. There is, however, one series of facts, namely, those bearing on recovery of colour by petals immersed in strong alcohol, which seems to throw doubt upon this conclusion. For, as several investigators have shown, increasing concentrations of alcohol exercise a progressively adverse effect on enzyme action. Thus, Hudson (1910),* working with invertase has expressed the effects of different concentrations of alcohol in the form of a regular curve. He finds that in 70-per-cent. alcohol invertase retains only 10 per cent. of its activity. We have studied the relation of the activity of maltase to alcoholic con- centration and find that this enzyme is even more sensitive to ethyl alcohol than is invertase; the activity of maltase ceasing in 60 per cent. With methyl alcohol a 40-per-cent. solution suffices to render the enzyme inactive. Our experiments with emulsin, which confirm those published in 1912 by Bourquelot, give results similar to the foregoing except in one important particular, The activity of emulsin falls rapidly as the concentration of alcohol increases to 40 per cent. After this point is reached the activity falls off more slowly and some activity may be detected in solutions containing 90 per cent. of alcohol ; in solutions containing from 40 to 90 per cent. of alcohol the activity of emulsin is proportional roughly to the amount of water present in the solution. For the purpose of investigating the effect of aleohol on oxydase we have made use of the Lovibond tintometer. We measure by means of this apparatus the depth of coloration—a mixture of red and yellow—produced by the action of bran peroxydase on guaiacol. The curve representing the amount of oxydase action—as measured by the tintometer—is similar to the curves which have been obtained for invertase, maltase and emulsin. As the alcoholic strength increases the activity of oxydase falls. In 50-per-cent. solutions it becomes very small and ceases altogether in 70-per-cent. alcohol. Alcohol causes a similar retardation of * Hudson, C. §., ‘U.S. Dept. of Agric., Bureau of Chemistry, Circular 58.’ 316 Prof. Keeble, Dr. Armstrong, and Mr. Jones. _ [ Feb. 3, the benzidine reaction ; but the colour of the latter is not suitable for tinto- metric estimation. The results obtained in test-tubes are opposed to the view that the recovery of colour in petals immersed in strong alcohol is due to the activity of oxydase. (General considerations, however, led us to suspect that although alcohol of 70-80 per cent. may prevent the action of the oxydase in solutions extracted from plant-tissues, it might prove less potent to retard the action of oxydases in the tissues themselves. We have confirmed the truth of this suspicion in the following way :— Petals of purple stocks were incubated with 99-per-cent. aleohol and when decolorised they were placed, some in 70 per cent., and others in 80, 90, and 95-per-cent. alcohol. Equal quantities of a solution of benzidine in water- free alcohol were added to each tube containing the petals, one drop of hydrogen peroxide was introduced into each tube, and the preparations were placed in the incubator at 37° C. Examination of the petals after half-an- hour showed that the petal treated with 70-per-cent. alcohol gave a well marked brown benzidine reaction for oxydase, the petal in 80 per cent. showed a very distinct reaction, that in 90 per cent. an equally good or even better reaction, and that in 95 per cent. a slight but distinct reaction in the veins of the claw. Whence it follows that the peroxydase of stock petals is capable—if peroxide be present—of bringing about the oxidation of benzidine even in a medium containing 95 per cent. alcohol; and we infer that what is true of this artificial chromogen is true of the natural anthocyanic chromogen, namely, that the latter may undergo oxidation even in the presence of 95-per-cent. alcohol. Thus the conclusion is reached that, although the experiments with oxydase extracted from plant tissues are adverse to the view that recovery of colour is due to oxidation, the more apposite and crucial experiments with the oxydases contained within the petals lend powerful support to that view. The series of observations described in the foregoing pages lead us to the following conclusions :— In concentrated alcohol the anthocyan pigments are reduced to the state of colourless chromogens. The reduction is brought about by reducing agents, the nature of which is unknown. The reducing agents may be specific chemical substances ; they may perhaps be of the nature of catalysts; they are probably not enzymes (reductases). It is interesting to observe that an effect similar to that exercised by the reducing agent contained in stocks is brought about by hydroquinone, though not by formaldehyde. When the concentrated alcohol is replaced by water, the oxydases, which 1913.] Formation of the Anthocyan Pigments of Plants. 317 are not destroyed by the former reagent, resume their activity, and colourless chromogen is converted into anthocyan pigment. The fact that the colours of the pigments thus produced are identical severally with those of the natural petals, indicates either that chromogens of more than one kind exist in the different colour-varieties of stocks, or —what for our present purpose is nearly the same thing—that one chromogen is present, and associated with it are substances which determine the colora- tion of the oxidised produce of the chromogen. The petals of plants such as stocks contain much larger quantities of chromogen than are used in the natural flower. Not only may the original depth of colour be recovered, but the pigment so formed may be removed from the tissues and further instalments of pigment may be produced. Whether the reserves of chromogen contained in the flower occur as such or, as would appear more probable, in the form of prochromogen, we cannot at present say. The factor which determines the direction in which the pigment-producing reaction shall go is the amount of active water present in the cells. As the amount of water decreases, the reducing agents of the cell become active and oxydase becomes inert; as the amount of water increases oxydase action comes into play and the reducing agents are either destroyed, or, if they persist, any action which they exert is masked by the superior and opposed activity of oxydase. The relations may be expressed diagrammatically by the following scheme :— ——-—_=- Oxydase } + Water Chromogen _ Anthocyan pigment. — Water Reducing agent. SNE se Ut The occurrence of reducing bodies side by side with oxydases in the anthocyan-containing tissue of plants, the antagonistic relation which obtains between the reducing and oxidising agents of the cell, and the relations which exist between the activities of these agents and the degree of hydra- tion of the cell are calculated to throw light, not only on the phenomena of pigment-formation and pigment-inhibition in plants, but also on others of wider import. Following the clue offered by these experiments we may hope perhaps to advance towards an understanding of the biochemical meanings of activity and latency of seeds, of the enforced and natural awakening of vegetation, and of cognate phenomena. A discussion of the foregoing facts in relation with these phenomena lies, however, beyond the scope of the present communication. VOL. LXXXVI.—B. 2k 318 The Formation of the Anthocyan Pigments of Plants. Part V.—The Chromogens of White Flowers.* By W. Neiison Jonzs, M.A., Lecturer in Botany, University College, Reading. (Communicated by W. Bateson, F.R.S. Received February 4,—Read February 27, 1913.) The series of communications of which the present paper forms a part (Part V) deals with the biochemistry and genetics of pigmentation in plants. Parts I and III of the series describe the véle of oxydases in the formation of the anthocyan pigments of flowers; Part IV gives an account of the chromogens which constitute the colourless antecedents of these pigments ; and the present paper has for its object the investigation of the chromogens in white flowers. The subject is of interest from the point of view both of biochemistry and genetics; for, as has been discovered by Mendelian research, the white flowers which occur so commonly in cultivated and wild plants belong to more than one category. The types of white flowers recognised hitherto are known respectively as dominant and recessive whites. As shown by Keeble and Armstrong, both dominant and recessive white flowers contain oxydase (or peroxydase). In the former the oxydase is inactive owing to the presence of an inhibitor; in the latter it is active. Inhibition of oxydase suffices to account for the absence of colour from dominant white flowers. In order to account for the absence of colour from recessive whites it is assumed that some part of the colour-forming mechanism—for example, the chromogen—is lacking from the flowers. It is, however, possible that lack of colour may be in some cases the consequence not of absence of an essential constituent of the colour-producing mechanism, but of the failure of these constituents—all of which are present in the flower—to come together and interact with one another. The results of experiments about to be described show that both kinds of recessive white flowers exist. As the result of treating the white flowers of Lychnis coronaria var. alba with alcohol (15-80 per cent.), chloroform, ether, or carbon disulphide, a brown pigment develops. * Parts I and II, Keeble and Armstrong, ‘ Roy. Soc. Proc.,’ 1912, B, vol. 85, pp. 214 and 460; Part III, Keeble and Armstrong, ‘Journal of Genetics,’ November, 1912 ; Part IV, Keeble, Armstrong, and Jones, ‘Roy. Soc. Proc.,’ 1913. The Formation of the Anthocyan Pigments of Plants. 319 This pigment. is at first limited to the veins, though subsequently the whole petal becomes distinctly coloured. The depth of colour is considerable, and the general aspect of the brown petals resembles closely that produced by the action of benzidine. There seems no doubt, indeed, that the brown coloration obtained by treating petals of Lychnis coronaria with an alcoholic solution of benzidine is due to this effect of the alcohol rather than to a reaction between oxydase and benzidine. That this is so is indicated by the fact that addition of hydrogen peroxide to a petal so treated causes a further and immediate darkening. If, however, petals are immersed in absolute alcohol from which the water has been removed by anhydrous copper sulphate, no browning occurs. This is to be expected if the browning is due to oxydases, for, as shown in Part IV, the oxydases are thrown out of action temporarily by dry alcohol. Petals transferred to water after soaking an hour or so in dry alcohol rapidly develop the brown colour; but petals that have been left several days in the dry alcohol form no brown pigment on transference to water, nor does the addition of hydrogen peroxide cause it to appear. If now it be assumed that the formation of the brown pigment under the influence of chloroform, alcohol, etc, is due to interaction between a _ colourless “chromogen” and an oxydase (kept apart in the intact petal but allowed to come together when the alcohol has destroyed the impermeability of the plasmatic membrane) ; then the failure of the pigment to develop in the case of petals that have been soaking some time in alcohol may be taken to indicate that the chromogen has been removed from the petals and diffused out into the alcohol. Failure of the brown colour to appear is not due to the destruction of the body that functions as peroxide, since addition of hydrogen peroxide is without effect; nor is it due to the destruction of the peroxydase itself, since the petal, after long immersion in alcohol, gives a good benzidine reaction for peroxydase. Ii the above view of what occurs be correct, the absolute aleohol in which the petals have been soaked should contain the chromogen in solution. In order to prove that this is the case, a considerable number (150) of Lychnis coronaria flowers were treated with 50-per-cent. alcohol, raised to boiling point in order to destroy the oxydase. After concentrating the extract to a small bulk, white Lychnis fiowers, soaked in dry alcohol as above to remove the chromogen, and known to contain peroxydase, were incubated at 36° C.in the solution. The flowers remained colourless while in the extract, but when they were transferred to water containing hydrogen peroxide a reddish-brown pigment appeared at 2A 2 320 Mr. W.N. Jones. The Formation of the [Feb. 4, once in those parts of the flower which contain peroxydase. Hence it is demonstrated that the petals of Lychnis contain a chromogen, which, when extracted from the flowers, is acted on by the peroxydase contained in the petals aud gives rise to a red-brown pigment. The peroxydases of Primula sinensis, Primula obconica, Dianthus sp., etc., were shown also to bring about —in the presence of hydrogen peroxide—an oxidation of the chromogen extracted from the petals of Lychnis. A similar chromogen has been extracted from the white-flowered variety of Anemone japonica. Like that obtained from the flower of Lychnis coronaria, it yields pigments when acted on by the oxydases of petals of various plants. The white flowers of eg. Jychnis coronaria thus yield an extract which can be used to demonstrate the distribution of oxydases in place of a benzidine solution. The experiments show, moreover, that these flowers of Lychnis coronaria, although they are white, contain both oxydase and chromogen.* It is therefore probable that these constituents are located in different cells or parts of the same cell, and that whiteness is due to the fact that the plant. lacks the means of bringing chromogen and oxydase into contact with one another. As has been mentioned already, the pigment obtained by the action of the peroxydase of the petals of Lychnis coronaria on the chromogen extracted from these petals is of a reddish-brown colour. It might, therefore, be urged that the chromogen which gives rise to this pigment is not that which in coloured flowers yields the red anthocyan pigment of the natural petals. The objection is weighty; but that it may be met is shown by the following considerations and experiments :— 1. It is known that changes in the chemical nature of the chromogen, the degree of oxidation, the conditions under which the reactions occur, and the presence of traces of other substances,{ affect the colour of the end product of oxidation. Too much weight, therefore, should not be attached to mere difference in colour as the colour is very susceptible to alteration. * Since browning of the fresh petal occurs under the influence of alcohol alone, the body that behaves as a peroxydase towards e.g. a-naphthol, can behave as an oxydase towards the natural chromogen. + In this connection it may be noted that if a pink bract of Bilbergia sp. be immersed in H,0,, the pink pigment becomes changed into brown, presumably as the result of further oxidation. In ‘U.S. Dept. of Agric. Bureau of Plant Industry Bulletin,’ No. 264, 1913 (received as the present paper goes to press), H. H. Bartlett records a red pigment of Dioscorea as becoming brown on oxidation. t Chodat, R. “Nouvelles Recherches sur les Ferments Oxydants. Les matiéres protéiques et leurs dérivés en présence du réactif p-crésol tyrosinase.” ‘Arch. Sci. Phys. Nat.,’ 1912 (LV), vol. 33, pp. 70, 225. 1913. ] Anthocyan Pigments of Plants. 321 2. The behaviour of flowers of Brompton stocks, as described in detail in Part IV of this series, provides a convincing proof that it is possible for all the mechanism for colour production to be present in a flower, and yet for the bodies concerned not to interact to produce pigment until the plasmatic impermeability has been destroyed. The fading and recovery of colour of petals of these plants was observed by the present writer during the preliminary experimental work in connec- tion with the above paper. The facts will be referred to here only in so far as they illustrate the point at issue. If coloured flowers of Brompton stocks are soaked in absolute alcohol, the contained pigment gradually fades; on transferring the colourless petals to water they quickly become coloured, the “ recovered colour” being of exactly the same shade as that of the fresh flower used. In the paper referred to evidence is presented that the fading of the coloured petal in alcohol is due to the reduction of the pigment to a colour- less state, as well as to its diffusion out into the surrounding liquid, and that the formation of pigment when the petal is transferred to water is the result of an oxydase converting into pigment a colourless chromogen substance contained in the petals in addition to a re-oxidation of the reduced pigment remaining. Thus, in stock, oxydase and chromogen are both present, and the condi- tions are naturally such as to allow a proportion of these two bodies to come together to produce pigment. In the white-flowered variety of Lychnis coronaria the natural conditions are never such as to allow any interaction between oxydase and chromogen. On treatment with alcohol the barrier is removed by the destruction of the plasmatic impermeability, and, as a result, pigment is produced. The method which serves in the case of Lychnis to bring chromogen and oxydase together, and causes them to interact with one another, serves with Brompton stocks to bring about a large increase in the amount of pigment present—which pigment is of the same colour as that occurring in the flower under natural conditions. By use of such methods the following types of white flowers have been demonstrated in the course of this investigation :— 1. White Flowers which contain an Oxydase and a Chromogen, eg. Lychnis coronaria, Anemone japonica, Chrysanthemum sp.—When petals of these plants are subjected to the action of alcohol, chloroform, etc., a colour change is produced. The colour may be brown, as in Anemone japonica, and appear more or less evenly all over the petal; or of a reddish tinge, as in Lychnis coronaria, where. the colour is located chiefly in the veins. In both these 322 The Formation of the Anthocyan Pigments of Plants. examples the depth of colour obtained is very considerable; in the case of Chrysanthemum the colour change is only slight. All the flowers belonging to this class give the characteristic peroxydase reaction with benzidine or «-naphthol solutions and hydrogen peroxide. 2. White Flowers which contain a Peroxydase and a Chromogen.—tThis type of white flower is illustrated by certain varieties of Dianthus caryophyllus (e.g. var. “ Mrs. Sinkins,”) and of Dianthus barbatus (Sweet William), ete. These flowers on treatment with dilute alcohol, chloroform, etc., show no development of colour, but the addition of H,O2 causes a rapid formation of pigment. In many flowers of this class colour is produced only locally in the petal on the addition of hydrogen peroxide. On testing such a flower with benzidine solution with the subsequent addition of hydrogen peroxide, a peroxydase reaction is obtained only in these same localities. The peroxydase is limited, therefore, to those areas that give a colour reaction when treated with alcohol, etc., and hydrogen peroxide.* Whether the chromogen, which contributes to the reaction occurring in alcohol is present in the regions from which peroxydase is absent is as yet undetermined. 3. White Flowers which contain a Peroxydase but no Chromogen.—The white- flowered varieties of Plumbago capensis and Swainsonia Tacsonia illustrate a third type of white flower. No colour reaction is given after treatment with chloroform or similar bodies even after the addition of hydrogen peroxide. Such petals, however, give in every case a reaction with benzidine and hydrogen peroxide. 4. White Flowers which contain no Oxydase or Peroxydase-—A fourth type may perhaps be inferred from the behaviour of a white variety of Sweet William investigated by Keeble and Armstrong, which was found to give no benzidine reaction, direct or indirect, and was therefore assumed to lack oxydase and peroxydase. The possibility of an inhibitor being present was not overlooked, but was not investigated. Information as to the occurrence of a chromogen in these flowers is also wanting. The interpretation suggested above of the behaviour of white flowers when treated with alcohol, etc., is the most obvious and simple one, but it is fully recognised that intermediate steps may occur of which no account has been taken. In Lychnis coronaria, it may be that a chromogen, as such, does not occur in the intact petals, but is split off from a body which one may term a pro-chromogen, after the plasmatic impermeability has been destroyed by treatment with alcohol. * In coloured varieties these same areas are often the only bas of the flower containing pigment. A New Ganglion in the Human Temporal Bone. 323 A modification of the interpretation on these lines, however, does not affect the general hypothesis as to the existence of several types of white flowers, or the inference that pigment is not necessarily produced although all the requisite ingredients for the production of colour may be present. On the Occurrence of a Ganglion in the Human Temporal Bone not hitherto Described. By Apert A. GRAY. (Communicated by Prof. J. G. McKendrick, F.R.S. Received November 26, 1912, —Read February 6, 1913.) (PLATE 6.) The existence of a previously unknown nerve plexus associated with a comparatively large ganglion embedded in the substance of the human temporal bone must be regarded as a somewhat surprising fact at this period in the history of human anatomy. It may be well, therefore, to describe in a few lines the process which led to the discovery. While making some preparations of the middle ear of animals according to my own method, I discovered the presence of a large plexus of nerves on the posterior surface of the bulla of the sheep. This plexus was found to be composed of bundles derived from the pneumo-gastric and the facial nerves. Since the preparation was only macroscopic, I was unable to ascertain whether nerve ganglion cells were present. Such a plexus has not been described in the human subject, unless the two minute bundles of fibres which pass between Arnold’s nerve and the facial nerve be dignified by the name of plexus. It seemed, therefore, highly probable to me that some corresponding structure might exist in the human subject which had not hitherto been described, and a search was accordingly made. The initial difficulty lay in the fact that this portion of the temporal bone is very different in man from that in the sheep. In the latter there is a large bulla, but no mastoid process; whereas in man there is a large mastoid process and no bulla. In man the only indication of a bulla is the little cul-de-sac which runs backwards from the lower, inner, and posterior corner of the tympanum. In the human subject a mass of bone fills the space which in mammals is occupied by the bulla. The plexus, 324 Mr. A. A. Gray. A Ganglion in the Human [Nov. 26, therefore, if present in the human subject, would probably lie embedded in bone. The second difficulty lay in the fact that the structures in the region involved do not have the same relationship to one another in the human subject as in the sheep. The stapedius muscle, for example, lies below, internal to, and in front of the facial nerve in man, not above it as in the sheep. The plexus which was seen in the sheep had a position a little below the lower termination of the stapedius muscle, and internal to the facial nerve. Since the stapedius muscle has, so to speak, been rotated downwards and inwards in man, as compared with the position which it occupies in the sheep, the most likely place for the plexus would be a little below the lowest point of the origin of the stapedius muscle. This clue was followed, and, as will be shown later, the inference was justified by the result. The petrous portion of a human bone was removed in the fresh state and a small piece, including the region indicated above, was decalcified and prepared for microscopic section. The preparation, after very complete decalcification, was embedded in celloidin, and sections in the vertical plane were made from before backwards. The sections were stained with iron-alum and hematoxylin. The series of sections was not complete, but it was quite sufficient to reveal the existence of a plexus, corresponding to that of the sheep, but very much smaller in size. Owing to the incompleteness of the series it cannot be definitely stated what is the exact origin of all the bundles which go to compose the plexus, but it is clear that they are derived from at least two sources: first, the facial nerve; second, the auricular branch of the pneumogastric. While it was interesting to discover this plexus in man, a much more remarkable fact was also revealed. This was the presence of a com- paratively large ganglion associated with the plexus, and, like it, embedded in bone. The first impression on finding this ganglion was that it was a portion of one of the ganglia of the glossopharyngeal or pneumogastric nerves. But closer examination showed that this could not be the case, because it was at a considerable distance from the trunks of these nerves, and, moreover, it was embedded in bone. A further search was then made among the writer’s macroscopic pre- parations of the temporal bone, and the ganglion was discovered to be present init also. It is shown in Plate 6, fig. 1, which is taken from the write1’s text-book on ‘ Diseases of the Ear.’ It will be seen that the ganglion, 9.., lies immediately below the inferior termination of the stapedius muscle, and about the same distance in front of the facial nerve in the vertical portion 1912. } Temporal Bone not hitherto Described. 325 of its course. The auricular branch of the pneumogastric nerve passes upwards from the jugular fossa through the bone towards the ganglion (Plate 6, fig. 1, a.p.). The general position of the ganglion having now been described, it remains to give a few details concerning its finer structure. This can only be done from a series of microscopic sections. Such a series was made, but the decalcification process was too energetic, and in some of the sections portions of the bone and even portions of the ganglion itself have been washed away. The ganglion is very irregular in shape, and is surrounded on all sides by bone. As a result of this irregularity in shape different portions of the structure come into view in different portions of the same section, so that at first sight it would appear that there are two or more ganglia. But when the series is studied carefully it is found that this appearance is merely due to the presence of outlying semi-detached portions of one ganglion. The name which I propose to give to the structure is “the Stapedial Ganglion.” It is situated close to the lowest point of the stapedius muscle in man, and the name suggested is, perhaps, as appropriate as any. The first section (fig. 2) passes through the anterior portion of the ganglion, Considerable portions both of the bone and of the ganglion itself have been lost in the course of preparation, but the upper and lower parts of the latter are seen, g.g. The posterior semicircular canal is seen to the left of the uppermost portion of the ganglion, and the jugular fossa is shown in the left lower part of the photograph. A bundle of fibres derived from Arnold’s nerve runs in the direction of the ganglion. f. facial nerve ; p. posterior semicircular canal ; g.g. ganglion ; 7. jugular fossa ; a. auricular branch of vagus nerve. 326 Mr. A. A. Gray. A Ganglion in the Human [Nov. 26, In the next section (fig. 3) the ganglion is seen to be much reduced in size, and now appears as one piece which is quadrilateral in shape. The facial nerve shows rather more in this section as it is beginning to turn downwards. =e De ma. aaa ies) SESS < 4 1 fe , “= ? Fig. 3. f. facial nerve ; p. posterior semicircular canal ; g. ganglion ; 7. jugular fossa ; a. auricular branch of vagus nerve. On coming to examine the minute structure of the ganglion there arises the difficulty that, in the human subject, the structures have undergone con- siderable post-mortem changes, owing to the fact that they cannot be put into the fixing fluid until a considerable time after death. Besides this difficulty another occurs in this special case, in that the fixing fluid takes some time to penetrate the bone which surrounds the ganglion. It is impossible, therefore, to give any satisfactory description of the minute intracellular appearances of the nerve-cells of the ganglion. As regards their general appearance, however, some points are obvious. As a whole the ganglion contains a rather large proportion of nerve-fibres relative to the number of nerve-cells, Plate 6, fig. 4. The nerve-cells are present in numerous groups which are separated by bundles of fibres, and it is quite impossible to say whether all the fibres are connected with the nerve-cells of the ganglion or not. As regards the cells themselves the majority are multipolar, and in this respect they differ from those found in the terminal ganglia of the eighth nerve, which are bipolar. In the stapedial ganglion there are also a few bipolar cells to be found in the upper portion (Plate 6, fig. 4), but this does not alter the general statement that the cells are multipolar in character. 2 < Roy. Soc. Proc., B, vol. 86, Plate 6. Gray. 1912.) Temporal Bone not hitherto Described. 327 It is hardly possible at present to speak of the physiological significance of this ganglion. Experiment and careful pathological and clinical observation alone can determine this. There is, however, one very interesting clinical fact which is probably worth consideration in this respect. There is a very common form of deafness, unassociated with middle ear disease, and termed otosclerosis. Among other symptoms of this affection are, a diminution in the secretion of wax in the external meatus, and a diminished sensitiveness of the tympanic membrane and the posterior wall of the meatus. These are the regions supplied by Arnold’s nerve which is formed by the union of bundles from the pneumogastric and facial nerves. As the stapedial ganglion is composed of fibres derived from both of these nerves it is very possible that autonomic fibres run from it to the parts mentioned. This would account for the diminished secretion of wax in the cases mentioned above. DESCRIPTION OF PLATE 6. Fia. 1.—The soft structure of the human middle ear. 2. incus ; f. facial nerve ; s. stapedius muscle ; g.g. ganglion ; a.p. auricular branch of vagus nerve ; ¢. nerve of Jacobson. Fie. 4.—Section through ganglion showing the nerve cells. x 250 ca. 328 Studies on Enzyme Action. XIX.—Urease: a Selective Enzyme. Il.—Observations on Accelerative and Inhibitive Agents. By H. E. Armstrone, F.R.S., M. 8S. BensaAmin and Epwarp Horton. (Received January 1,—Read February 20, 1913.) In the previous communication* experiments were described which had been made with the Urease present in the Soja bean proving that the enzyme is strictly selective in its action and that whilst its activity is much reduced by ammonia it is increased, in a remarkable manner, by the presence of carbonic acid: in other words, the two products of change affect the activity of the enzyme in opposite ways—a result altogether without precedent. In explanation of these results, the suggestion was made that Urease is a feebly acidic substance. Though it was obvious that the results were not to be harmonised with the views that were current as to the manner in which enzymes act, we refrained from comment, deeming it desirable to obtain more information before discussing the new situation that wascreated. In the interval, the behaviour of other enzymes has been under observation by Dr. E. F. Armstrong and ourselves and it is proposed to discuss the general outcome of the work, in a comprehensive communication, at an early date. Meanwhile, we desire to bring forward an account of further observations on Urease carried out with the object of ascertaining the manner in which the activity of the enzyme is affected by the presence of various substances together with the urea. Experimental Method.—In cases in which the substance to be added was easily soluble in water, solutions were prepared containing either one-half or one-tenth of a molecular proportion of the substance per litre. Having measured out 50 cc. of a half-molecular solution of urea into each of two 200 c.c. Jena flasks fitted with indiarubber stoppers, 50 c.c. of water were added to the one and to the other 50 c.c. of the M/2 (or M/10) solution of the substance of which the effect was to be determined; each flask received also 25 c.c. of Soja extract (prepared as described in our former communication) ; all operations were carried out as near as possible at 25°. As soon as the two flasks were charged, they were placed in an incubator which was maintained at 25°. After intervals of 5, 10, 15, 30, 45, 60, 75, 90 and 120 minutes, samples (10 cc.) were withdrawn from each flask by means of pipettes previously warmed to 25°; each sample was run into a * “Studies on Enzyme Action. XV.—Urease: a Selective Enzyme,” ‘ Roy. Soe. Proc.,’ 1912, B, vol. 85, p. 109. Studies on Enzyme Action. 329 200 c.c. Erlenmeyer Jena flask containing a measured volume (an excess) of standardised chlorhydric acid. In all the experiments, the carbon dioxide present was removed by bubbling air through the mixture to which the standard acid had been added, after a few drops of olive oil had been intro- duced in order to prevent the frothing which otherwise occurs. At the end of an hour the excess of standard acid present was determined by titration with standard baryta solution, using litmus as indicator. When dealing with substances of slight solubility (eg. benzaldehyde, methylic salicylate, etc.), 105 c.c. of a solution were prepared having a con- centration of M/4 as regards urea and M/20 as regards the substance to be added; 100 cc. of the liquid taken out with a pipette were introduced into the 200 c.c. Jena flask and treated with 25 ¢.c. of Soja extract as before. In most of the experiments made with the object of studying the action of carbon dioxide on more .concentrated solutions of urea (semi-molecular,. molecular, twice molecular and pentamolecular), in which samples were taken over a considerable period, the quantity of urea solution used was 200 c.c. together with 50 c.c. of Soja extract. Influence of Acid Compounds on the Activity of Urease. Not only strong acids but even the relatively much weaker carboxy- acids prevent the enzyme from acting, if present in appreciable amount. Thus no action took place in solutions of Aspartic and Salicylic acids of M/50 strength. Borie Acid.—This acid has a remarkable depressant action when present. in a solution containing the proportion B,03/50 as shown in Table A (see Graph No. 12).* Phenol.—_In M/25 strength phenol itself has little influence but it appears to be sufficiently “acid” in M/5 strength to exercise a marked retarding effect (Table A, Graphs 1 and 2). The influence of guaiacol and resorcinol is distinct from that of most other substances. At first these compounds retard the rate of hydrolysis but subsequently accelerate it slightly. Apparently, at the outset, they enter into competition with the enzyme and share the urea with it; as ammonia. is liberated, however, they also serve to neutralise this base and therefore promote the change (Graphs 4, 5 and 6). In the presence of quinol, action soon comes to an end (Table B). Only about 2 per cent. of change was effected when the solution was of M/25 * Apparently boric acid is singular in that it retards the action of urease even in very weak solutions; all other acids, if present in sufficiently small amount, accelerate hydrolysis. 390 Prof. H. E. Armstrong and others. (Jan. 1, Table A——Hydrolysis of Urea in M/5 Solutions containing Acidic Substances. Percentage of urea hydrolysed. Time In the | In the | In the In the In the (mins.). presence presence presence presence presence Alone. of Alone. of of of Alone. of phenol phenol resorcinol resorcinol guaiacol M/25. M/5 M/25. M/5. M/25. 5 10°8 9 °2 SES ol 9°4 5) 6°5 11°8 9-2 10 INy/ °¢ 15°5 15°8 | 11:9 16 °2 3 12°6 171 15°3 15 23°8 21 °2 20°6 | 15°9 22 4 8 176 22-7 20 °6 30 37°4 | 35°6 33°9 | 26°5 37 °7 8) 30 °9 36 0 34°8 45 49°5 | 47°8 44°9 | 35°3 50 “6 2 426 47-4 46 -2 60 60°8 | 58°4 54°2 | 43 °4 61-2 a) 53 °3 56 °7 56 9 75 69°7 | -68°8 63°5 | 50°7 71 °4 a) 63 °2 65 °8 66 °4 90 78:2 | 78-2 ak ay |) Be Cr 81°0 °3 721 742 751 120 94°4 | 94-4 86°0 | 71:2 956 “6 89 °9 88 °2 90°1 In the |} In the In the In the In the presence presence presence presence presence Alone. | of boric || Alone. of of of | Alone. of acid glycine | glycine glycine asparagine M/25. M/s. | 2M/5. 4M/5. M/5. 5 or 30 sts) iatoat 9°8 12°0 10 ‘0 10°8 8-3 88 10 16°5 6 °4 a) iol 16°5 21 °4 15°5 21:0 13 °4 14 °6 15 21°3 8°8 5 27-8 | 21°6 29-0 21°0 29 °4, 17-2 22 6 30 84°3 15 °6 2 46 °5 33 °9 48-1 || 33°4 49 °4. 27 :°7 39 °2 45 449 20 °9 “A; 61 ‘1 447 63 “7 44 °7 65°5 36 2 528 60 53°6 | 25°8 5} || 7/83 25 | 53°0 75°9 55 5 78 °7 43-9 64:°5 75 63°7 | 3071 2°2 | 85:0 64 °3 87 °7 63 °2 91 °4 52 °2 746 90 71°5 350 a2) 94-4; 720 96 °5 | 72:4 101 0 57-4 83 *4 120 86°3 | 41°8 2 | 99°7 87 °3 98 9 | 85:5] 101-4 69 8 irl Table B.—Hydrolysis of Urea in M/5 Solutions containing Quinol and Monomethylquinol. Time (minutes). Alone. 5 10 *4 10 17:0 15 225 30 85 °6 45 46°4 60 56:0 75 64-7 90 72-1 120 85:0 In presence of quinol M/500. COO Pb BA BB co DONSONNNHND OIBHAP wow FONDHAN AO ORDAIWAK SSO Percentage of urea hydrolysed. In presence In presence of of quinol Alone. monomethyl- M/25,. quinol M/25. 2°3 9°0 56 2-1 14°3 8:1 1:9 18°3 8-7 16 30°3 9-5 1°6 39-4 10:1 — 48 -2 10°9 1:9 55 6 11°8 1°4 62 °8 12°3 0-9 | 75-6 13°3 331 Studies on Enzyme. Action. 1913.] ‘g ydeay ‘g ydeay p ydeay ool 09 09 O2 (0) ba SILANIW NI AWIL J. 5. ry = 3 lov (@) an (=) mm > =x = oOo Sloe 3 w 0 (0} a SaeAe, TONINDTAHLAWONOW 40 3ONaS3ud NI y, ° ¥ G3aSAIONGAH V3"N JO SDVLN3ADa o a "¢ ydeuy G ydeay *T ydeay ‘ST ydeay ‘IT ydery ‘OT ydeay SALANIW NI AWIL (Jan, 1, GASAIOUGAH W38N JO JOVLNADY Prof. H. E. Armstrong and others. 332 6 ydeiy —— "8 ydery *y yea 1913. ] Studies on Enzyme Action. 333 strength and only about 4 per cent. when it was reduced to M/500; the solution rapidly darkened in colour. As no action takes place in presence of quinone (M/50), there can be little doubt that the effect produced by quinol is dependent on the production of this compound directly ammonia is present in sufficient amount to condition the oxidation of the quinol. A series of experiments were made with the monomethylic-derivative of quinol, CsH,(OH)* OCH:, a compound of some interest, as it is formed together with quinol when arbutin is hydrolysed by emulsin. The material used was that supplied by Kahlbaum. We were inclined at first to attribute its imhibitive power to the presence of quinol; we therefore purified it by distilling it 7m vacuo and made use of the inter- mediate fraction. The results given in Table B are those obtained with this product. We then digested the compound with ferric chloride, with the object of oxidising any quinol that might be present; after treatment with a little sulphite, to remove quinone, the residue was distilled in vacuo. As the substance thus purified was as active as the original material, we are inclined to think that in presence of ammonia and air monomethylated quinol is slowly converted into quinone and that this is the reason why it is so active an inhibitant (Graph 3). Glycine and Asparagine—These substances accelerate the rate of change as shown in Table A (Diagram 13). ; Though they are “neutral” compounds, they neutralise both acids and bases; their marked accelerative effect is probably due to the fact that they serve to neutralise the ammonia as it is produced by the hydrolysis of the urea. As the positive influence of glycine is no greater apparently in more concentrated solutions, it is not improbable that it acts in two directions, both serving to fix ammonia and combining also to some extent with the enzyme. Carbonic Acid.—A further series of observations carried out in presence of carbonic acid is given in Table C. The experiments were made in the manner already described (XV, p. 121). The results are represented by graphs in Diagram 15. The four graphs in Diagram 14 are drawn from data given in XV, Part I. They represent comparable results obtained in experiments carried out simul- taneously with the same sample of enzyme. It will be noticed that whilst the products of change taken together have but little influence, taken singly they are relatively very active but in opposite directions. The set of graphs marked ¢c (Diagram 15) show that when the proportion of urea is varied the difference observed in the absence of carbonic acid (XV, p. 117) is again apparent, the amount of change taking place in solutions * VOL. LXXXVI.—B. 2B 304 100 80 60 40 80 60 40 Prof. H. E. Armstrong and others. Dracram 13. PERCENTAGE OF UREA HYDROLYSED TIME IN MINUTES hea fo) 4 , 60 80 Studies on Enzyme Action. 335 1913.] ‘asno SI} Ur pash svi ourdzue YONUI sb oOTAT, 4 ‘quouittedxe styg ut quesord sv plow otuogituo ON | | 6. ese | 9- L8T 0 | & 6F% | &- 002 6. 801 OT Cis F- 008 = 6 | 6 86T -- | 9-461 = = G.8 | 2+ 861 4. 81 1-661 || &-86T &- L61 L. &6 8 a, — — | p. LST = ee 9-1 T- IST — 0-28T || T-18T &. PST 0: 8 4 9. 00% T- 991 §- IST Gg. TAT F- POL 0: 69T 8: PL, 9 9-002 | = 1. 98ST 6: 2ST. G. OFT F- 66 &- IST 1.9 g 8- 961 = _ — = - 66 — — go.) €- SST 6: 821 0: Z&T 8: T&T. 6: PCT Z- 66 = g. gg v = os = — a= F- 66 — — 9.8 Caco |e eon 1. FOT 8: LOT z: 66 8: 26 $- OOT 8: GOT 8. PF g — | — — — — 8: 64 = = — 9.% T- 91 GGL 4-92 0: 08 9. OL F- 99 — Piel, Bice smmoy g 2. 26 1.83 8: 6S g. 89 P49 I. FS T- &9 G. 04 & 9 & LZ 06 1. $9 Ieee CR G. 68 G. bP 0: &F L. OF G. 88 G. 2 0. OF T. 68 6: 02 : 09 = | = oan = = = ao = T- OF — = 0g F. SF 9. 82 9. 1% g. 08 G. 6% G- 8% = 6. FE 0: 82 G. ST OF 8. €8 9. 2% G. 12 8: 83 &. 2S Ms Its GB F 66 L. 82 Ny MO 0- 8 08 0: FZ 6: ST 8- FI 9. OT G. ST 0- ST G: ST &- 0% P- 0% &. OF 8:6 0z F. 61 0- I &- 21 GT L- BI 6: IT = = @- ST ¥- SI 1.8 ST L- BT 1-6 6-6 1-6 1-6 T-8 0-8 | 8. TE es IU 0: OT 0-9 OL &-6 G.g 8. ¢ G.g ag 6: F gg | 6.¢ 7.9 8-¢ 6.8 “SUIT G “WOT} “UOT? “mu0T4 “mory “u0ry “uory eco) U1 “u01ly “uOTy “Ory “mOry -R.IJU90U00 -B1}U9DTO9 -B.1J 090000 -eBt}U90T09 -BIZU90TLOD -BIZUeDTO0 -BIQUd0TO00 -B.1QU90T00 -RIJUdDTOD -G1j,U90TLOO “BIZUBDUOD aN W Wz W W /N WS W o/N W W ur vot t eas EEN) ut B20) ur essa UL Bop) ul Oa ul eri UL VOLE) UT BOL) Ul BAIQ, | UL VAI y ‘QUILT, ‘TI OT 6 8 Vi 9 g p g G “il : ‘gjdutes ‘o'0 QO] Aq post[vaqnou proe [eMAoUTOEp Jo oumNOA ‘OpIXOl] WOGAeD YAIA\ poyernges suornjog ut vetq jo stsAyoapAH— OTqey, Prof. H. E. Armstrong and others. 336 AY TAA fe ATdWVS 2201 Ad G3SIIVULNAN GIDV IWNYONIDAG 40 (2°) ANNIOA “Gl WVHSVIG p) q D Ov 08 02 09 002 V9 1S..| Studies on Enzyme Action. 337 of M/5 and M strength being almost the same, less change taking place in a solution of 5M strength. When change was complete in the weakest solution, only 1/5 of the urea in the solution of intermediate strength was hydrolysed and about 1/23 of that in the strongest solution. To ascertain the optimum strength of solution, a comparative experiment was made with solutions of M/2 and M strength (the two graphs marked 6). Again, the solution of molecular strength was found to be slightly the more active. On contrasting the behaviour of solutions of molecular and twice molecular strength (the two graphs marked d), it was found that the change took place at very nearly the same rate in each, being slightly more rapid in the weaker during more than half the period of change. Two experiments were made with solutions of molecular strength, twice the usual amount of enzyme being added to the one (the two graphs marked a) these gave results showing that the use of the larger proportion of enzyme is attended with a slight advantage. The striking fact brought out in all the graphs representing experiments made in presence of carbonic acid is the approximation of the rate of change to a “linear ” character. To secure a more rigid comparison, smooth curves were drawn carefully to a large scale from the data obtained in the experiments and the rates of change were deduced by finding the value of the;tangent at each of a series of points. The results are given in Table D. It will be noticed that the influence of the acid increases as the action proceeds and that the rates are not far from being constant over con- dx] dt siderable intervals. The values of the ratio = in no way correspond to those to be expected in the case of a change proceeding at unimolecular rate, which is commonly regarded as the rate to which such actions tend to approximate. Hydrogen Cyanide.—In view of the fact that hydrogen cyanide is a product of the hydrolysis of a considerable number of glucosides by “emulsin” and other enzymes, as well as on account of its remarkable physiological activity, it appeared to us to be important to study its behaviour towards an enzyme with which, presumably, it is not ordinarily brought into relation- ship. Our anticipation that it would act merely as a very weak acid and accelerate hydrolysis was proved to be correct. The results of a series of experiments with various strengths of the cyanide are given in Table E and as graphs in Diagram 16. It will be seen that the accelerative influence increases with the [Jan. 1, Prof. H. E. Armstrong and others. 338 8s0- T O2P- & 186-0 LE&- 0 £00: FT 6&9: 0 GLL. 1 916. T TOP: 0 O€Z. I 60F- 0 262. I CPP -T 8SP- 0 LS¥- TL 109. @ S1é- 1 108: T O16: 0 1G&- PT 196: 0 €L0- T L89- T GTS. 0 PSI- OL F00- T &1S-. 0 162. % LL1- 1 O&Z. GT 0ZE- 0 G60: O 86. T 09g. 0 116.8 €1L-0 898-0 ZOL-& 9GP- T 61. IT 882. 0 80: 0 E11. 09: 0 869. 4 GLS- O GIE- O GII-€ 90L-T LFS. 6 OLZ. 0 810: 0 668. & 899. 0 G96. 9 419-0 08%. 0 L¥9- € 16. T 926: 4 &aa- 0 PL0-0 68F- & Zéh- 0 260: & LP. 0 926. 0 980. 7 G13. % cog. 9 922: 0 TLO- 0 LES. & S6L- 0 618. € Scr. 0 TG. 0 USE. P OLP- & PLL. V S0z- 0 OL0- 0 8SS-@ G18: Q OPS. 6 OLE- 0 1&3. 0 6LE- F O8L- % ZS. & 161-0 80. 0 919. & ILT- T 26. T O&€€-. 0 G1é- 0 9EP- P 198.3 16: T OSZ- 0 SLI. 0 68S. € 629. T G9E9- O ‘amqit cod ‘ougty aod -oaqty aod ‘omgry cod : . emXzu9 00 OOP | eutdzue ‘o's QOg | eurkzue ‘oo QoF | outdzua ‘o'0 QOZ eae eas *(snoonbr) ae an *(snoonbr) renee) PEED THRHA|O) TNR EEHEC|O | poyemus) | wornyjos poyeanyes) | worntos (x) TOTANTOS val TW WOT}N OS Bat W © ‘OPIXOIp woq.reo ‘eprxorp woq.reo Pasay Berl) W wart) TW ae YIM pozVinjzes MOLN]OS VoL) TAL YGIA pozBinyes WOTYNOS vat TA Snes | i= =. E sou) 7. : | aa (p/P) Feige ame TOA = | roat wacn Sp soattned Fciiueen Eo eee ! 4) quororyooo A4r0070 A BoM FO § 0) ‘SjMsIoyjeon Aq100JeA pue very) Jo sisdforpAF Jo soyvy—'( Iq", 19138. Studies on Enzyme Action. 339 Table E.—Hydrolysis of Urea in M/5 Solutions in presence of Hydrogen Cyanide. Percentage of urea hydrolysed. Time | | (minutes). In presence of | In presence of | | In presence of | Alone. hydrogen hydrogen Alone. hydrogen eyanide M/25. | cyanide M/65. | cyanide, M. | | | 5 8°8 10°9 126 Ot 17°0 10 15 6 18 °5 21°9 16°5 31 °4 15 20 *4 24 ‘0 —_ 21°3 42-2 20 = == 36 °3 = 53°3 30 32-9 38 “6 49 6 Pe teaee 69-1 40 as Ms a | = 81°5 | 45 43 *4, 49 ‘8 63 °1 | 44-9 = 50 = = —_ = 89-7 | 60 52 °2 59 °7 75 4 53 6 94 °3 | 75 614 69 °2 86 °2 63 °7 98 8 90 68 °7 776 94-8 71°5 — | 120 83 ‘1 91°6 100 ‘0 86:3 —_ | (=) ui a) a o a4 (=) > <= < uu a4 =) ju e) uJ < = r4 Ld 1S) o 1 LJ oO 0 20 40 60 80 1090 120 concentration and becomes very considerable in solutions of molecular strength, 50 per cent. of the urea in such a solution being hydrolysed after an 340 Prof. H. E. Armstrong and others. [Jan. 1, interval of about 18 minutes, whilst in the absence of the cyanide this amount is changed only after 55 minutes. In this case it was noted that the solution became brown. It is noteworthy that hydrogen cyanide has relatively less influence than carbonic acid in the later stages of the change. Influence of Methylurea—In Part XV it is shown that methylurea has a definite retarding effect. We have therefore carried out a further series of experiments in which, in one case, only the amount of methylurea present was varied, whilst in the other the action took place in presence of carbonic acid. The results obtained with this substance when used in presence of carbonic acid are of special interest in view of the close relationship of urea and methylurea; they are given in Table F and in Diagram 17. It will be noticed that the addition to the M/®5 solution of urea of an equivalent amount of methylurea has a marked depressant effect and that when the urea and the methylurea are present in the ratio 1:4 the effect of the neutral substance is considerable. As practi- eally the same alteration in osmotic conditions would be produced by equivalent proportions of urea and methylurea, it is to be supposed that the influence exercised by methylurea is due in part to the fact that it shares the acid enzyme with the urea but it also interferes mechanically. The results obtained in presence of carbonic acid are similar to those obtained in its absence but action proceeds at accelerated rates. Table F.—Hydrolysis of Urea in M/5 Solutions in presence of Methylurea. Percentage of urea hydrolysed. Time rane Sak : | I all F (minutes). n presence In presence || In presence | In presence of | In presence o: Alone of of of methylurea methylurea ‘| methylurea | methylurea carbon | (M/d) and (4M/5) and M/5. 4M_/5. dioxide. | carbon dioxide. | carbon dioxide. |. 5 100 Ou 8°3 15 °9 15 °3 TL ¢/ 10 16-7 15 °9 13°4 27°8 25 ‘0 22 °0 15 22-9 21°4 18-0 36 ‘9 35 °4 30°9 20 = = — | 49 -O 46 -O 39 6 30 36 0 34.°3 28 4 66 6 62 °8 55 ‘1 40 = — ila G89 80 °6 70°8 45 AT “9 44, +1 37 °3 — _— = 60 57:0 53 °4 45 °2 | 97 “9 98 “6 94-1 75 66°8 62 +1 53 °0 = ra = 90 74:0 70-0 59°6 99 6 99 8 99 6 120 88 °4 83-9 70°6 — as = 1913. ] Studies on Enzyme Action. 341 Diaaram 17. a iu se —! re) ~ Q > ae < Wu c& co} Lo oO ud < pe z tu 1S) m~ J a O 20 40 60 80 100 120 Influence of Neutral Agents which depress the Activity of Urease. Alcohols.—Ethylic and propylic alcohols exercise moderate effects which may be attributed to the changes they produce in the osmotic conditions. Asin all other cases studied, the less soluble alcohol is the more active (Diagram 18). Saligenin, CsH4(OH):CH,(OH), is far more active than either of the paraffinoid alcohols (Graphs 10 and 11). Aldehydes.—In the presence of formaldehyde (M/25), action comes to an end when about 4 per cent. of change has taken place. Acetic aldehyde, benzoic aldehyde and salicylic aldehyde are moderately active depressants; the results obtained with these substances and with saligenin are given in Table G and in Graphs 7, 8 and 9. The observation that glucose has a slight retarding effect has been confirmed. Sodium salicylate and methylic salicylate have practically no action. We are inclined to think that the aldehydes are all, in some measure, chemically active towards urease and that even saligenin may be credited with slight chemical activity. It has been customary to regard the action of enzymes as subject to the 342 Prof. H. E. Armstrong and others. [Jan. 1, Dracram 18. \ x | | TIME IN MINUTES ) 20 40 60 80 100 120 “law of mass action” and to assume that the rate at which action takes place is such that it is proportional at any moment to the amount of substance left unchanged. It should be possible, therefore, to express the rate of such changes by logarithmic curves but it is recognised that the products of change have a more or less marked retarding influence and that, on this account, the actual curve expressing the rate of change always falls below the theoretical curve. It is further supposed that the action is reversible and that therefore, on this account, the change is never complete, though it may be very nearly so in dilute solutions. Lastly, it is recognised that when the enzyme is present in very small proportion, the action proceeds at a nearly constant rate: also that it is much retarded in very concentrated solutions—a result ascribed by some to the viscosity of such solutions. It appears to us that our results are not in accord with the views hitherto accepted and that it is to be supposed that enzymic changes would be found to take place at approximately constant rates were it not that they are subject directly and indirectly to considerable retardation by the products of change; indeed it is probable that the products of change have an affinity for the 1913. ] Studies on Enzyme Action. 343 Table G.—Hydrolysis of Urea in M/5 Solutions containing Alcohols and Aldehydes. Percentage of urea hydrolysed. : ; 7 - (aaa i a Tsai meaaen ay Gia Plus Plus_ || Plus Plus | | In presence Ya ethylic propylic || ethylic propylic | of enone. alcohol | alcohol || NOE alcohol aleohol | epone: | saligenin 1 mol. 1 mol. || 3 mols. 3 mols. | M/25. ee Bier isda’ 2 Ne ei 5 2-7 29 | 28 4:9 3-7 2:5) ietiez) | a7 9c8 10 70|\ 69 | 66 7-9 8-7 65 | 199) 16% 15 11°2 11°4 10°9 12 ‘4 12°2 8°8 26 287 30 22d 21°65 | 2 @) 23 °2 23°56 179 40°3 | 37-4 A5 32 6 31°8 | 30°8 348 331 26 °2 — | — 60 42-0 41 °5 40 °5 44,°6 42-1 | 34°65 64°3 | == 75 51 °2 50 °4 49 ‘0 53 °3 509) | 4-9 744 = 90 60 *2 570 | 573 61°8 58°9 49 6 84:0 80°6 120 75 °2 73°3 73:0 78 1 74:3 62 °7 97:4) 95°5 12 hrs.| 98:1 98 °7 98-7 set = = | | | | | | | In presence | In presence In presence ! | In presence | of of acetic of benzoic |) of salicylic Alone. | saligenin, || “1° | aldehyde, | “1M | aldehyde, | “!*) aldehyde, M/5. M/25. | M/25. M/25. | part | | 5 10°4 4°7 | 10°6 8°5 9°4 | 5°4 |} 875 | 56 10 16°7 9:2 | 16:1 13°1 16°0 8:9 | 14°6 9 °4 15 20°5 12°2 21:0 17 °2 20 °5 12°56 | Us) 83 12:0 30 32 °5 19°3 33°5 261 33 °°7 21°2 | 8l1‘l | 20:0 45 41 °5 261 43 °9 34°3 44.°3 27°8 | 41°9 | 26°8 60 50 °9 31°5 53 °6 40 °5 538°9 | 33-7 50°9 | 32:0 75 60 °3 37-7 63 *4 46 *4 62°9 | 39:3 59 °3 38-1 90 66:9 | 42°6 708 513 || 704 | 44:0 || 666] 4371 120 | 80°4 51°8 84 °8 60 2 ecOcOMs manok” | 81:0 52°53 In the experiments with ethylic and propylic alcohols in weight normal solutions, an amount of ammonia equivalent to one-tenth of that ultimately produced was added initially to the urea solutions. enzyme which is actually greater than that which obtains between the hydrolyte and the enzyme. It has often been suggested that the enzymes are colloids. The experi- ments carried out in the course of this series of studies appear to justify the belief that enzymic action takes place entirely at the surfaces of colloid particles suspended in the solution of the hydrolyte and not between substances which are all in true solution. The subject will be more fully discussed in the later communication to which we have referred. [The cost of this investigation has been partially met by a grant for which I am indebted to the Government Grant Fund of the Royal Society.—H.E. A.] 344 A Preliminary Note on the Fossil Plants of the Mount Potts Beds, New Zealand, Collected by Mr. D. G. Lillie, Biologist to Captain Scott's Antarctic Expedition in the “ Terra Nova.” By E. A. NEWELL ArBER, M.A., Sc.D. F.GS. F.LS., Trinity College, Cambridge, University Demonstrator in Paleeobotany. (Communicated by Prof. T. McKenny Hughes, F.R.S. Received February 17,— Read March 6, 1913.) [PLATES 7 AND 8.] In the present communication, I propose to discuss very briefly the first fruits, which have reached this country, of Captain Scott’s Second Antarctic Expedition (1910-13). A full account of the fossil flora in question must be reserved for a future occasion. At present I have only permission to contribute a preliminary note on the subject. It is well known that, during the winter months of the last two years, the “Terra Nova,” the ship of Captain Scott’s Second Antarctic Expedition, has been actively engaged in furthering scientific researches in New Zealand waters, returning, however, to the Antarctic each summer. My friend, Mr. D. G. Lillie, B.A., of St. John’s College, Cambridge, one of the biologists of Captain Scott’s Scientific Staff, who has been attached throughout to the “Terra Nova,’ has been busily occupied with various researches, partly biological and partly geological. During the short periods when he has been free to proceed with geological work, he has set himself the task of trying to clear up some of the doubtful points, which remain unsolved, in regard to the stratigraphical geology of New Zealand, more especially by means of the fossil floras of the rocks in question. As is well known, the precise geological age of many subdivisions of the stratigraphical sequence of these islands remains in doubt, and in some cases these questions have been matters of keen dispute in the past as at the present time. Among them, none has given rise to greater controversy than the doubt which has existed as to the precise geological age of the plant beds of Mount Potts, in Ashburton County, Canterbury. Do these beds contain Glossopteris, and perhaps a typical Permo-Carboniferous Glossopteris flora? Did New Zealand, as one would expect, in Permo-Carboniferous times form part of the great Southern continent, “Gondwanaland,” the home of the Glossopteris flora, like the greater part of Australia, South Africa, and South America ? These are the questions as yet in doubt. If, on the other hand, New Zealand, in Permo-Carboniferous times, formed no part of Gondwanaland, fi isk 00 Fossil Plants of the Mount Potts Beds, New Zealand. 345 we are obviously face to face with a conclusion of the greatest geological importance. This is one of the questions which Mr. Lillie has set himself the task of solving.