aps ph D ok mieatces eos eee ais ereahe epare hoe gt 9 oe a seg a oe Pierre here air ek Le Aone erarer nem barron BMC nw ie rt 2A] cane Ce Ee eed knee AME ed Ce ee ee | baste : wey weet) Aen er x a HE > yeaa eC ee tteat Sat ete te gE co Ce Ek Std tote aren eer ee i er er eae ee eet ee ele bet bee } AWA stopp eae ee Pt ee ee ee ee ees E Dacabhpudede tale taecrlsthddanetadarb emitted 11S to Toy gene tet aL ea me ime a toe Belt dekd to tithe ee ew 4 ME airtedy oF Aete or ” WOE EM Fy ee ee eee A ee a pee NO wher ee BRO Gs EA eal me ted tetel LUN her mStar mR Mt TT Uheentet he tottet HeRed dette 4 0h f are Pe ee Re ee ee ee re at the © . Peet a) + oe y we 6 O6e Ou tre tte Oe Cree | ealaters4 erie Ceindareedet 7,84 Oi Sar Sparen CAP ety) G14 0964 Sho dees peerese be > ea Ct Gaede sh whee t ae Vbenares me eee avian ne egy baa y A COA ek ire tahini a tern U8 fy . Uldeew teow aor hs ete ie . See eet ae 444 Cee ee parc Sorry ow i cr tee ar) Ob Ve * eo ok 44 hegeee tg hoe * renrea aces a4 bed Berd. 94 + Rt Se tell oe ap ALE + Pee? " KSEE AS ie Ve OEE AES gH VENOM teateme et eee 7 oP ec ASetew anc teh BASIN be Buln at oy She tere aes eA EY oy Ba helt Comet ee oe eeu Mio er gat PO TM So MEM ae bL due ee Be GO A ne We ee Nig fe 8 Yee S Heb 4s Hew edeayuuae . OO ee ee ey ryrs =H te 1 ee 1 Uy ' «an fava La ‘© toe ie. oe eee TO Ge Bete Sd ted Vheg ON ae a bs. ae ae ete te ee eee ee re oa ien wero wrk ke) ha) a4 ae 19:8 0we 5 11-9 SCOR MT Se faa RA 6 VATE CHG AG Som VME “ (eee daowe Jy bts. £9 8 758-8 F ora Pee F doted ha ot eee sane yea) AAS oe ere ce a) OPP Dit tas Moke AN A cabawe eta pes OC Pere i et Feeewey et y SA Ot eee ee ee Le tee DO teens) Veninee <4 Vda tok ee awl ’ ‘ peer wc ao et Deh nehdni arteeles Tere eta orn eee st alee ere ek vey Mad eg tT ht Beng h rtte ae otth CEA OS Oe Toit ae RAR S ATA * 5. aN a a he oy Ga et ie ee a ee ( ey OO Re Vee eee ase eae ean t Cte eee CP ren eo et eho Cd Ce wh aeiee Geld ton Fy Les era hw eee Wee Ge ee Ce a eet LNA Rt Ae We tom a ewe a Gh Cate ot ed Ged ht bee CCR ee ttcr eS Pein te ye oer ertet eohee te eed ae hs tosh Oe et Oe hee DEF ES Cre ge t elites wrona yWhh Weitere a Gate atone aoe { Sar cLentn ale bcd seers «a eat ea ee Pea Lae Ma Th i Teenage Ldn herpes Pyle yest dh iy Aer aye 4 REM By 4 tis ; yp ue Tale etn some’ Re Kane th) CICA AAR CO ek aU iees te ce tears ir ro te AN or Cee et ceveea cts OE Ee ce Re UTE Leb aad Met tte) eet ae die: Cea dog a CA OU ee OA Og CR delet GP aed Seqtetted COG Bet ek hee eee . Coated tee seo eee as oe, a enews FAA ee Bett a rhe Ure ark AAS iRad dH Ade tk Cait AEN ALO Fed Fe wey ® CREE CEN EE Dae & We VR ee Me Bee GT ig Se ae a tae ea ss CAS Cee a ay Fle ery hl BEA gem eire ue Swat Arde 0 Rohit ham te deat . Te Or ee ee a) 120.9. 979 9 Gogh 8 SWS 04 OO wh ee ST ENE ERC COL BE THEE AR OE NETRA TNS Ee ee oe eee EE ea AN Af 8 rate Sr ee * oath! . yee « wee “% propa A Mite Gothen Moh mG ated nds oN Dot! AE ae AR RR eH teu pehoee Waser Pte Mee tena tad ec erate ees Pec ca t Cee ee ee ee) RS Meda Wd Al HoT eee vit ” ‘ “ (OY HH One gs A Aen nry yea Re PAGEL Wt Woke MAA hed © wet H COSINE Sed hie tb ae aT Frei CAR HC Ee ERA eR At td ee OE ERT WE rhe here a he fare ae 44s: eee Rates te eed Pee ts U4 gwen eae et ROSA tata lh ie Paik Rema IODA oti Sal a it a Sete ta * as Par rah ee Wigan tern ae 9) ie ‘i Seda a are be ak how fey Sinan g Ek de WW deg Wf re ietee arta tele Prac ee Cre ee eisaetreny Ce * 14 eb tidet a ot te " GCI AER ee Ue Oma) td Vee EF te ad nee Aree eae Re er aL hte Gara sey) toge! ee tay CANON AD he 1 get nitetlp CEWEK LE Mott ee RCC, “oss COR re ier wt Write Mehra! Wath “ 7 ‘ Gove re c +i etaeeddawe oO tee Ve Vee pe wk 4 A Re eee wee ue revit e Sopa aie 42 9 9it W e dol H piprchch Aen ig! phot Seles Lens iene ey ake Rane cutee ete a aly «Sar eek re) an ‘Pee he ens rte DO RORIOC Koes LU hed oe te peared: Aas ane Cie Carn'y V4 4 ha ay TE wed etl detente dT 1A a* ‘ Ate 4 eee Site eat) CP LADO EES a tmnt iy «404 AMPH EER | Vietied Saha oe eee Ba Hee CARO al 0 het, fot 4d tat He ee £88 8 ON EO OE Bw e-9 Bd Oey We a Gee tee ae SEA eHOeT teak SGU Sete SY) » ha C48 SC a-Fabtit a Meh Cotes ew rei Si Arahat 4 ek oy hen Cee e ea Eve tk Cee heed eee EE ED CATs Came ee 4.35 09 gr #4 Dw oes PUR bee e ee se ae, Ma4e8 48 Kin wie ee oe ee a eee Y , any peace’: " LURE EES COGN ee 1. TA TAA KK Ae ite RRR FL AMG Y Sderot RUKA ACR Thi | 6 ee eee ee er er ee ee ek er ca, SINS. yee MeN Hele EM EM oe Eee 8 ee EH A hited Obert de Gitried te 4c4e fk eal ho Ria ae hotel hee’ Py apheah s ei we ues ae eae a dee eee WOW Geet OM ee dete Rh OEE EEE Ee les eur tie TO- $ SRE Ne ‘ wih TE EON et OE 8 Vb ew et eed eet ee Daeg ee oe 4 Weedon oe tee ett ei tee hea ¥, ceateee ots tad n ave * oe Ves en yew get tiasd Pee Vive) ewes Ce ae el eee +t he ok a EET He ETT * oy weak ty He 4 Gia COs | 4 HHH VL a ee ere CULES PLES ASH SF > PCR ee Se ei ret cused o4 14 A ee rs oe a peas oh. 4 ee Pate th +4 * " Wierd eet em nn CAG 4 Tai neh x$ey ‘hae 4Uae Mae hoy eee eg oes ae eet yee eelbes Gato’ a ChE KO etic iho CPL ge oe EAE bo 68 ted thd BON ee Cee Ee ag eg ele sete eet RO Otel RUE E VE MMe Pitot iid wiloe wah Wirt oe VE Sey CHAAR Nel ine Dida an eee Behe See UNECE WR LOS CRM Grad added al vA Pu Tantei NCie mk tert noe OCA CUA ene 4b ered ade CR een eo od ere ee Oe er i” ’ wae 4 a oe ea PS 4 TeVoay oh ey 8 A Oe nel Oe Aen wee O4 ee eee ar Wy he eri tron hk ee ory eb te 49.04 ORS Fede tres ne VAI Gea HE vee Sata * “tele eee ees} “ oe So KE 1a N ee Hare DA steht CARO ree 44g Hien dade feta dhe Lo Bae 4H ed ah eee e faa Qe ae ee he aM ee ae eee eee ek Viewed Chine ee SSAA ee see ts Sn Re er ¥ te Re ee wee ged ee ee ee ey she Liteus Pe Phiten ee ek ee oy ot “ TERR ER Re ee eee eee ee Cee Phe) Pee err e rut err VET ee ee eee eee eh ere vert Sh heen ere eke rine PNT TS NR Gt He Le a OF RET MENT BE TTY FR ae tegrtey sate Saeaisiy ew ON 4-4 noe we eewae eRe Ke ee ik Mie ra ok ke) He yekwasevat Ce seas Ot a RL mee is Mite rareed.s renin ke eee SERRE eee meee anne eeeu are ae MAS A ATES a AOR AY pon WMA At N64 5 Veh aw ares Wena. a4 ya 4 418 bs vee See OY SR Cee eo . Pere mur wiwieie emt ky pra ks ho os PR Mk A eae ee he ELNAM DAA WASH O ach Bisel ‘ aed Seen eaee ‘ pede anaes Rae Sy aagess CAPO os wer eee a) " War GA Sree Rear eS Peat st kate eA Rn +4 w os eh AaOree e Oaee rae eR Le ed Coat) SEW a dealt UUM GMa eats Corcoran oteees be Vea Pere POO ek eG Rae Cte 841 Oe 8 Gl ee © Oeste SS Fe Se Ore gh fort eh ere hei ad, eh EE NR ES ote Rat eink to Wt eR WD AW WE Herta aoe, Meds Gow Gee ea ee ae ee SS ee) Nee wee Ge RN RSG We UT BETTI Te Ctencmenl dc Oe th ee is la Font Pte eee $8 ner ese Coie ee eet 44 A ey Vee FST EEE re eee er Toe ine err tr) AO ah Mowe hae eee ene CR AS EO TEV EH 448 AEE Srey wea Gee Wt nee Aoe 6 wha Ch he er kim tt at ke ea CR ky Se ta inh ere a ret mitt her Lt wet CAAA A eh med HE Ge mee wee ‘eed are eh aed Ore tee . se Cow a o tos SU Septal feb i ct ont SI et ret arene wr te Ane f seed aitinen A beg braee oe igh eer trie GRC OEE ates 0 oOHe eee Ow Ay ets oe re Ce UR Oe hast Ok ne tle) ae OS ed ee Coe Wee, ws «th Sotbeds Ata BARN AS COEF OE UES 4 ee Ae ned Qed ld OMe ard gee a ve NAME TRE OTE TE OL EE ETH ATONE IE ® CUR ee rere PEAT Oe Me ka date et A « ee, aaa Cee me ee ae) WON oe Oe ae sae toe Cas SPOT TE SDs Orr ich ioe} eS dow eee) “8 Ske ooe eee EM CG Sg ae 5 at as ded peers teter eino 1 atgemeete Fo wee Not ra Seed ae ethane ee ' pats ee ee ae ” eo : sis See he aa eww ee ede ees CeCe ee ce ce ee ery arr Se eres Clee ad lee eee ae Deo neh Oc ic ik rick Pheer oO See een Rep - <> = aa Ea wick Ree yey ee a ea te OW NON ted eat CHT Attia Ae te: eee err hit re ei ra ek is CN a ew eer ares a oer nant See RR Css Sot ey os “acteye FU Se ea aa on Pu Oe Te PORE ERE RTE LE ee Oo ence yk ener tk ee ren POT He OR EE ERNE ET GE LN Te donee Odom Hew dered gu am nrard Slater CEP PeR REECE RU Ley Severn we wit ook ee reer: EASE totem erdete Cede Meta eae Se niet etn ee ~ St ere re Ser) cee rca i SONS VE a Ree ie rr ee ce & feta ok eee NS Eee Aerts 8! ere aot A Ree ee CWA dewey ts cerns ti Pah ner at ay ee hy nde torhara hy pte Penman eres et y one 6 Dates aiaeuarbriesroanth Cede wn ns ewe eee we ee eds we mete AH. Pte ey aCe ae ie ea 4 44 4 rte ete He Gowen < CPMGME SEK Sse aks oe een ear urre, ereriers ed ps soemc Vee are aay ee Ne eV ete ve ‘4-54 8 See AOE ABARAT Rew Sth eeeseagrees Fee Pee OM CARTER ey ere ws 4 64> Gcbeld 6-6 46 6 O06 4 4 oe NE ES ROR E em mr: ee: OURS AWTS hater Pg Gets ar eee: CONES ered eens hae ee eS FO wee eee Hr pew aie Bh ar enV @'e GeO Gre G-trecd, Oi d- OOo 6) 4 SG G cOereee 640. Ged. oe at PROCEEDINGS OF THE ROYAL SOCIETY OF LONDON SERIES B CONTAINING PAPERS OF A BIOLOGICAL CHARACTER VOL. LXXXVIII. TO WNEDIOANE: Printep ror THE ROYAL SOCIETY anp Soup By HARRISON AND SONS, ST. MARTIN’S LANE, - PRINTERS IN ORDINARY TO HIS MAJESTY, JUNE, 1915, LONDON : HARRISON AND SONS, PRINTERS IN ORDINARY TO HIS MAJESTY, ST. MARTIN’S LANE. CONTENTS. SERIES B. VOL, LXXXVIII No. B 600.—August 6, 1914. The Influence of Osmotic Pressure upon the Regeneration of Gunda ulve. By Dorothy Jordan Lloyd, B.Sec., Bathurst Student, Newnham College, Cambridge. Communicated by Prof. J. Stanley Gardiner, FURS. ...........csscvceveecscreseeesence Glossina brevipalpis as a Carrier of Trypanosome Disease in Nyasaland. By Surgeon-General Sir David Bruce, CB. F.R.S., A.M.S.; Major A. E. Hamerton, D.S.0., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Plate 1) CeO eww eee ee etre Hee HOPES EEO HEE HOE H HH HEHEHE H FOF S OE eae ED H SOE HHEH HEE eDeeEHEEHaEHeeseoee® Trypanosome Diseases of Domestic Animals in Nyasaland. III.—TZrypanosoma pecorum. Development in Glossina morsitans. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Plate 2) Trypanosomes found in Wild Glossina morsitans and Wild Game in the “ Fly-Belt” of the Upper Shiré Valley. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C. ; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasa- land, 1912-14) etme eee wet eet e eee e ween eta s ease estes e ns es eases ese as bOSessssstassessus ease ogres The Food of Glossina morsitans. By Surgeon-General Sir David Bruce, C.B., E.R.S., A.MLS. ; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) Bete rete tetera teen we eae tees se ees nesrseressast ess netarese esses Infectivity of Glossina morsitans in Nyasaland during 1912 and 1913. By Surgeon- General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) etme teers sees ea assests sasee The Various Inclinations of the Electrical Axis of the Human Heart. Part Ia.— The Normal Heart: Effects of Respiration. By Augustus D. Waller, M.D., F.R.S. (Plates 3-6) Pee eet ee ee eee tee eee ae tet Eee HOH Erte tee tase ae eer test eHssesssbasseenesses On the Relation between the Thymus and the Generative Organs and the Influence of these Organs upon Growth. By E. T. Halnan and F. H. A. Marshall. (With a Note by G. Udny Yule.) Communicated by Prof. J. N. Langley, F.R.S....... PAGE 20 38 4] 49 68 1V The Cultivation of Human Tumour Tissue 7 Vitro.—Preliminary Note. By David Thomson, M.B., Ch.B. (Edin.), D.P.H. (Cantab.), Grocers’ Research Scholar, and John Gordon Thomson, M.A., M.B., Ch.B. (Edin.), Beit Memorial Research Fellow. Communicated by Sir Ronald Ross, K.C.B., F.R.S. (Plate 7) ......... No. B 601.—August 27, 1914. Trypanosome Diseases of Domestic Animals in Nyasaland. Trypanosoma capre (Kleine). Part III.—Development in Glossina morsitans. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.0., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Com- mission of the Royal Society, Nyasaland, 1912-14.) (Plate 8) .............:...:.200 The Trypanosome causing Disease in Man in Nyasaland: The Liwonde Strain. Part I.—Morphology. Part I1.—Susceptibility of Animals. By Surgeon- General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. EH. Hamerton, D.S.0., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) ...............0- The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part I—Morphology. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S. ; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland) el 912143) @Platesto =i) meee seesee reese eerie errr aera The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part II1.—Susceptibility of Animals. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Com- mission of the Royal Society, Nyasaland, 1912-14)...............cccsecneeneeeeeeeseneeee The Vapour- Pressure Hypothesis of Contraction of Striated Muscle. By H. H. Roaf. Communicated! by Prot. C:1ShSherrimotomy HR oesesen assesses: aesieatseeesee ee nrenee On the Nutritive Conditions Determining the Growth of certain Fresh-water and Soil Protista. By H. G. Thornton (New College) and Geoffrey Smith, Fellow of New College, Oxford. Communicated by Prof. G. C. Bourne, F.R.S. (Plate 12) Pew w emer eee eee ee ewes ase nere eee Denet cates ssssesesessesesesstersseseeseseseeessseeatsnsees The Validity of the Microchemical Test for the Oxygen Place in Tissues. By Alan N. Drury, B.A., Shuttleworth Student of Gonville and Caius College. Com- municated; with) a) Note, by Wr Bi Hardy; sBaRAS) tyessesneseemeressstecseeecsissseareemee Studies on Enzyme Action. XXII.—Lipase (1V).—The Correlation of Synthetic and Hydrolytic Activity. By Henry E. Armstrong, F.R.S., and H. W. Gosney, BS Ciahiesears vactaetdviide ouSaemtechlessedasiiles daattoctiateladee asic ees aGeemaee cae cms iba seeictsgasemuttesease No. B 602.—September 15, 1914. Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland: The Human Strain (continued).—VI to X. By Surgeon-General Sir David Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, RRC. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) .........sccccssseseaeeeeeeeee PAGE 90 Oy 111 130 139 166 176 V The Trypanosome causing Disease in Man in Nyasaland. II. The Wild-game Strain. III. The Wild Glossina morsitens Strain. Part Il.—Susceptibility of _ Animals. By Surgeon-General Sir David Bruce, C.B,, F.R.S., A.M.S. ; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part I1].—Development in Glossina morsitans. By Surgeon- General Sir David Bruce, C.B., F.R.S., A.M.S. ; Major A. E. Hamerton, D.S.0O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912—-14)..............sceeeceseeeceeee ees The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part IV.—Experiments on Immunity. By Surgeon-General Sir David Bruce, C.B., F.R.S., A-M.S.; Major A. E. Hamerton, D.S.O., and D. P. Watson, R.A.M.C. ; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14) The Colouring Matters in the Compound Ascidian Diazona violacea, Savigny. By Alfred Holt, M.A., D.Sc. Communicated by Prof. W. A. Herdman, F.R.S. . Some Accessory Factors in Plant Growth and Nutrition. By W. B. Bottomley, M.A., Professor of Botany, University of London, King’s College. Communi- CaLeda yp Enon Hl. Whe Olivier, BeRAS: he. Seen cvctare ceete.casdesebecte selene: svesvecceeoeeeees Further Observations on the Changes in the Breathing and the Blood at Various High Altitudes. By Mabel Purefoy FitzGerald. Communicated by J. S. iB ig beste. INSISTS Se Sone etecarione code secote bore ee Hae OnE rere Co a Eee eee eer eEe Seer te a arene Constancy of the Optimum Temperature of an Enzyme under Varying Concentra- tions of Substrate and of Enzyme. By Arthur Compton, Imperial Cancer Research Fund. Communicated by Sir J. R. Bradford, K.C.M.G., Sec. B.S. ... A Theory of the Action of Rays on Growing Cells. By J. Joly, Sc.D., F.RS. ...... The Influence of Timbre and Loudness on the Localisation of Sounds. By Charles S. Myers. Communicated by Prof. C. 8S. Sherrington, F.R.S. ............cccseeeeeeeeees No. B 603.—December 1, 1914. A Comparative Study of Oxidation by Catalysts of Organic and Inorganic Origin. By Alfred J. Ewart, D.Sc., Ph.D., Professor of Botany and Plant Physiology in the Melbourne University and Government Botanist of Victoria. Commu- nveaheds byeeKOtadly Key NGlOSIGLEeNs MN Ssetteaea.c tess ceseosecererese-vectoccceancorss The Fixation of Arsenic by the Brain after Intravenous Injections of Salvarsan. By James McIntosh, Beit Memorial Research Fellow, and Paul Fildes, Assistant Bacteriologist to the London Hospital. Communicated by Prof. Weebselloch Bh sevice. ger: eitae kt ern, -obtuaect cfs omens 2 acco C CORE CaCiduEC Da SHEUDODEEREDOCE The Production of Anthocyanins and Anthocyanidins.—Part IJ. By Arthur Ernest Everest, M.Sc., Ph.D. (Lecturer in Chemistry, University College, Reading). Communicated by Prof. Frederick Keeble, F.R.S. .........c.c:sesccesesceeeesceeeeaeees CroontaN Lecture: The Bearing of Cytological Research on Heredity. By Edmund B. Wilson, Da Costa Professor of Zoology, Columbia University, New York ee eee eee eee eee ee eee weet e eee H ee tees eee e eee ease a ESHEETS EH EEE EEESTE ESE EtESEHESESEEae PAGE bo =) or 213 248 284 320 al No. B 604.—February 1, 1915. PAGE Observations on the Life-Cycle of a New Flagellate, Helkesimastix fecicola, n.g., n.sp.: Together with Remarks on the Question of Syngamy in the Trypano- somes. By H. M. Woodcock, D.Sc., Assistant to the Professor of Protozoology, University of London, and G. Lapage, M.Sc., Assistant Lecturer and Demon- strator in Zoology, Victoria University, Manchester. Communicated by Prot. (S.J. Hickson) Hanis. | (Plates) 13 and 14) peers-p see a-e eaten shes ase eee 353 The Antagonistic Action of Carbon Dioxide and Adrenalin on the Heart. By S. W. Patterson, M.D., Beit Memorial Research Fellow. Communicated by Prof. Starling, BuR.S...cio.datesvesieecdedterseestboesccomsedectesctatsssoneee eee aera 371 No. B 605.—March 1, 1915. The Influence of Salt-Concentration on Hemolysis. By W. W. C. Topley, M.B., M.R.C.P., Bacteriologist to Charing Cross Hospital. Communicated by Dr. BW... Mott. RiSsvictusateeccntiecenteh ee se deesen- es baece ee meee ete eeeeenee ees: 396 The Influence of the Hydrogen Concentration upon the Optimum Temperature of a Ferment. By Arthur Compton, M.B., D.Sc. (N.U.I.), Imperial Cancer Research Fund. Communicated by Prof. W. Bulloch, WLR.S: -cossekesr -<- ses 25-95--eeeeeneeeee 408 The Life-Cycle of Cladocera, with Remarks on the Physiology of Growth and Reproduction in Crustacea. By Geoffrey Smith, M.A., Fellow of New College, Oxford: Communicated! by Hi. \S) Goodrich) Be Rass leeree-sesceess sees seeee ee ee eee 418 Lepidostrobus kentuckiensis, nomen nov., formerly Lepidostrobus Fischeri, Scott and Jietirey= asCorrection» ‘By, D) He Scott, Bors Sec: Rss eeeteeeeeeeee ates ee eee eee eee 435 No. B 606.—April 1, 1915. Investigations on Protozoa in Relation to the Factor Limiting Bacterial Activity in Soil. By T. Goodey, M.Sc., Protozoologist, Research Laboratory in Agricultural Zoology, University of Birmingham. Communicated by Prof. F. W. Gamble, 0s Ril ae sce cE RID EC RPERBEOROSREOREOEH sae seeiane sac an00090 donqcocndeocapbaodanbnoaoocoo70c2009 132 437 On the Mesodermic Origin and the Fate of the So-called Mesectoderm in Petromyzon. By 8. Hatta (Sapporo, Japan). Communicated by Prof. E. W. MacBrides"FURAS. Gesiescibcbca veeds due seecaesnesoe ce SS oc Cee eee Coe eee eee 457 On the Variation in the Growth of Mammalian Tissue in Vitro according to the Age of the Animal. By Albert J. Walton, M.S., F.R.C.S., M.B., L.R.C.P., B.Sc. Communicated by Prof. W. Bulloch, F.R.S. (Plate 15) ...........2..000200cseescenes 476 No. B 607.—May 3, 1915. The Influence of Homodromous and Heterodromous Electric Currents on Trans- mission of Excitation in Plant and Animal. By Prof. J. C. Bose, M.A., D.Sc., C.S.1, C.LE., Presidency College, Calcutta. Communicated by Prof. 8. H. VAMOS SHIR IS), caeacescadercsueanacesarcnens sass cores ace uokie se ee cee Sea ee ee eee Ree ree ctia deer 483 The Measurement of Arterial Pressure in Man. I.—The Auditory Method. By Martin Flack, Leonard Hill, F.R.S., and James McQueen...........-..0:--1eeeeeees 508 The Measurement of Arterial Pressure in Man. II.—A Schematic Investigation. By Martin Flack, Leonard Hill, F.R.S., and James McQueen ................+-s0200 516 vu PAGE On the Mechanism of the Cardiac Valves: a Preliminary Communication. By A. F. Stanley Kent, M.A. Oxon., Henry Overton Wills Professor of Physiology in the University of Bristol. Communicated by Prof. C.S. Sherrington, F.R.S. (Plleis) 71@)),dosapnotgacundbamubeseoucodescacoeb ocd on cad CoCon SA bancEr Ane Aer OAneaenepAcRaAeee aE ancerere 537 The Effect of Functional Activity upon the Metabolism, Blood-flow, and Exudation in Organs. By Ji. Barcroft, ¥.R.S., and Toyojiro Kato...........-.....cs0csesseeeeree 541 Functional Gidema in Frog’s Muscle. By M. Back, K. M. Cogan, and A. E. Towers. Clommamonmieenerl loyy Ue IbeHAeONt, IDGIRISH, ‘sdoceaccobadsosdcnaedorooddooanebpobeudsoossoRebudod: 544 No. B 608.—June 1, 1915. The Effect of the Depth of Pulmonary Ventilation on the Oxygen in the Venous Blood of Man. By J. F. Twort and Leonard Hill, F.R.S. .............ccecsenee eens 548 On the Occurrence of an Intracranial Ganglion upon the Oculomotor Nerve in Seyllium canicula, with a Suggestion as to its Bearing upon the Question of the Segmental Value of Certain of the Cranial Nerves. By Geo. E. Nicholls, D.Sc., Beit Memorial Fellow (Zoological Department, King’s College, London). Com- imntummneR noel Joy, LRG: ANG ID ern lye IBSIRSS) GoosegoccoondesoceonebeosaeganospapaenoneanosdcabnoKec 553 The Osmotic Balance of Skeletal Muscle. By Dorothy Jordan Lloyd. Communi- GRILCOM ya iawis sw ELan Cyn H kus Stace cules a sees kei veldiiccrieieeal sede wac ee sitnt ve wertineiiaverice sousied 568 Surface Tension and Ferment Action. By E. Beard and W. Cramer. Communi- Gaul lov Sie IBichwabiGl Soe, IDEIRASI, ic coononoaaocoqodadnons dens doonod ogbepEduacndosneeGobDNe 575 Surface Tension as a Factor Controlling Cell Metabolism. By W. Cramer. Com- renmbnanle Mie oy Sine WiehwebRel Srelabnierg, IRIS So5gssdcesasdocoooedeonecondeescoarongseqadcdee 584 OBITUARY NOTICES OF FELLOWS DECEASED. ANTPULONIE TUTE son ap oouddadnescobodsec cua cobedenenpss bb osddRobesba dco seeadebocdoronnsponescreEorcneeaGe 1 A, Ob Lop Ge, Ci Mee (CHA TLOTABTENT)) o>. cocnangesoeneosadeoocHs0 cobocneoogabBsaxegeobooKdsedb oe xi \Wyelteaie Islolll orrovalie (Cris) alll Ss sassonanceacadsasuoasasdncoeny dediné cosas Hod sopabdosaanedasoBoRdaconeraa XXV sors amen amol ol sk Gree my tech era cen Meta oc hamette sa alan caerasiheaciteeed natin aca tiamcOcedmcwen XXxvi HNirvel Ocammmerst ene eerie ateiser cere iat catetaa ss ania ves inocolte cles enevie sist welds acne dlseilaciaw's setsoieineniadiens XXX1X PROCEEDINGS OF THE ROYAL SOCIETY,— Szecrion B.—BrotogicaL SCIENCES. i. The Influence of Osmotic Pressure upon the Regeneration of Gunda ulve. By Dorotuy Jorpan Luoyp, B.Sc., Bathurst Student, Newnham College, Cambridge. (Communicated by Prof. J. Stanley Gardiner, F.R.S, Received March 21,— Read April 30, 1914.) CONTENTS. PAGE He UTI tROMUCUION! manessacdgaedissscceaane cscs tesstoeecGsesesvecesceeecrcntscutscameutencoreeite 1 IJ.—Preliminary Experiments on Reduction and Growthin Whole Animals... 3 Gasparvation otek uly feds Worms chesscerectesssermetiscearerscsseccesseesscane 3 Duaticedinevols starveGuyWOLMisy ccnsesnct eae sscascsaseectedeaseeee es cseeek ene eens : 5 IIT.—Preliminary Experiments on Duration of Life of Whole Animals in Solutions of Different Osmotic Pressure ............:scsseseesceccescerescecees 5 IV.—Study of the Normal Course of the Regeneration of the Posterior End... 6 V.—Regeneration in Solutions of different Osmotic Pressure ................0600 10 RV ATED GCHSSIOMG osha svi pea ses sence store se neeaucot soul ona wees tse cade st stesestouactseseuncs 17 BVAIHE—SUIMMIMNAGY, “ese sanensassbanac ocsecencsaese dese saatoue casts aasscecevuatsicsssueancuseenseete 18 I. INTRODUCTION, This work was carried out during the summer of 1912 and the spring of 1913, at the laboratory of the Marine Biological Association, at Plymouth. The object of the investigation was to discover how far marine turbellarians could be made to undergo variation while regenerating lost parts, and, if possible, to correlate the variations with definite physical conditions. Gunda ulve was the only species used during the course of the investigation. It is a small Triclad and its structure has been described in detail by Wendt (16) and Iijima (5). VOL. LXXXVIII.—B. B 2 Miss D. J. Lloyd. Influence of Osmotic The length of the fully grown specimens obtained at Plymouth hes between 5 and 6 mm., though the worms are found sexually mature from a length of 4 mm. It occurs between tide-marks and near the borders of a small stream. Density determinations made on different occasions of the water in which it was found, show that it is normally exposed to a very wide variation in osmotic pressure. The lowest density value recorded was 1-001. v.e. a value corresponding to almost fresh water; the highest was 1:028. The mean density of the sea-water at Plymouth is 1023. This fact is an important one to take into consideration when placing the animal under experimental conditions in the laboratory. Before proceeding to the main part of the work, it seemed advisable to make three sets of preliminary experiments: the first of these was to determine the limits of osmotic pressure within which whole worms are capable of living; the second was to observe the morphological changes accompanying starvation in whole animals; and the third was to make a histological study of the regeneration in sea-water. These are all described in detail below. For the experiments in regeneration, only regeneration of the posterior end was considered. G.ulve differs from most Triclads in that it will only regenerate a new head in the presence of the cerebral ganglia(6). As it was wished to study the effects of removing large pieces, the experiments were necessarily limited to regeneration of the posterior end. To secure uniformity, the worms were all transected half way down the length of the body. G. ulve is capable of living for many months in a state of starvation, during “which it slowly diminishes in size. Nevertheless, worms kept without food until reduced in length from 5°5 to 1:5 mm., when bisected, restored their normal form as rapidly as did fully fed specimens. This point is emphasised here because in the experiments considered in this paper the animals were left entirely without food. This was done, firstly, to avoid contaminating the experimental waters, and, secondly, because worms regenerating in different solutions develop the new mouth at very different intervals after section, and thus the starvation experiments progress equally rapidly and more uniformly. A number of observations were made on the regeneration of G. wlvw under both starving and fully fed conditions, and the differences which are found consist only in the rather less degree of reduction that occurs when feeding is taking place while the worms regenerate. The figures of the whole worms are camera sketches made from living material. The sections were made from material fixed in corrosive acetic, and stained in picronigrosin and eosin. The sections used for studying the Pressure upon the Regeneration of Gunda ulvee. 3 cytology were stained with Heidenhain’s iron hematoxylin followed by orange G. TL Pretiminary EXPERIMENTS ON REDUCTION AND GROWTH IN WHOLE ANIMALS. (a) Reduction. A number of adult worms 5-6 mm. in length were washed in sterile water and placed in a wide-mouthed sterile bottle which was closed with a cork. Through this cork were passed two glass tubes forming the inlet and outlet of an air-circulation, both tubes having a cotton-wool plug in them to keep back chance organisms in the circulating air. The animals in this apparatus were kept in a state of starvation. At intervals animals were removed, measured, and fixed for sectioning. The worms were found to decrease fairly steadily in length under starvation. Curve A in text-fig. 1 shows the course of diminution in length in an experi- ment lasting 12 weeks. Microscopical examination of sections showed that Curve A, curve of diminution of length of starving animals. Curve B, curve of rate of growth of starved worms on feeding. Note that feeding was interrupted between 22nd and 37th day. starvation is always accompanied by absorption of the generative system. The yolk-glands appear to be absorbed first, but after the worms have been reduced to about 4-5 mm. in length degenerative changes appear in the ovary, testes and secondary sexual organs. By the time the worms are reduced B 2 4 Miss D. J. Lloyd. Influence of Osmotic to 3 mm. in length, 7.¢. after 10 weeks, the last trace of the generative system has vanished. Stoppenbrink (15) has recorded similar occurrences in starving Planaria. He states that the order of disappearance is :—Yolk-glands, copulatory apparatus, testes, and ovaries. In Gunda, while the yolk-glands disappear first, the rest of the genital system appears to be absorbed at a uniform rate. Stoppenbrink states that there is no phagocytosis in the absorption of the organs. In Gunda the fact that an organ undergoing reduction is always surrounded by a sheath of parenchyma cells (see fig. 4, B), some of which even penetrate into the various organs, makes it appear probable that the organs are destroyed by a phagocytic action of the undifferentiated parenchyma. Fig. 2 shows an adult testis, fig. 3 a testis undergoing degeneration. Figs. 4, A and B, show the changes in the ovary. a EEEEERSE RTs bes-- SSSDEE 5 og Fig. 2.—Testis of Adult Worm. g- gut. tes. testis. y.gl., yolk-glands. SEE! ar} Ts sie] 1 Q § NS i) ldalslole) aa Ae So Fic. 3.—Testis after 21 Days’ Regeneration of Posterior End in Sea-water. deg.sp.c., degenerating spermatocytes. es., testis. Pressure upon the Regeneration of Gunda ulve. 5 Fia. 4. A.—Ovary B.—Ovary after 25 days’ regeneration of posterior end in sea-water. g- gut. pa.c., parenchyma cells. ov., ovary. ov.d., oviduct. (b) Growth. Experiments were also made in rate of growth. Worms which had been reduced to 1:5 mm. in length by a long period of starvation were fed daily on scraps of the tail muscle of shrimps. The rate of growth was very rapid, and is indicated in curve B, in fig. 1. Unfortunately the experiment was interrupted, and the worms went without food from the 25th to the 37th day. It can be seen that this period was one of reduction in length. On the feeding being resumed growth again took place. At the beginning of the experiment the worms were without a trace of the generative system. This was first noticed in the sections at the 10th day, we. when the worms were 2°7 mm. long. The generative system was completely restored at a length of 46 mm.,7.e. 36 days after its first appearance. The generative system in G'unda, as in other Turbellaria, is formed by the localisa- tion and metamorphosis of nests of parenchyma cells. III. PRELIMINARY EXPERIMENTS ON DURATION OF LIFE OF WHOLE ANIMALS IN SOLUTIONS OF DIFFERENT OSMOTIC PRESSURE. A preliminary series of experiments was made so as to ascertain the limits of density, salinity, and osmotic pressure which whole worms are capable of withstanding. A table recording these results is given below (Table IJ). Conditions under which the worms can live for seven weeks were taken as capable of supporting them alive indefinitely. 6 Miss D. J. Lloyd. Influence of Osmotic Table I. ® = Oftienoetneses) | Length of life Composition of water. Density. | Salinity. ne of G. ulue (eigen, watlnc, in the solution 0°5 atmosphere). . grm. per litre. 1. Distilled water ............... 1-000 0 ‘00 = 1-18 hours. 2. Mapiwaterven nemene era ccc 1-000 0-00 — 6 hours-3 days. 3. 100 c.c. distilled water— + 65 c¢.¢. sea-water 1 ‘000 160 1:0 3-6 days. 4. “110. 5, x 1-001 aan | 2-0 Indefinitely. 5. i FD a, iy 1-002 5°70 3°5 ’ 6. BO) 34 1-007 11-4 7-5 fi Hs +100 ,, 1-012 17‘1 11-0 is 8. +200 ,, i 1015 22:°8 15-0 < 9. +500 ,, 3 — 28 °5 18 °5 By OD Seaswater....j):ccesccessetconce 1-028 34 °2 22:5 is 11. 100 c.c. sea-water— + 10 ¢.c.2°5M NaCl 1-030 44 °3 29-5 i 12. fs 0), xu 1-037 52°8 33°°5 14 days- Indefinitely. 13. + 30 ,, » 1-043 60 ‘1 40 °5 18-40 days. 14, 2 AD . 1-046 66°7 cS 5-20 ,, | 15. 5 30 i 1-058 83 °8 = eS, | 16. +200 ,, A 1:073 109-0 — 6 hours-1 day. The solutions used were made by mixing known volumes of sea-water with distilled water, or with 2°5M NaCl, calculating the salt content, and from this taking the osmotic pressure from the values given in Kriimmel’s ‘Handbuch der Ozeanographie’ (8). The salinity of the sea-water used was measured by titrating against standard silver nitrate and determining the chlorine value. Knudsen’s tables (7) give the formula for converting the chlorine value obtained into the salinity value. It will be seen that the worms are capable of living for an indefinite time in water which has an osmotic pressure of more than 2 and less than 33 atmospheres. Between these limits the worms remain perfectly normal in appearance. In solutions more dilute and in fresh and distilled water the worms swell up to four or five times their normal size. In water having a salinity of 66 per cent. or more they become shrivelled. ITV. Norma, REGENERATION OF THE POSTERIOR END. The changes which set in in G. ulv@ after section and which culminate in the restoration of the lost parts and production of a complete worm fall under the head of regulatory rather than of regenerative changes. There is no growth in the ordinary sense of the word, since the first product of Pressure upon the Regeneration of Gunda ulvee. 7 regulation is never larger than the fragment which produced it. Fig. 5, 1-4, show the stages in the transformation of an anterior end into a complete worm. It can be seen that there is in this case actual diminution in length. This is due to the fact that the regulation occurred under starving conditions. When the pieces are fully fed there is no reduction in length. es 1 2 3 ia. 5.—Regeneration of Posterior End in Sea-water. 1. 7days. 2. 14 days. 3. 28 days. 4. 49 days. 5. Side view of 3. 5 The chief rdle in the transformation is played by the parenchyma cells. These cells are simall migratory cells which are found in large numbers | throughout the whole length of the animal’s body. In regulation they have two functions— (1) They migrate in large numbers to the region of the wound, where they form first an undifferentiated mass of cells and later the new tissues. (2) They act as phagocytes, making a sheath round the old organs, ¢.7., brain and genitalia, and reducing them in size until they have become proportionate to the size of the new worm. The course of regulation in G. wlvew is essentially similar to that already described in various species of Planaria by Schultz (11), Flexner (4), and Bardeen (1). The wound is closed by the ectoderm creeping inwards as a thin acellular layer (see fig. 6, ac.ep). After the wound has healed (three to five days after section) the parenchyma accumulates at the hind end. This accumulation of parenchyma cells is chiefly due to migration. Mitotic figures are to be found, but are not common in the sections, showing that actual cell division plays only a small rdle in the building up of the new tail. This consists for a few days of undifferentiated parenchyma, but later differentiated cells appear. The formation of the muscular layer of the new tail takes place by differentiation 8 Miss D. J. Lloyd. Influence of Osmotic Fig. 6.—Regeneration of the posterior end of G. wlve in sea-water. 3 days. ac.ep., acellular extension of epidermis. ep.,epidermis. m.,muscle. u.c., parenchyma cell. ph. pharynx. ph.c., pharynx chamber. w., point of closure of wound. W : ; Fic. 7.—Regeneration of the posterior end of G. wlve in sea-water. 14 days. m., mouth. m.c., muscle cell. pa.c., parenchyma cell. pf., pharynx. ph.c., pharynx chamber. w., point of closure of wound. ‘ Pressure upon the Regeneration of Gunda ulve. 9 of parenchyma cells, one parenchyma cell forming one differentiated muscle fibre (fig. '7). The restoration of the circle of the nerve cords also takes place by parenchyma cells pushing their way into the cut end of the old nerve cords and becoming transformed into nerve cells. The new eut is formed by the cut ends of the two branches of the old gut fusing together behind the pharynx. The wound on the pharynx is also healed by the migration of parenchyma cells. The new mouth is formed by a perforation appearing emmeen the pharynx chamber and the exterior. The mouth perforates at a point just anterior to the point of closure of the wound, usually about nine days after section (fig. 7, m. and w.). While the constructive changes described above have been taking sess in the region of the wound, in the rest of the animal’s body reductive change has been simultaneously proceeding. The first system to show degenerative change is the generative system. The amount of degeneration that takes place is directly proportional to the amount of restoration that is required. In the present case, where regeneration is being considered after a cut that has removed half the body, reduction proceeds so far that the whole of the generative system is absorbed. The yolk glands are absorbed first, then the genital glands and secondary sexual organs. - Besides the absorption of the genital system it can be seen on referring to . figs. 5, 1-4, and 8, 1-4, that the growth in size of the posterior end is accompanied by a reduction in size of the anterior end. This external adjustment is found to apply also to all the internal organs of the animal, a.¢., as the new parts grow larger the old grow smaller until the proper equilibrium is reached. When these changes have restored the normal proportions of the worm, regulation is complete. The subsequent increase in size, and in the case under consideration, restoration of the generative system, are phenomena of normal growth (see Section II, 0). It might here be noticed that the reductional changes described above in a regenerating worm fragment are entirely identical with those described in Part II («) of this paper, where the reduction was the result of starving whole animals. It will be seen later that, when regeneration is inhibited, reduction is inhibited to the same degree. Absorption of the generative system takes 10 weeks in starving worms; in animals which have been bisected and are regenerating under starving conditions it takes 5-6 weeks. In animals which have been bisected and are being fed while regenerating, the genital system is greatly reduced, but never completely absorbed. 10 Miss D, J. Lloyd. Influence of Osmotic Fic. 8. Median longitudinal sections through worm regenerating in sea-water.—l. 7 days. 2. 14 days. 3. 21 days. 4. 56 days. Median longitudinal sections through worm regenerating in 100 c.c. sea-water+ 20 cc. 23M NaCl—5. 7 days. 6. 14 days. 7. 21 days. 8. 31 days. V. REGENERATION UNDER VARYING CONDITIONS OF OSMOTIC PRESSURE. A number of experiments in regeneration were set up in artificial solutions 4-11 of Table IJ, in order to see what effect was produced by raising or lowering the osmotic pressure of the medium. It can be seen that a range of pressure of from 2 to 40 atmospheres was examined. The results of such experiments show that moving the osmotic pressure in either direction from a definite optimum value just below that of sea-water results in a decrease in the rate of regeneration. This decrease of rate culminates, at 2 and 40 atmospheres respectively, in complete inhibition of the restorative processes. Fie. 9 shows a number of worms on the 42nd day of regeneration, which have been taken from different experimental media. Pressure upon the Regeneration of Gunda ulve. 1d We, 3 4 Fic. 9.—Worms 42 days after section from— 1. 100 cc. distilled water+50 ¢.c. sea-water. 3. 100 c.c. distilled water + 200 ¢c.c. sea-water. water. water+10c.c. 24M NaCl. Table II. 2. 100 «cc. distilled water+100 cc. sea- 4. Sea-water. 5. 100 c.c. sea- Time required. No. of Osmotic solution (see | pressure in 5 : Tove Course of regeneration. Table 1). |atmospheres.| For Healing perforation ae of mouth. 4 2°0 Wounds never — All pieces die under 21 days. Both close gonads show degeneration. Gut | cells swollen and vacuolated. | 5 3°5 8-10 days | Mouth never| Testes continue functioning for perforates 24 days after section, forming accumulations of sperm. Gut cells vacuolated. Regeneration never complete. 6 75 6-10 days 18 days - | Testes function for 6 weeks after section. Disintegration of gonads begins in 8 weeks. Regeneration complete in 12 weeks. 7 11°0 6-8 ,,: 14, Testes only function for a few days after section. Gonads very re- duced in 8 weeks. 8 15:0 5-7, 1@ ,, Gut cells always remain normal. Regeneration complete in 6 weeks. Secondary sexual organs appear. 9 18 °5 Bg Sin Regeneration complete in 4-5 weeks. Secondary sexual organs appear. 10 225 G8) gy Os Regeneration complete in 7 weeks. (sea-water) No trace of generative system present. 11 29 5 6-9 ,, Ze ays Regeneration takes 8-9 weeks. Genital system entirely absorbed. 12 33 °5 6-10 ,, Mouth never| Gut cells in these tissues appear perforates shrunken, with very dense proto- plasm. “Tailless” forms pro- duced. 13 40 °5 Wounds never — All pieces die without any regene- close ration in 21 days. 12 Miss D. J. Lloyd. Influence of Osmotic The decrease in rate of regeneration can be measured proportionately by a consideration of the time taken to reach some definite stage. Two such points are considered here: (1) the time taken for the healing of the wound ; (2) the time taken for the perforation of the new mouth. These figures are summed up in Table II on p. 11. It can be seen that regeneration is most rapid in solution 9. This solution consists of 100 c.c. distilled water +500 c¢.c. sea-water. Regeneration in this solution is accompanied by reduction (absorption of the yolk-elands, etc:), but the regenerative processes proceed very rapidly, and, at the moment when the normal form is restored, reduction has not gone so far as to have removed the whole of the genital system. Under these conditions the secondary sexual apparatus is redeveloped in the new tail. This also occurs in solution 8 (100 c.c. distilled water+200 c.c. sea-water). In more hypotonic solutions, in sea-water and in hypertonic solutions, restoration of the normal form is a much slower process. In these solutions, the reduction due to regeneration + the reduction due to the longer starvation, make it impossible for the worm to redevelop the secondary sexual organs, even after the normal form has been restored. In normal sea-water or hypertonic water, the removal of the posterior end of the worm acts as a check on the production of sperm. Sperm present in the testicles at the time of section remains there, and, though the activity of the spermatocyte cells does not immediately cease, it is greatly diminished, and, as regeneration proceeds (under starvation conditions), the whole generative system is slowly absorbed to feed the growing parts. In worms regenerating in hypotonic water, the production of sperm continues for some time longer, and the sperm produced leaves the testes, passes down to the cut end of the vas deferens, where, being unable to escape, it collects in sinuses, one on either side. This condition is shown in fig. 10. In figs. 11 and 12, A and B, are shown camera drawings of gut cells (g.c.) of animals regenerating in hypo- and hypertonic water, and also, for comparison gut cells of normal whole animals. In the strongly hypotonic solutions the cells are swollen and vacuolated, and the cell boundaries are hard to distinguish. In the hypertonic solution the cells are shrunken. This fact suggests that in the one case water has actually been absorbed by the tissues, and in the other case extracted from them. From this it appears that the epidermis in Ganda must be a highly protective membrane, since whole animals placed in similar media show no such changes. A histological study of the tailless forms produced at 33°5 atmospheres was made by sectioning the worms at various stages. These sections show certain peculiarities in the behaviour of the parenchyma cells. In sea-water after Pressure upon the Regeneration of Gunda ulvee. Fie. 10.— Regeneration in 100 c.c. distilled water +50 c.c. sea-water. 43 days g- gut. sp., sperm, ves., vesicle. ~~ Fie. 11.—Normal gut cells. g-c., gut cell. v., vacuole. 13 14 Miss D. J. Lloyd. Influence of Osmotic B Fie. 12. A.—Gut cells from worm regenerating in 100 c.c. distilled water +50 c.c. sea-water. 43 days. B.—Gut cells from worm regenerating in 100 cc. sea-water+20 cc. 24M NaCl. 32 days. g.c., gut cell. v., vacuole. section, the parenchyma cells congregate in the region of the wound (see above), but in the solution of high osmotic pressure they only travel to the wound very slowly. The first result of this slow migration appears in the greater length of time required to heal the wound (6-10 days as compared with 3-5 days in sea-water). The next point to be noticed is that the parenchyma cells do not collect together at the hind end to form the mass which is subsequently to become the new tail (figs. 13 and 14). These tailless forms never develop the perforation for the new mouth. They develop a thin muscle layer under the new epithelium, but the circuit of the nervous system is not restored. From the third week after section they develop large vesicular spaces (see fig. 8, 7 and 8, v.) in the gut, or in the pharynx chamber, after 7-8 weeks these become filled with masses of disintegrating tissue, and the animals subsequently die. A similar retardation of the movements of the parenchyma cells is found in the strongly hypotonic media. This checking in the processes of restoration is also to be found to a parallel degree in the processes of reduction. For instance, in sea-water an anterior half is completely restored in 8 weeks, and during restoration every Pressure upon the Regeneration of Gunda ulvee. 1G Fia. 13.—Regeneration in 100 cc. distilled water+50 c.c. sea-water. 15 days. ph.c., pharynx chamber. w., point of closure of wound. Fic. 14.—Regeneration in 100 c.c. sea~-water +50 c.c. 25M NaCl. 6 days. pa.c., parenchyma cell. ph. w., point of closure of pharynx wound. w., point of closure : of wound. trace of the genital system is absorbed. In the tailless forms 8 weeks after section it is found that the genital system is not reduced to any great extent. 16 Miss D. J. Lloyd. Influence of Osmotic The slight degree to which the reduction processes are carried is reflected in the external form of these pieces, which only suffers very slight changes during the time (c¢/. figs. 5 and 15), Bz, i 6 gy Fie. 15.—Stages of Regeneration in 100 c.c. sea-water +20 c.c. 24M NaCl. 1. 10 days. 2. 14 days. 3. 28 days. 4. Side view of 3. Many sections were examined for mitotic figures, and a comparison was made between the mitoses found in worms regulating in sea-water and in water with a higher or lower osmotic pressure. Text-fig. 16 shows the type of mitoses found. At osmotic pressures of 7:5 and 33 atmospheres the mitoses appear to be a little irregular, but in all cases mitotic figures are so rare that it would not be justifiable to ascribe a principal role in the failure to regenerate, to any such irregularities. CS OOD ue 6 fA 5) 6 Wi, Q en ‘ 2) rf 8 g) ZO ail 12 Fie. 16.—Mitotic Figures from the Parenchyma Cells. 1-4. From tissue regenerating in sea-water. 5-7. From tissue regenerating in 100 cc. distilled water+50 c.c. sea-water. 8-12. From tissue regenerating in 100 c.c, sea- water +20c.c. 24M NaCl. Pressure upon the Regeneration of Gunda ulvee. 17 VI. DISscUSSION. The results given in this paper show that for G, wlve there is a definite value for the osmotic pressure of the surrounding medium, at which regenera- tion of the hind end occurs most rapidly. This point lies inside the range of variation of osmotic pressure to which the animals are subjected in their natural environment. Above or below this point the rate of regeneration diminishes, until finally it is entirely inhibited. A similar phenomenon has been obtained by Child (2) in Planaria by adding doses of alcohol, ether, and potassium cyanide to his cultures, or by raising the temperature to a point just below that which causes the death of the tissues. Child has ascribed this change of rate of regeneration, with its final production of tailless forms, to the lowering action of the external agent employed upon the “ rate of action” of the regenerating fragment as a whole. In the present case the morphological cause of the reduced rate of regeneration can be localised in the parenchyma cells. These normally migrate in large numbers to the region of a wound and there build up the new tissues. When, however, the osmotic pressure of the surrounding medium is removed from the optimum value, this migration is checked and finally inhibited. In the latter case, the wound never heals, and the fragment ultimately dies. In the case where the osmotic pressure is just short of completely preventing the migration of the cells, the wound slowly closes but no new organs are formed. This produces the so-called tailless forms. The swelling that occurs at the hind end in these cases is not due to the formation of tissue but to the development of large cavities in the gut or pharyngeal chamber, or, in hypotonic media, to the absorption of water. In Planaria maculata, Curtis (3) describes the development and absorption of the genital system as a regular seasonal change. (. uwlvw, however, is found sexually mature at Plymouth all the year round. Once it has reached maturity, the genital system is only re-absorbed under conditions which make it necessary for the animals to use their own tissues as a source of energy. These conditions are—(1) hunger, (2) regeneration after a cut which has removed a fairly large proportion of the animal’s body. (Doubt- less, after loss of small amounts, some reduction takes place, but not enough to be easily recognised.) In the regulation of G. ulvw, after a loss of some of its parts, reduction and regeneration go hand in hand, and in some cases where regeneration is inhibited by some external factor, eg., high osmotic pressure, reduction is inhibited to the same extent. As Stockard and others have already pointed out, regenerating parts have a potent influence on the old body. This fact is again well illustrated in VOL. LXXXVIII.—B. C 18 Miss D. J. Lloyd. Influence of Osmotic the sections shown in fig. 8. It can be seen that where a new part is growing, an old part is correspondingly diminishing (fig. 8, 1-4). In cases where there is no formation of new parts (fig. 8, 5-8), there is correspondingly no diminution of the old ones. Thus, when no tail grows out, and where the pharynx also shows no growth, the part anterior to the pharynx remains with unchanged proportions. The value of the irregular mitoses found in the parenchyma cells under the conditions unfavourable to development remains to be considered. Under normal conditions regenera- tion takes place without the parenchyma cells increasing to any great extent in number. For this reason it seems probable that the failure to restore the lost parts under conditions of very high or very low osmotic pressure is not due to failure of the parenchyma cells to divide, but that both checked migration and irregular mitoses are an expression of unfavourable change in the physiological conditions. This suggestion receives support from the appearance presented by the gut cells, which in strongly hyptonic solutions are swollen and vacuolar, while in hypertonic solutions they are contracted and dense. The most favourable conditions for the regeneration of (. ulvw are in a mixture of 100 parts of distilled water to 500 of sea-water, and at an osmotic pressure of 18 atmospheres. Regeneration occurs, though less rapidly, in solutions having an osmotic pressure as low as 7:5, or as high as 29°5 atmospheres. Above or below these two points the water-content of the tissues appears to have been altered to such an extent that the cells, notably the parenchyma cells, can no longer perform their normal functions. SUMMARY. 1. G. ulve is capable of living indefinitely in water having an osmotic pressure of more than Z and less than 33 atmospheres. 2. The rate of regeneration of the posterior end in G. wlve depends on the osmotic pressure of the medium. This osmotic pressure has an optimum value for regeneration at 18 atmospheres, 7.¢., just below that of sea-water, and limiting values at 5 and 33:5 atmospheres. 3. Restoration of Jost parts in G. wlvw is brought about entirely by the undifferentiated parenchyma cells which migrate to the region of the wound and build up the lost parts. 4. For values of the osmotic pressure lying between the optimum and the limiting values, this migration of the parenchyma cells is retarded, and the rate of restoration is retarded to a similar degree. At the limiting values of the osmotic pressure there is no migration of the parenchyma cells and no restoration of lost parts. Pressure wpon the Regeneration of Gunda ulve. 1g) 5. Under starvation conditions, @. wlve undergoes reduction. This consists in (1) absorption of the genital system, (2) general reduction in size. Both these are brought about by the activity of the parenchyma cells. 6. During the process of restoration of lost parts the same reduction processes occur as in starvation. Where the restoration of lost parts is retarded, eg., by raising or lowering the osmotic pressure, reduction is retarded to precisely the same extent. 7. In sea-water or hypertonic solutions removal of the posterior half of the body inhibits further production of sperm. In hypotonic solutions sperm continues to be produced for a varying length of time. 8. In strongly hypertonic solution examination of the gut cells shows that these have diminished in size and become more dense. In strongly hypotonic solutions they have increased in size and become vacuolar. I should like to take this opportunity of thanking the director and staft of the Plymouth laboratory for the promptness with which they have suppled me with material, and for their unfailing kindness during the whole time I have worked at the laboratory. JI should like also to acknowledge my indebtedness to the Royal Society, the Zoological Society, and the University of Cambridge for the use of their tables at Plymouth. Finally, I wish to thank the Trustees of the Balfour Fund for a grant which made it possible for me to take up the work in 1912. BIBLIOGRAPHY. 1. Bardeen, C. R., “Embryonic and Regenerative Development in Planarians,” ‘ Biol. Bull.,’ vol. 3, p. 262. Child, C. M., “ Experimental Control of Morphogenesis in the Regulation of Planaria,” ‘ Biol. Bull.,’ vol. 20 (1911). 3. Curtis, Winterton C., “The Life History, the Normal Fission, and the Reproduc- tive Organs of Planaria maculata,” ‘Bost. Soc. Nat. Hist. Proc.,’ vol. 30, p. 515 (1902). 4. Flexner, Simon, “ Regeneration of the Nervous System of Planaria torva,” ‘Journ. Morphol.,’ vol. 14, p. 337 (1898). 5. Jijima, I., “Untersuchungen iiber den Bau und die Entwicklungsgeschichte der Siisswasser Dendrocoelen (Tricladen),” ‘ Zeitschr. f. Wiss. Zool.,’ vol. 40, p. 359 (1884). 6. Jordan Lloyd, D., “The Influence of the Position of the Cut upon the Regeneration of Gunda ulve,” ‘Roy. Soc. Proc.,’ B, vol. 87, p. 355 (1914). Knudsen, Martin, ‘Hydrographical Tables,’ London and Copenhagen, 1901. Kriimmel, Otto, ‘Handbuch der Ozeanographie,’ vol. 1, p. 240, Stuttgart, 1907. Morgan, T. H., “Growth and Regeneration in P. lugubris,” ‘Arch. Ent. Mech., vol. 13, p. 179 (1901). 10. Przibram, H., “Equilibrium of Animal Form,” ‘Journ. Exp. Zool.,’ vol. 5, p. 259 (1907). bo So) Go Eu G 2 20 Sir D. Bruce and others. G. brevipalpis as a 11. Schultz, E., “ Aus dem Gebiete der Regeneration bei Turbellarien,” ‘ Zeit. f. Wiss. Zool.,’ vol. 72, p. 1 (1902). 12! “ Uber Reduction,” ‘ Arch. Ent. Mech.,’ vol. 18, p. 555 (1904). 13. Stevens, N. M., “A Histological Study of Regeneration in Simplicissima, Maculata, and organi,” ‘Arch. Ent. Mech.,’ vol. 24, p. 350 (1907). 14. Stockard, C. R., “Studies in Tissue Growth. IJ.—On the Rate of Regeneration and the Reaction of the Regenerated Tissue on the Old Body,” ‘Journ. Exp. Zool.,’ vol. 6, p. 483 (1909). 15. Stoppenbrink, E., “Der Einfluss herabgesetzer Ernihrung auf den histologischen Bau der Siisswassertricladen,” ‘ Zeitschr. Wiss. Zool.,’ vol. 79, p. 496 (1905). 16. Wendt, A., “Ueber den Bau von Gunda ulve,” ‘Arch. f. Naturgesch., 54 Jahrg. vol. 1, p. 252, Berlin, 1888. Glossina brevipalpis as a Carrier of Trypanosome Disease m Nyasaland. By Surgeon-General Sir Davip Brucs, C.B., F.R.S., A.M.S.; Major A. E. HAMERTON, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912 -14.)* (Received March 25,—Read April 30, 1914.) [PuatE 1.] INTRODUCTION. Glossina brevipalpis (Newstead) is found at many spots along the west. shore of Lake Nyasa. The nearest point to the Commission’s camp at Kasu where they were at all common was at the mouth of the Lingadzi river (138° 27’ S. lat., 34° 19’ E. long.). This was 50 miles away, but with the aid of a motor-cycle specimens of these flies were brought up to the camp. It is proposed in this paper to give: (1) a short account of the habits of this tsetse fly; (II) the results of the dissection of the flies; (III) the infectivity of the wild flies; and (IV) the result of various transmission experiments. One of the members of the Commission camped on the Lake-shore from April 26 to May 10, 1913, to superintend the catching and sending to the camp of these flies, He has supplied the following account of their habits :— * Major D. Harvey, R.A.M.C., took part in the work described in this and the four following papers; but, having left the Commission in September, 1913, before the reports were drawn up, his name does not appear in the titles. Carrier of Trypanosome Disease in Nyasaland. al TI. Hasits or G. BREVIPALPIS. G. brevipalpis is found frequenting the roads in a small area of country around Markho village on the west shore of Lake Nyasa, at the mouth of the Lingadzi river. This tract of country, which may be roughly estimated at about 5 square miles in extent, is broken up by swampy hollows and streams forming the delta of the Lingadzi. It is covered with palm forest and a dense undergrowth of high grass and bush, and is traversed by two main roads, one between Markho and Lingadzi estate running east and west, and another running parallel to the Lake-shore and about 1 mile to the west of it. The roads are hoed tracks about 3 yards broad, cut through the palm forest and walled in by high grass and almost impenetrable bush. G. brevipalpis is crepuscular in its habits and quite unlike other Nyasa- land tsetse flies. During the daytime, from dawn till about an hour before dark, one may pass along the roads or wander in the surrounding jungle without encountering a single fly of this species. But as evening approaches odd flies suddenly appear sitting motionless in the middle of the hard-trodden path all the way along the road between Markho and Kasache, a distance of 2 miles, and for about 2 or 3 miles up the Lingadzi road. They do not follow or settle upon passers-by like other tsetse flies, and they would pay no attention to a dog which was repeatedly walked through their haunts in the evening. In the dim light they are difficult to see, and resemble little bits of earth on the path, but the searcher’s attention is attracted by the sound of them buzzing up as they are disturbed by his footsteps. They immediately settle on the path again and are easily caught, for if missed by the first stroke of the net they at once resettle near the same spot. The hard-trodden surface of the path seems to have an irresistible attraction for them. They do not move about in search of food or chase each other, but remain motionless for several minutes, and when they move it is only to fly up and immediately settle again in the middle of the path. They were never seen on the roads at dawn, as the mornings at this time of the year were invariably cold and misty. Out of the 500 flies caught and examined on the spot all were males. Some of them were found on dissection to contain mammalian blood. Game, such as buffalo and several species of antelope, is common in the district. Flies kept in captivity remain dormant on the sides of their cage during the day. At night they are very active and buzz incessantly in their efforts to escape. If the side of the cage be applied to the skin of a goat or a dog © they will feed with avidity at any time of the day or night. A few wild flies were dissected in order to ascertain their natural food. 72, Sir D. Bruce and others. G,. brevipalpis as a In 50 flies seven contained recognisable blood. This in six cases was antelope blood, in the seventh probably human. It is curious the great preponderance of males over idles It seems to be a habit of the former to frequent paths in the evening, while the females presumably hide in the thick jungle. The same thing obtained to a lesser extent with G. morsitans. Among all the G. brevipalpis dissected at the laboratory, amounting to several hundreds, only four females were found. II. DISSECTION OF WILD G. BREVIPALPIS TO ASCERTAIN WITH WHAT SPECIES OF TRYPANOSOMES, IF ANY, THEY ARE JNFECTED. Four hundred and ninety-six wild flies were examined; 44 of these were positive and 452 negative. Table I gives the result of the dissection of the positive flies. | Among the 44 flies which contained flagellates, it was possible to make a more or less correct guess at the species in 19. These are Z. brucei vel rhodesiense, 1; T.. pecorum, 9; T. sume, 1; and TZ. grayi, 8. In 10 flies the flagellates were considered to belong to a pathogenic type, species unknown. No opinion could be expressed about the remaining 15. It is curious that G. brevipalpis should contain flagellates resembling T. grayt, which is so often found in G. palpalis. Now one thing common to G. brevipalpis and GC. palpalis is that they both live alongside water. This would point to the vertebrate host of 7. yrayi, if there is one, being some water animal, such as the crocodile or iguana, or some water bird. Conclusion. Wild G. brevipalpis are naturally infected with 7. brucei vel rhodesiense, T. pecorum and T. simie. Carrier of Trypanosome Disease in Nyasaland. 23 Table I. | Proboscis. | No. Proventri- | | Fore- | Mid- | Hind-| Salivary Ee of fly.| tT abial ign: cae, «| | Coe gut. | gut. | gut. | glands. cavity. | pharynx. | | | i | | 1912. | | | June 14 ... Se = = _ | a = oh GUC el a = - } = — ++ > +47 Perla <.| 33 & A 4 i 1) dlc ae Nae ae = pap ab | dee |) ab ae aA 4 he Git = = se | pa ees 6) = = + ap 5. CAs 7c] aul, | aa | ++] ++4 | ” 28 .. 8 | ey, ay 45 4° ++ a AeMacall as WN es = fog |e de Sept. 12 10 _ — ++ Tee | Sat 1913 Mar. 5 11 Sh eae + + ++ +44 | ee | +) 18 | 12 | ae =, ar SF = fnes: LS 13 _ - ++) 44 ) 44 — opal 14 _ + = es + = April 4 15 = = = + + + — ” 22 16 | Ta a + | | Wiehe eel ey da “| eeensT | 18 + Sutera fest .../ 19 = = in Bs | June 2...) 20 - _ te = | 9 PAA 21 ze a = fos: 5 22 ap ae a ar APE |) eae) |) ab ot 3 9 | 23 = ary } a= ee U2 ill 4 ++ ++ ++ }++ | ++] ++ _ ns, OBS cet PAE - Pee | ee pee ++ ed. 26 se eee $e | ee fl ek 5 ZB cool) AU = | = + ae co PAB cents =a} = £ ES eeoG...| 20 el A a8 aatgies j daly 1...) 380 shee SW) cere t+oot4+ ) $4) x 6) % 31 | —- — | ae = 9 11 ve 32 — = | + p= PAG eel oe ae ar |) _ | Voce ae i eee pus » 14...) 34 = a te = BP LA Nl! Bh ++ _ Fae | a oj AUG She] YD PS _ (+ | is 3 UG sc 37 | ar = } | -+ | do ds = BA AG) seal S8y ie se | + oe Bel fies se = Bit | aay oe er | ace = i 4 eee) 2B) el = Tee | ; + | we Sede | 41 | _— | _ | | + oa U2 MENA Tiana Ly ms | eck = SOP 2c! RASS WR. 8 - li seem = | Bee iGO... 44 = = [ine fos i | | Ill. THe Inrecriviry or THE WILD G. BREVIPALPIS. A few experiments were made to test the natural infectivity of G. brevipalpis by feeding them on healthy animals. When the flies were brought up from the Lake-shore they were fed on a 24 Sir D. Bruce and others. G., brevipalpis as a monkey, dog, and goat. Five experiments in all were made ; four negative, one positive. The following Table gives the result :— Table I1.—Feeding Wild Glossina brevipalpis. Monkey. Dog. Goat. No. . F é 3 : 5 *° 2 § “3 SS) S aS o ~ S88] 8) 8 | S888) 8) 8s see = = a | 5 2 Wt s S| Sj = 3 & | 5 S 1912. | | June 14... 42 = =} = | = yeah ese lh ea pe ee Ml = 1913. Mar. 18 ... 146 = =i] = | = = as alee llr Apr. 29...) 541 _ +/—-|]-— — ei] =) = = 35./l] ela a May 7... 90 - -—-|—-|- June 25 ... 276 = hee es = SoH ees tile as Ua har ele oes Total...) 1095 LV. TRANSMISSION EXPERIMENTS. Several experiments were carried out with GC. brevipalpis, to ascertain if this species of tsetse fly can act as a carrier of the various pathogenic trypanosomes found in Nyasaland. These experiments were made, not with laboratory-bred but with wild flies, and this of course takes away much of their value. It was found impossible, on account of the distance from the Lake-shore and the scarcity of flies, to attempt the breeding of . brevipalpis, in order to obtain laboratory-bred flies. The species of trypanosomes experimented with were: (1) 7. brucer vel rhodesiense, the trypanosome causing disease in man in Nyasaland; (2) T. brucei, Zululand, 1913 ; (3) 7. pecorwm; and (4) 7. capre. 1. The Development of T. brucei vel rhodesiense in G. brevipalpis. (a) Feeding Wild G. brevipalpis first on Animals infected with the Trypanosome causing Disease in Man in Nyasaland and then on Healthy Animals, to discover uf this Species of Trypanosome passes through a Cycle of Develop- ment in this Species of Tsetse Fly. Two hundred and thirty-two wild G. brevipalpis were used in six experi- ments, but in no case with a positive result. Only fifty-three flies were dissected and eight infected flies found. If all had been dissected probably about 30 to 40 infected flies would have been found. Carrier of Trypanosome Disease in Nyasaland. 25 Table III. | | : - No. Expt. * . No. of No. of days Date. Expt. | of flies | positive or Renee infected flies before flies | used. | negative. re ; found. became infective. | | in| OTS: | | Mar. 6 ...| 1986 | 2 ~ 1 0 Hea 2O)<..-1 2062 | 10 - 6 (@) June 2...) 2201 70 - 8 2 co pe 2213 60 — 19 2 lnc: .2-| 12232 50 - 18 4 | ‘Tuly 23 2310 40 | - 1 0 x (b) Details of the Six Negative Experiments. Table IV. | | Expt. Day of Procedure. Remarks. | expt. | | | i | | 1986 1 Flies fed on infected Guinea- Only 2 flies used; one escaped and the | pig 1658. other died on the 15th day of the ex- | | 2-4 Starved. periment; on dissection it was found | 5-15 Fed on clean Guinea-pig 1907. negative. | | 2062 1-4 Flies fed on infected Monkey 10 flies used. 6 were dissected; all 970. negative. 5 Starved. | 6-28 Fed on clean Dog 2041. q | 2201 1-6 Flies fed on infected Monkey 70 fiies used. Only 8 were dissected ; | 2156. 2 found infected, one with 7. grayi, the | Uf Starved. other with a pathogenic type of trypano- | 8-33 Fed on clean Dog 2211. some. | i | 2213 1-12 Flies fed on infected Monkey 60 flies used. 2 found infected. Only | | 2151. 19 dissected. | | 13-14 Starved. | 15-44 Fed on clean Dog 2237. | 2232 =| 1-6 Flies fed on infected Monkey 50 flies used; 4 found infected. 18 flies | 2152. dissected. The 4 infected flies showed | ; 7-9 Starved. infection of the gut alone | | 10-42 Fed on clean Dog 2244. ' | 2210 1-3 Flies fed on infected Monkey 40 flies used. Only one fly dissected ; 1792. negative. 4-5 Starved. 6-39 Fed on clean Dog 2315. (c) Result of the Dissection of the Infected Flies. As will be seen from Table III, eight infected flies were found among the Gr. brevipalpis which had fed on animals infected with the trypanosome causing disease in man in Nyasaland. The following Table gives the result of the dissection of these eight flies :— 26 Sir D. Bruce and others. G. brevipalpis as a Table V. | | | Expt. Time, days. | Proboscis. | Alimentary tract.| Salivary glands. 2201 30 - + = 2201 37 = + = 2213) | 338 =) | = eee 22138 46 + ++ = 2232 37 | + = 2232 37 | + | — 2232 44 = + = 2232 44 — oF = In Experiment 2201 there were two infected flies found. One of these appeared to be infected with 7. grayi, the other with a trypanosome of a. pathogenic type. In Experiment 2213 there were also two infected flies. In one the development was restricted to the proboscis and the alimentary tract; in the other the salivary glands as well as the intestine were found to be swarming with trypanosomes. This fly, which was dissected 33 days after it had fed on an infected monkey, had the whole lumen of the glands filled with active motile trypanosomes, which came pouring out of the broken end of the glands. It was thought at the time that these must be infective forms of the trypanosome causing disease in man in Nyasaland, but a part of the salivary glands and contents of gut injected into a white rat failed to infect it. In spite of this negative experiment, however, it is probable that. this represents a true development of the Nyasaland trypanosome in G. brevipalpis, the development not having reached the infective stage. A fuller description of the morphology of these salivary forms will be given under the next heading. It may be noted here that salivary glands infected by Z. brucei vel rhodesiense or by T. brucei, Zululand, 1913, seem to be more crowded with trypanosomes than in the corresponding infection of G. palpalis by T. gambiense. The salivary glands in the former case appear to be swollen and bursting with the flagellates. In Experiment 2232 there were four infected flies found, but in none was there any invasion of the salivary glands. (d) Morphology of the Trypanosomes found in the Salivary Glands of a Wild G. brevipalpis which had fed on a Monkey infected with the Trypanosome causing Disease in Man vn Nyasaland. This fly, as described above, was found in Experiment 2213, and failed to infect the clean Dog 2237 which it had been fed upon. Part of the salivary = Carmer of Trypanosome Disease in Nyasaland. 27 glands and contents of the intestine also failed to infect Rat 2234. It would appear, then, that this fly had not yet reached the infective stage. It is now proposed to describe the different forms of the trypanosomes found in the salivary glands of this fly somewhat in detail, and to bring forward a theory in regard to a stage which seems to occur in the final development of this trypanosome in the salivary glands. The different forms and apparent stages in their development could be more easily made out in G. brevipalpis than in G. morsitans, on account of its greater size. On Plate 1 this evolution of the trypanosome causing disease in man in Nyasaland in the salivary glands is represented. Figs. 1 and 2 represent the long, slender developmental forms of trypano- somes found in the intestine of the fly, from the mid-gut to the proventriculus. It is this type of trypanosome which invades the salivary glands. Fig. 3 shows the change in shape which the intestinal forms undergo on entering the salivary glands. The posterior extremity lengthens somewhat and the micronucleus and flagellum pass forward. This appears to be the commencement of the change to the crithidial type. Fig. 4 represents the fully developed crithidial form of the parasite. The micronucleus and flagellum have passed further forward until they lie anterior to the nucleus. The anterior portion of the parasite has broadened out ; the posterior has become attenuated. Fig. 5, the parasite is still crithidial in form. The anterior half has become still broader, the posterior half elongated and further attenuated. Fig. 6, the attenuated posterior portion has shortened. Figs. 7, 8 and 9 represent further stages in the evolution. The long attenuated posterior extremity has disappeared and the parasite has become contracted and thickened. Fig. 10 shows the last phase of tle crithidial stage. Here the parasite has assumed a rounded form and the flagellum is folding on itself. Fig. 11 represents a group of parasites in the encysted stage. In this form they are found massed together in the lumen of the salivary glands. Fig. 12, in this group the encysted forms are just unfolding. Fig. 13 shows a single encysted form breaking open. The micronucleus is now posterior to the nucleus; the crithidial has become the trypanosomal. Figs. 14, 15, 16, and 17 demonstrate further stages in the unfolding of the encysted form. The parasite is now assuming a trypanosome shape. Figs. 18 and 19 show the fully developed salivary-gland form of the trypanosome. This constitutes a reversion to the “ blood form” from which the cycle of development began and is the only infective form. On comparing these figures with the developmental forms of the 28 Sir D. Bruce and others. G. brevipalpis as a trypanosome causing disease in man in Nyasaland in G. morsitans,* or with the same forms of 7. brucei, Zululand, 1913,+ it will be apparent that in all probability this is a true development of this trypanosome in CG’. brevipalpis. It is true wild flies are being dealt with, but in this district it is only trypanosomes of this type which invade the salivary glands, so that T. pecorum, T. simiw, and JT. capre are excluded. It is very unlikely that 7. grayi invades the salivary glands. Conclusion. G. brevipalpis is capable of acting as a carrier of 7. brucei vel rhodesiense, the trypanosome causing disease In man in Nyasaland. 2. The Development of T. brucei, Zululand, 1913, in G. brevipalpis. (a) Feeding Wild G. brevipalpis first on Animals infected with T. brucei, Zululand, 1913, and then on Healthy Animals, to ascertain if this Species of Trypanosome passes through a Cycle of Development in this Species of Tsetse Ely. Hable var LEDs aMhs No. of Ou ; No. of No. of days Date. Expt. | flies BSD Dee ne oe infected flies before flies used pire gare aera found. became infective. 1912. | May 12 ...| 2130 80 = 10 i July 1...) 2250 60 4 2 0 51 elie a2 299 60 = 0 0 (b) Details of the Three Experiments. Table VII. Expt. Day Of Procedure. Remarks. expt. 2130 1-2 Flies fed on infected Monkey 1970. | 80 flies used ; 1 found infected; 3 Starved. only 10 dissected. 4-35 Fed on clean Dog 2142. 2250 1-4 Flies fed on infected Dog 2240. 60 flies used; only 2 dissected, 5-6 Starved. both negative. Dog 2276 showed 7-60 Fed on clean Dog 2276. trypanosomes on the 58th day. 2299 1-10 Flies fed on infected Dog 2254. 60 flies used ; none dissected. Il Starved. | 12-61 | Fed on clean Dog 2314. * “Roy. Soc. Proc.,’ B, vol. 87, p. 516 (1914). + Ibid., B, vol, 87, p. 493 (1914). Carrier of Trypanosome Disease 1 Nyasaland. 29 Two hundred wild G. brevipalpis were used in three experiments—one positive, two negative. Only twelve flies were dissected, one of which was found to contain in the intestine long, ribbon-like trypanosomes of apparently a pathogenic type, but it was not possible to place it. It is to be regretted that all the flies had not been dissected as was the rule in these transmission experiments. Dog 2276 became infected 58 days after the flies had fed on the infected animal. This would point to the development of 7. brucei, Zululand, in the flies. If there had been a fly in the cage naturally infected with the Nyasa- land strain, then the animal fed upon ought to have shown trypanosomes in its blood earlier than 58 days. It may therefore be held as highly probable that Dog 2276 was infected by a G. brevipalpis in which a development of T. brucei, Zululand, 1913, had taken place. Conclusion. G. brevipalpts is capable of acting as a carrier of 7. brucei, Zululand, 1913. 3. The Development of T. pecorum in G. brevipalpis. (a) Feeding Wild G. brevipalpis first on Animals infected with T. pecorum and then on Healthy Animals, to ascertain if this Species of Trypanosome passes through a Cycle of Development in this Species of Tsetse Fly. Table VIII. No. Expt. , No. of No. of days Date. | Expt. of flies positive or | pene g | infected flies before flies used. negative. | = ; found. | became infective. i | | 1912. | May 28...... 2190 80 + av) 6 | 29 June 2...... 2207 50 — | 16 | (0) One hundred and thirty wild G. brevipalpis were used in two experiments— one positive, one negative. Only 33 flies were dissected ; six found infected. (b) Details of the Two Experiments. Table IX. Da of | | Expt y | Procedure. Remarks. | expt. | 2190 1-3 | Flies fgfl on infected Goat 2186. | 80 flies used; 6 infected flies found ; 4-5 | Starved. only 17 dissected. Goat 2202 showed | 6-36 | Fed on clean Goat 2202. trypanosomes on the 36th day. 2207 1-6 | Flies fed on infected Goat 2126. | 50 flies used ; only 16 flies dissected; 7 | Starved. all negative. | 8-46 | Fed on clean Goat 2210. | 30 Sir D. Bruce and others. G. brevipalpis as a (c) Result of the Dissection of the Infected Flies. Table X. Proboscis. pes be Alimentary Salivary Expt. Time, days. pea olands, Labial cavity. | Hypopharynx. 2190 23 = = + 2190 23 4p ap Sr + 2190 36 = = + - 2190 | 36 + ++ + - 2190 36 + ap ar + = 2190 36 + ++ + = In Experiment 2190 six infected flies were found. In four of these the hypopharynx was blocked with small “blood forms” of 7. pecorum. Taking into consideration the time which elapsed between the feeding on the infected goat and the appearance of an infective fly in the cage—23 days— and, further, the number of flies found infected with 7. pecorwm—A4 in 33— it must be admitted that a development of 7. pecorwm has taken place in G. brevipalpis. Conclusion. G. brevipalpis is capable of acting as a carrier of 7. pecorwm. 4. The Development of T. capree in G. brevipalpis. (a) Feeding wild G. brevipalpis, first on Animals infected with T. capre, and then on Healthy Aimimals, to ascertain if this Species of Trypanosome passes through a Cycle of Development in this Species of Tsetse Fly. Three experiments were carried out with wild flies. One was negative and two positive. On examining the positive experiments, they were found to be 7. pecorum infections, and not 7. caprw. The animals were probably infected by naturally-infected G. brevipalpis. No infection by 7. capre took place, but one of the flies, on dissection, was shown to have an undoubted development of 7. capre in the labial cavity and hypopharynx. Table XI. | | No. of ative | WN . No. of No. of days | Date. Expt. | flies BaD eae | Be care infected flies | before flies became | used 8 a4 ~ ; found. infective. | | \ = alongs | April11...| 2071 | 12 | = 3 1 | May 30...) 2199 | 60 + | Sumes 1 20 edmlyet (en 22000 lee + ) ) 24 | | Carrier of Trypanosome Disease in Nyasaland. 31 (b) Details of the Three Experiments. Table XII. Expt. Bay a Procedure. | Remarks. expt. | —— | — -= 2071 1-8 Flies fed on infected Goat 1912. 11 flies used; 1 infected fly found. | 9-10 Starved. | Only 3 dissected. | 11-12 Fed on clean Goat 2103. 2199 1-8 Flies fed on infected Goat 1912. | 60 flies used; 1 Fs Hectiet fly found. | 9 Starved. 25 dissected. Goat 2212 showed | 10-28 Fed on clean Goat 2212. | trypanosomes on the 27th day. | | 29-30 | Starved. | Goat 2245 negative. 31-50 Fed on clean Goat 2245. 2277 1-5 Flies fed on infected Goat 2220. | 50 flies used. No flies dissected. | 6-7 Starved. | Goat 23887 showed trypanosomes on : 8-32 Fed on clean Goat 2287. | the 31st day, and Goat’ 2362 on the | 33-34 Starved. 49th day. 35-49 Fed on clean Goat 2362. (c) Result of the Dissection of the Two Infected Flies. Table XIII. Proboscis. | a | Alimentary | Salivary Expt. Time, days. | Late eieeiss Labial cavity. | Hypopharynx. | | | | 2071 9 | & | aoe | = = | | | 2199 45 | ++ | - | ++ In Experiment 2071 there is evidently a development of 7. capre in the fly found infected. The intestine and salivary glands are free, whereas the hypopharynx is crammed with numerous short trypanosomes of the 7. capre type. In the labial cavity one large colony of large flagellates of a crithidial type was seen. In Experiment 2199 the intestine and labial cavity of the infected fly were found to have a heavy infection of trypanosomes. As the intestine was also involved, this is probably a natural infection of the fly with 7. pecorum. The animal the flies were fed on was found to be suffering from a 7 pecorwm infection. In the third experiment—2277—none of the flies were dissected, but, as the animal the flies were fed on became infected, as in the iast experiment, 32 G. brevipalpis as a Carrier of Trypanosome Disease. with 7. pecorum, and not with 7. capre, it is probable that no complete development of the latter had taken place. Conclusion. G. brevipalpis is capable of acting as a carrier of 7. capre. GENERAL CONCLUSION. G. brevipalpis is capable of acting as a carrier of ZT. brucer vel rhodesiense, T. brucei, Zululand, 1913, 7. pecorum, and T. capre.- DESCRIPTION OF PLATE. (See also p. 27 above.) Plate 1, Trypanosoma brucei vel rhodesiense, the trypanosome causing disease in man in Nyasaland. Figs. 1 and 2, intestinal forms. Fig. 3, intestinal form after entering salivary glands. Figs. 4-10, crithidial forms. Figs. 11 and 12, groups of parasites in the encysted stage. Figs. 18-17, encysted forms opening out. Figs. 18 and 19, fully developed salivary-gland forms. This constitutes the end of the cycle of development, which ends where it began, in the “blood form” of the vertebrate host. Stained Giemsa. x 2000, e others. hoy Soc. Proc B.vol, S811. “s § > ’ i J tA 5 4 ay ; L : bea a Sst j { S. if S t AO IZ 48 AD Trypanosome causing Disease wm Mar in Nyasaland. Devtdopment in Glossina brevipalpes. ieGibbous. del, % us X 2000. 33 Trypanosome Diseases of Domestic Animals in Nyasaland. Ill. — Trypanosoma pecorum. Development im Glossina morsitans. By Surgeon-General Sir Davip Brucg, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.0., and Captain D. P. Watson, R.A.M.C.; and Lapy Brucz, R.R.C. (Scientific Commission of the Royal Society, Nyasa- land, 1912-14.) (Received March 25,—Read April 30, 1914.) [PLATE 2.] INTRODUCTION. In a previous paper* the morphology of this species of trypanosome and its action on animals were described. In this it is intended to give an account of its development in Glossina morsitans. This trypanosome belongs to the group in which the development takes place first in the gut, then passes forward into the labial cavity of the proboscis, and finally reaches the hypopharynx, where the trypanosomes revert to the original “ blood-forms” and become infective. There is no infection of the salivary glands. THE DEVELOPMENT OF T. PECORUM IN G. MORSITANS. Seven experiments were carried out with laboratory-bred flies. Five were positive and two negative. Table I—Laboratory-bred Flies. | No. of Expt. No. of | No. of days | Date. Expt.| flies | positive or| infected flies before flies | Ae | used. | negative. found. became infective. aired ; 1912. | | | May 16... 546 22 4 53 | 69° F. (20°5°C.). July 2... 524] 20 + 2 37 65° F. (18 :3° C.). TENE. | | | Jan. /3...| 1732.| © 60 = 0 au | 84° F. (28 '8°C.). | ” 7...| 1737 | 40 + 3 19 | 84° F. (28 °8°C.). Heb. 10...) 1853 | 25 + 5 | 24. | 84° F. (28 :8°C.). | eae 950). 33 a 6 | 21 84° F. (28 ‘8° C.). April 29... 2115 | 40 = 4, == 84° F. (28 °8°C.). | | | | * “Roy. Soc. Proc., B, vol. 87, p. 1 (1913). VOL. LXXXVIII.—B. D 34 Sir D. Bruce and others. Trypanosome Two hundred and forty flies were used and twenty-four infected flies found—10 per cent. The first two experiments were carried out at the ordinary temperature of the laboratory ; in the others the flies were kept in the incubator. Details of the Five Positive Experiments. The following Tables give the principal details in the carrying out of the five positive experiments. They were all carried out with laboratory- bred flies :— Table II. Expt. ees | Procedure. Remarks. | — = = — — | 546 | 1-3 22 flies fed on TZ’. pecorum- | Goat 559 became infected on the 60th day ; infected dog. | Dog 880 on the 82nd day. All flies dis- 4 | Starved. sected ; 4 found infected. 5-62 | Fed on clean Goat 559. 63 Starved. 64-82 Fed on clean Dog 880. | | 524 | 1-5 | 20 flies fed on 7. pecorwm- | Trypanosomes first appeared in blood of infected rat. : | Dog 541 on the 44th day. All flies dis- 6 Starved. | sected; 2 found infected. 7-44 Fed on clean Dog 541. 1737 1-3 40 flies fed on 7. pecorwm- | Trypanosomes first appeared in blood of infected dog. | Dog 1750 on the 26th day. All flies dis- 4 | Starved. sected ; 3 found infected. 5-27 Fed on clean Dog 1750. 1853 1-3 25 flies fed on Z. pecorum- | Trypanosomes first appeared in blood of infected dog. | Goat 1903 on the 31st day. 25 flies dis- 4 Starved. sected; 5 found infected. 5-25 Fed on clean Goat 1903. 1950 1-8 33 flies fed on 7. pecorum- | Trypanosomes first appeared in blood of | infected goat. | Dog 1973 on the 28th day. All flies dis- 9 Starved. | sected; 6 found infected. 10-29 Fed on clean Dog 1973. It would appear from these five positive experiments that a period of from 19 to 53 days may elapse before the cycle of development of Trypanosoma pecorum in G. morsitans is complete and the fly becomes infective. Details of the Two Negatiwe Experiments. The following Table shows the method of procedure in carrying out the two negative experiments :— ey Diseases of Domestic Armmals in Nyasaland. 335) Table III. Expt. Dayne Procedure. | Remarks, expt. 1732 1-2 60 flies fed on 7. pecorum- | Dog 1736 never showed trypanosomes. infected dog. 25 flies remained alive; used for another 3 Starved. experiment. Only 12 flies dissected; all 4-39 Fed on clean Deg 1736. negative. (Experiment stopped.) : 2115 1-4 40 flies fed on 7’. pecorum- | Monkey 2066 never showed trypanosomes, infected rat. All flies dissected; 4 found infected. 5-6 Starved. Experiment stopped on account of death 7-8 Fed on clean Monkey 2066. of most of the flies. (Experiment stopped.) RESULT OF THE DISSECTION OF THE INFECTED FLIES. The following Table gives the result of the dissection of the infected flies found in the positive experiments. The second column gives the number of days between the first infected feed of the fly and its death and dissection :— Table [V.—Laboratory-bred Flies. Positive Experiments. | Time, : Proventri- i Fore- Mid- | Hind- | Salivar Bede days. EXONS: culus. CoP. gut. | gut. | gut. sland Rs ce stine : y 546 | 30 _— = - + + ar = 546 | 64 = _ - an oP ++ ap = 546 | 84 = = = ap ar ar ar aP oF = 546 | 84 ++ 3p ae ++ ++ ++ + — 524 | 27 _ - Peo - ++ ar or = 524 | 55 + + - ++ ++ t+ = Labial | Hypo- cavity. | pharynx. 17387 | 27 ++ ++ ++ ++ ++ ++ ++ - 1737 | 28 + = + - ++ ++ + | _ 1737 | 30 ++ ++ ++ _ ++ ++ ++ - 1853 | 17 ++ ++ ++ - ++ ++ ++ - 1853 22 — — - _ + ++ + - 1853 | 23 + = + - ++ ++ a7 ar = 1853 | 25 ++ ++ | + - ++ ++ ++ - 1853 | 26 ++ apr + = ++ ++ 45 ar = 1950 | 17 - - | + — +/+ ++ ++ — 1950 | 19 ++ +4 - - +4 ++ ++ — 1950 | 24 = = = = = + = = 1950 | 26 ++ ++ + - ++ ++ ++ - 1950 | 31 ++ ++ | + _ ++ ++ ++ - 1950 | 31 are ap or ar Sr = ap ar ++ 3 SF = In Experiments 546 and 524 there was no special examination of the hypopharynx ; it is included in the general term “ Proboscis.’ It was only iD 2 36 Sir D. Bruce and others. Trypanosome after the importance of the hypopharynx became evident that an examination of these separate parts of the proboscis was made. _ In Experiment 546 only one infective fly was found. In Experiment 524 two infected flies were found: in one of these the development was incom- plete, in the other complete. In 1737 two flies were infective, in 1853 three, and in 1950 four. In not a single fly was any invasion of the salivary glands noted. The following Table gives the result of the dissection of the infected flies in the negative experiments :— Table V.—Laboratory-bred Flies. Negative Experiments. i; % ee 7 7 ae Tay 1 Proboscis. | | | Time, |— Proventri- | Fore- Mid- Hind- | Salivar Expt | Crop. y 1 CEB | Labial | Hypo- | culus. P gut. gut. gut. glands. | cavity. ai p | 2115 | 9 = = pall le = = + = = 2115 | 9 - =a)“ vee - — ++ - - 2115 9 + | = | ar = + | ae ae ++ — 2115 11 ae = + = SP) ap ae + = In the negative Experiment 1732 all the flies were found to be negative. In Experiment 2115 four infected flies were found, but in none of these had the development reached the hypopharynx; none of them were infective. From a consideration of these tables it will be seen that 7. pecorwm belongs to the same group as 7. simew as regards its development in G. morsitans. This development takes place at first in the intestine, then passes forward into the labial cavity, and finally invades the hypopharynx and there is completed. THE Type of TRYPANOSOMES FOUND IN THE INFECTED FLIES. Plate 2 represents the developmental forms of 7. pecorwm in G. morsitans. In regard to the forms found in the intestine, it may be said that these are indistinguishable from the developmental forms of other pathogenic trypano- somes, and what was written in regard to 7. simic* is equally applicable to T. pecorum. Figs. 1 and 2 are forms from the proventriculus, and represent the dominant intestinal trypanosome forms passing forward to the labial cavity. Figs. 3-8 represent early forms found in the labial cavity. These were seen adhering singly by their flagella to the labrum. * ‘Roy. Soc. Proc.,’ B, vol. 87, p. 65 (1913). Ne bruce & others. et 27 ‘ 28 29 Trypanosoma pecorune Development trv Gloss morstases. ME. Brace, det. X 2000. Diseases of Domestic Animals in Nyasaland. 37 Figs. 9-11 are the ordinary forms found clinging by their flagellar ends to the labrum. It will be seen that they have assumed the crithidial stage, a stage which seems to be a sie qud non in the final stages of the cycle of development of all the pathogenic trypanosomes, and the interpretation of . which is still obscure. Figs. 12-19 are various forms other than “ blood forms” which have been squeezed out of the proboscis of a living infective fly. Fig. 15 appears to be encysted. Figs. 20-29 are “blood forms” from the hypopharynx of dead infective flies, and also from living flies induced to salivate on a cover-glass. They represent the final stage in the cycle of development and are the only infective forms. CONCLUSIONS. 1. That 7. pecorum is capable of passing through a cycle of development in G. morsitans, the flies becoming infective some 20 days after feeding on an infected animal. 2. That 7. pecorum belongs to the same group as 7. simie, the development taking place at first in the gut and afterwards passing forward into the labial cavity and finally into the hypopharynx. 3. That the final stage of the development only occurs in the hypopharynx, where the trypanosomes revert to the “ blood form” and become capable of setting up infection if injected under the skin of healthy animals. DESCRIPTION OF PLATE. (See also pp. 36 and 37 above.) Figs. 1 and 2, trypanosome forms from proventriculus. Figs. 3-8, early infection of the labrum; the flagellates still retain the trypanosome characteristics. Figs. 9-11, ordinary crithidial forms found adhering in masses to the labrum. Figs. 12-19, other forms from labial cavity. Figs. 20-29 represent the final stage of the development in the hypopharynx—the infective or ‘‘ blood form.” Stained Giemsa. x 2000. 38 Trypanosomes found in Wild Glossina morsitans and Wald Game on the * Fly- Belt” of the Upper Shiré Valley. By Surgeon-General Sir Davin Brucs, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 18, 1914.) INTRODUCTION. In June, 1913, one of the members of the Commission went to the Liwonde district to identify and isolate the various species of trypanosomes infecting the “ fly” and wild game in the “ fly-belt”’ which extends along the Upper Shiré River valley from Lake Pamalombe to the Murchison cataracts. This “fly-area” is, roughly speaking, 100 miles south of Kasu and the “Proclaimed Area.” It is separated from the extensive “ fly-area” of the plains on the west shore of Lake Nyasa, of which the “ Proclaimed Area” forms a part, by a range of hills and high plateaux where the “fly” is absent, although there is nothing to prevent trypanosome-infected wild animals wandering from one “ fly-belt” into the other. The valley of the Upper Shiré is thickly populated, and the “ fly-area” is erossed by two of the most frequented roads in Nyasaland, the grand trunk road running from Zomba to the north and the main road from Liwonde to Fort Johnston. Although thickly populated, human trypanosome disease, though probably existing, has not yet been discovered in this district.* The natives, however, can keep no cattle, and their goats and dogs are constantly destroyed by trypanosome diseases, so that they have to con- tinually import these animals from the highlands. Game is very abundant in this district, particularly in the dry season, when herds of eland, koodoo, waterbuck, and impala concentrate in the vicinity of the river. In the wet season elephant and buffalo wandering about the country frequently remain for many weeks in the impenetrable thickets and swampy “dambos” along the river banks. A characteristic feature of the flora of this district is the extensive forests of “sanya ” trees, open forests of medium-sized trees, devoid of undergrowth, but carpeted with short wiry grass. Large herds of impala are always to be found in these forests, and.tsetse flies are everywhere, being particularly numerous * Since this was written cases of trypanosome disease in man have been found. Trypanosomes found in Wild G. morsitans and Wild Game. 39 along the dusty tracks made by the antelope and around the pools where they drink. Nandumba’s village (14° 40° S. lat., 35° 10’ E. long.), on the banks of the Shiré River, was selected as the locality for the camp in which to carry out experiments of feeding flies on healthy dogs and goats, monkeys being unobtainable. The experiments were carried out between the dates of June 19 and July 25, 1913. Metuops EMPLOYED. The method employed in the feeding experiments was the same as described in a previous paper in the ‘ Proceedings,* except that monkeys were unobtainable, and the flies were fed only twice on each animal. All infected animals were subsequently taken to Kasu, the usual pre- cautions being taken to prevent re-infection on the way, and the trypano- somes found in them were compared with the species and strains of trypanosomes obtained from human beings, various animals, and the flies in the Proclaimed Sleeping Sickness Area. Special attention and study were devoted to the comparison of the strain of the trypanosome causing disease in man in Nyasaland—TZvrypanosoma brucei vel rhodesiense—from Nandumba’s, with strains obtained from human beings, various animals, and the tsetse flies in the Proclaimed Area. : The following Table gives in the first column the date the tsetse flies were first fed on the experimental animals, the second column the number of flies fed, and the signs plus and minus show the result of feeding the flies on the dog and goat. Table I.—Infectivity of Wild Glossina morsitans in the Liwonde District. Dog. Goat. No. of | _ | Pee i | Date. flies ag N | ass | gs | Cell Roa ekeS SANS S| iil ages > S'S Sy Ps SSS Se S Sis | & Ss Si Ss SS & | | | | | | (ee — - eee + + + 150 | + e - > =| = A = + 160 | + Bare hee ae + = + 450 + = =- |= | - + - + 650 | + Spee ele | a 650 + - |-|- — t+ = - UOT |, Sep eal se cts eae Pecan | ea pee (ea ae | | = ©Roy. Soe. Soc.,’ B, vol. 86, pp. 422 and 423 (1913). 40 Trypanosomes found im Wild G. morsitans and Wild Game. It will be seen that the “fly ” in the Upper Shiré district carries the same four species of trypanosomes as those found at Kasu in flies from the. Proclaimed Sleeping Sickness Area: TZ. brucei vel rhodesiense, T. pecorum, TL. sume, and 7. capre. Here, in a series of seven experiments, all the animals on which the flies. were fed developed trypanosome disease. In six experiments the flies. infected the dogs with 7. brucei vel rhodesiense ; in the second and fifth there was a double infection with 7. pecorwm; and in the seventh an infection with 7. pecorwm alone. None of the goats were infected with T. brucei. Six goats were infected with 7. pecorwm, one with 7. somiw, and five with 7. capre. It will be noticed that the smallest batch of flies used, . a batch of 73, infected the dog with 7. brucet, and the goat with 7. pecorwm, T. sumie, and T. capre. EXAMINATION OF THE BLOOD OF WILD ANIMALS IN THE LIWONDE DISTRICT. Whenever wild animals were killed their blood was examined for trypano- somes, which were identified by the microscope in, stained films of the blood. The following Table gives the results :— Table II. F ihe = ais =i wo \ Species of trypanosomes found. ‘Animal. | | mae tea is 7 7 | T. brucei vel | es 0 Ti | pitodleianee. T. pecorum. | T. simiz. . Capre. é Ug tase : ot J Feb) Vb aS Oana Hap soe BaD SOhEES = = = = = Then A, — sonooocaaca sso = = = = = saagiih, bean Shae cane - | = = = = siete = = — af a OES bhaenachnisenetamtays Fy COO ICCior ra = _ — — Pen IDE tee eee reac — + = = = PR) atindascebabaeencoc - + = a mY PR EEOOSICMOCOOOOn OOo a = _ — = IKoodoomnnvenccnrerterce - — = pe es ppl!) NgeodacoBesdno8s.ox0 _ — _ + pa Waterbuck ............... - _ = a pass BoudabooudecobG = 4P _ — = ” sodbo cp anoa0d = Gi _ — = ” 0 9) | laielnin\e\isinyoinin,n'nis or =_ — — + 99 tea eee ees cat = = _ | — Fait). “s boppdogoomDodad _ - — at | ES | This Table, even in so small a series of animals examined, indicates that T. pecorum occurs frequently in the wild game, such as the impala and The Food of G. morsitans. A} waterbuck. 7. brucei vel rhodesiense was found in only one animal out of the sixteen, 7’. simi@ in none, and 7. capre in three. CONCLUSIONS. 1. The trypanosomes found in the wild G. morsitans and wild game of the Upper Shiré “fly-area” are identical with those found 100 miles farther north in the Proclaimed Area. 2. The trypanosome causing disease in man in Nyasaland—Z. brucei vel rhodesiense—is frequently met with, so that it is probable cases of this form of sleeping sickness will be found among the natives of this district. The Food of Glossina morsitans. By Surgeon-General Sir Davip Bruce, C.B., F.R.S., A.M.S.; Major A. E. HAMERTON, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 18, 1914.) Five hundred flies, freshly caught in the Proclaimed Area, were killed by chloroform and the gut of each was roughly dissected out, smeared on a slide, fixed by osmic vapour and alcohol, and subsequently stained by Giemsa. The flies were all caught in the bush, away from the paths, the fly-boys proceeding in single file and catching the flies with gauze nets as they circled round, or settled on the boys or the grass. The proportion of male flies to females caught was roughly two to one. But only 30 females were used in the present experiment, the majority being sent to the breeding-station at Chunzi. Of the 500 flies examined, 288, or 57°6 per cent., were found to contain mammalian blood in a recognisable state. No measurements were made of corpuscles, which in most cases were much altered by the digestive processes, but the small type of cell appeared to predominate, such as occurs in the hartebeeste, waterbuck, and other antelope. In only three cases were nucleated red corpuscles found, and in two of these there was only a small proportion of nucleated blood mixed with a large amount of mammalian. In the third case the blood was all nucleated. Thus, of those flies which contained recognisable blood, only 1:0 per cent. AQ The Food of G. morsitans. contained nucleated blood. From measurements, it seems highly probable that in all three cases the blood was avian, not reptilian. The average length of corpuscles and nuclei of blood from several different reptiles was measured and found to be—corpuscles 15 microns, nucleus 5-9 microns; while the blood of several different birds gave as the average—corpuscles 11:8 microns, nucleus 4°6 microns. In the three cases under consideration the average of the corpuscles was 10°5, 10:0, and 10:0 microns respectively, and that of the nuclei 47, 48, and 44 microns. Probably the size of the nucleus is the better guide than that of the whole corpuscle, as being less altered by digestion. In no case was vegetable matter noted in the intestinal contents. Trypanosomes were found in 14 flies—2°8 per cent.—but many of the smears were so thick and so much obscured by the fat-body and other structures of the fly, that probably trypanosomes were present in other cases. Of the 30 female flies examined, 13, or 43°35 per cent., contained mammalian blood, and there was nothing to suggest that they differed in their feeding habits from the males. From experiments with flies in the laboratory, it was found that blood is recognisable in stained specimens for two to three days after a feed, but not beyond the third day. Hence it may be inferred that, roughly, half the flies examined had fed within, at most, three days of their capture, and that therefore the flies feed naturally at least once every six days. Conclusions. 1. The food of Glossina morsitans consists mainly of mammalian blood (99 per cent.), chiefly from species of antelope, and what appeared to be avian blood (1 per cent.). 2. There is no difference in the feeding habits of males and females. 3. Probably the flies feed once in five or six days. SN ee 43 Infectivity of Glossina morsitans in Nyasaland during 1912 und UG) Lei. By Surgeon-General Sir Davin Brucg, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 18, 1914.) INTRODUCTION. The object of this paper is to attempt to set up a rough standard of the proportion of infected to non-infected tsetse flies in an ordinary “ fly-area ” where wild game abounds. It is thought that a standard of this kind may prove useful in the future. The flies were collected in the low country lying near the Commission’s camp at Kasu, in what is known as the “ Proclaimed” or Sleeping-Sickness Area of Nyasaland. This bit of country swarms with Glossina morsitans and wild game, the latter highly infected and well protected. In 1912 a total of 1975 flies were dissected between the months of January and November. Of these 129 were found to be infected with trypanosomes—6'53 per cent. Males, 86 per cent:; females, 14 per cent. In 1913, 1060 flies were dissected, of which 91 were infected— 8°58 per cent. The following Tables give the details :— Table I.—1912. No. of fly. | Proboscis. EOE | Crop. Intestine. Salivary | Part of fly injected. | faunal Result. culus. s glands. | injected. il | ott ++ ++ ++ | | 2 = - — + _ Intestine | Dog = 3 me =a at Ganats | » ” = 4 — + - a Goat | - 5 - ++ A Dog | = 6 — ++ ¥ Goat = 7 - ++ ++ — Shea oe + ++ D | Dog = 9 ++ - = Proboscis | Goat = 10 + + 47 oF = 26 KW as = 11 = + 12 = + = | 13 | = ++ - Tntestine | Dog - 14 } ae sh ie fh base | WG je - ++ = | 16 | — ++ — 17 + _ = - - 18 - ++ = 19 a> = | | 44 Sir D. Bruce and others. Infectivity of . Table I.—1912—continued. No. of fly. | Proboscis. ea Crop. Intestine. ae Part of fly injected. guacee Result. 20 => = ++ = 21 + 22 + + = 23 _ + — 24 GP oP ap Gr ap 4P = 25 +4 — 26 _ =_ ++ _ 27 ++ ++ + +4 _ 28 + = ++ = | 29 — ++ _ Intestine and sali- | Dog = vary glands 30 — — - + Tntestine 2) — 31 _ + = Qs ” ” 22 ah = oF { Proboscis p - 33 det + 34 | — ++ - Intestine Goat — 35 | _ ++ An Dog - 36 | - ++ - 5 Goat = Si, | = ++ fe de = 5 and _sali- a = | vary glands 38 = + Intestine oF = 39 | + — Proboscis 20 _ Intestine 2 oS ay b i { Proboscis Dog — Al — + - Intestine of) = 42 + — | Proboscis Goat — 43 | - ++ Intestine 5 =— 44 | + - Proboscis 9 = 45 — + Intestine @., - 46 = ar ” Dog oe; AT — Proboscis Goat — 48 es fare | Intestine Dog = ” ” 49 ane ve Te { Proboscis 59 - 50 of a { Intestine 5) _ Proboscis Bs = 51 — - ++ | — Intestine Goat _ 52 + + 5 Dog = Be a 3 | { Proboscis ” = 54 - — — + ! — Intestine 39 = 55 + _ Proboscis Goat - 56 a + | — Intestine 9 = 57 _ + | 58 — dap | — » » a 59 = ee 60 — + | 1) ” oA 61 + = ” ” ad 62 + + 63 _ ++ + = 64 + | = 65 — + = 66 | + ote a 67 | + — 68 + = | 69 + ++ = 70 = = ++ — | 71 | + - | = 72 | = fe he ++ = Intestine Monkey — 73 + ++ ++ ++ = sie o Pe G. morsitans 7n Nyasaland during 1912 and 1913. 45 Table I—1912—continued. No. of fly. | Proboscis. ean Crop. Intestine. ea | Part of fly injected. | eee Result. - | a ee a SM 7A = ++ 75 = - + ++ = 76 + = 77 ++ ++ — Proboscis Monkey - 78 + + 79 + ++ - 80 + = 81 - = ++ - 82 + aP ae = 83 + = 84 = ++ ap oar — 85 - — — ++ 86 = de + es | 87 + = - | | 88 + — 89 — Siete 90 = + — 91 — + —_ 92 + = 93 — ++ 94 +t ++ ++ ap 45 = 95 = ear 96 = + + — 97 ap 4p = 98 | + - 99 | se 100 a | + a 101 — jt fe = 102 ++ ++ — 103 = ++ ++ - | 104 ++ ++ ++ = | 105 = ss 106 + ++ 107 fo fh ++ - Intestine Dog = 108 + + + + = 109 + ++ ++ i and pro- | Goat + boscis T. simie 110 ap on = = = Proboscis i — 111 + + + ap Ae ae ar 55 and intes-| Dog - | tine 112 — ++ | - Intestine aa = 113 ao qe SE | — | a Me — 114. = ++ t+ | 115 cs te | 116 _ — te = 117 = eer | 118 + ap ar ++ ah SP = Intestine and _ sali- 50 — vary glands 119 = ++ Intestine a — 120 + eis wal n ay = aaa ig 9 te a 122 = ++ — + | — 128 ate Se Sita I a » and: = goat — 124 + ++ ++ —_ 125 = + Intestine Dog 32 126 = = a aF or = » Goat a 127 + + + = 128 = + = andr 4 ar Salivary glands Rat + T. brucei 129 = + _ ++ ++ Ms - ae | | T. brucei 46 Sir D. Bruce and others. Infectivity of It will be seen from the above table that 60 attempts to determine the infectivity of the flies were made by injecting emulsions of the infected organs into healthy animals. In only three cases did the animals become infected: once with Yrypanosoma simiw and twice with TZ. brucei vel rhodesiense. The usual experiment was to inject the contents of the intestine into dogs or goats, which is known now to be useless, as the developmental forms in the intestine are not infective. Doubtless more positive results could be got at present with more knowledge of the laws which govern infectivity. Only in two cases were the salivary glands found to be invaded. This infection, of course, could only be Z. brucec vel rhodesiense, and this was confirmed by injecting the glands into rats. In 1912 no attempt was made to diagnose directly the species of trypano- somes with which the flies were infected, but in 1913 this was done, as by that time a good deal of experience had been gained. For example, invasion of the salivary glands could only be 7. brucei vel rhodesiense ; invasion of the intestine, labial cavity and hypopharynx meant 7. pecorum or T. simiw, and size would distinguish between the two. Lastly, if only the labial cavity and hypopharynx were seen to contain flagellates, then 7. capre was indicated, and here also the size and character of the trypanosomes in the hypopharynx would assist in the diagnosis. Table II.—1913. Proboscis. ae 4 i Z : alivar ecies of Novot fy: Tnbassine: pisses teypadecomel Labial cavity. | Hypopharynx. 1 + + T. pecorum. 2 ere — | DT. capre. 3 + a | FF) 4 cr aie | ) 5 i a | 7. simie. Gi A — - T. capre. a - + + - 8 = = + 9 + + | T. simie. 10 ap = T. capre. il + _ 9 12 - ++ — 13 = + - 14 _ + = 15 + ++ T. simie. 16 * + i 17 + T. capre. 18 a + + = T. pecorum. 19 + + 20 + + a 21 — - ++ ar T. brucei, 22 = + - | 23 + = T. capre. | G. morsitans in Nyasaland during 1912 and 1913. 47 Table II.—1913—continued. No. of fly. Proboscis. Labial cavity. Hypopharynx. Intestine. Salivary glands. Species of | trypanosome. | I] Il sp + 11++t p+teti +] lar ll Warf ee ee ee | + + a se ee Se) ee ee a ee || |} ++ ] ae apse i) Spar arse || | se Il ar | + +++ | | LT. capre. T. pecorum. T. simie. | ”? T. pecorum. T. simie. ” | T. capre. | tr} T. simié. 32 T. capre. T. pecorum. T. capre. | LT. simie. 48 Infectity of G. morsitans in Nyasaland during 1912-13. Table I1.—1913—continued. Proboscis. | Salivary Species of glands. trypanosome. No. of fly. aw Intestine. Labial cavity. | Hypopharynx. | T. pecorum. | | T. simie. ee | [| ak ap ap ae [] ab ae ae Ht wey yak | aeae ar (| i} il ap ar ar || Ho ae th In 1913 no injections of the contents of organs were made into healthy animals. The direct diagnosis of the species of trypanosomes by examination of the fly took the place of inoculation. From the above table it will be seen that in 1060 flies 7. brucei vel rhodestense was found once, 7. pecorwm six times, 7. simi 12 times, and T. capre 14 times. It must, however, be confessed that the margin of error in this calculation may be large. CONCLUSION. In 1912, 6°53 per cent. of the G. morsitans found in the “ Proclaimed ” or Sleeping-Sickness Area, Nyasaland, were infected with pathogenic trypanosomes ; in 1913, 8:58 per cent. 49 The Various Inclinations of the Electrical Axis of the Human Heart. Part Ia.—The Normal Heart: Effects of Respiration.* By Aucustus D. WALLER, M.D., F.R.S. (Received March 30,—Read May 14, 1914.) (PLATES 3-6.) In my first account of the electrical action of the human heart,t I made no allusion to the influence upon the electrical pulse of the movements of respiration. I noticed that influence indeed which is especially well marked in my own case (where the heart happens to be of the complete horizontal type) but only as disturbing the demonstration, and in some cases rendering the direction of the pulse uncertain. I noticed in particular that when demonstrating the transverse effect from the two hands, best effects were shown by holding my breath in expiration, and that these effects were markedly diminished during deep inspiration. I imagined at that time that the effect was due to a disfavouring of current spread from the heart by reason of the distended lung, but was puzzled by the fact that with the axial lead (right hand and left foot) the electrical pulse was augmented during inspiration instead of diminished as was the case with the transverse lead. I did not, however, follow up the clue afforded by this discrepancy, and it was only much later, 2.e. after the introduction by Einthoven of the string- galvanometer and the observations of Einthoven, Kraus and Nicolai, Samojloff and others, that the meaning of the discrepancy and with it the whole mechanism of the respiratory effects became clear. The variations of ampli- tude are, if not entirely, almost entirely due to the rise and fall of the diaphragm, raising and lowering the heart as a lever hinged at the aortic end and thus widening and narrowing the “axial angle.” . (By axial angle I mean the angle formed with the vertical by the current axis of the heart or line of greatest potential difference at right angles to the equator OO.) In 1889 I represented this angle as being 45° to the left (and 45° to the right in cases of situs viscerwm inversus), and drew two series of curved * The method followed was the same as that followed for the observations of Part I. A Bock-Thoma galvanometer was used for these observations, the deflections being standardised whenever necessary by a millivolt switched into the circuit. For the careful comparison of values in different leads or between the values obtained at different times, a standard millivolt deflection was of course taken, and the proportional correction applied, if necessary. But in comparisons taken between the two sides of the body the standard deflection was found to be so invariable that it was often omitted. . t+ “On the Electromotive Changes connected with the Beat of the Mammalian Heart, and of the Human Heart in particular,” ‘Phil. Trans.,’ B, vol. 80, p. 169 (1889). VOL. LXXXVIII.—B. E 50 Dr. A. D. Waller. Various Inclinations of the equipotential lines at opposite ends of the heart. To-day I represent the angle as 30° to the left (it can range in healthy subjects between 10° to the right and 100° to the left), and draw straight equipotential lines, parallel with the equator, at right angles to the current-axis CC (fig.1). To-day as in 1889 I regard as essential the distinction between “strong” leads (left superior \< “6 SS A Be J x YO ARS XS es ARS ASAES volt 1000 (Seaes XS < . \ Ed ay as “a cS es co OO is the equeor, CC the current-axis. Equipotential lines drawn parallel to the equator, and at right angles to the current-axis at regular intervals of 1/10000th volt. a indicates the “axial angle” formed with the vertical MF by the current-axis CC. tan a= ati where R and L are the electromotive values of the Right and Left inferior ventricular spikes. ML, and right inferior RF) and “weak” leads (right superior MR, and left inferior LF) (see figs.). The present diagram (fig. 1) is drawn so as to exhibit the relative values of potential difference between the led-off points R, M, L, F, by the number of equipotential lines which they include. Each unit = 1x 104 volt. Thus eg. in fig. 1 it is to be seen that (approximately) MR=1°5; ML=5°5; RL= 40; RF = 9:0; LF = 5:0 ten-thousandths of a volt. At pages 518-9* of the communication to which the present paper belongs it is stated that a full consideration of the effects of respiration upon the (amplitude of the) electrocardiogram is a necessary preliminary to the due understanding of the physiological and pathological departures from the * “Roy. Soc. Proc.,’ B, vol. 86 (1913). Electrical Axis of the Human Heart. 51 normal type. The consideration of the effects of respiration appeared to me at that time as a side-issue to be cleared up by a few carefully planned observations, whereas it now presents itself as an extremely simple exercise in elementary trigonometry on the main line of the principal argument. The essential distinction between favourable or strong leads (left superior and right inferior) and unfavourable or “ weak” leads (right superior and left inferior) which was the principal result of my first investigation of the subject, afford, when their data are converted into simple ratios, sinusoidal curves which plotted upon millimetre paper exhibit directly the quantitative relations between the electrical effects of the heart, whether horizontal or vertical (as described in Part I) or at various inclinations in accordance with respiratory alterations of the cardiac axis. The facts will be best presented by a detailed account of two typical cases. The Case of B. O. B.—An Oblique Heart. Fig. 2 (Plate 3) gives the transverse, right inferior and left inferior records taken simultaneously with the record of respiration slow and deep so as to emphasise the effects upon the heart, which in this subject had been determined as having an electrical axis forming an angle not exceeding 30° with the vertical during quiet breathing. The amplitude of the spike in the transverse record varies between a maximum of 20 mm. in expiration and a minimum of 15 mm. in inspiration. The waxing and waning is very regular and at once suggests a sinusoidal curve. In the right inferior record the respiratory variation of amplitude is much less pronounced and regular within a range that may be taken as 27°5 mm. in expiration and 24 mm. in inspiration. In the left inferior record the variation is regular and large and suggestive of a sinusoidal curve. The range is measured as between the values = 6 mm. immediately after the culmination of expiration, and = 14 mm. during inspiration. (In these three records the deflection by 1/1000 volt through body and galvanometer was 18 mm.) From these values right and left below the heart we calculate that :— 27:5—6 The expiratory tan« = 27546 —=— NS 4 ee OA The inspiratory tana = 2 aot = 058, 2 3 = WS. Diff = 24°. In the right superior record the respiratory variation is small and fairly regular between values that may on the average be taken as = 5 mm. in expiration and = 7:5 mm. in inspiration. E 2 52 Dr. A. D, Waller. Various Inclinations of the In the left superior record the variation is large and fairly regular between values = 26 mm. in expiration and = 18 mm. in inspiration. The right and left measurements give for the superior angle a above the heart the following values :— In expiration— = 26—5 =e 21 Sys : — AUS Re = Fe = a == (ae, a ey Se Tn inspiration— . tana = US ria ek EU = (és 2 2 = Be”. Ditty pie From these results we see that in the case of an oblique heart the effect of inspiration is to weaken the strong leads and to strengthen the weak leads. But this rule, while affording a useful mnemonic, is applicable only to oblique and vertical hearts.* The inspiratory diminution of the superior angle has in this case come out as 12°, as compared with 24° for the inferior angle, This is as might be expected from the fact that the basal attachment of the heart cannot be rotated to the same extent as its apical free portion, Obviously we cannot expect that the angle calculated above the heart from the three points M, R, L, should necessarily be identical with that calculated below the heart from the three points F, R, L. (The records of this case incidentally afford a striking example of the variations of pulse-frequency that are associated with the two phases of respiration, sometimes in the human subject, always in the dog. I have discussed them elsewhere under the title “Dog Pulse.”+ Their consideration does not belong to this portion of the subject.) The Case of A. D. W.—A Heart of the Horizontal Type. Fig. 3 gives the transverse, right inferior, and left inferior records taken simultaneously with a record of respiration, slow and deep so as to emphasise the effects upon the heart, which in this subject had been determined as having an approximately horizontal electrical axis. The spikes of the weak leads (right superior and left inferior) are accordingly reversed, as was explained in Part I. In the transverse records the respiratory variations of the spike are regular and large, between values = 15 mm. in expiration and = 10 mm. in inspira- tion. It may be noticed in this record that inspiratory diminution goes on * I speak of hearts as “vertical,” “oblique,” and “horizontal,” according as the axial angle a is between 0° and 30’, 30° and 60°, 60° and 90°. The contrasted types are vertical and horizontal ; oblique hearts belong to the vertical type. tT ‘Physiol. Soc. Proe.,’ June 28, 1913. = OF Roy. Soc. Proc., B, vol. 88, Plate 3. Waller. “WBIDOIPIVIOIZOOTA OT} YI ATSHoowEy[NUMIS pepdooad ore WOTyeaIdsear Jo seseyd oy, ‘ToHeudxe SuLINp UBy UoyeIIdsUL SULIMp 1048018 ST (S}op UeeAyoq ouTT UTGy ey Aq UMOYsS se) osvo styq ur Aouenbe.y-ospnd oxy, ‘uoryerdsur Sump peseedour ATWOUTYSIP SI 41 (eUT, LeMoT) pea, Lorafue afa) ay) UT TOMBAIdSUT SULIMp PeYysTUTUIp Toy ayy Wo st 41 (OUT e[pprur) peey wowalue yybrs ayy UT ‘uoreadsur SuLInp peysrurup ATOUTZSIp Si 1A oeyxids repnorsqUEA ayy (eur, aeddn) prox asvaasun.g ayy UT ‘spve] Jonafur pfo7 pure sorafur ry br osveasun OY4 UI (Aouenbez-ospnd oy} uodn puv) wea«sorpreoosqoeye 049 uodn uoeridset Jo yooyy -gavoy onbyqouy ‘gO ‘q 40 aSVO HHL —s ‘DIT ae : —— qa SS Fe oS aed seep REE HPT { I aiule aSdamin, 1 1 1 '/i000 VOLT - "sgn dana P Roy. Soc. Proc., B, vol. 88, Plate 4. Waller. ‘uoeidsur SuLmMp posvertout ATQYSITS st (pxooed ostoasmery oY} UL ATUO poyeorpur) Aouonba.azj-ospnd oxy, ‘uorqeaidsut Sutanp poeysturarp st eytds oarqesou sty} pure ‘oatesou st oytds repnorqUeAa 949 (@UIT TaMoT) pwoy Lorafue 2fa7 949 UT ‘TO1ZVIdsuUl SULINP pesvo.1OUL ST 41 (UIT 9[PprIut) peo] Lorlafue gyre og UT ‘uoryedidsul SULIMp poeysturatp ATouTysIp st oytds TeMorsyUeA oy] (eur, 1eddn) poy aswaasunuy oY UT ‘SPBOT LOLLOJUT 4FoT PU LOTAIOFUI YYSLA ‘ostoAsuR1) Oy UL (AoTONbIAT -osjnd oy} uodn pur) wmersorpazes0ajooto oy} uodn uoryerIdser Jo yoo “yaevoy [ejyUOZTIoOY VW “M ‘@ “VW #0 SVD GH] —'¢ ‘Ply E 1 i 1EGS i I I iy 1oxe 1 i : | ee j Eee ee 0 ‘dXa ‘A 100-0 ES oe ee ee [exiess | Wea peo ERE STARTER py oe | ss Te bh “4X3 Hibs ies ws Gas — a > . t rn i 7 1 s ' 7 1 i, 1 r - o t ' Th - t y - + s x ? = 5 ea es = s Electrical Axis of the Human Heart. 