* Plant-remains in the Mount Potts beds were first discovered by A. McKayt in 1878. The collection was examined by Hector} in the same year, and he stated that it contained examples of Glossopteris and Schizoneuvra, this assertion being repeated in 1886. These conclusions subsequently led to considerable controversy between Hector, Haast, McCoy, and others, the details of which it is unnecessary to enter into here. The whole question has turned on the identification of the fossils, and the evidence as to their stratigraphical position. Further, until Mr. Lillie’s visit, the collections from this region appear to have been small and very fragmentary, and even these had not been examined at any time by European specialists in paleobotany. The fine collections made, in November, 1911, by Mr. Lillie, in conjunction with Mr. R. Speight of Canterbury College,in very wild and difficult country, appear, however, to settle this question once and for all. Glossopteris itself is not present, nor is the flora a typical Glossopteris flora. The most characteristic and striking plant represented is, however, one which resembles Glossopteris in habit. It has the same tongue-shaped, entire frond, with a well-marked midrib, but the lateral nerves, instead of anasto- mosing as in Glossopteris, do not unite with one another.§ (Plate 7, figs. 1 and 4.) One member of this genus has already been described from the Rhetic beds of Chili The specimens in question were termed Lesleya Steinmanm.|| It seems to me very doubtful whether these leaves belong to the Paleozoic genus Lesleya. I should be inclined to place the New Zealand * This paper was written before the news of the sad disaster to the Polar party of the Expedition reached this country. It is, however, only fair to the memory of the late Capt. Scott, whose death I deplore most sincerely, to point out that the work which Mr. Lillie and others have been engaged in, during the winter months in New Zealand, was part and parcel of the scientific intention of his expedition, to be fulfilled during the times when the “Terra Nova” would be useless in the Antarctic, but could be profitably employed in New Zealand waters. + McKay, ‘ Rep. Geol. Explor. Geol. Surv. N.Z.,’ 1878, pp. 91-109. ¢t Hector, ‘Proc. N.Z. Inst.,’ 1878, vol. 10, p. 533, and ‘Cat. N.Z. Court, Indian and Colonial Exhibition, London,’ 1886, p. 77. § As will be seen from the upper part of the photograph on Plate 7, fig. 1, there appear to be at least some anastomoses, but I am convinced that, in this as in other cases, these are false and not real, and that they are due either to the partial removal of the film of carbon, or to the fact that the normal distribution of the nervation had become disturbed just before or during preservation. Such false anastomoses are by no means infrequently met with among fossil impressions. || Solms, ‘ Neues Jahrb.,’ 1899, vol. 12, Beil. Bd., p. 596, Plate XIII, figs. 5-7. 346 Dr. E. A. N. Arber. Fossil Plants of the [Feb. 17, fossils, at any rate, in a new genus Linguifoliwm, and to regard them as a new species Z. Lillieanum, so named in honour of Mr. Lillie. They are certainly specifically distinct from the Chilean plant. This plant may be also compared with the Copiapaea plicatella of Solms* from Chili, and the Blechnoxylon talbragarense of Ktheridget from New South Wales, the latter believed to be a Paleozoic fossil. In addition to Lingusfolium Lillieanwm, a number of other well preserved species occur. There is a new species of Chiropteris (Plate 8, fig. 6), the distal margin of which is lacerated, and which I propose to name C. lacerata sp. nova. Leaves of a species of Bazera (Plate 7, figs. 1, 2, and 3), closely similar to but perhaps distinct from Baiera paucipartita, Nathorst, from the Rheetic of Bjuf, Sweden, are also associated. A Dictyophyllum, which may be closely compared with Dictyophyllum acutilobum (Braun), is present. Other fronds are those of Thinnfeldia lancifolia (Morris) (Plate 8, fig. 7) and Cladophlebis australis (Morris). Numerous examples of a Z'eniopteris, which is no doubt 7. Daintree’, McCoy, occur (Plate 8, fig. 5). Among the Equisetaceous remains, pith-casts are represented which resemble those of Phyllotheca or Schizonewra, but, in the absence of foliage, it is impossible to refer them to the one genus rather than the other. However, the small detached leaf-sheaths of a Phyllotheca are undoubtedly present. Finally, associated with the above plants, are many examples of the Indian (Gondwana) Palissya conferta (Oldh.) (Plate 8, fig. 5). From this brief review of this interesting flora it is obvious that it is of Mesozoic age, and belongs either to the late Triassic (Rhetic) or to the early Jurassic period. Linguifoliwm, which is a homeeomorph of Gilossopteris, just as Lonchopteris is of Alethopteris, or Dictyozamites of Zamites, is already known from Rhetic rocks in Chili. Chiropteris is at present confined to the Rheetic, some species occurring in the Triassic rocks of Europe, and also, as Prof. Seward has shown, in the Stormberg Series (Rheetic) of South Africa. Baiera pauci- partita, Nath., and Dictyophyllum acutilobum (Braun) occur in the Rhetic of Europe. Zhinnfeldia lancifolia (Morris) is found chiefly in the Rhetic, though it no doubt also occurs in the Jurassic. Cladophlebis australis (Morris) is known both from the Jurassic and Rhetic in the Southern Hemisphere. The only two plants which, at present, are exclusively Jurassic are the Gondwana (Rajmahal) Palissya confertat (Oldb.), and Toeniopteris Daintreer, * Solms, zbid., 1899, p. 594, Plate XIII, figs. 8-11. t Etheridge, ‘Rec. Austral. Mus.,’ 1899, vol. 3, p. 135, Plates XXIV-X XVII. { It is interesting to find that Dr. Halle has just described this plant from Graham Land in the Antarctic. (‘ Wissen. Ergebn. Schwed. Stidpolar-Exped., 1901-1903,’ 1913, vol. 3, Part 14, p. 86, Plate VIII, figs. 26-40.) Roy. Soc. Proc., B, vol. 86, Plate 7. Arber. Photos by W. Tams. FOSSIL PLANTS FROM NEW ZEALAND. Roy. Soc. Proc., B. vol. 86, Arber. Photos by W. Tams. FOSSIL PLANTS FROM NEW ZEALAND. 1913. | Mount Potts Beds, New Zealand. 347 McCoy, which, in Australia, is essentially a Jurassic type, though perhaps it may also occur in the kheetic. From this rapid survey of the Mount Potts flora we see that, while its affinities are essentially Rheetic, a few Jurassic types also occur, and thus the age of the beds may be either Rheetic or Lower Jurassic. At present we are unable to distinguish clearly between a Rhetic and a Lower Oolite flora, so this point need not be laboured here. There is little doubt that the Mount Potts beds are, geologically, the oldest, plant-bearing series as yet discovered in New Zealand, and as we have seen they are of Rheeto-Jurassic age. Paleozoic sediments with marine invertebrates undoubtedly occur in the islands, but so far there is no evidence of any land floras of older age than the Rheetic. Not only is Glossopteris unknown from New Zealand, but no land plants of Palaeozoic age of any description have ever been found there. There is thus no evidence that New Zealand ever formed part of Gondwanaland, and this is a conclusion of great theoretical interest. EXPLANATION OF THE PLATES. (All the photographs are by Mr. W. Tams, Cambridge. Nearly all the figures are enlarged. PLATE 7. Fig. 1.—Linguifolium Lillieanwm gen. et spec. nova. On the left, an almost entire leaf, the apex being wanting however. On the right Baiera sp., the base of a leaf. Enlarged x3. Fig. 2.—Baiera cf. Baiera paucipartita Nath. An almost complete leaf. Enlarged x3. Fig. 3.— Baiera cf. Baiera paucipartita Nath. A median portion of a leaf. Nat. size. Fig. 4.—Linguifolium Lillieanum gen. et spec. nova. Two fragments of leaves, one nearly apical, the other median, showing the nervation clearly. Enlarged x3. PLATE 8. Fig. 5.—Teniopteris Daintreei (McCoy) above, and Palissya conferta (Oldh.) below. Enlarged x3. Fig. 6.—Ohiropteris lacerata sp. nova. A nearly complete leaf, showing the incised apex. Enlarged X32. Fig. 7.—Thinnfeldia lancifolia (Morris). Nat. size. 348 On the Nature of the Toxic Action of Electric Discharge wpon Bacillus coli communis. By J. H. Prigstiey and R. C. Knicur. (Communicated by J. Bretland Farmer, F.R.S. Received February 13,—Read April 10, 1913.) Introduction. In a recent paper, Thornton* has drawn attention to some results he had obtained in experiments upon the bactericidal action of electric discharge. Plates of agar were infected with bacteria of various species, and subjected, under different conditions, to the discharge from an electrified point. The plates of agar were subsequently incubated and observations taken of the development of colonies from the surviving bacteria. From experiments upon these lines he concluded that the ionised air, ze. the small current (the whole of the current passing from the point was about 4 micro-ampéres) produced by his discharge methods, proved fatal after longer or shorter periods to all the species of bacteria subjected to it. This conclusion is of considerable interest, suggesting, as it does, the possibility of electrical treatment of tissue attacked by pathological bacteria, with a view to retarding bacterial action. Our attention was attracted to this paper by the fact that its conclusions seem at variance with some conclusions previously arrived at by one of us in conjunction with Miss E. M. Lee, in an investigation carried out at the University of Bristol, of which only a brief preliminary note has so far been published,t pending the further experiments which Miss Lee hopes to be able to carry out. In this research cultures of the sour-milk bacillus, B. Bulgaricus, were subjected to small electric currents, and observations were made to determine the effect of such treatment upon their vitality. Contrary to Thornton’s experience it was found that current densities below about 58 micro-ampéres per square centimetre served to increase both the fermentation power of the bacteria as determined by electrical conductivity, and also the rate of growth as determined by countings. The fact that the current density required to produce any inhibitory effects in these experiments had to be greater than about 60 micro-ampéres per square centimetre may have been due to the fact that in these cases the electric current was derived from * “Ynfluence of Ionised Air on Bacteria,” ‘ Roy. Soc. Proc.,’ 1911, B, vol. 84, p. 280. + “The Influence of Electricity on Micro-organisms,” J. H. Priestley and EH. M. Lee, ‘Brit. Assoc. Report,’ 1911, p. 603. Toxic Action of Electric Discharge upon B. colt communis. 349 a source of comparatively low voltage, and transmitted to the nutrient medium through the ordinary form of Kohlrausch platinum electrode which was immersed in it. But this suggestion immediately raises the question as to whether the effect detected by Thornton bore any relation to the direct action of the current, or was connected with the chemical changes produced in the atmosphere surrounding the discharge point. Thornton considers that the fatal result of the discharge may be wholly attributed to “the direct influence of, and contact with, ions in the electric wind.” It is hardly conceivable, however, that mere ionic bombardment could be responsible for such deep-seated action as was observed, especially in con- sideration of the fact that ions have practically no penetrating power in the presence of water, a film of which must have always intervened between the organism and the discharge. Foulerton and Kellas,* as the result of experiments carried out along lines similar to those described by Thornton, employing in many cases the same species of bacteria, had previously arrived at the conclusion that electric discharge itself was not deleterious to the organisms. They found that “emulsions” of bacteria in water became sterile after subjection to the discharge in air and in various artificial atmospheres, but considered that the fatal effect was due, not to the current, but to the products of the discharge, viz., nitric and nitrous acids in air and hydrogen peroxide in hydrogen. Qualitative and quantitative tests of distilled water, after sub- jection to the discharge, revealed the fact that these substances were indeed present in measurable quantities, and subsequent trials showed that such concentrations of them were fatal to bacteria, independent of the discharge. It is possible that the results obtained by Foulerton and Kellas cannot be directly applied to explain Thornton’s experiments, because of the different electrical conditions. In their experiments the bacteria were contained in water in a test-tube and the current was discharged from the points of a platinum brush suspended over the surface, earth connection being made through a platinum wire sealed into the bottom of the tube. In all cases the high-tension discharge from the brush of platinum points was oscillatory in character, and it might therefore be expected that any effects produced by the action of the discharge upon the atmosphere would be enhanced, while effects due to direct action of an electric current should be far less apparent. The results obtained by Thornton with the apparatus depicted in his fig. 2 suggest that the products of discharge, and not the ions, were the active factor. In these experiments the current passed, not through the bacteria- * “Action on Bacteria of Electrical Discharges,” ‘Roy. Soc. Proc.,’ 1906, B, vol. 78, p. 60. VOL. LXXXVI.—B. 2¢ 350 Messrs. J. H. Priestley and R. C. Knight. Toxic [Feb. 18, infected plate to earth, but through the air between two metal points above the culture, and in this case, where the current through the bacteria was a minimum, “the (sterilising) action was much stronger than in the first arrangement,” in which one point discharged directly on to the culture. The investigation, of which an account is presented below, was therefore commenced with the intention of attempting to ascertain whether current densities of the order used by Thornton, obtained from a high-tension source, could still prove toxic when the influence of all toxic substances produced by the chemical action of the discharge had been eliminated. Experimental. Bacillus coli communis, being found by Thornton to be one of the least sensitive organisms he employed, was chosen for the experiment. The high- tension discharge was obtained from the ordinary 100-volt direct-current circuit by leading this current after interruption by a mercury break through the primary of a large induction coil. The alternating discharge from the secondary was then obtained as a continuous positive and negative charge at either side of a spark gap by leading the alternating discharge through five Lodge valves arranged in series; these valves act as rectifiers, only permitting the current to pass in one direction owing to the structure of the electrodes. This apparatus, which was purchased from a special research grant obtained from the Board of Agriculture and Fisheries, was available during the intervals when not required for other experiments in progress in the Department. By this method it was then easy to maintain as long as required a difference of potential of some 70,000 volts between the poles of the spark gap, one pole was then connected to earth and the other to the discharging point. The current passing from the discharge point was measured by placing a plate of tinfoil of definite area beneath the discharge point, and connecting this by a carefully insulated wire, shielded by an outer metal tube connected to earth, to a Paul micro-ammeter, which was carefully screened hy an earthed metal cover. In this way it was ascertained that the current density of the discharge to which the bacteria were subjected was of the order of from 107° to 107° amperes per square centimetre. The method of treatment of the bacteria was aimee! identical with that employed by Thornton, viz., a Petri dish containing a sterilised agar medium infected with the bacillus was supported on a small metal tripod, which itself stood on an earth-connected metal plate. Dish and tripod were then covered with a small bell-jar fitted with a rubber stopper, through which passed a glass tube, open at the upper end, and with a platinum wire sealed into the - 1913.] Action of Electric Discharge upon B. coli communis. 351 lower. By means of mercury, connection was made between this wire and the cable from the discharge set, and so the platinum point discharged down- wards towards the Petri dish and metal plate. 1. Discharge in Air.—A repetition of the original experiments seemed first desirable, and accordingly cultures were exposed as described above. In 30 minutes the plates were almost cleared, subsequent incubation producing only a few colonies around the edge. In one case one side of the bell-jar was inadvertently wet, and instead of a continuous discharge, intermittent sparking down that side to the metal plate took place. The Petri dish after- wards showed sterilisation over about one-third of its area, and that on one side only. Exposure of 40 minutes or more always resulted in complete sterilisa- tion of heavily infected dishes. Since, then, the discharge in air was definitely deleterious to the organisms, and as it did not seem likely that ionisation effects could be the cause, the réle played by the products of discharge needed investigation. These would be chiefly ozone and nitrous and nitric acids, which would be carried well on to the infected surface by the electric wind, thus providing every facility for their absorption. To test the action of these products, apart from the direct action of the discharge, a Petri dish of distilled water was exposed to the discharge under conditions identical to those’ obtaining in the original experiments with infected agar. The liquid was then removed and heavily infected, plate subcultures being made from it after an hour, In no case was there any development in these subcultures upon incubation, even if the water, previous to infection, were exposed to the discharge for only 20 minutes, thus confirming the idea that the products of discharge alone proved fatal. Particular investigation of these products was now carried out by means of qualitative and quantitative tests of the distilled water after exposure. Abundance of NOz: radical was present, as shown by the diphenylamine test, whilst addition of starch and potassium iodide solutions produced a deep blue coloration, due to nitrite or ozone, or both, also the liquid gave a distinctly acid reaction, the acidity being measured in a few cases -by titration :— No. of Acidity as grammes of nitric acid experiment. per c.c. per hour. 1 0:00034 0:00072 0:00070 000053 0:00067 0:00060 - . 000054. ID OP w ro 202 352 Messrs. J. H. Priestley and R. C. Knight. Towic [Feb. 13, The following experiments show that solutions of about this strength of acidity, and containing nothing but nitric and nitrous acids, are capable of destroying the bacteria. A solution containing approximately 0-002 grm. of nitric acid and 0:003 grm. potassium nitrate per cubic centimetre: was made up and its acidity determined. From it were made solutions corre- sponding respectively to No. Grammes of nitric acid per c.c. 1 0:0007 2 0:0003 3 0:0005 4 0:0003 Each was then infected with the bacillus, well shaken, and after an hour subcultures were made. All the plates proved sterile upon incubation, indicating the failure of the organisms to exist in such solutions. Controls with untreated distilled water were carried out simultaneously with the above experiments, and plates infected from these showed luxurious growth. Thornton, in his paper, intimates that the criticism had been made that hydrogen peroxide might be responsible for the sterilising action of the discharge, but on discharging on to a test solution of potassium titanium sulphate, which detects minute quantities of hydrogen peroxide by the formation of yellow titanium peroxide, we found that no measurable amount of that compound was formed. Indeed, it seems unlikely that hydrogen peroxide would exist in the presence of excess of ozone, the two tending to interact with mutual reduction :— H202+ Oz = H,0 + 20r. 2. Discharge in Hydrogen.—It is chiefly upon his experiments in hydrogen that Thornton bases his conclusions as to the direct instrumentality of the current in the bactericidal action. These experiments have therefore been carefully repeated, using the same form of electrode by means of which a continuous stream of hydrogen was caused to enter the discharge vessel by sweeping past the discharging point. Pains were taken to obtain the hydrogen in a comparatively pure state, since discharge in the unpurified gas resulted in the formation of a film of metallic appearance, possibly arsenic, on the object discharged upon. Therefore all hydrogen, after leaving the cylinder containing it under pressure and before being used, was passed slowly through three U-tubes containing respectively soda-lime, silver nitrate crystals and lumps of a mixture of lime and mercuric chloride, and, finally, through wash-bottles of strong sulphuric acid and potassium pyrogallate solution. In the latter, solutions of caustic potash and pyrogallic 1913.] Action of Electric Discharge upon B, coli communis. 353; acid were mixed, after all air had been replaced by hydrogen, by means of repeated exhaustings with a Geryk pump and refilling. When the potash and pyrogallol were allowed to mix they constituted both a test and an absorptive agent for oxygen, very little of which was present, judging from the very faint coloration of the solution. After this treatment the hydrogen was led to the bell-jar, which for these trials was fitted with an exit delivery tube. This exit tube was connected through two more wash-bottles, the first being another test of pyrogallate, and the second merely to prevent diffusion of air back into the first. The bell- jar was rendered air-tight by sealing it to the metal plate with a stiff wax. The apparatus was then exhausted by means of a Geryk pump and slowly re-filled with purified hydrogen, this being done three times, after which the bell-jar was found to be free from oxygen. The discharge was now switched on, a stream of hydrogen being kept continually passing through the apparatus during any exposure. : Continuous discharge upon infected agar, for periods varying from 30 minutes to 2 hours, failed to produce any toxic effect, the colonies developing as quickly and in as great a number after exposure as normally. This result is the reverse of that obtained by previous investigators, Thornton; and also Foulerton and Kellas, having stated that the discharge proved fatal in hydrogen as well as in air, though the latter give no indication that any attempt was made to exclude oxygen completely. They attribute their result to the formation of hydrogen peroxide, which, by quantitative tests and subsequent trials with definite concentrations, they show to be produced in quantity sufficient to destroy the bacteria. Thornton, on the other hand, assumes at the outset that no hydrogen peroxide was formed in his experiments, but makes no statement as to any test employed to detect it. It may have been that the compound was indeed formed, and that it was responsible for the sterilisation. Such a state of affairs is probable if the hydrogen atmosphere contain small quantities of oxygen, as was shown by some experiments of ours with such mixed atmospheres. Infected plates exposed to the discharge for 40 minutes, under such conditions, show after incubation a small clear space immediately beneath the discharging point, but the effect never approaches that obtained in air. Quantitative determinations, made by a series of comparative colour tests with the above-mentioned titanium solution, disclosed the fact that the presence of oxygen induced the formation of hydrogen peroxide in varying quantities. 354 Toxic Action of Electric Discharge upon B. coli communis. _. Total amount of H,O, -. Total amount of H,O, - No. per hour. No. _ perhour. | 1 0:000197 grm. 8 0:000099 grm. 2 0:000085 ,, 9 0000114 ,, 3 0:000051 ,, Sept 0:000029_,, 4 0:000019 _,, ' 11 0:000174 ,, 5 0:000046 _,, 12) | 0000677. 6 0000112 ,, 13 - 07001459 ,, 9 0:000195 _,, ! : » The variation is probably due to the different amounts of oxygen present, and although quantitative determinations of the relative proportions were not made, the amounts of oxygen present it in Nos. 12 and 13 were certainly larger than in the other cases. Herein, then, may lie the explanation of the disarcoanen alladed to above, the unsuspected presence of small quantities of oxygen serving to produce hydrogen peroxide in amount sufficient to cause the death of the bacteria. From the foregoing results we are led to conclude that in the destructive action of the discharge upon. bacteria, the current itself plays no part, but that the gaseous products of such a discharge in air are the actively toxie agents, causing the death of the organisms, independently of the current.. Summary. 1i.; | Electric: ischial in air is fatal to bacteria, exposed to its pchon ae 2. The effect is due to the products of the interaction of the constituents. of the air, namely nitric and nitrous acid and ozone. 3. Discharge in air-free hydrogen has no deleterious effect on fhe organisms, but the presence of small quantities of air allows the formation of a toxic substance, probably hydrogen peroxide, which ai exerts, a pects) action. ile 4, It,, therefore, ‘ollona that electric discharges in. which the ee density does, not exceed 10—° ampere per square centimetre do not exert any directly toxic action upon’ micro-organisms, a result which.,is contrary to the statements made by some previous investigators.’ . ie betaine 355 Experiments on the Kidneys of the Frog. (Preliminary Communication. ) By F. A. Bainsripcs, 8. H. Coutins, and J. A. Menzizs. (Communicated by Prof. C. J. Martin, F.R.S. Received March 27,—Read April 24, 1913.) Introduction. As is well known, the glomeruli of the frog’s kidney are supplied with blood only by the renal arteries, whereas the renal tubules have a double supply. On the one hand, they receive blood by way of the renal portal veins; on the other hand, the efferent vessels from the glomeruli open into the capillary network round the tubules. The whole of the tubule receives blood from each of these two sources, so that the capillary network around the tubules can be fully injected either from the renal arteries or from the renal portal veins. Taking advantage of this fact it has been shown by Beddard and one of us (F. A. B.) that after complete occlusion of the glomeruli the tubules, when adequately supplied with oxygen, maintain their normal histological appearance, and may secrete urine. In the present experiments an attempt has been made to determine the function of the glomerulus and to ascertain whether the tubules possess the capacity to absorb water and solids. Methods. 1. Hxperimental.All the experiments were carried out on fully pithed frogs. In the earlier experiments the following method was adopted :— Ligatures were tied round the fore limbs, the heart was exposed, and the right aortic arch tied off. The ventricle and auricles were freely opened, and a cannula connected with a perfusion bottle was tied into the left aortic arch. The arterial perfusion was started at once, and the ventricle and auricles were then excised. This procedure was carried out as quickly as possible after the frog was pithed, and usually took five or six minutes. It is of importance to commence the arterial perfusion at the earliest possible moment. When most of the blood was washed out of the circu- lation the anterior abdominal vein was tied in two places and divided, the hind legs were tied off, and a cannula was placed in the inferior end of the anterior abdominal vein and connected with a perfusion bottle. The fluid leaving the kidneys was collected by means of a cannula placed in the beginning of the vena cava just beyond its origin from the renal veins. Finally, cannule were placed in the ureters. This latter operation was 356 Messrs. Bainbridge, Collins, and Menzies. _[ Mar. 27, much more easily carried out in the male than in the female frog, and the former were almost invariably used. Frequently the mesenteric artery was: also ligatured in order to limit the extent of the perfusion. In the later experiments the aortz were exposed immediately above the kidneys, the right was tied off, and a cannula was placed in the left; the mesenteric artery was ligatured. Cannule were then placed in the vena cava just above the renal veins, the inferior end of the anterior abdominal vein (after tying off the legs) and the ureters. The testes were removed by the cautery. The advantage of this method is that the perfusion is practically confined to the kidneys, and the arterial pressure can be more readily gauged and adjusted than in the earlier experiments. The arterial perfusion through the aorta was made at a pressure of 20—24 cm. of water; the venous perfusion pressure varied from 10 to 14 cm. of water. Solutions Used.—The following perfusing fluids were used: (1) Normal Ringer’s solution (NaCl 0°65 per cent., KCl 0:02 per cent., CaCle 0°03 per cent.) ; (2) Hypotonic or hypertonic solutions of sodium chloride containing also potassium chloride 0:02 per cent. and calcium chloride 0°03 per cent. —these are subsequently termed hypotonic or hypertonic Ringevr’s solution ; (3) Hypotonic or hypertonic Ringer’s solution with the addition of 0-1 or 0:2 per cent. sodium sulphate (anhydrous). The solutions were fully oxygenated and were filtered before being put into the perfusion bottle. Frequently oxygen was also bubbled through the perfusing fluid in the bottle. The perfusion bottles were provided with a Mariotte tube. 2. Physical—The greater part of the work which required analytical determinations of the materials used was carried out by means of the refractometer. The instrument used was of the Pulfrick type without water cooler. To enable the instrument to work with less than one drop of liquid, a small flat bottom tube was placed in the refractometer cup. Between the top of the prism and the bottom of the flat bottom tube there was a thin film of the liquid tested; in the tube was a little water with a thermometer. As neither the Pulfrick angles nor the corresponding indices convey much meaning in the present communication, all the results are returned as having a refractive index equal to a solution of sodium chloride of some special strength. As has been shown before,* the refractive index of solutions is proportional to their concentration. The refractive index of water on the particular instrument used is at 20° C. + 1° = 67° 12:7’ + 0°8’ with a probable error of one determination of +0°65’.. The value of NaCl is 1’ = 0:0443 per cent. NaCl, and the determinations of strength of sodium * B. Walter, ‘Ann. Phys. Chem.,’ vol. 38, p. 107; ‘Journ. Chem. Soc.,’ 1890, A, ‘p. 202. 1913.] Experiments on the Kidneys of the Frog. 357 chloride between 4 and 2 per cent. of pure NaCl showed a probable error of + 0:02 per cent. NaCl in such solutions. The actual figures given in the communication must be considered as having that degree of error. Although other salts, as KCl, CaClo, were used, the amounts taken were too small to introduce any appreciable error on this account. Since the solutions used were very nearly solutions of one single chemical substance, the refractometer readings, like specific gravity, give the molecular concentration of the solution. 3. HistologicaltImmediately after each experiment the kidneys were removed and placed in a fixing solution. This was generally alcohol, but in some cases formalin (10 per cent.) and in others Flemming’s fluid was used. After being hardened the kidneys were embedded in paraffin and a series of sections was taken from the middle of each kidney. When the blood-vessels. had been perfused with a mercuric salt and ammonium sulphide the sections were mounted unstained. In other cases stains were used, generally hematoxylin and eosin. In one case complete serial sections were made of the pair of kidneys. Injections of the blood-vessels were also made as described below and serial sections prepared. | Results, Histological.—The validity and significance of most of the experiments to be described rests upon the proof that the whole of the capillary network round the tubules normally receives blood both from the efferent vessels of the glomeruli (that is, by way of the arterial system) and from the renal portal vein. In order to demonstrate this we have injected a number of kidneys on the one hand from the aorta, after ligature of the renal portal veins, and on the other hand from the renal portal veins after occlusion of the arterial circulation. The injections were made from a perfusion bottle under a pressure approximately equivalent to the normal blood-pressure in the frog’s kidney, namely 20-24 cm. of water for the arterial perfusion and 10-12 cm. of water for the venous perfusion. In all the experiments the venous outflow was unobstructed. The fluids used consisted of (a) Berlin blue and (0) carminate of ammonia in Ringer’s solution. A few double injections were made, carmine by the arteries and Berlin blue by the renal portal vein. One gelatin double injection was also made, but in this case the pressure used was obtained by means of an injection syringe and pressure bottle and was higher than usual. It was found that whether the single injection was made by the artery or by the renal portal vein the whole of the intertubular capillary network appeared to be injected. In the case of the double injections the glomeruli and efferent vessels were filled with the 358 Messrs. Bainbridge, Collins, and Menzies. [Mar. 27, arterial injection fluid, whereas the intertubular capillaries showed, some the arterial fluid, some the venous, and some a mixture of both, We confirmed Beddard’s observation that the perfusion of Berlin blue solutions at a low pressure through the renal portal vein leaves the glomeruli completely uninjected. We found, however, that if the venous perfusion were made under an abnormally high pressure (eg. 35 em. of water) the coloured solution eventually made its way into some, at least, of the glomeruli. It was further shown by arterial perfusion with Berlin blue that the glomeruli will withstand a perfusion pressure of at least 40 cm. of water. Other evidence that the renal portal blood supplies the whole of He tubules was obtained by perfusing 1/10,000 mercuric chloride at 10 cm. pressure for three to five minutes through the renal portal vein, and then perfusing through the vein under the same pressure first saline solution for a few minutes and then a very weak solution of ammonium sulphide in saline solution. The whole of the tubules showed a deposit of mercuric sulphide while the glomeruli remained free from it in the vast majority of cases. It seems clear, therefore, that a poison reaching the kidney by way of the renal portal vein will come in contact with the whole of the tubules and yet leave the glomeruli practically or quite intact. The only risk is that a diffusible poison, if not quickly rendered inert or washed out of the kidney, may gradually reach the glomeruli by direct diffusion. This risk is minimised by having a simultaneous arterial perfusion and by using the poison in a concentration which is just adequate to kill the tubules when, brought into immediate contact with them. Whether the arterial. blood supply alone provides the tubules with a sufficiency of oxygen has yet to be determined by experiments on the living animal. _ It is known, however, that the venous supply alone will maintain their nutrition in the living frog ‘provided that the frog is kept in an atmosphere of oxygen. There can be little doubt that in these experiments, in which the perfusing fluid was fully oxygenated, both the supply of oxygen to the tubules in a venous perfusion and that to the glomeruli in an arterial perfusion were amply sufficient to maintain their vitality so long as the rate of perfusion remained normal. Experimental.—In. most of the experiments to be described the pencen was made with normal or hypotonic Ringer’s solution, and the experiments made with hypertonic Ringer’s solution will only be referred to incidentally. (1) The Normal Kidney—tThe rate of arterial perfusion varies considerably in different experiments, doubtless as a result of the varying tone of the glomerular vessels, and more particularly the efferent vessels; it is apt also ‘to decrease in the course of a single experiment. Since the oxygen supply 1913.] Experiments on the Kidneys of the Frog. 359 to the kidneys varies directly with the perfusion rate, a slow perfusion leads to an inadequate oxygen supply to the glomeruli or tubules. The amount of fluid escaping from the renal veins on an arterial perfusion alone varied from 20 to 60 c.c. per hour in different experiments ; an average rate was about 30 c.c. per hour. On a venous perfusion alone the rate of perfusion was more constant and averaged 60-70 c.c. per hour. Flow of Urine—The amount of urine obtained from the normal kidneys ‘on an arterial perfusion alone varies directly with the rate of perfusion, and under favourable circumstances as much as 1°5 c.c. may be obtained ‘in less than an hour. The concentration of the urine is almost always notably less than that of the perfusing fluid when the latter is hypotonic Ringer’s Solution; if the kidneys are perfused with normal Ringer’s solution the urine may be isotonic with, but is usually hypotonic to, the perfusing fluid. The urine obtained on a simultaneous arterial and venous perfusion does not, so far as we could determine, differ in amount from that obtained on an arterial perfusion alone; a simultaneous arterial and venous perfusion, however, seems to be more conducive to the formation of a very dilute urine than is arterial perfusion alone. Table I.—Urine from Living and Dead Kidneys. Concentration. Experiment. i Porfusine Rud Urine from normal Urine from dead 8 ; kidneys. kidneys. per cent. per cent. per cent. per cent. | 1 0-59 per cent. NaCl (a) 0°42 (5) 0°46 2 0°57 x *: | (a) 0°40 (4) 0°38 3 055, Se (a) On38 (6) 0-40 4 0°53" - ,, 3 0°25 5 0-42 «, i 0°30 0:40 6 0°72 5 (a) 0°55 (6) 0°49 7 0°83 * * (a) 0°73 (4) 0 "68 (az) 0°83 (5) 0°83 The letters (z) and (4) refer to successive samples of urine. In Experiment 7 the perfusing fluid contained 0:1 per cent. NazSO,; in the others the perfusing fluid was simply normal: or hypotonic Ringer’s solution. ' On a venous perfusion alone no urine was secreted with any of the perfusing fluids used. In some of these experiments the arterial circulation was excluded by tying the aortic bulb at the outset and allowing the ‘glomeruli to become infarcted; in others, the glomeruli had previously been ‘perfused with Ringer’s solution and the arterial perfusion shut off. 360 Messrs. Bainbridge, Collins, and Menzies. [Mar. 27, (2) The Dead Kidney.—The vitality of the kidney was destroyed by perfusing it through the aorta either with weak (1/10,000) corrosive sublimate or with boiled Ringer’s solution. After the former procedure the arterial perfusion with Ringer’s solution was resumed under normal pressure. The rate of perfusion and the amount of urine obtained were always much less than in the normal kidney, and sometimes, with a very slow perfusion rate, the flow of urine entirely ceased. The urine was usually isotonic with, but occasionally hypertonic to, the perfusing fluid, the latter only in experi- ments in which the formation of urine was extremely slow and scanty. (3) Since the urine obtained in all these experiments comes solely from the glomeruli (the tubules secrete no urine), it is natural to suppose that the difference in the character of the urine formed by the intact and dead kidneys. respectively depends upon one of two causes. On the one hand, the glomeruli may normally form by filtration a urine which is isotonic with the perfusing fluid, and the absorption of salt may be effected by the tubules as the glomerular filtrate passes along them. On the other hand, the tubules may possess no absorptive power for sodium chloride or other salts, and the glomeruli may possess the capacity to secrete a hypotonic urine. In attempting to decide between these two possibilities, two methods have: been used. (a) The tubules were poisoned by perfusing 1/10,000 mercuric chloride through the renal portal vein for three to five minutes, and then Ringer’s. solution was perfused for a few minutes through the renal portal veins, to wash away the mercury in the blood-vessels. The arterial perfusion of oxygenated Ringer was maintained throughout the experiment. The urine obtained both before and after the poisoning of the tubules was examined, and at the end of the experiment the mercury was fixed in the tubule cells. by perfusing dilute ammonium sulphide through the renal portal veins, and the kidneys were examined histologically. It was found in most of the experiments that the glomeruli remained free from,mercury, and that mercuric sulphide was present in the whole of the tubules. This was also: the case in control experiments in which the mercury was fixed by ammonium sulphide immediately after it had been perfused through the renal portal vessels. Experiments in which mercury was present in the glomeruli were- rejected. The following protocol illustrates the character of these experi- ments :— 1913.] Experiments on the Kidneys of the Frog. 361 Protocol I.—Pithed Male Frog. Cannule’ in left aorta, origin of vena cava, anterior abdominal vein and ureters. Hind legs tied off. Mesenteric artery and right aorta ligatured and testes removed. Fluid : ie escaping ie Concentration o Time by Urine ithe: vy. cava. | C.c, Simultaneous arte- ( | 3.15-3.30 22 | 8.10-3.30 R. K.0O‘le.c. | 0°33 per cent. NaCl. | rial and venous | R. K.0°2 ,, | 0°32 i Fs perfusion oxyg. SES tese a0 | 8.30-8.404 J LE OPI 5 POPee) ‘ rita | Ringer’s solution. 3.45-4.0 21 3.40-3.57 { B. . one ¥ | ; mn - e 3.58-4.4, Perfused 1/10,000 HgCl, through renal portal veins. Arterial perfusion main- | tained. Then perfused Ringer’s solution for 5 minutes through renal portal veins, to wash | the mercuric chloride out of the vessels. | Simultaneous per- } | fusion, art. oxyg., venous non-oxy¢., but containing O'l p.c. caffein. + A. press. 24 cm. V. 12 cm. Ringer = 0°53 per cent. NaCi J | Finally perfused weak ammonium sulphide through renal portal vein. Kidneys cut in serial section; no mercury observed histologically in the glomeruli. ‘D8 per cent. NaCl ‘52 ” ” | | | | The ‘venous perfusion of corrosive sublimate causes considerable vaso- constriction, and the efferent vessel from the glomerulus which ends in the tubular capillary network appears to be particularly affected. This, at least, is our interpretation of the extremely slow perfusion through the glomeruli which is met with under these circumstances and which is associated with a lessened flow of urine. We have attempted to overcome this vaso-constric- tion by perfusing through the renal portal veins Ringer’s solution containing a trace of acetic acid or 0:1 per cent. caffein sodium benzoate, but without complete success. We have obtained, however, a rate of perfusion through the glomeruli which was adequate to maintain their vitality, and in some experiments was for a time equivalent to that of the normal kidney, although it eventually became very slow. The employment of 1/10,000 mercuric cyanide caused less vaso-constriction, but also less definite histological evidence of the complete poisoning of the tubules. The results of a number of these experiments are shown in Table II. 362 Messrs. Bainbridge, Collins, and Menzies. [Mar. 27, Table Il.—Effect of Poisoning the Tubules with Corrosive Sublimate. The perfusing fluid was in all cases hypotonic Ringer’s solution. l Concentration of urine. | Concentration of | : F De perfusing fluid. Histological result. Normal kidneys. | Poisoned kidneys. | per cent. | per cent. 1 | 0°38, 0:29, 0°40 0°53, 0°52 0 ‘53 per cent. NaCl) 2 0-48, 0°43 0-60 0:58 ar Glomeruli intact, 3 0°30 0-40 0°42 ~—Cy 33 i tubules all show Hg. 4 0:47, 0-40 0°51 0°57. ,, ir Different figures represent separate samples of urine. (b) The kidneys were first perfused with oxygenated Ringer’s solution simultaneously by the aorta and renal portal veins. Then the glomeruli were perfused with boiled Ringer’s solution in order to kill them, while the tubules were still receiving by the veins an adequate supply of oxygen. A typical experiment is shown in the following protocol :— Protocol II.—Pithed Male Frog. Cannule in left aorta, anterior abdominal vein, inferior vena cava and ureters. Mesenteric artery and right aorta ligatured. Hind legs tied off. Simultaneous arterial and venous perfusion. Arterial pressure 24 cm., venous 12cm. Molecular concen- tration of Ringer’s solution (refractometer) = 0°55 per cent. NaCl. Ringer’s solution contained 0°5 per cent. NaCl, 0°02 per cent. KCl, 0:03 per cent CaCl, in distilled water. ‘ Escape from j : Pigtate Time. Sanita Urine, Concentration of urine. | c.c R.K.0-1. ce. 0-40 per cent. NaCl | 1147-124 31 11.40-12. 10{ t KoW8 Dabines 12.4-12.28 51 anes R.K.0°15 ,, 0-330 { 12.28-12.47 37 12.10-12.48 { hese a Meee z 12.47-1.17 33 12.48-2.15, both kidneys 0° oT c.c.| 0°36 ,, a 1.17-1.47 . 27 : LAT-2.17.5 | 23 2.17-2.47 H 25 ‘ 12.50 onwards. Glomeruli perfused with boiled Ringer’s solution. It will be noticed that the cutting off of the oxygen supply to the glomeruli lessened the rate of arterial perfusion and also the amount of urine formed. The general result of a number of such experiments is shown in the following table :— 1913, | Experiments on the Kidneys of the Frog. 363 Table III. Concentration of urine. i ae as 2 | Concentration perfusing expt | fluid. Normal kidney, Glomeruli killed. 1 (a) 0°48 per cent. NaCl | f(z) 0°45 per cent, NaCl} 0°56 per cent. NaCl (0) 0°45 ” ” (0) 0-45 ” ” (a) 0°43 ” ” 2 (6) 0°34 ” »” 0°36 ” ” 0°55 ” ” (c) 0 33 ” ” 8 (a) 0 “48 ” ”? (a) 0 “42 ” ” 0 50 ” ” (b) 0°38 ” iy (d) 0°43 ” ” mn { (a) 0°46, 66) (a) 0°48, % 0°53, 7 (4) 0°36, + (6) 0:48, 4 In all the experiments the perfusing fluid contained sodium chloride plus 0°02 per cent. KCl - and 0:03 per cent. CaCl). The letters (a) and (0) refer to separate samples of urine. (ce) Caffen—Barcroft and Straub have shown that caffein sodium benzoate greatly diminishes the consumption of oxygen by the mammalian kidney, and they used it for the purpose of poisoning the renal tubules. We have carried out a number of experiments to determine the action of caffein on the frog’s kidney. In our early experiments 1 per cent. caffein sodium benzoate was perfused through the renal portal veins for five minutes, the- arterial perfusion of oxygenated Ringer’s solution being simultaneously- maintained ; the collection and examination of the urine was not begun until, from one to two hours after the perfusion of the caffein, and it was found to: be hypotonic to the perfusing fluid. At that time we believed that caffein, (in the dose given) permanently poisoned the tubules, and we therefore- regarded the glomeruli as capable of secreting a hypotonic urine. Further: experiment seems to show, however, that the poisoning effect of caffein is merely temporary. The immediate effect is to render the urine isotonic with the. Table 1V.—Action of Caffein Sodium Benzoate. Concentration of fluid in terms of NaCl. | Urine after poisoning tubules with Expt. caffein Perfusing fluid. | Normal urine. whi Flay. i | Immediately after. | 1-2 hours after. per cent. per cent. iper cent. per cent. 1 0°53 y 0-25 0°54 = 2 0°55 0-40 0°55 — 3 0°51 | == = | 0°22, 0°30 364 Experiments on the Kidneys of the Frog. perfusing fluid, but this effect gradually passes off and, unless more caffein is injected, the urine once more becomes hypotonic to the perfusing fluid. We found that if 0°1 per cent. caffein were continuously perfused through the renal portal veins the urine remained isotonic with the arterial perfusing fluid throughout the experiment. Unfortunately it is impossible to trace the caffein histologically into the tubule cells, and we do not know whether it attacks the whole length of the tubules or not. We regard these results, therefore, as merely subsidiary to and confirmatory of those in which the tubules were poisoned with mercury. The formation of a urine which is hypertonic to the perfusing fluid has occasionally been observed after poisoning the tubules with corrosive sublimate and even in the dead kidney. It occurs only when the formation of urine is extremely slow ; we are not yet satisfied as to its significance. Summary and Conclusions. When the frog’s kidneys are perfused through the aorta and the renal portal veins with oxygenated normal or hypotonic Ringer’s solution, the urine formed is hypotonic to the perfusing fluid and is derived entirely from the glomeruli, since the tubules secrete no urine under these circumstances. When the tubules are poisoned with corrosive sublimate or (temporarily) with caffein, the urime becomes isotonic with the perfusing fluid. On the contrary, if the glomeruli are killed by the arterial perfusion of boiled Ringer’s solution, while the tubules still receive an adequate supply of oxygen through the renal portal veins, the urine formed continues to be more dilute than the perfusing fluid. These results suggest, first, that the glomeruli form by filtration a urine isotonic with the perfusing fluid, and, secondly, that during the passage of the glomerular filtrate down the tubules sodium chloride is absorbed by them. Whether any water is also absorbed we do not know. The expenses of this research have been defrayed by a grant from the Government Grant Committee of the Royal Society. REFERENCES. (1) Barcroft and Straub, ‘ Journ. Physiol.,’ 1910, vol. 41, p. 145. (2) Beddard, ‘Journ. Physiol.,’ 1902, vol. 28, p. 20. (3) Bainbridge and Beddard, ‘ Bio-chem. Journ.,’ 1906, vol. 1, p. 255. (4) Walter, ‘Ann. Phys. Chem.,’ vol. 38, p. 107; ‘ Journ. Chem. Soc.,’ 1890, A, p. 202. 365 The Effect of the Lability (Resilience) of the Arterial Wall on the Blood Pressure and Pulse Curve.—II. By LeonarD HILL, F.R.S., and Martin Fiack.* (Received March 31,—Read April 10, 1913.) (From the Physiological Laboratory, London Hospital Medical College, London Hospital Research Fund.) In a paper published in the ‘ Proceedings of the Royal. Society,’ 1913, B, vol. 86, p. 180, Russell Wells and Leonard Hill brought forward evidence to show that the nature of the arterial wall has an important effect in modifying the conduction of the pressure waves from the heart to those arteries where the pulse is usually explored, such, for example, as the radial, where sphygmograms are recorded and readings of arterial pressure taken with the sphygmomanometer. They concluded that the conduction depends on the greater or less “resilience” of the arterial wall, using the term “resilience” to express “the ease with which an elastic tube distends with a rise and recoils with a fall of pressure of the contained fluid” ; thus a rubber tube with a wall of 0°2 mm. thick is more “ resilient” than one with a wall 0°4 mm. thick, the thinner, more “resilient” tube yields with the rise and recoils with the fall of pressure more than the “ harder,” thicker-walled tube. A glass tube, in this sense, has no resilience, and the same may be said of rubber pressure tubing. As the arterial wall contains muscle, its “resilience” will be altered by a more or less contracted state, also since the degree of contraction and “resilience” may vary locally it is to be expected that the curve of blood- pressure may likewise vary, ¢.g. in the brachial and in the femoral arteries. We have found this to be the case under certain conditions, namely, in cases of aortic regurgitation.t In such cases the systolic pressure reading for the leg is much higher, 100 mm. or more, than in the arm arteries. Also in normal men a difference in the systolic pressure in the two radial arteries may be observed when the heart is made to beat forcibly by a short period of hard exercise and after one elbow has been placed in hot and the other in cold water. The artery relaxed by heat gives the lower systolic pressure. Russell Wells constructed a schema by means of which a known rhythmically changing pressure could be passed (1) through rubber tubes of the same calibre, but varying thickness, e.g. 0°8, 0:4, 0°2 mm., (2) through various lengths of the * During tenure of Eliza Ann Alston Research Scholarship. + L. Hill, with Martin Flack and W. Holtzmann, ‘ Heart,’ 1909, vol. 1, p. 73; L. Hill and R. A. Rowland, ‘ Heart,’ 1912, vol. 3, p. 222. WOlb, LOS aA— Ss, 2D 366 Messrs. L. Hill and M. Flack. Effect of [Mar. 31, same tube, (3) through the same tube and same length of tube, but with increasing amplitude. With an entering pressure of 160 mm. systolic and of 40 mm. diastolic, a curve taken with the Hiurthle manometer from the 0°8 mm. tube had all the characters of a “low ” pressure sphygmogram—ereat amplitude, sharp rise and fall and very well-marked dicrotic wave; while with exactly the same entering pressure the curve taken from the 0-2 mm. tube took on all the characters of a “high” pressure sphygmogram—slow rise, flat top, slow fall and slightly-marked dicrotism. Using the same tube, it was found that the higher the pressure and the more the “resilience” of the tube was brought into action, the nearer together were the diastolic and systolic pressures at the end of the tube. A length of 30 cm. of 0°2 mm.-thick tube with an entering pressure of 78 mm. Hg diastolic and of 148 mm. Hg systolic gave an almost continuous pressure of 104 mm. diastolic and 107 mm. systolic at its farther end. The same length of 0°8 mm.-thick tube gave a much more discontinuous pressure at its farther end. Lengthening a given tube had a like effect, approximating the diastolic and systolic pressures at its farther end, e.g. by increasing the length of a 0°8 mm. tube from 15 to 30 cm. the difference between systolic and diastolic pressures was diminished from 66 to 44 mm. . The use of the word “resilience” in the sense given above is not in accordance with the meaning given to this word by the physicist, and in the discussion which followed the reading of the above paper it was suggested by the President of the Royal Society that the word “lability” might be suitable. We propose to adopt this word and thus free ourselves from the charge of ascribing to “resilience” an equivocal significance. By the “lability” of an artery, then, we mean the ease with which it distends with a rise and recoils with a fall of pressure. We have investigated the effect of the “ lability” of arteries on the pressure curve, (1) by means of a simple schema, (2) by interpolating a length of artery between a Hiirthle manometer and the carotid artery when recording the blood-pressure curve in the living animal. The schema consists of a piece of rubber pressure tubing, connected at one end to a Hiirthle mano- meter and branching into two channels at the other end by means of a T-piece. A short length of rubber pressure tubing forms the channel on one side, and an equal length of artery that on the other side. The two channels are connected by another T-piece to a short length of rubber pressure tubing, which is closed at its farther end. This tube is rhythmically pulsed between the thumb and finger, the whole schema being filled with water to a pressure equal to that of the normal arterial pressure. Clamps are arranged so that, in turn, either the rubber pressure tubing or the artery is made to conduct 1913.] Lability of Arterial Wall on Blood Pressure. 367 the pulse to the manometer. We rhythmically pulsed the tube as hard as possible between thumb and finger, so that the pressure curves produced were made of approximately equal amplitude. Fig. 1 shows the curve obtained—A, when 5 inches of rubber pressure tube conducts the pressure-wave to the manometer; B, when the conduction is by 5 inches of femoral artery (human), all the branches of which have been tied so that there is no leak. The “ lability” of this length of artery has a very great effect, as is seen by the diminution of the systolic wave and the absence of “ overswing ” secondary waves. Fig. 2 shows—A, the curve obtained from the carotid artery of a cat, conducted to the manometer by a 23-cm. length of pressure tubing ; B, when the manometer is connected by a 23-cm. length of eat’s artery, made up of the aorta together with the part of one carotid and one femoral artery. The arrangement was such that the pulse curve could be transmitted alternately by the rubber pressure tube or by the artery. Fig. 3 shows the effect of transmitting the pressure curve from the carotid of a cat through—A, 8 em. of rubber pressure tubing; B, through 8 em. of excised cat’s carotid. Fig. 4 shows the effect of transmitting the curves—A, through 6 cm. of rubber pressure tubing; B, through 6 cm. of cat’s carotid. Fig. 5 shows that, after hardening the 6-cm. length of carotid artery in alcohol, the same result is obtained as with rubber pressure tubing. Fig. 6 demonstrates the effect of 4:5 cm. of cat’s carotid (B) on the conduction of the pressure curve. Even this short length notably alters the curve, particularly in diminishing the dicrotic wave. Fig. 7 shows the effect of connecting one end of an 8-cm. length of carotid artery, closed at the other end, to the rubber pressure tube through which the pressure curve was being conducted. This brings down the amplitude and approximates the systolic and diastolic pressures just as effectually as if the conduction were wholly through the same length of carotid. Fig. 8—A shows the curve taken from the right and left carotid of a cat. On the right side the cannula was inserted as near the aorta as possible, while on the left side the cannula was inserted as far from the aorta as possible. The dicrotic wave is much less marked in the tracing taken from the longer length of carotid. B shows tracings taken from the right and left carotid, both being as short as possible. There is no noteworthy difference. We give this tracing as a control, to show that the conduction by the length of artery is the factor which makes the difference in A. Fig. 9 shows the pressure curve of the cat taken—A, from the long length of left carotid still embedded in the tissues; B, from the short length of right carotid. In A, the dicrotic wave is less evident, and the curve has a flatter top, but the difference is less than in fig. 8. 