53 into the beginning of expiration, and that the expiratory augmentation goes on into the beginning of inspiration. This lag is probably mechanical, the pneumograph begins to indicate sooner than the heart begins to be raised and lowered. ‘The effects are similar to those seen with the vertical heart, viz., inspiratory diminution, expiratory augmentation. In the right inferior record the respiratory variations of amplitude are regular and relatively large, 7.c. in relation to the absolute magnitude of the spike, which in the horizontal heart is small. The range is approximately between 6 mm. in expiration and 9 mm. in inspiration. The left inferior record exhibits the negative spike characteristic of the horizontal or soft heart, and with respiration well marked augmentation during expiration and diminution during inspiration. The values as read on this record reach to —13 mm. at the end of expirationand to —6 mm, at the end of inspiration. The angles as calculated from these right- and left-hand values are :— In expiration— pH Os 4s ; Menino LN pes eas mi? ee oe = 95> In inspiration— Fane 2 Ore a == OO 6+6 i The values for the right and le/t superior leads of this subject (of which tracings are not reproduced here) were as follows :— _ Right superior. | Left superior. tana. a. | ° mm, mm. In expiration ......... —5°0 7°5 5:0 79 In inspiration......... —2°5 10 ‘0 1°67 59 Reviewing the results of this case of horizontal heart as to the general effects of respiration in comparison with the results obtained for the oblique or vertical heart, it appears that these effects with inspiration are as follows :— B. O. B. A. D. W. Vertical (and oblique). Horizontal. Transverse spike ...... Decrease from 20°0 to 15°0 mm. | Decrease from 15°0 to 10°0 mm. Right inferior spike 5 5 20 » PAD — \hormro =, OO , OD | Left H 3 Increase ,, 6°0 ,, 14°0 ,, | Decrease ,,—-13°0 ,, —6°0 ,, Right superior ,, 7 Se LOMO ClkiOniacn 26 SOOM suai) Is Left vi 5 Decrease ,, 26°0 ,, 18°0 ,, |Increase ,, (i OnmelOk oles, 54 Dr. A. D. Waller. Various Inclinations of the - There is at first sight a flagrant discrepancy between the effects of respira- tion upon the two types of heart. The only case in which the effect is similar in both is that of the transverse lead. The provisional rule framed above for the vertical heart — inspiratory weakening of strong leads and strengthening of weak leads —is evidently inapplicable to the horizontal heart. The cases of the right superior and left inferior (decreased negative spike) are indeed amenable to it, since decreased negative as well as increased positive amount to electrical strengthening. But the cases of the left superior and right inferior, where after some uncertainty it became clear that a positive spike is increased for the horizontal, decreased for the vertical heart, are in apparent contradiction of the rule. Graphically expressed the angles in the two subjects B. O. B. and A. D. W. in inspiration and in expiration are as follows :— F F B.O.B. A.D.W Fie. 4.—Diagrams to indicate the value of the axial angle a in inspiration and in expiration of the subjects B. O. B. and A. D. W., calculated from the electromotive values of the right and left superior spikes and of the right and left inferior spikes. B.0.B A.D. W tan a. Gs || tan a. a. | E ° ° ; 18—7°5 _ 10°5 10+2°5 12-5 CEnepeeen ut a =0-41 | 22 aye |) 50 SapE anep 18475 25°5 10255) ED 56s ol 7-645 1255 Nex po ae -2! ~0-+68 | 34 2 BAe 25 79 SDE CIP 2645 31 75=5 0 255 Tap ansps eo 7) eee Day | ag | 22S 20 90 24414 38 6+6 275-6 43 9+13 44 ir exe ae ay | gp | 2 spade = aca lnos BE LexD 27546 33°3 9—13 = But the general relations between the varying “strengths” of leads in different hearts and in different phases of respiration will become most clear Electrical Axis of the Human Heart. 55 from a consideration of the simple trigonometrical ratios of a varying axial angle. The transverse lead is the simplest to consider, and its simplest case is that of the horizontal heart in which the axial angle during quiet respiration is 90°. In this position, regarding the heart as forming (electrically) a horizontal lever, it is evident that with this type of heart the transverse electrical effect has its maximum value, and that the effect must be diminished by either rise or fall of the diaphragm; the diminution will be proportional to the sine of the altered axial angle. Taking as unity or | 10 this maximal value with the chest at rest, we shall have the altered values = 9°8, 9-4, 8°7, 7°7, 6:4, 5:0, if with fall of the diaphragm the angle 90° is changed to 80°, 70°, 60°, 50°, 40°, 30°. Thus the theoretical alterations of magnitude in the transverse lead with alterations of the axial angle must be in ratio with the numerical values of the sines of the altered angle. And as the reference curve to which to com- pare our observations we have the simple curve of sines :— 0 1736 3420 5000 6428 7660 8660 9397 9848 10000 Ogee Om 20 9 305) A0P 5077 60 — 70" 7. 80°» 80° The superior leads, right and left, come next in order of simplicity. The former (mouth and right hand), as stated in 1889, is “weak,” being most nearly in accordance of direction with the direction of the equator; the latter is “ strong,” being most nearly in accordance of direction with the direction of the current-axis. The angle RML (fig. 1) is taken as being = 90°. In the case of the horizontal heart with the axial angle = 90° it is evident that the electrical effects along MR, ML, are equal and opposite, with values = + RL(cos 45°)? = + 50. With the axial angle altered + 10° by respira- tion the effects become :— Along MR = 100 cos (45° + 10°) x cos 45° = 40, and along ML = 100 cos (45°—10°) x cos 45° = 58. With a perfectly vertical heart, 7.c. with an axial angle = 0°, the effects are again equal, viz., 100(cos 45°)? = 50. The theoretical alterations of magnitude of the superior leads with alterations of the axial angle must be in numerical ratio with the cosines of the altered angle. We have as the reference curve to which to com- pare our observations, a simple sine curve formed by the tabular values of cosines multiplied by a constant factor to correct for projection between RL and the two sides MR, ML—in this case multiplied by cos 45°. The electromotive values for the superior leads for values of the axial angle from 0° to 90° are thus :— 56 Dr. A. D. Waller. Various Inclinations of the Left superior Right superior cos 45° x cos (45°—a). cos 45° x cos (45° + a). fe) 0) 5000 5090 10 5793 4056 20 6409 2988 30 6829 1830 40 7014 0616 45 7071 0000 50 7014 — 0616 60 6829 : — 1830 70 6409 — 2988 80 5793 — 4056 90 5000 —5000 Tt will be noticed that the right-hand values steadily diminish as the axial angle increases from 0° to 90° from their maximum of 50 to their minimum of —50, passing through the value 0 when «= 45; that the left-hand values between « = 0° and « = 90° increase from 50 to a maximum of 71 (7071) at 45° and then decrease again to 50. The right-hand lead is “weak,” the left-hand lead is “strong.” In the strong lead the first ventricular wave Vy; is positive in both types of heart—vertical and horizontal. It is increased by inspiration in the horizontal heart, decreased ‘by inspiration in the vertical heart. In the formule given above, for the calculation of the numerical values of right and left superior leads, we have taken the vertical angle at M = 90°, so that the semi-vertical angle M/2= 45° [and that the formula for calculating « from known values of R and Lis: tan a = (L—R)/(L + B)]. Generalising for any value of M the formule for the superior leads become :— For left-hand values— cos + x cos (5- 4) ; For right-hand values— —_cos e X Gos (S+ “| ; and for calculation of « from known values of R and L, M L—R t = cot — . ——.. an « = CO eae The right and left inferior leads (right hand and either foot, left hand and either foot) are at first sight somewhat less simple, but they are readily simplified. As was shown in my first observations of 1889, the two feet are practically isoelectric, and we may therefore regard as being electrically identical the two right inferior leads (axial and right lateral) and the two left inferior leads (equatorial and left lateral). The two feet Electrical Axis of the Human Heart. 57 are thus taken to be represented by a single point F. The right-hand lead RF, being most parallel to the normally oblique current-axis, is the strong lead; the left-hand lead, being least parallel to the current-axis, is the weak lead. The general formule for the inferior leads are :— For right-hand values— x cos (F/2—2), cos = cos F/2 1 r left- — = 2 For left hand values cos F/2 x cos(F/2+ 2), and for calculation of « from known values of R and L ee) a2 ony eel D TRaeT oS In Part I the angle F has been taken as 53°, so that F/2 = 26°5° and tan a = 2 os The values of R and L at different values of « are now— : ; if h alae a Ol For the right side Gos 9659 * 08 (26°5°—«), 1 50 For the left side— cos D653 * 08 (26°5° + a). The results come out as follows :— Right inferior | Left inferior | cos (26 *5°—a)/cos 26 “5°. cos (26 °5° + a)/cos 26 *5°. | ° ) . 1000 1000 10 1071 898 20 1110 769 30 1115 617 40 1086 445 50 1024. 261 60 932 68 70 811 —126 80 665 —317 90 500 —500 Similarly, we may work out the values of the leads which have been assumed above as identical, ze. right lateral and axial, left lateral and equatorial. But it would be tedious and unnecessary to give this in detail, since, as will presently be seen, the results are most easily and quickly obtained by a geometrical model, which gives immediately the results that have been considered up to this point. This model is intended also to render evident the meaning of the apparent discrepancies between vertical and horizontal hearts as regards the effects of respiration, and to give 58 Dr. A. D. Waller. Various Inclinations of the geometrically the theoretical values of the first ventricular spike Vy in all leads at all values of the axial anele «. The quadrilateral figure (fig. 5) RMLF, in which CM = 4(LR) and LR = CF, so that cot 4 the vertical angle at M = 1, and cot} the vertical angle at F = 2, represents at its points the leading-off points mouth, hands, and feet. It is pinned by its centre to the centre of a field ruled in parallel (equi- potential) lines; each division in this figure represents 1 x 10~* volt, so that 10 divisions are equivalent to 1 millivolt. Lines at right angles to the equi- potential lines (not drawn) would represent lines of force; the arrow CC through the centre of the field represents the current-axis, the line OO at right angles to CC represents the equator. The figure and the field can be rotated in relation to each other round the centre C, so that the current-axis CC is placed at any given angle from the vertical. The figure being weighted so as to remain vertical, the field when tilted gives any desired inclination of the current-axis, and the points R, M,L, F, oceupy positions upon the field that indicate directly the potential differences between them, 7.¢., in the different leads. Thus, ¢.., if the current- axis 1s inclined 30° the values in relation to O of the four points R, M, L, F, will be respectively + 2°5,-+ 4:7, — 2°5 and — 8°6 above and below zero, and the potential differences in the various leads will be given by the differences of level between various pairs of points, as follows :— Current-axis = 30° with vertical (as in fig. 1). MR (right superior) ......... 43—25 = 18x 10~° volt MIL (left superior) ...-5..2--.- 43+25= 68 , is by (HEIR VEHSE)) 44 -ceodnoe A068 2 Dt Oo Oe RF (right inferior)............ SO4E25 = ibl IGE Gettamierion) mas ae S020) — a Olan MF (longitudinal) ............ 86-43 = 1297; Obviously as regards the two sides of the body, MR and LF are relatively “weak” leads, ML and RF are relatively “strong” leads. Similarly the potential difference in the several leads can be determined for any value of « by pricking off the positions of the points M, R, L, F, and measuring off the differences of level between pairs of points. It will be realised at once that with high values of « the direction of the weak leads is reversed— e.g. at 80° the right superior potential difference is seen to be —40 and the left inferior potential difference —32. At 90° both weak leads = —50 and both the strong leads = +50. . Plotted out upon squared paper the values thus obtained give the curves represented in fig. 6, which are obviously sinusoidal curves, most obviously [To face p. 58.] ‘GOA YIOOOOT/T Jo Sdoqs quoseadaa yoru soury [eyue30d mba jo punorsyorq oy} uodn yo poxyorad oq weo vg ‘Sy jo “A “T “Yy ‘ Sputod oy} Jo uotisod oY} 919U00 SIq? puNor poqzeqor WY AA ‘SaInsy og oy Jo oaqueo oy} Ysnoryy uid vw Aq g ‘sy uodn posodsodns puw Yo yno eq prnoys vg ‘Ss OL, "4x04 04} UI poqiiosep sv ‘p o[Suv [wIXe OY} Jo Son[VA SUTATBA qu ‘A oxtds aepNoTAyWAA oY} JO SoNTeA 9ATZOIOTZeTO OYY JO WOTYe[NoTVo oTYVULOJNY ot} OF [OpOU-WIBASeI—'Vg PUB G ‘SPI ‘VG ‘DI — 4JOA . tAC'S Pines, Electrical Axis of the Human Heart. peste ce ale oe —_—_> Fia¢.6.—Curves giving the electromotive values of the several heart leads with values of the axial angle a from 0° to 90°. The value of, e.g. the transverse lead in- creases with increase of a; it decreases with inspiratory decrease of a. The value of, e.g. the left inferior lead decreases with increase of a and becomes nega- tive when a is above 64°. The negative left inferior spike of a horizontal heart is diminished by a deep inspiration and can sometimes be reversed to positive. The posi- tive left inferior spike of an oblique heart is increased by inspira- tion and can sometimes be reversed to negative by a maximum effort of expiration. See figs. 7 and 9. 60 Dr. A. D. Waller. Various Inclinations of the so for the transverse values, which can be directly read off from a sine table; for the other leads the curves can be verified by working out their appropriate formule. Fig. 6 is useful as showing at a glance the theoretical values to which the observed values must approximate in the several leads with hearts of various inclinations, and as regards the present paper affords the readiest mode of explanation of the otherwise somewhat perplexing effects of respiration upon the amplitude of the electro-cardiogram. — Inspira- tion by reason of descent of the diaphragm and elevation of the ribs causes a clockwise rotation of the electrical axis of the heart, z.c.a diminution of the angle « above and below the heart. Reading the appropriate line on fig. 6 from right to left, 7.c.in the direction of inspiratory decrease of «, we see by its rise or fall whether and how much increase or decrease of the ventricular spike is to be anticipated. Thus the transverse must be decreased with inspiration (or increased with expiration); the longitudinal, the right superior, and the left inferior increased. And as regards the troublesome case of the two strong leads we realise at a glance how ‘it happens that at high values of a, 7c. with a “horizontal” axis, we find inspiratory increase of the strong leads; whereas at low values of a, ze. with an axis at less than 30°, we find inspiratory decrease. This figure also supplies an explanation of the following two experiments, one or other of which can nearly always be repeated with success upon a heart of which the electrical axis happens to be of suitable inclination :— Experiment 1.—With an approximately horizontal heart the normally negative left inferior spike may be abolished and rendered positive during maximal inspiration. An example of this experiment is given in fig. 7. Experiment 2—With an oblique heart (« = 30° to 45°) the normally positive left inferior spike may be abolished and rendered negative during maximal expiration. An example of this experiment is given in fig. 8. [Vote.—It is not always easy to decide what are the actual measurements to be taken for calculation. In simple cases, 7.¢., in cases of vertical heart _ where V,; is large and positive on both sides, there is no difficulty in determining these values. Nor is there any serious ambiguity for the case of the clearly horizontal heart where Vj; is large and at once negative. But the not infrequent cases where V,; is double, composed of a small positive followed by a large negative, or of twosmall positives separated by a negative movement, cannot be dealt with with the same degree of certainty. The indicator does not take up a decided position, the current axis resultant from opposed and nearly balanced forces fluctuates to and fro, and the obvious Roy. Soc. Proc., B, vol. 88, Plate 5. Waller. MOT} e11dsUL oUleI}XS UI ZZ = PUL UOwAIdxe oUIEI}XE UI COT = S¥ 4NO souIOD TA Jo sonTeA TT pu y oY} WoOIJ poze[Noeo ‘v opSue perxe oxy, ‘pequeurone st ‘eargisod st yorym “» xvod repnormme oy out) cures 919. 9V “uoryeatdsur deep Suramp satztsod peropuea st ‘goolqns styq ur eayesou ATeuttou st yotyM ‘TA oytds pexoqey 4Joy OU, ‘tornvatdser deep SutiMp pesverout st 1, oyids yeroqey ay Str Uy, *sp1090. (1orreyzUt) Teraqe] 9Jo] pus gusta oyy uodn aoyeadsur deep jo syooym -qav0y [ejuozl10y VW “[ queurtodxgq “MQ ‘GW 40 asVO SHE—'), ‘SIT | | | ! | | sens bons! \ : i wine | Mee a wy | “Wn USNI gy sal alec Roy. Soc. Proc., B, vol. 88, Plate 6 ‘moTyerIdxe ouIe.4 x9 UL 98 = pus ‘uoeadsur otmedjxe UT .Z] = SB yNO soto ‘IA Jo sonTBA rT pur Y ayy WoAy poye[Noreo ‘y ofSue jerxe qOLTazUt OTL ‘uorqeridxe etmeijxe Aq SATpeSoU potopuod st ‘aatqtsod Ayyeuraou ‘IA oyrds [e199 979] au, ‘peysturMip uoyy ‘posvertour ATIYSI[S st 1A oxids [eroge] 4YSII ONY, ‘sp.100er (1OTTeyUT) [e197¥] 49] pue yysrr ogy uodn uowesidxe omesyxe Jo yoo “qaeey onbyqo uy ‘gquemmedxq “MM ‘Of £0 aSVO HET—'g ‘Dl Waller. Electrical Axis of the Human Heart. 61 fluctuation produces in the mind of the observer a corresponding state of indecision as to what value is to be taken for difference of potential during the three to four hundredths of a second of the presystole while the chemical and electrical changes preparatory to contraction are proceeding. I have . given much thought to such doubtful cases, and made many attempts to synchronise fluctuating records and to calculate the fluctuating angle at definite points during the initial period of systole, but without any satis- factory result; the fluctuations have proved to be too rapid to allow of any satisfactory establishment of corresponding points in time between different records, even when such records have been taken simultaneously. therefore for the present abandoned the attempt, and taken for these doubtful cases maximal and minimal values, whether positive or negative, and calculated separate values of a from such values, duly noting, of course, their doubtful character, the case of Dr. E. I have not found it possible to make any satisfactory correction for asynchronism between the initial and culminating points of the transverse and lateral spikes, and have taken into formula only the values of the right and left lateral (inferior) spikes, of which, according to my observations, the I have An example in point is given in Part I at p. 520, asynchronism is so small as to be negligible. ] Values of V; in the several Leads at Different Values of the Axial Angle. (Taken by direct readings of the model to the nearest millimetre.) 1. II. III. IV. We Wile Transverse, | - gine well "ipso suet Longitudinal. inferior. inferior. superior. superior. fo) @) 0 100 100 50 ‘0 50 150 10 17 107 90 40 °5 58 148 20 34 111 77 30 °0 64, 141 30 50 111 62 18-0 68 130 40 64 109 45 6°0 70 115 50 77 102 26 — 6:0 7 96 60 87 93 7 —18°0 68 75 70 94 81 —13 —30°0 64 51 80 98 66 —32 —40°5 58 26 90 100 50 —50 —50°0 50 0) 62 Dr. A. D. Waller. Various Inclinations of the Values of Vz in the Inferior Leads, taking into reckoning a Difference of Potential between the two Feet. Left inferior. Right inferior. Left Right Inferior Left Richt longitudinal. jonehe transverse, 1 i Equatorial. 18 Axial. BERNER ateral. lateral. (0) 100 100 100 100 150 150 0-0 10 91 89 106 108 149 147 iL 7/ 20 79 75 109 1138 143 139 3°4 30 64 58 109 113 132 128 5-0 40 48 Al 105 111 118 112 6°4 50 30 22 99 106 100 93 eds 60 dil 2 89 98 79 el 8°7 70 —8 —18 76 86 56 AT 9 °4 80 —27 —37 62 72 31 21 9°8 90 —45 —55 45 55 5 —5 10 ‘0 t I have repeated these two experiments many times and have rarely failed to bring off either the first or the second upon the subjects who have submitted themselves to one or other of the two. I have failed to effect complete reversal in only one or two cases where the electrical axis was above the horizontal or very nearly vertical. With a slight modification the model given in fig. 5B serves to indicate the nature and amount of the slight differ- M ences observable between the two pairs of inferior leads in consequence of the slight differences of potential that are produced at the two feet by the systolic spike V, The potential-difference between the two feet in what may be termed the inferior transverse lead is in the same direction as that of the hand-to-hand potential-difference, but of rouch lower value. The inferior transverse = 1/20 to 1/5 of the (superior) transverse ; as an average for the purpose of calculation it is taken here = 1/10, and represented on the fig. 5B by the horizontal line 7/ 1 cm, long across the point FL If now the positions of the two points 7, l (representing right and left foot), are calculated for values of « from 0° to 90°—or pricked off on the model— and plotted as before, the curves given in fig. 9 are obtained showing Electrical Axis of the Human Heart. 63 Fie. 9.—Curves con- structed in a similar manner as those given in fig. 6, to show the slight differ- ences of the electro- motive value of the spike V; in the in- ferior transverse lead and in the two right inferior leads (axial and right lateral) and in the two left in- ferior leads (equa- torial and left lateral). The numbers along the curves in this fig. as in fig. 6 express 1/100000ths of a volt, e.g.. 108 signifies 0:00108 volt. volt 1000 Go ae Ba leole ee oN BES 64 Dr, A. D. Waller. Various Inclinations of the clearly the theoretical values of the differences in the several inferior leads—axial and right lateral, equatorial and left lateral—which in fic. 5A were assumed to be identical under the designations right inferior and left inferior on the assumption that the two feet might be regarded as isoelectric.* By reference to this figure it is easy to satisfy himself what order of error can arise by regarding the two feet as isoelectric, and to understand without effort certain minor but otherwise puzzling discrepancies met with in certain hearts when the several inferior leads are compared by means of a highly sensitive instrument. Thus eg. it is obvious at a glance that such dis- ‘erepancies become more sensible in “horizontal” than in “ vertical ” cases, and that eg.a 64° heart offers the paradoxical instance of a positive left lateral and a negative equatorial spike. And the observation which I have frequently made without thoroughly understanding it to the effect that an equatorial spike is smaller than the left lateral when positive, but the larger of the two when negative, is rendered obviously intelligible. Likewise that mentioned without comment in Part I, p. 510, of this paper that the left is slightly larger than the right longitudinal spike.t The two experiments described above can be repeated with advantage in view of this second diagram model, upon consideration of which it will be apparent that in repeating Experiment 1 the left lateral is slightly more favourable than the equatorial lead, and that in repeating Experiment 2 the equatorial is slightly more favourable than the left lateral. The differences of value between axial and right lateral and between equatorial and left lateral are obviously small, and their absolute measure- ment involves a large relative error. For instance, in the case of J. C. W. the values come out as follows :— * In my first communication (‘ Phil. Trans.,’ 1889, p. 191) the two feet are given as being isoelectric, although I was well aware of the fact that theoretically there should be a slight P.D. between them. + In that connection, speaking only of the four leads between the mouth and four extremities, I used the expression right and left superior and inferior, whereas for right and left inferior I should have used the designations right and left longitudinal, in accordance with the terminology of the present paper, where right and left inferior refer to hands and feet. The table given on p. 510 should read accordingly :-— Right superior ....... amare 3 (= 0:00023 volt) Left superior .......2.....-. 15°3 (= 000118 ,, ) Right longitudinal ...... 16°55 (= 000126 ,, ) Left longitudinal ......... 175 (= 000185 ,, ) Electrical Axis of the Human Heart. 65 DransVverses 6. ss: caer woccis 0:0020 volt. Right laterals. cesnatnee: 0:0024 ASmIAll Soy a peattnionsstadeeaee 0:0028 _,, etuelatera lanes ee eece O00 ID(TPU EOE Aa Mee een ronnaee J00I25 Inferior transverse ...... 00002 ,, Pathological Applications. The various applications of the electro-cardiogram to clinical diagnosis of heart lesions fall into two chief divisions: Class A, in which the indications are certain ; Class B, in which they are uncertain. Class A is represented by the arhythmiz; no mistake is possible e.g. as to alterations of frequency and rhythm, coupled beats, auriculo-ventricular dissociation. Class B includes among many others the supposed electrical signs of right and left ventricular hypertrophy and of partial interruptions of auriculo-ventricular conduction. It is as regards Class B that I believe it to be most necessary to pay attention to the physical relations of the normal heart. The electrical signs that are accepted by clinical authorities as associated with right ventricular hypertrophy as expressed in clinical language are “small R,, large R,,,,” or, as I prefer to express it, small transverse, large left lateral spike. These are, as has been explained above, physical evidence of an approximately vertical electrical axis. The great majority of clinical observers agree in stating that this combination of small R, and large R,,, is common in mitral disease, and that it signifies hypertrophy of the right side of the heart. I have seen many cases during the last three years that are In agreement with this statement, but, on the other hand, I have during the last 20 years met with still more numerous cases of apparently perfectly normal persons that presented this combination, and have. only inferred that they possessed vertical hearts. I have become accustomed to expect to find this combination in infants and in any tall healthy young man accustomed to take plenty of open-air exercise. Therefore, without presuming to express any opinion as to the clinical value of this sign of right ventricular hypertrophy, I do venture to say that in the first instance our reasoning from the sign should be limited to the conclusion that in its presence the electrical axis of the heart must be vertical or directed to the right, and bear in mind that this indication is presented by the hearts of many normal persons. The diagnosis of right ventricular hypertrophy has to be established on independent clinical grounds. The electrical signs that are presented as being significant of left ventricular hypertrophy are, in clinical language, large R,, small or reversed VOL. LXXXVIII.—B. F 66 Dr. A. D. Waller. Various Inclinations of the Ryp or, as I express it, a large transverse and a small or reversed left lateral spike. Precisely similar considerations apply to this sign in relation to diagnosis as have been just stated as regards the right side of the heart. I have seen cases that agree with the clinical statement (but others that do not), and I have seen far more numerous cases of normal persons with small transverse and negative left inferior spikes, and have inferred therefrom that the electrical axis of the heart was approximately horizontal. I have met with this sign at all ages and in all conditions of health, and have become accustomed to expect to find it in anzemic young women and in aged persons of either sex. I associate it in my mind with a soft or flabby heart muscle, but possess no confirmatory post-mortem evidence of that impression. The electrical signs that are presented as being significant of interruption of the right (or left) branch of the auriculo-ventricular bundle of Kent and His consist essentially in a reversed and prolonged R,,, resembling a left ventricular extra-systole, but occurring in sequence to an auricular con- traction. All the cases hitherto reported which have been confirmed by post-mortem examination have been on the right side, and have been charac- terised electrically intra vitam by a negative left lateral deflection, which has been accepted as an indication of ventricular contraction initiated on the left side. I shall not venture to deny the possible accuracy of the chain of . argument upon which the diagnosis of the interruption depends, but in estimating probabilities I think it should be clearly realised that a negative left lateral deflection is of frequent normal occurrence. In Part I of this paper it has been shown that the inferior angle « varies within a very wide range (--10° to +100°) with the shape and position of the heart. A presumably “soft” heart, of which the muscle is deficient in tone, is sessile upon the diaphragm, and its electrical axis is approximately horizontal. With “hard” muscle the heart, even during diastole, is more nearly erect upon the diaphragm, and its axis is more nearly vertical. The axial angle is decreased by inspiration, increased by expiration; it is decreased by muscular exercise, increased by repletion of the stomach. I think that it is extremely probable that to this series of statements it may be added that, pathologically, the angle is decreased by engorgement of the right side of the heart (as occurs eg. in mitral disease), so that the electrical axis may be vertical or actually directed to the right, and increased by hypertrophy of the left side especially (as occurs ey. in aortic disease), so that the axis may become horizontal. But since both these conditions, i.e. axis to the right and axis horizontal, are compatible with a normal state, I do think that either a remarkably large left lateral spike or a reversed left Electrical Axis of the Human Heart. 67 lateral spike is to be admitted as affording per se proof of the existence of right or left hypertrophy or dilatation. Considering, further, that a negative left lateral spike is of frequent occurrence in the normal as well as in the diseased heart, it cannot be admitted as affording per se proof or even evidence of an interruption of the right branch of the auriculo-ventricular bundle. The facility with which in certain hearts a positive left lateral spike can be rendered negative is such as to forbid us from admitting a temporary reversal as an indication of temporary interruption of conduction to the right ventricle. [Note added May 23, 1914.—By courtesy of Dr. Part of the National Hospital for Diseases of the Heart, I am able to complete the account of the cases of A. D. W. and of J. C. W. by the reduced skiagraphic outlines (fig. 10) of their hearts in the positions of deep inspiration and expiration. The outlines of the heart and diaphragm indicate approximately the anatomical alterations corresponding with the electrical alterations given in figs. 7 and 8. There is, however, in these cases no absolute correspondence of axial angles to be made out between the anatomical and the electrical estimates. The skiagrams required the breath to be held for several seconds in inspiration and in expiration respectively. In general, the correspondence between the anatomical and the electrical axial angle is not very close. Asa rule the ae Insp. Insp. Fie. 10. right and left electrical effects of the infantile heart which is mesial may be expected to be about equal, while in the senile heart which tends to become horizontal the left hand inferior spike is more usually negative. But cases occur of apparently normal as well as of diseased hearts (e.g. mitral diseases) where the left inferior is larger than the right inferior spike (implying an electrical axis directed to the right and a reversed transverse spike), but where the anatomical axis is distinctly directed to the left. Cases also occur of reversed left inferior spike where the electrical axis comes out in @ a 68 Mr. Halnan and Dr. Marshall. as greater than 90°, zc. directed upwards to the left, but where the anatomical axis is distinctly less than 90°, ze. directed downwards to the left.] Corrigenda in Part I, ‘ Roy. Soe. Proc.,’ B, vol. 86. Page 512. In the second line of the footnote tan = should read tan a =. hot should read tan a = 2 i Page 520. The record of the right superior lead is placed upside down. The first ventricular wave is actually negative. Page 525. The numbers 36 and 47 in the last column (15th and 16th from bottom) should be transposed. Page 514. Last line, tan a = On the Relation between the Thymus and the Generative Organs and the Influence of these Organs upon Growth. By E. T. Hatnan and F,. H. A. MarsHaty. (With a Note by G. Upny YULE.) (Communicated by Prof. J. N. Langley, F.R.S. Received April 4,— Read June 18, 1914.) Calzolari was the first to show that in castrated male animals the absolute weight of the thymus is larger than that of the same gland in normal animals. The experiments were made upon six rabbits, which were castrated when between one and three months old and killed at various periods afterwards up to nine months, each rabbit being compared with a control. Subsequently Henderson carried out.a statistical investigation upon the weight of the thymus in cattle, and showed that in these animals castration caused a persistent growth and a retarded atrophy of the gland. Henderson also records two experiments upon guinea-pigs by Noél Paton, and the results of these are confirmatory of the observations upon cattle. The possible reciprocal action of the thymus upon the testis was investigated by Noél Paton, who removed the former organ from 24 young guinea-pigs and killed them when they attained weights varying from 115 to 355 grm. These animals were compared with 23 normal guinea-pigs kept as controls. The conclusion reached was that in guinea-pigs below 300 grm. (ze. prior to the time when the thymus usually atrophies) thymectomy is followed by a more rapid growth of the testes. In guinea- pigs above 300 grm. Paton found that the difference in weight of the testes in thymusless and normal animals was not manifest. The figures upon Relation between the Thymus and the Generative Organs. 69 which these conclusions are based are dealt with statistically in an appendix to this paper by Mr. G. Udny Yule. Noél Paton’s conclusions have been challenged by Soli, who worked upon guinea-pigs and upon fowls. The guinea-pigs, except in the case of two. pairs, were killed when weighing considerably over 300 grm., so that the results have no bearing upon Paton’s assertion that in guinea-pigs below that weight thymusless animals tend to have larger testes than normal individuals. In the two pairs killed below 300 grm. the testes of the operated guinea-pigs were slightly lighter than those of the control animals. In the experiments upon fowls Soli found that in two cases the thymusless birds had heavier testes than the controls, but in 11 cases the testes of the operated individuals were lighter than those of the unoperated. LPaton,- however, points out that in certain of these the rate of growth was below the normal, and that the small size of the testes might have been due to inferior nutrition. Soli found that castration produced hypertrophy of the thymus or arrested atrophy in that organ. In the unoperated birds the average weight of the thymus was 0°6 grm. to each kilogramme of body weight, whereas in the capons its weight was 1:16 grm. to each kilogramme of body weight. Gellin, Klose and Vogt, Marrassini, and Squadrini have also found that castration tends to enlargement of the thymus, or arrests the normal involution of the gland. As a result of a further series of experiments, Paton has concluded that the thymus and the testis do not act antagonistically to one another, but that each organ has a stimulating influence upon growth, the one organ compensating for the removal of the other by undergoing hypertrophy. Paton found that castration alone without thymectomy had no effect upon the growth of young guinea-pigs, neither had thymectomy alone any influence upon the rate of growth. On the other hand, castration and thymectomy performed simultaneously in very young guinea-pigs was found to check growth. Considered in the light of our experiments to be described below we are of opinion that this effect may have been consequent upon the double operation, which very possibly lowered the resistance of the animals towards disease. Paton describes further experiments showing that in six castrated males and four castrated females the average weight of the thymus was greater than in control animals. Basch, and also Klose and Vogt, describe extirpation of the thymus in dogs as producing a softening of the bones or retarding the growth of the bony tissues, besides causing other pathological phenomena. Similar changes resulting from thymectomy are described by Matti. Soli also has confirmed 70 Mr. Halnan and Dr. Marshall. these results for rabbits, but failed to confirm them for guinea-pigs, in which thymectomy is a simple operation. It is not improbable, therefore, that the inhibitory effects of the growth observed by Basch and others were merely post-operative, since they only occurred when the operation of thymectomy was a severe one. Gudernatsch states that tadpoles fed upon thymus extract grew to an abnormal size and postponed undergoing metamorphosis. In some cases they did not change into frogs at all, but remained as giant tadpoles. Stotsenburg, working upon the effect of ovariotomy on the growth of albino rats, found that this operation caused an increased rate of growth. This appeared to be the case not only after the age of sexual maturity was reached, but prior to the attainment of this age, since the removal of the ovaries appeared to induce an accelerating effect forthwith. Lastly, Miss Hewer in a recent paper states that it is possible to induce a hyperthymic condition in rats by feeding these animals upon fresh thymus or upon thymus tabloids, and that this condition is accompanied by partial or complete sterility, the spermatogenetic tissue in the testes ceasing to be active or even undergoing atrophy. It is to be noted that these results are directly contrary to Paton’s theory of a compensatory mechanism between the thymus and the testis. Record of Experiments with Guinea-pigs. The experiments described below were undertaken to put on a more quantitative basis the results obtained by Noél Paton in a previous paper, and form, with a few minor differences, a repetition of his work on the subject. In the control thymectomies (pseudo-thymectomies) the thymus glands were exposed without being removed, and the neck afterwards sewn up in the usual way so as to make the operation resemble as nearly as possible an actual thymectomy. In the vasectomy experiments a portion of each vas deferens was removed, but the vascular supply of the testes was not interfered with. The guinea-pigs were housed in roomy wire cages and given a liberal diet of oats, bran, and roots, varied occasionally with green food as circumstances permitted. The normal animals were housed in the same cages as the experimental ones, thus ensuring uniformity of external conditions for both sets of animals. With the exception of one experiment all the animals used were males. Experiment 1: Effect of the Removal of the Testes upon the Weight of the Thymus and the Growth of the Animal.—tIn this experiment, 12 male guinea- Relation between the Thymus and the Generative Organs. 71) pigs were used. The experiment started on December 20, 1912. On February 7 the testes from six guinea-pigs were removed. On March 4 the thymuses of the entire set were extirpated. The animals were killed on May 9. The number of animals taken was too few to establish any possible erowth effects due to removal of the thymuses and testes. The animals — retaining the testes grew less in the later stages of the experiment, this possibly being due to the castration effect on growth observable in all castrated adult animals. With regard to the effect on the thymus, small though the number of animals is, the evidence is quite clear. Moreover, the fact that the animals castrated had all attained puberty before castration took place shows that the arrested atrophy and subsequent hypertrophy of the thymus cannot be explained by Paton’s theory. Days after first weighing. | | , | Weight Animal No. | | | of thymus. 0 49, | 7A. | 140 | | Control Animals. erm. | grm. erm. germ. erm. 1 237 | 355 475 580 0 :276 | 2 192 335 354 a 0 °342 3 215 358 442 497 0 :370 4 214 342 452 617 0 337 5 202 357 377 588 0-400 | Average— | 4 animals ...... = = 436 570 9) -d0ga00 | 212 349 | 4.20 | 0 °345 Operated Animals. 1 212 355 | 402 615 0-430 2 240 , 400 470 682 0 °395 3 184: 345 375 567 0-470 4 187 360 380 610 0 564 5 225 375 447 674 0-680 A\SSENEYRD on no0000 860 209 367 415 | 629 0 508 Experiment 2 (figs. 1, 2,and 3): Effect of Removal of the Testes on the Weight of the Thymus, of Removal of the Thymus on the Weight of the Testes, and upon the Growth of the Animal.—This set contained 10 normal animals, 7 castrated, and 6 thymectomised. The animals were operated upon on February 24~27, and first collectively weighed on March 4. On May 9 the experiment terminated. a2 Mr. Halnan and Dr. Marshall. _ Reference to the data below shows that thymectomy has no effect upon growth, and that castration has little, if any, effect. | Days after first weighing. | F Ne es ins | Weight when ees Weight of , BEEMAN NO: | | killed. aaah thymus. 0. 63. ive hao Control Animals. se grm. | grm. gym. erm. grm. 1 202 349 875 1°25 0 253 2 157 365 380 1°43 0-395 a 3 164 | 385 420 2°05 0 °392 j 4 182 310 320 1°02 0-197 5 127 275 290 0-376 0-390 6 170 412 420 1 ‘67 0 *350 | til 180 395 410 2:08 0 °380 | 8 185 ~ | 410 425 1°82 0 °300 9 221 452 470 2°38 0 °395 10 191 | 452 | 468 2°43 0 °390 Average......... 168 | 380 | 398 1°651 0 344 Castrated Animals. 1 214 401 430 — 0 580 | 2 211 405 420 — 0°750 3 155 320 | 330 — 0 *405 4 185 377 423 — 0 :507 5 235 462 485 — | 0 680 6 220 435 | 470 — | 0-710 7 204. 292 335 — | 0 416 Average......... 2038 | 385 413 — 0°578 Thymectomised Animals. 1 135 440 450 2 °185 2 154 395 410 1 °560 3 154 315 325 1 :200 4, 167 384 390 1 “705 5 142 847 300 0-835 6 183 394 400 1-260 Average......... 156 379 : 388 1. qe ele oven Relation between the Thymus and the Generative Organs. 73 Experiment 3 (fig. 1): Effect of the Removal of the Testes on the Weight of the Thymus and the Growth of the Animal.—This set contained originally 7 castrated and 9 normal animals, but owing to deaths only 4 castrated and 6 normals are recorded. The experiment commenced on the day of | operation, March 17, and finished on June 24. Here castration seems to have had a positive effect upon growth, but the number of castrated animals (4) indicates the tentative nature of this result. As in Experi- ments 1 and 2 the effect of castration on the weight of the thymus is well marked. | | | Days after first weighing. PectaalN Ps De ' < Weight of testes § Weight of oe | + epididymes. thymus. 0. | 96. | Control Animals. | grm. grm. | grm. grm. 1 93 283 1°225 0-190 2 109 322 1 +235 0-265 3 | 107 | 285 0-870 0°175 4 165 | 420 2-100 0 °295 5 109 336 2.°154 0°'217 6 165 | 390 1°750 0-275 | Average 2.20... 124 | 339 | 1 ‘556 bee Castrated Animals. 1 148 | 412 | —_ | 0-610 2 | 140 | 346 | ae | 0-452 3 | 111 372 — 0-430 4 124 391 | — 0-540 | Average......... 131 | 380 | = | 0508 | : Experiment 4 (fig. 1): Effect of the Removal of the Testes on the Weight of the Thymus and on the Growth of the Animal—To annul the operative effect the control animals were vasectomised. The experiment started on May 28 and finished on August 1 with 6 vasectomised and 5 castrated animals. The evidence in this experiment with regard to growth is con- tradictory to Experiment 3. 74 Mr. Halnan and Dr. Marshall. Days after first weighing. : Animal No. Weight of thymus. 0. | 65. Vasectomised Animals. erm. erm. grm. il 245 387 0 433 2 | 230 A424. 0 -400 3 268 467 0-500 4 173 282 0-254 5 208 373 0-300 6 170 400 0-393 | Average...... | 215 389 | 0 :380 Castrated Animals. 1 243 330 0-573 2 280 405 0 :607 3 210 353 0°738 4 147 320 0 545 5) 157 336 0 °614 Average...... 207 349 | 0-615 400 400- @ = Castration. © = Controls. 350 300 300 250 Weight in grams. 200 Time in days. 150! 10 20 30 40 50 60 76 80 Fic. 1.—Effect of Castration on the Growth of Guinea-pigs. (The curves are formed from the averages of the individuals of Experiments 2, 3, and 4.) Relation between the Thymus and the Generative Organs. 75 Experiment 5 (fig. 2): Effect of the Removal of the Thymus on the Weight of the Testes and on the Growth of the Animal—To annul the operative effect the control animals were pseudo-thymectomised. The experiment started on May 27 and finished on July 8. There were 7 thymectomised animals and 6 pseudo-thymectomised in this set. The results obtained support the contention that thymectomy has no effect upon growth. | ] 5 . Fee pall Weight | Weight of testes Weight seal Ne: Weszht, May 27 | after 42 days. | + epididymes. of thymus. | | | Control Animals. grm. | grm. grm. grm. 1 210 325 1 °420 0 °325 2 | 177 316 1-155 0-510 } 3 190 267 0-610 0°315 4 } 151 238 0 °470 0 +235 5 162 290 1012 0-320 6 | 190 338 1°380 0-283 | | | 7 : | PAVEXA PO.) ceca ecscse 180 296 1-008 0-331 | Thymectomised Animals. 1 141 | 265 0-770 | 2 227 | 307 1-112 | 3 128 238 0-313 | 4 153 257 0°57. 5 197 287 0-880 6 150 245 0-527 7 187 330 1 °363 Average wove 169 | 276 0-791 | , | I Experiment 6 (fig. 2): Effect of the Removal of Thymus on the Weight of the Testes and on the Growth of the Animal.Nine normal animals and 14 thymectomised were used in this set. The operations were performed June 19-26. The experiment started on July 1 and ended on July 25. The evidence supports Experiment 5 with regard to growth. The effect on the testes is here in accordance with Noél Paton’s results. 76 Mr. Halnan and Dr. Marshall. p Weight, Weight Weight Weight Weight 2ST Te) J aie 1. after Oe days.| when killed. of erent of then Control Animals. grm. grm. | grm. | grm. grm. 1 133 187 200 0 320 0:277 2 213 317 322 | 1-250 0-410 3 201 267 272 1°120 0 °322 4 151 245 252 0 ‘676 0 °350 5 136 193 207 0-320 0 *320 6 171 245 235 0 “620 0 ‘408 a 226 316 340 1125 0 -480 8 170 268 307 0-970 0 506 9 166 252 _ 240 0 °633 0 *460 | Average... Pere 24 | 264 o-7s1 | 0-392 Thymectomised Animals. 1 233 347 352 1:43 2 156 219 221 0-454, 3 238 358 380 1-800 | 4 168 226 212 0 *425 | 5 167 215 235 0-580 6 201 260 245 0-930 | 7 163 233 215 0 °620 8 205 296 300 1°170 9 157 238 258 0 623 10 197 301 310 1 *450 iit 225 316 364 1°702 L 12 205 256 237 0-835 13 152 230 265 0 -640 14 158 234 257 0 ‘630 Average............ | 187 266 275 0-949 } 400 400 350 350 300 300 250 2s0 Weight in grams. 200 200 150 150 Time in days. 100: 10 20 30 40 so 60 70 80 Fic. 2.—Effect of the Removal of the Thymus on the Growth of Guinea-pigs. (The squares represent the averages of Experiment 5, the triangles Experiment 6, and the circles Experiment 2. In all cases the blacked figures represent the operated animals.) Relation between the Thymus and the Generutive Organs. 77 Experiment 7 (fig. 3): Effect of the Simultaneous Removal of the Testes and Thymus on the Growth of the Animal—tThis experiment was undertaken to investigate the effect of castration and thymectomy together on growth. In order to annul the operative effect the controls were vasectomised and ~ pseudo-thymectomised. The experiment commenced with 9 animals, 4 operated, 5 controls, on July 16, and with 16 others (8 a side) on July 25. Reference to the protocol will show that this double operation has no effect upon the growth of guinea-pigs, a result directly contrary to that of Noél Paton. Set 1. | | Weight after— Animal No. | Weisht, | | | Tes 9 37 75 | 86 LOU 25 | 160s ola days. | days. | days. | days. | days. | days. | days. | days. | Control Animals. grm. grm. erm. grm. | grm. | grm. | grm. | grm. | grm. 1 180 210 306 342 3388 | 357 437 542 607 2 178 202 270 290 265 292 328 433 460 3 185 216 300 372 370 | 402 442 543 | 582 4 192 220 301 340 340 347 397 459 509 5 105 139 237 286 285 | 317 383 458 | 485 | Average ......... 168 197 283 326 | 319 | 342 397 492 | 528 Operated Animals. 195 221 285 331 320 345 387 474, 502 160 191 282 | 318 319 322 367 445 492 197 223 311 | 343 354 345 370 480 465 186 215 280 | 335 344 360 375 445 472 B cob Average ......... 184 212 289 | 332 334 343 375 461 483 78 Mr. Halnan and Dr. Marshall. Set 2. | | | | Weight after— Weight, | Animal No. | Send | | | | July 25. | 28 | 66 79 | 116 141 168 | days | days. days. | days. days. days days | | Control Animals. grm. grm. | grm. erm. grm. grm. grm. grm. | il 155 278 352 337 365 408 512 537 2 160 250 343 309 367 448 525 555 3 260 231 298 272 290 334 407 445 4 133 212 286 285 300 378 450 417 5 i155) 258 355 368 370 450 534 574 6 126 221 276 287 312 353 377 357 i 220 321 395 393 394 452 522 573 8 186 256 | 336 330 340 419 478 AT5 Average ...... 174 253 | 330 329 342 405 475 490 Operated Animals. ' ll 181 |. Dasr/ 318 323 327 882 | 428 490 2 190 286 346 363 382 437 480 552 3 | 136 235 291 302 327 358 407 390 4 | 175 268 340 337 340 400 450 487 | 5 | 156 270 377 383 410 A380 | 517 570 6 173 250 268 273 297 sav 88s 395 7 170 277 386 380 410 435 | 512 567 8 | 212 275 336 346 287 379 423 470 Average ...... 174 264 333 338 347 393 450 490 Experiment 8 (fig. 3): Hfect of the Simultaneous Removal of the Testes and Thymus on the Growth of the Anvmal.—The animals in this experiment were treated as in the previous experiment, the operations extending from July 30 to August 5. The first collective weighing took place on August 22. There were 5 operated and 5 control animals. The results obtained in this experi- ment confirm the findings in Experiment 7. Relation between the Thymus and the Generative Organs. 79 4 Weight after— : Re ————————————————_——————————————EEEE Animal No. q Bai re 2e 22h ek 51 64 ee Ol ane aiee | 152 days. days. days. days. | days. days. days. Control Animals. grm. | grm. grm. isaa, |) ei, [| fe, ff, fafa | 1 2038 | 309 340 397 ago | 498 | 577 | 590 | 2 285 374 | 396 | 422. | 484 | 572 | 587 | 547 3 De) 320 335 355 | 452 520 | 567 548 4 iy aa 3 180 232-9 asl9) 9409). | 4651 | “465. | 5 183 | 330 360 380 462) 587 585 | 575 Average ....,. 221 | 296 | 322 357 | 442) 607 || 546 1) 545 Operated Animals. 1 243 esse) 43500) 407 469) 537 607 |; 622 2 1385 | 176 177 | 235 | 284 | 354 | 3650 360 3 230 373 AQoellaedao ets) |) 552)01|) 1607 | | 600 4 152 | 286 320 365 | 402 | 485 || 520: | 565 5 217 ~+| 360 402, | 387 | 482 | 525 606 | 615 | | | | | } a Westnet aS Se in are Average ...... 195 | 315 | 347 | 366 425 | 491 588 | 6552 550 6 E g t ion) f ws <= & uv 3 250 O= Pseudothymectomy + Vasectomy. e-= Thymectomy + Castration. Zale Days after operation 150 0 20 40 60 80 100 120 140 160 189 Fre. 3.—Effect of Thymectomy and Castration on the Growth of Guinea-pigs. (The curves are produced from Experiments 7 and 8, the averages of Experiment 8 and of the second set in Experiment 7 being interpolated to give the approximate average weights.on the desired intervals after operation. _The separate portion at the beginning of the curve is due to the fact that the weighings of certain animals only commenced 20 days after operation.) 80 Mr. Halnan and Dr. Marshall. Experiment 9 (fig. 4): Effect of Semi-Castration on the Weight of the Thymus.— In this set 7 animals were semi-castrated. The results, in so far as these related to the thymus weights, were too variable to admit of any importance being attached to them. : : Weight of : Ren ehNGe pelea Weight after srancinaiia Weight of ugust 13. 51 days. lesti thymus. estis. | erm. grm. grm. grm. i 117 310 0-652 0 °355 2 134 358 1:214 — 0 480 3 136 350 1-048 0-380 4 117 363 1-206 0 550 5 154 | 362 1°312 0 °385 6 123 | 368 1-100 0 546 7 148 | 356 1 -400 0°252 | Average... 133 352 = | 0-421 250 300 359 400 450 500 Fie. 4.—Effect of Removal of Testis on the Weight of the Thymus. (Blacked circles, castrated animals; semi-blacked circles, semi-castrated animals; clear circles, control animals. Vertical, weight of thymus in grammes; horizontal, body weight in grammes.) Relation between the Thymus and the Generative Organs. 81 Experiment 10: Effects of the Removal of the Ovaries on the Weight of the Thymus.—Five females were castrated August 11-14, and compared with 6 controls. The experiment extended from August 11 to October 9. Reference to the data given will show that castration in the female, as in the male, leads to an arrested atrophy and continued growth of the thymus gland. . Weight, Weight after Weight of Astor, 6, August 11. 59 days. thymus. Control Animals. erm. grm. germ. 1 182 3384 0 °305 2 170 246 0-190 3 160 326 0 363 4 177 320 0 :280 5 158 365 0-400 6 181 350 0°320 Average ...... 171 323 0-309 Operated Animals. 1 162 424 0 °805 2 160 326 0 °405 3 153 406 0-720 4 170 415 0-780 5 145 341 0 590 Average ...... 158 382 0 660 Experiment 11 (fig. 5): Effect of Removal of the Thymus on the Weight of the Testes and on the Growth of the Animal.—Nine animals thymectomised October 21, 9 controls. The experiment extended from October 21 to November 18, by which time 6 operated animals and 7 normal animals were over 300 grm. in weight. The evidence here confirms the findings of Experiments 2, 5 and 6, VOL, LXXXVIII.-—B, ( 82 Mr. Halnan and Dr. Marshall. Animal No. Weight, Oct. 21.| After 28 days. Test, and epid. Thymus. | Control Animals. grm. grm. erm. grm. iL ATT 332 1-650 0-330 2 253 308 0 °820 0 °330 3 217 233 0-340 0°120 A 298 335 1-065 0-290 5 205 282 0 520 0°375 6 260 315 0°770 0 340 7 348 392 2 556 0-290 8 272 370 1-695 0 °250 9 312 350° 2-230 0 °235 INGE yao00000" 271 324, 1-294 0° 284 Operated Animals. grm. grm. erm. il 270 342 1-010 2 277 350 1174 3 227 303 1 °535 4 287 281 0544 5 207 278 0-470 6 320 355 * 1:°186 7 370 A425 2 °504 8 276 362 1°610 9 240 298 1°125 ; Average 275 332 1-239 200 250 300 350 400 450 500 Fig. 5.—Effect of Removal of Thymus on the Weight of the Testis. (Clear circles, control animals ; black circles, operated animals. Vertical, weight of testes and epididymes in grammes ; horizontal, weight of animal in grammes.) Relation between the Thymus and the Generative Organs, 83 Experiments 12 to 17: Effect of the Removal of the Thymus on the Weight of the Testes.—In these experiments, the chief intention was to investigate the effect of thymectomy on the weight of the testes. In order to cut out the enormous variation in the weight of the testes obtained as the animal approaches puberty all animals were killed when they attained the weight of about 260 erm. Reference to fig. 5, on which the results obtained from this and other experi- ments are plotted, shows that thymectomy does not lead to a compensating acceleration in the growth of the testes. Animals used. | Initial average weight in grammes. Hxperiment. = Normals. Operated. | Normals. Operated. 12 8 6 188 166 13 — 5 — 164 14, 7 5 185 191 15 6 5 139 | 145 16 9 7 166 187 17 a 4 — 174 Experiment 12. is 6 Final Weight of Initial Final Weight of Kennel engl weight. testes, weight. weight. testes. Control Animals. - Operated Animals. grm. grm. grm. grm. grm. grm. 187 277 0 620 153 257 0 °550 222 270 0-720 207 268 0 °590 170 254 0-465 174 275 0-795 175 254 0-514 138 266 0-550 187 255 0 456 162 255 0 °460 202 267 0 603 162 257 0 ‘655 200 267 0-590 163 260 0 “795 Average...... 188 263 0 °598 166 263 0 -600 Mr. Halnan and Dr. Marshall. 84 Experiment 13. Initial weight. Final weight. Weight of testes. Operated Animals. grm grm. | grm. 165 260 0-480 170 277 0°515 135 266 0 -490 160 269 0 600 190 257 0 *580 Average ...... 164 266 0583 Experiment 14. ae : | Final Weight of Initial Final Weight of Ibaeialol wich Me | weight. testes. . weight. weight. testes. Control Animals. Operated Animals. grm. grm. grm. erm erm grm. 137 261 0 °470 210 258 0540 177 255 0370 227 267 0 568 240 252 0 520 182 258 0512 177 256 0-385 172 255 0-710 197 253 0-370 165 267 0 663 — 252 0 °545 = 258 0 °572 Average ...... = 255 0 462 191 261 0 598 Experiment 15. “ps : Final Weight of Tnitial Final Weight of bao aegis weight. testes. weight. weight. testes. Control Animals. Operated Animals. grm. grim. grm. grm. grm. grin. 165 262 0 460 155 253 0-710 108 262 0-660 107 253 0 590 107 253 0 670 210 255 0 560 130 277 0 °505 132 257 0 °520 165 251 0 564 120 252 0-530 160 262 0 390 Average ...... 139 261 0 °541 145 © 254 0 582 Relation between the Thymus and the Generative Organs. 85 Experiment 16. Tritinliaveroht Final Weight of | Initial Final Weight of Bare eavele ne. weight. testes. || weight. weight. | testes. | Control Animals. Operated Animals. rm. rm. rm. grm. grim. grm. 128 °358 6 *390 132 267 0 *350 150 253 0520 209 267 0 ‘976 129 254 0°510 207 264 0 330 128 255 0:710 215 255 0 °330 160 262 0 °520 167 255 0-800 135 256 0 580 202 259 0 507 219 273 0-580 183 205 0-480 236 278 0°740 215 279 0 ‘570 Average ...... 166 263 0 *569 187 260 0 °5389 Experiment 17. Initial weight. Final weight. I | | Weight of testes. | Operated Animals. erm, grm. grm. 198 253 0 5038 172 262 0-790 152 255 0-465 — 253 0680 174 256 0 609 Summary of Conclusions. From the evidence given in the above set of experiments, where, in investigating growth effects, the authors were careful to compensate for any possible operative effects, are drawn the following conclusions :— (1) Removal of the thymus in young guinea-pigs does not affect the growth of the animals. (2) Removal of the testes and epididymes in young guinea-pigs does not affect the growth of the animals before sexual maturity. (3) Simultaneous removal of the testes and thymus in young guinea-pigs does not affect the growth of the animals before sexual maturity. (4) Thymectomy is not followed by hypertrophy of the testes. (5) Castration leads to an arrested atrophy and subsequent hypertrophy of the thymus gland, as found by other investigators. 86 Mr. Halnan and Dr. Marshall. (6) There is no evidence of the existence of a compensatory mechanism between the testes and the thymus. The work was carried on at the Field Laboratories, Cambridge. The operations were done by F. H. A. Marshall; the weighings and the chief part of the other work by E. T. Halnan. The expenses were defrayed by a grant made by the Board of Agriculture and Fisheries out of funds placed at their disposal by the Development Commission. Note by G. UpNY YULE. In view of the disagreement with Prof. Paton’s conclusions, Dr. Marshall asked me to investigate the probable errors of some of the comparisons made, with especial reference to the alleged effect of extirpation of the thymus on the growth of the testes. The problem was not an easy one. A glance at Prof. Paton’s figures, or at the corresponding data given by Halnan and Marshall, will show how exceedingly variable are the weights of the testes and how much caution must consequently be used before basing any conclusion on a small difference between the average weights for two groups of some 20 to 30 animals. Considerable differences might be shown even by the averages of groups treated in precisely the same way. Were the animals adult, the “probable error” of the difference between any two observed averages—the amount which it would be as likely as not to exceed owing to mere fluctuations of sampling—might be readily obtained in the ordinary way. But the animals are not adult; the weight of the testes increases very rapidly with the weight of the animal, and the weights of the different individuals themselves vary greatly, so that the two groups of operated and controls are not strictly comparable as a whole. What I finally decided to do, therefore, was this: To obtain, by known methods, equations expressing as closely as possible the relation between mean weight of the testes and body-weight, for operated and for normal animals, and to see whether the constants in these equations differed more than could be expected owing to the chances of sampling alone. As in Prof. Paton’s data the weight of testes did not seem to be a linear function of the body-weight, and it was these data that I first investigated, the logarithm of the testes-weight was substituted for the actual value, and this seemed to give an approximately linear relation, judging from the diagram (fig. 6). The two equations, with the probable errors of the constants which I finally obtained from Prof. Paton’s data, including all the 23 normal Relation between the Thymus and the Generative Organs. 87 Fic. 6.—Effect of Removal of the Thymus on the Weight of the Testes (Prof. Paton’s data). (Vertical, logarithm of the testes-weight in decigrammes; horizontal, body-weight in grammes. Unbroken line, regression line for normal animals ; broken line, regression line for operated animals. Clear circles, normal animals; black circles, operated animals.) and 24 thymusless animals, were, ¢ being the testes-weight in decigrammes and 6 the body-weight in grammes :— Thymusless......... log? = (0:00489 + 0:00043) 6—0-4069 + 0-1065, Control stagicece on log t = (000503 + 0:00025) b—0°5193 + 0:0608. If the three normals and four thymusless whose body-weights are over 300 grm. be excluded, the results are :— Thymusless......... log¢ = (0:00539 + 0:00051) b—0°5097 + 0:1203, Controls oie. .ss. logé = (0:00501 +0-00030) b—0°5156 + 0:0815. The probable errors must not be regarded as too precise, since they are obtaimed on the assumption of normal correlation, but they are likely to give a fair guide to the possible magnitude of fluctuations. That of the coefficient of the body-weight (the regression of the logarithm of the testes-weight on body-weight) is the known value 06745 Tuc) 02 SN : while for the constant term I find the probable error 2 2(1— 2 0:67.45 (= (11?) +? ac o g 88 Mr. Halnan and Dr. Marshall. where o, and o2 are the standard deviations of log¢ and body-weights respectively, 7 is the correlation between them, 7 is the number of observa- tions, and Z is the mean body-weight. Further, it may be noted that there is a high negative correlation between errors in the regression and in the constant term. It is clear from the probable errors given that no stress can be laid on the differences observed, which lie well within the range of differences likely to occur owing to fluctuations of sampling alone; equally unlikely or more unlikely differences might have occurred, I find, even had both groups been normal, once in some seven or eight trials. Applying the same method to Halnan and Marshall’s data, I find for all the 65 controls and 70 thymusless animals — Mhymislessis.. ces log¢ = (0°00319 +0-00020) J—0:0384 + 0°0597, Wontrolsi enaysae log ¢ = (0:00367 + 0:00015) b—0:2032 +0:0441 ; and for the 43 controls and 49 thymusless under 300 grm., Mhymusless.. 2.72 - log ¢ = (0°00210 + 0:00069) 64 0:2195 + 0:1775, Controls) ees log t = (0:00364 + 0:00057) 6—0:2098 + 0:1465. The difference between the constant terms in this last case looks large, but the probable errors are also very large, and the difference is less than twice its probable error, viz., 0°2301. Summarising in the same way as before, I find differences as improbable as those observed might have arisen owing to fluctuations of sampling once in some five or six trials. Halnan and Marshall’s data, it may be noted, do not include any animals under 200 grm. and few under 250 erm. and give a low correlation between body-weight and log (testes- weight) for the rather narrow available range of the non-adults. Within the short range of body-weight 250-259 grm., there are 22 thymusless and 17 controls, and it may be desirable to give a simple comparison for these to emphasise the magnitude of the probable errors. For controls the mean testes-weight is 0569, with standard deviation 0-101 grm.; for thymusless, mean 0°519, with standard deviation 0:103. The difference, 0:050, is therefore in the direction indicated by Prof. Paton’s views, but no stress can be laid on it, as it is only 2°25 times the probable error of the difference, viz., 0:0222. Taking Paton’s and Halnan and Marshall’s data as a whole then, it seems impossible to regard any effect of extirpation of the thymus on the growth of the testes as proved; if there is any such effect it seems clear that it is small. The data stand in complete contrast with those relating to the effect of castration on the growth of the thymus. Within the limits of body-weight in Halnan and Marshall’s data, there seems to be little relation between weight of thymus and body-weight, so the means may be compared directly. I find :— Relation between the Thymus and the Generative Organs. 89 21 castrated animals— mean thymus weight, 0°557 erm.; s.d., 0°1104 orm. “7 controls—mean weight of thymus, 0°331 erm.; s.d., 0:0785 grm. The difference is 0°226 grm., and is 11°8 times ne probable error of the difference, viz., 0:0192. In the case of Prof. Paton’s data respecting the effect of simultaneous removal of the thymus and the testes on the rate of growth, the differences observed between operated and control animals seem to point to definite causation. If his Tables III and IV are pooled together, giving 9 operated animals and 12 controls, the difference between the mean gains in weight (viz., 92°2 and 149:2 erm.) is 41 times its probable error. For Lot 4 of Table V there are only five animals a side, but the results are unusually uniform. The mean gains in weight are 65 and 120 erm., and the difference 6°6 times its probable error. This result seems in direct conflict with Halnan and Marshall’s experiments; they have pointed out above a possible cause for the divergence. In the preceding, one or two results in probable errors have been given without proof. I hope to publish the proof elsewhere shortly. REFERENCES. ' Basch, ‘Jahrb. f. Kinderheil.,’ 1906, p. 64; 1908, p. 68. Calzolari, ‘ Arch. Ital. de Biol.,’ vol. 30, p. 71 (1898). Gellin, ‘ Zeitschr. f. Exp. Path. und Ther.,’ vol. 8, p. 71 (1910). Gudernatsch, ‘Arch. f. Entwick-Mech.,’ vol. 35, p. 457 (1913). Henderson, ‘Journ. Physiol.,’ vol. 31, p. 222 (1904). Hewer, ‘Journ. Physiol.,’ vol. 47, p. 479 (1914). Klose and Vogt, ‘Klinik und Biol. der Thymusdriise,’ Tiibingen, 1910. Marrassini, ‘ Arch. Ital. de Biol.,’ vol. 53, p. 419 (1910). Matti, ‘Ergeb. der Inneren Med. und Kinderheil.,’ vol. 9, p. 1 (1912). Paton, ‘Journ. Physiol.,’ vol. 32, p. 28 (1905). Paton, ‘Journ. Physiol.,’ vol. 42, p. 267 (1911). Paton and Goodall, Tear n. Physiol.,’ vol. 31, p. 49 (1904). Soli, ‘Arch. Ital. de Biol.,’ vol. 47, p. 115 (1907). Soli, ‘Arch. Ital. de Biol., vol. 52, p. 353 (1909). Squadrini, ‘ Pathologica,’ ann. 2, p. 10 (1910). Stotsenburg, ‘ Journ. Anat. Record,’ vol. 7 (1913). VOL. LXXXYVIJI.—B. H The Cultivation of Human Tumour Tissue in Vitro.— Preliminary Note, . By Davin THomson, M.B., Ch.B. (Edin.), D.P.H. (Cantab.), Grocers’ Research Scholar, and JoHn Gorpon THomson, M.A., M.B., Ch.B. (Edin.), Beit Memorial Research Fellow. (Communicated by Sir Ronald Ross, K.C.B., F.R.S. Received April 4,—Read May 14, 1914.) (From the Marcus Beck Laboratory, Royal Society of Medicine, London.) [PLATE 7.] On two occasions the authors have definitely succeeded in cultivating human tumour tissue i vitro, The tissue was obtained at operations performed by Sir John Bland-Sutton at the Middlesex Hospital, and conveyed in sterile Ringer’s solution in a thermos flask to the laboratory, where small portions were immediately inoculated into the culture medium. (a) Intracystic Papilloma of the Ovary (not truly malignant)—This tissue was grown’in a medium composed of fowl plasma 1 part, Ringer’s solution (containing 0°5 per cent, of glucose) 1 part, and extract of the tumour in Ringer’s solution 1 part. On the third day of incubation at 37:5° C. definite buds of new growing tissue appeared. On the fifth day these were more distinct and on,the eighth day the amount of growth had increased considerably (fig. 1, Plate 7). This growth consisted of a solid extension of epithelial cells. As the growth increased it caused some liquefaction of the medium, which was of a gelatinous consistence, and in the more liquefied parts the new growing cells were scattered (fig. 2), but as a rule they remained in contact with each other by means of long fine protoplasmic connections (fig. 3). The new actively proliferating cells varied markedly from the cells of the original tissue planted in the medium. The former were large and amoeboid, with long processes which communicated with each other, and they also contained large highly refractile granules. The original cells, on the other hand, were much smaller ; they showed no amceboid processes, did not exhibit amceboid movement and they contained few or no refractile granules. This tumour was a very soft one and appeared to contain little or no fibrous stroma. It was composed entirely of epithelial cells, and it will be noted that the new growth also consisted of epithelial cells only. (b) Carcinomatous Gland from the Neck (secondary to carcinoma of the floor of the mouth).—Small portions of this tumour tissue grew most success- Roy. Soc. Proc. B., vol. 88, pl. 7. Thomson & Thomson The Cultivation of Human Tumour Tissue in Vitro. 91 fully in a medium composed of fowl plasma 1 part+ extract of embryonic ehick 1 part. Unlike the previous tumour, this one was somewhat tough and fibrous, due to the presence of a considerable amount of connective tissue stroma, and it is interesting to note that the new growth in this case con- sisted of both tissues. After 44 hours’ incubation at 37°5° C., long branching stroma cells appeared growing out from the original tissue. After five days, there appeared in several places solid buds composed of epithelial cells, and these increased in size day by day. Fig. 4 represents a microphotograph of the live tissue after nine days’ incubation and shows clearly the outgrowth of stroma cells and also new buds of epithelial cells of the cancer. Fig. 5 shows definitely a solid outgrowth of cancerous epithelial cells after nine days’ incubation, and fig. 6 shows the marked increase of the same portion after 13 days’ incubation. Growth ceased after 15 days. As in the case of the papilloma of the ovary the new growing epithelial cells were again much larger than the original. They were amceboid and were filled with highly refractile granules. It is interesting to note that these human tumour tissues were cultivated in a medium composed chiefly of fowl blood plasma, or, in other words, the human tissue proliferated in a nutrient material obtained entirely from a bird. This is contrary to what was previously believed, since it was considered that the tissue of a certain animal could only grow in a medium composed of the blood plasma of the same species of animal, Fuller details of this work, and parallel researches on the cultivation of the normal tissues of other animals, will be published in the Proceedings of the Royal Society of Medicine. EXPLANATION OF PLATE. (All the figures represent microphotographs of the live growing tissue.) Fig. 1—Papilloma of ovary ; tissue after eight days’ incubation. Note the outgrowth of processes composed of epithelial cells. x50 diameters Fig. 2.—Another portion of the same after nine days’ incubation. The medium is becoming liquefied and the new cells are somewhat scattered. x 90. Fig. 3.—Higher magnifications of the new growing cells. Note the ameboid proto- plasmic processes and the highly refractile granules. x 870. Fig. 4.—Cancerous lymphatic gland ; tissue after nine days’ incubation. Note the buds of epithelial cancer cells and also the outgrowth of connective tissue stroma cells. x80. Fig. 5.—Another portion of the same, after nine days’ incubation. Note the solid ; outgrowth of epithelial cancer cells. x 80. Fig. 6.—Same portion after thirteen days’ incubation. Note the marked increase of the outgrowth of epithelial cancer cells. x 80. The Cultivation of Human Tumour Tissue in Vitro. 91 fully in a medium composed of fowl plasma 1 part+extract of embryonic chick 1 part. Unlike the previous tumour, this one was somewhat tough and fibrous, due to the presence of a considerable amount of connective tissue stroma, and it is interesting to note that the new growth in this case con- sisted of both tissues. After 44 hours’ incubation at 37°5° C., long branching stroma cells appeared growing out from the original tissue. After five days, there appeared in several places solid buds composed of epithelial cells, and these increased in size day by day. Fig. 4 represents a microphotograph of the live tissue after nine days’ incubation and shows clearly the outgrowth of stroma cells and also new buds of epithelial cells of the cancer. Fig. 5 shows definitely a solid outgrowth of cancerous epithelial cells after nine days’ incubation, and fig. 6 shows the marked increase of the same portion after 13 days’ incubation. Growth ceased after 15 days. As in the case of the papilloma of the ovary the new growing epithelial cells were again much larger than the original. They were amceboid and were filled with highly refractile granules. It is interesting to note that these human tumour tissues were cultivated in a medium composed chiefly of fowl blood plasma, or, in other words, the human tissue proliferated in a nutrient material obtained entirely from a bird. This is contrary to what. was previously believed, since it was considered that the tissue of a certain animal could only grow in a medium composed of the blood plasma of the same species of animal. Fuller details of this work, and parallel researches on the cultivation of the normal tissues of other animals, will be published in the Proceedings of the Royal Society of Medicine. EXPLANATION OF PLATE. (All the figures represent microphotographs of the live growing tissue.) Fig. 1—Papilloma of ovary ; tissue after eight days’ incubation. Note the outgrowth of processes composed of epithelial cells. x50 diameters. Fig. 2.—Another portion of the same after nine days’ incubation. The medium is becoming liquefied and the new cells are somewhat scattered. x 90. Fig. 3.—Higher magnifications of the new growing cells. Note the amceboid proto- plasmic processes and the highly refractile granules. x 870. Fig. 4.—Cancerous lymphatic gland ; tissue after nine days’ incubation. Note the buds of epithelial cancer cells and also the outgrowth of connective tissue stroma cells. x80. Fig. 5.—Another portion of the same, after nine days’ incubation. Note the solid outgrowth of epithelial cancer cells. x 80. Fig. 6.—Same portion after thirteen days’ incubation. Note the marked increase of the outgrowth of epithelial cancer cells. x 80. VOL. LXXXVIIL.—B. \ift 92 Trypanosome Diseases of Domestic Animals in Nyasaland. Trypanosoma capre (Kleine), Part III.—Development in Glossina morsitans. By Surgeon-General Sir Davin Brucs, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 18, 1914.) [Puate 8.] INTRODUCTION. In a previous paper* the morphology and action on animals of this species of trypanosome were described. In this is given an account of its develop- ment in Glossina morsitans. Trypanosoma capre belongs to the 7. vivax group, in which the develop- ment of the trypanosomes is restricted to the proboscis. THE DEVELOPMENT OF T. CAPR IN G. MORSITANS. Six experiments were made with laboratory-bred flies. Five were positive and one negative. Table I.—Laboratory-bred Flies. No. of | Experiment ; No. of days Date. | HExpt.| flies positive or Nong ee before flics t hie. used. negative. une- | became infective. | "°™Peravunre- 1912. April16| 444] 12 # il 16 71° F. (22'1°C.) June 3] 617 33 — (0) — 65° F. (18'3° C.) » 9&| 1215 22 ar 1 21 65° F. (18°3° C.) 1913 Jan, 18| 1777 | 35 4 11 19 84° F. (288° a » 22] 1784 35 + 20 19 84° F. (28'8°C. April 1} 2046 33 + 13 20 84° F\. (28'8° C.) One hundred and seventy laboratory-bred flies were used and forty-six infected flies were found—27:1 per cent. The first three experiments were carried out at the ordinary temperature of the laboratory ; in the last three the cages containing the flies were kept in an incubator. It is difficult to understand the difference in the number of infected flies found. In Experiments 444 and 1215 only 8 and 5 per cent. respectively of the flies became infected, whereas in the last three experiments, an average of more than 40 per cent. was found. The flies in the second group were kept, it is * “Roy. Soc. Proc.,’ B, vol. 86, p. 278 (1918). Trypanosome Diseases of Domestic Armals mm Nyasaland. 93 true, at a temperature similar to that which they would find in summer in the low country, while the first three experiments were done in winter and at the ordinary temperature of the laboratory. This no doubt would explain the difference to some extent. Again, goats and sheep infected with | T. capre are unsatisfactory animals to feed flies on. One day the try- panosomes are present in small numbers in the blood, the next day it may be impossible to find any; very seldom are they in any numbers. It is quite possible, then, that flies may feed on an infected goat or sheep without taking in a single trypanosome. Details of the Sia Experiments. Five Positive, One Negatove. The following table gives the principal details in carrying out the six experiments. Laboratory-bred flies were used in all. Table II. Expt. Day of expt. Procedure. Remarks. 444, 1-4 12 flies fed on infected | Trypanosomes appeared in blood of Goat 339. Goat 419 after 23 days. All flies 5-6 Starved. dissected; 1 infected fly found. 7-24 Fed on clean Goat 419. Goat 339 contained few trypano- somes in its blood. 617 1-4 33 flies fed on infected | Trypanosomes never appeared in Sheep 347. blood of Goat 628. All flies dis- 5 Starved. sected; all negative. Sheep 347 6-63 Fed on clean Goat 628. was unsatisfactory; one day its blood contained a few trypano- somes, the next day none. 1215 1-3 22 flies fed on infected | Trypanosomes appeared in blood of Goat 979. Goat 1219 after 28 days. All flies 4 Starved. dissected ; 1 infected fly found. 5-29 Fed on clean Goat 1219. 1777 1-5 35 flies fed on infected | Trypanosomes appeared in blood of Goat 1746. Goat 1803 after 26 days. All flies 6 Starved. dissected ; 11 infected flies found. 7-27 Fed on clean Goat 1803. 1784 1-4 35 flies fed on infected | Trypanosomes appeared in blood of Goat 1746. Goat 1812 after 26 days. All flies 5 Starved. dissected ; 20 found infected. 6-27 Fed on clean Goat 1812.) 2046 1-5 33 flies fed on infected | Trypanosomes appeared in blood of Goat 1912. Goat 2102 after 27 days. All flies 6 Starved. dissected ; 13 found infected. 7-18 Fed on clean Monkey 2066. 19-20 Starved. 21-23 Fed on clean Goat 2102. 24-25 Starved. 26-36 Fed on clean Monkey 2066. 94 Sir D. Bruce and others. Trypanosome It would appear from the five positive experiments that an average period of 19 days elapses before the cycle of development ot 7. caprw is complete in G. morsitans and the fly becomes infective. RESULT OF THE DISSECTION OF THE INFECTED FLIES. Table I1].—Laboratory-bred Flies. Positive Experiments. Time, 3 Proventri- Fore- Mid- Expt. days Proboscis. eae Crop. ob aus 444, 25 + = = = = 1215 32 + = = = — 1i77 7Al + — = — - 1777 26 + _ = - — 1777 30 + = = = — 1777 30 + =_ = — - WEY 30 + - = — - 1777 30 + _ = — — 1777 30 + = = = _ 1777 30 + — = _ - 1777 30 + — = — - 1777 30 + - = - _ 1777 30 + = = = = 1784 19 + — = = _ 1784 21 + = = = = | 1784 23 + = = = = 1784. 24; + = = = = 1784 29 + - = = = | 1784 | 29 + = - _ = 1784 29 + = = = = 1784 29 + = = = = 1784 29 oF = = = = 1784 29 + = = = a 1784 29 + = = = — 1784 29 + = = = 1784 29 + = = = = 1784 30 + = = = = 1784 30 + = = = = 1784 30 + = = = = 1784 30 + = = = = 1784 30 + = = = = 1784 30 + = = = od 1784 30 + = = i = Labial | Hypo- cavity. | pharynx. 2046 | 238 + + = = = 2 2046 | 23 + - = = = = 2046 | 24 + ++ = = = = 2046 | 26 + + ++ = = = = 2046 | 28 £ = = = = 2046 | 28 + = - = - 2046 29 ++ aa = = = = 2046 | 29 + ++ = = = = 2046 29 + ++ = = = = 2046 | 29 + ++ = = = = 2046 29 aP ar ar ar = = = = 2046 29 ++ + = = = = 2046 30 + ++ = = = = Hind- gut. Diseases of Domestic Animals in Nyasaland. 95 It will be seen from the above table that it was not until the last experi- ment that the labial cavity and hypopharynx were examined separately. In the previous experiments the presence or absence of trypanosomes in the proboscis as a whole was noted. In the first two experiments, only a single infected fly was found in each. In Experiment 1777, 11, and in 1784 as many as 20 were found. In regard to the number of trypanosomes in the labial cavity, this may vary greatly. Sometimes the lumen of the tube will be seen to be densely crowded ; at other times a single colony will be seen. For example, in Experiment 1777 the first infected fly, dissected on the 21st day, is noted to have had the lumen of the proboscis swarming with clusters of torpedo- shaped flagellates attached to the labrum by their flagellar ends, a few swimming free. In the seventh infected fly, dissected on the 30th day, only three colonies, in the eighth one, and in the ninth two small colonies, are noted. In the same way the hypopharynx may contain few; at other times it is seen to be densely packed with swarms of actively moving trypanosomes. In unstained specimens the difference in size and shape between the trypanosomes in the labial cavity and those in the hypopharynx is quite manifest. It may be stated here that, exceptionally, flagellates may be seen in the cesophagus, or that part of the alimentary tract anterior to the pro- ventriculus. Among the 46 flies described above, this was noted twice. In the first instance they are reported as being very scanty, in the second as being active and in large numbers. But from Table III the broad fact stands out boldly—that in this species of trypanosome the development is confined to the labial cavity and hypo- pharynx, and does not take place in any other part of the fly. Tue Type oF TRYPANOSOMES FOUND IN THE INFECTED FLIES. No attempt has been made by the Commission to study the development of 7. capre in G. morsitans in the earliest stages. This can only be done if a large number of laboratory-bred flies are available, and this was not the case at Kasu. Plate 8 represents some of the developmental forms found in the labial cavity and hypopharynx of infected flies. Fig. 1 represents a torpedo-shaped organism taken from a single cluster growing near the bulb on the 19th day after the first infected feed. Figs. 2 and 3 are similar shaped flagellates, also from a single group growing near the bulb on the 21st day. Figs. 4-10 are drawn from 24-day flies. 96 Trypanosome Diseases of Domestic Animals in Nyasaland. Figs. 11-19, 29 days. Fig. 19 has an encysted appearance. Figs. 20-22, 30 days. It will be seen that most of the flagellates found in the labial cavity are crithidial in type. They are generally ribbon-shaped, with well-defined nuclei and micronuclei and free flagella. Figs, 23-30 are from the hypopharynx and have been obtained, as a rule, by causing the fly to salivate on to a cover-glass. They represent the final stage in the cycle of development—the reversion to the infective or “blood form.” They are smaller than those found in the blood of the vertebrate host, but resemble them closely in every other way. CONCLUSIONS. 1. Trypanosoma capre is capable of passing through a cyele of develop- ment in G. morsitans, the flies becoming infective some 19 days after feeding on an infected animal. 2. Trypanosoma capre belongs to the same group as 7. var and T. uniforme, the development taking place only in the proboscis. 3. The final stage of the development takes place in the hypopharynx where the trypanosomes revert to the original “blood form” and become infective. DESCRIPTION OF PLATE. Figs. 1-3.—Common type of torpedo-shaped flagellates found attached in small single groups or clusters to the labrum, near the bulb, after 19 to 21 days. Figs. 4-22.Various other developmental forms found in the labial cavity in flies dissected 24 to 30 days after their first infected feed. They are mostly crithidial in type. Figs. 23-30.—“ Blood forms” from the hypopharyux. These represent the final stage in the cycle of development. Stained Giemsa. x 2000. OAL ET ag AIRS Re EF APRs UE IY SOO ov Trypanosoma capre Development ti Glossina morsitares. X 2000. a. ee. 97 The Trypanosome causing Disease iw Man in Nyasaland: The Inwonde Strain. Part 1—Morphology. Part I1.—Suscepti- bility of Ammals. By Surgeon-General Sir Davip Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 25, 1914.) INTRODUCTION. This strain was obtained in the “ fly-area” of the Upper Shiré Valley, in the Liwonde district, which is situated about 100 miles south of the “ Proclaimed Area.’* At the time it was procured no cases of trypanosome disease in man had been reported from this district ; lately, a case has been discovered at Mpimbi, in the south of the district, about 150 miles south of the “ Proclaimed” or Sleeping Sickness Area. Three dogs infected with the wild Glossina morsitans strains of this trypanosome were brought to Kasu, and other animals—monkeys, dogs and rats—were inoculated from them. For purposes of description, measurement and comparison, only trypano- somes from rats were used. I. Morphology of the Liwonde Strain. A. Living, Unstained. All three strains agree in being actively moving flagellates, but with little or no translatory movement. B. Fixed and Stained. The blood films were fixed, stained and measured as previously described in the ‘ Proceedings.’f Before proceeding to describe the three strains in detail, it may be stated here that in general appearance, shape, position of nucleus, size of micronucleus, contents of cell, and undulating membrane, all three strains were similar, and in no way differed from the various strains of the trypanosome causing disease in man in Nyasaland which have already been described. * “Trypanosomes Found in Wild Glossina morsitans and Wild Game in the ‘ Fly-belt’ of the Upper Shiré Valley,” ‘Roy. Soc. Proc.,’ B, vol. 88, p. 38 (1914). + B, vol. 81, pp. 16 and 17. 98 Sir D. Bruce and others. Trypanosome MORPHOLOGY OF THE LIWONDE STRAIN I. Length.—The following table gives the length of this trypanosome as found in the white rat—500 trypanosomes in all. Table I.—Measurements of the Length of the Trypanosome of the Liwonde Strain I. ne In microns. 0. Date. of Animal, eee of ae thod of expt fant Re, Average | Maximum | Minimum length. | length. length. 1913. Aug. 18...) 2878 | Rat ............ Osmic acid | Giemsa 24 °8 28-0 19-0 DT GSH V287 8 in OS Ay Be B i: 24:3 27-0 21-0 OT Bia |82 3 (On |e aee nea . 24 °6 27-0 22-0 5) 19...| 2378 Cpe seeemapaeanen 59 di 23 °4 27 0 200 fe, Oi | SST ONlnmene ten ans i 23-4 27-0 20 ‘0 D 19...) 2378 Spot npbaeeanneasD £5 5 24:7 27-0 Zio) 55 20...| 2378 Sol, Mbboomennnaees 29 mH 22 °3 25 °0 19 ‘0 a On ORE cake ae a 21-4. 25-0 18-0 9 20...| 2378 Bee See sesn aes “5 is 21°3 24, °0 18-0 5 21 2378 Sa. pooasaabonne 36 5 22-1 26 0 16:0 col OTD Te glh beeen eet is i 22-1 26 -0 19-0 ia oes DO TOR la meen metas leet is * 21-2 25-0 18 ‘0 ‘ 22...) 2378 Bid apetoaneD eS » oH 23:3 28-0 20°0 +AU), SOOM OS Om Mee cnet i i. 231 26-0 20-0 a 22 2378 pf aadbeseaaecs 59 50 23 °2 28-0 20°0 99 23 2378 By Heaeeeseiaee 55 0 23 °3 28 -0 21°0 hat OS URANO RI SM ie mecid ee at 20 a _ < 24-0 28 0 20-0 RRO Oc) Hola Ream ‘ hove 24-2 30-0 20-0 5 24...| 2378 Hy) eoaaadeadoro 2» 23 -0 28 -0 20 °0 ry 24...| 2378 Bp) laddepbotanan 3 6 23 -1 29-0 19-0 ial. (OATS 7S My Ai Rabi oe Sut onl 5 ‘, 229 25-0 190 3p 25...| 2378 Bint cod son deotne PS p 225 27 °0 18-0 op 25...| 2378 fH \aasseaoddeds yD 99 220 26-0 18-0 RN Ih ORIIS. | ys heen i: i 22:0 250 18-0 s 26...| 2378 opt idence spores 3 es 21°8 24-0 18 °0 ] 230 30-0 16 ‘0 ! causing Disease in Man in Nyasaland. 99 Cuart 1,—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Liwonde Strain I, taken on Nine consecutive Days from Rat 2378. ia [3 [it [iS [v6 17 [ie [ro [zo|at[2a[23]24]o5[a6lo3 |28]29]30] 51 [52 [53 ge es i JE es AS Uy v Aes) Gs) L rs i] 0) & o oa This curve differs from the ordinary Wild G. morsitans curve, but is similar to Strains II, IV and V of the Human strain.* Breadth.—The following table gives the breadth of this trypanosome in the rat—500 trypanosomes in all. Table I1.—Measurements of the Breadth of the Trypanosome of the Liwonde Strain I, measured across the Widest Part, including the Undulating Membrane. In microns. Experiment Animal Number if Nie. AERA Average Maximum Minimum breadth. breadth. | breadth. at ean: - 2378 I NEY ieatenpaededs 500 30 | 4°50 | 1°25 * “Roy. Soc. Proc.,’ B, vol. 86, pp. 288, 295, and 298 (1913). 100 Sir D. Bruce and others. Trypanosome Table I1]—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Liwonde Strain I. Dee aso Wei Percentage among short 0, and stumpy forms. 1913. Aug. 18......... 2378 TRAYS Soooonon0 2 Meret Ore ee 2378 be eral ae 23 op BD sotssoonee 2378 im batter 17 By atceoben 2378 pp! sdodepanno 4 59 ZBroneaacee 2378 piu, Banas 12 $9, ZBoranoonn0 2378 fp) eannqouerp 16 PW A cece 2378 Bp Bansbauth 28 6 Z2Don0006000 2378 Sp; onbueende 15 pis Onee- ibe 2378 pA. Sondaaden 19 PN ZU Baonebons 2378 Fy), dd00no00p 22 Average.:....... 15 °8 MorPHOLOGY OF THE LIWONDE STRAIN II. Length.—The following table gives the length of this trypanosome as found in the white rat—500 trypanosomes in all :— Table [V.—Measurements of the Length of the Trypanosome of the Liwonde Strain IT. im In microns. Date of | Amimal, | Method of | Method of ; = } : fixing. staining. | Average | Maximum | Minimum ext: length. | length. | length. 1913 Aug. 18...) 2363 | Rat ............ Osmic acid Giemsa 20 °3 31:0 17:0 ety 1Gce 25651 eee wen i e 19-0 23-0 14-0 ETS ts POSGan hee ae en _ 3 20°3 26 ‘0 18-0 i GE oaGael) ae ae eee is x 19 2 25 0 180 Re Deal Oe MIO 5 GOs | eee ay eens ‘ é 206 25-0 17-0 CPO PAPO S Gaal, he nennnee le ‘1 ‘ 19°7 24-0 170 9 ADacall ZBISB |p), © son ond 600020 0 5 20-7 25 0 18 0 LR IESTON al ORY |r cn Ae ir et, is i 21-2 29 0 18 ‘0 Mo 0s PEECBG eel bts of is x 20-0 24-0 17-0 ORO SGA ie, tale ae cee ¥ - 190 24-0 16-0 oF, Vel 2AGSal sec ese s i 19-9 28 -0 13-0 pd Ee IMZOO OI ete ene aceeeris 99) i 19°8 260 17-0 pe Pacall 283 ||" by peoeesnaas00 a a 219 27 ‘0 18-0 eH OOM D Goan atau rn: - is 21°9 28 0 18-0 erg POR GET ht ma een ate is ‘ 22-9 27-0 200 GO) (SSHALDaGae ln phika leas sau i ‘ 22-6 33 0 18-0 POO ata NOSGS MN meas 2 We i ¢ 22.9 34:0 18-0 seer D3 mal GSES al ha see eee te 4 a 21-0 30:0 17-0 bey TIA ORY EY NCR MeN ea i f 20°9 35-0 17-0 oe DA PANOSG OM tee ane ta es * i 22-2 31-0 16-0 Boca EER) pe, obsonsdoood 5p 9p 20°1 31:0 16°0 5 i SOSA ew ate aae 3 is 20-9 28-0 16-0 Se 5 snl MOG63 @ uk peer es is be 20° 28 -0 15-0 HOOP O LTA Sige Shaun ee is is 20 6 27-0 12-0 » 26...) 2863 99. eokGon000000 9 7p 20°1 31°0 17-0 20°7 35 0 120 causing Disease in Man in Nyasaland. 101 Cart 2.—Curve representing the Distribution by Percentages, in respect to Length, of 500 Individuals of the Liwonde Strain II, taken on Nine consecutive Days, from Rat 2363. | | Sari ren Seat ate aa - Breadth.—The following table gives the breadth of this trypanosome in the rat—500 trypanosomes in all :— Table V.cMeasurements of the Breadth of the Trypanosome of the Liwonde Strain II, measured across the Widest Part, including the Undulating Membrane. In microns. Experiment cena Number No. are measured. Average Maximum Minimum breadth. breadth. breadth. 2363 Rati occ scsessan 500 3°0 | 4°75 | 1°25 102 Sir D. Bruce and others. Trypanosome Table VI.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Liwonde Strain IT. Date: ae ae Asante Percentage among short 0. and stumpy forms. 1913. JHE, WE) coconaco 2363 Rate eee | 25 We Os eee 2363 poinaeeaonalas 13 OO ean | 2363 ie tie 14 1 a acseeseen | 2363 sae Beowaneeee 16 9 Le scanso000| 2363 Pie eraceance | 12 Fp enaneeere | 2383 Say Rctattact | 23 1), UDA Sis acrean 2363 OOF ese esse 35 Ph ZO science | 2363 pi aceoes: 37 Ie cheonat sth | 2363 See ee 14. ie Lerawtoas | 2363 bya hieadetton 23 Averagers...---: 21-2 MOorPHOLOGY OF THE LIWONDE SrraIn III. Length—The following table gives the length of this trypanosome as found in the white rat—500 trypanosomes in all :— Table VII.—Measurements of the Length of the Trypanosome of the Liwonde Strain III. N In microns. Date of Animal wienagoal or icing! os ; ; : Bea } fixing. staining. | Average| Maximum | Minimum 1p length. | length. length. 1913. ANT, WSs) ZEVO || TR 200 ccoena ace Osmic acid | Giemsa 22-9 29°0 18-0 TFT, A287" | ant oth ere eS 3 22-1 30-0 17-0 Salen oar" (ace. seen ceaeen 4 S 22-6 31-0 18-0 ft S19) 370) | aes ne ane 5 ‘ 19°7 24-0 18-0 | ina ab pel 9/2370! |e ep ene - 198 23-0 18-0 on (Os 2370) |e eee a t 19 ‘9 28-0 18-0 UT 20) Nioa70\) be eee! es 20-9 25-0 19-0 ake 2056 MOTO ie oe teeee * én 21-2 30-0 18-0 Seo DOW LOS TON 5 eee i x 226 32-0 19-0 Gens Wel (ORT (T) oN anata ehe tin He - i 20-0 24-0 17 ‘0 Mag DUELS 70.0\Ku). shes lien i fs 21-0 28 -0 17-0 Seo MOS TOu hn 0 ae nene ‘s is 20°3 28 0 17 ‘0 sel Peal BUI) leapt castes nee “ J 19 °9 29 -0 17-0 SAM nrOomal oir) h(i i i a 201 25-0 15-0 aie VOD MEDS ZO) sit «see nee i a 20 °5 29 -0 17 ‘0 28 aoa z0d| ieee nie sae ‘i ig 23-7 300 19-0 Same sel 82 570) eee Meee eh is . 23-9 32-0 18-0 Ke SOB AOR YOM eae wan iameees is - 23-0 300 17-0 Pe OUM oar ee canine 3 ‘ 21°7 33-0 17-0 reg OGR BOS70 iltgemeeres fea aa is “ 23-0 33-0 18-0 G4 ROL RALOIA (OP || cpu Ay i 22-2 27-0 18-0 fe D5 Cea 704 (pan ee i i 21°7 31-0 18 ‘0 potas ROBYN pine Baar A i i 21-0 30-0 18-0 ead BM) hs “Goagaseusces p 51 20 °9 29-0 17°0 Basal y(n lhe ioe team ea ks f 19°7 26-0 16-0 21 °4 33 °0 15 0 causing Disease in Man in Nyasaland. 103 CuArtT 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Liwonde Strain III, taken on Nine consecutive Days, from Rat 2370. MiicTrons [ie [rs ]is [ip [1a [9 [a0 [21 [2>.]23 [24 25 [26] 27 [28] 29130 [ot [32 ]33] 54135 [36]37|>8 | PEEEEEISECCEEEEEEEEEEEEEEE te | aS = SASS eee ee ene HERES a a eee ee eee pf-LS z eS Se ee aes A Se —j- —H a See oS eases [aes ee Seneat i aoe a a Ea PEELE EERE eat EPP Sas Bae Loe POPE G5 AAPRREW Een ace eeeeseee JOE eens Breadth.—The following table gives the breadth of this trypanosome in the rat—900 trypanosomes in all :— Table VIII.—Measurements of the Breadth of the Trypanosome of the Liwonde Strain III, measured across the Widest Part, including the Undulating Membrane. In microns. Experiment Astin Number — No. i measured. Average breadth. breadth. breadth. Maximum | Minimum 2370 TBENB Goaenoconnbe 500 2°6 | 4°75 | 1°25 104 Sir D. Bruce and others. Trypanosome Table [X.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Liwonde Strain III. Experiment < Percentage among short 8: nel sation. and. sbemipy (os 1913. | AMES U8 sonsea00 2370 Rat aeeaes| 4 SO BAO heae 2370 wd Tae Oe 44, pr eeO Leste 2370 }- obagye beers | 36 | 99. | Bibosocsesae 2370 Pee oetod | 33 Sra 05) ea 2379 pa So | 22 Sit Wace oES 2370 lei 8 4 1 Gis ee Oy oer a 2373 ses OB | 6 [Pe Pee Direetemen. 2370 PA anc ceacer 12 Wer ene BOOK emer 2370 Sv ers ers 30 i Oy ae Ae 2370 os oe io 33 | Average......... 22-4 COMPARISON OF THE LIWONDE STRAINS WITH ONE ANOTHER. Table X.—Measurements of the Length of the Trypanosome of the Liwonde Strains. In microns. E ; Number of Date. a ~ | Strain. | Animal. | trypanosomes et eae measured. Average | Maximum| Minimum length. length. length. 1913 2378 I Rat ...... 500 230 30°0 16 °0 1913 2363 Il (Rabie 500 20°7 35 ‘0 12°0 1913 2370 iil Rat) ecce. 500 21°4 33 0 15°0 PLT 35 °0 12°0 causing Disease in Man in Nyasaland. 105 Cuart 4.—Curve representing the Distribution, by Percentages, in respect to Length, of 1500 Individuals of the Trypanosome of the Liwonde Strains. Mi cromns 12 [03 [14 [15 [16 ]17 [18 [19] 20 [21 [22] 23 [24 [25126 | 27 |2a]29 130] 31 | 32] 33]34 [35] 36 137] 81 i SEQEESSMEEREE UE || 252 See ee IEEE EEE EEE 4 Ber see emiseeer eer ttt ttt ea | a | ea er a a pa oD ee ee a a LN Stele era sd) oe Ne le ee Seen Eee eeeeeeee Table XI.—Measurements of the Breadth of the Trypanosome of the Liwonde Strains. In microns. enone Number of Date. 2 N Strain. | Animal. | trypanosomes | | ra aera | measured, Average | Maximum) Minimum | | | breadth. | breadth. | breadth. | | | 1913 2378 I TEA open: 500 30 4°50 | 1:25 | 1913 2363 II Rat ...... 500 30 4°75 | 1:25 1913 2370 Til Rat <..... | 500 2°6 4°75 | 1325 2:9 4°75 1°25 106 Sir D. Bruce and others. Trypanosome Table XII.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Three Liwonde Strains. Date. Experiment No. Strain. Animal. Percentage ea and stumpy forms. 1913 2378 I Rat 15°8 1913 2363 Ii F 21-2 1913 2370 III ‘. 22-4 Average......... 19-8 COMPARISON OF THE LIWONDE STRAIN WITH THE HUMAN, WILD-GAME, AND WILD GLOSSINA MORSITANS STRAINS OF THE TRYPANOSOME CAUSING DISEASE IN Man IN NYASALAND. Table XIII.—Average Length of the Trypanosome of the Human, Wild- game, Wild Glossina morsitans, and Liwonde Strains. In microns. Number of Strain. trypanosomes Animal. measured. ° Average | Maximum | Minimum length. | length. length. Ja Ub HN : ngopocodeanobanuadaco 5500 Rat ......| 23°5 38 0 14:0 Wild-game.................. 2500 By hfatmeawal et22icG, 35 -0 15 °0 Wild G. morsitans ...... 2500 pee oot nae 226 35-0 15:0 Miwondel cn. Sessee.geccet: 1500 ebmercboe: 21°7 35 °0 12:0 22°6 38-0 120 causing Disease in Man in Nyasaland. 107 Cuart 5.—Curve representing the Distribution, by Percentages, in respect to Length, of 1500 Individuals of the Trypanosome of the Liwonde Strain, 50U0 of the Wild- game and Wild Glossina morsitans combined, and 5500 of the Human Strain, all measured from Rats. Liwornde Strain. =ecoe Wild Garne & Wild G morsitans — —-— Human Straer Table XIV.—Average Breadth of the Trypanosome of the Human, Wild- game, Wild Glossina morsitans, and Liwonde Strains. In microns, Number of Strain. trypanosomes Animal, 3 measured. Average Maximum | Minimum breadth. breadth. breadth. JSRRETEN Soeoscpoousohse 1500 Ratipeeescne: 2-6 50 1°25 Wild-game............ 1500 sede igen dit 3-2 5°75 1°50 Wild G. morsitans 1500 Siiisacetaee 2°9 5°25 1°25 IU ROC vocceocseooaee 1500 a neiecmaes 2°9 4°75 1°25 2°9 5°75 1°25 VOU. DXXXvili.—F K 108 Sir D. Bruce and others. Trypanosome Table X V.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Human, Wild-game, Wild Glossina morsitans, and Liwonde Strains. al Date. Strain. Animal. Janeen ge HGS SLOT! and stumpy forms. 1912 JEQVEAEY 5c ecc0coo020000 Rat......... 17°8 1912 Wald= gamer arenas. ns ubteenenee 26 °2 1912 Wald SG S7no7,suic7isiee aera ee 12°5 1913 Thiwondeleeneresetee Pesan oRe ao 19°8 Average......... 19 ‘1 II. Animals susceptible to the Trypanosome of the Liwonde Strain. The following tables give the incubation and duration of the disease in goats, monkeys, dogs, guinea-pigs, and white rats, which were sub-inoculated with Strains I, II, and III from dogs infected in the Liwonde district :— Liwonde Strain I. Table I. No. of Period of | Duration Date. et t Source of virus. | incubation, | of disease, Remarks. san in days. | in days.* Goat. 1913. Aug. 12...| 2382 | Monkey 23840... 6 52 | Died of Strain I. Wal) CREB OE o340) 6 44 F > Monkey Aug. 7...| 2839 | Dog 2322 ...... 4 27 Died of Strain I. oy . Wooo) PRED Weg LB2B sccoo 4 2 eae i cs Dog. July 16...| 2322 | Wild flies ...... 6 23 Died of Strain I. Aug. 12...| 2380 | Monkey 23840... 6 46 D - so DP Agec| | IIL esto 6 45 & z Guinea-pig. Aug. 12...{ 2384 | Monkey 2340... 13 78 | Died of Strain I. pe 12) 2386 mie os40nl)) Hes 114 5, y Rat. Aug. 12...| 2378 | Monkey 2840... 6 35 Died of Strain I. 5 12/2379 ietoe40 6 44 y i * Duration includes the days of incubation ; it dates from day of inoculation. causing Disease in Man in Nyasaland. 109 Liwonde Strain LI, Table IT. NOWOR 3 Period of | Duration Date. o3 ‘ Source of virus. | incubation, | of disease, Remarks. eae in days. | in days.* Goat. 1913. | Aug. 12...| 2366 | Monkey 2336... 6 39 Died of Strain IT. » 12...| 2367 » 2836 ... 9 41 B Z Monkey. Aug. 6 2335 | Dog 2323 ...... 5 82 | Died of Strain IT. . 2336 5 OREN Baek 5 43 is is Dog. July 5...) 2353 | Wild flies ...... 3 — Killed July 28. » 28...| 2323 | Dog 2353 ...... 9 17 Died of Strain IT. Aug. 12...) 2364 | Monkey 2886... 6 23 FS BS » 12...| 2365 R836 6 27 - _ Guinea-pig. Aug. 12...) 2368 | Monkey 2336 ... 49 94. Died of Strain IT. enee | 2369 | as 2336 ... — — Never showed trypanosomes. Rat. Aug. 12...| 2391 | Monkey 2336... 5 22 Died of Strain IT. Reo 23863 Wy 28560. 6 21 i y Liwonde Strain ITT. Table III. No. of Period of | Duration Date. a Source of virus. | incubation, | of disease, Remarks. i in days. | in days.* Goat. 1913. | Aug. 12...| 2874 | Monkey 2338 ... 6 10 Died of Strain ITI, 1 Whed| PBS 93388) 4) 9 12 i a Monkey. Aug. 6...) 2337 | Dog 2321 ...... { 5 32 Died of Strain III. Peer ONecasae ies Osa 5 38 | ms f Dog July 8...| 2356 | Wild flies ...... 4 = Killed July 28. » 28...| 2321 | Dog 2356 ...... 9 16 Died of Strain ITI. Aug. 12. 2372 | Monkey 2388 ... 6 39 BA “ » 12...| 2373 i Pea 6 12 fi i Guinea-pig. Aug. 12. 2376 Monkey 2388 ... 20 39 Died of Strain III. Reina he Ae wo osag ly 78 93 st # Rat. Aug. 12...) 2370 | Monkey 2338 ... 6 20 Died of Strain III. i Uaodll BRYA 90) 2838 6 17 it fi * Duration includes the days of incubation ; it dates from day of inoculation. i Wy 110 Sir D. Bruce and others. Trypanosome Disease Set Up in Various Animals by the Trypanosome of the Liwonde Strain. The infection set up in the various animals by the.Liwonde strain gave rise to symptoms and appearances during life, and pathological changes in the various organs after death, alike and similar in évery way to those caused by the Human strain, Wild-game strain, and the Wild G. morsitans strain of T. brucer vel rhodesiense found in the “ Proclaimed Area.” COMPARISON OF THE THREE LIWONDE STRAINS IN REGARD TO THEIR VIRULENCE TOWARDS VARIOUS ANIMALS. Table IV.—The Average Duration, in Days, of the Disease in various Animals. : Strain. Goat. | Monkey. Dog. Guinea-pig. White rat. a aaerpecccen 48 49 38 96 39 1 bes eec 40 37 22 94 21 METS aeees 11 | 35 22 66 18 No recoveries took place among the experimental animals. Table V.-The Average Duration of Life, in Days, of various Animals infected with the Liwonde Strain. | Goat. | Monkey. | Dog. Guinea-pig, | White rat. Average duration, in days ... 33 41 28 84 26 Number of animals employed 6 6 9 5 6 COMPARISON OF THE LIWONDE STRAIN WITH THE HUMAN STRAIN. Table VI—The Average Duration of Life, in Days, of various Animals infected with the Human and the Liwonde Strains. The letter R stands for “refractory.” Strain. | Ox. Cee Baboon. | Monkey.) Dog. | Rabbit. Seca 1 Average dura- | Human | 134 42 R. 26 34 28 67 30 tion, in days. Average dura- | Liwonde| — 33 _— 4] 28 a 84 26 tion, in days. causing Disease in Man in Nyasaland. 111 CONCLUSIONS. 1. The three wild G. morsitans strains from the Liwonde district resemble each other closely, and all belong to the same species of trypanosome. 2. The Liwonde strain belongs to the same species as that occurring in man, wild game, and wild G. morsitans inhabiting the “Proclaimed Area,” Nyasaland—7. brucei vel rhodesiense. 3. Hence it would appear that wild G. morsitans occurring in a district 100 miles south of the “ Proclaimed Area” are infected with the trypanosome which causes the human trypanosome disease of Nyasaland. The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part I1.—Morphology. By Surgeon-General Sir Davin Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 7,—Read June 25, 1914.) [Piares 9-11.] INTRODUCTION. This strain differs so much from the others that it is doubtful if it should be included among the various strains already described, Human,* Wild- game.f Wild Glossina morsitans,* Mzimba,§ ete. It has only been found on three occasions and, curiously enough, each time in a native dog. The three dogs suffering from trypanosome disease were brought up to Kasu from the “ Proclaimed Area,” where they had probably been naturally infected by the wild G. morsitans, hence the name “The Naturally Infected Dog Strain.” All the infected dogs coming from this area did not show this strain ; for example, Dog 553 was infected with a trypanosome resembling the ordinary Human strain. e * “Roy. Soc. Proc.,’ B, vol. 85, p. 423 (1912), and vol. 86, p. 285 (1913). + Ibid., B, vol. 86, p. 394 (1918). t Ibid., B, vol. 86, p. 408 (1918). § Ibid., B, vol. 87, p. 26 (1913). 112 Sir D. Bruce and others. TJrypanosome ° If this Naturally Infected Dog strain had been found in the blood of the wild game and in the wild G. morsitans, then it would have been legitimate to make a new species of it. But it would be unjustifiable to make a new species of a strain which, up to the present, has only been found in three chronically infected dogs. The Commission have therefore decided to consider this strain as belonging to the species described as the Trypanosome causing Disease in Man in Nyasaland—Trypanosoma brucei vel rhodesiense— and not as a new species. If this is correct, then it is curious how much a species can vary in disease-producing power. For example, it will be shown that this Naturally Infected Dog strain is almost harmless to monkeys and guinea-pigs, whereas the parent species kills these animals without fail. Not only does it differ in virulence, but even its morphology is apparently somewhat changed. There is a comparative absence of the blunt-ended posterior-nucleated forms, which are sometimes so marked a feature in the parent species. Not that they are altogether absent, but they are not so prominent, do not strike the eye so readily. It will therefore be interesting to describe this strain as fully and completely as possible. causing Disease in Man in Nyasaland. MorrPHOLOGY oF THE NATURALLY INFECTED DoG STRAIN. 113 STRAIN I. Dog 48. Table I—Measurements of the Length of the Trypanosome of Naturally Infected Dog. Strain I. Dog 48. In microns, Date of. Necrat Method of | Method of : rapa : r fixing. staining. | Average} Maximum| Minimum Sees length. | length. length. 1912. Heb ysemplierets |) LOM Ox Os ese ceak es Osmic acid | Giemsa 25 °5 31:0 16:0 pense OG male Shee piercer 5 es 26 °4 30°0 21:0 dat, BD onal. G3 1) IDe es Socosscesce ws s 23 °6 29-0 19-0 » 24...) 48 Sh eee S < 24 °2 31:0 16°0 Hemi ABh | oo ee s ss 25-7 35-0 15-0 » 15...) 140 Ao eecbedeees iy - 24°5 30°0 17-0 » 19...) 139 fa Th cee Sie is Re 29°8 33-0 26-0 » 19...) 139 im ocees atc Fs ‘ 32:3 35-0 25-0 OPE GOngl ene ft é 301 330 26-0 OG oles ot ese sks. a s 28 °5 32-0 23 -0 yo SOM es 4 A 228 32-0 19-0 OO OO ee ees. « ‘ 201 23-0 18-0 yall 16, Al) G0 ae Se eee = ‘ 20:1 31-0 18 ‘0 jp WD rel) BLES pet Sa en Fe 3 23 -2 31:0 19-0 2a ool Boul nc eccone 5 55 22 6 31°0 16:0 » 8...| 889 | Rabbit......... 50 3 we) |) PA) <0) 15 °0 » 15...) 389 Sik pa condo eee 5 ¥p 26°0 | 29°0 18-0 1B ESO ie tsi. - 22°83 35-0 15-0 ea =| 389 a MLE i i. es 24,°3 30-0 170 Hebel Stele 605): Raticsscc sn: . | eee 28 -0 18-0 Poel Od Shere renerinen 5 20°9 31°0 17°0 sy seal) GI Be Goes cose 5 - 22 °8 31°0 18-0 5 ilesa| = Or Pe netanrene ee tay Ms cA 19-8 30-0 17-0 5p a) aaol| ast) hace Sourea ten 3 = 24-9 30 °0 18:0 op. HES el OKSTO Ye eee és 256 29-0 170 oy AD Sco aE) aCe ee ») s 28 6 5°0 21:0 oe One ba Leen c rs he SO 34-0 18 -0 ~ =O. 67 eosin aa Ree 5 Ee | 23°2 33 °0 13 °0 si! IST aah CrP ges a ae ee a th BO 31:0 16-0 Mar. 11...) 67 A tM al ae Bey BS z | 20°7 30-0 16:0 errors West cece x eat B35 32-0 18-0 yD oSoi BUA a crcrc raiet eS o | 2a oy 32-0 18:0 Ce SOI Hie Aken Se fs pee a 22008 30-0 16-0 2 AS esaUGE Ol Upstate iene | i i 24°°8 32-0 16-0 ME OR GO9F No os tae is 25 °°8 32-0 17-0 See eee ESN (Oe a . ss 25-5 32-0 16-0 Ar rilee lee inh Sil ovag ONY Oh es 29:5 35-0 19-0 PPL S12 Ba ak Sek 3 | * as 25-4 320 16-0 99 UP coal) SUR Re eer es | ri ys 20°8 31:0 15 °0 PREP OP STOR owed cate ty is 239 36:0 19-0 OMe aOle Gr vente st | - “ 27 6 32-0 19-0 5p Wa cod| BSYA ioe ie sere . - | 24°3 30:0 18-0 WORN CN OMINE Psy nth a ig s 2 21-9 32-0 18-0 = le 1 pd ale ae a . 4 PAL 37 330 19-0 » 18...| 391 ee ee ae eal! ie 5 29 °3 35-0 19-0 Perl oye S02 mp lacey ese. # i 23°7 36-0 18-0 Weld E407 alt) eek tens if 23-0 33-0 160 » 15...) 312 3p ~~ pag DoeoCOgeNnD ” ” 22-1 33 0 18-0 pie LS saa SE |e ene ee af . 25-0 33-0 19-0 3 LO. | 892 PM betitgeve ses 5 1s 21°8 350 19-0 oy WG coal) CHL rp eenereeeneeees 5 . 21°8 32 0 17-0 LOE Roo Sh. agcado anceps ” ” 19-7 28:0 7/30) | 24-2 | 36-0 15-0 114 Sir D. Bruce and others. Trypanosome Table I1—Measurements of the Length of 500 Specimens of the Trypano- some of Naturally Infected Dog, Strain I, Dog 48, taken on nine consecutive days, from Rat 1218, after passage through rats for seven months. No. Date. of expt. 1913. Sept. 5 ...| 1218 ep en leas Sega ders Poy elots 3 6 ...| 1218 5 6 ...| 1218 = 7 ...| 1218 $3 Vian take) aie es 5 8 ...| 1218 i 8 ...| 1218 x 8 ...| 1218 5 9 ...| 1218 5 9 ...| 1218 SoS) seiS ay UO 550] zak} a NO) s2|) 1S} Ones 2s ay Lab Gael] zal} fy eT sol} aalts} yy tll docll LAS} eles | el 2is 6D es UR sf eas a8 LS 1218 In microns. : Method of | Method of | ; Animal. xi renee ane: Smee Average | Maximum | Minimum length. | length. length. Rati stenoses Osmic acid Giemsa 25°8 31:0 170 Si uedaeeseaee 3 | 3 27°3 33 0 19-0 3) Weis seeiecek es 45 _ 25 +1 31-0 18-0 ee see 5 33 23-0 31-0 16°0 pan) wetaracarees 5 yy 21-7 32-0 16-0 shal Wachagaeetes cs * 23 °7 31-0 18-0 Ree deedac ete 5 | 55 26 -4 30-0 21:0 sue adeestoldenes 5 os 26 °0 30°0 20-0 ie aren . is 25-2 31-0 18-0 ie Reaietee anes - 26 -2 32:0 21:0 So eee desea is - 25 °3 31-0 18-0 FM Concnareccs = 5 26-0 30-0 19-0 aomeeeecasteh Ee » 28-1 32-0 24:0 So Ree 5) . 27°8 32-0 21-0 See ee fe 27-0 31-0 21-0 5 Sates ae ye a 226 31-0 170 Pe cocaine 5 0 20 °8 30-0 16-0 5 * 5 22 °9 32-0 18-0 go Gebers 25 50 21-2 29-0 18-0 Se as eh: ig bs 20 4 23-0 17-0 Jo. | fabhpaiantios bs 5 20°9 26-0 18-0 sp Re gaeeieate is sate xp i$ 23 °0 33-0 18°0 6 gon Tia & if 20°1 23-0 17-0 23 °2 320 16-0 9 0 190 16 ‘0 causing Disease in Man in Nyasaland. 115 Table I11.— Measurements of the Length of 500 Specimens of the Trypano- some of Naturally Infected Dog, Strain I, Dog 48, taken on nine consecutive days from Rat 2471, after passage through rats for two years. Series of 46 animals. In microns. No. . Date. of Animal. aPuee of ee of | expt. ne: stamng- | Average| Maximum Minimum length. length. | length. 1913. IDCs PAD seal) PHYA Aen Geos cecbane Osmic acid Giemsa 29:5 34:0 22:0 ye 26...| ZAK Sh BCD Te a 5; 5 28 6 32:0 25-0 » 26...) 2471 Sym Mateteneies 5 nA 29-1 33°0 25 °0 Meo eda telin Se den cae A = : 28-4 31-0 26-0 og loon) CATAL eds Meee asta is 5 29-0 81°0 26-0 2 Q7...| 2471 Bae AS cee asO eae 6) a 28:0 33 ‘0 19-0 DOCH 247 Ir | ere Pea a * 30°3 350 20-0 EDS OAM ed etd ; i 31-0 350 27-0 RR 2 Sis AG bs ee hs Dosen dae, sh :5 29°8 36 ‘0 22:0 1914. ania Lea DA TUE ks WME semua 5 3 28:7 32-0 240 . TN scale Orel es Sa if i 28-7 31-0 25-0 . = gl ACA LV ea a 4 - 27°8 30-0 19 ‘0 if 2 el Oil ead ae ead - i 29:3 330 23-0 : See ROO he SMa a ack i a 30 °4 340 25:0 6 Sea SAA IGM, we Canes kobNeels A < 30°4 34:0 26°0 WR Ay Agel ele goon ante. i i 268 33-0 20-0 Fp 4...) 2471 34 abeaaaehabine 6 a 27°6 32:0 19-0 3 4...) 24°71 Ae laeae tas Satta ae i 27°9 320 19-0 ap 5...| 2471 NHN MEtOROGROM REN Fh a4 28 °2 33:0 20:0 0 5...) 2471 SliesnonaeatTat i ¥ 26 6 32:0 18-0 cp ES ALT | sont were ats ' 3 27°3 33°0 19-0 5 CZ Ae ya ek nena. Fy i 27°9 32:0 19:0 b 6...| 2471 ett er RGM Te a4 5 28:0 330 24:0 5 6...) 2471 er Te ea RE rs ; Ps 28 °7 33:0 20:0 3 te | MANTA Nae ie ee a 4 ‘ 26°9 32:0 19:0 28°6 36:0 18-0 The average length of the trypanosome of Naturally Infected Dog, Strain I, Dog 48, taken from Tables I, II, and ITI, is as follows :— Table [V.—Average Length of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48. In microns. Number of ta Species of animal. trypanosomes measured. eeraolon acl Maximum Minimum | 8 ay length. length. (Ope Meecencanna saeeee 20 25 °5 31°0 16°0 Sheep, e-kec toes 20 | 26 *4 300 21-0 SD) O Die eo meetes cate 260 | 25 -2 35 0 15:0 Ralbbitigeeeeece rie 80 23 °9 35°0 15-0 0 1 ep aancitemaniare | 660 | 23-7 36 116 Sir D. Bruce and others. Trypanosome Table V.—Average Length of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, after passage through rats for seven months. Number of Species of animal. trypanosomes measured. IR BUBAEM cose ene 500 In microns. Average length. Rea | ae 24-1 | 33-0 | 19-0 Table VI.—Average Length of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, after passage through rats for two years. Series of 46 animals. In microns. Number of Species of animal. trypanosomes measured. Avévassuen ete Maximum Minimum 8 hee length. length. Rath cnesawasnnraedse 500 28 6 | 36 -0 | 18 °0 Cuart 1.—Curve representing the Distribution, by Percentages, in respect to Length, of 1040 Individuals of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, - taken at random from various animals. hs [ig Tus fui [8 [ro [20 as [zo] 23 [24 ]25 [26 [29 [2s [29 [0 [ai [n Jos [5a 135 [56 137 aa Es causing Disease in Man in Nyasaland. 117 This curve is made up of measurements from 20 specimens of trypano- somes taken from the ox, 20 from the sheep, 260 from the dog, 80 from the rabbit, and 660 from the rat. Cuart 2.—Curve representing the Distribution, by Percentages, in respect to Length, of 660 Individuals of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, taken at random from several rats. This is the curve of a markedly dimorphic type, and may be compared with the Wild-game strain J,* or with the Wild G. morsitans strains IV and V.f The above curve shows the strain as it appeared in the rat in February, 1912, when it was first obtained. The next curve shows the same strain as it appeared in the rat in September, 1912, after it had passed for seven months through a series of eight rats. * ©Roy. Soc. Proc.,’ B, vol. 87, p. 395 (1913). + Ibid., B, vol. 87, pp. 415 and 417 (1913). 118 Sir D, Bruce and others. Trypanosome Cuart 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, taken on nine consecutive days from Rat 1218, after passing through a series of eight rats. ial Gor © are Sie 13] 14 }15 [16 | 17 | 18 | 19} 20 | 21 122) 23)24 125 | 26|27 128/29 | 30/3! |32z| 33 34135136 |3 os 18 1 } MIE mes S 50) Percen Tages —- ~Pwp us oa H This curve still shows a markedly dimorphic type, but the proportion of the long forms is increasing. The next curve shows the same strain at the beginning of 1914, after passing for two years through a series of 46 rats. causing Disease in Man nv Nyasaland. 119 Cuarr 4,—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, taken on nine consecutive days from Rat 2471, after passage through rats for two years. n (7) oO ON | » £ (0) 3) » o a 8) 8 1 6 5 4. 3 2 i} The curve is now practically monomorphic, The short and stumpy and the intermediate forms have almost disappeared, and only the long and slender survive. But, it may be objected, perhaps this curve from Rat 2471 is a mere accident due to some peculiarity in the rat; it is possible that if another rat is inoculated from it the curve will be found to be as dimorphic in type as that of Rat 1218 on Chart 3. That this is not so will be seen by the following chart, which represents the gradual change in type which . takes place in this trypanosome by passage through rats. The first rat, 67, was inoculated from the original dog; the second rat, 312, was inoculated from Rat 67; Rat 407 from Rat 312, and so on through a series of 20 rats. The unbroken line represents the percentage of the long and slender forms, the broken line the short and stumpy. For example, Rat 67 has 30 per cent. long and slender and 70 per cent. short and stumpy. In Rat 670 the long and short forms are almost equally divided; Rat 786, 82 per cent. long and 18 per cent. short. The percentage of the long and slender gradually increases until at the end of 17 passages it reaches 100 per cent., so that 120 Sir D. Bruce and others. Zrypanosome from a dimorphic type with 70 per cent. short forms the type gradually changes into a monomorphic type which has lost almost all the short forms and nothing but the long remain. This seems to show how fallacious it is to reason from laboratory types of trypanosomes to the wild natural types, and probably accounts for the showers of new species which are constantly falling about our ears. Cuart 5.—Curves representing the Gradual Change of this Trypanosome from a Dimorphic Type to a Monomorphic. Rats ts) MN }i2Z}iSjI4) IS} 16 4 18 | 19 |Zo n u re iy) p < v 12) - @ a Ae ees a f 6 as CRE, esa] Ie sa vl aa ve OT [TSS a eae Long « Slender SOr772s meses Short & Stumpy Sor7s Breadth.—The following table gives the breadth of Strain I, Dog 48 :— causing Disease in Man in Nyasaland. 121 Table VII.—Measurements of the Breadth of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48. In microns. , Number of a Date. Hay ae Animal. | trypanosomes i measured. | Average Maximum Minimum breadth. breadth. breadth. 1912 1218 Rat 500 2-90 | 5 00 1°25 | Table VIII.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48. Experiment “ Percentage among short Date. Ee Animal. and aera forms. 1912. April 22 .......-. 407 IMENT: cenbctloce 3 Bae OLAS a ee te 407 iioes bane aee 2 SOA eae 407 Haieebeachars 1 May ADE eect 407 aebre stn teen 4 $e Ghee: 407 ne Cor 3 2 Qik car 407 Ay mies mentee 8 ey als Rea waae 407 Beare saan 6 er wllaie Nara 407 py tnocoGeEo 8 Meo Oa car 407 pin ets 4 ae SLIM ee 407 een 23 eam A ah 407 Cees 15 Average! i... .5: 7°0 Table [X.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, after passage through rats for seven months. Date. Experiment No. Animal ; Percentage among shor and stumpy forms. t SCrFOCOCOCOORWN = wd 122 Sir D. Bruce and others. Trypanosome After passage through rats for two years the short and stumpy forms have nearly all disappeared, and with them the posterior-nucleated forms. In Rat 407, on Table VIII, there are a fair number of posterior-nucleated trypanosomes, on one day as many as 23 per cent., but this is exceptional. MorPHOLOGY OF THE NATURALLY INFECTED Doc Strain. Strain II. Doe 690. Table X.