368 Messrs. L. Hill and M. Flack. fect of | [Mar. Fig. 1. NS NS Ne van te WANN i OND ae ean Oe Pa AW WAZ AE MEA Awe Fie. 6. Pe ee A er ~)_ , dl, 1913.] Lability of Arterial Wall on Blood Pressure. 369 Fic. 7. Fia, 9. Fig. 10 shows the pressure curve taken from the carotid artery of a dog— A, through the rubber pressure tubing; B, through 5 inches of human femoral artery. The greater approximation of systolic and diastolic pressure and the diminution of the dicrotic wave is evident. These results confirm our previous conclusion that the nature—* lability ” —of the arterial wall notably affects the conduction of the pressure wave, and therefore both the form of the sphygmogram and the readings of pressure 370 Effect of Lability of Arterial Wall on Blood Pressure. obtained with the sphygmomanometer. Much of the systolic force of the heart is stored as potential energy in the distension of labile large arteries, to be given up again on their recoil during diastole; a part is spent in over- coming their resistance to distension; the greater the lability the less will be the amplitude of the systolic wave which reaches the peripherally placed arteries ; the higher also will be the diastolic wave, the closer the approxima- tion of the diastolic and systolic pressures, the less marked the dicrotic wave. It follows from these considerations that, supposing the force of the heart remains constant, what has been termed a “high” or a “low” pressure form of sphygmogram does not depend only on the resistance in the arterioles, but may be- obtained according to the greater or less “lability” of the conducting arteries—eg. aorta—subclavian—brachial—radial. So, too, will the readings of systolic and diastolic pressure vary with the “lability ” of the conducting arteries. Observations on cases of aortic regurgitation have shown us that the systolic readings may be 100, even 150, mm. Hg higher in the leg than in the arm arteries. This great difference is entirely due to conduction modified by the “lability” of the arteries. The arm arteries in such cases are more “labile” than the leg arteries—the latter are con- tracted, “ harder,’ more rigid, and conduct the systolic wave from the aorta with far less diminution of force. This is to ensure a circulation through the capillaries of the leg, to compensate for the great fall of diastolic pressure. By local modification of the contractile state of the arteries either the full hammer-like stroke of the heart may be delivered to the capillary vessels with a wide variation between the systolic and diastolic pressures, or a more uniform pressure be conveyed with the systolic variations of pressure approximated. In reading the systolic or diastolic pressure by means of the sphygmo- manometer we read, not the actual pressure produced by the heart, but this pressure: as conducted by the arteries to that artery selected for observation. In the normal young man, placed in the recumbent posture, the arteries so conduct, that approximately the same readings are obtained in the arm and in the leg arteries. The contraction of the muscular coat of the arteries is controlled so as to effect this. In cases of aortic regurgi- tation the conduction is widely different. A similar difference may also pertain in conditions of functional activity, eg. the leg arteries may be more contracted and give higher readings of systolic pressure after running up a flight of stairs. High readings of systolic pressure do not necessarily indicate any greater systolic force exerted by the heart. They may indicate, and probably often do indicate, less “lability ”—arteries held in the con- tracted state so as to conduct the systolic wave with almost undiminished ee See ee Probable Value to B. coli of “ Slime” Formation in Soils. 371 force. The character of pulse curves, taken either with a Hiirthle or other spring manometer, placed in direct communication with an artery, or by means of the sphygmograph, depends very largely on the “lability” of the conducting arteries. It is arterial “lability,” not reflection of waves, which modifies the form of the pulse curve taken in different arteries. While the pressure waves produced by the heart may remain the same, the form of the sphygmogram may be altered, and what has been termed a “high” or a “low” pressure curve may be produced by variation in the “lability ” of the conducting arteries. The wall of an artery is supported by the surrounding tissues and skin, the whole being permeated with blood; it will be a matter for further consideration as to how far the lability is affected by the condition of the surrounding tissues. Comparison of figs. 8 and 9 shows how large a part the tissues normally take in supporting the arteries. On the Probable Value to Bacillus coli of “ Slime” Formation in Soils. By Ceci Revis. (Communicated by Sir J. R. Bradford, K.C.M.G., Sec. R.S. Received April 8,— Read April 24, 1913.) During the course of an investigation into the causes of variation in the physiological activity of Bacillus coli, a number of experiments were started, in which soils, either virgin or mixed with cow dung or human excreta, were inoculated with cultures of £&. coli, together with cultures of various soil organisms so different from the colon organism that they could not be mistaken for it on plating out. The requisite quantity of soil was placed in a layer about ?-inch deep in large flat litre-bottles, and the cultures were added in the form of emulsions in physiological salt solution, made from agar slopes. Sufficient water was also added to make the soil visibly moist. The bottles were closed with cotton-wool plugs and kept at ordinary room temperature in the dark. Controls which were inoculated with all the organisms except the B. coli were started at the same time. The soils were examined from time to time by withdrawing about 5 grm. by means of a sterile tube, shaking this up with 50 c.c. of sterile water, spreading plates directly on to ordinary agar and incubating at 20° C. It was found very difficult to isolate the B. coli in this way because of the rapid and expansive growth of the other organisms present, and because the 372 Probable Value to B. coli of “ Slime” Formation in Sorls. experimental organism did not usually grow in a typical manner, but in large watery colonies, which were at first not recognised as B. col, and were also soon involved with other growths on account of their spreading nature. It was therefore necessary to employ the usual method of pre- liminary inoculation into bile-salt glucose broth, followed by plating out into ordinary (+1), or bile-salt agar. In this way B. coli was always readily isolated, but the tendency to form large, moist, slimy colonies was still marked, a characteristic to which I have directed attention before. The results were not of any great interest from the point of view of variation. From time to time, during the first 18 months of the investigation, apparently typical cola were isolated which refused to grow in peptone water or to attack any of the test substances. In many cases, the original culture from the plate failed to attack dulcitol and sometimes mannitol, but these failures were not of a permanent character. Towards the end of the experi- ment quite typical organisms only were obtained. The necessary use of bile- salt broth possibly is adverse to the separation of atypical organisms in the presence of a preponderance of typical forms. There was not apparently during the whole course of the experiment (which lasted three years) any marked diminution of the original B. colz, as it could be recovered in all cases from at least 0:00001 germ. of the soil. The remarkable point of the investigation lies, however, in the fact that throughout the course of the experiment no further addition of water was made to the flasks. The control flasks, which did not contain B. coli (though all the other soil organisms were present), dried up within a few months of the start. In all the flasks which contained B. coli not only did the flasks retain their moisture for three years, but during the first 12 months of the experiment had evidently taken up large quantities of moisture from the atmosphere, and in one or two instances the soil became completely water-logged. It seems evident that this extraordinary behaviour is connected with the B. coli, and in view of the fact, which I have constantly noticed, that this organism can easily produce “slime” (without the presence of sugar), and that when grown in this manner in soil it certainly does so, it seems reasonable to attribute the water-absorption of the soil to this curious property. These results possibly give at the same time some explanation of the well-known power of many organisms which occur in soil, especially the “nodule ” bacteria, to form “ slime.” The viability of B. coli for such a long period is also remarkable, but cannot, of course, be taken as true for ordinary soils, as there are bactericidal influences at work in such, which have been destroyed by the initial sterilisa- tion necessary in these experiments. SS 373 Variation in Bacillus colii—The Production of Two Permanent Varieties from one Original Strain by means of “ Brilliant Green.” By Crcin REvIs. (Communicated by Sir J. R. Bradford, K.C.M.G., Sec. R.S. Received April 8,— Read April 24, 1913.) In a former communication* it was noted that a profound change in physiological activity could be brought about in some strains of Bacillus coli by growth in the presence of Malachite Green. This change consisted in the complete loss of power to produce gas in the usual test media, the action ceasing at the acid stage. Similar experiments have been carried out using Brilliant Green in place of Malachite Green, and many interesting results have been obtained. The following is of particular interest in that two permanent varieties arose from one original organism under precisely similar circumstances. The method employed was exactly the same as in the former experiments. The culture used was repeatedly plated until it was evident that the final culture used had arisen from one single organism. This procedure is undoubtedly better and less liable to error than some of the more complicated methods which have been proposed (cp. Kisenbergt). The properties of the original culture were :— Mn series ead kde ee Coagulated in 48 hours. tLactose peptone water ...... qe op oe > 7, Sucrose i eh ee — Adonitol MAM Dat P RARER —-- Duleitol A) ee mere ++++ on 5th day. Inulin APRA DTT —— Glucose RAR PHE, dit Sesey +++ in 48 hours. Salicin a Gee =e on 5th day. Mannitol RERS Lisa bers 2 +++4+ in 48 hours. This selected culture was then inoculated into nutrient broth (+1) containing 0:004 per cent. of Brillant Green. Two different inoculations * Revis, ‘ Roy. Soc. Proc.,’ 1912, B, vol. 85, p. 192. + Eisenberg, ‘Cent. fiir Bakt.,’ 1912, Abt. I, vol. 63, p. 305. { The positive sign indicates the production of acid and gas in the test medium, using 5 c.c. medium and a gas tube ;3, inch diameter. Thesigns +, ++,+++,and +++4++ indicate 3, , $, and ? inch of gas in the tube. No change, thus VOL. LXXXVIL—B. Dies 374 Mr. C. Revis. [Apr. 8, were started in order to provide a control for any changes which might occur. This procedure has, however, been found subsequently to be of little avail, as in many cases in which two inoculations from the same cultures have been so started they have not behaved in the same manner under the same circum- stances. In this particular case one culture experienced much greater - difficulty in growing in the presence of Brilliant Green and eventually died out. The other which survived was finally plated out, when it had been trained to grow in broth containing 0:05 per cent. of the dye, and the organism was then developing well in it. The plating out was made on ordinary nutrient (+1) agar and the plates were kept at 20° C. ; Two types of colony arose:—(A) very small, (B) large and arborescent. Both varieties grew rather slowly at 20° C. on the plates, as did also sub- cultures from them. They were tested in the usual peptone water media with the following results :— CAS) Mitt eect serene sere eit Acid in 7 days, but no coagulation at all. Lactose peptone water... ++ in 48 hours. Sucrose a oo Adonitol a aoe Duleitol i .. *A,, sl. G., about the 20th day. Inulin * oo Glucose is ... +4 in 48 hours. Salicin i ... A., sl. G., on the 7th day. Mannitol a ae + in 48 hours. (B) was very similar, but milk coagulated in 48 hours, dulcitol was not attacked at all, and only A., sl. G. occurred in mannitol. It is noteworthy that the organism which showed the greater vitality {judged by growth on agar) had suffered the greater loss of fermentative power. This has been found to be the rule in other cases. Both of these organisms (A and B) were carried on in Brilliant Green broth (0-05 per cent.) but (A) soon succumbed, while (B) grew as well as before. This will now be called Culture B. The original culture in Brilliant Green broth from which these (A and B) were obtained was kept going in the same medium (Culture C) and an Inoculation was also made from it into Malachite Green broth (Culture D). In all cases these cultures were re-inoculated into fresh tubes once a week and were all kept at 37° C. * Indicates acd reaction, but only very small bubble of gas. 1913. | Variation in Bacillus coli. By Gin Culture C. This was plated out several times at intervals of about two months, but only one type of organism was obtained in every case. This type coagulated milk usually in 48 hours, produced only A., sl. G. in lactose, A., sl. G. to +reaction in dulcitol (in 8 to 12 days), + +reaction in glucose and A., sl. G. to +reaction in mannitol (in 48 hours). In salicin, acid usually appeared about the 7th day. This variety was quite permanent. Every endeavour was made to restore the original activity of the organism, without success. It does not differ markedly from the original culture, but tested side by side the difference is decided. Culture D. This was the inoculation of Culture C made into Malachite Green broth (0°05 per cent.), as stated. It underwent no greater physiological changes than did C, milk did not coagulate until about the 15th day in most cases, and in lactose never more than the A., sl. G. reaction was obtained. Physiologically the organism was the same as Culture C and was as permanent ; culturally it showed marked differences. Its power to grow at 20° C. was greatly inhibited. No growth was apparent until six days had elapsed, and then a very watery viscid streak appeared. Growth at 37° C. was quite normal. Culture B. This organism, on the other hand, became profoundly modified. After several re-inoculations in B.G. broth it was plated out and the colonies obtained on testing in the usual media gave: Milk, acid 48 hours, coagulated in about five days; lactose, glucose, and mannitol, acid in 48 hours, but no gas at any time; dulcitol, not attacked at all; salicin, sometimes acid after several days and sometimes not attacked. Every attempt to restore the original activity failed, the above features being quite permanent. One colony of this type was then started in Malachite Green broth (0:05 per cent.), in which it soon grew well. After several re-inoculations, the colonies, on plating out on agar (+1) at 20° C., came up very slowly and were watery and very large (4-inch diameter). On testing, acid production was delayed in lactose and mannitol till the 5th day, but glucose was rendered acid in 48 hours. Sub-culture on agar at 37° C. restored acid production in lactose to 48 hours, but not in mannitol, which was still delayed. Gelatin sub-cultures at 20° C. were rather weaker physiologically than the original culture from which they were taken. It must be carefully noted that only changes in physiological activity had occurred. The ordinary growth of all the organisms described above was 376 Variation in Bacillus colli. quite normal and strong at 37° C. and was quite strong at 20° C., there being only a great tendency to delay in development. In every case before testing in the various peptone waters a strong growth was first obtained and heavy inoculations were used, and there was no trace of the dye stuff added to the test media with the culture. The effect which was produced in each case described above had been brought about by some impression made on the protoplasm as it was transmitted unchanged through successive cultivations on ordinary media. The point which stands out clearly from the above results is that, from an original typical culture of B. coli obtained from a single cell, two strains have arisen, (1) a strain slightly modified by the dye stuff, but in a permanent manner and refusing to be further affected; (2) a strain gradually under- going profound and increasing change in the same environment and resulting in an organism entirely different from the original culture, the strain being also of a permanent character. It is important to notice that from one original organism there have arisen, by a simple process of cell division, at least two organisms, one of which is practically resistant to its environment, while the other has become greatly and progressively modified. It has been held that all such individuals should behave alike under similar circumstances, but it has been my constant experience that this is not the case. Failure to recognise this has no doubt led to the impression that organisms do not show variation. Further, granting that these fermentative changes are brought about by enzymes present in the bacterial cell, it is evident that these are not an intrinsic and integral part of the protoplasmic substance. They may be entirely lost or greatly modified in activity, and, supposing that two enzymes at least are necessary to bring about the complete fermentation of the test substance, 1t is also evident that those which bring about the acid change may subsist while those which produce gas, etc., are completely lost. Under these circumstances, it is not too much to suppose that in the life of the organism itself, the opposite phase may occur, and that, as under certain circumstances fermentative power is lost, so also, under some other set of circumstances, it may be acquired when it does not already exist. 377 Further Researches on the Extrusion of Granules by Trypano- somes and on their Further Development. By W. B. Fry, Major R.A.M.C., and H. 8. Ranken, M.B. (Glasg.), M.R.C.P. (Lond.), Captain R.A.M.C. (With a Note on Methods by H. G. PLmimer, F.R.S.) (Communicated by H. G. Plimmer, F.R.S. Received December 23, 1912,—Read February 6, 1913.) [Puatres 9-11.] Introduction. In March, 1911, in the course of some work on trypanosomes carried out at the Wellcome Tropical Research Laboratories, Khartoum, the extrusion of certain granules from trypanosomes was observed by one of us (W. B. F.). The Director of the Laboratories, Dr. Andrew Balfour, was informed of these observations, and he himself shortly after observed a somewhat similar extrusion of granules from Spirochetes (spirochetosis of fowls), an account of which he published.* In June, 1911, a preliminary note on the subject was communicated to the Royal Society by one of us (W. B. F.). Since then, a great deal of work has been done on the subject by us conjointly, but for the most part independently ; by one of us (W. B. F.) at Khartoum and in London, by the other (H. S. R.) at Yei in the Lado Enclave. As will be seen, the results recorded in the course of this paper go far to confirm the conclusions arrived at in the preliminary note, ze. that the phenomenon is one connected with a stage in the life-history of the parasite, especially in chronic trypanosomiasis, in which it is found that the trypano- somes disappear from the blood of an affected animal for considerable periods. These observations offer, too, an explanation of the infectivity of fluids, blood for example, which, while showing absolutely no trace of trypanosomes, will infect susceptible animals, a fact that all workers on trypanosomiasis are acquainted with; they also throw light on that condition which has been spoken of as “a possible ultra-microscopic stage ” in these diseases. Methods. In the earlier part of these investigations two methods were principally used, they were :— * ‘Brit. Med. Journ.,’ April 1, 1911. VOL, LXXXVI.—B. 2F 378 Major W. B. Fry and Captain H. 8. Ranken. [Dee. 23, (1) Dark-ground illumination used in the ordinary way, but with the addition of a practically monochromatic light, which improved the definition. (2) A method of “vital staining,” the stain used being 0°75-per-cent. toluidin blue in physiological salt solution. This was mixed with blood, gland juice or other fluid to be examined, in a capillary pipette, blown on to a slide, covered with a cover-slip, and ringed with vaseline. With solid organs an emulsion in salt solution was used. The proportion of stain varied with the material from 1-3 to 1-8, according to the rate at which it was felt desirable to cause the staining to take place. New methods of fixation and staining have also been used; these form the subject of a note appended to this paper. Both of these processes give practically the same results. I. On Granules in General in Trypanosomes. Besides the nucleus and blepharoplast there are other bodies in many trypanosomes which may, in the ordinary acceptation of the term, be called “ oranules.” There are certainly two classes of granules to be seen in trypanosomes: (1) those with which we are concerned—probably of nuclear origin and of infective nature ; and (2) others which probably represent stored food material. The latter are of importance for us only because of the possibility of their being confused with the former, and here it may be stated that it has been found possible to fix and stain preparations so as to show a difference between them in staining reaction. Further evidence in favour of this differentiation was met with as a side issue in the course of experiments with hypertonic and hypotonic salt solutions, to be described later. It was found that when trypanosomes swelled up under the influence of these solutions many granules disappeared, leaving evident only from one to three. The inference seems to be that the granules which disappeared, owing to altera- tion in osmotic conditions, are of a quite different nature. In this paper the word “ granule ” connotes those first mentioned, whilst the food granules are ignored in our descriptions, unless specifically mentioned. The following varieties of trypanosomes have been available for study, and granules have been observed in all :— (1) 7. gambiense (Sudan), (2) 7. rhodesiense, (3) T. brucei, (4) TL. evansi (Sudan), (5) 7. nanwm (Sudan), (6) 7. pecaudi (Sudan), (7) Z. lewisi. In all cases the granule, as seen by dark-ground illumination, is a small, sharply defined, highly refractile body, and on vital staining it takes up the toluidin blue rapidly, and shows as a deeply stained, more or less circular body, which contrasts with the lighter tint of the trypanosome body. 1912.] On the Extrusion of Granules by Trypanosomes. 379 The number of granules apparently varies in different species of trypano- somes. They may also vary in size and number in the same species. ¢.g. a strain of 7’. nanwm, obtained from cattle, was carried on by passage through gerbils. For two and a half months many of the trypanosomes contained a single large granule. At the end of this period the granule became multiple, and three or four could be seen; at the same time there was a great diminution in their size. It was noted that, coincidently with this increase in number of granules, the virulence of the strain became greater. We have found that granules are not necessarily always present in trypano- somes. At present we can only generally indicate the stage at which granules may develop, and are unable to say what conditions determine their appearance ; but the following details are the result of our observations :-— T. brucei was investigated in gerbils, in which the disease was fatal in six weeks, and during its course they showed at least two or three exacerbations with a large number of trypanosomes in the blood, with corresponding latent periods when they were absent. When trypanosomes first appeared in the blood, whether at the beginning of the infection or after a latent period, it was observed that they did not ‘contain granules; the latter developed about the fourth day after trypano- somes were first seen, and increased in size and number. For about 24 hours trypanosomes with granules were numerous. After this period, when free granules were numerous in the blood, the proportion of trypanosomes containing granules steadily diminished, till finally, though an enormous number of trypanosomes might be present, granules could not be found in any of them. ‘This condition usually preceded a latent phase, or the death of the animal. We have thus a definite sequence of events during an exacerbation of the disease :— ; (1) Trypanosomes without granules. (2) Trypanosomes showing granules which gradually become larger and very evident. (8) Many free granules. (4) Many trypanosomes but no contained granules. (5) Trypanolytic crisis, or death of the animal. This was also found to hold good with 7. nanwm and T. evansi (Sudan). In the case of a goat inoculated with 7. brucei, which lived for 133 days, and whose blood from the end of the first fortnight was always infective, no trypanosomes were at any time discoverable in the blood, which was examined daily for the first two months of the illness. In all the specimens 2¥F 2 380 Major W. B. Fry and Captain H. 8S. Ranken. [Dec. 23, of blood examined granules have been found. Similarly in the case of guinea-pigs and rabbits the blood has been found to be uniformly infective during the so-called latent periods, when no trypanosomes can be found in the blood by microscopic examination. - Il. L£atrusion of Granule. The original observations have been repeatedly verified during the past 18 months, and we have been able to satisfy ourselves completely that extrusion of granules is a constant feature of trypanosomal infections. The phenomenon has been observed in all species of trypanosomes studied with the exception of 7. lewisi. We were able to assure ourselves of the presence of granules in that trypanosome, but the movements are so active that definite extrusion was never witnessed by either of us. On account of the high degree of motility the species was unsuitable for work on this subject, and prolonged observations were not made. The mechanism of extrusion has been studied in detail in 7. nanwm and T. gambiense. (1) 7. nanwm.—The strain was obtained from infected cattle from the White Nile district, and, for the purpose of these observations, was kept up by passage through gerbils. This type of trypanosome is very convenient for: the study of this process, as the granule is large and very evident and the trypanosome, whilst evincing active lashing movements, does not progress across the field of the microscope, but remains more or less stationary, so: that there is no difficulty in watching the same trypanosome through all the phases over a period of several hours, if necessary. Further, an animal can be selected at a period when extrusion is a frequent occurrence. When extrusion is about to take place the granule begins to work its way slowly, but quite distinctly, from the centre of the trypanosome towards the poufted extremity. Arrived there, it makes its way back to the centre. This takes place quite often—as many as seven or eight such movements: having been observed. During these passages the granule can be seen distinctly bulging the periplast as if becoming more and more superficial— this bulging being strikingly apparent at the pointed extremity. Probably this movement is largely due to the movements of the trypanosome itself. Finally, the granule, stretching the periplast to a greater extent, is extruded: suddenly from the pointed extremity and becomes a free element in the surrounding medium. Plate 9, fig. 1, illustrates all these stages. (2) TZ. gambiense—Here the preparations were made direct from cases of human trypanosomiasis. In this species the granules are multiple and move rapidly backwards and forwards in the iong axis of the trypanosome. They exhibit also a dancing. 1912.] On the Extrusion of Granules by Trypanosomes. 381 movement and appear to throw themselves against the periplast and rebound from it. Sometimes the granules approach the surface, and in so doing may actually cause a slight protrusion on the covering membrane. This seems to be preliminary to extrusion, as afterwards the granule may be shot out with a certain degree of force into the free fluid to some distance from the host. In this species extrusion is not as a rule effected from the extremity, but from some point near the middle of the trypanosome body. In infections running a very rapid course—such as 7. brucei and T. rhodesiense in white rats—extrusion is readily observed, whereas in sleeping sickness in man, a very chronic infection, prolonged search may be necessary. Certain intermediate types of the disease are particularly suitable for study of this subject—for instance, 7’. brucei in gerbils, as described above. In the course of this infection granules are not extruded when the trypanosomes first appear. At a later stage the phenomenon is easily seen, and again it cannot be observed just before disappearance of trypanosomes from the blood. These facts tend to confirm our opinion that extrusion occurs at a definite period in the life of an adult trypanosome. Extrusion can be stimulated by the administration of drugs and by certain mechanical effects such as variations in osmotic conditions. Reference is made to extrusion induced by varying strengths of salt solution in a later section of this paper. Under ordinary circumstances extrusion of granules does not appear to have a prejudicial effect on the trypanosome. In warm wet preparations it can be seen to continue its movements and it apparently lives as long as the others. On the other hand, it has been shown above that extrusion of eranules, if occurring generally, apparently heralds a disappearance of trypanosomes from the blood and is, in fact, the precursor of a trypanolytic crisis. Under unfavourable circumstances, ey. after treatment, extrusion is followed by rapid disintegration of the trypanosomes. IIL. Effect of Drug Treatment on Extrusion. Certain phenomena in connection with the liberation of granules have been observed after treatment with antimony. Cases of sleeping sickness were given an intravenous injection of metallic antimony, and gland-puncture wet preparations made at short intervals after treatment, 3 minutes, 5 minutes, and so on. These were examined by dark-ground illumination and the results of the observations are here described. Extrusion of granules is more frequent. The exaggerated motility is one factor, and the protoplasm, and more particularly the periplast, seem to lose elasticity, with the result that the granules can get free more easily. If a 382 Major W. B. Fry and Captain H. 8. Ranken. [Dec. 23, granule is forcibly ejected by energetic movements of the trypanosome it is flung out into the free fluid to some distance ; this is the most usual method. Some trypanosomes seem to be acutely poisoned by the antimony, and death and complete dissolution occur very rapidly. This is more frequently seen when the preparation is made 5 to 7 minutes after injection of antimony. The trypanosome becomes anchored, its lashing movement slows down and comes to a standstill, the body swells and becomes bloated, losing its characteristic form. In this condition it is devoid of energy and can no longer forcibly extrude granules, but the latter have not suffered so severely and may still show an excited dancing movement inside the degenerate trypanosome body, which appears to give way before this activity, and the granule may ultimately work its way clear of the degenerate protoplasm and inelastic covering of the now dead, or dying, trypanosome. In other instances the trypanosome does not die so rapidly and the granules, after continuing this dancing movement inside it for some time, gradually come to rest before the trypanosome has reached so advanced a state of degeneration as to permit a dancing granule to escape by its own efforts. The degeneration of the trypanosome continues till it has lost outline and refractility and can only be recognised as an _ ill-defined “ghost,” enveloping the granules which are held in position—more or less in the original long axis of the trypanosome—by this viscid protoplasm. This is the last stage that can be seen in a dark-ground preparation where the objects are at rest. In the living subject, however, it is probable that this degenerate protoplasm would not be allowed to remain at rest, but would be broken up by the active currents and eddies and the granules would thus be set free. Thus there are probably three methods by which a granule may be liberated from the parent trypanosome :— (1) By the activity of the trypanosome—forcible extrusion. (2) By the active movement of the granule in a rapidly degenerating trypanosome. (3) By outside agencies, eddies, currents, etc, which may break up a degenerate trypanosome when the contained granules are unable to effect their escape. In some cases extrusion occurs rapidly. A trypanosome has been seen to extrude two large granules and immediately afterwards break up—the whole process being complete in 20 minutes. In the early preparations (3 minutes) the exaggerated motility is a prominent feature and forcible extrusion is most commonly seen; in the later films (7 minutes) the antimony has had longer time to act and the 1912.] On the Extrusion of Granules by Trypanosomes. 383 phase of hyperactivity has passed. It is then more usual to see the more gradual escape of the granule, and as the trypanosomes are “ anchored” they can be kept under observation more easily. On several occasions where death has occurred slowly we have been able to watch a trypanosome for periods up to four hours. In the 20-minute preparations trypanosomes have never been found, but granules are very numerous. The activity of the freshly extruded granules after antimony is much greater than the movements of granules seen before treatment. IV. The Free Granule. The granule free in the blood or fluids is seen to be a small spherical or pear-shaped body. In dark-ground preparations it is seen to be highly refractile, and by its activity it causes considerable disturbance in the surrounding fluid; with vital staining this young granule takes on the stain rapidly and uniformly, and seems to be undifferentiated. It frequently remains near its former host for some little time before showing independent movement. At first only a dancing movement may be seen ; this, however, is a preliminary phase, and soon the granule begins to move slowly across the field, turning over on itself. There is no doubt as to the motility: they have often been observed to move out of a microscope field in preparations where there was no question of currents, etc. In our opinion a pseudopodial protrusion appears early, which at first is short and rather thick. In animal infections and in cases of sleeping sickness in man, granules are found in the blood, glands, and internal organs. They are, of course, much more numerous in animals in which the adult parasites appear in great numbers. In experimental animals granules have been found in the proximal glands 24 hours after inoculation. This fact seems to be of great importance. The criterion in the recognition of granules must be motility,* but their greater affinity for such stains as toluidin blue is of undoubted assistance in distinguishing them from the countless small bodies seen in wet preparations, e.g. blood-platelets and leucocyte granules. V. Further Development of Granule. So far we have shown that the trypanosome discharges living elements endowed with motility, and showing the same reaction as nuclear material to toluidin blue. The further stages are more difficult to follow, as all stages * The addition of a small quantity of cherry-gum solution to the preparation will differentiate between Brownian and vital movement. It stops the former and slows the latter.—H. G. PLrer. 384 Major W. B. Fry and Captain H. 8. Ranken. [Dee. 23, cannot be seen in any individual preparation. We have endeavoured, so far as possible, to correlate the various appearances met with; at the same time we cannot be sure that we set them out in exact chronological order. In the first place granules, like many other free bodies in the blood plasma, are liable to undergo phagocytosis, and have been seen in all conditions within polynuclear leucocytes. Granules somewhat older have also been seen in hyaline mononuclear and in endothelial cells, but in cells of this type, on the other hand, the contained granules are quite unchanged, and we are unable to say that they are being destroyed. It is possible that they may be entering on an intra- cellular phase of existence. They have been very well seen by one of us (W. B. F.) in a large mononuclear leucocyte during examination of the blood of a cat infected with 7. nanwm. They have also been seen in endothelial cells in liver puncture preparations from cases of sleeping sickness. The first change seen in the free granule is a slight enlargement and elongation, rendering it more definitely pear-shaped. Then one begins to note a slight differentiation of structure into a central area staining a dark blue or purple, and a peripheral zone which is only faintly tinted blue. The enlargement is progressive, and the body becomes more uniformly blue, while a small dark blue or purple spot is visible, varying in position from the centre of the body to the apex. This may be assumed to be the earliest differentiation of cytoplasm from nuclear material. At this stage there is sometimes a definite flagellum-like projection which is usually short and rather thick, and more like a pseudopodium (Plate 11, fig. 2). The same early forms have been studied in dark-ground preparations from bone-marrow in animals infected with 7. nanwm and are illustrated in Plate 9, fig. 2, A to I. From this point the body enlarges, and the flagellum-like body becomes relatively, if not actually, reduced in size, so that forms are seen as in Plate 11, figs. 3 and 4. Later on the mass of chromatic material divides, and two are seen—one much smaller than the other. The body then becomes more rounded. Some are regularly spherical, while others show projections from various points, and have on surface view a roughly triangular appearance. At this time of their development they resemble very closely the Leishman- Donovan bodies in Kala-azar ; they are found sometimes in enormous numbers in lungs, bone-marrow and spleen. Death in acute trypanosomiasis is caused by plugging of the cerebral capillaries with these forms. This cause of death is very similar to that in pernicious malaria. From this stage—the binucleate body—there appear to be two directions 1912.] On the Extrusion of Granules by Trypanosomes. 385 in which the further development may proceed. The body may enlarge slightly, develop a true flagellum from the neighbourhood of the micro- nucleus, and then become longer. This increase in length continues, and the macro- and micro-nucleus in this process become further separated; the flagellum comes to lie along the margin, and this form can now be recognised as an early immature trypanosome. There is no undulating membrane, but development proceeds till the adult form is reached. ; On the other hand, the circular form may enlarge to a greater degree, and show a larger amount of a pale-blue staining cytoplasm that seems characteristic of young forms. The nucleus and micronucleus then undergo division by schizogony, but remain within the single mass of cytoplasm. The time of appearance of the flagellum seems to be variable, but ultimately all the pairs of macro- or micro-nuclei come to have a flagellum with a fan- shaped origin usually projecting beyond the margin of the cytoplasm. Plate 11 shows forms with two, four, and eight macro- and micro-nuclei and flagella. We have seen indications of similar forms in vital pre- parations, but the latter cannot show the same detail as fixed and stained preparations. We have no knowledge as to the conditions which determine either of these events—possibly in the latter case there may be some sexual process either in the cells or fluids. Many of these bodies are identical with the Plimmer and Bradford bodies, which they described in 1902,* and we have found them in preparations made from many different animals and from man, of glands, internal organs and bone-marrow. They show when living undoubted motility, but the early granule shows much more active movements than these later forms. The fact of their showing this vital property, however, precludes any possibility of their being degeneration forms. In a few cases of sleeping sickness in man some other bodies have been seen by the vital method in fluid obtained by liver puncture. In the majority of instances some blood was mixed with the liver juice; this diluted the fluid and the bodies were very scanty, but the appearances presented suggested that some process of division was going on. Protoplasmic masses were seen containing four or eight small ovoid bodies taking on nuclear stain, but there was no nuclear differentiation. These were seen only in wet preparations, and could not be preserved. Another form was seen as a fusiform body lying round a segment of the periphery, apparently of a mononuclear cell. It suggested an immature trypanosome, and this idea was confirmed by the presence of similar bodies free in the liver juice showing slight sluggish movement. * “Quart. Journ. Micros. Sci.,’ February, 1902. 386 Major W. B. Fry and Captain H. 8S. Ranken. [Dec. 23, VI. Fixed and Stained Specimens. The foregoing sections have dealt with living trypanosomes, but we were not able by the ordinary methods to make permanent preparations showing the various stages and forms, and demonstrating the staining reactions of the granule from its origin as a nuclear bud onwards. Mr. H. G. Plimmer, F.R.S., has appended a note describing special fixing and staining methods devised by him, and we wish to state that it is only by the use of these methods that we have been able to confirm the appearances we have described in unfixed wet preparations, together with the differentiation between vital and nutritive granules. In regard to the granule within the trypanosome, films have been stained showing the granule taking origin from the macro-nucleus itself as a small bud with characteristic chromatin reaction. All the stages of separation have been seen till the granule is a small, independent, dense, chromatin- staining mass in the cytoplasm (Plate 10, figs. 1-6). The granules, as stated, vary in number, and are most frequently seen between the macro- and micronucleus. They stain a deep red and show a remarkable contrast to the food granules which have taken on the iodine reaction from the fixation, and are visible as bluish-staining bodies or sometimes as a fused mass. This can be better seen in certain bird trypanosomes on account of their large size. In a certain number—probably the larger number—of instances the granule at some stage of its development is surrounded by a faint-staining hyaline circular or ovoid area. It is probable that in such cases the granule is really within a vacuole (Plate 10, fig. 8). Sometimes the granule appears to be spherical, but in other cases, even when being budded off from the nucleus, it already shows as an elongated, pear-shaped body; this is well seen in Plate 10, figs. 2-4. Granules can be seen actually causing a protuberance on the periplast and evidently on the point of being extruded. Others have been fixed when half-way out, while free granules, which have just effected their escape, have been seen lying close to the parent trypanosome. The early free granule takes on the chromatin stain deeply, and is identical with the body observed by the vital method. The observations as to phagocytosis have been confirmed and more advanced forms, showing a macronucleus, micronucleus, and flagellum, have also been seen within polynuclear cells, and rounded forms resembling the Leishman-Donovan body have been demonstrated in large mono- nuclear cells. All stages have been seen from the early free granule; the protoplasm 1912.| On the Extrusion of Granules by Trypanosomes. 387 becomes more visible, and increases in amount; the nuclear material becomes differentiated from it and more concentrated, and then we are able to see early forms with a macronucleus and micronucleus. The macronucleus in the circular forms may be spherical or may become elongated and spread out along the periphery. Some forms show much more protoplasm : it stains a pale blue and sometimes shows some faint pink granules. The flagellum varies in length, but is relatively much longer than that of the adult trypanosome, In the internal organs, and especially in the lung, there may be enormous numbers of these small rounded bodies with macro- and micro- nuclei, with or without flagella, sometimes separate and sometimes massed | together. A further stage has been observed in these masses; they have been seen just on the point of disruption, some of the small bodies were separating, and lay at varying distances from the main mass. Each showed the two nuclear elements with a small body of homogeneous cytoplasm. In addition, forms such as mentioned on p. 385 have been seen—large masses of protoplasm with two, four, or eight macronuclei, and corresponding micronuclei, which are, as a rule, placed close to the macronuclei, and stain very densely. The flagellum can be seen arising from a line equal in length and close to the micronucleus, in a fan-shaped collection of very fine filaments which unite to form a flagellum (Plate 11). In smears of blood or organs advanced single forms—i.ec. with one macro- nucleus, micronucleus and flagellum, and a relatively large amount of protoplasm—can be seen, and all stages from this to the adult trypanosome (Plate 11). A series has been prepared showing an almost imperceptible gradation from the granule stage up to adult trypanosomes. Up to this point we have only referred to the work of Bradford and Plimmer in their paper on Trypanosoma brucei and its development. In this paper and in the plates they have described and figured the granules within the trypanosomes, the free early bodies, the more advanced single forms called “amoeboid ” and the disrupted schizogonous bodies called “ plasmodial masses.” Our work was carried out at a time when we had no access to the paper, and this makes it all the more remarkable that the forms we describe should so closely resemble, and indeed confirm, many of the appearances described in 1902, and we feel that in many respects we can add little to the original work, beyond demonstrating the vital properties of these bodies. We should like to draw attention to the fact that early granules, forms with short flagella and small round forms, are figured by Mott* in his * “Reports of the Sleeping Sickness Commission of the Royal Society,’ No. VII December, 1906. 388 Major W. B. Fry and Captain H. S. Ranken. [Dec. 23, “ Histological Observations on Sleeping Sickness and other Trypanosome Infections.” VII. Some Animal Experimental Work in Reference to Granules. A number of experiments were undertaken to ascertain if it were possible to infect animals by granules alone. To do this, fluid containing granules and no trypanosomes was required. It was thought possible that the granules (if reproductive elements) might prove more resistant to changes in their environment than adult trypanosomes. In order to test this, blood showing a heavy infection was added to a hypertonic salt solution, up to 2 per cent. It was found on mixing one volume of infected blood with two to three of salt solution and keeping it at temperatures between 34° and 38° C., that after standing for 5 to 10 minutes individual trypanosomes began to swell up and become globular and the contained granule or granules to become active, moving about in the now spherical trypanosomes ; after a short period the granules escaped from the containing membrane and became free. The remnant of the trypanosome was left as a faintly discernible spherical body with no characteristic features. This process of escape of granules continued until no formed trypanosomes could be found; at the end of from half to three-quarters of an hour the process, as a rule, appeared complete. There are apparently several factors which influence the occurrence of this phenomenon, the temperature, the hypertonicity of the solution, the stage of development of the trypanosomes, and the strain worked with. If, whilst looking at one of these sides during the process, an individual trypanosome be watched, it will be noticed that its active movements suddenly become slowed, and then, as though blown steadily out by some entering fluid, the trypanosome, in the course of about 3 to 10 sees., is changed from its usual shape to that of a round body in which the granule or granules are freely motile. The escape of the granule takes place, as a rule, a few minutes after this. Infection was obtained repeatedly, and the following are details of two positive results :— No. I. November 5.—Gerbil (F. 10) injected with about 0:2 ec. of treated blood obtained from a gerbil infected with 7. nanwm (heavy infection). The injected blood was treated with sodium chloride solution 2 per cent. and sodium citrate 1 per cent. for one hour. At the moment of injection no living trypanosomes could be distinguished ; sphere forms and free granules very numerous. November 10.—Trypanosomes first found in blood. November 11.—Trypanosomes very numerous. November 14.—Gerbil found dead ; spleen very large. 1912.) On the Haxtrusion of Granules by Trypanosomes. 389 No. II. November 5.—Gerbil (F. 11) injected with blood (0:2 ¢.c.) obtained as above, but after two hours’ standing no trypanosome could be seen, only round forms and free granules. November 12.—Trypanosomies first found in blood. November 14.—Trypanosomes very numerous. November 15.—Gerbil found dead ; spleen very laige. The average time of infection in gerbils is four to six days after ordinary inoculation. Similar results were also obtained with dogs. These experiments are, of course, not absclutely conclusive, but so far as could be ascertained microscopically the granules were the only discernible remnants of the trypanosomes which retained their characteristic form. Further experiments were also made to trace if possible the fate of granules so injected into animals. Inoculations were made with solutions containing large number of free granules, and the animals were killed before trypanosomes could be found in the blood. Granules and the later forms in various stages of development were found in the proximal glands, also in the internal organs, Note on a New Method of Blood Fixation. By H. G. Primer, F.R:S. During some years of work on the blood of animals, many methods of fixation have been tried, principally with the view of obtaining a better fixation of blood parasites. The method described below has fulfilled this object better than any other, and is more faithful than even osmic acid. The use of iodine for the fixation of unicellular organisms dates from the work of Kent in 1881 on the Infusoria, but the application of it to blood is, so far as I know, new. I have used iodine in two forms, in vapour and in solution, and each has its special advantages. When a blood-film is exposed wet to the vapour from a solution of iodine in chloroform, the fixation of the various elements is practically instantaneous, as the penetrative power of iodine in this form is greater than that of any other fixative known to me; there is less alteration both in form and size of the cellular elements and parasites than with any other fixative. When used in solution several things happen which are of value in enabling very fine structures to be more easily made out. If blood be mixed with a solution of iodine in salt solution containing iodide of potassium, certain elements and parasites, especially trypanosomes, swell up so that the finer parts of their structure, for instance the nucleus and blepharoplast, are much clearer and more definite than with the ordinary 390 Major W. B. Fry and Captain H. 8. Ranken. [Dee. 23, methods. The nucleus shows as clearly as, if not clearer than, when Flemming’s solution and iron-hematoxylin have been used. There is the clear space containing the karyosome, and surrounding this, in many cases, are seen a number of granules, some of which can be seen budding off. The blepharoplast is clearly seen as a structure quite distinct from the micronucleus, and the earlier stages of division of a trypanosome, 7.c. the division of the blepharoplast and the formation of a second undulating membrane extending down the body of the trypanosome and forming eventually a second flagellum, can be seen and followed easier than with any other mode of fixation. For the smaller forms found in spleen, glands, and marrow of animals with chronic trypano- somiasis, this method, by causing swelling of the elements, renders the very small forms distinct, and renders their nuclear structures much more visible. Both these methods are also the best I have found for avian and reptilian blood containing parasites, e.g. filaria, malaria, hemogregarines, ete. The steps of the two methods are here detailed. Either slides or cover- glasses can be used, but in all blood-work the best results are obtained with cover-glasses. After the Giemsa or fuchsin staining the definition is greatly increased by the use of a green monochromatic screen, such as Wratten’s No. 19, which shows the picture in blacks and greys. I. Vapour Method. 1. Expose the thinnest possible film whilst wet to the vapour of a solution of iodine in chloroform for 10-15 seconds until it is distinctly yellowish. ; A hollowed glass block does for cover-glasses, and a glass cylinder of suitable height, with the iodine and chloroform in a small vessel at the bottom, does for the slides. In cold places the vessel should be warmed in order to get the vapour given off freely. 2. Place the film when it has become just surface dry (a dead, mat surface, not really dry) in chloroform, or in alcohol and ether, equal parts, for two hours. I use chloroform for cover-glasses and alcohol-ether for the rougher slides. 3. There will now be no free iodine left in the film, and it can be stained in many ways. I use the following :— A. a. Drop 3-8 drops of Giemsa’s solution on the film, and immediately after double the number of drops of distilled water. Leave for from 2 to 12 hours. Wash well with tap-water. Drop on 2-8 drops of orange-tannin solution and leave for 15 seconds. Wash thoroughly with tap-water, up to two minutes. Dry with filter-paper. Mount in cedar oil or liquid paraffin. Carbol-fuchsin for from 2 to 12 hours. Wash in tap-water. Alcohol until free from bulk of stain. Differentiate in clove oil saturated with orange G. Stop when desired by washing in xylol. Mount in cedar oil or liquid paraffin. Sas Sk BS ass 1912.] On the Extrusion of Granules by Trypanosomes. 391 C. Iron-hematoxylin may be used in any of the ordinary ways. Kernschwarz for 24 hours gives very delicate results. II. Solution Method. 1. Make a saturated solution of potassium iodide in 0°8-per-cent. salt solution and add iodine to saturation. 2. Mix 5-6 drops of this with 10 c.c. of salt solution. 3. Mix in a marked pipette equal parts of this and the blood to be examined. In the case of organs small pieces may be crushed in an equivalent quantity of the iodine solution to form an emulsion. 4, Take large drops and make a thickish film. Wait until the surface has begun to dry (as in I), and place in alcohol and ether for two hours. 5. Continue as under 3. DESCRIPTION OF PLATES. Puate 9. Fig. 1.—Series to illustrate mechanism of extrusion of granules in 7. nanwm (see p. 380). », 2.—Developmental forms of 7. nanum, seen in bone-marrow; the progressive tendency towards the characteristic shape of the adult trypanosome is shown. Dark-ground illumination, Leitz +; objective, N.A. 1.30, compensating eye- piece. x8. The earliest form, A, shows no evidence of a protoplasmic envelope and has the appearance of a well-developed granule just after extrusion. In B the cytoplasm is cleariy evident and the separation of the micronucleus has commenced. C shows a well-developed form, of circular shape, with the nuclei shown at a distance from each other. D, E, F, and G show the progressive increase of protoplasm, the last form being almost trypanosomal. H is a young trypanosome, and I an older one in which a flagellum is evident. These forms were all living when drawn. Puates 10 anv 11, All the figures are drawn under a Zeiss 3-mm. apochromatic objective, N.A. 1.40, with compensating ocular, x12. PLateE 10. Figs. 1-8.—T. rhodesiense in rat’s blood, showing granules from their origin to extrusion. Figs. 9-16.—From blood and liver of rat infected with JT. rhodesiense. Figs. 17-22, 24, and 26.—Are from the spleen of a guinea-pig infected with Nagana which lived three months, and showed no trypanosomes in the blood for some time before death. Figs. 23 and 25.—From a lymphatic gland of a cat infected with Nagana. Fig. 1.—Four granules are seen in the trypanosome-body, and another is in an early stage of being budded off from the macronucleus at the right upper angle. 2.—Two granules are seen coming off the macronucleus. The one on the left is still attached and shows the elongated form. ” 392 Major W, B. Fry and Captain H. 8. Ranken. [Dec. 23, Fig. 3.—A similar elongated granule is seen completely separated from the nucleus. There is a faint indication of a halo surrounding it. 4.—A large elongated granule is seen between the macro- and micronuclei, lying close to the periplast. 5.—Several granules are present ; one is just being detached from the macronucleus. 6.—Two granules are seen on the point of escaping from the trypanosome ; the larger looks as if it is nearly extruded. 7.—A recently extruded granule is seen near the trypanosome. The macronucleus shows two deeply stained points—probably granules becoming differentiated in its substance before being budded off. 8.—Two granules, lying between the macro- and micronuclei, are each seen to be surrounded by a well-defined clear hyaline area. Two others are almost com- pletely separated from the macronucleus. 9.—Free granule ; no differentiation. 10.—Free granule, larger, and with a faint rim of cytoplasm. 11.—Ring-shaped nucleus with micronucleus coming off ; definitely more protoplasm than the previous form. 12.—Early form with macro- and micronucleus and pale blue-staining cytoplasm. 13.—Similar form, larger. 14.—The nucleus has divided in this specimen, while there is only one micronucleus seen. 15.—Both macro- and micronuclei are divided. 16.—Micronuclei only have divided ; macronucleus in process of division. 17-26.—All are similar forms. They vary in shape and correspond closely with the forms seen by vital staining of emulsions of internal organs. 20 and 22.—Show division of the micronuclei. 21 and 22.—Show the third chromatin body described. 25.—Shows division of macro- and micronuclei. 27.—A single form, with macro- and micronucleus, and a very long flagellum. PLATE 11. Figs. 1-12.—The specimens were found in smear preparations from the liver and kidney from rats infected with 7. rhodesiense. They show dividing forms in various stages. Figs. 13-20.—Blood from liver of rat infected with 7. rhodesiense. - Immature trypano- somes are shown gradually merging into adult forms. Fig. 1.—Early stage of division. There are already two micronuclei, but the macro- 9 nucleus is just beginning to divide. 2.—This shows similar division to fig. 1, but a little further advanced. The macro- nucleus is now in the stage of mitosis. 3.—Complete separation of macro- and micronuclei, but the flagella are not yet separated. 4,—Two form with uuclei and flagella completely divided ; one flagellum is much longer than the other and lies round the margin of the body. 5.—Two form beginning to divide into two independent bodies which are identical with the early immature forms shown in figs. 13-15. 6.—Two form. The nuclei have moved to some distance from each other. A thick fan is seen in the shorter of the two flagella. FRY & RANKEN. Roy. Soc. Proc. B., Vol. 86, Pl. 9. Fig. 1. Fig. 2. it Na Os Ftoy Soe roe. Bvol &6. Pl 10 Roy Soc. Proe Brol 86 Pi 11 Warmer osivatipien iors mule caie, 1912.| On the Extrusion of Granules by Trypanosomes, 393 Fig. 7.—Four form (early), the macronuclei have evidently recently divided. The two lower are moving away from each other; the upper have not completely separated. », 8.—More advanced four form ; all the pairs of macro- and micronuclei have moved away from each other. » 9.—Eight form, a large body of cytoplasm whose margin shows a few indentations as if there might later be division of the whole mass at these situations. All the macro- and micronuclei and flagella can be seen. » 10.—Eight form beginning apparently to divide; the cytoplasm shows lines of cleavage along the lower part of the outline. ,», 11.—Mass of 16 bodies breaking up. These resemble the Leishman-Donovan heey. ; each has a macro- and micronucleus, but no flagellum. », 12.—A large form with single macronucleus and large micronucleus showing ent shaped origin to flagellum. », 13.—The body is rounded and has a clear blue-staining cytoplasm. The flagellum shows the fan-shaped origin well and stands straight out from the body. The micronucleus lies close to the macronucleus. », 14 and 15.—The body is longer, and the flagellum is lying along the margin ; the micronucleus is now moving away from the macronucleus. ,», 16, 17, and 18.—These features are more marked, and the specimens show gradual approximation to adult type. The flagellum is seen to be separated at some point from the outline of the trypanosome body, the earliest stage in the development of an undulating membrane. », 19.—The undulating membrane is now clearly present, but the trypanosome can still be recognised as immature by the fan-shaped origin of the flagellum and the pale homogeneous cytoplasm. », 20.—An early adult trypanosome ; the flagellum no longer shows the fan-shaped origin, and is much longer. Early granules can be seen in the cytoplasm. VOL. LXXXVI.—B, _ 7 (G 394 Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland.—The Wild-game Strain. By Surgeon-General Sir Davip Bruce, C.B., F.R.S., A.M.S.; Majors Davip Harvey and A. E. Hamerton, D.8.0., R.A.M.C.; and Lady Brucg, Tee, )A(C), (Scientific Commission of the Royal Society, Nyasaland, 1912.) (Received February 24,—Read April 10, 1913.) Introduction. Trypanosomes of this species, isolated from five antelope, are here described and compared with the Human strains which formed the subject of a previous paper.* The Wild-game strains were isolated by injecting blood of antelope into susceptible animals. The blood was, as a rule, injected into a healthy goat, monkey, and dog, and from these other animals were inoculated. Trypanosomes from the following species were studied: reedbuck, water- buck, oribi, and hartebeeste. In these experiments, with the exception of the oribi, the three inoculated animals became infected. In the case of the oribi the blood was inoculated into a monkey and dog, no goat being available, and the monkey alone took the disease. I. Morphology of Strain I, Reedbuck. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest :— Table I.—Measurements of the Length of the Trypanosome of Strain I, Reedbuck. In microns. P Method of Date. Method of fixing. 56 : : SETS Average Maximum | Minimum length. length. length. 1912 Osmic acid Giemsa 21-7 | 34:0 | 16°0 * Supra, p. 285. Trypanosome causing Disease in Man in Nyasaland. 395 Table I1—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain I, Reedbuck. In microns. jeenliG: Wie 18. iG} 20. 21. | 22 23. 24. | | | | | | | | SNOUEM Fedoneespveeceraeoctaredos | 5 30 57 90 81 | 53 2 17 16 | Dil he | | | | | | Percentages .....4......ss00000 Plc OmGcO ie 18:0 | 16°2 | 10°6 | 5:4 | 3°4 | 3:2 | et | | ell | | | In microns. - — | d | laa! ee 25. 26. | 27. | 28. | 29. | 30. | 81. | 32. | 33. | 34. | | | | | | | | LUGUGII-gegeodnpeneeenensyoteedseos 18 23 | 18 25 | 18 9 | 8 | 2 | 2 li || | | Ss | | | | } | EVCenbagese etc casecsaeaness 3:6 | 4°6 | 3°6 | 5:0 | 3:6/1:8|16|0-4); 0-4 | 0-2 | | | Cuart 1.—-Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain I, Reedbuck, taken on nine consecu- tive days from Rat 847. cpa tapes aes ET sr ee 396 Sir D. Bruce and others. Zrypanosome [Feb. 24, Table I1I.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain I, Reedbuck. Da apeen Arcana Percentage among short 0. and stumpy forms. 1912. | juiliy se /feeeteneer 847 Rat 4 py) CAO) bocodacac 847 5s 2 Si Doe eee 847 Zs 2 5 28) cooecnond 847 x 4 jy Mode eneese 847 fs 9 Se eS eaosaadan 847 % 10 Poet aiBons tee 847 3 19 SOT mea oe 847 ‘6 22 inom abdancans 847 3 4, AN IGTENS) aon sgtoacca0 8 °4 II. Morphology of Strain II, Waterbuck. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest :— Table [1V.—Measurements of the Length of the Trypanosome of Strain II, Waterbuck. In microns. . Method of | oe A hae es Average Maximum | Minimum length. length. length. 1912 Osmice acid Giemsa 23 °5 | 33 °0 | 16°0 1913.] causing Disease in Man in Nyasaland. 397 Table V.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain IT, Waterbuck. In microns. fae 2 16. 17. 18. 19. | 20. 21. 22, | 23. 24, | MCE Genpennericdoaso sal 2 8 26 | 59 74 58 | 58 44, Percentages ...... 0:2 0°4 1°6 5°2 11°8 | 14°8 | 11°6 | 11°6 8°8 } In microns. 25. 26. 27. 28. 29. 30. 31. 32, 33. | MOtalieeceecaeces =. 27 34 33 26 17 | 19 9 3 2 Percentages ...... 5 °*4 6°8 6°6 5°2 3°4 3°8 1°38 06 0°4 Cuart 2.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain II, Waterbuck, taken on nine consecu- * tive days from Rat 1220. fy asic ais eles ae ae ee eee Fas a | a a i a a a A LL 16 [Feb. 24, Sir D. Bruce and others. Trypanosome 398 Table VI.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain II, Waterbuck. Dite. | Beperent Wanraal: | Percentage among Bhort oO. | and stumpy forms. 1912 Septarwieeeseaces 1220 Rat 7 Pete) toaecsoc 1220 D 2 Mees 1220 sf 12 ees re Wek 1220 & 48 yO choedanss 1220 ~ 39 A She es 1220 ss 45 a) 1220 i 21 ee OMeees | 1220 3 50 Fy. 233) conten 1220 . 36 BEN RES 1220 * 47 ANSITEYGD sosneqocaae7 30°7 III. Morphology of Strain IIL, Oribi. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest :-— Table VII.—Measurements of the Length of the Trypanosome of Strain III, Oribi. | In microns. Date. Method of fixing. | ace of 8: Average Maximum | Minimum length. Jength. length. | | | | 1912 Osmic acid Giemsa 21°6 | 33 °0 16°0 1913. | causing Disease in Man in Nyasaland. 399 Table VIJI.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain ITI, Oribi. . Jn microns. 16. 17. 18. 19. 20. 21. 22, | 23. 24. ¢ mm te fal Metal es coasceetsc: \ et 10 | 22 | 7 | 109 | 90 | 87 28 19 | __- 2 (ome | tay | Percentages ...... 0-2 | 2°0 | 44 Tea || Paltshel| akes qa voaki lee! 5°6 3°8 | ae | | Tn microns. 25 26 27. | 28 29. 30 31. | 32 33 Sigh hee ener Ba | 15 | one | 14 | 6 5 op hase | | Percentages ...... BeG A acOn braze |. 28 2 On iO cs — | OE? | I | Cuart 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain III, Oribi, taken on nine consecutive days from Rat 992. BaEcEEEECECEEEE EE HE fie 400 Sir D. Bruce and others. Trypanosome [ Feb. 24, Table 1X.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain III, Oribi. | Date Experiment Ieee) Percentage among short _ | : No. : and stumpy forms. | 1912 | | INVERSE MW pemcsdoee 992 Rat | 12 yy. Bihestibcens 992 3 14 Pf TiG)) idbanadoce 992 5 10 We isieg eas 992 is 34 gle GO) ssaiewene 992 . 15 5 2d SS | 992 . 42 pe, OMe cae 992 29 0 LOWS scee | 992 . 52 J8 {eee 992 A 42 PA eB aso stacy 992 53 | | ATEREYR9 aanocososoce | 30°3 | | | IV. Morphology of Strain IV, Hartebeeste. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest :— Table X.—Measurements of the Length of the Trypanosome of Strain IV, Hartebeeste. | pe ; | In microns. vee | ee ‘staining. Average | Maximum | Minimum | length. | length. | length. 1912 | Osmie acid Giemsa 23 °5 | 35:0 | 18-0 1913. | causing Disease in Man in Nyasaland. 401 Table XI.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain IV, Hartebeeste.. In microns. Wirt | } if) |} 19 } 2. | 2. | 22.,| op | 24. | 25. | 26. | | | Tiare lane || a2 pases) | | es | 46 | 45 | 28 | | | | Percentages ...... 0-2 | 2-4 | 10°6 | 160 | 184 106| 92 90 56 | In microns 27. | 28. | 29. | 20. | 31 32 | 33. | ot | 35 | | | | | Tots) Avene zeieset if top 19 Gui 6 6 1 | } 7 : Percentages ...... 5°0 | 4°2 | 2°0 3°8 1-2 | TZ ie Vian OS 0:2 | | Cuart 4.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain 1V, Hartebeeste, taken on nine con- secutive days from Rat 849. Microns 3 [4 [5 [is [7 [va] [20a [aeles [ze] 25 [25] 27] 20] 20] 20] 1 [se] 25 [2 [as|e5] 7] 00 / BEB ASns es | A ee Piece NIAN tea SEERA 402 .Sir D. Bruce and others. Trypanosome [Feb. 24, Table XIJ.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain IV, Hartebeeste. | | { | | | Date Beperent Aiowal: Percentage among short | 0. and stumpy forms. | 1912 Uy Oleeep ere 849 Rat 15 By hn Oeacsee ses 849 1 18 BZD aseeta ee 849 i 8 BC laceetaes 849 3 16 be DA cao 849 x 26 Sn OM edocs 849 49 Pe 24) donsoniceo 849 % 16 PSO onopaabae 849 xp 38 pees! Ul saeangede 849 50 52 Are tl eer ese 849 . 45 AVELA SE cas. n ecco 28 °3 V. Morphology of Strain V, Hartebeeste. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest :— Table XIII.—Measurements of the Length of the Trypanosome of Strain V, Hartebeeste. | | In microns. | Date. | Method of fixing. | Method of 3 | | SA Average Maximum Minimum | length. length. length. 1912 Osmie acid. Giemsa 226 | 34-0 | 15°0 i 1913.] causing Disease in Man in Nyasaland. 403 Table XIV.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain V, Hartebeeste. In microns. toe RTGS M edie Hes ID.) BO OT. | Bo Bao) Uae. MBoballires: cz.21.cesecece to od i || ala 30 | 47 79 51 51 44 37 / | P Percentages ......... | O52) | O-2 | 2:2 | 6:0 | 9-4 | 15-8} 10-2 | 10-2) 8:8 | 7-4 —————— = — | == = —— In microns. | | 25. 26. PAE 28. 29. 30. 31. 32. | 33. | 34. | | le Potelie ces fa oo -| 36 Shy |) 143 94 | 1 3 8 1 = 2 | ! | | Percentages ......... Miwa dO 8 0-648: 22h) 60°64) DG O24 0 —— | 0-4 | Cart 5.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain V, Hartebeeste, taken on nine con- secutive days from Rat 1022. pea ic Wi SL Se SS 22) Lt eT A puliaiz es (IPD SP 2 OD = JE TOT AE e Ces SSS Sonor Ree fee Nes I ere Peh se | e T FF PES PNY 2 OF PP pe 404 Sir D. Bruce and others. Trypanosome [Feb. 24, Table XV.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain V, Hartebeeste. Experiment c Percentage among short Det. PN O. avunel, and nas fone 1912 INTE CA 38 orcas 102 Rat 18 Sao aan 1022 is 7 PROS Mine 1022 i | 11 ny bicereraee 1022 3 * -30 Ey ot emennen 1022 is 42 ao eae 1022 fe 55 ce ROS Rea 1022 ‘ 34 49 DO PMN 1022 < | 54 SEMin ZA cecoooase 1022 5 | 50 ANVOLEIAS) aocactiate rac 33 °4 Comparison of the Wild-game Strains with one another. Table XVI.—Measurements of the Length of the Trypanosome of the Wild- game Strains. In microns. Experi- N From Date.| ment Animal. ee gq.) What ere, No. See | aa Average Maximum , Minimum | length. | length. | length. 1912 783 | Reedbuck | 500 Rat | 21°7 340 160 1912 | 1180 Waterbuck 500 235 33 0 16:0 1912 | 863 | Oribi | 500 i 21°6 33 0 16-0 1912 | 799 | Hartebeeste 590 on 235 35°0 18 °0 1912) 957 | as 500 if 226 34-0 150 | ks ne 2500 Average 22 6 35 “0 15 ‘0 Comparison of the Curves of the Wild-game Strains. Unlike the curves of the Human strains, these are ali remarkably alike, and there can be little doubt that the same species of trypanosome is being dealt with in all five of the Wild-game strains. The Wild-game curves resemble Strains II, IV, and V of the Human strains, described in a former paper, and also those found by Kinghorn and Yorke in the Luangwa Valley. 1913. ] causing Disease in Man in Nyasaland. 405 Table X VII.—Distribution in respect to Length of 2500 Individuals of the Trypanosome of the five Wild-game Strains. In microns. 20. 21. | 22. 23. 24. | | Total seotanne 1 & | 58-4) 118 | 252 381 | 348 | 285 | 200 | 162 | | Percentages ......... | Seas: | 21 | 4-7 | 10-0 13-9 | 11:4! 8-0 | 675 In microns. a ] | 25. 26h | eee | BEY || CAE BIO BEES || SPE SBE |) BEL || ebP | | peal | i HMR atal eect asec vhs. ac | 149 | Sele Om G2 it bp. | 53a) 12 7 3 | 1 | nar | | Percentages .........| 6-0 | 5°4 | 5°0 | 4°4 | 2°5 | 2°2 | 15|05/|03]01) — |} | Cuart 6.—Composite Curve representing the Distribution, by Percentages, in respect to Length, of 2500 Individuals of the Trypanosome of the Wild-game Strains. __ Din Bie MEpSanononeeeeceeenoeneocan aaa EGaae ams aa mai ARSE Eeec oe el 1 ae ee PA mS eles | ee ee 7 | JUDE ens ae eae sees + GESOS0) G00 Gee PASCECE (a a pb tialal aleal wtb ad dhe hcbidl ollie) eee | Ea This composite curve resembles the human Strain Il, E——_. 406 Sir D. Bruce and others. Trypanosome [Feb. 24, Table XVIII—Comparison of the Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosomes of the five Wild-game Strains. Devi: Experiment earns Percentage among short No. and stumpy forms. | 1912 783 Reedbuck 8 °4 1912 1180 Waterbuck 30°7 1912 863 Oribi 30°3 | 1912 779 Hartebeeste 28 °3 | 1912 957 a 33 “4 | 2. re ae | ANISHEIL® — paeacns0s ec 26 °2 It ‘is evident from these tables and charts that the various strains of this trypanosome, as they occur in wild game, are remarkably alike. This is what might be expected. Here the trypanosome is at home: it is leading a natural life. It may be supposed to be saved from variation by constantly passing and repassing between the antelope and the tsetse fly. Comparison of the Human Strain with the Wild-game Strain. Table XIX.—Average Length of the Trypanosome of the Human and Wild- game Strains. In microns. | Number of Strain. trypanosomes Animal. | measured. Average Maximum Mininum length. length. length. TaQwiee Way Soncooroec00 3600 Rat 24-2 38 ‘0 15 ‘0 Wild-game......... 2500 an 226 35 ‘0 15 °0 The length of the trypanosomes of the Human strain found in white rats only is included in this table, in order to permit of comparison with the Wild-game strain, which is also taken from rats. The curves (Chart 7) differ from each other in such a marked manner as to be of no use in deciding as to the identity of the Human and Wild-game strains. In spite of this, however, by a comparison of the two strains morpho- logically and by the susceptibility of the different experimental animals to ' their pathogenic action, the Commission are driven for the present to the decision that the two strains belong to the same species of trypanosome. 1913. | causing Disease in Man in Nyasaland. 407 Cart 7.—Curves representing the Distribution, by Percentages, in respect to Length fo) d of 3600 Individuals of the Trypanosome of the Human Strain, and 2500 of the Wild- game Strain. Microns EE 20|21 |z2|23|24]25|26]27]28|29|30 |: |az|33]34]35|26 sal a SES ee ee eee jf [| \ | es W/o Game Strain. elisa ~--— Human Strain. Percenta Table XX.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Human and Wild-game Strains. ) | . Average, Maximum, | Minimum, Date. Strain. percentage. percentage. percentage. | | 1912 Human 21 ‘1 2°0 1912 Wild-game 26 -2 8-4 | Conelusions, 1. The five Wild-game strains resemble each other closely, and all belong to the same species of trypanosome. 2. The Wild-game strains and the Human strains, although they differ to some extent, also belong to the same species. 3. This species is 7. rhodesiense (Stephens and Fantham). 4. There is some reason for the belief that 7. rhedesiense and J. brucei (Plimmer and Bradford) are one and the same species 408 Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland.—The Wild Glossina morsitans Strain. By Surgeon-General Sir Davip Brucz, C.B., F.R.S., A.M.S.; Majors Davip Harvey and A. KE. Hamerton, D.S.0., R.A.M.C.; and Lady Bruce, R.R.C (Scientific Commission of the Royal Society, Nyasaland, 1912.) (Received February 24,—Read April 10, 1913.) Introduction. These strains were obtained by bringing tsetse flies (Glossina morsitans) to the laboratory from the neighbouring “ fly-country ” and at once allowing them to feed on healthy animals. The term “wild” is used to distinguish these caught flies from flies bred in the laboratory. The first strain was obtained by feeding the flies on a monkey, the remaining four by feeding on dogs. As soon as the healthy animal was found to be infected other animals were inoculated from it. But, as in the case of the Wild-game strains, only trypanosomes from a single rat were used for purposes of measurement and comparison. I. Morphology of Strain I, Wild Glossina morsitans. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest. ‘Table I—Measurements of the Length of the Trypanosome of Strain I, Wild Glossina morsitans. In microns. Method of staining. Date. Method of fixing. Minimum length. Maximum Average length. length. 1912 Osmic acid Giemsa 22 °7 | 34 °0 | 16 °0 Trypanosome causing Disease in Man in Nyasaland. 409 Table I1.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain I, Wild Glossina morsitans. In microns. 16. 17. 18. | ‘woh 20. | a 22 | 23 24, 25. StH -eGbooRsddoseqacoo 3 ii i 25 | ee 62 | .75 53 pa 28 44, Percentages ......... O06 | 2:72 | 5:0 | 11-2 | 12:4] 15:0) 106 | 6°2 | 5-6 | 8:8 In microns. eo | 27. | 28. | 29. 30. 31. 32. 33. 34. MINGLE)” -cecgoeesnacdodoee ae 23 26 | 22 12 6 4, 1 2 Percentages) ......... 3°2 4-6 5:2 4°, 2-4 1:2 0°8 | 02] 04 CuHart 1.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain I, Wild Glossina morsitans, taken on nine consecutive days from Rat 655. aS = ep) of -O! @ 25) = @ 1S) MS ic} a - S&F aan © wo VOL. LXXXVI.—B. 410 Sir D. Bruce and others. Trypanosome [Feb. 24, Table I1I].—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain I, Wild Glossina Morsitans, Date Ons eae Percentage among short 0. and stumpy forms. 1912 June 18 ......... 635 Rat 6 6 1) corono08e 655 5 10 py AD) cosaogon0 655 " 11 AAD pane. 655 & 17 SOD eee 655 “ 0 PEE epacboond 655 » 16 Bree ataatoco 655 5 28 by, AB cooaao008 655 %) 26 np) 20 seoesoooe 656 A; 23 pea ot 655 4 0 SO) Receatere 655 7 AOHEO) ax a 655 % 4 | Average ............ | ; 12°3 Il. Morphology of Strain II, Wild Glossina morsitans. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest. Table [V.—Measurements of the Length of the Trypanosome of Strain II, Wild Glossina morsitans. Date. 1912 In microns. Method of fixing. | Method of sree: Average Maximum Minimum length. length. length. Osmie acid Giemsa 24,°5 | 840 | 170 ee 1913. ] causing Disease in Man in Nyasaland. All Table V.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain II, Wild Glossina morsitans. In microns. ari BY. rsa iz. | 18 19. 20. 21. 22. 23. Py, || BA | Ba | | PROG, estes venesice 1 4 20 43 67 44, 48 34 42 | | Percentages ...... 0:2 0°8 4:0 8°6 13 °4 8°8 9°6 6°8 8 °4 In microns. 26. 27. | 28. 29. | 30. 31. 32. 33. 34, | | PUOtAU Soc issscen se | 35 33 43 28 23 19 13 Ps eats etl e | y | | | Percentages ...... | 7:0 66 86 56 4°6 3°8 2°6 04 | 02 | Cuart 2.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain II, Wild Glossina morsitans, taken on nine consecutive days from Rat 656. | Microns el —EE 22|23|24|25|26|27|z6|2a|30|31 |32/33|34|35|36|37\38 7 ie Bee REG co Agee eee KE 2H 2 412 Sir D. Bruce and others. Trypanosome [Feb. 24, Table VI.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain II, Wild Glossina MOTSUANS. Date. HES ene haniiale Percentage among short 0. and stumpy forms. 1912 Timiae UW oacsoonoe 656 Rat 8 5 Sh ieee 656 5p 12 Ps Olea del ae 656 a 13 NO usd teers 656 a 22 Di aa here: 656 - 43 5 EDD lee 656 44, i Phe 0000800 656 99 27 Pn WABI cauceored 656 Dp 28 sc MRO Giaer ft2it 656 2 45 Be AU Seep 8b8 656 % 47 Average ...... 28 °9 Ill. Morphology of Strain III, Wild Glossina morsitans. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the length of the longest and shortest. Table VII.—Measurements of the Length of the Trypanosome of Strain III, Wild Glossina morsitans. In microns. Method of = Date. Method of fixing. ctepsineins, Average | Maximum Minimum length. | length. length. 1912 Osmic acid Giemsa | 205 | 31-0 | 16-0 1913. | causing Disease in Man in Nyasaland. 413 Table VIII—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain III, Wild Glossina morsitans. In microns. 16. li 18. 19, 20. Ze 22. 28) Bigeale tes. secsenetcaee, LOMO 2 le 84 85 Ase |W ican Aor | 33 . | — Percentages ........... 3:8 14-4 16°8 17-0 8°8 4°8 8-0 66 In microns. 24, | 25. | AG, Pr RR | BEB ee IL ee "Gere Ne aeeeeet nea ae 27 | 18 | 21 12 11 3 4 3 | = bs encentaees ....curcene: 5°4 3°6 | 4-2 24 2-2 0°6 0°8 0°6 ' | CHART 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain III, Wild Glossina morsitans, taken on nine consecutive days from Rat 657. Microns = 22|25|24|25)| 26) 27|28|29)30 |3! |$2 |33\34|35 | 36/37 \38 14.}15]16117 |18}i9 20 |2 (yi ee! NH oP HAHA OD + =F AAEEEE aoe CI ) a a ony isi ime | : Eve | 414 Sir D. Bruce and others. Trypanosome [F eb. 24, Table IX.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain III, Wild Glossina mMorsitans. Experiment . Percentage among short PEN: No. soci and stumpy forms. | 1912 | dfmag) W7/ songoncae 657 Rat | 4 Pil nt Keli tapeadoeon 657 90 9 Soom 657 i 13 se Olseeneoucs 657 O09 4, St Die cee inen 657 a 2 Ooo Reatenle 657 ns 15 6 pon walt Seacmeeee 658 % 16 Be Sancncisae 658 5 8) a5) eedibeeennneee 658 5) a ROY et 658 * 7 By AD eooesono0 | 658 3 8 mT ie Ces 658 4 0 3) 2S) daganu0ar 658 e 10 pae9)). cea aO Soe 658 5 0) uly each 658 a 0) pai) eee rewcenees 658 ch (0) | SN. LOM sauedgunt 658 - 0 ANS ORDEES --00.00000 | 4°5 V. Morphology of Strain V, Wild Glossina morsitans. The following table gives the average length of this trypanosome as found in the rat, 500 trypanosomes in all, and also the fengeh of the longest and shortest :-— Table XIII.—Measurements of the Length of the Trypanosome of Strain V, Wild Glossina morsitans. In microns. . | ‘Method of one Pt aot | pate: Average Maximum | Minimum length. length. length. 1912 Osmie acid Giemsa | 22°8 | 35°'0 | 15°0 1913. ] causing Disease in Man in Nyasaland. 417 Table XIV.—Distribution, in respect to Length, of 500 Individuals of the Trypanosome of Strain V, Wild Glossina morsitans. | In microns. | | | 15. 16. | agepeae. | 19. }e20; | 21. | 82 23. | 24 is ene ees : et ! “Tie arses 6 | 4.1) 27 | 57 | 94 | 49 | 37 | 22 | 14 | 1B | | ] Percentages ............ 11:2|/0:8| 5°4 | 11-4 | 18:8] 9°98 | 7-4 | 4-4 | 2°38 | 2°6 | In microns. | | | i |) 2b. | 262 Bie | 28: | 29. | 30. | 31. | 82. | 83. | 34. | 35. 27 18 eee 2 | | | Percentages ............ 2:2 | 3:8 leo “46 | 5°8 54|s6| 2-0 | 1-4 | 0-6 | 0°4 Cuart 5.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain V, Wild Glossina morsitans, taken on nine consecutive days from Rat 660. HERE EEEEEEE am a2 Vee Boe eae 3 SM Se EN a fa ETIGe (leske ees ee 418 Sir D. Bruce and others. Trypanosome [ Feb. 24, Table XV.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain V, Wild Glossina morsitans. Date, Bera Avimal: Percentage among short | 0. and stumpy forms. 1912. Uibbarey WS) oan sono 660 | Rat (0) nr) eG) eabessene 660 m) @) 5h LO ae eee 660 > ) a, oat (SNe AN ae | 660 | os 8 at, JOD) ata 660 | a 19 staal eoeceal 660 5 0 Hy 28) Lorcocced 660 6 Pi sZOr weer cee 660 3) 4 wt 20 Be os 660) a> | , 6 TL ds) Scgnaetoe | 660 | 5 6 FS) coop seine: 660 | % 2 Vuh? 2) coasaopor 660 | i 2 Hew Phicadeaasce| 660 | 5 il | AV EFALC) ssp oeccs +20 4°2 Comparison of the Wild Glossina morsitans Strains with one another. Table X VI.—Measurements of the Length of the Trypanosome of the Wild Glossina morsitans Strains. | In microns. age | os Date! ee bea weceeaais yous From li O: CASTE anes Average Maximum Minimum length. length. length. | 1912 523 Monkey 500 Rat 22 7 34:0 | 16:0 1912 542 Dog 500 As 24°5 34°0 17:0 1912 | 551 eT 4500 i | 20°5 31-0 16 ‘0 1912 | 595 iy 500 if |) Ber 34-0 150 1912 | 602 i 500 r, | 228 35-0 15-0 2500 Average... 22 °6 35 °0 15-0 1913. | causing Disease in Man i Table X VII.—Distribution in respect to Length of 2500 Individuals of the Trypanosome of the five Wild Glos n Nyasaland. sina morsitans Strains. In microns. | | 15. | 16. | 17. 18. HG). 20. 21. | 22. | 23 | lr ae re Pate ae | | MROUAME .20s desivesiv sc octcacouniis yh 31 148 230 326 252 | 237 | 184 | 143 Percentages .....3....0000..- Oe} | Ie) Ge || @e4 | aeiak yt) ioe | Seat 983 || 8) oe | 4°6 | In microns. | 25. 26. | 27.} 28: |, 29: | 80: | 381. | 32. | 83. | 34. 130 | 110 | 127 | 133 | 113 96 | 54 | 44; 11 ~T Percentages .....0.cesscesecss 5°2| 4°41 5:1 | 5:3 | 4 | slas|22 18/04 0:3 a Cuarr 6.—Curve representing the Distribution, by Percentages, in respect to Length, of 2500 Individuals of the Trypanosome of the Wild Glossina morsitans Strains. Microns i516 [7 fe] 3 [eo [a [ze es] ze|25)26|27| o Percentages ee | yn &Bt YAN @ Ww alia At HT tt} tt OAC Aaa ae Bees eee sees (oa a Ba a a [ai LE UT HERE 420 Sir D. Bruce and others. Trypanosome [Feb. 24, Table XVIII.—Comparison of the Percentage of Posterior-nuclear Forms. found among the Short and Stumpy Varieties of the Trypanosome of the five Wild Glossina morsitans Strains. Dare USS enna: Percentage among short | No. and stumpy forms. 1912 | 523 Monkey 12°3 1912 542 Dog 28-9 1912 551 Bs 12°8 1912 595 % 4°56 1912 602 i 4:2 ANVIGHENE®) oon 500000 200 12°5 Comparison of the Human Strain with the Wild-game and Wild Glossina. morsitans Strains. Table XIX.—Average Length of the Trypanosome of the Human, Wild- game, and Wild Glossina morsitans Strains. | In microns. Number of | Strain. trypanosomes | Animal. measured. Average Maximum Minimum | | length. length. length. | ISEPERNAFA . son50c500000 3600 Rat 24 °2 38:0 15 ‘0 | Wild-game ......... 2500 %5 226 | 35 ‘0 15-0 Wild G. morsitans 2500 ” 226 35 ‘0 15:0 It isremarkable that the Wild-game strains and the Wild Glossina morsitans strains should come out exactly alike. Why the Human strains differ to some extent from the other two cannot at present be explained. The curves of the Wild-game strains and the Wild Glossina morsitans strains are so similar that there can be little doubt the same species of trypanosome is being dealt with. The Human strain differs so much, that the suspicion must present itself that in some way more than one species is being dealt. with. The three Human strains which differ most from the Wild-game and Wild Glossina morsitans type are Stephens and Fantham’s case of Armstrong, in Liverpool; Strain I, Mkanyanga; and Strain III, Chituluka. On examin- ing these three strains, however, from every possible point of view, nothing, except the difference in the type of curve, can be found to justify this suspicion. 1913. | causing Disease in Man in Nyasaland.” 421 Cart 7.—Curves representing the Distribution, by Percentages, of 3600 Individuals of the Trypanosome of the Human Strains, 2500 of the Wild-game Strains, and 2500 of the Wild Glossina morsitans Strains. | Fees sec ee el sl [pe ak a a a leas Mmm mmm Te Tee Ios] Wild Game Strain. aera » G. Morsitans » ee Human Strain. Table XX.—Percentages of Posterior-nuclear Forms found among the Short aud Stumpy Varieties of the Trypanosome of the Human, Wild-game, and Wild Glossina morsitans Strains. | Percentages. | Date. Strain. | | | Average. Maximum. | Minimum. | a | IQS.) “oman ieo enw 21-1 52-0 20 | 1912 Waldroamesoetes..cyes-.: 26 -2 33 °4 8-4 | 1912 | Wild G. morsitans ...... 125 28 °9 | 4-2 Conclusions. 1. The five Wild Glossina morsitans strains resemble each other closely, and all belong to the same species of trypanosome. 2. The Wild Glossina morsitans strain, the Human strain, and the Wild- game strain, belong to the same species. 3. This species is 7’. rhodesiense (Stephens and Fantham). 4. It is probable that 7. rhodesiense and 7. brucei (Plimmer and Bradford) are identical. 422 Infectivity of Glossina morsitans in Nyasaland. By Surgeon-General Sir Davin Brucs, C.B., F.R.S., A.M.S.; Majors Davip Harvey and A. E. Hamerton, D.S.0., R.A.M.C.; and Lady Brucz, IRC! (Scientific Commission of the Royal Society, Nyasaland, 1912.) (Received February 24,—Read April 10, 1913.) Introduction. The object of this paper is to put on record the proportion of tsetse flies found in this district of Nyasaland to be infected with disease-producing trypanosomes, and, further, to identify the species of trypanosomes with which the flies are infected. It must be understood that this paper only deals with the district lying near the camp at Kasu and not to every part of Nyasaland. The geographical position and other features of this district have already been described in a previous paper.* It is known as the “ Sleeping Sickness Area” or “ Proclaimed Area,” and in it almost all the cases of the Human Trypano- some Disease of Nyasaland have occurred. This disease is caused by Trypanosoma rhodesiense, and the natives give it the name of “ Kaodzera,” but there is a good deal of evidence accumulating that 7. rhodesiense is in reality 7’. brucei, in which case the disease would be known as “ Nagana.” Definitions. In this paper an infective fly is one which contains trypanosomes which have reached the final stage of their development and are capable of giving rise to disease. An infected fly is one in which the development of the trypanosomes may not have reached this final stage, and where, therefore, it may not as yet be disease-producing. An infective fly must be infected, whereas an infected fly need not be infective. “Fly” means the tsetse fly, and in this paper the species dealt with is Glossina morsitans. Methods Employed. The method employed in studying the infectivity of the flies was simple. Native boys were employed in catching the flies, which were brought up to Kasu in small cages by a native cyclist. Each cage of flies was fed on three healthy animals—the first day on a monkey, the second on a dog, and the * Supra, p. 269. Infectivity of Glossina morsitans in Nyasaland. 423 third on a goat. To ensure, as far as possible, that each animal was fed on by every fly, the flies were fed nine times—three times on each animal. The number of flies brought up each day would probably average about 60, and as each animal was fed on by three cagefuls, then each monkey, dog, and goat ran the gauntlet of some 180 flies. It is therefore impossible to arrive at any very precise knowledge of the proportion of infective flies in each cage. Infectivity of the Flies. As will be seen from Table I, every experiment, with the exception of one, was positive, and on two occasions a goat was infected by all the four species of pathogenic trypanosomes occurring in this neighbourhood. The Commission showed in a previous paper* that one in three of the wild game found in this district is infected with trypanosomes, and recom- mended that the animals should be destroyed. If this were done and a year allowed to elapse, the proportion of infective flies then found would be an index of the usefulness or futility of such operations. The Commission is of opinion that the wild game is the principal factor in the spread of trypanosome disease, and that, for practical purposes, the smaller mammals, birds, and reptiles need not be taken into account. The following table gives, in the first column, the date the first cageful of flies was fed on the monkey, the second column the number of flies fed; the signs plus and minus show the result of feeding the flies on the monkey, dog, and goat. The four species of trypanosomes carried by the “fly” in this district are T. brucer vel rhodesiense, T. pecorum, T. simiw, and T. capre. The first and second of these attack all three animals, the third the monkey and goat, being harmless to the dog, whereas the fourth only produces disease in the goat. Where no plus or minus sign occurs it means that an animal was not available. For example, the experiment beginning on January 20 shows that the monkey was infected by 7. simie, the goat by 7. brucei, and that no dog was available. The experiment on February 21 shows that neither the dog nor goat became infected by the bites of 170 flies, and that no monkey was available. * Supra, p. 277. AQ4 Sir D. Bruce and others. Infectwity of — [Feb. 24, Table I—The Result of Feeding 10,081 Tsetse Flies (@. morsitans), caught in the “ Proclaimed Area,” Nyasaland, on Monkeys, Dogs, and Goats. | Monkey. Dog. | Goat. S ie z Number | @ = é ae \! aes 4 1912. | ‘of fies | Fs) Soy | PSs (|) ee eS fed Ss) 5/8) 8) 8315) 8] &) ss) see S28 Sees Si] SL} Sts &}/ ss ef Ss SI | Se | | S|) s | SS | 8] 4) 8 | ss ie dein, 240) nocosc 296 —_ —/+]— oe en (dhl yh LEN eke 370 = —|+]-— = oth heen et eho aeaves 280 _ —/|+ _ = a, as |) = Feb DN ds ae 295 = =| + —_— — + + + x) Bh oceee 220 _ _— + _ = aL fe as Bea PS iareacae 200 _ +/+] — = alee er ” LGr egy 195 + — _— — + — I = i Le ae 170 — =i/=> i/o = Pn es ll Cr E26 ae ae 170 — = |) = | = = =) ee es Milas 62e0 con 140 = Sey ls ae = oe ese ames ” 9 winale sie) 165 _ + —_— —_ — — = + sete Ate hares 100 = ee ete = oe lee Pee fi aac os 160 = ~ | = | = = 45 "| >a »” PPA ae Kn 205 = + _— — + + + + ANS 3} spo 0ce 135 _ | oo | = = Bes ee 5p a LO Meteor 278 + —|/+]— + eo ey = = aah SPR ” LO, cess 330 = — | + _ _— + — — = e + ae pias io yaa rg 200 _— —-/)/+)]—-— _ 2 = |) = = ato eal ee 3 Stes: 180 _ —/;+]-— _ a a =: =| eee ees PRPC D aneee 230 — —|/ +] —- _ oe = eae, I) it Mt (aoln nnee 140 — +/—-]- — oe a = Bea eee || re 26 100 _ ee ee — & i} = | = + +4 = PEEL Sesoas 260 — —|/+]- +) —]—-— = fife |] 4k May Siac | 155 + |+ 4]/—-]/ - |+)/-|]-|] - |-|-]+4 ” O31 enee dss 96 = —);- _— — + — = i + = = ” SF ae 330 oP = | at = + = = == = + + ” 9 rate aleleds 120 = = | + — — + — a eats + au + ” 13. eee 50 toe =| > = + — —_— — — — + + ye WLAN eGR ee 250 = —j|;+]-— _ 2 |p ee | = fe ae ae a ” i7/ ddance 190 + + _ — — — + +. ba aa eee 113 EUW Rt eee eee Wee eh ial | a co AS) eascon 120 = =|=]= = +)/—}]— - +) —-]- itch teat ahh 230 — —|/—]—- ik =] = eas ae ae 7 Ae ” 29 weenee 320 + + — _ — + = + as AOR conebus 240 — +/—]=— — +] —]— = Bee |) se ” 29 serene 100 _ + — — — + + + Ay OILY eaetes 175 aP — ae = + — — = ah te, a June Tie heten 300 — + — = — a J ae p 6G) eae 210 = +/—]— — fe ee Pe — tt alee ARETE see 2380 + —|—-|/- + +}/—]— + fe |) = ee Oe ee 160 _ =a |} o> |) — == ae = arene ees 0) Oe oy ls} « Kesees 135 — —-;-|]- = | Sf = = ae | ae sph ac umagses 90 + —-;-|/- aL = |=) = == fs Wes | ct Sith? Bonde 95 = oe tee ee = een (We ee Sept. 25) ...... 70 — +/—] — Breed bncsneees 25 + =)}/=>] = Octie29 ere 87 + =—/}=—/—= + —|— |= +/—| + WOW 8 ceooe 145 = =]}—= |) = = cs ret gl Lion Saeed 150 — =| & | = a ppl re |e = Bip ee ert aealnldotmreroae 157 — = | os |) = = abo es | = a calles recaps 95 J o= —/—|{]— ae =} S} es su ee eee es PAVE CoO Bbc oG 180 _ —|/+]— + =j]/=/= + Seales i IDO; Gasens 180 — ap ap |) = - +/—-]— _ == ee oD cect 198 + 3) ae |S + = |=) = + —/|—-—|+ peat hoes 156 — —|/+ti— + =|] = + pe A extol sp Gi eens: 1138 - | | = a ge ar] eos Seal eet aes Total......| 10,081 9 9 | 26 14 | 34 11 85 | 17 | 35 1913. ] Glossina morsitans in Nyasaland. 425 Table I1—The Number of Times a Monkey, Dog, and Goat became infected with Trypanosoma brucei vel rhodesiense, T'. pecorum, T. simic, and T. capre in a Series of 56 Experiments, averaging 180 Tsetse Flies each. ey ee te T. pecorum. T. simie. T. capre. . I [Fe me | ra 7 | 7 aes Teal Monkey. Dog. | Goat.| Monkey.| Dog. acey Monkey.| Dog. ieee Monkey.| Dog. ee 9 14 | 11 9 | 34. | 35 26 | 0 17 0) | (0) 35 : i This shows that the monkey is less susceptible to 7. brucea and 7. pecorum than the dog, whereas it is remarkably so to Z. simiw. The dog is not susceptible to 7. simic, and neither the monkey nor dog to 7’. capre. Table IIl—The Proportion per 1000 Tsetse Flies, caught in the “Sleeping- Sickness” Area of Nyasaland, found to be Infective with Pathogenic Trypanosomes. ene ye) T. pecorum, T. simia. T. capre. Per 1000. Per 1000. Per 1000. Per 1000. 2-0 4°6 3:4 3°5 This is only allowing one infective fly to each series of flies fed on the experimental animals, and is therefore the irreducible minimum. The average number of flies fed on each animal was 180, and it might well be that there were present in the same batch several flies infective with the same species of trypanosome. Ten thousand flies gave rise to 135 infections, and taking it for granted that no fly was infective with more than one species of trypanosome, then 13:5 per 1000 flies are infective with one or other of the disease-producing trypanosomes of this district. Table [V.—Nuinber of Times the Species of Trypanosomes under consideration were found in 56 Experiments. Pee ne ee ae T.simie. | T. capra. 20 | 46 34. 35 35°7 per cent. 82 ‘1 per cent. | 60°7 per cent. 62-5 per cent. VOL. LXXXVI.—B. bo 426 Infectivity of Glossina morsitans in Nyasaland. This means that in experiments carried out in the manner described T. brucei may be expected to turn up once in every three series, 7’. pecorum eight times in ten, and 7’. simie and 7. capre six times in ten. Months and Seasons. On examining Table I it will be seen that these infective flies occur all the year round, and are just as numerous during one season as another. It will also be seen that no experiments on the infectivity of the flies were carried out during July and August. This was due to the fact that all the energy of the Commission was devoted during these two months to the study of the wild game. Conclusions. 1. The tsetse flies (Glossina morsitans) caught in the “ fly-country ” near Kasu are infected with four species of disease-producing trypanosomes— T. brucei vel rhodesiense, T. pecorum, T. simi, and T. capre. 2. The proportion of infective flies is 13:5 per 1000. 3. The proportion of flies infective with 7. brucei vel rhodesiense, the cause of the Human Trypanosome Disease of Nyasaland, is 2 per 1000. 4, The flies are found infective all the year round. 5. To prevent the infection of tsetse flies it is proposed that the experiment should be tried of destroying all the wild game in the “ Proclaimed Area” of Nyasaland. The Excystation of Colpoda cucullus from its Resting Cysts, and the Nature and Properties of the Cyst Membranes. By T. Goopry, M.Sc. (Birm.), late Mackinnon Student of the Royal Society. (Communicated by A. D. Hall, F.R.S. Received February 28,— Read May 8, 1913.) (From the Rothamsted Experimental Station, Harpenden.) Introduction. Having had occasion to make use of the resting cysts of Colpoda cucullus in the course of my work on the protozoa of the soil* I was interested by the way in which the organisms escape from the confines of the cyst membranes. The processes involved were by no means obvious on somewhat casual observation, and it became necessary to study them in considerable detail before they could be fully elucidated. Moreover it was thought that by working out as fully as possible the conditions involved in excystation some light might be thrown on the activity or inactivity of the protozoa in the soil. The water-content, available food supply, and temperature of any soil are obvious external factors in determining the possibility of protozoal activity, but that these were all the determining factors was by no means clear. There was the possibility that certain peculiar external influences were required for excystation, and if these could be determined it was possible that one would be able to say whether they were present or absent in a soil normally containing protozoa. In studying the literature bearing on the subject it was soon evident that although a considerable amount of work had been done by earlier workers, particularly by Rhumbler} on the excystation of Colpoda and the different kinds of cysts formed by it, and by Fabref on the properties of cyst membranes, little or no work was recorded dealing with the processes involved in the emergence of active forms from resting cysts. I therefore set out to investigate the matter as fully as possible. Methods——The experiments have been carried out in hanging-drop * Goodey, T., ‘A Contribution to our Knowledge of the Protozoa of the Soil,” ‘ Roy. Soe. Proc.,’ 1911, B, vol. 84, p. 165. + Rhumbler, L., ‘Zeitschr. fiir wissen. Zool.,’ 1888, vol. 46, p. 571. t Fabre-Domergue, P., “Sur les propriétés dialytiques de la membrane du Kyste des Infusoires,” ‘Compt. Rend., 1885, vol. 101; “Recherches anat. et physiol. sur les Infusoires ciliés,’” ‘Ann. des Sci. Nat.’ (7s.), ‘ Zoologie,’ 1888, vol. 5. 21 2 428 Mr. T. Goodey. The EHaxcystation of [Feb. 28, preparations. A drop of hay-infusion or other liquid medium* is taken on @ loop of platinum wire and spread out on a clean cover-slip; into the liquid is placed a number of cysts of Colpoda, and the cover-slip is then inverted over the hollow of a cavity slide and the edges waxed down by painting round with the hot wick of a candle. It is then allowed to incubate. In order to experiment further with the excysted organisms, the cover-slip is carefully lifted by means of a needle, and after the necessary treatment it is replaced and the edge waxed down again. Where not otherwise stated incubation has’ been carried out at 30° C., this being the temperature at which excystation is most rapid. Material—The investigations have been principally carried out with the resting cysts of Colpoda cucullus. A culture of this species, of a particular strain measuring about 45 w in length and having a rather pointed anterior end, was obtained free from any other ciliated protozoa. By making sub- cultures of this from time to time in sterile 1-per-cent. hay-infusion, and allowing the Colpoda to go on multiplying and encyst, quantities of resting cysts were obtained. For convenience the cysts were collected on small filter papers and kept in an air-tight glass dish. A quantity of cysts of a larger strain of Colpoda cucullus was also obtained, but these did not prove to "be so convenient for manipulation as the smaller strain. It was found that if the cysts were kept for some weeks on the filter paper in a quite dry condition, their power of excysting rapidly was con- siderably diminished. Excystation was most rapid in the case of cultures made from cysts recently filtered and almost air dried. Influence of Temperature on Rate of Excystation—Hanging-drop cultures were made with cysts in hay-infusion and put into incubators at different temperatures, 40°, 30°, 25°, and 20°C. The cysts were all from the same collection and the hay-infusion from the same stock solution, so that conditions were all similar, except for temperature. The results were as follows :-— 40° C., none excysted after several hours. 30° C., many active after 1 hour incubation. 25° C., a few active after 1 hour 17 minutes. 20° C., a few active after 2 hours 12 minutes. It is evident that differences in temperature have a considerable influence on the rate of excystation. It was not possible to determine exactly the * Distilled water, tap-water, and aqueous soil-extract were used at different times. Excystation takes place quite freely in distilled water. 1913. ] , Colpoda cucullus from its Resting Cysts. 429 optimum temperature for excystation, owing to there being no incubators at other temperatures, but it lies close to 30° C. Meunier* obtained active Colpoda two hours after moistening resting cysts, and Fabref also obtained active forms in the same time. Influence of Alkaline, Acid, and Neutral Media on Excystation—The ordinary 1-per-cent. hay-infusion used for cultures of Colpoda and other protozoa is made slightly alkaline in reaction to litmus by the addition of caustic soda solution and contains 0°01 per cent. NaOH. Colpoda cucullus excysts very freely in this liquid and also in neutral 1-per-cent. hay-infusion. When, however, the hay-infusion was made slightly acid in reaction to litmus, excystation appeared to be completely inhibited, and from these preliminary trials the inference was tentatively drawn, that excystation was inhibited in an acid medium. ’ It was necessary, however, to determine more accurately the critical percentage strength of acid and alkali at which excystation is inhibited. NaOH.—1-per-cent. hay-infusion was put up in 10 cc. lots, each with a different percentage strength of NaOH, starting at 0°01 per cent. and going up in hundredths of 1 c.c. to 0:2 per cent. Excystation was rapid and free in all the cultures below 0:15 per cent. NaOH. At this strength, however, only a few were active after two hours’ incubation and these were apparentiy rather uncomfortable. At 0°16 per cent, one or two were trying to excyst, 0-17 per cent. showed one or two revolving within their endocysts but unable to get out quickly, at 0°18 per cent. there was no motion, though the contractile vacuoles in one or two were dilated, 0:19 per cent. gave the same appearance as 0:18 per cent., and at 0-2 per cent. there was no indication whatever of excystation. After 21 hours’ incubation active forms were found in all, up to and including 0°18 per cent. and 0°19 per cent., but none were active in 0:2 per cent. This then is the critical percentage strength of NaOH which inhibits excystation, whilst 0-18 per cent. may be taken as the critical strength for short-period incubation. HCl.—For hydrochloric acid a number of 10 c.c. lots of 1-per-cent. hay- infusion were put up, each with a different percentage strength of HCl. starting at 0°001 per cent. and going up to 0-01 per cent. in thousandths of 1 cc. and then in hundredths of 1 c.c. up to 0-1 per cent. After one hour’s incubation at 30° C. there were many active in all the cultures up to 0°01 per cent. At 0-08 per cent. there were a few active, * Meunier, V., “Sur la résistance vit. des Kolpodes ‘encystés,’” ‘Compt. Rend.,’ 1865, vol. 61. + Fabre-Domergue, P., loc. cit. 430 Mr. T. Goodey. The Excystation of , [Feb. 28, at 0:09 per cent. only one or two, and at 0-1 per cent. none were active; 0:1-per-cent. HCl is then the critical strength which inhibits excystation. Colpoda excystation can therefore take place within fairly wide limits in an alkaline medium containing 0°18 per cent. or 0:19 per cent. NaOH and in the presence of 0-09 per cent. HCl. Tests for the Nature of the Ecto- and Endo-cysts. Fabre found that the ectocysts of resting cysts could withstand the action of concentrated sulphuric acid for a long time, and also caustic potash solution. He failed to obtain a cellulose reaction with iodine and sulphuric acid, whereas Stein obtained a wine-red coloration in this manner with Vorticella microstoma, and believed that the cyst-membranes of this organism were composed of a substance combined with cellulose, which could be dissolved out by the action of caustic potash. In the course of the present investigation the following tests have been applied in order to determine the nature and characteristics of the ectocyst and endocyst membranes of Colpoda cucullus. Solubility in Water—Both ectocyst and endocyst are insoluble in cold water and in water or 1-per-cent. hay-infusion at 95-100° C. Acids.—Sulphuric acid, strong, does not affect either ectocyst or endocyst in the cold. Acetic acid, 90 per cent., has no action on ectocyst or endocyst in the cold. Hydrochloric acid, strong and cold, causes the ectocyst to swell up slightly, but does not dissolve the endocyst. On gradually heating up cultures containing ectocysts and endocysts with 2-per-cent. hydrochloric acid, and keeping the temperature at about 97° C. for half an hour, the endocysts disappeared, whilst the ectocysts remained somewhat swollen. Alkalies.—Caustic soda: 1 per cent., 2 per cent., and 4 per cent. do not dissolve ectocysts or endocysts in the cold, though they penetrate freely into the endocysts and attack the Colpoda within, causing them to swell up and become transparent. At 30° C., 1 per cent. and 2 per cent. still do not attack the endocyst membrane, though 4 per cent. causes its solution at this temperature. Twenty-per-cent. caustic soda, acting in the cold, causes the ectocyst to swell up considerably and become transparent, though there still remain indications of the layers making up this membrane. Caustic potash: 1 per cent. and 2 per cent. do not dissolve ectocysts or endocysts in the cold. Fat Solvents: Alcohol, 95 per cent., ether, toluene, and chloroform do not dissolve the ectocyst or endocyst when added to a culture which has been exposed to the action of osmic vapour for a few seconds in order to kill the active and excysting organisms. 1913. | Colpoda cucullus from its Resting Cysts. 431 Formalin: 40-per-cent. formalin does not dissolve either ectocyst or endocyst in the cold. Prowazek* mentions that Kutscher found paramylum was soluble in formalin. Protein Tests—Xanthoproteic test: cultures containing ectocysts and endo- cysts were carefully heated with concentrated nitric acid and then ammonia was added, but no coloration resulted. Millon’s reagent was used on two or three occasions, but there was no red coloration of the cyst-membranes, though by this method protein was detected in the protoplasm of Colpoda by the brick-red coloration. Starch and Cellulose Tests——Iodine in potassium iodide solution does not stain ectocysts or endocysts, though it passes through the latter very readily and stains the Colpoda light brown. Preparations treated with iodine are not affected when strong sulphuric acid is added. Ammoniacal cupric hydrate does not dissolve either ectocyst or endocyst. Corallin soda solution does not stain the endocyst pink. These negative reactions show that the cyst- membranes are not composed of starch or cellulose. The following carbohydrates occur in certain protozoa :— Glycogen} has been observed in a number of ciliated protozoa, Opalina, Paramecium, and Vorticella. It stains light or reddish brown with iodine. Paraglycogent occurs as refractive spherules in certain of the Sporozoa. With iodine it stains brown, which changes to wine-red or violet on the addition of 70 per cent. sulphuric acid. It is soluble in hot water. Paramylum§ is a carbohydrate nearly related to cellulose. As prepared from Huglena viridis by Biitschli it did not stain with iodine in potassium iodide solution nor was it affected by the addition of 70-per-cent. sulphuric acid. It is insoluble in cold and hot water and is hydrolysed by continued boiling with strong sulphuric acid. It is evident that the cyst membranes of Colpoda are not composed of glycogen or paraglycogen. They resemble paramylum in their reactions to iodine and in the fact that the endocyst disappears on being heated up with acid. It will be shown later on, however, that they are not composed of paramylum. Reactions to Stains—The following substances were tried in order to determine what staining reactions are given by the cyst-membranes :— Picric acid, a strong aqueous solution : membranes not stained. * Prowazek, S. von, ‘ Einfiihrung in die Physiologie der Einzelligen (Protozoen),’ Berlin, 1910, p. 13. + Prowazek, S. von, loc. cit. ; Biitschli, C., loc. cit., pp. 1469-72. i Biitschli, O., loc. cit., pp. 1469-72 ; Minchin, E. A., “The Sporozoa,” ‘A Treatise on Zoology, Pt. 1, 2nd Fas., 1903, p. 182. § Biitschli, O., “ Kenntniss des Paramylons,” ‘ Arch. fiir Protist.,’ 1906, vol. 7, p. 199. 432 Mr. T. Goodey. The Excystation of [Feb. 28, Methyl green, saturated aqueous solution + 1 per cent. acetic acid: membranes not stained, contents of endocyst immediately stained. Iodine green: same result as with methyl green. Safranin, strong aqueous solution: membranes not stained, cyst contents freely stained. Gentian violet, strong aqueous solution: membrane not stained, cyst contents stained. Eosin, strong aqueous solution: membranes not stained, cyst contents stained. Carbol fuchsin, the strong bacteria stain: membranes not stained, contents of endocyst stained deep red. Hematoxylin (Heidenhain’s): film-preparations made by hatching out Colpoda in 1-per-cent. hay-infusion + egg white and killing the excysting forms with osmic vapour, were stained with iron hematoxylin. The ectocysts stained dark blue or purple; the endocysts stained the same tint but not so intensely. Hzematoxylin (Delafield’s): in films prepared as for Heidenhain’s, but stained with this preparation, the ectocysts stained purple, the endocysts a pale reddish or bluish purple. From these records it will be seen that only the hematoxylin stains touch the cyst-membranes. This is interesting, for Fabre considered that many aniline stains affected the cyst-membranes of ciliates. The Nature of the Process of Excystation.—Bitschli* says that littleis known on the manner in which the cyst-contents are reorganised prior to excystation. No doubt water penetrates into the interior causing the organism to swell up and the contractile vacuole to begin pulsating. The cilia reorganise them- selves, but how they do so is not accurately known. The imbibition of water doubtless plays an important part in the rupturing of the ectocyst. He also suggests that instead of Colpoda escaping from the temporary division-cysts through the narrow aperture, they may perhaps have the power of dissolving the membrane in a particular place. Many observers have recorded the fact that the ectocyst is caused to rupture, as Biitschli suggests, by the increased volume of the cyst contents owing to the imbibition of water. The exact manner, however, in which the organism manages to get out of the endocyst does not appear to have been determined. It is on this interesting point that the present investigation throws light. In carefully watching the movements of Colpoda within the transparent endocyst, it is noticed that the organism rotates freely and that the endocyst gradually increases in diameter; the wall becomes thinner and thinner until, . * Biitschli, O., ‘Bronn’s Klassen des Thierreichs,’ “ Protozoa,” Abt. III, p. 1664. 1913. | Colpoda cucullus from its Resting Cysts. 