—Measurements of the Length of the Trypanosome of Naturally Infected Dog, Strain II, Dog 690. In microns. No. : Date. of |] yAmimal: wll yuan co Sioa ee ‘expt. Bee sraining- | Average | Maximum | Minimum length. | length. length 1912. July 26...) 911 | Rat ............ Osmic acid | Giemsa 25 °3 32 0 18-0 Gia! WONT Bien oeueeees i cs 236 30 °0 18 ‘0 eG. bl ONT Ml) Seer kibee conn : ¢ 250 31-0 19 °0 pecs 911 ppl poonsasancane 0 oA 24-9 320 15°0 Ben arate 911 SF haat ss crests * 99 25°9 32 0 18-0 Oe OL ot) Utama mre a f 27-0 32-0 17-0 3. SSO a O11 "|| emanates ee i Ki 24.2 31-0 18 ‘0 802: | SOI =|: aero i i 25-4 30-0 18-0 BO) cle Q1 y|| ee numeemnenne ee a “i 25-0 31-0 18-0 “Us 3ir eel (OL ie} Une meanee es am i 259 32-0 19 0 » ol 911 PAP noopoaredeso 59) ” 25-2 31°0 16-0 AB | HOLL Mae mneetn ens 5 i 26 °9 32 0 18-0 Kae | OTT | Meee i 2 23-4 32-0 16-0 5 We soalft LIL pjubdinetels ae abel 5) 2 23:9 33 0 17-0 5 OMIA, cecmieedaecl Dee ‘ . 24-2 36-0 170 S00 ape MO TTS | EME et - z 21°7 27 0 17-0 ih prbal Bons || Se aaenann etn Bs i 21-0 300 16 ‘0 So COMES OT ie hana eae * bs 23-4 31-0 16-0 FHS cada) Diane: 2ee s Es 21°5 31-0 16-0 p 3...| 911 Ty lgdagnanabece 1p Bs 21°6 31°0 16:0 cyte malaga als G6 0h e NCE Oe i “3 21°7 29-0 160 SME sie FF Shite ea ee eae ee ah - 21°9 32-0 17-0 . &) ool} SNOL Prt ueaSreeniaacce x 90 23 °2 29:0 17:0 Bek COAT Be anette a 8.0 m6 ee 221 30 0 17-0 Bee ee a iy aera ‘i 3 24-2 32-0 17-0 23°9 36-0 150 causing Disease in Man in Nyasaland. 123 Cuart 6.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Naturally Infected Dog, Strain II, Dog 690, taken on nine consecutive days from Rat 911. tccerone PEN eN cma | 3] 14 ]is [ug [iy] 18 [to [20[2 [22 ]25]24[25]26]27 [2829 [30 [s1 [32 [33 ]54]35 [36 [37 [6 | Ea ee ea PEEREEEEEEEEE EEE EEE cai AHHH HH LH Perce ntages = eA Vea The curve of Strain II, Dog 690, is also eminently dimorphic, so much so that the presence of two species might be suspected, one with a maximum of 18 microns and the other with a maximum of 28 microns. If such were the case it could be argued that in Chart 3 of Strain I the long species had driven out the short. But it will be shown later that this is probably not so: that the difference is merely due to dimorphism and not to the mixture of two species of trypanosomes. Breadth—The following table gives the breadth of Strain II, Dog 690 :-— Table XI—Measurements of the Breadth of the Trypanosome of Naturally infected Dog, Strain II, Dog 690. | In microns. | Experiment EV HISEE OS Date. | P N Animal. trypanosomes | | ee measured. Average | Maximum | Minimum | breadth. | breadth. | breadth. | | | | | 1912 911 Rati tao..: 500 Zeman line Seiten el 25 VOL. LXXXVIII.—B. L 124 _ Sir D. Bruce and others. Zrypanosome In regard to posterior-nuclear forms in this strain, there are practically none. MoRPHOLOGY OF THE NATURALLY INFECTED Dog STRAIN. SrTrRarn III. Doc 2033. Table XIJ.—Measurements of the Length of the Trypanosome of Naturally Infected Dog, Strain III, Dog 2033. In microns. No. | | 55 29 °5 33 0 25°0 cee LIBOOS 1G Gey ge eee if F. 296 36-0 24-0 1 Bec ypal aS eadl DOS TE Cab eet eee eee ; i 29 °7 350 25-0 | ig) BERN ZOS7TAN Bys. teeeeeecaemes oF X 28 -2 33 -0 25-0 es 2037 Fb bocagaanaode * 6 29°6 33 °0 25°0 MAAS |S2037>| coals a * 27:8 34-0 24-0 sn) MAG 2OSTE || a anereeeneeerene 5 297 37-0 250 i 14222057) | ae areas a * 278 33-0 22-0 Ae DOST | has Sattiaaeteteat at “ ~ 28-1 32-0 25-0 cubs palleimeel| AOR” 8 ace aioe ae i 286 32-0 24.0 PEM ROOST: | Tee eer 2 27-5 31-0 25-0 » 16...) 2087 | 4, «...2 ee y 6 27-9 33 °0 24:0 1 IGE R2OS Ci Seen ce aurea ; 5 28-1 33°0 | 24°0 pe eee| 20375 a eeie | Z 4 27°5 32:0 | 23-0 Cs SIERO DT all aytic eae dhe ‘ ti 268 31-0 21-0 Re lla 2080) ll eas oe anes | 4 ¥ 26-1 33-0 20-0 Fp SPT eA ZOS Tilia coke ee CORE he By 26-1 29 0 23-0 | sy) PLO MZ0ST el fa. eeteenecss . _ 27 °6 32°0 | 240 De ph Ts | SOT a) eines é * 28-5 36-0 25-0 lp he CERO MANSO OSI Bue UE weds 3 . 28-3 34:0 25-0 [ iaeigs | AUF BBOS TA panies hest ake. as fe 27-9 31:0 | 22-0 | | | ; 28°4 37°0 20 °0 causing Disease in Man in Nyasaland. 125 Cuart 7.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Naturally-Infected Dog, Strain III, Dog 2033, taken on nine consecutive days from Rat 2037. ————————— 202i [22]25]24]25 [26] 27 [2a] 29 [30 [si [52] 5524 |35] 367 [22] Pererremenm rr rrr yt ela ae we Fal yo a 2 a a pica ~ 10 0 ecenlta Oo This curve was taken from a rat which was inoculated directly from the naturally infected Dog 2033, and had therefore passed through no series of rats. Yet the curve is the same as that shown in Chart 4 after two years’ passage through rats. It might be argued that this is vealty an infection with the larger of the two hypothetical species, the one having a maximum of 28 microns: It is to be regretted that this strain has died out, so that no further experimenta- tion with it is possible. It would have been interesting to pass it through other animals, in order to learn if any reversion to the short and stumpy form would take place. But if it is not possible to do this with Strain II, it is with Strain I, which was seen to change from a dimorphic type to a practically monomorphic type after two years’ passage through rats. When this almost monomorphic rat strain, as shown in Chart 4, is inoculated into a dog, a reversion to the original dimorphic type is brought about, as will be seen from the following chart :— 126 Cuart 8.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Naturally Infected Dog, Strain I, Dog 48, taken from Dog 2498, This dog was inoculated from Rat 2471, which showed Sir D, Bruce and others. 95 per cent. long and slender forms. mM 1C 7 Of Ss 13 nh) =n & Percentaoes 5 14 5| 16] 17 |18 19 |20/21 | 22)23) 24) 25 Trypanosome yp Ww PAA un ® © 26|27|28|29|50|31 |32| 353 [ In the blood of Rat 2471, Chart 4, the trypanosomes almost all belonged to the long and slender type. Now this is reversed, and the majority are short and stumpy. This goes against the theory that two species are being dealt with in this strain. Passage through the rat favours the production of a long and slender monomorphic type of trypanosome, whereas passage from the rat to the dog at once changes this to a dimorphic type, in which the short and stumpy form the greater number. Breadth.—The following table gives the breadth of the trypanosome of Strain III, Dog 2033 :— Table XIII.—Measurements of the Breadth of the Trypanosome of Naturally Infected Dog, Strain III, Dog 2033. In microns. Deak \ Number of Date. No ; Animal. trypanosomes | | : i i measured. Average | Maximum | Minimum | breadth. | breadth. | breadth. | | 1913 2037 Rat 500 2°61, | 4-00 | 1°50 | causing Disease in Man in Nyasaland. 127 In Strain III there are no posterior-nucleated forms. This is not to be wondered at, as there are almost no short and stumpy forms, and it is only, or almost only, among them that posterior-nucleated trypanosomes are found. COMPARISON OF THE THREE NATURALLY INFECTED DoG STRAINS WITH ONE ANOTHER. Table XIV.—Measurements of the Length of the Trypanosome of the Naturally Infected Dog, Strains I, II, and IIL. | | | In microns. Expt ae | , | Number of | Date. No. | strain. | Animal. | trypanosomes | | measured. Average | Maximum} Minimum | | | | length. length. length. | | | 1912 — I | REND coonee 660 23-7 36-0 15 °0 1912 S11 IL aay ealeoeeee 500 23:9 | 36:0 15-0 1913 2037 | Til a | 500 28 °4 | 37 °0 20:0 | | | Table X V.—Percentages of Short and Stumpy, Intermediate, and Long and Slender Forms in the Three Strains of the Naturally Infected Dog. eee | | Number of | Short and| Inter- Long and Date. | No. | Strain. | Animal. | trypanosomes | stumpy, | mediate. | Slender, | | measured. 15-21. 22-24, 25-37. | | per cent. | per cent. | per cent. 1912 = I | TRO Sonne 660 | Al7 1. 10-2 42-7 1912 911 1a | Mlanetiars shea 500 38-2 9-2 526 1913 2037 III [eee oni Sacer 500 0°6 4°8 94 °6 | | Strains I and II are similar, but Strain III differs so much from them that it would be useless to combine the three into one curve; the result would be misleading. In Strain III, as will be seen from Table XV, almost all are long forms. Table XVI.—Measurements of the Breadth of the Trypanosome of the Naturally Infected Dog, Strains I, II, and ITI. In microns. | | Expt 2 Number of | | Date. ie, Strain. Animal. | trypanosomes | | | | : measured. Average | Maximum | Minimum | | breadth. | breadth. | breadth. | | 1912 1218 iL Rates | 500 2-90 WOO) |) | al 25 1912 911 it pr eereees | 500 2-719 4°75 1°25 1913 20387 O08 rye oneced | 500 2°61 4-00 1°50 128 Sir D. Bruce and others. Trypanosome Table XVII.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Naturally Infected Dog, Strains J, II, and ITI. Date: Experiment Spa Tea Percentage among short No. and stumpy forms. 1912 407 I Rat 7-0 1912 911 II " 0-1 1913 2037 III if 0-0 COMPARISON OF THE NATURALLY INFECTED Doc STRAIN WITH THE HUMAN, WILD-GAME, AND WILD GLOSSINA MORSITANS STRAINS OF THE TRYPANO- SOME CAUSING DISEASE IN MAN IN NYASALAND. Table XVIII.—Measurements of the Length of the Trypanosome of the Human, Wild-game, Wild Glossina morsitans and Naturally Infected Dog Strains. In microns. Number of Strain. trypanosomes Animal. ‘ measured. Average | Maximum | Minimum length. length. length. IBEGUITENA sa0cca90000000000 5500 Rat 23°5 38 °0 14:0 Wald-oame renee tees 2500 a 226 350 15 ‘0 Wild G. morsitans ... 2500 H 226 35-0 15-0 Naturally infected dog 1660 iy 25 °5 37-0 150 Table XIX.—Measurements of the Breadth of the Trypanosome of the - Human, Wild-game, Wild Glossina morsitans and Naturally Infected Dog Strains. | In microns. Number of Strain. trypanosomes Animal. measured. Average | Maximum | Minimum breadth. breadth. breadth. BGP cescgosnqondbar000 1500 Rat 2°6 5 00 | 1-25 Wild-game .............:. 1500 = 3:2 5°75 1°50 Wild G. morsitans ... 1500 Hs 2:9 5°25 1°25 Naturally infected dog 1500 rs 2°8 5 00 1°25 causing Disease in Man in Nyasaland. 129 Table XX.—Percentages of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of the Human, Wild-game, Wild Glossina morsitans, and Naturally Infected Dog Strains. Date. Strain. Animal, |) Percentage eee RON and stumpy forms. TIGR) |) JELU SAHA CooonoooonnsobeeoqooN JM iscooncous 17°8 1912 | Wild-game ............... mab ormenee 26 2 1912 | Wild G. morsitans ...... Sa) Stee SeWE 12°5 ss ee 2-4 1912 | Naturally infected dog COMPARISON OF THE MORPHOLOGY OF THE NATURALLY INFECTED DoG STRAIN WITH THE OTHER STRAINS OF THE TRYPANOSOME CAUSING DISEASE IN Man In NYASALAND. At the outset it may be stated that it is impossible to separate the Naturally Infected Dog strain from the other strain by microscopical examina- tion. As far as can be made out it is identical in shape, size and position of nucleus and micronucleus, contents of cell, and disposal of the undulating - membrane. Three plates are given at the end of this paper to illustrate the morphology of this strain, and if they are compared with the plates given of the other strains* this statement will be borne out. On the other hand, there are very few posterior-nuclear forms, although in one instance they ran up to 23 per cent., and, as a rule, the thick, blunt- ended type is not so common in this strain as in the others. But for all practical purposes it must be concluded that the Naturally Infected Dog strain is so similar in appearance to the others that it would be impossible to separate it by morphology alone. How this aberrant strain arose in these three chronically infected dogs it is impossible to say. If it had been found anywhere else—in man, game, or fly—the position would have been simplified. But in none of them did anything like the Naturally Infected Dog strain appear. It was thought that perhaps the long sojourn in the blood of the dog had modified and weakened this strain, and attempts were made to prove this, but without success. All the dogs inoculated with the ordinary strains died in a few weeks, and inoculations from those which lingered longest showed no signs of weakening or change of any kind. * ‘Roy. Soc. Proc.’ B, vol. 87, p. 35 (1918). Jbid., B, vol. 87, p. 493 (1914) “ Description of a Strain of Trypanosoma bruce from Zululand.” 130 Sir D. Bruce and others. Trypanosome CONCLUSIONS. 1. The Naturally Infected Dog strain differs slightly from the other strains of the trypanosome causing disease in man in Nyasaland, in that there are fewer of the posterior-nucleated, blunt-ended forms which are sometimes so much in evidence in the ordinary strains. 2. Taking into consideration the fact that this strain was only found in three chronically infected dogs, it is concluded that it is an aberrant strain of the widely spread species 7. brucez vel rhodesiense, the trypanosome causing disease in man in Nyasaland. DESCRIPTION OF PLATES. Trypanosome of Naturally Infected Dog. Plate 9.—Short and Stumpy, Non-flagellated Forms. Plate 10.—Intermediate Forms. Plate 11.—Long and Slender Forms. x 2000. The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part I1.—Susceptibility of Animals. By: Surgeon-General Sir Davin Bruce, O.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received April 16,—Read June 25, 1914.) INTRODUCTION. In a previous paper* the morphology of the three strains of this trypano- some, from three naturally infected dogs, was described, and the strains compared with each other and with the Human strain. This paper describes the action on various animals of the three strains and tabulates a comparison with the Human strain. The first strain—Dog 48—was studied in a fairly large number of animals, but the second and third in few, as both were accidentally lost. * ‘Roy. Soc. Proce.,’ B, vol. 88, p. 111 (1914). Bruce and others. Roy. Soc. Proc., B, vol. 88, Plate 9. London Stereoscopic Co. imp. SHORT AND STUMPY FORMS, Bruce and others. Roy. Soc. Proc., B, vol. 88, Plate ro. ( r ‘Gun f me London Stereoscopic Co. imp. INTERMEDIATE, Bruce and others. Roy. Soc. Proc., B, vol. 8&8, Plate rz. London Stereoscopic Co. imp. LONG AND SLENDER FORMS. causing Disease in Man in Nyasaland. 131 _ SUSCEPTIBILITY or ANIMALS TO THE NATURALLY INFECTED Doc STRAIN. I. Strain I, Dog 48. | | Table I. NGOS | Period of | Duration | Date oF ‘4 Source of virus. | incubation, | of disease, | Remarks. eae in days. | in days.* Cattle. 1912. | | Mar 6... 314 | From Dog 210 ......... == = | Never showed trypanosomes. Maras. | 315 = PII) ccocner ~~ Lo= 54 5 fn April13.... 314 | From Rat 311 ......... 12 | = | Still alive after 335 days. aa Sis, 315 Ig SH cae _- | -— Never showed trypanosomes. Goat Mar. 6...) 275 | From Dog 210......... | = = | Never showed trypanosomes. BLO a = sh GEN eet April 5...) 275 | Brom Rat 312 ......... | 24, — Accidentally killed. eos ay | de a een — = Never showed trypanosomes. c 20... 427 | Bs BOOM rt 10 = Still alive after 277 days. ” 20.. | 432 ” 392 edosdonct 26 hae ” ” ” | | | Ions, | | | Mar. 21... 2008 5 GEA cosncone | = = Never showed trypanosomes. pp Albani ADO) as WOO oconocons 27 — Still alive after 251 days. op Lbbscall AOU) 5 OO eesceree = == Never showed trypanosomes. op illecs!) ZAONLAL * TOO wees 17 — Died of pneumonia. pp Calbecolp PAD} 5 OIE. ccooneane = = | Never showed trypanosomes. ” 21, | 2013 { »” 1991 wee eeenee| an a | ” ” ” lee 2014 3] a UO sccodacen 24 = | Still alive after 251 days. ay 21. 2015 7 1991 Sanadanonl| 1 ‘aaa ” ” a” » 21...! 2016 , Te scogaoaen a == _ Never showed trypanosomes. op alk coull ZAOUERS 5 TOM soca = = 0 5 5 Sheep igPS | | | April 20... 456 | From Rat 392 ......... | 5 at Still alive after 340 days. pp AUknell | SIG 5s Ci Loaatinecn 10 64 Died of Strain I. Antelope. 1913. | May 21 2059 | From Rat 2024 ......... | 13 | — | Still alive after 250 days. Monkey 1912. | | Maree Ghee 318 | From Dog 210......... | = = | Never showed trypanosomes. April 5.. 318 | From Rat 312 ......... | = == be » » Pe i20n 50 | From Dog 3817 ......... | a a ae x - » 20...) 458 | From Rat 392 ......... —— a UP eae s5 » Oct. 29...) 1533 ¥ D490 cscccal — = | 6 ” » '» 29...) 1584 MS PAST | 6 — _| Still alive after 148 days. Noy. 22... 1629 | From Monkey 15384... — — | Never showed trypanosomes. fy ACceol| IKGB0) % 1534...) 10 _- | Still alive after 124 days. * Duration includes the days of incubation : it dates from the day of inoculation. 132 Sir D. Bruce and others. Trypanosome Table I—continued. Now OE Period of | Duration Date. seen Source of virus. incubation, | of disease, Remarks. Be in days. | in days.* Monkey—continued. 1913, Jan. 22...) 1792 | From Rat 1741 ......... 5 | — Still alive after 186 days. 9 sca) IS bs VAN ea anor 5 — 5 i ss » 22...| 1794 a ANTAL ee ae = — Never showed trypanosomes. » 22,..| 1798 | From Monkey 1630 — — me 5 Feb. 28...| 1794 | From Rat 1945 ......... — —_— Ni ies 4 5p » Boss 17S te TIGY GS pseoodoos — | — autos i 3 May 22...) 1794 | From Monkey 2131 ... — | — | leans 53 = 22) 1798 a PAB eal) = . ie June 11...) 1794 A 2184 . —- —_— MF is a pall 1798 5 2184 — — * i ss Dog. 1912; | | Feb. 17... 210 | From Rat 67 ......... 8 25 Died of Strain I. Mar. 6.... 317 | From Dog 210......... ae = Never showed trypanosomes. 5p Baooll SBIL || Wigoren TRE BY coocoenen — _— be - 5 April 6... 317 a SUZ iieses 12 16 Died of Strain I. % Weel eel | a Gull cpenauees 12 46 n * 55 1 One 458 | From Dog 317 ......... — —_— | Never showed trypanosomes. ey eles 459 ys BUT eject 19 — Still alive after 224 days. Sept. 6...) 1253 | From Rat 1218 ......... 6 — is 5 Oe Oct. 29...) 1525 | rs WAOW, 22. divest 13 30 Died of Strain I. 9 Beco! . UBD | sh Aiea? cece 23 53 5 3 oy eda. llayearh si AQT se cetict — = Never showed trypanosomes. mea aes i. TAQ er cance 13 21 Died of Strain I. 2 Olli 29 ae TAQ ee 13 47 . > Sp 2Bocol ERO) 35 WAL coonavooc 16 — Still alive after 148 days. | 19138. | Jan. 22...| 1795 | From Rat 1741 ......... 5 29 Died of Strain I. hop coll UB 5 THM, seo nosione| 8 35 5, F 3» ball Lay | Pe geo BON ee | 22 | — Still alive after 1381 days. April 14...) 2054 | Laboratory-bred flies... 7 30 Died of Strain I. |May 29...| 2197 | From Dog 2054......... | il 16 AS 5 |_>, 29...) 2198 a 20 SA | 4 13 99 p | Dec. 30...) 2483 | From Rat 2471......... 14 34 zs i 1914 |Jan, 26...) 2498 | From Dog 2488 ......... | 4 11 Died of Strain 1. | | | | eens Pea | Average...... | 11°6 29°0 Rabbit | 1912 | [Peter | Mar. 30...| 389 | From Rat 67 ............ | 9 256 Died of Strain I. lp BOs) 80 DEG ab sah een ae 109 ys . | [I Average...... 13 0 1825 Guinea-pig. Mar. 6... 313 | From Dog 210......... a == Never showed trypanosomes, April 5...| 313 | From Rat 312......... = = » ” my Win 460 5 BO) snocosese | = = ” ” 9 20 461 ” BOL: Ciera ime aa 2 oP) Oct. 29 1531 ” AQT eee saan aa | ” »” » 29... 1582 “ TWAIN geo | = — | e 29 Nov. 13...) 1531 PORE RIAOD Ey aan — = | 5 ” noniat | laR i TGP salva — — . * Duration includes the days of incubation; it dates from day of inoculation. causing Disease in Man in Nyasaland. Table I—continued. 133 Period of | Duration | Date. Ne: e : Source of virus. incubation, | of disease, | Remarks. aD in days. | in days.* Guinea-pig—continued. 1913 Jan. 17 1775 ae nV GY ee = — Never showed trypanosomes. June 17.. 2228 ” PPANGY o6 sabe = aia ” 2»? lines) 2229 4 PEAS eooscosce 6 = Still alive after 246 days. July 22...| 2307 is PAPAS ind onbaoo _ — Never showed trypanosomes. ” 22. 2308 » 2285 BOOCOUDIOG ora oe ” ” ” 31.. 2307 > PASS) ceoasceco | rar aa ey) 2? ” olen 2308 ” PYTSIS) Godenbboo | a ae 2 2” Rat. 1912 Mar. 26... 311 | From Rat 67......... 13 18 Died of Strain I. sp. _ Abas 312 3 Oiedaenoaces 9 60 PP » Pe Os.. 391 #, Ciizion 9 21 » | 30: :. 392 3 Cli ereskcacs 9 21 » ” April (se 407 ” Gil eesenmane 10 54 »” » ap ess 462 4 AQ Tes econ 10 25 ”» » May 25 585 ” 407 AocsooD54 5 28 ” ” June 4 670 x BEB: cesocoos 6 27 » ” July 2 786 ” 670 Hocmen occ 6 45 ” ” Aug. Qe. 1020 ” hkoececanee 6 30 ” ” co Deno] Laalfs} 5 ACH) o-ceoouce 3 94: ” ” Oct. 19... 1492 0 WMS) ssoaboacs 9 54 ” ” Dec. 12... 1687 | ” AD Dirt: 4, 16 ” ” »” 28. . 1719 | ” 1687 ee eceenee 9 32 oF) ” | 1913. Jan. @iba0 1734 | ” 1570 Bua oocno 9 36 2 29 op) Ball BES") s NBO" .cceose-, 6 39 3 » 33 Usea Ltknke | Pr UT LON see 4 15 ” ” 5 S)acall dlgAss) 8 WN sagesoooe 4 18 » » oy Mlooal| AGHIZ) | 35 MTA es aseuec a 16 » ” Feb. 10...) 1855 | Ps US canacsee 3 20 ” > ap PPxcoll IGE is TEED ccocdsose 5 17 55 56 Mar. Une 1985 op) OA OE eS. 6 20 ” ” » 20...) 2022 an UBB) cooconeeo 6 16 5) 5 20...| 20238 3 IGE coscoucce 6 18 ” ” » 20...) 2024 ie 1985 6 19 »” » April LOR 2070 ) PLP. son tacos 4, 12 ” ” Pee LOD) 58 FOV ssaceceoa 6 20 ” » May Uso 2124 ” ANOS) Beccnq one 3 8 ” 2? » 18...) 2184 | From Dog 2054......... 2 12 % » » 25... 2188 | From Rat 2188 ......... 3 Q » ” Gira], BIG s DUDA sack: a 17 3 ” ” Uo 213838 From Dog 2054 vo okk 0. 6 10 » ” June Owe 2214 99 QNG Torrens. 3 10 ” ” » See 2230 From Rateae Aiea: 8 10 ” 2 Aug. 18...) 2889 A ZPASO) oatacedoc 8 48 > » Sept. 6... 2409 ” PBS) seoono oad 5 49 29 ” » 16...) 2413 x 2409 ......-.. 6 70 a % Oct. 28...) 2425 a 79 I 5 64 » » Noy. 12... 2432 op DAD es oeceten 6 38 ” eB Dee. 20...) 2471 5 2432.00.20... 6 22 5 3 1914 Jan. 16 2484 sy) BAT Messornasst a 16 ” 2» Average...... 6:2 28 6 * Duration includes the days of incubation ; it dates from the day of inoculation. 134 Sir D. Bruce and others. Trypanosome II. Strain II, Dog 690. Table II. | No. of Period of | Duration Date. pea Source of virus. incubation, | of disease, Remarks. | ip in days. | in days.* Monkey. 1913. | i Sept. 13...) 1314 | From Dog 690 ......... = = Never showed trypanosomes. Fy UBsco|| dienes | » G80) oncessac = = 5 25 Dog. 1912. | | June 17...| 690 | Naturally infected ...... Bin | — Recovered. Guinea-pig. 1913 | Sept. 18...| 1316 | From Dog 690 ......... | = — Never showed trypanosomes. ay 18%.) ely MO E90 ee, [ase «23 = * I, Rat. | 1912. July 18...| 911 | From Dog 690 ......... 7 30 Died of Strain II. * Duration includes the days of incubation ; it dates from the day of inoculation. Ill. Strain ITI, Dog 2033. Table III. | No. of | | Period of | Duration | Date. ao | Source of virus. | incubation, | of disease, Remarks. pee | | in days. | in days.* Goat. 1913. | May 21...) 2174 | From Rat 2089 ......... 12 71 Cause of death uncertain. > AL...) 2075 < BOSS) sscoogoes — — Never showed trypanosomes. ” 21. 2176 ” 2089 a eceeeees or Pit ” ” mp Zilcooll PAUzr/ 5 ADEE) coarse = are » ” tas) PN fe OSS aoe 12 — _| Died of pneumonia. Monkey. May 14...) 2161 | From Rat 2091 ......... = | = Never showed trypanosomes. | ae Te SOOT eh CHa |e Recovered. » L14...] 2168 Ay FAO ME osandon — —_ Never showed trypanosomes. , 14...) 2164 ne XO ole « a — s 53 Aree |eziliGo a PROSE Sanco nn 12 — Recovered. June 14. 2161 | From Dog 2157 ......... — — Never showed trypanosomes. » 14...| 2164 ey 2157) ee aa = te) oh * Duration includes the days of incubation ; it dates from day of inoculation. causing Disease in Man in Nyasaland. 185 Table [L1—continued. | No.of Period of | Duration | Date. o: t Source of virus. incubation, | of disease, | Remarks. ele | in days. in days.* | | Dog. Mar. 28...| 2033 | Naturally infected......| ? | ? | Died April 1. May 14...) 2156 | From Rat 2091 ......... 15 40 | Died of Strain IIT. » 14...| 2157 nih DOO hones 8 33 x 5 Mere iss |, © 2091... 8 26 if i (A ae 1) eee 15 93 fe c » 14...| 2160 | on 74.08) Weare pace 15 102 ee a SSS SSS | | Average...... 12 °2 58 °8 Guinea-pig, Mar. 28...) 2039 | From Dog 2088 ......... = — Never showed trypanosomes. Ps 2040. | Be os USE eat = = | a Z May 21...) 2180 | From Rat 2089 ......... 22 — Recovered. oy Palesd Palka x PAU Sh), socie Soene — — Never showed trypanosomes. Rat. Mar. 28... 2037 | From Dog 2088 ......... { 13 | 73 | Died of Strain III. Apr. 16...| 2089 | From Rat 2037 ......... | 5 | 15 | Ks Be » 16...| 2090. | , DOBT es 5 ERR Szaah eB H's & ‘s 5 16...) 2091 i DOB Yl scocenece 5 | 28 | x * May 14...) 2167 in DOSie 8 avevee i . Oct. 23...) 2426 | From Guinea-pig 2180 11 | 19 FF i | Average...... 7-8 | 46 °5 * Duration includes the days of incubation; it dates from the day of inoculation. Disease set up in various Animals by the Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Ox.—This trypanosome does not appear to be virulent to the ox. Four experiments were made. The trypanosomes appeared in the blood of one of the oxen, and it was returned as “ Recovered” after being under observation for 335 days. The parasites were only seen on three occasions in this ox, and then only in scanty numbers. Goat.—The trypanosome also has little effect on goats. Twenty-one were inoculated. Of these 12 proved refractory; five showed the trypanosomes in their blood on one or two occasions in very scanty numbers, and were returned as “Recovered” after being under observation for nearly a year; four died, one from the result of an accident, two from pneumonia, and the remaining one only once showed the trypanosomes, and as no post-mortem examination was made it is impossible to say what was the cause of death. It may therefore be said that not a single goat of the 21 died of the disease. 136 Sir D. Bruce and others. Trypanosome Sheep.—Two sheep were inoculated. One recovered; the other died after 64 days, probably of the disease. Monkey.—This trypanosome has little or no effect on monkeys. ‘I'wenty- seven were used as experimental animals. Twenty-one proved refractory ; the remaining six were returned as “ Recovered” after being under observa- tion for several months. Dog.—This strain has become, after several passages, virulent to dogs. Twenty-eight were used for experiment. Nineteen died, on an average, in 36'8 days (11 to 102); four never showed trypanosomes in their blood ; and five recovered. The post-mortem appearances are the same as those found in Nagana: enlargement of the spleen, gelatinous cedema about the vessels at the base of heart, petechize of mucous membranes, and corneal opacity. Rabbit.—Only two were inoculated. Both died, one after 109 days, the other after 256 days. Both showed corneal opacity and presented the same symptoms as those described in Nagana rabbits, but in a much milder degree. Guinea-pigs— The guinea-pig, like the monkey, is almost refractory to this strain. Twenty-one animals were inoculated. Nineteen of these proved refractory, and the remaining two only showed trypanosomes on one occasion and appear to have recovered. Rats inoculated with their blood remain unaffected. White Rat.—This strain is virulent to rats. Forty-eight were inoculated, and all died, on an average, in 30°8 days (7 to 94), with enormous enlarge- ment of spleen, and the blood swarming with trypanosomes. COMPARISON OF THE THREE STRAINS OF THE TRYPANOSOME OF THE NATURALLY InrecteD DoG StTRAIN IN REGARD TO THEIR VIRULENCE TOWARDS VARIOUS ANIMALS. Table IV.—The Average Duration, in Days, of the Disease in various Animals ot the three Strains. The letter R means that the animal is refractory. Strain. | Ox. | Goat. | Sheep. | Monkey. | Dog. | Rabbit. | Guinea-pig. | White rat. I R. R. 64 R. 29 | 182 R. 29 II oe ee ae R. Ber R. 30 111 a R. = R. a | = R 46 causing Disease in Man in Nyasaland. 137 Table V—The Average Duration, in Days, of the Disease in various Animals of the three Strains combined. The letter R stands for “ refractory.” 4, | @uinea- White Ox. | Goat. | Sheep. Ee Monkey. hae Rabbit. | oie a | | \ | Average duration,| R. oC: 37 182 Tit off BL | in days | Number of animals, 4 21 | Be 19 2 Bil | 6) | employed | | | | | Table VI—The Percentages of Recoveries in various Animals infected with the Naturally Infected Dog Strain. Three strains combined. | 7 7 | | Ox. | Goat. | Sheep. | Monkey. | Dog. | Rabbit. Crea White | | | [Oe Ee } | | | \ | | | Percentages ......... 100 100 50 =| 100 21 0) 100 OR | Number of animals. 1 7 Zig 6 24 2 2 48 | employed | | | | | From Table VI it will be seen that the Naturally Infected Dog strain is not fatal to oxen, goats, monkeys, or guinea-pigs, whereas it killed 79 per cent. of the dogs and 100 per cent. of the white rats. COMPARISON OF THE TRYPANOSOME OF THE NATURALLY INFECTED Doc STRAIN WITH THE TRYPANOSOME CAUSING DISEASE IN Man IN NYASALAND (TRYPANOSOMA BRUCEI VEL RHODESIENSE). Table VII—The Average Duration of Life, in Days, of various Animals infected with the Naturally Infected Dog Strain and the Human Strain. The letter R stands for “refractory.” Strain. | Ox. | Goat. | Monkey. | Dog. | Rabbit. Guinea-pig. | White rat. | ass es | ae | Naturally infected R. | R. R. | Byf |) Ica ta R. 31 dog | | | JEIRITEN, eaphoceedncocee | 134 42 26 | 34 | 28 | 67 30 It is curious that this strain, although evidently harmless to oxen, goats, monkeys, and guinea-pigs, is quite as virulent as the Human strain to dogs and rats. 138 Trypanosome causing Disease in Man in Nyasaland. Table VIII.—The Percentages of Recoveries in various Animals infected with the Naturally Infected Dog Strain and the Human Strain. TEETER! soonannobeobons | 80 | 0) Strain. | Ox. , Goat. | Monkey. | Dog. | Rabbit. | Guinea-pig. “White rat. | | | | | | Naturally infected | 100 100 | LOO tee Zee 0 100 | (0) dog | | ae 50 0 0 0 | 0 This shows the great difference in regard to action on animals which exists between the Naturally Infected Dog strain and the Human strain,* and if similar tables referring to other strains—for example, the Zululand 1913 Straint—be compared, the same difference is found. It might be said that this alone is sufficient to make it rank as another species, and, as already mentioned, if this strain had been found among the wild game and wild Glossina morsitans in Nyasaland, this would have been justified. It was, however, only found in three chronically infected dogs, and so it is thought best with our present knowledge to include it among the strains of Trypanosoma brucet vel rhodesiense. If in the future it should be decided to give it specific rank the name T. anceps is suggested. This name seems appropriate on account of the uncertainty which exists as to the classification of this trypanosome. CONCLUSIONS. 1. The Naturally Infected Dog strain is fatal to dogs, rabbits, and white rats, but oxen, goats, monkeys, and guinea-pigs appear to be refractory. 2. The Commission is of opinion that this is an aberrant or exceptional variety or strain of the trypanosome causing disease in man in Nyasaland— T.. brucei vel rhodesiense. * © Roy. Soc. Proc.,’ B, vol. 87, p. 35 (1913). + Ibid., B, vol. 87, p. 493 (1914). 139 The Vapour-Pressure Hypothesis of Contraction of Striated Muscie. By H. E. Roar. (Communicated by Prof. C. 8. Sherrington, F.R.S. Received April 16,— Read June 18, 1914.) (From the Laboratory of Physiology, St. Mary’s Hospital Medical School.) In 1854 Graham suggested that muscular contraction might be due to osmotic influences (6). In 1878 FitzGerald suggested, on the other hand, that muscular contraction might be due to changes in tension at the surface of the fibrils of muscle (4). The application of these two hypotheses to the problem of muscular contraction is still under investigation and discussion. Two principal objections have been raised to the osmotic explanation :— (1) “ Neither theoretically nor practically is it possible to construct a model in the manner imagined by McDougall, which will, on being distended, produce anything near the shortening which is observed in living muscle. Living muscle may contract certainly to one-third of its length” (25). (2) “It is impossible to conceive that water will flow into the sarcostyles from the ssarcoplasm, not in a third or a tenth of a second, only, but as in the case of the wing muscles of insects, in less than one-two thousandth of a second” (10). These two objections seem to be shared by Bernstein (2), Macallum (9), and Schafer (24). On the other hand, Macdonald(11), Macdougall(13), Pauli(15), and Zuntz (27), support the hypothesis that muscular contraction is due to a rise of osmotic pressure. The structure of striated mammalian. muscle is generally agreed to be a ‘series of fibrils suspended in sarcoplasm, the whole surrounded by the sarco- lemma. The fibrils consist of alternating bands of anisotropous and isotropous ‘substance, the former corresponding to the portion of the fibril which is contained in the dim band, the latter to the light band of the fibril (26). Accepting this, the model that I wish to describe assumes that during contraction lactic acid is formed, causing the portion of the fibril contained in the dim band to swell and become spherical (see figure). Such a process is exemplified by a parchment paper osmometer containing a protein solution. If the osmometer is placed in a dilute solution of acid, the acid diffuses in and causes an increased absorption of water, with a rise of pressure. One explanation of this rise of pressure is that, as the acid diffuses into the VOL. LXXXVIII.—B. M 140 Mr. H. E. Roaf. The Vapour-Pressure osmometer, it unites with the protein to form an ionising salt. Of this salt the protein ion cannot pass through the parchment paper membrane, and the acidic ion is held back by the opposite electrical charge on the protein. The pressure inside the osmometer consists of the sum of the protein and acidic ions and the free acid; the pressure outside the osmometer consists of the free acid alone. Therefore the excess pressure inside is due to the protein and acidic ions (14, 17, 18, 20). If in calculating the rate of contraction we assume that the lactic acid is liberated in the isotropous substance, and that it must diffuse into the anisotropous substance, we are using the most unfavourable conditions for the ’ Diagram illustrating to scale the osmotic con- traction of two segments of four fibrils of frog’s sartorius muscle. Dim band 0°94 long ; light band 0°9 » long and fibrils 0°3 p diameter. A, relaxed ; B, contracted condi- tion. S, sarcolemma ; shaded portion aniso- tropic substance ; unshaded portion isotropic substance. osmotic hypothesis. The anisotropous substance is considered to be an ellipsoid, but calculations assuming that the anisotropous substance is cylindrical give practically the same results. In calculating the extent and rate of contraction the results will depend upon the dimensions of the structures concerned. Measurements made by Dr. F. O’B. Ellison on frog’s sartorius show that there the length of the dim band is 0°9 y, the length of the light band is 0-9 w, and the diameter of the fibril is approximately 0:3. As the diameter of the fibril and relative lengths of the dim and light bands may vary in different muscles, the calculations are made so as to show what variations these differences may allow in the extent and rate of contraction. In the Table (p. 146), therefore, Hypothesis of Contraction of Striated Muscle, 141 the calculations are made for ellipsoids 0°9 long and 0°4, 0:3, 0:2, and 0-1 » in diameter ; that is with half-axes, a and 0, 0°45 » and 0:20, 0:15, 0:10, and 0:05 wu respectively. We can assume that the ellipsoid swells to become a sphere of the same surface area, that the ellipsoid is really an extended sphere with the walls thrown into folds, or that it has inextensible longitudinal fibres and extensible circular fibres. The first assumption is not valid, as one cannot imagine a surface area of constant magnitude which would be so mobile as to change from the surface of an ellipsoid to that of a sphere. The other two assumptions give the same results and they form the mechanical basis of the ensuing calculations. We can simplify the description by dealing with one dim band and an adjacent light band, since the result is the same even if one dim band is associated with the two adjacent half light bands. Extent of Contraction. The quotation given above refers specifically to Macdougall’s hypothesis (13), but the same criticism seems to be tacitly applied to all osmotic hypotheses. Assuming that the ellipsoid becomes a sphere, the longitudinal perimeter of the ellipsoid (Table, p. 146, column 3) will be the circumference of the sphere. The dim band, therefore, shortens in the ratio of the length of ellipsoid to diameter of sphere, that is as 2a is to 2r, Not only does the dim band shorten but the area of the dim band increases, The fibrils being closely packed together, the increase in area of dim band will be proportional to the squares of the radii of the ellipsoid and sphere respectively, or as 0? is to7?. The relative volumes of the dim band in the two conditions are given by this ratio multiplied by the corresponding length of the band. These relative volumes, 2a x b?, and 27 x 7?, are given in the Table (columns 4 and 8). The volume of the dim band is more than doubled as the result of contraction (column 9). The extent of contraction can be estimated by assuming—(a@) that, as measured above, the light band is exactly the same length as the dim band, or (0) that the length of the light band is such that, during contraction, the whole of the muscle can be just absorbed by the dim band. These figures are given in the Table (columns 10, 11, and 12). The length of the muscle fibre which can be absorbed by the dim band is easy to calculate, as the area of the light band is exactly the same as that of the dim band before contraction, hence the relative length is the number of times that the volume of the contracted dim band contains the volume of the dim band before contraction. M 2 142 Mr. H. E. Roaf. The Vapour-Pressure A model to illustrate the above principles was made(22) by enclosing four rubber balloons in silk bags and suspending them from a disc ; a rubber cylinder was placed round these, extending from the supporting dise to a similar dise below. The whole was filled with water: a certain amount of water was removed from the cylinder surrounding the balloons and the same volume was injected into the balloons. The total volume was thus unchanged and the model contracted to 50 per cent. of its original length. By more careful construction there is no doubt that a greater degree of contraction can be produced. The diagram shows, to scale, the contraction when ellipsoids 0°94 long and 0:3 w in diameter become spheres of the same circumference, the light bands before contraction being of the same length as the dim band. The results of the calculations in this section are independent of the actual dimensions, but they depend on the relative dimensions. The results in the following section, however, depend on the absolute values. Rate of Contraction. Graham pointed out that “in minute microscopic cells the osmotic move- ments should attain the highest velocity, being mainly dependent upon the extent of the surface” (7). This statement can be amplified by caleulating the rate at which the ellipsoids would swell to form spheres. The amount of liquid that is absorbed (column 13) is the difference between the volume of the sphere (column 6) and the volume of the ellipsoid (column 2). This amount of liquid passes through a surface which can be taken to be of the same extent as the surface of the sphere (column 7). To calculate the rate at which absorption of water may take place through the membrane we may utilise the data which I obtained in some direct determinations of the osmotic pressure of hemoglobin solutions (17, 20). The surface area of the parchment paper was 19 sq. cm. and osmotic equilibrium was attained in about three days by the absorption of about 5 c.c. of water. If the rate of diffusion into the anisotropic substance is the same as the rate of diffusion through the parchment paper, the length of time until equilibrium will be reached will be directly proportional to the volume of liquid passing in, and inversely proportional to the surface area through which the liquid can pass. The increase in volume when the ellipsoids become spheres (column 13) and the surface of the spheres (column 7) are given in the Table. In order to compare the increase in volume and surface of the osmometer with the corresponding measurements of the ellipsoids, the measurements in Hypothesis of Contraction of Striated Muscle. 143 centimetres must be converted into microns: that is,5 must be multiplied by (10000)? and 19 by (10000). The ratio, 5 x (10000)? | Increase in volume of ellipsoid 19 x (10000)? — Surface of sphere denoting the relative surface through which unit volume passes into the anisotropic substance compared with unit volume through unit surface in the osmometer, is given in column 14 of the Table. In calculating the rate of diffusion we must distinguish between two processes, namely, the rate at which lactic acid diffuses from the isotropous into the anisotropous substance, and the rate at which water diffuses from the isotropous into the anisotropous substance. In each case the rate of diffusion must depend on the driving force, that is, the difference of pressure divided by the distance. Let us assume that in the osmometer there is perfect mixing, so that the distance is the thickness of the parchment paper; 43 thicknesses of dry parchment paper were measured in a screw micrometer, their thickness was 3°7 mm, ; each piece is, therefore, at least 0086 mm. thick. As the parchment paper swells when wet, the actual thickness must be greater than this. Further, let us assume that the anisotropous substance is a gel, and let us take the greatest distance, namely, the radius of the sphere (column 5), as the distance that determines the driving force in the ellipsoid. If the difference of pressure between the anisotropous substance and the isotropous substance were the same as that between the contents of an osmometer and its surround- ing fluid, the driving force (being inversely as the distances across which the force acts) would be, as shown in the Table (column 15), very much greater in the case of the muscle fibre; in other words, the osmotic gradient would be much steeper. If acid is placed with the hemoglobin inside the osmometer, equilibrium is reached in two to three days (17, 20), but when the acid is placed outside, the pressure reaches its maximum more gradually, depending on the rate at which acid enters the osmometer. Let us assume that the latter arrange- ment corresponds with the conditions existing inside a muscle fibre; let us take three days as the period required to reach osmotic equilibrium once the pressure has been produced, and let us take 14 days as the period required for the acid to diffuse into the osmometer and cause a rise of pressure. As the parchment paper is saturated with water the rate at which water enters the outer surface of the parchment paper determines the rate at which water enters the osmometer ; that is, the rate of entry is proportional ~ 144 Mr. H. E. Roaf. The Vapowr-Pressure to the osmotic gradient. The time required to reach osmotic equilibrium in the fibril (column 17) is calculated by dividing three days, in seconds, by the relative surface through which unit volume passes (column 14) and also by the relative driving force (column 15). In the case of the diffusion of acid, the acid must diffuse through the thickness of the paper before it can affect the contents of the osmometer. Therefore, in calculating the time required to reach equilibrium, the distance must be taken into account a second time. For this reason the equilibrium for acid will be reached in inverse proportion to the square of the distance (square of column 15). The length of time required to reach equilibrium for lactic acid in the muscle (column 16) is found by dividing 14 days, mm seconds, by the relative surface through which unit volume passes (column 14), and also by the squares of the figures in column 15. In order to control the above calculation the rate of diffusion of lactic acid was measured by allowing it to diffuse into gelatine containing an indicator. A 2-per-cent. solution of gelatine was coloured by Congo red, and tubes 3 mm. in diameter were filled with it. After the gelatine had solidified, the tubes were cut into short lengths and placed into 25 cc. of solution. The length of tube in which the indicator had changed colour was measured at definite time-intervals. Two tubes were placed in each flask, two flasks of each solution were used, and there were two ends to each tube, therefore the measurements are the average of eight determinations in each case. 0:05 N lactic acid. 0:02 N lactic acid. 0:01 N lactic acid. hours. mm. mm. mm. 1 3°4 2°6 2°0 4 7:0 (2x 3°5) 55 (2x 2°75) 4-0 (2x2°0) 25 17:0 (5 x 3°4) 14:0 (5 x 2 °8) 10°1 (5x 2:0) A second experiment was carried out, but neutral red with sufficient alkali to cause a yellow colour was used. The diffusion was slightly slower, possibly because some of the acid was lost by neutralising the alkali. 0°05 N lactic acid. 0:02 N lactic acid. 0:01 N lactic acid. hours. mm mm. mm. 1 3°2 2°3 1°6 4 62 (2x3°1) 4.°6 (2x 2°38) 3:4 (2x1°7) 25 | 15 °4 (5x 3°1) 12 ‘1 (5 x2°4) 9-9 (5x 20) The results show very clearly that the time required for diffusion is Hypothesis of Contraction of Striated Muscle. 145 proportional to the square of the distance, therefore we can calculate the rate of diffusion of lactic acid in muscle fibre. Taking the rate of diffusion as between 2 and 3 mm. for the first hour, we can calculate the time for the acid to diffuse one micron as follows:—A ~ micron is 0:001 mm., therefore the acid would diffuse one micron in between (1/2000)? and (1/3000)? of an hour, or between 0:0009 and 0:0004 of a second respectively. The radius of the spheres is about one-third of a micron, therefore the time for the acid to diffuse into the anisotropic substance is one- ninth of the above, or 00001 and 0:00004 of a second respectively. These times are somewhat less than those given in the Table (column 16). In the Table the calculation is for the establishment of equilibrium, but these ‘figures show the time for diffusion of just sufficient acid to cause a change of colour of the indicator. Considering this difference and the different ways in which the times were calculated the agreement is very satisfactory. If we sum the two time-intervals (columns 16 and 17) we get the total time for complete osmotic.contraction of muscle, if the two processes are successive (column 18). As the processes would be more or less concurrent, the time would actually be less. In spite of the fact that in these calculations, wherever there was any doubt, the assumptions were made to the disadvantage of the view now defended, it will be seen that if the maximum contraction of frog’s sartorius requires at least 0°04 second there is more than sufficient time for the contraction to be the result of osmotic swelling of the anisotropic substance. The calculation for the wing muscles of insects has not been attempted owing to the lack of data. Although the fibrils are coarser the presence of minute canals in the anisotropous substance must be remembered, so that probably the results would be equally convincing. If it were permissible - to consider the rate of diffusion of water as proportional to the square of the distance the ellipsoids would become spheres with sufficient rapidity to allow ample time for the contraction of insect’s muscle. Absolute Force of Muscle. In the preceding discussion the term osmotic pressure has been used, but any other process which produces a lowering of vapour pressure could be used as the force by which water is moved into the anisotropous substance. In order, however, to make the discussion more concrete, I may quote the following experiment as showing the effect of lactic acid on the osmotic pressure of a protein (14, 18). As the proteins of muscle are difficult to obtain in an unaltered condition I have used hemoglobin. It is more sensitive to reagents than the proteins of serum (21). Mr. H. E. Roaf. The Vapour-Pressure 146 G10. 0 100: 0 67000: 0 966 00S82 €860- 0 ¥6-Z 08- 61 v 8620: 0 620-0 — 87000- 0 V8sG OOZTE L160- 0 6. OT 9S. G € LLZ0. 0 GLO. O L¥000- O 896 OOLSE L960- 0 6-16 &6-@ SG PEZ0- O 0&0. 0 &¥000- 0 PSG OOFPT S80: 0 L. PE ¥6- 1 I My W@2__. EL ee ; , Uraetatl SL Te) aghast] wuisizo | 6 109 2g ee IE) Foes | pene eecis0) | ce OOOO EN GY ayy Jo a8uque0 Se LT ‘199 +91 “1°D skp ¢ (GT 109) XFL 100 | — ssouyorty, «(O000T) x & Z 1-9 TD) rod Stuumoypoy oxy WIp poqovigu09 UOTZOR.IGMOD ‘oroqds skvp FI “TOJOULOULSO OT]4 “1994 aULOU1SO -oroyds 09 0q [ITA 101 &q peqrtosqe epNosnUL JO worIsue4sTp “oTOSNUE UL 4vq} oy} Ur sovzans yun | F S2NLODE | deren ong Spurq Ajeqe duos eq eyeTduroo «oF [eurxeur sonpoad| ur umiaqiymbe | yar pereduroa | ysnoaryy Sutssed prosdryje tae]a) [PER ENG exe) UBd FOIA purq spmooes 04 oanssead U1e4}8 09 plow LoF | ]IAqy ut ems @UINIOA 4IUN 07 ways STANIOA | Gu4 fq pogrosqe | WIP Pus qu sty UL Olly [LO], DTJOUSO LOF spuooes ut ewry, | -setd oyouso | ojosnur ur sossed UL osvotouy {jaqojdmoo oq poqyovaqu09 Spuooes Ul OUI, JO qustIpeas | owmyjoa 4run yor ueo oposum ejoym | “UFO Ygsuey OATIVION Ysnomyy oovyzans oya vq y48u9T yons [PLC BEK|O) Bae cued Jo st puuq USI] FL “ST “LT ‘OT “ST ‘FL eI “GL ‘TI Y- GE 00: 2% F6FO- O ¢90- T O&OT- 0 16d: 0 200: 0 €8- T L¥00- 0 $0.0 GP-O | ¥ - 2-88 81-9 9S¢0- 0 GST- T S91T- 0 £08. 0 0600- 0 806: T 8810. 0 O1-O SFO | § L.G& GG. & 0990- 0 G6z. T G8EI- 0 TE: 0 £020- 0 G0: & SCPO- O SI-0 GhO | @ G-L& 9T- & PLL0- 0 OFP- T OT9T- 0 8&&- 0 090: 0 ST-& S¢40-0 | 02-0 SFO | T { | Diy wlth ela $ wu 7) 72 i OOT x 42 Pp [09 + 8 ‘TOD Tee ‘aroyds Jo | ‘eroyds Fo | ‘eroyds axe ; —y48ueq yeutsi10 | -wormorryuoo exojoq aor SRD eovgAng eWNPOA — FOSMIPPT | roryovaquos | “woroas gous | Ue ee ae ase [eheseKgy WAND) EAE) AO Jo pue L 9 4 eL0Foq yeoudiye | ‘prosdryje HN tors gyzoose dee orNOTTOF | Naeqy Cy GOL oe LOO N T reg wan _| paeq mrp jo yo taal CA areas goeP Oe epee Jo oulnjoa JO ouUINTOA | LoJoUITIeg | oWN]OA os -w0o “yysue, jenbo | wip oy} Fo eunjoa oaTqETONy Ay aanrpey |. | Fo ore pueq wp oY} JO o1neyy ‘ prosdiyje se pue purq 4 SI] 37 aouedozUIMoAIO oumes TALIM eLoydg ‘OT 6 ‘8 v ‘$ G “i Be ee ee ee Eh ‘potmmbod O18 SJUSWIOINSBOU OINTOSGe 94 UOLJOVIJWOOD JO OFVI OYJ SuTyepNoTeo url ynq ‘pormbeaz ore 4eyy [[v 91’ SUOISMOUIIP OATIVIOL OY} WOTZORIMOO FO 4UO}xXe oYy Suyepnoyeo uy «(wut [90.9 = 7) Ajoatoadser 7 pur ‘71 ‘/ Ul UdATS OLB OUUNTOA puUe ‘odvJINS ‘yysuey Jo syuouLomMseeul [Ty ‘QPOSNJAL JO SOMIIGISsog oYOUISE OY} SUTPeIYSNIII 9[q eT, Hypothesis of Contraction of Striated Muscle. 147 Sheep’s blood was centrifugalised and the corpuscles washed once with 0-9-per-cent. sodium chloride. The corpuscles were diluted with an equal volume of distilled water and the resulting solution of heemoglobin was placed in osmometers. | | [ | : ‘ | Pressure per Solution outside ; ipa eime, | EO NEO | 1-per cent. Maximum pressure. | in days, to | of organic : 5; osmometers. : of organic | maximum. matter. BTR | | | | mm. 600 c.c. water oo... 108 | 2 4 Si ime 600 ce. water ooo. 100 2 146 | 6°85 600 c.c. 001 N lactie acid 248 | Average in- | 8 14°5 17-2 600 c.c.0-O1N lactic acid) 266 J crease 153 7 140 | 19-0 This experiment illustrates how the anisotropous substance may swell as the result of an increase in osmotic pressure due to the formation of lactic acid. Bernstein (2) states that the force of frog’s muscle is about 600 grm. per square centimetre, which amounts to a pressure of 445 mm. of mercury. The percentage of protein in muscle is about 18, so this is equivalent to 248 mm. per 1 per cent. of protein, a pressure not much greater than that recorded above with a strength of acid which is less than that formed in muscle during contraction. Macallum (8) and Macdonald (12) have shown that the inorganic con- stituents of muscle are mainly confined to the dim band. Some years ago I endeavoured to find out if the proteins were also mainly confined to the dim band. Crab’s muscle was stained by immersion in dilute Millon’s reagent, copper sulphate followed by alkal, and other protein reagents. In each case the dim band appeared to stain more deeply. The experiments were discontinued, as the appearances might have been due to the colour seeming darker because of the opacity of the dim band. If the proteins are unequally distributed less pressure per 1 per cent. of protein would suffice for muscular contraction. Another way to calculate the osmotic force of muscular contraction is to regard the increase of pressure as due to the amount of lactic acid combined ‘as an ionising salt with protein, the increase in pressure being considered to be due to the addition of the lactic ion to the protein. The increase in pressure (153 mm.) is equivalent to a 0:009 normal solution. This added to the 0:01 normal solution, the amount of uncombined acid which should be equally distributed inside and outside the osmometer, 148 Mr. H. E. Roaf. The Vapour-Pressure gives a concentration of 07019 normal. To produce this concentration in the 25 c.c. of solution in the osmometer would require 5 c.c. of a decinormal solution of lactic acid. It would be difficult to show the disappearance of this amount from the 600 c.c. of. N/100 solution outside the osmometer. In the last osmometer of the experiment 4°9 c.c. had disappeared, but such a close result is probably a coincidence. If we wish to make a similar comparison in muscle we must calculate the relative volume of the anisotropic substance to the whole muscle in the uncontracted and contracted states. A disc of paper was divided into as many smaller discs as possible by a cork borer. These smaller discs collec- tively weighed 0°581 grm., and the residue weighed 0°224 grm. It was found, however, that there were certain larger pieces from the edge of the large disc that were too small to form small discs. If these pieces were excluded the weight of the residue was 07153 grm. The larger the whole disc the less relatively is the waste at the edges, so that we can consider the weight of the smaller discs as between 79:2 and 72:2 per cent. of the whole; these weights represent the relative area of the combined smaller dises to the whole disc. These figures give the relative volume of small cylinders inside a larger cylinder of the same length. The volume of an ellipsoid is two-thirds of a cylinder of the same length and diameter. Therefore the volume of the ellipsoids will be two-thirds of the above, namely, between 52°8 and 48:1 per cent. of the volume of the dim band. As the volume of the light band is. approximately the same as that of the dim band, the ellipsoids will occupy half the above amount, that is between 26:4 and 24°1 per cent. of the uncon- tracted muscle. In the contracted condition the whole of the muscle is contained in the dim band, and the ellipsoids become spheres. In the preceding paragraph we have calculated the relative volume of cylinders inside a larger cylinder, and since the volume of a sphere is two-thirds of the volume of a cylinder surrounding it, we see that the spheres occupy between 52°8 and 48-1 per cent. of the whole contracted muscle. If as in the osmometer experiment the concentration of lactic acid in the isotropous substance remains as low as 0:01 normal, and the concentration in the whole muscle is equivalent to 0°025 normal (5), we can calculate the amount of lactic acid in the anisotropous substance. If the muscle is prevented from shortening, the concentration would be about 0°07 normal. If, on the other hand, the muscle is allowed to shorten, the concentration of lactic acid in the anisotropous substance would be about 0:04 normal. These figures are obtained by subtracting from the totai concentration of lactic acid Hypothesis of Contraction of Striated Muscle. 149 the amount contained in the isotropous substance and concentrating the remainder in the anisotropous substance. That is [0:025 —0:01 (100—X)/100] . 100/X, where X is the percentage of the whole muscle formed by the anisotropic substance as calculated above. If we subtract the amount of uncombined lactic acid (0:01 normal) from the above we have left 0:06 normal and 0:03 normal respectively, as the amount of lactic acid as ionising salt in the anisotropic substance. These would give osmotic pressures of 1018 mm. and 509 mm., both of which exceed the 445 mm. which Bernstein states is the maximum required (2). By the formation of a lactic-acid salt of protein a negative potential would be produced. This is apparently the explanation of the negative potential produced when a muscle contracts (1, 3, 16, 19). The time relations of acid production are such that the acid can be considered as the cause of contrac- tion (23). There is one further point. It is claimed that if the contraction of muscle is due to changes in surface tension one explanation suffices for it and for amceboid movement. Granting that amceboid movement is due to changes in surface tension, may it not be that the changes in surface tension are due to electrical charges, the result of acid production? Thus amceboid movement and muscular contraction may be related, but in a different way from that advocated by those who claim that muscular con- traction is due to changes in tension at surfaces of separation. We do not yet know enough about unstriped muscle to suggest how its contraction is brought about. The above outline gives a definite conception of muscular contraction which accounts for all the known facts. If an equally tangible explanation could be furnished based on surface-tension changes, the two hypotheses -might be compared one with another, and the two together might lead to a conception closer to the truth than can be arrived at from either hypothesis taken by itself. Summary of Conclusions. 1. The contraction of striated muscle can be explained on the hypothesis that lactic acid is set free, and that this combines with protein to form a salt, with a consequent rise of osmotic pressure. 2. Muscle can shorten by osmotic processes until its length is somewhere between 37 and 3 per cent. of its original length (Table, columns 10 and 12). 3. The osmotic process can occur in frog’s sartorius in less than 0-04 of a second (Table, column 18). 150 Contraction of Striated Muscle. 4. In order to determine whether all cases can be explained by this hypothesis it is necessary to have measurements of the structures concerned. Insect’s muscle, for instance, should be the test as regards the rapidity of contraction. 5d. The amount of lactic acid formed during muscle contraction can cause sufficient rise in osmotic pressure to account for the force exerted by muscle contraction. 6. The electrical changes in muscle can be explained by the formation of a protein salt of lactic acid. Many of the experiments on which this paper is founded have been carried out by the aid of grants from the Government Grant Committee of the Royal Society. REFERENCES. 1. W. M. Bayliss, ‘Roy. Soc. Proc.,’ B, vol. 84, pp. 243-248 (1911). 2. J. Bernstein, ‘ Pfliiger’s Arch. f.d. ges. Physiol.,’ vol. 109, p. 323 (1905), and vol. 128 p. 136 (1909). 3. J. Bernstein, ‘ Electrobiologie,’ pp. 87-107 (Vieweg and Sohn, Braunschweig, 1912). 4, G. F. FitzGerald, ‘Trans. Roy. Dublin Soc.,’ vol. 1, p. 95 (1878). 5. W.M. Fletcher and F. G. Hopkins, ‘Journ. Physiol.,’ vol. 35, p. 247 (1906). 6. T. Graham, ‘Phil. Trans.,’ 1854, p. 177. 7. T. Graham, Jbid., p. 227. 8. A. B. Macallum, ‘Journ. Physiol.,’ vol. 32, p. 111 (1905). 9. A. B. Macallum, ‘Journ. Bioi. Chem.,’ vol. 14; ‘Proce. Amer. Soc. Biol. Chem.,’ p. i (1912). 10. A. B. Macallum, Jo7d., p. xvi. 1l. J.S. Macdonald, ‘Quart. Journ. Exper. Physiol., vol. 2, p. 5 (1909). 12. J.S. Macdonald, Jbid., p. 78. 13. W. Macdougall, ‘Quart. Journ. Exper. Physiol.,’ vol. 3, p. 53 (1910). 14. B. Moore and H. E. Roaf, ‘ Kolloid-Zeitschrift, vol. 13, p. 133 (1913). 15. W. Pauli, ‘Kolloidchemie der Muskelkontraktion’ (Theodor Steinkoff, Dresden and Leipzig, 1912). 16. W. Pauli, Jéed., pp. 5 and 6. 17. H. E. Roaf, ‘Quart. Journ. Exper. Physiol.,’ vol. 3, p. 75 (1910). 18. H. E. Roaf, bid., p. 171. 19. H. E. Roaf, Jbid., p. 178. 20. H. E. Roaf, ‘Quart. Journ. Exper. Physiol.,’ vol. 5, p. 131 (1912). 21. H. E. Roaf, ‘Journ. Physiol.,’ vol. 38, Proceedings, p. ii (1909). 22, H. H. Roaf, Zbzd., vol. 43, Proceedings, p. xxxviii (1912). 23. H. E. Roaf, Jbcd. (in course of publication). 24, HK. A. Schifer, ‘Quart. Journ. Exp. Physiol.,’ vol. 3, p. 63 (1910). 25. EH. A. Schafer, Jbid., p. 69. 26. EH. A. Schiifer, ‘Text-book of Microscopic Anatomy,’ p. 175, e¢ seg. (Longmans, London, 1912). 27. N. Zuntz, ‘Die Kraftleistungen des Tierkérpers, pp. 21-26 (Paul Parey, Berlin 1908). 151 On the Nutritive Conditions Determining the Growth of certain Fresh-water and Sow Protista. By H. G. Tuornron (New College) and Grorrrey SmitH, Fellow of New College, Oxford. (Communicated by Prof. G. C. Bourne, F.R.S. Received April 20,—Read May 14, 1914.) [PLATE 12.] It is well known that in ponds and lakes cycles of development, occur, in which various kinds of animals and plants replace one another in succession, but the conditions are usually so complex that the succession rarely repeats itself with regularity from year to year, and it is impossible to assign, with any certainty, the successive phases to their determining causes. The same kind of cyclical development occurs in artificially made organic infusions, where bacteria, alge, flagellates and ciliates replace one another in irregular sequence. The object of this paper is to indicate certain lines of experiment upon which it may be possible to attack this problem. Woodruff has contributed some data for studying the underlying causes of these successive events, and the work of numerous authors has added to our knowledge of the factors regulating the growth of alge and diatoms. Amongst these may be mentioned the work of Oswald Richter* on the nutrition of fresh-water algz, and that of Miquel,t and more recently of Allen and Nelson,t on the culture of diatoms. The work of these authors tends to show that, even in the case of alge and diatoms in which nutrition appears to be holophytic, the presence of some organic matter in the culture medium is of great assistance to the growth of the organisms. The experiments with Huglena viridis were carried out with the object of investigating the nature of this organic matter which exerts a beneficial influence on the growth of apparently holophytic protista. The method employed is to use a culture medium containing a constant proportion of the inorganic salts necessary for the nourishment of a holophytic organism, and to supply the organic matter in the form of * Oswald Richter, ‘Die Ernihrung der Algen,’ 1911. + Miquel, ‘Le Diatomiste,’ 1892. {t E. J. Allen and KH. W. Nelson, “On the Artificial Culture of Marine Plankton Organisms,” ‘Journ. Marine Biol. Assoc.,’ vol. 8, No. 5, 1910. 152 Messrs. H. G. Thornton and G. Smith. chemically pure organic compounds instead of the indefinite composition of an organic infusion. The method employed in the cultures of Huglena viridis has also been used to study the minute bacterial feeding flagellates living in the soil. Expervments with Kuglena viridis. In culture experiments with Huglena gracilis, Zumstein* found that a much improved growth could be obtained if a little organic matter was added to the culture medium. By increasing the amount of organic matter in the medium, he found that Huglena gracilis could be induced to change its mode of nutrition, living solely as a saprophyte. Under these conditions the Euglena passed into an Astasia-like form, the chlorophyll disappearing and leaving only the colourless leucoplasts. When living saprophytically, the organism could thrive in the dark as well as in the light. Zumstein found that the green coloration was reassumed if the Astasia form was brought back into a solution containing only a small amount of organic matter and kept in the light. Treboux} was able to grow Huglena gracilis in solutions containing citric acid, but found that Huglena viridis could not be grown under these conditions. Thus, there appears to be a marked physiological. difference between these two species of Euglena, a fact which is emphasised by the earlier work of Klawkinet on Huglena viridis. His experiments showed that it was impossible to make Huglena viridis thrive well in the dark. When kept in the dark in a medium containing organic matter, the Euglene remained alive, but did not lose their chlorophyll or show a perceptible increase. Our experiments with this species of Euglena have confirmed the results obtained by Klawkine, and show that Huglena viridis is not able to thrive in the absence of light, even when placed in. the optimal culture medium and in the presence of suitable organic matter. It is thus evident that Huglena viridis is a more essentially holophytie organism than Euglena gracilis, a fact which tends to simplify the issue when we come to study the physiology of its nutrition. By appropriate methods, a culture of Huglena viridis has been kept in active growth in test-tubes by inoculation from tube to tube, for a period of about two years. * H. Zumstein, “Morphologie und Physiologie der Huglena gracilis,” ‘ Pringsheim’s Jahrbiicher f. wiss. Botanik,’ vol. 34, p. 419 (1899). + O. Treboux, “Organische Siuren'als Kohlenstoffquelle bei Algen,” ‘Ber. d. D. B. Ges.,” vol. 23, p. 482 (1905). t W. Klawkine, “Recherches biologiques sur l’Astasia ocellata et Y Huglena viridis,” ‘Ann. des Sci. Nat., Zool.,’ série 6, vol. 19, and série 7, vol. 1. On the Growth of certain Fresh-water and Soil Protista. 153 The medium used for growing these organisms has been a mixture of inorganic salts given by Miquel in his paper on the growth of diatoms.* To this medium, which contains all the elements necessary for the growth of a green plant, it has been found necessary to add some organic material in order to obtain an active growth of the organism. The composition of Miquel’s fluid is as follows:— Solution A. Solution B. USS Odea. tec nrsa suid ss as 10 grs. Sodium phosphate............ 4 gers, IN (a © Recs trae vats. cbiaae IO Calcium chloride .:..:..:.... 4, Sodium sulphate ...... Be iss Hydrochiloniciacidie-nsceeee: 2 cc. Ammonium nitrate ... Ula Perchloride of iron, sat. sol. 2 ,, Potassium nitrate ...... eee “AY AEN E) char is A reg ae A era SOmes Sodium nitrate......... Dis Potassium bromide ... 02,, Potassium iodide ...... OFIL WWiathenta ss uacueenoatees 100 rh) To make up the fluid, 40 cc. of solution A and 10 cc. of solution B are added to 500 cc. of tap water, and the mixture is filtered. It was necessary at first to determine the strength of Miquel’s fluid best adapted for growing the Euglena. In a number of preliminary experiments it was found that the best growth could be obtained when 4 cc. of the above Miquel solution were added to 6 cc. of tap water, the organic matter being supplied by 1 cc. of hay infusion. The experiment was performed by inoculating the tubes with a very small amount of the stock culture, introduced by means of a capillary pipette. The tubes were kept in diffuse daylight at room temperature; attempts to hasten the growth by incubation at 75-80° F, were unsuccessful, the Euglena dying, or at any rate failing to flourish at this temperature. In the absence of any organic infusion, the Euglena either failed to develop or else multiplied very slowly, and the fluid in the tube never became crowded with free-swimming organisms so as to appear opaque and green. The addition of 1 or } ¢.c. of hay infusion, on the other hand, caused a thick growth which, after the lapse of 10-14 days, filled the tube with myriads of free-swimming individuals, giving a totally different appearance to the con- dition seen in the tubes to which no organic matter had been added. It was found, however, that the efficacy of the hay infusion varied very greatly according to the length of time during which bacterial growth had continued in it. Thus, a fresh hay infusion, after being sterilised, was found * Dr. Miquel, ‘Le Diatomiste,’ No. 9, June, 1892. 154 Messrs. H. G. Thornton and G. Smith. to have a much feebler effect in stimulating growth than an infusion which had been kept for some weeks and in which bacteria had been allowed to multiply before it was sterilised. On the other hand, the same infusion, after being left for several months, lost much of its previous efficacy. Similar results were obtained with other vegetable infusions. A typical experiment, showing the effect of the various dilutions of Miquel solution, and of the presence and absence of organic matter in the medium, may be seen in Table I. Table I—Cultures inoculated on April 21. Tube ot : Growth on Growth on No. Composition of Medium. April 29: May 21. 1S el OxcrcrMirquelysolutionsnreereetst te teeter ereeteeee ree en ere eeer eee None None 2 8 c.c. i » +2 c.c. distilled HO ............... 3 50 3 6 c.c. 55 » +4c.. 5 Pane toe BoeR AED . of 4, 4 c.c. a 9 tr Ce@ re Fy EC eee 5 5 5 2 c.c. rs on ae. GG ae OT aie seer emo 9 oD 6 10 C.C.. ” ” \ ee q ( ” ) 7 8 c.c. e 9 AP) OL a jy teal | eS ke ; 8 6 e.c. ” ” +4 c.c. ” 900 a S: 3 ” | ” 2 | 4 C.c. Ls hs ae c.c. - a ; ne ( Slight growth Be | C.c. Ms - C.c. : eae, ; Wit |} 3} Ce, 3 . +2 c.c. tap water ig 1 AS eee None ‘é 12 | 6yere: 55 » +4 ce.¢. Rt mace eee rae one aie 39 5 13 4 c.c. na 6+ Gree. Plaats tar cies ncHonetia in Slight growth “ 14 2 c.c. , +8 cc. BOE at Se aA Ros Soncodoue 45 | 15 | Scie. ms A +2 c.c. i eae a il None 5) 16 6 c.c. ee » +4 ce. PME apne: |“ 8 || Slight growth i 17 4 ¢.c. » +6 c.c. ie tee e en 3 | Very strong | Very strong | ase | growth growth 18 2 cc. 60 » tore ee ean Dooce J +°7 | Strong growth | Good growth These experiments show that the best growth was obtained in tube No. 17, in which the culture medium consisted of 4 ec. Miquel solution + 6 ce. tap water, to which 1 ¢.c. hay infusion was added; while in those tubes to which no organic matter was added, growth was either totally absent, or else a slight growth was observed in those tubes in which the proportions of Miquel and tap water approached the optimum. It must be noted that when tap water is replaced by distilled water, the growth is either prevented altogether or else is very slight, even when the necessary elements for growth are given in the Miquel and organic matter. This can be seen in Table I in tubes Nos. 7-10, as compared with tubes Nos. 15-18. A similar result was noticed when cultures were made with Miquel fluid that had been made up with distilled water instead of tap water. The improved growth in tap water might be due either to the On the Growth of certain Fresh-water and Soi Protista. 155 difference in osmotic pressure or to the influence of some constituent, organic or inorganic, in the tap water. An attempt was made to discover which of these was the determining influence. An artificial tap water was made up from an analysis of Thames water furnished us by Mr. W. W. Fisher. This contained :— Parts per 100,000. INGO Trieste terse roca 2°8 INGING@ atic ieke ine cs 0:7 IMCS Og eee cnincptetsndee 1°4 CaSO pre ceesttiacccnit 2°8 CaCO3 svetstefelstolctatetohFoneclereys 22°3 SOs heat semcctaaueeres 1:0 Motalasolidisase-eence 31:0 Somewhat conflicting results were obtained when this artificial Thames water was used to replace natural tap water: on the whole the growth obtained was not so good as when natural tap water was employed, but since it was possible to obtain quite good growths with the artificial medium, it must be concluded that the superiority of the media containing tap water is due to some slight alteration in the proportions of the inorganic constituents. Having determined the optimum conditions for growth as far as the inorganic constituents are concerned, namely, 4 c.c. of Miquel solution + 6c.c. natural tap water, the question of the nature of the organic matter used by the Euglena was then taken up. In the experiments given below, the ordinary method of inoculation by means of a capillary pipette was employed, and, in addition, another method which gives more rapid results. In this second method, an old culture tube, in which Euglena has been growing for a long period, is taken. In such a tube there is a ring of encysted Euglena adhering to the glass at the surface of the liquid. The liquid is poured away and the encysted Euglena, which is very firmly attached, may then be thoroughly washed with tap and distilled water. In this way a practically pure culture of Kuglena may be obtained on adding the appropriate culture medium, though in no case has it been found possible to obtain a sterile culture free from bacterial contamination. The following chemically pure substances were added to the “ optimal Miquel” mixture and the tubes inoculated with Euglena, with the results subjoined. VOL, LXXXVIII.—B. N 156 Messrs. H. G. Thornton and G. Smith. 1. Deatrose— In media, in which 0°5-1 c.c. of a 1-per-cent. solution of dextrose was added to 10 ec. of the “ optimal Miquel” solution, no growth of Euglena was observed, but in all cases there was a considerable growth of fungus, probably derived from spores from the stock tube of Euglena. It is probable that the great development of the fungus inhibited the growth of the Euglena, since a slight growth in the optimal Miquel solution was expected. 2. Cane Sugar— The addition of 05-1 cc. of a 1-per-cent. solution of this substance to 10 cc. of the “optimal Miquel” solution did not inhibit the growth of the Euglena to the same extent as the dextrose, but only rarely and after a long period did any noticeable growth appear. The growth of fungus in these tubes was either absent or very slight. 3. Tartarie Acid— The addition of 1 cc. of a 1-per-cent. solution had a purely negative effect, no growth of Euglena but a strong growth of fungus being observed. It is evident from these results that the stimulating element in the organic infusion is not in the nature of a carbohydrate. 4, Peptone— The addition of 1 ec. of a 1-per-cent. solution of peptone to the medium invariably gave rise to a very strong bacterial growth, the bacteria being no doubt introduced with the Euglena, on inoculation. Under these conditions the Euglena scarcely developed at all, although it is not entirely killed off, a slight ring appearing at the top of the fluid. 5. Amido-acids* — : Tyrosin.—As this compound is very insoluble in water, a saturated solution was made up in distilled water. The saturated solution when cold contains the salt in the proportion of 1 in 2400 of water. In the earlier experiments 1 or 2 ¢.c. of the above solution were added to 10 cc. of the “optimal Miquel” mixture. A very strong growth was obtained in this medium, indeed superior to that obtained by means of the addition of a natural organic infusion. The marked difference in the growth of the Euglena in a tube containing this minute trace of tyrosin (1 : 24,000) as compared with a culture in a medium free from organic matter, may be seen in the photograph (fig. 1). * A number of experiments have been made by Loew, Bokorny, and others on the growth of alge in amido and fatty acids. A full literature of this work will be found in Oswald Richter’s ‘ Die Ernahrung der Algen,’ 1911. On the Growth of certain Fresh-water and Soil Protista. 157 In cultures containing 1-2 c.c. of the tyrosin solution, it was found that after a period of about six to eight weeks the Euglena ceased its active erowth and became encysted upon the walls of the tube and especially round the surface. For example, a tube containing 4 c.c. Miquel solution, 6 c.c. tap water, and 1 cc. tyrosin solution, was inoculated with Euglena on November 25. By December 3 there was a noticeable growth, which became very thick by December 30. By January 30 all free-swimming forms had disappeared, and nothing remained but a ring of encysted a i 4 t a eG Fic. 1.—Photograph to show the Growth of Euglena in a Tube (B), containing the optimal Miquel mixture, to which tyrosin solution was added, as described. Tube A shows the slight growth in a control culture containing the optimal Miquel solution, but with no organic solution. Photograph shows a 10-days growth. Euglena. It was found that by replenishing the culture medium in this tube, the growth of the Euglena could at once be revived: within two days after replenishment a thick growth of free-swimming forms was obtained. This suggested that the tyrosin was used up after a certain period. To test this hypothesis, cultures were made in the optimal Miquel mixture, to which tyrosin was added in the solid form, so that as soon as the dissolved tyrosin was used up, fresh tyrosin might go into solution. Cultures grown in this medium showed a very rapid growth of Euglena during the first fortnight or three weeks, but after that the increased development of N 2 158 Messrs. H. G. Thornton and G. Smith. bacteria in the culture usually interferes with the Euglena growth. The following culture may be regarded as typical of the growth of Euglena in a medium of this nature :— Cultures with Tyrosin Media, inoculated February 2. Composition of tube. Growth on Feb. 5. | Growth on Feb. 18. 4 .c.c. Miquel tap + 6 c.c. tap water + solid | Very strong Euglena} Euglena dead or |. tyrosin growth. encysted ; numerous bacteria. 4 c.c. Miquel tap + 6c.c. tap water +1 c.c. | Slight growth of] Very strong Euglena tyrosin solution Euglena. growth; very few bacteria. To avoid the excessive growth of bacteria in the tyrosin and at the same time to ensure the continuous supply of tyrosin, the following culture method was devised. The Euglena was grown in a tube containing the optimal Miquel mixture alone, and the trace of tyrosin was supplied from another tube containing a saturated solution of this substance connected by means of a capillary tube with the Euglena culture. In this way the culture medium is continually supplied with traces of tyrosin solution, but the diffusion is too slow to cause an excess of tyrosin in the tube containing the Euglena. It was found that by this method a strong growth of Euglena could gradually be obtained, nearly free from the bacteria and minute flagellates which always appeared in cultures to which solid tyrosin was added. Of all the culture media employed the thickest and most successful growths of Euglena have been obtained with optimal Miquel mixture to which tyrosin is added. Glycocoll.Cultures were made in optimal Miquel mixture to which was added 1 e.c. of 1-per-cent. solution of glycocoll. These cultures invariably gave a strong growth of bacteria and, at first, a greatly retarded growth of Euglena, though subsequently the Euglena increased. In no case did these cultures compare in strength of growth with the cultures in tyrosin media. It is probable that this retardation was due to the bacterial growth, and this subject will be dealt with in the latter part of the paper. Alanine.—1 c.c. of a 1-per-cent. alanine solution was added to the Miquel mixture as usual. The cultures invariably gave a very strong bacterial growth, and very frequently a bacillus producing a vivid apple- ereen coloration appeared. This green colouring matter was shown not to be chlorophyll, as it was developed more rapidly and to a greater degree in the dark than in the light. At first, as in glycocoll, the Euglena failed to multiply, though after a long period, viz. about three weeks, tubes inoculated On the Growth of certean Fresh-water and Soil Protista. 159 with a ring of encysted forms, as described above, produced a considerable growth. | The great superiority of the tyrosin solutions over the solutions of glycocoll and alanine was very marked. Jt was at first thought possible that this was due to the presence of the benzene ring in the tyrosin, especially since alanine is similar in composition to tyrosin, except that in the former substance the oxypheny] ring is absent. OH ¢) CH;3.CH(NH2).COOH. \ / CHe2.CH(NH2).COOH. Alanine. Tyrosin. With a view to testing this hypothesis, the phenyl compounds of alanine and glycocoll were employed in the culture media. In media containing pheny! glycocoll, YAN \ / CH(NH,).COOH, it was found that no growth took place, even the development of the bacteria being prevented. But, in the media containing phenylalanine, c \ 7 CH2.CH(NH2).COOH. a very strong growth of Euglena was produced, bacterial growth being at first slight, but increasing after some time. Since this compound resembles tyrosin in being very insoluble, it was added to the media in a solid form. Attempts to grow Euglena in saccharin, ci eeu NZ CO ‘ showed that this substance prohibited all growth of the organism. The negative results obtained with phenyl glycocoll and with saccharin showed that, at any rate, the mere presence of the benzene ring was not the essential factor for the growth of the Euglena. Since the substances that are most successful for the propagation of Euglena, namely, tyrosin and phenylalanine, are only very slightly soluble, so that exceedingly weak solutions are used, and since, on this account, bacterial growth in these solutions is very slight compared with that which occurs in the stronger solutions of alanine and glycin, it seemed possible 160 Messrs. H. G. Thornton and G. Smith. that the strong growth of the Euglena might be connected with the slight bacterial growth. To test this hypothesis, lesser amounts of alanine and glycocoll were added to the Miquel mixture. From 0:2 cc. to 05 ec. were added to 10 cc. of the optimal Miquel mixture. In cultures started from a ring of encysted forms it was found possible to obtain extremely good growths of Euglena by this means, the bacterial growth being very much lessened. It is thus obvious that the Euglena can use alanine and glycocoll, as well as tyrosin and phenylalanine, provided its growth is not inhibited by the rapid development of bacteria, such as always takes place in the glycocoll and alanine solutions when they are too strong. This result is of importance as indicating that the amido-acids are used as such by the Euglena, and not after being decomposed by bacterial growth. Thus, attempts made to grow Euglena in tubes containing an alanine medium in which bacterial decom- position had proceeded for a long time were entirely unsuccessful. It is interesting to notice in this connection the fact mentioned above that, in the case of tyrosin, the addition of the solid substance to the tubes causes a considerable bacterial growth, which, after about three weeks, was sufficient to inhibit the proper growth of the Euglena. Nencki* has shown that, . under the influence of anaérobic bacteria, tyrosin is converted into oxyphenyl- propionic acid, OH () \ / CH2.CH2.COOH, and it is probable that, under the aérobic conditions met with in the culture tubes, further decomposition into oxyphenylacetic acid and phenol takes place. Decomposition along similar lines occurs when phenylalanine is subjected to bacterial growth, phenylpropionic acid and phenylacetic acid being formed. Thus, when tyrosin and phenylalanine are added in the solid condition, their solutions are sufficiently strong to allow a growth of bacteria, which decompose them into phenol derivatives that are harmful to the Euglena growth. We may suppose that, in the same way, harmful products are produced by bacterial action on alanine and glycocoll, so that the Euglena is prevented from developing in the solutions of a strength adapted to the growth of bacteria. Other nitrogenous compounds have been tried, ¢.g. urea, uric acid, and * Nencki, ‘Ber. d. Deutsch. Chem. Gesellsch.,’ 1874, p. 1593. On the Growth of certain Fresh-water and Sow Protista. 161 allantoin. All these substances gave negative results, and no growth of Kuglena could be obtained in optimal Miquel mixture to which these substances were added. Thus, no substances other than compounds of the amido-acid type, have been found suitable for stimulating the growth of the Euglena. If we enquire into the part played by the amido-acids in the nutrition of Euglena, it may first be noted that the Euglena is obtaining the greater part of its nutriment from the CO». of the air and from the mineral substances in the Miquel mixture. This was readily proved by keeping a control tube, containing the optimal Miquel solution to which tyrosin had been added, in the dark, in which case the growth of the Euglena was at once arrested (fig. 2). It must also be pointed out that the amcunt of amido-acid present in the optimal culture medium is exceedingly minute; eg. in the case of tyrosin Fia. 2.—Photograph showing the Growth of Euglena in Miquel Mixture and Tyrosin, in the dark (tube C), and in the light (tube D). Photographed three weeks after inoculation. solution the amount of the salt is only 1 part in 24,000 of liquid. It is very remarkable that so minute a trace of organic matter can make so great a difference in the rapidity of growth and reproduction in an organism as shown in the first photograph (fig. 1). It would appear that the organic substance acts more as a stimulant than as a direct source of nutriment. 162 Messrs. H. G. Thornton and G. Smith. The facts observed in the culture of Euglena may be summarised as follows :— (a) In solutions containing no organic matter, the Euglena increases very slowly. (6) By the addition of a trace of organic infusion to the solution of inorganic salts, a good growth of Euglena can often be obtained. (c) The efficacy of the natural organic infusion in stimulating the growth was very variable. (dZ) Minute traces of amido-acids added to the inorganic solution had a remarkable effect in stimulating the growth of the Euglena. (e) Stronger solutions of amido-acids were less successful owing to the rapid development of bacteria in the medium. (7) The Euglena does not appear to live saprophytically on the amido-acid, since it cannot be made to thrive in the absence of light. 2. Eaperiments with Soil Protozoa.* The method of growing Protozoa in solutions containing a mixture of Miquel in tap water to which various organic compounds are added was also -_ applied with a view to studying the protozoal fauna of various soils. The mode of procedure was similar to that employed in the experiments on Euglena. The cultures were made in sterilised test-tubes to which the optimal Miquel solution was added, the solutions also being carefully sterilised. Various organic solutions were added to the various tubes, which were inoculated by adding a small amount of soil to each tube. This method was found to be particularly suited to the culture of the minute soil flagellates, more especially Prowazekia terricola described by Martin.+ The following Table shows a typical series of cultures conducted as described above :— * See Dr. Russel and Dr. Hutchison, “On the Effect of Partial Sterilisation of Soil on the Production of Plant Food,” ‘Journ. Agric. Sci.,’ vol. 3, part 2 (1909); also Goodey, ‘ Roy. Soc. Proc.,’ B, vol. 84, p. 165 (1911). + C. H. Martin, ‘Zool. Anzeiger,’ vol. 41, No. 10 (1913). A flagellate monad, similar to that described by C. H. Martin (‘ Roy. Soc. Proc.,’ B, vol. 85, 1912), was found in small numbers in our cultures. On the Growth of certain Fresh-water and Soil Protista. 163 Composition. pe Observations on March 13. 4 c.c. Miquel tap +6 c.c. tap— +1 c.c. cane sugar solution \ re Few bacteria only. 5 i ontrols. Not : EB ldieraienzs |) meat Wicca +1c.c. cane sugar ......... ) A few soil flagellates. Some ciliates. | Very large numbers of soil flagellates and of soil ameebe. Large numbers of flagellates ; a few f L| ciliates. +solid tyrosin ..............5 l In BenluteaN RE +solid phenylalanine ...... stale manure A few soil flagellates. Very large numbers of flagellates. Large numbers of flagellates. Very few flagellates. Inoeulated with +solid tyrosin ............... leat should +solid phenylalanine ...... +1 c.c. cane sugar ......... 9 11. +solid tyrosin [eae meeu es ee No flagellates. 12. +solid phenylalanine ...... pious Very few flagellates. 13. +1 c.c. cane sugar ......... [eno 4 No flagellates. ¢ | J +lc.c. cane sugar ......... } | el QE esi dE ChE CT 14. +solid tyrosin ............... soil under Very few flagellates. 15. +solid phenylalanine ...... grass land. Fair number of flagellates. The above Table illustrates the fact that while the minute soil flagellates thrive best in tyrosin or in phenylalanine solutions, yet they are able to develop in solutions containing cane sugar. Cultures were made with the object of ascertaining the effect of various other organic substances on the erowth of the flagellates. The flagellates in these cultures were derived for the most part from a stock tube of Euglena culture in tyrosin, in which Prowazekia was also very abundant. The following list embodies the results obtained with various organic substances. Save where otherwise mentioned, the organic compounds were added in the proportion of 1 ¢.c. of a 1-per-cent. solution to 10 ¢.c. of the optimal Miquel mixture in tap water. Growth of the flagellates. ROP LOMC wana canter woust tern aden es se ccpatasoncaicnn BA ete ee Good growth. Tyrosin (OH.CgH;.CH2CH(NH2).COOH) ............... Very strong growth. By rosin (adde dU solid) en sys sees acest sees wcidsciseddnccisas Optimum growth. Phenylalanine, CgsH;.CH2.CH(NH2)COOH (added solid) Strong growth. Alanine CH3.CH(NH2).COOH (0°5 c.c. of 1-per-cent. SOlUUIOMD amet mane coe sieges ics wna hie ea tast us yous: Fair growth. Cisroowolll, Cielo ndely COOLE fsAscenosnocasenanoacnboceacaect Fair growth. Phenylglycocoll, CsH;.CH(NH2).COOH .................. No growth. PAU IN GORI Y eesa inva iartea pe rsee onec ec Nccoe Ne cali lea ec ae erattatina Hs No growth. S ACCA asec aeen emia se ceric Sens Saray iw tiecatuselcine once Susana No growth. ame ssu Catenin cc ste ne ka eon anon eine acisaat nase Good growth. MMancanicraciG aren scare cess sou saisestars nS adcmwansscmosagtaactat « No growth. These cultures showed that the soil flagellates were able to grow in tubes containing a large variety of organic substances, in many of which Euglena is 164 Messrs: H. G. Thornton and G. Smith. unable to thrive. This is the result of the holozoic mode of nutrition of the flagellates, which feed greedily on the bacteria in the culture and are always to be found in greatest abundance in the bacterial scum at the surface. The development of the soil flagellates in the culture is evidently dependent upon the bacterial flora in the tube. In the tyrosin media the bacterial growth reaches its most favourable degree for the development of the flagellates. In the case of cultures in media containing alanine it is frequently found that the flagellates fail to attain their maximum growth, being probably swamped by the excessive numbers of bacteria. In order to discover whether the Miquel salts were necessary for the growth of the soil flagellates, a culture medium was made up by adding solid tyrosin to 10 c.c. of tap-water, and this was inoculated with a strong culture of Prowazekia terricola. This culture entirely failed to develop, remaining almost entirely free even of bacteria, which were evidently unable to develop satisfactorily in the absence of the salts of the Miquel solution. The Prowazekia were observed to flourish in cultures containing very © varied types of bacteria. In order to discover whether the flagellates exercised any selective faculty when feeding upon the bacteria, a number of smear preparations were made from different cultures. The films were fixed with corrosive acetic or with osmic vapour, and were stained with iron hematoxylin. These preparations showed that bacteria of all the types surrounding the flagellates were ingested in quite a promiscuous manner (see Plate 12). The cultures inoculated with various soils, both in the test-tubes and in. drop cultures which were also made, show the enormous abundance and wide distribution of these minute flagellates as compared with other soil protozoa. Although ciliates and amcebee often fail to appear in tubes inoculated with a very small quantity of soil, yet all the types of soil that have been tried have yielded at least some Prowazekia when inoculated into the appropriate culture media. The organism has also been found in tap water and in water from an open-air tank. The very rapid increase of these minute flagellates is also very noticeable. Under the optimum culture conditions it has been found possible to obtain a strong growth of the flagellates within 48 hours of the time of inoculation. On the other hand the larger protozoa, such as the ciliates, do not become even noticeable in the tubes until a week or so has elapsed. The great abundance and wide distribution of the minute flagellates, taken in conjunction with their rapid powers of increase, suggest that in all probability they are of much greater importance than the larger soil protozoa as a factor in the destruction of soil bacteria. Thornton and Smith. Roy. Soe. Proc., B, vol. 88, Plate 12. “ On the Growth of certaan Fresh-water and Soil Protista. 165 The ease with which Prowazekia can be grown in culture media containing tyrosin suggests the possibility of investigating its distribution in various types of soil. Experiments in this direction are at present very incomplete, but as far as they go they tend to show that rich manure soil or leaf mould contains a considerably greater number of the minute flagellates than less rich soils. This is well seen in Table II. (Compare Nos. 4-9 with Nos. 10-15.) In summing up the points observed in the cultures of soil flagellates we notice the following facts :— (a) As compared with Euglena they are able to live in cultures to which organic compounds of very varying natures have been added. (0) This comparative impartiality is the result of the holozoic mode of nutrition, the development of the flagellates being absolutely dependent on the bacterial growth. (c) The presence of the Miquel salts in the solution is necessary for the growth of the soil flagellates and for the proper development of the bacteria upon which they feed. (d) The flagellates can feed upon a variety of different types of bacteria. DESCRIPTION OF PLATE. Soil Flagellates from a Cuiture containing a Mixed Bacterial Flora, showing various Types of Ingested Bacteria. x 2000. Figs. 1-2.—Two individuals containing ingested bacilli. » 93-6.—Individuals containing cocci of two kinds. », ¢-8.—Two individuals containing partially digested bacteria. 166 The Validity of the Microchenncal Test for the Oxygen Place in Tissues. By ALAN N. Drury, B.A., Shuttleworth Student of Gonville and Caius College. (Communicated, with a Note, by W. B. Hardy, F.R.S. Received April 25,— Read June 18, 1914.) (From the Physiological Laboratory, Cambridge.) In the last few years endeavours have been made to locate precisely certain reactions known to occur in the living cell; and much work has been done especially on the reduction place, the oxygen place, and the position of oxidases and peroxidases. The reaction relied upon to indicate the position is always one involving a colour change ; as an example, Unna’s* method of fixing the oxygen place in the cell may be chosen. He uses for this purpose a solution of rongalit white,+ which is a solution of the leucobase of methylene blue kept in a state of reduction by excess of rongalit, an adsorption product of formaldehyde with sodium sulphite. This solution is not affected by air or by light, but is, according to Unna, a test for active oxygen. He places the section in a solution of rongalit white for one minute, then washes in a large volume of water; when, the reducer having been washed away, the tissue is able to show its ability to oxidise, and all the tissue elements which are capable of effecting an oxidation are blued owing to the oxidation of the methylene white to methylene blue. Unna has noticed that it is possible to abolish the staining of the oxygen place by rongalit white, by the action of heat, neutral salts, alcohol, phenol, and other protoplasmic poisons, while an ordinary nuclear stain is not thus affected. Also the intensity of staining can be altered by previous treat- ment with alcohol, formalin, or gum, while the action of a nuclear stain is unaltered. He, therefore, asserts that the stained portions of the tissue are the oxygen places. It is to be noticed that these comparisons are made between a very complex mixture, namely, rongalit white, and a simple solution of a dye, so that they are of little value unless controlled by an exact determination of the influence of the constituents of rongalit white on the absorption process. * “Tie Reduktionsorte und Sauerstofforte des tierischen Gewebes,” P. G. Unna, ‘ Arch. fiir Mikr. Anat., vol. 78 (1911). + “Zur Chemie der Haut. 6.—Hautreagentien,” Unna and Golodetz, ‘Monatshefte f. Prakt. Dermat.,’ vol. 50, p. 451 (1910). Microchemical Test for the Oxygen Place in Tissues. 167 Unna claims also that methyl green picks out the oxygen foci of the cell,. and on experinients with this and other dyes he bases a claim that staining is controlled by the oxidising or reducing properties of the substances exposed to the stain. Neither experiments nor conclusions are above criticism. The problem of dyeing is to discover the conditions which control the condensation of the dye on to a surface separating a solid from a fluid. The presence of oxygen must affect the process, since, ike any other chemical substance, the oxygen will contribute to the chemical, electrical, and mechanical potentials which determine the degree of condensation ; the purpose of this paper, however, is limited to the proof by exact physical experiment that there is no special connection between the presence of oxygen and the process of dyeing. Experiments with Suk. The first substance tried was silk, and the procedure was as follows: A tube was arranged which had a three-way tube joined to it, down one of which nitrogen gas could be passed, down the next rongalit white solution, and down the third nitrogen water, so that they could be changed one to another by closing or opening taps. The rongalit white was freed from oxygen by passing a stream of nitrogen through it for about six hours, the nitrogen itself having been passed through potash bulbs and bottles containing alkaline pyrogallate, to ensure that it did not contain oxygen. The water was boiled for 10 minutes and was cooled, while a stream of nitrogen passed through it. The silk was fixed in the tube and some nitrogen water was passed over it; the water was then turned off and nitrogen gas was passed over it; after an hour or so the gas was turned off and water again allowed to flow over, this again being replaced by nitrogen gas. Such a procedure was carried on for 5-7 hours. The rongalit white was then allowed to flow into the tube, and, after it had covered the silk for 1-2 minutes, the excess of rongalit was removed by a stream of nitrogen water flowing for 5 minutes. The tube was then opened to the air. In no case was any blueing observed until the tube was exposed to the air. The experiment divides itself into three stages :— 1. The exposure of the silk to the rongalit white solution (under nitrogen). 2. The washing off of the excess of the rongalit still under nitrogen. No signs of blueing were observed in these two stages. 3. The exposure of the silk to the air, when the methylene white is oxidised to methylene blue. The staining, as it is usually understood, namely, the condensation of the 168 Mr. A. N. Drury. The Validity of the _solute on to a surface, takes place in Stage 1, so that it is obvious that this stage can occur in the absence of free oxygen—that is, of oxygen other than what may be still clinging to the silk surface. The final effect, namely, the development of colour, is no part of the staining process, and is not an indication that the surface of the silk has any special affinity for oxygen, or is a place where oxidation is taking place. Such experiments as this, however, and similar ones, with gelatine, agar, and the gel of silicic acid, leave the fundamental proposition, which seems to be the basis of Unna’s work, untouched, namely, that the condensation of a basic substance with a high avidity for oxygen, such as the reduction product, the leucobase of methylene blue, or of the fully oxidised coloured body methylene blue, occurs either where oxidation is taking place or where there is a condensation of oxygen. To disprove this contention it is necessary to show that condensation may occur on to the surface of a body already fully oxidised, and completely freed from the film of condensed oxygen, which adheres so tenaciously to solid surfaces which have been exposed to air. As is well known, a solution of a basic dye filtered through a layer of sand is decolorised, the dye being condensed on to the sand particles ; sand, therefore, was chosen. Experiments with Sand. The procedure was as follows :—About 2 inches of sand were packed in a small combustion tube, having a pad of asbestos at one end to prevent the sand from washing through, when the various solutions were passed through. Hydrogen, prepared in a Kipp apparatus, was washed in water to remove acid, passed through a strong solution of alkaline pyrogallate to remove oxygen or traces of acid still remaining, and finally passed through a long tower of calcium chloride to remove water. The apparatus used was similar to that used for silk, save that the three-way tube was fixed on to a combustion tube which contained the sand. The sand was heated to redness in a furnace, and the purified hydrogen passed over it while it was in this condition. The heating was followed by cooling, and this by heating again, still, of course, in a stream of hydrogen, and this was followed by cooling once more in a stream of hydrogen. These series of operations were carried on for three to four hours, so as completely to burn off the oxygen.* After the sand had cooled in the atmosphere of hydrogen, rongalit white solution, freed from oxygen by passing nitrogen gas through it for six hours, was allowed to flowin. It was allowed to stay * Compare “Contact Electricity,” F. M. Spiers, ‘ Phil. Mag.,’ 5th ser., vol. 49, Part 1 (1900). Sn Microchemical Test for the Oxygen Place in Tissues. 169 in the tube for one or two minutes, after which it was washed through by a stream of nitrogen water. The sand was then treated in one of two ways—it was either kept in the tube, the two ends of which were open to the air, or it was washed through by the- nitrogen water on to filter paper. The object of the second procedure was to eliminate the possibility that it might be only the solution of rongalit white trapped between the sand grains that underwent oxidation, the methylene blue so produced being taken up by the sand. If any such trapped solution exists, it is rapidly taken up by the filter paper. The results which the first method gave are interesting. It was noticed that no blueing occurs for a considerable time, although the tube has been | full of air. That is to say, the hydrogen which has replaced the oxygen on the surface of the sand continues in possession of that surface for some time after it is brought into the presence of oxygen. A similar condition is met with in the case of iron; if the oxygen which is normally condensed on the surface of the iron is completely replaced by a layer of hydrogen, the potential difference between the iron and another metal plate is changed. When the iron is brought imto the presence of air the reversion to the original potential difference is very slow, thus showing that the hydrogen is only slowly displaced from the surface of the iron.* Details of an Experiment. 3.10. Sand heated to redness and hydrogen passed through. 3.40. Allowed to cool in a stream of hydrogen. 3.50. Again heated to redness in hydrogen stream. 5.0. Allowed to cool in stream of hydrogen. 5.15. Rongalit white solution passed through and allowed to remain in contact with the sand for one minute. There was no appearance of any colour at allt Nitrogen water was then passed through for three minutes ; again there was no sign of colour in the sand. The nitrogen water was then driven through by a stream of nitrogen gas. 5.25. Tube opened to the air and shaken to disturb the gas inside the tube. 5.40. No colour. 6.0. No colour. 6.30. Slight blue colour beginning to develop. © From this time onwards the colour gradually developed and deepened, till at 12.0 it had become dark blue. EOen Cite, Pade + Ordinary sand when placed in rongalit solution turns a green-blue colour. 170 Mr. A. N. Drury. The Validity of the The method of spreading the sand on to dry filter paper gave similar results. 1.30. Sand heated to redness, stream of hydrogen passed over. 2.0. Sand allowed to cool in hydrogen. 2.10. Sand again heated in hydrogen. 2.40. Sand allowed to cool in hydrogen. 2.50. Sand again heated in hydrogen. 3.20. Sand allowed to cool in hydrogen. 3.30. Rongalit white solution allowed to flow in and to remain in tube for one minute. This is followed by a stream of nitrogen water. There was no sign of colour during these two stages. 3.45. Sand washed out on to filter paper ; no colour. 4.0. Slight blue-green tinge through the sand. 5.15. Sand had become much deeper blue, the intensity of which continued to increase for some hours. This experiment agrees with the former in every particular, except in the time taken for the first appearance of the blue colour. This is a difference rather to be expected than otherwise, as in the washing out on to the filter paper the surface would be very much disturbed, and consequently the condensed hydrogen would be more rapidly displaced. That the oxygen-free surface clings tenaciously to the rongalit white condensed on to it appears from an experiment carried out in the same way as the preceding. The sand was freed from oxygen, and then the oxygen-free rongalit-white solution was passed over for one minute. After this the nitrogen water was allowed to flow through until the water as it emerged con- tained only very minute traces of methylene white. The sand inside the tube was then exposed to air on filter paper, and it developed a quite appreciable blue colour. There is thus no doubt that sand freed entirely from oxygen not only condenses methylene white on to its surface, but also holds it with a certain degree of pertinacity. Experiments on the Hffects of the Gases condensed on the Surface on the Condensa- tion of Methylene Blue. 1. Oxygen.—A small combustion tube was filled with two inches of sand, having at one end a plug of asbestos to prevent the sand from being moved by the solution as it passed through. Through this tube a solution of methylene blue was allowed to flow by gravity. The effluent was at first colourless, but with lapse of time as the Microchemical Test for the Oxygen Place in Tissues. 171 sand became saturated to the dye the colour increased to an intensity indis- tinguishable from that of the solution sent in. 2. Hydrogen.—The sand was alternately heated and cooled in an atmosphere of hydrogen gas completely to remove the oxygen. When it had cooled down ° in the hydrogen gas a solution of methylene blue of the same strength as was used above, but which had been made up in carefully boiled water, and had had nitrogen gas passed through it for six hours, was allowed to flow through by gravity as before. Samples of equal volumes were collected at the other end, and these showed an exactly similar graduation from colourless to the colour of the solution sent in. If there was any quantitative difference in the amount of condensation of methylene blue in the two cases, it could not be shown by such an experiment. This point will be dealt with later. It will be noticed here that sand whose surface is freed from oxygen and occupied by a film of condensed hydrogen will condense methylene blue from a solution so as completely to decolorise it. In the face of these results it is difficult to lay more importance on the results obtained by Unna than that he is merely picking out the basophile portions of the tissue with the rongalit white, the staining being modified, as might be expected, by the presence of the alkaline reducing substance, rongalit. The methylene white and the rongalit would both saturate the tissue with which they are brought in contact, but the rongalit is more easily dislodged than the methylene white, so that the basophile parts of the cell to which the latter clings would show a blue coloration owing to the oxidation of the methylene white to methylene blue. Quantitative Hapervments on the Effect of Oxygen upon the Amount of Methylene Blue condensed on to Sand. The preceding experiments show that the presence of oxygen at a surface is not necessary for the condensation of either the highly oxidisable leucobase, or of methylene blue. We now proceed to the further question, how far does the film of condensed oxygen favour condensation or the reverse? It will be seen that it actually lessens condensation of methylene blue. The following experiment was made. A solution of methylene blue was made and divided into two parts, one of which was freed from oxygen by passing nitrogen through it. Two combustion tubes were filled with similar lengths of sand, and were heated in a furnace. Over one was passed a stream of hydrogen to replace the oxygen, and over the other a stream of air was passed. The tubes were VOL. LXXXVIII.—B. O 172 Mr. A. N. Drury. Zhe Validity of the heated for three hours and one hour respectively and were allowed to ccol in their respective gases. The sand was then turned into the methylene blue, the hydrogen sand into the nitrogen methylene blue, the air sand into the air methylene blue. _ After they had remained in the methylene blue for equal periods of time, the solution was decanted off, and the sand carefully dried, the intensity of blueing of the sand showing a very appreciable difference even to the eye. The solutions of methylene blue were compared by means of the colorimeter with the original solution. The sand was heated strongly to vaporise the methylene blue on its surface, and was then weighed. Account of an Experiment.—Sand was heated in a hydrogen atmosphere for three hours, and cooled in an atmosphere of that gas; it was then put into nitrogen methylene blue solution for five minutes. The sand was decanted off and was compared with the original methylene blue solution by means of the colorimeter. The sand was dried, heated to volatilise the methylene blue, and weighed. A similar quantity of sand was likewise heated in a stream of air for three hours, cooled and put into ordinary methylene blue solution of the same strength for five minutes, decanted off and compared with the original solution by means of the colorimeter. The sand was dried, heated, and weighed. The same volume of solution was used in both cases :— Wt. Sol. Orig. sol. A.—Sand heated in hydrogen ......... 175 0:8 0°5 B— , ja hein al Sse Es 1:03 0°75 0:5 The result can be represented as columns of methylene blue solutions on the same base and containing the same amount of dye. Original lsolutionl se eee eee ee 5 Solution sA 2a peewee cece ener 75 NOlUEOn NB TET wee ae ve con eens 12:5 This experiment shows that the gas condensed on the surface plays an important part in the depth of staining which the surface undergoes. An explanation is, perhaps, to be found in the alteration of the electrical potential of the surface. The Effect of Certain Chemical Substances on the Amount of Methylene Blue Condensation on Sand. In the following experiments the sand was shaken up with the substance to be tested, and washed in water. The methylene blue solution was then Microchemical Test for the Oxygen Place in Tissues. 173 added and shaken for one minute, allowed to settle for one minute, after which it was decanted off; the solution was then compared with a sample of the original solution by means of the colorimeter. The same volume of the methylene blue solution was used in every case; the sand was finally dried, heated to volatilise the methylene blue condensed on the surface, and weighed. The following results were obtained. The ordinates represent the heights of columns of methylene blue solutions on the same base and containing the same amount of methylene blue, and consequently the relative desaturation of the original solution. A. The height of the column of the original methylene blue solution with which the various solutions were compared. B. Ordinary sand. C. Sand previously treated with gum. chloroform. formalin. : ‘rongalit. mercuric chloride. soap. octyl alcohol. caprylic ‘acid. 5 F p-cymol. aieal tenes I seals These results show, as might be expected from theory, that the previous treatment of the sand has a large influence on the amount of methylene blue condensed on the surface. 174 Mr, ALN. Drury. The Validity of the Summary. Experiments were made which prove that the results obtained by Unna with rongalit white do not justify his assumption that it is a specific stain for the oxygen place in tissues; consequently his theory of staining by oxidation and reduction is not proven. Further experiments were performed to find the effect of the gases condensed on to a surface upon the depth of staining. Altering a surface by preliminary treatment with various chemical substances also has a marked effect upon the subsequent condensation of methylene blue. I should lke to express my thanks to Mr. Hardy for help and criticism. The expenses of this research were defrayed by a grant from the Thruston Memorial Fund, Gonville and Caius College, Cambridge. [Note by W. Bb. Hardy.—The fundamental uncertainty in all microchemical tests, and perhaps especially in those for oxidation places, may be put as follows :—It is easy, as the writer found many years ago, to get diserimi- nating colour reactions in sections and unfixed cells with oxidisable bodies such as Wiirster’s tetra-substance, which is a singularly delicate test for what is called active oxygen—that is to say, for oxygen whose chemical potential is raised above that of atmospheric oxygen by, ¢.g., ionisation or the formation of peroxide. When oxidised it becomes a vivid purple and the purple reaction is given very definitely by, ¢eg., the basophile granules of leucocytes when the cells are exposed to a trace of the substance. But the tetra-substance is itself unfortunately a basic substance and would therefore be condensed from solution by the basophile granule in the ordinary process of staining. There appear to be three possibilities, and experiment seems unable to choose between them :— 1. That the basophile granule is in fact a region where active oxygen is produced, ¢.g., in the course of some local oxidation process. 2. That the tetra-substance is oxidised indiscriminately about the section during manipulation, in the course of which it is probably exposed to the combined influence of evaporation and light;* and that it is subsequently condensed on to the basophile granules by a simple staining process. 3. That in the process of condensation of an oxidisable body by surface energy its chemical potential is raised, so that oxidation, which would not otherwise occur in the presence of atmospheric oxygen, actually does occur. It must be noted that the condensation is due solely to surface forces, * D'Arcy and Hardy, ‘Journal of Physiology,’ vol. 17, p. 390 (1894). Microchemical Test for the Oxygen Place in Tissues. 175 and has nothing to do with the particular surface being one specially prone to oxidation or reduction. In this connection I am not forgetting that, when once completely condensed, the chemical potential of the oxidisable substance is no higher than (it is in fact identical with) what it is in the solution; but many instances show that during transition the molecules are under stresses which may find relief in exceptional chemical activity. For instance, the exceptional electrical and chemical properties of gases entering or leaving the surface of platinum ; the high chemical potential of condensing oxygen in Prof. Bone’s experiments on surface com- bustion, while by contrast a fully condensed film of oxygen on metallic iron does not oxidise the iron. Roberts’ experiments on the volatilisation of metals, perhaps, are also a case in point. During condensation local heat changes occur, the sign being determined by whether the solution of the substance condensed is endothermic or exothermic. Let heat be liberated when condensation occurs, the relation of solvent and solute being such that heat is absorbed during solution. The local liberation of heat will oppose condensation and the velocity of condensation becomes a function of the rate of dissipation of heat. In the well known case of an over-cooled fluid phase the dissipation of heat may be so slow as completely to arrest the change of phase.* If the substance which is being condensed under these conditions is chemically unstable, chemical change of the nature of oxidation, reduction, dissociation or association may be caused locally by the enormous molecular stresses. This third possibility considers a surface not as specially a place of oxidation because, for instance, oxidation of Wiirster’s tetra-substance occurs there, but as a surface which condenses basic substances. In this process oxidation of a basic body may occur, but an equally oxidisable acid substance would escape change. . Mr. Drury’s experiments clear the ground for further discussion to this extent—they prove conclusively that the condensation of a so-called test substance for “active” oxygen or a simple basic dye not only will take place on to a surface wholly devoid of oxygen, but is actually hindered by the existence thereon of a film of oxygen It must always be remembered that an oxidation place is also a reduction place, and it is to be called the one or the other according to the particular zero which is chosen. A convenient zero is the chemical potential of atmospheric oxygen, and a place would be an oxidation place if oxygen, whose chemical potential is = that of atmospheric oxygen, is condensed to the intra- molecular state. Such a region would then be a reduction place for chemical * H. A. Wilson, ‘Camb. Phil. Proe.,’ vol. 10, p. 25 (1898). 176 Prof. H. E. Armstrong and Mr. H. W. Gosney. compounds in which the oxygen potential is = that of atmospheric oxygen, and an oxidation place for substances in which it is less than that of atmospheric oxygen. In the absence of some agreement as to the zero point the discussion is likely to be as confused in the future as it has been in the past. | BIBLIOGRAPHY. Unna, P.G. “ Die Reduktionsorte und Sauerstofforte des tierischen Gewebes. Festschr. Waldeyer,” ‘ Arch. fiir Mikroskop. Anat.,’ vol. 78, p. 1 (1911). Unna, P.G. ‘ Biochemie der Haut,’ Jena, 1913. Spiers, F. M. “Contact Electricity,” ‘ Phil. Mag.,’ 5th ser., vol. 49, Part 1 (1900). McDonagh, J. E. R., and Wallis, R. L. M. “The Chemistry of the Leucocytozoon syphilidis and of the Host’s Protecting Cells,” ‘ Biochemical Journal,’ vol. 7 (1913). Kite, G. L. “Studies on the Physical Properties of Protoplasm.