433 finally, it becomes invisible, at which time the Col/poda swims away. The whole process only occupies from 5 to 10 minutes, when incubation has --- BcFocyst taken place at 30°C. It seems as if the endocyst is gradually dissolved or digested so that the enclosed organism may be liberated. In order to determine as far as possible the nature of the process, the following methods were adopted :— Killing Excysting Colpoda with Different Reagents—Hanging-drop cultures were made in the usual way, and when there were several Colpoda rotating within their endocysts the cover-slips were lifted and the cultures exposed for a short time to the action of the vapour of some particular reagent capable of killing the organisms already liberated and those still excysting. The cover-slips were then waxed to the slides again. When volatile anti- septics were used, such as xylene, toluene, chloroform, and carbon disulphide, the organisms were killed within 30 seconds, and then it was observed that the endocyst still continued to increase in diameter, the wall becoming gradually thinner and thinner until finally it disappeared. This was determined accurately by taking measurements of the endocyst at different times with an eye-piece micrometer. In some cases the endocyst disappeared in the course of 10 to 15 minutes, and in other cases after the lapse of a few hours. When osmic vapour and vapour of 40 per cent. formalin were used the Colpoda were killed in a few seconds. It was then noticed that the endo- eysts did not continue to swell up and disappear, but remained exactly as they were, in diameter of cyst and in thickness of wall, at the moment of killing. There was no alteration even after 24 hours. This interesting difference between the effect of volatile antiseptics and osmic acid and formalin vapour indicates that the process of excystation from the endocyst is normally effected by the secretion of a solvent or digestive ferment. The antiseptics, toluene, chloroform, ete., kill the excysting organisms but not the enzyme which it is secreting, and thus. the 434 Mr. T. Goodey. The Hucystation of [Feb. 28, process of endocyst-digestion continues. On the other hand, formalin and osmic acid vapour kill the organism and also the secreted enzyme, hence the digestion of the endocyst is stopped. After the organism has been killed by a volatile antiseptic the further digestion of the endocyst is not so rapid as when the organism is actively swimming about within, but this may be due to the fact that normally an excysting Colpoda has a considerable mechanical effect upon the thin endocyst wall as it swims about, no doubt aiding in its more rapid destruction. Tests on Soluble Starch.—In order to test the effect of the enzyme secreted by Colpoda on soluble starch an agar medium was made in the following proportions+100 e.c. 0°05-per-cent. solution of soluble starch, agar-agar 0°8 grm. Two sterile Petri dishes were poured with this medium, and each was inoculated with cysts of Colpoda which had been well teased out in sterile distilled water. There were ten or twelve areas of inoculation on each plate, each about 3 mm. in diameter. After incubating at 30° C. for an hour and a quarter, active and excysting Colpoda were found on many of the inoculated areas. One or two Colpoda were seen to emerge from their digested endocysts, and at these spots traces of iodine in potassium iodide solution were added. A uniform blue coloration resulted over the whole of the region; there were no clear uncoloured zones in the region of the digested endocysts. As controls, traces of the iodine in potassium iodide solution were added in close proximity to the inoculated areas. These gave exactly the same tone of blue as in the areas where Colpoda had excysted. It is evident therefore that the enzyme secreted by Co/poda during excystation is not capable of digesting soluble starch. Tests for the Digestibility of the Endocyst by Different Ferments.—Having determined that the endocyst is digested by enzymic activity, it was deemed advisable to test the digestive powers of pepsin and trypsin upon it. Hanging-drop cultures were made in the usual way with resting cysts and the excysting organisms were killed at the proper moment with osmic acid vapour. Two or three cultures were then supplied with a platinum loop or two of 0:04-per-cent. pepsin in 0°2-per-cent. hydrochloric acid,* and two or three supplied with an equal quantity of 0°04-per-cent. trypsin in 0°5-per- cent. caustic soda solution. All the cultures were then waxed down again and put into the incubator at 40° C. Examinations were made at different times to ascertain the effect of the * The pepsin and trypsin were made at this strength according to the instructions in Hoppe-Seyler’s ‘Handbuch der Physiol. und Pathol. Chemisch. Analyse,’ 7th Kdit., 1903. 1913. ] Colpoda cucullus from its Resting Cysts. 435 ferment, but in no case was the endocyst digested either by pepsin or trypsin even after remaining for several days at 40°C. Pepsin had the effect of turning the Colpoda almost black, whether they were free or enclosed within the endocysts. Trypsin on the other hand gradually digested the Colpoda already free and those within the endocysts, rendering them very transparent and causing complete solution and digestion of all the organism except certain very refractive granules. These two digestive ferments were tried on many occasions but at no time were they found to attack the endocyst. Diastase.—The diastase used was prepared from fresh pale-barley malt by the method described by O’Sullivan.* According to Brown and Escombe,t diastase acts best in very dilute acid solutions and they recommend 0:006-per- cent. formic acid. The diastase used was therefore dissolved in formic acid of this strength. One lot was heated up in a water bath at 62-65° C. for about 20 minutes in order to kill off any cytase which might be present. Two solutions of diastase were thus obtained, one heated and the other unheated. Hanging-drop cultures were made and the excysting Colpoda killed with osmic vapour. One or two platinum loops of the diastase solutions were added and then the cover-slips were waxed down again and the culture put into the incubator at 40° C. Twenty-four hours after the addition of the diastase no endocysts could be found even after careful search; they had all been digested, both hy the heated and the unheated diastase. Similar results were obtained by repeating the experiment on other occasions. Ptyalin.—Saliva was collected and diluted slightly with distilled water. It was then filtered and a small crystal or two of thymol added to prevent putrefaction. A platinum loop or two of the filtered liquid was added to hanging-drop cultures, which were then allowed to incubate at 40° C. When perfectly fresh saliva was used the endocysts were digested somewhat slowly, ae. in 48 hours. Some of the cultures made did not show the digestive action, whilst others did. It is possible that a certain amount of change must first be effected by the enzyme secreted by the Colpoda within, before ptyalin can act on the endocyst. The digestive powers of this ferment on the endocyst are therefore uncertain. From these experiments it is evident that the endocyst consists of a substance which is not digested by pepsin or trypsin, but which is * O'Sullivan, C., ‘Trans. Chem. Soc.,’ 1884, vol. 45, p. 2. t+ Brown and Escombe, “On the Depletion of Hordewm vulgare during Germination,” ‘Proc. Roy. Soc.,’ 1898, vol. 63. 436 Mr. T. Goodey. The Hacystation of [Feb. 28, completely digested by diastase and in some cases by ptyalin. It is therefore highly probable that this endocyst material is of a carbohydrate nature. This raises a point of considerable interest, for the endocyst was digested by the heated as well as by the unheated diastase, thus showing that the seat of the enzymic activity was not confined to any cytase which might have been present in the diastase as an impurity.* Diastase is generally considered to act only upon starch and glycogen. The substance under consideration is certainly neither starch nor glycogen and yet it is digested by diastase. In this respect also it differs from paramylum, which is untouched by diastase. Paraglycogen is soluble in hot water, and the solution is hydrolysed into dextrin and a trace of reducing sugar, thus differing from glycogen, which yields dextrin and plenty of reducing sugar when hydrolysed with the salivary ferment. It is because of this difference that Biitschlit called it paraglycogen. Since diastase digested the endocyst, chemical tests for carbohydrates were sought. Attempts were made to obtain an osazone from the liquid in which Colpoda had been caused to excyst, after proper treatment with sodium acetate and phenyl-hydrazine hydrochloride solutions, but only negative results were obtained. It was concluded that the carbohydrate produced by the digestion of the endocyst was present in far too small a quantity to be detected by this method. Colour Reactions for the Detection of Carbohydrates—A methodt was found by which dextrose, lactose, saccharose, starch, and cellulose, when heated up with strong hydrochloric acid and scatole, yield a violet coloration, the reaction still showing at a dilution of 1: 300,000. Fifteen grains of scatole were obtained, and tests were made with dextrose, starch, lactose, saecharose, cellulose, and hay-infusion, by heating them in test-tubes with strong hydrochloric acid and small quantities of scatole. In all cases a violet coloration of the liquid resulted, even when dextrose diluted 1 : 300,000 was used. Hay-infusion could not, therefore, be used for the excystation of Colpoda from cysts in this experiment. Cultures were therefore made in distilled water containing 0:01-per-cent. caustic soda, this being the percentage of alkali in the hay-infusion commonly * Occasionally one or two fibres of cellulose from the filter paper on which the cysts had been collected were introduced into the hanging drops along with the cysts. These fibres never showed signs of corrosion or solution, thus showing the freedom of the diastase from cytase. + Loe. cit., p. 1484. } ‘Jour. Chem. Soc.,’ 1907, vol. 92, Pt. 2, Abst., p. 308 (“Colour Reactions of Carbohydrates with Indole and Scatole ” ). 1913.] Colpoda cucullus from its Resting Cysts. 437 used, The object of the experiment was to obtain a colour reaction for the presence of a carbohydrate in the culture-liquid containing the products of the digestion of the endocysts by the enzyme secreted by the Colpoda. The cultures were made in the hollows of cavity slides, and the cysts of the large variety of Colpoda cucullus were used. After a large number of organisms had become active, the cultures were unsealed, and the liquid was taken up in a capillary pipette and transferred to a small glass tube. This liquid was practically free from ectocyst membranes, though it is possible there may have been a few present, for it was impossible to examine the liquid under the microscope. There may also have been one or two fine threads of cellulose present from the filter paper on which the cysts had been collected from the original culture of Colpoda. To the liquid was added an equal volume of strong hydrochloric acid, and the mixture was boiled for half an hour on the water-bath, in order to hydrolyse any maltose present to dextrose. A little scatole was added and a drop or two more of strong hydrochloric acid. On gently heating this mixture over the Bunsen flame a pale purple coloration resulted, indicating the presence of a carbohydrate in the liquid. On heating up some of the original culture-liquid, ze. distilled water containing 0:01 per cent. caustic soda, with strong hydrochloric acid and scatole, no coloration was obtained, nor was any coloration produced when distilled water alone was tested in this way. The liquid, then, which had been used for the excystation of Colpoda clearly contained a carbohydrate. That the carbohydrate was there as a product of the digestion of the endocyst of Colpoda cannot be definitely asserted, since there may have been present in the liquid a trace of cellulose, or ectocyst membrane. Whether this supposed cellulose and ectocyst would be sufficient to account for the colora- tion produced cannot be stated, but I consider it negligible, and think it does not vitiate the result of the test. No special emphasis, however, is laid on this test as an indication of the carbohydrate character of the endocyst. The fact that it is digested by diastase and fails to show any reactions to protein and other tests is sufficient to show that it is a true carbohydrate. Note on the Endocyst of Gastrostyla steinii, A few cultures were made with the resting cysts of Gastrostyla steinii, and several active organisms were found after three hours’ incubation at 30° C. There appear to be ectocysts and endocysts in the resting cysts of this organism as in Colpoda. The process of excystation is slower than in Colpoda cucullus, and many 438 Mr. T. Goodey. The Excystation of [Feb. 28, organisms appear to rotate and cause digestion of the endocyst whilst still within the confines of the ectocyst, afterwards making their exit from the latter by forcing their way through a rupture in the wall. Some free endocysts were tested with iodine in potassium iodide solution but they did not stain, nor was any alteration in colour or shape effected by the addition of 70-per-cent. sulphuric acid to the preparation. This indicates that the endocyst of Gastrostyla steinii is in all probability of the same nature as that of Colpoda cucullus. Note on the Initial Stages of Encystation in Colpoda cucullus. On examining a hanging-drop culture in which Colpoda had been active for two days many organisms were found to be encysting, probably com- mencing to form resting cysts, since there was not an abundance of food. An opportunity therefore presented itself for the study of the earliest stages of encystation. Whilst examining under the oil-immersion a group of three rounded Colpoda which were revolving, each within a limiting membrane, a free- swimming Colpoda came into the field. It did not move away but began to revolve, the cilia moved very violently, and the organism began to lose its characteristic form and outline and became more rounded. It appeared as though in this initial stage of encystation some jelly-like substance or mucilage were exuded from the ectoplasm, and that because of this the cilia had great difficulty in moving. They seemed to move in bunches which gave the appearance of waves on the surface of the organism. Shortly after this there was a definite limiting membrane within which the organism was revolving, clearly propelled by means of its cilia, which were easily visible under the oil-immersion. This is in accord with Rhumbler’s observations. He watched the forma- tion of the cyst membranes in division cysts and resting cysts of Colpoda, and was convinced that rotation within the cyst was due to ciliary activity. Summary and Conclusions. 1. The cyst membranes of Colpoda cucullus consist of the outer ectocyst and the inner endocyst. The ectoeyst is insoluble in strong acids, gives no reaction with iodine and strong sulphuric acid, only stains with hematoxylin, is insoluble in alcohol and ether, and is only dissolved by 20-per-cent. caustic soda. In all these properties except the absence of any reaction to lodine, it resembles the outer cyst wall of Huglena viridis, investigated by Biitschli, which is composed of a nitrogen-free carbohydrate compound. 2. The endocyst is composed of a transparent substance which is insoluble 1913. | Colpoda cucullus from its Resting Cysts. 439 in cold or hot water, strong acids, fat solvents like alcohol and ether. it is soluble in 4-per-cent. caustic soda at 30° C., and fails to give any reaction with iodine and strong sulphuric acid. It is not digested by pepsin or trypsin but is completely digested by diastase acting at 40° C., and is sometimes slowly digested by the salivary ferment, ptyalin, also at 40° C. It is thus of a carbohydrate nature but differs from all other carbohydrates which have been found in the protoplasm or secreted by the protoplasm of protozoa. It seems to consist of a substance allied to glycogen, para- glycogen and paramylum. The name Cystose is proposed for it. There is evidence to show that the endocyst of Gastrostyla steinii is composed of a similar substance. 3. During the normal process of excystation the endocyst is set free by the rupture of the ectocyst and the Co/poda liberates itself by the rapid digestion of the endocyst by means of a powerful enzyme which it secretes. This enzyme is put out of action by killing the excysting organisms with formalin or osmic acid vapour, but continues to act on the endocyst when the organisms are killed with toluene, ether, chloroform or carbon bisulphide vapours. It ean act when the medium surrounding the cyst contains 0°19 per cent. of caustic soda on the one hand and 0:09 per cent. of hydrochloric acid on the other, and may therefore be said to act in alkaline, neutral and acid media. It cannot act at 40° C., since excystation is inhibited at this temperature, but it appears to act best at 30°C. In these respects it is remarkable and differs strikingly from diastase, the activity of which is retained even as high as 68° C., but is entirely checked in the presence of less than 0:04 per cent. of caustic soda. A further point of difference from diastase is found in the fact that the new enzyme fails to digest soluble starch. The name Cystase is proposed for it. General Conclusions—The endocyst membrane of Colpoda, and probably of other ciliated protozoa forming double-walled resting cysts, is composed of a carbohydrate which is different from all carbohydrates previously described. To this substance the name “Cystose” is given. During the process of excystation the endocyst is digested by a powerful enzyme secreted by the enclosed Colpoda. The name “Cystase” is proposed for this ferment. My best thanks are due to the Lawes Agricultural Trust for allowing me to carry on the work at this Laboratory. I am also indebted to various members of the Laboratory staff and other friends for helpful suggestions, particularly to Mr. W. A. Davis, Organic Chemist here, for suggesting the names “ Cystose” and “ Cystase.” 440 Protostigmata in Ascidians. By A. G. Huntsman, B.A., M.B., Biological Department, University of Toronto. (Communicated by Prof. A. B. Macallum, F.R.S. Received March 3,—Read April 24, 1913.) CONTENTS. PAGE Gradestofserotostiouabaneseree sec eeeeeeneeret tern eer eee Ee 44] Protostigmatic Condition in various Groups .............+. 442 HactorsiotestlemabOSenesis)eaeradereracreeteee rece eeeeeeere teen: 444 Origin of Stigmata in various Groups ..........-sceeseeeeeeee 445 SUMmMamyeam deluiberacuee nee een te-ceeecsecer comet er eee ee Rete 452 Introduction. It has been shown that at a certaiu stage in the ontogeny of Ascidians there is usually a definite number of stigmata or gill slits, which are elongated dorsoventrally and arranged in a longitudinal series on each side of the pharynx. To these the name protostigmata was given by Garstang in 1892. The stigmata of the adult differ from these in being usually very numerous, indefinite in number and elongated antero-posteriorly. In 1904 Julin and Damas made two separate proposals to base the classifi- cation of the Tunicates on the condition of the protostigmata. As these proposals differed rather widely from each other and from the current classifications based upon the adult condition, J have taken up the question as far as it concerns the Ascidians and have investigated the origin of the stigmata in a large number of genera. I have been able to confirm for myself very many of the observations that have been published regarding the origin of the stigmata in the genera Holozoa [| Distaplia], Clavelina, Ciona, Corella, Ceswra [Molgula], Botryllus and Dendrodoa (Styelopsis). In addition J have investigated the genera Amaroucium, Polycitor [Distoma], Sycozoa | Colella], Ascidiopsis (very near Ascidiella and Phallusia, which have been studied by others), Chelyosoma, Styela, Boltenia, Pyura [Cynthia pars] and TLethyum [Cynthia pars]. The last three genera belong to the family Tethyide [Cynthiidze], no members of which have been previously investigated as to the origin of the stigmata. My interpretation of the course of development in these genera tends to show that the protostigmatic condition supports the classification advanced by Seeliger and revised by Hartmeyer.* ‘ * Bronn’s ‘KI. u. Ordn. des Tierreichs.’ Protostigmata in Ascidians. 441 Grades of Protostigmata. In fig. 1, C represents diagrammatically the left side of the pharynx of a young Ciona. It can be clearly seen that there are two grades of protostig- mata, Counting from the anterior end (left side) the first, fourth, and fifth are longer than the second, third, and sixth. The former are the first to appear and they perforate the pharyngeal wall independently of each other. They give rise to the latter three as indicated in the figure, by their ventral ends extending and turning upward. They then divide attheangles. De Selys and Damas (1901) have proposed the term primary protostigmata for the three arising by independent perforation and the term secondary proto- stigmata for the six resulting from their division. But the fourth protostigma differs from the first and fifth in having its concave face forward and in turning forward and upward before division Nl (tte Fie. 1.—Various types of protostigmatic condition. The left side of the pharynx is shown, the anterior end being toward the left. instead of backward and upward. Willey (1893), De Selys (1901) and Damas (1904) have recorded instances in Ciona and Ascidiella, in which the first and fourth protostigmata were continuous ventrally (before the second and third had been formed). Willey concluded from this that the first four protostigmata were derived from a single primary gill slit. This view seems to be justified by the facts. The fourth behaves as if it had been derived from the first in the same way that the sixth is derived from the fifth. The fourth is equal in size to the first because it is derived from the latter at such an early stage, before perforation. They are both able to repeat the process and thus form four stigmata before the change occurs that brings about the formation of the definitive stigmata. The fifth is able to divide once only. I conclude therefore that in Ciona the stigmatic rudiment or anlage divides into two parts placed one behind the other. They correspond in position to VOL. LXXXVI.—B. 2K 442 Mr. A. G. Huntsman. [ Mar. 3, the first and fifth. These may be called the protostigmata of the first order or primary, using the term proposed by de Selys and Damas but in a some- what different sense. They correspond to a certain extent with the “ primary gill slits” of Willey and the “ fentes branchiales” of Julin (1904). This early division of the stigmatic rudiment produces elements with a similar orientation. When they elongate to form the protostigmata, their concave surfaces are posterior. The subdivision of the two primary protostigmata produces four proto- stigmata of the second order or secondary. They correspond in position with the first, fourth, fifth, and sixth. The two members of each pair are oriented differently, one being the reverse of the other. The process of subdivision is repeated in the case of the first pair, giving four protostigmata of the third order or tertiary, namely, the first, second, third and fourth. Here also of each pair one is the reverse of the other. There are no Ascidians in which there has been shown to be further sub- division to form new protostigmata. The protostigmata are formed in two ways. First, by simple subdivision ; second, by modified subdivision, resulting in the intercalation of new stigmata. By these same processes acting in the same order, the proto- stigmata are converted into the rows of definitive stigmata of the adult. Frequently the two processes are not altogether distinct. An intermediate method of subdivision may occur. The Protostigmatic Condition in the Various Groups. Julin and Damas have blazed the way for a comparison of the various. kinds of stigmatogenesis to be found in Ascidians. Both, however, have based their classifications on definite stages, and have more or less ignored the differences in the process. Damas (1904) uses as a basis the final protostigmatic condition, and divides the Tunicata into the Mono-, Di-, Tetra-, Hexa- and Poly-pro- stigmata. The Cesiride [Molgulide] are separated from the Tethyide [Cynthiidee], Styelide, etc., and placed with the Cionide, Phalluside, ete. And, as I shall show subsequently, some of the Tethyidz would have to be placed in the Hexa- and others in the Poly-prostigmata. Damas also. places the Pyrosomatide and Doliolide with the Tethyide, ete. These are unnatural groupings. Julin (19042) uses as a basis the number of stigmata that perforate independently and calls them “fentes branchiales.” He makes three distinct. groups: those with one pair of “fentes branchiales,” those with two pairs, and those with three pairs. To the first group belong the pelagic Tunicata,. 1913. ] Protostigmata in Ascidians. 443 Appendicularians, Salpide, Doliolide and Pyrosomatide ; in the second group are the Krikobranchia and the Perophoride; and in the third group all the simple Ascidians. These groups are natural enough, but I cannot agree with his interpretation of the development in many forms. Having to prove that all the stigmata of the adult are derived from a limited number of first stigmata, he puts a rather forced construction on many of the facts. If his method were followed out fully, all the stigmata in each and every Tunicate could be derived from a single pair of stigmata. I think that this is a legitimate conception, but it does not lead to any divisions within the group. The Appendicularian is without doubt a primitive form. The Doliolidz, Pyrosomatide and Ascidiaceee show remarkable diversity in the methods by which a large number of stigmata are derived from a single pair corresponding with those of the Appendicularian. It is by studying these methods that we get an insight into the affinities of the various forms. Mere numbers at any definite stage are of subordinate importance to the process. The rather definite separation of the process into two parts, (1) the formation of a longitudinal series of protostigmata, and (2) the trans- formation of these into transverse rows, makes it possible to institute a comparison of the early stages alone. In fig. 1 I have represented diagrammatically various types of protostigmatic condition, indicating not only the number but also the method of origin. A represents the condition in the Krikobranchia, a group consisting of the majority of the compound Ascidians. I agree with Julin and Damas in considering that two protostigmata are represented in this group. True protostigmata are never seen. The process of division into definite stigmata begins before perforation, as will be shown subsequently. The two protostigmata are evidently of the second order, and represent a single primary protostigma. B represents the condition in Perophora, modified from the description by Damas. Four protostigmata of the third order are present, representing a single primary one. Here also typical protostigmata do not occur, and the interpretation is doubtful, as will be shown subsequently. C represents the condition in other Dictyobranchiates (Ciona, Phal- lusia, ete.). This condition has been described above. There are four tertiary protostigmata, representing a single primary, and behind these two secondary ones, representing another primary. These forms are peculiar in possessing two primary protostigmata. D shows the condition in the Cesiride and some Tethyide (Pyuwra and Tethyum [Cynthia]). There are six secondary, representing three primary protostigmata, 2K 2 444 Mr. A. G. Huntsman. ~ [Mar. 3, In E is shown the condition in some Tethyide (Boltenia) and some Styelide (Styela). There are four secondary, followed by an indefinite number of primary. F represents the condition in some Styelide (Dendrodoa (Styelopsis)) and in the Botryllide. There are an indefinite number of primary and no secondary protostigmata. Julin has given an entirely different account of the origin of the protostigmata in Dendrodoa (Styelopsis) from that given by Damas. My investigation of this genus is in accord with the account given by Damas. It seems possible that Julin’s material has been wrongly identified as belonging to this genus. Damas (with Garstang and Seeliger) considers the single row of stigmata in the Pyrosomatide and Doliolide as being a series of protostigmata. Julin (with Lahille) considers the row as homologous with one of the transverse rows of stigmata in an Ascidian. I favour the latter view. Damas’ objection that the longitudinal bars of Pyrosoma would, in that case, not be homologous with those of Ascidians is valid, but they are readily homologised with the internal transverse vessels of Holozoa and of many simple Ascidians, as done by Lahille (1890). The series of fig. 1 shows gradation in two respects. There is a continuous increase in the number of primary protostigmata, and, excepting A, there is a continuous decrease in the ability of the protostigmata to subdivide, this decrease being greater posteriorly than anteriorly. The variation shown in the families Tethyide and Styelide is noteworthy. The former family contains some forms with a limited number (three) and others with an indefinite number of protostigmata. Up to the present, a relatively small proportion of the existing genera have been investigated. Further studies will doubtless show even greater variations. The Factors of Stigmatogenesis. In interpreting the various stages that may be examined, it is important to keep in view the factors involved in the process. The Ascidian pharynx offers a splendid field for a study of the processes of development, since great variations are shown in the different groups, and the processes are practically confined to two dimensions, 7c. one plane. As the basis of the process we have, of course, the properties of the cells and their interaction with each other and with the surrounding water. These determine the ways in which the cells divide and arrange themselves to give the gross characters that are seen. Certain gross factors (depending upon the above), which vary in the different groups, may be mentioned. 1913. ] Protostigmata in Ascidians. 445 1. Growth.—This takes place chiefly at the ends of the stigmata, and results in elongation. 2. Perforation.—This is the appearance of an opening in a fused area of the pharyngeal and atrial walls. It may occur early or late. 3. Subdivision.—This is the division of a stigma by the growth across it of a bridge. It may occur early or late in the growth of a stigma; it may occur at a varying point on the stigma; it may occur before or after perforation. 4. Orientation or Direction of Growth.—This factor determines the shapes and the positions of the stigmata with reference to each other. It depends upon equality or inequality of growth of the cells on either side of a stigma. Great variation is shown in this factor. By variations in these four factors there results the infinite diversity in the stigmata of Ascidians. A comparatively slight variation may make a great difference in the final picture, and a similar end result may be attained by following quite different paths. One should avoid giving too much signifi- cance to certain differences shown. It is certainly astonishing to see, in a young colony of Botryllus, the first blastozoid provided with four rows of minute definitive stigmata that have arisen by independent perforation, and the oozoid beside it with six large protostigmata that never become converted into definitive stigmata. The Origin of the Stigmata in the Various Groups. In fig. 2 the methods of development of the protostigmata in the various groups are schematically represented. They all are seen to be derived from a common starting point, a single stigma or rudiment not yet perforated. The stigmata represented in solid black are those that have not yet perforated. Many of them are hypothetical. Usually a dark mass can be seen at the point where a stigma will shortly perforate. Perhaps the simplest method is the fourth of the figure, that occurring in the Botryllide and some Styelide. The protostigmata appear in order beginning in front. Each stigma is formed by independent perforation at a point behind the middle of the previously formed one. Julin (1904a) has maintained that in Dendrodoa (Styelopsis) grossularia the protostigmata arise in a manner analogous to that of Ciona described above. Both Damas (1904) and Fechner (1907) describe their origin by independent perforation and give excellent figures. In an allied species (Dendrodoa carnea) I have found no indication of intercalation or of new protostigmata being formed from pre-existing ones, the conditions being the same as figured by Damas and Fechner. 446 Mr. A. G. Huntsman. [Mar. 3 \ / \ a NO \ ‘ / \ 7; \ \ / \ \ / ‘ \ / SOnguamnre wef peer omeinaes (2 Boa \ / 00 oo Oe 00 \ .---\V 5 : (o} (e} . at ee \ y Une = Oh nace 67 = = 2 eee Ye fo} fo) ‘ 0 0 i 5 oS / See aN VA 2 S \ >» L fle ATs: peels a2 Or 42: - VeaNes ates) 0000 DOCO \ LG eeghe ete RO Se oe aS crts : PPP hails siege Masten ten? \ oy Ommol Ono \ nT hoes ) alt) \ \ es. @ . \. Cel pide aya i. LO NOROOK 5.70 On ©) CCl AA eeocee eeee On LOD TOO Cotta oe ee oO O G © Veg fF OF e e ee ¢ 1 ha whe Fie, 2.—Schematic representation of variations in early stages of development of stigmata in 1913. ] Protostigmata in Ascidians. 447 Cesira ae Meritt siei@elole aiclelsvervierese:s Hugyra i; Ceesiridee 7 Pyura > | PL fdas cnaets ines Tethyum roe | Perce ers er ese eseeereceee Boltenia sia Geeisantbeit dee | Styela | Styelide oe a | Dendrodoa Botryllus Botryllidx J errs Chelyosomatide y Ascidiopsis | Mliglaiwilelle eiele(elisielele v.c.clale'sie eels < eS , Phallusiidee | Phallustopsis : | Ciona Cionide Pe PE ile c sata Perophora Perophoride J Clavelina ne Beer nAT oes ease een 72 Cha inaendia } Clavelinide 5 Holozoa 7 200 AGE a SEE ea Cystodites ‘ Krikobranchia. Sycozoa (Polycitor) Polycitoride | Polycitor J | ee) Amaroucium Synoicide J Ascidians. The stigmata of the left side are shown, the anterior end being toward the left. 448 Mr. A. G. Huntsman. [ Mar. 3, The number of protostigmata is indefinite. Damas gives more than 15, Julin 12 at least. In the Botryllide the number is smaller ; from 5 to 7 have been observed. Their method of formation has not been established. The majority appear about the time of metamorphosis, when observation is difficult. I can affirm independent perforation for the sixth in a species of Botryllus from Naples. The others in size show no evidence of intercalation. In the Cesiride we have in many respects the other extreme. The number is definite and intercalation is regular. Three- stigmata perforate in the usual order and each one by turning upward at the ventral end and dividing gives rise to a second placed immediately behind it, as is shown in the first series of stages in the figure. Six protostigmata arranged in three pairs are thus formed. The same method is followed by Pyura haustor, one of the Tethyide. I have observed in this species the derivation of the second and fourth from the first and third, but not of the sixth from the fifth. Only six protostigmata are formed and their first subdivision to form the definitive stigmata is the same as that which has been described for the Ceesiridee. A slight modification of this method is to be found in another Tethyid, Tethyum [Cynthia] pyriforme. The turned up portion of the stigma is cut off very early, probably before the opening 6f the stigma has extended into it. Ihave not observed all the stages but can affirm that the second is intercalated from the first and that only six protostigmata are formed. Also in the earliest stage examined only one stigma is present. A further modification occurs in the genera Boltenia (Bb. ovifera and B. villosa) and Styela (S. gibbsiz) of which I have examined a full series of stages. The second and fourth are formed by the abstraction of the ventral growing tips of the first and third. Subsequently these perforate and extend dorsoventrally. The fifth and all succeeding protostigmata arise in order by independent perforation at points along the middle of the side of the pharynx. They give no indication of being derived from pre-existing ones. Their number is indefinite. I have observed stages with 11 in Styela and with 9 in Boltenia. The two genera differ greatly in the method by which the definitive stigmata are formed from the protostigmata. The method just described for these two genera is an approach to the method of Dendrodoa and the Botryllide. The third primary protostigma has lost its ability to form a new one by intercalation. If the first and second likewise lost that power we would have the condition in Dendrodoa. All the forms so far considered belong to the Order Ptychobranchia. This group is therefore characterised by having three or more primary protostigmata, which perforate in regular order. Secondary protostigmata may 1913)] Protostigmata in Ascidians. 449 or may not be formed and are never more numerous than six: This Order is sharply marked off from the other two Orders. Inside the Order, the four methods of formation of protostigmata do not correspond with the four families of the Order. The Tethyids exhibit three methods and the Styelids two. But comparative anatomy has shown that some of the Tethyide approach the Cesiride and,.others approach the Styelide. And in a similar way the Styelide show affinities with both the Tethyide and Botryllide. The protostigmatic condition therefore corroborates the classification based upon the adult condition. There are two distinct methods in the Dictyobranchia. ‘The most general ‘one is that described above for Ciona. The first stage is the appearance of two stigmata of about the same size. From their occasional connection with each other and from their subsequent behaviour, they are very evidently halves of an original single element which is doubtless homologous with the first stigma of the Ptychobranchiate. They are secondary protostigmata, corresponding with one primary. Their probable history before perforation is indicated in the figure. The original element has turned up at its lower end in the usual fashion and then divided into two. The two secondary protostigmata subdivide into four tertiary, and at the same time a second primary appears and divides into'two secondary ones. The new stigmata are connected for a short time with the old in Ciona, and this has been shown in the figure, In gy “sessheaonasooooedecouber | 3°0 71:1 | SPIO) Ce INYO)” oy) Tespobopanoboccucconoo cc 3°8 | 63 “4 oft TI) GOS INET RAS eso cob bon cceeodecnunasec 6:9 33 °6 SETI) GG, TN CNY 2 eaocanecosececsordccaccs | 10 °4 0 bd) @xGs INA fh, aesoobenor caedborednoree | 10-4 0) 474 Mr. 8. B. Schryver. Investigations on the __[Mar. 6, ‘Other Methods of Obtaining Clots. The General Character of Clot Formation. The experiments already quoted seem at first sight to fully confirm the theory as to the general action of ferments and inhibitory substances indicated in the introduction. They do not, however, indicate the actual chemical process of clot formation, and the whole subject is somewhat com- plicated by the fact that caseinogen itself is very labile. Further investiga- tions were therefore necessary to ascertain the nature of the chemical processes involved in clot formation, and to determine whether the change of caselnogen was a necessary preliminary, and whether calcium or any other alkaline earth was an essential constituent of the clot-producing system. During the course of these additional researches, it was found that it was possible to produce a clot by other methods. It was found that the calcium caseinogenate solution alone, and without the addition of another calcium salt, could produce a clot on the addition of rennet. This clot formation, which can take place at room temperature, was, however, completely inhibited when the caseinogenate solution was diluted with a equal volume of milk serum, 7.c. when the milk serum constituents were in only half the concentration in which they exist in milk. In the presence of such quantities of inhibitory solutions, the presence of an additional calcium salt is necessary to produce a clot. As caseinogen can be converted into metacaseinogen by water, it is conceivable that a similar reaction would take place more readily in the presence of rennet. In this case calcium caseinogenate should form a mixture of free metacaseinogen and calcium metacaseinogenate and there would then exist in the system more metacaseinogen than is necessary to saturate all the calcium present, for, as has already been shown above, a saturated solution of this substance in % sat. Ca(OH), contains only about a third of the amount of nitrogen contained in a corresponding solution of caseinogen. As a matter of fact there is evidence, which is given in greater detail below, that the clot formed in the presence of rennet is produced from metacaseinogen. It is conceivable, therefore, that the clots are formed from the free caseinogen or meta- caseinogen directly and not from the calcium salts. This supposition is also confirmed by other facts. By reference to the table given on p. 464 it will be seen that } saturated lime water dissolves less than half the amount of caseinogen or metacaseinogen dissolved by half-saturated lime water, whereas completely saturated lime water dissolves only very little more than double the amount dissolved by the z sat. Ca(OH): Repeated experiments on the solubility of natural caseinogen in ¢ sat. Ca(OHL)2 gave solubility numbers varying between 8 and 11 instead 1913. ] Phenomena of “ Clot” Formations. 475 of the number 17 or 18, which would have been expected, had the solubility been one-half of that in 4 sat. Ca(OH)2. As these figures indicate hydrolytic dissociation of the sesh aat salt in low dilutions, it is not unreasonable to suppose that the saturated solution of caseinogen in 4 sat. Ca(OH), undergoes a similar hydrolysis when the temperature is jibes Now attention has been called to the fact that caleium chloride does not produce a clot with calcium caseinogenate at room temperature, but only when the mixture is slightly warmed (e.g. to 25°). If, however, the calcium caseinogenate is first treated with carbon dioxide gas, which produces by itself no precipitate, the subsequent addition of calcium chloride rapidly produces a clot in the cold. Finally, it is possible to produce a clot from sodium caseinogenate solutions in the absence of calcium salts. Rennet by itself produces no clot from such solutions; if, however, they are treated first with carbon dioxide, the addition of the ferment solution causes clot production after a short interval, even in the cold. There is evidence that the sodium salt does not readily undergo hydrolytic dissociation (see table on p. 464). Carbon dioxide can apparently decompose the sodium salt and set free sufficient was caselnogen to allow the clotting process to take place. Saturated solutions of metacaseinogen (prepared by the treatment of “natural” caseinogen with water for 24 hours at 37°) in } sat. Ca(OH). do not clot on the addition of rennet, and yield only a very faint precipitate on addition of the ferment after previous treatment with carbon dioxide. The action of calcium chloride on these solutions has been already described. By means of these reactions, metacaseinogen can be readily distinguished from caseinogen. Caseinogen, after solution in alkalis and reprecipitation with acetic acid, can yield solutions in } sat. Ca(OH)s, which clot on the addition of rennet or calcium chloride. preenenmiore. clots can be produced by the same method from metacaseinogen which has been reconverted into caseinogen by treat- ment with alkalis (with the usual precautions), provided that the former has not been changed too much by very prolonged action of water either at 37° or at higher temperatures. Where such additional changes have been brought about, there is evidence of the partial scission of phosphoric acid from the caseinogen molecule. In no case was it found possible to reconvert a metacaseinogen into a caseinogen with quite as high a solubility in 4 sat. Ca(OH), as “natural” caseinogen. Alkali appears, therefore, to exert some slight subsidiary action. * Cf. Brailsford Robertson. VOL. LXXXVI.—B. 2N 476 Mr. 8. B. Schryver. Investigations on the [| Mar. 6, The Chemical Nature of the Clot. As already stated, clots, whether produced by calcium chloride alone, or by rennet alone, can be distinguished from metacaseinogen or caselnogen pre- cipitates by the fact that, when moist, they yield an indiarubber-like mass on treatment with alcohol, whereas the unclotted material undergoes no visible change. A systematic examination was made of the clots prepared by various processes with the object of determining their relationship to caseimogen and metacaseinogen. The rennet clot was formed by adding 1 c.c. of a com- mercial rennet solution to 50 cc. of a saturated solution of “natural” caseinogen in $ sat. Ca(OH )s. The mixture was made in a high cylinder in the cold, and was then placed in an incubator. After about 10 minutes, the whole had set to a solid clot, which was broken up, filtered off from the liquid, washed with ice-cold water, alcohol, and ether, and then air-dried. It was then boiled with absolute alcohol for about 10 minutes to destroy the ferment, and analysed. A portion was then dissolved in weak sodium hydroxide solution and reprecipitated, after filtration through pulp, by acetic acid, rapidly washed with ice-cold water after precipitation, and freed from water in the usual way. The clot produced by calcium chloride was prepared by mixing the calcium caselnogenate solution with an equal volume of N/25 calcium chloride solution, and incubating the mixture until the clot had formed. This was then filtered off, washed with 50-per-cent. alcohol until the washings were free from chlorine, then with absolute alcohol and ether, and then air-dried. To free it from calcium it was treated in the same way as the rennet clot, we. redissolved in alkali (NaOH), and reprecipitated with the usual precautions. It was found that the clots produced by rennet alone and by calciuin chloride alone differed in one important particular, for whereas the calcium chloride clot (purified by solution in alkali and reprecipitation) gave with 4 sat. Ca(OH). a milky solution of high solubility (more than 30), which clotted on addition both of rennet and calcium chloride, the purified substance from the rennet clot yielded with the lime water of the same concentration only an opalescent solution (solubility 125), which gave a precipitate with an equal volume of N/25 calcium chloride solution, and no clot on addition of rennet. ; The rennet clot differed therefore both from metacaseinogen and caseinogen, in that, even after re-solution in alkali, it no longer gave rise to a product capable of giving with 4 sat. Ca(OH): an opaque highly concentrated solution, from which clots can be formed. 1913. ] Phenomena of “ Clot” Formations. | 477 Experiments were next ‘carried out to ascertain whether the treatment with alcohol had caused the difference between the calcium chloride and the rennet clots. For this purpose a calcium chloride clot was obtained in the form of a dry powder, by the method described above, and then boiled for 10 minutes with alcohol, and purified by solution in alkali and repre- cipitated. The preparation thus obtained also gave with $ sat. Ca(OH), an opaque solution of high concentration (29°7) which did not clot on addition of rennet, and gave only an incomplete clot on addition of calcium chloride. Alcohol, therefore, alters both the calcium chloride and the rennet clots. The fact may be recalled that caseinogen on heating with alcohol is not altered, but yields a solution im $ sat. Ca(OH)s, which readily clots both on addition of rennet and calcium chloride. The clot produced by calcium chloride alone, however, on re-solution in alkali, is readily converted into easeinogen, whereas the clot produced by rennet alone, even after redis- solving in alkali, yields a product of low solubility in } sat. Ca(OH). from which no clot could be produced. It was unfortunately necessary to heat the rennet clot with alcohol to destroy the ferment, and as there is evidence that the calcium chloride clot is also altered by this treatment, the question as to whether the rennet clot can be reconverted into a clottable caseinogen remains for the present unsolved. Numerous other experiments were carried out with the object of preparing a clottable calcium caseinogenate solution from the rennet clot. The ferment in one experiment was destroyed by heating with water, and then after drying with alcohol and ether was redissolved in sodium hydroxide solution and reprecipitated by acid. In another experiment the clot, after washing with ice-cold water, was directly dissolved (whilst still moist) in sodium hydroxide solution, reprecipitated, dried in the usual way, and then heated for a few minutes with boiling alcohol, a treatment which causes no change in caseinogen. In no case was a product of greater solubility than 17 to 20 in $ sat. Ca(OH), obtained, which is only about half that of natural caseinogen. The solutions were in all cases translucent and yielded no clot either with calcium chloride or with rennet. The evidence obtained so far tends to indicate that rennet alters caseimogen in such a way that it is not reconvertible into caseinogen. The fact that both descriptions of clot are altered by alcohol, whereas caseinogen is not, indicates that the latter undergoes some change during the process of clot formation. The fact, however, that the rennet clot, even after re-solution in alkali, gives a product of low solubility in lime water, whereas the calcium chloride clot gives one of high solubility (even after treatment with hot alcohol), indicates that the caseinogen under the influence of rennet 2N 2 478 Mr. 8. B. Schryver. Investigations on the __[ Mar. 6, is converted into metacaseinogen, which undergoes some further alteration, whereas in the absence of rennet it can form a clot without undergoing this change. The clot produced by the direct action of rennet on milk, which was prepared in the same way as the rennet clot from pure calcium caseinogenate solution, behaved in the same way as the latter, yielding after purification a product of low solubility in lime water. It contained, as the analyses* indicate, products other than those derived from caseinogen, carried down from the milk. The analyses show furthermore that the nitrogen—phosphorus ratios in caseinogen, metacaseinogen (unless heated for too long a period), and in the clots produced both by calcium salts and rennet are the same. There is no evidence, therefore, of any proteoclastic digestion produced by the rennet. The change of caseinogen into meta- caselnogen is not an essential for clot formation, which can, furthermore, be inhibited by the presence of various adsorbable substances. Owing to the lability of caseinogen, especially under the action of rennet, it is not possible to conclude from the experiments on milk-clot formation that the ferment exerts a direct antagonising influence on the substances inhibiting aggrega- tion, although such an action is not by any means improbable. Nor is it possible from the above experiments to directly formulate the chemical changes which take place in clot formation. The evidence points to the fact that it is the free caseinogen which is changed. This substance, as a complex polybasic acid, can conceivably undergo many changes by the simple scission of the elements of water, and although it is not possible to express in the form of an equation the conversion of caseinogen into casein, such a change does not appear to be at all mysterious when considered from a chemical standpoint. Summary and Conclusions. 1. A preliminary account is given of the action of calcium salts on sodium cholate (cholalate). When solutions of these substances are mixed, a clot is formed on heating. Investigations were carried out with the object of determining the relationships between the clotting time and the amounts and characters of the calcium salts. It was found that, im the case of . those salts which raise the surface tension of water, the greater the con- centration of the salt the shorter was the time required for clot formation. In the case of salts which lower the surface tension, on the other hand, increase of concentration decreased the clotting time only up to a certain limit of optimal concentration. Above this limit the clotting time was diminished, or the clot formation inhibited entirely. The more a salt * These analyses are not published in this communication. 1913. ] Phenomena of “ Clot” Formations. 479 lowers the surface tension of water the narrower the limits of con- centration within which clot formation is impossible. The inhibition of intravascular clotting after peptone injection is probablya similar phenomenon. 2. This inhibition of clotting is probably due to the adsorption of simple molecules by the more complex colloidal substances, which are thereby inhibited from aggregation to form a clot. The results suggested that in other cases, such as that of milk, the materials necessary for clot forma- tion pre-exist, but that aggregation is prevented by the adsorption of simpler molecules from the system. The conception was formed that a ferment, for which the colloidal substances could act as a substrate, could clear the surface of such substances of adsorbed bodies and thus allow aggregation (clot) formation to take place. If such an action of ferments takes place it might be possible to explain the function of the intracellular ferments. If they act in the manner suggested, an aggregation equilibrium in the system—colloids (proteins, ete.), simpler adsorbable substances (extractives, etc.), ferment—would be maintained and would be probably necessary for the maintenance of the normal functions of the cell. There would, in this respect, exist a contrast between the “solid” tissues and the fluids of the body. 3. In attempting to apply this hypothesis to explain the clotting of milk, efforts were made to obtain a “natural” caseinogen. It is already known that caseinogen forms with alkalis solutions of very acid salts, and con- siderable differences were found in the individual preparations with regard to the amount of caseinogen dissolved by alkalis. The solubility in half- saturated lime water was employed as the criterion for differentiating the various preparations. It was found that if caseinogen is prepared in such a way that it is allowed to remain for as short a time as possible with acetic acid used for its precipitation (1 in 1000), a product is obtained which gives an opaque milky fluid containing nearly 8 per cent. of caseinogen. If such a preparation is heated with water at 37°, or allowed to stand with the acetic acid (1 in 1000) at room temperature, it gives rise to a product, the solubility of which in lime water is only about + of that of natural caseinogen. This has been designated “ metacaseinogen,”’ the solution of which in half- saturated lime water is opalescent and not opaque. Metacaseinogen can be reconverted into caseinogen by solution in sodium hydroxide and precipita- tion with acetic acid, provided that the precautions are taken that the precipitate does not remain too long in contact with the acid. The solvent capacity of sodium hydroxide approximates to that of an equimolar (not equinormal) solution of calcium hydroxide. 4. The action of calcium chloride solutions on a saturated solution of 480 Mr. 8. B. Schryver. Investigations on the [ Mar. 6, caseinogen in sodium hydroxide equimolar with half-saturated lime water “was investigated. It was found that a precipitate was formed (apparently by double decomposition) only when the concentration of the calcium salt was within certain definite limits. The reactions form an “irregular series” similar to many others where one of the reacting substances is a complex colloid. If rennet is added to a mixture in which precipitation is inhibited by excess of calcium salt an action takes place, and a precipitate is formed shortly after the addition of the ferment. 5. If the optimal amount of calcium salt is present, precipitate formation can also be inhibited by the presence of milk serum, Witte’s peptone, or even glycine. The addition of rennet to such mixtures can cause precipita- tion, provided that not too much inhibitory substance is present. The amount of calcium salt also influences the reaction, which depends therefore on the relative quantities of various products present in the system. The relative influence of these substances has not yet been investigated in detail. The precipitating power of calcium salts other than the chloride has also been investigated. 6. If solutions of calcium chloride or the salts of another alkaline earth are added to a saturated solution of natural caseinogen in 4 sat. Ca(OH): no precipitate is formed at room temperature. If, however, the mixtures are warmed slightly (to 25°), a typical clot is formed within certain limits of concentration of the calcium chloride. This shrinks, and gives on treatment with alcohol an indiarubber-like mass, and behaves, generally, in a manner characteristic of the milk clot obtained by rennet. In optimal concentrations of the calcium salt, clot formation can also be inhibited by the presence of milk serum, Witte’s peptone and glycine. The addition of rennet to mixtures containing these inhibitory substances can cause the clot to form directly. 7. There is reason to believe that the clot is formed from free caseinogen or metacaseinogen and not from the calcium salt. The chief are :— (a) The clot after addition of calcium chloride forms only on warming, and there is evidence that the calcium caseinogenate, under these conditions, undergoes hydrolytic dissociation. (6) The clot can, however, form in the cold, if the caseinogenate solution is previously treated with carbon dioxide. (c) The clot can be formed from sodium caseinogenate solutions, in the absence of calcium, if the latter are treated first with carbon dioxide and then with rennet. 8. Calcium (but not sodium) caseinogenate solutions clot on addition of rennet in the absence of calcium chloride. Clot formation under these conditions is, however, inhibited by relatively low concentrations of milk serum. : 1913. | Phenomena of “ Clot” Formations. 481 9. Solutions of calcium metacaseinogenate are, under optimal conditions of concentration, only incompletely precipitated by calcium salts, and do not clot on addition of rennet. 10. There is evidence that the clot formed on addition of rennet alone is formed from metacaseinogen, as it has a low solubility in half saturated lime water, whereas that formed by addition of calcium chloride alone is formed from caseinogen. Rennet appears also to cause some further change, as up to the present all attempts to reconvert the clot into natural caseinogen (by action of alkalis, etc.) have failed. In this respect, the casein differs from metacaseinogen. Clottable caseinogenate solutions can, however, be readily prepared from clots produced by the action of ealeium chloride alone. 11. It is not possible to formulate accurately the relationships of caseinogen, metacaseinogen, and the clots to one another. Analyses negative the suggestion of anything of the nature of proteoclastic digestion on addition of the rennet. The products are possibly formed from one another by the scission or addition of the elements of water from or to the acid hydroxyl groups, and possibly the various products bear the same relation to one another as do, e.g. the pyro-, ortho-, and metaphosphates. 12. Although many of the facts appear to support the hypothesis as to ferment actions given above (para. 2), the same cannot be said to be definitely proved by the facts elicited in the study of the phenomenon of milk clotting. The process is rendered more difficult of comprehension by the peculiar instability of caseinogen. There is no doubt, however, that in milk the clot formation depends upon the presence of four series of substances in the system, viz, simple inhibitory substances, colloids, ferment and calcium salt, even if their relative actions cannot be formulated in as simple a manner as that suggested. 482 On the Action of Radium Rays upon the Cells of Jensen's Rat Sarcoma. By S. Russ, D.Sc.,* and HELEN CHamBeErs, M.D. (Communicated by H. G. Plimmer, F.R.S. Received March 14,— Read May 29, 1913.) [Puates 12 anp 13.] The experiments of B. H. Wedd and one of ust have shown that if freshly excised portions of mouse carcinoma are exposed to X-rays or the B-rays from a few milligrammes of radium for a comparatively brief period (crea 1 hour), the irradiated material will not grow on subsequent transplantation. This line of experimental work has here been extended to Jensen’s rat sarcoma, the initial material for which was kindly provided by the Imperial Cancer Research Fund. From several points of view this tumour provides excellent material for the investigations in question. Inoculations are successful in practically 100 per cent., for out of 125 inoculations into normal rats, 124 gave growing tumours; the rate of growth is rapid, tumours measuring 2 x 2 cm. frequently being obtained in 15 days after the inoculation of 0-1 c.c. of tumour emulsion. Spontaneous absorption of the tumours, however, is not uncommon; five disappearances have occurred in 53 rats, all of which were under observation for a minimum period of 20 days. The Inhibitory Effect of Irradiation by B-Rays. Thin slices of rat sarcoma from a rapidly growing tumour were exposed between sterile sheets of mica to the 8-rays from a source of radium, having an intensity of 1:58 megrm. per square centimetre. Small pieces of the irradiated material were then inoculated into the right axille of a number of normal rats, into the left axille of which small pieces of non-irradiated tumour were also inoculated. The amount of tumour tissue which can be irradiated in this way is necessarily small and the inoculations were, therefore, made with a hypodermic needle and stilet, the pieces of tumour tissue being as nearly as possible of the same size. This procedure was adopted for the same tumour material for three different periods of irradiation, 7.e. 30 minutes, 14 hours, and 3 hours, * Part of this work was done during the tenure of a Beit Memorial Fellowship. t+ ‘Journ. Path. and Bact.,’ 1912, vol. 17. —s-- Radium Rays and Cells of Jensen’s Rat Sarcoma. 483 six animals being inoculated for each series. They were examined at frequent intervals, and fig. 1 gives in outline one half the actual sizes of the ’ tumours which formed. DA YS AFTER why NOCULATION. 10 '% Uf 22. Ti MOD MER SJE keke Fre. 1. It will be seen that as a result of irradiation for 30 minutes there is a slower growth of the irradiated than of the untreated tumour. When the period of irradiation is increased to 90 minutes, the inoculated material, although apparently increasing in size for some days, was in all cases eventually absorbed in the animal which was simultaneously supporting the growth of the control tumour. Extension of the period of irradiation to three hours ensures the progressive and complete absorption of the tumour cells. There is a close similarity in the action of the @-rays in retarding or preventing the growth of rat sarcoma tissue to the results recorded for mouse carcinoma by Wedd and Russ in the paper to which reference has been made. The Action of Radium Emanation on Tumour Tissue. Preliminary experiments were made by mincing a tumour with a Haaland mincer and adding to it sufficient normal saline for the mixture to flow into a small glass bulb. Radium emanation was then supplied in a concentration of about 0°5 millicurie per cubic centimetre. After 32 minutes 0-1 cc. of 484 Drs. Russ and Chambers. Action of Radiwm [Mar. 14, the irradiated emulsion was inoculated into each of six rats, 01 ce. of a control portion of the emulsion being inoculated at the same time into the opposite axille of the same animals. No tumours developed from the irradiated material, but in each animal a rapidly growing tumour formed from the control emulsion. An experimental series was undertaken on similar lines in order to determine the dosage of irradiation necessary to prevent the growth of the sarcoma tissue. Tumour emulsion was exposed to 0°275 millicurie per cubic centimetre, samples being withdrawn after 15, 30, and 60 minutes; 0-1 cc. of each sample was inoculated into each of six rats, 01 cc. of the original emulsion being inoculated into an equal number of other rats, making 24 in all; the average weight of the rats in these four series was 39, 34, 42, and 43 grm. respectively. Measurements of the tumours resulting were made by means of callipers at frequent intervals. In fig. 2 are recorded the sum of the superficial areas of the growths in = ] in zg Cs : & i ed i: z= 1 a g é 8 ; _ < A a! / ConTROLS ze ja ——— NE SI j K @ LT @) q K / Z lene | a Aiea mS a Oa Ge Gesaal cee pe, pees ee eae eS ee ef ws a LZ £ O DAYS 70 20 Fie. 2. the six rats of each series till 27 days subsequent to the inoculations. The tumours in the control animals showed a vigorous growth; in Series I, ae. after irradiation for 15 minutes, a phase of apparent inactivity lasting 1913.] Rays upon the Cells of Jensen's Rat Sarcoma. 485 about eight days was followed by an almost equally rapid growth. For the more prolonged periods 30 and 60 minutes, zc. Series II and III respectively, the initial reaction, resulting in easily measureable nodules, showed gradual signs of absorption, until after 27 days there was every indication that complete disappearance of the nodules would result. One rat of Series II was killed at this stage, the small nodule was found to consist almost entirely of sarcoma cells (vide Plate 12, microphotograph 1). Continued observations, however, showed that in three of the rats of Series II, nodules remained palpable for a prolonged period and eventually they developed into tumours. In Series III this occurred in one rat and resulted in a fairly rapidly growing tumour. The gradual decline in size of the initial nodules and the subsequent growth of the tumours are depicted for each of these four animals in fig. 3. The full-line curve marked III a Fy es ; oO fi ce) 4 A &%, a : G a ; < en f W fe L 4 leaersens/: i é 2 ‘the r Al } Fre. 3: corresponds to the single occurrence in Series III and the dotted curves to the three rats of Series II. The nodules in the two remaining rats in Series II disappeared after 24 and 81 days. On re-inoculation with 0:1 cc. of tumour emulsion the first was refractory and the second yielded a growing tumour. 486 Drs. Russ and Chambers. Action of Radium ([Mar. 14, Of the five remaining animals in Series III one died, the growths in two of the rats disappeared in 30 and 38 days, the animals proving refractory to subsequent inoculation. The nodules in the two remaining were palpable for 116 days, when they were excised and found to consist of fibrous tissue. To attempt to interpret the course of events illustrated in figs. 2 and 3 two alternatives may be considered. If it be supposed that all of the irradiated cells suffer some damage dependent upon their time of exposure, then the irradiated series might be expected to show a general quantitative sequence, and this appears to be the case. On these lines the tumour cells appear to overcome the effect of their irradiation after a prolonged period in the animal body. On the other hand if it be supposed that some cells are unaffected by the rays, the delay in the apparent onset of growth would be proportional to the time of exposure, and would depend on the number of cells left undamaged. Although the observations do not allow of a decision between the alterna- tives, they show that the irradiated cells increase at a slower rate than do the controls. The areas of the tumours were found by actual measurement, and if these areas are raised to the three halves power, numbers are obtained which are proportional to the volumes of the tumours. On the simplest assumption of continuous cell proliferation and reckoning from the time when growth has certainly started, it is found that the mean life period (T) of the tumour cells in the animals of the different series vary in the following manner :— Control cells (mean of 6 animals) ... aos ne) = Sides: + hour irradiated cells (mean of 6 animals) co p=) onc me $ 55 ‘ 3 nf Lee oOo mae 1 ms i (one animal) ae LO ade 0 aa Attempted Re-activation of Irradiated Tissue. The changes produced in the tumour tissue by the irradiation may be internal or external to the cells, or the cell boundaries may be affected. If a tumour be minced by a Haaland mincer, as has been done in this work and the emulsion be vigorously centrifugalised, about 1 ec.c. of fluid may generally be pipetted off from 20 grm. of tissue. If the failure of the tumour to grow after irradiation be due to changes occurring external to the cells, such changes might possibly be counteracted by taking tissue that had been irradiated sufficiently long to prevent its growth and adding to it fluid obtained from non-irradiated tumour tissue in the manner indicated. This re-activation test has been put into operation three times, the extent of irradiation having been 1 hour 4 mins. 2 hours 35 mins., and 5 hours 1913.] Rays upon the Cells of Jensen's Rat Sarcoma. 487 40 mins., to a concentration of 0°37, 0:40, and 0:37 millicuries per cubic centimetre respectively. The technique followed was practically the same in each case. To some of the irradiated tumour emulsion an equal volume of fluid obtained from non-irradiated tumour tissue was added and after allowing the fluid to permeate the irradiated tissue for about 1 hour, 0-1 cc. of the emulsion was inoculated into a number of normal rats (7.2. 6 or 8). To another portion of the irradiated tumour emulsion an equal volume of normal saline was added, and 011 ¢.c. of the mixture inoculated into (6 or 8) other rats to serve as controls. At the same time 01 cc. of fluid only was injected into a number of rats. No reaction was detected when fluid only was injected. In no one of the cases was the attempt at re-activation successful to the extent of the ultimate production of a growing tumour; indicating that the changes occurring in the tumour tissue as a result of irradiation cannot be counteracted by the action of non-irradiated tumour fluid, and that the irradiation probably causes some change in the cells themselves. Charts of the animals show that 17 days subsequent to the inoculation of the irradiated emulsion treated with fluid, 16 animals out of 22 showed palpable nodules, compared with 4 out of 21 of the control animals. This result suggests that normal tumour fluid has some action upon the irradiated cells, which delays their absorption by the animal although ineffective in re-activating them, Histological Examination. To study the histological changes which occur in the irradiated material after inoculation, three series of rats (36 in all) were inoculated on one side with 0-1 cc. tumour emulsion, and on the other side with 0:1 cc. of the same emulsion which had been exposed to a concentration of about 0:45 millicurie per cubic centimetre for periods of 20 minutes (q@), 80 minutes (0), and 24 hours (c). These times of exposure ensure that the grafts will (a) be slightly delayed in growth, (4) just fail to develop into tumours, and (¢) show no signs of proliferation, respectively. An animal from each series was killed each day for the first week, and then at intervals until the 22nd day after inoculation, the control and irradiated tumours were excised and sections prepared. Microscopical examinations of the emulsions, after irradiation and before their inoculation into the rats, failed to establish any differences between them and the non-irradiated portions. 488 Drs. Russ and Chambers. Action of Radiwm [Mar. 14, Control Grafts. The day after inoculation the tissue of the graft is almost entirely necrotic, only those sarcoma cells at the edge look in good condition. There is extensive invasion of the graft with leucocytes and much inflamma- tory cedema of the surrounding tissue. On the second day active proliferation of the sarcoma cells at the periphery is evident, and they soon form an encircling ring of growth around the graft. Sarcoma cells extend outwards into the connective tissue and also invade the central necrotic area. By the 6th and 7th day the mass is completely solid and has the structure of the fully developed tumour (vide microphotoeraph 2). Irradiation for 24 Hours. On the day after inoculation the graft, as in the control, is almost entirely necrotic. The inflammatory changes and cedema set up in the surrounding tissue are less marked; there is also less invasion with leucocytes, and the sarcoma cells are all apparently degenerate. On the third day they can still be detected, but they show no signs of proliferation. On the fourth day the sarcoma cells have completely disappeared. The graft now consists of granular structureless material, a few leucocytes and nuclei alone being left (vzde microphotograph 3). There is commencing vascularisa- tion and fibrous tissue formation at the periphery. Irradiation for 80 Minutes. On the day after inoculation the reaction of the surrounding tissue to the implanted graft is again less marked than for the controls, and the graft is largely necrotic. Sarcoma cells in good condition can, however, be found at the edge. By the sixth day the graft is almost completely vascularised and many of the sarcoma cells at the edge appear to have proliferated to a slight extent. The condition shows (vide microphotograph 4), a distinct contrast with the preceding. Ata later stage the graft is largely replaced by fibrous tissue but the sections can be distinguished from those of the 24-hour irradiation series by the presence of a few large sarcoma-like cells embedded in the fibrous tissue. Irradiation for 20 Minutes. Both the experimental and control grafts formed tumours, the former being the smaller. The sections of the experimental grafts are similar to the controls throughout this series, except that in the former the rate of proliferation of the cells is delayed. Moreover 11 days after inoculation the irradiated Russ and Chambers. Roy. Soc. Proc., B, vol. 86, Plate 12. ve a ; x HW 5 4 I t i aS My a i= i ' f uu S Ro ’ a reas re a oe ae : Roy. Soc. Proc., B, vol. 86, Plate 13. Russ and Chambers, 1913.] Rays upon the Cells of Jensen's Rat Sarcoma. 489 tumour contains numerous very large sarcoma cells, a few of which are multi- nucleated ; they are not found in the actively growing tumours. Cells of this kind haye been described by Clunet* and others as occurring in tumours which have been irradiated in vivo. The histological changes indicate that after a long period of irradiation the cells of the growth are killed and are rapidly absorbed. With shorter periods of irradiation, even in cases where no tumour develops, the cells remain at the site of inoculation fora long time, but their capacity for proliferation is diminished. This inability to proliferate is not due, as in immune animals, to failure of the connective tissue to vascularise the graft, but is due to some change in the cells themselves. Conclusions. 1. Jensen rat sarcoma when exposed in vitro to the @-rays from a source of radium of intensity 1:63 mgrm. per square centimetre for 90 minutes, or to radium emanation of concentration 0°53 millicurie per cubic centimetre for 45 minutes, will not grow upon inoculation into normal rats. 2. The sarcoma cells which have been irradiated may remain in the animal body for more than 60 days before giving evidence of growth. 3. Histological evidence shows that failure of the irradiated sarcoma cells to produce a tumour does not necessarily indicate their destruction at the time of inoculation. * Clunet, ‘Tumeurs Malignes,’ 1910. 490 On Light-Sensations and the Theory of Forced Vibrations. By Georce J. Burcu, M.A., D.Se. Oxon, F.R.S. (Received April 19,—Read June 26, 1913.) Every hypothesis, whether mechanical, photo-chemical, or ionic, concerning the connection between the light-waves and the sensations they evoke, must of necessity rest ultimately on the theory of forced vibrations. It seemed probable, therefore, that a model illustrating the production of forced vibrations over a range comparable with that of a light-sensation might be of service for teaching purposes, and might prove suggestive in studying the phenomena of vision. The apparatus now described was made for me in May, 1909, by Mr. H. Davis, the Assistant for Manual Instruction at University College, Reading, and was used in my lectures both in Reading and in Oxford, but no account of it has been published. From a light wooden bar, A (fig. 1), pivoted at the two ends, hang a number of grey silk ribbons, varying in length from 11 to 44 cm. The longest, which represent the red end of the spectrum, have therefore a period twice as long as that of the shortest, which correspond to violet, Each ribbon is weighted at the end with a strip of lead. The red sensation is represented by a scarlet ribbon occupying the Ta Fig. 1 Side Elevation of Oscillating Bar A, with Silk Ribbons of different lengths acting as Resonators to the Pendulum with Movable Weight B. On Light-Sensations and the Theory of Forced Vibrations, 491 position corresponding to Fraunhofer’s O-line, the green sensation by a green ribbon near the Fraunhofer 0-line, and similarly the blue and the violet by a ribbon of each colour in the corresponding region of the spectrum Tn the actual model there are five grey ribbons between every two coloured ones—in the diagram only three are drawn, for clearness. The bar A is made to oscillate by a heavy pendulum B, with a movable bob fixed by a thumb-screw, If this is tuned to the period of the green ribbon, although those on either side respond more or less, the maximum amplitude is attained by the green, which, according to the theory of Thomas Young, is most strongly excited by light of a certain wave-length. If the pendulum is tuned for the yellow, then the grey ribbon under neath the yellow portion of the bar indicates by the amplitude of its oscilla- tion the position in the spectrum of the impressed vibration, while the red and green, representing the colour-sensations, swing moderately, being both, as Young said, excited, though to a less degree, by yellow ight. It makes the experiment more striking if the bar A is painted with the complete series of spectral colours, and the principal Fraunhofer lines marked upon it in black in their proper positions. For some reason, perhaps connected with the fact that the ribbons swing so near each other and that eddies are formed by their edges, friction is relatively greater at small than at large amplitudes, so that to some extent the effect of its increase on the range of resonance may be seen as the oscillations subside. For greater differences I use a set of oscillators with lighter weights. For purposes of demonstration it 1s necessary to show the effect upon such a system of resonators of light of more than one wave-length. To do this two heavy pendulums, C and D (fig. 2), are connected by light wooden rods, E and F, to the ends of the cross-link G, from the middle of which a third rod H leads to a crank-pin on the bar A. The heavy bob of the pendulum B being removed, the stem of it, which is quite light, serves as an index to show the compound nature of the motion imparted to the bar A, the movements of which often appear strikingly irregular. But each pendulum is responded to by the resonators in tune with it; as though the others were still. Quite apart from any theories of colour-vision, the apparatus demonstrates in a striking manner the phenomena of forced vibrations. The change of phase according as the natural period is greater or less than that of the impressed force is well shown, and it is particularly instructive to start with a considerable difference of period between the pendulums C and JD, and gradually bring them into unison. VOL. LXXXVI.—B. 20 4992 Dr. G. J. Burch. On Light- [Avon toe Wis 2 End Elevation of Oscillating Bar A, showing its connection, by the link-work, E, F,G, H, with the Pendulums C and D, the two Resonators in tune with which are oscillating violently, and the rest scarcely at all. The periodic variations of amplitude in the response which die out after the impressed force has been in action for a little while are easily seen, and, in fact, the apparatus affords an excellent illustration of §47 and §48 in Volume 1 of Rayleigh ‘On Sound’ For my present purpose the main interest of the apparatus les in the possibility of utilising it to elucidate one of the most difficult problems of colour-vision—the problem, namely, of white light. Newton’s discovery of the physical fact that the prism separates the beam into rays of different refrangibilities and different colours affords no explana- tion of the physiological fact that any one of these colours, even the most brilliant, should disappear with absolute completeness in presence of the others. Hering’s theory of the antagonism of red and green, and of blue and yellow, is, from this point of view, a most natural one. It would be simple enough, on Young’s theory, to explain why all colours tend to white by very strong light, for each of his three hypothetical “nerves” is assumed to be affected, though to a different degree, by light of all wave-lengths, so that we have only to suppose that, with a sufficiently strong stimulus, all three “nerves” are almost equally excited. But this leaves us without any valid explanation of that other fact, that by very feeble light all colours tend to grey. 1913.] Sensations and the Theory of Forced Vibrations. 493 In some respects the phenomenon of colour is more striking than that of whiteness, and probably, if it had been possible for the spectroscope to be invented before any theories on the subject were thought of, the problem would have rather been to explain why, with a certain intensity of illumina- tion, light of different wave-lengths should appear brilliantly and variously coloured, whereas with less light or with more the colours fade. The existence of this optimum intensity for the excitation of the sense of colour is very noticeable when calibrating an ordinary students’ spectroscope with the Fraunhofer lines by direct sunlight. The act of measuring produces sufticient fatigue to make the spectrum appear pale. If now the instrument is directed towards the sky or a cloud, the colours instantly become rich and brillant, though it may be necessary to open the slit wider before there is light enough to see them. I propose to show how this entire range of phenomena—from the scarcely visible band by feeble light to the brilliantly coloured spectrum with optimum intensity, and the washed-out colour with bright sunlight—is in strict accordance with the laws of forced vibrations. It is, of course, understood that any hypothesis as to the manner in which the ethereal vibrations we think of as light give rise to the sensations we know as light would naturally be expressed, in the first place, in terms of the electromagnetic theory. Inasmuch, however, as the same laws apply to all kinds of harmonic oscillations, whether electrical or mechanical, it will be more convenient to retain the phraseology proper to the mechanical model. Evidently, the action, whatever it be, must take place through those processes which we commonly regard as chemical—processes, namely, in which some rearrangement of atoms, either within the molecule or from some other molecule, is brought about. There is a very clear statement by Kiihne of the Opto-chemical Hypothesis in Hermann’s ‘ Handbuch der Physiologie,’ vol. 3, p. 327. According to him, the opto-chemical hypothesis regards the visual cells as carriers of chemically decomposable materials called visual substances, which, however, have no effect upon the visual cells as long as they are undecomposed. But the hypothesis ascribes to the decomposition products resulting from the action of light on these substances the power of chemically exciting the protoplasm of the visual cells. This excitation might conceivably result from the act of decomposition by light of the visual substances, but inasmuch as the effects do not instantly cease with the removal of the stimulus, it would appear that the cause is to be sought rather in the material action of the decomposition products. In applying to this theory the elementary principles of chemical dynamics, 202 A94 Dr. G. J. Burch. On Lnght- [Apr. 19, we must carefully separate the initial electro-chemical action from the resultant physiological action. The production of the active decomposition products which excite the retina takes place, doubtless, according to formule capable of exact and more or less simple numerical expression. The regeneration of the sensitive material, whether due to a natural tendency to revert to its original condition or whether effected, as seems more probable, by processes of metabolism, is still conceivably capable of numerical expres- sion. But between these actions, which may be classed as electro-chemical, and the resultant sensation there is a whole group of modifying causes, which must, for the present, be kept outside our attempts at analysis. There arises, in this connection, a question of general interest in physiology that has not, so far as I know, been fully dealt with. Is the excitation of a tissue by a chemical compound a process of the same character as a chemical reaction? I do not refer to such crude foreign substances as dehydrate or coagulate, or otherwise damage the tissue, but such as excite quite normally its characteristic functions. There must be some end to the activity of the exciting substances in the eye, else, once separated, they would go on exciting sensation indefinitely. Hither they recombine or are washed away or their energy is transformed into that of the sensation. I do not attempt to decide this problem, but have so stated the theory that it would not be affected by it. I. Opto-chemical Processes. Let » = the number of molecules of the visual substance not yet acted on by light. the number of molecules of the decomposition products capable fo I of exciting the retina. e = the number of molecules of decomposition products used in pro- ducing sensation. The known data are insufficient for a complete statement of the problem— for instance, it would be necessary to know whether any molecules of « remain’ ultimately unaccounted for by e—whether they are neutralised, destroyed, or simply washed away by the circulation and dispersed. For our present purpose it is sufficient to note that each of the variables is subject to conditions of equilibrium governed by the ordinary laws. Then dn/dt = the rate of supply of the visual substance, and dx/dt = the rate of demand upon it. Also de/dt = the physical intensity of the excitation. If dn/dt is greater than dz/dt the store n of molecules capable of being 1913.| Sensations and the Theory of Forced Vibrations. 495 acted on by light increases, and we have, when da/dt is zero or very small, the condition of dark-adaptation, in which m reaches its maximum. In the dark-adapted eye we must regard the compound in which the forced vibration is produced, as in a state of maximum concentration, and by all the analogies of electro-chemical action, this must correspond with increased resistance, or, in the mechanical model, with increased friction. In order therefore to obtain a graphic representation of the meaning of dark-adaptation, we may compare together a series of curves of forced vibrations with different values of the coefficient of friction. Let it be assumed that the quantity of the exciting substance set free from the sensitive material may be taken as corresponding to the kinetic energy of the forced vibration induced by the action of the light, the expres- sion for which is phan ee) Se ay) P See where an a ee A? + 4h?’ nu i 4 p being the natural period of the resonator, 7 that of the impressed force, and k representing friction.* The ratio p/n corresponds to the musical interval between the natural period of the resonator and that of the forced vibration, and as the expres- sion for A is symmetrical with regard to the ratio /p it follows that if values of B are taken as ordinates and ratios 1/p as abscissz, we shall have for each value of p a curve symmetrical about its apex. This consideration naturally suggests plotting the spectrum by what may be called the “ keyboard ” system, in which, as in the piano, equal horizontal distances correspond to equal musical intervals. Except in general illustra- tion of the relation between ultra-violet and infra-red rays I do not think this method has been used. It seems to offer certain advantages in dealing with problems of colour-sensation. Fig. 3 shows four hypothetical colour-sensation curves plotted on this Resonance Curves of the Colour Sensations Red, Green, Blue, and Violet, with the coefficient of friction, 4 = 0-2. For the dotted curve 4 = 0:1, but the ordinates are reduced to one-quarter of their full value by the shunt factor. * Barton, ‘ Handbook on Sound,’ p. 145. 496 Dr. G. J. Burch. On Lnght- [| Avpirey ho system. They represent the four colour-sensations described in my paper on “ Artificial Temporary Colour-Blindness.”* Inasmuch as according to the theory these curves must be symmetrical, T assumed the apex of the green sensation to lie midway between the two ends of it, and similarly with the blue sensation. Then that part of the yellow where neither red nor green predominates must be where the tangents to the two curves are equal and opposite. In this way I got a position for the apex of the red sensation, and in like manner for the violet sensation by means of the blue. It may serve to fix our ideas if I refer to the illustration of the musical scale. Taking E as the apex of red, green would be approximately at G, blue at A, and violet at B. It is noteworthy that green, blue, and violet are practically equidistant and much closer together than red and green—red in fact might almost be taken to correspond with Eb. In order to assign a value to the coefficient of friction I was guided by the keyboard interval over which the forced vibration must extend. After a number of trials I decided to take 4h? = 0:16 as a trial value, giving each of the four colour-sensations the same coeflicient of friction. It will be observed that the transition from blue to violet occurs at G, that from green to blue at F, and that from red to green a little above the D line—or exactly at the D line if the centre of the red is placed at E>. In this case, too, the red would meet the blue about the 6 line, as it does during artificial green-blindness. Fig. 4 shows the effect of increasing 4k? to 0°60. Each one of the curves is lower, but it extends over a greater range of wave-lengths in proportion to its height. This agrees perfectly with the well-known fact that “by feeble light all colours tend to grey.” Fig. 4.' Resonance Curves of the Colour Sensations Red, Green, Blue, and Violet, with the coefficient of friction, & = 0°4 nearly, z.e. 447 = 0°60. In my paper on “ Colour-Vision by Very Weak Light ”+ I have noted the changes which occur in such cases between the boundaries of the colour- * ¢Phil. Trans.,’ B, vol. 191, pp. 1-34. + ‘Roy. Soe. Proc.,’ 1905, B, vol. 76, p. 214. 1913.] Sensations and the Theory of Forced Vibrations. 497 sensations owing to the fact that they are not all affected in the same proportion, and also the general effect of the increased extent in the spectrum of the several colour-sensations. “The colour must look pale under feeble illumination owing to the presence of three if not of all the constituents of white.” If this view is correct it would explain a rather puzzling fact that I have noted in my investigation of cases of colour-blindness.* I found it necessary to specify not only those cases in which a colour-sensation was deficient, but those also in which it was of greater extent than usual. The Swiss girl therein referred to had practically but one colour-sensation, because her green so greatly exceeded the normal in spectral extent, yet by suitable fatigue it could be reduced so as to reveal her possession of the other sensations. We have only to suppose that the supply of sensitive material to the green end- organs was so copious as to cause a quite unusual concentration, with corres- pondingly large coefficient of friction. In such a case the eye would be with respect to green in a condition of dark-adaptation, with this difference, that no ordinary demand could overtake the supply and reduce the store, n, of molecules of visual substance to the normal amount. In other words di/dt—dz/dt was positive for all ordinary values of dz/ dé. But it is clear that this might arise either from dn/dt being large, or from dz/dt being abnormally small. I consider that I have met with cases of both kinds. In one, of which I shall give a full account in my next paper on “Cases of Colour-Blindness,” there was monochromatic vision in one eye only. This enabled me to ascertain that the sensation was unmistakeably one of whiteness, with a little colour at the two ends of the spectrum. But the whole intensity of the light-sensation was extremely low although the transparent tissues were perfectly clear. In the same way we may explain the fact noted in my paper on “ Artificial Temporary Colour-Blindness ”+ that with some people the “overlaps” of the colour-sensations are very large, and with others almost non-existent, and also that with some the overlaps are large between red and green and small at the other end of the spectrum, so that to them yellow is an important colour, and with others there is little or no overlap between red and green, and large overlaps in the green, blue, and violet region. It seems to indicate that the coefficient of friction is constitutionally large in these people either for all the colours or for those to which the large overlaps belong. And by the expression “constitutionally large,” I mean large not because any temporary cessation of the demand dz/dt for the exciting molecules has allowed a large * ©Phil. Trans.,’ B, vol. 199, pp. 289 and 250. + ‘Phil. Trans.,’ 1899, B, vol. 191, p. 1. 498 Dr. G. J. Burch. On Lnght- [Apr. 19, store, 7, of visual substances to collect, but because the metabolism is so active that the rate of supply dn/d¢t of the visual substance is greater than in most people. In quite a number the rate of supply of the visual substance for the blue is so great that it is not recognised as a separate sensation, but confused with violet. According to this hypothesis, the store 1 of molecules of the visual substance would be governed by an equation of equilibrium between the causes tending to use, destroy, or disperse the exciting molecules, and those by which they are formed, so that it would have a definite value for every constant intensity of illumination and of physiological condition. In ordinary dark-adaptation we should therefore have a large accumulation 7 of molecules of the visual substance, so that on coming into a quite moderate light the rate of formation of exciting molecules dz/dt may for a short time be greatly in .excess of its ordinary value for that degree of illumination, producing a sensation vividly described by the French word ¢blowissement.* This gradually passes off as the store m falls to its equilibrium value for that rate of demand dz/dt, when the eye reaches its new condition of adaptation, the curves of the colour-sensations changing into those belonging to the coefficient of friction proper to the concentration represented by the new value of n. As the intensity of the light increases the equilibrium value of 1 grows less—the resonance becomes more free and the curves grow peaked, so that a condition is reached represented by fig. 5. Each colour is_ brilliantly distinct and separated from the rest, dominating its own region of the spectrum. This is the optimum intensity for the excitation of the sense of colour. What happens when the light is still brighter will be described in the next section. A BC D b F G HK Fic. 5. Resonance Curves of the Colour Sensations Red, Green, Blue, and Violet, with the coefficient of friction, £ = 0°15. Before passing on, it must, however, be noted that according to the opto- chemical hypothesis we must imagine the decomposition products to be formed by the action of the light, independently of whether or when they are used. * See also, for the physiological aspect of this, p. 501. 1913.] Sensations and the Theory of Forced Vibrations. 499 The effect of a sudden flash must be to set free a definite quantity of exciting molecules which will continue to produce sensation until their power is exhausted, or they are dispersed. This may occupy several minutes even after a momentary flash, and much longer for a more prolonged exposure to light. The phenomenon may be detected as a very brief after-effect even with light of low intensity. Using the terms already employed :— dx{dt is conditioned by the intensity of the illumination and the store n of exciting molecules. de/dt represents the positive after effect so soon as dz/dt becomes zero, i.c. when exciting molecules are no longer produced. This agrees with Fechner’s theory as described by Helmholtz. And at this point it becomes necessary to take into account also the physiological elements -of the visual sensation. Il. Physiological Elements of the Visual Sensation. Thus far, the conditions discussed relate to one process only of the opto- chemical action, namely, the setting free of the active decomposition products which excite the retina from the bland visual substances which have no action upon it. The other process, necessary to complete the visual effect, is physiological. It is complex, including the excitation of the sensitive elements of the retina by the active decomposition products, the transmission to the central -organ of the response, and its translation into conscious sensation. Moreover, there is strong evidence of the existence of a protective mechanism whereby the intensity of the stimulus is regulated. This physiological process is curiously distinct from the first-described -opto-chemical process. The rate at which the exciting substance can be produced far exceeds that at which it can be used up. This can easily be shown by means of a photographic exposing shutter fixed in the window of the dark room. The shutter is set to give an exposure of known length, and the observer looks through it at the sun’s dise reflected in the mirror of the heliostat. The positive after-image so produced lasts many times longer than ithe flash. I have made some attempts to establish a relation between length -of flash and duration of after-image, but have had to relinquish the work for the present before getting complete data. The store of exciting molecules set free by the action of the light is used up at a rate depending apparently on the concentration, so that it is rapid at first and dwindles down to nothing. It follows from this that during 500 Dr. G. J. Burch. On Light- [Apr continuous illumination the sensation of light at any imstant is not due exclusively to the light falling on the eye at that instant, but includes also the remainders left over from previous illumination. Now in the electrical stimulation of nerve, and nerve-muscle preparations, after a quite moderate strength of stimulus has been reached no further increase causes any modification of the response. This was abundantly evident in the experiments by Gotch and myself with the capillary electro- meter—and, moreover, there was no sign of “fatigue” of nerve unless the excitation was excessively great. It is fair to conclude that when, under the action of light, the exciting decomposition products reach a certain concen- tration, no further increase adds anything to the intensity of the resultant. light-sensation. This may be far stronger than is pleasant, just as the muscular contractions. duting cramp may be excessively paintul, but it does not seem probable that. any compound produced as the result of a normal physiological process can be of such character or of such concentration as to destroy the tissues in which it originates. We should thus have a maximum limit to the possible intensity of the light-sensation—a limit independent of the particular physiological condition of the eye at the moment. Were it not for this the theory of forced vibrations would indicate that with intense and long- continued illumination the several colour-sensations must stand out more and more distinctly instead of becoming paler and tending towards white. But there is evidence of another factor, purely physiological, which I may term the shunt-factor, whereby the strength of the sensation is governed. I am inclined to think that this factor, manifesting itself under various. conditions, affords the explanation of several quite different phenomena. Thus Charpentier’s bands may be taken as evidence that the sudden onset of a fairly bright illumination over a large surface results in a sensation of intermittent intensity. Shelford Bidwell’s experiments with pigments and my own with spectral colours show that each colour-sensation acts inde- pendently of the others in this respect. Purkinje’s recurrent images, especially in the striking form described by MecDougall.* exhibit the same thing in connection with the positive after-effect. Some controlling mechanism is. set in action whereby the positive after-effect, as it dies away, is periodically shut off like the sound in the swell box of an organ. And this action, like that of the bands of Charpentier, gives fairly rapid alternations. I take it. to be a spasmodic excitation of the negative after-effect. A similar but much slower periodicity may be observed in connection with the fusion of binocular images, especially when both are needed to complete: * ‘Journ. Psychol., 1904, vol. 1, part 1, p. 91. 1913.] Sensations and the Theory of Forced Vibrations. 501 some familiar outline, eg. a cross. The phenomenon occurs not only with the direct image but with its positive after-effect, the waxing and waning of which was known long before the time of Thomas Young. When making my experiments on artificial temporary colour-blindness I particularly noticed that the positive after-effect of a brilliant mono- chromatic light over a large retinal area showed nothing of this periodicity but died out steadily and gradually. Putting together these facts I have suggested in my paper on “ Areal Induction,”* that they indicate the existence of a protective arrangement in the retina by which the eye is shielded from the sudden effects of too strong a light. I now submit that there is sufficient evidence to identify this protective action with the negative after-effect of which it is a special case, and that the other phenomena to which I have just referred come under the same category. The negative after-effect is so generally referred to as synonymous with fatigue that some reference to that aspect of the problem is imperative. The sensitive materials being used wp and exhausted, the eye is supposed to be rendered locally less sensitive for a while. But the photo-chemical conditions indicate a state of increased activity, for if the store 1 is exhausted, dn/dt reaches its maximum, and during continuous steady illumination da/dt = dn/ dt. But it is easy to show that during a bright summer day the quantity of sensitive material used without causing any sensation of visual fatigue greatly exceeds the consumption during the production under suitable con- ditions of very considerable retinal “fatigue.” The explanation, therefore, must be sought among the physiological rather than the photo-chemical conditions of the problem. We may consider a few examples. The following phenomenon is instructive, and probably familiar to most people. On a rather misty afternoon, on coming to the window, the eyes rest on a bright gap in the clouds. At first it appears too bright to look at, and the after-image is very black. But if we persist, after a few seconds the sun’s disc appears sharply defined in the midst of the dazzling hght. The retina can hardly be regarded as fatigued in the sense of being exhausted and having its activities impaired, since we may continue watching until the eyes have become completely adapted to the more brilliant illumination, and further details—spots on the sun—isolated wisps of cloud in the clear space—become visible, which were at first hidden in the glare. In the subdued light of the room the store m of sensitive material had become large. * “Roy. Soe. Proc.,’ vol. 69, p. 129. 502 Dr. G. J. Burch. On Lnght- [-Ajoreelge On first glancing at the cloud gap the rate dx/dt at which exciting sub- stances were formed was greater than that di/dt at which sensitive material could be secreted. Accordingly, after a short time of almost painful brillianey, the store 7 was reduced to zero, and a hand-to-mouth condition of things set in, during which dz/dt = dn/dt, i.e. just as much exciting substance was formed as could be furnished by the sehsitive material secreted. But, during the interval, dz/dt was so far in excess of the normal that maximum sensation was produced both by the light of the clouds and by the sun itself. And it can hardly be denied that during the so-called retinal fatigue, after the details became visible, a condition of very great activity existed.* On the other hand, the sensation was undoubtedly less, and became less as the eyes got accustomed to it. Similar phenomena may be seen in a furnace—a blinding glare in which details of flame and of molten metal with slag floating on it gradually appear —or with far less intensity of illumination in the “ faces in the fire” in the hot coals of an open grate. There is at first just the same sense of being dazzled—the same gradual perception of details, and in the end the same quiet contemplation of what has become comfortably visible, whether the experiment is made with the hot coals of the open grate or the far greater intensity of the evening sun. Yet the actual intrinsic luminosity of the hot coals is a good deal less than that of the newspaper which we read out of doors in the sunshine on a summer's day. The name retinal fatigue for this state is not very apt. It is a condition in which a powerful stimulus produces a reduced effect, not because the organ is in any way deteriorated or used up, but because it works best in that way. It suggests strongly the use of a shunt with a galvanometer. And if we add the idea that with the eye it takes some little time to put the shunt on, and still longer to take it off again, we have a fairly accurate description of the facts. The condition in which a shunt factor too strong for the exciting light persists is called the negative after-effect. Thus, if S; = the shunt-factor at time ¢, de[dt S = the streneth of the sensation at that moment. This view of the negative after-effect agrees with the theory of Fechner on the subject, as described by Helmholtz.t I have been unable to consult * Of. Waller, ‘Phil. Trans.,’ 1897, B, vol. 188, p. 65, note, “The retina resembles nerve with respect to its inexhaustibility.” t Helmholtz, ‘Handbuch der Physiologischen Optik,’ 2nd ed., p. 534. 1913.] Sensations and the Theory of Forced Vibrations. 503 Fechner’s original papers, and do not know what was his opinion on the point, but I consider that there is strong evidence that the shunt-factor is a function of the retina rather than of the brain. The form of the curves of colour-sensation during very bright light may now be discussed. The supply dn/dt of sensitive material is limited by the rate of metabolism possible to the tissues, but the rate at which it is converted into exciting substances, dx/dt, depends on the intensity of the light. Therefore long exposure to a bright light will make the concentra- tion of the sensitive material, and consequently also the molecular resistance, tend toa minimum. But when friction is low the amplitude of the forced vibration is slightly greater whatever be the period of the impressed vibration, and very much greater as this approaches the period natural to the resonator, so that the curve develops a very sharp central spike. This spike cannot be reproduced in the corresponding curve of colour-sensation because of the separation of the opto-chemical from the physiological functions of the eye. The active material, however concentrated, cannot do more than excite a maximal response. But the excess of active material may excite the protective arrangement, or shunt function, to stronger action, so that the maximal response may only send through to the central organ a quite moderate sensation. This state of things is illustrated in fig. 3, where the dotted line represents the first beginnings of artificial green-blindness, before the sht has been opened wide enough to dazzle the eye. The response is well within the limits of possible sensation, so that there is no truncation of the curve of resonance, although by the shunt-factor its ordinates are reduced to one quarter of their normal value. It will be noted that the effect is to lessen the apparent extent of the green sensation. And this is precisely what happens. The red sensation on the one side, and the blue sensation on the other, encroach on the green. A powerful light is not required to show this. My method of testing for colour- blindness is based upon it. After looking at the d-lines in a spectrum of quite ordinary intensity for 30 seconds the boundary between red and green is found to have shifted from 100 to 400 A.U. nearer the green, if the observer possesses a normal green sensation. The height of the apex is inversely proportional to the coefficient of friction. When, therefore, this is further reduced by the action of a strong light, a stage is reached when the curve is truncated. For the height of the apex of the resonance curve is limited because in the first place dx/dt, the rate of production of the exciting substance, cannot exceed dn/dt, the rate of secretion of the sensitive visual material, and in the second place because de/dt, the 504 Dr. G. J. Burch. On Lnght- [ Ape Loy strength of the response, reaches its maximum under the action of smaller quantities of the exciting substance. Consequently the strength of the resulting sensation, when reduced by the shunt factor, will be represented by a more or less flat-topped curve. From various data I judge that in artificial colour-blindness of quite moderate degreee, a shunt ratio of 1: 100 before and after the exposure of the eye to light is well within the mark. I described in 1897 an experiment bearing upon this subject which seems to have escaped notice.* When the spectrum is viewed by intermittent light of great intensity, the flash ratio being 1:3 or 1:4, as soon as the eye gets accustomed to the strong light it is seen that the continuity of the colours is gone, and that there are now visible four bands of strong colour, viz., red, green, blue, and violet, upon a pale but brightly illuminated ground. Between the intense red and the rich green is a space of a colour between yellow-ochre and raw sienna, but very pale. Between the green and the bright blue is a region of pale greenish-blue, passing into pale steel-blue. Beyond the bright blue is a pale lilac space. Farther than this cannot be seen while the red end of the spectrum is in the held, because of the overpowering intensity of the light in the neighbourhood of the yellow-green. But by shifting the prisms so that the b-lines come on the extreme Jeft, while the G-line occupies the middle of the field, the region about H and K appears of a deep rich violet of great intensity, contrasting strongly with the pale llac between it and the blue. I suggest the following explanation. Owing to the periodic intervals of darkness during which the visual substance collects, the rate of formation of exciting substance is maximum over the whole range of the resonance. But the light is so strong that the shunt ratio is very high. Consequently, where two colour-sensations overlap, as in the yellow, we get the pale colour of the binary blend, and on either side, where only one colour-sensation is effective, the full rich tone proper to it. The question as to how and by what mechanism the shunt function works, and where that mechanism is situated, is of the greatest interest and importance. I have already mentioned that I do not hold with those who refer it to the central organ. For one thing, it would add to the complexity of the optic nerve, as I have pointed out in my paper on “ Areal Induction.’”+ And, for another thing, it would leave the end organs in the retina unpro- tected. I should therefore look for the shunt mechanism in the retina itself. If Charpentier’s bands and Purkinje’s recurrent images are to be regarded as due to spasmodic applications of the shunt factor, they would seem to * “Journ. Physiol.,’ vol. 21, p. 481. t+ ‘Roy. Soc. Proe.,’ vol. 69, p. 125. 1913.] Sensations and the Theory of Forced Vibrations. 505 indicate that it depends in some way upon contractile movements within the tissues. It would be quite easy to imagine an arrangement whereby a sensitive point might be dipped into or drawn out of a swarm of exciting molecules set free by the action of light on a surface close to it. The actual range necessary for such a movement to be effective would be extremely small, so that it might easily escape notice. Each retinal element would act independently of the rest, the protection from over-stimulation would obviously be complete, and spasmodic action could easily take place. Areal induction, by which the illumination of one portion of the retina affects the surrounding areas, might result from secondary cross-connections between the retinal elements. Note on the Laws of Weber and of Fechner. The question arises whether the idea of a shunt factor, by which all sensations are reduced in the same ratio, is compatible with the received view that the light-sensation varies according to the logarithm of the stimulus. Weber’s law is the statement of an experimental fact, viz., that “the smallest perceptible difference of luminosity is a constant fraction of the whole intensity of the light.” And this is nearly true, though, according to several observers, not accurately true, over a considerable range. Fechner’s law is on an entirely different footing. It is based on mathe- matical theory, and claims to establish a numerical scale of sensation :-— “The intensity of the sensation varies as the logarithm of the stimulus.” If the sensation of light were a continuous function of one variable from subliminal threshold to maximum, there could be no two opinions as to the validity of Fechner’s law when Weber’s law is taken for granted. It is a simple, obvious, elementary application of the calculus. But the sensation of light is a function of at least three variables, viz., the production by light of the exciting substance, the stimulation by this of the end-organs, and the regulation by the shunt factor of the strength of the resulting sensation. Moreover, the perception of a difference between two sensations, which is necessarily implied in the statement of Weber’s law, involves the judgments of the central organ, thus introducing a possible fourth variable. Under such conditions we are not warranted in ascribing to Fechner’s law the cogency of a result obtained by the calculus as physicists use it. The experiments of Ebbinghaus quoted by Helmholtz* go to prove that Fechner’s law fails within a quite moderate range of intensities. My own experiments extending over the last three years confirm this view, and bring out, in addition, a point of considerable interest. * Helmholtz, ‘Physiol. Optik,’ 2nd ed., p. 392. 