—I,” ‘Amer. Journ. Phys.,’ vol. 32 (1918). Studies on Enzyme Action. XXII.—Lipase (IV)—The Correlation of Synthetic and Hydrolytic Activity. By Henry E, ArMstTRONG, F.R.S., and H. W. Gosney, B.Sc. (Received and read April 30, 1914.) In the previous communication on this subject, in which the behaviour of Lipase towards ethereal salts generally was discussed, it has been argued that the enzyme is specially fitted to determine the hydrolysis of the insoluble, oily, glyceric salts of the higher fatty acids but is not suited to act in aqueous solutions: we expressed the opinion that interaction must be supposed to take place at and between surfaces separated only by a thin film of water at most—in other words, that water in excess is inimical to the occurrence of change. The results we advanced, in conjunction with those deduced from the study of other enzymes, notably urease, also led us ‘to conclude that it is impossible to apply the laws of mass action directly to the interpretation of the changes effected by Lipase. Previously we have directed our attention only to the hydrolytic activity of the enzyme: numerous observations are on record which prove that, whether of animal or vegetable origin, it can act reversibly but no com- parative study of the two processes has been made hitherto in the case of fats.* * (1) Kastle and Lowenhardt, ‘Amer. Chem. Journ.,’ vol. 24, p. 491. (2) Hanriot, ‘Compt. Rend.,’ vol. 132, p. 212 (1901). (8) Pottevin, zbzd., vol. 136, p. 1152 (1908) ; (4) ‘Bull. Soe. Chim.,’ III, vol. 35, p. 693 (1906). (5) Dietz, ‘Zeit. Physiol. Chem.,’ vol. 52, Studies on Enzyme Action. Wy 7 In view of present ignorance of the manner in which fats are formed in the organism and the desirability of determining the extent to which their synthesis can be effected, under various conditions, we have carried out a series of parallel experiments to ascertain the limits within which the two. opposing changes take place in presence of different proportions of the interacting substances and of water. In the first series of synthetic experiments, 4°84 grm. of the fatty acids from olive oil (the amount equivalent to 5 germ. of the oil) was used, in each case, together with the quantity (0°53 erm.) of anhydrous glycerol that would be required if the whole of the fatty acid were to be converted into triglyceride. The acid and glycerol were weighed into a 50-c.c. Jena glass flask together with 0°5 grm. of the enzyme preparation and 0°5 c.c. of toluene. The flasks were closed with rubber stoppers and kept slowly rotating in an incubator+ maintained at 30° C. during the times stated. Alcohol was then added and the residual acid titrated with a normal solution of caustic soda. Each determination was made in duplicate and control experiments were carried out simultaneously with a preparation that had been boiled with water to destroy the activity of the enzymes. The results are given in the following table :— Table I.—Synthesis of Fat from three Molecular Proportions of Acid to one of Glycerol. Time Meidityio® contra! Acidity of mixture containing | Percentage of acid enzyme combined hours 1 17-00 15°72 15°71 8-0 2 U-O7, ; 14-90 15 -03 12 °4 4 17 06 13 09 12 92 23 °9 8 17 04 11°37 11 -29 33 °8 17 16 93 10 “61 10 ‘61 37°9 30 16°91 10°53 10 -67 38 0 50 16 “65 10°48 10°33 39 °2 70 16°70 10 ‘47 10 *62 38 °3 p. 279 (1907). (6) Hamsik, zbd., vol. 59, p. 1 (1909). (7) Bradley, ‘Journ. Biol. Chem.,’ vol. 8, p. 251 (1910). (8) Taylor, ‘Univ. California Pub. Path.,’ vol. 1, p. 33 (1904) ; (9) ‘ Journ. Biol. Chem.,’ vol. 2, p. 102 (1906). (10) Fokin, ‘Chem. Rev. Fett-u.-Harz. Indust.,’ vol. 13, p. 238 (1906). (11) Welter, ‘ Zeit. angew. Chemie,’ vol. 24, p. 385 (1911). (12) Dunlap and Gilbert, ‘Amer. Chem. Soc. Journ.,’ vol. 33, p. 1787 (1911). (13) Kransz, ‘Zeit. angew. Chemie,’ vol. 24, p. 829 (1911). (14) Jalander, ‘Biochem. Zeit.,’ vol. 36, p- 485 (1911). (15) Bournot, zdzd., vol. 52, p. 172 (1913). + That described in the previous communication (‘ Roy. Soc. Proc.,’ B, vol. 86, p. 589), It may be noted that the figure there given is printed upside down. 178 Prof. H. E. Armstrong and Mr. H. W. Gosney. To discover whether a true equilibrium had been reached or whether the action had ceased owing to the destruction of the enzyme, 0°5 grm. ot enzyme was added to the system after the expiration of 24 hours and the mixture was titrated at the end of a second period of 24 hours. Experiments were also made in which 0°5 and 1 erm. of the enzyme were allowed to act during 48 hours before titrating the residual acid. Percentage of acid combined 0°5 grm. enzyme during 24 hours............ 37 *4 0°5 AS OU TE ceqeteceh 37 °7 10 AB io), fectieniegaate 33 6 0°5 24 34-9 Together with 0°5 grm. during a second 35-3 period of 24 hours The slightly lower activity observed in the experiments with 1 germ. of enzyme may have been due to the slight amount of water introduced with the preparation. Further evidence that a true equilibrium had been reached was obtained on hydrolysing olive oil by the theoretical minimum amount of water, ae. three molecular proportions to each molecular proportion of triglyceride or 5 germ. of oil and 0°53 erm. of water, quantities equivalent to those used in the synthetic experiments. As in the reverse case, the equilibrium was quickly reached and the acidity of the system was approximately the same as that observed in the experiments in the reverse direction. Table Il—Hydrolysis of Fat by three Molecular Proportions of Water. Time Percentage of acid liberated hours 30 °6 2 45 °5 4 56 ‘0 8 61°3 17 62-0 30 §2°9 50 62 6 68 62-0 The addition of even a small amount of water influences the equilibrium to a marked extent and also has a retarding effect—to an increasing extent, moreover, as the amount of water is increased. This is shown in the following table, in which are recorded the results obtained by the synthetic action of 0°5 grm. of enzyme on mixtures of 4:84 grm. of fatty acid from Studies on Enzyme Action, 179 olive oil and 0°53 erm. of glycerol, together with from 0°31 to 3:1 grm. of water, z.c. from 3 to 30 molecules per molecule of glycerol. Table I1I—Synthesis of Fat in Presence of various Molecular Proportions of . Water—showing Percentage of Acid combined.* Time No water 3 mols. 6 mols. 15 mols. 30 mols. | | hours il 8°0 7°83 | 4°7 2 12 °4 12:0 | 78} 3°6 | 4 23°9 16°3 10°3 2°4 8 33 °8 19 *4 12 °4 5°8 3°5 17 37 °9 22°6 13 °8 6:0 2°9 30 38 0 22 °2 5-9 3°4 50 39 °2 22°83 6°7 70 38 °3 21°5 15 °2 6°8 3°9 Water also has a marked retarding effect on the rate at which the hydrolysis is effected, as shown in the following table, in which is given the percentage of acid formed on hydrolysing 5 grm. of olive oil in presence of from 3 to 24 molecular proportions of water per molecular proportion of glyceride. In these experiments, the difference between duplicate observa- tions was somewhat greater than in the case of the synthetic experiments. Table 1V.—Hydrolysis of Fat in Presence of various Proportions of Water— showing Percentage of Acid liberated. Time 3 mols. 6 mols. 9 mols. 15 mols. 24 mols, hours | il 30 °6 | 27:2 19 *4 13-1 | 9°6 2 45 °5 | 40 ‘6 27 °4, 19°1 | 15 °9 4 56 ‘0 61°7 46-7 36 °2 23 +2 8 61°3 7/63 Al 56 °5 49-9 36 °7 17 62:0 UW 74°8 63:0 55 ‘2 | 30 62 °9 83 “6 WG) FL 66 ‘1 50 62 °6 85 °2 80-2 | 47:2 | 70 62-0 78 °7 84°8 82 °6 81:2 | It will be noticed that, in presence of 3, 6 and 9 molecular proportions of water, when equilibrium is reached, the acidity of the system is approxi- mately the same as that observed in the corresponding synthetic series: but that when more water was present the effect on the enzyme was such that the equilibrium was not reached during the experiment. * The results recorded are in all cases the means of duplicate experiments which differed by about 1 per cent. at most. 180 Prof. H. E. Armstrong and Mr. H. W. Gosney. The effect of glycerol on the synthetic action is similar to that of water on the hydrolytic change, the equilibrium point being so shifted that more acid is removed from the system. Excess of glycerol retards the rate of change in a very noticeable manner. The amounts of acid which entered into combination in a mixture of 482 orm. of fatty acid and 1:06 grm. of glycerol, z.e. three molecular proportions of acid to two of glycerol, are shown in the following table. Table V.—Synthesis of Fat in Presence of an excess of one Molecular Proportion of Glycerol. Ti Acidity of system c.c. normal Percentage of acid ime 3 : alkali combined hours 16°01 16 ‘03 6°2 2 15-10 15°18 11°3 A 13 26 13 °46 21°8 8 10-45 10 ‘84: 37 °7 i7/ 8 58 8°51 49-9 30 Oe 8°17 52 °7 50 7°56 7°53 55 °8 70 7°57 7°55 55 °7 When three or more molecular proportions of glycerol are present to every three molecular proportions of acid, the retarding effect is so pronounced that no equilibrium point is reached within a reasonably convenient time, the acidity of the system falling slowly after 70 hours. Thus— Glycerol, mol. props. | Acidity after 50 hours | Acidity after 70 hours 3 45 °3 44 °5 5 47-3 A2°3 10 44°8 0 The effect of glycerol on hydrolysis is similar, as is shown in Table VI, in which is recorded the amount of acid liberated from 5 grm. of olive oil by 0°5 grm. enzyme and 0°31 cc. water, in presence of 0°53 grm. of glycerol. Studies on Enzyme Action. 181 Table VI.—Hydrolysis of Fat in Presence of one Molecular Proportion of Glycerol. Time Acidity of system in c.c. Percentage of ; normal alkali. acid liberated. hours. 2 3°54 3 52 20°6 4 4°98 5°08 29 °4 8 6-05 6°13 35 6 17 7:06 7 42 42 °3 30 7°73 7°38 AA °3 50 7°73 45 °3 70 7°70 7°64 44-9 When more glycerol is present hydrolysis takes place to a reduced extent, and still proceeds slowly even after 70 hours. Glycerol, molecular Percentage of acid Percentage of acid proportions. after 50 hours. after 70 hours. 2 29°8 30°9 A 11°0 12 °6 The results of the experiments described are summarised in Graphs 1, 2 and 3, the synthetic observations in Graph 1, the hydrolytic in Graph 2, the parallel series of observations in the two opposite directions in Graph 3. The manner in which water affects both the rate of the change and the extent to which this takes place in one or the other direction is brought out in a very striking manner in these diagrams: it will be noticed especially how much less rapid is the approach, both from the hydrolytic and the synthetic side, to an equilibrium as the amount of water present is increased. Whilst the retardation of the hydrolytic change must be ascribed to a direct interference of the water, which presumably prevents the enzyme and the oil from coming into effective contact, the retardation of change in the opposite direction, especially the diminution of the extent to which synthesis takes place, must be ascribed rather to the withdrawal of glycerol from the system through its dissolution in the water: in this connexion, it is remarkable that synthesis is not entirely prevented even by the presence of thirty molecular proportions of water to one of glycerol, whilst in absence of an excess of water, an excess of glycerol beyond two molecular proportions has but little effect in increasing the proportion of fat synthesised. o 3MOLS. ACID: 1 GLYCER | Percentage of acid combined. - ACID : 1GLYCEBOL : 30H, sm (aS cio: rowvcenon: eon, | | 20 }MOL GLYCE gece ee 30 ACID : eee ROL: I50H, ae 40 50 60 70 a GLYCERIDE : [3 OH, “Tig0Re 60H Percentage of free acid. 0 HOURS !0 YSERIDE + | |MOL.CLYCEROL 3 0H2 183 Studies on Enzyme Action. VL, Ov 0€ 0¢ ol SunoH 0 alow" TOW : itt + TONSOATS “TOW 1) + JGINADATD TOW! () “HOE|+ GIOV IGNE+ 10YNX9DA19 “IOWII = “HOO |+ AGIYADAT Dies aie ee ea = oo cay ae tS ae Ol ee ate “HO los IOWE + 108 F30A19 TOWI= “HO 6 + 3GINFOA1N 10W | "ploe eaij Jo eseyusoI1eg 184 Prof. H. E. Armstrong and Mr. H. W. Gosney. In so far as our results can be brought into comparison with those of previous workers, they appear to be in harmony with their observations: but the activity of the enzyme we have had at our disposal, thanks to Tanaka’s important discovery, appears to have been in excess of that used by others. As it was obvious that if the limit reached in our synthetic experiments (about 40 per cent. when equivalents are used) were to be exceeded, the water produced in the interaction must be removed as it is formed, we endeavoured to secure this end by carrying out the synthesis im vacuo in a flask connected with drying apparatus. The results obtained have been uniformly unsatisfactory, inferior, in fact, to those obtained under ordinary conditions. Apparently, as pointed out by us previously, the intervention of a film of water is necessary at the interface of the system, where interaction takes place ; if this be removed, action comes to an end. In working with the Tanaka preparation, it is noticeable that the activity varies considerably, in a manner which is difficult to understand at first. On more than one occasion we have found that an enzyme which was quite active hydrolytically was inert when used as a synthetic agent with a mixture of acid and glycerol free from water: ultimately, this behaviour was traced to “ overdrying,” as on the addition of a very small amount of water the enzyme became active. Enzyme which has been used and then recovered, by washing it free from oily matter by means of light petroleum, is found, as a rule, to be still active but usually less active than it was originally: the variable behaviour of such preparations is not surprising, however, in view of the colloid nature of the material and the effect which alterations in the state of aggregation and of surface conditions must have. It is noteworthy that the hydrolytic activity—in presence of a relatively large excess of water—of the enzyme is much more reduced by such treatment than the synthetic activity. Nature of the Products of Change.—In order to ascertain whether the product of the synthetic action of Lipase is a nearly pure triglyceride like the natural fats and oils, the amount of glycerol uncombined in each experiment of the first series was determined, following the directions given by Lewkowitsch. After titration, the contents of each flask was washed into an evaporating basin, boiled to expel most of the alcohol and then just acidified by sulphuric acid. After heating the liquid to the boiling point, the solution of glycerol was filtered off and the fatty acids and enzyme on the filter were then well washed with hot water: the filtrate was purified by addition of a solution of basic lead acetate and the glycerol estimated in the clear filtrate by Hehner’s method (oxidation by an acid solution of potassium bichromate). Studies on Enzyme Action. 185 The method is probably one which is affected with a considerable error, so that the results have only qualitative significance. Table VII. Glycerol found im Percentage Mols. of acid | Time of acid combined Blank series* Series A Series B combined. per mol. glycerol hours grm. grm. grm. | i 0°51 0:08 —_— 8-1 1°6 2 0°52 0-11 0°11 12°4 ° 1°8 4 0°55 0°18 0°19 23 °9 2:05 8 0 °545 0:27 0°28 33°8 2:0 17 0 °495 0°28 0°28 37 °9 2°15 30 0 °5385 0-295 0°28 38 0 2:1 50 0 *495 0°28 0-305 39 2 2°1 70 0°515 0 °305 — 38 °3 2°0 | * The amount found should be about 0°53 grm. In the same manner, estimations were made of the amount of glycerol liberated on hydrolysing olive oil by the enzyme in presence of 24 molecular proportions of water. It was found that, at first, the acids liberated were slightly in excess of the glycerol, an indication that a small quantity of a lower glyceride was formed: but as the action continued, the whole molecule was hydrolysed. Table VIII. Glycerol found Percentage of acid combined Mols: oftacid Time ; liberated Series A Series B Series A Somers || HEPC Ghycaml! hours grm. il 0 055 0-045 SES 8°5 2°8 2 0°075 0 065 15 °9 13 °8 3°4 8 0-165 | 0-165 36 °2 34°9 34 17 0°275 | 0°25 56-2 53 ‘0 3°3 30 0 °325 | 0°33 65 °2 64.°4 3°] 50 0°41 0 °395 17°53 76°3 3°0 70 0-43 0 425 80°8 80 °9 3°0 The amount of glycerol liberated, however, is less in proportion to the acid when the hydrolysis is brought about by a small proportion of water, showing that under these conditions mono- and di-glycerides are produced to a greater extent. 186 Prof. H. E. Armstrong and Mr. H. W. Gosney. Thus, on hydrolysing 5 grm. of oil by 0°31 ec. of water :— Time Percentage of Percentage of Mols. acid per mol. acid liberated glycerol liberated glycerol hours 1 31:0 245 3°8 2 45 6 36 °8 3°7 8 61°8 46 °3 4-0 Conversely, when the synthesis is effected in presence of water, less glycerol is combined than when no water is added; that is to say, the glycerides formed are more saturated. Thus, in presence of 0°62 ce., 2.é., 6 molecules of water :— Time Percentage of Percentage of Mols. acid per mol. acid combined | glycerol combined glycerol | hours 7°7 12-0 1g) 8 11°9 15-0 24 70 15-2 18 °5 2°5 An excess of glycerol not only alters the equilibrium so that a greater proportion of acid is combined but also influences the nature of the product, which then contains a smaller proportion of acid: thus the composition of the product of the interaction of two molecular proportions of glycerol and three molecular proportions of acid was found to be as follows :-— Table IX. Ti Percentage of Percentage of Mols. acid combined ame acid combined glycerol combined per mol. glycerol hours il 6°3 71 1°3 2 111 9°5 1°8 4 21-2 | 17-0 19 8 38 *2 30 °2 1°9 if 49 ‘7 41°0 1°8 50 55-7 Ad 4 1°8 From these results, it 1s not improbable that the main product is a diglyceride: in other words, that, as is to be expected, the two primary hydroxyl groups of glycerol are first affected. Some of the product of the interaction of the acids from olive oil with an excess of glycerol was isolated by evaporating off the alcohol after neutralising the unchanged acid and extracting the soap solution with ether. About Studies on Enzyme Action. 187 18 erm. of a pale yellow oil was thus obtained, which became turbid on standing, slowly clearing again on heating to 30° C. The saponification value of this oil was 183°5, that of the olive oil used being 191°7; on acetylation the saponification number was increased to 248°7, the “acetyl - value” being 786. These data favour the assumption that the oil contained a high proportion of diolein. It is obvious, however, that no final conclusion is possible until experiments have been made with definite acids and the products have been isolated and characterised. In view of our results, we venture to call attention to several directions in which the fats now deserve renewed attention. Our knowledge of the manner in which they are absorbed and utilised under vital conditions is at present very vague in character and much of the evidence on which reliance is placed appears to be open to question. It is generally believed thaty when ingested, fat is rapidly hydrolysed, under the influence of the pancreatic secretion and that derived from certain tracts of the intestine, this change being regarded as a necessary preliminary to its passage through the walls of the villi prior to entry into the circulatory system. Lipase appears to be widely distributed throughout the organism. Apparently, whenever fat is to be transferred across cell membranes, it is hydrolysed: assisted by the emulsifying influence of the biliary fluid, the fatty acid that is liberated during digestion of fatty food can penetrate tissues that are impermeable to the fat but it is held that on entry into the villi the fatty acids are rapidly re-associated with glycerol and pass into the lacteals as fat. In fact, all fat that is stored is supposed to be fat that has been reconstituted from fatty acids. In the normal heart and other tissues, however, the fatty acids are not present as glycerides but apparently are combined in such a way that their histological behaviour is different from that of fats—the discriminative staining agents being without effect in such cases. If the vital mechanism be such that only fatty acids can pass through, it is clear that in presence of lipase fats would undergo complete hydrolysis readily, under natural conditions, if the acids were removed as they were liberated, as reversal would be prevented. Our observations appear to show that hydrolysis would be most rapid in presence of a minimum amount of water; they therefore favour the conclusion that conditions which would tend to reduce the concentration of the cell fluid would promote the conservation of fat—a conclusion which is perhaps applicable in explanation of. the obesity which apparently is a frequent consequence of the indulgence in large quantities of weak alcoholic fluids such as Lager beer. VOL.. LXXXVIII.—B. P 188 Prof. H. E. Armstrong and Mr. H. W. Gosney. But in view of our observation that. under 40 per cent. of fatty acid is convertible into fat, even when no water is present, it is difficult to under- stand how the fatty acids are completely reconverted unless there be some mechanism whereby the fat is separated from the fatty acid as it is formed— or some means by which the acid is held in abeyance until it is required. May it not be that the clue is afforded by the observations above referred to with reference to the presence of fat in the tissues in a cryptic form? Lipase apparently is a “carboxylase” which has the power of determining the hydrolysis of the ethereal salts of all the very weak carboxylic acids and, within limits, is more effective the less soluble the acid and the alcohol from which the salt is derived; presumably, the argument applies equally to the synthetic activity of the enzyme. It is therefore probable that, under the influence of lipase, fatty acid may become associated with hydroxylic centres in the protoplasmic complex and that such withdrawal may be the cause of its cryptic existence in muscular tissue. The effect on health of an absence of fat from the diet, to which Arctic travellers have called attention, is noteworthy from this point of view. Stefansson, in his recent book ‘My Life with the Eskimo, * states that the symptoms that result from a diet of lean meat are practically those of starvation; during the winter period, even when gorged with caribou meat free from fat, he and his party felt continually hungry; the dogs, though they got more meat than dogs usually get, were nothing but skin and bones. Previously, when they had lived practically on oil alone, taking a teacupful of oil a day, there were no symptoms of hunger ; they grew each day sleepier and more slovenly, he says, but at the end of their meal of long-haired caribou skin (to give bulk) and oil felt satisfied and at ease. On the assumption that fat is not always laid down as such but frequently reconstructed in situ, the presence of glycerol in the necessary amount at the seats of synthesis has to be accounted for. Owing to the solubility of this substance, it cannot well be supposed that, when fat is hydrolysed, the fatty acid and glycerol always remain together in the required proportions: it is more probable that the glycerol becomes separated from the acid to a greater or less extent and that the deficit is derived from carbohydrate : it is on this account, at least in part, perhaps, that it is desirable that a certain minimum ratio should be preserved between fat and carbohydrate in our food. We are indebted to the Hull Oil Manufacturing Company, Ltd., for having placed at our disposal Indian castor seed of recent growth for the purpose of this inquiry. * Macmillan and Co., London, 1913, pp. 140-141. Studies on Enzyme Action. 189 [Note added June 18.—In a communication which came to our notice only when the work we have described was completed Bournot (15) has called attention to the activity of the lipase present in the seeds of Chelidoniuwm mayus, the common Celandine, a papaveraceous plant. Having been able, through | the courtesy of Messrs. Parke, Davis and Co., to obtain a sample of the seed, we have contrasted its activity with that of our Ricinus lipase and have confirmed Bournot’s statement that it is not necessary to treat the seed with acid to render it active. According to Bournot, Cheladonium lipase differs from icinus lipase in being most active in a neutral medium, even N/ 50 acid having an inhibitory effect. But as is shown in Part II, when once liberated from its zymogen Ricinus lipase is also sensitive to acid: in our experience, it has maximum activity when the acidity does not exceed that of oleic acid. The enzymes from the two sources both hydrolyse and synthesise glyceric oleate with about the same ease and give rise to mixtures similar in composi- tion at the equilibrium poimt. But weight for weight, the Tanaka Aicinus preparation is less active than Chelidoniwm seed (free from oil) in effecting the synthesis of isoprimary butylic oleate. Thus in an experiment in which 41 per cent. of the acid was combined by the agency of the Aicius enzyme, about 80 per cent was etherified by Chelidoniwm seed. Apparently, the alcohol has a specially marked effect on the Ricinus preparation, as olive oil is hydrolysed only to a small extent in presence of a molecular proportion of isobutylic alcohol to one of the oleate. Sunilarly, on hydrolysing isobutylic oleate, whereas, in presence of a single molecular proportion of water, 9°3 per cent. of change was effected in 17 hours eye by the Chelidoniwm enzyme, the ficinus preparation caused only 2°4 per cent. of change. The difference was less marked on using 10 times as much water, as 16°7 per cent. was hydrolysed by the one and 13:0 per cent. by the . other “enzyme ”: in this case, the effect of the aleohol was reduced apparently by the presence of the excess of water. In our opinion, such differences as are observed are to be regarded, provisionally at all events, as consequences of differences in the “condition ” of the enzyme in the different seeds. At present, as it is impossible to arrave at any estimate of the “ concentration ” of an enzyme or to allow for differences in its distribution, we cannot well make any valid comparison of the enzymes of like function derived from different sources. ] VOL. LXXXVIII.—B. 190 Morphology of Various Strains of the Trypanosome causing Disease in Man in Nyasaland: The Human Strain (continued).-—VI to X. By Surgeon-General Sir Davip Bruce, C.B., F.R.S., A.M.S.; Major A. E. HaAmerTON, D.S.O., and Captain D. P. Warson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received May 5,—Read June 25, 1914.) INTRODUCTION. In a former paper* five strains of this species of trypanosome, obtained from man, were described. In this it is intended to describe another set of five. A small quantity of blood was taken from the sick native, brought up- to the laboratory, and inoculated into a monkey or white rat. Strains VII, VIII, and X were directly inoculated into the rat from which the drawings and measurements were made ; in Strains VI and IX a single monkey intervened. These five strains have therefore been obtained under fairly similar circumstances, and should prove useful for purposes of comparison. This is put in tabular form in Table I, and the history of the previous five strains is also given, as this was omitted in the former paper. It is possible that the type of a strain may be changed by passage through different animals before it reaches the rat. Table I—Showing the Passages through Animals between Man and the Rat whose Trypanosomes are Drawn and Measured. Strain. Man. | Monkey. | Dog. Rat. I, Mkanyanga ....:. JON TW Seo cecnaticoage TLl, Chituluka ......... f IV, Chipochola ...... WW, CUO coecococene VI, Manakumpara ... WADL, WOOrDON 5 aacaccoene VIII, Mekka ............ IX, Mkanthama ...... | X, Dongolosi ......... | | | | | | terorwr | BREE Re eee | Pre es tay tow bo to = cow woo | i From this table it will be seen that little time was lost in inoculating the rat, the trypanosomes in whose blood were drawn and measured. * ‘Roy. Soc. Proc.,’ B, vol. 86, p. 285 (1918). = sa Trypanosome causing Disease in Man in Nyasaland. 191 On comparing the curves from the 10 strains it cannot be said that the passage from dog to rat, or from monkey to rat, or direct from man to rat, has had any marked influence on the character of the curve. But in these cases only a single monkey or dog intervened. In the case of Strain I, Mkanyanga, the trypanosomes were taken at random from several species of animals;* 600 trypanosomes from four different rats were measured. The passages are as follows:—(1) Man, monkey, Rat 38; (2) man, dog, rat, rat, Rat 37; (3) man, monkey, guinea- pig, monkey, Rat 256: (4) man, monkey, guinea-pig, monkey, Rat 235. VI. MorpHoLocy or Strain VI, MaNAKUMPARA. The following Table gives the average length of this trypanosome as found in the white rat, 500 trypanosomes in all, and also the longest and shortest :— Table I11—Measurements of the Length of the Trypanosome of Strain VI, Manakumpara. | : ; | In microns. | Expt. ; Method of | Method of ;—— Date. I Awimneil, |) oA Sait | No. fixing. staining. Average | Maximum |} Minimum 3 | length. length. length. | 1913. June 30...) 2239 | Rat. ...... Osmic acid | Giemsa 21-4 27-0 18 0 erm 2239 SE i ie | ce a 21°6 290 18 ‘0 <9 80) onell 22BG) ie bean | 3 s 22-1 27-0 20-0 Sully, Mesa eC al irae eae | f - 19-0 230 70 ny onal’ PED) eee | S a 18 ‘6 23 0 15 ‘0 ee] 92239 ke eee | A i 18 6 230 16 0 PPO eEDOSO, lk ee , i 19-8 22-0 17-0 22239 sie tea ef a 20°6 25-0 17-0 Nein suey Sheela 2239 Bo ae i‘ ie 19-3 21-0 17-0 Wee Tone be eeeoe ree ac vk: ecw hee es 21°1 27-0 17-0 |} 5, &...| 2289 ped aN - nt 21°8 260 17:0 Pe aise 8 -otele2230 Perce? | ¥, 226 30:0 17-0 5p Gh nell P2BO) a eres as | st . | 28:0 26-0 17-0 nea ees D939 pe ae | i | 2B 29-0 18-0 Pea 2239 pera | Fs a | 230) 28 -0 19-0 Rani owt 12039 at cei id i 22-4, 28 0 170 Waite Die gelpe OOO. Wramey oi aoe i i 21°8 30-0 17-0 Lo AB sed DRE eae ad . i. 23-8 29-0 18-0 my © soci) 2PBY) Aa: | i a 23 °5 28-0 21-0 eee 7h esl, 22280 A eae: | . °F 22-9 30-0 170 Prom eet 7h elp 2039 aa a | i ‘ 24.°5 320 19:0 (enor Sty <2080 ae REET e i 23-0 28 -0 19-0 Loe By cea, PRB) cee eee | x 22:6 290 170 fe oe Seen s0089 Ft asesaee i e 21-2 27-0 15:0 | eC) ait 2 0) el Pare ee % is 20-6 260 15-0 | | 21-7 32-0 15:0 * “Roy. Soc. Proc.,’ B, vol. 85, p. 427 (1912). Q 2 192 Sir D. Bruce and others. Trypanosome Table IL].—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain VI, Manakumpara, from Rat 2239. In microns. 13 16.) 17.| 18. 18, 20. | 21 | 22, | 23.) 24. 20. 26.| 27.| 28.| 29.) 30.) 81.) 32. | | l l otal simeepercee | 3 5 | 22 | 40 | 46 | 74 82 62 | 40 | 37 | 28 | 16 | 23 18] 5 3 |—]1 Percentages Dieliae aaa ee 9°2)14°8/16°4/12°4)8-0|7-4)4°6|3°2/4:6 3°6/1°0;0'6| — |0-2 | Cuarr 1.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain VI, Manakumpara, taken on nine consecutive days from Rat 2239. Microns i3 [va [is [ie] 7 [ie [is [zo]2i[z2[2a]24]25[z6]27|28 2s 30] 31 ]32]33]>4]>5]36 |37|38 it it i | i 18 1 [ 5 Lt 17 — + { | 6 4 (aa 14 iB} = ”n ov 2 oar (eos oO S10 C9 v u 6 c 7 v a6 5 4 5 a Lt LJ Ba) Table 1V.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain VI, Manakumpara. | | Experiment | : Percentage among short Date. “ie. | Seauamal and cay feinie. | | 1913. | June 30 ......... | 2239) |) IRR sapeaotss . 4, Pike Il) secenp von 2239 Wh Poniisaire 16 ANGE is Han BOSON ail a aie 16 A ae tae DOSG, Oly 1d) aan ee 14 Pine aerate 2239 Rae oda 9 SPMD th TE 2239 Bae 21 Piensa ilgabaor aot 2239 Sy aRie ae 8 FG os ere ae 2239 ihe tienes 13 Riko ioe a secrels 2239 ie Game 17 se uMlOLsan ee 2239 SA ee | 25 Average ......... 14 °3 causing Disease in Man in Nyasaland. 193 Breadth.—Vhe following Table gives the breadth of the trypanosome of Strain VI, Manakumpara :— Table V.icMeasurements of the Breadth of the Trypanosome of Strain VI, Manakumpara. | | | Nupibewioe In microns. | Date | cde as | Animal tr aa a Pcl No. ; pI eae ae | Average | Maximum | Minimum | | measurec- | breadth. | breadth. | breadth. | 1913 2239 | Rat Lee | 500 | 2-76 | 450 | 125 | In regard to shape, contents of cell, size and position of nucleus and micro- nucleus, disposal of undulating membrane and flagellum, no difference can be made out between the Strains VI to X and the first five strains. VII. MorpHouocy or Strain VII, YoRAMU. The following Table gives the average length of this trypanosome as found in the white rat, 500 trypanosomes in all, and also the longest and shortest :— Table VI—Measurements of the Length of the Trypanosome of Strain VII, Yoramu. | | In microns. Expt. | : Method of | Method of |— ae oat DENS. a ssaene fixing. staining. | Average | Maximum| Minimum length. | length. length. | 1913. | June 80 ...| 2236 | Rat ...... Osmic acid Giemsa - 25°8 28:0 23:0 »° 80 SOSGhs a emer. - B 26°7 33-0 21-0 = 80 cod) PARE SVR Saree y i 26-9 an @ || 2aw Fulya mie 2286 allie cu se - 27-0 32-0 | 18-0 ibe ORR MOP RT al Maa is py BO ip BBO a Baw Ay i Nae PD i ees * ark alk Obs Gi) || DLW ope rosso rh on e Ns 20°3 240 18-0 SLO. le Pee eo eens ne i 21°5 26 -0 18-0 eo ie DONG rns a e it 21°9 29-0 190 5 soall. 2288 ee ceee 3 33 20°5 | 28:0 16-0 SoC) IMR S iV ea ‘ i 20'2 29-0 16-0 SE See MDDS Giese ein if . 19°4 26 -0 16-0 Pe ie BOD SGe Ub eect % id 19-4 23 -0 17-0 | oy) 4h 50 2236 Of | obob00 2 ) 19 ‘9 24-0 | 18°0 | » &...| 2236 a ganas y ; WG PRO) alls} XG) Ve SP aa ie al OP 1s) abd a fs : 22-0 2770 | 18-0 5) ccel! © 2B Sone s i 21°7 30-0 17:0 ee Sete 11.2256 Me een. s ‘: 21:9 30-0 180 SG) Ml OOSG ane hy is i 21-7 28-0 17-0 | 5 © cod) 2PRG Ao Roa a 3 il Bs 28-0 17-0 SOG MOOS GL Name Gas i i 223 28 -0 18-0 a o ten|eZoo a eoneoen a 3 22:5 ABO) |) TS} 0) wy ousll, PPBB pat al aM fyi a om 28:0 | 18-0 mth doc 2238 ey Sente G Ms ne 220 29-0 18-0 Pits pe lee doa Gea means A i 22-4 34-0 | 18-0 | 225 34-0 | 16-0 194 Sir D. Bruce and others. Trypanosome Table VII.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain VII, Yoramu, from Rat 2236. In microns. | | | | | 16. 17 | 18. 19 | 20. | 21.) 22.) 28.] 24.| 25.) 26.| 27 | 28, 29.) 30.} 31 | 22 33. | 34. ell | | | | | Totals ............ 3 | 9 | 40 | 84 | 67 | 38 | 47 | 27 | 28 | 31 | 33 | 37 | 21 | 13 |13) 5 | 2 | 1 Hf Percentages ...|0°6|1°8|8-0 eae 7°6|9°4|5°-4|5-6|6-2|6:6/7-4 a2 he 2°6)1°0/0°4,0°2)0-2 | | | | Cuart 2.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain VII, Yoramu, taken on nine consecu- tive days from Rat 2236. Miligcinvovimis: 13] 14] 15] 16) 1718119 [20] 21 |22]23|24]e5|26)27|26]29|30| 31 [32/33 ea hazel eel ES EEE aaah lEAe == ; — : \ 1g) { r - ; 17 | | a] f + 16 ea i + [ | | 15 4 14 - i 7 | aul t ; 1 ov 2 iL = Di t | ele | 2 ol - _— — --- |Ecag' | 4. ) | | us ] T L, i | IL uv a cL Lo IL. 5 4 4__} i | 3}; , L 2]; 1 io T = Pyaisis Eel | Table VIII.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain VII, Yoramu. | | DEN, | peer ley Srey Percentage among short | 0. and stumpy forms. - 1913. | June 30 ......... 2236 |, Rath esse | 0) Oye eases 2236 [ks gtsye gaeteene 1 pte oe ieah aadae 2236 Pe unt saee | 15 APMIS? BORr eos | 2236 7h i Baa | 0 Peele cn eaters 2236 9 ego sonded 2 op Pei Rarneanee: 2236 PM reraRacan: 6) Sin sos 2236 jf Gauasblens 10 By) “bob on000 2236 Ao eaasoaeHs 36 Bite ete aR eM ee 2236 Benno arte 34 LOO see ete | 2236 Se Soo 36 Average ......... 13 °4 - causing Disease in Man in Nyasaland. 195 Breadth.—The following Table gives the breadth of the trypanosome of Strain VII, Yoramu :— Table [X.—Measurements of the Breadth of the Trypanosome of Strain VII, Yoramu. wea In microns. | Experiment | Number of | Date | a | Animal. trypanosomes | | | °5 | measured. | Average | Maximum) Minimum | | breadth. | breadth. | breadth. | | 1913 | 2236 | Rati oo... 500 | 2°51 | 4°50 | 1°25 VIII. MorpHotocy or Strain VIII, Mexxka. The following Table gives the average leugth of this trypanosome as found in the white rat, 500 trypanosomes in all, and also the longest and shortest :— Table X.—Measurements of the Length of the Trypanosome of Strain VIII, Mekka. | In microns. eke | Ear Wee ATE Method of Method of | | ese Bene staining: | Average) Maximum] Minimum | | length. | length. length. | | | | | | | 1913. Aug. 27...) 2800 | Rat ............ Osmie acid Giemsa | 21-2 26-0 190+ | Par Oe NORCO eke i Re NU” aie 29-0 20:0 (| BA oT POSOO \inhh Lape i, i 21-1 28 -0 18:0 | ROS aODO0N | 5 a ce Btroos - 22-2 27-0 19-0 OS MMOS OO het ew oe | a Ss 22-4 26-0 18:0 | ROG AEOSO0) | is fiat RVC H is 23-4 29-0 20-0 AGO OW R2SO0 nly a ume se 3 21 °4 26 0 20:0 9. Bsool| ZEODAN py" Vessdoages 3 5 23 -0 28 :0 21-0 ay OL IMZ300n ius tonalite: ‘5 ty 21-0 24-0 18:0 » 30...| 2800 Pe qanopcewapesc of 9 20 °4 23 °0 18-0 Se SOV NMOS OO A MONI. iB 22-1 28-0 18-0 » 80..,| 2800 oh). pndedaoenb bac ; 5 20-7 25 °0 18 °O Mes ESOO yy. ltr nte i i 20°7 24-0 18-0 A SIM EOS00 on ie 21 °4 29-0 18-0 pp) lec) ZOO oy" sesuoaanonne 5 r 21-1 27-0 SEO Seyi, Wsccl| HOO |) 5)” cehoantacces “6 is 24:8 32-0 30:0 | 5p MA eBOO) Wiss ocean hones 5 5 23 °8 33 °0 20:0 SE) TEAS OO) ME eee | ‘ a 24-1 23-0 190 i Peel CRO MN oy Ns 23-9 300 19-0 OME S00) || Mk angie :| 6 is 23 6 29-0 16-0 Ae a DO SOO) Mere oak é s 23-2 30-0 20-0 MEE SPA MOSO0 dA oe. : | 24-0 32-0 20-0 Sh pt | SOROTO NMP ae ae i if 23-1 29-0 20-0 90 Bseq] ZBOD 45° coonne Sane . "9 235 30 °0 21-0 9p A GAMASOORIT Ren cade vsccse ne | 59 | rH 21°3 27°0 18 ‘0 | | | 22-4 33 0 160 I 196 Sir D. Bruce and others. Trypanosome Table XI.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain VIII, Mekka, from Rat 2300. Tn microns. | 16, W7 18. 19. | 20 | 21. | 22. 23. | 24. | 25. 26.| 27 25, 29.| 30. 31. 32.| 33. | | | | | | | | | Motalsy ese. 1|— | 19 | 30; 69 89 | 80 68 | 49 | 34 | 20 | 8 OM SSR ie 2 2 i Percentages |0°2| — a8 6°0|13°8 U7, 8) BRO 13 °6 el 4°0|1°6/2°0|2°6 10/0 0°4/0°2 | ! | Cuarr 3.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain VIII, Mekka, taken on nine consecu- tive days from Rat 2300. IM Gir © in So ] a] Seniienil Ailas T T L US fle) 5 te [17 [ie 9 20) 21 |22)23 24|25|26!27 28/29 30) 31 32/33 34 | 35 | 36 |37|38 T 7 II | | | 15 | | Percentages Omani BE dt ro La - NO PU aN © oO ft Table XII.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain VIII, Mekka. Mates sg eee Percentage among short 0. and stumpy forms. | 1913. | Wend ullivae2i/aeen sees 2300 Rateerenes 5 eS Nae ae de | 2300 PEN Paani | 8 PP WeD tas eens 2300 fie wage | 22 JN 8 soococoe 2300 A Ua BS Ha soHs | 9 MNS Oe cee 2300 peo as 8 | 13 Pr ot ipeceen en: 2300 PAL aS BCAN 38 Ae AWE Scie eee | 2300 ji UR eae | 4) | Lame! Real 2300 A he ead 41 | mall Avital 82800 i 0 PE Baa 37 | reel any cep | 2800 el Oe | 29 | Average ......... 24 °2 | causing Disease in Man in Nyasaland. 197 Lreadih—The following Table gives the breadth of the trypanosome of Strain VIII, Mekka :-— Table XIII.—Measurements of the Breadth of the Trypanosome of Strain VITI, Mekka. | | | | In microns. | Experiment | < | Seer | | Date. P No Animal, | trypanosomes a | | measured. Average Maximum Minimum | breadth. | breadth. | breadth. ] | | | 1913 2300 | Rat ...... | 500 | 2-68 | 5-00 | 12500)! IX. MorpPHoLoey oF Strain 1X, MKANTHAMA. The following Table gives the average length of this trypanosome as found in the white rat, 500 trypanosomes in all, and also the longest and shortest :— Table X1V.—Measurements of the Length of the Trypanosome of Strain IX, Mkanthama. In microns. Dystie. ee? UM arvnale sates of | Method of l ie aoe eal SE Average | Maximum | Minimum length. | length. length. 1913. Aug. 18 ...) 2386 | Rat ............ Osmic acid Giemsa 23 °2 31:0 170 ree, al Shas a TSY3)2 fe le aR i is 26-7 33-0 18-0 as YBis16) Il ise Bese tane 60 251 320 180 «| UY ZISGC ile ys os ssacee eeu 50 5 19:0 240 17-0 ROE eS p2SRG al tee #3 vs 18 °5 26-0 16-0 SOTO ESS SG rlin ute if i 18-2 23-0 16-0 , 20 ZSSOMIM tre, vce ceee cae: a 19-4 240 17-0 SenZ OPA 28860 ln ‘ f 19°8 30-0 16-0 eeO ZING)" || ogy Visaquoesacooe 1) 19) 8) 30-0 16:0 a raile P-fet9) II! 55) “Geo onnandese bp % 17-7 24:0 15 ‘0 SOT OSS Hlinee acer eEnE iM i 17°8 22-0 14:0 fy CAL Seal 210 we Boouuacasend 50 3 175 20-0 16 0 io Pca) PRTO Seeman i % 19 *4 31-0 150 eZ, ZOSOI he ere alosenthas. iH % 20°5 280 150 oy DB call CREM = & s ss 21-8 29-0 15-0 5 2B ll PRBS a ‘ 23-5 29-0 17-0 Sy 28} Spal] ZBI) Mogomapeindencn m9 5 24,2 320 16:0 PUGo a es I 2ORG ER WEEL or. es i ‘ 23-4 31-0 16-0 sl OE NEN SORRY AR ose ants ‘ ss 24-1 33-0. | 18-0 fo, Le all PRs eae s " 23 *4 31-0 16-0 ERO Ae nD SOGHIIE eee en x is 21-6 30-0 17-0 prod [SOE AMEE Urs, lee Bane a i 22-0 29 -0 17-0 SRDS IL ORC GNINGS, iain rae” c 21°8 31°0 170 SAM CAPO REO I gt eta ee i = 21-2 30-0 15 ‘0 » 26 ZEB Mlb en vame one ~ » 196 26-0 17-0 | 21:2 33:0 | 14:0 | | 198 Sir D. Bruce and others. Z’rypanosome Table XV.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain 1X, Mkanthama. In microns. 14.| 15. 16. 17. | 18. | 19. | 20.| 21.) 22.| 23.) 24.) 25.| 26.| 27.| 28.| 29.] 30.| 31. 32.) 33m \ | ; { | | | | | Totals ...| 1 | 9 | 22) 64 | 79 | 70 | 40 | 23 | 21 | 29 28 | 24 | 28 16] 13 | 15 12" 9) | 952m Percent- | 0°2|1°8 4/28 15°8| 14:0 oe 4°2/5°8/4°6/4°8/ 4-6/3 °2)2-6/3-0/2-4/1°8|1 0/04) ages | | | | i Cuart 4.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain IX, Mkanthama, taken on nine consecutive days from Rat 2386. Mb ie ©) in Ss # ia fia]isliel iz 6 [9 [20] 2zj23 [24 25|26|27|28|29/30|31]32[33]>¢ 35 | 16 137/58 eral aT EEL 16 a = ir 17 16|}— i ea 15] — = | 14|/— a(t Lol n | Cy) A Di } C0) | a '0/F te © oll v u 8 = v 7 a6 5 4 3 ll 2 |_| . i) LI 1 Table XVI.—Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain IX, Mkanthama. | Experiment ; Percentage among short ate. | “No. Zane and sane foeen | | 1913. Wilkie JUS aocegoss 2386 |i MERGER f teste 13 Hei in ee 2386 [Gees tae enemies | 32 pH eOe, xetenekes | 2386 Sanne ee 18 AION 5 Bik, 2386 bee 11 PA nee ae del 2386 Rey aenacene | 19 Pe ee 2386 Sra Sacre 27 Yk ry decrees || 2386 ere LO | 34 | SDD) cations 2386 Pe Peed ts 33 [$9 eI a come 2386 aor Nias sea 45 PARMAR sect: 2386 | Be ee ns en 54 | | Average ......... 28 °6 causing Disease in Man in Nyasalai Breadth.—The following Table gives the breadth of t Strain IX, Mkanthama. Table X VII.—Measurements of the Breadth of the T Strain IX, Mkanthama. id. 199 he trypanosome of rypanosome of In microns. | eerie: | Number of | Date. aS | Animal. trypanosomes | 4NO. : | Dao measured. Average Maximum) Minimum | | breadth. | breadth | breadth. 1913 2386 Rab ©... 500 2°56 5 00 1°25 X. MorPHOLOGY OF STRAIN X, DONGOLOSI. The following Table gives the average length of this trypanosome, as found in the white rat, 500 trypanosomes in all, and also the longest and shortest :— Table XVIII.—Measurements of the Length of the Trypanosome of Strain X, Dongolosi. In microns. Date. a etriri ee, of Meu tot | oe rah SyalnIne Average Maximum) Minimum | | length. | length. | length. | | 1913 Nov. 27 2437 | Rat ...... Osmic acid Giemsa 24.°8 30-0 20-0 Prog i CY oe et ‘ 7 24-6 29-0 21-0 = BY Te (ul byes ie is ‘ 25-0 29-0 21-0 , 28 OY Gy fa A ane A ; 19-9 22-0 17-0 hs pe a aa Re 9 F, 19°9 23-0 18-0 gs cai Mlge sal ae a eat . > 19 6 22-0 18-0 Derrek OtS7N| ce * s 23-2 28 -0 17-0 | oe nie tae 2437 Batt. ¥ 22°0 29-0 ifs}0) 7 | ja Oy a | ee Bi 23 2 28-0 18:0 | on CEE 2437 Of hasseee op “5 24-0 30-0 20-0 Hed) SY Ssh eee 5 4 37 29-0 20:0 | 2) 2S aay eal alien gai ie e408 lee 30.0 19-0 SN er? a Ne re z i ieuleeoSco) 1 127-0 19-0 Airmray A Sasa ee & ¢ 232 | 300 19-0 » 4 ...| 2437 ve aheee ks 5 Pp MN ka g0) 21-0 Eee ere OAT aia dear 4s 5 24-3 28-0 | 20-0 ye ieee aes ye tees 55 Bs 23 °9 29-0 19-0 Feige) at eis 2437 ORD ae y j 24 °4 28-0 21-0 ie 5G ally Dario Res eddie is b 24-9 30-0 20-0 ay fora oe ae i 4 25-2 30-0 22-0 et Gia Fal yadda ed tape | ‘ tt eal Wied ars 28-0 21-0 Pie pies evil et ere Paes if fe MARGE os icey 31-0 2-0) ae Ovi pages z 4 25-0 30-0 20:0 | al cal eoaayeg pin eo Sole a 5 24-8 32-0 200 Bt SiG) okt ROO aaNm ett» e PB ly B30 18-0 | | i | | | 23°75 32-0 Tico | 200 Sir D. Bruce and others. Trypanosome Table XIX.—Distribution in respect to Length of 500 Individuals of the Trypanosome of Strain X, Dongolosi. In microns. | ! | | | | 17. | 18.| 19.| 20.! 21. | 22. | 23. | 24. | 25. | 26, 27 | 28. | 29. | 30. aL 32 | i Sa | | | | | | | | otal sieeeceeeee 2 | 9 | 80 | 86 | 51 | 52 69 | 68 | 61 | 48 | 25 | 23 | ah | | Py 1a 5 | Cae ty St es 13 °8 | 13 °6) 12:2 eee” 2|2-4/0-4| 0-2 | | | | Cuarr 5.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain X, Dongolosi, taken on nine consecu- tive days from Rat 2437, at the beginning of the infection. IL Microns - - — 13 [14] 15] 16 [17 [18 [19 |20]21 |22|23/24|25]26|27]28/29|30 31 [52 ]33 34/95/36 |37|38 SPEEEEE EE Eee = H = Sa a (ial aa : ee ieareeee oe a a JAE 3S Percentages - NOAA WAN @ © FE PEE : Cuarr 6.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain X, Dongolosi, taken on the 33rd to 41st day of disease from Rat 2462. Microns Teall T 13 | 14] 15] 16] 17 |18] 19 |20) 21 |22)23 |24|25|26|27|26|29]30/51|32/33]34/ 95] 36 |>7 |] 98 (f secesede' eeeuereeeeeeaerd ft ie Zia ata | ne FEEEEEEEEE ‘Sie LI ia Eee | = JI is cS) if Percentages ———— -Nvoreu — ie [ A [ at aa aie causing Disease in Man in Nyasaland. 201 It is evident that in this case there is little difference in type in the trypanosomes taken during the first nine days and the last nine days of an infection. In the former there are 25 per cent. short and stumpy forms, 37 per cent. intermediate, and 38 per cent. long and slender; in the latter 22, 43, and 35 per cent. respectively. ‘ + Table XX.— Percentage of Posterior-nuclear Forms found among the Short and Stumpy Varieties of the Trypanosome of Strain X, Dongolosi. Experiment | F Percentage among short DEES. iNet Hyena and stumpy forms. F 1913 1 INOWAUS) Mee eacion ta 2437 jad ata igen neee 1 ont Fir eevee 2437 Bie ee Re 3 DEC 2. Sera 2437 suet -pee: 0 5 EMR e? 2437 Ss cared cots 98 1 Palanan tabeeer act 2437 Bice ante 10 Pie Duiak none 2437 Sta saetetest 3 ree ONG were ae 2437 oe iacsece 4 sn ae 2437 Pibiaceceoadc Gi RECUR) Muck TB Te gl tee ee 10 4s pgtmeetcnre: 2437 srigpesecadbrea” 11 4 oO 8 9 iQ o or fo) Breadih—The following Table gives the breadth of the trypanosome of Strain X, Dongolosi. Table XX1—Measurements of the Breadth of the Trypanosome of Strain X, Dongolosi. f | | In microns. Experiment | Number of | ——_—— = Date | PNo Animal. | trypanosomes | | | | ; measured. Average | Maximum) Minimum | | | breadth. | breadth. | breadth. | | | | | 1913 | 2437 Rataeeeeee 500 2°71 4,50 1°25 To ASCERTAIN THE TYPE OF TRYPANOSOME WHICH ARISES FROM A SINGLE TRYPANOSOME. There is always, when dealing with a dimorphic type of trypanosome, a danger of there being two species present. Some experimeuts were, therefore, made by inoculating animals with a single trypanosome, to find out if the original dimorphic type would appear. The single trypanosomes were picked out in the usual way, by means of dilution and capillary tubes. The blood of 202 Sir D. Bruce and others. Zrypanosome an infected rat was diluted with normal saline solution until a volume one-sixteenth of an inch in length of a fine capillary tube was found to contain a single trypanosome. This was then inoculated into arat. Six experiments were made, three of which were successful. In each case the long and slender type of trypanosome was isolated and injected. » As will be seen from the following curves, from the single long and slender type, short and stumpy, intermediate, and long and slender forms resulted. Among the short and stumpy there was a large percentage of the blunt-ended posterior-nucleated forms, which are a feature of this species of trypanosome. From these three experiments, then, it may be concluded that in the dimorphic trypanosome causing disease in man in Nyasaland, a single species is being dealt with. Cuarr 7.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain X, Dongolosi, taken on nine consecu- tive days from Rat 2493, which had been infected by a single trypanosome of the long and slender type. Microns. | 19 [14] 15] 6] 17 [18 [19 [20] 21 [22]23 [2425] 26|27|26 |29[20]31 foe [93] 24|>5 [96 [37] 36 2 + Pie Seite : : | LI fall 1 a ec 4 oe aE Big PEEP EES EEEECCCE EEE causing Disease in Man in Nyasaland. 203 Cuart 8.—Curve representing the Distribution, by Percentages, in respect to Length, of 500 Individuals of the Trypanosome of Strain X, Dongolosi, taken on nine consecu- tive days from Rat 2489, which had been infected by a single trypanosome of the long and slender type. 7 as - T T }13 [re ]15] 15] 17 [18] 19 [20] 21 ]22)23|24/25]26|27|28| 29 30]31 [52 [33134] 9596 [97 38 ee (eens el a | é a ea al Eel A oL Lu cai 14) + + — | I —t rb] +—+ ==: > ho | fal al | c | ae | letotel aaa ! ~ Pe o i JL A | la fa CI eee: FA eae eer LL COMPARISON OF THE HuMAN Srrains VI To X. Table XXII.—Measurements of the Length of the Trypanosomes of the Five Human Strains VI to X. The trypanosomes have been taken from the rat alone. In microns. | | _ Number of | Date. Strain. | Name. | trypanosomes | | | | measured. Average | Maximum) Minimum | | | length. length. | length. | | | 1913 | VI | Manakumpara 500 21-7 32-0 15-0 1913 LVAD ea eVioramueee eee | 500 | 22°5 34°0 16°0 1913 Vague Mekka ......... | 500 aoe 35g OM LOE Olan 1913 IX | Mkanthama ... 500 pile BScOm le 14-0) | | X | Dongolosi ...... 500 CB |) S20 4 eo | | | | | | | | | | | | BB | BAG |) MO | The average length of Strains I to V, taken from rats alone, is 242 microns ; maximum 38, minimum 15. This gives an average for the 10 strains of 23:2 microns; maximum 38, minimum 14. Sir D. Bruce and others. 204 Trypanosome Table XXIII.—Distribution in respect to Length of 2500 Individuals of the Human Strains VI to X. The trypanosomes have been taken from the rat alone. In microns. | | | | | ~F 14,| 15. 16. 17.| 18.) 19. oY 21.) 22.| 238. 2. 29: | 26. 2 28. | 29, | 30.| 31.| 32.) 33.) 34 | Seas = ; [ty - [ = | Totals 1 | 12 | 31 | 97 | 187/260 286 283 262 | 233 205 178) 140 109| 85 | 57 | 45 | 18) 11] 4] 1 Per- 0:05) 0°5 | 1°2)3°9| 7-4 Lo aa is cai is 3/8°2/6°9)/5°8/ 4-4/3 -4)2°3/1°8)0°7/0-4)0°2|0 05 centages | | | | | | | | | | Cuarr 9.—Curve representing the Distribution, by Percentages, in respect to Length, of 2500 Individuals of the Human Strains VI to X of the Trypanosome causing Disease in Man in Nyasaland, taken from the Rat alone. a IM) Ve te © fal So 13 | 14] 15] 16] 17 [18 | 19 [20] 2) |22}23|24) 25 26[27 26 |29/30} 31 [32] 33/34/35 36 |37] 38 ee | a i oe ; | t i | 17 |} + + 16) 4 = | 15 14 f + t iB) + - | 2) wu 12 — + +————} Du r =! | an | | { | S 9 dl =| }__| al ) u 8 j = + i, = amalaat | o7 ime) T a6 | | { 5 T | 4 r + | 3) - ~ ca | | | i 2||4 sale 4 H = + | ae => This curve is very similar to that made from Strains I to V,* except that it lies a little to the shorter side. Table XXITV.—Comparison of the Percentages of Posterior-nuclear Formsfound among the Short and Stumpy Varieties of the Human Strains VI to X. * “Roy. Soc. Proc.,’ B, vol. 86, p. 301 (1913). Experiment : | * 2 Percentage among short Desi, No. Se, | time and stumpy forms. 1913 | 2239 | VI, Manakumpara ...... |) JR ceadan oan 14°83 1913 | 2236 VII, Yoramu............... Ep aenade 13-4 1913 | 2300 WADE, WIGNER, Gas ucdaceen aor ay) pbocoNGoo 24-2 | 1913 2386 IX, Mkanthama ......... Peon ees 28°6 | 1913 2437 X, Dongolosi ............ Aor tai 340) | | Average ............ | Wf “il The average percentage of the Strains I to V was 17°8 microns. causing Disease in Man in Nyasaland. 205 Table XX V.—Comparison of the Measurements of the Breadth of the Trypanosomes of the Human Strains VI to X. | In microns. Date. pepernent Strain. | Animal. | (eRean ot | Average |Maximum) Minimum | breadth. | breadth. | breadth. | | | 1918 2239 VI, Manakumpara...| Rat...) 2°76 | 4°50 1-25 1913 22386 VII, Yoramu............ mp. cooseal) BIL SD fe aL rhs) 1913 2300 VALI kcal emer ievagirecnt 2°68 | 5:00 1-25 1913 2386 IX, Mkanthama ...... [Rese earer eee 2°56 5 ‘00 1°25 1913 2437 X, Dongolosi ......... If Busi eee ee 27) 47-50 1°25 2°65 5-00 1°25 CONCLUSION. These further five strains of this trypanosome, isolated from five natives in Nyasaland, belong to the same species, Trypanosoma bruce vel rhodesiense, the trypanosome causing disease in man in Nyasaland. The Trypanosome causing Disease in Man in Nyasaland. Il. The Wuld-game Strawn. Ill. The Wild Glossina morsitans Strain. Part I1.—Susceptibility of Animals. By Surgeon-General Sir Davip Bruce, C.B., F.RS., A.M.S.; Major A. E. HAmERTON, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received May 5,—Read June 25, 1914.) INTRODUCTION. In previous papers* the morphology of these strains of trypanosomes was described, and it was concluded that they are identical with the try- panosome causing disease in man in Nyasaland, the Trypanosoma rhodesiense of Stephens and Fantham, the 7. brucei of this Commission. This paper tabulates the action on animals of the two strains, and they are compared in this respect with each other and with the Human strain. * ©Roy. Soc. Proc.,’ B, vol. 86, pp. 394 and 408. VOL. LXXXVIII.—B. R 206 Sir D. Bruce and others. Trypanosome ANIMALS SUSCEPTIBLE TO THE TRYPANOSOME CAUSING DISEASE IN MAN IN NYASALAND. Il. The Wild-game Straim. Table I. No. of | Period of | Duration Date. ae Source of virus. | incubation, | of disease, | Remarks. ae | in days. | in days.* Goat 1912. | July 1...) 718 | Hartebeeste 7/79......... a 67 | Died of Wild-game strain. Bye eae E Oo7ee ne 13 601 i . Sy alse dy Ra OO OM Neer 9 oy) | # i Aug. 21...) 1126 | ‘otha TAD area 5 leas mea i i Pee LZ SEilanclel 202s eerste 16 82 Also showed 7’. pecorum. » 24...) 1127 | Waterbuck 1180 ...... 9 4y Died of Wild-game strain. of CASiono| dLanl es) ah PAO) sepao coo 19 49 | Also showed 7. capre. Average......| 8:6 | 456 Monkey. June 19...| 785 | Reedbuck 783............ 8 iy | Died of Wild-game strain July lie 776 | Hartebeeste 779......... a 30 = . » Sool SG) |) OrBllot HEB sscccos90 200 000006 7 | 59 | * 4 oy bese 961 Hartebeeste 957......... 13 21 09 m0 Sy focal UCIT i OOO 9 | és : Aug. 24... 1181 Waterbuck 1180 ...... 9 40 : es 5 onal TALI ee ono 8 | S vs Sept. 13.... 1348 | Reedbuck 1347 ........ 10 q| 19 b fs ‘ 5 Bred 1S pa Ass aaee 10 ennGGe mull i fe | eS meer i Average...... 9:0 39 9 Dog June 30...| 784 | Reedbuck 783 ......... 8 11 Died of Wild-game strain. July 1... 733 Hartebeeste 779......... i 58 .; 5 se LO 846 | Monkey 785 ............ | 5 A8 i >» 10... 848 Dog (33) mebsilelet tiie cieraierhctate i 8 43 | > ” ny) BoA 892 Ilartebeeste 957......... | 6 31 h ” SZ Ore ool aaleMlonkey, 864i heme ey | 6 48 x am » 27...| 1002 | Hartebeeste 1000 ...... 9 59 fe ths Aug. 21...) 1144 é dope | 1 [re 28 in is » 24...) 1182 Waterbuck 1180 ...... | 9 40 ins A BOSE ate i WPAN edsceesp | 8 | ) 5) Sept. 7...| 1266 " TOG ue aaemnlG 64 | x # » 18...| 1849 | Reedbuck 1847 ......... | 10 38 | = a 5 OB. WAR LE Magar: eth 10 42 st < Average...... i) 41-2 | Rat July 10... 847 | Monkey 785 ............ 5 30 | Died of Wild-game strain. aLOM. 849 ID Voy 7/833 bo5 cn coansolsdacas 8 39 a 39 ; Ra IZAGHog 992 Monkey 864 ............ 6 21 is 3 Aug. 13...| 1070 1D o¥5 WOO), seecea sseoandee 6 17 if mS LO 1022 ate SOD) py aia chs aly ees 3 30 | 5 op Sept. 3 1220 Mronikelyan AG Npeerereeneee 3 54 ie 3 Average...... 52 31°8 * Duration includes the days of incubation; it dates from the day of inoculation. causing Disease in Man in Nyasaland. 207 In making up the average incubation and duration, mixed intections are not included. It must be admitted that these averages are only approximate, as it is impossible to deal only with animals of the same age, weight, health, and powers of resistance. Dogs, for example, fall away very much during ' the rains, when biting flies and ticks are numerous. Table I11—The Average Duration, in Days, of the Disease in Various Animals caused by the Wild-game Strain of the Trypanosome causing Disease in Man in Nyasaland. Goat. Monkey. | Dog. | White rat. | | Average duration, in days............ | 46 | 38 | Al 32 No. of animals employed ............) 5 | 9 | 13 | 6 | | Compare this with the following Table :— Table I1I.—The Average Duration of Life, in Days, of Various Animals infected with the Human Strain of the Trypanosome causing Disease in Man in Nyasaland. Goat. | Monkey. | Dog. | White rat. Average duration, in days............ | 42 | 25 | 24. | 30 | 25 | 21 | No. of animals employed ............ 29 | 20 Table IV.—The Percentages of Recoveries in Various Animals infected with the Wild-game Strain of the Trypanosome causing Disease in Man in Nyasaland. | Goat. Monkey. | Dog. | White rat. TEE OC NES aed o0bsc0 cod cpabecesdece oped (0) 0) | 0 0 No. of animals employed ............ 5 9 | 13 6 Compare this with the following Table :— 208 Sir D. Bruce and others. Trypanosome Table V.The Percentages of Recoveries in Various Animals infected with the Human Strain of the Trypanosome causing Disease in Man in Nyasaland. Goat. Monkey. Dog. White rat. | iPercentacestarereecreneeee eer eee | 0 0) | 0 (0) No. of animals employed ............ 29 20 | 25 21 | Ill. The Wild Glossina morsitans Straim. Table VI. | NaWOE | | Period of | Duration Date. nee + Source of virus. | incubation, | of disease, | Remarks. 1p | in days. | in days.* | | Cattle. 1912 April13....) 437 | Dog 325 ......... 9 — Recovered. el She nt .3 Sar | eee one 9 = - Goat. | Jan. 21... 385 | Wild flies ......) 19 35 Died of wild fly strain. Feb. 1... 117 | Monkey 20 ......| 11 50 Also showed 7’. simiz. op Bool) — AO AYSIU S55 onc 6 24 Died of wild fly strain. Weal | BON | Dey NB 2 505050| 12 27 i e 9 Ldccul| AOU ag LG seta 9 40 3 5 April13...) 421 VOB 5k wae 19 130 3 “3 op lescal! 4B} oh py BID), Set enoc00 | 9 82 5 3 = Wel 2O4 ||. FEB 9 9J . i May 15 416 | Wild flies ...... 5 19 Also showed Z. simiz and T. pecorum. June 12...) 637 | Rat 543 .........| 29 82 Died of wild fly strain. 2 Fh 68S 0 aos en 12 46 a i" Be Aes 639 Fy MOSSY ie attias 12 — Still alive after 224 days. Ly Boos! NG a Vale ths. shoo 10 60 Died of wild fly strain. | Oct. 38L . 1538 PEt noone 8 24 Also showed 7’. pecorum and T. caprz. Noy. 23 1626 Any a Wi et obe 7 56 Died of wild fly strain. veel LOSS said SN tense | 15 64 Also showed 7. pecorum. Dec. 5...) 1667 Mie p Wes sais 11 80 Be T. pecorum and T. capre. 7 oool| aa) Gok eh aro ocal 4 40 48 T. capre. fy Boal) ales) EN le nod 7 38 rs T. pecorum. ee, | April16..., 2084 | Rat 2020......... | 5 41 Died of wild fly strain. jG BOS [apy HOBO GR ccccen 19 32 x ia i HG) dil) 2OS6 14s w2020 Wee 8 32 4 i 1 LG.ool| AGSi7 $9 AOPAD) ogso0050e 8 39 i Gs |) BORE |, ORD prdconos 8 44 € cs Average...... 11°8 | 54°33 Pig. 1912. | | Nov. 25...| 1636 | Wild flies ...... 14 98 Also showed 7. simiz and T. gecorum. | | | 1913 | Jan. 21...) 1781 Fee ae atin) 6 230 s T. pecorum. April12...| | 2074 =I Me oR 11 |} 24 ss T., simize. May 14...| 2169 Be aig eaB ehoyst 8 | 25 ‘a " * Duration includes the days of incubation ; it dates from day of inoculation. 2 causing Disease in Man mm Nyasaland. 209 Table VI—continued. | NOLL Period of | Duration Date. 0. O" | Source of virus. incubation, | of disease, Remarks, exe: in days. | in days.* Monkey. 1912 | | Feb. 26 2a7alDoe daz) on. 7 28 | Died of wild fy strain. Pl Scat PSG Nene ol. tanunee 6 52 | Also showed 7’. simiz. Mar. 13 SB |) Cron NG sopteodee 8 31 Died of wild fly strain. April 13...| 406 | Dog 325 ......... 9 71 z n ee e402, |",,. 4369 a: 5 63 be i May 8.. 523 | Wild flies ...... 4, BY Also showed 7’. simiz. pos ae 60L | et ewe bus uns 6 31 : T. pecorum. June 7.. 625 | Sea ee 9 83 Died of wild fly strain. nL 2s, 739 | ae lol bales 5 45 3 5 July 24 970 | Rat 658 ......... 8 = Still alive after 162 days. Sept. 27...) 1459 Wild flies ......| 9 13 Died of wild fly strain. Oct. 29 .. 1536 | PD ky i nao nee 10 12 tD y 1913. Jan, 13...| 1757 Hal 4 Sesh: 6 32 - May 14 2151 | Rat 2082......... 5 30 ; , wy digas) eel te > ere ae 5 41 d miata orss. |; 20822... 5 22 . j PAP O1B4e N55 2082 IH. 6. | 5 28 ik : ues 2155 | ,, 2082......... 8 33 x Average...... 6-9 38 °7 Dog. 1912. | Feb. 1...) 116 | Monkey 20 ...... 8 23 Died of wild fly strain. Pot Ohen meee Zier Wit GiHies) 5 veneer 6 — Killed March 2. anil, WESO |} Dey IDG) sen coc sae 9 11 Died of wild fly strain. » LZ...) 248 Pe Gy ercaseee 5 23 A 33 ay fora! | AE allies tele 5 23 hs ie Mar. 9... 325 | Monkey 286 ... 9 41 oy * April12...| 436 | Wild flies ...... 7 36 5 0 18) 440% || Dor 825 «..... 5 29 3 ‘ » 13...) 441 ODO eres. 9 60 3 > MOTE AOS. Se ha SOI nic! 5 60 4 Mi May 10...) 525 | Wild flies ...... 3 8 a s Ramergtl | 542 BS in 6 42 i K Pemnt ues 549 | Monkey 523 10 51 Also showed 7’. pecorum. op les 551 | Wild flies ...... 8 25 3 x » 29...) 595 ala eres: 3 18 5p Pol... 602 Set Map anaes 5 — Still alive after 175 days. June 8...| 626 ae Wake oe 8 53 Died of wild fly strain. 6 A@oadt CAE Patino 5 26 5s 3 July 24...) 971 | Rat 6d8 ......... 8 32 3 ee Oct. 30...) 1537 | Wild flies ...... 5 30 3 5 Nov. 22...) 1625 EEO Pere 4 30 * if 5 PBwdl NERY BES ALN rin] 12 38 if Dec. 7...) 1675 SRRDENA abe Seren 6 25 a 55 Meso “1684 rin Bee as rs 19 is fr 1913. Jan. 22...) 1782 | Wiid flies ...... 9 | 21 i Pa May 14...) 2146 | Rat 2082......... 5 17 & : » 14...| 2147 5 AUEPS sescboboe 8 17 : » ides) 2048 ele woOea) mil. 8 17 é PU TACROT49 Ea cognl | 5 11 Es emia | 2150 es 2082 5 24 a i. Average...... 6°4 28°6 * Duration includes the days of incubation ; it dates from day of inoculation. 210 Sir D. Bruce and others. Trypanosome Table VI—continued. | | | lanauae Period of | Duration Date. ax t Source of virus. | incubation, of disease, | Remarks. DY: in days. | in days* | Rabbit. 1912. i April 13 439 | Dog 825 ......... a == Never showed trypanosomes. Dec. 14. 1543 | Pig 1636 ......... 19 | 13 | Died of wild fly strain. Pe TUN AE ed bp GR Brie) 19 eo 3 i son tl4e 1545 op HIGBXS ahcosooc: 33 | 90 | ms Ps Average...... 23°7 | 47 °3 Guinea-pig. Feb. 17. 239) | Dog sGieesece | 16 80 Died of wild fly strain. Pal ee 240 5 Pel Giese 16 72 | es a April 13. 442 1 ,, 325 ......... | — — | Never showed trypanosomes. May 14 544 | Monkey 492 ...! — = 5 x June 14. 676 | Dog 549 ae eel —_— — = 56 se 1 Bil Nita E28) ee ce eer == fs if peels COS MW) cg ED asa Ssonne lp 13 jo elliG Died of wild fly strain. OA. GiO wey eee et) oa F ie 53). LA: 680 re ar Batt eS | 13 39 Also showed 7. pecorum. As 681 DO ye orate 20 | 42 x 4 sy) LA: 682) ol pee Ooms eee — — | Never showed trypanosomes. 5 CY, 683 | Bath) saeeeree| = = Fe ; a 1913. Jan. A...| 1731 | Wild flies ...... 15 |} 100 | Died of wild fly strain. Mar. 28...) 2034 | Guinea-pig 1781 10 53 ie re » 28...| 2035 | 5 1731 10 89 fs s April16.... 2077 | Rat #020......... a5 61 ¥ . ey Ghsie 2078 Fe OAD Cane 12 | 61 *5 .. » 16...) 2079 #2020) ..cdcced 8 72 s . | Average...... 13 °5 80°8 Rat. 1912. | Web. 17... 241 | Dog 116 |....... 5 17 _| Died of wild fly strain. capil ps ie a ee Os Meea as 5 ic | n i April 13.. 443 a OLD M ero 9 24 | 5 : eS AA yal are Geer 12 14 : May 7.. 519 5) PAOD RY 6 12 . rs po WA 543 | Monkey 492 6 29 “ é oy Eso! GD TREE GUNG) sccscoce 6 12 . 3 June 11.. 655 | Dog 549 ........., 5 36 5 : Ua Ts | = GiGuel we essai Gene 6 194) J re i Aenea il eee G5oG eds UDI eee 13 31 Also showed 7. pecorum. Pee bi Bevan CHS |) sy SEY Saoasonos 8 43 | Died of wild fly strain. pe dle G60) Ge) G02 ene 5 70 Also showed TZ’. pecorum. Dec. 3 1664 Monkey 970 13 40 | Died of wild fly strain. 1913 Jan. 13::.| 1755 | Rat 1664......... 7 71 5 os Mar. 25...; 2020 By oO meses 13 22 ; 25 Des orbs Nie Se nit ec. 13 22 ‘ : April16...! | 2080 ng PAUPAW Sagconce- 1 27 : » eS -AGhs | 20ST ls, 2020) ee ee 5 31 0 x 5 16...) 2082 A220 em ates 5 28 ; a ee LG 2083 93 LO ZOOM ee ares 5 27 | 35 - May Ubi), 266) || 5, 2082). ce 8 Ta , 5 | =—>— > —. | Average...... 7°3 26°3 * Duration includes the days of incubation ; it dates from the day of inoculation. causing Disease in Man in Nyasaland. 211 Table VII.—The Average Duration, in Days, of the Disease in Various Animals caused by the Wild Glossina morsitans Strain of the Trypano- some causing Disease in Man in Nyasaland. | | | | | Winter | Ox. | Goa iMionkes ite Dorsal MRaboiee yecmmcts |) White | PIs |e radi. i] | | | | | Average dura- | Rec. 54 | 38 2B) | 47 81 26 tion, in days | | | Nowomanicalciy 20) aelGo ees Tee i 25 | es) | to) | * a9 employed ; | Compare this with the following Table :— Table VIII—The Average Duration of Life, in Days, of Various Animals infected with the Human Strain of the Trypanosome causing Disease in Man in Nyasaland. | Ox. Goat. | Monkey. Dog. | Rabbit. Gutreas Rite | pig. rat, | Average dura-| 134 42 | 26 | 34 28 | 67 30 tion, in days | | No. of animals. 1 om | oy |) oe 7 15 21 employed | | | Table [X.—The Percentages of Recoveries in Various Animals infected with the Wild Glossina morsitans Strain of the Trypanosome causing Disease in Man in Nyasaland. | | | Ox. Conta Monkey. | Dog. Rabbit. Suunieas White pig. | rat. I | | Percentages ...; 100 6 7 | 4 | 0) 0 0) No.of animals) 2 17 TS pha ahezera linc 10 19 employed | | oe Compare this with the following Table :— 212 Trypanosome causing Disease in Man in Nyasaland. Table X.—The Percentages of Recoveries in Various Animals infected with the Human Strain of the Trypanosome causing Disease in Man in Nyasaland. Ox! |) Goat. || Monkeys |) Dos) ||) Rabbit |) GlI2e sue pig. rat. Percentages ... 80 (0) 0) 0 0 0) 0 No. of animals) 5 BO 25 7 15 21 employed | | | | COMPARISON OF THE WILD-GAME AND WILD GLOSSINA MORSITANS STRAINS WITH THE HUMAN STRAIN OF THE TRYPANOSOME GAUSING DISEASE IN Man IN NYASALAND. Table XI—The Average Duration, in Days, of the Wild-game, Wild Glossina morsitans and Human Strains of the Trypanosome causing Disease in Man in Nyasaland, in regard to their Virulence towards Various Animals. Strain. | Ox. Goat. Monkey. Dog. | Rabbit. Giuinen- yy litte | pig. rat. | | | | | | | TSG pRENERN oo4 se sop09 26 |} avy | ae 26 34, 28/ | | 67 lente Wild-game ......... Niet 46 38 a a — | pe Wild G. morsitans| Ree. Oe) as 29 | 47 81 | 26 | Table XII.—The Percentages of Recoveries in Various Animals of the Wild- game, Wild Glossina morsitans and Human Strains of the Trypanosome causing Disease in Man in Nyasaland. i : | Strain. Ox. Goat. | Monkey. | Dog. Rabbi || aoc ee | plg. rat. TENWOAGHA 55+ yo se000s000 | 80 0 | (0) 0 0 0 (0) Wild-game ......... | = 0 0 0) — _ 0 Wild G@. morsitans| 100 6 7 4, 0) 0 0) CONCLUSIONS. 1. The pathogenic action on various animals of the Human strain, the Wild-game strain and the Wild G. morsitans strain is so much alike, that it may be concluded that they all three belong to the same species of trypanosome. 2. This species is 7. brucei vel rhodesicnse, the trypanosome causing disease man in Nyasaland. 213 The Trypanosome causing Disease in Man in Nyasaland. The Naturally Infected Dog Strain. Part IIL.—Development in Glossina morsitans. By Surgeon-General Sir Davi Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.0., and Captain D. P. Warson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received May 5,—Read June 25, 1914.) INTRODUCTION. In previous papers* the morphology and action on animals of this strain of trypanosome were described. In this a short account of its development in Glossina morsitans is given, in order to compare it with the development of the Human strain of the trypanosome causing disease in man in Nyasaland.t It is to be regretted that more material is not available, but, scanty as it is, there is enough to show that this strain develops in the alimentary tract and salivary glands of G. morsitans in the same way as the Trypanosoma brucei and gambiense group. The Commission aimed at having five positive experiments in every series of transmission experiments, but in this case failed. The failure was principally due to the difficulty of procuring laboratory-bred flies, and also to the fact that this strain of trypanosome does not readily develop in G. morsitans. THE DEVELOPMENT OF THE NATURALLY INFECTED DoG STRAIN IN G. MORSITANS. Eleven experiments were made with laboratory-bred flies. Two were positive and nine negative. Three hundred and seventy-six flies were used and fourteen were found infected—3-7 per cent. This small percentage is partly due to the fact that . In some of the experiments few or none of the flies were dissected. There is the same discrepancy to be noted here as in 7. brucei, Zululand, 1913. In some of the experiments not a single infected fly was found, whereas in Experiment 20184 there were seven in a cageful of 36. * “Roy. Soc. Proc.,’ B, vol. 88, pp. 111 and 130 (1914) + Ibid.. B, vol. 87, p. 515. 214 Sir D. Bruce and others. Trypanosome Table I.—Laboratory-bred Flies. | | No. of | Experiment | No. of | No. of No. of days Temperature | Date. Expt.| flies | positive or | flies dis-| infected before flies at which flies | used.| negative. | sected. jue found. | became infective. kept. 1912. Sept. 7 | 1257 | 25 - | 26 (0) 84° F. (29° C.) | Oct. 23 | 1499 26 = | 26 1 » Dec. 3) 1668 | 10 = 10 0 93 1913. | Jan. 13 | 1753 50 = eels 0 84°F, (29°C.) Mar. 24 | 2018 55 = 18 2 2 April 7 | 2067 30 + 30 3 24 ” | May 9 | 20184) 40 = | 36 7 ” | June 13 | 2226 40 - | 8 1 ”» | July 21 | 2303] 40 - | 29 0 » | Aug. 30 | 2394 | 40 + 0 0 58 és Nov. 12 | 2433 | 20 = 25 O 4 | Details of the Two Positive Experiments. The following Table gives the details in the carrying out of the two positive experiments. They were both carried out with laboratory-bred flies. Table II. | Day | Result. Date. of | Procedure. Remarks. CxpE: Positive. | Negative. Experiment 2067. 1913 April 7-| —) ” 8 1 ” g| 21180 flies fed on infected ” 10) 3f| Rat 2024. ees aera ne 5) ,» 18 6 | Starved. TAN oe eles 5 le 2 9 7) LO 59 al lf il Oh eee 5 BOIS fe GaL aA » 22} 15+ Fed on clean Dog 2054 ... + Trypanosomes appeared » 23} 16 in blood of Dog 2054 ae 24a LT, on the 31st day. Gs || | | , 26} 19 | , wt | BO | 5 S| Ol | oy CAS) 22 | | 5 OP PAL) causing Disease in Man in Nyasaland. Table I 1—continued. 215 Dawe Result. “Ny Date. of Procedure. Remarks. expe Positive. | Negative. Experiment 2067—continued. May 1] 24) Be) | 5 Ree 26 22 : a ie Fed on clean Dog 2054 ... + | Trypanosomes appeared MG) |126)| | in blood of Dog 2054 paw |, 30 | on the 31st day. ” 8 31) >» 9| 82 | Starved. S10" 33 | ral | Fed onclean Monkey 2131 | + | Trypanosomes appeared eee 35 | in blood of Monkey ls 36 | Starved. | 2181 on the 42nd day. | eelanl 937) ise) 38"| » 16} 39!]Fed on clean Guinea- — » 17| 40°] pig 2145 Poise pe ai | sy 1) |) 23) » 20} 43 | Starved. p21) 46) Bee Ul wedionicleaa Duiker!2059 Tr d | 23 | 46 ( ed on clean Duiker + rypanosomes appeared | - a4 | 47 } | in blood of Duiker | eo NSi Starved! 2059 on the 60th day. ». 26| 49) » 27] 504 28) || 51} ” 99| 52: Fed on clean Monkey 2184 + Trypanosomes appeared ” 30 53 | in blood of Monkey 2 31 54, | 2184 on the 68rd day. i 4 All flies dissected and | three found infected. | Experiment 2394. Aug. 20 | 1-6 | 40 flies fed on infected to Rat 2389. Aug. 26 Aug. 21 7 Starved. Aug. 22 | 8-65 | Fed on clean Dog 2404 ... + Trypanosomes appeared to in blood of Dog 2404 Oct. 24 on the 65th day. No | | flies dissected. Experiment 2067 was a successful experiment, as all the animals the flies fed on became infected with the exception of the guinea-pig, and it will be remembered that the guinea-pig was found to be refractory to this strain. Experiment 2394 also infected a dog, but as none of the flies were dissected none were found infected. From these two experiments it would appear that a period of from 24 to 216 Sir D. Bruce and others.. Zrypanosome 58 days may elapse before the cycle of development of the Naturally Infected Dog strain is complete in G. morsitans and the fly becomes infective. Details of the Nine Negative Experiments. The following Table shows the method of procedure in carrying out the nine negative experiments. In each of them laboratory-bred flies were used. Table IIL. Expt. Day of expt. Procedure. Remarks. | 1257 1-5 .| 25 flies fed on infected Dog 690. All flies dissected ; all 6 Starved. negative. 7-44 Fed on clean Dog 1313. 1499 1-3 26 flies fed on infected Rut 1218. All flies dissected ; one 4 Starved. found infected. 5-48 Fed on clean Dog 1500. 1668 1-3 10 flies fed on infected Monkey 1630. | All flies dissected ; all 4-5 Starved. negative. 6-23 Fed on clean Monkey 1670. 1758 1-3 50 flies fed on infected Monkey 1584. | 18 flies dissected; all 4 Starved. negative. 5-40 Fed on clean Monkey 1778. 2018 1-8 55 flies fed on infected Rats 1985 and | 18 flies dissected ; 2 in- 2023. fected. 9 Starved. 10-30 Fed on clean Monkey 2056. 31 Starved. 32-45 Fed on clean Dog 2112. 2018A 1-8 40 flies fed on infected og 2054. 36 flies dissected; 7 9 Starved. found infected. 10-35 Fed on clean liog 2172. 2226 1-6 40 flies fed on infected Rat 2214. 8 flies dissected ; 1 in- 7 Starved. fected. 8-37 Fed on clean Dog 2233. 23038 1-2 40 flies fed on infected Duiker 2059. | 29 flies dissected; all 3-13 Fed on infected Rat 2280. negative. 14 Starved. 15-44 Fed on clean Dog 2319. 2433 1-3 20 flies fed on infected Rat 2425. 15 flies dissected; all 4, Starved. negative. 5-38 Fed on clean Dog 2435. causing Disease in Man in Nyasaland. 217 RESULT OF THE DISSECTION OF THE INFECTED FLIES. Table IV.—Laboratory-bred Flies. Positive Experiments. Time, in days, means the number of days which elapsed between the first infective feed and the death and dissection of the fly. Hear Time, Prcwoucs Proventri- ae Fore- Mid- Hind- | Salivary Lae days. Tae culus. eh gut. gut. gut. glands. | 1 a : : : 2067 42 _ | | + | + + = 2067 50 us aoe uals os a ale: 2067 51 | | igue icl - | | | : | | In Experiment 2067, 30 flies were used; all were dissected and three found infected—10 per cent. This experiment was carefully carried out from beginning to end, and yet only one fly was found with invasion of the salivary glands. This one fly seems to have done all the mischief, infecting one after the other a dog, monkey, antelope, and finally another monkey. There is some doubt about this last monkey, as the fly died 50 days after its first infective feed, and the monkey had only come into use the day before. This particular fly, however, is reported not to have fed on that day, no fresh blood having been found in its intestine. It may be that it attempted to feed and so infected the monkey, but was unable to draw blood. If this fly did not infect the last monkey, it is difficult to explain its infection, as all the remaining flies were dissected on the 54th day and all found to be negative. In Experiment 2394 none of the flies were dissected. Table V.—Laboratory-bred Flies. Negative Experiments. Time, La Proventri- Fore- Mid- | Hind- | Salivar BS 0 days. SHOOROE, culus. CHD: gut. gut. | gut. | Mander | ; [ Ganamer ne j 1499 |} Il = are | se TS ar | = 2018 | 24 — ++ ++) ++ ++ — 2018 | 30 = + + — ZU18A 13 = = ++ aP oF + = 20184 14 - = + + ++ + — 20184 15 - = + + _ 20184 17 dod; if db + - 20184 18 - ef eter fi db = 20184 35 _ + a | 2018 35 - + ee 2226 32 | ab sp ap or + = One hundred and eighty-five flies were dissected and 11 found infected —59 per cent. In none was there found any development in the proboscis nor invasion of the salivary glands. 218 Trypanosome causing Disease in Man in Nyasaland. From an examination of these Tables it will be admitted, in spite of the paucity of the material, that the Naturally Infected Dog strain belongs to the same group as 7’. gambiense and 7’. brucei vel rhodesiense in regard to its mode of development in G. morsitans. | Tur Type OF TRYPANOSOME FOUND IN THE INFECTED FLIES. A number of drawings of the developmental forms of the Naturally Infected Dog strain of trypanosome was made from the alimentary tract and salivary glands of the infected tsetse flies. In the intestine the same type of trypanosome was found which has already been described and figured in previous papers.* In the only infected fly which showed a development in tne salivary glands, the trypanosomes were described in the living unstained preparations as being exceedingly numerous, small and active. In the stained preparations the trypanosomes were seen to be typical “blood forms” and absolutely identical to those figured in the development of the trypanosome causing disease in man in Nyasalandt and also in that of T. brucei, Zululand, 1913.f It is therefore unnecessary to figure them again. CONCLUSION. The trypanosome of the Naturally Infected Dog strain belongs to the same group as 7. gambiense and T. brucei vel rhodesiense, the trypanosome causing disease in man in Nyasaland, and is probably merely a weak strain of the latter species. * ‘Roy. Soc. Proc.,’ B, vol. 83 (1911). + Ibid., B, vol. 87, p. 516. t Ibid., B, vol. 87, p. 493. 219 The Trypanosome causing Disease in Man in Nyasaland.— The Naturally Infected Dog Strain. Part 1V.—Hxperiments on Immunity. By Surgeon-General Sir Davip Bruce, C.B., F.R.S., A.M.S.; Major A. E. Hamerton, D.S.O., and Captain D. P. Watson, R.A.M.C.; and Lady Bruce, R.R.C. (Scientific Commission of the Royal Society, Nyasaland, 1912-14.) (Received May 5,—Read June 25, 1914.) INTRODUCTION. The following experiments were undertaken to find out whether the Naturally Infected Dog strain of the trypanosome causing disease in man in Nyasaland would protect against the other strains. These different strains have been described in previous papers as the “ Human,” the “ Wild Game,” the “Wild Glossina morsitans,’ “ Zululand, 1913,’ ete., and here they will be known by the same names. “Human” will therefore mean a strain of this species of trypanosome coming from man, “ Wild G. morsitans” from a tsetse fly, and so on. These immunity experiments were necessarily one-sided, as it was, with three exceptions, only animals which had recovered from the weaker Naturally Infected Dog strain which were available. There are practically no recoveries from the Human and other strains. One goat apparently recovered from the Mzimba strain, and a goat, monkey and dog from the Wild G@. morsitans strains, and are included. There are, therefore, no completed eross-inoculation experiments—or at least only one unsatisfactory one, Experiment 17—as would have been carried out if material had been forthcoming. It will be seen from the following experiments that the Naturally Infected Dog strain failed toimmunise animals against the Human, Wild G. morsitans, and Zululand, 1913, strains; but it is not known whether these strains, on the other hand, would have immunised animals against the Naturally Infected Dog strain or not :— 220 Sir D. Bruce and others. Trypanosome Experiment 1, Dog 459, Naturally Infected Dog Strain. | Strain of | Soulot. OE queen Date Expt. Source of virus. | incubation, i pein: | trypanosome. Soe RTEY. days, or | oa a recovery. / 1912. | | April 20...) 459 | From Dog 817............ | Naturally infected 19 R. | Dog 48 | Noy. 22...) 459 | From Guinea-pig 1333 | Human ............... 4 59 Remarks.—Dog 459, which had recovered from Naturally Infected Dog Strain 48, when inoculated with a Human strain died in 59 days. Conclusion.—The Naturally Infected Dog strain does not protect against the Human Strain IV, Chipochola. Experiment 2, Monkey 1792, Naturally Infected Dog Strain. | ] | | : Duration of | olan Period of | - . | Date. | Expt. Source of virus. Strain ot incubation, CHSEASE, i” | trypanosome. in days days, or : recovery. | | | 1913. | | | Jan. 22 1792 | From Rat 1741 ......... Naturally infected 5 | R. | | | Dog 48 | June 24 .... 1792 | From Rat 2167 ......... Naturally infected — | = Dog 2033 | 1792 Hrom Rat 2235 0... | JELWAMR .cposono sao une 9 61 July 12 Remarks.—Monkey 1792, which had recovered from Naturally Infected Dog Strain 48, and proved immune to Naturally Infected Dog Strain 2033, succumbs to the Human strain. Conclusion.—The Naturally Infected Dog strain does not protect against the Human strain, Yoramu. Experiment 3, Monkey 1793, Naturally Infected Dog Strain. | ; Peonlee Duration of = | Q : Strain of : ; disease, in Date. Expt. Source of virus. Noa incubation, d ry panosome. sea ays, or | in days. | recovery. { f 1913. | | Jan. 22 | 2793 | Brom Rat W741 7)... Naturally infected 5 R. | : Dog 48 | June 24 .... 1793 | From Rat 2167 ......... Naturally infected | — — | | Dog 2033 | July 12 .... 1793 | From Rat 2235 ......... ISQWHANAT 545200008 a0000 5 9 | | i} Remarks.—Monkey 1793, which had recovered from Naturally Infected Dog Strain 48, and proved immune to Naturally Infected Dog Strain 2033, succumbs in nine days to the Human strain. Conclusion.—The Naturally Infected Dog strain does not protect against the Human strain, Yoramu. causing Disease in Man in Nyasaland. 221 Experiment 4, Monkey 2164, Naturally Infected Dog Strain. ! Strain of Pewee OF ee a Date. Expt. Source of virus. t incubation, | J rypanosome. ate al days, or in days. : recovery. 1913. May 14 ...| 2164 | From Rat 2091 ......... Naturally infected — | — Dog 2038 June 14...) 2164 | From Dog 2157 ......... Naturally infected — — Dog 2033 July 31...) 2164 | From Rat 2285 ......... | Naturally infected — = Dog 48 | Dec. 17 ...| 2164 | From Rat 2437 ......... 1alNTER ccdcccoseconed 7 Alive 1.3.14 Remarks.—Monkey 2164, which was proved to be immune against the Naturally Infected Dog Strains 2033 and 48, reacts readily to the Human strain. Conclusion —Immunity to the Naturally Infected Dog strain does not imply immunity to the Human strain, Dongolosi. Experiment 5, Dog 690, Naturally Infected Dog Strain. | | . Strain of Period of | ‘Qivease, in Date. Expt. Source of virus. t incubation, mee rypanosome. sp a days, or in days. recovery. 1912. | : | July 17 .... 690 | Naturally infected ...... Naturally infected ? R. Dog 690 Noy. 22 ...| 690 | From Rat 1492 ......... Naturally infected — — Dog 48 | Dec. 20...) 690 | From Dog 1675 ......... Wild G. morsitans 10 43 H | Remarks.—Dog 690, which had recovered from Naturally Infected Dog Strain 690, and proved immune to Naturally Infected Dog Strain 48, succumbs in 43 days to the Wild G@. morsitans strain. Conclusion.—The Naturally Infected Dog strain doesnot protect against the Wild G@. morsitans strain. Experiment 6, Dog 1530, Naturally Infected Dog Strain. | | : Strain of Pemou 2 ieee Date. Expt. Source of virus. t incubation, peur rypanosome. se Gl | days, or in days. recovery. | | faite : Oct. 29...) 1530 | From Rat 1491 ......... Naturally infected 16 R. Dog 48 1913. March21...| 1530 | From Rat 1991 .........| Naturally imfected = — i Dog 48 April 11...) 1530 | From Guinea-pig 2034 | Wild @. morsitans 10 23 Remarks.—Dog 15380, after having recovered from Naturally Infected Dog Strain 48, and shown to be immune on reinjection, readily succumbs to one injection of the Wild @. morsitans strain. Conclusion.—The Naturally Infected Dog strain does not protect against the Wild G. morsitans strain. VOL. LXXXVIIT.—B. iS) 222 Sir D. Bruce and others. Trypanosome Experiment 7, Monkey 1534, Naturally Infected Dog Strain. Strain of EOE Or ee Date. Expt. Source of virus. ae incubation, 2 ry panosome. de days, or in days. recovery. 1912. Oct. 29...) 1534 | From Rat 1491 ......... Naturally infected 6 R. Dog 48 1913. March 21...| 1534 | From Rat 1991 ......... Naturally infected = = Dog 48 April 11...) 1534 | From Rat 2020 ......... Wild G. morsitans 6 17 Remarks.—Monkey 1534, having recovered from Naturally Infected Dog Strain 48, and proved to be immune to the same strain, succumbs in 17 days to the Wild G@. morsitans strain. Conclusion.—The Naturally Infected Dog strain does not protect against the Wild G. morsitans strain. Experiment 8, Monkey 1630, Naturally Infected Dog Strain. é Duration of a . Strain of Period of disease, in Date. Expt. Source of virus. A incubation, rypanosome. ; P days, or in days. recovery. 1912. Nov. 22 ...| 1630 | From Monkey 1534...... Naturally infected 10 R. Dog 48 1913. March 21...) 1630 | From Rat 1991 ......... Naturally infected —_ = | Dog 48 April 11...) 1630 | From Guinea-pig 2084 | Wild G. morsitans 13 85 Remarks.—Monkey 1630, having recovered from Naturally Infected Dog Strain 48, and shown to be immune on re-injection, suecumbs to the Wild G. morsitans strain. Conclusion.—The Naturally Infected Dog strain does not protect against the Wild G. morsitans strain. Experiment 9, Goat 427, Naturally Infected Dog Strain. 3 Duration of D , Strain of JPost of disease, in ate. Expt. Source of virus. tee incubation, Al = ry panosome. rin GSS ays, or recovery. ' / 1912. April 20...) 427 From Rat 392 ............ Naturally infected 10 R. Dog 48 1913. Jan. 22%.) 427 From Rat 1741 ......... Naturally infected — = Dog 48 Feb. 11...) 427 From Rat 1375 ......... Naturally infected = == Dog 48 Feb. 28...) 427 From Dog 1906 and} Zululand, 1913...... 38 54 Rat 1832 Remarkes.—Goat 427 has recovered from Naturally Infected Dog Strain 48, and has shown no reaction to two re-injections, but when inoculated with 7. brucei, Zululand, 1913, takes the disease and dies in 54 days. 1913. Conclusion.—The Naturally Infected Dog strain does not protect against 7. brucei, Zululand, causing Disease in Man in Nyasaland. 223 Experiment 10, Goat 432, Naturally Infected Dog Strain. Period of || Duration of E R £ vir Strain of : ae ti disease, in Date. xpb. ource of virus. trypanosome. ee ion, days, or eee recovery. © 1912. | | April 20 ...;| 432 From Rat 392 ............ Naturally infected 26 R. Dog 48 1913. | Jan. 22 ...) 432 From Rat 1741 ......... Naturally infected — — Dog 48 | Feb. 11 ...; 482 From Rat 1735 .........| Naturally infected — — Dog 48 Feb. 28 ...| 432 From Dog 1906 and | Zululand, 19138...... 31 143 | Rat 1832. Remarks.—Goat 432 has recovered from Naturally Infected Dog Strain 48, but succumbs to T. brucei, Zululand, 1913. | Conclusion.—The Naturally Infected Dog strain does not protect against 7. brucei, Zululand, . 1913. Experiment 11, Sheep 456, Naturally Infected Dog Strain. | Strai Period of Duration of Date. Expt. | Source of virus. av Os incubation CSGESE, We eo trypanosome. 5 z days, or | JI in days. ie recovery. 1912. | April 20...) 456 From Rat 392 ............ Naturally infected 5 R. | Dog 48 1913. March 21...) 456 Homie hate lO Olea e: Naturally infected oo — Dog 48 1914. Jan.7 ...| 456 From Rat 2470 ......... Zululand, 1913 ...... 10 Alive 1.3.14 Remarks.—Sheep 456 has recovered from Naturally Infected Dog Strain 48, but reacts when exposed to the virus of 7’. brucei, Zululand, 1913. Conclusion.—The Naturally Infected Dog strain does not protect against 7. brucei, Zululand, 1913. : Experiment 12, Dog 1253, Naturally Infected Dog Strain. | % | - Ge Duration of Date. Expt. | Source of virus. t Sinan Ot incubation, cliseusie, Un | Yypanosome. : days, or in days. | | recovery. | 1912. | | Sept. 6) «..|' 1253) || From Rat 1218 ......... Naturally infected 6 R. Dog 48 1913. Jan. 4 ...| 1253 | From Rat 1570 ......... | Naturally infected 5 R. | | Dog 48 Jans l7 i.) 1253) | Brom Rat 173840 1... | Naturally infected a — | | Dog 48 Hebs28iees i 253 From Dog 1906 and Zululand, 1918...... 6 48 Rat 1832 - Remarks.—Dog 1253 has recovered from Naturally Infected Dog Strain 48, but suecumbs to T. brucei, Galuland, 1913. Piece Naturally Infected Dog strain does not protect against 7’. brucei, Zululand, : Ss 2 224 Sir D. Bruce and others. Trypanosome Experiment 12, Monkey 1794, Naturally Infected Dog Strain. Date. 1913. Jan. 22 ... Heb. 28i--- May 23 ... June 11 ... Heh? Ait Expt. 1794 1794 1794 1794 1794 : Strain of Period of Source of virus. t incubation, rypanosome. : in days. From Rat 1741 ......... | Naturally infected — Dog 48 From Rat 1945 ......... Naturally infected — Dog 48 From Monkey 2181...... Naturally infected — Dog 48 From Monkey 2184...... Naturally infected — Dog 48 From Guinea-pig Zululand, 1913...... 9 Duration of disease, in days, or recovery. 42 Remarks.—Monkey 1794 has been proved to be immune to Naturally Infected Dog Strain 48, but when exposed to the Zululand strain succumbs. Conclusion.—Immunity to the Naturally Infected dog strain does not imply immunity to T. brucei, Zululand, 1913. Experiment 14, Monkey 1798, Naturally Infected Dog Strain. . Duration of ; Period of : . ; Strain of Fi é disease, in Date. Expt. Source of virus. t incubation, rypanosome. | days, or in days. recovery. 1913. Jan. 22 ...| 1798 | From Monkey 1630...... Naturally infected — — Dog 48 Feb. 28 ...) 1798 | From Rat 1945 ......... Naturally infected — — Dog 48 May 22...) 1798 | From Monkey 2131...... Naturally infected — — Dog 48 June 11 ...) 1798 | From Monkey 2184...... Naturally infected = — Dog 48 June 24 ...| 1798 | From Guinea-pig 2225 | Zululand, 1913...... 9 15 Remarks.—Monkey 1798 has been proved to be immune to Naturally Infected Dog Strain 48, but succumbs to 7’. brucei, Zululand, 19138. ; Conclusions.—The Naturally Infected Dog strain does not protect against 7’. brucei, Zululand, 1913. causing Disease in Man im Nyasaland. 225 Experiment 15, Monkey 2161, Naturally Infected Dog Strain. { { : Bsaeliee | Duration of : Strain of < é disease, in Date. Expt. Source of virus. incubation, trypanosome. id days or in days. aa recovery. 1913. : May 14 ...| 2161 | From Rat 2091 ......... Naturally infected — — Dog 2083 June 14 ...!| 2161 | From Dog 2157 ......... Naturally infected — — ; Dog 2033 July 31...) 2161 | From Rat 2285 ......... Naturally infected — — Dog 48 Dec Lijec 2LGLe | hromyRiat 2451s sete Zululand, 1918 ...... Ul Alive 1.3.14 Remarks.—Monkey 2161, proved to be immune to Naturally Infected Dog Strains 2033 and 48, shows no immunity to 7. brucei, Zululand, 1913. Experiment 16, Goat 639, Wild G. morsitans Strain. | | : Porinnlor Duration of | : Strain of : 5 disease, in Date. Expt. Source of virus. t incubation, rypanosome. : days, or in days. recovery. 1912. June 12 ...| 639 From Rat 543 ............ Wild G. morsitans 12 R. 1913. | Jan. 22 ...| 689 From Rat 1741 ......... | Naturally infected — — Dog 48 | Feb. 11...) 639 From Rat 17385 ......... | Naturally infected — — Dog 48 Feb. 28 ...| 639 From Dog 1906 and | Zululand, 1913 ...... 10 48 Rat 1832 Remarks.—Goat 639 has recovered from the Wild G. morsitans strain, has shown no reaction when inoculated with Naturally Infected Dog Strain 48, but is killed in 48 days by Z. brucei, Zululand, 1913. : ° Conclusion.—The Wild G. morsitans strain, combined with the Naturally Infected Dog strain, has no protective power against 7’. brucei, Zululand, 1913. 226 Trypanosome causing Disease in Man in Nyasaland. Experiment 17, Monkey 970, Wild G. morsitans Strain. Strain of Hen OL ©: ee Date. Expt. Source of virus. t incubation, y rypanosome. cn aoe days, or Ne: recovery. | | 1912. July 24 : 970 From Rat 658 ............ Wild G. morsitans 8 R. 1913. Jan. 2 970 From Rat 1664 ......... Wild G@. morsitans -- — Spe wal? 970 From Rat 1740 ......... Wild G. morsitans = _— Feb. 4 970 From Rat 1814 ......... Naturally infected 9 R. Dog 48 28? 970 From Rat 1945 ......... | Naturally infected — — Dog 48 March15...; 970 From Guinea-pig 1657 | Human ............... 9 gal Remarks.—Monkey 970, after recovering from the Wild G. morsitans strain, shows a reaction to Naturally Infected Dog Strain 48, and finally succumbs to a Human strain. Conclusion.—The Wild G. morsitans strain does not protect against the Naturally Infected Dog strain, nor does the combination of the two against the Human Strain V, Chibibi. Experiment 18, Dog 602, Wild G. morsitans Strain. | : Duration of tet Period of 5 : . Strain of 2 6 disease, in Date. Expt. Source of virus. t | incubation, ry panosome. sae days, or in days. recovery. 1912. May 31...) 602 Walditives ee-ce eee Wild G. morsitans 5 R. Nov. 22 ...| 602 From Guinea-pig 1833 | Human ............... 6 28 | Remarks.—Dog 602, which had recovered from the Wild G. morsitans strain, when inoculated with a Human strain died in 28 days. ; Conclusion.—The Wild G. morsitans strain does not protect against the Human Strain IV, Chipochola. CONCLUSIONS. 1. The Naturally Infected Dog strain does not protect animals from the Human, Wild @. morsitans, and Zululand, 1913, strains. 2. The Wild G. morsitans strain and the Naturally Infected Dog strain do not protect animals from the Human or the Zululand, 1913, strain. 3. The Wild G. morsitans strain does not protect against the Human strain. 4. In spite of the damaging evidence of these experiments, the Commission still holds the opinion that the Naturally Infected Dog strain is a weak strain of the trypanosome causing disease in man in Nyasaland, 7’. brucei wel rhodesiense. —————— ta 227 The Colouring Matters in the Compound Ascidian Diazona violacea, Savigny. By Atrrep Hott, M.A., D.Sc. (Communicated by Prof. W. A. Herdman, F.R.S. Received May 12,— Read June 18, 1914.) 1. Experimental Observations. The present investigation on the colouring matters of the compound Ascidian Diazona violacea, Sav. (= Syntethys hebridicus, Forbes and Goodsir), had its origin in an observation of Prof. W. A. Herdman while dredging in the neighbourhood of the Outer Hebrides in 1912. Some specimens of this rare Ascidian were collected by Prof. Herdman, which showed whilst alive the green tint described by Forbes and Goodsir,* but on placing them in alcohol for purposes of preservation it was found that after a few hours the alcohol had acquired the original green colour of the Ascidian, while the organism itself had changed to violet, a shade nearly complementary to that of the living animal. A description of these specimens has already been published. During another expedition to the Hebrides in the summer of 1913, Prof. Herdman obtained so many specimens of this organism that it was possible to use some of the material for an examination into the nature of the green and violet pigments, but a complete study has been impossible owing to the minute quantity of pigment found in any one Ascidian colony, and to the fact that no fresh and living material was at my disposal. The green alcoholic solution obtained from the specimens collected in 1912 had been examined, and a. brief account of the results was given in the above-mentioned paper in the Journal of the Linnean Society, but it will be useful to begin by recapitulating them here, and also to add some further information. The green colour of the solution was not unlike that due to chlorophyll, and it exhibited weil marked red dichroism. The absorption spectrum consisted of a broad band in the orange red, which was characterised by a more distinct edge towards the red than towards the yellow, and there was also practically complete absorption at the blue end of the spectrum. The band had the greatest intensity about 7» =620 wp, and the absorption in the blue and violet began at >) = 470 up, and continued downwards. Though not identical, this spectrum is not unlike that of true chlorophyll, * “Roy. Soc. Edin. Trans.,’ vol. 20, p. 307. t+ ‘Linn. Soe. Zool. Journ., vol. 32, May, 1913. 228 Dr. A. Holt. Colouring Matters in the in so far that the general absorption towards the violet begins at almost the same point, and that there is a very definite band in the orange red. The two spectra are shown in the accompanying figure, where their points of agreement and disagreement are more immediately visible. In 1875, Sorby* obtained a green alcoholic solution from the Gephyrean worm Sonellia viridis, which, when examined spectroscopically, gave an absorption spectrum which also resembled chlorophyll in some respects. In neutral solution there was a very pronounced band at \ = 636 py and distinct bands at X% = 587, 520, and 490 up, but he does not record any general absorption for 7X<470mp. His spectrum (“bonelleine”) is also reproduced in the figure for the sake of comparison. Judging solely from these observations, it appears very probable that ‘the green solutions from Diazona and Bonellia contain either chlorophyll (for chlorophylls from different sources have not identical spectra), or some closely related chlorophyll body. On cutting a violet, aleohol-preserved specimen of Diazona in two, it was found that the tint only extended a short way (about 1 cm.) beneath the surface, by far the larger mass of the colony (perhaps 20 cm. across) remaining a pale greenish yellow. The coloured outer portion of a specimen was therefore removed, and extracted with absolute alcohol. A blue-green solution was slowly obtained, but only a very minute portion of the violet pigment appeared to pass into solution. On cooling the alcoholic extract the blue-green colour became both paler and more yellow, though the original tint was restored on reheating, while after standing in the cold for some days a very small quantity of a substance having a violet tint identical with that observed in the outer portions of the colony was precipitated, the solution from which it had separated being now yellow-green. Extraction of the inner portions of a colony with absolute alcohol gave a solution of an almost pure yellow tint, scarcely any trace of yreen being detectable by the eye. No change took place either on cooling or after standing for some time. The spectrum of the blue-green solution appeared to be similar to that already described, there being absorption in the red, and general absorption in the blue and violet, but the yellow solution from the interior of the colony gave no distinctive band, but only a general absorption of the more refracted rays. On standing for several days these solutions became somewhat paler in tint, but attempts to concentrate the colour by distilling off the alcohol resulted in such turbidity that spectroscopic observations were impossible. The original green solution obtained from the 1912 specimens has scarcely altered in tint (April, 1914), but on concentration it also becomes turbid. * *Quart. Journ. Microsc. Sci.,’ vol. 15, p. 167. 700 B Cc 6000 E 500 F - G 400 a. Diazona (green solution) in alcohol. b. Chlorophyll in alcohol. ce. Bonelleine, green neutral solution in alcohol. Sorby. d. Diazona (yellow solution) in alcohol. e. Chlorophyll (yellow pigment) in alcohol. f. Diazona (purple pigment) in acetylene tetrachloride. g. Purpura (purple pigment) in acetylene tetrachloride. h. Bonelleine, purple acid solution in alcohol. Sorby. 230 Dr. A. Holt. Colouring Matters in the A few of the 1913 specimens had been preserved in formaldehyde solution instead of in alcohol. These had retained their natural green colour, and on treatment with alcohol gave a pale yellow-green solution, the organism itself becoming practically colourless, no trace of purple being observed. This pale yellow solution showed a faint absorption in the blue and violet, but no band in the orange-red. All the above greenish or yellow solutions were shaken with carbon disulphide to see whether any separation of the colouring matter could be effected. The green 1912 solution after repeated shaking changed to a yellow, or brownish yellow, both the aleohol and carbon _ disulphide layers being coloured to about the same tint. The green solution from the outer portion of a colony gave a green carbon disulphide extract, the alcoholic solution becoming yellow. From the strength of the colour it appeared that there was far more green than yellow pigment present. The yellow solution from the inner portion of the colony gave an exactly similar separation, but there appeared to be but little green pigment, as the carbon disulphide became only slightly coloured. The greenish yellow solution from the formaldehyde-preserved specimens was quite unaltered by shaking with carbon disulphide, this solvent seeming to dissolve no colouring matter. The green carbon disulphide extracts all showed an absorption band > = 620 ww and general absorption for \ < 470 pp, the yellow alcoholic portion exhibiting only a faint general absorption in the blue and violet. No satisfactory chemicai observations could be made with any of these solutions. Acids and alkalies gave no very characteristic reactions, as the addition of either merely caused the solutions to become somewhat more yellow. This was particularly the case with alkalies, acids often appearing to have no action. Neutralisation of the alkaline solution did not restore the original colour. Acid or alkaline hydrogen peroxide was also without visible action. Saturated barium hydroxide solution gave a greenish precipitate with the green solutions, the supernatant liquid being yellow, while with the yellow solution the precipitate had a yellow tint, though the liquid was not com- pletely discolored by the barium salt. There was too little precipitate to try the action of an alcohol-glycerine solution of boric acid. The unsatisfactory nature of these reactions is mainly due, no doubt, to the great dilution of the solutions employed, but the fact that neither the specimens of Diazona nor the alcoholic extracts were fresh may be a con- tributing factor, for chlorophyll bedies are not very stable. It may be mentioned that fresh chlorophyll is altered in a non-reversible direction by acids, whereas the pigment from onellia, as described by Compound Ascidian Diazona violacea, Savigny. 231 Sorby, changed to purple when strongly acidified, but regained its original shade on neutralisation. From all the above observations it must be concluded that the green colour of Diazona probably results from some chlorophyll-like body. Though the spectra are not exactly those of ordinary plant chlorophyll there is quite a resemblance, and the association of separable green and yellow pigments from the alcoholic solution is also very suggestive. If it is a chlorophyll body one is driven to the view that the green colour arises from a symbiotic alea, as chlorophyll does not appear to be a likely pigment for a marine animal. In a monograph on the compound Ascidian Fragaroides awrantiacum, by Charles Maurice,* the’ pigmentation of the test is described, and the author concludes that there the yellow pigment cells are in reality alge (a Protococcus), which contain chlorophyll. He comments on the fact that these alge when free show colours ranging from green to yellow, and that their cells during the period of reproduction resemble most closely the colour and structure of the globules in the test of the Ascidian. There are, however, three possible objections to this view. Firstly, Diazona has been collected from a depth of 60 fathoms, and it would appear to be most improbable that sufficient actinic light would penetrate to that depth to cause the formation of chlorophyll. Secondly, there is the evidence of Pizon} that in certain Tunicata which show very similar pigment cells to those of Diazona the yellow or yellow-green pigments result from the waste products of the organism, and are gradually excreted from its surface. Chloro- phyll would hardly be a waste product. Thirdly, the pigment cells in Diazona are far smaller than the algal cells in known cases of symbiosis. The cells appear as minute spheres filled with one or more drops of an oily substance (judging by their high refractive index), and do not appear to show the structure of an algal cell. Until it is possible to work with some fresh, living colonies of Diazona, nothing more definite concerning the green pigment can be said than that it resembles chlorophyll in many respects, but is not identical with that ordinarily obtained from plants. It is, however, more like chlorophyll than the green pigment obtained by Sorby from bonellia, and possibly represents algal cells. Extraction with alcohol having shown that the purple pigment was all but insoluble, some 400 grm. of the organism were worked up on the lnes employed by Friedlander in the case of the Molluse Murex brandaris.t The material was first ground with sand and then digested for several * “Arch. de Biol.,’ Liége, 1888. + ‘Compt. Rend.,’ 1899, p. 395, and 1901, p. 170. {£ ‘Ber.,’ 1909, p. 765. 232 Dr. A. Holt. Colouring Matters in the hours with hot dilute sulphuric acid, fresh quantities of acid being used till the yellow tint at first imparted to it was no longer visible. The mixture of animal matter and sand was then boiled with several portions of water and ‘filtered. It was next extracted with alcohol. Finally, it was treated in a Soxhlet with ethyl benzoate or acetylene tetrachloride. These solvents acquired a fine blue colour, and exhibited a strong purple-red dichroism. On cooling, the solutions gradually deposited a fine purple-black powder, which after recrystallisation and washing with ether did not greatly differ in tint from the violet colour of the Ascidian colony when preserved in alcohol. When quite dry this purple powder had a distinct coppery lustre. It was insoluble in water, alcohol, and ether, but soluble in aniline, pyridine, quinoline, nitro-benzene, ethyl benzoate, and acetylene tetrachloride, though the solubility varied with the liquid. Thus, though easily soluble in hot acetylene tetrachloride, it was almost entirely reprecipitated on allowing the cooled solution to stand for some hours. In every case the solution was blue with a greenish shade when very dilute, changing to pure blue, and subsequently violet blue, on concentration. When hot both the violet colour and the dichroism were more pronounced. The absorption spectrum was determined in both hot and cold ethyl benzoate and acetylene tetrachloride. The ethyl benzoate solution in the cold showed an absorption band with a maximum about >A = 611 yup, though absorption began at 7 = 617 py. When heated, the maximum was shifted to ) = 605 my, the band being very indefinite towards.the green, the total absorption ranging from 617 wy to 2598 uy. In acetylene tetrachloride the maximum absorption both when hot and cold was shifted towards the green, the maximum when cold being A = 606 wy, and when hot A = 598 py. The pigment dissolved in concentrated sulphuric acid to form at first a pinkish solution, which rapidly changed to a dirty purple colour. On standing, or more rapidly on warming, this colour changed to a brown tint with a green shade in it, or if sufficiently strong to a dullgreen. The pink colour appeared to be of a transient nature, depending for its stability on concentration and low temperature. Addition of water to the cold acid solution precipitated the pigment, so it must be concluded that it does not form a soluble sulpho-salt, as is the case with indigo, but after heating, the addition of water caused the separation of a dull green flocculent precipitate, not the original pigment. The colouring matter was insoluble in alkali, but gave a colourless solution with an alkaline reducing agent. Owing to the small quantity available it was impossible to try its action on cotton satisfactorily, but a cotton cloth in Compound Ascidian Diazona violacea, Savigny. 233 which an Ascidian colony had been wrapped during preservation was found after drying and exposure to air and light to be dyed with a pale pink, not very fast, colour. (Qualitative examination showed the presence of a halogen, apparently bromine, for after treating a few milligrammes of the dyestuff by the Carius method a pale yellow silver precipitate was obtained which did not appreciably darken in sunlight and which was slowly soluble in excess of ammonia. The general behaviour of the colouring matter was thus seen to resemble a dibromindigo, which has been shown by Friedlander to be the dye in the ease of the Mediterranean Murex brandaris. The chief point of difference appeared to be the bluer shade of tint in all the solvents employed, the greater solubility in ethyl benzoate or acetylene tetrachloride, and except in strong, hot solutions the displacement of the maximum of the absorption band somewhat towards the red. As living specimens of JZurex brandaris in quantity were not available, for the sake of comparison, the pigment from the closely related British Mollusc Purpura lapillus was therefore examined. The purple pigment of this mollusc has already been studied by many chemists.* In the present instance the colouring matter from material collected at Port Erin, Isle of Man, was extracted in exactly the same way as described by Friedlander for Murex brandaris, and was obtained in a pure crystalline condition from solution in ethyl benzoate or acetylene tetrachloride. It will suffice here to say that its appearance and reactions agreed in every particular with the dye from Murex brandaris :—66' dibromindigo. Solutions in various solvents were more red purple than those from Dzazona, and its absorption band in hot acetylene tetrachloride gave AX = 584 wy. The other three isomeric symmetrical dibromindigos have recently been described by Friedlander,{ and their absorption and behaviour in concentrated sulphuric acid are given for reference from the above mentioned paper. Compound. A. | Colour in concentrated sulphuric acid. bye. 44! dibromindigo ...... 613 Blue. 55/ avid Bea HL ERA 621 Blue. 66’ tele en Mised ee 585 Dull violet brown. a Were Maassanes 606 Greenish blue (peacock blue). * Bancroft, ‘ Philosophy of Permanent Colours,’ 1803 ; Negri, ‘Gaz. Chem. Ital.,’ 1875; Schunk, ‘Chem. Soc. Trans.,’ 1879 and 1880 ; Letellier, ‘Compt. Rend.,’ 1889. + ‘Ann. Chem.,’ vol. 388, p. 23 (1912). 234 Dr. A. Holt. Colouring Matters in the It will be observed that while the pigment from Diazona in some respects agrees with the 66’ body, in other respects it more resembles the 77’ isomer, which gives blue solutions in solvents, the colour being not unlike that of indigo. Possibly the Diazona pigment is some other isomer, or an indigo with a different number of substituted hydrogen atoms, but it 1s impossible to decide this point without far larger supplies of material. For the sake of comparison, in the figure (p. 229), the positions of the absorption bands for the violet or blue solutions obtained from Diazona, Bonellia, and Purpura are shown. 2. Orin and Formation of the Violet Pigment. The experimental evidence so far available does not enable one to ascribe any certain origin to the violet pigment nor to account for its development in such different organisms as Mollusca (Murex and Purpwra), Vermes (Bonellia), and Tunicata (Diazona). Nevertheless it may be useful to collect such evidence as there is at present to hand. In the case of Murex brandaris it seems to be well established that the colour has a photogenetic origin, but in M/wrex trunculus this is not the ease, according to Negri (/oc. cit.). In Purpura lapillus the pigment is produced both by the action of sunlight, and by hydrochloric acid in the dark, this latter- observation agreeing with the behaviour of Bonellia viridis according to Sorby. Further, the pigments produced photogenetically in these organisms are uniformly insoluble in alcohol, while those resulting from the action of acids are soluble. In the case of Diazona it is by no means certain that the pig- ment has a photogenetic origin. Prof. Herdman has recorded the gradual production of violet colour in the living organism under the influence of sun- light, the original yellow-green tint changing first to blue green, then indigo blue, and finally a dull or dirty violet, but he is of opinion that this colour- change attends a moribund condition. There is, however, no evidence that this change would not have taken place in the dark. Natural violet-coloured specimens of the Ascidian have been obtained alive in the Mediterranean, near Naples, and also grey-green specimens which have remained unaltered after preservation in alcohol* but these natural violet specimens do not appear to be healthy, and hence the formation of the dyestuff may accompany or result from a metabolic change. In alcohol the colour is produced in the dark, for the specimens were placed in a closed tank immediately after they were collected. Microscopic examination of an alcohol-preserved specimen shows the purple colouring matter apparently precipitated in the spherical pigment cells of the test, these cells in the inner * See Herdman, ‘ Linn. Soc. Journ.,’ 1913. ess Compound Ascidian Diazona violacea, Savigny. 235 portion of the animal being filled, as already mentioned, with a bright yellow-green oily-looking substance. It is possible that the action of the alcohol may be merely that of precipitant, for if the dyestuff (which gives a blue solution) were dissolved in this oil the resultant colour would be the green of the living organism. The chloroplasts, as mentioned above, appear to contain a substance the solution of which in alcohol has an absorption spectrum resembling a yellow chlorophyll body, but it does not follow that they only contain this com- pound. Some solvent for the indigo derivative may quite possibly be present in them as well. Now if this solvent is miscible with alcohol its removal would precipitate the colour body in its solid, violet-tinted form. The alcoholic solution, however, would still have a greenish tint, since some of the pigment would dissolve in the alcohol-solvent solution, exactly as in the case of other three-component systems, and so add its blue colour to the yellow of the chlorophyll-like substance. Hence the presence of pigment in the alcoholic solution need not necessarily imply the existence of a second colour body soluble in alcohol, and this indeed is believed to be the origin of the traces found in some of the alcoholic extracts examined during this investigation. It was remarked that the traces of colouring matter thus obtained were insoluble in absolute alcohol, a fact quite in accordance with the above view, since none of the natural animal solvent would then be present. It must, however, be pointed out that if the green colour of these extracts was due to the presence of a minute quantity of the violet pigment one would expect the spectrum to exhibit an absorption band about X% = 606 and not at X= 620 uu. It is of course a possibility that the animal solvent may shift the absorption band this amount towards the red, though this seems somewhat improbable in dilute solution in alcohol. It is far more likely that the traces of violet pigment found in the alcoholic extract had their origin in a disinte- gration of parts of the organism during extraction from purely mechanical causes. Though the production of the violet colour could thus be explained when specimens are preserved in alcohol, this precipitation theory seems scarcely sufficient to explain all the observed facts. According to this view the gradual production of the colour as observed by Prof. Herdman in living specimens must arise from its production in such quantity that it can no longer be kept in solution by the solvent in the pigment cells, yet there is no evidence that there is more colouring matter present under these circum- stances than when an ordinary green healthy colony is placed directly in alcohol. Further, it affords no explanation why the pigment is produced only 236 Colouring Matters m Diazona violacea. on the exterior, and not throughout the mass of the colony. It may also be remarked that a minute quantity of the pigment causes an intense coloration of its solvents, so much so, that if all the violet colouring matter in a colony was in solution during life the colour of the organism would almost certainly be a bright blue, not yellow green, as it would entirely mask the yellow of the chlorophyll body. The non-production of violet colour in the formaldehyde-preserved specimens is what one would expect from an indigo derivative, for the reducing action of the aldehyde would certainly produce colourless indigo- white derivatives, if indeed the whole molecule was not split up. The recorded phenomena can, however, be explained if we suppose that in the healthy animal the pigment is present dissolved in the pigment cells in its reduced condition as a chromogen. Owing to its natural tendency to oxidation the animal by maintaining it reduced could use it as an oxygen carrier, and, since the only available oxygen is in the surrounding water, its presence would only be expected on the exterior of the colony, though the green-yellow chlorophyll-like pigment is present throughout its mass. As soon as the animal became moribund or unhealthy metabolic processes would change and oxidation would begin, with the consequent production of colour. In the dead animal oxidation would be complete, and the colourless body converted into the violet pigment. The same change would occur in alcohol, which by killing the animal would allow oxidation to proceed rapidly. The colour results no doubt from the action of an oxydase, which in the formaldehyde-preserved specimens would be destroyed, and hence no colour would result. Until it is possible to experiment with living colonies one cannot express a definite view, but it appears more probable that some such series of changes as is outlined above takes place, rather than that the body in its fully oxidised condition is present in solution in the pigment cells of the live animal. With only preserved colonies available it is not possible to prosecute further this enquiry as to these green and violet pigments or to express any opinion as to their possible relationship, as regards function in the organism, to those found in Bonellia and various Mollusca. In conclusion my thanks are due to Prof. Herdman for providing several. complete colonies of Diazona and the green alcoholic solutions obtained directly from the living organisms, and for suggesting to me that a chemical investigation might throw further light on the colour relations of the violet Diazona violacea and the green condition known as Syntethys hebridicus. 237 Some Accessory Factors in Plant Growth and Nutrition. By W. B. Borromury, M.A., Professor of Botany, University of London, King’s College. (Communicated by Prof. F. W. Oliver, F.R.S. Received May 29,—Read June 18, 1914.) Recent research has demonstrated the importance of the presence of minute amounts of certain substances as accessory factors in normal dietaries of man and animals. The most striking examples of the influence of these substances are seen in their curative effect on the diseases of beri-beri and scurvy, and their stimulative effect on the growth of young animals. Beri-beri is caused by the deficiency in a diet of polished rice of a nitrogenous substance, small amounts of which are essential for the metabolism of the nervous system. The curative substance is found in rice husks, barley, wheat, lentils, yeast, ege-yolk, milk, etc., and is pre- cipitated from an aqueous solution of an alcoholic extract of these bodies by phosphotungstic acid. It is effective in very minute amounts, an addition to the diet of 0°02 grm. of the active fraction of the extract curing polyneuritis (beri-beri) in pigeons. Scurvy also is caused by a diet adequate as regards proteins, carbohydrates and fats, but deficient in some constituent, small amounts of which are essential. This anti-scorbutic substance is found in lime-juice, fresh vegetables and fruits, and, like the curative substance of beri-beri, is precipitated by phosphotungstic acid. The special importance of small amounts of substances of unknown com- position in the metabolism of growing animals has been demonstrated by the recent researches of Osborne and Mendel* and Hopkins.t These investiga- tions have.shown that young rats, fed on a diet consisting of a mixture of pure proteins, carbohydrates, fats, and inorganic salts, failed to grow, but. on the addition of a very small amount of certain substances obtained from milk growth was normal. Hopkins found that the fraction obtained from a phosphotungstic acid precipitation of proteid-free milk contained the active substance and gave excellent growth results. As a result of his experiments he states that “the presence of minute traces of certain * Osborne and Mendel, Carnegie Institution Publication No. 156, Parts I and II, 1911. + F. G. Hopkins, ‘Journ. Physiol.,’ vol. 44 (1912). VOL. LXXXVIII.