506 On Light-Sensations and the Theory of Forced Vibrations. ° I have, since 1910, in the practical classes which I have conducted for Prof. Gotch, made a considerable number of people arrange from 7 to 10 strips of white paper at distances varying from 4 metre to 2 metres from a candle in a dark room, so that viewed from a certain point they presented a series of apparently equal gradations of luminosity. The results were very instructive, and in one respect unexpected. Most of the men repeated the experiment two or three times in order to get what they considered a good result. Almost without exception the first attempt of each person showed considerably higher values for the ratios at the two ends of the series. In the majority of cases the difference was less in subsequent experiments, but it was evident that, to the unbiassed judgment, the eye is less sensitive to differences between the brightest objects visible and also between the faintest objects visible at any one time than between those that are moderately illuminated. Although practice reduces this divergence from Weber’s law it does not do away with it. I still make the ratios at the two ends of the series higher than those near the middle of it. But with monochromatic red light any series that looks right to me under a feeble illumination looks right also in light 100 times as strong, as it should do according to Weber’s law. I am inclined therefore to accept Fechner’s law with the same caution that Waller expressed in reference to the relation between strength of stimulus and the retinal currents of the frog’s eye. “The curve plotted from the data comes out concave towards the abscissee and not unlike an ordinary logarithmic curve.” If, in judging the minimum perceptible difference of brightness, we instinctively make use of the shunt function only, that would lead to some- thing not far removed from the logarithmic law, within the range covered by the shunt. The greater part of the experimental work connected with this paper has been done in the Physiological Laboratory, Oxford, and the expenses have been defrayed out of the Government Grant Fund. 507 The Various Inclinations of the Electrical Axis of the Human Heart. Part I.—The Normal Heart. By A. D. Water, M.D., F.R.S. (Received March 6,—Read May 8, 1913.) (From the Physiological Laboratory of the University of London, South Kensington.) CONTENTS. Page 1. Introductory— Weak and Strong Leads .........:....ccsessesneecnccnsececeneeueenes 507 2) Determination of the Blectrical Axis’ ..0......c.cccsesescsoccsecscorscocscsescccccsns 511 3. Proof of the Formule, tana = (L—R)/(L+R), tana = 2(R—L)/((R+L)... 513 a AMO SEIANES TWEE! scooccooccodoséace scassonqecoeecbecoB0 pagabanocobenedeppadoncsadaddao000n5d 516 5. Influence of Position and of Respiration ............:.csceeeecececneesereenscesecues 517 Gua Ronerotpilea ti lUSCLE i. sameaerisceaccersiaceies se tesebicctscecclcice cs svececishn + slcasincealene 518 7. Influence of Muscular Exercise ..............:sccscecccscceccucetectececesceseecseceses 519 Sale N COLO OOM! .eeieciacs eat deatens «nies ac cetdsdeee Sauk so celslalostd de aeoatlddessttdddeccee as 522 GMECHESTPINCAGS IN sdartalsccesecuaton a teaemetee selec ctioncicebsvins'sasbessudenncstion nese weaansje as 522 1. Introductory— Weak and Strong Leads. The “normal hearts” dealt with in the present communication are those of the regular workers in this laboratory and of ordinary visitors who have wished to see the electrical effects of their own hearts and who have been good enough to allow me to make use of their records ; these have been taken when possible by right and left superior and by right and left lateral leads, so as to afford data for the calculation of the position of the current-axis above and below the heart. No clinical examination of any kind was made in the case of visitors, so that strictly speaking normality of the heart in their ease has not been verified. We may, however, assume as probable that the heart of an ordinary active person does not depart widely from the normal. And as a matter of fact the electrocardiograms taken upon such persons are in themselves sufficient evidence of normality to an observer familiar with the signs of abnormality. In the entire series of normal persons (amounting to about 200) I have examined during the last three years I have only met with five cases present- ing “ abnormal” cardiograms ; of these five, three have volunteered an interest in their own cases that has permitted a careful examination of the heart to be made, in the other two cases no attempt at verification has been made or suggested. One of the “normal” cases has proved to be particularly interesting, that namely of Thomas Goswell, my former laboratory man, upon whom I made VOL. LXXXVI.—B. 74 1 508 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, repeated observations during the year 1887 and whose current-axis I then estimated, or rather guessed, as forming an angle of 45° with the vertical line. His normality was, of course, verified at the time. And I have been fortunate enough a few days ago to obtain records of the four necessary leads from which the angle of the current-axis can be calculated. It happens to come out at the value of 45°, which in itself is, in my opinion, proof of normality. And the photograph that was taken of this subject last week as compared with the sketch of the same subject that was used asa class-diagram 27 years ago is in itself sufficient proof of normality as to the state of the heart. In 1887 the idea occurred to me that it should be possible to utilise the limbs as natural electrodes in relation with more or less opposed aspects of the heart, and so to obtain information concerning the absolutely intact organ, With the aid of Lippmann’s capillary electrometer I surveyed all the different pairs of leads that I could think of, and as the outcome of this survey, divided the human body into two unequal parts by means of an imaginary line or equator cutting at right angles a second imaginary line or current-axis crossing the chest obliquely in the direction of the anatomical axis of the heart—itself an imaginary or at least an indefinite line. I figured the current-axis as forming an angle of 45° with the vertical and taking the leads two by two I found that they fell into two sets which I called “favourable” and “unfavourable.” On review of these observations it became apparent that in the “favourable” cases, ze. those in which the electric pulse was obvious, the two leads were on opposite sides of the equator, while in the “ unfavourable” cases, z.e. those in which little or no pulse was visible, both leads were on the same side of the equator. I ascertained by direct observation that of the six possible leads afforded by the four extremities taken two by two, three are favourable (1, 2, 3) and three unfavourable (4, 5, 6) as regards the demonstration of the electrical pulse; and that of the four possible leads from the extremities taken in conjunction with the mouth, three are favourable (7, 8, 9) and only one unfavourable (10).* Finally the proof of the relation between the normal obliquity of the heart and favourable and unfavourable leads was completed by the investigation of two cases of situs viscerum inversus where in correspondence with the reversed obliquity of axis, the transverse effect between the two hands was observed to be reversed; the right superior and left lateral leads to be favourable ; the left superior and right lateral to be unfavourable. * Waller, “On the Electromotive Changes connected with the Beat of the Mammalian Heart, and of the Human Heart in particular,” ‘ Phil. Trans.,’ 1889, p. 169. The principal facts were demonstrated, in 1887, at the First Congress of Physiology, at Bale. 1913. ] Electrical Axis of the Human Heart. 509 The facts have been confirmed by all subsequent observers, more or less completely according as they have reviewed more or fewer of the 10 leads, but with one notable exception, viz., the left lateral, as to which I am stated to have been mistaken. Prof. EHinthoven in particular has attributed my having classified it as an “unfavourable ” or “weak ” lead to the comparative slowness of the capillary electrometer ;* as a matter of fact I did not make the mistake attributed to me, and I expressly adhere now to my original classification of the left-hand lead with either foot as a “weak” lead. The classification of leads as “strong” and “weak” forms indeed the basis of this paper, in the course of which it will be made apparent that the relatively weak left-hand effect below the heart is an index of the degree of obliquity of the cardiac current-axis. R. hand. L. hand. ils Transverse. 4 4. L. lateral. | L. hand. L. foot. R. hand. L. hand. Dee Atxdallisesnese se { 5. Equatorial L. foot R. foot. R. hand. R. foot 3. R. lateral oH 6. Inferior ...... \ R. foot. L. foot. Mouth. f Mouth. 7. L. superior = 10. R. Magee L. hand. R. hand. Mouth. 8. L. inferior... \ L. foot. ( Mouth. 9. R. inferior... | R. foot. CC denotes the current-axis, O O denotes the equator. When the survey of these ten leads has been completed, we may reduce them in number for further systematic investigation. The two feet are practically iso-electric: this pair may therefore be neglected. And if we regard the two feet as electrically indifferent, the axial and right lateral leads are equivalent, and one of them may be left out; also, on this. assump- tion, the left lateral and the equatorial are equivalent, and we may drop one out. Then as to the mouth; it is not necessary in a cursory survey to take * Einthoven (‘ Pfliiger’s Archiv,’ 1908, pp. 551-2). In my first three cases of 1887 the angles (measured in 1913) are 45°, 51°, and 86°. In Einthoven’s two cases of 1908 (measured from the published records on pp. 554-5) the angles are—Hi. = 22° and 1h Sy. 2 DY 510 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, all the three “favourable” leads: it is sufficient to compare the “good” hand (left) with the “bad” hand (right) by taking the right and left superior leads. And so I reduce my system of leads to five, as under :— I. Transverse. IV V II. Right lateral. III. Left lateral. II I IV. Right superior. V. Left superior. Note—In adopting this simplification we should, however, not lose sight of the fact that, for certain finer determinations, the P.D. between the two feet must be taken into reckoning, and that we may not always take as equivalent axial with right lateral and equatorial with left lateral, although for all ordinary purposes this may be done. Also, in the case of the mouth leads, where we shall content ourselves with taking the left superior as the strong lead and neglect the other two strong leads, viz., left and right inferior, we may find it necessary to take into account the small differences that obtain between the three strong leads with the mouth. An idea of their order of magnitude is given by the following example, in which they were carefully measured :— Right superior ......... 3 mm. (= 0:00023 volt) heft Superion.ceseseseess 15:3 — (= 000118) 55) Right inferior ......... NG 5iee— O00 260) Left inferior ............ WS 55 (= OCOD 5, )) The study of the subject was subsequently taken up by Einthoven and his pupils, whose principal publications appeared in 1895, 1900, and 1908. In 1900 he extended the inquiry to the abnormal heart. In 1903 he devised his string-galvanometer, which, by reason of its superior rapidity, is preferable to the capillary electrometer for these observations. At the same time, Einthoven reduced the six possible leads from the extremities to three, and promulgated what is known as “ Einthoven’s equation,” viz. :— Lead II (axial)—Lead I (transverse) = Lead III (left lateral), and represented in the form of an equilateral triangle, of which the heart is the centre. And quite recently* Einthoven has given a construction by which he * W. Einthoven, ‘“ Ueber die Form des menschlichen Elektrocardiogramms,” ‘ Pfliiger’s Archiv,’ 1895, vol. 60, p. 101; W. Einthoven and K. de Lint, “Uber das normale menschliche Elektrocardiogramm, und iiber die capillar-elektrometrische Untersuchung einiger Herz-Kranken,” ‘ Pfliiger’s Archiv,’ 1900, vol. 80, p. 139; W. Einthoven, “Die 1913. ] Electrical Axis of the Human Heart. 511 synchronises the three records, in order to calculate from their corrected value the magnitude of the angle « that must be formed with the horizontal by a potential difference passing through the centre of the triangle. From this basis Einthoven has calculated that in a particular subject (“Bak”) the heart had rotated in the chest around a sagittal axis during the movement of expiration by an angle of —36° (ie. from « = 76° to a = 40°, referred to the horizontal). 2. Determination of the Electrical Azis. I have approached this problem of the angle from a different standpoint. namely, from the point of view of my first observations of 1887-9, and the distinction between “favourable” and “ unfavourable” leads, or, as I now call them, “strong” leads and “weak” leads. In order to calculate the obliquity of the cardiac current-axis, I took values between a mesial point and the two sides of the body, in fulfilment of the rough notion of a balance of which the two unequally loaded arms, R and L, give an angular deflection of the indicator in relation with a weight-difference between R and L. In electrical analogy with this idea, I took for calculation the values of the potential differences between leads from the mouth and right hand and between the mouth and left hand. According to this picture, the electrical pivot or zero is an electrode in the mouth, and the weights are the potential differences between the electrode M and the right hand R on one side, and the left hand L on the other, to form the other electrode. These two leads may be referred to as the right superior and left superior respectively, in distinction from analogous leads between hands and feet, which will be referred to as right and left lateral. The formula for calculating the angle a between current-axis and vertical is very simple: tan « = (L—R)/(L+R), where L and R represent respectively observed magnitudes of potential difference at the outset of systole by the left and right superior leads respectively ; for example, with L = 9 and R = 3, Cas Oe Similar considerations apply to the inferior (or posterior) half of the body, the feet (or either foot if we admit that the small P.D. existing between the galvanometrische Registrirung des menschlichen Elektrocardiogrammes, zugleich eine Beurtheilung der Anwendung des Capillar-Elektrometers in der Physiologie,” ‘ Pfliiger’s Archiv,’ 1903, vol. 99, p. 472; “ Le Télécardiogramme,” ‘ Archives Internat. de Physiol.,’ 1906, vol. 4, p. 132; “ Weiteres tiber das Elektrocardiogramm,” ‘ Pfliiger’s Archiv,’ 1908, vol. 22, p. 517; “The different Forms of the Human Electrocardiogram and their Signification,” ‘ Lancet,’ March, 1912, p. 853. * Or, more precisely, 26° 34’ but in this connection a will be given to the nearest degree only. 512 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, two feet may be treated as negligible; but by reason of the greater acuteness of the angle subtended by the R and L leads at the inferior lead F, as com- pared with the superior lead M, a new factor must be introduced into the formula, which is now taken as tan 2 = 2(R—L)/(R+L), where R and L represent respectively observed magnitudes of the potential differences in the right and left lateral leads. If, for example, the values have been observed, R= 12, L = 4, then 12-4 _ 16 : z TEA See Hy oe Put into words, the general conclusion expressed by the two formule* for calculating the direction of the cardiac current-axis from the right and left potential difference existing at the outset of systole above and below the heart, is as follows:—The tangent of the angle formed by the current-axis with the vertical is proportional to the difference between the potential differences of the strong lead and the weak lead divided by the sum of these two differences. The error of angle by error of measurement of the spike is, with most records, inconsiderable, except in the case of the left lateral lead, when its direction is doubtful. When it consists of a small positive followed by a large negative peak we may hesitate whether to take into our formula the small positive or the large negative value; in such case of doubt it is advisable to calculate the angle for both values. If, in a doubtful case, the right lateral is smaller than the transverse spike, the negative value of the left lateral should be taken in the formula; if the right lateral is larger than the transverse, the positive value of the left lateral should be taken. But I do not attach much value to the numerical result in such a case (vide infra fig. on p. 520, the case of Dr. E.). In most normal records, where the ventricular spike is clearly positive, there is no difficulty in measuring out its values for the right and left hands. I have not taken into reckoning the small preliminary negative movement (“Q”) by which the positive movement is more or less distinctly preceded. This negative movement, while forming part of the systolic change of potential, occurs before the ventricle has begun to contract. As far as it is possible to judge from simultaneous records by the R and L lateral leads, taken on a film travelling at the rate of 130 mm. per second, the two spikes R and L, when both are positive, culminate synchronously. tana * The two formule might have been expressed by a single general formula, as follows: tan = cot 6(S—W)/(S+W), where 6 stands for the angle taken at M or at F, and S, W for the values of strong and weak leads respectively. But the two special formule are preferable for practical calculations with a conventional direction of currents. 1913. Electrical Axis of the Human Heart. 513 Where the spike is positive on the right side and negative on the left, the positive maximum was reached 0-006 sec. earlier than the negative minimum, and, as far as can be judged from the record, the two events appear to commence synchronously. From the inspection of the left lateral spikes of intermediate type, + —,it appears that the transition from 1 to 4 is by a retrograde encroachment of — upon +. 3. Proof of the Formule: tana = re R-L The normal difference between “strong” and “weak” leads is evidently dependent upon the normal obliquity of the heart. The principal electrical event of the beat, ze. the systolic spike, is the resultant along a line BA or CC representing the current-axis, of all the component differences of potential existing in the heart muscle at the outset of systole. Comparing the magnitudes of the spikes of the two sides it is obviously in correspon- dence with the normal tilt of the heart’s axis that the left superior is the “strong” lead and the right superior the “ weak ” lead. From the observed magnitude of the two spikes, right and left, it is easy to construct a geometrical figure by means of which the numerical value of the obliquity can be expressed in terms of the angle made by the current axis CC with the vertical line MV. In the triangle MRL, the sides MR, ML indicate the right and left superior leads, M being the mouth, R the right hand, L the left hand. Actually, in relation to the heart, these leading points are to be regarded as sections through the neck and shoulders, and for convenience of calcu- lation we shall take the angle at M=90°, so that MV =4RL=RY. Taking, ¢.g. the right-hand spike = 1 and the left-hand spike = 3, we have at M the potential = 0, at R the potential = 1, at L the potential = 3. Projecting the potential of M on to the horizontal line LR produced to O and taking the position of the point O such that the length OR shall represent the potential difference between M and R, and OL the potential difference between M and L, we have MO as the line of zero potential. A 514 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, line VP, drawn perpendicular to this zero line MO, gives the position of the current-axis CC forming with the vertical MV an angle, which call «. The angle MOV = «; tane = tan MOV = MV/OV=4=05. There- fore « = 26° 36’; or otherwise: since MV = }(L—R) and OV = 3(L+R), we may write tan « = (L—R)/(L+R), or, in words, the required angle a is an angle having for its tangent the fraction of which the numerator is the difference between the spike of the strong lead and that of the weak lead, and the denominator their sum. The formula holds good for negative values of the weak lead, as can easily be shown by a geometrical construction. But an example will be sufficient. Let —1 be the value observed by the weak lead R, and +3 that for the strong lead L. nn 2 = = = = = 2, a = 64° Finally, we may mention two particular cases that are occasionally observed. If the weak lead R = 0, then tan 2 = ed = il, oe If the weak lead R is negative and greater than the strong lead we shall have « greater than 90°, ze. a current-axis upwards from right to left. This case may also occasionally be met with as an extreme case of the “cor breve et molle,” eg. let R = —2 and L=1, al Sempre ea a 2 eS JUG An essentially similar principle of calculation is applicable to the data afforded by leads from the four extremities—the two hands and the two feet —with, however, certain modifications. The modifications to be taken into account are: (1) that the two feet are assumed to be equipotential and represented by a single foot F at the inferior angle of the triangle RLF in analogy with the point M of our first triangle; (2) that therefore we are to regard as equivalent the right lateral R Vv L re) with the axial leads, and the left lateral with the equatorial; (3) that we have to remember that with the extremities, R is the strong lead, L the weak lead ; and (4) that the conditions require us to take at F a more acute angle than in the case of the superior triangle we have taken at M. F In the figure representing the inferior triangle the length VF has been taken as equal to the length RL, ae. twice RV. The formula now runs: tan « = 2(R—L)/(R+L), where R and L stand for the 1913. ] Electrical Amis of the Human Heart. 515 magnitudes of the systolic spikes of the right lateral and left lateral leads respectively. (In the illustration we have taken R = 3 and L = 1.) It is scarcely necessary to give the geometrical construction from which the formula is derived. As in the previous case we have tan « = the difference R—L divided by the sum R+L and multiplied by FV/RV, the cotangent of 4 the angle RFL. For the superior triangle we took MV/RV = 1, for the inferior angle we take FV/RV = 2. The application of our formula may be illustrated by examples. It has been observed, eg., that R = 3 and L=1; then 3-1_4 t = —— = — | uo = Ae. an a 341 4 5 a 9) It has been observed, ¢.g., that R = 3 and L = —1; then oe 8 ° ili =? — = — bck — 5 an a males 4, a6 We shall now apply our formula to some actual cases :— The Case of B. 0. B. (Inspiratory Values.) ° , 22 Right superior ...... 75 M Left superior ...... 175 R L (Transverse ......... 16) Right lateral ...... 23°5 Left lateral ......... 155 L+R 175+75 25 R=L _ 235-155 _ 16 _ R+L "2354155 39 516 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, The Case of J. C. W. Right superior ...... —2 o M 6l Left superior ...... a R L (Transverse ........000 11) Right lateral......... 14 o Left lateral ......... 7 oe F Sup. tana = disse a 18, oy =O, 745) 14-7 _ 2 ° Z = 9 = = = 0'67 aS = Sb, Inf. tana way a 67, a The Case of A. D. W. a M Right superior ... -2 60 _ Left superior ...... 75 R L (Transverse ......... 12°5) 86° Right lateral ...... 10 Left lateral......... 75 F 7T54+2 95 = ° SS aa SSS 3 = Sup. tana os ae 1°73, a = 60 Tne frame ee ea ae BG: 4. Apparatus Used. In these observations I have used :— (1) An early model by Edelmann of Einthoven’s galvanometer with a platinum fibre of 2000 ohms resistance, and platinum electrodes dipping in normal saline. (2) An oscillograph (Bock-Thoma model) by Thoma, of Munich, with a coudenser of large capacity (10 and 20 mf.) in circuit, and, as before, platinum electrodes in saline. Each of these instruments possesses advantages and disadvantages; for the special purposes of the present observations the second instrument has proved to be the more convenient. By the use of platinum electrodes 1913. ] Electrical Axis of the Human Heart. BAU) (which in themselves act as a condenser of considerable polarisation capacity, é.g. in one measured case about 10 mf,), supplemented by an added condenser in circuit, we are rendered independent of alterations of resistance; the galvanometer is practically converted into an electrometer. 5. Influence of Position and of Respiration. The influence of slight alterations of position of the body upon the electrical record both as regards form and as regards amplitude is trifling. I did not notice any such influence in my first observations. Einthoven has subse- quently found that in the recumbent posture turning from the left over to the right side alters the form of the first ventricular wave from simply + to +—. But in view of the calculations to be made from the relative amplitudes of the left and right records, I thought it necessary to re-try this point in order to learn whether alterations of position great or slight cause alterations of amplitude. I found that slight alterations are negligible, but that great alterations, such as from standing to sitting, and lying either on the back or on the face, or on one or other side, alter the amplitude and the angle. I have therefore taken all observations from persons in the most convenient position, 7.e. sitting. The effect of considerable alterations of position is evident from the following observation, in which the transverse, right and left lateral records were taken of the subject B. O. B. in the standing, sitting, and lying positions. The points that come out most clearly on review of this group of records are that lying on the left side as compared with lying on the right side diminishes the angle a, as shown by increase of the left lateral spike and diminution of the transverse spike. This alteration is also brought about by muscular exercise (vide infra), and the reverse alteration, viz., increase of the angle z, is caused by distension of the stomach. B. O. B. (Feb. 26, 1913). | | } | | ¢ alse Transverse. R. lat. L. lat. | a.¥® requency. | | Insp. Exp | Standing ........00... vgn 14 o 4 | 7°5 31 (29 45) | Sith RA ada eseeeck ester 64 15 14 5 43 (40 53) | Lying on back ......... 62 15 18 | 75 40 (39 50) | Lying on right side...| 62 17 18 5 48 (42 50) | Lying on left side .... 60 il 16 10 25 (25 27) | | * The values of a given in parentheses are the approximate inspiratory and expiratory numbers calculated from maximal and minimal values of the R and L spikes. 518 Dr, A. D, Waller. Various Inclinations of the [Mar. 6, There is no more than a general correspondence of inclination between the line drawn through the heart to represent its anatomical axis and the line or lines representing its electrical axis. This is hardly surprising when we realise how indefinite is the anatomical axis and upon what data and assump- tions the determination of the electrical axis depends. I regard the latter as the more definite of the two lines; it gives expression to the functional resultant at the outset of contraction of the living organ, and indicates by considerable and definite variations changes of axis that are difficult to settle either post mortem or by skiagram during life. 6. Tone of the Heart Muscle. I attribute considerable importance to the tone of the heart muscle as regards position of the heart and of its electrical axis. With soft muscle the heart is sessile on the diaphragm, and the axis, as calculated from right and left lateral spikes, is approximately horizontal. (See fig. on p. 520, the case of Dr. E.) With hard muscle the heart is more nearly erect on the diaphragm, and the axis, calculated as before, is more nearly vertical. (See fig. on p. 521, the case of Dr. D.) Influence of Respiration—I paid no attention in my first observations to the effect of respiration upon the electrocardiogram. This effect has since been carefully studied by Einthoven, who has shown that with forced inspiration the amplitude of the record of Lead III (left lateral) is increased, while that of Lead I (transverse) is diminished ; the changes in Lead II (axial) he describes as very slight. My observations are on the whole in harmony with those of Einthoven, but their full discussion must be post- poned. For the purpose of the present communication it will be sufficient to state that I have found it necessary for any exact estimation of the angle to take into account the phases of ordinary respiration, which briefly are of the following character :— With inspiration the left superior and the right lateral records are dimin- ished, the right superior and the left lateral records are increased. In taking out values of R and L fora careful determination of « by the appropriate formule it is therefore necessary to take for its inspiratory value the smallest values recorded of the left superior and right lateral leads and the largest values recorded of the corresponding right superior and left lateral leads. For the expiratory value we must take out the largest values of the left superior and right lateral and the smallest values of the right superior and left lateral. But this troublesome correction is in most cases superfluous. It can, however, occur that normal respiration brings out differences of amplitude that lead to differences of angle of 10°, eg. the case of J. B. F., 1913. ] Electrical Axis of the Human Heart. 519 where I first noticed that the correction might be required, and I have therefore thought necessary to say that I have been alive to the error that might be made by neglecting the correction where it was evidently of moment. The type of the electrical pulse by different leads is remarkably individual and constant. Thus it has not altered, as far as I can tell, in the cases of A. D. W. and A. M. W. and T. Goswell between 1887 and 1913. Nevertheless, temporary variations do occur in a given individual—varia- tions of frequency of course, but also variations of form and variations of the angle « I think that such variations are attributable to greater and lesser repletion of the stomach and of the several cavities or of the two sides of the heart. In at least two instances where the angle « has been found greater in the same individual at one time than at another, I have associated the value of the angle with the state of health. But the dis- cussion of this important point cannot profitably be entered upon until the effect of respiration has been fully considered ; and it properly belongs to a future communication on the pathological significance of the angle «.* 7. The Influence of Muscular Exercise. As was to be expected, the frequency and character of the electrical pulse are altered with muscular exertion. Electrocardiograms afford indeed the most convenient available means of exactly counting the pulse, and of measuring out the working-time of the heart from varying relations between length of systole and length of diastole. But the primary object of this paper is to study variations of the angle «, and in this connection it appears that any variations that might be expected to result from variations of repletion of the several cavities is masked by the large variations caused by deepened respiration— especially inspiration. The detailed consideration of the influence of exercise must therefore be subordinated to that of the influence of respiration, and at present it will be sufficient to give the results of observation on one subject (B. O. B.) in illustration of the fact that muscular exertion indirectly through modified respiration, and perhaps also directly by modifying the repletion and shape of the heart, does actually bring about considerable modifications of the electrocardiogram and of the angle « as calculated from its right and left hand values. * A further complication arises where the left lateral lead is negative ; the correction has then to be taken in the opposite sense, because whereas a positive left lateral is increased with inspiration, a negative left lateral is diminished. The same holds good for a negative right superior record. But the discussion of these points must be postponed. 520 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, Dr. £.—The superior angle a can be estimated without ditticulty from the values of the right and left superior spikes— Dek } epee a ark Ge 125-75 5 But the mixed character of the right and left lateral spikes, viz., positive followed by negative, does not afford assured data for calculation (see text, p. 512). Taking into the formula the positive values 75 and 5, the angle comes out as 22°. Taking a positive value 7°5 and a negative value —25, a = 105°. Taking the negative values —12°5 and —25, a = 34° of reversed current direction. But I place no reliance on these figures. Right superior. _ Left superior. Transverse. Right lateral, Left lateral. EOS. | Electrical Axis of the Human Heart. 521 Dr. D.—In this case the electrocardiograms were taken with a simultaneous record of the respiratory movements, and the measurements are given for the same (inspiratory) position in each case. Upon another occasion the R. and L, values were found to be 20 and 15, ze, the inferior a came out = 16°, but no attention was then paid to the phase of respiration. In ’the accompanying records the values are as under :— 75-5 ° Re tan pe kat Sup an a 754 a 1 i 22°5 —20 Inf. ti c= 9 == aI aS nf ana 995.90 OO Left superior. Transverse. Right lateral. Left lateral. 522 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, Pulse frequency. Transverse.| R. lat. L. lat. a. Insp. Exp. | Insp. Exp. | Insp. are, Insp. Exp. Atirest! oie ekeccasoemsen Sooner 66 TOW FL 13 «15 7 31° 45° Immediately after exertion, 2000 kgrm. in 35 secs....... 146-126 775125] 12 14 LOSS 10° 44° Two or three minutes later...| 98-96 10 15 13° (15 o -& 20° 45° | Pulse frequency. R. L. a. Normal ate sseeweeenes 96 to 86 11 2°5 52° | Exp. Insp. | Exp. Insp. Second minute ...... 120 12 °5 5 to 10 | 41° to 12° Fifth minute ......... 104 | 14 Seat EO 52° ,, 438° Thirtieth minute ... 72 12°5 2°5 53° The inferior angle a is diminished in consequence of muscular exertion. 8. The Influence of Food. From measurements taken on the same individual before and after food, the angle « has come out greater in the latter than in the former state. But the numerical estimation of this difference cannot be discussed with profit apart from the consideration of the respiratory variations of angle. In the case of B. O. B. the difference has come out = 10°, the actual measurements having been as under :-— ¢ Pulse Transverse.| R. lat. L. lat. tan a. a. requency. | | | Before dinner ......... 62 11 12°5 6 0-70 35° After dinner ............ | 70 1174 955 12°5 | 4, 1:02 45° | j The left lateral is smaller with a full stomach. The right lateral is not appreciably altered. The transverse is increased. The angle a is increased. 9. Thoracic Leads. The preceding considerations dealing with leads from the mouth and extremities apply to the current-axis in the frontal plane. Similar con- siderations can be applied for measurement of the current-axis in the antero-posterior (sagittal) plane, and we may calculate superior and inferior values of « by the same formulz as those taken for their values in the frontal plane. 1913. | Electrical Axis of the Human Heart. 523 To compare effects in the sagittal plane, the leads to be taken are :— Mouth to precordium and mouth to back for the upper part of the body ; back to foot and precordium to foot for the lower part. We can calculate the value of an angle « formed by the current-axis and the vertical in the antero-posterior or sagittal plane, by taking the values of the two upper and two lower spikes. And without laying stress upon the precise values obtained for the angle it is satisfactory to find that the arrow representing its position and direction comes out from the calculation conformably with what might be anticipated. To complete our picture of the current direction we may compare effects horizontally in transverse section across the body ; this can be done roughly by the leads from precordium and back to the right and left hands, or better, by taking the effects from four symmetrical leads encircling the chest on a level with the heart. This symmetrical arrangement of electrodes gives a more satisfactory set of values than is afforded by the two hands with the front and back of the chest, which gives values in an oblique and principally frontal plane, as is illustrated by the following group of measurements :— B. O. B. connected to an oscillograph by large (100 x 75 mm.) electrodes on the right and left sides of the chest, and by the right and left hands dipping in saline, gave the following values :— Right hand a eee i’) 0: Right side Left hand be Left side } a Right hand 22°5, Teeeaide } eee Left hand > +2°5. Right side i =, At first sight these results appeared rather perplexing, but their obscurity disappears when we reflect that leads from the hands are, as regards the heart, equivalent to leads from the shoulders, and plan the results accordingly along the lines of a frontal diagram. We then realise that of the four leads, the 3rd, being “axial” as regards the heart, must be the “strongest” lead, and the 4th lead, being “ equatorial,’ must be the “ weakest ” lead. In a transverse section of the thorax, z.e.in the horizontal plane with the body in the erect posture, similar differences obtain between the two sides. Taking leads from the precordium and first the right then the left hand, we VOL. LXXXVI.—B. 2Q 524 Dr. A. D. Waller. Various Inclinations of the [Mar. 6, find that the first lead is stronger than the second. Taking leads from the back of the chest, and first the right then the left hand, we find that the Back Mouth POSE Ank. Right Lefr Fronr Foor first lead is positive, the second small. Taking, eg. the values observed for these four leads in the case of B. O. B. the angles come out as follows :— no = EBs i) = (30, ENE — 120% 16+7°5 eNO oo = pote = p, or ches Ho The angle formed above the heart by the current-axis with the vertical in this plane can be estimated from the relative values of the spike im the two leads mouth-precordium and mouth-back. Thus, eg. in the subject B. O. B., the values of the spike in these leads are: anterior = 25, posterior = 10, from which our formula for the superior triangle gives tan « = (25—10)/(254+10) = 15/35 = 0-43, so that the required angle a = 23°. The angle below the heart can be obtained by calculation from the values observed in the two leads precordium-foot and back-foot. In the case of B. O. B. these values are: anterior = —10, posterior = 10; 10+10 3 t = 2 ce so) ee an a 10-10 oe a = 90 This result may be taken as indicating that for the superior part of the heart and body the algebraic sum of current is to be pictured as an arrow directed forwards and downwards at an angle of 23° with the vertical ; while for the inferior part of the heart and body the current-axis is repre- sented by an arrow directed horizontally forwards at an angle of 90° with the vertical. j I have tabulated, from my laboratory notes and records of “normal” subjects, the values of right and left hand records, and of the angle « calculated from them. 1913. ] Electrical Axis of the Human FHeart. 525 Values of « calculated from Workers in and Visitors to the Physiological Laboratory. x - Trans- | Right Left | « | Right Left a o 8 | verse. superior. | superior. superior. lateral. lateral.| inferior. | 1 | Thomas Goswell | 53 8 0 8 45 9 3 45 Zi PAG DOW es ns cacawoss 56 12 —2 dad 60 10 — 75 86 DOM Nason 12°5 —5 10 72 7:5 |—15 108 yee PAC IVES. Wry cecncic dt | 538 —3 8 66 13 3 AT Aiea PAC Ge, Wise accseeat 27 5 6 8 8 16 18 —10 GH NVYG VWs Seis 24 8°5 0 9°5 45 10 ail tb 70 ORS MED EW .sccccsclis 26 11 2 10°5 65 |— 8 93 (I Ole KO Ni oeesaneeen ae 21 il —2 a 61 14 a 34 San NE Ob Be 20 14 7°5 17°5 22 23°5 15°5 22 | (Insp. values) 29 9 3 45 SR PAIS So cvzsccskec couse +1 53 5 0 5 45 5 { iia ae 5 10 8 13 TOME NV Tis 88 © hace { 2 ie ‘ 39 io F aa TE lia 2 \ia Creech Re 46 18 5 175 29 8 —4 81 Insp. 12 —10 80 WOW Weak e, 18 4 16 31 { te ie \Le ee TM Geet insadsvccstsss vc. 29 2 75 30 22°5 17°5 14 TSP ISU eieecatcswcce 0 15 45 17°5 10 28 DOM ra Sisk valor: 15 75 33 Do | { Exp. 20 75 42 oe Insp. 17 | 11 23 VAD ES Gon ees aclaces vos 5 10 18 12°5 15 —10 ID OV > Se eae ere 5 7 11 10 10 (0) ie opeee RW? ts. 6: 12 —2°5 75 64 5 { ae e UG | Dye, 1315 JD), oneotone 5 5 75 11 20 15 16 IDLO EP RPEReree: 22:°5 20 u (Insp. values) Lenten: oh aie. =i || ies 76 7 { ee i 102 + 7:5 |+ 5 DOR rics: 20 —7°5 12°5 76 { Sips Webs 105 1S}. || Dyes: ais abate 5 15 16 —4 US) 1 TDS Det Bena eaat eee —5 10 72 5 —10 99 FAQ) |) INATRHE@ 18s GSnncnene 20 a 12 6 34 ALE R WR aes. 8 20 6 15 | 12 12 ap Exp. 22°5 | 12:5 30 | sec 48 { eerie | ae 7 DeHGe emlat he Ar 45 25 10 59 { eae 3 ie Bea ied vee ee —4 12 64 | 7H | Sire ID IS ssonnenes 60 6 0 6 45 + 2 |+ 1 34 7A ||| Nes TSS NAYS sboosacee 55 —7°5 12°35 76 15 7°5 34 26 | Dr. G. B 10 4, 40 Zip (Gs ONG ieee ca | 30 7°5 2 10 34 12°5 6 35 WC Rccneesaconcenr: | 15 6 Al 28 | Miss F 25 9 2°5 85 29 12 7 28 AS) || Wie, Er” S.cesoccosne 7 | 6 9 SOM Se as caceacn 10 12 —10 (Situs inversus) | | 31 | Captain E. ...... | 9 — 4 87 copll anya ee a | : Insp. 15 ey Aas) 34 | TRUE o.. 86 25 | 10 al { Epis | 5. | 48 33 G. R. M. " | Insp. 12°5 7-5 Insp. 27 Bee ce 29 10 4 24 { ee ait ise SH IDR SES coacoorne 45 12 Vee 1S 58 526 Inclinations of the Electrical Asis of the Human Heart. No Vas Trans- | Right Left a Right Left a ; 8° | verse. | superior. superior. | superior. lateral. lateral. | inferior. : Insp. 8 4°5 29 35 | Miss B. B. G. R. 10 1 8 38 { eS ae a 36 | Miss M. F. H. ... 12 —4 12°5 63 9 Al 70 37 (| Miss M. M. A 15 —2 14 52 iG -—7 90 38 | Miss M. G. B 9 2 10 34 12 A, 45 39) MassieNe Ue ieee 10 —2 10 60 9 2 52 40 | Miss P. W. ...... 5 Ad i 33 8 6 16 41 | Miss H.6&.......... 3 2°5 6 22 10 6 27 42 | Miss H.H. ...... 7°5 0 8 45 i 4°5 23 43 | Miss M. EF. ... 5 5°5 0°5 40 9 A 38 Ae Ge Wie Meena 50 8 9 3 45 Exp. 10 6 27 AB NT) uae Moule 35 6 { OS 2 mi Exp. 10 5 34 AGI) Minst Wittenmerecee: 7 { ee 9°5 6 24 Exp. 8 4, 34, Alen Nir Agave an 6 { Tae. pee E 35 ASH Sir) oe eeaeee 15 —2 14 53 75 |— 6°5 88 CS Mba AY Deal BA) Bret ae Sart 3 9 27 15 10 22 G0) | TEAyove, IEE) soe onoee 10 2 10 34 13 5 42 OL | Dr eeepc 46 6 2 75 30 17°5 14 12 GPA IDI IG See soseodee 6 —2 6 64 + 5 + 3 27 G3} |) Ire, Er, Os. Seocwoace 12°5 —3 13 58 7 — 5 85 O41) |-Erok give eee rence 54 6 —0°5 5 50 10 A Al BO) DET IKE, istectsi nee 6 2 7 29 15 10 22 G0}. IDS WAYS cgecansnoace 4 2°5 5 18 6 3 34 BS) || Wabi ISIS scone 20 5 —0°5 5°5 50 9 4 38 58 | Miss G.H. ...... 20 | 7-5 1 9 eee | a 59 | Miss TL. .......... 2.20 | 6 1 5 a4 {ee a Sol) eee (GO). ) IDI 18Is seoneeacavis ec. 50 8 5°56 |— 2°5 37 527 Acineta tuberosa: A Study on the Action of Surface Tension in Determining the Distribution of Salts in Living Matter. By A. B. Macatuum, Ph.D., Se.D., LL.D., F.R.S., Professor of Bio-chemistry in the University of Toronto. (Received February 19,—Read May 29, 1913.) [Puates 14 anp 15.] CONTENTS. PAGE I. Introduction: The General Effects of the Action of Surface Tension...... 527 II. The Material and Methods of Investigation................:scsesecscesceeecnceees 535 MPR CRENCSULGSH cteccces tec ceceseaceeeemcemetac cence tscccsatns dacstertcemsossdaccecdertecree 536 Veen GCN era li ODSELVALIONNS ys. cae nacsreseeecese sare ses ccsecencscs caves -\cinseseciceeiacssseease 542 V. Summary of Results and General Observations ...............:eceeeeeeneeeenees 547 AVAIEMMBINT GET AUT Cl: farce ates ccecocsea mente ccteee teat ccisecaeav ack cceitetedcesecssesescsindsordencscscs 548 WIL TDsgelbmeni@n OW TEENS) | Soaceacccosocecoounoo ac encoLesood doSnOOBudSoH SBE abeoHoseubsobosnn: 549 I. Introduction: The General Effects of the Action of Surface Tension. The distribution of salts in living matter is supposed, in the current conception of the subject, to be on the whole the same as in ordinary fluids. Living matter is generally regarded as a semi-fluid, semi-viscid material in which the conditions, though not fully typical of those which obtain in a fluid like water, are, nevertheless, such as to allow the substances that are dissolved in it to be distributed uniformly throughout it. The only obstacle to this distribution may be presented by a membrane such, for example, as that which encloses or surrounds the cell nucleus. Elsewhere throughout the cytoplasm there is, it is believed, a free play of the force that determines the diffusion of the substance or substances dissolved, until uniformity in their dispersion obtains throughout the volume occupied by the cytoplasm. This force is that postulated in the van ’t Hoff theory of solutions extended to include the Arrhenius theory of dissociation. In this composite theory, as 1s well known, the material dissolved in a fluid is supposed to be in a state analogous to that of a gas, that is, in as rarefied a condition as if its molecules were isolated from each other and occupying alone the volume filled by the solution itself. The molecules of the solute and their ions, when they are dissociated, are thus supposed to be in translational motion, and the resulting pressure—the osmotic pressure—which they give, acts on the surfaces enclosing the fluid as the molecules of a typical gas act on the walls confining it. At every point in the system there would be, on this VOL. LXXXVI.—B. 2k 528 Prof. A. B. Macallum. Acineta tuberosa: [Feb. 19, view, the same pressure and, in consequence, the number of molecules per given volume of the solution in any portion of it would be uniform. On this conception of the force determining the distribution of salts in fluids there were based a number of views which have played a part in explaining physiological processes. Of these the most important is that which postulated that all diffusion, whether in the cytoplasm of a cell or through a living membrane, is due to osmotic pressure acting as a driving force, the ultimate result of whose action would be to equalise the pressure throughout the cytoplasm or on both sides of the membrane. This reduced the processes of secretion and excretion, as well as the diffusion into and from cells, to the operation of gas laws. This explanation of the force and the conditions that make for the diffusion of solutes, not only in physical solutions but also in living matter, has obtained and still obtains a wide acceptance. The very simplicity of it, the support it derived from a considerable range of experimental evidence, and the unifying effect it appeared to exercise in a large number of phenomena manifested by solutions and gases, told very strongly in its favour, and eventually revolutionised the aspect from which all the problems of osmosis and diffusion were viewed. Criticism, however, was not silenced. It was seen that there were phenomena which not only could not be explained in that way but also were irreconcilable with the explanation. These were physical and physiological. Of the physiological only is there concern here. In one of these, that observed in renal excretion, the concentration of the urine is much greater, ordinarily, than that of the blood plasma from which it is derived through the activity of the kidney tubules. In other words, the osmotic pressure of the product of renal action is greater than that of the blood. . This cannot be explained by the van ’t Hoff-Arrhenius theory, the only conceivable result of which would, in such a case, be approximately an equality of pressure or that the urine formed would not exceed in concentration, and, therefore, in osmotic pressure, the blood plasma itself. The failure of the van ’t Hoff-Arrhenius theory to explain this and other physiological results of a like character does not put the theory out of court in explaining many physical phenomena. It still may be regarded as of value in accounting for these, though even in this respect it may be looked upon as beset with limitations. In the physiological sphere its application is of much less service and, were it here the last word in the way of an explanation, the causation of a few physiological phenomena would ever remain an insoluble problem. In recent years the aid of other factors in explanation of certain physio- —————- = - 1913. | A Study on the Action of Surface Tension. 529 logical phenomena has been sought for, and, as a result, attention has been specially directed to the principle of surface tension. The participation of this force, although considered as a factor in the causation of amoeboid and contractile movement ever since 1869, was not ‘suggested as influencing the distribution of salts in living matter till 1910, when the author advanced the view that surface tension plays an all-important 7d/e in determining the localisation of salts and other solutes in cells and tissues, and in controlling the diffusion through living membranes that brings about the formation of éxcretions and secretions. It was the observations derived from the microchemical study of the distribution of potassium salts in living cells that led to this view. The compounds of this element are amongst the most soluble of all known salts. Only two of its salts are under certain conditions insoluble in water, and these are the triple salt, the hexanitrite of cobalt, sodium and potassium, and the double salt, the potassium platinum chloride. Neither of these is found in the natural world, and therefore the salts of potassium found in living tissue, when ageregated in masses or layers in a cell, cannot be so localised as a precipitate. Some other explanation for this localisation had to be sought for, and the author, after full consideration of all the facts involved, claimed that the localisation observed is a condensation due to the influence of surface tension. How such a condensation may develop through surface tension may be recoonised on an examination of the results of the action of the Gibbs— Thomson principle of surface concentration. This law or principle may be ‘stated in a few words. It is to the effect that when a substance in solution increases the surface tension of a fluid system (eg., a drop of water) it is less concentrated in the surface layer than in the rest of the system, while a substance that lowers the surface tension of the system 1s more concentrated in the surface film than it is in the rest of the system. It has also been found by Lewis and others that solutes which raise the surface tension at a water—air interface as well as those which lower it, also lower it at a liquid— liquid interface and at a liquid—solid interface and undergo condensation there, as a result, it is understood, of the operation of the Gibbs-Thomson principle. At such interfaces the degree of condensation depends on the extent of the diminution of surface tension as well as on the concentration of the solute throughout the fluid system, but, assuming the application of the gas laws to dilute solutions, the concentration as deduced from Gibbs’ formula for this value would be hak Ci, de, ieecRT ac: 2R 2 530 Prof. A. B. Macallum. Acineta tuberosa: [Feb. 19, where S is the surface excess per unit of surface area of the part affected, C the concentration of the solute throughout the fluid, co the surface tension value, 7 R the gas constant, and T the absolute temperature.* . The value of S as experimentally ascertained was in a great many instances very small. Forch (3) found that in a normal solution of sodium chloride, which raises the surface tension of water, the deficit in the surface film was 0:°024 mgrm. per square metre.. Whatmough (16) with a normal solution of acetic acid, which lowers the surface tension of water, determined the surface excess to be 0°2 merm. per square metre, and this concentration increases by less than 15 per cent. even when the concentration of the acid throughout the system was increased eight-fold. Milner (13) estimated that in a sodium oleate solution of 0:00204 gramme-molecular strength the surface concentration of the sodium oleate was 0'4 merm. per square metre over that of the solution generally, but from the data furnished by Reinold and Riicker (15) regarding the conductivity of films made from a solution of 1 part of sodium oleate in 60 parts of water, Milner estimated’ the surface excess therein to be 2-4 merm. per square metre. The results of Benson (1) obtained with aqueous solutions of amyl alcohol of 0:0375 molar value gave a considerably higher value, the surface excess of amyl alcohol reaching a concentration of 0:0394 molar value, involving an increase of about 5 per cent. Were these the only values to come into consideration, surface concen- tration, as a result of the action of surface tension, would be negligible, except for the solution of certain problems of very limited interest. There are, however, other experimentally determined values which make it plain that surface concentration is, under certain conditions, a very great factor in influencing the distribution of salts in solutions. These values were recently determined by W. C. M. Lewist (6, 7, 8) who, to ascertain them, employed ingeniously devised methods. The surfaces on which the condensations were studied were those of aqueous solutions in contact with hydrocarbon oil or with mercury. The oil or mercury was in the form of droplets or spherules of uniform size, the surface area of each of which was calculated from data derived from the total quantity of oil or mercury used and the total number of droplets or spherules formed. The hydrocarbon oil and the mercury were employed because they do not absorb or dissolve in themselves a trace of the solute from the solutions bathing the surface of the droplets or spherules. * For the development of this formula from the original values of Gibbs see S. R. Milner, op. cit. + From the Muspratt Laboratory for Physical Chemistry of the University of Liverpool. 1913. ] A Study on the Action of Surface Tension. 531 The hydrocarbon oil was employed in two ways: as an emulsion with the aqueous solution or as droplets, which, being lighter than the solution in which they were liberated, were allowed to ascend through a long vertical column of the solution. In the first instance a definite quantity of the oil was mixed with a known quantity of the solution and the mixture agitated for some hours in a shaking apparatus. The droplets of oil in the resulting emulsion were of approximately like diameter, which was in a large number of cases measured under the microscope. The average volume of each was determined, and from this and the total quantity of oil used the united surface areas of all the droplets present in the emulsion were calculated. The quantity of the solute in the emulsion, apart from that on the surfaces of the droplets, was ascertained by the application of the drop-pipette or stalagmometric method. As the concentration of the original solution was known, the difference between it and that ascertained stalagmometrically was the amount condensed on the united surface areas of all the droplets. In the second method the hydrocarbon oil was allowed to ascend in droplets through a long column of the solution in a cylinder which terminated above in a broad cup-like receptacle. The apparatus was so arranged that the connection between cylinder and receptacle formed of rubber tubing could be cut off by a pressure clip. The oil droplets could freely ascend to the interior of the receptacle, but the flow of the contents of the latter back into the cylinder was reduced to a minimum. The quantity of oil used was known, and the total number of droplets which ascended through the column of liquid was determined from an average of counts made for several selected test periods. .When the droplets had all ascended the connection with the cup-like vessel above was cut off by compression of the rubber tube below it, and the concentration of the fluid in the cylinder, whose volume was known, was determined. The difference between this concentration and that originally present, the total volume in the cylinder, and the united surface areas of all the droplets were the factors from which was deduced the concenttation of the solute on each cm.? of surface area. When mercury instead of hydrocarbon oil was used, the droplets, all uniform in size, fell through the fluid in a cylinder which ended below in a reservoir from which it could be cut off by the closure of a glass tap. As the volume of the mercury used was known and the number of droplets also ascertained through an average of the counts made for that purpose, the united areas of the surfaces of all the droplets were determined. The original concentration and volume of the solution being known and the final concentration ascertained, the amount of the solute condensed on each cm.” of the surface area of the droplets was calculated. 532 Prof. A. B. Macallum. Acineta tuberosa: [Feb. 19, Out of a large number of solutes used in these experiments only two, caffein and anilin, gave values which approximated the values postulated by C de ART aC; mentally ascertained values greatly exceeded the theoretic values. Thus, in a sodium glycocholate solution of 0°25-per-cent. concentration, the surface condensation on the droplets of hydrocarbon oil was 5 x 107* grm. per cm.?, that is, the directly ascertained value was about 70 times the theoretic value. When mercury was used the ascertained value exceeded the theoretic about 25 times. In sodium oleate solutions the directly determined value was about 100 times the theoretic. With Congo red, methyl orange, and sodium hydrate the experimentally ascertained values were respectively 25, 43, and 20 times the values derived from the Gibbs equation. Even if errors in calculation were allowed for in every case, the excess of adsorbed solutes was still so great that Lewis suggested, as a possible explanation for the great discrepancy, that the adsorbed material is in a gelatinised or colloidal condition on the surfaces of the droplets of oil or mercury. the equation S = — while in the case of the others the experi- Lewis found also that in the case of inorganic salts the quantity adsorbed exceeded the theoretic amount, but it was chiefly the cation that was so affected. In silver nitrate the silver adsorbed was 5 times in excess; in cupric chloride the copper was 17; and in potassium chloride the potassium was more than 30 times the calculated amount. This excess of the cation is, in Lewis’ opinion, probably due to electrical effects, since the oil used is negatively charged, and the potential difference between the oil and the water is approximately 0:05 volt. The diameter of a water molecule is, according to Kundt and Warburg (5), 3°39 x 1078, and, consequently, the diameter of its sphere of attraction at the oil-solution interface would be 6°78x10~® cm., but Lewis, accepting the range as equivalent to that postulated by Parks (14), namely, 13-4 x 107° cm., and using also the value 54x 107° erm. for the amount adsorbed per em.” from a 0:25-per-cent. sodium glycocholate solution as experimentally determined, calculated that the concentration of the bile salt in the superficial layer at the oil—solution interface was 40:3 per cent., or 160 times that obtaining in the rest of the solution. If, however, we assume that the range of molecular attraction is smaller than that postulated by Parks then the concentration of the bile salt at the interfacial surface must be extraordinarily high. How such a concentration can obtain we do not know, but an explanation may be tentatively suggested. It is not certain that the solute adsorbed 1913. ] A Study on the Action of Surface Tension. 