—B. aly 238 Prof. W. B. Bottomley. organic substances are, without doubt, essential for the proper nutrition of growing animals.” Very little is known as to the nature and composition of these substances. Unfortunately, the active substance appears to be largely destroyed by chemical manipulations, and it is difficult to obtain sufficient to study its chemical constitution and properties. Funk,* by a complex fractionation of the phosphotungstic precipitate of anti-beri-beri substance, succeeded in isolating a substance, melting at 233° C., which in amounts of 0:02 to 0:04 grm. cured polyneuritis in pigeons. This substance he considered to be of the nature of a pyrimidine base. Hopkins, however, states that the additions in his growth experiments were free from amino-acids, purine and pyrimidine bases. It is possible that these substances belong to a new group of nitrogenous compounds, which exist only in small amounts in food materials, but are so extremely active that minute quantities are sufficient to supply the needs of the organism. Although these substances have been found to occur chiefly in plants, there is no record of any investigations concerning the part, if any, they play in the metabolism of the plant itself. During the summer of last year (1913) a number of experiments were made at the Royal Gardens, Kew, on a series of plants, to test the manurial value of Sphagnum peat which had been incubated with a mixed culture of aérobice soil organisms for a fortnight at a temperature of 26° C. It had been discovered that by this bacterial treatment the humic acid in the peat is converted into soluble humates, and this bacterised peat, after sterilisation, forms an excellent medium for the growth and distribution of nitrogen-fixing organisms. As the experiments progressed it was evident that, in addition to the ordinary plant-food constituents, there was present in the bacterised peat a substance which stimulated growth in a remarkable manner. Further experiments showed that this substance was soluble in water, and was effective in very small quantities. Dr. Rosenheim, of King’s College, found that seedlings of Primula malacoides potted up in loam, leaf-mould and sand, and treated twice with a water extract of only 0:18 grm. of bacterised peat, were, after six weeks’ growth, double the size of similar untreated plants, and it was noted that flower production and root development were promoted equally with increase of foliage. These results suggested that the growth-stimulating action of the bacterised peat might be due to the presence of a substance or substances similar in nature to the accessory food bodies concerned in animal nutrition. * C, Funk, ‘Journ. Physiol., vol. 45 (1912-1913). j f 1 Some Accessory Factors in Plant Growth and Nutrition. 239 These accessory substances essential to animal nutrition are known to be soluble in water and alcohol, and, in order to ascertain as rapidly as possible whether there are present in the bacterised peat such water- and alcohol- soluble substances which have a similar effect on plant growth, an experiment | was made to test the effect of an aqueous extract of the alcohol-soluble material of the bacterised peat on the growth and fixation of nitrogen by Azotobacter chroococcum. The bacterised peat was extracted with absolute alcohol in a shaking machine for three hours, and the extract evaporated to dryness wn vacuo. The residue was taken up in warm distilled water, the liquid filtered, and the clear filtrate diluted until it contained the extract of 10 grm. of peat per litre. Portions consisting of 100 c.c. of this liquid were then transferred to each of 12 conical flasks, and the contents of six of the flasks boiled briskly over a Bunsen burner for five minutes; 100 c.c. portions of distilled water were placed in each of six similar flasks; to each of the 18 flasks of the series were added 1 grm. mannite, 0°2 erm. K2HPO,, 0°02 grm. MgSO,, and 0:2 grm. CaCOs, and each was inoculated with 1 ¢.c. of a uniform suspension of Azotobacter chroococcum. The contents of two flasks from each of the three series of six were analysed at once to serve as controls, while the remaining four of each series were incubated for eight days at 26° C., at the end of which period they were analysed by the Kjeldahl process for their nitrogen content. The results are given in the following Table :— Table I. Series. Nitrogen content. EUROS AUCEO EUR i) fixation. fixation. mgrm. mgrm. mgrm. I. Complete food ............ 1. Control 0-4 Mean, 2. 0°4 f 0°4 mgrm. 3. Culture 4°6 42 4, 4 °4, 4-0 3 5. 3-6 3-2 278 6. 44; 4:0 II. Complete food +alcoholic | 1. Control 2°6 Mean, extract of bacterised peat | 2. 2°3 J 2°5 mgrm. 3. Culture 20°7 18 °2 4. 20 ‘5 18-0 : 5. 199 17-4 380 6. 20°9 18 *4 III. Complete food + boiled | 1. Control 2:3 Mean, alcoholic extract of bacte- | 2. 2°5 [ 2°4 mgrm. rised peat 3. Culture 19°6 17-2 ‘ 4. 19 0 16 °6 E 5. 20 °6 18-2 ee 6. 19 *4 17 ‘0 240 Prof. W. B. Bottomley. The more rapid growth of the organism in Series II and III was rendered apparent by the fact that a scum was visible on the surface of the liquid in each flask of these series after 24 hours, while the pellicle formed in Series I only after the lapse of 72 hours. The results obtained indicated clearly that there is present in the bacterised peat a substance which stimulates plant growth, and that this substance is of a fairly stable nature is shown by the fact that almost as good results were obtained with the extract which had been boiled for five minutes as with the unboiled extract. In order to test whether the active substance is present as such in the original peat, or whether it is produced in the bacterised peat as a result of treatment, an extract of the raw peat was made in precisely the same manner and in the same concentration as described for the bacterised peat. Two series of cultures were prepared, the one containing complete food substances in distilled water, the other complete food in alcoholic extract of raw peat. The controls were analysed at once, while the cultures were incubated for eight days at 26° C. as before. No increased growth was apparent in the cultures containing alcoholic extract of raw peat, while the results of analysis, as given below, indicate the absence of any stimulating substance :— Table IT. | Series. Nitrogen content. Berens Mean nitrogen xation. fixation. mgrm. mgrm. mgrm. I. Complete food ............... 1. Control 0°3 Mean, ; 2. 0-4 { 0°4 mgrm. 3. Culture 3°8 3°4 4, 3°9 3°5 ae 5. 3°8 34 6. 4:0 3°6 II. Complete food+alcoholic | 1. Control 2-4 Mean, extract of raw peat 2. 2°8 } 2°6 mgrm 3. Culture 4:0 1°4 4., 4°6 2°0 19 5. 46 2°0 6. 4°8 2°2 The active substance is evidently produced in the bacterised peat as a result of treatment, and since this treatment consists essentially in the production of soluble humates by bacterial action, a test was made to ascertain whether the chemical production of soluble humates would be equally effective. Two equal portions of raw peat were saturated with solutions containing 1 per cent. of their weight of sodium carbonate, and were stirred at frequent intervals for Some Accessory Factors in Plant Growth and Nutrition. 241 several hours. One portion was allowed to dry slowly at room temperature, an alcoholic extract taken and evaporated in vacuo as before, the residue being made up in aqueous solution to a concentration of 10 grm. of carbonated peat per litre. The other portion was leached with water until the washings were © colourless, the liquid filtered, and the aqueous extract thus obtained was evaporated im vacuo to dryness. The residue was extracted with alcohol, and the alcoholic solution again evaporated to dryness in vacuo, the residue being taken up with water, filtered, and the clear filtrate diluted to the proportion of the extract of 10 grm. of the original peat per litre. The effect of both these extracts was tested on Azotobacter, three series of cultures being incubated—one containing complete food in distilled water, the second complete food in alcoholic extract of carbonated peat, and the third complete food in alcoholic extract of water-soluble substances from carbonated peat. Again no appreciable effect was observed on the growth of the cultures, while the results of the analyses as given below failed to reveal any stimulation of the organism :— Table III. Series. Nitrogen content. ieogen Moor eeeen xation. fixation. megrm. mgrm. mgrm. TI. Complete food............... 1. Control 0°4 Mean, 2. , 0°5 J 0°S mgrm. 3. Culture 4°8 4°3 4, 4:0 3°5 . 5. 44 3-9 430 6. 48 4°3 II. Complete food +alcoholic | 1. Control 2:2 Mean, extract of carbonated | 2. 2:2 { 2°2 mgrm. | peat 3. Culture 4°6 2-4 | 4A, 5°0 2°8 2°4 5. 4-4 2-2 6. 4°6 2°4 III. Complete food + alcoholic | 1. Control 1:°9 Mean, extract of water-soluble | 2. 1°7J1°8 mgrm. substances from carbon- | 38. Culture 4°4 : 2°6 ated peat 4, 4,°2 2°4 2-7 5. 4°6 2°8 | Ge 5-0 3°2 The results thus far obtained tend to prove that the active stimulant of plant growth which is present in bacterised peat does not exist as such in the raw peat, nor can it be liberated by a chemical production of soluble humates. It has been obtained only as a result of bacterial action. Cooper and Funk* in 1911 showed that their curative substance was * Cooper and Funk, ‘ Lancet,’ p. 1267 (1911). 242 Prof. W. B, Bottomley. entirely precipitated by phosphotungstic acid from an aqueous solution of the dry residue from the alcoholic extract of rice polishings, and Hopkins* also states that he obtained the best results upon growing rats with the fraction from the crude phosphotungstic acid precipitate of protein-free milk. Conse- quently an experiment was made to determine how far the phosphotungstic acid fraction of the bacterised peat extract was effective in stimulating plant growth. The bacterised peat was extracted with absolute alcohol as described above, and the alcohol evaporated off 1m vacuo. The residue was taken up in water, filtered, and to the filtrate sulphuric acid was added, until the concen- tration of the latter reached 5 per cent. A slight precipitate of humic acid was filtered off, and to the filtrate an excess of 30-per-cent. solution of phosphotungstic acid was added. The whole was then left to stand overnight, when the liquid was decanted off through a filter, the precipitate repeatedly washed with a 5-per-cent. solution of sulphuric acid, and finally decomposed with an excess of baryta. The liquid was filtered off from the precipitate of barium phosphotungstate, and the filtrate, freed from the last traces of baryta . by means of a very dilute solution of sulphuric acid, was evaporated to dryness in vacuo. From 7 kgrm. of bacterised peat the amount of dry substance obtained from the phosphotungstic acid fraction amounted to 12:0096 grm., and since this was made up for experimental purposes into a solution containing the fraction from 10 grm. of peat per litre, the proportion of the dry phosphotungstic acid fraction in the final solution employed consisted of 17 parts per million. This fraction was tested upon wheat- seedlings in conjunction with Detmer’s complete food solution. Ten seeds were germinated in clean sand in each of nine pots, which were arranged in three series of three pots each. Series I was treated with a complete food solution, Series II with complete food plus alcoholic extract from 10 grm. of peat per litre of solution, and Series III with complete food plus phospho- tungstic fraction from 10 grm. of peat per litre of solution. The food solution employed contained nitrogen, phosphorus and potash, estimated as NH, P20;, and K,O in the proportion of 400, 200, and 1220 parts per million: respectively, so that in addition Series III had 17 parts per million of dry substance obtained from the phosphotungstic fraction. Hach pot was treated with 100 c.c. of its solution one week after sowing the seed, and the treatment repeated once weekly for five weeks, at the end of which period the plants were uprooted, washed, dried, and weighed. The results were as follows :— * Hopkins, ‘ Brit. Med. Journ.,’ vol. 2, p. 463 (1913). Some Accessory Factors in Plant Growth and Nutrition. 248 Table IV. Series. Weight of 30 plants. | Increase over Series I. grm. per cent. I. Complete food solution .................0008 11-94 = Ii. = 5 » +talcoholic extract 14°46 21°1 IL ) 3 » +phosphotungstic 15°45 29 °4 fraction | The results thus obtained indicate that the stimulative substance in bacterised peat is precipitated by phosphotungstic acid, and that this phosphotungstic fraction is quite as effective as the original alcoholic extract of the peat. Funk* found that, upon further fractionation of his phospho- tungstic acid precipitate with silver nitrate and baryta and, elimination of the reagents, he obtained a relatively pure crystalline substance, to which he gave the name “ vitamine,” and this he considered to be the specific curative substance. In order to determine how far the growth stimulant in bacterised peat resembled these so-called “vitamines,’ a further fractionation was carried out along the lines described in his paper. The phosphotungstic acid precipitate was decomposed, as before described, with baryta, and the last traces of baryta eliminated by means of sulphuric acid. To therfiltrate from the barium salt silver nitrate was first added, and then baryta, until no further precipitate was produced. The brownish precipitate was filtered off, well washed, suspended in dilute sulphuric acid, and decomposed with sulphuretted hydrogen. The filtrate from the silver sulphide was then exactly neutralised with baryta, the clear liquid filtered off from the ‘precipitate of barium sulphate, and evaporated to dryness in vacuo. The weight of dry substance obtained from the silver fraction from 7 kgrm. of bacterised peat amounted to 0°2452 grm., and, since this also was made up for experiment into a solution containing the silver fraction from 10 grm. of peat per litre, this solution contained the dry substance from the silver fraction in the proportion of 0°35 part per million. This fraction was also tested, concurrently with the phosphotungstic acid fraction, upon wheat seedlings; 15 seeds were germinated in clean sand in each of nine pots, which were arranged in three series of three each. Series I was treated with complete food solution, containing nitrogen, phosphorus, and potash, estimated as NH3, P20;, and K,O, in the proportion of 400, 200, and 1220 parts per million respectively. Series II was treated with a similar solution, containing in addition 17 parts per million of the phosphotungstic fraction, * Funk, ‘Journ. Physiol.’ vol. 45 (1912-1913). TAA: Prof. W. B. Bottomley. and Series III with the complete food solution + 0°35 part per million of the silver fraction. The pots were first treated one week after sowing the seed, and after that each pot received once weekly 100 c.c. of its food solution for seven weeks. At the end of that period the plants were washed, dried, and weighed, and, after the gross weight had been taken, the plants were all dried in the steam oven at 100° C. until their weight was constant. The results are as follows :— Table V. SEriee! Gross weight, | Increase over ' Dry Increase 45 plants. Series I. weight. over I. grm. per cent. grm. per cent. I, Commolene 10006! coocosoegosesnnosnononsn 64.°5 — 13 +3480 — Il. se » + phosphotungstic 96 ‘8 50-0 16 -3818 22-7. fraction JOLIE, se » +Ssilver fraction ... 96 °5 49 -6 15 °7148 Wi 27 The silver fraction from the bacterised peat extract, corresponding with the “vitamine” fraction of Funk, having thus given results approaching those of the phosphotungstic fraction, a preliminary investigation was made to test its effect on the growth of wheat seedlings in water culture. Two sets, each consisting of 18 similar seedlings, were carefully selected, each set being originally of equal weight, viz., 473 grm. Hach set was divided for purposes of water culture among three similar bottles of 200 c.c. capacity, six plants being inserted through notches in the corks of each bottle, so that the roots dipped into the culture solution. The three bottles of set I were filled with a nutrient solution of pure salts in physiologically pure distilled water, in which the proportions of NH3, P20; and KO were 400, 200, and 1220 parts per million respectively; while those of set II con- tained a precisely similar solution which had received, in addition, 0°35 part per million of the silver fraction of bacterised peat extract. The bottles were aérated daily, and the solutions changed twice a week, while at the end of every 16 or 17 days the plants were taken from the Jars, moisture removed from their roots by means of blotting paper, and weighed. The results obtained are shown in Table VI. The change brought about by the addition of the silver fraction is represented by the accompanying curves, in which the dotted line represents the change in weight of the series in pure food, while the unbroken line shows the progressive increase in weight obtained upon the addition of this substance. Some Accessory Factors in Plant Growth and Nutrition. 245 Table VI. Series. I. Pure food solution ............... II. Pure food solution + silver fraction from bacterised peat Weight in grammes Qu ie) is 2) ve ° IO Weight of set of 18 plants. grm. Original weight......... 4°73 After 16 days............ 5°42 After further 17 days 5-29 ” ” ” 4°33 Original weight......... 4°73 After 16 days............ 5 57 After further 17 days 6°65 7°33 ” ”? ) 20 30 Time in days Percentage increase on original weight. per cent. ACRES © Up to a certain point the two series of plants increased in weight to an almost equal extent, but beyond this point the seedlings growing in pure food solution appeared to be unable to utilise the food elements supplied to them; a condition which was apparently corrected by the addition of the silver fraction. Experimenting with guinea-pigs in 1909, Fiirst* demonstrated that seeds of barley, oats, peas and flax contained no curative substances for scurvy, but that during the germination of these seeds anti-scorbutic substances developed, which were quite as effective as extracts from green vegetables. * Fiirst, ‘Zeitschr. f. Hyg. u. Infekt.,’ vol. 72, p. 121. 246 Prof. W. B. Bottomley. These facts indicate the possibility of the development, during germination, of special growth substances which enable the young embryo to utilise the food material present in the seed. If this is so, the removal of the source of these growth stimulants by the cutting off of the seed as soon as possible after germination should render the effect of an addition of such substances in the food solution all the more marked. In order to test this hypothesis, two series of wheat seedlings, similar to those used above, but in a rather younger stage, were taken, and before the removal of their seeds the two sets were of equal weight, viz.: 3:97 grm. Their seeds were carefully removed, injury to the plants being avoided, and after this process the two sets weighed respectively 3°2 and 3°17 grm. These were treated in precisely the same manner as before, the first set being given complete food salts, and the second food salts with the addition of the silver fraction from bacterised peat. The weights of the two sets at various dates are shown in the following Table :— Table VII. Series. Weight of set of 18 plants. Rencen tae ce sepia: | grm. per cent. I. Complete food solution ............ Original weight ......... 3°2 — After 16 days ......... 3°37 5°3 After further 17 days 3°20 0-0 ” ” ” 2°85 —10°9 II. Complete food solution + silver | Original weight......... 3°17 = fraction from bacterised peat | After 16 days ......... 3°63 14°65 After further 17 days 4°29 35 3 ” ” ” 5 05 59 °3 The following diagram shows the variation in weight of the seedlings throughout the experiment, the dotted curve representing the series in pure food, while the unbroken curve shows the effect of the addition of the silver fraction. These results indicate that during the germination of wheat seeds certain substances are formed which enable the young embryo to utilise the food materials present. The supply of these substances formed by the seed during germination is sufficient to establish the embryo as an independent seedling, then some other source is necessary. It has been shown that these accessory food substances are produced when peat—decayed vegetable matter—is acted upon by certain soil bacteria, and the natural inference is that during the bacterial decomposition of organic matter in the soil, that is, during humus formation, these substances are formed, hence the beneficial effect on crops of Some Accessory Factors in Plant Growth and Nutrition. 247 farmyard and other organic manures. The specific action of these accessory substances is not known. They may be concerned in the metabolism of Weight in grammes |0) 10... 20 O O O Time in aes ai 2 phosphorus, they may act as catalytic agents, or may be a definite constituent of plant food—a “ bau-stein.” Experiments to test these various hypotheses are in progress. In conclusion I wish to acknowledge my indebtedness to Miss F. A. Mockeridge, B.Sc., for her valuable help in the chemical part of this investi- gation. 248 Further Observations on the Changes in the Breathing and the Blood at Various High Altitudes. By Manet Pureroy FirzGERa.p. (Communicated by J. 8. Haldane, F.R.S. Received June 2,Read June 25, 1914.) In a previous investigation carried out in connection with the Anglo- American Pike’s Peak Expedition (1911), the changes in the breathing and . the blood at high altitudes were recorded at atmospheric pressures ranging from 625 to 458 mm. of mercury. Lack of time prevented further observations being made, and in the graphic representation of the gases of the alveolar air and of the percentage of hemoglobin in the blood subsequently published,* the supposed values for atmospheric pressures ranging from 625 to 760 mm. of mercury were indicated by a broken line. To complete the records, experiments were made by me in North Carolina, U.S.A., during the months of July, August, and September of 1913. Three localities were chosen in the Southern Appalachian chain, approximately between 35° and 35° 6’ N. latitude, and 82° 5’ and 83° 25’ W. longitude: Highlands (altitude 3850 feet), the highest village east of the Rocky Mountains, situated in the Blue Ridge Mountains; Waynesville (altitude 2645 feet) in the Balsam Mountains, and Asheville (altitude 2210 feet) situated in the valley of the French Broad River, with the Blue Ridge Mountains lying to the south and east, and the foot-hills of the Unaka Mountains to the west and north. Experiments were made with 43 residents. Care was again taken to exclude the unhealthy, and those who, on account of recent change of abode, might be unacclimatised. Observations were also made on myself at each locality visited. The research is based upon 206 CO, determinations and 52 hemoglobin percentage determinations. The subjects were adult men and women, of from 18 to 70 years of age. The number of subjects corresponding to each decade, or part thereof, were as follows :— Age. Number. Between 15 and 19 years ......... 8 3 2045. 229 eal eta setae 15 ;; BO io BOW RE aera 14 5 AD jee AD Se cleat sscocmcenes 1 3 5 Oe. : Ome ae meade 2 s 60 fo vines iectesaee 4 * Phil. Trans.,’ B, vol. 203, pp. 351-371. Changes in Breathing and Blood at High Altitudes. 249 In the majority of cases, the subjects were natives of the respective localities. With the exception of two subjects, one of whom had been at sea-level three weeks prior to the experiment, and the other, one month before, at an altitude of about 1000 feet, no subject had left the place at which he or she was living for a considerable time. The methods of determining the alveolar CO, percentage, and of calculating the alveolar CO2 and O2 pressures, were the same as in the previous investiga- tion. A carefully standardised Gowers-Haldane hemoglobinometer was used for determining the hemoglobin percentage, and pure CO was used for saturating the blood solution. At Asheville and New York the readings for the barometric pressure were obtained from the local offices of the United States Weather Bureau; at Highlands and Waynesville readings were taken from an aneroid barometer compensated for temperature and checked by comparison with the readings of the Weather Bureau. The altitude records were taken from the bench marks of the United States Geological Survey, with the exception of that for Highlands, in which case the elevation recorded (3850 feet) is for the Highlands Camp Sanatorium, where the experiments were made, and not for the village proper. The mean values for the hemoglobin percentage, the alveolar CO. percentage and pressure, and the calculated alveolar oxygen values for men, at altitudes varying from 658 to 711 mm. of mercury are given in Table I, and similar values for women in Table II. Normal mean values for near sea-level (Oxford) are included for comparison in each table. é The results obtained in the present and the former investigation are indicated graphically in Charts I and II. It will be seen that taken in conjunction with the barometric pressure the values for the alveolar CO2 and O2 pressures decrease progressively with the increase of altitude. Since the CO2 values, with the exception of those for Waynesville, correspond closely with the supposed values indicated by the graph in the preceding paper* additional support is given to the statement then made, that the lowering of the CO2 pressure is in direct proportion to the diminution of the barometric pressure and amounts to about 4°2 mm. or 10°5 per cent. of the sea-level value for each 100 mm. of diminution of barometric pressure. There is a corresponding progressive fall in the oxygen pressure of about 16 mm., or 16 per cent. of the sea-level value for each 100 mm. fall in the barometric pressure. In each of the two charts a curve is also plotted, as in the former paper, * Ibid., Chart 1. Further Observations on the Miss M. P. FitzGerald. 250 ‘(1061) gos “4 ‘9g ‘Joa ,“Jorskyq “Bano, “ouBpreHT oag + (GO6T) 98F “d ‘ge ‘joa ,“jorskyg ‘uanog , ‘ourp(vyyT pur preteyzqiy cag y [P29] 96 18 SIT |40-00T| 96-66 9Z- FL O€T LE 9.38 | &- PP | G- 6E 6g. SPL 006 -BO8 BOT y“PIOFXO V 86 GIL | 8: 901 6: 6 GO: FL 81 9 g.sg | 4-17 | $-88 | 94-9 ( 1007 OlZ@e jo eT Aoysy ‘M0 G.Q poq0e1200) 6 16 LIT | 9. 80T P. 96 L9. FL OV SI | 9-1Tg | & 48] 0-FE | 6.9 969 (peqoerrt00un) 6 16 ZIL | 9- 801 c- 6 L9- VT OV SI | §-1€ | 0-28 | 8-8 | 2-9 169 op9g | oytasoude ay (am110j8URBg L 66 | OIL | ¥- O11] 6-98 %- PL rE OL | 0-1 | &68 | &- FE | 29-9 899 osse | dwep) spuvlqsry ‘SAY “ULuL ‘Ul | “XV | Urey | “O48 9% “ae ‘suoryeurur | “S300! UI | XBW | uBey_| “Usp, *SUOIqVUIUL poyemyes | relooaye “1099 “ang “118 -19j0p aeaxepooaye| xp utr XBTOOATB jo ut oansseid | osequoored | “suoryeurTmIej0p ‘) LE Aap ut ‘omssord | og ay xroquin jy "99009010 q MBOUL uBouL jo pue sqool 4B poyeinyzes IIe esequeo | o1ajzouL01eq ais : “fqI[woorT peyenoleg | poyznoyep | -qns Jo zoqumyy | azBpooaye ul cansserq | -10g Uva, pugnly “Urqo[souas xe) “00 “UO, OF pourejyqg synsey uve —T qe, 251 des. gh Altitu a d Blood at H mg an Breath ges mm Chan ‘(1O6T) 98% “d ‘9g “104 ,“ Torsha ‘uanoe, “ouepreyy eeg 4 (S061) 987 d ‘Ze ‘Joa “Joskyg ‘uanop, ‘ouspley pus pletepHzqty 20g [P49] GI 18 96 40-68 9- SOT LL. VT 8&T ce | v-08 | 0-1h | &-98] ZL 6rL 003 =| -B08 teoT 4 *p10FxO, v s6 POL | 8-26 0- OOT 80- ST GS G 6-08 | I-S€ | §-3e | 16-7 ( Or (RG ONES EN “W10 G. QO poqoeat00) 3S S6 66 16 T- 86 06. ST 9 G I-08 | 6-1€ | 0-1g | 18-F | &69 (poq0eat000n) 3S S6 66 16 v- 16 06. ST 9 G 8-66 | 4-18 | 8-08 | 18-7 889 ep9z | eptrasoude (ami10jeu0eg L G-06| $6 | 8-16] 4-06 GL. FI 8% LE | %-2z | 2:92 | 0.c8 | 6L-S 099 osse | dup) spuv[ystH ‘Sq ‘mur “UIA | Xt | UVeW | “0 OLE 1% ee pe ceaes ae WL | XV | UV | “AOTAL *SMOT]BUIUL paqeamyes | aepooare | “721°C “ans “ae -1ayap qreaejosaye | Aap ur | IvpooaTB jo utomssead | odequeored | ‘suoryeuru.1e}op *O Le Aap ut romssord | uy requan Ny -enRyua018 uveul uWeoUl jo puz sqgoofqns qe poyeinges ate =| esequeo | olqgeul0.eq mee : “AqITB00TT peye[noTRQ | pozepnopsg Jo soquin yy IB[OIATe UI oanssaig | -10g uBoyy Pryaly “UIqo[s0ue FT 20 eye) “USULO AA IO poutejqgQ si[nsoy ueoW_—'T] QR], 252 Miss M. P. FitzGerald. Further Observations on the showing the values for the barometric pressure when the mean temperature of the air column between sea-level and the heights indicated is assumed to be 15° C. ‘For the calculated values and graphic representation of the percentage Gas pressure Altitude mm.of 800 750 700 650 600 6550 600 450 400 350 300 250 200 Ft. me, ee ee eo (ee ee 31,000 HY re 29,000 28000 26,000 Cole Sdure A 140 24,000 23.000 22,000 21,000 20.000 19,000 18,000 800 750 700 650 600 550 500 450 400 350 300 250 200 Atmospheric pressure tn Inim. of mercury. Chart I. composition of the alveolar gases at atmospheric pressures ranging from 760 to 250 mm. of mercury the reader is referred to the earlier paper.* The idea was put forward in that papert that at atmospheric pressures greater than 625 mm. of mercury the straight line representing the CO2 tension might perhaps be replaced by a curve, and that this would flatten as * Ibid., pp. 359-360. + Ibid., p. 360. — Changes in Breathing and Blood at High Altitudes. 253 a pressure of 760 mm. of mercury was approached, and continue as a straight line parallel to the abscissee with increased pressure. If such alteration oceurs, it is evident that it can only be at pressures higher than 760 mm. of mercury. From the earlier experiments of Haldane and Priestley,* Hill and Haemoglobin. ° Alt tude oa 800 750 700 650 600 550 600 450 400 350 300 W = 26000 Bae ss caf [|r 170 24000 ‘ yal 160 = Y 22000 a 21000 150 | | 20,000 19.000 fe be ; 140 18000 | fly 130 2 16000 lag Lio E fave 120 4 fi 14,000 13.000 110 100 cee ia 90 ds : iz 70 6 50 QO 700 300 Prisha Bh, dea in mm Be Ae. Chart II. 12,000 11000 10000 Le ape tL al iL a 9.000 i. 8,000 7,000 6,000 9,000 al 4000 3.000 ° Greenwood,f and Haldane and Boycott, we know that on short exposure to increased pressure the alveolar CO2 remained at normal value, even up to * ‘Journ. Physiol ,’ vol. 32, p. 225 (1905). + ‘Roy. Soc. Proc.,’ vol. 70, p. 455 (1906). { ‘Journ. Physiol.,’ vol. 37, Nos. 5-6, p. 355 (1908). VOL. LXXXVIII.—B. ‘254 Miss M. P. FitzGerald. Further Observations on the seven atmospheres in the case of Greenwood. With long exposures, however, the result may well be different. The percentage of hemoglobin in the blood in men increased progressively with the fall in barometric pressure. The mean percentage values correspond closely with the supposed values at similar pressures indicated in Chart III of the previous paper* and support the statement there madet that “for every 100 mm. fall of atmospheric pressure, there is an average rise of about 10 per cent. in the hemoglobin.” Similar, but less regular, increase occurred in the percentage of hemoglobin in the blood of women. An unusually low value {91°8 per cent.) was recorded at Highlands. Insufficient data together with varying physiological condition may account for the irregularity in the records for women. At each locality greater uniformity in the individual determinations was met with than in the previous investigation at the higher altitudes. At Waynesville and Asheville, in four of the five men subjects aged between 52 and 64 years, the hemoglobin percentage was less than 100, which possibly indicates that the compensatory increase of hemoglobin in the blood wanes with age at the less high altitudes. With regard to the influence of age on the fall in the alveolar COz pressure, the tendency previously noticed{ for the fall to be less in those under 30 than at later age periods was again observed. Only one subject complained of the ill-effects of living at high altitudes (5850 feet). A marked difference was observed in the degree of “nervousness” in the subjects, this being much less than in residents at altitudes of 5000 feet and over. Observations were made upon myself at each locality (see Table III); 24 hours were allowed to elapse before the alveolar CO, determinations were made. My mean normal alveolar CO, pressure at Oxford and at New York is 34 mm. of mercury (barometric pressures respectively 753 and 759 mm. of mercury). The journey of about 33 hours to Highlands (altitude 3850 feet) was made direct from New York, where the previous nine months had been spent. At Highlands, a stay of over eight weeks was made. During the last two and a half weeks I slept at a point about 400 feet higher than the sanatorium, where work was conducted during the day, and through the eighth week was more or less stationary at the higher altitude (4250 feet). Contrary to the former experience at altitudes of 5000 feet and higher, the alveolar COz pressure did not fall, but remained, with slight variations, at the normal value for sea-level (34 mm. of mercury). In consequence, the alveolar * Ibid , p. 362. + Ibid., p. 361. » Lbed., p. 365. a. 255 Changes in Breathing and Blood at High Altitudes. ‘(G06T) 98F “4 “2g “Joa “ors “Wanof , “oULP[eAT PU’ PIMlEHzITY 9g x | | yO x v 148 | 26 | 68 €-80I | &&-ST ra I- 8 | 9-98 | 1-58 | 64-4 6S, [PAeT-wog | May “Jaaey-tag = ai 68 G- LOT 6I. ST I[QV[LVAB | Jou sp/oooy | QO. FE | 18-7 SSL AOS " DLOFXO T ae a 96 &- 16 &L. FL 8 L-&& | 8: PE | FFE |} O6-G 802 OT6Z | of laoysy :; (‘wo ¢. 0 peqat100) T s Tae &6 8: &6 OS: FL v L.P& | T-SE | 6 PE | 6&9 69 (peqoeat00un) I — i &6 T- &6 OS: FI V F-VE | 8-0 | 9- PE | GES 689 GROG eee o|[tasouds Wy (wnt0zLUutg g 16 86 40) 6- 18 | 86. VL 1& €-§§ | 9-G8 | &. 7s | LS-S 9, £99 osse | dup) spurpysry SET ‘wur | | Ul | XV | Uveyy ‘uve, | “UROL ‘ULL | XB | UBey| “UveT Die | Sy perocties ao | te Tee raramerep xajooaye eee gD jo TR[OOATL UT oo = “jo ‘O LE Aap ut ‘ounssotd 4 Vann | “on RyUdole T ammssord sae i =o Taq uN Ny qe poqwanges aie | aduquoa | ongautoanq | 72°F WE “AgI[vo0'y | peqenoye9 ACA) AB[OIATC UL aINsso'g | -lag UvdIAT STAY “uIqo[souta 20 “00 . oe Bl ol JAN CRON! DS TONNERIAO SHIRE WAC A= IIE SUED eee 256 Miss M. P. FitzGerald. Further Observations on the oxygen pressure was lower than it otherwise would have been. In general I was not conscious of being at a higher altitude than usual, but signs of being at a physiological disadvantage, due to the low alveolar oxygen pressure, were manifested by the constant feeling of great fatigue and, except during the stay at 4250 feet, when work was in part lessened, by poor and unrefreshing sleep.* From Highlands the journey was made via Lake Toxaway (seven hours’ drive and thence by rail) to Waynesville (altitude 2645 feet), one night being spent on the way at Asheville (altitude 2210 feet). Four days were spent at Waynesville (2645 feet). Stormy weather prevailed. The mean alveolar CO: pressure was found to be 34:6 mm. of mercury with a barometric pressure of 689 mm. of mercury (corrected barometric pressure 694 mm. CO» pressure 34-9 mm. of mercury). At Asheville (altitude 2210 feet), where a week was spent after leaving Waynesville, the mean COs: pressure in the alveolar air was found also to be within the variation of normal sea-level values, t.e. 34-4 mm. of mercury (barometric pressure 708 mm. of mercury). The return to sea-level (New York) was made direct from Asheville, a journey of 22 hours and involving a change of altitude of slightly over two thousand feet. Fourteen hours after arrival, the alveolar CO, pressure was found to be 33:1 (barometric pressure 766 mm. of mercury), a slightly lower value than usual. On the third day at sea-level the CO» pressure was 34:0 mm. of mereury. The values obtained during the subsequent three weeks varied from 33:2 mm. of mercury to 35°6 mm. of mercury, the normal mean value obtained for New York being 34:1 mm. (mean barometric pressure 759 mm. of mercury). Thus at altitudes ranging from 4000 feet to 2000 feet, and at barometric pressures ranging from 663 to 708 mm. of mercury, there appears to be no respiratory reaction in M. P. F. G. Entirely different behaviour of the respiratory centre was therefore met with below 4000 and above 5000 feet, for in the latter experience (5000-14,000 feet), although the response of the respiratory centre to want of oxygen was slow, it was nevertheless apparent. From the present series of experiments, and in spite of a stay of eight weeks at 3850 feet, the alveolar CO pressure remained at sea-level value (34 mm. of mercury) at barometric pressures ranging from 759 to 663 mm. of mercury. The sea-level value of 34 mm. of mercury falls within the lower limits of * In spite of this, however, there was a general improvement in condition as evidenced by increase in bodily weight. Excellent food was provided at the sanatorium, and the daily consumption of food was greater than usual. It must be borne in mind that fatigue was easily produced since I had been weakened during the previous winter by a chronic Staphylococcus infection. + Ibid., p. 367. Changes in Breathing and Blood at High Altitudes, 257 alveolar CO, pressure observed for normal women. Whether the fact of my COs pressure remaining at the normal value in spite of the barometric pressure varying from 759 to 663 mm. of mercury is an idiosyncrasy, or an indication that with persons in whom the alveolar CO, pressure is naturally low a | marked decrease of barometric pressure (i.e. more than 100 mm. of mercury) is required to produce a lowering of the threshold value of the COs pressure, can only be determined by further experiments. In contrast to the absence of change in the alveolar COs, the hemoglobin in M. P. F. G. rose, as before,* with decreased barometric pressure and fell as sea-level pressure was approached. From an initial value of 89 per cent. at New York, the hemoglobin had after three days at Highlands (altitude 3850 feet ; barometric pressure 668 mm. of mercury) risen to 96 per cent. It oscillated during the following four weeks between 91 and 93 per cent. and was 98 per cent. during the eighth week, after a fortnight had been spent at an altitude of 4200 feet. At Waynesville (altitude 2645 feet) it fell to 93 per cent., and was again 96 per cent. at Asheville (altitude 2210 feet and barometric pressure 708 mm. of mercury). Three days after the return to New York a lower value than usual was recorded, 87 per cent. The hemo- globin reached 92 per cent. a few days later, and then fell to the normal value of 89 per cent. Conclusions. 1. In persons acclimatised at altitudes up to 3850 feet, the partial pressure of COz2 in the air of the lung alveoli is invariably lower than at sea-level, so that the lung ventilation is correspondingly increased. The results of the present investigation are in accord with those obtained with persons acclima- tised at altitudes of 5000 to 14,000 feet, and support the conclusion previously published that “the lowering of the COs. pressure is in direct proportion to the diminution of the barometric pressure, and amounts to about 4:2 mm. or 10°5 per cent. of the sea-level value for each 100 mm. of diminution of barometric pressure.” 2. It is again found that in women, as at sea-level, the alveolar COz2 pressure is about 3 mm. lower than in men. _3. As at higher altitudes, in persons acclimatised at altitudes up to 3850 feet, the percentage of hemoglobin in the blood is increased. The present observations support the view previously published that “for every 100 mm. fall of atmospheric pressure the percentage of hemoglobin in the blood is increased by about 10 per cent. of the normal value for men at sea-level.” * [bid., p. 369. 258 Mr. A. Compton. Optimum Temperature of an In women, as at sea-level, the values are about 11 per cent. lower than for men, but greater irregularity is observed. Graphic representations and tables of the results are given. To render possible a complete survey of the alveolar gas pressures and the hemoglobin percentages recorded for acclimatised persons at varying atmospheric pressures and heights above sea-level, the values previously published* are included in the graphs. : . In conclusion, I wish to express my cordial appreciation of the kind help and hospitality received during the investigation. My thanks are specially due to Dr. Mary Lapham, of Highlands, Dr. Stokes and Mr. J. Tull, of Waynesville, and to Dr. George Purefoy and Messrs. Taylor and Johnstone (U.S. Weather Bureau), of Asheville. My sincere thanks are also due to Dr. J. S. Haldane for his advice and for the loan of standardised instruments, and to Prof. Yandell Henderson, of Yale University, for the further loan of apparatus. Constancy of the Optimum Temperature of an Enzyme under Varying Concentrations of Substrate and of Enzyme. By Artuur Compton, Imperial Cancer Research Fund. (Communicated by Sir J. R. Bradford, K.C.M.G., Sec. R.S. Received June 10,—Read June 25, 1914.) In a recent paper} a new enzymic relation is recorded. For the enzymic hydrolysis of salicin—by the enzyme which Gabriel Bertrand and the author} have named salicinase—it is found that, in an action of fixed duration,§ the temperature of greatest activity of the ferment is always the same, whatever — the dilutions of substrate and of enzyme adopted for the determination. In other words, the duration of the action being constant, the optimum tem- perature of the ferment is independent of the concentration both of the substrate and of the enzyme. The observation is suggestive: if true of one enzyme it may be true of all, and possibly becomes the enunciation of a general law. Herein, for the moment, lies its main interest. * Phil. Trans.,’ B, vol. 203, pp. 351-371. + Arthur Compton, ‘ Roy. Soc. Proc.,’ B, vol. 87, p. 245 (1914). { Gabriel Bertrand and A. Compton, ‘Comptes Rendus,’ vol. 157, p. 797 (1918). § For the variation of the optimum temperature of an enzyme with the duration of the enzyme action, see Gabriel Bertrand and A. Compton, ‘Comptes Rendus,’ vol. 152, p. 1518 (1911) ; ‘Ann. Inst. Past.,’ vol. 26, p. 161 (1912). Enzyme under Varying Concentrations. — 259 In the present paper further experimental evidence for this hypothesis is given, in the case of another hydrolytic enzyme, the maltase of Aspergillus. oryze (taka-diastase). . For the extract of Aspergillus oryze used, the Imperial Cancer Research’ Fund is indebted to Messrs. Parke, Davis and: Co., who placed at my» disposal one of their most active preparations. This preparation, after bemg freed from insoluble constituents and purified by a technique to be detailed elsewhere, consists of a white powder, entirely soluble in water, whose activity in maltase is double that of the original preparation. The maltose used was Kahlbaum’s. It was purified by successive recrystallisations from watér, the mother liquor impurities being removed after each recrystallisation by pressing the crystals in an hydraulic press between several layers of clean dry linen. Eventually, after powdering in a mortar and drying for about a week im vacuo over sulphuric acid, a specimen of pure maltose, containing one molecule of water of crystallisation, was obtained. It gave an optical activity [«]? = +130-4°, and its reducing power, determined by Bertrand’s method,* was as set out in Table I. ry Table I. Weight of maltose. Weight of copper. mgrm, mgrm., 20-0 21:0 40-0 42-0 60-0 62°0 80 ‘0 83 -0 100 ‘0 103 °5 These numbers, allowing for the molecule of water of crystallisation present, correspond exactly with those given by Bertrand (207d.). That the optimum temperature of the ferment is independent of the concentration of the substrate is shown by the following experiments :—Four series of eight clean Jena glass test-tubes were prepared containing respec- tively 360, 180, 90, and 60 mgrm. of maltose dissolved in 4 cm.* of water which had been specially purified by redistillation under diminished pressure. Then into each tube was introduced in portions of 1 cm. a solution of the enzyme, prepared a half to one hour previously, containing 10 merm. per cm.*. The substrate concentrations in the four series of tubes are M/5, M/10, M/20, and M/30. The tubes, after being closed with clean sterile corks, were plunged into water-baths kept at known temperatures. After * ©Bull. Soc. Chim.,’ (8), vol. 35, p. 1285 (1906). 260 Mr. A. Compton. Optimum Temperature of an od 16 hours’ incubation the tubes were withdrawn, the corks removed, and each rapidly washed with 1 cm.* of water, the washings being carefully added to the contents of the corresponding tube. The tubes were next heated for five minutes in boiling water to stop the enzyme action, they were then cooled, and the contents of each diluted to a known volume, such that 20 em.® of the diluted mixture corresponds to 36 merm. of maltose. The proportion of maltose hydrolysed was estimated by the increase of reducing power as determined by the method of Bertrand (idzd.). The numbers obtained are recorded in Table IT. If the percentage of maltose hydrolysed be plotted against the mean tem- perature of the experiment these numbers give the series of curves represented by fig. 1. Each curve shows a maximum at or about the same point, +47°. Hence, under the conditions of the experiment, the optimum temperature of the ferment is constant, and independent of the variations in the concentration of the substrate. That the optimum temperature is also independent of the concentration of the enzyme is shown by the following experiments :—Four solutions of the enzyme were prepared containing 10, 30, 60 and 100 mgrm. dissolved in 10 em’. 90 90 ® (S) 80 3 TO Co) [e) (ey) [e) is ie) iS Maltose hydrolysed (7%) —~ a fe) Maltose hydrolysed (%) —> S S oO fe) T T 20 20 10 Ior (ea ali) LOMO AOS omicoc’ 10 |) Zo’ P40) op soieugt Temperature —— Temperature > Fia. 1. Fig. 2. Fia. 1.—Substrate concentrations M/5 to M/30. Enzyme concentration 2x 107% grm. per em.* Fig. 2.—Substrate concentration M/20. Enzyme concentrations 2 x 10~ to 20x 10-4 grm. per cm.? Enzyme under Varying Concentrations. Table II. 261 Temperatures at the beginning and end of each experiment. Maltose hydrolysed per cent. with the following substrate concentrations. M/s. M/10. M/20. 60 °3 40 6 Table LIL. Temperatures at the beginning and end of each experiment. Maltose hydrolysed per cent. with the following enzyme concentrations in grammes per cm.* = ~T oy) 4 ~I nr 2x1074, 6x 10-4. 12 x 10-4. 20 x 10-3. 16°9 32 °8 262 Dr. J. Joly. of water, which, after standing from a half to one hour, were introduced in portions of 1 em. into four series of test-tubes containing 90 mgrm. of maltose dissolved in 4 cm.* of water. The concentration of the substrate in this experiment is M/20, while the enzyme concentration varies between 2 x 107* and 20x 107* grm. per cm.%. After 16 hours’ incubation, the action was stopped, and the quantity of maltose hydrolysed in each tube was determined as before. The numbers obtained are set out in Table III. On plotting the percentage of maltose hydrolysed against the mean tem- perature of the experiment the curves of fig. 2 are obtained. _ Here, again, each curve shows a maximum in the same region of temperature, +47°. Consequently, the optimum temperature of the enzyme is independent of the enzyme concentration. Thus it is found, for the maltase of Aspergillus oryze—as for the salicunase of sweet almonds—that the optimum temperature of the ferment is indepen- dent alike of the concentration of the substrate and of the concentration of the enzyme. A Theory of the Action of Rays on Growing Cells. Bye OLYe SC) yakiakins: (Received May 28,—Read June 25, 1914.) The recent accessions to our knowledge of the nature of y- and X-rays bring the treatment, by these rays, of malignant and morbid growths, into continuity with the treatment of lupus, ete., by the Finsen light or by other actinic radiation. The pathological effects of the shorter and more penetrating waves have been described by experienced observers as stimulative of the morbid growth when the administered radiation is feeble in intensity and as inhibitive of growth when the radiation is sufficiently intense. Here there is plainly an effect produced by the short waves upon the growing cell, and the question arises if from this and allied observations we cannot gain some insight into the nature of the activity which characterises the malignant and morbid cell. The well ascertained facts of photo-electricity show that, in all cases, the phenomena of direct light effects classed under that head are ascribable to the expulsion of electrons as a result of the vibratory energy communicated from the ether. The loss of electrons is attended by ionisation of the atomic or molecular systems from which they are derived, the abstraction of the A Theory of the Action of Rays on Growing Cells. 263 negative charge leaving the positively electrified ion behind it. This is the sufficient explanation of many phenomena collected under the name of photo- electricity. It has been ascertained that the velocity of the electron at the moment of its expulsion is the greater the shorter the wave-length of the | radiation concerned. The swift-moving §-rays represent the electronic dis- charge excited by y- and X-radiations. Some years ago I endeavoured to explain the nature of the events taking place in the photographic film in terms of photo-electric activity. The theory has recently been republished and oe by Mr. H. Stanly Allen in his book on photo-eléctricity. According to my view the latent image is formed of molecular systems which have been subjected to loss of electrons and which remain as ions positively charged in presence of these electrons, the nature of the medium being responsible for the maintenance of the static attraction between electron and ion. In development these ions and electrons are discharged, and as a consequence of the chemical reaction thereby effected between the developer and the ionised photo-system the metallic atom is liberated, constituting the visible image. The phenomenon of the reversal of the latent image under excessive light stimulus is well known. On the theory this significant event is explained as the result of the increasing electrostatic stress attending over-exposure, whereby ultimately the resistance to recom- bination breaks down and the original molecular structure is restored. The luminous stimulant will now begin to re-form the latent image. “A succession of such constructive and destructive effects is obviously possible according to the theory, and is, in fact, matter of observation. The theory can be shown to explain the facts respecting the different types of reversal as ascertained by R. W. Wood. Classifying the modes of formation of the latent image as (1) by pressure, (2) by X-rays, (3) by light-shock (very brief flashes, as by light from an electric spark), (4) by lamp light, Wood found that the latent impression produced by any one of these can be reversed by subsequent exposure to any other following it on the list but not by any one preceding it. He found that Becquerel rays (y-rays) behaved like X-rays. For the manner in which Mr. Allen applies the photo-electric theory to these obser- vations I refer to his book. My object in referring to the photo-electric theory of photographic actions is to show that on the assumption that growth in the cell, generally, is attended and conditioned by ionic activity, there is sufficient resemblance between the effects of stimuli on the plate and on the cell to lead to the belief that there must be, physically, much in common between the actions in each case. Prima facie the formation of the normal latent image by 264 Dr. J. Joly. moderate light stimuli is parallel with the stimulation of growth by feeble X- or y-radiation. The photographic reversal by greatly increased illumination compares with the inhibition of growth by the heavy doses of y-radiation now employed in the treatment of cancer. The analogy when further pursued must take account of intrinsic differences prevailing in the two cases. In the living cell there are continuous molecular movements and chemical interchanges accompanied by and attending the ionisation. The static conditions reached in the latent image can only prevail for a brief period which terminates when the ions and electrons find fresh combinations. The image-forming and reversing activities of the plate become respectively represented in the cell by the following events :— (a) Increased liberation of electrons (8-rays) and attendant formation of ions under the y- or X-rays. This increases the metabolism, and, in the case of morbid growths, promotes the evil it is intended to cure. (>) With increasing radiation sudden and excessive electrostatic stress (or over-ionisation) brings about immediate reversion to the original molecular state so that molecular changes and reactions are stopped and metabolism ceases. The maintenance of this condition may lead to complete modification of the cell and ultimately to its absorption by the more stable normal cells which are not so readily influenced by the radiation. An alternative view, less in line with the photo- graphic analogy, is to suppose that, with increased density of electronic radiation emanating from all parts of the tissues, an ion freshly formed in the metabolic substance of the cell is almost instantly neutralised by a B-ray, so that the time required for the molecular movements attending metabolism is not given and growth ceases. In another particular we find the cell behaving in a similar manner to the photo-sensitive plate. Physicians ascribe the origin of malignant growth in certain cases to continued local irritation. Here we have a parallel with the photographic plate; for the latent image, 7.c. the ionisation and electronisation of the film, may be obtained by various mechanical stimuli, such as pressure, friction, ete. The inhibition of the growth so produced in the tissues by y-rays compares with the reversal of the pressure or friction marks of the film by light shock. The selective action exhibited by the morbid cell towards the radiation, so that these cells are soonest affected by the rays, is significant. The thera- peutic value of the rays depends on this action. To what may it be due ? Let us suppose the morbid cell characterised by less stable molecular systems than occur in the normal cell. In other words that the conditions obtaining in it are abnormally favourable to ionisation like a highly “ripened ” photo-sensitive emulsion. A feeble radiation will accelerate the activity of A Theory of the Action of Rays on Growing Cells. 265 the morbid cell and yet scarcely affect the normal cell, the latter corre- sponding to a “slow” photo-sensitive film. Increased radiation which only attains the point of accelerating interchange in the normal cell may be attended by a sufficiently dense @-radiation to inhibit. the metabolism in the ~ morbid cell in the manner already suggested. In other words—to revert to the analogy with the photo-sensitive salt—the amount of ionic and electronic stimulus which builds the latent image in the “ fast” film is insufficient to affect the “slow” film and as the stimulus is increased the latent image of the first suffers reversal at a point which builds up the latent image in the second. This appears to be just what is observed in the case of radiation treatment, the success of the method depending upon a lag in the effects arising in the normal tissues, as compared with those arising in the morbid tissues. It may also be urged for the present view that if the effects of y-rays on the growth of the cell are not of a photo-electric character, and so productive of ionisation, we must recognise in them some quite new reaction between matter and light. This seems a needless course when there does not appear to be any a priori objection to urge against the unification of our views respecting the photo-stimulation of the sensitive salt and the effects of y-rays on the molecular systems existing in the cell. Assuming a real basis for the approximation of the two processes, the question as to how the peculiar constitution of the morbid cell may arise deserves more careful consideration than I am competent to give to it. Upon the photographic analogy we might reason thus :—If, in the life of the cell, ions are naturally always being formed, the absence of a “restrainer ” might lead to morbid ionisation; or, again, the presence of a “sensitiser ””—the former to limit the ionising activity either physically by its inert properties, or chemically ; the latter to accelerate it by removing the products of reaction as fast as they are formed. Dr. Lazarus-Barlow, however, has found notable and excessive quantities of radium in certain tumours. If this was general to all spontaneously arising cancers we might find here a sufficient cause of excessive lonisation. In this connection it is perhaps significant that the study of the distribution of cancer has been found to follow in a notable way the nature of the soil constituents of the district. Thus it is stated that cases of cancer are more frequent in clay-covered areas than in calcareous regions. Now calcareous rocks are almost without radioactive constituents, whether of the uranium-radium series or of the thorium series. The amounts of emanation continually being exhaled from such soils must be very different. It would be interesting to directly examine the several districts for soil- emanation. 266 A Theory of the Action of Rays on Growing Cells. Again, the well-known prevalence of cancer among chimney sweeps may be associated with the fact that charcoal and other forms of carbon, which must enter largely into the composition of soot, absorb radium: emanation readily from the atmosphere. It is improbable that sweep-cancer is ascribable to skin irritation only, seeing that many other occupations (e.g. stone working, cement making) are exposed to even greater risks from that source. On the theory that the cancer cell is the seat of excessive ionisation, we may ask if it is possible to control its activity. The latent image, although not possessed of the progressive fluxional characters of the cell, is, potentially, such an active configuration. It may be destroyed: (a) By such a light stimulus as will bring about reversal. The radioactive treatment of cancer is—on the present theory—an application of this fact. (0) By development, i.e. by such a chemical treatment as serves to discharge the ionised systems. The finding of a reagent which would act similarly on the morbid cell is, perhaps, not impossible. It would have to act selectively towards the less stable cell and must itself be ionised or become so in process of application. It would discharge the function of diverting the ionising activity to the formation of inert and harmless products. In a sense we may regard development as continually progressing in the organic system, much asif a light-sensitive salt were maintained submerged in a developer while exposed to light. From this point of view it might be better to seek the intervention of a “restrainer” which would either retard molecular motions of diffusion, etc., in a mechanical way, 7.e. by viscosity—as many restrainers are believed to do—or by chemically altering the nature of such conditions as result in growth and metabolism. If such remedies could be applied through the circulatory system, so as to reach metastases, depressing and lowering the abnormal ionic activity or directing its results into harmless channels, curative treatment might be attainable. { The theory here suggested for the processes going on in a cancer cell is a physical one, or, it may be said, takes account of the physical aspect primarily, and would involve the probability of successful treatment by experiments directed along physical and chemical lines. But it is not suggested that the origin of, or predisposition towards, abnormal ionic activity may not be founded in biologic causes. Nor does it enter into, or take account of, the probably extremely complex nature of the events progressing within the cell as leading to, or resulting from, the physical actions referred to in the theory. 267 The Influence of Timbre and Loudness on the Localisation of Sounds. By Cares S. Myers. (Communicated by Prof. C. 8. Sherrington, F.R.S. Received June 3,—Read June 25, 1914.) I. Introductory. In analysing the factors determining our localisation of sounds, it will be found convenient to distinguish “laterality” from “incidence.” By the laterality of a sound I mean its apparent position in relation to the median vertical front-to-back, or “sagittal,” plane; thus, a sound may give the impression of rightward or leftward laterality, or it may appear to have zero laterality— that is to say, its position may seem to be in the median plane. By the incidence of a sound I means its apparent position in relation to the horizontal “interaural” or “coronal” line, thus, a sound may give the impression of more or less upward, downward, forward, or backward incidence, or it may appear to be directly sideward, neither above nor below, neither in front of nor behind, the interaural line—when the incidence is of zero value. I consider it important to distinguish at the outset these two elements in localisation, since they are dependent on very different factors. In normal subjects, that is to say, in persons who have normal binaural hearing, the one certain and obvious determinant of laterality consists in binaural differences of intensity. A sound is localised on the side of that ear which receives the stronger stimulus; it is localised: in the middle line, midway between the two ears, when they are equally stimulated by the sound.* But such binaural differences of intensity must clearly fail as a basis of our determination of incidence. Whether a median sound les immediately in ‘front of or behind us, or whether it is placed immediately above or below our * Another determinant of laterality, binaural differences of wave phase, was suggested in 1907 by Lord Rayleigh (‘ Phil. Mag.,’ vol. 13, pp. 214-231, 316-319) ; but, taking into consideration the physiological fact that, owing to the bone conduction of sound across the skull, it is impossible to stimulate one ear without stimulating the other, I have indicated, in collaboration with H. A. Wilson [‘ Roy. Soc.'Proc.,’ A, vol. 80, pp. 260-266 ; ‘Brit. Journ. Psychol.,’ vol. 2, pp. 363-385 (1908)], how the effects of binaural phase differences are ultimately explicable in terms of the differences in binaural intensity to which they may be supposed to give rise. ord Rayleigh has since [‘ Roy. Soc. Proc., A, vol. 83, pp. 61-64 (1909)], allowed that “for the moment the choice between the competing views [as to the manner in which phase differences at the two ears produce their effect] is likely to depend upon-preconceptions as to the manner in which the nerves act.” 268 Mr. C. 8. Myers. The Influence of head, it must stimulate the two ears with the same intensity. It is just under these conditions that our localisation becomes erratic. As is well known, a sound coming from in front is apt to be localised behind, and vice versd. So, too, in regard to sounds placed before and behind the ear: a sound produced midway between the front and the side of one ear is often localised midway between the back and the side of that ear, and so on. It has been found that, although extremely erratic, our determination of the incidence of a sound is capable of enormous improvement by practice, and, seeing that our accuracy is greater with sounds richest in overtones,* it has been supposed that our awareness of incidence is dependent on the variations of timbre which occur with variations in the angle at which the sound waves impinge on the auricle. Now, if it be true that variations in timbre are responsible for our determination of the incidence of a sound, it should be possible to put this assumption directly to the test by experimentally varying the timbre of a given sound while its position is kept constant, and by observing what changes, if any, in its apparent position are produced thereby. Such has been the main purpose of the experiments described in this paper, and, as will be seen, they afford a striking proof of the correctness of the assumption. Two other possible factors affecting sound localisation have yet to be mentioned. It has long been recognised that sounds coming from in front of the subject’s auricle are better heard than those coming from behind. The auricle is so inclined and is so formed as to “catch” forward sounds better than rear ones.t Such variations in loudness, according to the relative positions of the sound and of the ear, may conceivably help in determining the incidence of the sound. The other possible factor, assisting the determination of laterality and incidence, consists in the tactual sensations which vibrations of sound may conceivably evoke by their contact with the auricle, the external meatus, or the tympanic membrane. The experiments described in this paper also afford some estimate of the value to be attached to these two factors. Il. Lapervmental Methods, The experiments were conducted in a sound-proof room (R, fig. 1), the walls and ceiling of which, composed of stone, peat-moss, and cork * Angell and Fite, ‘University of Chicago Decennial Publications,’ vol. 3, part 2 (1902). + How the ear “catches” sounds is quite unknown. The old explanation of reflection of the sounds from the concha to the tragus, and thence into the meatus, is untenable in view of the disproportion between the size of the ear and the length of the sound waves. 269 . tal w ws < aS} S w > “8 s 3 D rS Ss 5 S v = ol 3 : : S Fy S nw wH o < [Ss S = S Ss 2 )| @2 ee: : y 5 1, Sew | 2 NN: 2 0 = —al IN i qj Ss l\ ie = > 5 = | — reel Vie ng \ x \\ My SI ‘tL —=5 : 5 270 Mr. C. 8. Myers. The Influence of composition, were covered on their inner surface with a thick layer of horsehair. The floor, also isolated from the rest of the building, was similarly covered. By this means the reflection of sounds from the walls and floor was reduced to a minimum. The subject sat blindfold in the centre of the room. A — sound perimeter (P, cf. also fig. 2) was constructed for these experiments. It consisted essentially in an arm M, silently rotatable about Z by means of the handle L, and carrying a funnel-shaped mouth F, which was connected by a flexible pipe with the sound-producing apparatus in a neighbouring room. The centre about which the arm of the perimeter rotated was always placed over the point midway between the two ear-holes. The perimeter could be turned round the axis A so as to give sounds in the vertical as well as in the horizontal plane. With the perimeter arranged as in fig. 2 the sound could be presented at any point in the median vertical (sagittal) plane, ae. directly in front of (= 0°), above (= 90° v.) or behind (= 180°) the subject, or at any inter- mediate point. The sound could also be presented, as in the arrangement of the perimeter shown in fig. 1, at any point in the horizontal plane at the level of the ears, ae. directly in front of (= 0°) or behind (= 180°) the subject, or exactly opposite the right or left ear (= 90° h. or 270° h.) or at any intermediate point. The pipe connecting the funnel with the sound-producing apparatus was enveloped with layers of cotton wool and bandages, and passed through a tube in the wall of the sound-proof room to a very wide-mouthed horn H, such as is used in connection with the phonograph when heard in large halls. Before this horn were arranged four of Stern’s tone-variators, Vi, Vo, V3, Va, blown by wind from a hydraulically worked organ bellows. These tone-variators, one of which, Vi, was unenclosed, produce relatively pure (overtone-free) tones ; they are essentially bottle whistles, each consisting of a mouthpiece fixed over a metal air-containing cylinder. The pitch of the note depends on the height and circumference of the cylinder, the base of each cylinder being movable so as to adjust the pitch accurately. ‘The first or largest tone-variator was arranged to emit a (fundamental) tone of 215 vibrations per second. The second variator gave the first overtone, the third gave the second, and the fourth gave the third overtone, of this fundamental, ze. they emitted tones of 430, 645, and 860 vibrations per second respectively. In order to reduce and to vary at will the loudness of these overtones, the three corresponding variators, Vo, V3, V4, were each enclosed in a wooden box, open at one end. The open end of each box could be more or less completely closed by means of an adjustable slide, thus allowing the intensity of the overtones (and hence the timbre of the Timbre and Loudness on the Localisation of Sounds. 271 total sound) to be experimentally varied.* Three positions of the slides were adopted: position B, the middle or “normal” position of the slides, which was used for practising the subjects in sound localisation ; position A, in which the slides nearly closed the open erids of the boxes ; and position C, | in which the slides were drawn well out so as to produce relatively loud overtones. The loudness of the whole sound (fundamental and overtones) was varied by moving the horn nearer or farther from its middle or “normal ” position. The subjects were practised in sound localisation with the horn at its “normal” or B position. Subsequently, the loudness of the sound was decreased by moving the horn farther from the tone-variators (the A position of the horn) or increased by moving the horn nearer to the tone-variators (its C position). The use of the tone-variators and of varying positions of the slides or horn, just described, necessitated the presence of an assistant in the room in which the sounds were produced. Communication between him and the experimenter, who sat with the subject in the sound-proof room, was effected by means of loud-speaking telephones and an electric bell, so that by pre-arranged signs the assistant might give the sounds at the desired moment, and vary their timbre or loudness in the desired order. Experiments were also carried out in which the sound was produced by the experimenter within the sound-proof room by means of a telephone buzzer or an electric bell placed at the position of the funnel F, at the free end of the rotating arm of the perimeter. Except when otherwise stated, the mode of procedure was as follows: The subject, seated in the chair C in the sound-proof room, was blindfolded, and a head rest was adjusted to the back of his head in order to prevent, so far as possible, any movement. The perimeter was then arranged so as to allow of sounds being given in one or other of the two planes (vertical or horizontal), and the slides and horn were set at their respective B positions. Several sittings were given by each subject for practice in localisation, and later the positions of the slides and horn were irregularly varied for the study of the effects of variations in the timbre and loudness of the sound in one of the two planes, before similar experiments were made in the other plane. The production of the sound which the subject was required to * In a number of preliminary experiments, I employed resonators at variable distances from the variators, but the tones conducted from the resonators by narrow rubber tubes to the sound-proof room were too weak for my purpose. I also tried a loud-speaking telephone for conducting the tones, but, owing to the unsatisfactory timbre and incon- stancy of the resulting sound, I had to abandon this method likewise. x Z 272 Mr. C. 8. Myers. Zhe Influence of localise was preceded by a warning “ Now”; and the sound was allowed to last for about two seconds. Immediately after each sound was given the subject was required to indicate its supposed position. In the early stages of practice the sounds were given at any position within the half circle (from 0° to 180°) of the plane concerned, and the subject’s forefinger was armed with a large graduated quadrant carrying a freely movable index, so that when he pointed to the apparent direction of the sound the index registered the angle at which the sound appeared to be placed. In the later experiments, when only three positions of the sound in any one plane were employed, and the subject was either being instructed in correct localisation or (still later) being tested for the effects of variations in timbre and loud- ness, he learned to return his answers orally in terms of the angle from which the sound appeared to come. Eleven subjects were investigated, seven male and four female, all under 40 years of age. Each sitting lasted about 40 minutes, and each subject gave from three to six sittings, making from 200 to 400 judgments of localisation. IIL. Bepervmental Results. 1. Localisation before Practice— (a) For Sounds in the Median Vertical Sagittal Plane.—Without practice the complex sound from the variators proved extraordinarily difficult to localise. Whatever the actual position of the sound, some subjects localised it in front, others localised it behind, others were unable to give any consistent localisation. As one subject remarked, “I could put it anywhere ; I seem to think out where it might be and then it seems to be there.” Another subject reported, “When you tell me where it comes from, I see it can do so, and can place it there.” When the variators were replaced by a telephone buzzer before the horn, no appreciable difference in the certainty or accuracy of localisation was observable. It was always difficult to arrange the apparatus so that the sound appeared exactly in the middle line. At first the difficulty was traced toa slight leakage of the sound through the flexible tube which conducted the sound from the inlet pipe in the wall to the funnel-shaped opening borne on the perimeter, But even when this difficulty was surmounted, the slightest error in the position of the perimeter in regard to the sagittal line of the head immediately occasioned lateral (right or left), instead of purely median, localisations. Wondering whether any possibly still remaining leakage of sound during transmission could be responsible for the extreme difficulty and inaccuracy Timbre and Loudness on the Localisation of Sounds. 273 of localisation of the variator sounds, I replaced the funnel-shaped mouth first by an electric bell, later by a telephone buzzer, in the expectation of obtaining more accurate and certain localisation when the sound was generated on the perimeter instead of being conducted to it from the room outside. . The same diverse and erratic localisations were maintained. Some subjects never localised the sound behind 90° v. if it was placed at 180°; others never localised in front of 90° v. a sound given at 0°. Here, for example, are the records of two subjects, J. and Ss., for sounds of the buzzer (affixed to the perimeter) at 0°, 90° v. and 180° :— J. Ss. OR O2a OL rROS Oe" S02 HOP SOme elS0s as02 ORV ete JOR ORO O90 90 * 80 jou lisoy so SOM Sees. ok OD OD SO Oo 90 90 130 200 200 Before I had obtained evidence of these striking individual differences in localisation, I wondered whether inequalities in the reflexion of the sound from the four walls of the room could be responsible for the gross errors met with. Accordingly, on several occasions, I reversed the position of the subject, testing him now with his face, now with his back toa given wall. With these changes one subject was tested with the variator sounds, two subjects with the buzzer sound conducted from the room outside, and one subject with the bell ringing on the perimeter. But in no case was any change in localisation relatively to the subject observable. If he had localised all sounds to his rear in one position, he continued to localise all sounds to his rear in the reversed position, and so on. This result is striking evidence of the influence of natural tendencies and prejudices on the part of the subject in his localisation of sounds placed in the median sagittal plane. The influence of expectation was also clearly demonstrable by directing the subject’s attention forwards or backwards at the moment of the production of the sound, whereupon the apparent position of the sound was generally changed in the sense of such direction of the attention.* (b) For Sounds in the Horizontal Plane.—But if the ability to localise fore and aft sounds in the median vertical sagittal plane is so defective, we should not expect to be better able to localise fore and aft sounds placed along the horizontal plane; for both kinds of localisation are instances of what | * The following conversation between subject (S.) and experimenter (E.) will serve to illustrate this feature :—S. “I expected a sound behind and I got it [sound given at 0°] there.” HE. “ Now try to imagine it in front ”[sound at 0° repeated]. S. “ Yes, I certainly get it there, too.” E, “Now try to imagine this sound [at 180°] behind.” S. “Yes, certainly it is there, but when I change my idea to its being in front, I get it there too.’ Eee Mr. C. 8. Myers. The Influence of have termed incidence. Laterality can only concern whether the sound is placed to the right or left of the subject or somewhere in the median line; and, as I have already said, errors in laterality were never found in these experiments, provided that the auditory acuity of the subject’s two ears was normal and that the position of the sounds relatively to the two ears was such’ as to produce the required binaural difference (or equality) of loudness. On the other hand, whatever factors are responsible for’ the determination incidence should hold for the horizontal, as well as for the median sagittal, plane. . Experiments carried out on five subjects with vertavien sounds given in the horizontal plane reveal just the same inaccuracies as have been described for the median sagittal plane. The first of these subjects localised all sounds— whether fore (45° h.), side (90° h.), or aft (135° h.)—behind_ his ear, the second localised them all in front of his ear, the third localised fore and aft sounds in front of his ear, while the fourth and fifth subjects gave too variable a localisation to allow of any more general statement than that they showed total inability to distinguish fore, side, and aft sounds from one another. . Two questions naturally arise :-—How is it that previous observers, while recognising a liability to err in the localisation of such sounds, have not laid stress on the initial grossness of the errors of localisation revealed under the conditions of these experiments ? How is it that these errors do not play an equally prominent part in our everyday life? We are all aware of occasional errors in fore and aft localisation, but it is relatively seldom that they are brought to our notice. Now, one important factor consists in familiarity with the sound, As we shall see, with practice every subject learned to localise correctly. Another important factor employable in everyday life, but’ eliminated to a very large extent in these experiments through the use of a head rest, consists in head movement. On several occasions in the course of these experiments I expressly instructed my subjects to move the head while they were listening to the sound, whereupon their errors in localisation were immediately and often quite accurately corrected. In some experiments, moreover, performed in the open air, in which I acted as subject, where the vowel E was spoken by an assistant and his position had to be ascertained, I localised both fore and aft positions forward, but when the experiments were repeated with a small head movement carried out during the production of the sound, I at once eaeaget the localisation of the rearward sounds from fore to ait. ; Obviously, by turning the head, the sound is alterable in intensity ; for, Timbre and Loudness on the Localisation of Sounds. 275 as I have already mentioned, the position of the auricle is adapted for “catching” sounds coming from in front (and in consequence of which our auditory acuity is keener for forward than for rearward sounds). But turning the head alters, too, the timbre of the sounds; a forward sound appears to the ear not only louder than, but also of a different timbre from, the sound placed to the rear.* It is noteworthy, however, that, whereas change in the position of the head while the sound was being heard was remarkably effective in correcting errors of localisation, change in the position of the sound while the head was at rest proved of little or no advantage for such correction. It generally resulted in an interpretation of increased or diminished loudness, or of increased or diminished distance of the sound; less frequently, a movement of the sound was detected, but the direction of the movement was not always correctly given, and the initial error in localisation failed tu be corrected by the detected movement of the sound. This difference in effect between what may be conveniently termed “active” and “passive” change in the position of the sound is of considerable interest in relation to the associated function of the semicircular canals and (in animals) of the movements of the auricle. Two other factors which are conceivably of importance in determining the incidence of sounds, but which were almost wholly eliminated in these experiments, may be briefly mentioned. Of these the influence of expecta- tion has been already alluded to on p. 273, and was almost always success- fully ruled out by the noiseless movement of the perimeter. On several occasions I expressly asked the subjects if they had any notion of where the sound was coming from, and they generally replied that they had no idea. In everyday life, however, and, perhaps, in many of the experiments other- wise condugted, various cues may determine a favourable attitude of expectation in the subject. The remaining factor, the effect of sound reflections from the ceiling, walls, and floor, was prevented by the peculiar construction of the sound-proof room (pp. 268-270). But in everyday life and after brief practice in experiments, conducted under ordinary conditions, there are indications that such reflections are taken into account and thus assist in determining the incidence of the sound. * Tt is practically impossible to increase the loudness of a sound (ze. the intensity of the fundamental and its overtones) without altering its timbre (the relative intensity of the fundamental and its overtones). Even if this could be physically realised, the varying position of the peculiarly formed auricle relatively to the sound may be expected to influence the ease with which it takes up the different overtones contained in the sound. + Since writing this, I have examined two subjects, first in an ordinary room, and later (after a rest) in the sound-proof room, using the perimeter with an attached 276 Mr. C. 8S. Myers. The Influence of 2. Localisation during Practice— (a) For Sounds in the Median Vertical Sagittal Plane-—The practice experiments were carried out during several sittings, the number (cf. p. 272) depending on the rapidity of improvement in the subject’s accuracy of localisation. The subjects were now told when they were right or wrong, and only three positions of the sound were given—at 0° directly in front, at 90° v. directly above, and at 180° directly behind an imaginary line between the two ears. The final result was always to establish absolute accuracy in the localisa- tion of the sounds. But the three positions were not learned with equal ease ; consequently the total number of right answers varied with the position of the sounds, the figures during relatively late stages of practice with the variator sounds being— For (eli A Me 80 per cent. of answers correct. ” 90 Ws 650000 TZ +P) 2 He odie OP RE east ske 67 ”? be) The criteria apparently employed during the subject’s practice, in order to distinguish these three positions, were (i) so-called “ tactual” experiences ; (ii) right or left laterality ; (a1) difference in timbre, loudness, or nearness. Of (i) further mention will be made later (pp. 280-283). Reliance on (ii) was only possible when the sound was not accurately produced in the middle line ; if, for example, the rotating arm swung a little obliquely from before backwards, the subject came to realise that when the sound was heard (say) to his left it was placed (say) behind him, whereas when it appeared to his right it lay to his front. In regard to (iil) various subjects stated that at 0° the second was “ fuller,’ “more voluminous,” or “ more open,” while at 90° v. it was “duller,” “drearier,” “more drony,” or “more booming,” and at 180° it sounded “rather like an echo,” “ faint,” “lacking in assurance,” “fuller than at 90° v., though very like it,” yet “duller and more distant than 0°.” Similar results were obtained during practice when the telephone buzzer took the place of the variators before the horn. (b) For Sounds in the Horizontal Plane.—In these experiments only three subjects received practice for sounds placed at 45° h., 90° h., and 135° h., but the results were precisely similar to those obtained for the sounds in the vertical plane. Two subjects thought that at 90° h. they could distinguish electric bell and buzzer. Despite the practice gained in the ordinary room, their errors increased by about 50 per cent. in the sound-proof room, showing clearly the influence of the strange environment. Over 300 judgments were obtained. Timbre and Loudness on the Localisation of Sounds. 277 (i) “ tactual ” sensations. Of course in the positions used in the horizontal plane (ii) laterality could afford no clue as to the fore or aft localisation of the sound. The remaining factor (iii), differences in timbre, loudness, and nearness, proved the most important criterion in learning to localise the sounds correctly. One of the subjects complained of special difficulty in distinguishing sounds at 90°h. and at 45° h.; another of special difficulty in distinguishing sounds at 90° h. and at 135° h. At 135° h. the sound seemed to two subjects “more remote and weaker,” “less clear and less distinct,” than at 45° h. or 90° h.; the third subject, however, observed that at 135° h. it was “nearer and more rounded,” “ not so veiled,” as at 45° h. 3. Experimental Variations in the Timbre and Loudness of the Sounds— (a) For Sounds in the Median Vertical Sagittal Plane—lIn the case of two subjects, after being thoroughly practised in the correct localisation of sounds given at 0°, 90° v., and 180°, instructive results were obtained by experi- mentally varying (i) the timbre and (1) the loudness of the variator sounds. In both subjects variations in timbre (produced by varying the position of the slides) yielded less striking errors of localisation than variations in loudness (produced by varying the position of the horn). Thus in one subject, who had just given 17 of 18 answers correctly for the normal or B position of the slides and horn, variations in the position of the slides produced one doubtful and one ambiguous* answer in 11, while variations in the position of the horn gave one wrong and four doubtful or ambiguous answers in nine. On another occasion the same subject, after giving two doubtful or ambiguous answers in 18 for the normal position of the slides and horn, gave 10 wrong answers and one doubtful answer in 18 when the position of the horn was varied. The disturbing uncertainty thus produced was in some degree carried over to the subsequent experiments immediately carried out with variations in the position of the slides, when three wrong and two doubtful or ambiguous answers in 13 were returned. The second subject, who was examined only for the positions 0° and 90° v., gave five wrong and two doubtful or ambiguous answers in 27 for the normal or B position of the slides and horn, followed by three wrong and two doubtful or ambiguous answers in 18 when the position of the slides was varied. On another occasion, when the same subject had just given 12 con- secutive right answers for the normal or B position of the slides and horn, six wrong answers in 21 were obtained by varying the position of the horn. In both subjects it was found that, whereas the sounds at 0° suffered least, * An answer is “doubtful” when the subject is obviously uncertain ; it is “ambiguous” when the subject ascribes to the sound alternative positions, of which one is correct. 278 Mr. C. 8. Myers. The Influence of those at 90° v. suffered most in the accuracy with which they were localised under the above conditions. These results may be tabulated thus, the figures showing the percentages of error, doubtful or ambiguous answers being counted as half errors, wrong answers as whole errors, and the two vertical columns for each subject repre- senting the results respectively obtained from the two sittings at which each was examined :— Subject I. Subject IT. Horn and slides in B position..................++- jG QD -O) Horn in B position, slides in varied position... 14 31 22 — Slides in B position, horn in varied position... 33 58 — 29 Six times, the near or O position of the horn caused a sound at 90° v. to be located at 0°, and on two occasions, one at 180° to be locatedat 90° v. Three times, immediately. following a sound given with the open or C position of the slides, a sound given at 0° with the B position of the slides was located at 90° v.; and on two occasions, immediately following a sound given with the B position of the slides, a sound given at 90° y. with the C position of the slides was located at 0°. The following answers.illustrate the difficulties in which the subjects found themselves, and indicate the bases of their judgments of localisation :— Sound given. Subject’s reply. fo} Slides C 90 vy. 0°. “ Because it was so full; yet it seemed perfectly vertical and hit me on top of the head.” Hor 0 90 v. 0° or 90° y. ‘‘It seemed loud, hence front; yet far away, hence top.” | py ea 0) ? 90° y., hesitation. “Because, though not so weak as a top sound, yet it does not seem so direct as a front one.” pe AL 0 90° v. ‘‘ Because it is so faint.” real (0) ? 0° “It is rather weak, though, for a front sound.” Horn C 90 v. ? O°. “It has the character of a front sound in coming from a distance, but it’s so drony and dreary.” Oe 90 v. 0°. “Its character resembled the previous sound [B, 90° v.], yet it came from so short a distance as to seem front.” Bh AS 0) 90° vy. “It’s drony, yet it’s rather too loud for top.” (b) For Sounds in the Horizontal Plane.—The influence of changing the timbre and loudness of the sounds on their localisation is not less marked for sounds in the horizontal plane, although certain differences are to be noted. In the following record of one of my subjects the first two columns give the actual, and the. third column gives the apparent positions of the sound, the observations of the subject being given in footnotes :— Timbre and Loudness on the Localisation of Sounds. 279 Sound given. Subject’s reply. Sound given. Subject’s reply. ° CG ce} ° Slides A 135 h. 45 h. Horn B 90 h. ? 90 h. mie 135 h. 135 h. Slides A 135 h. 45 h.t RB. 45 h. - 45 h. wegen 3) 90h. 4 90h. cot ml} 90 h. 90 h. eet a ©] 185 h. 135 h. BOB 135 h. 135 h. epee) 45 h. 45 h. sonal 0} 45 h. 45)hy Horn A 90 h. 90 h. LA nO) 45 h. 135 h. reese) 90h. | 90 h. Bir 3 135 h. 135 h. pane dexen 45h. |} 45h. Satay B 90 h. - ? 90h. Slides A 45h. |} 45h. ay Peal B3 135 h. 135 h. eke 45 h.: 45 h. or 90h. Horn A. 90 h. pts Horn A 90 h. 45 h. or 20 h. het vB 45h. 45h. Wh og svi 45h. | 45 h. oy Bl BB Tap 135 h. | eee tS a eels Open eeu musi ops ee Bie eltsonhe 135 h. _ || Slides A 90h. |} 45 h. + B 90 h. 90h. | aD 90:h. - | 45 h. or 90 h. Slides C 45 h. 45 h. ptean 01 45h.” | 45 h. * “Tt has the quality of 45° h., but it is not so far back, I think, nor so distant as 90°.” yt “That’s the 45° h. all right!” That is to say, for 19 estimations in the normal or B position of the horn and slides, only four doubtful or ambiguous answers occurred (11 per cent. of errors), whereas for six estimations in the A or C positions of the horn there were two such answers (17 per cent. of errors), and for seven estimations in the A or C position of the slides there were four wrong answers (57 per cent. of errors). Thus the effect of varying the loudness of the sound was to reduce the certainty of this subject’s answers, while the effect of varying the timbre of the sound was to change the apparent position of the sound. It will be noticed that whereas changing the positions of the horn produced greater confusion in the vertical plane, changing the position of the slides produced greater confusion in the horizontal plane. We might be inclined to conclude from this that localisation is based on differences in loudness for sounds in the vertical plane, and on differerices in timbre for sounds in the horizontal plane. But we have to remember that changes in the position of the horn must have affected not only the loudness but also, though much less markedly, the timbre of the sound; and that changes in the position of the slides must have affected not only the timbre but also, though much less markedly, the loudness of the sound. We have also.to bear in mind that in the vertical plane we were dealing with sounds placed at forward (0°), topward (90° v.), and backward (180°) positions, while in the horizontal plane, the sounds were given half- forward (45° h.), to the side (90° h.), and half-backward (135° h.). We may, I think, legitimately conclude that for sounds given at 0°, 90° v., and 180°, our localisation is based principally upon differences in loudness, 280 Mr. C. 8. Myers. The Influence of whereas for sounds given at 45°, 90°, and 135° in the horizontal plane our localisation is based principally upon differences in timbre ; “ principally ” because changes in the position of the horn must have affected not only the loudness but also, though much less markedly, the timbre of the sound, and because changes in the positions of the slides must have affected not only the timbre but also, though much less markedly, the loudness of the sound. This conclusion is in harmony with other considerations. There are enormous differences between sounds at 0°, 90° v., and 180°, as regards the favourableness of their position for being “ caught up” by the pinna. The pinna catches sounds coming from the front better than it catches those coming from the rear; as is well known, auditory acuity is keener forwards than behind. It is hence not surprising that we learn to distinguish fore, aft, and top sounds principally by differences in loudness. On the other hand, sounds given at 45°, 90°, and 135° in the horizontal plane must differ little in loudness ; the difference between the extreme positions, 45° and 135°, is much less than that between. positions 0° and 180°; 45° and 135° are almost, although not quite, equally favourable positions for the sound to be caught up by the pinna, and indeed 135° is the angle most suitable for the direct entry of the sound into the meatus. 4. The Réle of Tactual Sensibility in Auditory Localisation— These experiments appear to prove conclusively not only that variations in timbre and loudness are responsible for our determination of the incidence of sounds but also that cutaneous sensibility can play no part whatever in sound localisation. That cutaneous sensations can play no part so far as concerns laterality is shown by the well-known fact that whereas we are able correctly to localise two simultaneous tones of clearly different pitch, placed one on each side of our head, whatever be their relative loudness, our localisation of two tones thus placed, when they are of identical pitch, depends upon their. relative loudness; if the two tones are equally loud, the sound is localised in the median plane; as soon as they become of unequal loudness, the sound is immediately localised in that ear which receives the stronger stimulus.* Now, if the sounds falling on each ear gave rise to tactual sensations, there can be no reason why, whatever their pitch and relative loudness, two such tones should not be correctly localised, one on one side of the head, the other on the other. On the other hand, it is quite clear that sound localisation rests on an auditory not on a tactual, sensory basis, since when the tones are of identical pitch only a single sound is heard and its localisation is * I omit for simplicity’s sake the consideration of phase difference here (see, however, footnote to p. 267). Timbre and Loudness on the Localisation of Sounds. 281 accordingly ascribed to a single position, median or lateral, instead of to two lateral positions. Further, when the tones are of different pitch, it is impossible to see how tactual sensibility can be the basis of their separate localisation. For suppose that one pinna, meatus or drum receives a series of tactual stimuli from the one tone, and that the opposite pinna, meatus or drum receives another series from the other, it is inconceivable how the subject can refer these two series of tactual stimuli to their respective tones; how can he decide which tone to allot to which ear merely on the basis of his tactual sensations? Again, suppose that a subject has become absolutely deaf in both ears; why on the hypothesis of tactual localisation should he not still be able on request to localise successfully a sound stimulus though unable to hear it assound? Yet this is inconceivable save in the case of the very lowest tones, the stimuli of which evoke tactual as well as (indeed ultimately in place of) auditory sensations. Moreover, the unimportance of the tympanic membrane in sound localisation is shown by the preservation of localisation in cases where the membrane has been removed through disease, and in cases of tinnitus aurlum where the sensatious although localised arise subjectively, within the inner ear. That tactual stimuli received by the pinna play a part in the localisation of sounds in the median sagittal plane is rendered highly improbable by a& priori considerations. The following experiment, moreover, appears decisive. After preliminary practice, I acquired correct localisation of sounds in this plane; whereupon I placed a short piece of narrow rubber tubing in each ear, the result of which was to make an obvious change in the loudness and timbre of the sounds heard. Now if the pinna had been responsible for the previously correct localisation, no change should have resulted from the insertion of the rubber tubes into the two meatus. But in point of fact, I was quite unable to localise the sounds correctly, and had to start afresh re-learn- ing them. There was no doubt in my mind that I had based my previously correct localisations on changes in the relative loudness and timbre of the sounds dependent on their position in regard to the ears. The results of this experiment confirm those already described in this paper showing the definite changes in localisation produced by definite changes in the loudness and timbre of the sounds. Nevertheless, the belief that auditory localisation is, at bottom, of tactual origin dies hard. Started by Weber* and perpetuated by Wundtt and others, the tactual hypothesis has been recently invoked by Hocart and McDougallt * *Ber. d. Kgl. Siichs. Ges. d. Wiss.,’ 1848, p. 237; 1851, p. 29. + ‘Grundziige der Physiologischen Psychologie,’ 5th ed., vol. 4, p. 487. { ‘Brit. Journ. Psychol.,’ vol. 2, pp. 386-405 (1908). 282 Mr. C.S. Myers. The Influence of to account for their experimental results. In my own experiments, were I to trust the introspective data of several of my subjects, additional evidence could be supplied in favour of this view. Thus for sounds given in the median sagittal plane, one subject in her early stages of practice described those at 0° as follows: “It hit my head just above the forehead,” “it hit me just above the forehead,” “it hit me just in front of the top of my head,” while the 90° v. sounds seemed to have “a more vertical feeling,” “a straight-downward feeling.” Now to this subject all sounds (in the median sagittal plane) at first appeared to come from the ceiling cornice in front of and above her. Most descended and hit her forehead and vertex, while a few others remained there. Sounds at 180° were accordingly described thus: “It remained on the ceiling, but pointed to the forehead” ; “it was located in front of and above me at the ceiling cornice, but it struck me at once on the vertex”; “it seemed a little lower than the rest, but it hit me in the middle of the forehead.” Another subject, who at first ascribed a backward position for all sounds in the median sagittal plane, described the 0° and 90° v. sounds as hitting him at the occiput, and the 180° sounds as hitting him at the nape of the neck. Yet another subject reported on 90° v.—< that reached my eye instead of my ear.” In the case of sounds in the horizontal plane similar examples may be quoted. One subject, who at the start ascribed a position of 45° h. to sounds given at 45° h. and at 135° h. and a position of 22° h. or 45° h. to sounds given at 90°h., at a later stage of practice mentioned that sounds at 90° h. “seemed several times to end up opposite my ear, possibly giving a touch sensation.” Yet when the perimeter arm was moved from 135° h. to 90° h., while the variators were sounding, this subject replied that the sound ‘seemed to move from 30° h. to 45° h.,” and that “I felt something blowing on my skin at 45° h. in front of my ear.” In the face of such evidence it seems incredible that tactual sensibility plays any important part in sound localisation. Only a few of my subjects reported its presence, and these agreed that ultimately they discovered the only reliable basis of localisation to consist in differences of timbre and loudness. We seem forced to conclude that the localisation of such tactual sensations is to be regarded as resulting from, instead of giving rise to,, determinations of sound localisation.* * It would be rash to assume that auditory stimuli do not give rise to tactual sensa- tions: the longest sound waves unquestionably do. Nevertheless, it is unlikely that the shortest waves excite tactual sensations, and it seems certain that whatever tactual sensations an auditory stimulus may evoke, they play no part in determining sound localisation. / ’ Timbre and Loudness on the Localisation of Sounds. 283 This view of the illusory function of tactual sensations in sound localisa- tion receives support from other data afforded by my subjects. Many of them, at their early stages of practice, seemed compelled to objectify the sound in tactual visual terms. Thus to one subject a sound at 90° v “appears to come from a central point,’ while one at 180° “appears to come from different sides as if entering the ear by various rays instead of by a central one.” Another subject said “I can’t attend to the sensation as such ; I have to fancy a motor cycle behind me,” or “I fancy myself in a wood with the sound (given at 180°) low down at the end of the path before me.” A third subject observed “I give each sound a body, and each is generally circular—a ball.” A fourth subject, who had described the variator sounds as generally coming down and hitting her, observed that the buzzer sounds “seemed in many cases to remain in their place and to throw out a sort of pseudopod, like a wriggling worm pulling its tail through.” In view of these descriptions, need we hesitate to ascribe localisations of tactual sensation, when they occur, to an inevitable tendency to treat localised sounds as if they were external objects describable in visual and tactual language, and as if they hit the ear, face, vertex, or occiput according to their localisation determined on the basis of timbre and loudness ? IV. Conclusions. 1. The “laterality” of a sound (zc. its estimated position in relation to the median “sagittal” plane) is determined by binaural differences or equality of intensity of the sensation.* Experimental changes in the timbre or loudness of a sound make no difference in its laterality. As soon as an infant begins to take notice of sounds, their laterality is at once appreciated. There are no trial movements of the head, this way or that, for sounds placed to one side of the median sagittal plane. The reception by one ear of a stimulus stronger than that reaching the other ear at once determines in the infant a movement of the head and eyes to bring the latter towards the source of the sound. 2. On the other hand, even in adult life, the grossest errors are made in determining the incidence of a sound (ze. its estimated position in relation to the horizontal “interaural” line), unless the subject has been practised in the changes in timbre and loudness produced by such changes of incidence, or unless he is allowed to make movements of the head, the effect of which is to vary the timbre and loudness of the sound while it is being heard. * And, according to Lord Rayleigh, by binaural ditferences or identity of phase of the sound waves (see, however, footnote to y 267). VOL. LXXXVIII.—B. a 284 Prof. A. J. Ewart. Comparative Study of Oxidation by 3. The “incidence” of a sound is hence determined by its timbre and loudness. Experimentally produced changes in the timbre or loudness of a sound lead to marked changes in its apparent incidence. 4. Tactual sensibility appears to play no part in auditory localisation. Localised tactual sensations evoked by auditory stimuli are generally the outcome of interpretations by the subject, resulting from his natural tendency to treat sounds as material objects, and to refer to them a localisation based on solely auditory data. A Comparative Study of Oxidation by Catalysts of Organic and Inorganic Origin. By ALFRED J. Ewart, D.Sc., Ph.D., Professor of Botany and Plant Physiology in the Melbourne University and Government Botanist of Victoria. (Communicated by Prof. J. Reynolds Green, F.R.S. Received January 14,— Read February 19,—Received in revised form July 25, 1914.) CONTENTS. ; ’ PAGE ihe BrownineyorAipplestands Potatoes saeesea-n erect reeeree cee eet eee eee eee eee eeeE Eee 285 Ua Thorslwvesnes Gie IOS OMS; Oya, ENON AMMEN 4.0:icaongodoncHnodaonooHboonoHoDBcADOBGAIHI0oNI002909" 286 Inorganic Oxidases 5 occ. 4.0 cxcige dee eee a con tenes ae eae ele CeCe EERE See eee 287 The -Naturerof Oxidases x aseiinsaitacsinec ad duen en seneacnseeace ten case acon aeaee Clete eee 290 Oxidase Sensitisersjamds Inlhibibonsreeseraeeee deere ate ee eer ter ene tener eee nee eee eR Eee 292 Sensitisers and Antagonisers to Plant Oxidases .......................ceeceeeneee eee eeees 296 The Relation between the Action of Metallic Oxidases and that of an Enzyme 298 Apple Oxidase iis cencaeteseaes acinscesentrsinn lence cone nt gece teeer ca RE ae ees an eee ee 299 hel ChromogentofithewAtpplerwa- easaee eee recone eeee cee ERR eee eee ee eer eee REP CREEEE 300 hey Chromogenkotebhe sz ouabOeereee epee: eae eee eee eee e er eee eee eeecr eee et eeeE eee eneneee 302 Potato OxIdase «issn tskijeneatilotetenssina-aines/caucteee ree e oie epitesiaee nea cee h oer ee eee EEE 303 ine Hixtraction othe Oxidase serene: see teeee eee e eee eee eeRE Ce ee err EE eRe eEEEe 303 The Distribution of the Oxidase in the Cell 20.0.0... cece cecneececneteten ener 305 Coloursand OxidaseiA ction.) iin tia aiosaaecascetltosestimtncie se tdcn ter ise aeee ee ee eeeees 306 Mhe Destruction of |Oxidaseybiys Hieatecne- eet eeee-eeseces ee ee ese eee ert eee eee eee eee ere 307 The Resistance of Oxidase to Drying and Keeping ....................eeeeeeeeeee sense 307 Paraphenylenediamine Test for Oxidase... .--.0.....02c.sceeceeneee sos eeserserecescanessee 308 UnrsolM@artrate Mestiton mulomiin wenmerpeeds-e cc cer cheer seee Cresta reece eseraiets eee eee eee eeeee 309 The Action of Paraphenylenediamine on the Oxidases of Apple Sap and Potato 310 The Potassium Iodide and Starch Test for Oxidase in Living Tissues ............ 311 The Influence of Anesthetics on Oxidase ACtion...............:0cseseceeeeeeeneeeeneneeee 312 The Oxidases of the Lemon and Orange .................00ccceeeesecceee eee sen neces eee ene 314 TheiOxidaselofwthelCarnob meee. serie eee eee eee cece eect aecnetcecm ens eeeee ee 315 TheiOxidase of Red Beetroot sins -eememeceseerce tances se sacteench eh be er cecal eee eeere 315 The Oxidase, of the Parsnip is. cc sc saves seine suc soe sso ore ones ebe eee te caer eee 316 po) ulloa) §2)2 i Zee Avon ae See ease Raden noise Suuqd odbm hobo bed ooddnaudanooHuauepocuDpanesouoaséuboDdacoe 317 The present paper is the outcome of the work carried out on the influence of poisoning on apples and potatoes, and its progress has necessitated a —————e— tO Catalysts of Organic and Inorganic Origin. 285 general revision of the oxidase ferments, and in particular a general comparison with metallic oxidases. As is well known, oxidases are widely distributed in plants, and are frequently responsible for the changes of colour in extracted plant juices or in plant tissues after death. In the case of the apple and potato, the curious fact that browning took place in pulp killed by immersion in poisonous metallic solutions, but not when killed by heat, demanded special investigation. The Browning of Apples and Potatoes. It is generally assumed that this is due to the action of an oxidase ferment upon a chromogen present in the pulp cells, such as tannic acid in the apple* and tyrosin in the potato, but the oxidase is not necessarily the same in either case. The term “oxidase” in fact rather represents a result than a particular substance, and many “oxidase” actions are not necessarily due to organised ferments or enzymes at all. In a previous paper it has been shown that apple pulp immersed in solutions of metallic poisons may still turn brown, although ordinary ferments are destroyed by such poisons. Furthermore, the reasons why the browning takes place on death, but not in the living cell, and not when the cell is killed in certain special ways need investigation. Pro- chromogens or zymogens may exist in the living cell which decompose into interacting chromogen and enzyme on death, or the latter may be kept apart in the living cell by semipermeable membranes which lose their imperme- ability on death. In the latter case the localisation of the chromogen and enzyme in the cell becomes a problem of special importance. According to Grusst “ Antioxidases ’{ capable of arresting oxidase reactions exist In various plants, and if the oxidase and the “antioxidase” balanced, a chromogen and its oxidase might exist in contact in the living cell and the mode of death might determine. whether browning occurred or not. Behrens§ considered that the browning of the apple pulp is due to a direct oxidation of tannic acid to form a leathery compound with the proteids of the cell, without the aid of an oxidase. {. \The first point needing full investigation was the influence of poisons on browning, particularly in regard to the time factor and the rapidity of penetration. * Lindet, ‘Compt. Rend.,’ vol. 120, p. 370 (1895). t ‘ Biologie und Capillaranalyse der Enzymen,’ p. 56 (1912). t To avoid possible confusion, the word “inhibitor” may be used instead of this term. § ‘Centralbl. f. Bakt.,’ 2 Abth., vol. 4, p. 514 (1898). Vo 2 286 Prof. A. J. Ewart. Comparative Study of Oxidation by The Influence of Poisons on Browning. As is well known, apple and potato pulp if killed by dropping into boiling water, remains colourless in the presence of oxygen for an indefinite length of time, and this according to Bourquelot is due, in the case of the apple, to the destruction of the oxidase responsible for browning. When portions of apple pulp are immersed in very dilute sulphuric or tartaric acids, the pulp turns brown, whereas in stronger solutions it remains colourless. This might be due to the stronger acid inhibiting or destroying the oxidase ferment, or preventing its formation if only present as a zymogen in the living eell. Pulp pounded in its own volume of 1-per-cent. H2SO, remains colourless and gives no guaiacum test and no distinct decomposition of H2O». If pounded and allowed to brown before adding the sulphuric acid, no oxidase reactions are shown after, but active ones before the addition of the acid. Slices of fresh pulp decompose H2O. actively and also turn guaiacum blue. If pulp pounded with 10-per-cent. H2SO, is neutralised with ammonia and tested, it gives no oxidase reactions. Apparently, therefore, the sulphuric acid acts directly by destroying the oxidase present in the living cells. Pieces of apple pulp immersed in poisonous metallic solutions develop a brown colour on drying, and the same is shown whatever the concentration. In 1 and 5-per-cent. solutions of lead nitrate, however, the browning is fainter than usual and is mainly confined to the veins. Lead nitrate destroys oxidase ferments and hence apparently the production of browning and the presence of oxidase are not exactly parallel. The addition of dilute ammonia rapidly turns the pulp a deeper brown, but the immediate addition of dilute H2SO, or HCl restores the original pale colour. Hence the browning produced by ammonia is not quite the same asthe permanent brown produced in slowly dying pulp cells. If the pulp is soaked in dilute ammonia for some hours, acids will not, however, entirely remove the brown colour. When pieces of pulp are soaked in a poisonous solution, a certain time elapses between the first penetration of poison and the death of each cell, and this time-interval will be greater in the case of deeply seated cells than of superficial ones. This is well shown when prepared potatoes are immersed in 5 or 10-per-cent. solutions of lead nitrate. Chromogen oxidation only takes place towards the inner boundary of the diffusion zone. To eliminate the time factor the pulp was rapidly pounded in a mortar with the poisonous solution and then tested with guaiacum and H,O02. French crab apples were used. Catalysts of Organic and Inorgane Origin. 287 Poison. Chore ebanee Guaiacum test. | paetay tae (Winimeaitedteeeccnce:aakeisa se Brown Blue Active. slmpercentwkis lo) ee rcacdeta cs: None Pale blue Feeble. il ip OWSO)n-conesanoaos 08 : Deep blue Strong. | ul Be DANO necro ) No blue None. | 5 5 lelbby\Ol "Soenonioce nos Pale greenish or Pale blue Fairly active. | yellowish brown 1 op JNBINOG Gooveesadaos Blackens Strong blue especially None. with H,O0, 1 H morphine sulphate Brown Pale blue Feeble. 10) 5 strychnine ......... 3 _ 6 | 0) brucin nitrate s i 9 | 2 an Bal CLOW jencacnsacs None i | 2 45 JIEKOLG cetedvosseaueds Rapidly changing Deep blue Active. | to dark brown | Absolute alcohol ............... None Faint blue Doubtful. Tstonllel Frwy) Goosen peaodangs abe Nil Nil Very feeble. | With fresh juice or pulp from potatoes the following results were obtained :— PaeOn Colour change Cudiaeun tent, Decomposition in air. of H,0.. 5 per cent.:lead nitrate ......... Nil Nil Nil. 5 mercuric chloride | ni Blue Very feeble. 5 ‘4 copper sulphate ...| 6 Deep blue Active. WimtageEHeel “.conasceonveedessadoounr Brown Strong blue Active. TRYOTEC Seacddoba honseacraAceonnneanee| Nil Nil Very feeble. At first sight these results seem to show that browning is not closely With absolute alcohol, however, if the pulp is allowed to stand for a short time the oxidase reactions entirely disappear. Apparently the alcohol first weakens and then destroys the oxidase and the oxidation of tannic acid seems to require a more powerful oxidase action than is necessary to produce a blue with guaiacum. related with the presence of the oxidase. Inorgame Oxidases. In addition the influence of the metallic poisons must be taken into account. According to L. Meyer,* salts of manganese such as the chloride and sulphate can act as strong oxidases, and salts of copper, iron, and cobalt have the same power but progressively decreasing, whilst the least oxidase action is shown by salts of nickel, zinc, cadmium, and magnesium. It is not, however, clear as to whether a strong metallic oxidase oxidises all substances . * “Ber. Chem. Gesell.,’ vol. 20, p. 3085 (1887). 288 Prof. A. J. Ewart. Comparative Study of Oxidation by capable of oxidaticn with the same relatively greater intensity than does a feeble oxidase, or whether a strong oxidase to one substance may be a feeble oxidase to another as in the case of the organic oxidases. Other points needing elucidation are as to whether oxidase action is solely due to the metallic base, and what influence a metal such as iron will have when present as an acid. Also as to whether peroxide of hydrogen only accelerates oxidase action when the oxidase salt decomposes it. In the following table the results of a series of tests are given using guaiacum, ursol tartrate,* hydroquinone, pyrogallic acid, gallic acid, tannic acid, and tyrosin as oxidant substances. In order to render possible a comparison of the relative activities in each case, the oxidase was present in one-tenth the molecular concentration of the oxidant, except in the case of the guaiacum, where the alcoholic solution is best allowed to float as a thin layer on the oxidase solution. The exposure to air was continued for one day, and if the colour was still the same as the test solution the result is given as nil. A rapid reaction is indicated by three positive signs (+++), slower ones taking one or more hours to become distinctly perceptible by two signs (+ +), a very slow one taking the full 24 hours by one (+). In the case of guaiacum, owing to the mode of application of the test, the time factor does not enter to the same extent, but a difference in the strength of the oxidase is indicated by the depth of the blue coloration (strong = +++, weaker blue =+ +, feeble blue = +). | of Co j 3 3 3 cS "E | 2g 5 = < Selb mer eI | 6. S) q s @ fl - || 5 eS | | aS oS a2 Ie) ost || a) $x a | SS “3 o8 ‘S oF = ='5 S | ¥ 3 Gy ns | So = 33, u 3 I | ta is | Po oS oS S | So | Pp a | & ido} 4s) SI | } | Pie Cn at TG ak | j ; | Cupric chloride...... ++ af + | Do. +H,0, ...) ++ tet) tet) tte | t+ | ett + | | Cupric sulphate ... | st Do. +H,0, ...