533 is confined to a deposit within the range of attraction of the molecules on the interfacial surface. It is not improbable that there is no sharp break between the concentration at the interfacial surface and that in the solution generally—that, in effect, there is a shading off between the two. The amount adsorbed does, indeed, depend on the surface tension of the solution at the interfacial line, and the effects of this surface tension do not extend beyond the range of molecular attraction of the molecules on the interfacial line, but in the case of the solutes which are adsorbed in such extraordinary excess as to suggest the formation of a colloidal or gelatinous deposit, the latter must tend to produce successively superposed interfacial deposits until equilibrium is attained, in which case the deposit may extend with lessening concentration through a distance from the interfacial surface equivalent, it may be, to several, if not many, times the range of molecular attraction. These facts make it evident that in solutions in which interfacial surfaces, numerous as in emulsions and consequently of very great areal value, exist, the concentration of the solutes may fall very considerably through condensation of the solutes on the interfacial surfaces. This must result in lowering the osmotic pressure. The dissolved molecules free in the solution are fewer, and their pressure is less than that of the molecules in the simple homogeneous solution unaffected by surface tension, except at its boundaries or limiting surfaces. The osmotic pressure of a solution of potassium chloride contained in a beaker is, therefore, different in value from that of the same solution and like concentration in which numerous interfacial areas obtain through the presence of very numerous foreign non-soluble systems. The elemental unit of living matter, the cell, is constituted of unhomo- geneous material in which the essential constituents are chiefly in colloidal condition, that is, the constituents are minute particles or dispersoids, separating which is a fluid containing in solution the salts and other compounds characteristic of living matter. Such dispersoids present inter- facial surfaces of very great areal value, and if the surface tension of the fluid at the fluid—dispersoid interface is lowered by the solute or solutes there must be condensation of them on these interfacial surfaces. This would lower very considerably the concentration of the solutions in the fluid diffused through the living matter, and thus the osmotic pressure, otherwise due to the presence of such salts in living cells, would be nearly correspondingly reduced. Surface tension, therefore, through the operation of the Gibbs-Thomson principle determines in a very considerable degree the distribution of salts 534 Prof. A. B. Macallum. Acineta tuberosa: [Feb. 19, in the living cell. Further, between each cell of a tissue or organ and the lymph that bathes its surface there is an interface where the salts of the lymph may lower the surface tension, and in consequence undergo more or less of condensation there. This serves to lower the general concentration in the lymph, and the osmotic pressure is diminished accordingly. That such surface and interfacial condensations of solutes in living tissues and organs can occur, and thus modify the distribution of their soluble constituents, though definitely indicated by the results of physical investiga- tions, would largely remain of theoretical interest if there were not more direct evidence to this effect. Such direct evidence comes from investigation by microchemical methods of the distribution of salts in living cells. The microchemical localisation of salts in cells and tissues is as yet not advanced enough to enable us to determine the finer distribution in living structures of all its inorganic constituents, but along several lines it is complete enough to permit us to demonstrate a condensation of salts due to the action of the Gibbs-Thomson principle. This localisation is most effective in the case of the potassium salts. Eight years ago the author found that the hexanitrite of cobalt and sodium represented by the formula CoNa;(NO2)s, in an appropriately prepared solution, instantaneously precipitates potassium from its solutions as Co(NO2)3,3(K/Na)NO2+7H.20, in which the amount of the potassium ranges, according to K. Gilbert’s (4) determinations, from 16°31 to 18-21 per cent. The sensitiveness of the reaction may be understood from the fact that, in a solution so dilute as 1 part of potassium in 275,000 parts of a solution formed chiefly of the reagent, the crystals of the triple salt, the hexanitrite of cobalt, sodium, and potassium, are formed. In tissues the potassium is not always abundant enough to give with the reagent crystals of the triple salt, and in the vast majority of preparations the triple salt formed is evident only as a yellowish reaction, and, conse- quently, not sharply delimited in tissue preparations. If, however, the reagent is completely washed out of the preparation with ice-cold water, the application of ammonium sulphide, reacting as it does to form the black cobaltous sulphide, brings out, in an extraordinarily sensitive way, the distribution of the triple salt, and, therefore, of the potassium in it. This black reaction, as observed under the microscope, makes the distribution of the triple salt sufficiently sensitive to diagnose the presence of potassium when it is but 1 part in over 1,000,000 of tissue mass. With this method of localising the distribution of potassium, the author succeeded in finding that, in a number of animal and vegetable cells, the potassium is condensed on surfaces in such a way as to make it very highly eee, 1913. | A Study on the Action of Surface Tension. 535 probable that surface tension is the only immediate factor involved in this condensation. Of these he has given an account in recently published communications (10, 11, 12). More recently, however, the author has found a unicellular animal form, in which the action of surface tension in influencing the distribution of salts in it is, to all appearance, placed beyond doubt. This organism is Acineta tuberosa, a Suctorian Protozoan of marine habitat, a form which permits readily of technical manipulation and microscopic examination, especially for microchemical purposes. Some, at least, of the salts of sea water penetrate its cytoplasm, and amongst these certainly are those of potassium, of which the chloride is apparently the most abundant. The distribution of this element in the cytoplasm of this organism was carefully investigated, and the results were found to be of such significance as to justify the detailed description of them given in the following pages. Il. Methods of Investigation. The specimens of Acineta tuberosa used in these observations were found in abundance attached to brown filamentous alge growing on wooden wharves and floats, at and just below the surface of the water, near the Marine Biological Station of the Biological Board of Canada, in the Bay of St. Andrews, New Brunswick, in July and August of 1911. It was easy at any time during that period to get an abundant supply of these specimens by collecting a mass of the alge, which was carried to the laboratory in a quantity of the sea water of the immediate locality. In nearly all cases the material so collected was used within a few minutes—20 at the most—after it was collected. For the determination of the distribution of potassium in the Acinete, a mass of the algze was lifted with a forceps from the sea water, and allowed to drop into a quantity of the solution of the hexanitrite of cobalt and sodium. For the method of preparing this reagent, the reader is referred to an earlier article of the author’s (9). At the end of five minutes the mass of filaments was removed from the reagent and placed in ice-cold distilled water, which was renewed every three to five minutes, until, at the end of half-an-hour, all the uncombined reagent was completely extracted, and the only demonstrable cobalt compound present was that in the form of the triple salt, the hexanitrite of cobalt, potassium and sodium. The filamentous mass was now placed in a quantity of a mixture constituted of equal parts of glycerine and fresh or recently prepared ammonium sulphide, which gave the preparation a black reaction, due to the formation of the black cobaltous sulphide compound. 536 Prof. A. B. Macallum. Acineta tuberosa: [Feb. 19, In the transference from one fluid to another, goose-quill points, glass needles, or platinum points were used, in order to avoid contamination of the preparations with iron or other metallic salts, which would tend to give a bluish-black, black or brown reaction with the ammonium sulphide. The preparations were now made ready for examination under the micro- scope. For this purpose minute portions of the material subjected to the treatment described were teased out on a slide in a drop of a mixture of five parts of glycerine and one of ammonium sulphide, a cover-glass was added, and, after all the glycerine-sulphide mixture not included under the cover- glass was carefully removed, the edges of the cover-slip were luted to the slide with benzol balsam to prevent evaporation and to facilitate examination under high magnification with the microscope. Preparations so made have been found after 16 months to have retained all their original value and distinctness. For revealing the structural and other characters of the Acinete fresh material was placed in 10-per-cent. formalin solution, in which also it was kept. This material was treated with a saturated solution of scarlet red in 70-per-cent. alcohol to show the distribution of fat in these organisms. Further fresh material was treated with Zenker’s fluid, Flemming’s chrom- osmio-acetic mixture, and with saturated aqueous solutions of mercuric chloride. The material so prepared was used to reveal the minute structure of the organisms. IIL. The Results. The structure of an 3 ” ” zy 12 hrs. after last meal... 5 | 18 hrs. after last meal... Dose of thyroid and length of feeding. 1 lobe twice a day for 1 day 1 lobe once a day for 2 days 3 lobes three times a day for 1 day 2 lobes on first day and 3 lobes the second day Liver glycogen of thyroid-fed animals. per cent. trace Liver glycogen of normal controls. per cent, 0°79 3°03 3°88 2°96 7°22 Table I1].—Effect of Single Administration of Thyroid on Rats. Dose of thyroid. | Expt. State of digestion. | | | 1 3 hrs. after last meal.. 2 6 hrs. after last meal.. 3 9 hrs. after last meal.. i 4 11 hrs. after last meal.. 1 lobe 17 hrs. previously ” ” Liver glycogen | Liver glycogen of thyroid-fed of normal rats. controls. per cent. per cent. ial 4°2 1°6 3°7 0°5 1°8 @yal 556 Messrs. Cramer and Krause. Carbohydrate [June 10, The Effect of Thyroid Feeding on the General Metabolism of Carbohydrates. In the further study of this phenomenon we have been led by the following considerations. Two possibilities, which appear to be diametrically opposed to each other, suggest themselves as being capable of furnishing an explana- tion of this action of the thyroid hormone on the liver glycogen. Either thyroid feeding primarily increases the oxidation of carbohydrates in the organism or it primarily inhibits the function of the liver to form and store glycogen. In the former case one would expect to find the following com- bination of symptoms: (1) A formation of glycogen in the liver soon after a meal rich in carbohydrates followed by a disappearance of glycogen more rapidly than in the normal animals, (2) an increased tolerance for glucose, (3) a diminution of the blood-sugar, especially in the fasting animal. An inhibition of the liver function, on the other hand, would be reflected, (1) in the relative absence of liver glycogen even soon after a meal rich in carbohydrates, (2) in a diminished tolerance for glucose, (3) in a rise in the blood-sugar especially after a carbohydrate-rich meal. Information on the behaviour of the liver glycogen in fed and fasting animals can be gathered from Table I. It will be seen that the effect of thyroid feeding on the glycogen-content of the liver is independent of feeding. Experiments on the tolerance for glucose after thyroid feeding were made on dogs. A detailed account of these experiments will be published later ; but the results may be briefly summarised here, as showing that thyroid feeding produces a slight but distinct lowering of the tolerance for glucose. A dog, for instance, which normally could assimilate 100 grm. glucose without any glycosuria supervening and only began to excrete sugar in the urine after the administration of 110 grm. glucose, had its limit of assimilation reduced to 90 grm. glucose after three days’ feeding with fresh thyroid. The behaviour of the blood-sugar is at present being investigated in this laboratory by Mr. R. J. M. Horne. These observations are not yet completed but they are sufficiently far advanced to show that there is at any rate no diminution in the sugar-content of the blood, but rather the reverse. One must conclude, therefore, that the internal secretion of the inyeotd gland, when administered to normal animals, has an inhibiting influence on the carbohydrate metabolism. But since the utilisation of carbohydrates by the organism is not markedly affected, as is shown by the comparatively slight lowering of the tolerance for glucose, it follows that the thyroid hormone acts specifically on only one aspect of carbohydrate metabolism in so far as it inhibits the formation and storage of glycogen in the liver. 1913.] Metabolism in its Relation to the Thyroid Gland. 557 The Effect of Thyroid Feeding on the Protein Metabolism. This conclusion helps to explain a difficulty, already alluded to in the introduction, in the interpretation of the influence of the thyroid hormone on protein metabolism, One distinguishes at present with Folin two forms of protein metabolism,. the constant endogenous metabolism, which is independent of the protein taken in in the food, and the exogenous metabolism, which varies with the intake of protein in the food. Since thyroid feeding produces an increased protein metabolism even in the fasting organism, it follows that the thyroid hormone acts on the endogenous and not on the exogenous protein metabolism. One would expect, therefore, to find after thyroid feeding a marked increase in the excretion of uric acid and of creatinin, since both these substances are supposed to represent end-products of endogenous protein metabolism. We found, however, that the increased nitrogen excretion after thyroid feeding is accounted for almost entirely by the increased excretion of urea and ammonia,. while the excretion of creatinin and of uric acid is either not increased at all or only very slightly. This difficulty would appear to find its explanation in the fact that thyroid feeding affects carbohydrate metabolism in the manner described above. For the distribution of the urinary nitrogen after thyroid feeding is very similar to that which presents itself when carbohydrates are withheld from the diet.. In the latter case, too, there is a marked increase in the nitrogen excretion,* even when no protein is given in the food, and here, too, the increased nitrogen output is due, almost entirely, to an increase in the excretion of urea and ammonia.t ‘ ‘ ; 3 : ; A : mee say The Action of Adrenin on Veins. (Preliminary Communication.) By J. A.GUNN and F.B.CHAVASSE : ee : : 25 192 An Apparatus for Liquid Measurement by Drops and Applications in Counting Bacteria and other Cells and in Serology, etc. By R. DONALD, B.Sc. (N.Z.), D.P.H. (Oxf.) : =: 2 : : e 198 ~A Preliminary Report on the Treatment of Human Trypanosomiasis and Yaws with Metallic Antimony. By H.S. RANKEN, M.B. Glasg., M.R.C.P. Lond., Captain R.A.M.C., Member Sudan Sleeping Sickness Commission 203 The Liberation of Ions and the Oxygen Tension of Tissues during Activity. (Preliminary Communication.) By H. E. ROAF, M.D., D.Sc. : i w2ES Sao - - _ >. . : fis PRINTED FOR THE ROYAL SOCIETY AND SOLD BY \_ Or a ee ee HARRISON & SONS, 45, ST. MARTIN’S LANE, LONDON, W.C. Prue Three Shillings. No. B 586. March 6, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers’ are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. MSS. must be type-wnitten or at least written in a legible hand, and properly prepared as copy for press. Type-written transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should. be drawn on a large scale in Indian ink on a smooth white surface, with lettering adapted to a reduction in scale. Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.” —Statute VI, Cap. xii. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the ‘Proceedings,’ separately paged, and shall afterwards be republished in the “ Year-Book.’ The ‘Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. On application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for binding may be purchased, price Is. 6d. The ‘ Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The ‘ Philosophical Transactions’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in a cloth case for 2s. 6d., or the cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The * Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers ; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. Tha Proceedings” may also be Abenen by subscription, for either or both series, paid in advance, at special rates as under :— Series A, 15s. per volume. Senes B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately on publication, or as completed volumes, bound in blue cloth. a a ee ee ee ee ee SS 3 4 4 , PROCEEDINGS OF THE ROYAL SOCIETY. Series B. Vol. 86. No. B 587. BIOLOGICAL SCIENCES. CONTENTS. Page Reciprocal Innervation and Symmetrical Muscles. By C.S. SHERRINGTON, F.R.S., Professor of Physiology, University of Liverpool : , w= akg Nervous Rhythm arising from Rivalry of Antagonistic Reflexes : Reflex Stepping as Outcome of Double Reciprocal Innervation. By C. S. SHERRINGTON, F.R.S. ‘ : - : 2 : A 233 Herbage Studies. I].—Variation in Lotus corniculatus and Trifolium repens (Cyanophoric Plants). By H. E. ARMSTRONG, F.R.S., E. FRANKLAND ARMSTRONG, and EDWARD HORTON . : 3 : ‘ ee 2G2 The Trypanosomes found in the Blood of Wild Animals Living in the Sleeping-Sickness Area, Nyasaland. By Surgeon-General Sir DAVID BRUCE, C.B., F.R.S., A.M.S.; Majors DAVID HARVEY and A. E. HAMERTON, D.S.O., R.A.M.C. ; Dr. J. B. DAVEY, Nyasaland Medical Staff ; and Lady BRUCE, R.R.C. : ‘ ; : : . 269 Trypanosome Diseases of Domestic Animals in Nyasaland. — I].— Trypanosoma capre (Kleine). By Surgeon-General Sir DAVID BRUCE, C.B., F.R.S., A.M.S.; Majors DAVID HARVEY and A. E. HAMERTON, D.S.O., R.A.M.C.; Dr. J. B. DAVEY, Nyasaland Medical Staff; and Lady BRUCE, R.R.C. (Plate 5) p 2 5 ; : : K+ 278 Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland. JI.—The Human Strain. By Surgeon-General Sir DAVID BRUCE, C.B., F.R.S., A.M.S.; Majors DAVID HARVEY and A. E. HAMERTON, D.S.O., R.A.M.C. ; and Lady BRUCE, R.R.C. . . 285 PRINTED FOR THE ROYAL SOCIETY AND SOLD BY HARRISON & SONS, 45, ST. MARTIN'S LANE, LONDON, W.C. Price Four Shillings. No. B 587. April 7, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. IMSS. must be type-written or at least written in a legible hand, and properly prepared as copy for press.. Type-written transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. Authors are requested to refer to a Memorandum on Mathematical Notation published in these “ Proceedings,’ 1909, Series A, vol. 82, p. 14, and to adhere to the suggestions therein contained, so far as possible. Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should be drawn on a large scale in Indian ink on a smooth white surface, with temporary lettering in pencil. Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.”—Statute VI, Cap. xii. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the “ Proceedings,’ separately paged, and shall afterwards be republished in the ‘ Year-Book.’ The ‘ Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. On application to Messrs. Harrison and Sons, 45, St. Martin's Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for binding may be purchased, price Is. 6d. The ‘ Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The * Philosophical Transactions ’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in a cloth case for 2s. 6d., or the cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The ‘ Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers ; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. The ‘ Proceedings’ may also be obtained by subscription, for either or both series, paid in advance, at special rates as under :— Series A, 15s. per volume. Series B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately on publication, or as completed volumes, bound in blue cloth. fs PROCEEDINGS OF THE ROYAL SOCIETY. Series B. Vol. 86. No. B 588. BIOLOGICAL SCIENCES. CONTENTS. ‘ JAS. LOCHHEAD The Formation of the Anthocyan Pete of Blne een IV.—The Chromogens. By FREDERICK KEEBLE, Sc.D., Professor of Botany, University College,- Reading; E. FRANKLAND ARMSTROMG, D.Sc., Ph.D. ; and W. N. JONES, M.A., Lecturer in Botany, eee College, Reading The Formation of the eee “Picnests af Be Part Vi-—ihe Chromogens of White Flowers. By W. NEILSON JONES, M.A., Lecturer in Botany, University College, Reading . On the Occurrence of a Ganglion in the Human Taine Bone not hitherto described. By ALBERT A. GRAY. (Plate 6) Studies on Enzyme Action. XIX.—Urease: a Selective Enzyme. II.— Observations on Accelerative and Inhibitive Agents. By H. E. ARMSTRONG, F.R.S., M.S. BENJAMIN and EDWARD HORTON A Preliminary Note on the Fossil Plants of the Mount Potts Beds, New Zealand, Collected by Mr. D. G. Lillie, Biologist to Captain Scott’s Antarctic Expedition in the “ Terra Nova.” By E. A. NEWELL ARBER, M.A., Sc.D., F.GS., F.L.S., Trinity College, Saale University Demoruaton in Balcony: (Plates 7 and 8) . On the Nature of the Toxic Action of Electric Discharge upon Bacillus coli communis. By J. H. PRIESTLEY and R.C. KNIGHT . Experiments on the Kidneys of the Frog. (Preliminary Gaara) By F. A. BAINBRIDGE, S. H. COLLINS, and J. A. MENZIES . The Effect of the Lability (Resilience) of the Arterial Wall on the Blood Pressure and Pulse Curve.—llI. By LEONARD HILL, F.R.S., and MARTIN FLACK On the Probable Value to Bacillus a Ge ‘ Slime” F ormation in Soils. GEGCIESREVISaes Variation in Bacillus coli. ae Production f Two Permanent Vitictis from one Original Strain by means of “ Brilliant Green.” By CECIL REVIS PRINTED FOR THE ROYAL SOCIETY AND SOLD BY HARRISON & SONS, 45, ST. MARTIN’S LANE, LONDON, W.C. Price Three Shillings and Sixpence. No. B 588. Page Contributions to the Biochemistry of Growth.—The Glycogen-content of the Liver of Rats bearing ae New Growths. By W. CRAMER and 302 308 318 323 328 344 348 355 365 371 373 “May 22, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. MSS. must be type-written or at least written in a legible hand, and properly prepared as copy for press. Type-written transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. Authors are requested to refer to a Memorandum on Mathematical Notation published in these ‘ Proceedings, 1909, Series A, vol. 82, p. 14, and to adhere to the suggestions therein contained, so far as possible. Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should be drawn on a large scale in Indian ink on a smooth white surface, with temporary lettering in pencil. ~Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.’ —Statute VI, Cap. xii. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the ‘ Proceedings, separately paged, and shall afterwards be republished in the “ Year-Book.’ The * Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. On application to Messrs. Harrison and Sons, 45, St. Martin's Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for PLES may be purchased, price Is. 6d. The ‘ Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The * Philosophical Transactions’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin's Lane, these will be bound in a cloth case for 2s. 6d., or the cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The ‘ Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers ; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. The ‘ Proceedings’ may also be obtained by subscription, for either or both series, paid in advance, at special rates as under :— Series A, 15s. per volume. Series B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately on publication, or as completed volumes, bound in blue cloth. r) Series B. Vol. 86. PROCEEDINGS OF THE ROYAL SOCIETY. BIOLOGICAL SCIENCES. CONTENTS. Further Researches on the Extrusion of Granules by Trypanosomes and on their Further Development. By W. B. FRY, Major R.A.M.C., and H. S. RANKEN, M.B. (Glasg.), M.R.C.P. (Lond.), Captain R.A.M.C. (With a Note on Methods by H. G. PLIMMER, F.R.S.) (Plates 9-11) Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland.— The Wild-game Strain. By Surgeon-General Sir DAVID BRUCE, (GB; vE-RiS:,, AeM:S: = Majors DAVID HARVEY and A. E. HAMERTON, D.S.O., R.A.M.C. ; and Lady BRUCE, R.R.C. . Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland.—The Wild Glossina morsitans Strain. By Surgeon- General Sir DAVID BRUCE, C.B., F.R.S., A.M.S.; Majors DAVID HARVEY and A. E. HAMERTON, D.S.O., R.A.M.C. ; and Lady BRUCE, IRV IRS Infectivity of Glossina morsitans in Nyasaland. By Surgeon-General Sir DAVID BRUCE, C.B., F.R.S., A.M.S.; Majors DAVID HARVEY and A. E. HAMERTON, D.S.0., R.A.M.C.; and Lady BRUCE, R.R.C. The Excystation of Colpoda cucullus from its Resting Cysts, and the Nature and Properties of the Cyst Membranes. By T. GOODEY, M.Sc. (Birm), late Mackinnon Student of the Royal Society Protostigmata in Ascidians. By A. G. HUNTSMAN, B.A., M.B., Biological Department, University of Toronto ; i On the Origin of the Ascidian Mouth. By A. G. HUNTSMAN, B.A., M.B., Biological Department, University of Toronto PRINTED FOR THE ROYAL SOCIETY AND SOLD By HARRISON & SONS, 45, ST. MARTIN’S LANE, LONDON,. W.C. Price Four Shillings and Sixpence. No. B 589. N & B 589. Page 377 394 408 422 427 440 454 June 12, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. MSS. must be type-written or at least written in a legible hand, and properly prepared as copy for press. “Type-written transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. Authors are requested to refer to a Memorandum on Mathematical Notation published in these ‘ Proceedings, 1909, Series A, vol. 82, p- 14, and to adhere to the suggestions therein contained, so far as possible. Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should be drawn on a large scale in Indian ink on a smooth white surface, with temporary lettermg in. pencil. Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.” —Statute VI, Cap. xu. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the ‘ Proceedings,’ separately paged, and shall afterwards be republished in the ‘ Year-Book.’ The ‘ Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. Onapplication to Messrs. Harrison and Sons, 45, St. Martin's Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for binding may be purchased, price Is. 6d. The ‘ Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The ‘ Philosophical Transactions ’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in a cloth case for 2s. 6d., or the cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The ‘ Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers ; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. The ‘ Proceedings’ may also be obtained by subscription, for either or both series, paid in advance, at special rates as under :— Series A, 15s. per volume. Series B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately cn publication, or as completed volumes, bound in blue cloth. = SS = 3 Series B. Vol. 86. PROCEEDINGS OF EHH ROYAL SOCIETY. BIOLOGICAL SCIENCES. CONTENTS. : Page Some Investigations on the ‘Phenomena of “ Clot” Formations. Part I.— On the Clotting of Milk. By S.B.SCHRYVER . 400 On the Action of Radium Rays upon the Cells of Jensen’s Rat Sarcoma. By S. RUSS, D.Sc., and HELEN CHAMBERS, M.D. (Plates 12 and 13) 482 On Light-Sensations and the Theory of Forced Vibrations. By GEORGE Ii BURCH, M.A., D.Sc. Oxon, F.R.S. . ‘ : i 3 : 490 The Various Inclinations of the Electrical Axis of the Human Heart. Part 1—The Normal Heart. By A. D. WALLER, M_.D., F.R.S. 507 Obituary Notices of Fellows Deceased :—Lord Lister (with 2 oe SS Robert Milne Murray ; Adam Sedgwick : : i1—xxix PRINTED FOR THE ROYAL SOCIETY AND SOLD BY HARRISON & SONS, 45, ST. MARTIN'S LANE, LONDON, W.C Price Four Shillings and Sixpence. No. B 590. No. B 590. July 30, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. IMSS. must be type-written or at least written in a legible hand, and properly prepared as copy for press. Type-wnitten transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. Authors are requested to refer to a Memorandum on Mathematical Notation published in these “ Proceedings,’ 1909, Series A, vol. 82, p. 14, and to adhere to the suggestions therein contained, so far as possible. | Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should be drawn on a large scale in Indian ink on a smooth white surface, with temporary lettering in pencil. Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.”—Statute VI, Cap. xi. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Library will be closed for cleaning from August I8 to August 50 inclusive. All books borrowed are returnable by August I, and no books can be borrowed during August. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the © Proceedings,’ separately paged, and shall afterwards be republished in the “ Year-Book.’ The * Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. On application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for binding may be purchased, price Is. 6d. The * Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The ‘ Philosophical Transactions ’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in a cloth case for 2s. 6d., or the®cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The ‘ Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers ; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. The ‘ Proceedings’ may also be obtained by subscription, for either or both series, paid in advance, at special rates as under :-— Series A, 15s. per volume. Series B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately on publication, or as completed volumes, bound in blue cloth. ees ee eS ee oe a ee ee, ee ee ee ee eee PROCEEDINGS OF THE ROYAL SOCIETY. Series B. Vol. 86. No. B 591. ‘BIOLOGICAL SCIENCES. CONTENTS. Page Acineta tuberosa : A Study on the Action of Surface Tension in Determining the Distribution of Salts in Living Matter. By A. B. MACALLUM, Ph.D., Sc.D., LL.D., F.R.S., Professor of Bio-chemistry in the University of Toronto. (Plates 14 and 15} . ‘ : : : ; ; > y527 Carbohydrate Metabolism in its Relation to the Thyroid Gland.—The Effect of Thyroid Feeding on the Glycogen-content of the Liver and on the Nitrogen Distribution inthe Urine. By W. CRAMER and R. A. KRAUSE 550 Studies on the Processes Operative in Solutions (XXX) and on Enzyme Action (XX).—The Nature of Enzymes and of their Action as Hydrolytic Agents. By E. FRANKLAND ARMSTRONG and H. E. ARMSTRONG, F.R.S. - 5 : F : 5 + 5 3 565 Studies on Enzyme Action. XXI.—Lipase Bre By H. E. ARMSTRONG, F.R.S., and H. W. GOSNEY, B.Sc. 5 4 : : s . 586 Index . : : ; ‘ F 3 ' : “ ; ie SKXG Title, Contents, etc. See INstitya ofp A “3 Office (\pracy: =* In PRINTED FOR THE ROYAL SOCIETY AND SOLD By HARRISON & SONS, 45, ST. MARTIN'S LANE, LONDON, W.C. Price Four Shillings and Sixpence. No. B 591. August 26, 1913. NOTICE TO AUTHORS AND COMMUNICATORS. The Council have had under consideration the rapid increase of the Society's expenditure on publications. In view of the necessity for economy, authors of papers are urgently requested to see that their communications are put in as concise a form as possible. Delay in decisions regarding publication, as well as subsequent trouble to authors, is often caused by diffuseness or prolixity. MSS. must be type-written or at least written in a legible hand, and properly prepared as copy for press. Type-written transcript should in all cases be carefully revised by the author before being presented. It is desirable that authors should retain copies of their MSS. for reference. _ 5 Authors are requested to refer to a Memorandum on Mathematical Notation published in these “ Proceedings, 1909, Series A, vol.'82, p. 14, and to adhere to the suggestions therein contained, so far as possible. — Authors are further requested to send in all drawings, diagrams or other illustrations in a state suitable for direct photographic reproduction. They should be drawn on a large scale in Indian ink on a smooth white surface, with temporary lettering in pencil. Great care should be exercised in selecting only those that are essential. Where the illustrations are numerous, much time would be saved if the authors would indicate in advance those which, if a reduction of their number is found to be required, might be omitted with least inconvenience. “Tt shall be the duty of each Fellow or Foreign Member to satisfy himself that any letter, report or other paper which he may communicate, is suitable to be read before the Society.” —Statute V1, Cap. xu. NOTICES TO FELLOWS OF THE ROYAL SOCIETY. The Library will be closed for cleaning from August I8 to August 350 inclusive. All books borrowed are returnable by August I, and no books can be borrowed during August. The Council have directed that the Minutes of the Meetings of the Society shall be sent out as an inset in the ‘ Proceedings,’ separately paged, and shall afterwards be republished in the ‘ Year-Book.’ The ‘ Proceedings,’ both the Physical and the Biological Series, are sent in the ordinary course by post to every Fellow of the Society who resides within the limits of the Postal Union. On application to Messrs. Harrison and Sons, 45, St. Martin's Lane, these will be bound in volumes, in cloth, for 2s. 6d., or the cases for binding may be purchased, price Is. 6d. The ‘ Philosophical Transactions’ are delivered, in volumes, in cloth covers, only to those Fellows who call for them, or who send a written application to the Assistant Secretary. Such an application may, if so desired, be filed as a standing order. The ‘ Philosophical Transactions’ are also delivered in the form of separate Papers, post free, immediately on publication, to those Fellows who desire to have them in that form. And, on application to Messrs. Harrison and Sons, 45, St. Martin’s Lane, these will be bound in a cloth case for 2s. 6d., or the cloth cases for binding may be purchased, price Is. 6d. SUBSCRIPTION TO THE PROCEEDINGS OF THE ROYAL SOCIETY. The ‘ Proceedings’ are on sale to the public in numbers, on publication, at the prices appearing on the wrappers; or as completed volumes, bound in cloth, Series A, 20s., Series B, 25s. per volume. The ‘ Proceedings’ may also be obtained by subscription, for either or both series, paid in advance, at special rates as under :— ; Series A, 15s. per volume. Series B, 20s. per volume. For this advance subscription, each series may be had in separate numbers, immediately on publication, or as completed volumes, bound in blue cloth. ea ee ee es oe ¥ a a ee ee ee ae Pee ee eee! le a ere PHILOSOPHICAL —TRANSACTIONS 1910— 1912. SERIES A. CONTAINING PAPERS OF A MATHEMATICAL OR PHYSICAL CHARACTER. Vol. 210 (1910). Pp. 418. Price £1 3s. 6d. Vol. 211 (1912). Pp. 486. With 7 Plates. Price £1 6s. 6d. SERIES B. CONTAINING PAPERS OF A BIOLOGICAL CHARACTER. Vol. 201 (1910). Pp. 392. With 36 Plates. Price £1 15s. 6d. Vol. 202 (1912)> Pp. 378. With 20 Plates. Price £1 Ils. 6d. SOLD BY HARRISON & SONS, 45, ST. MARTIN’S LANE, W.C. A. 487. A. 488. A, 489. A. 490. A. 491. A. 492. A. 493. B. 291. B. 292. B. 295. B. 294, B. 295. B. 296. RECENT PAPERS IN THE ‘PHILOSOPHICAL TRANSACTIONS.’ Sold separately by Dulau & Co., Ltd., 37, Soho Square, London, W. PHYSICO-CHEMICAL DETERMINATIONS AT HIGH PRESSURES BY OPTICAL METHODS. (Research executed in the Davy Faraday Research Laboratory of the Royal Institution of Great Britain, London, W.) By WALTER WAHL, Ph.D., Rosenberg Research Fellow of the University of Helsingfors. Pp. 32. Price Is. 6d. A NEW TREATMENT OF OPTICAL ABERRATIONS. By R. A. SAMPSON, eke Sueiiepso 7 exiceras: ON THE VARIATION WITH TEMPERATURE OF THE RATE OF A CHEMICAL CHANGE. By A. VERNON HARCOURT, M.A., Hon. D.Sc. Oxon., F.R.S. With an Appendix. By WILLIAM ESSON, M.A., F.R.S., Savilian Professor of Geometry in the University of Oxford. Pp. 18. Price Is. ON THE EMISSION VELOCITIES OF PHOTO-ELECTRONS, By A. LL. HUGHES, M.Sc., B.A., Research Scholar of Emmanuel College, Cambridge. Pp. 22. Price Is. 6d. ON THE APPARENT CHANGE IN WEIGHT DURING CHEMICAL REACTION. By J. J. MANLEY, Hon. M.A. Oxon., Daubeny Curator, Magdalen College, Oxford. Pp. 34. Price Is. 6d. OPTICAL ROTATORY DISPERSION. PART I.—THE NATURAL AND MAGNETIC ROTATORY DISPERSION IN QUARTZ OF LIGHT IN THE VISIBLE REGION OF THE SPECTRUM. By Dr. THOMAS MARTIN LOWRY. Pp. 37. Price 2s. THE EFFECT PRODUCED BY AN OBSTACLE ON A TRAIN OF ELECTRIC WAVES. By H. M. MACDONALD, F.R.S., University of Aberdeen. Pp. 39. Price 2s, ON THE DISTRIBUTION OF THE NERVES OF THE DENTAL PULP. By J. HOWARD MUMMERY, M.R.C.S., L.D.S. Pp. 13; 1 Plate. Price 2s. THE HUMAN ELECTROCARDIOGRAM: A PRELIMINARY INVESTIGA- TION OF YOUNG MALE ADULTS, TO FORM A BASIS FOR PATHOLOGICAL STUDY. By THOMAS LEWIS and M. D. D. GILDER. Pp. 26; 2 Plates. Price 2s. TITLE, CONTENTS, INDEX, &c., Vol. 202. Pp. 10. Price 6d. A METRICAL ANALYSIS OF CHROMOSOME COMPLEXES, SHOWING CORRELATION OF EVOLUTIONARY DEVELOPMENT AND CHROMATIN THREAD-WIDTH THROUGHOUT THE ANIMAL KINGDOM. By Captain C. F. U. MEEK, M.Sc., F.L.S., F.Z.S. Pp. 74; 5 Plates. Price 5s. PETRIFACTIONS OF THE EARLIEST EUROPEAN ANGIOSPERMS. By MARIE C. STOPES, D.Sc., Ph.D., F.L.S., Fellow of University College, Londen. Pp. 26; 3 Plates. Price 2s. 6d. A PETRIFIED WILLIAMSONIA FROM SCOTLAND. By A. C, SEWARD, F.R.S., Cambridge. Pp.26; 4 Plates. Price 2s. 6d. aa EIB. SI ems dae PUBLISHED BY THE ROYAL SOCIETY. > SoLD By HENRY FROWDE, AMEN CORNER. Impl. 4to. Price 3 guineas. THE SIGNATURES IN THE FIRST JOURNAL-BOOK AND THE CHARTER- BOOK OF THE ROYAL SOCIETY, being a Facsimile of the Signatures of the Founders, Patrons, and Fellows of the Society from the Year 1660 down to the Present Time. A reduction of price to Fellows of the Society. Crown 4to. 483 pages. Price 15s. THE RECORD OF THE ROYAL SOCIETY OF LONDON. Third Edition, entirely revised and rearranged. A reduction of price to Fellows of the Society. Crown 4to. 455 pages. Price 15s. CATALOGUE OF THE PERIODICAL PUBLICATIONS IN THE LIBRARY OF THE ROYAL SOCIETY, 1912. A reduction of price to Fellows of the Society. Sotp BY Harrison & Sons, 45, St. MARTIN’s LANE, W.C. OBITUARY NOTICES OF FELLOWS DECEASED, 1898—1904 (being ‘ Proceedings,” Old Series, vol. 75), pp. 381, with Portraits ; on sale separately in cloth orcs price 10s. net. THE SCIENTIFIC CAREER OF LORD KELVIN. By Prof. Larmor, Sec. R.S. Pp. Ixxvi, with 3 collotype portraits. (From the Obituary Notices of the Royal Society.) Price 4s. THE ATOLL OF FUNAFUTI: BORINGS INTO A CORAL REEF AND THE RESULTS. 1903. 4to, cloth boards. Pp. 428, with numerous Plates and Diagrams. Price 30s. ; to Fellows, 20s. REPORT ON THE PEARL OYSTER FISHERIES OF THE GULF OF MANAAR., By W. A. HerpMAN, F.R.s. _ With Supplementary Reports by other Naturalists upon the Marine Biology. 4to, cloth boards. 5 Parts, 1903—1906. Price per Part, 20s.; to Fellows, 15s. REPORTS OF THE SLEEPING SICKNESS COMMISSION OF THE ROYAL SOCIETY. Demy 8vo. No. I., 1903, Price Ts. 6d. No. II., 1903, Price 7s. 6¢. No. III., 1903, Price 6s. No. IV., 1903, Price 6s. 6d. No. V. , 1905, Price 3s. No. VI., 1905, Price 4s. 6d. No. VII., 1906, Price 4s. 6d. No. VIIL., 1907, Price 2s. 6d. No. IX. 1908, Price ls. 6d. No. ee 1910, Price 3s. * REPORTS OF THE COMMISSION ON MEDITERRANEAN FEVER. 8yo. Part I., 1905 (out of print). Part II., 1905 (out of print). Part III., 1905, Price 2s. 6¢. Part IV., 1906, Price 3s. 6d. Part V., 1907, Price 2s. 6¢2. Part VI., 1907, Price 3s. Part VII., 1907, Price 5s. REPORTS TO THE EVOLUTION COMMITTEE OF THE ROYAL SOCIETY. Demy 8vo. Report1. By W. BaTEson, M.A., F.R.S., and Miss SAUNDERS. Pp. 160. 1902. Price 10s. Report 2. By W. BATESON, M.A., F.R.S., Miss SAUNDERS, R. C. PUNNETT, M.A., and C. C. Hurst, F.u.s. Pp. 154. 1905. Price 7s. 6d. “Report 3. By W.. BaTESON, M.A., F.R.S.. Miss E. R. SAUNDERS, and R. C. PunneTT, m.a._ Pp. 53. 1906. Price 2s. 6d. Report 4. By the same, with Miss DurHAM, L. DoncASTER, etc. Pp. 60. 1908. Price 2s. 6d. Report 5 Gonclneiney: By M. WueEvpatg, D. C. E. Marrvart, and I. B.-J. Sotras. Pp. 7: 1909. Price 3s. 6d. Title page, Contents, and Index to above 5 Reports. Price 6d. The 5 Reports may also be obtained bound complete in one volume, Price 20s. REPORT OF A MAGNETIC SURVEY OF S. AFRICA. By J. C. Beatriz, D.Sc. 4to, cloth boards. 1909. Pp. 235, with 9 charts and numerous diagrams. To be obtained from the Cambridge University Press Warehouse, Fetter Lane, E.C. Price 20s. net. Contents of PROCEEDINGS, No. A 597. October, 1912. Price 3s. Page A STANDARD MEASURING MACHINE. By P. E. Suaw, B.A., D.Sc. (Plate 11) 385 ON THE DETERMINATION OF THE ABSOLUTE UNIT OF ee BY ALTERNATING CURRENT METHODS. By Avsert CAMPBELL, B.A. .. 391 TRICHROMATIC THEORY OF COLOUR VISION. =e pee Ss OF RETINAL FATIGUE. By Sir W. pE W. Apney, K.C.B., F.R.S. . 415 ON THE DISCHARGE BETWEEN CONCENTRIC AI ES any SES AT LOW PRESSURES. By F. W. Aston, B.A., B.Sc., A.1.C. 55 428 ON THE INFLUENCE OF THE NATURE OF THE CATHODE —_ 28 LENGTH OF THE CROOKES DARK SPACE. By F. W. Asron, B.A., B.Sc., A.C. 437 A SPECTRO-PHOTOMETRIC COMPARISON OF THE EMISSIVITY OF SOLID AND LIQUID GOLD AT HIGH TENE SLURSS Wine ree ON A FUE nok By C. M. Srusss, M.A., and E. B. R. PripEAux, M.A., D.Sc. .. 451 PHILOSOPHICAL —TRANSACTIONS 1912 — 19135. SERIES A. CONTAINING PAPERS OF A MATHEMATICAL OR PHYSICAL CHARACTER. Vol. 211 (1912). Pp. 486. With 7 Plates. Price £1 6s. 6d. Vol. 212 (1913). Pp. 436. Price £1 3s. 6d. SERIES B. CONTAINING PAPERS OF A BIOLOGICAL CHARACTER. Vol. 202 (1912). Pp. 378. With 20 Plates. Price £1 Ils. 6d. Vol. 203 (1913). Pp. 381. With 21 Plates. Price £1 5s. SOLD BY HARRISON & SONS, 45, ST. MARTIN’S LANE, W.C. RECENT PAPERS IN THE ‘PHILOSOPHICAL TRANSACTIONS.’ A. 492. A. 493. A. 494. A. 495. A. 496. A. 497. B. 297. B. 298. B. 299. B. 500. _B. 301. B. 502. Sold separately by Dulau & Co., Ltd., 37, Soho Square, London, W. OPTICAL ROTATORY DISPERSION. PART I1—THE NATURAL AND MAGNETIC ROTATORY DISPERSION IN QUARTZ OF LIGHT IN THE VISIBLE REGION OF THE SPECTRUM. By Dr. THOMAS MARTIN HOW RY.) Pp: 37. Price’Zs: THE EFFECT PRODUCED BY AN OBSTACLE ON A TRAIN OF ELECTRIC WAVES. By H. M. MACDONALD, F.R.S., University of Aberdeen. Pp. 39. Price 2s. ON THE FOUNDATIONS OF THE THEORY OF ALGEBRAIC FUNCTIONS OF ONE VARIABLE. By J. C. FIELDS, Ph.D., Associate Professor of Mathematics in the University of Toronto. Pp. 35. Price Is. 6d. ON THE SCATTERING AND ABSORPTION OF LIGHT IN GASEOUS MEDIA, WITH APPLICATIONS TO THE INTENSITY OF SKY RADIATION. By LOUIS VESSOT KING, B.A. (Cantab.), Late Scholar of Christ’s College, Cambridge ; Lecturer in Physics, McGill University, Montreal. Pp. 59. Price 3s. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 212. Pp. 8. Price 6d. ON THE REFRACTION AND DISPERSION OF THE HALOGENS, HALOGEN ACIDS, OZONE, STEAM, OXIDES OF NITROGEN AND AMMONIA. By CLIVE CUTHBERTSON, Fellow of University College, London University, and MAUDE CUTHBERTSON. Pp. 26. Price Is. 6d. THE EFFECTS OF STIMULATION AND EXTIRPATION OF THE LABYRINTH OF THE EAR, AND THEIR RELATION TO THE MOTOR SYSTEM. PART I.—EXPERIMENTAL. By J. GORDON WILSON, Northwestern University Medical School, Chicago, and F. H. PIKE, Columbia University, New York. Pp. 34; 4 Plates. Price 2s. 6d. NOTES ON THE LIFE-HISTORY OF TRYPANOSOMA GAMBIENSE, WITH A BRIEF REFERENCE TO THE CYCLES OF TRYPANOSOMA NANUM AND TRYPANOSOMA PECORUM IN GLOSSINA PALPALIS. By MURIEL ROBERTSON, M.A. Pp. 24; 5 Plates. Price 2s. 6d. PHYSIOLOGICAL OBSERVATIONS MADE ON PIKE’S PEAK, COLORADO, WITH SPECIAL REFERENCE TO ADAPTATION TO LOW BAROMETRIC PRESSURES. By C. GORDON DOUGLAS, B.M., Fellow of St. John’s College, Oxford ; J.S. HALDANE, M.D., LL.D., F.R.S., Fellow of New College and Reader in Physiology, Oxford ; YANDELL HENDERSON, Ph.D., Professor of Physiology, Yale University Medical School ; and EDWARD C. SCHNEIDER, Ph.D. (Yale), Professor of Biology, Colorado College. Pp. 134. Price 6s. 6d. TRANSMISSION OF ENVIRONMENTAL EFFECTS FROM PARENT TO OFFSPRING IN SIMOCEPHALUS VETULUS. By W. E. AGAR, Glasgow University. Pp. 32. Price Is. 6d. THE CHANGES IN THE BREATHING AND THE BLOOD AT VARIOUS HIGH ALTITUDES. By MABEL PUREFOY FITZGERALD. Pp. 21. Price Is. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 203. Pp. 10. Price 6d. PUBLISHED BY THE ROYAL SOCIETY. SoLD By HENRY FROWDE, AMEN CORNER. Impl. 4to. Price 3 guineas. THE SIGNATURES IN THE FIRST JOURNAL-BOOK AND THE CHARTER- BOOK OF THE ROYAL SOCIETY, being a Facsimile of the Signatures of the Founders, Patrons, and Fellows s of the Society from the Year 1660 down to the Present Time. A reduction of price to Fellows of the Society. Crown 4to. 483 pages. Price 15s. THE RECORD OF THE ROYAL SOCIETY OF LONDON. Third Edition, entirely revised and rearranged. A reduction of price to Fellows of the Society. Crown 4to. 455 pages. Price 155. CATALOGUE OF THE PERIODICAL PUBLICATIONS IN THE LIBRARY OF THE ROYAL SOCIETY, 1912. A reduction of price to Fellows of the Society. SOLD BY HarrRIsON & SONS, 45, ST. MarTIN’s LANE, W.C. OBITUARY NOTICES OF FELLOWS DECEASED, 1898—x1904 (being ‘ Proceedings,’ Old Series, vol. 75), pp. 381, with Portraits ; on sale separately in cloth hoards: price 10s. net. THE SCIENTIFIC CAREER OF LORD KELVIN. By Prof. Larmor, Sec. R.S. Pp. Ixxvi, with 3 collotype portraits. (From the Obituary Notices of the Royal Society.) Price 4s. THE ATOLL OF FUNAFUTI: BORINGS INTO A CORAL REEF AND THE RESULTS. 1903. 4to, cloth boards. Pp. 428, with numerous Plates and Diagrams, Price 30s. ; to Fellows, 20s. REPORT ON THE PEARL OYSTER FISHERIES OF THE GULF OF MANAAR. By W. A. HEeRDMAN, F.R.S. _With Supplementary Reports by other Naturalists upon the Marine Biology. 4to, cloth boards. 5 Parts, 1903—1906. Price per Part, 20s.; to Fellows, 15s. REPORTS OF THE SLEEPING SICKNESS COMMISSION OF THE ROYAL SOCIETY. Demy 8vo. No. I., 1903, Price Ts. St No. II., 1903, Price 7s. 6¢. No. III., 1903, Price 6s. No. IV., 1903, Price 6s. 62. No. V., 1905, Price 3s. No. V , 1905, Price 4s. 6d. No. VII., 1906, Price 4s. 6d. No. VIII, 1907, Price 2s. 6d. No. IX. 1908, Price ls. 6d. No xe 1910, Price 3s. REPORTS OF THE COMMISSION ON MEDITERRANEAN FEVER. 8vo. Part I., 1905 (out of print). Part II., 1905 (out of print). Part TII., 1905, Price 2s. 6d. Part IV., 1906, Price 3s. 6d, Part V., 1907, Price 2s. 6¢. Part VI., 1907, Price 3s, Part VII., 1907, Price 5s. REPORTS TO THE EVOLUTION COMMITTEE OF THE ROYAL SOCIETY. Demy 8vo. Report 1. By W. BaTEson, M.A., F.R.S., and Miss SAUNDERS. Pp. 160. 1902. Price 10s. Report 2. By W. BaTEson, M.A., F.R.S., Miss SAUNDERS, R. C. PUNNETT, M.A., and C, C. Hurst, F.ut.s. Pp. 154. 1905. Price 7s. 6d. Report 3. By W. BaTEson, M.A., F.R.S., Miss E. R. SAUNDERS, and R. C. PunNETT, M.A. Pp. 53. 1906. Price 2s. 6d. Report 4. By the same, with Miss DurHam, L. Doncaster, etc. Pp. 6 1908. Price 2s. 6d. Reportis (concluding): By M. WHELDALE, 20), Gh 1D Marryat, and I. . J. Sovvas. iBpau 1909. Price 3s. 6d. Title page, Contents, and Index to above 5 Reports. Price 6d. The 5 Reports may also be obtained bound complete in one volume, Price 20s, REPORT OF A MAGNETIC SURVEY OF S. AFRICA. By J. C. Beattiz, D.Sc. 4to, cloth boards. 1909. Pp. 235, with 9 charts and numerous diagrams. To be obtained from the Cambridge University Press Warehouse, Fetter "Lane, E.C. Price 20s. net. Contents of PROCEEDINGS, No. A 602. March, 1913. Price 3s. 6d. Page * IONIC SIZE IN RELATION TO MOLECULAR PHYSICS, TOGETHER WITH A NEW LAW RELATING TO THE HEALS OF ees OF SOLID, LIQUID, AND IONIC MOLECULES. By W.R. Bousrietp, M.A., K.C. 147 ON THE NEW THEORY OF INTEGRATION. By W. H. Youne, Sc.D., F.R.S. 170 oe ue EOE aon oF USUALLY CONVERGENT FOURIER SERIES. . H. Youne, Sc.D. Bebe e's on. AG CASSEGRAIN. “REFLECTOR WITH CORRECTED FIELD. By . A. Sampson, F.R.S. (Abstract) F 188 ON. ‘A NEW ANALYTICAL EXPRESSION FOR THE ‘REPRESENTATION OF THE COMPONENTS OF THE DIURNAL VARIATION OF TERRESTRIAL er eeree By Grorce W. Waker, M.A., A.R.C.Sc._.. % Sy SL: A SPECTRO- PHOTOMETRIC COMPARISON OF THE EMISS\| TY OF SOLID AND LIQUID COPPER AND OF LIQUID Shek AT HIGH TEMPERAT ES WITH THAT OF A FULL RADIATOR. By C. M. Stusss, M.A., M.Sc. 195 ON THE SPONTANEOUS CRYSTALLISATION AND THE MELTING- AND FREEZING-POINT CURVES OF TWO SUBSTANCES WHICH FORM MIXED CRYSTALS AND WHOSE FREEZING-POINT CURVE EXHIBITS A TRANSITION POINT.—MIXTURES OF p-BROMNITROBENZENE AND p-CHLORNITROBENZENE. By Miss Florence Isaac : 205 THE "EXCITATION OF y-RAYS BY THE «a-RAYS OF IONIUM AND RADIOTHORIUM. By J. Cuapwick, M.Sc., and A. S. RussELL, M.A 217 RE-REDUCTION OF Oe TIDAL OBSERVATIONS, "1883-84, ETC. By Epwarp Roberts, I.S.O., 230 STUDIES OF THE ’ PROCESSES OPERATIVE IN SOLUTIONS. KV. THE INFLUENCE OF NON-ELECTROLYTES ON SOLUBILITY. THE NATURE OF THE PROCESSES OF te ea AND PRECIPITATION. By H. E. -ARMSTBONG! Seine ee se J. Varcas Eyre, P STUDIES OF THE ‘PROCESSES OPERATIVE IN SOLUTIONS. 20,4 1p Rees THE DISTURBANCE OF THE EQUILIBRIUM IN eS Oe me icate se BY S81” AND BY NON-ELECTROLYTES. By HE. E. Waker, B.Sc. 3 . 246 — o re sae ee a PHILOSOPHICAL TRANSACTIONS -1912— 1913. SERIES A. CONTAINING PAPERS OF A MATHEMATICAL OR PHYSICAL CHARACTER. Vol. 211 (1912). Pp. 486. With 7 Plates. Price £1 6s. 6d. Vol. 212 (1913). Pp. 436. Price £1 3s. 6d. SERIES B. CONTAINING PAPERS OF A BIOLOGICAL CHARACTER. “ Vol. 202 (1912). Pp. 378. With 20 Plates. Price £1 Ils. 6d. Vol. 203 (1913). Pp. 381. With 21 Plates. Price £1 5s. SOLD BY HARRISON ®& SONS, 45, ST. MARTIN’S LANE, W.C. RECENT PAPERS IN THE ‘PHILOSOPHICAL TRANSACTIONS.’ A. 494, A. 495. A. 496. A. 497. A. 498. B. 298. B. 299. B. 500. B. 3OI. B. 502. B. 303. B. 504. B. 305. Sold separately by Dulau & Co., Ltd., 37, Soho Square, London, W. ON THE FOUNDATIONS OF THE THEORY OF ALGEBRAIC FUNCTIONS OF ONE VARIABLE. By J. C. FIELDS, Ph.D., Associate Professor of Mathematics in the University of Toronto. Pp. 35. Price Is. 6d. ON THE SCATTERING AND ABSORPTION OF LIGHT IN GASEOUS MEDIA, WITH APPLICATIONS TO THE INTENSITY OF SKY RADIATION. By LOUIS VESSOT KING, B.A. (Cantab.), Late Scholar of Christ’s College, Cambridge ; Lecturer in Physics, McGill University, Montreal. Pp. 59. Price 3s. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 212. Pp. 8. Price 6d. ON THE REFRACTION AND DISPERSION OF THE HALOGENS, HALOGEN ACIDS, OZONE, STEAM, OXIDES OF NITROGEN AND AMMONIA. By CLIVE CUTHBERTSON, Fellow of University College, London University, aad MAUDE CUTHBERTSON. Pp. 26. Price Is. 6d. ON A CASSEGRAIN. REFLECTOR WITH CORRECTED FIELD. By R. A. SAMPSON, F.R.S. Pp. 40. Price 2s. NOTES ON THE LIFE-HISTORY OF TRYPANOSOMA GAMBIENSE, WITH A BRIEF REFERENCE TO THE CYCLES OF TRYPANOSOMA NANUM AND TRYPANOSOMA PECORUM IN GLOSSINA PALPALIS. By MURIEL ROBERTSON, M.A. Pp. 24; 5 Plates. Price 2s. 6d. PHYSIOLOGICAL OBSERVATIONS MADE ON PIKE’S PEAK, COLORADO, WITH SPECIAL REFERENCE TO ADAPTATION TO LOW BAROMETRIC PRESSURES. By C. GORDON DOUGLAS, B.M., Fellow of St. John’s College, Oxford ; J.S. HALDANE, M.D., LL.D., F.R.S., Fellow of New College and Reader in Physiology, Oxford ; YANDELL HENDERSON, Ph.D., Professor of Physiology, Yale University Medical School ; and EDWARD C. SCHNEIDER, Ph.D. (Yale), Professor of Biology, Colorado College. Pp. 134. Price 6s. 6d. TRANSMISSION OF ENVIRONMENTAL EFFECTS FROM PARENT TO OFFSPRING IN SIMOCEPHALUS VETULUS. By W. E. AGAR, Glasgow University. Pp. 32. Price Is. 6d. THE CHANGES IN THE BREATHING AND THE BLOOD AT VARIOUS HIGH ALTITUDES. By MABEL PUREFOY FITZGERALD. Pp. 21. Price Is. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 203. Pp. 10. Price 6d. ON THE COMPARATIVE ANATOMY AND AFFINITIES OF THE ARAUCARINE/AE. By ROBERT BOYD THOMSON, University of Toronto. Pp. 50; 7 Plates. Price 4s. ON ONYCHASTER, A CARBONIFEROUS BRITTLE-STAR. By IGERNA B. J. SOLLAS, Newnham College, Cambridge. Pp. 12; 2 Plates. Price Is. AN AUTOMATIC METHOD FOR THE INVESTIGATION OF VELOCITY OF TRANSMISSION OF EXCITATION IN MIMOSA. By Prof. J. C. BOSE, M.A., D.Sc., C.S.I., Presidency College, Calcutta. Pp. 35. Price 2s. iE RRA PUBLISHED BY THE ROYAL SOCIETY. SoLD BY HENRY FROWDE, AMEN CORNER. Impl. 4to. Price 3 guineas. THE SIGNATURES IN THE FIRST JOURNAL-BOOK AND THE CHARTER- BOOK OF THE ROYAL SOCIETY, being a Facsimile of the Signatures of the Founders, Patrons, and Fellows of the Society from the Year 1660 down to the Present Time. A reduction of price to Fellows of the Society. Crown 4to. 483 pages. Price 155. THE RECORD OF THE ROYAL SOCIETY OF LONDON. Third Edition, entirely reyised and rearranged. A reduction of price to Fellows of the Society. Crown 4to. 455 pages. Price 155. CATALOGUE OF THE PERIODICAL PUBLICATIONS IN THE LIBRARY OF THE ROYAL SOCIETY, 1912. A reduction of price to Fellows of the Society. So_p BY HARRISON & Sons, 45, St. MAarrTIN’s LANE, W.C. 2 REPORTS OF THE NATIONAL ANTARCTIC EXPEDITION, 19014. METEOROLOGY.—PART II. COMPRISING DAILY SYNCHRONOUS CHARTS, 1901—4. Prepared under the Superintendence of Commander M. W. Campsett Hepwortu, C.B. Royal 4to. Cloth. Price 2ls. Now Reapy. OBITUARY NOTICES OF FELLOWS DECEASED, 1898—1904 (being ‘ Proceedings,’ Old Series, vol. 75), pp. 381, with Portraits ; on sale separately in cloth Bouts price 10s. net. THE SCIENTIFIC CAREER OF LORD KELVIN. By Prof. Larmor, Sec. R.S. Pp. Ixxvi, with 3 collotype portraits. (From the Obituary Notices of the Royal Society.) Price 4s. THE ATOLL OF FUNAFUTI: BORINGS INTO A CORAL REEF AND THE RESULTS. 1903. 4to, cloth boards. Pp. 428, with numerous Plates and Diagrams. Price 30s. ; to Fellows, 20s. REPORT ON THE PEARL OYSTER FISHERIES OF THE GULF OF MANAAR,. By W. A. HeERpDMAN, F.R.S. _ With Supplementary Reports by other Naturalists upon the Marine Biology. 4to, cloth boards. 5 Parts, 1903—1906. Price per Part, 20s.; to Fellows, 15s. REPORTS OF THE SLEEPING SICKNESS COMMISSION OF THE ROYAL SOCIETY. Demy 8vo. No. I., 1903, Price Ts. 62. No. II., 1903, Price Ts. 6¢. No. III., 1903, Price 6s. No. IV., 1903, Price 6s. 6¢. No. V., 1905, Price 3s. No. VI., 1905, Price 4s. 6d. No. VII., 1906, Price 4s. 6d, No. VIII. *3 1907, Price 2s. 6d. No. ieee 1908, Price ls. 62. No. xe 1910, Price 3s. No. XI, 1911, Ts. No. XII, 1912, 5s. 6d. REPORTS OF THE COMMISSION ON MEDITERRANEAN FEVER. 8yvo. Part I., 1905 (out of print). Part II., 1905 (out of print). Part III., 1905, Price 2s. 6¢. Part IV., 1906, Price 3s. 6d. Part V., 1907, Price 2s. 6d. Part VI., 1907, Price 3s. Part VII., 1907, Price 5s. REPORTS TO THE EVOLUTION COMMITTEE OF THE ROYAL SOCIETY. Demy 8vo. Report 1. By W. BaTEson, m.A., F.R.S., and Miss SaunpErs. Pp. 160. 1902. Price 10s. Reports By W. BATESON, om. ae , F.R.S., Miss SAUNDERS, R. C. PUNNETT, M.A., and C. C. Hurst, F.u.s. Pp. 154. 1905. Price 7s. 6d. Report 3. By W. Bateson, mM.a., F.R.S., Miss E. R. SaunprErs, and R. C. Punnett, m.A. Pp. 93. 1906. Price 2s. 6d. Report 4. By the same, with Miss DurHam, L. Doncaster, etc. Pp. 60. 1908. Price 2s. 6d. Repor» oncludine): By M. WHELDALE, DiC Marryart, and I. B. J. Souias. Pp. 7: 1909. Price 3s. 6d. Title page, Contents, and Index to above 5 Reports. Price 6d. The 5 Reports may also be obtained bound complete in one volume, Price 20s. Contents of PROCEEDINGS, No. A 606. July, 1913. Price 2s. 6d. Page THE CAPACITY FOR HEAT OF METALS AT DIFFERENT pee es TURES. By E. H. Grirritus, Sc.D., F.R.S., and Ezer GrirFitus, B.Sc. (Abstract) ss 549 ON FOURIER SERIES aD eI EIS OE SOE VARIATION. By Prof. W. H. Youne, Sc.D., F.R.S. . £61 ON A CONDITION THAT A Pee ee ee as SERIES Shae See A CERTAIN FORM. By Prof. W. H. Youne, Sc.D., F.R.S 589 ON A METHOD OF MEASURING THE VISCOSITY OF THE VAPOURS OF VOLATILE LIQUIDS, WITH AN APPLICATION TO BROMINE. By A. O. Rankine, D.Sc. .. 579 Bue HALL EFFECT IN LIQUID ELECTROLYTES. By A. E. Oxtzy, M.Sc., = INDEX . ate Se Src Fisch anie ounlaates oA Pon eae Jee ete se) KT TITLE, CONTENTS, ETC : "48280 i i f b ey: AW f ~ , TIL 8 01306 4555