| ++ oo] + gpapar || aearep || apap ar + | Copper oxychloride | + | Do. +H,O, ... + | + ++ + + +++ + + | Copper acetate and | + + + | | subacetate | | Do. +H,O, ...) +* | +++ + ++ ++ ++ + + | Ferric chloride ...... +++) +44 oP oP | Do. +H,O.'.... +++ |) +++ ) +++] +++ | +4+47) 44? +? + | Ferrous sulphate ... | Do. +H,O, ...) ++4+f) +4 +4 ++ ++ | +4+P | +P + * Cu30,Cl.,4H,0. + Flat black shining needles separate out on standing in the coJd, but no distinct oxidase action is shown. ft Liquid yellow, then slowly brown precipitate. * Paraphenylenediamine tartrate. Catalysts of Organie and Inorganic Origin. 289 Katalase action. Ursol tartrate. Pyrogallol. Hydro- quinone. Gallic acid. Gallotannic aeid Ferrous chloride ...! + Do. +H,O, ...) +++4%) +4 | +4 ++ a ar Potassium _ ferro- | cyanide | Do. +H,O, ... +t +++ ++ + + Potassium _ ferri- | + | ++ 4 ee |; o+ cyanide DY, FE TELOs oon tet | tet) t44 t+ | +4 | Manganese chloride + | | (MnCl) | Do. + H,O, see! Manganese sulphate (MnSO,) Do.. +H,0z ... t+ | +4 + + + Potassium perman- | | desk or de th ae ee ete ++ | ++ ganate | | | West 1EO cool ap ab se |] Sp ae ay AP ate ap ar ++ ea eie gare! Pelee gear Black oxide of man- | fob + +++ 444 5 + Hy ats tll =| ganese | | Do. +H,O, ...) +++ | +++] +++) ++ ++ , + Chromium chloride | | Do. +H,O, ...| Nil | Chromic acid ...... + Do. +H,O.f bb | + Potassium bichro- + mate Do. +H,O.t +++] +++] +++] +4+4+ t+ | ++ +++ + “Neutral” potas- | + sium phosphate§ Do. +H,O, ...| Nil Nil + | + faut Nitric acid ............ +++ + +++ qeap | ap tp te Do. +H,0, ... Lead acetate ......... Do. +H,9, ... + +44 Seat lay scts SEE AAU ey kek + ave wo + + + + aa + + | + * Ferric salt formed, hence the oxidase reactions with H,O, the same as with ferric chloride. + H,O, converts ferrocyanide partly into ferricyanide. Hence ferrocyanide and H,O, give similar oxidase reactions to the ferricyanide. Similarly traces of ferricyanide appear slowly in a solution of ferrocyanide exposed to light and the liquid becomes a deeper yellow, while ultimately prussian blue separates out. According to Sarthou (‘ Journ. Pharm. Chim.,’ vol. 1, p. 482, 1900), the bark of Schinus molle contains a ferment “schinoxidase” which converts potassium ferro- cyanide into ferricyanide. It is, however, very doubtful in this case that we are dealing with an oxidase ferment at all. { With dilute solutions the colour change with hydroxyl is easily distinguished from the oxidase change by using controls. A mixture of dilute CrO; and H,O, becomes colourless again on long standing. § Made by adding potassium carbonate to a boiling solution of acid potassium phosphate until imperceptibly acid or alkaline to litmus. The foregoing Table shows clearly that an inorganic oxidase is not necessarily a “katalase,” nor a katalase an oxidase, and that hydrogen peroxide may accelerate the oxidase action of substances incapable of decomposing it. In the case of nitric acid, chromic acid and potassium permanganate the 290 Prof. A. J. Ewart. Comparative Study of Oxidation by oxidation is, in part at least, a direct one and hydrogen peroxide diminishes the oxidising action. If hydrogen peroxide is added to very dilute potassium permanganate, a colourless liquid is formed and the evolution of oxygen soon ceases. With stronger solutions the liquid is brown, and contains an oxide capable of continuous oxidase and katalase action. The Tables show further that chromium and iron can act as oxidases when present in the form of acids. In general a feeble oxidase acts feebly on all the substances tested, and the order of sensitivity to oxidases is: guaiacum, ursol tartrate, pyrogallol, hydroquinone, gallic acid, gallotannic acid, tyrosin. There are, however, various exceptions. Thus chromium chloride and manganese chloride give a blue with guaiacum, and copper sulphate does the same in the presence of H202, but all three give direct oxidase reactions with ursol tartrate, pyrogallol, etc. Where there isa strong tendency to precipitation between the oxidase and oxidant as in the case of lead acetate or of ferric chloride and hydroquinone, the oxidase action may be retarded or prevented. In the case of potassium phosphate the feeble oxidase properties are evidently due to the phosphoric acid and not to the potassium. Further, the chlorides, nitrates or sulphates of the same metal are not necessarily equally powerful oxidases, chlorides apparently surpassing sulphates (see copper) and nitrates chlorides. Thus cobalt chloride shows no oxidase properties with or without H.202. Cobalt nitrate slowly browns pyrogallol and hydroquinone in the presence of H2O2 but is inactive to guaiacum, ursol tartrate, and tannic acid. Lead nitrate shows no oxidase action, whereas lead acetate exhibits a peroxidase action. Yellow potassium chromate has similar oxidase properties to potassium bichromate except that it causes tannic acid to brown rapidly in the absence of H202, probably owing to the alkaline nature of the basic chromate. The Nature of Oxidases. The fact that certain plant oxidases contain oxidase metals, such as manganese in laccase, has long been known and certain oxidases, as for instance, tobacco oxidase, can be boiled without being destroyed. Woods* considers that this is due to the oxidase existing as a zymogen from which on cooling the oxidase is reproduced. The supply of zymogen can, however, hardly be unlimited, and, since the boiling can be repeated more than once without destroying the oxidase, it must itself be resistant to heat. According to Bach and Chodat+ the oxidases form three distinct groups of ferments, namely :— * «Bull. U.S. Dept. Agric.,’ vol. 18, p. 17. t ‘Biochem. Centralbl.,’ 1903, p. 141. Catalysts of Organic and Inorganic Origin. 291 (1) Oxygenases, proteins which absorb molecular oxygen forming per- oxides ; (2) Peroxidases, which increase the oxidising power of peroxides and can only act in their presence ; (3) Katalases, which destroy peroxides with an evolution of oxygen. An oxidase which turns guaiacum blue without hydrogen peroxide being added is a mixture of ferments of the first and second class. Thus Bach* considers tyrosinase to be a mixture of a specific oxidase and a specific peroxidase. Moore and Whitley} conclude that all oxidases are peroxidases acting in the presence of a peroxide, such as may be present in certain solutions of guaiacum, or an organic peroxide derived from the juice of the plant tested. The peroxides can be removed from the juices and from guaiacum by adding animal charcoal and filtering, and they are destroyed by heating to 55°-60° C. for several hours. After such treatment potato juice will only give a blue with guaiacum on adding H20p. As a matter of fact the action is merely to attenuate the oxidase, so that the addition of an accelerator such as H202 is necessary to render the oxidation of the guaiacum perceptible. The facts that potato juice decom- poses peroxides strongly, and that the juice and pulp give negative results when tested with decolorised magenta, and show no effervescence until H2O2 is added, are hardly in accord with Moore and Whitley’s explanation, and a study of the preceding Tables shows that, so far as metallic oxidases are concerned, too much importance can easily be attached to the influence of peroxide of hydrogen on oxidase action. Thus in some cases the same substance may be a katalase, an oxidase, and a peroxidase. In other cases the same metallic salt may be an oxidase to one reagent, a peroxidase to another, ineffective to another, and may or may not at the same time bea katalase. Finally, the mere addition of H2,O2 may convert a weak oxidase (“peroxidase”) into a stronger one, which will then act in the absence of H202 (ferrous salts and potassium ferrocyanide). Moore and Whitley found that hydrochloric acid of 1/800 normal concen- tration nearly destroyed potato oxidase, and quite destroyed carrot oxidase, while sodium hydrate of similar concentration had no effect. On the other hand, Na,HPO, had a stronger destructive action than NaH2PO,y The reaction may, however, be prevented without the oxidase being destroyed. Thus the addition of hydrochloric acid (or sulphuric) to ferric chloride removes its power of giving a blue with guaiacum, while tartaric and citric * ‘Berichte, vol. 39, No. 10, p. 2126 (1906). + ‘Biochem. Journ.,’ vol. 4, p. 136 (1909). 292 Prof. A. J. Ewart. Comparative Study of Oxidation by acids hinder or decrease the blue reaction, which is however given strongly even in the presence of 1-per-cent. oxalic acid. Hydrogen peroxide is: partially antagonistic to this action, and it is possible to obtain acidified solutions of ferric chloride which will give a blue when peroxide of hydrogen. has been added, but not when it is absent. The effect probably depends upon the ionic condition of the iron in solution, but the disappearance of an oxidase reaction on the addition of acid does not necessarily mean that the oxidase has been destroyed, any more than ferric chloride is destroyed by the addition of hydrochloric acid. Further, in Moore and Whitley’s experiment with expressed potato sap heated to 55° C. for some hours, the sap becomes more acid, and this might in itself be a sufficient explanation of why an addition of peroxide of hydrogen then becomes necessary to obtaim an oxidase reaction. Further, the case of cobalt chloride and ammonia shows that the addition of alkali to certain plant oxidases might greatly increase their oxidase activity or might convert a non-oxidase combination into an oxidase one. Thus, with a small quantity of ammonia, cobalt chloride forms a green precipitate, slowly oxidising to brown; but with slight access of ammonia a nearly colourless liquid is formed, oxidising to brown from the surface. If the two liquids are diluted, the latter gives a blue with guaiacum directly, and the former an intense blue on adding H:O2; but without HO. no blue is given, or only a faint trace on long standing. In other words, an oxidisable substance can: act as an oxidase, and ammonia, by accelerating the rate of auto-oxidation, | also increases the intensity of oxidase action and converts a peroxidase into an oxidase. According to Porodko,* per salts (-ic) of iron, copper, manganese, and chromium give a blue with guaiacum in the absence of peroxide of hydrogen, proto salts (-ous) only when it is present. This is, however, by no means a general rule. . Ferrous chloride gives a blue without H2O2, but not cupric sulphate or manganic chloride in moderate dilution. Lead nitrate gives no blue in the presence or absence of hydrogen peroxide, while lead acetate gives a strong blue in its presence. Further, the addition of HzO: converts the non-oxidase ferrous sulphate or potassium ferrocyanide into ferric compounds, . which give a blue with guaiacum in the absence of hydrogen peroxide. Oxidase Sensitisers and Inhibitors. Various neutral salts may exert a powerful action on metallic oxidases: either as sensitisers or retardants. Thus an old test for a soluble copper salt . given by Purgotti is: Add salt and pour on top an alcoholic solution of * © Bot. Centralbl.,’ Beihefte 2, vol. 16, p. 1 (1904). Catalysts of Organic and Inorganic Origin. 293 guaiacum—strong blue colour. Hither sodium or potassium chloride will cause copper sulphate to give as deep a blue as in the presence of H2Oz, 7.e. converts it from a “peroxidase” to an oxidase. The action is not solely due to cupric chloride being present in the mixed solution, since the blue is deeper than with cupric chloride alone. The intensity of the action decreases slowly with increasing dilution. It is given by a dilution of 1 erm. of copper sulphate in 100,000 e.c. of water, faintly in a dilution of 1 in 250,000, very faintly by 1 in 100,000, and not at all in a solution of 1 in 10,000,000; ae. guaiacum is about as sensitive in the presence of salt to the oxidase action of copper sulphate as the pulp cells of apples are to its poisonous action. Sodium and potassium phosphates are also able to act as sensitisers to such oxidases as potassium ferricyanide, and the influence of a sensitiser may show with some but not necessarily with all test substances (oxidants). The accelerating action of phosphates is particularly marked with tannic acid, if sufficient is added to leave a clear solution. An excess produces a purplish white precipitate and naturally interferes with oxidation. Acid potassium phosphate is a less active oxidase sensitiser than the neutralised solution of the same salt. Neutral potassium phosphate accelerates the oxidation of tannic acid by potassium permanganate, but not by black oxide of manganese, and it acts as a retardant to those soluble metallic oxidases which it pre- cipitates. Water itself may act as a sensitiser as well as an oxygen carrier. Thus, if potassium ferricyanide is dissolved in pure boiled glycerine and guaiacum dissolved in absolute alcohol added, even after long standing only a faint blue or none at all appears at the junction of the two liquids, which rapidly inten- sifies on adding a little water. Nasse and Fram* have even gone so far as to ascribe the oxidation entirely to the hydroxidation of water without the presence of free oxygen being necessary, but Porodkot has shown that this is not the case. The addition of a neutral solution of potassium phosphate to potassium ferrocyanide does not cause any ferricyanide to appear, but causes it to behave as a weak oxidase to guaiacum, ursol tartrate, pyrogallol and hydroquinone, and as a “peroxidase ” to gallic acid, tannic acid and tyrosin. The action in the presence of hydrogen peroxide is, however, in part due to its partial con- version into ferricyanide. Neutral potassium phosphate intensifies the oxidase reaction of potassium ferricyanide and converts it from a non-oxidase to tannic acid and tyrosin into an oxidase to the former and a “ peroxidase’ to the latter. * ‘Pfliiger’s Archiv,’ vol. 68, p. 203 (1896). t ‘Bot. Centralbl.,’ Beihefte, vol. 10, p. 1 (1904). 294 Prof. A. J. Ewart. Comparatwe Study of Oxidation by | : , z . ma -S ea| & Sill FE Salleme F | Boe 3) g S L 8 & $ a | Se Ss ae G0 2 oe a) Ss # 3 3 a os S | oo = = 5 9 aS 5 a? eSemen p ht tess (er: 3 3 3 ww o =) a a Oo.) e A ernie | Acid potassium phos- phate Do. +H,O, ......... + + Neutral potassium phos- 2 phate 1Dyoy > Gl BL {OSy Gasepcees + + + Potassium ferrocyanide ? ? and acid phosphate IDO, EIEIO) Soancacce + +++ ++ + + + sa Potassium ferrocyanide + ap tr + and neutral phosphate Dos sO; See + ++ ++ ++ ++ +4 ++ + Potassium ferricyanide ++ + + ? ++ ++ and acid phosphate IDYoy, = EEO}, sone se +++] +44 + | + + +4 Potassium ferricyanide +++) +44 ah ar + ap apap |) ar a ae and neutral phosphate Doe We Hs O wade + fears ff] searsp || aeaeap | deere Wl gear sp i) sk sk tp | se | * Yellowish colour due to formation of ferricyanide. To some extent acid potassium phosphate and peroxide of hydrogen are antagonistic in their action on potassium ferricyanide, and hence the oxidation produced when both are added may be no greater or even less than when either is added singly. In the case of tannic acid, the addition of sodium or potassium phosphate seems not so much to accelerate the action of the oxidase as to render the tannic acid more liable to oxidation. The chlorides and bromides of sodium and potassium act as strong sensitisers to certain metallic oxidases but not to others, and even with the former the sensitising action is not the same to all oxidants. By themselves these salts exhibit no oxidase properties with any of the oxidants mentioned. A detailed comparison is given beneath of the influence of potassium chloride upon the oxidase action of relatively inert oxidase salts such as : ibe le |S ; Sy o 1% | ® | = | 6 od = = a 2) 3 a 25 | = 3 |; o ig oS s [=| | Gea | oS ic] Ss od @ So or ay S at tJ} ged 2 aye D 3 2 1 Stet UN Ney a itey | ee a ° eS Se ae ree ees sil aaa lee are os ad taal | m& a a it ts (o) | > Ay | Oo }A 1a | | | | = , i Copper sulphate and sodium chloride ...... ape i) ar ar ar + a | GF Ferrous sulphate and potassium chloride | + + tee + A flel Chromium chloride and sodium chloride + + Wares * Wading again on long standing. + A pale violet colour darkening from the surface possibly owing to oxidation to ferric salts. Catalysts of Organic and Inorganic Origin. 295 copper sulphate, ferrous sulphate and chromium chloride, which act as oxidases in the presence of hydrogen peroxide but not in its absence. Copper sulphate, in the presence of salt and HO», rapidly causes an oxidase browning in tannic acid, a slight browning is slowly produced with tyrosin, and a full colour sequence with ursol tartrate. The oxidase reactions of copper sulphate and H20. with pyrogallol, hydroquinone, and gallic acid are approximately the same in the presence as in the absence of salt. In the presence of H,Oz, salt slightly accelerates the oxidase action of chromium chloride on hydroquinone and pyrogallol, and a deep blue is given with guaiacum, particularly if bromide is used instead of chloride, but a weaker blue if no chloride or bromide is present. In general sodium and potassium bromides are slightly stronger sensitisers than the chlorides, the iodides are less active* and the fluorides still more so or may even exercise the reverse action. Copper acetate, ferrous chloride and potassium ferricyanide, which slowly give a pale blue with guaiacum in the absence of H2Os, give a stronger blue rapidly in the presence of salt nearly as well as when H20z is added. Salt is, however, unable to produce a blue in the absence of H2O2 with manganese sulphate or chloride, with copper oxychloride or with potassium ferrocyanide. . The relative mass of the oxidase and sensitiser is of importance. Thus if equal masses of ferrous sulphate and of KCl, KI, or KF are present, and the solutions fairly strong, no blue is given with guaiacum, but if the ferrous sulphate is present in relatively dilute solution, a rather pale blue is given with KCl, weaker with KBr and KI, and faint or imperceptible with KF. Hence if the oxidase and sensitiser are not present in the proper proportions some oxidase actions may be prevented or overlooked. Copper sulphate, however, even when present in excess gives direct oxidase reactions in the presence of sensitisers, possibly because unlike ferrous sulphate it has no tendency to auto-oxidation. On the other hand, a sensitiser does not act with all oxidase tests. Thus neither KI, KBr, KCl nor KF, whether relatively dilute or concentrated, give ferrous sulphate any oxidase action on ursol. tartrate or hydroquinone. The double fluoride of sodium (NaFHF) inhibits the oxidase action of ferric chloride on guaiacum, ursol tartrate, pyrogallol, hydroquinone and tyrosin, and also its power of decomposing H:02. It strongly retards the oxidase action of potassium ferricyanide and of manganese sulphate and H2O2, and although a blue is still given with guaiacum it is much paler. gradual attenuation, the same oxidase became first a “peroxidase ” and finally a pure “ katalase.” Paraphenylenediamine Test for Oxidase. As is well known, this substance forms an exceedingly sensitive test for oxidases, and goes through a remarkable series of colour changes under their action. The full series of colour changes is green, then blue, then brown, then violet, darkening, and in strong solutions forming a black precipitate, but according to circumstances, or if the oxidase action is very intense or very feeble, one or more of these changes may be omitted or modified. The chief objections to the reagent are the readiness with which decomposition or oxidation takes place naturally and its excessive sensitivity. Gruss* recommends the use of the tartrate of paraphenylenediamine (ursol tartrate). This dissolves readily in water, and a pinch of the dry salt can be dissolved in water for each test. Any colour change in the clear solution is readily perceptible, and there is no alcohol present to interfere with the reaction. Further the dry tartrate keeps indefinitely. It is, however, not so sensitive and responds more slowly. On the other hand, it will often give a full colour series, where the alcoholic solution of paraphenylenediamine gives a single colour change only, which, when slow, may be confused with its natural slow darkening on exposure to ait. Neither alcoholic paraphenylenediamine nor the tartrate respond to all oxidising agents. Thus nitric acid appears if anything to exercise a reducing rather than an oxidising action. It does not produce any colour sequence, and if a little dilute nitric acid is added to potato pulp turned green or blue by ursol tartrate and peroxide of hydrogen, the pulp imme- diately becomes pale in colour. The reactions with those metallic salts capable of turning guaiacum blue are of interest. Silver nitrate forms a grey precipitate, slowly darkening, with alcoholic paraphenylenediamine, but in the presence of hydrogen peroxide the colour sequence, green, brown, ruby, violet is given, and the same is given with silver nitrate and ursol tartrate, whereas in the presence of hydrogen peroxide the change is from green to brown only. Ferric chloride gives the full colour sequence (green, brown, violet, or purple) with alcoholic paraphenylenediamine, * ‘Biologie der Enzyme,’ 1912. Catalysts of Orgame and Inorganic Origin. 309 but in the presence of hydrogen peroxide gives a reddish-brown at once. With the watery solution of the tartrate the colour sequence is also given. The presence of free nitric, sulphuric, hydrochloric, citric, or tartaric acids prevents or delays the production of an oxidation colour sequence with ferric. chloride, but this takes place readily in the presence of 10 per cent. oxalic acid. With soluble copper salts (sulphate, acetate, chloride), alcoholic paraphenylenediamine darkens directly without showing any colour sequence, but if the solutions are dilute, and salt and hydrogen peroxide are present, a partial colour sequence from brown to violet or purpie is shown. With the watery solution of ursol tartrate no colour change is given with copper acetate, sulphate, or chloride in the presence or absence of sodium chloride. With the sulphate and acetate an apparent colour sequence of green to brown is given on the addition of hydrogen peroxide, but this is partly due to the fact that the peroxide gives a greenish colour with copper, and ursol tartrate slowly browns in the presence of hydrogen peroxide. With copper chloride, however, a violet or purple tinge ultimately appears, and a full colour sequence (green, brown, violet, or purple) is given with copper sulphate and copper acetate in the presence of salt and hydrogen peroxide. If, however, the solutions are very dilute the colour change is slow, and is direct to brown. Gruss* suggests that the direct oxidation to brown is due to molecular oxygen, and that the colour sequence is the result of the action of atomic oxygen. The data given above, however, yield no support to this view. The colours produced seem to depend to some extent upon the relative degrees of dilution and intensity of action. Colour may be intramolecular or extramolecular, ze. due to the absorption or modification of light rays at the surfaces of molecules or of molecular aggregates. In the latter case if the peculiar aggregation is broken up when the material is in solution the colour may disappear or be modified. It is quite possible that the colour sequence with paraphenylenediamine is the result of temporary molecular ageregations during the process of oxidation which react differently to light rays, and whose production depends more upon relative mass action than upon any other factor, this determining molecular aggregations-of material in various stages of oxidation. Ursol Tartrate Test for Lignin. This delicate and striking reaction is best shown with boiled or dead tissues by placing them in the watery solution. It is a reaction comparable with the phloroglucin test, and is shown in the absence of free oxygen, acid, %7 LGC, Cilter, J, Whe 310 Prof A. J. Ewart. Comparative Study of Oxidation by or light. The colour given is brown or brownish red. It picks out the wood vessels in a slice of boiled potato or carrot in brownish red without affecting the other tissues. The bundles or vascular network on the inner surface of orange or lemon peel are coloured bright red on a white ground, looking like blood vessels injected with carmine. Conifer wood or a match also colours bright red. As a direct test, it is simpler to apply than any other lignin test, and the colour is confined to the walls of the vessels or tracheides. In testing tissues for oxidase this reaction must be borne in mind. The Action of Paraphenylenediamine on the Oxidases of Apple Sap and Potato. Apple pulp turns violet, in a few minutes rapidly darkening to blue, with alcoholic paraphenylenediamine in the absence as well as in the presence of peroxide of hydrogen. In the former case the blue colour remains permanent for an indefinite length of time, whereas in the presence of peroxide of hydrogen the oxidation is ultimately completed to a dark brown or black. Potato pulp remains colourless with alcoholic paraphenylenediainine for one or more hours, but on long standing the liquid acquires a brown colour, tinged with violet, and the pulp a weak but distinct violet tinge. In the presence of peroxide of hydrogen a violet colour is rapidly produced, but changes to brown in the presence of an excess of peroxide of hydrogen. Using a watery solution of ursol tartrate, apple pulp develops a violet or blue colour in the absence and presence of peroxide of hydrogen, appearing first in the veins and persisting for a long time. With potato pulp and in the presence of peroxide of hydrogen a green colour is shown passing” through blue rapidly to a slate colour. In the absence of peroxide of hydrogen potato pulp remains colourless, gradually acquiring a slight brown colour in two days, but with no signs of any colour sequence, and the brown is hardly stronger than that produced in pieces of boiled egg albumin used as a control. In all cases no colour sequences were produced by boiled apple or potato pulp. In needing hydrogen peroxide to produce a colour sequence with ursol tartrate, potato oxidase therefore resembles copper sulphate and salt, and, similarly, both the vegetable oxidase and the metallic oxidase give a blue colour with guaiacum in the absence of hydrogen peroxide. That is they are oxidases to guaiacum, “ peroxidases” to ursol tartrate. On the other hand, apple oxidase resembles ferric chloride in its ability to: produce oxidase colour changes with both ursol tartrate and guaiacum in the absence of hydrogen peroxide. Catalysts of Organe and Inorganie Origin. Bil The Potassium Iodide and Starch Test for Oxidase in Living Tissues. Moore and Whitley consider that where a plant extract gives a blue with ‘euaiacum without the addition of hydrogen peroxide being necessary, this is ‘due to the production of peroxides by the dying protoplasm during extraction or to their presence in the guaiacum solution. If H2Oz2 is added to a solution of potassium iodide, iodine is liberated and gives the usual blue with starch. On applying potassium iodide to the freshly cut surface of a potato a blue is also slowly formed, as well as with the cut surface of an apple or carrot smeared with starch. Bach and Chodat* consider this to prove that the living cells develop peroxides. If, however, the material is pounded to a fine pulp and potassium iodide applied to the surface no liberation of iodine takes place, and yet in freshly pounded pulp any peroxides produced by drying cells should be more abundant than in a freshly cut surface. Further, the pounded pulp gives a strong blue with guaiacum without the addition of hydrogen peroxide being necessary. With strong potassium iodide, pounded pulp browns, and the starch grains swell but remain uncoloured, although staining readily when free iodine or when hydrogen peroxide is added. As a matter of fact the liberation of iodine from potassium iodide appears to be due to the oxidase present in the tissues used. Ifa slice is boiled and a fresh surface cut no liberation of iodine is shown. Actual tests showed that slices soaked in hydrogen peroxide contained some of the latter undecomposed after again boiling. Certain metallic oxidases such as ferric chloride, black oxide of manganese and potassium ferricyanide will also liberate iodine in a solution of potassium iodide. Hydriodic acid is a substance which readily undergoes oxidation with a production of free iodine, and dilute hydrochloric acid liberates free iodine at the surface of a solution of potassium iodide, giving a blue colour in the presence of starch. Bach and Chodatt have shown that the oxidases in the sap of plants can decompose hydriodic acid, although Asot considers this action to ‘be due to the presence of nitrates or nitric acid. The solution of potassium lodide we may suppose to contain in addition to ions and undissociated molecules of KI also KHO and HI. The latter would be liable to oxidation by organic oxidases when appled on one side of a semipermeable membrane. The action is naturally favoured by the presence of free acid, and is only shown by tissues rich in oxidase. The apple, potato and carrot, which are all acid, give the change readily and the iodine is liberated first over the parts rich in * «Ber. d. D. Chem. Gesell.,’ vol. 35, p. 2464 and p. 3943 (1902). + ‘Ber. d. D. Chem. Gesell.,’ vol. 37, p. 36 (1904). { ‘Bull. Coll. Agric.,’ Tokio, vol. 5, p. 481 (1903). 312 Prof. A. J. Ewart. Comparative Study of Oxidation by oxidase, such as the phloem and cortex ot the carrot, the veins of the apple and potato. A potato kept until the tuber was watery but still acid, and in which the oxidase had nearly disappeared, showed no power of liberating iodine from potassium iodide. The cut surface of the parsnip is neutral or feebly alkaline and, although rich in oxidase, a cut surface shows only a feeble power of liberating iodine from potassium iodide over the phloem ring and outer cortex after many hours’ exposure to air. The acid pulp of the orange and lemon, which contains no oxidases, is unable to produce any liberation of iodine, nor does the wood cylinder of the carrot, which is usually more acid than the cortex but contains hardly any oxidase. This action is evidently due to the oxidase and not to the free acid. The extracted oxidase, however, like pounded pulp is unable to produce any liberation of free iodine from potassium when tested in the usual way. Possibly this 1s because any iodine liberated would at once attack and destroy the plant oxidase where this was in immediate contact with potassium lodide. Free iodine does actually destroy potato oxidase. Hence to produce any progressive liberation of iodine sufficient to stain the starch the oxidase and potassium iodide would need to be separated by a semi- permeabie or colloidal membrane, such as is formed by the cell walls on the cut surface. If pounded potato pulp or filter paper pulp saturated with a glycerine extract of oxidase is covered by a layer of gelatine containing starch or of starch paste, and a little potassium iodide poured on top when the colloid layer has set, after one day a more or less prominent violet line appears on or close to the pulp. Apparently the oxidase is only able to liberate iodine from potassium iodide when the latter diffuses slowly to it, and this is possibly a question of relative mass action and osmotic separation. In any case the liberation of iodine from potassium iodide on the cut surface of a living tissue can be used as a confirmatory test for the presence of an oxidase. It does not indicate the presence of hydrogen peroxide or of any special “ iodoxidase.” The Influence of Anesthetics on Oxidase Action. Ether.—Small cubes of potato soaked in saturated ether water for a day and then exposed to air darkened distinctly. Pulp triturated with ether darkened slightly, and gave strong oxidase reactions and decomposed H»2O». The clear ether extract had no oxidase properties. Apple pulp pounded with excess of ether turns a deep brown, but a little more slowly than in the absence of ether. The ether extract is yellow, not owing to tannin but Catalysts of Orgamc and Inorganic Origin. Bile: to etiolin. The pulp gives strong oxidase reactions, but only decomposes H202 feebly or not at all. If allowed to dry in the air the oxidase reactions are feebler, but the decomposition of HO. a little more active. If the potato pulp ground up with ether is left in contact with it and tested at. hourly intervals, it ceases to give distinct oxidase reactions in the following order :—Ursol tartrate and H2O2, guaiacum, ursol and H2O», guaiacum and H:O.2, decomposition of H2O:. Hence a substance which is at first a “ peroxidase,’ an oxidase and a “ katalase,” as it is attenuated, becomes a “ peroxidase ” and “ katalase,” and finally a “ katalase” only. Chloroform.—Apple pulp pounded with an excess of chloroform turns a yellowish-brown, deepening slowly on exposure to air. The pulp does not decompose H,O.2; it gives feeble or doubtful oxidase reactions, which in the case of guaiacum are rendered more distinct by the addition of H2O2, but not in those of ursol or its tartrate. If the chloroformed pulp is dried in air and powdered up, it regains a weak power of decomposing H202, and shows stronger but still feeble oxidase reactions. Potato pulp triturated thoroughly with excess of chloroform, after the latter had been allowed to evaporate, gave no oxidase reactions with ursol or with the tartrate and H2Os, a pale blue with guaiacum on standing, given at once in the presence of H»Os, and produced a very feeble decomposition of H2,02. The chloroform apparently attenuates or retards oxidase action much more than ether does. Neither chloroform nor ether inhibits the action of metallic oxidases such as copper sulphate and salt, ferric chloride, black oxide of manganese, potassium permanganate, or potassium ferricyanide, but in certain cases. chloroform retards or inhibits the decomposition of hydrogen peroxide. Thus if a mixture of copper sulphate and salt is shaken up with an excess of chloroform a temporary precipitation film hike an exaggerated surface tension. film forms on the surface of the chloroform, and on adding H2O2 an occasional large bubble may form beneath this film, lifting it up lke a skin, but in the liquid above no decomposition of the H2O2 takes place. If, however, the chloroform is removed by evaporation or the liquid warmed to start the decomposition it continues indefinitely. Chloroform itself does. not decompose H.O», and saturation with ether slightly lessens the decom- position without arresting it. Similar results were given with ferric chloride, except that the action of the ether is stronger, and if the ether or chloroform is removed by boiling the liquid becomes reddish-brown and loses the power of decomposing H2O2, whereas if removed by evaporation at a low tempera- ture the power of decomposing fresh hydrogen peroxide is regained. Chloroform added to potassium ferrocyanide and H:O. merely changes a rapid stream of small bubbles into a slow stream of occasional larger 314 Prof. A. J. Ewart. Comparative Study of Oxidation by bubbles, and still less influence was exercised upon the decomposition produced by potassium permanganate and black oxide of manganese, although in the latter case some remarkable surface tension effects were exercised. Hydrogen peroxide is readily soluble in ether, which will in fact remove it from a watery solution. It is only sparingly soluble in, chloroform, for, although the chloroform solution wili not give any blue with chromic acid, it gives a feeble reaction with a watery solution of starch and potassium iodide or with ferrous sulphate. Chloroform prevents the ether-chromic acid reaction for hydrogen peroxide being given but not by destroying the hydrogen peroxide. In fact the hydrogen peroxide can be shaken with chloroform and the latter then boiled off without the former being destroyed. The retarding action on peroxide decomposition produced by ether might depend upon whether the “ katalase” salt dissolves in it as well as in water, since otherwise the hydrogen peroxide might be removed from katalase action except at the contact surface. In the case of chloroform any bubbles produced form mainly below the surface tension film, although both hydrogen peroxide and the katalase salt may be present in abundance in the liquid above. The chloroform apparently acts as an “anesthetic” to “katalase” chemical action. The Oxidases of the Lemon and Orange. According to Moore and Whitley there are no “ peroxidases” in the pulp or rind of these fruits. ‘This is hardly the case, as no allowance was made for the effect of the acid in the pulp or of the oils in the skin. Quarters of the pulp were squeezed dry in a press between blotting paper, and collected until a sufficiency of clean matertal free from acid was obtained. This gave no reaction with guaiacum alone and none with ursol tartrate except that the fragments of the tracheee coloured brownish-red. On adding a drop of peroxide of hydrogen a pale but distinct blue was given with guaiacum and a slow change to violet with ursol tartrate. The pounded pulp does not decompose hydrogen peroxide appreciably. On dissecting out the vascular bundles and applying ursol tartrate and peroxide of hydrogen, all the veins right down to the stalks of the endocarpal hairs turned green, then brown, then violet, but no other parts. They also showed a feeble power of decom- posing HO». After soaking in orange or lemon juice for some hours or after boiling no oxidase reaction was given but the walls of the trachee gave a bright red lignin reaction, making the bundles look like blood-vessels injected with carmine. Catalysts of Organic and Inorganic Origin. 315 The Oxidase of the Carrot. According to Moore and Whitley the “peroxidase” of the carrot is most abundant in the protoxylem. They were either misled by the lignin reaction or mistook the central wood cylinder of the carrot for pith. With both guaiacum and ursol tartrate in the presence of H2O. the oxidase reaction is given first in the cambium, cambium segments and phloem. The central wood cylinder is the last part to show any true oxidase reaction. All parts decompose H,Os, and in ursol tartrate alone a green colour slowly appears along the line of the cambium, while the vessels in the wood within colour reddish-brown. The latter colour appears in a boiled section but not the former. We are evidently dealing with a somewhat weak oxidase, most abundant in the cambium and phloem, next in the outer cortex and least of all in the central wood cylinder. According to Moore and Whitley the cut surface of a carrot rapidly develops peroxides and will then give a blue with guaiacum without any addition of H2O»,. It is difficult to see how this explanation can apply to a tissue like that of a carrot which rapidly decomposes peroxides, or at least peroxide of hydrogen, or to a section of carrot immersed in a large quantity of a watery solution of ursol tartrate, in which the reaction is slowly given by the uncut cells beneath the surface, and where any peroxides formed in the uninjured cells would be washed away. Actual tests failed to detect any peroxide of hydrogen in living or dead carrot tissue. The Oxidase of Red Beetroot. In spite of its red colour the expressed sap of the beetroot shows a strong reaction with guaiacum, but is difficult to use with other oxidants. Hence the sap was squeezed out, the pulp washed, the excess of water squeezed out, and the residue pounded with glycerine, the first portion of which was thrown away. In this way a pale pink strongly active oxidase was obtained, which closely resembled potato oxidase. It reacted to guaiacum and tyrosin in the absence, but to ursol tartrate only in the presence, of hydrogen peroxide. It has no action on tannic acid by itself and only a feeble one in the presence of sodium phosphate. It has a weaker power of decomposing hydrogen peroxide than potato oxidase and the peroxide appears to inhibit its action on pyrogallol, but acts as a sensitiser in the case of ursol tartrate. The pounded pulp reacts strongly and rapidly to ursol tartrate in the presence of hydrogen peroxide but only slowly and faintly or not at all in its absence. Neither the pulp nor the expressed sap appears to contain any perceptible amount of chromogen capable of oxidation. VOL, LXXXVIII.—B. i 316 Prof. A. J. Ewart. Comparative Study of Oxidation by According to Bertrand* the sugar-beet contains an oxidase capable of oxidising tyrosin which he terms “tyrosinase,” and this, according to Gonnermann,f oxidases tyrosin to homogentisinic acid, which darkens rapidly by direct oxidation to red, brown, or black. In the red beet the amount of tyrosin present appears to be too small to appreciably affect the neutral red colour. It may undergo oxidation, while the plant is living, and hence be unable to accumulate. ; The Oxidase of the Parsnip. The parsnip differs from all the other vegetables used, in that a cut surface is neutral or faintly alkaline instead of acid, and it resembles the carrot in containing no chromogen oxidising on death. Neither carrot nor parsnip oxidase will directly brown boiled potato or apple pulp, but if a little sodium phosphate and H.O2 is added, they will cause tannic acid, apple pulp and apple juice to brown distinctly and with fair rapidity. In both carrot and parsnip the oxidase is mainly present in the phloem ana _ outer cortex, and that of the carrot appears to be a little more abundant. Hence of similarly prepared watery or glycerine extracts the former isa little more active than the latter. | | Beetroot oxidase.| Carrot oxidase. | Parsnip oxidase. | | (EqDOC Hid scoosaosa00 and sonaonoduaee abdso0000 a op oF ++ ah se DOs, “(HES O ON Fe otk sec canaataceenees +++ +++ | +++ Uinsoliitantrate leer eee- creer cree eeeecreeen | Wor ses Os a latenceaesesenesscctnnet + + | + IPROGANO .conadooucoobaceseonopoea9054000000 | ap 3e | + i Doren oi Opa tencesardseancnasnbaocasecd| + +++ | +44 | Tabs @bYoe(UDNOIN® 34 nnqnaodonasarnboooroberonor | | Dos, ASH OSM saan ceeaccsente ctrl + + + Gallltieracidl, toh atepascrcateceine seskitcn ahapeeee | Dor WASH O rite ssucnwusansontcnseess | + | + | ap tb at Manni CAC dy Fact cepeuctsacssetsate seas 1D oye gece S GO) aia oun arencnanadadacdoos6e Tannic acid and sodium phosphate ... 1D oie on o (D Paenencepnapesdeeneeasonaa dae | | + + IN ROSS IAV.cclaae oon nonbeBoanbosinage se domocauce ++ DG.) oT OSY ssiscurcesey dente eeeemeeeeces + Both ursol tartrate and hydroquinone when applied to a cut surface of a carrot or parsnip show signs of oxidation particularly over the phloem ring. This is probably because the oxygen is more concentrated at the surface and also a greater mass action is exercised upon the inwardly diffusing oxidant. No assumption of the production of peroxides in the tissue is necessary to explain the action. The addition of magnesium sulphate or of potassium. * ‘Bertrand, ‘Compt. Rend.,’ vol. 122, p. 1215. + Gonnermann, ‘Pfliiger’s Archiv,’ vol. 82, p. 289 (1900). I is Catalysts of Organic and Inorganic Origin. 317 phosphate to carrot or parsnip oxidase causes it to give a faint trace of browning in one day with tannic acid which is not increased by the addition of hydrogen peroxide. None of the other salts which could be made up from the ash constituents (excluding iron salts) exerted any sensitising oxidase action. The oxidase of the beetroot and potato appear to belong to one class © (betase, potatase, dahliase, russulase), those of the carrot and parsnip to another, while the chief peculiarity of apple oxidase, namely, the readiness with which it oxidises tannic acid, appears to be due to the presence of a sensitiser such as potassium phosphate. Apple oxidase appears to have some resemblance to a weak form of laccase which is also able to oxidise tannic acid. Summary. Plant oxidases form a class of substances of great importance in plant metabolism, but which are known merely by the reactions they cause, and whose exact nature is quite unknown. According to Bach and Chodat they form three distinct classes of ferments namely :— (1) Oxygenases, substances which absorb molecular oxygen forming per- oxides. (2) Peroxidases, which increase the oxidising power of peroxides and can only act in their presence. (3) Katalases, which destroy peroxides with an evolution of oxygen. It has long been known that certain of the reactions supposed tc characterise oxidase ferments could be produced by certain inorganic metallic salts.* As the result of the detailed investigation of the oxidase action of various metallic salts of copper, iron, chromium, manganese, lead, etc., upon guaiacum, paraphenylenediamine, hydroquinone, pyrogallol, gallic acid, tannic acid, and tyrosin, the conclusion has been formed that the correspondence between the action of organic and of inorganic oxidases is extremely close. It was also found that the oxidase action of a metallic salt varies according to its acid combination, and that in the case of certain salts, such as sodium or potassium ferrocyanide, ferricyanide, phosphate, or chromate, the oxidase action was due to the acid and not to the base. Jn addition, oxidase action may be accelerated in the presence of sensitisers such as the chlorides or phosphates of sodium or potassium, or retarded or prevented by a variety of antagonisers. The addition of a sensitiser may cause a “ peroxidase” to act in the absence of hydrogen peroxide. This applies to both organic and inorganic oxidases, and determinations of the minimal amounts of metallic * Bertrand, ‘Compt. Rend.,’ vol. 122, p. 1032 (1896). bo > bo 318 Prof. A. J. Ewart. Comparative Study of Oxidation by oxidases required to produce progressive oxidation in the presence of a sensitiser indicate that their action can be considered as closely akin to that of any enzyme. H. E. and E. F. Armstrong* have shown in a series of valuable papers, and particularly in the hydrolysis of raffinose by acids and enzymes, that a close correspondence exists between the action of organic and inorganic hydrolysing agents. The same appears to hold for organic or inorganic oxidases. In general, oxidases, whether organic or organic, may vary from strong to weak. The former will cause direct oxidation from the oxygen dissolved in a watery solution. The latter will transfer oxygen from labile oxygen compounds such as hydrogen peroxide, or will use dissolved oxygen in the presence of sensitisers such as the chlorides or phosphates of sodium or potassium. Various intermediate grades of activity are shown. ‘There is no reason for separating oxidases and peroxidases as distinct classes of ferments, and peroxides do not necessarily take part in all oxidase actions, although water does. The supposed separation of oxidase and peroxidase by fractional precipitation with alcohol may be merely the result of attenuation. An oxidase may be a “ peroxidase” to certain oxidants or may become so when attenuated. Metallic oxidases act as ferments in that a small amount may produce considerable oxidation, especially in the presence of sensitisers such as salt with copper sulphate, sodium phosphate with potassium ferricyanide, etc., and in that the oxidase appears to act as an intermediary in the chemical change. Hydrogen peroxide may influence oxidase action :— (a) By providing a supply of labile oxygen. (b) By converting a feeble oxidase into a strong oxidase (ferrous salt into ferric, ferrocyanide into ferricyanide). (c) By acting as a sensitiser to the oxidant substance. (ad) By acting as an inhibitor or antagoniser in some cases. Various salts may act as sensitisers (sodium and potassium chlorides, bromides, and phosphates) or as inhibitors (barium chloride, sodium fluoride, organic or inorganic acids), and in some cases, with increasing concentration, the action of the former is reversed, while a substance which is a sensitiser with one oxidant may act as a reducing agent with another (copper sulphate and salt on indigo carmine). Strong metallic poisons will arrest the action of organic oxidases or destroy them (apple, potato, carrot, parsnip) if immediate contact or rapid * * Roy. Soc. Proc.,’ B, vol. 82, p. 349 (1910); vol. 80, p. 312 (1908), ete. Catalysts of Organic and Inorganic Origin. 319 penetration is assured. Hence the organic oxidases are possibly proteids, with or without oxidase metals, in basic or acid combination. There is no justification for the use of such terms as “peroxidase,” “katalase,”’ “cenoxydase,’ or “ tyrosinase,” to indicate specific substances, ferments, or groups of ferments. The “tyrosinase” of the potato is also a “katalase,” a “peroxidase,” a “pyrogallase,’ a “hydroquinonase,” and a “ paraphenylendiaminase.” It is, however, permissible to use such terms as katalase action or peroxidase action, and such names as laccase, russulase, potatase, carrotase, etc., as temporary names to indicate the origin of the substances, whose chemical nature is yet unknown. Since, however, their > ¢¢ oxidase powers will be only one of many properties, it will never be advisable to name them according to these properties alone, any more than it would be in the case of the metallic oxidases. Comparison with metallic oxidases shows that we are not even on safe ground in assuming the existence of specifically distinct classes of plant oxidases, such as phenolases, aminoxidases, and iodoxidases. The chlorides and phosphates of potassium and sodium are able to act as oxidase sensitisers, and thus may influence special oxidations, or respiration in general. It is possible that they may exert a stimulatory or controlling action on plant metabolism, and that the sodium chloride always present in the ash of plants may not be an entirely useless constituent. This may explain partly why small doses of salt stimulate the growth of many plants, and why phosphates, in addition to being food substances, may act as stimuli to growth. The stimulating action of many metallic salts on growth may be partly due to their oxidase action. Ursol tartrate turns lignified walls red or reddish-brown. This is not an - oxidase reaction, but is an admirable test for lignin, especially valuable for demonstrating the wood elements in pulpy tissue. Chloroform strongly, and ether more feebly, retard or inhibit katalase action, but they do not suppress oxidase action. After prolonged contact, however, the organic oxidases are slowly attenuated and destroyed. The liberation of iodine from potassium iodide may be used as a test for the presence of oxidases in living tissues, but does not indicate the existence of any power of producing peroxides. Dried organic oxidases may retain their properties for three weeks or more, and a glycerine extract for five or more months. Where organic oxidases are destroyed by boiling, this is probably the result of proteid coagulation. The oxidases of the beetroot and potato appear to be related to one another, and to be among the strongest plant oxidases, and the nearest analogies to them are perhaps afforded by ferric salts and ferricyanides. If 320 Messrs. J. McIntosh and P. Fildes. the special action of apple oxidase on tannic acid is due to the presence of a phosphatic sensitiser, it would be a feebler oxidase of the same type. Carrot and parsnip oxidases are a grade feebler, but still react to guaiacum in the absence of a peroxide. Malt diastase is still weaker, and papain feebler still, while pepsin may show a weak “ peroxidase ” reaction with guaiacum, but not any other oxidase action. The Fixation of Arsenic by the Brain after Intravenous Injections of Salvarsan. By James McInrosu, Beit Memorial Research Fellow, and PAUL FILDEs, Assistant Bacteriologist to the London Hospital. (Communicated by Prof. W. Bulloch, F.R.S. Received July 8, 1914.) (From the Bacteriological Laboratory of the London Hospital.) During the period of probation of salvarsan as an anti-syphilitic remedy, a number of toxic phenomena were reported which led to the belief that this drug had particular neurotropic properties, and was therefore to be used with the greatest circumspection. These fears were very largely founded upon the well known effect of the related drug atoxyl in producing optic atrophy. Subsequent experience has, however, shown that the supposed neurotropic action of salvarsan was due to certain technical errors in its administration. In 1911 we published(1) an observation which combated the view that salvarsan had neurotropic qualities. We submitted the organs of an infant who died after administration of salvarsan to Dr. W. H. Willcox for analysis, and he reported to us that the brain in this case contained no arsenic, although considerable quantities were present in other organs. We then applied the law of Ehrlich, “corpora non agunt nisi fixata,” and argued that, since the brain was free from arsenic, salvarsan could have no neuro- tropic action. Exactly similar conclusions were arrived at by Ullmann in 1913(2). In the course of a very extensive investigation upon the distribution of arsenic in the body after salvarsan injections, he made it quite clear that the brain never contained more than traces of arsenic, and “this fact was evidence against the neurotropic action of salvarsan.” Similarly, Morel and The Fixation of Arsene by the Brain. 321 Mouriquand (3) found the brain and cord in five animal experiments to be free from arsenic, although other organs contained much. They also concluded that salvarsan has no neurotropic effect. On the other hand, Mouneyrat (4), after the injection of salvarsan into animals, found arsenic — “in very appreciable quantities in the liver, muscles, and brain, distinctly more than the infinitesimal amounts met with in the control animals, which had had the same food but no injection.” Thus Mouneyrat considered that salvarsan was “particularly neurotropic.” In criticism of this paper it must be stated that the author gives no evidence of having made quantitative estimations, by which he might have discovered that the liver and kidney, for example, contained vastly more arsenic than the brain. Further, the fact that he was able to demonstrate arsenic in the brains of normal animals must lead to a certain suspicion of his other results. The observations of Ullmann, Morel and Mouriquand and ourselves were aecepted by Ehrlich (5) as showing that salvarsan has no “ Vorliebe” for the brain. As a result of subsequent investigations upon the minute anatomy of the cerebral vessels, however, we were not entirely satisfied that the deductions we had drawn were correct. It appeared that absence of arsenic from the brain might be due to two distinet factors. Firstly, the arsenic might not be “fixed” by the brain as already suggested, or, secondly, it might not gain access to the brain, owing to some peculiarity of the cerebral vessels. In order to test this possibility we conducted experiments in vitro, applying neosalvarsan directly to fresh brain substance, and then found that neosalvarsan was in fact “fixed” by the brain. Experiment 1.—100 grm. of fresh human brain and 100 grm. of human liver were minced and placed in two separate glass bottles of 1000 cc. capacity. These bottles were filled with saline solution, shaken, allowed to stand, and the supernatant fluid removed. The tissue was thus freed from excess of blood; 500 cc. of saline solution, containing 0°15 erm. of neosalvarsan, were then added to each, and the bottles were shaken for one hour. At the end of this time the supernatant fluid was removed by decantation and the bottles filled with saline solution eight times, shaking on each occasion and removing the washings by decantation. By this method it was hoped that all “unfixed” neosalvarsan would be removed from the tissues. Estimations of the amount of neosalvarsan in the washed tissue, the supernatant fluid after fixation, and the final washing fluid were then made. In every case the material to be tested was heated with concentrated nitric 322 Messrs. J. McIntosh and P. Fildes. acid and sulphuric acid, the residue extracted with water and tested in the Marsh apparatus. The technique employed was that advocated by Chittenden and Donaldson (6). The rings of arsenic obtained in this way were compared with a standard series of rings made from known weights of neosalvarsan, and the amount of arsenic was expressed as grammes of neosalvarsan. In no case was any attempt made to obtain accurate quantitative results, and our figures merely indicate roughly whether a particular sample contained much or little neosalvarsan :— Gramme of neosalvarsan. Moist brain residue after washing, 100 grm. ............ contained 0-004 Supernatant fluid after fixation with brain, 500 c.c. be 0-1 uma) swashime slug nace seececn eee ee eee ee ee eeree ee aeeree a 0 Moist liver residue after washing, 100 grm................ 5, 0:015 Supernatant fluid after fixation with liver, 500 c.c.... 3 0-1 TDN ASI MAMEM AUC Sid ooccooddocanaaomaanooboaosanoecdsoaaas s 0 From this experiment it appears that brain substance will fix considerable quantities of neosalvarsan i vitro. Experiment 2.—Repetition of Experiment 1. Gramme of neosalvarsan. Moist brain residue after washing, 100 grm. ...... contained 0-01 Supernatant fluid after fixation, 500 e.c.......... is 0:05 Bina avyasininostuid eee eeee eee aeerer ee ee eeaeee eS 0 Moist liver residue after washing, 100 grm. ...... - 0-01 Supernatant fluid after fixation, 500 cc.......... eh 0°05 NTA AVIVA AGGIE o5ancdinsadabcnoboAoppadocece son ME 0 Experiment 3.—Repetition of Experiment 1, but twice the quantity of brain and liver used, viz., 200 grm. The results were similar to those obtained in the other experiment, although the brain substance appeared to fix more neosalvarsan than did the liver, thus :— Gramme of neosalvarsan. Moist brain residue, 200 grm. contained ....,....... 0:02 Moist liver residue, 200 grm. BAM Sie Meee ten Act 0:004 It may be objected to these experiments, that the arsenic was merely entangled mechanically with the minced tissues and was not chemically fixed ; we therefore repeated the experiment after replacing the animal tissues with fragments of Doulton filter (asbestos) and finest vegetable charcoal. The Fixation of Arsenic by the Brain. 323 Experiment 4.—50 grm. of broken-up filter candle were placed into 250 c.c. of saline solution and 0075 grm. of neosalvarsan added. Further technique as in Experiment 1. Result.— Gramme of neosalvarsan. Moist filter residue after washing, 50 grm.......... contained 0 Supernatant fluid after fixation, 250 c.c.......... e 0-04 Biinvcalarras mimo tine Seca . Catnonacclacisin ccc eee ¢ 0 Experiment 5—Repetition of Experiment 4, but with finest Venetian charcoal instead of Doulton filter. Air was removed from the charcoal by boiling and the use of the vacuum pump—0°15 erm. of neosalvarsan were added. Gramme of . neosalvarsan. Moist charcoal residue after washing, 50 erm.... contained 0-001 Supernatant fluid after fixation, 250 c.c. ...... is O11 Mirialewas Win CoM ike cats ct eamtonceoe snes sivnsiat an 3 0 Although a certain degree of absorption of the neosalvarsan was observed in the case of charcoal, it was not so great as with liver and brain tissue, and this fact, taken with the absence of fixation by the filter, suggests that the presence of arsenic in the latter after washing was a true combination and not a purely mechanical phenomenon. The conclusion may therefore be drawn that the absence of arsenic from the brain after intravenous injection of salvarsan is not due to a lack of affinity between the drug and the brain, but to an inability of the drug to penetrate into the proper brain substance. As is well known, arsenic is not usually found in the cerebrospinal fluid after intravenous injections of salvarsan, and this has been suggested as an explanation of the lack of effect of this drug upon certain syphilitic affections of the brain. On other data, however, we have formed the opinion that the inability of arsenic to penetrate the brain has no relation to its absence from the cerebrospinal fluid, but is due to a peculiarity of the cerebral capillaries. If, howevér, neosalvarsan is introduced directly into this fluid by lumbar puncture, then, as shown by Wechselmann, marked toxic symptoms appear. Experiment 6. To Demonstrate the Toxity of Neosalvarsan to the Nervous System.—In a typical experiment two rabbits were used in addition to a control. 0:05 grm. of neosalvarsan were dissolved in 10 c.c. of saline solution, and 324 Messrs. J. McIntosh and P. Fildes. of this solution 0:2 ¢.c. were injected into the cerebrospinal fluid of a rabbit weighing 500 grm., through the posterior occipito-atlantoid membrane. The effect of the injection was to produce convulsions immediately, and death the next day. A second rabbit (800 grm.) received 0:1 c.c. and was found to be paralysed on the following day. As a control, 0°2 c.c. of saline solution was It thus follows that about 0°0005 germ. of neosalvarsan is the toxic dose to a rabbit of 800 grm. weight when injected into the cerebrospinal fluid. It may be assumed that the drug is rapidly absorbed from the cerebrospinal fluid into the brain and is there fixed as in our experiment conducted i vitro. Thus salvarsan may be said to be as much “organotropic” to the brain as to the liver, but this effect is not | apparent after its therapeutic administration in the ordinary manner. The inability of salvarsan to penetrate into the brain explains the lack of success which often attends the treatment of syphilitic lesions of the brain, and in particular the parenchymatous varieties (dementia paralytica and shown to be innocuous. tabes dorsalis). We next considered whether this inability might be due, in some measure, to the rapid fixation of the drug by the liver and kidneys and its removal from the blood. We have indeed found, experimentally, that the blood is practically free from arsenic two days after an injection, as shown in the following experi- ments Experiment 7—Four rabbits were injected intravenously, each with They were killed 3, 6, 24, and 72 hours after Specimens of blood were collected and the animal transfused 0:15 grm. of neosalvarsan. the injection. with saline to remove the blood from the organs. The blood and organs con- tained the following quantities of neosalvarsan :— Rabbit A, Rabbit B, Rabbit C Rabbit D. 3 hrs. 6 hrs. 24 hrs 72 hrs. eats ae —- “ | | Grammes of neosalvarsan. | Brain (ca. 10 grm.) ......... | ? trace ? trace 0) 0 | Ibe (UO Gre) sooaranca sab o5e 0 0005 0 -00075 0 0012 0 -00025 | TBool (5 CG.) soowoneocveavee 0-008 @) 9006 0 0004 (0) We next endeavoured to satisfy the affinity of the liver and kidneys for arsenic by repeated injections of neosalvarsan, in the hope that a sufficient content would be maintained in the blood to allow of penetration into the brain. Experiment 8.—Three rabbits were injected, each twice daily for four days ce iw) LI The Fixation of Arsenic by the Brain. with 0:05 grm. of neosalvarsan. They were killed 12, 36, and 60 hours after the last dose and transfused as before. The following results were obtained :— Rabbit A, Rabbit B, Rabbit C, 12 hrs. 36 hrs. 60 hrs. Brain (car lO'prmt) 1.1). ...cc0eecse ee | 0 -0001 | 0 00004. 0 Piers Cl Ovenras yc: eee aneneeeaen | 0 0005 | 000075 0 00075 cod (ore ne en | 00003 | 0-0001 re eg Experiment 9.—Repetition of Experiment 8, but all the animals killed 24 hours after the last dose. Rabbit A. Rabbit B. Rabbit C. Grammes of neosalvarsan. Thieme, (ez, UO Groen) Jsconenodccaseococcsd (0) (@) (0) Lies (IO gam)" cpeacueasseqnsoedb eeeanooe 0 90012 | 0 :0001 0 0005 1Bilowye| (5), @5G:)) “aooscgcoo oovasuecnoocongneg 0 0-001 (0) From these experiments it appears that no definite penetration of the ‘drug into the brain can be obtained by a repetition of the injections, although in one case a small quantity was observed 12 hours after the last dose. Similar results were obtained with Prof. Ehrlich’s new copper-salvarsan ‘combination (/3). Experiment 10.—Technique as in Experiment 8. The dose of ks was 02 erm. and the animals were killed 24 hours after the last injection. Rabbit A. Rabo Ban | Grammes of neosalvarsan. Brain (ca. 10 grm.)............ @) a) Ibias (IO) fanane)) Gcaboteracseoae 0 -00002 0 0005 181 loaysl: (C45) C-@)))ssocesnoandedbacc: 0 6) Conclusions. 1, After intravenous injections of salvarsan and neosalvarsan in man and animals no arsenic can be found in the brain. 2. This phenomenon is not due to a lack of affinity between the brain and the drugs, but to an inability on the part of the drugs to penetrate into the substance of the brain. ii 326 Dr. A. E. Everest. 3, Fixation of arsenic by the brain occurs as readily as by the liver, as shown by experiments in vitvo and the toxic effects of intrathecal injections. 4, Penetration of neosalvarsan into the brain cannot be obtained even by frequently repeated intravenous injections. REFERENCES. 1. J. McIntosh and P. Fildes, ‘Syphilis from the Modern Standpoint,’ p. 220, London, 1911. 2. K. Ullmann, ‘ Archiv f. Dermatol. u. Syph.,’ vol. 114, p. 511 (1913). 3. A. Morel and G. Mouriquand, ‘Lyon Méd.,’ vol. 120, p. 315 (1913). 4. A. Mouneyrat, ‘Compt. Rend. de Académie des Sciences,’ vol. 154, p. 284 (1914). 5. P. Ehrlich, ‘Abhandl. tiber Salvarsan,’ vol. 3, p. 546 (1913). 6. R. H. Chittenden and H. H. Donaldson, quoted by A. W. Blyth in ‘ Poisons,’ p. 568,, London, 1895. The Production of Anthocyanins and Anthocyandims.—Part IL. By Artuur ERNEST Everest, M.Sc., Ph.D. (Lecturer in Chemistry, University College, Reading). (Communicated by Prof. Frederick Keeble, F.R.S. Received July 15, 1914.) The author of the present paper* and other investigators} have coneluded as the result of their investigations that the red pigments obtained as the result of careful reduction of the yellow flavonol derivatives are identical with the natural anthocyans of plants. Prof. Willstatter, on the other hand, holdst that these artifically produced red pigments are different from the natural anthocyans.§ That the point at issue is important may be judged from the facts that, the complete synthesis of flavonol derivatives having been realised by Kostanecki, the production of anthocyans from them as described by the author (doc. cit.), Watson and Sen,|| and Combes,{1 completes the synthetic production of these compounds. Moreover, the production of anthocyan glucosides—anthocyanins * © Roy. Soc. Proc.,’ B, vol. 87, p. 444 (1914). + Of, ibid., p. 446. { ‘Sitzungsber. K. Akad. Wiss. Berlin,’ vol. 12, p. 402 (1914). § Since the present paper was forwarded for publication, Wheldale and Bassett, ‘J. Genie? vol. 4, p. 103 (1914), have published a criticism of the author’s views, in which they also take up a similar attitude. || “Chem. Soc. Journ.,’ vol. 105, p. 389 (1914). | ‘Compt. Rend.,’ va 157, pp. 1002 and 1454. The Production of Anthocyanins and Anthocyanidins. 327 —from yellow glucosides of the flavonol series by reduction in acid solution as described by the author (Joc. cit.), is of vital interest to those studying the production of these pigments in plants. Willstitter bases his criticism of the author’s conclusions upon the following observations :—(1) The red pigments obtained by reduction are unstable, becoming decolorised even in acid solution. (2) The decolorised solutions of these pigments do not recover their colour on acidification, even with considerable excess of acid. The anthocyans do not behave in this manner, they are stable in acid solution, and, in neutral solution—with a few exceptions—become decolorised by isomerisation, but the decolorised solution on acidification regains its colour. The author has carefully examined the red pigments obtained by him, and compared them with cyanin. The glucosides and the one non-elucoside examined gave results in every way similar to those given by cyanin and eyanidin respectively, and support the conclusions reached in the author’s previous paper (Joc. cit.), viz.: that these pigments are true anthocyanins and anthocyanidins. It is perhaps necessary to point out at this stage that, for the comparisons made by Willstatter and now to be discussed, the anthocyanins—g¢lucosides— give far more reliable results than the anthocyanidins—non-glucosides— which latter compounds are more unstable in solution, even when faintly acid, than the corresponding anthocyanins. This fact was first observed when working with cyanin; moreover, the decolorisation by isomerisation and return of colour by the action of acids is far more difficult to carry out satisfactorily with the anthocyanidins, particularly if they are somewhat impure, than with the anthocyanins. With the anthocyanins this series of changes is most characteristic and very easily carried out, even when the pigment is in quite a crude and impure condition. As Willstitter appears to have used the non-glucosides only in making his comparisons, the foregoing considerations may perhaps explain the il eaies | | WAN Nou aed NAN +H,0 \WAc/ Son On 7 reduction. (jy ‘C 2 OH i" O OH H H (Myvriectin.) Colourless or faintly coloured Anthocyan pigment. Flavonol derivative. intermediate product. already put forward by the author* to explain the formation of the * Loe. cit., p. 449. 330 Dr, A. E. Everest. anthocyans from flavonol derivatives. Willstatter agrees with the author with regard to the structure of the anthocyan molecule, and further, his analytical results (doc. cit.) have shown that the colourless iso-forms are produced by the addition of one molecule of water to the anthocyan molecule ; this is in agreement with the above scheme. As flavone derivatives, which also occur in plants, yield somewhat similar red pigments, it seems probable that further examination of the natural anthocyans may yield compounds related to this class of pigment. Again, no natural anthocyan pigment has as yet been examined in which the hydroxy- groups in the phenyl group attached to the benzo-pyran residue are in the meta-position to one another. As morin, a flavonol of this type, occurs naturally, and on reduction yields a red pigment, it is possible that natural anthocyans of this type may be among the fruits of further research on these pigments. It is quite likely that such compounds would differ slightly from those at present investigated. Apart from the anthocyan question, the observations recorded. appear to be of interest, in that, as in every case where the author obtained an anthocyan by reduction of the flavone or flavonol present in flower extracts, he was able to show that the pigment was an anthocyanin—glucoside—it would follow that in each case the flavone or flavonol derivative in the plant must have been present in the form of a glucoside. Experimental. The methods of obtaining anthocyanins and anthocyanidins from flavonol derivatives by reduction have already been described by the author (Joc. cit.). The method which gave most satisfactory results, viz. : reducing the compound dissolved in a.mixture of 5 vols. absolute alcohol and 1 vol. concentrated HCl by means of magnesium (ribbon or turnings), was then used only when pure or nearly pure yellow compounds were obtainable. By slight adaptation, however, it has been found possible to use it with advantage in many eases when dealing with plant extracts. It is only necessary to add to the extract of the yellow pigment in 2N HCl its own volume of concentrated HCl and then to the whole 4—5 times its volume of absolute alcohol, for the reduction with magnesium to proceed readily. In some cases it has been found advantageous to add a drop or two of concentrated HCl from time to time. A disadvantage of this method is the somewhat considerable dilution which results. Aqueous solutions of the red pigments produced may be obtained from the aqueous alcoholic solutions above mentioned, by the addition of much ether, and if necessary, a small quantity of dilute acid. The red pigments are left The Production of Anthocyanins and Anthocyandins. 331 in the aqueous acid layer, provided that sufficient ether is added. The ethereal layer contains much of the unchanged flavonol derivatives and these may be more completely removed from the aqueous acid solution by repeated extraction with ether. In the preparation of extracts from flowers, dried and powdered petals are more convenient to use than fresh flowers. With the red pigments—all of which were shown by the amyl-alcohol reaction to be glucosides—obtained by reduction from extracts of the following flowers, viz.: daffodil, primrose, viola (yellow) and wallflower (lemon-yellow), the following experiments were carried out, viz. :— (1) The stability of the pigments in aqueous or alcoholic acid solutions was examined. In every case they were stable, no decolorisation was observed, even after standing for many days. (2) An acid (HCl) solution (aqueous or alcoholic) of the pigment was separated into two portions: (a) kept for comparison ; (0) shaken with excess of powdered calcium carbonate, whereby all excess of acid was removed, colour passed from red to violet, solution then became decolorised—all traces of red or violet disappeared, solution being a faint brownish yellow. ‘The red colour was restored quantitatively on acidification by 2-5 drops of concentrated HCl (concentrated used to prevent unnecessary dilution) whether the acidification took place immediately after decolorisation or after allowing the decolorised solution to stand over 1 ight. (3) The absorption spectra were examined, with results already mentioned. Thanks to the generosity of Prof. A. G. Perkin, F.R.S., in placing at his disposal a quantity of crude rutin—a pigment consisting of quercetin combined with a disaccharose residue, and identical with viola-quercitrin*— the author was enabled to carry out similar experiments using the pigment derived by reduction from this flavonol disaccharide of known constitution. The red pigment remained almost quantitatively in aqueous sulphuric acid (about 2N) when shaken with amyl alcohol, only traces passing into the alcohol. When hydrolysed by boiling with hydrochloric acid (about 20 per cent.) it yielded a sparingly soluble non-glucoside, which, when shaken in dilute sulphuric acid solution with amyl alcohol, passed quantitatively into the alcoholic layer, yielding a fine red solution. On shaking the solution with sodium acetate, the colour changed to violet, remaining, however, in the , alcoholic layer. On shaking with sodium carbonate solution it became blue- zreen and passed quantitatively to the aqueous layer. The series of observations (1, 2, and 3) as above was repeated with the pigment obtained from rutin, and in every way similar results obtained. * Of, A. G. Perkin, ‘Chem. Soc. Journ.,’ 1910, p. 1776. VOL. LXXXVIII.—B, 2B 332 The Production of Anthocyanins and Anthocyanidims. From the specimen of rutin supplied by Prof. Perkin, a sample of the crystalline pigment was prepared according to his instructions, and from this also the red pigment obtained and subjected to similar treatment. It yielded exactly similar results. This case is of particular interest, for, if the disaccharose is attached to the flavonol molecule in the correct position, the pigment produced should be identical with cyanin. For comparison with the above glucoside pigments, the same series of experiments was carried out with cyanin chloride, and gave exactly similar results. From the red pigment obtained by reduction of quercitrin (Kahlbaum) or from rutin, the non-glucoside pigment was obtained by hydrolysis, collected, dissolved in a mixture of equal parts by volume of ethy! alcohol (absolute) and 2N hydrochloric acid (aqueous), and of the solution obtained one part was retained for comparison, the other shaken with excess of finely powdered calcium carbonate to remove acid. The colour changed to violet, and after filtration and warming became decolorised—lost all violet colour, only faint brownish-yellow colour remaiming. On acidification with 2-3 drops of concentrated hydrochloric acid and warming, the red colour returned, but not quantitatively. A comparative experiment using cyanidin chloride gave exactly similar results. In solution—even faintly acid—this non-glucoside pigment is not so stable as are the glucosides above mentioned; thus, after standing for a week the colour is considerably diminished, and this diminution is due to destruction. Cyanidin, however, behaves similarly, if somewhat impure. The cyanin and cyanidin chlorides used were obtained from a specimen of cornflower pigment containing about 10 per cent. of pure pigment. The author desires to express his thanks to Prof. A. G. Perkin, F.R.S., for supplying him with the sample of rutin used in some of the above experiments. -CroontAN Lecrure: The Bearing of Cytological Research on Heredity.* By Epmounp B. Winson, Da Costa Professor of Zoology, Columbia University, ~ New York. (MS. received and Lecture delivered June 11, 1914.) The privilege of speaking ;in this historic centre of learning was first accorded to me more than 30 years‘ago, through the extraordinary kindness of Prof. Huxley to a young and unknown student. I would like to think it more than a fancy that to the same source, possibly, I may trace the distinguished honour of having been invited, after the lapse of many years, to speak here once again on the subject of cytology in its bearing on heredity. Of all Huxley’s wise and felicitous sayings none has more per- sistently lingered in my memory or appealed to my imagination than one which vividly pictured, 35 years ago, the basic phenomenon that the eytologist seeks to elucidate. Suggested, no doubt, by the researches of Hertwig, Strasburger, and Van Beneden, then but recently made known, this well-known passage is as follows :— “Tt is conceivable, and indeed probable, that every part of the adult contains molecules derived from both the male and the female parent; and that, regarded as a mass of molecules, the entire organism may be compared to a web of which the warp is derived from the female and the woof from the male. And each of these may constitute an individuality, in the same sense as the whole organism is an individual, although the matter of the organism has been constantly changing” (1878). The advance of modern cytology has been in some important respects a development of the germ contained in these words. For the aim of cytology, in so far as it bears directly upon the problems of heredity, is to trace out in the individual life the history of maternal aud paternal elements originally brought together in the fertilisation of the egg. And the drift of latter-day research, while it has not precisely confirmed Huxley’s conception, has never- theless been quite in harmony with the essential thought to which he gave such picturesque expression at a time when the labours of cytology were but just begun. This thought has been most nearly realised through the study of the cell nucleus, and in particular of the bodies known as chromosomes. I ask attention especially to these bodies in connection with certain problems of * It has been impracticable to reproduce here the original photographs and some of the other figures by which the lecture was illustrated. 2B 2 334 ; Prof. E. B, Wilson. genetics, not because the chromosomes are the only elements concerned in heredity, but because they offer the most available point of attack and have in fact yielded the most definite results. The limitations of time compel me to take a good deal for granted, and to pass over, for the most part, the historical and controversial aspects of the subject. I must be content, in the main, to state briefly what I believe to be established or indicated by the evidence. My task is much lightened by Prof. Farmer’s earlier presentation of many important aspects of the subject in his Croonian Lecture of seven years ago. Permit me, nevertheless, for the sake of present clearness, to indicate briefly some of the essential facts determined prior to the re-discovery of Mendel’s law in 1900. (1) The work of cytology in its period of foundation laid a broad and substantial basis for our more general conceptions of heredity and its physical substratum. It demonstrated the basic fact that heredity is a consequence of the genetic continuity of cells by division, and that the germ-cells are the vehicle of transmission from one generation to another. It accumulated strong evidence that the cell-nucleus plays an important rd/e in heredity. It made known the significant fact that in all the ordinary forms of cell- division the nucleus does not divide en masse but first resolves itself into a definite number of chromosomes ; that these bodies, originally formed as long threads, split lengthwise so as to effect a meristic division of the entire nuclear substance. It proved that fertilisation of the egg every- where involves the union or close association of two nuclei, one of maternal and one of paternal origin. It established the fact, sometimes designated as “Van Beneden’s law ” in honour of its discoverer, that these primary germ- nuclei give rise to similar groups of chromosomes, each containing haif the number found in the body-cells. It demonstrated that when new germ- cells are formed each again receives only half the number characteristic of the body-cells. It steadily accumulated evidence, especially through the admirable studies of Boveri, that the chromosomes of successive generations of cells, though commonly lost to view in the resting nucleus, do not really lose their individuality, or that in some less obvious way they conform to the principle of genetic continuity. From these facts followed the far-reaching conclusion that the nuclei of the body-cells are diploid or duplex structures, descended equally from the original maternal and paternal chromosome- groups of the fertilised egg. Continually receiving confirmation by the labours of later years, this result gradually took a central place in cytology ; and about it all more specific discoveries relating to the chromosomes naturally group themselves. All this had been made known at a time when the experimental study of The Bearing of Cytological Research on Heredity. 335 heredity was not yet sufficiently advanced for a full appreciation of its significance; but some very interesting theoretical suggestions had been offered by Roux, Weismann, de Vries, and other writers. While most of these hardly admitted of actual verification, two nevertheless proved to be of especial importance to later research. One was the pregnant suggestion of Roux (1883), that the formation of chromosomes from long threads brings about an alignment in linear series of different materials or “ qualities.” By longitudinal splitting of the threads all the “qualities” are equaily divided, or otherwise definitely distributed, between the daughter-nuclei. The other was Weismann’s far-seeing prediction of the reduction division, that is to say, of aform of division involving the separation of undivided whole chromosomes instead of the division-products of single chromosomes. This fruitful suggestion (1887) pointed out a way that was destined to lead years afterwards to the probable explanation of Mendel’s law of heredity. (2) Such, in bird’s-eye view, were the most essential conclusions of our science down to the close of the nineteenth century. A new era of discovery now opened. As soon as the Mendelian phenomena were made known it became evident that in broad outline they form a counterpart to those which cytology had already made known in respect to the chromosomes. Characters and chromosomes alike are singly represented (haploid or simplex) in the gametes, doubly represented (diploid or duplex) in the zygote and its products. In the formation of new germ-cells both alike are once more reduced from the diploid to the haploid condition. A parallelism so striking inevitably suggested a direct connection between the two orders of phenomena. And the hope was thus raised that the mechanism of heredity might be susceptible of a far more searching analysis than had yet been thought of. It is a rather striking coincidence that almost at the moment of the re-discovery of Mendel’s law, and apparently quite independently of it, microscopical studies were establishing the cytological facts upon which its _ explanation probably rests. Guyer’s studies on hybrid pigeons led him, in 1900, to suspect a disjunction of maternal and paternal chromatin-elements in the reduction division, a conclusion which he developed further in 1902. But the real basis for an explanation of Mendel’s law was laid by two conclusions announced in 1901 by Boveri and by Montgomery, independently of each other and apparently without knowledge of the Mendelian phenomena. Familiar as these conclusions are, I will dwell upon them for a moment, since they are fundamental to all that follows. Boveri's masterly experiments on dispermic sea-urchin eggs gave the first conclusive proof that the chromosomes directly affect the process of 336 Prof. E. B. Wilson. development, and that they are qualitatively different in respect to their individual influence. Eegs into which two spermatozoa are caused to enter develop into larvee that are almost always pathological, deformed or monstrous. The first cleavage of such eggs is by a tripolar or quadripolar division, and the cytological examination proves that this involves initial and apparently irreversible aberrations in the distribution of the chromosomes to the embryonic cells. Boveri's analysis, carried out with characteristic sagacity and thoroughness, seems to leave no escape from the conclusion that the abnormal combinations of the chromosomes thus produced are the cause, and the only cause, of the abnormal forms of development. The chromosomes must therefore be qualitatively different. This conclusion has been confirmed and rendered more specific by many later researches. It was proved, for instance, that in certain animals one of the chromosomes, or a small corresponding group of chromosomes, stands in some special relation to the determination of sex and the heredity of sex-linked characters. The study of hybrid sea-urchin larve by Baltzer, Herbst, and others, gives strong reason to conclude that many of the aberrations which they show in respect to the combination of maternal and paternal characters result from corresponding aberrations in the distribution of maternal and paternal chromosomes. In the evening primroses, the researches of Lutz and of Gates have shown that the gigas type of mutant has arisen in association with a doubling of all the chromosomes ; recently the same observers show that the /ata type is characterised by, and probably has arisen through, the presence of a single extra chromosome. To the still more recent important results of Gregory on the Chinese primrose I will presently refer. The conclusion of Montgomery was not less important, but failed at first to receive the consideration that it deserved. Among the suggestions that immediately followed upon Weismann’s speculations concerning the reduc- tion division, one of the most fruitful was that of Henking (1891), that the reduction of the number of chromosomes in the germ-cells is initiated by their conjugation two by two in pairs during synapsis, to be followed by their disjunction in the reduction division. Montgomery drew the bold conclusion that in this process each chromosome of paternal descent unites with a corresponding or homologous one of maternal descent; and he suggested that this process, though occurring at the very end of development, might be regarded as the final step in the fertilisation of the egg. This surprising conclusion was based on a comparative study of the size-relations of the chromosomes in the diploid and haploid nuclei. JI well remember the scepticism with which I, like many others, first received it. The conjugation The Bearing of Cytological Research on Heredity. 337 of chromosomes, to say nothing of paternal and maternal homologues, has been obstinately contested; it must be admitted that the proof is still far from complete for the chromosomes generally. Nevertheless, in spite of all scepticism, the drift of later research has been, I think, steadily in its favour. Both in plants and in animals the diploid nature of the chromo- some vroups in the somatic cells is often clearly visible to the eye, owing to conspicuous size-differences among the chromosomes. In such cases, as was first urged especially by Montgomery and by Sutton, the chromosomes may be sorted out into pairs according to their size. In a few cases, of which the Diptera offer the most striking examples, the sorting out is performed by nature, all the chromosomes being actually grouped side by side in pairs according to their size (fig. 1).* The conclusion here becomes highly probable that each pair includes a maternal and a paternal member, and that these are destined to conjugate in synapsis. In the case of the sex-chromosomes, to which I shall return, the probability becomes a certainty. a G FF Fie. 1.—Exact drawings of the diploid chromosome groups in various Diptera, showing the chromosomes grouped in pairs ; @, 0, c, e, from Stevens ; d, f, from Metz. a, Calliphora vomitaria, $ ; b,the same, 2 ; c, Sarcophaga sarracina, 2 : d, Droso- phila amena, § ; e, D. ampelophila, 2 ; f, D. funebris, 2. * These facts were illustrated by photographs of the chromosomes in Drosophila and Musca, from preparations by C. W. Metz, who has for some time been engaged with this problem in my laboratory. 338 Prof. E. B. Wilson. The proof of the reduction division likewise remains incomplete for the chromosomes generally, and is fully demonstrative only in case of certain special kinds of chromosomes, in particular the sex-chromosomes and the “m-chromosomes ” of the coreid Hemiptera. Strong confirmatory evidence of both conjugation and disjunction has, however, been afforded for the chromo- somes generally by studies on the maturation-process in hybrids, especially in Drosera by Rosenberg, in Cinothera by Geerts, and in Lepidoptera by Federley and Doncaster. The promulgation of the conclusions of Boveri and Montgomery opened the modern period of cytological inquiry and, as has been said, provided a substantial basis for the cytological explanation of Mendel’s law. This explanation follows in the most simple and natural manner from the observed facts. It assumes primarily that the Mendelian phenomena result from the shuffling (to employ the phrase of Farmer) of chromosomes that are concerned in the determination of the so-called unit characters. More specifically, the main assumption is that Mendelian allelomorphs are borne by corresponding pairs of chromosomes, each consisting of a maternal and a paternal member. By the conjugation of the homologous members of these pairs two by two, to form bivalents or gemini, as assumed by Montgomery, the maternal and paternal homologues assume such a grouping that they may be disjoined in the succeeding reduction division (in general accordance with Weismann’s early prediction); and from this follows the disjunction or segregation of the Mendelian allelomorphs which these chrormosomes bear. The independent distribution or assortment of different units is explained by the assumption (in favour of which definite evidence now exists) that the bivalents behave independently of one another. The explanation as here outlined was first clearly and logically developed by Sutton in 1902-3, when a student in my laboratory. Naturally enough, however, several others came independently to more or less similar conclu- sions nearly at the same time—in particular Guyer, Correns, Boveri, Cannon, and de Vries. As will appear later, Sutton’s elegant hypothesis was too simply framed to account for all the facts, and has had to undergo some modifications. In its main principle, however, it has received cumulative substantiation by later work in many directions. An important confirmation of the fundamental assumption is given by a discovery announced by R. P. Gregory before this Society only a few weeks ago. In certain plants of the Chinese primrose the usual number of chromosomes is doubled in both the gametes and the somatic cells. The genetic evidence obtained from such plants indicates that all the Mendelian units or factors thus far examined are correspondingly doubled. ‘This result weighs strongly, I believe, in favour of The Bearing of Cytological Research on Heredity. 339 the view that these factors are borne by the chromosomes, and may open the way to its crucial experimental test. The full force of the hypothesis only becomes apparent when we come to closer quarters with the facts. I shall attempt to illustrate this by consider- ing certain phenomena which now stand in the foreground of interest and bring home to us the intimacy of the relation that has been established between cytology and genetics. (3) I first ask attention to certain facts relating to the cytological basis of sex, a subject with which my own researches have been especially engaged during the past ten years. To the cytologist the interest of the phenomena extends far beyond the special problem of sex. Nature has here performed a series of experiments which gives a crucial test of many of our earler conclusions, provides a secure basis for further advances, and at the same time brings vividly before us the connection of the chromosomes with heredity. I will here touch only upon the main facts, especially in their bearing upon the phenomena of linkage, to which, I believe, they give the cytological key. That the chromosomes are involved in the determination of sex was first suggested, in 1902, by McClung, who argued on @ priort grounds that the so-called “accessory chromosome,” which enters but half the spermatozoa, is a sex-determinant. A substantial basis for this conclusion was provided in 1905, when the late Dr. N. M. Stevens and myself, working independently on Coleoptera and Hemiptera, discovered that in some of these insects the sexes differ in the composition of the diploid chromosome-groups. In the simplest type, first worked out in the Hemiptera, the “accessory ”—or, as I have preferred to call it, the X-chromosome, or sex-chromosome—is unpaired in the male, but paired in the female. Since the sexes are identical in respect ° to the other chromosomes, the latter may be disregarded, the sexual formulas being written simply as XX for the female and X (or X0) for the male. All of the other chromosomes are paired; hence the male possesses an odd number of chromosomes, one less than that of the female. Thus is explained the fact, first discovered in 1891 by Henking in Pyrrhocoris, that in the reduction division of the male the X-chromosome passes undivided to one pole, so that two classes of spermatozoa are formed, one with X and one without. In the female, on the other hand, the two X-chromosomes conjugate to form a bivalent, as usual, and then disjoin in the reduction division, so that every ege receives one X. This fact, at first inferred from the other relations, was soon afterwards demonstrated by direct observation, first by Morrill in insects, on rather scanty evidence, later fully established by Boveri, Gulick, Mulsow, and Frolowa in nematodes. It thus became clear that fertilisation 340 Prof. E. B. Wilson. of the egg by the X class of spermatozoa will produce the female combination XX, by the no-X class the male combination X (or X0) (fig. 2). From these relations, and those found in a second type, described below, Miss Stevens and I concluded that the determination of sex in these animals depends upon which class of spermatozoon enters the egg, and that the X-chromosome plays some special ré/e in the process. In a general way this substantiated McClung’s earlier suggestion, though he reversed the actual significance of the two classes of spermatozoa. I discovered in my first researches on Hemiptera a second type, inde- Protenor Type Lygaeus Type _—__A__—__| =r“ 3 Diploid Nuclet XX Pee XX a Fertilization Gametes : Zygotes XX XO XX IN etc etc Fie. 2.—Diagram of the relations of the sex-chromosomes to sex-production, showing the: two main types represented by the Hemiptera Protenor (Y-chromosome absent) and. Lygeus (Y-chromosome present). In either case random union of the maternal and paternal gametes reproduces the original forms (males and females) in equal numbers. pendently found and fully worked out in the Coleoptera by Miss Stevens, in which the X-chromosome of the male is accompanied by a mate of different type, often much smaller than X, which I called the Y-chromosome. This was the first discovery of a heterogeneous chromosome-pair in any animal or plant. In this case X and Y conjugate and disjoin like any other chromosome pair—a fact here shown with incomparable clearness—so that half the spermatozoa receive X and half Y. The X-class are, as before, female-producing, while the Y-class are male-producing; and the sexual formulas become XX for the female and XY for the male (fig. 2). Owing to. the small size of Y, these differences are in many species conspicuously visible in the diploid nuclei, despite the fact that the sexes here possess the same number of chromosomes. The two types are connected by a series of inter- ———————————<=— The Bearing of Cytological Research on Heredity. 341 mediate conditions, varying with the species, and X may consist of two or more separate chromosomes. The Y-chromosome, on the other hand, is single in all cases thus far accurately known.* At the time these conclusions were announced it seemed unlikely that the fertilisation of the egg by the two classes of spermatozoa could ever actually be followed out. This has, nevertheless, been accomplished recently by Mulsow in the case of a nematode, Ancyrocanthus, where the chromosomes remain distinct in the mature spermatozoa and can readily be counted, even in the living object. Both classes were here traced into the egg, and the sexual differences were clearly shown in the germ-nuclei at the time of their union. I will not enter upon the many interesting modifications of detail which these phenomena exhibit. In principle, the facts are the same in many insects and nematodes, probably in the myriapods and arachnids, perhaps also in the mammals and in man, though the demonstration here still leaves much to be desired. An extremely interesting series of researches by Morgan, von Baehr, Schleip, Doncaster, and others have proved that the same principle applies also to the parthenogenetic forms, such as the aphids, bees, and ants. Hardly less interesting are the investigations, especially of Boveri, Schleip, Kriiger, and Zarnik, which show that this principle may even be extended to certain types of hermaphrodites. The results of genetic experiments on Lepidoptera and on birds lead us to expect the existence in these forms of a different cytological type, in which the eggs, instead of the spermatozoa, are of two different classes; but the cytological facts have not yet become sufficiently clear to warrant any definite conclusion. In the case of birds, indeed, a conspicuous contradiction still appears between the cytological and the genetic results ; but the cytological observations have not. yet produced evidence that can compare in cogency with that available in case of the insects or the nematodes. I turn to the broader significance of the cytological facts that have been made known in this field. They constitute a very definite advance upon Boveri's general demonstration of the qualitative differences of the chromo- somes; for it is impossible to doubt that the X-chromosome stands in some special causal relation with sex-heredity. A powerful argument for this is. given by the facts of sex-linked heredity, which I shall presently consider. The riddle which this form of linkage presents is solved by a cytological phenomenon to which I first drew attention in 1906. The Y-chromosome, * Jn an extreme case, now under investigation by Mr. Goodrich in my laboratory, the X-element consists of not less than eight distinct chromosomes, opposed by a single Y. The females here show seven more chromosomes than the males (photographs). 342 Prof. E. B. Wilson. when present, is confined to the male line, and hence always passes from father to son. The X-chromosome, on the other hand, always passes from father to daughter (because sperms of the X class produce females), while the sons receive their single X-chromosome from the mother (because the male- producing sperms are of the no-X or Y class) (fig. 2). I will show a little later that on this curious fact probably depends the “ criss-cross ” type of sex-linked heredity in which the sons are like their mothers, the daughters like their fathers. The cytological phenomena of sex-production lend strong support to the theory of the genetic continuity of chromosomes. ‘They give unquestionable proof, in case of a particular chromosome pair (XY), of the conjugation and subsequent disjunction of corresponding maternal and paternal chromosomes. They thus substantiate the conclusions of Henking and Montgomery and confirm Weismann’s earlier conception of the reduction division. They are in full accord with genetic studies, which prove that one sex is homozygous, the other heterozygous, with respect to a sex-determining factor. They give the first direct evidence of a difference of nuclear constitution between the homozygous and the heterozygous conditions, and of corresponding gametic differences. And finally, in the case of a particular chromosome pair, they fully substantiate the general cytological explanation that has been offered of Mendel’s law. (4) The facts just considered now lead us to some of the most intricate and interesting of current problems. No phenomena appealed more strongly to the interest of earlier naturalists than those of correlation. A very interesting light is thrown upon this problem by the phenomenon now widely known as gametic coupling or linkage, and it is here, perhaps, that we may best appreciate to what an extent cytology and genetics reciprocally illuminate each other. In the second of Sutton’s original papers (1903) he pointed out what seemed to be an obstacle in the way of his own hypothesis, of which much has been made by later critics. The number of chromosomes is probably always much smaller than that of Mendelian units in any given case; hence each chromosome must bear many such units. From this it follows that if the composition of the chromosomes be fixed, or even fairly constant, the units should cohere in definite groups, equal in number to that of the chromo- somes ; but the earlier studies on heredity gave little definite evidence that such was the fact.* Sutton did not, I think, meet the difficulty, which he * Jn a general way, of course, this fact was known to earlier observers, e.g., ‘““ We appear, then, to be severally built up out of a host of minute particles, of whose nature we know nothing, any one of which may be derived from any one progenitor, but which are The Bearing of Cytological Research on Heredity. 343 himself had pointed out. It was perhaps the same difficulty that led Correns (1902) and De Vries (1903), in their attempts to explain Mendel’s law, to treat the chromosomes as of quite secondary importance. ‘Their explanations operated almost wholly with smaller elements, of which the chromosomes were supposed to consist. It is now clear, however, that only when the chromosomes are taken into account do all the facts fall into line. For the most recent studies in genetics have produced indubitable evidence that Mendelian units are often, in fact, more or less definitely linked together in groups, as they should be under the chromosome theory. Linkage was first clearly recognised in sex-limited or sex-linked heredity, to which I have already referred. A form of linkage having no relation to. sex was brought to lght a little later by Correns and by Bateson and Punnett in certain plants, and is now known to be of rather wide occurrence. I will confine my attention mainly to the case in which both these forms of linkage are now most accurately known, that of the fruit-fly, Drosophila ampelophila. In this species a very extended experimental analysis of the genetic phenomena has been carried on during the past four years in the laboratory of Columbia University by my colleague, Prof. T. H. Morgan, and his pupils and co-workers, Sturtevant, Bridges, and many others, from the investigations of whom the following results are reported. Drosophila (to paraphrase the words of Lacaze Duthiers) seems made for the experimental study of genetics. It passes through a complete generation from egg to ege in about twelve days. A single female not infrequently produces upwards of a thousand eggs. These fortunate circumstances have made it possible, in the course of four years of continuous study, to accumulate a prodigious mass of data, far surpassing in extent any others thus far made known. During this period these flies have given rise to more than a hundred definite mutations which are inherited in accordance with Mendel’s law. They are of many different kinds, affecting the colour, shape, and structure of the body -and of the eyes, the structure of the wings, legs, antenne, and so on. Up to the present time 72 of these characters have been more or less completely tested as to their behaviour when crossed with the normal or “ wild” form, and with one another. The all-important fact which these tests have established is that the characters fall into four definite linkage-groups, of which the first now includes 31 characters, the second 23, the third 17, the fourth, so far, but a single one. These numbers represent of course only a. beginning. They steadily increase as observation continues. As the elaborate experimental analysis has proceeded, carried on by a usually transmitted in aggregates, considerable groups being derived from the same progenitor” (Francis Galton, ‘ Natural Inheritance,’ 1889). 344 Prof. E. B. Wilson. number of specially trained co-operating observers, it has been more and more conclusively demonstrated that the units of each group are more or less firmly linked together in heredity, while those belonging to different groups are quite independent. This at once suggests that the units of each group (or corresponding things on which they depend) are borne by a particular chromosome which constitutes their common vehicle of transmission, and that to this fact is due their cohesion or linkage in heredity. Conversely, the several groups are independent of one another, because of the independence of the chromosomes which bear them. This hypothesis would have been a plausible one even were the number of chromosomes in Drosophila unknown. In point of fact, however, the gametic number of chromosomes in this species (or of chromosome-pairs in the diploid groups) is actually the same as that of the linkage-groups, namely, four (fig. 1, d,e). It is at least an odd coinci- dence that one of these chromosomes, like one of the lnkage-groups, is extremely small. One is tempted to guess that this may explain why for a long time but three linkage-groups could be identified, and that the fourth thus far contains but a single character, recently discovered by Muller. Thus far, admittedly, the hypothesis presents a somewhat speculative aspect; but fortunately there is a means of testing it specifically, for the cytological evidence demonstrates that one of the four chromosomes is definitely connected with the determination of sex. One of the four groups of units, therefore, should likewise exhibit some special relation to sex. And this is in accordance with the facts, for every one of the 31 characters of the first group exhibits sex-linked heredity, of the same type as that which appears in colour-blindness or hemophilia in man. It was pointed out, in 1910-11, by Morgan, Gulick, and myself that the heredity of sex-linked characters of this type exactly follows the course of the X-chromosome; that is to say, that the history of such characters is precisely such as it should be if they were dependent upon factors borne by this chromosome. Like the latter, the sex-linked units are always simplex or haploid (hence heterozygous) in the male; and they zigzag between the sexes in exactly the same way. No other group shows this relation. Without entering far into the detail, let me illustrate these phenomena by a single example, that of the so-called “criss-cross” heredity. The normal Drosophila possesses red eyes; a common mutant has white eyes, recessive to red. If a pure-bred white-eyed female be paired with a normal red-eyed male, all the resulting sons are white-eyed, like their mother; all the daughters red-eyed, like their father. Exactly analogous results appear when, instead of white eyes, any other units of the first group, such as yellow body colour or miniature wings, are similarly tested. The results at once lose their The Bearing of Cytological Research on Heredity. B45 apparently bizarre character if we assume that the production of each sex- linked character depends upon something (which we may call a “ factor” or a “oen”) that is borne by the X-chromosome. I have already emphasised the fact that the sons derive this chromosome from their mother. In the cross just considered the sons therefore inherit with this chromosome the white eyes of their mother ; the daughters, on the other hand, are red-eyed like their normal father, because they receive from him in every case a normal X-chromosome bearing the factor for the dominant red colour (fig. 3, A). A B 2 é 9 6 Diploid Nuclei x KY xX xf Aone . Um Cua Via x Bre HN Meas hee | Zugotes xX xY xx Xx xY XY ¥ie. 3.—Diagram of the relations of the sex-chromosomes to sex-linked heredity. Any normal (dominant) sex-linked character (e.g., red eyes) is assumed to depend on the presence of a particular factor contained in the X-chromosome. Loss or modification of this factor produces a corresponding recessive (e.g., white eyes), X now becoming «. A. Criss-cross heredity, when the double recessive, white-eyed female (vx) is paired with the normal, red-eyed male (XY). The heterozygous daughters (7X) are red-eyed, with white-eye recessive ; the sons (zygotes x Y) white-eyed. B. History of the following generation, produced by crossmg wX and zY. The offspring of both sexes are now indifferently white-eyed (vz, rY), or red-eyed (Xx, XY). Gametes This has been tested in many ways, with results always in accordance with the hypothesis. Very convincing evidence in its favour has recently been obtained by Bridges through the study of a particular race of Drosophila, which regularly shows about 10 per cent. of exceptions to the “criss-cross” type of heredity. This holds true for all the sex-linked characters thus far tested. Bridges’ analysis—too intricate to be entered upon here—led to the conclusion that these exceptions are due to a failure of the two X-chromosomes to undergo disjunction in the reduction division of the female, so that the mature egg sometimes receives both X-chromosomes, sometimes neither. Eggs of the XX type might be expected to produce females even if fertilised by the no-X type of sperm, the XX combination (characteristic of the female) 346 Prof. E. B. Wilson. being supplied solely by the egg. Sex-linked characters shown by such females must be derived from the mother. On the other hand, eggs of the no-X class fertilised by the X class of sperm (normally female-producing) should produce males, and these should show only sex-linked characters derived from the father. This hypothesis was first tested, very ingeniously and thoroughly, by combining different sets of sex-linked characters derived respectively from the mother and the father. The results uniformly sustain the hypothesis. Very recently Bridges has tested his assumption cyto- logically. The expectation is that eggs of the XX type fertilised by sperm of the X class should produce females with three X-chromosomes, while if fertilised by sperms of the Y class the females should possess two X’s and a Y. The cytological examination has demonstrated that certain females of this race actually possess three of these chromosomes. Taken as a whole, the foregoing evidence gives almost crucial proof in favour of the conclusion that both the sex-determining factor and the sex- linked ones are borne by the X-chromosome. Sex-linked heredity of the type seen in birds or in Lepidoptera requires an explanation somewhat different in detail but similar in principle (Spillman, Castle). In the facts of sex-linkage generally the chromosome hypothesis finds, I think, its strongest support, for the linkage of sex-linked factors with one another is of quite the same type as that which appears in other groups that are inde- pendent of sex, and the conclusion can hardly be avoided that in both cases linkage is due to the same cause. We now take a final step in order to consider a seeming difficulty which introduces us to the most recent inquiries in this field. If our hypothesis is correct, how does it come to pass that linkage is not complete? How can we explain the variations in the so-called strength of linkage? Let me again illustrate by a single example taken from Drosophila, showing the heredity of two pairs of sex-linked characters of the first group. One pair comprises the normal grey body colour (G) and its recessive mutation yellow (Y), the other the normal red eye-colour (R) and its recessive mutation white (W). Let the pure-bred dominant female RG be crossed with the pure-bred recessive male WY and the hybrid offspring be inbred. Were linkage complete we should expect to find in the grandchildren the same combina- tions as those which entered the hybrid, and these alone—that is to say, 2G or WY. In point of fact, this expectation is nearly always realised, but about one individual in eighty shows one or the other of the new combina- tions RY or WG. Genetically this means that R and G, or W and Y, are strongly linked, yet now and then may dissolve their union and recombine. Cytologically it means that the original X-chromosomes, bearing in one case The Bearing of Cytological Research on Heredity. 347 RG, in the other WY, usually maintain their original constitution, yet occasionally may undergo an exchange of units so as to produce the new eombinations observed. How is this possible ? Haploid Group Diploid Group Gamete Homozygote (¢) po eee eS RE res ed ee A Al |A B BI IB Cc CIC : H O Ay | O} }O ] I} ]1 P P D J Q D} |D sila alla K R KI IK R} |R cE Le F aie es Sse F M T Vv File omilm ocltr v| lv G N U WwW G| |G NIN U, JU WIilWw TN I aii V XX Il Til WwW Chrornosomes Chromosome Pairs Fia. 4.—Diagram. of chromosomes and linkage-groups, based on the relations observed in Drosophila ampelophila. Heavy vertical lines represent chromosomes (or the chromatin-threads from which they arise), letters different factors or gens, assumed to be aligned in linear series in the threads. In the diploid groups corresponding chromosomes are paired side by side in the position assumed by them during conjugation or synapsis. Each of the four series A-G, H-N, O-U, V—W, forms a definite linkage-group in which the factors tend to cohere, while independent of all other factors. In the male diploid groups one X-chromosome is missing, its plave being taken by a Y-chromosome (Lygeus type). The nature of the latter is still unknown. In attempting to answer this, it is necessary to bear in mind that the recombinations with which we are dealing affect units that are usually linked together, and hence belong to the same group—in the example just given to the sex-linked or X-group. The recombination or exchange of units must accordingly take place between the corresponding or homologous chromo- somes of a pair (here the X-pair). It therefore becomes probable that the exchange is effected during the intimate association of these chromosomes in conjugation or synapsis. It was long since suggested by Boveri that an exchange of particular elements might take place between conjugating chromosomes as between conjugating Protozoa, but this suggestion is too vague for our present purpose. The basis for a more specific explanation was offered by Janssens in 1909 in his theory of the “chiasmatype,” more recently VOL. UXXXVIII.—B. 2 348 Prof. E. B. Wilson. elaborated in a remarkable manner by Morgan and by Sturtevant. This explanation is as follows :— It has long been known that subsequent to the process of conjugation— sometimes, as now seems probable, during this process—the two chromosomes of each pair, while still in the form of long threads, often twist about each other, thus producing the so-called “ strepsinema” stage. Janssens concluded from a careful study of the facts in the Amphibia that the double spirals thus formed may in some cases come into contact and fuse at certain points where the threads cross, and then may be separated again by a straight longitudinal split through the points of fusion. Such a process would lead to an exchange of certain regions between the two threads. Janssens named this the “chiasmatype,” and briefly called attention to the possible bearing of such a process on the Mendelian phenomena. Morgan afterwards very ingeniously developed this thought as follows: The “determining factors” or “gens” of unit-character are assumed to be aligned, as Roux long since suggested, in linear series in the chromatin-threads and in a definite order. In the process of conjugation corresponding or allelomorphic gens (large and small letters in the diagrams) are assumed to lie opposite one another in the two threads, as 1s shown in diagram in figs. 4 and 5. If in a certain proportion of cases Diploid Group : Diploid Group Complete Heterozygote (9) Partial Heterozugote (9) oe —— Aj ja Aj ja By] |b By] |B Cj Ic cl|C H] |b O} jo Hy }h 0} |0 T]fi i} |] Dj }d Pile D P JT hj Ql Iq B i} lJ q| |Q Ky |k R} |r Ky |K R E] Je L {I S elle Wjb S| |5 alle Wy calm atl in dee Wally File omiim tl fr yy Gl ig Nj Jn UlJu Wilw 6} IG n{ jn uj fu wi lw XX Il SINE SAVERWNE KX apt im WwW Fie. 5.—Diagram illustrating heterozygous conditions. Homologous or allelomorphic factors (e.g., C and ¢) occupy corresponding loci or levels in homologous chromo- somes. The condition at the left is heterozygous for all factors; at the right, heterozygous for some (Aa, gG, etc.), and homozygous for others (BB, ee, etc.). the two threads (linear series of gens) twist together, unite, and separate in the manner described by Janssens, the result will be an exchange between a The Bearing of Cytological Research on Heredity. 349 them of certain gens, as shown in fig. 6. A simple and elegant solution of the problem of recombination or of “ crossing over ” is thus given. It should be pointed out that Janssens’ actual observations on the chiasmatype still lack adequate confirmation, that the strepsinema has been observed in the longitudinally split single chromosomes of the somatic divisions, and that nearly all cytologists have hitherto believed the twisted threads to untwist again before actual separation takes place. I have not yet been able fully to satisfy myself concerning the facts; but there is no doubt that in the Amphibia many of the appearances seem to be in favour of Janssens’ conclusions. The most ingenious part of the explanation relates to the varying strength of linkage. It is obvious that if the twisting be not too close, the likelihood of a chiasma taking place in the interval between any two units (and hence of their separation or dissolution of linkage) increases with the distance between them in the thread. Conversely, the nearer together two units le the greater Ath T}ji eeh-=2= eres EL oT Mile oleh deeret che iY jl Ji JN iW il IQS ch dd Wien HIPS | PUORESETEIN Ne © HU se IIe Reg viesiatve ha cae che [JL iL LH} JL IL Lill L} {1 LUe EW mi{M miiM ml IM mi |M Mi lm Mi fm m{|M mj |M niIN nliN niIN nliN Niin) NiIn niiN nlIN nd Se | ea sey (2) (3) (4) (5) in Fig. 6.—Diagram of the exchange of factors (dissolution of linkage) as explained by the chiasmatype hypothesis. The second chromosome-pair of fig. 5 is here employed as an example. The original condition is assumed to be that shown above (1). The lower figures (2-5) show a few of the many possibilities of exchange or “crossing over” between‘the two homologous linkage-groups or chromosomes, H—N andh-n. In each case the point of crossing, fusion, and subsequent splitting is shown at the left, with the result at the right. The position of the chiasma indicated in each case by ch. In (2) and (8) but one chiasma is present, in (4) two, and in (5) three. A very large number of recombinations is possible, even with so small a number of factors as is here represented. 350 Prof. E. B. Wilson. the probability of their remaining in association ; in other words, the greater the “strength of linkage.” Hence the astounding possibility which this suggests of using the “strength of linkage” as an index of the serial order of the units in the threads and the relative distances between them. This, it seems to me, is the most remarkable result to which these researches have led, for it opens the possibility of a detailed experimental analysis of the nuclear organisation—almost, we might say, of the topographical anatomy of the germ-plasm. By the application of this method to an immense body of experimental data, Morgan and his co-workers, Sturtevant in particular, have actually plotted the location of most of the units in each chromosome, and constantly use the diagrams thus obtained as working models for further analysis. The order and relative distances of the units in each linear series, once deter- mined, are found to be remarkably constant when tested by varied experiments designed to this end. The practical value of the hypothesis is attested by the fact that when the distance (strength of linkage) between any two units, A and B, is known, and also that between B and a third unit, C, the relation between A and C may be predicted with con- siderable accuracy. This fact gives reality to the assumption that each unit has a definite locus in the lnear series (chromatin-thread), and that allelomorphie units occupy corresponding loci in homologous chromosomes. Further corroboration is found in the mteresting phenomena presented by multiple allelomorphs, of which an example is given by the four eye-colours : white, eosin, cherry, and the normal or red colour, all of which belong to the first or sex-linked group in Drosophila. Any two of these are allelomorphic to each other, and the important fact is that all exhibit the same strength of linkage with all other sex-linked units. The inference is that each of the units in question must occupy the same locus in the sex-chromosome. Since no two can occupy the same locus at the same time, it follows that not more than two of them can co-exist in any particular female, and not more than one can be present in the male. And this corresponds with the facts as actually observed. To those not actually engaged in such investigations this hypothesis will, perhaps, seem of highly speculative character. But is it more so than many working hypotheses of experimental physics or organic chemistry that have proved themselves fruitful in the past? I will not pretend to answer. There is no doubt that it provides us with a simple, easily intelligible and effective means of handling enormous masses of intricate data, of devising new experi- ments, of predicting results. Such an hypothes’s, venturesome though it may seem, is something more than a speculation. The Bearing of Cytological Research on Heredity. 351 I have endeavourea to show how the chromosome-theory, first outlined in very general form, has been more and more specifically developed until it has become an important instrument for the detailed analysis of intricate genetic phenomena. I am well aware that some eminent students of genetics are still reluctant to accept this theory, at least in its more detailed applications. Iam not disposed to reproach them for such scepticism. The cytologist suffers under the disadvantage of working in so unfamiliar a field that some of his conclusions, even among those most certainly established and most readily verifiable, are apt to give a certain impression of unreality, even to his fellow naturalists. It is undeniable, too, that in this subject, for better or for worse, hypothesis and speculation have continually run far in advance of observation and experiment. It is quite possible that some of my hearers may consider some of the views I have touched upon as a fresh illustration of this fact. If so, I beg them to bear in mind that no conclusion which I have considered has been reached as a merely logical or imaginative construction. I have endeavoured to limit myself to matters of observed fact, and to conclusions that are either demonstrated by facts or directly and naturally suggested by them. To those who have had opportunity to come into intimate touch with both cytological and genetic research the conclusion has become irresistible that the chromosomes are the bearers of the “factors” or “gens,” with the investigation of which genetics is now so largely occupied. What are these gens? How do they operate? We do not know what they are. We assume only that a gen is something that is necessary to the development of a particular character. We do not know how they operate; for, despite all that experimental cytology and embryology have taught us concerning development, we are still without adequate understanding of its mechanism. We may nevertheless guess that gens play their several rdles by virtue of their specific chemical nature, and that the study of chemical physiology as applied to development is destined to take an important part in the future investigation of this problem. In the meantime it would be well to drop the term “determiner” or “determining factor” from the vocabulary of both cytology and genetics. What we really mean to say is “differential” or “differential factor,’ for it has become entirely clear that every so-called unit character is produced by the co-operation of a multitude of determining causes. Embryologists long since demonstrated by direct experiment that the cell-protoplasm as well as the nucleus is concerned in the determination of development. Our whole study of the cell leads us to the conclusion that it is an organic system, in the operation of which no single element can be wholly dissociated from the rest. When, therefore, we speak of nuclei or VOL. LXXXVIII.—B. 2D 352 The Bearing of Cytological Research on Heredity. chromosomes as the “bearers of heredity” we are employing a figure of speech. They are such just to the extent that they are necessary to develop- ment and heredity; but how far this conclusion carries we are as yet unable to say. Genetic experiment has already given some ground for the conclusion that definite types of hereditary distribution may be immediately dependent upon elements contained in the protoplasm. Recent advances in our knowledge of the “ chondriosomes ” or “ plastosomes” provide this conelu- sion with at least 2 possible cytological basis. Our conceptions of cell-organisation, like those of development and Hevedite, are still in the making. The time has not yet come when we can safely attempt to give them very definite outlines. It is our fortune to live in a day when the business of observation and experiment leaves little time or inclination for a priori speculations concerning the architecture of the germ- plasm or of the cell. Nevertheless it is impossible not to be struck with the fact that recent advances in cytology and genetics are in certain important respects in line with theoretical views put forward nearly 30 years ago by Roux, Weismann, and de Vries. These views were, it is true, almost purely imaginative or logical constructions. Some of them, especially as applied to the mechanism of embryological development, have been experimentally disproved, others are incapable of verification, and hence have fallen into disrepute. We have become chary of theories which assume all parts of the cell to be built up of ultimate, self-propagating, vital units, such as “gemmules,” “pangens,” or “ biophores.” The working hypothesis that has here been considered must not be identified with those far-reaching specula- tions; it is at once more limited in scope and more flexible in form. And yet, as far as the cell-nucleus is concerned, those visions of a bygone speculative era are now beginning to seem more real than would have been thought possible by some of us ten or even five years ago. We read in the latest productions of cytology and genetics of the division and genetie continuity of factors or gens, of their linear alignment in the chromatin threads, of their conjugation and disjunetion, of their linkage or independent distribution, in heredity. We find such conceptions no longer treated as belonging to an age of cytological romance, but employed every day in the most matter-of-fact way as practical instruments of laboratory experiment, analysis, and prediction. We are bound to nospeculative systems or extrava- vances of an earlier day if we recognise in this, let the outcome be what it may, a triumph for the men who first endeavoured to bring cytology and the experimental study of heredity into organic relation. 353 Observations on the Life-Cycle of a New Flagellate, Helkesimastix* feecicola, n.g., n.sp.: Together with Remarks on the Question of Syngamy in the Trypanosomes. By H. M. Woopcock, D.Se., Assistant to the Professor of Protozoology, University of London, and G. Lapacs, M.Sc., Assistant Lecturer and Demonstrator in Zoology, Victoria University, Manchester. (Communicated by Prof. 8. J. Hickson, F.R.S. Received July 31, 1913.) [PLarEs 13 anp 14.] This new flagellate occurs frequently in goat-dung ; we have found it also in sheep-dung. It is a “ passenger,’ being carried passively through the alimentary tract, im an encysted condition. When the dung is moistened with water—probably, in nature, when it is deposited on damp grass or earth—the flagellate emerges from its cyst and goes through its life-cycle, ultimately encysting again. The cysts are doubtless swallowed by the goat with its fodder. We have cultivated Helkesimastiz under various conditions, which will be described in our full account later. In order to obtain this form in large numbers and study its life-history without any fear of being misled by stages in the life-cycle of other flagellates, we succeeded in isolating it from the other forms occurring in simple dung- cultures and cultivating it on agar-media, on which it multiplies rapidly. The medium which we have used principally is weak Lemco-agar, 7.c. the same medium as used for blood-agar, but considerably diluted. For our con- tinuous observations we have used hanging-drop preparations in sealed cells ; in these cases we used very dilute Lemco-broth, without any agar, as the medium for the development of the flagellates, because in the denser agar-medium it is very difficult to see the flagellum. In all these media the flagellate forms the protozoan component of a “mixed culture,” since there is, of course, an even greater and more rapid bacterial development. Commencing the account of the life-cycle with the permanent cyst, this is a small spherical or slightly ovoid body about 3-3} w in diameter (figs. 1-3). The cyst-wall is well defined, but not very thick; it appears to consist of a single membrane, there being no differentiation into inner and outer envelope such as is found, for example, in many Ameba cysts. Some- times bacteria are adherent to the cyst-wall (fig. 3). The protoplasm is * The generic name is formed on the analogy of the Homeric epithet, éAxeourem)os, trailing mantle or cloak. We are indebted to Prof. Minchin for suggesting this appro- priate name. VOL. LXXXVIII.—B, 2h ip 354 Dr. H. M. Woodcock and Mr. G. Lapage. fairly homogeneous, or contains fine granules. There is frequently, however, a conspicuous, somewhat refringent grain, situated near the periphery (Plate 13, figs. 1 and 2), but no vacuole is present. The nucleus can usually be seen as a clearer area, and at times the contained karyosome can be distinctly made out as a dull body in the centre. No division takes place inside the cyst, which is therefore a resting or “Dauer” cyst and not a multiplicative one. Excystation.—We have followed the process of excystation in cysts which had been for about 22 hours in fresh medium (Liebig broth). This is not quite the minimum period required, before the flagellate will emerge from its cyst, for in this particular preparation some half-dozen individuals were already active by this time, though the great majority were still encysted. The period which must elapse depends, we consider, upon how soon the multiplication of the active bacilli in the new environment has taken place to a sufficient extent to produce, in sufficient quantity, the ferment or chemical substance in solution which is either directly or indirectly the cause of the dissolution of the cyst-wall. For we are strongly of opinion that the explanation of both the encystment and the excystation of Helkesimastiz (and probably equally of other dung- and infusion-flagellates) is to be found in the view put forward by Cropper and Drew* in their recent important and suggestive work on the causative factors of the corresponding phenomena in Amebe. We hope to study our new flagellate from this biological standpoint subsequently, and will here merely indicate in passing certain facts we have observed which bear upon the question. We have found excystation occurring only in those cultures, plates, or observation-prepara- tions in which there was a plentiful development of an (or of more than one) active, markedly aérobic bacillus, which we have not yet more closely deter- mined. We have kept cysts in ordinary cover-slip preparations in different media (the preparations being sealed, of course, to prevent evaporation), for several days, without the emergence of any active individuals taking place ; and in such preparations, the medium not being in contact with free air, no noticeable development of active bacilli took place. In a cyst from which the flagellate is about to be liberated the wall is noticed to become gradually less and less obvious, until at length one can no longer say that there is any envelope present, apart from the delicate mem- brane or pellicle limiting the body-protoplasm (fig. 4). There is no definite rupture or bursting of the wall at any point, through which the flagellate could pass out to the exterior; in other words, when the creature has again become active and moved off, there is no cyst-wall left behind. We were the * ‘Researches on Induced Cell-reproduction in Amcebe,’ London, John Murray, 1914. Observations on the Infe-Cycle of Helkesimastix feecicola. 355 more convinced of this because we had previously seen the excystation of a species of Bodo from other dung-cultures, and in that case the rupture of the eyst-wall was very evident, and when the Bodo had moved away the empty cyst-wall was left behind. To return to Helkesimastix, when a sphere from which the cyst-wall had disappeared was watched it was seen gradually to change its shape, and in about seven minutes had elongated somewhat and become ovoid (fig. 5). Near one extremity a small vacuole had made its appearance, which represented the contractile vacuole; active metabolism had again begun. Very slight, spasmodic movements were noticed at intervals, and then suddenly it was seen that a very short, delicate flagelium was present, apparently emerging from about the middle of the body (fig. 6). This waved slowly to and fro, though we do not feel at all certain that it was causing the jerky movements. The flagellum could at first be made out only when it was projecting from the side of the creature, and not when it lay over the body-protoplasm. Meanwhile the flagellate was growing in size rapidly, and five or six minutes later it appeared as in fig. 7: the flagellum had lengthened considerably and was more prominent and stouter, and its point of origin was now near to the anterior end. The nucleus had doubtless also passed towards the anterior end, but unfortunately it could not be clearly made out; this was owing partly to the jerky move- ments and partly to the fact that the protoplasm was becoming very granular. The jerky, to and fro movements of the body, producing no displacement of the creature, are very characteristic of certain phases ; we describe them as “nicking ” movements (cf. below, p. 358). Ten minutes later the flagellate was beginning to make short, gliding movements of progression, and these alternated for some time with the knicking movements until, after about an hour, it moved steadily for the first time out of the field of vision. During this period the same process had been going on throughout the preparation, and many actively gliding individuals were now present. These were all growing rapidly before beginning to multiply, and attained a size considerably larger than the average individual size found later on; the protoplasm was usually full of refringent granules. Form and Structure—The body of the ordinary active individual of Helkesimastix fecicola is typically elongated and fairly cylindrical, with one end (the anterior one) bluntly rounded, the other (posterior) one tapering away more and being at times somewhat pointed (figs. 8, 10, 13). The broadest part of the body is generally nearer to the anterior end. We can most aptly compare the form with that of the fleshy part of a small carrot. Sometimes the hinder extremity is drawn out into a narrow prolongation (fig. 11), This hinder part of the body is often very plastic and irregular, 2E 2 356 Dr. H. M. Woodcock and Mr. G. Lapage. and reminds us somewhat of the posterior extremity of Cercobodo (Cerco- monas), with its long cytoplasmic “ tail,” though in Helkesimastiz it is never, in normal conditions, drawn out to anything like the same extent. The body is usually about 64 to Tw long by 24m or 2u broad. Under other conditions the shape of the body may be oval or slightly pyriform (figs. 9,18); this is frequently the case in smaller individuals, and also when the flagellates are sluggish, in a rather denser medium than usual. There is a single flagellum, generally about two and a half to three times as long as the body, or even longer, inserted at, or very near to, the anterior end. This new flagellate is remarkable in having its flagellum always directed backwards, z.e. it possesses only a trailing flagellum, whence the generic name. In life, the flagellum is usually contiguous to the body for practically the entire length of the latter, and the proximal portion is always in very close contact with about the anterior third or so of the body, from which it never becomes free. This is seen very clearly when a steadily progressing individual makes a sharp turn and goes off in another direction. The proximal portion of the flagellum also turns immediately, along with the body, but. the remain- ing part is for the moment free from the latter and turns more gradually into the new direction (cf. fig. 14). The flagellum lies along the middle of the upper (dorsal) side of the creature; in ordinary conditions it is never on the under side. Sometimes the flagellate turns on its side, when the close adherence of the flagellum is well shown (fig. 10). Yet, in spite of this close contact, there is certainly no attaching membrane developed, for, in individuals killed by osmic acid, the flagellum is sometimes seen to stand off completely from the body (¢. fig. 12). Frequently there is a definite row of three or four conspicuous granules along the line of contact of the body with the flagellum (figs. 10, 11). The nucleus can generally be seen as a clear, spherical area near the anterior end ; in individuals in motion it is sometimes difficult to see the karysome, but in those quiescent, or in individuals killed by osmic acid vapour, this body can also be made out (cf. fig. 12). Helkesimastix is not a binucleate ; it has no kinetonucleus.* There is certainly no cytostome or definite mouth-aperture present. We are not quite certain, however, whether the creature does or does not ingest solid particles, such as small cocci, etc.; we are strongly inclined to think it does not. In this connection, a peculiar mode of food-ingestion which we have observed in Cercomonas (or Cercobodo) longicauda is of interest. While * We may add that, as seen in stained preparations, the nucleus is of the usual flagellate character, consisting of a nuclear membrane, a clear zone (the nuclear sap), and a large central karyosome, connected with the membrane by delicate radiating fibrils. Observations on the Life-Cycle of Helkesimastix feecicola. 357 an individual was moving sluggishly along, the posterior, plastic part of the body would come into contact at some point with a small extraneous particle. A small portion of the protoplasm of the flagellate in the imme- diate neighbourhood of the point of contact would remain adherent to the still stationary particle for a few seconds, gradually engulfing it, while the Cercomonas moved calmly on. Thus most of the body was quickly separated from the small portion of cytoplasm remaining behind, until often only an extremely thin thread or line of protoplasmic substance still joined the two parts; this thread might be as long as the whole body of the flagellate. Suddenly the tension of the connecting thread overcame the resistance of the stationary particle, and the latter, together with the small portion of protoplasm surrounding it, was safely hauled up again into the main body. We often thought the thread must break, but it never did! As our new flagellate shows a resemblance in some respects to Cercomonas, we have watched particularly for the occurrence of anything corresponding, which, however, we have never seen, although analogous appearances are seen during conjugation (cf. below). We have never been able to satisfy ourselves that Helkesimastix does engulf solid particles at the hinder, plastic end, but do not deny the possibility of it doing so. We certainly consider, however, that its principal mode of nutrition is by osmosis. The natural habitat of the creature, namely, moist dung, is, of course, rich in organic matter in solution, when the bacteria have been active fora little time. In Helkesimastix, therefore, we have an instance of a form which is, at all events, mainly saprozoic. The contractile vacuole is usually small, that is to say, it contracts before it attains a large size. It is generally situated in the hinder part of the body, to one side (figs. 8,10). Now and again, however, owing doubtless to some condition of the medium, the contractile vacuole becomes very large (fig. 15). Immediately after it has ruptured the protoplasmic wall and its contents have passed out to the exterior, the body of the flagellate presents for a short time the curious appearance of fig. 16 ; the hinder end is forked, like the two arms of a V. After a little while it becomes triangular (fig. 17), and ultimately assumes again its normal shape. Another point illustrating the looseness or plasticity of the body is a method of turning round often shown by asluggishly moving individual. It will come to rest, and the body becomes more ovoid (fig. 18). Then the side with which the flagellum is in close contact begins to show what we term “ working” movements ; at first the peripheral protoplasm moves in slow, short, irregular waves to and fro, the line both of the contiguous part of the flagellum and of the row of granules (if these are present) becoming meanwhile indented and 358 Dr. H. M. Woodcock and Mr. G. Lapage. uneven (fig. 19). These peripheral movements of the protoplasm can perhaps be compared, on a very small scale, with the wave-like movements of the ectoplasm of Ameba verrucosa. Next, the anterior portion of the body- protoplasm on this side is moved as a whole backwards, around the central part, as it were, carrying with it the anterior part of the flagellum and the granules, and also, doubtless, the nucleus (cf. fig. 20). Finally, the creature elongates again, having the anterior part of its body now where the posterior part formerly was (fig. 21), and is ready to swim away in just the opposite direction. Movements.—Helkesimastiz possesses two distinct and characteristic methods of locomotion. One, the more usual mode, is very interesting, because it is very difficult to explain; in fact, we do not know quite how to explain it. The creature glides forwards with a steady, almost unwavering movement, the flagellum trailing behind in a straight line. While the rate of pro- gression is not: as rapid as that of a Monas or a Bodo, for example, the movement cannot by any means be called slow; indeed, it is often surprisingly fast, considering how little there seems to be to account for it. There is certainly no vibration of the free, distal part of the flagellum at all; in this method of movement, the flagellum does not act as a pulsellum. The creature is always at the surface of the medium when moving in this way (or it is gliding along the under surface of the cover-slip), the flagellum in both cases being uppermost. This fact leads us to think that surface-tension plays some part in this type of movement. The body is, as it were, suspended along its length to the flagellar thread, as a gymnast may be suspended along a tight rope. The body is often seen to swing side- ways partially around its flagellum, appearing then as in fig. 10; but 1t never swings right above the flagellum. The only movement of the body which can be noted is a very slight “knicking” movement of the anterior end, 7c. the anterior end may make little tentative jerks from side to side, perhaps caused by slight contractions of the anterior end of the flagellum. But the anterior end of the body is not displaced laterally by more than half its width, if as much, and the strength of these slight movements appears wholly insufficient to produce the steady forward progression ; moreover, the creature may glide for quite a considerable distance without even these. The other characteristic mode of progression occurs when an individual is in the middle of the fluid, 7.c. completely surrounded by it. Then it performs vigorous, more or less undulatory movements of its whole body and flagellum, the latter, as regards its applied portion, never becoming separated from the body, and with its free, distal part lashing actively behind. The movement of the body is very like that of a fish’s tail, and not at all unlike the Observations on the Infe-Cycle of Helkesimastix feecicola. 359 movements of certain trypanosomes. Yet, in spite of all this activity, the rate of progression is no greater than is obtained by the quiet surface gliding. One other variety of movement—not of progression—remains to be noted. An individual which has been gliding about will become anchored by the end of its flagellum to some particle of débris or small clump of bacteria. The body will then execute sharp bending or knicking movements about its narrow, posterior end, where the flagellum becomes free, often turning through nearly 180°, and then turning back again. This anchoring process recalls the anchoring of some Bodos by the trailing flagellum, but the body movement in Helkesimastix is not at all of the vibrating or dancing character seen in the Sodo, because it lacks the anterior, vibratile flagellum. Multiplication—After the flagellates have been active for some hours multiplication begins. We have not observed it occurring up to six hours after excystation has taken place, but by the end of 20 hours it was proceeding actively and had evidently been going on for some time. An individual about to undergo division always comes to rest in the first place. The body becomes ovoid and then practically spherical (figs. 22, 24, 27). The subsequent course of events is not always quite uniform, though the variations are only slight. In the great majority of cases, by the time the body-form has become rounded, or even after it has been ovoid for a short while, the flagellum is no longer visible. The free part has entirely disappeared, and we are strongly inclined to think that the attached part also goes, though we cannot write with absolute certainty because it is extremely difficult to detect a motionless flagellum lying over, and closely applied to, the body. However, we think that, in many cases, it is probably entirely withdrawn and absorbed, especially as, just prior to this stage, small “ working” movements of the peripheral protoplasm occur at the side where the flagellum lay. At this stage the clear nuclear area can frequently be seen to be elongated and to possess now two karyosomatic bodies (fig. 22), the parent-karyosome having already divided. Next, the body begins to elongate again (figs. 23, 29), and in a minute or two more becomes slightly dumb-bell shaped (fig. 25). The two daughter-nuclei are practically reconstituted and have begun to separate by this time (figs. 23, 29); they can usually be seen because the body remains motionless during this period. (We hope to give information with regard to the cytological details both of multiplication and conjugation in a subsequent memoir, when we have studied fixed and stained preparations). The time of appearance of the first new (daughter) flagellum varies somewhat. As a 360 Dr. H. M. Woodcock and Mr. G. Lapage. rule, it does not appear until the dumb-bell stage is reached, when it can suddenly be seen projecting out from the side of the body, a short distance from one end (fig. 23), and waving slightly to and fro. Occasionally, however, it can be seen while the creature is still rounded (fig. 24), and in such a case it may represent the old flagellum, which has been only partially withdrawn. The constriction at the middle of the elongated body now rapidly increases (figs. 25, 30), and about this time the second flagellum appears, always some distance away from the first and in the other half of the body, not far from the second nucleus. Very generally, the second new flagellum projects out from the side of the body opposite that where the first one is. The flagella increase in length and the body undergoes little irregular, jerky movements. Its middle part becomes narrower and more drawn out, the whole body having now the appearance of a double pear. Usually the two flagella have their free ends directed towards the middle, the bluntly rounded extremities becoming the anterior ends of the two daughter-individuals; though sometimes the second daughter-flagellum (the later developed one) starts from near the middle of the body, ze. near to the constriction connecting the two halves, and points outwards. Ultimately, helped by the movements of the flagella, the two halves of the body are drawn still farther apart and the small daughter-individuals at length separate, gliding away in opposite directions. The cytoplasmic “ tail,” which each at first possesses, rapidly contracts and the typical body-form is attained (fig. 31). ; In the above type of division, which is the most usual one, we regard the cytoplasmic fission as being approximately transverse to the original long axis, so that in this case we have the flagellate undergoing transverse division of the body. The whole process is fairly rapid. From the time when an individual has become ovoid and practically motionless to the time of separation of the two daughter-individuals only 10 to 15 minutes usually elapse. Now and again, however, the process is slower, taking upwards of 25 minutes ; but this isof rare occurrence. In such a case, moreover, we have noticed that the body becomes divided before the second flagellum is formed, so that one daughter-individual swims away, leaving the other motionless for a few minutes longer, until it has developed its flagellum. On one or two occasions we have observed a modification of the above method of division, a second flagellum being formed while the old one is still present, having only been withdrawn (shortened) a little (fig. 32). The shortening proceeds further (fig. 33), but in this case the old flagellum is not entirely absorbed but forms the basis of one of the daughter-flagella. In this variety of division we regard the fission of the body as being more in the long Observations on the Life-Cycle of Helkesimastix feecicola. 361 axis, rather than transverse. This may represent a more primitive mode, which has been largely relinquished in favour of the other. Syngamy.—tIn a fresh culture (either agar-plate or observation-preparation), after the flagellates have once emerged from their cysts, multiplication goes on, often at first with amazing rapidity, for two days or so, until by about the third day—or sometimes even earlier—an epidemic of conjugation sets in. The only important point in the whole life-cycle in regard to which we are not yet certain is whether definite conjugating individuals, gametes, morpho- logically distinctive from the usual forms, are developed, and, if they are, whether these are anisogamous or not. Our difficulty arises from the fact that in a culture in which conjugation is beginning, the flagellates present show more or less the customary variation in size and form ; and, further, we have not yet succeeded in seeing two individuals actually come together and unite. In most of the cultures in which we have observed conjugation, the majority of the individuals belong to one of two slightly different types. One of these is rather characteristic and distinctive, we think, of this period. It has the posterior end of the body gradually tapering and always turned definitely to one side (figs. 34, 35), usually, though not invariably, the right side, when the flagellum is dorsal or uppermost. The curved tail-portion differs from the irregular extension sometimes seen at the hinder end in ordinary individuals (cf. fig. 11), in being fairly constant and not so changeable or metabolic, now retracted and now drawn out, as in that case. The other type is distinctly smaller, but is not so readily distinguishable from an ordinary individual (fig. 36); it is oval in shape, and the hinder end is usually more bluntly rounded. In an observation-preparation in which syngamy has begun, two individuals are often noticed to come into contact and glide along together for a short while (fig. 37). The members of such a pair are very frequently—though, again, not invariably—of the two distinct forms just noted. The larger one of the two often appears to stick, or become attached to the other by its curved, hinder end (fig. 38); when this happens, both individuals get very excited and actively jerk themselves about for a moment or two; then they will either separate abruptly or glide along together for a short distance again, and then move apart. Unfortunately, we have never seen this process followed by actual union, and therefore cannot say whether it represents a tentative seeking out of each other by definite gametes. Only in one case, up to the present, have we caught the two gametes in close contact, with the body-protoplasm of each still separate just for an instant before joining up (fig. 39) ; and, in this case, so far as could be gathered from our momentary impression before the two protoplasmic masses ran together, as it were, into 362 Dr. H. M. Woodcock and Mr. G. Lapage. one, the two conjugating individuals were not very dissimilar. We leave the matter there for the present, but hope to settle it before publishing our detailed account.* Of course, the actual coming together and uniting is a matter of a few seconds, and therefore difficult to catch; but the gradual fusion of the two gametes and the subsequent development of the zygote is a long process, and we have observed every stage in it, on many occasions. The actual cyto- plasmic union is lateral (fig. 40), and immediately after it has occurred one would think that a single protoplasmic entity was now constituted. But the subsequent behaviour is amazing and unique, so far as we are aware, in the history of conjugating elements, and well illustrates the looseness of the first union and the fluidity of the protoplasm in Helkesimastiz—and, we doubt not, in other of these dung- and infusion-flagellates. As soon as the actual cytoplasmic union has occurred, the definite form of the two gametes is practically lost, and there is for some time a remarkable lack of attraction, or reluctance to unite, between the two essential parts, if we may thus regard the nuclei and associated elements together with the portion of cytoplasm immediately surrounding them. (This is probably because the nuclei have not yet undergone a process of maturation.) We cannot do better than describe the sequence of form-changes undergone by the zygote immediately following the union, in the instance referred to. The conjugants of fig. 40 had not been joined for more than a minute or so before they separated again in the anterior region, one individual being desirous of steering off to the right, while the other preferred to keep straight on (fig. 42). This little difference being composed and the two individuals joined up again, one began to lag behind the other, the com- bined body appearing now like an irregular rhomboid (fig. 43). The conjugants were progressing forwards, more or less steadily, all the time. In the next instant the smaller half had slipped still further back, and the body had now the appearance of fig. 44. A few seconds later it was quite behind the larger half (or individual), the two being connected together only by a narrow cytoplasmic thread (fig. 45). After progressing thus for a short distance the smaller half rapidly overtook the one in front by a kind of “slithering” movement along it, and the protoplasm of both again joined up along the side (fig. 46). A few seconds later a very characteristic stage was reached, in which the combined body of the two conjugants * We have since reconsidered the above point and now think that the tendency to adhere in couples in this way is probably purely a matter of surface tension or attraction. We have observed the same phenomenon in observation-preparations of a non-conju- gating strain of a closely allied species (vide p. 366, et seq.). Observations on the Infe-Cycle of Helkesimastix feecicola. 363 was almost square, the whole zygote, with its two flagella trailing along near the outer sides, having almost the appearance of a procession banner (fig. 47). (Ata later stage the body becomes more definitely rectangular and banner-like.) In this particular instance the zygote remained thus for a couple of minutes or so, moving along steadily, the two “halves” with their flagella at times approximating slightly, causing the common cytoplasm to sag, as it were, just as a banner does when its pole bearers do not keep their line. Now and again the zygote turned on its side, when it appeared as in fig. 41. At length it altered again completely, a portion of the body, apparently about a third of the bulk, advancing quickly in front of the remainder (fig. 48), until the two portions were only connected by a very narrow thread of cytoplasm, which was at first extended along and in contact with. the greater part of the flagellum (fig. 49). We have no doubt that in this remarkable dissociation of the zygote into two portions, one nucleus goes with each half, just as the origin and proximal part of the flagellum can clearly be seen to do. Unfortunately, we could not make out the nuclei at all during these living observations of the conjugation-processes ; this was due partly to the fact that in the earlier stages the conjugating pairs are very active and constantly undergoing form-changes, and partly because, in the later stages, the nuclei are most probably undergoing maturation prior to nuclear fusion. This separation, which may amount almost to disruption, of the common cytoplasm into two portions reminds us of the behaviour of the protoplasm of Cercomonas (Cercobodo) above alluded to, but in Helkesimastix it is not a passive leaving behind of a certain amount of protoplasm, but a separation actively brought about by a difference in behaviour of the two gametic energids. Here, again, in more than one instance (c/. fig. 54), we felt sure that actual rupture was going to occur, but we have observed this remark- able process on several occasions and the two halves never broke loose. To return to our particular zygote, the small leading portion suddenly turned right round, its flagellum becoming at the same time mostly free from the cytoplasmic thread (fig. 50), and then joined up again to the main portion (figs. 51 and 52). The zygote next assumed the form of a pear (fig. 53), and after another minute or less the stage of fig. 45 was repeated, with the difference that the smaller half was this time in front. The hinder part then “slithered” up along the other and the banner form was again arrived at. All the above changes took place in a period of about 14 minutes from the instant of first union. This time the banner was fairly permanent, and after following it for a little while longer we left it, as we knew exactly what the subsequent behaviour would be from numerous earlier observations. The above described remarkable behaviour of the two conjugants—or, at least, of 364 Dr, H. M. Woodcock and Mr. G. Lapage. the two essential portions, since the cytoplasm is apparently indiscriminately divided at times—is not at all an exceptional occurrence; indeed, we think something similar usually happens during the earlier stages of Ssyngamy. On other occasions we have seen both the “slithering” of two slightly unequal portions and the almost complete separation of the two halves which remained connected only by a delicate thread, just as in the above case. The joining up again in such a case occurs, we consider, as a result of the rapid contraction of the connecting thread, in just the same manner as the lagging portion of protoplasm was suddenly hauled up into the main body of the Cercobodo (cf. above). Further examples of these early conjugation stages are given in figs. 54-57, from different zygotes. Even after the banner has been definitely formed for as long as an hour, we haye seen it suddenly break down into two portions, one behind the other. Different forms assumed by the banner, when at length permanent, are seen in figs. 58-62. As time goes on the banner gradually loses its square shape and becomes oblong (fig. 63), the two parallel flagella also gradually coming to lie nearer to the mid-dorsal line, and so to each other. Most provably, by this time, the maturation processes of the gamete-nuclei are completed. Further, the two flagella are slowly shortening in length. In its. earlier period, however, the banner is still capable of performing the active, undulating (free) movement, as well as the more characteristic gliding move- ment. From an oblong the banner now turns to an oval—really, of course, to an ovoid—which is ultimately almost as deep as it is broad. We have followed a banner from the time when it had about just become permanent (eg. as in figs. 61 and 62), up to the definite oval (fig. 65), the time occupied being about three hours. In an observation-preparation in which syngamy has been going on for some time, banners and ovals are quite numerous ; this will be apparent when it is remembered that every cyst formed is a zygote- cyst. Ata later stage the two flagella have become much shorter (figs. 66 and 67), and the flagellate moves very sluggishly ; it soon ceases to displace itself and only performs little turning and “knicking” movements. The protoplasm becomes contracted and somewhat denser, and the oval nearly always leaves the under surface of the cover-slip by this time and sinks to the lower level of the medium. The flagella are at length quite absorbed and the body becomes spherical. We have followed an individual oval with two flagella close together, as in fig. 65, up to the time when it became a motion- less, rounded body at the lower level, and this change took about six hours. The whole process up to this stage takes from 9 to 10 hours, according to our actual observations; of course, this may not be the minimum period required, though we should say it is not far from it. Observations on the Infe-Cycle of Helkesimastix feecicola. 355 We have not actually watched a motionless, rounded zygote through its encystment period, but we noted the position of the above individual and of others at the same time late in the evening, and on looking at them next morning we found they had become the very characteristic “shrinkage” cysts — (figs. 68 and 69). The cyst-wall has been formed and the protoplasm has continued to contract, so that a space is left at one side between the body and the wall. We are inclined to think it is a space containing no liquid, because it is always very clear. The space appears of different size, according to the age of the cyst (figs. 68 and 69). When small the space is spherical, but when it attains its maximum size it is lens-shaped. A curious fact is that after the cyst has shown this condition of shrinkage for some time—the period may vary from two days or less up to longer—the space disappears entirely and the cyst becomes a permanent cyst, as described at the com- mencement of this account. This finishes the life-cycle. Biological Notes—As, in a normal culture, the great majority of the flagellates form cysts, one could not have a more convincing and readily obtained demonstration of the regular occurrence of syngamy in the life- eycle of these simple dung-flagellates. Although all the above observations on Helkesimastix have been described either from plate-cultures or from observation-preparations, this is only because, on account of the more rapid multiplication and development under those conditions, the individuals are very much more abundant than in a simple, moistened dung-culture, and therefore the different stages are more readily found. We have not the slightest doubt that the life-cycle is exactly similar under the more natural conditions, the only difference being that it is probably slower, ic. a longer period may elapse before the whole life-cycle is completed. For instance, it is usually three or four days before this flagellate is observed in the active condition in a dilute dung-culture, and it may persist in the active condition for a week or more before forming cysts. The reason for this is, we consider, because the watery medium is not nearly so rich in nourishment as the beef- extract medium. The bacterial development is not nearly so great as in the latter case, nor is the multiplication of the flagellates so abundant. Therefore, on the one hand, a somewhat longer time most probably elapses before the eyst-wall is dissolved ; and, on the other hand, the medium does not so soon become excessively full of the “toxic” products, or whatever chemical substances induce the cessation of multiplication and the tendency to conjugation, whether formed from the flagellates, or the bacteria, or from both. As, however, we are at present engaged in making a full study of this interesting flagellate from a biological standpoint we will not further discuss these questions at present. 366 Dr. H. M. Woodcock and Mr. G. Lapage. Loss of Syngamy and the Power to form Cysts—We have discovered one very interesting and important fact, however, which deserves to be mentioned. By sub-culturing the flagellates* while they are still all in the active condi- tion, v.e. before cysts have been developed, on to a fresh plate of the same medium, we have found that multiplication will continue for a further period (of two days or so), before conjugation sets in. The important fact, however, is that, after a certain number of sub-cultures have been thus made, the flagellates, althotgh they are still able to thrive and multiply actively on each successive transference, no longer undergo syngamy, followed by cyst-formation, and have, so far as can be seen, entirely lost the power to do so, at all events under the existing conditions.t Up to the present time (October), we have thus kept a “strain” for more than 20 weeks, through 35 sub-cultures on to fresh “ constant” medium, and in each sub-culture the ’ flagellates have multiplied enormously. In the sub-cultures up to the fourth a very few isolated cysts were still found. But in none of the subsequent ones has a single cyst ever been seen. The flagellates, instead of conjugating and forming great numbers of cysts, as usual, degenerate and die off, only a small proportion remaining still alive. Yet, at the end of 10 or 12 days, sometimes more, a few individuals still persist alive, and the transference to a fresh sub-culture can be successfully made. The flagellates which remain alive are rounded, granular, very sluggish forms, with the flagellum sticking straight out and almost motionless; but, in the fresh medium, they are capable of quick recuperation and renewed multiplication. During the earlier weeks, after this new strain was fairly started, we never observed any biflagellate forms, comparable to the conjugating pairs above described, although thousands of individuals must have passed under our eyes. Then, during the 8th and 9th weeks, respectively, we observed on two occasions, in different observation preparations, a single biflagellate individual. After this, we looked carefully for others, at intervals, but no more were seen until the 14th week (in the 23rd sub-culture), when another was observed. Since that time, in most of the sub-cultures, a few biflagellate individuals (banner-like forms) have been found and the number of these has gradually tended to become less infrequent, although they still constitute a very small * This experimental work, it may be mentioned, has been carried out upon another species of Helkesimastix, with which we have worked latterly. This form (H. major, n. sp.) also occurs both in sheep and goats, and differs only from H. facicola in the larger size of the adult individuals and also of the cysts, rendering it more convenient for study. The life-cycle is similar-in both. + We tried the experiment of re-introducing the non-conjugating “strain” into the original dung-culture (sterilised) with the object of seeing whether the ability to undergo syngamy would he restored. This also gave negative results. Observations on the Infe-Cycle of Helkesimastix feecicola. 367 proportion only of the total number of flagellates present. We have been able, on several occasions recently, to follow the further behaviour of these forms, and have ascertained the important fact that they always divide, and never become rounded-off and form cysts. Moreover, we have observed a certain number of these forms which were very large and possessed not two, but three, or even four or five flagella (and nuclei, as ascertained from permanent preparations). We have seen an individual with three flagella, and also one with five, actually divide into two rather unequal portions ; in the former case, one half had two flagella and the other one, and in the latter case, one had three and the other two flagella (and nuclei). We will assume, for the present, that these forms represent a union of two or more individuals, rather than a long-delayed division.* When we first observed the isolated instances of biflagellate forms in this strain, we con- sidered that they represented true syngamy, which, although occurring very rarely was apparently still “latent.” We now know (1) that these “unions” may be either binary or multiple, apparently more or less indifferently ; (2) that these forms always divide, the nuclei and flagella being apportioned out, often unequally, between the two halves, and that they never proceed to form cysts—the invariable sequel, normally, to conjugation in Helkesimastiu ; (3) and, lastly, that these forms, or the products of their division, are equally incapable of persisting ultimately (unless, of course, they are transferred to fresh medium). Hence, we feel certain that none of these unions represent true syngamy. Bearing in mind the extremely plastic character of the cytoplasm of this creature, we consider these unions are due to physical, rather than “vital” factors, and result from continued cultivation and the condition of the culture at the time. In our opinion, these peculiar cases do not invalidate the following general conclusion which we wish to draw from our observations on this non-conjugating strain. The “intensive” culture of this flagellate has resulted in the loss of the power to undergo true syngamy and to form cysts. The further existence of this “strain” is now dependent on continued transference to fresh “constant” medium. Minchin, in his ‘Introduction to the Protozoa’ (London, Arnold, 1912), in the chapter on “Syngamy and Sex,” has pointed out (p. 161) that “intensive culture, whether artificial or * We cannot yet write with certainty upon this point, because we think, from the evidence obtained so far, it is possible that these forms represent a long-drawn-out mode of division, due to the effect of the cultural conditions, in which the nucleus and flagellum divide, it may be more than once, before the cytoplasm does. There are other cases known of a corresponding behaviour under conditions which are probably not quite normal ; an example which is particularly interesting in relation to the present discussion is that of multiple longitudinal fission in certain lethal trypanosomes. 368 Dr. H. M. Woodcock and Mr. G. Lapage. natural, as in parasitism, seems to diminish the necessity for syngamy ”; from this stage it is only a step further to find the capacity for syngamy lost. Bearing on the Case of the Trypanosomes.—It appears to us that this remark- able experimental fact has an important bearing on the question of the trypanosomes, and may afford an explanation of why conjugation (or syngamy): in these parasites, though assiduously sought, has not been observed in any authenticated instance. Probably it never does occur, because these forms have lost the power to conjugate. Just as, in the case of the above-discussed strain of Helkesimastix, the rapid, successive transferences to fresh, non- toxic medium at first removed the necessity for conjugation (and encystment), and then have led to the loss of the ability to undergo syngamy (and, in this case, to form cysts), so a similar development has very likely occurred in the trypanosomes and related forms. As is obvious, the conditions of life in these parasites are readily comparable to those above described. The trypano- somes live in a rich nutritive medium, namely blood, in the vertebrate, or blood in various stages of digestion, in the invertebrate host; and, as is well known, rapid multiplication in both hosts is usually found. In the inverte- brate we get, more or less frequently, replenishment, 7.e. a fresh supply of the “ constant” medium (namely blood), together with removal of toxic products. In the blood of the vertebrate, where there is at first abundance of medium, but not any “fresh” supply, it is very interesting to note that, in the case at all events of many lethal trypanosomes, which, as Minchin has pointed out, are not yet completely adapted to their hosts, we get the pro- duction eventually—after active multiplication has gone on for some time— of the well known “involution” forms. These, at any rate in our opinion, as in that of many French workers, simply die off after a time, if left to themselves; but they are capable of being “rejuvenated” and of again multiplying actively if passed into a fresh host. And a similar state of affairs has been met with by one of us in cultures of avian trypanosomes. Now it seems to us that these two cases present a very close parallel to what we have observed in Helkesimastiz. It may be pointed out, also, that the non-conjugating strain of Helkesimastix is the result of an abruptly. originating and more or less artificial intensive culture, whereas in the case of trypanosomes and allied parasitic flagellates the corresponding conditions have operated naturally and over a long duration of time. On the supposition that syngamy has now been entirely lost in the hamo- flagellates, the explanation which we here put forward is much more probable, we think, than the idea, which has also been suggested, that the stimulus afforded by the change of hosts is accountable for the loss of this process in the trypanosomes. We have always considered that there are two or Observations on the Infe-Cycle of Helkesimastix feecicola. 369 three serious objections to this latter view. Briefly stated, these are as follows :—-(1) Trypanosomes can be successfully inoculated, without limit, into fresh vertebrate hosts, zc. the same “constant” physiological medium, without showing (so far as is known) conjugation. (2) Binucleate flagellates of insects alone (¢.g. certain leptomonads) have no alternation of hosts, and syngamy is apparently just as little likely to be found in these parasites as in the trypanosomes themselves. (We are inclined to think, indeed, that syngamy may have been lost in the ancestral trypanosome form, before ever it acquired an alternation of hosts.) Lastly, we have the important case of the Hmosporidia, intracellular parasites which all have the change of hosts, and in all of which conjugation regularly occurs. Why should not the physiological stimulus afforded by the change of environment have influenced them also? And, on the other hand, there is no reason to doubt that the primitive ancestors of the trypanosomes underwent a pro- cess of syngamy, since it appears likely that many of these lowly proto- monadine forms, among which the origin of the trypanosomes is to be sought, possess this feature. We consider, therefore, developing the ideas expressed by Cropper and Drew (Joc, cit.) along the line indicated by the experimental facts adduced above, that the loss of syngamy is due to the surfeit of nutrition, together with the non- toxicity of the medium, that is to say, the absence (in excess) of the chemical substance or substances to which the flagellates react normally by the cessation of multiplication and the onset of conjugation. It will be readily apparent, from what has been just pointed out, how these factors have pre- vailed in the case of the trypanosomes. And we think that a similar explanation can be applied, not only to the case of insectan Binucleata, but probably to that of many other parasitic flagellates as well. bo is] VOL. LXXXVIII.—B. 370 Observations on the Life-Cycle of Helkesimastix feecicola. EXPLANATION OF FIGURES. All the drawings, with the exception of fig. 12, are from sketches made directly at the time of the living observation. Fig. 12 is a camera lucida drawing of an individual fixed with osmic acid vapour. The magnification throughout is approximately 2250 times linear, arrived at after a comparison of other individuals fixed with osmic. We are indebted to Miss Rhodes for kindly tinting and sharpening up the drawings. In our figures, we have shown the nucleus or omitted it, according as to whether we were able to observe its position in the particular individual represented, or not. As mentioned in the text, we were unable to make out satisfactorily the nuclei in life, in the conjugating pairs. Note with regard to the flagellum :—As mentioned in the text, the length of the flagellum varies in individuals otherwise similar. In order to save space, we have drawn the flagellum usually short, except in fig. 10; the flagellum of the individual of fig. 12 is, of course, natural length. Puate 13. Figs. 1-3.—Cysts. Figs. 4-7.—Excystation and the development of the flagellum. rey Figs. 8-11.—Different forms of individuals. Fig. 12._Individual fixed with osmic acid vapour, to show that the flagellum is not actually attached by any membrane to the body. Figs. 13 and 14.—To illustrate the behaviour of the flagellum in turning of the body. Figs. 15-17.—Rupture of a large contractile vacuole and form-changes of the body. Figs. 18-21.—To illustrate a mode of turning of a stationary individual (see text). Figs. 22-26, 27-31, and 32 and 33.—Different instances of division (binary fission). Puate 14. Figs. 34-36.—Different forms predominating in a culture about to start conjugation. Figs. 37-62.—The whole process of conjugation (syngamy), showing the different form- changes during the progress of the fusion. Figs. 63-67.—The gradual contraction of the body and absorption of the two flagella of the zygote, prior to encystment. Figs. 68 and 69.—The cysts as first formed, so-called “ shrinkage” cysts, with a vacuole or space of varying size. ~ Woodcock and Lapage. Roy. Soc. Proc., B, vol. 88, Plate 13. ©) Cinat \ = See ye \= x i) oOo 5S Roy. Soc. Proc., B, vol. 88, Plate 14. | Woodcock and Lapage. 371 The Antagonistic Action of Carbon Dioxide and Adrenalin on the Heart. By 8. W. Patterson, M.D., Beit Memorial Research Fellow. (Communicated by Prof. Starling, F.R.S. Received August 14,19 14.) (From the Institute of Physiology, University College, London, and the Physiological Institute of the University, Berlin.) Although a great volume of work on asphyxia has been published, it is only comparatively recently that attempts have been made to dissociate the influence of various factors in the production of the phenomena observed. Kaya and Starling (1) were the first to differentiate the effects of lack of oxygen and excess of carbon dioxide in the spinal animal; and their work was elaborated by Mathison (2, 3), in whose papers a full discussion of the previous literature will be found. He found that during nitrogen adminis- tration no increased output of the heart is seen in the early stages of asphyxia, aud attributed the increase in output noticed in ordinary asphyxia to the presence of increased tension of CO: in the blood, which Jerusalem and Starling (4) had shown to increase the systolic output of the cat’s heart. He also observed an acceleration of the heart beat about the time of the primary blood pressure rise in asphyxia, which occurred even after removal of the upper part of the spinal cord. Since the work of v. Anrep (5) and Itami (6), a third factor, variations in the secretory activity of the suprarenal glands, must be taken into consideration ; and the present paper contains an account of an investigation of the action of carbon dioxide and adrenalin on the heart isolated from the nervous system. Methods. The experiments reported in this paper were carried out mainly on dogs, a few on cats, the animals being anesthetised by inhalation of chloroform and ether mixture, in the case of dogs after a preliminary hypodermic injection of morphine. The isolated heart-lung preparation was made as described by Knowlton and Starling(7). The systemic blood was taken off by a cannula in the brachiocephalic artery after ligation of the left subclavian artery and the aorta beyond; and was returned to the heart by a cannula tied in the superior vena cava after ligation of the azygos vein. The mean blood pressure was recorded by a mercury manometer connected to the side of the innominate cannula, and the side pressure of the blood from the venous return by a water manometer connected with a cannula tied in the inferior vena cava close to its entrance into the right auricle. To the top of the water 2F 2 372 Dr. 8. W. Patterson. The Antagonistic manometer a small piston recorder was attached for graphic records; 0:1 grm. of hirudin was added to the blood to prevent clotting. In the case of cats, the systemic schema (consisting of the arterial resist- ance and venous reservoir) used was that described by Knowlton and Starling ; while the other experiments were made at various times with the different modifications of method which have marked the evolution of the systemic schema, and which have been reported in the communications from the University College laboratory during the last two years. References to these will be made in the discussion of results, as the separate points brought out are considered. In all cases the systemic output was measured directly into a graduated vessel by means of a stop-watch, and is recorded in cubic centimetres per 10 seconds, the output per minute being this observed figure multiplied by 6. When the heart volume was recorded, a glass plethysmograph on the ventricles was used similar to that described by Henderson (13), and the cardiometer was connected by air conduction with various recorders. In some experiments the mean pressures in the left auricle were recorded froma water manometer connected with a small cannula tied in the appendage of the left auricle. The adrenalin (Parke, Davis and Co.) was mixed with the blood in the venous reservoir, while the carbon dioxide was administered from a gas bag. The figures given of the percentages of CO2 in the gas bag can only approxi- mately show the percentage in the air at the trachea, as various types of artificial respiration apparatus were used, and were connected to the trachea by sometimes considerable lengths of piping, and as the side slot of the tracheal cannula was more or less open. Results. In the heart-lung preparation, as in the intact animal, there are three circles of blood from one side of the heart to the other, one from the right ventricle to the left auricle through the lungs and two from the left ventricle to the right side of the heart, one of which passes through the schema and represents the systemic circulation in the whole animal, and the other passes through the coronary circulation. The total output of the left ventricle consists, therefore, of the systemic output, which was measured directly in our work, together with the output through the coronary vessels; so the question of the relation of the coronary output to the systemic output under normal conditions and under the influence of carbon dioxide and adrenalin must first be considered. Relation of Coronary Sinus Flow (as Index of Total as onary Output) to re 2 Action of Carbon Dioxide and Adrenalin on the Heart. 373 Systemic Output.—Evans and Starling (8) found a constant relation between the flow from the coronary sinus and the total coronary output in the proportion of 3:5; Markwalder and Starling (9) proved that the coronary flow depended on the arterial pressure, and, using the above figure as a basis for calculation, showed that the total output from the left ventricle was constant for a given venous inflow and independent of the arterial resistance within very wide limits. They have confirmed also the observations (10) made previously, that adrenalin causes an increased coronary flow. Table I* contains the results of an experiment carried out on the heart-lung preparation in which a Morawitz cannula was introduced into the coronary sinus, and the blood flow through the coronary sinus and that through the systemic part of the schema were measured at the same time. The figures in column 10 (Total coronary circulation) were obtained by multiplying the observed coronary sinus flow by the factor 5/3. It will be seen that in the _two series with the normal heart, the coronary output depends on the aortic pressure, while the total output is independent of the arterial resistance within wide limits and is conditioned only by the venous inflow. During the period of administration of carbon dioxide the coronary flow was the same as in the normal condition: but it was increased markedly during the recovery period from carbon dioxide, while the heart was returning to its normal state. This imcrease was probably due to the accumulation of ‘metabolites ’ in the heart during the action of COs. Adding adrenalin to the circulating blood caused a great rise in the coronary flow, and this occurred also when adrenalin and carbon dioxide were given together. We have thus a guide to the interpretation of the results obtained in other experiments where the systemic output only was measured. Systemic or Effective Output with Carbon Dioxide and Adrenalin. Experiments on Dogs.—The figures in column 7 of Table I show that the administration of CO. may cause a marked falling-off in the output per minute. This diminution may be small with low percentages of COs, but I have never observed an increased output during the administration. The diminution becomes more marked as the percentage of COs» is increased ; so that if the CO: is strong, or a moderate percentage is administered for a * Jn this and the other Tables, the following abbreviations are employed :—A.R. = pressure in mm. Hg in air chamber surrounding the thin rubber tube forming the arterial resistance. B.P. = meanarterial pressure in mm. Hg as measured in the cannula in the innominate artery. I.V.C. = pressure in inferior vena cava in mm. H,O. V.S. = venous supply. =, +, — = maintained constant, increased or diminished. Systemic output = output in c.c. per 10 secs. as measured on the venous side of the artificial peripheral resistance. IE SSIES ~SS~S=S‘~S=~ — Con OL: 0 920. 0 Lg Gee | Oa TPO: 0 0-19 a 0: 67 ree | Oot 868: 0 gg0.0 | &% 008 (oy 880. 0 0. LF = 0. €¢ — OOT STE: 0 010: 0 0g ‘QPLIUGA 4Jo] PUL OPLIN 4Jo] UL efNUUL f “WIS gg JIvey “WISY ¢.¢ ‘Soq—'Z yuomLodxG 08 | os €10: 0 P- 89 S6I- 0 9. 1S 060: 0 Oor G86: 0 a G.pE | pauterexps {pee 200 OL OS L¥0: 0 ¢. 19 G08: O G. 8 88I-0 | OOL O6F- O =e 1S. |e poem “onus kL 7NNQ) OOT (0}¢) OFO: O y- 19 LG-0 1-8 OLT- O OOT OFF: O == SOT AGRI lioos akin derma ae [BULI0 NT Sy wu ‘wut 008 “998 998 008 ‘QIUAUGA 4JoT PUL OPTINe go] UL efnUUO ‘4 yavey “WISy Gg.) ‘Soq—| quouedxy Dr. S. W. Patterson. ] l | “QINSSOT | “OUT, “quad Log | | leew : : : : een -o[ouquea | | em queo tag uly, qua Jog | Gaon ul | | | ‘[BAteqgur aqBi aanssord ‘ajo ho SA-SV | qrvoyy | WIMNULIX’ FY ported o1jout0sy 4 210}SVI 4 210}8hg qavay JO UOT}eING 388 “TOTVEXLOT PUL WOTJOVIGUOL) Jo 9oIOT pue oyeT—TA AQeT, 389 +S S S s SS) S Ss S ~ ‘ajoko away] Jo Lepureutet = o]ojseIp {seayva Areuownd 0 o14L0v Jo aaNsojO 07 TOIyoRAyUOD Fo SutuUIdeq = 9f09shQ y La) | | jl 3 | | = €9 Ov | 9F0-0 €- 6¢ 9-0 L. OV 691-0 OOT STF. 0 = | Ste xele) [PKS PHAGE NY > 09 OF | SFE0-0 | 0-19 | VES. 0 Y- 8E 9FT-O oor | O8&-0 are | eer wag horeeqa\y SS GI) | Op | GabO | BOS | Tieo 9-8h | 99T-0 OOL | 08-0 = | = ap NX 8P OF | *S0-0 | 6-9 | 822. 0 8. SV | CEs. O OoT | &I%-0 es | aro ea [BULLO NY ea OF 790-0 | &-9rF | 082-0 9.69 | §82-0 | O0OL | se¢-0 Soil ee | “00 S or ' 9c0-0 | %2S | 682-0 8-7. | 912-0 | O01 | 909-0 = = Pee [8ULL0 NT = ‘Aroqae Arvuowynd pur sp1yuea qysts {wis OY yrvoy “WUsy ¢.g ‘So(j—'g quowiodxy Ss aS 8 @L1 ¢9 1Z0:0 a) |) JO | ieee | CeO | OOK | GBD | Ga url[Bueipy | = 921 Ga) || SOO) Pa || ERO | RAG i AAO | Con OO | = | Ges perverse PHEOIN I EHEC SE e ‘URIAR[OGNS Ifo] PUL OPOTIQUOA QJoyT UI BNUUR f “WIS Gy qavey “UISY G.g “Soq—') juoutIedxy S Ge fake |) | F10-0 | e9 | SOT. 0 | ES) = L610 |e NOOK | 296-10 60 | Ul[vueIpy & Oi i 2 | O90 | zs €1Z-0 Sp | 261-0 | OOL | O1F-0 | [BULIONT Se ‘Q/OLIYUIA JO] PUL OPIN JJo] UL BpMUULD {WI G) 4LRaY “WLISY C.J aes queued xy S) s 68 ST 220. 0 Copel) 1910 1/5 e938 680: 0 00T OLSO qs | 7 || OO) wis unereriny -o SS | go: ST 020: 0 G09 | 4481-0 | 0.0% 060: 0 OOr 822-0 | Silt Stevan aie a view]? F = @- LT ST GPO: O c-€9 | 686-0 ; 9-9 8&1-0 OOT LLE. 0 Se NCO Gees lt [GUILON NX ¥- OL SI 9700); O88) Oce10 | OnGr. , O9T- 0 Oot O8?- 0 | aoe OGmEl ce ei “00 | 6 ST GFO-0 | 0-09 | 812-0 | 0.0% 6VI-O oot CHO) jp = SG es he, ae [BULL NY “LOJOWOIPAVI PUL Y[OLIGUEA FYSII UI BNUULO f “WIS ZO qavey “WIsy g.J ‘Soq—'g quowtiodxy 200 Dr. S. W. Patterson. The Antagonistic Fia. 8.—Left Auricle and Ventricle before and after Adrenalin. Read from right to left. Experiment 6, Table VI. (a) T. 35°5, B.P. 84, 1.V.C. 20, O.P. 194 cc. in 10 secs., rate 24°5 in 10 secs. (normal). (b) T. 36°4, B.P. 84, L.V.C. 10, O.P. 194 cc. in 10 secs., rate 38 in 10 secs. (after 0°1 mgrm. adrenalin). Action of Carbon Dioxide and Adrenalin on the Heart. 391 and after injecting adrenalin, diastole was 63 per cent. of the total heart cycle. The shortening of the time of the contractile period is so great that, even with the increased rate of heart-beat, the heart is in a state of contraction for 3°7 seconds in 10 seconds, as against 4°8 seconds in 10 seconds. in the preceding normal period. In the heart cooled by cooling the inflowing blood, the proportion of diastole to total cycle is 50 per cent. Evans and ‘Ogawa (17) have found the metabolism of the heart to be increased by about three times after injecting adrenalin. Since the time per 10 seconds during which the heart is in a state of contraction is shorter than in the normal. period, adrenalin seems to have a specific effect in mobilising the “contractile substance,” or in exaggerating the changes taking place on the “active surface” of the muscle fibres during contraction. That the heart behaves in quite another manner after injecting adrenalin is shown in fig. 8, the protocols of which are given in Table VI, 6 and 7. They show the more rapid development of tension in the isometric period, the great rise of maximum pressure (116 to 314 mm. Hg) as the blood is shot violently out of the ventricle, and the relatively short duration of the whole contractile process. For a similar effect on the right ventricle, see fig. 10 (2), protocols in Table VI, 5. When CO: and adrenalin are administered together, there is a slightly lengthened time to get up a certain tension in the isometric period; the maximum pressure in the ventricle attains a greater height than before; and there is marked lengthening of the diastolic period, thus allowing time for greater filling before the next systole begins [figs. 9 and 10 (3)]. Output per Beat.—During the inhalation of CO2 in percentages from 3 to 11, the output per beat is equal to, or even greater than, before; but with percentages of 12-20, or more, the output per beat is diminished, and there may be no flow of blood at all. When the CO, is removed and ordinary air breathed again, the output per beat is greater than before the COs, since the efficiency of contraction of the heart recovers before the rate again becomes normal, : With adrenalin, the output per beat is diminished. Inhalation of CO, combined with addition of adrenalin, gives an output per beat, if the dosage is adjusted, equal to, or greater than, the normal (fig. 5). As-Vs Jnterval.—Adrenalin shortens the a—v interval. Small percentages of COz in the air breathed cause little or no alteration in the interval, but with larger percentages (above 12 per cent.) the interval is lengthened ; when combined with adrenalin, COs, still has not much effect ; but sometimes with large amounts of CO2 the heart rate drops to half, and a 2:1 heart-block is set up. Lewisand Mathison (18) found heart-block to be of regular occurrence 392 Dr. 8. W. Patterson. The Antagonistic Fie. 9.—Effect of Adrenalin, and Adrenalin combined with CO,, on Left Auricle and Ventricle. Read from left to right. Experiment 2, Table VI. (a) T. 35, B.P. 96, I.V.C. 70, O.P. 100 ¢.c. in 10 secs. (after adrenalin). (6) T. 35, B.P. 96, I.V.C. 70, O.P. 2 (adrenalin and CO,). i(1) Fie. 10 (1), (2), and (3).—Right Ventricle and Pulmonary Artery. Read from right to left. Fig. 10 (1).—Effect of CO,, Experiment 8, a, Table VI. (a) T. 365, B.P. 92, L.V.C. 22, O.P. 106 cc. in 10 secs. (normal). (6) T. 36°5, B.P. 92, I.V.C. 90, O.P. 2 (6 per cent. CO,). . \ Action of Carbon Dioxide and Adrenalin on the Heart. 393 Fig. 10 (2).—Effect of adrenalin, Experiment 8, b, Table VI. (a) T. 26°6, B.P. 92, I.V.C. 38, O.P. 128 cc. in 10 secs. (normal). (6) T. 36-6, B.P. 92, I.V.C. 12, O.P. 2 (after 0-1 mgrm. adrenalin). (3) Fig. 10 (3).—Adrenalin and CO,, Experiment 8, c, Table VI. {a) T. 36:6, B.P. 92, T.V.C. 12, O.P. 2 (adrenalin alone). (6) T. 36°7, B.P. 96, 1.V.C. 22, O.P. 124 cc. in 10 secs. (adrenalin with 6 per cent. CO,). VOL, LXXXVIII.—B. 2H 394 Dr. 8. W. Patterson. The Antagonistic in asphyxia and independent of inhibition ; and Mathison (19) found the cause to be due to lack of oxygen rather than excess of COz in the spinal animal ; but observed that heart-block may occur with large doses of COs, even in the presence of sufficient oxygen. Summary and Discussion of Results. Recent work on contraction of skeletal muscle has tended to establish more and more the view that the phenomena of contraction can best be described by reference to alterations of the surface energy of the muscle elements, and the length of the muscle fibres is a measure of the surface of action. In the heart the mean volume is the guide we have to the length of the muscle fibres, and it has been shown (20) that the heart reacts to increased work by increasing its mean volume, whether the increased work is evoked by greater diastolic inflow or greater arterial resistance. It will be seen, from the results given above, that COs in all doses appears to have a. depressant action on the functions of the heart; the contractile stress is developed more slowly, the heart taking a greater mean volume to carry out its work, but if the COs is continued, the observed output diminishes. Since the coronary circulation is unaltered this indicates a diminution of total ventricular output, and the venous pres- sure rises owing to the damming back of blood in the veins. Adding adrenalin to the blood circulating through a heart which is capable of responding, causes increased rate of contraction and ‘rate of development of contractile stress; the heart can develop the requisite tension more easily and from a position of shorter initial length, so the mean heart volume is shifted to the systolic side. Adrenalin seems to have a specific action in mobilising the “contractile substance” and increasing the energy changes taking place at the surface of the muscle fibres during contraction. The result is that the ventricle contracts violently and the blood is expelled under great pressure into the aorta. The coronary perfusion is greatly increased and the nutrition of the heart improved. Relaxation takes place rapidly, and since there is no resistance to the inflowing blood the venous pressure falls; but the onset of the next systole comes so early that the filling of the heart, and consequently the output per beat, are less than normal, but the total output of the ventricle per minute is equal to or greater than normal. With CO, and adrenalin combined in proper doses we still obtain greater rate of contraction and relaxation, but the whole diastolic period is lengthened; thus there is time for greater filling, and there is increased output per beat and per minute. This increased observed systemic output Action of Carbon Dioxide and Adrenalin on the Heart. 395 is accompanied by an increase in the coronary output, so that the total ventricular output is above normal. The slower contraction is also more effective in driving a mass of blood into the aorta instead of firing it out suddenly. Cannon (21) has recently summarised the evidence of the significance to ~ the organism of the function of the adrenal medulla in times of great emergency. We have found that the heart muscle is not only better nourished by increased coronary supply, but the contractile process is also strengthened in a specific manner. We have probably obtained in the heart- lung schema with maximum venous inflow and proper proportions of COz and adrenalin, the conditions occurring in short severe muscular exercise, where the muscles of the arms and abdomen are contracted, while the legs are active, all aiding the venous return to the chest, the increased depth of respiration also assisting the venous return. The small excess of COz2 is both the call to the secretion of the adrenals, which dilates the coronary vessels and strengthens the cardiac contraction, and the cause of the lengthened diastole and time for greater filling, so that the maximum output of the heart can be obtained. Conclusions. 1. Carbon dioxide alone depresses all the functions of the isolated heart. 2. Adrenalin, besides dilating the coronary vessels, has a specific action in increasing the rate and strength of ventricular contraction. 5. The effect of carbon dioxide and adrenalin combined is still to allow of more rapid and stronger contraction and rapid relaxation, and also to lengthen the diastolic period. Thus greater filling of the heart takes place and the heart is in a better condition for putting out a maximal output. I have much pleasure in recording my thanks to Prof. Piper, of Berlin, for his assistance with the endocardiac pressure tracings, and to Prof. Starling, of London, for the initiation and successful carrying out of the whole work. REFERENCES. Kaya and Starling, ‘Journ. Phys.,’ vol. 39, p. 347 (1909). Mathison, ‘ Journ. Phys.,’ vol. 41, p. 416 (1910). Mathison, ‘Journ. Phys.,’ vol. 42, p. 283 (1911). Jerusalem and Starling, ‘ Journ. Phys.,’ vol. 40, p. 279 (1910). VY. Anrep, ‘Journ. Phys.,’ vol. 45, p. 307 (1912). Itami, ‘ Journ. Phys.,’ vol. 45, p. 338 (1912). Knowlton and Starling, ‘Journ. Phys.,’ vol. 44, p. 206 (1912). Evans and Starling, ‘Journ. Phys.,’ vol. 46, p. 413 (1913). VOL. LXXXVIII.—B. 21 poe NS) ES eae atea) iss 396 Mr. W. W. C. Topley. 9. Markwalder and Starling, ‘Journ. Phys.,’ vol. 48, p. 348 (1914). 10. Markwalder and Starling, ‘Journ. Phys.,’ vol. 47, p. 275 (1913). 11. Ketcham, King, and Hooker, ‘Amer. Journ. Phys.,’ vol. 31, p. 64 (1912-18). 12. Fiihner and Starling, ‘Journ. Phys.,’ vol. 47, p. 286 (1913). 13. Henderson, ‘Amer. Journ. Phys.,’ vol. 16, p. 325 (1906). 14. Piper (unpublished). 15. Piper, ‘Archiv f. (Anat. u.) Phys.,’ p. 343 (1912). 16. Wiggers, ‘Amer. Journ. Phys.,’ vol. 33, p. 382 (1914). 17. Evans and Ogawa, ‘ Journ. Phys.,’ vol. 47, p. 446 (1914). 18. Lewis and Mathison, ‘ Heart,’ vol. 2, p. 47 (1910-11). 19. Mathison, ‘ Heart,’ vol. 2, p. 54 (1910-11). 20. Patterson, Piper, and Starling, ‘Journ. Phys.,’ vol. 48, p. 465 (1914). 21. Cannon, ‘ Amer. Journ. Phys.,’ vol. 33, p. 356 (1914). The Influence of Salt-Concentration on Hemolysis. By W. W. C. Toptey, M.B., M.R.C.P., Bacteriologist to Charing Cross Hospital. (Communicated by Dr. F. W. Mott, F.R.S. Received November 25, 1914.) (From the Bacteriological Department of Charing Cross Hospital.) The question of the effect of salt-concentration on the phenomena involved in hemolysis has already received a considerable amount of attention. Nolf originally showed that the presence of certain salts, in definite con- centrations, inhibited hemolysis, and his observations have been repeatedly confirmed. Markl, working with acid sodium phosphate, showed that the introduction of this salt into a hemolytic mixture caused complete inhibition of hemolysis when a certain concentration was reached. He was also able to show that the presence of this salt did uot prevent the combination of the antibody with the red cells. He therefore concluded that its action consisted in so influencing the csmotic relations of the cell membrane that the complement could not be fixed upon it. He found that this action was not specific fer acid sodium phosphate, but could be observed with other salts, notably with hypertonic solutions of sodium chloride itself. These results were confirmed by Ehrlich and Sachs; but these authors, interpreting their findings in the light of the side-chain theory, believe that the action of the increased saline concentration is produced by preventing the chemical union of the amboceptor and complement, and not by any change in the osmotic relations of the cell membrane. The Influence of Salt-Concentration on Hamolysis. 397 It is of interest to note that Muir and Browning have shown that the addition of sodium chloride to fresh serum, in quantities sufficient to inhibit the hemolytic action of complement, also prevents the retention of the complement in the pores of a Berkefeld filter. The following investigations were undertaken in the hope of throwing further light on the part played by salt-concentration in preventing the union of red cells and complement. Attention has been confined throughout to the. action of varying strengths of sodium chloride. Sheep corpuscles, three times washed, have been employed as the test red cells, and fresh guinea-pig serum as complement. The inactivated serum of a rabbit, immunised by repeated intravenous injections against sheep’s red cells, provided a powerful hemolytic antibody. It seemed desirable to commence by repeating Markl’s experiments, amplifying them in certain directions. Without entering into details of the experiments performed, it may be said that Markl’s conclusions were entirely confirmed, and that no action of the increased salt-concentration, other than the prevention of the combination of the complement with the sensitised red cells, could be demonstrated. Markl also showed, in the course of his experiments, that by adding increasing amounts of a hemolytic serum it was possible to produce lysis in the presence of increasing amounts of acid sodium phosphate. He, however employed an active serum containing hemolytic antibody and complement, and made no attempt to investigate these two factors separately. An attempt was therefore made to estimate quantitatively how far the anti-hemolytic ? action of the increased salt-concentration could be counteracted by increasing the concentration of the hemolytic antibody or of the complement. The following experiment is a typical one, and the results obtained in it were repeatedly confirmed. Lxperiment.—tIn this and similar tests, since it was desired to determine the increase in hemolysis resulting from an increase in complement or anti- body content consisting of a definite number of heemolytic doses, the guinea- pig serum was first absorbed for 2 hours at 0° C. with excess of sheep corpuscles; and a preliminary series of tests was put up to determine the minimal hemolytic dose of this absorbed serum. 398 Mr. W. W. C. Topley. M.H.D. of Hemolysin M.H.D. of Complement ... Amount of Test Corpuscles Hemolysis in 1 hour at 37° C. 0:05 ce. of a 1/250 dilution. 0:03 ee. 0:05 c.c. of a 50-per-cent. suspension. oni 1. I, 1. 1. il, 2°5 5. 10. | -, vei ‘ Biv eee sole nM eve Mae Se i peers tan 1. 2°5. 5. 10. 100. | il, | i. i J | i Salt concentra- | tion— 0°8 per cent. | Complete} Complete | Complete | Complete | Complete | Complete | Complete | Complete 1% npr heels in siel ve | nepeamhie «ae 12 | Slight | complete |» » » 04 cobalt! | eel ae 1°4 i None | None care 5 i Slight | Moderate i ol d 16 » D | 9 None | ee s | None Slight | Moderate 1°8 z. | as | es | = None s Pe | None Slight 2:0 9 a | 90 » 99 0 yy | . None 2°2 BS x | * 56 - | Marked 9 | 34 is Pe ait Pah A + Sight eee ae A 2°6 : a 5 = - Trace ¥ i es 2°8 55 5 5 5) | None 2 | 0 % Notr.—The two series are not strictly comparable, because, while it was always possible to add the doses of hremolysin as 005 c.c. of a saline dilution, the amount of complement added necessarily varied. The last series was the only one in which the amount added was sufficiently large to seriously influence the salt concentration, and here the values, instead of varying from 0°8 to 2°8 per cent., ranged approximately from 0°8 to 2°38 per cent. In the series containing 100 M.H.D. of hemolysin, the tubes which showed little or no hemolysis showed marked agglutination of the red cells; but this also decreased, and finally disappeared as the saline concentration increased. Having thus confirmed Markl’s observation that increased salime con- centration inhibits the combination of the complement with the sensitised red cells, and, further, that this inhibitory influence can, to a certain extent, be counterbalanced by increasing the amount of hemolytic antibody present, it was natural to attempt to determine whether, in hypotonic solutions, the amount of antibody necessary to cause lysis would be decreased; and, finally, whether it was possible for complement to combine with red cells without the intervention of hemolytic antibody in mixtures containing little or no ionisable salt. ‘ The Influence of Salt-Concentration on Hamolysis. 399 In considering the results obtained during this part of the investigation, it is necessary to bear in mind the action of hypotonic or salt-free media on complementary sera, especially the phenomenon known as complement- splitting. An observation made by Sachs and Terruuchi in 1907, whilst studying the — inactivation of complement in a salt-free medium, has an important bearing on the problem in hand. These observers found that guinea-pig serum alone was capable of causing a more or less marked degree of lysis of ox corpuscles in a salt-free medium, made isotonic by the addition of 7:8 per cent. saccharose, while the same serum had no lytic action in normal saline solution. They were not, however, prepared to admit that this phenomenon was really comparable to the hemolysis which takes place in a mixture of red cells, hemolytic antibody and complement. They found that, while the complement alone produced hemolysis in saccharose solutions, but not in normal saline, complement acting together with a specific hemolysin always produced more lysis in saline than in saccharose solution ; and, in some cases, lysis was entirely absent when a complete hemolytic system was allowed to act im a saccharose medium, but well marked when the same system inter- acted in normal saline solution. They noted, however, that inactivation for 30 minutes at 55° C. destroyed the power of the complement to produce this abnormal hemolysis, and suggested that it might be due to “a peculiarity of the normal hemolysin.” A large number of experiments were undertaken in order to further investigate this phenomenon. It is only necessary to present here the main results obtained; but ‘it should be noted that different specimens of serum obtained from presumably normal guinea-pigs gave very varying results, while a single given specimen often showed marked changes within 24 to 48 hours though stored on ice. The specimens of serum vary in two ways, in their power of inducing lysis in salt-free media and in their resistance to the anti-complementary action of such media; and these two variations do not run parallel. Thus,a given specimen of serum may be actively lytic in a salt-free medium, but ‘be easily inactivated by this same medium in the absence of red cells, while another specimen may be feebly lytic but suffer only a slight degree of inactivation, a third being actively lytic and markedly resistant to the inactivating action of the salt-free medium. The specimens of guinea-pig serum used in these experiments were absorbed at 0° C. with sheep’s red cells for a period varying in different experiments from one to two hours, in order to remove any trace of normal hemolysin that might be present. It was found that the great majority of sera so treated 400 Mr. W. W. C. Topley. | had a definite lytic action on sheep corpuscles when allowed to act on them for one hour at 37° C. in a salt-free medium, the hemolysis varying with different sera from the merest trace to complete solution. A few sera which, when untreated, produced hemolysis, failed to do so after the preliminary absorption, while a few others produced no lysis before or after treatment. The lysis was never very rapid; thus in no case was complete solution ever observed in less than 30 minutes. Inactivation at 55° C. for 30 minutes completely destroyed the activity of every specimen of serum examined in this way. In a preliminary series of experiments fresh guinea-pig serum was absorbed with sheep corpuscles, and then varying amounts were added to 0:05 c.c. of a 5()-per-cent. suspension of sheep’s red cells contained in 1 c.c. of 7'8-per-cent. saccharose solution. The following results may be taken as typical :— | Experiment.—A series of tubes was put up, each containing 1 ec. of 7'8-per-cent. saccharose solution and 0:05 c.c. of a 50-per-cent. suspension of sheep corpuscles. Varying amounts of the complementary serum were then added and the whole series incubated at 37° C. for one hour. A control tube, containing the same amount of red cells suspended in 7-8-per-cent. saccharose solution to which had been added 0:8 per cent. of sodium chloride,* and the maximum amount of complement employed, was put up; also a similar tube with the addition of 4 M.H.D. of hemolysin. Amount of complement Hemolysis in 1 hour added to salt-free tubes. at 37° C. C.c 0:0 None. 0:05 Moderate. 0:075 Marked. | Ol Slight. 0°125 Trace. O15 None. | 0°175 mR | 0-2 ‘ | 0-225 im 0:25 us if : x Hemolysis in 1 hour Controls. | at 37° C. In 1 ee. of 7°8 per-cent..saccharose contaiming 0°8 per cent. sodium chloride :— (1) 0°05 ¢c.c. red cells +0°25 c.c. complement .................0605 No hemolysis. (2) 0:05 cc. red cells+0°25 c.c. complement+4 M.H.D. | Complete hemolysis. hemolysin * This medium is hereafter referred to as “‘ normal saline-saccharose solution.” q : ; . The Influence of Salt-Concentration on Hamolysis. 401 It is clear from this experiment that only a certain definite amount of complement will cause heemolysis, an increase or decrease in this amount leads to a rapid reduction and final disappearance of the lytic action. These results are wholly unlike those obtained when working with a complete hemolytic system reacting in normal saline solution. It seemed * possible that they might be explained by the fact that in increasing the amount of complementary serum we necessarily increase the amount of electrolytes, and that hence we might pass the limit at which complement could be absorbed by the red cells in the absence of a hemolytic antibody. Experiments carried out to test the validity of this explanation gave very definite results. Experiment—A series of tubes was put up, each of which contained 1 c.c. of 7-8-per-cent. saccharose and 0:03 c.c. of a 50-per-cent. suspension of washed sheep corpuscles. After the addition of the constituents mentioned below, the whole series was incubated for one hour at 37° C. | | Hemolysis in Tube. 1 hour at 37° C. Each tube contained 1 ¢.c. of 7 °8-per-cent. saccharose solution + 0-05 ¢.e. red cells 1 +p OSObY@IOs Contos Baobascceiccsevood sbadon ene cogopuderBosodeconotouoe | Almost complete. 2 +0°10 cpl | agri, eisgbededot boosodonoddsnodenconedsiedcsoncdo csbaddnot None. 3 +0°25 Pp, 1 | | ekera6do8G ¢o0 CG OBoC Be obeSn ober ceoaaater os eecopon cnn 55 4 +005 0 +0 -05 ¢.c. 0 °8-per-cent. saline ............ | Slight trace. 5 +0°05 ft +0°20 ,, 0°8 mee eae eke ’.| None. 6* +005 3 OL Ob mmm nemo lysiniey.eeseseee eee eran er | Complete. 7 +0°05 RS +0°20 ,, Be Seas seats Aes degwee eee 9 8 +0°05 c.c. complement + 0°05 c.c. hemolysin + 0°05 c.c. | None. inactivated complement | * The hemolysin was diluted with normal saline so that 0-05 ¢.c. contained 2 M.H.D. This experiment shows that it is the raised salt concentration which pre- vents the hemolysis in the tubes containing the higher quantities of comple- mentary serum. Thus, 0:05 c.c. of complement causes almost complete lysis while 0:25 cc. causes none. But 0:05 cc. of complement + 0°20 cc. of 0'8-per-cent. saline also produce no lysis. It will be noted also that in this experiment the addition of a hemolytic serum causes increased lysis, but if we also add 0:05 cc. of the same serum which was employed as complement, previously heating it at 55° C. for 30 minutes, no lysis results. Here, again, the effect of increased saline concentration in combating hemolysis is clearly seen. These results help to explain those obtained by Sachs and Terruuchi, which led those observers to deny the identity of the hemolysis produced 402 Mr. W. W. C. Topley. by complement alone in salt-free media with that produced by a specific hemolysin in normal saline. The dilution of a complementary serum with 10 times its volume of 7°8-per-cent. saccharose solution, and the subsequent incubation of the mixture for one hour at 37° C., destroys the activity of the complement, but, if we add 0:075 cc. of complement to 1 cc. of 7:8-per-cent. saccharose solution, containing 0°05 ¢.c. of a 50-per-cent. suspension of sheep corpuscles, and then incubate at 37° marked degree of hemolysis. This can only mean that the complement which has become attached to the red cells is no longer subject to the destructive action of the salt-free medium. Thus, it is clear that the addition of a hemolytic serum will only tend to increase hemolysis, if the increased tendency to combination of red cells and complement, caused by the hemolytic antibody it contains, is greater than the decreased tendency to combination resulting from the increased saline concentration. For unless combination is fairly rapid the destructive action of the markedly hypotonic medium on the complement will come into play. It follows that C. for one hour, we shall obtain a more or less we should expect very powerful hemolytic sera to increase lysis under these conditions, and weak ones to decrease it. In general, the active sera, which are obtained by immunising rabbits against sheep corpuscles, tend to markedly increase hemolysis in saccharose solutions ; but this is not always the case, and in some experiments it was found that more lysis occurred with complement alone than with complement and hemolysin, thus confirming the observation of Sachs and Terruuchi. The addition of serum, or serum diluted with saline, seems always to act more powerfully in inhibiting hemolysis in salt-free media than does the addition of an equal amount of normal saline solution alone; and, although the hemolytic sera here employed were always greatly diluted, it 1s possible that this factor came into play. Keeping these facts in mind we may pass to the consideration of experi- ments in which varying amounts of hemolysin were allowed to act in vary- ine strengths of saline in 7’8-per-cent. saccharose solution, in the presence of 0:05 c.c. of a 50-per-cent. suspension of sheep corpuscles and 0°05 cc. of complementary serum. Experiment.—Each tube contained 1 ¢.c. of 7-8-per-cent. saccharose solution to which had been added varying amounts of sodium chloride. To each tube was added 0-05 c.c. of a 50-per-cent. suspension of sheep corpuscles and 0:05 cc. complement. To each tube was then further added the amount of hemolysin indicated in the left-hand column. This hemolysin was added as 0°05 c.c. of a dilution of the required strength in the saline-saccharose solution. Incubation was carried out for 1 hour at 37°C. 403 A 55 > La} 9S = ~ = S < ES = 3 iS = D i) 5 & 3 7a) > isd) oS S S > SS g ss 49 ce “ce ce “ ce a3 Pye = - 0- OT ajotdurog ee | agotdutog so & “6 a: iy g ayotdutoy § 0.3 oqvaopoyy | ogvxopopy, | poytvyy | ogotdutog | eqe,tdurosy | oyo;dut0p & ajeduioy | ayoTdutop ajo, dutop 0-1 me aovly S poyteut | oje,dutoo wey ajotdutoa | ayazduoo | eja;dutoo ee WSIS WTS SAS PSN S00 A qsowpy SHIELD) qsoupy qsoupy 4soully 02. 0 i ‘ ‘ WSUS « peytont | pay | PCE | poe | paMUIE | poy 6%. 0 789 ss 6 aovay, qysITQ | eqvtepoyy | ozvaopop ayRAopoypy JUST Fy sts EDT pl L-0 | a4B.lopoyL ouo NT ouo NT auoN alo NT QUO N euONN = OUONT auoN auON auo NT 0 — | = i. 2 0:0 6-0 ‘93-0 GE: 0 ‘86-0 PP. 0 *G.0 ‘98.0 9-0 GL. O “8:0 ‘adHw “HOTJNLOS PSOLVIPOIVS 9 UL VPMO[TTo WMIpos Fo soduquod10 gq ““sisATOUee ET UL UISATOULA FT 404 Mr. W. W. C. Topley. From this experiment it is clear that :— (1) When no hemolytic antibody is added, hemolysis only occurs in the tube which contains no salt, except the small quantity present in the 0:05 c.c. of guinea-pig serum employed as complement. (2) When amounts of hemolytic antibody are added below the minimal hemolytic dose (previously determined for hemolysis occurring in normal saline-saccharose solution), the degree of lysis at first increases as the salt concentration decreases. For instance, while 0:1 M.H.D. produces only a. trace of hemolysis in 0°8-per-cent. saline it produces moderate heemolysis in the tubes containing 0°56 per cent. to 0-44 per cent. sodium chloride ; and, while 0°5 M.H.D. produces marked hemolysis in the presence of 0°8 per cent. sodium chloride, it produces complete hemolysis in the 0°5-per-cent. solution. (3) When, however, the lower concentrations of saline are reached, the anti-complementary factor comes into play and the hemolysis again decreases. The hemolysis in almost all cases reaches its minimum in the tube containing 0-2 per cent. sodium chloride, probably for the reason that this amount of salt is sufficient to prevent or delay the union of red cells and complement, unless a considerable amount of hemolytic antibody is present, while the hypotonicity is sufficiently marked to rapidly destroy the activity of the complement. It will be noticed that, although the hemolysis in the great majority of the tubes presents an ordered variation, certain tubes show irregular results ; in the above experiment, for instance, the tube containing 1 M.H.D. in 0°72-per-cent. saline and that containing 10 M.H.D. in 0:2-per-cent. saline. Several similar experiments, however, all yielded results agreeing in all substantial particulars with those indicated above, though in all there were individual tubes which showed irregular hemolysis. One factor which accounts for this is that the cells tend to sink rather rapidly in the saccharose solution, and sometimes undergo agglutination, while in a few cases the corpuscles show marked agglutination in the saccharose solution before any serum is added, though they remain evenly distributed in control saline tubes. Having established the fact that, in passing from markedly hypertonic to markedly hypotonic saline solutions, less and less hemolytic antibody is needed to bring about the union of red cells and complement, and that, when a completely salt-free medium is reached, this union takes place unaided, it seemed necessary to enquire further into the nature of this direct or unaided combination. We know that when dealing with a mixture of red cells, hemolytic antibody, and complement, reacting in a medium of normal saline solution, The Influence of Salt-Concentration on Haemolysis. 405 the red cells and antibody combine at a temperature of 0° C., while the further combination of complement with the red-cell-antibody complex only occurs when the temperature is raised. An attempt was made to determine whether the combination of red cells and complement in a salt-free medium conformed to this rule, as we should expect if the union were really a direct or unaided one. This observation is of considerable importance, since it would obviously be theoretically possible for an antibody to be involved, which was present in the guinea-pig serum, but which only acted in salt-free media, and which therefore was net removed when the undiluted serum was absorbed on ice with sheep corpuscles. The following experiment yields a definite indication as regards this possibility :-— Eaperiment.—A number of tubes were prepared, each containing 1 c¢.c. of a 7-8-per-cent. saccharose solution. To one, red cells and complement were added and incubation was carried out for one hour at 37° C. This tube was controlled by another containing the red cells and complement in 1 c.c. of the normal saline-saccharose solution. To further tubes of saccharose solution were added in some cases red cells and complement, and in others complement alone, and these tubes were kept for one hour at 0° C. Some of these tubes then rapidly centrifugalised, and the supernatant fluid and deposit treated as indicated. The red cells were added as 0°05 ec. of a 50-per-cent. suspension in saccharose solution; 0-075 c.c. of complementary serum was added in every case. | | Hemolysis Incubation for 1 hour at 37° C.— | (1) Complement + red cells in saccharose ...........0.00sseceeceeseeereeeeeee Very marked. | (2) ‘ e Fe +0°8 per cent. NaCl ......... None. 1 hour at 0° C. followed by incubation for 1 hour at 37° C., after the following treatment— | (3) Tube containing red cells + complement shaken ..................000++ Very marked. | (4) Deposit from one (red cells+complement) tube + supernatant es o fluid from another | (5) Supernatant fluid from one (red cells + complement) tube+fresh | _,, 5 | red cells | (6) Deposit from one (red cells+complement) tube+1 c.c. of | None. saccharose solution | (7) Supernatant fluid from a tube containing complement alone+ | Very marked. fresh red cells | (8) A tube containing complement alone shaken + red cells ............... Ps 5 From these results it follows that :— (1) The red cells do not absorb the whole complement at 0° C., since when they are separated by centrifugalisation and, after being suspended in fresh 406 Mr. W. W. C. Topley. saccharose solution, are incubated for one hour at 37° C., they undergo no heemolysis. (2) No antibody capable of combining with red cells at 0° C. is involved, since the supernatant fluid after one hour’s absorption on ice is capable of hemolysing fresh red cells. It is important to note that in this experiment the complement is not appreciably affected by the dilution with salt-free saccharose solution at 0° C., since when a tube containing this mixture was centrifugalised the supernatant fluid was capable of hemolysing added red cells on further incubation at 37 C, A somewhat different result was, however, often obtained. While the deposit from a tube containing red cells and complement in saccharose solution, which had been kept for one hour at 0° C., never showed any trace of hemolysis when suspended in fresh saccharose solution and incubated for a second hour at 37° C., it frequently happened that the supernatant fluid tailed to produce more than a trace of hemolysis when incubated at 37° C. with fresh corpuscles. In such cases it was always found that, if the tube containing complement alone in saccharose solution was similarly kept for one hour at 0° C., then centrifugalised and the supernatant fluid added to fresh red cells, only a trace of hemolysis occurred upon further incubation at 37°C. It is therefore clear that the failure of lysis is due to the action of the salt-free medium on the complementary serum, and not to any fixation of the complement by the red cells in these cases. At the same time, it was found that if the supernatant fluid from one of these tubes containing complement only was added to the deposit from a tube which, during the preliminary treatment at 0° C., had contained both red cells and complement, then a degree of lysis resulted during the second hour’s incubation at 37° C., which corresponded with the hemolysis which occurred when a tube containing red cells and complement in saccharose was incubated at 37° C., or kept on ice for one hour and then shaken and incubated for another hour at 37° C. This clearly indicates that some part of the serum has been precipitated and is present with the red cells in the deposit. In other words a certain degree of complement-splitting has taken place. From other experiments performed during this part of the investigation this does not seem to be the whole explanation. It was found in many cases that, when a tube containing complement alone in saccharose solution was kept on ice for one hour and then well shaken, the contents failed to hemolyse fresh red cells, though still producing lysis of the deposit from an iced saccharose tube containing red cells and complement. It seems clear, The Influence of Salt-Concentration on Hamolysis. 407 therefore, that the presence of the red cells prevents the action of the salt- free medium on the complementary serum from progressing beyond a certain point. Experiments carried out to test the hypothesis that this might be due to a combination of the mid-piece with the red cells gave no definite results, for reasons which need not be entered into here. The fact remains that those complementary sera which are unaffected by simple dilution with salt-free saccharose solution at 0° C. show no evidence of any absorption of the complement by red cells at this temperature. We are, therefore, justified in saying that complement, combining directly with red cells in a salt-free medium, behaves in the same manner as complement combining with red cells in a medium of normal saline solution under the influence of a hemolytic antibody, in that no combination occurs at a temperature of 0° C. In the experiments described above, and in the discussion of the results obtained, the terms “complement” and “hemolytic antibody” have been employed in their usually accepted sense. It is obvious, however, that the sera employed contain many other substances, and the recent cross- absorption experiments of Thiele and Embleton show that a hard and fast division into complement and antibody is not permissible. It is, therefore, possible that not the whole complement, but only some specialised part of it can combine, unaided, with red cells in the absence of electrolytes; but this consideration does not affect the main thesis that a combination, that is impossible in a salt-containing solution without the addition of a special antibody, can occur in its absence in a salt-free medium. Conclusions. In the case of the hemolysis of sheep corpuscles by guinea-pig comple- ment, it is found that :— 1. The presence of an excess of an electrolyte (sodium chloride) above the normal limit in a hemolytic mixture prevents the combination of the com- plement with the red-cell-antibody complex. 2. If the concentration of the antibody be markedly increased, it is possible, up to a certain point, to counteract the effect of the increased salt concen- tration. 3. If the salt concentration be decreased, a decreasing concentration of antibody serves to produce the union of red cells and complement. 4. In an almost completely salt-free medium the combination occurs in the complete absence of antibody. 408 Dr. A. Compton. The Influence of the Hydrogen It only remains for me to record my great indebtedness to Dr. 8. G. Platts for the assistance which he has rendered me throughout a considerable part of this investigation. REFERENCES. Ehrlich and Sachs, “‘ Ueber den Mechanismus der Amboceptorenwirkung,” ‘ Berl. Klin. Wochschr.,’ vol. 39, p. 492 (1902). é Markl, “Ueber Hemmung der Himolyse durch Salze,” ‘ Zeitschr. f. Hygiene,’ vol. 39, p- 87 (1902). Muir and Browning, “On the Filtration of Serum Complement,” ‘ Jour. Path. and Bact.,’ vol. 13, p. 232 (1909). Nolf, “ Le Mécanisme de la Globulolyse,” ‘ Annales Inst. Pasteur,’ vol. 14, p. 656 (1900). Sachs and Terruuchi, “Die Inaktivierung der Komplement im Salzfreien Medium,” ‘Berl. Klin. Wochschr.,’ vol. 44, p. 467 (1907). Thiele and Embleton, ‘The Evolution of the Antibody,” ‘ Zeitschr. f. Immunitats. u. Exper. Therap.,’ vol. 20, p. 1. (1918). The Influence of the Hydrogen Concentration upon the Optimum Temperature of a Ferment. By ArtHur Compton, M.B., D.Sc. (N.U.I.), Imperial Cancer Research Fund. (Communicated by Prof. W. Bulloch, F.R.S. Received December 17, 1914.) The present investigation is an outcome of previous work,* resulting in the discovery that the optimum temperature of any ferment, or ferment function, occurring In a given enzymic preparation, is independent of the. concentration both of the substrate and of the enzyme, the duration of the action being constant. To follow up this observation it was felt desirable to investigate in a similar way the influence, if any, of the reaction, that is of the hydrogen ion concentration of the medium, on the optimum temperature of enzyme action: the more so because enzymes, as regards their activity, are known to be extremely sensitive to this factor—some requiring an acid, others a neutral, and others again an alkaline medium in which to act. The question, moreover, is of special interest on account of the fact that Sorensen, in his classical researches on the véle of the ionic concentration of the medium in activating enzymes, alludes to it, and predicts in regard to it that, when investigated, the optimum temperature will no doubt be found to vary with the hydrogen ion concentration of the medium.t That * Arthur Compton, ‘ Roy. Soc. Proc.,’ B, vol. 87, p. 245 (1914) ; B, vol. 88, p. 258 (1914). + 8. P. L. Sorensen, ‘Comptes Rendus des Travaux du Laboratoire de Carlsberg,’ vol. 8, p. 148 (1909). Concentration upon the Optimum Temperature of a Ferment. 409 opinion, being of a speculative nature, was not deemed a sufficient answer to the question; its experimental investigation therefore became the more needed. The enzyme chosen was the maltase of Aspergillus oryzw, the same preparation being used as had already been studied in a former communica- tion,* where its optimum temperature, in an action of 16 hours’ duration, and H* concentration that of the preparation simply dissolved in redistilled water, is shown to be +47°. Two stages occur in the investigation: (1) A determination of the optimum reaction curve of the enzyme in an action of 16 hours’ duration at +47°, for chosen dilutions of substrate and of enzyme, in presence of progressively increasing quantities of acid and of alkali added to the reaction mixture ; (2) separate determinations of the optimum temperature of the ferment under the same conditions of substrate concentration, of enzyme concentration, and of duration of the experiment, for different hydrogen ion concentrations of the medium, corresponding to various points on the above optimum reaction curve. : The substrate concentration chosen was M/20, or 18x107* grm. of hydrated maltose per cubic centimetre of the reaction mixture ; and the enzyme concentration was 6 x 10~* germ. of the enzyme preparation in powder per cubic centimetre of the reaction mixture. For the determination of the optimum reaction curve the practical details were as follows :—First, a solution of the enzyme was prepared by dissolving 30 mgrm. of the preparation in 10 cm.? of redistilled water. A clear pale amber-coloured solution resulted, which, after standing from a half to one hour at the ordinary temperature, was introduced in portions of 1 cm.? into a series of eight clean test-tubes (Fischer’s extra resistance glass) already containing 90 mgrm. of hydrated maltose and either 4 cm.’ of pure water or a solution containing a known quantity of acid or alkali. The tubes, after closing with clean sterile corks, were plunged into a water bath regulated at +47°. After 16 hours’ incubation they were withdrawn, the corks removed, and each rapidly washed with 1 em.’ of water—the washings being carefully added to the contents of the corresponding tube. Next, the tubes were heated for: seven minutes in boiling water to stop the enzyme action, after which they were cooled and the contents diluted to 50 cm. The proportion of maltose hydrolysed was then determined, by Bertrand’s method, on 20 cm.’ of the diluted mixture. The numbers obtained are set out in Table I. * Ibid. + ‘Bull. Soc. Chim.,’ (3), vol. 35, p. 1285 (1906). 410 Dr. A. Compton. The Influence of the Hydrogen Table I. Quantities of acid and alkali added per 5 em.’ of the reaction mixture. Makeue nydhelyset. Per cent. 0-25 em.? M/100 HS, .....0.cecccceeeeceeees 28 2 0-15 ,, if ale URL Ona Tania leg 63°5 010 ,, cet eta 00 ve RCS ae 75 1 0-05 ,, Kita Pitnotid et Meee Salt, BS Za hcp 76°8 0°00 ,, (natural reaction ; control) ...... 84°3 OWS 55. IWYMCD INB(COs sonsasscosonocastenos 142 O20 FPA aCe opascccodachecorts 3 °2 When the percentage of maltose hydrolysed’ is plotted against the quantities of reagents added, the curve indicated in fig. 1 is obtained. a(%)—> ! @ (6) Fig. 1. Maltose hydrolyse S 100 H*SO* Mjioo Na*CO o-6 Od 9) O1 o2 Oo4, 03.02 Oi | CNP per 5CMS reaction mixture. > 020 O17 ok O10 O07 _ o0% fe) 005 O07 CM® per MGRM. cf engyine approsc. = -3: =3: -3: -47 =: ~62 F2 7. 162) do?) Mies tol io 40°10 5 10° noe _ H* concentration of medium. This curve shows that under the conditions of the experiment the addi- tion of the merest trace of alkali is detrimental to the enzyme action ; while on the contrary the addition of acid increases the efficiency of the enzyme, until a certain point is reached, corresponding to the presence of 0:07 em.? M/100 H2SO, in the 5 cm.* reaction mixture, beyond which the addition of more acid is in turn detrimental to the action. In other words, the action passes by a maximum situated in the acid region. Concentration upon the Optimum Temperature of a Ferment. 411 Measurement of the hydrogen ion concentration of the medium, resulting from the addition of various quantities of acid and of alkali per 5 cm.? of the reaction mixture, containing 3 mgrm. of dissolved enzyme, were made by the colorimetric method of Sérensen (/oc. cit.). The results are contained in Table II, and for convenience of description, in what follows, they have been reproduced on the base line of fig. 1, underneath the respective quantities of M/100 H2SO, and M/100 NazCO3 which give rise to them. Table II. ] Quantities of acid and of alkali added per 3 mgrm. of | : 2 ; A : Corresponding H+ concentrations. | enzyme contained in 5 cm.’ of the reaction mixture. | E 8 | 060 M/100 H.SO, ...... ) Os citrate, methyl orange. 4 = 31018 ae La Baa tea | at concentrations greater than ae a® 2 2p 0°20 - pis | dhe pian ot Hs Te 10-47; phosphate, neutral red. | 0°15 PEL ee J : Ores 56 3 0:07 BR Wa Mets 3 H+ concentrations equal to the | 10-%?; ss as optimum of fig. 1. | 0 04 Be clade Ht : NO=88 5 % » | : concentrations less than the —n5 0-00 (natural reaction) optimum of fig. 1. 1077 2; i J 0:05 M/100 Na;CO,...... © 1077 i *, The second stage of the enquiry, which consists in determining the optimum temperature of the enzyme for a series of hydrogen ion concentrations corresponding to different points on the optimum reaction curve, figured in fig. 1, was next undertaken. Nine different H* concentrations of the medium TENS HNOS SuNCeCle WO OPS) WOH Oa, IO eH. IO) Se aver and: [Om >, The practical details of the first determination in the series may be given as an example, to show how these optimum temperature determinations were carried out. A solution of the enzyme was prepared containing 3 merm. of the preparation per cubic centimetre and, after standing for a half to one hour, was introduced in portions of 1 cm. into eight clean test-tubes, already containing 90 mgrm. of maltose, 0°6 cm.’ of M/100 H.SO, and 3-4 em.’ of redistilled water. Such a mixture, according to Table II (or fig. 1), corresponds to a H* concentration of 10~*°. The tubes were closed with sterile corks, plunged into water thermostats at known temperatures, and incubated for 16 hours, when the enzyme action was stopped and the quantity of maltose hydrolysed in each tube determined as before. The numbers found are set out in Table III, together with the numbers obtained for the other members of the series. VOL. LXXXVIII.—B. DK A12 Dr. A. Compton. The Influence of the Hydragen Table LI. Maltose hydrolysed per cent. for the following H* concentrations Temperatures at SP Ae cd the beginning and end of each experiment. | 49~3.0 | 49-32, | 10-84, | 10-*7,| 10-56, | 10-62, Ply fe) 332 7s ae see} 21°1 at | aN (op) aS a DAD AD Aad or Xe) TL on ee (0 ive) | | On plotting the percentage of maltose hydrolysed against the mean temperature of the experiment, these numbers give a series of optimum temperature curves from which the optimum temperature corresponding to each particular H* concentration of the medium may be read. The curves are summarised for purposes of description in two figures (figs. 2 and 3). In fig. 2 are collected the curves: corresponding to H* concentrations greater than the optimum reaction of fig. 1; while in fig. 3 are collected the curves corresponding to H* concentrations equal to, and less than, the optimum of fig. 1. Consider fig. 2, Here we have a series of curves of varying altitudes, which is what one would expect from the results already recorded in fig. 1 ; as the H+ concentration of the medium is increased beyond the optimum value the enzyme is gradually rendered less efficient. And, in this respect, fig. 2 indicates, further, that the diminution in the activity is true for Concentration upon the Optimum Temperature of a Ferment. 413 practically all temperatures, although varying in amount from temperature to temperature, subject to the influence—to be explained presently—of the hydrogen ion concentration of the medium. As to the influence, if any, of the H* concentration of the medium on the optimum temperature of the enzyme, fig. 2 clearly shows that for each H* concentration there exists a perfectly definite optimum temperature ; also, the optimum temperature is seen to fall progressively as the H* concentra- tion is increased beyond the optimum reaction of fig. 1. In fact, the locus Fig. 2. g Maltose hydrolysed (Z) $ g 1O 20 30 40 50 60 _ Temperature Hconcentrations varying between 16° 107° of the maxima of these curves is a straight line, the mathematical equation of which is approximately y = 656ea—200-44, The significance of this straight line is that it shows that the fall which occurs in the optimum temperature of the ferment as the H* concentration is increased—beyond its optimum value—is proportional to the fall in activity (or disablement) which the ferment undergoes at the physical optimum point. Fig. 2 further shows that the temperature of destruction of the enzyme also depends on the hydrogen ion concentration of the medium; the greater 2K 2 A14 Dr. A. Compton. The Influence of the Hydrogen the acidity the less heat is the enzyme able to support before being entirely disabled. Herein lies the explanation of the existence of the two lower curves of fig. 2, in spite of the indications of fig. 1, which shows that for the H* concentrations of 10-*? and 10-*° no activity on the part of the enzyme seems possible. The reason is, that what was the optimum temperature for the natural reaction, +47°—at which temperature the results set forth in fig. 1 were determined—is no longer the optimum temperature for the H* concentrations giving rise to the curves in question, but is instead a temperature of destruction. ) 3 -eerep tet thee ere EE EEE EF EFLY (eo) (o) ydrolysed (' ee Maltose h g 30 10 20 BO 40 50 60° Temperature ——> Htconcentrations varying between 1097610 That the optimum temperature of the ferment diminishes as the H* concentration is increased beyond that of the chemical optimum may be stated otherwise thus: As the H* concentration is diminished from ~ extreme values to values bordering on that of the chemical optimum, the optimum temperature steadily increases. An interesting question now arises. What would be the effect on the optimum temperature of diminishing the H* concentration beyond that of the chemical optimum ? Would the optimum temperature under, these circumstances continue to increase, in view of the fact that further diminution of the H* con- Concentration upon the Optimum Temperature of a Ferment. 415 centration of the medium must, in accordance with fig. 1, be sisienolet by disablement of the enzyme? . Fig. 3 answers these questions. Fig. 3, as expected, also shows a series of curves of varying altitudes ; and, as before, a perfectly definite optimum temperature is seen to characterise each hydrogen ion concentration studied. But here an unlooked-for result is discerned. For Ht concentrations situated between the optimum and the natural reaction the optimum temperature rises still higher, to fall again as the natural reaction is overreached ; the rise in the optimum temperature passes by an optimum value. This is evident from the locus curve drawn through the maxima of the several optimum temperature curves figured in fig. 3; and it is still better seen in fig. 4, where a curve is plotted with the various optimum temperatures recorded in figs. 2 and 3 as ordinates, and the corresponding amounts of acid and alkali present in the reaction mixture as abscisse. Fig. 4. j Vv aT = 8 ey oy 5 = 3 kul: Moo H*S0* : | M/too Na2CO3 <_< —_—_—__> os os O3 oO2 oO o1 O2 oa CM? per 5CM> aera mixture An optimum temperature of +49° thus seems possible in a reaction mixture containing 0:032 cm.* M/100 H.2SO, But it might be supposed that this is the optimum reaction of the ferment for that temperature, since, in accordance with the work of O’Sullivan and Tompson* on the enzyme sucrase, the optimum reaction of that ferment is known to depend on the temperature serving for its determination? This question it is proposed to investigate. To illustrate in a striking way the essential fact established by the investigation, that the optimum temperature of the enzyme depends largely on the hydrogen ion concentration of the medium, it is only necessary to reproduce side by side on the same diagram the optimum temperature curve from fig. 3 corresponding to the natural reaction and that from fig. 2 * “Chem. Soc. Journ.,’ vol. 57, p. 859 (1890). 416 Dr. A. Compton. The Influence of the Hydrogen corresponding to the addition to the reaction mixture of 0:2 em.3 M/100 H2SO,4 per milligramme of enzyme. This is done in fig. 5. Fig. 5. drolysed (72) ee le) le) ly iN) le) S ) Maltose h Temperature ——— Fig. 5 shows at a glance that by the simple ‘process of increasing the hydrogen ion concentration of the medium in which the enzyme acts from 1077? to 107% it is possible to change the optimum temperature of the ferment from +47° to +35°5°,—z.e. through a range of 11°5°. This result alone shows how important it is, when stating the optimum temperature of an enzyme, to point out at the same time the H* concentration of the medium serving to determine it. Thus, the optimum temperature of the maltase in question, in an action of 16 hours’ duration, may be expressed as [35°5°] 10°-*" or [47°] 10°77". Fig. 5 is further interesting from another aspect. The parallelism or similarity of the two curves is very striking; they are almost superposable. Were they exactly so, and obviously that is only a question of sufficient experimental patience, it would mean that the activity of the enzyme is the same at corresponding temperatures over two equal although different ranges of temperature,—the one curve being in that case simply a translation in the plane of the paper of the other. It is proposed to term this, the phenomenon of corresponding states. From a consideration of the locus lines of figs. 2 and 3, itis evident that for H* concentrations such that the maxima of the result- ing optimum temperature curves are situated at the same level on these lines, the enzyme should for the one and the other H* concentration be in “ corre- sponding states.” The main conclusion of the present paper, in the insight which it affords into the influence of the H* concentration of the medium on the physical optimum of enzyme action, cannot perhaps be summarised to more advantage at present than by placing it in its appropriate place in a differential table briefly setting out our present knowledge of the relation of the physical and the chemical optima of enzyme action to the experimental conditions involved in determining them. Concentration upon the Optimum Temperature of a Fermeit. OrrimuM TEMPERATURE. (Physical Optimum.) i. Is independent of the concentration of the enzyme (Compton). Maltase, salicinase. ii. Is independent of the concentration of the substrate (Compton). Maltase, salicinase. iii. Is dependent on the H.* concentration of the medium (Compton). Maltase. iv. Is dependent on the duration of the experiment serving to determine it (Bertrand and Compton).* Amygdalinase, amygdalase. 417 Optimum H* ConcENTRATION. (Chemical Optimum.) i. Ditto (Sérensen). Sucrase. me i. Is dependent on the concentration of the substrate (Van Slyke Zacharias).t and Urease. ae ili. Is dependent on the temperature of the experiment (O'Sullivan and Tompson). Sucrase. iv. Ditto (Sédrensen). Catalase, pepsin, sucrase. <4 . Is dependent on the age of the enzymic preparation (Bertrand and Comp- ton). Amygdalinase, amygdalase. * “Comptes Rendus,’ vol. 152, p. 1518 (1911). + ‘Journ. Biol. Chem.,’ vol. 19, p. 205 (1914). t ‘Comptes Rendus,’ vol. 159, p. 434 (1914). 418 The Infe-Cycle of Cladocera, with Remarks on the Physiology of Growth and Reproduction in Crustacea. By Georrrey Suiru, M.A., Fellow of New College, Oxford. (Communicated by E. 8S. Goodrich, F.R.S. Received December 18, 1914.) 1. Experiments on Daphnia pulex. In a paper on the life-cycle of Moa rectirostiis, published in 1913 (5), it was shown by the late Mr. G. H. Grosvenor and myself that it was possible to inhibit entirely the production of the sexual forms by isolating the parthenogenetic parents soon after birth, and keeping them at a constant high temperature of 25-30° C. It was proved that for a succession of eight generations the isolated parents at this temperature gave no males or ephippial females, while parents of the same generations kept crowded at a temperature of about 14° C. or 5° C. gave about 50 per cent. males. We were unable to determine how the effect of isolation and crowding of the parthenogenetic parents influenced the production of the sexual forms, but two alternative suggestions were made, either that in the crowded glasses the animals were unable to obtain sufficient nutriment and were partially starved, or else that some excretory matter accumulated in the crowded glasses which influenced the production of males and sexual females. : In order to confirm the above results and to throw some light on the processes involved, breeding experiments have been carried on for some time with another species of Cladocera, the common Daphnia pulex. Mr. Robert Gurney very kindly gave me some dried mud from a pond which was known to contain the resting eges of these animals, and, on placing the mud in a bowl of water, after some weeks some young Daphnia hatched out. One of these was kept until it had produced voung, and the offspring of these young ones were used to start the first experimental generation. & D. pulez does not flourish on the food used for Moina, viz., manure infusion, but I had previously found that they could be cultivated with great ease if some green Alga, such as Protococcus, is added to the water in which they are kept. In order to have a constant supply of the Alga, stock cultures were made in a nutrient medium of inorganic salts to which a small amount of organic material was added. The best medium for growing the Proto- coccus was found to be a certain dilution of the mixture recommended by Miquel for growing Diatoms, which Mr. H. G. Thornton and myself have used for cultivating Euglena(8). By adding a pipette-full of the green growth to each glass in which the experimental animals are kept, it is possible The Life-Cycle of Cladocera. 419 to ensure that there is always an excess of nourishment, because the culture consists only of the Alga and not of a mixed assemblage of bacteria, some of which may be useless as nourishment. The scheme of the experiments is as follows :—In each generation a certain — number of the individuals are isolated soon after birth in separate glasses ; some of these are placed in an incubator at 27° C., others are stood in a water tank with circulating water at 10-17° C. Other individuals are kept crowded together in the same glass in numbers of 10-15, and of these crowded glasses some are again placed in the incubator and others in the circulating water at 10-17° C. Thus in each generation we have individuals subjected to four different conditions:—(1) Isolated at 27° C.; (2) crowded at 27° C.; (3) isolated at 10-17° C.; (4) crowded at 10-17° C. All are supplied with excess of Protococcus. ; In Table I is given the result of breeding under these various conditions for eleven successive generations. This Table does not give the numbers of ephippial females which appeared among the parents, but it may be stated that ephippial or sexual females only appeared among the parents kept crowded at 10—-17° C., which also gave a high percentage of males. It will be noted that in this Table, besides the numbers of male and female offspring produced in each generation, a column is devoted to the number of parents, whether isolated or crowded, which were used for breeding. This factor, viz., the number of parents used, is one that must not be lost. sight of, since, in order to prove that the production of males and sexual females is not simply a question of chanee, it is clearly necessary to use a sufficient number of parents in each generation and under each condition to ensure that the effects of chance are ruled out. In as many cases as possible four broods were taken from each female. It was not found that there was any tendency for later broods to produce more sexual forms than early broods. By consulting Table I it will be seen that neither in the isolated nor crowded individuals at 27° C. did any sexual forms appear throughout the eleven generations. Adding the totals of the isolated and crowded at 27° C. together we have, 90 + 150 = 240 parents gave 1188 + 643 = [831 partheno- genetic females and no males. Eighty-six individuals isolated at 10-17° C. gave 1860 parthenogenetic females and 56 males, or about 3 per cent. males, while 420 individuals crowded at 10-17° C. gave 3564 parthenogenetic females and 256 males, or about 7 per cent. males. These results are in agreement with our previous experiments with Moina, showing that the effect of isolation and high temperature is to suppress the production of the sexual forms (5). 420 Mr. G. Smith. Table I. Number of offspring. Number of offspring. Generation Number of ' Number of % parents. parents. ‘ Female. Male. Female. Male. Isolated at 27° C. Crowded at 27° C. 1 | 5 35 (0) 20 47 10) 2 | 13 43 (0) 20 65 0 3 | 13 80 0 — — 10) 4 | 4 14 (0) 10 20 (0) 5 1 22 @) 16 28 0 6 | 11 108 ) 20 yall 0 " 12 331 0) 16 | 147 0 | 8 9 253 (0) 24 105 (0) | 9 7 120 0 16 20 0 10 6 75 0 | 8 40 (0) fat 9 112 0) | os = = | WOKS. —scocannod 90 1188 (0) | 150 643 (0) Generation. Isolated at 10-17° C. Crowded at 10-17° C. , | 1 5 | 129 ) 20 PaO 7 | | 2 8 | 204 0) 33 1035 100 3 10 319 (0) 70 482 15 4 | 10 | 157 0 27 242 8 5 | 9 168 0 30 190 0 | 6 7 144 0) 65 415 0 | a 3 49 0 75 452 36 | 8 10 250 (0) 30 470 65 9 | 11 | 2538 56 15 40 0 10 | 6 | 99 Opal 80 87 25 11 7 88 0 25 60 0 Total Siaeeeeeeee 86 | 1860 56 420 3564 256 af Since it was found that the individuals crowded at 27° C. produced few young and did not flourish, the experiment was subsequently continued in a rather different way, only two kinds of conditions being employed, viz., isola- tion at 27° C. and crowding at 10-17° C. In this second .experiment as nearly as possible equal numbers of parents in each generation were used in the isolated and crowded condition. Also a careful observation was kept to see how many of the parents used became ephippial or sexual. This experi- ment was made some time after the first with individuals that had been propagating by parthenogenesis, so that the first lot of parents used were about the 35th generation from the beginning of the cycle, 2.e., the original winter egg from which the first individual emerged. The Life-Cycle of Cladocera. 421 The result which is given in Table II shows that during eight successive generations 117 isolated parents at 27° C. produced 2564 parthenogenetic females, no males, and in no case became ephippial, while 129 crowded parents became ephippial in 17 cases and produced 1147 parthenogenetic females and 26 males. Since nearly equal numbers of parents were used and more offspring were produced from the isolated parents than from the crowded, it is impossible to ascribe the production of sexual forms by the crowded individuals, and their entire absence in the case of the isolated parents, to chance. If we add together the results for the isolated at 27°C. and for the crowded at 10-17° C. in the two experiments given in Tables I and II, we see that in nineteen generations 207 isolated parents at 27° C. gave 3752 parthenogenetic females and no males, while 549 crowded parents at 10-17° C. gave 4711 parthenogenetic females and 282 males, or about 6 per cent. males. It is also to be observed that, while no ephippial females appeared among the isolated parents at 27° C., about 10 per cent. of the crowded parents at 10-17° C. became ephippial. Table II. Isolated at 27° C. Crowded at 10-17° C. Genera- | Number Number tion. Number | Parents of offspring. Number | Parents of offspring. of | becoming 2 of becoming ——_—__ —__--— parents. ephippial. parents. | ephippial. | Female. | Male. Female. | Male. | 35 8 @) 132 (0) 10 2 63 104 36 16 0 334 0) 20 im 117 0) 37 19 0) 414 10) 20 0) 222 (@) 38 16 0 350 0) 10 1 111 0 39 13 0 397 0) 20 4, 116 @ | 40 . 9 (0) 153 (0) 10 (0) 264 On| 41 16 0) 644 0) 10 2 59 i 42 20 0) 140 10) 29 a | 195 15 Totals...... 117 (0) 2564 0) 129 17 1147 26 The general result of the above records is to show that in D. pulex, as in M. rectirostris, it is possible to inhibit entirely the appearance ot males and sexual females by isolating the parents soon after birth and keep- ing them at a temperature of 27° C. But if we look into the numbers given for each generation in Tables I and II, we find that the converse of the above statement does not hold good, 7e., it is not the case that crowding at 10-17° C. always results in the production of the sexual forms. ‘Thus, to 422 Mr. G. Smith. take some instances in Generation 6, 65 crowded parents gave 415 partheno- genetic young and no males; in Generation 37, 20 crowded parents gave 222 parthenogenetic young and no sexual forms. It must be concluded from this that there is some factor involved in the production of the sexual forms other than external conditions, viz., an internal factor. That this internal factor’ isa regular rhythmical cycle, such as Weismann originally suggested (1), which runs on without any regard to external conditions, is obviously not true, but there is this very important element of truth in Weismann’s view, namely, that a species such as D. pulex never produces as many sexual forms per cent. as a species like I. rectirostris ; and, as far as we know, no alteration of the external conditions would make it do so. The facts suggest that for each species of Cladoceran there is a maxima! limit to the numbers in which sexual forms may be produced, and that this number cannot be readily increased: but it can be decreased, or entirely abolished, by external conditions such as isola- tion and high temperature combined with abundant nutrition. If we suppose with Woltereck (3) that the production of parthenogenetic and sexual forms is due to the presence of two substances, a parthenogenetic substance and a sexual substance, then we should say that the relative amounts of these substances are fairly rigidly fixed for each species, and that, whereas the amount of the sexual substance cannot be easily increased, its operation can be indefinitely suspended by the action of external conditions. By this inter- pretation of the facts it is possible to retain the really important part of Weismann’s theory, that the proportions in which the sexual forms are produced in each species is fixed in its upper limit in accordance with the adaptive necessities of the species, though we must maintain that these proportions can be altered by the suppression of the sexual forms through external conditions. W. E. Agar (7) in a recent paper, after summarising the results of previous workers, comes to the conclusion that there is no such thing as an internal rhythm in Cladocera, and that the production of sexual or non-sexual forms is entirely controlled by the environment. J am in agreement with Agar in thinking that there is no hereditarily fixed rhythm, or that the production of the sexual forms is rigidly fixed on to particular generations or particular broods of these generations, but I find it impossible to believe that it is purely due to environment that a species like M. rectirostris, under any conditions, produces a far greater percentage of sexual forms than aspecies like D. pulea. In other words, there is an internal factor concerned in the production of the sexual forms; this factor varies in different species of Cladocera ; its operation can be entirely suspended by external conditions so that no sexual forms are produced ; but there is no experimental justification for the view The Life-Cycle of Cladocera. 423 that the production of sexual forms can be provoked at will in any or every generation of a particular species by alterations in the environment. It is important to consider whether the factor of crowding can ever operate in a state of nature in the same way as under cultural conditions. There can be no doubt that the way this factor exerts its effect is through the presence of some excretory material in minute quantities, because in our cultures of Daphnia, which were fed on nothing but Protococcus, it was possible to ensure that there was in all cases an excess of food, so that the crowding could never cause a shortage of food. Since the animals were cultivated in glasses containing about 100 c.c. water, and the presence of 10 individuals constituted the normal crowded condition, it is clear that the reaction must be a very delicate one, due to the presence of extremely minute proportions of the substance in question. Now, in a state of nature, the small pools inhabited by many “ polycyclic” species of Cladocera, ¢.g., Moina, are often far more intensely crowded with individuals than under our cultural conditions. But, quite apart from small pools, it is frequently to be observed that large ponds are often so thickly populated with species like D. magna and pulex as to be coloured blood red, and I have met with cases where farmers have been afraid to water their horses at a pond on account of the extraordinary colour of the water. I have also observed that even in lakes, certain areas of water may be intensely crowded with some species of Cladocera, and it appears to me probable that the factor of crowding may play its part in the production of the sexual forms even in the largest bodies of water. The interesting report of Dr. Viktor Langhans on the Cladocera of the Hirschberg Lake in North Bohemia (4) shows that the various species of Cladocera inhabit for the most part quite localised areas of the lake, and, moreover, that the appearance of the sexual forms usually either coincides with or follows closely after the greatest activity in parthenogenetic reproduction, when crowding would be at its height. 2. The Storage of Fat and Glycogen in its Relation to Growth and Reproduction in Cladocera. In the course of the breeding experiments described above, a contrast was noticeable between the individuals isolated at 27° C. and those kept crowded at a lower temperature. It was observed, even on inspection with the naked eye, that the young or fully grown individuals isolated at 27° C. were always of a pale, translucent green colour, while those crowded at the lower tempera- ture were generally bright reddish orange, or, at any rate, showed a good deal of this colour. On examining the two kinds of individuals under the 424 Mr. G. Smith. microscope it was found that the reddish orange individuals owed their colour to the abundance of coloured fat globules present round the gut and ovary and at the bases of the Jimbs, while the pale green individuals were either entirely devoid of any fat or else possessed a few globules in the neighbour- hood of the ovary. It was shown in a previous paper (6) that by placing living specimens of Cladocera, such as Moina or Daphnia, into a vessel of water in which a small quantity of neutral red is dissolved, it was possible to distinguish after a few hours certain bodies which took up the stain with great avidity. These bodies which stain intensely intra vitam with neutral red are distributed in three chief situations—(1) as very small granules in the polygonal cells of the chitogenous ectoderm (fig. 1) (in the case of ephippial females the chitogenous cells of the ephippium, which is formed of very thick chitin, contain much larger masses of glycogen, see fig. 2); Fic. 1. Fie. 1.—Chitinous areas on carapace, representing chitinogenous ectoderm cells, with small granules of glycogen stained with neutral red intra vitam. Fie. 2.—Chitinous areas on carapace in region of formation of ephippium, showing large lumps of glycogen, stained with neutral red. Fia. 4. Fic. 3.—A group of cells from the gut, showing small glycogen granules occupying each cell, stained with neutral red. Fic. 4.—Four large subcutaneous glycogen cells from base of limb, showing irregular masses of glycogen in periphery of cell, stained with neutral red. The Life-Cycle of Cladocera. A25 (2) as small granules in the cells of the gut (fig. 3); and (3) as much larger, irregular-shaped masses in the connective tissue at the sides of the gut and at the bases of the limbs (figs. 4 and 5). There can be no doubt that these bodies are reserve material of the nature of glycogen, because the areas in which they occur are the same as those in which glycogen is known to occur in the higher Crustacea, and they exhibit the same appearance and staining reactions as the glycogen deposits in higher Crustacea. The fact that the so-called glycogen deposits of Crustacea stain so intensely intra vitam, and also after fixation with neutral red, suggests that they are not pure glycogen, or, at any rate, not identical with the glycogen found in the liver of warm-blooded animals, because neutral red does not show any particular affinity for these latter deposits. That they are largely composed of glycogen is shown, however, by their giving the iodine reaction both microchemically and, in the case of the higher Crustacea, after extraction with hot water in a test-tube reaction. Fie. 5.—Parthenogenetic female, isolated at 27° C., showing reserve material present almost entirely as subcutaneous glycogen, with a few fat globules in neighbour- hood of ovary. It seems that the glycogen deposits in Crustacea consist of glycogen plus some nitrogenous material, probably a proteid derivative, which is responsible for the special affinity for neutral red. Leaving the exact chemical nature of these amylaceous deposits aside, it is to be observed that the pale green translucent individuals of D. pulex which have been kept isolated at 27° C. exhibit practically all their reserve substance in the form of this 426 Mr. G. Smith. glycogen material (fig. 5), while the reddish-orange individuals which have been crowded at a lower temperature have the greater part of this glycogen replaced by orange globules of fat. This does not apply to the small granules in the skin and gut, which are invariably present in all categories of individuals, but to the subdermal connective tissue masses round the gut and at the bases of the limbs. The connective tissue cells which store the reserve material have thus two alternatives: they may store glycogen, as in the case of the pale translucent individuals, or they may store preponderantly fat, as in the case of the crowded individuals at low temperatures. By following the course of events occurring in the parthenogenetic females under the conditions of isolation at 27° C. and crowding at 10-17° C., it can be shown that the quite young individuals soon after birth in both cases have their reserve material distributed typically in the way shown in fig. 6. There Fig. 6.—A young parthenogenetic female, showing distribution of glycogen and the few large fat globules. are a few large globules of fat, represented by the dark circles, and a large supply of glycogen at the bases of the limbs. Now, as growth proceeds, the individuals kept isolated at 27° C. retain their reserve material in the form of glycogen and do not develop fat in any quantity (fig. 5); they grow and moult very rapidly and may reach maturity in three or four days. The individuals kept crowded at 10-17° C., on the other hand, tend to lose their glycogen deposits and to deposit large quantities of fat, and they grow and come to maturity much more slowly than the isolated individuals at 27° C. It is important to note here the coincidence of glycogen storage and rapid growth on the one hand, and of fat storage and.retarded growth on the other. Now the question arises, is it possible to connect this difference in behaviour The Infe-Cycle of Cladocera. 427 relative to reserve storage with the occurrence and non-occurrence of the sexual forms ? An examination of the condition of the sexual forms strongly suggests that an affirmative answer can be given to this question. The ephippial females are always bright orange in colour owing to the abundant presence of fat in all the subdermal tissue at the sides of the gut and at the bases of the limbs, while the ephippial ovary is characterised by the presence of closely packed globules of fat in the eggs and nurse cells. The appearance of an ephippial female with its abundant reserve fat and opaque ovary loaded with fat is shown in fig. 7. In contrast with this the parthenogenetic female, even under the Fie. 7.—An ephippial or sexual female with ephippium of two egg chambers and opaque ovary full of fat. The reserve material present is in the form of very numerous fat globules of an orange-red colour. crowded condition at low temperature, never exhibits so much fat in its reserve deposits, while the parthenogenetic ovary contains a very large quantity of amylaceous matter in ‘addition to the comparatively sparse large fat globules init. The adult males, as shown in fig. 8, resemble the ephippial females in the abundance of fat present as reserve substance. Another point to be noted is that the ephippial females are inhibited in their growth and never attain to the same size as the parthenogenetic females kept isolated at 27° C., while the males are even more stunted. We thus see that there is a remarkable coincidence between storage of glycogen and rapid growth on the one hand, and fat-storage and inhibition VOL. LXXXVIII.—B. 21 A428 Mr. G. Smith. of growth on the other; that the parthenogenetic females which are kept crowded at low temperatures tend to store fat in place of glycogen and to be retarded in growth; that this tendency reaches its maximum in the sexual forms; and that these sexual forms are produced only by the crowded parthenogenetic females which have a tendency to store fat and to be retarded in growth. The conclusion to be drawn from this series of facts is that the induced fat-storage and retarded growth of the parthenogenetic females crowded at low temperatures are the causal forerunners of the production of the sexual forms. If we regard the parthenogenetic mode of reproduction as being essentially veri % Fie. 8.—Adult, fortnight old, male, showing reserve material present in the form of fat. a form of discontinuous growth or budding, we may observe that this is favoured by the conditions which also induce rapid growth in general, namely, preponderant storage of glycogen under the conditions of isolation and high temperature. The production of the sexual forms, which grow slowly and reproduce with extreme tardiness, is accompanied by a pre- ponderant storage of fat, under the conditions of crowding and low temperature. It is claimed, therefore, that the manner in which external conditions determine the continuance of parthenogenesis or the production of sexual forms is as follows: The condition of isolation and high temperature favours the storage of glycogen as opposed to fat, and this storage of glycogen leads The Life-Cycle of Cladocera. 429 to rapid growth and to continuous parthenogenetic reproduction, which is to be looked upon as a mode of growth by budding, The condition of crowding and low temperature, on the other hand, stimulates the storage of fat as opposed to glycogen, and this storage of fat tends to inhibit growth and to call forth the production of the sexual forms of male and female, which are pre-eminently characterised by abundance of fat-storage and retarded growth and reproduction. Stated in a short and summary fashion, it is claimed that conditions which favour glycogen metabolism lead to rapid growth and parthenogenesis, while conditions which favour fat-metabolism lead to inhibition of growth and the production of sexual forms. The way in which the factor of crowding leads to fat-storage, inhibition ot growth, and the production of sexual forms is still somewhat obscure. But it is clear that the crowding does not act through partial starvation, because in all cases there was an excess of the food material present upon which the Daphnia were known to be feeding. This was ensured by feeding the animals on a pure culture of green Protococcus, which constituted the sole food of the organisms. The only other way in which crowding can be conceived to exert an effect is by the accumulation of some excretory product in the water as the result of the presence of numerous individuals. It is reasonable to suppose that this excretory matter might act in some- thing the same way as phosphorus on a warm-blooded animal, namely, by stimulating the production of fat. All attempts at isolating or collecting this supposititious excretory matter have hitherto failed, and it would appear that it is easily destroyed, possibly by oxidation or bacterial action. 3. The Storage of Fat and Glycogen in its Relation to Growth and Reproduction in Decapod Crustacea. We may now consider how far the theory of the connection between reserve- storage and growth and reproduction in the Cladocera harmonises with what we know of these processes in the higher Crustacea. Ever since the writings of Claude Bernard and the more recent work of Vitzou (2), it has been known that the growth and moulting of the higher Crustacea is accompanied by a remarkable heaping up of glycogen in the liver and subdermal connective tissue. If we take sections through the liver of a crab, such as Carcinus meenas, which is about to cast its skin in the course of a day or two, it will be found, by staining the sections with iodine or neutral red, that the liver cells are crammed with small round granules of glycogen, to the exclusion of almost any other material (fig. 9). At this period there is practically no fat and the protoplasmic content of the cells is small. Besides these storage cells of the liver, the ferment cells, with darker protoplasm and larger nuclei, will be 430 Mr. G. Smith. « seen. In addition to the greatly increased glycogen deposits in the liver, cells containing large masses of elycogen are abundant in the subdermal connective tissue and in the tissue between the liver cells. If the liver of a crab in this condition is extracted for glycogen with hot potash solution, and the amount estimated as sugar by titration, the percen- tage of glycogen will be found to be very high, far higher than at any other time in the crab’s life-history. If now we take sections through the liver of a crab that has recently Fie. 9.—Section through portion of liver, tube of Carcinus just about to moult. The storage cells are crammed with small round glycogen granules. Fre. 10.—Section through portion of liver tube of ditto, some days after the moult, Reserve material is almost entirely absent from storage cells, which are full of protoplasm. completed its moult and has the shell soft and flexible, we shall find it in the condition shown in fig. 10. The storage cells are now almost depleted of glycogen, and consist of protoplasm in which a few globules of fat, especially at the basal ends, are beginning to appear. The subdermal glycogen will also be found to have very much diminished in quantity. It is clear that the glycogen deposited in the liver and subdermal tissues just before the moult has been used up in the formation of the new skin and tissues during the rapid process of growth which follows the moult. If, finally, we take sections through the liver of a hard-shelled crab at a period intermediate between two moults, when growth is not proceeding, we The Lufe-Cycle of Cladocera. | 431 obtain the appearance shown in fig. 11. Here the storage cells are seen to be filled with large and numerous fat globules, the only considerable stores of glycogen being found in the connective tissues outside the liver. Three such connective tissue cells with glycogen are shown in fig. 11. Quantitative estimations of the glycogen and fat in the liver and connective tissues under these various circumstances confirm the result obtained by histology, namely that during the moult there is abundance of glycogen (10 per cent.) and very little fat (3 per cent.), immediately after the moult there is very little glycogen Fig. 11.—Section through ditto, about mid-way between two moults, when growth processes are in abeyance. The storage cells are crammed with fat globules. Three connective-tissue glycogen cells are shown outside the liver. (0-1 per cent.) or fat (5 per cent.),and that between the moults there is abun- dance of fat (15 per cent.) and a rather small amount of glycogen (1°5 per cent. (The figures given here are only rough average approximations, but they give a trustworthy idea of the relative proportions of fat and glycogen in the liver under the various conditions.) We conclude, therefore, that just as in the Cladocera, so in one of the higher Crustacea such as Carcinus, the period of active growth is accompanied by glycogen- as opposed to fat-metabolism, while the fat-storage in the liver 432 ae Mr. G. Smith. - takes place in the intermediate periods when growth is in abeyance and. the reproductive organs are maturing. In Carcuvus monas it was shown in a previous paper that the male and female differed in respect to their fat-metabolism (6). Thus in the female, maturing its ovaries, it was shown that the blood became flooded with lutein and fatty material to a much greater extent than the male, whose blood instead of becoming yellow with lutein is charged with the pink colouring matter tetronerythrin, but is not so heavily charged with fatty material as in the case of the female. Coincidently with this it was pointed out that the female does not grow to the same size as the male, and this is again probably due to the fact that the preponderant fat-metabolism of the female for the nourishment of the eggs exerts a check on growth. Finally it has been clearly shown that individuals of Carcinus infected with the parasite Sacculina do not, increase in sizé at all after once the parasite has established its system of roots in the body of the crab, and this must certainly be ascribed to the fact that the Sacculina induces a most pronounced fatty habit in the liver of the crab, while the glycogenic function is permanently depressed. This has been completely proved by a series of quantitative estimations and histological examinations of the livers of infected crabs. The condition of the sacculinised crabs, both physiological and morphological, is converted by the action of the parasite into that of mature females in the act of ripening their ovaries, and this, as we have seen, consists in a pronounced fatty habit and the inhibition of growth. The above considerations on the processes occurring in normal crabs and in those infected with Sacculina enable us to perceive that there is a marked agreement between these processes in the higher Crustacea and the conditions observed in the Cladocera. In both cases active growth (in which partheno- genetic reproduction is included) is accompanied by storage and use of glycogen for building up the new tissues and skin, while inhibition of growth and sexual reproduction is accompanied by storage and use of fat for the nourishment of the sexual products. This storage and use of fat is more pronounced in the case of the female than of the male, and it is clear that ‘the fat-metabolism in the two sexes proceeds along different lines. In the ease of the female the fatty material developed and stored in the metabolic organs is merely transferred across to the ovary, while in the case of the male it appears that the fat storage is less pronounced and that the fat is not transferred as such to the testis in anything like the same quantity, but is broken down and used for other purposes. By thus bringing the processes of growth and reproduction in the Cladocera and in the higher Crustacea into agreement we obtain a certain The Life-Cycle of Cladocera. 433 insight into the physiological basis of the antagonism between growth and sexual maturity which undoubtedly exists in the Crustacea, and the principle apples with modifications to organisms in general. This antagonism is seen to be due to the necessity for the mature organism to produce a special kind of nutriment for the reproductive organs, so that there is a corresponding — lack of the suitable reserve substances for the purposes of growth. In the Crustacea and at some phase of the reproductive period in all organisms, the elaboration of fat for the supply of the ovary or accessory organs of repro- duction is a marked feature of the metabolism in the mature female, and the diversion of reserve material in this form and for this purpose inhibits erowth. In the male it is less obvious in what special form the reserve material for the nourishment of the reproductive organs is prepared, but here, again, it is probable that fat plays an important part, though the manner of its utilisation is certainly different from the comparatively passive transference which occurs in the female. The alteration in the metabolism thus brought about at sexual maturity, differing in its mode of operation in the male and female, we hold to be responsible for those morphological and physiological changes in the body which often accompany sexual maturity and are known as correlated secondary sexual characters. The view developed here as to the nature of sexual maturity and its antagonism to growth has an interesting bearing on the meaning of sex in general. Speaking broadly, the onset of the sexual mode of reproduction in organisms occurs under conditions when continued growth or asexual multiplication is hindered either by lack of appropriate food or accumulation of excretory matter or by some internal weakening of the assimilative capacity. Under such conditions the organism responds by laying up reserve material for a special kind of resting reproductive cell instead of continuing to expend it in growth. The sexual mode of reproduction is thus a means of lying dormant during conditions unfavourable to continued erowth. The differentiation into male and female may be looked upon as an economy or division of labour by which the female reproductive cell stores up compactly a mass of reserve material to be used for the nourishment of the next generation, but thereby loses the power of division, while the male reproductive cell retains the kinetic energy for division but relies on the female cell to supply the material for development. Summary. 1. By isolating the young Daphnia at birth and keeping them at 27° C. it has been possible to breed them for 19 generations without the appearance of males or ephippial females,3752 parthenogenetic females having been produced. 434 The Infe- Cycle of Cladocera. 2. Parallel cultures to the above, when the parents are kept crowded to the number of 10 in a glass and at a temperature of 10-17° C., produced about 7 per cent. males and 10 per cent. ephippial females. 3. The crowding does not directly influence the supply of food, but appears to act by the accumulation of excretory matter in the glasses. 4, The parthenogenetic females kept isolated at 27° C. grow and repro- duce more rapidly than those crowded at 10-17° C., and they store up reserve material almost exclusively in the form of glycogen, while the crowded parents at a lower temperature tend to store up fat instead of glycogen and are inhibited in their growth. 5. The storage of fat as opposed to glycogen is especially characteristic of the males and ephippial females; hence it is judged that the fat-storage induced experimentally in the crowded parthenogenetic females at 10-17° C. is causally connected with the production by them of the sexual forms. 6. We may conclude that the habit of glycogen-storage leads to rapid growth and parthenogenesis, which is a form of discontinuous growth, while the habit of fat-storage leads to inhibition of growth and sexual mode of reproduction. 7. In the higher Crustacea the act of growth and moulting is accompanied by heaping up of glycogen in the liver storage-cells as opposed to fat, while in the periods between moults fat-storage preponderates. 8. Preponderant fat-storage in the liver is characteristic of female crabs maturing their ovaries and of crabs infected by Sacculina, and in both these cases growth is inhibited. 9. We thus find that both in Cladocera and Decapoda growth on the one hand, and sexual maturity on the other, are accompanied by a different type of reserve storage, which is also distinct in the case of the male and female. This is the physiological fact at the root of the antagonism between growth and sex. 10. Sexual reproduction is a reaction to conditions when continued growth is disadvantageous or impossible. Sexual differentiation is an economy or division of labour by which the female reproductive cell stores the material for development and thereby loses the power of division, while the male cell retains the power of division but relies on the female to supply the material for development. , ‘ Lepidostrobus kentuckiensis, nomen nov., formerly L. Fischeri. 43 LIST OF LITERATURE. 1, Weismann, A., “ Beitriige zur Naturgeschichte der Daphnoiden,” ‘ Zeitschrift f. Wiss. Zool.,’ vols. 27-33 (1876-79). 2. Vitzon, A., “Recherches sur la Structure et la Formation des Tegumens chez les Crustacés décapodes,” ‘ Arch. de Zool. Expér. et Génér,’ vol. 10, p. 451 (1882). 3. Woltereck, R., “ Veriinderung der Sexualitiit bei Daphniden,” ‘ Internationale Revue der Gesamten Hydrobiologie,’ vol. 4 (1911). 4. Langhans, V. H., “Der Grossteich bei Hirschberg,” ‘ Monographien zur Inter- nationalen Revue der Gesamten Hydrobiologie,’ vol. 3 (1911). 5. Grosvenor, G. H., and Smith, G., “The Life Cycle of Moina rectirostris,” ‘Quart. Journ. Micro. Sci.,’ vol. 58, p. 511 (1913). 6. Smith, G., “Studies in the Experimental Analysis of Sex.—Part X,” ‘Quart. Journ, Micro. Sci.,’ vol. 59, p. 267. W. E. Agar, “ Parthenogenetic and Sexual Reproduction in S/mocephalus vetulus and other Cladocera,” ‘Journal of Genetics,’ vol. 3 (1914). 8. Thornton, H. G., and Smith, G., “Conditions of Nutrition in Protozoa,” ‘ Roy. Soe. Proc.,’ B, June, 1914. =I Lepidostrobus kentuckiensis, nomen nov., formerly Lepidostrobus Fischeri, Scott and Jeffrey : a Correction. By D. H. Scorr, For: Sec. B.S. (Received January 14, 1915.) In a paper by Prof. Jeffrey and myself, published in the ‘ Philosophical Transactions , last vear,* we described a new species of Lepidostrobus from the Waverley Shale of Kentucky, under the name, Lepidostrobus Fischert. My friend, Prof. R. Zeiller of Paris, has now kindly pointed out to me that the specific name Fischeri is not admissible, another fossil cone having been described in 1890 by M. B. Renault, under the same name, Lepidostrobus Fischeri.t |. am sorry to have overlooked this reference, an oversight for which I am solely responsible. Our fossil must now receive a new name and it is unfortunate that it is no longer possible to record in the specific designation the name of the discoverer, Mr. Moritz Fischer. The name I now propose for our cone is Lepidostrobus kentuckiensis, after the State in which the plant-bearing deposit occurs. The diagnosis is briefly repeated below. * D. H. Scott and E. C. Jeffrey, “On Fossil Plants, showing Structure, from the Base of the Waverley Shale of Kentucky,” ‘ Phil. Trans.,’ B, vol. 205, pp. 315-373 (1914). + “Etudes sur le Terrain Houiller de Commentry.—Flore Fossile, 2me partie,” ‘ Bull. Soc, Industr. Min.,’ 3e Série, 1V, 2me Livr., p. 526, Plate 61, fig. 3 (1890). VOL. LXXXVILI.—B. 2M 436 Lepidostrobus kentuckiensis, nomen nov., formerly L. Fischer. Lepidostrobus kentuckiensis, nomen nov. Lepidostrobus Fischert, Scott and Jeffrey* (non Kenault). Cone large (4 em. in diameter to outer end of sporangia). Sporophylls in about 35 vertical series. Stele with a large “pith” of prosenchymatous cells, surrounded by a somewhat narrow ring of xylem with prominent angles. Leaf-traces with definite, confluent sheaths. Inner (or middle) cortex narrow, with an interwoven structure, but no gaps. Outer cortex very wide, prosenchymatous. ‘ Pedicels of sporophylls triangular in section, with a groove and median ridge on the upper surface; vascular bundle (rarely preserved) lying in soft tissue above the median ridge. Sporangia reaching 17 mm. in length, with a palisade-wall and distal crest. Microspores in tetrahedral tetrads. Tetrads about 96, individual spores about 60 x 48u in diameter, smooth. From the base of the Waverley Shale, near Junction City, Boyle County, Kentucky, U.S.A. * ©Phil. Trans.,’ B, vol. 205, pp. 354-363, Plate 29, phots. 15-21; Plate 39, figs. 20-23 (1890). snsénian [on 437 _APRIB 19: ha Investigations on Protozoa in Relation to the Factor Inmiting Bacterial Activity m Soil. By T. Goopey, M.Sc., Protozoologist, Research Laboratory in Agricultural Zoology, University of Birmingham. (Communicated by Prof. F. W. Gamble, F.R.S. Received December 3, 1914.) Introduction. In the course of my work on soil protozoa particularly in relation to the question of the partial sterilisation of soil, I had occasion to work with some of the old stored soils kept at the Rothamsted Experimental Station, Harpenden, at which laboratory the work here recorded was commenced. These soils are remarkable for the length of time they have been stored, 67 years being the longest period, and for the fact that in many cases the original samples put up in large bottles have remained untouched since the day on which they were bottled. Preliminary cultures of some of these soils in hay-infusion were begun in 1912 to ascertain the character of the protozoan fauna, if such still persisted in them. From these cultures it was found that in a mixed sample from Broadbalk, bottled in 1846 and containing about 3 per cent. of water by weight, no protozoa were present, whilst in another mixed sample taken from six bottles of Barnfield soil, put up in 1870 and containing about 10 per cent. of water by weight, amcebe and flagellates but no ciliates were present. Quantities of these two soils were taken and were submitted to partial sterilisation treatment in order to find out if the limiting factor usually eliminated by partial sterilisation was present in them, The results obtained by bacterial counts over a period of about 281 days showed that in the 1846 soil no factor limiting bacterial activity was present, whilst in the 1870 soil the limiting factor was present.* As a result of this work I decided to use some of the 1846 soil for inoculation with different species of protozoa obtained from soil, in order to test if possible their power to act as the factor limiting bacterial activity. The protozoa selected for culture and inoculation into separate samples of soil were the following :—Colpoda cucullus, Col. maupasii, Col. stein, and Vorticella microstoma. Ameba sps.? and Flagellate sps.? were obtained by culture from the 1870 soil which, as already mentioned, had been found to contain * This work was carried out in collaboration with Dr. H. B. Hutchinson at Rothamsted. Wl, WOOO fila, 2N 438 Mr. T. Goodey. Investigations on Protozoa in the limiting factor, presumably the amcebe and flagellates in it, according to Russell and Hutchinson’s hypothesis. Besides the above series of samples the set was made up to include a bottle of untreated soil, and one inoculated with a culture of bacteria representative of the bacterial flora added with the cultures of protozoa in the other samples, so as to serve as a check against them ; and a bottle to receive 10 per cent. of the 1870 soil, thus making nine bottles of soil in all. Another set of soils was experimented upon at the same time. This consisted of seven bottles of fresh Hoosfield soil, partially sterilised first by toluene and then by heating to 65° C., so as to eliminate the limiting factor, and then inoculated again with cultures of protozoa obtained from the untreated soil. The series consisted of the following :—Untreated, Toluened, Toluened + Untreated, Toluened + Ciliates, Toluened-+ Amcebe, Toluened + Flagellates, Toluened+ Bacteria. The bacteria used for the last-named inoculation were representative of the bacterial flora of the other cultures. In each set of bottles the water content of the soil was finally brought to about 18 or 20 per cent. by weight ; this being about the water content at which many of Russell and Hutchinson’s* soils have been maintained, and at which they have found the limiting factor to be active. I decided at the outset to make periodic bacterial counts by the gelatine- plate method in order to determine the numbers of bacteria in the soil as nearly as possible once a month and to carry on the experiments for a long period. Methods. A mixed sample of soil from six bottles of 1846 soil was taken and divided into nine lots of 400 grm. in each. Each lot of soil was put up in a quart bottle which had previously been sterilised and plugged with cotton wool. In the case of the Hoosfield soil seven 400 grm. lots of slightly air-dried soil were taken after having been passed through a 3-mm. sieve. These were bottled in exactly the same way as the 1846 soil. The soil was first toluened by the addition of 2 per cent. of toluene, which was allowed to remain in the soil for two days, after which the soil was spread out on sheets of paper so as to allow the antiseptic to evaporate. Hay-infusion cultures of the toluened soil were made, and as it was found that flagellates developed in the cultures the bottles of soil were submitted to steam heat at a temperature of 63°-65° C. for three to four hours. This operation was * Russell and Hutchinson, ‘ Jour. Agric. Science,’ vol. 3, Part II (1909), and vol. 5, Part II (1913). Relation to Factor Linuting Bacterial Activity im Soil. 439 carried out in a steamer, the temperature of which was regulated by means of a thermostat. Hay-infusion cultures were made after this second treat- ment, and it was then found that no flagellates cropped up. The cultures of protozoa used for the inoculation of the bottles of soil were obtained in the following manner: Hay-infusion cultures were made’ from fresh soil and from old cultures containing cysts which I had on hand. By the use of fine capillary pipettes it was possible to isolate ciliates, which were then sub-cultured in hay-infusion. I found it best to use hay-infusion already containing active bacteria for the sub-culture of isolated forms, the bacteria serving immediately as a source of food for the protozoa. Cultures of flagellates from the 1870 soil were obtained in the same manner, and for the culture of amcebe from the 1870 soil I made use of cysts from pure cultures on agar plates which I had by me. In this way pure cultures of the following protozoa were obtained for use with the 1846 soil :—Col. cucullus, Col. maupasii, Col. steinii, Vort. microstoma, Ameba sp. ?, and Flagellate sp. ?. The protozoa for inoculation into the treated Hoosfield soil were obtained by isolation and sub-culture of forms cultivated in hay-infusion from the untreated Hoosfield soil, so that the forms added should represent as nearly as possible the fauna originally present in the soil. The cultures of protozoa thus obtained were one of Amba sp.?, one of Flagellates sp. ?, and one of Ciliates, including Col. cucullus, Col. stein, and Col. maupasii. The small Ciliate Balantiophorus minutus or elongatus also occurred in ‘the cultures made from the untreated soil, but as I was unable to obtain this free from flagellates, the culture of ciliates did not include this form. In order to obtain mass cultures of the protozoa in sufficient quantity to serve for inoculation into the soil the following method was employed: 80-grm. lots of washed and sterilised sand were put into large sterile petri dishes or glass cylinders and covered with hay-infusion, which was then infected with a pure culture of protozoa, and the latter were allowed to multiply and populate the culture. In this way a large quantity of each kind of protozoa was obtained for the inoculation of the soil. This process was carried out by spreading the soil on sheets of sterilised brown paper and then mixing the sand-hay-infusion culture of protozoa into it by means of a sterilised spoon, the whole of the soil, sand, and hay-infusion becoming thoroughly well mixed together and thus ensuring an even distribution of the protozoa throughout the soil. In the case of the 1846 soil the inoculation was carried out in a glass-house which had been steamed down in order to allay dust and thus minimise the chance of infection. The soils were left exposed in this house for some days 2N 2 440 Mr. T. Goodey. Investigations on Protozoa in in order to allow the bulk of the water to evaporate off, and a heating lamp was put into the house in order to accelerate slightly the evaporation. The reason for thus driving off the bulk of the water from the soil was that I desired to bring about in the soil all the conditions possible for aiding the excystation of the added encysted protozoa, for I had found in experi- menting with cysts that if they were slightly air-dried and then moistened, excystation was more rapid than where no slight drying had been allowed. The water-contents of the inoculated soils, when bottled again at the end of these few days of air-drying, were as follows:—Untreated, 6:2 per cent. ; Untreated + Bacteria,* 2°7 per cent.; U. + Col. cwcullus, 7-4 per cent.; U. + Col. steinit, 722 per cent.; U. + Col. maupasiz, 6-9 per cent.; U. + Vort. microstoma, 68 per cent.; U.+ Amabe, 7-5 per cent.; U. + Flagellates, 75 per cent.: U. + 1870, 6°5 per cent. Sterile distilled water was next added to all the soils, in sufficient quantity to bring up the water-content of each to 18 per cent. | In the case of the seven bottles of Hoosfield soil, the samples were inoculated with protozoa from the sand-hay-infusion cultures in exactly the same way as described above. These lots were spread out on sheets of sterile brown paper and left in the drying room for five hours at a temperature of about 20° or 22° C. in order to drive off the bulk of the added water and bring about conditions favourable to the excystation of ptotozoa after re-moistening. The water-content of the various soils after drying was as follows :—Untreated (U.), 7-4 per cent.; Toluened and heated (T.), 5°9 per cent.; T. + Untreated (IT. + U.), 58 per cent.; T. + Ciliates (T. + C.), 3:1 per cent.; T. + Ameba (T.+ Am.), 2°7 per cent.; T. + Flagellates (T. + Fl), 3 per cent.; T.+ Bacteria (TI. + B.); 1:6 per cent. Sterile distilled water was then added, as in the case of the 1846 set, in order to bring up the water-content of each lot to 18 per cent. by weight. Both sets of bottles were then left in a small warmed glass-house, the temperature of which varied between 45° and 55° F. Later on they were taken from the glass-house and kept in the laboratory in a room at about 12-15° C. At various intervals during the course of the work the water- content of each soil has been determined in order to estimate the loss of water by gradual evaporation, and at these times the loss from each has been made good by the addition of sufficient sterile distilled water to bring up the water-content to 18 or 20 per cent. by weight. In attempting to estimate the numbers of protozoa present in the soils the following methods have been employed :— * This lot was treated in the same manner as the inoculated Hoosfield soil described further on. Relation to Factor Inmiting Bacterial Actunty in Soil, 441 A dilution method was first used which Dr. H. B. Hutchinson had devised and which he and I had used in 1910 in the course of some joint work. Ten grm. of soil are shaken up for four minutes with 100 cc. of 1-per-cent. hay-infusion, either sterile or containing an active growth of soil bacteria. By means of sterile 1 c.c. pipettes varying quantities of this soil suspension are taken out and placed in sterile tubes; three tubes of each dilution being put up. In the case of the smallest quantities of soil suspension more hay- infusion is added in order to give a sufficient quantity of liquid for purposes of manipulation. The scheme of dilution is as follows :— grm. of soil. 10 ‘cc. original soil suspemsion ...............0.cseseeeeer nee = 5 f PENMAN td rk Rem Cok t as Bec = 0°5 2 50 Fee Viti co) MR GHEOOSCEDaSEE MDE C RCE RBEER STE = 0:2 1 5 pin’ URL \gdagboncedecbnaobencespobeseds = Oil 0°5 _ LCL Mes nutcene ce pscuemeaceices stern = 0°05 0:2 # Aston tn HAR AMIS AM ea noaatins dace sacs cbs = 0°02 0-1 i FHS Wiate Wideanacobeocaocnors sae oceenerCne = OO)! 0°5 cc. mixture of 1 ¢.c. original+9 c.c. hay-infusion ... = 0°005 0°2 55 3 sp i 1 OOO? Orl 55 %) m0 1 ee —sOROOL The cultures thus obtained are allowed to incubate for about a week and microscopical examination is made of the surface layers by taking out drops on a sterile platinum loop and examining them on glass slides. If protozoa occur in the cultures of any particular dilution, then one infers that they are present in the soil in numbers equal to the factor required to raise the particular dilution to 1 grm. Thus if they occur in all three cultures of 0001 grm., then there are at least 1000 protozoa per gramme of soil. The method possesses the advantage that one deals with a comparatively large quantity of soil, viz. 10 grm., and should thus be able to overcome the difficulties of any irregular distribution of protozoa in the soil itself, provided a good suspension is made. However, in working with it I have obtained most irregular results, which I have not been able to explain, and for this reason I have practically given up using it in favour of an agar-plate method. Before leaving the hay-infusion method, however, I may add that I carried out a series of experiments in order to ascertain if the violent agitation of the soil and liquid, in making the suspension by shaking for four minutes, had any injurious effect on living protozoa. I found, when soil was added to a hay-infusion culture containing innumerable active ciliates and apparently no encysting forms and the mixture was shaken violently for four minutes, that on making a series of dilution cultures from this suspension protozoa 442 Mr. T. Goodey. Investigations on Protozoa in cropped up in abundance throughout, thus showing that they had suffered no damage. The agar-plate method which I have used is as follows:—Sterile petri dishes are poured with nutrient bouillon agar of about 0°5 to 1 per cent. in strength, and, when cool, the surface of the agar is inoculated with a weighed quantity of the soil the number of protozoa in which it is desired to ascertain. Three plates, as a rule, are inoculated with each weight of soil and the follow- ing are the weights of soil which have been used—1, 0°5, 0°2, 0:1, 0°05, 0°02, 0°01, 0°005, 0:002, 0-001, 0:0005, 00002, 0:0001 grm. The plates are allowed to incubate for a few days and then the surface of each is examined under the microscope for the presence of protozoa. The method entails the use of a sensitive balance and is limited by the difficulty of manipulating such small quantities of soil as are produced in weighing in the region of 0°0001 grm. However, the results which I have obtained with it are fairly consistent and are more trustworthy than those of the hay-infusion method, I think. It was my hope at the beginning of the experiment to obtain evidence, by means of the counts of protozoa, concerning their activity and multiplication if such were proceeding. Counts of protozoa were therefore made at the beginning and towards the ends of the experiments. The hay-infusion method was used in the first counts and the agar-plate method for the later ones. As I have pointed out, the latter method gives higher counts and more trustworthy results, and one cannot, therefore, strictly compare the evidence afforded by the two methods. For this reason I have not found it possible to obtain sound evidence as to whether the protozoa have multiplied since being added to the soil. See footnote, however, on p. 454. The Bacterial Counts. The results of the periodical deterrhination of the numbers of bacteria by the gelatine-plate method are tabulated below and the curves obtained by plotting these results are shown in figs. 1, 2, 3, 4, and 5. In order to simplify matters I have arranged certain curves together, for the whole nine curves when plotted all together present a confusing array and do not lend themselves to easy elucidation. Fig. 1 shows the curves obtained from the untreated, bacteria, and V. microstoma inoculated soils. The most noteworthy feature is the extraordinarily high bacterial count in the Bacteria soil at 32 days and the subsequent drop in the numbers of bacteria to a level below that of the untreated soil. This low bacterial Relation to Factor Limiting Bucterial Activity im Soil. 443 Bacteria in millions per gramme. Atbegin- | After After | After After After After After ning. | 32 days. | 63 days. | 92 days. | 124 days. | 153 days. | 181 days. | 232 days. Untreated ......... 6°6 391 291 307 280 203 261 118 Bacteria. csc... 02 15 1504 472 440 259 98 102 59 Col. cucullus ...... 235 464 386 312 315 174 266 159 Col. steinit......... 218 379 314 271 289 165 324 142 Col. maupasit...... 253 501 262 313 260 191 264 116 Vort. microstoma| 186 453 216 201 156 65 119 84 Amoeba ...........- 178 599 412 274 409 193 242 178 Flagellates ......... 82 374 327 416 319 127 188 132 WTO es ccie=s00 8°5 204: 154 169 128 116 159 103 After After After After After After After 284 days. | 324 days. | 360 days. | 383 days. | 419 days. | 486 days. | 519 days. Untreated ......... 50 55 lost 115 76 87 57 Bacteria......:...... lost 27 | 40 36 20 45 33 Col. cucullus ...... 96 102 143 104 95 134 70 Col. steinii......... 77 138 | 112 224 133 138 130 Col. maupasii ... 75 116 | 126 131 114 133 91 Vort. microstoma 35 86 39 Qovelt 67 91 58 PACE DB Si raeteen soc | 160 135 129 242 160 168 118 Flagellates......... 54 86 93 120 75 124 123 Weak: (Unto peccerace 84 83 86 111 76 93 68 VORTICELLA~ a ROS Fo. °, 4 ~ content was maintained over a very long period—366 days—and is perhaps the most surprising and unexpected result of the whole investigation. 444 Mr. T. Goodey. Investigations on Protozoa in The V. nicrostoma curve is also very interesting and shows the influence of some factor which had become operative by the end of 63 days and which subsequently kept the numbers of bacteria in check, though only at about the same level as the untreated soil. The untreated curve also shows that the bacteria have gradually decreased ‘in numbers after reaching and maintaining a high level for 181 days. These results are very interesting when considered in relation to the number of protozoa in the soils. In the untreated 1846 soil no protozoa are present, so that the gradual decrease 'in the number of bacteria cannot be due to the activity of the protozoa. The V. microstoma soil, however, contained, a few weeks after inoculation, about 300 vorticellze per gramme. By December, 1913, however, all the vorticellee had died out, for they failed to appear in cultures of the soil made at that time and on all occasions since.* Flagellates and amcebee are present in this soil, probably due to infection of the mass culture or during the initial air-drying of the inoculated soils, to the extent of 1000 flagellates and 100-200 amcebe per gramme. It is probable that these lead an active trophic existence in the soil and so might be considered responsible for the limiting action on the bacteria. This can scarcely be the case, however, when we consider this soil in relation to the bacteria-inoculated soil. In the latter there are flagellates present to the extent of about 100 per gramme. If now the lmiting factor in this soil is considered as due to the action of these flagellates, we should expect to find not so great a decrease in the bacterial numbers as in the vorticella soil, where the flagellates are more than ten times as numerous, © and where there are, in addition, 100-200 amcebe per gramme. The reverse of this is the ascertained result, and clearly negatives the idea that the flagellates and amcebz are responsible for the limiting action on the bacteria. Another point of interest is that the two curves for the untreated and * This dying out of the Vorticella mucrostoma is very interesting. At first I thought its failure to appear might be due to an unsuitable culture medium. I, therefore, tried to obtain it again, taking care of the reaction of the hay-infusion, but with no better success. It also failed to appear on a nutrient bouillon agar, favourable to the growth of all the other protozoa under consideration. I afterwards remembered that in some earlier experiments I had failed to obtain Vorticella from a soil which had been kept in the laboratory for some months and from which I had obtained a very fine culture of the organism when the soil was fresh. At another time too I failed to get the excystation of V. microstoma from cysts which I had obtained in a hay-infusion culture and which had been stored for a few months. From this evidence it would appear that V. microstoma in its encysted condition does not retain its vitality for more than a few months, and the dying out in my soil is easily accounted for on this supposition. Bacteria in Millions per Gram. ex c o Relation to Factor Limiting Bacterial Activity in Soil. 445 V. microstoma soils are, during the greater part of their course, so closely akin that within the limits of experimental error they may be considered identical. In one of them no protozoa are present, whereas in the other amcebe and flagellates are present. If the latter are the limiting factor they have failed to reduce the numbers of bacteria below | the level of the untreated soil containing no protozoa. Fig. 2 shows the curves for the three samples of soil inoculated with the m ZOE > 8 three different species of Colpoda, viz., Col. cucullus, Col. steinwi, and Col. mawpasii, together with the curve of the untreated soil. There is a marked similarity between all four curves; all of them are of the same type and show no very pronounced differences. On the whole, the bacterial content of the three inoculated soils has remained higher than that of the untreated soil, in spite of the fact that each contained many hundreds of protozoa per gramme. They thus fail to indicate any action by the protozoa of a limiting character on the bacterial population of the soil. In the Col. cucullws soil there are, roughly, about 750 Col. cucullus and 1000 flagellates per gramme, the latter only being found in the later determinations by the agar-plate method. The Col. steinii soil contains about 100 Col. steinii and 1000 flagellates per gramme, whilst the Col. maupasw soil contains about 1000 Col. maupasii, 200 amcebe, and 100 flagellates per gramme.* One would have expected that had the protozoa been capable of acting as a check on the growth of bacteria in the soil, they would have brought their numbers to a level well below that of the bacterial content of the untreated soil. Instead of this, however, we find after 519 days all three soils showing a higher bacterial content than the untreated soil. * The numbers given for the protozoal counts are those obtained in the last determination. The amcebz which occur now in these soils are due most probably to infection either of the soil samples during the initial air-drying or of the mass cultures. 446 Mr. T. Goodey. Investigations on Protozoa in This is good evidence, I think, that the protozoa have not functioned as the limiting factor. In the case of the soils inoculated with cultures of amcebe and flagellates obtained from 1870 soil, the curves for the bacterial counts of which are shown in fig. 3, the general inference to be drawn is that after a period of x s i) 8 gs 8 € 8 Bacteria in Millions per Gram ooo 18 months in which to act, the protozoa have not exerted a limiting action on the bacteria in their respective soils. At 519 days the bacterial content of both inoculated soils is well above that of the untreated soil. The amcebe in the sample of soil specially inoculated with them are present to the extent of 10,000 per gramme, whilst in the flagellate-inoculated soil there are 10,000 flagellates and about 2000 amcebse per gramme. These results are very interesting, for they indicate that even when protozoa are present in the soil in such large numbers and under conditions favourable to active existence they do not exert a depressing effect on the bacteria. The amceba curve is especially significant, for it shows that even in the presence of 10,000 amcebze per gramme of soil the bacteria can maintain a higher level in numbers than in the original soil containing no protozoa. The curve for the flagellate-inoculated soil does not call for much comment. It is practically identical with that for the untreated soil during a great part of its length, but on the whole is at a higher level and indicates that the 10,000 flagellates and 2000 amcebe per gramme are not capable of bringing the bacterial content down below the level of the untreated soil. Fig. 4 shows the curves for the untreated soil and for the sample to which 10 per cent. of 1870 soil was added. At the end of 519 days the untreated soil is at a lower level than the U.+1870, though from 232 days onwards the two curves are very similar, and from 360 days to the end of the experiment may be considered as identical. There are about 2000 amcebe and 2000 flagellates per gramme in the mixed soil, and the curves show that these have not been able to bring down the bacteria to a level below that of the Relation to Factor Limiting Bacterial Activity im Soil. 447 untreated soil (see footnote, p. 454). The Untreated +1870 curve is, on the whole, very even, showing no marked fluctuations up or down, and it is very UNTREATED Bacteria in Millions per Gram 8 38 Oo o OO ALF = heal ea NI el IS nS a ae IN a ea te 50 v 232 284 4 519 DAYS Fic. 4. interesting to note that during the first 232 days the bacterial counts are below the counts for the untreated soil. This seems to indicate that some factor was introduced with the 1870 soil which, for a time, checked the rapid growth of bacteria and prevented their increasing to the numbers attained by the bacteria in the untreated soil. It is conceivable that this was owing to the action of the protozoa added with the 1870 soil, but the fact that after 232 days the bacterial numbers for the untreated soil came down below the level of the mixed sample, and that the two curves during the last 160 days are practically identical lends no support to this idea. Whatever the influence was which during the first 232 days checked the growth of the bacteria in the mixed soil, I am inclined to regard it as con- nected with some other property of the 1870 soil than the presence of protozoa in it. To be more explicit: The 1870 soil was bottled in a com- paratively moist condition and received no drying about 1881 as did the 1846 soil along with many others. This drying seems to have effected a very important change in the 1846 soil and, taken in conjunction with the pro- longed period of storage has produced in it a condition comparable with partial sterilisation. At any rate, the 1846 soil along with other dried and stored soils which I have examined gave very high bacterial counts when moistened, whereas the 1870 and other soils which have been stored in almost the same condition as they were in when taken from the field give low bacterial counts and indicate the presence of the limiting factor. The following experiments illustrate my point. Bacterial Counts in other Samples of Old Stored Soils. Three soils were taken for this piece of work, two of them from bottles of Broadbalk soil stored since 1856 and 1865, and one from a bottle of Geescroft 448 Mr. T. Goodey. Investigations on Protozoa in soil stored since 1865. The two from Broadbalk were dry when taken from their original bottles, whilst the Geescroft sample was in a comparatively moist condition. The water-contents of these soils were not determined immediately after taking them from their respective bottles, because it was not my intention at that time to use them for bacterial counts but merely to ascertain the character of the protozoan fauna. As far as I could judge, I should say that the Geescroft soil contained about 10 per cent. of water, whilst the Broadbalk samples were of about the same degree of dryness and contained about 2 per cent. or 3 per cent. of moisture. It was evident from the appearance of the soils that the Broadbalk soils had been taken out and dried along with many other soils in 1881, whereas the Geescroft soil had been left untouched and closely resembled the Barnfield 1870 soil, containing about 10 per cent. of water, which I had used in earlier work. I found on examining these soils culturally that the Broadbalk 1856 con- tained no protozoa, the Broadbalk 1865 contained amcebe and flagellates and the Geescroft 1865 also contained amcebe and flagellates. A weighed quantity of each of these soils was taken and after making initial counts to determine the bacterial content they were all moistened to 20 per cent. water-content. After a period of 148 days the soils were remoistened to bring up the water-content to 20 per cent. again, owing to the gradual loss of moisture by evaporation. Bacterial counts by the gelatine-plate method were made at different intervals and the results are set out in the table below. Bacteria in Millions per Gramme. At After After After After After | beginning. | 68 days. 102 days. 130. days. 198 days. | 2381 days. 136 | 169 73 200 158 Bd. 1856 ...... 4 Bd. 1865 ...... | 4 66 | 93 55 120 124 G. 1865 ...... 2 25 11:4 | 5 10°7 18 ‘6 9 The curves obtained on plotting these results are shown in fig. 5. The most noteworthy feature of these curvesis the high bacterial content of the Broadbalk soils and the low bacterial content of the Geescroft soil. The former have all the appearance of curves of partially sterilised soils, whilst the latter presents the usual appearance of an untreated poor soil, containing a limiting factor. The drop at 130 days in the two Broadbalk soils may be accounted for by Relation to Factor Ianuting Bacterial Activity im Soil. 449 the loss of moisture, and the subsequent rise in the bacterial content may likewise be attributed to the more favourable conditions occasioned by remoistening the soils. Considered in relation to their protozoan fauna, these results are very instructive. In the Broadbalk 1856 there are no protozoa. In the Broad- . N So oOo . 180 u, NS = é c / Wh Sian \\ : je) Y,, 2 . Css 0 1404 A \ of a fi «, o ; yy z= 120 @ 9; 5 / \ S wr = 100 yy : sf sb” — Ga = Fe eiN oe = 80 2G \ 4 . oe ~ (e / Z \ V Oy = Z \ 7 a4 A fia) 604 fe of Sa ic 7 Y 40} 4 = ZA. ee lie B 204 ; A Fr 1862: YF oEEScRO 0 20 40 60 68 80 10002 +120 30 40 160 180 198 220 231 ~«. DAYS. Mice 15), balk 1865 there is a rich protozoan fauna to the extent of about 5000 amcebee and 5000 flagellates per gramme. In the Geescroft 1865 there are about 500 amcebee and 500 flagellates per gramme. Thus the curve for the Broadbalk 1865 shows that in spite of the presence of this large number of protozoa, the bacteria can maintain a high level in numbers even after 231 days during which the protozoa should have been reducing them. It may be suggested that the bacterial counts for Broadbalk 1856 are higher than those for Broadbalk 1865 because in the former there are no protozoa present to check the growth of the bacteria. I would point out, however, that the two Broadbalk curves are practically of the same order as compared with the Geescroft curve. One would have thought that in the presence of such a large number of protozoa as in the Broadbalk 1865, had these been capable of functioning as the limiting factor, they would have checked considerably the multiplication of the bacteria and brought them down to somewhere near the level of the bacterial content of the Geescroft soil. My point is to show that the drying to which the Broadbalk soils were submitted has brought about a change in them strictly comparable with the change usually produced by partial sterilisation, and at the same time has 450 Mr. T. Goodey. Investigations on Protozoa in produced this change in one of them without killing off the amcebe and the flagellates, The Geescroft soil remained undried and has a much scantier protozoan fauna than the Broadbalk 1865 soil, yet the curve for:the bacterial counts in this soil would be interpreted as showing the presence of the usual limiting factor. I have no evidence on which to base a suggestion as to what the real character of the change is which has been produced in the Broadbalk soils by drying. I do suggest, however, that it has an intimate relation to the high bacterial counts which I have obtained on remoistening the soils. I am quite prepared to admit that the merely negative evidence furnished by the above results does not help forward very much the final solution of this elusive problem, but at the same time I think that it is useful. It points to the fact that much more information is required than is at present available on the changes brought about in soil by rapid air-drying or by drying at temperatures sufficiently low to avoid the killing of protozoa in the soil. Russell and Hutchinson give details of several experiments on this particular line of investigation in their second paper (p. 166), but there is room for still more research on these points. Hoosfield Inoculated Soil Bacteriai Cownts. The results of the bacterial counts for this set of soil samples are tabulated below, and the curves obtained by plotting these are shown in figs. 6, 7, and 8. As in the case of the 1846 set of soils I have arranged certain curves together for the sake of simplifying matters. It is necessary to point out at the outset that in attempting to interpret these results there are two standards of comparison, viz., the curve for the untreated soil and that for the toluened soil. For this reason I have intro- duced each of these curves into all three graphs. Fig. 6 shows the curves for the Untreated, Toluened, T.+ Ciliates, T.+Ameebe, and T.+4Flagellates. It will be seen that the untreated soil exhibits a normal low bacterial content ; the limiting factor is here exerting its full influence. Compared with the untreated, the curve for the toluened soil shows that the usual partial sterilisation effect has been obtained, the bacterial numbers rising to and maintaining a level at about 50,000,000 or 60,000,000 bacteria per gramme. Examining now the curves of the three inoculated soils represented in this graph and comparing them especially with the curve for the toluened soil, we find that after the lapse of 487 days the bacterial contents are higher than that of the toluened soil. Leaving i Relation to Factor Linuting Bacterral Activity in Soil. 451 Bacteria in millions per gramme. At begin-| After After | After | After After After | After ning. 32 days. | 60 days. | 93 days. | 125 days. | 151 days. | 173 days. | 208 days. Untreated (U.) ...) 14-4 18 38} 11°4 9 12 135 8 Toluened (T.)...... 9-2 73 60 61 48 40 53 49 ~ POR Fine cecenerccues 11°3 49 6L 43 70 19 39 45 T.+Ciliates ...... 4°5 371 292 296 181 56 70 64 T.+Amebe ...... 3 285 185 141 188 7A 72 90 T.+Flagellates ...) 27 247 214 22/7 368 196 108 104 T.+ Bacteria ...... 2°3 500 341 311 296 181 196 151 After After After After After After After 259 days. | 301 days. | 337 days. | 350 days. | 386 days. | 454 days. | 487 days. Untreated (U.) ... 6 13 lost 8 13 12°6 12 Toluened (T.) ... 42 55 70 67 43 51 56 4h ab Ul a eieadeceenee 23 39 44 51 33 33 48 T.+Ciliates ...... 59 57 65 59 57 57 73 T.+Amebe ...... 57 65 68 62 50 41 57 T. + Flagellates ... 77 104 92 94 81 35 113 T. + Bacteria ...... 116 151 85 105 138 115 150 3604 IN Bo] | vx o | 5280 a iS} & eee oo [s} 8 <) S ~AMoe T+ CILIATES a od ee Bacteria in Millions a o o rt o8s out of account for the moment the high bacterial counts during the first 125 days and the drops at about 150 or 170 days in all three cases, we may say that the protozoa added to the toluened soil in each sample have not reduced the bacteria to a level lower than that of the toluened soil alone. Judged in relation to the protozoa in these soils this is a very instructive result. The T.+ Amoebe contained at the end of the experiment 10,000 amcebee, perhaps more, and about 5000 flagellates per gramme. The T.+Flagellates contained 10,000 flagellates, perhaps more, and about 1000 amcebee per gramme, whilst the T.+ Ciliates contained about 3000 each of Col. stent and Col. maupasvi, about 500 Col. cucullus, and about 1000 452 Mr. T. Goodey. Investigations on Protozoa in flagellates per gramme. These are large numbers of protozoa, and it is probable that the amcebe and flagellates have occurred in the active condition. In no case, however, have these protozoa been able to reduce the bacterial contents of their respective soils to a level permanently below that of the toluened soil. Cultures were made at the end of the experiment to ascertain if protozoa were present in the toluened soil, with the result that about 5000 flagellates and about 10 amcebe per gramme were found in it. Now the curve for the toluened soil is quite a normal one, and shows the usual partial sterilisation results when compared with that of the untreated soil. Whatever be the limiting factor eliminated by the process of toluening and heating this soil, resulting in the rise of the bacterial content from 10,000,000 or 12,000,000 to 50,000,000 or 60,000,000 bacteria per gramme, that factor evidently has no connection with the flagellates, for these have resisted the action of the antiseptic and heat and, though much reduced in numbers at the beginning of the experiment, have succeeded in repopulating the soil. The results obtained from these inoculated soils accord with those obtained from inoculated samples of 1846 soil. In those it was found that the ciliates, amcebe, and flagellates failed to reduce the bacterial content below the level of the untreated soil. In this soil they have not brought down the numbers of bacteria lower than those of the toluened soil. The only inference which I can draw from these results is that the protozoa have not functioned as the limiting factor on bacterial activity. The curve representing the counts for T. + Bacteria as compared with those for the toluened and untreated soils is shown in fig. 7. It is 500 oo Db nN o Q c--} o u (=) So Bacteria in Millions per Gram. x ia) o oo oo Relation to Factor Limiting Bacterial Activity in Soil. 458 evident from this that the bacterial content of the T. + Bacteria has main- tained a consistently high level, the numbers never going down to those of the toluened soil, though there is a decided drop between 32 days and 337 days. It might, at first sight, be supposed that high bacterial content was due to the absence of protozoa from the T. + Bacteria soil, the bacteria of the soil left after treatment, together with those added, having no preying organisms around them to check their growth. The T. + Bacteria soil does, however, contain protozoa, no doubt the offspring of those which withstood the partial sterilisation treatment, to the extent of about 3000 flagellates and 100 amcebee per gramme. Thus there are almost as many flagellates and many more amcebee per gramme of this soil than in the toluened soil. Yet in spite of these numbers of protozoa the bacteria have maintained a much higher level than those in the toluened soil. The high bacterial counts obtained during the first 160 days in all the four soils, viz.: T. + Bacteria, T. + Ciliates, T. + Amcebe, and T. + Flagel- lates, together with the decided drop in all cases, are very interesting. At first sight it might be assumed that in the case of the three soils inoculated with protozoa the drop was due to the limiting action of the latter becoming well established. This would appear to be sound reasoning if it were not for the fact that a similar drop occurs in the T. + Bacteria soil, where no protozoa were added. Moreover, the protozoa found at the end of the experiment in the T. + Bacteria and in the toluened soils are quite com- parable, and if we were to assume that the drop in the bacterial content in the T. + Bacteria soil was due to the activity of the protozoa surviving partial sterilisation, we should be confronted with the difficulty that in one soil the surviving protozoa were exerting a limiting action, whilst in the other they were not doing so, though conditions for trophic existence were equally good in each case. The high counts during the first 160 days may probably be explained by the fact that hay-infusion and very large numbers of bacteria were added to the soils in inoculating them and in this way the conditions brought about were very favourable to extreme bacterial activity as compared with the toluened soil, to which only sterile distilled water was added, and which consequently exhibits no exceptionally high bacterial figures. In the same way the fall in the bacterial counts after about 160 days in these soils may, perhaps, be accounted for by assuming that the food supplies added with the hay-infusion became exhausted, and as a result of this the bacteria dropped to somewhere near the level of the numbers present in the toluened soil. Fig. 8 represents the curve for the T. + 5 per cent. untreated soil along with those for the toluened and untreated soils. It is evident that after VOL. LXXXVIIIL—B. 20 A5A Mr. T. Goodey. Investigations on Protozoa in 151 days the bacterial content of the moculated soil remained ata lower level than that of the toluened soil. At the same time it appears to be pretty clear that the two curves are of the same order. The T. + 5 per cent. untreated does not show a continuous and persistent decline in the numbers of bacteria, as one would expect if the limiting factor were due to the growth and activity of protozoa added with the untreated soil. The two curves are practically parallel from 259 days onwards, and it is obvious that the same set of conditions was affecting the bacteria in each soil. The numbers of protozoa present in each soil are as follows:—In the T. + 5 per cent. untreated there are at least 10,000 flagellates, about 500 amcebe, and an almost negligible number of ciliates—5 or 10—per gramme.* In the toluened soil there are about 5000 flagellates and about 10 amcebe per gramme. There are thus very many more protozoa per gramme in the inoculated soil than in the toluened soil, and the lower bacterial content of the former soil is thus easily accounted for 1f we assume that the protozoa act as the limiting factor. If we only had these two curves and that for the untreated soil on which to base our conclusions, the above inference might be considered correct. But when we take into con- sideration the points already mentioned in connection with the other inoculated soils, it is scarcely possible to assume that this is the real explanation. It has been shown that in the case of the toluened and the T. + Bacteria soils that the flagellates can be left out of account, so far as any possibility of functioning as the limiting factor is concerned. So that if the lower bacterial * This soil affords evidence of the activity of the amcebe added in the 5 per cent. of untreated soil. Amoebe are present in the latter to the extent of about 3000 per gramme, and assuming that this number was present at the time the soils were mixed, we can reckon that in each 100 grm. of the mixture there were 15,000 amcebze or 150 per gramme. There were present at the last protozoal count 500 per gramme, thus showing that the original 150 had increased to 500 per gramme. Similarly the 1846+10 per cent. 1870 soil discussed on p. 447 gives evidence of the amcebze added in the 10 per cent. of 1870 soil having increased from 50 per gramme at the beginning to 2000 per gramme at the end of the experiment. Relation to Factor Limiting Bacterial Activity in Soil. 455 content of the T. + 5 per cent. untreated soil is due to protozoal activity it must be the 500 amcebe and about 10 ciliates per gramme which are respon- sible for it. It seems to me highly improbable that this is the true explanation when we consider the enormously larger numbers of amcebe and ciliates present in the T. + Ameebee and T. + Ciliates soils, where the protozoa ~ obviously have not effected a limiting action on the bacterial contents of their respective soils. It is clear, however, that some factor has been added to the toluened soil in the 5 per cent. of untreated which acts as a check on bacterial growth. I cannot find support in these results, however, for the assumption that this limiting factor is the protozoa. I would suggest that the influences checking the growth of bacteria are connected with some other property of the added soil than its contained protozoa, General Discussion. The introduction of large numbers of bacteria into the samples of soil along with the added protozoa must be a source of disturbance to the bacterial flora, and for this reason the experiments dealt with above cannot be considered as showing a clear issue between protozoa on the one hand and bacteria on the other. I sought to reduce this source of error to a minimum, however, by the continuation of the experiments over a long time, thus allowing the disturbed bacterial floras to settle down so that any influence of the protozoa should be judged after this steady point had been reached, «ec. after about 160 days in both the 1846 and the Hoosfield soil. In order that the protozoa should have conditions, as near as I could bring them about, favourable to excystation I partially air-dried all the soils after they were inoculated. In this way I hoped to meet the criticism which might be brought against the experiments that the protozoa had failed to function. Moreover the soils were all kept under conditions of temperature, water-content and aération exactly comparable with those under which Russell and Hutchinson kept their soils. If protozoa therefore could act as they supposed them to do in their soils they had every chance of doing so in my soils. Another point calls for some comment. Martin and Lewin* have found evidence of an abundant fauna of active amcebz and flagellates devouring bacteria in certain soils which they have tested. They suggest that these have probably some influence on bacterial numbers and thus on soil fertility. Their results are very important direct evidence of the activity of protozoa in * “Some Notes on Soil Protozoa,” ‘Phil. Trans.,’ B, vol. 205, pp. 77-94 (1914). 20 2 456 Protozoa in Relation to Bacterial Activity in Soil. the soil, but this does not prove that the amcebe and flagellates are functioning as the limiting factor in the sense in which that term is used by Russell and Hutchinson. Before this can be shown to be true it will be necessary to correlate protozoal activity with a decrease in the numbers of bacteria in a given soil specially inoculated with protozoa. . I have shown above (p. 446) that the presence of 10,000 amcebz per gramme of soil is not sufficient to reduce the bacterial content of a soil to the level of a similar soil containing no protozoa even though the soil be kept under conditions of moisture, etc., favourable to the trophic existence of amcebe aud flagellates. Conclusions. The results of the experiments described above lead me to the conclusion that the protozoa, including ciliates, amcebe, and flagellates, added to the soil have not been able to act as a factor limiting bacterial activity in the soil. Inferentially, therefore, the ciliates, amcebe, and flagellates obtainable from ordinary soil under cultural conditions do not function as the limiting factor. This is in accord with and extends the conclusion put forward in my earlier paper,* viz., that the ciliated protozoa are present in soil only in an encysted condition and cannot function, therefore, as the factor limiting bacterial activity. There is evidence, however, in the case where a small quantity of untreated soil is added to a partially sterilised soil that some factor comes into action which keeps down the level of the bacterial content. The results obtained, however, do not lend support to the hypothesis that it is the protozoa added in the untreated soil which have this influence. It is shown in the case of Broadbalk 1865 soil, in which an abundant protozoan fauna of amcebe and flagellates is present, and presumably active, that the numbers of bacteria maintain a high level. This soil exhibits a clear case of partial sterilisation being effected without the elimination of protozoa. It is not within the province of this paper to attempt to rebut the very weighty indirect evidence put forward by Russell and Hutchinson as to the biological character of the detrimental factor. The results obtained, however, warrant the conclusion that ciliates, amcebe, and flagellates cannot be included in that biological factor. * © Roy. Soe. Proc.,’ B, vol. 84, p. 165 (1911). 457 On the Mesodermic Origin and the Fate of the So-called Mesectoderm in Petromyzon. By 8. Harra (Sapporo, Japan). (Communicated by Prof. E. W. MacBride, F.R.S. Received December 10, 1914.) Introductory. About 20 years ago v. Kupffer (85) described in the embryos of Petromyzon an epithelial structure extending, between the ectoderm and the somatic plate of the mesoderm, from the head to the posterior boundary of the branchial region, and described it under the name of the neurodermis; subsequently, he bestowed on it the name branchiodermis. Seventeen years later the same structure was again discovered by Koltzoff (02), who identified it with the mesectoderm which was described by Miss Platt (94) in Necturus embryos. Subsequently, so far as Petromyzon is concerned, nothing was published until last year, when a paper by Schalk (13) appeared, although the corresponding layer of cells was described by A. Dohrn (02) in Selachii and by Brauer (04) in Gymnophiona. For a long time the origin and fate of the layer in question engaged my attention. Last summer I was able to re-examine my sections and to confirm observations which I had previously published in a paper entitled “ Die Bildungsweise und erste Differenzierung des Mesoderms beim Neunauge (Lampetra mitsukurw, Hatta),’ in which the origin and differentiation of the so-called mesectoderm are described and illustrated by a series of microphotographs. To my regret the paper, which was ready for press when the great war broke out, could not be sent to the editor of a certain scientific journal in Belgium, who had promised to publish it in his journal. The present note is an attempt to communicate some of the principal points of that paper which relate to the mesectoderm. The other organs dealt with in the above-mentioned paper have already been described in preliminary notes or In my previous papers. ; 1. Origin of the Mesectoderm. The previous authors who deal with mesectoderm invariably assume its ectodermic origin. The layer was so named by Platt, because, in spite of its supposed ectodermic derivation, it takes on in its further differentia- tion features like a mesodermic structure. About the mode in which the layer arises from its assumed mother-layer, positive evidence has not as yet been given by any of the authors, except Schalk, who endeavours to make 458 Mr. 8. Hatta. Mesodernic Origin and the Fate of the intelligible, by means of figures and descriptions, the precise mode in which it originates. According to v. Kupffer and Koltzoff, the mesectoderm is formed of cells liberated partly from the medullary cord, but in the main from the ectoderm forming the lateral walls of the head and of the branchial section of the body. The cells of the branchial region are regarded by them as beimg proliferated from the ganglionic placodes. These cells, whether medullary or ectodermic in origin, are classed together by the authors and designated as ectodermal, because the cells from the two sources become so much intermingled as to be indistinguishable, when once they have left their mother-layers. But the cells of the mesectoderm become differentiated, as their later history shows, in two directions: viz. into the cephalic nerves with their internal ganglia on the one hand, and into the tissues which give rise to the cartilaginous branchial basket and the connective tissue on the other. ' For this reason probably the mesectoderm has been. designated either as neurodermis or as branchiodermis, until Koltzoff (02) identified it with the corresponding structure found by Platt in Wecturus lateralis, Neither v. Kupffer nor Koltzoff were able to distinguish in the mesectoderm the nerve cells from the other elements, but each states that both the nerves and connective tissue are derived from one and the same source. Koltzoff distinguishes in the mesectoderm the dorsal division, which is found above and within the cephalic ganglia, and the ventral division, which extends below them towards the ventral surface. This attempt amounts to nothing and is founded only on histological distinction; the dorsal division is a network of the strings caused by the connection of the cells with one another by their protoplasmic processes, while the ventral division of the layer consists of a typical epithelium of columnar cells. But this histo- logical difference is a temporary one, and does not indicate, as Koltzoff believes, a differentiation of the nerve cells from the other elements of the mesectoderm. The principal point in which the results by Schalk differ from those of the authors above mentioned consists in the origin of the mesectodermal cells, which, according to him, is confined absolutely to the ectoderm; he denies positively that any of these cells have a medullary origin. He says, however, nothing definite about the nerve cells or the mesoderm somites, except that the sclerotomes give rise to the trabecule and parachordals. If I understand Schalk correctly, there are two phases in the liberation of the mesectodermal cells from the ectoderm. In embryos about ten days old selected from his material, the heads of which have just begun to be raised a So-called Mesectoderm in Petromyzon. 459 above the yolk, more or less conspicuous groups of cells are proliferated from the ectoderm, at about the level of the chorda and uninterruptedly from the eye backwards along the whole extent of the branchial tract of the enteric canal, and these cells push their way ventrally between the ectoderm and mesoderm. The cells of this first phase are, the author believes, identical. with those of the branchiodermis of Kupffer. In quite young embryos the production of these cells goes on throughout one continuous streak, but in a little more advanced stage it is concentrated in certain centres, which Schalk believes to have been detected by him and which resemble the nerve placodes of Kupffer. This concentration indicates the beginning of the second phase of cell production. In the second phase of the formation of mesectoderm which Schalk describes, each centre of cell production is found close behind each visceral pouch, except the hyomandibular, which is destitute of such a centre. The statements are illustrated by his text-figs. 16 and 17. The centres are produced by local thickenings of the ectoderm, which appear from before backwards one after another and proliferate the cells in a continuous layer, which becomes pushed backwards so as to be mixed with those of the branchiodermis, so that the cells of both lots can no longer be distinguished one from the other. But he says nothing definite as to whether the branchiodermis, or the cells directly descended from the placodes, represent the formative elements for the branchial cartilage bars. He says merely quite indefinitely: “Wenn nun auch jene kleinen Plakoden moglicherweise an der Bildung der Branchialnerven Anteil haben, so glaube ich doch behaupten zu k6nnen, dass ein Teil der aus jenen Epidermisplakoden auswandernden Zellen bei der Bildung der Kiemenknorpel verwandt werden ” (14, p. 55). Schalk is correct in so far as he excludes the medullary cells from the origin of the mesectoderm; in other respects the results given by him do not add anything to what everybody had assumed. I have observed all the occurrences which Schalk gives in his figures and found that they have nothing to do with the mesectoderm. How and when the cell proliferation of the placodes in the branchial region is closed, we cannot learn from the statements given by him at all. But his text-fig. 18 shows a great resemblance to a section from a series of frontal sections through an embryo, 12 to 13 days old, in my possession, which is just hatched or is about to break the chorion. At such a stage as this the mesectoderm is, of course, already fully established and has commenced even to be differentiated, as is clearly seen in the figure referred to; the con- tinuous epithelium is divided by the outgrowth of the visceral pouches into 460 Mr.S8. Hatta. Mesodermic Origin and the Fate of the branchiomeres, and the proximal part of each branchiomeric piece is swollen up (Schalk’s text-fig. 18, p. 6), giving rise to the cartilaginous visceral bar.* The ectoderm lying in close contact with the thickened part of the mesectodermal epithelium, which is shown in the figure by Schalk as continuous with the ectoderm, is a thickening produced by active multiplica- tion of the component cells of the ectoderm (text-fig. 1, A). This ectodermal thickening assumes an oval outline with its long axis . vertical and grows inwards (text-fig. 1, B), pressing against the rudiment of the cartilaginous visceral bar. But the conical bottom of the entodermal visceral pouch in front pushes its way laterally and backwards, and presses upon the invaginating ectodermic pouch, finally fusing with it. On the 14th day this spot becomes perforated and forms an oval slit with its long axis vertical, and the gill slit is thus established (text-fig. 1, c). The rudiment of the cartilaginous visceral bar is found close behind this orifice. Now, the thickening of the ectoderm which Schalk saw was the developing gill-cleft and had no genetical relation to the cartilaginous visceral bar. A thick section might induce one to assume continuity of the thickened ectoderm and the likewise thickened mesectoderm ; but, in reality, both the parts are separated by a sharply defined boundary line, as careful observations of thin sections prove without difficulty. In spite of his efforts, Schalk could not detect, as he says, a corresponding placode in the hyoid segment. Here, in fact, no ectodermal thickening for the gill-cleft takes place at all, because the hyomandibular pouch in front of this visceral area does not break out to the exterior, but is transformed into the velar cavity; it has no direct respiratory function, but performs an auxiliary service to it. I may be permitted to add a few words on the placodes of nervous nature in the branchial region, in order to avoid possible confusion of them with the ectodermal thickening for the gill-cleft above stated. V. Kupffer (94) was the first who described the ganglionic placodes, termed by him the epibranchial placodes. The placodes of the epibranchial ganglia are situated at the level of the dorsal edge of the lateral plates, consequently at a much higher level than that at which the ectoderm thickens for the gill-clefts, and the placode appears in each branchiomere from the vagus segment, which represents the fourth branchiomere, counting from the premandibular arch to the hindmost branchiomere behind the last or eighth visceral pouch. These placodes are, of course, purely nervous and have nothing to do with the mesectoderm ; they represent the posterior section of what I have called * The cartilaginous visceral bar in Petromyzon is not to be confounded with the true visceral arch formed in higher craniota. ec. r.be. rvca. bel. Texr-ric. 1.—A and B show frontal sections, of which B shows a little further advanced stage. C represents a | transverse section through a postotic visceral arch of a much advanced larva. The nuclei of the mesecto- derm cells and the structures derived from the mesectoderm are shaded. a.m., muscular artery ; ao., aorta ; bc., branchial cleft ; bcl., branchioccele ; ch., chorda ; ct., connective tissue ; ec., ectoderm ; eg., epibranchial ganglion ; en., entoderm ; ep., epidermis ; m., hypoglossal muscle ; ma., mesodermal visceral arch ; m.ad., adductor muscle ; mec., medullary canal; me., mesectoderm ; ph., pharynx ; 7.be., rudiment of branchial cleft ; 7.va., rudiment of vascular arch ; 7.vea., rudiment of visceral cartilaginous arch ; s.ct., subchordal connective tissue ; sm., scleromyotome ; ta., truncus arteriosus ; thy., thyroid gland ; va., visceral vascular arch ; v.¢., cardinal vein ; veb., visceral cartilaginous bar ; v.j7., vena jugularis impar ; vp., visceral pouch. in my paper above referred to (140) the ventral series of cephalic ganglia. The epibranchial placodes are not only cut off from their mother-layer, but have been already transformed into the definitive nervous system of the branchial apparatus, when the ectoderm commences to be thickened for the gill-clefts. About the fate of the foremost placode, which is situated, as seen in text-fig. 14 by Schalk, immediately behind the optic cup, the author gives no definite account. Judging from the position in which it is found, and from what he says about it, the placode must be taken to be the rudiment of the lens which belongs to the trigeminal region. Here the circumstances are not so simple as shown in the figure. The ophthalmic and trigeminal placodes and the placode for the lens have coalesced at their bases and are distinguishable from one another only by the divergence of their distal parts, and they embrace between them the second mesodermic somite and a part of its mandibular fold, being closely apposed to one another. The 462 Mr. 8. Hatta. Mesodermic Origin and the Kate of the placodes are purely nervous in nature and have no genetic relation to the mesectoderm at all, although they are at ‘certain stages in close contact with the latter. In the course of the seventh day, therefore, before the first appearance of the epibranchial placodes, the placode for the trigeminal group is constricted off from the ectoderm. Finally, the origin of the branchiodermic cells in their first stage, of which Schalk speaks, seems to be unintelligible. Judged from his text-fig. 12 and the accompanying statements, the cells are brought into their position from the ectoderm not by cell-multiplication going on in this germinal layer but simply by liberation of some of those composing the layer. If such a case as given by Schalk really occurs, it might be looked upon as a case of delamination. But,the occurrence of delamination, even for the formation of the ventral parts of the mesoderm, as W. Scott (82) and, later, Mollier (06) assume, or for the origin of the pericardium and endocardium, as asserted by Shipley (87), has been disproved, and, according to my experience, occurs in no case at least in the development of Petromyzon. In the series of sections in my possession I find no similar case to that of Schalk, except some sections frontaly cut through the lower margin of the well-established mesectoderm. According to the results obtained by myself the so-called mesectoderm is not so peculiar a structure as it appeared to the previous observers, but it is the mesoderm itself, a part of the somites or, as we may term them, the scleromyotomes. It is represented at its first appearance, as observed in the mandibular arch, by scattered free cells, which later coalesce for the most part, to form a typical epithelium. ‘In the postotic region the mesectoderm is, on the contrary, from the first epithelial. In early stages there are found only two kinds of free cells: the blood vascular cells and the mesectoderm cells. The cells of both kinds appear almost at the same stage ; at about the fifth day from the fertilisation a few vascular cells are to be observed in the space below the chorda, and between the floor of the pharynx and the ectoderm which represents in these early stages the ventral wall of the body.* On the contrary, the mesectodermic cells, the earlier traces of which are seen already to the close of the fourth day, appear as a rule between the ectoderm and the somatic plate of the mesoderm. When established the mesectoderm is confined to the head and the branchial extent of the body. What interests us is that the mesectoderm is in the postotic region nothing else than the ventro-lateral edge’of the scleromyotome which has * As to the full account on the characteristics of the blood vascular cells and on the development of the vascular system I refer to my other papers (00, 07, 14a). So-called Mesectoderm in Petromyzon. 463 grown downwards by active cell-multiplication (text-fig. 2). The growth is produced not by the rearrangement of free cells cast off, but by the out- erowth of a continuous epithelium a single cell thick, which pushes its way between the somatic layer of the lateral plates and the ectoderm until the ventral edge of this epithelium reaches the mid-ventral line of the thyroid groove. The epithelium thus produced is what is called the mesectoderm. The downward growth of the mesectoderm can readily be traced step by step. On the fifth day, where the growth of the layer begins, the ventro- lateral edge, for instance, of the fifth scleromyotome* is produced a short distance downwards and is wedge-shaped in cross-section, The cutis layer of the scleromyotome passes over into the muscle plate round the apex of the wedge. On the sixth day the mesectoderm in its anterior portion is so broad that its lower margin is found as low as the thyroid groove, while farther backwards the layer is narrowed so that it is represented in the seventh scleromyotome by a short wedge-shaped process of the latter. It is only in the course of the eighth day that the mesectoderm is fully estab- lished in the posterior branchial region. The formation of the mesectoderm is, therefore, commenced in the anterior region and goes on backwards (text-fig. 3). In its early stages of formation the mesectoderm is an epithelium composed of flattened cells; but it thickens gradually as its component cells assume a tall columnar character which is, doubtless, brought about by their mutual pressure resulting from repeated cell-multiplication within the layer. On a cross-section through a visceral arch six layers of epithelium are now seen: the innermost is the entodermal pharynx wall; the next outer feeble layer represents the first rudiments of the vascular arch; then follow the splanchnic and somatic layers of the lateral plates representing the meso- dermic visceral arch; and between the latter layer and the outermost layer, the ectoderm, intervenes an intensely stained epithelium of tall columnar cells, which represent the mesectoderm. The mesectoderm is gradually diminished in thickness from the level of its middle height downwards into its sharp-edged lower margin. Although the mesectoderm is only a single cell thick in most parts, at the dorsal edge, where it is in connection with the ventro-lateral edge of the scleromyotome, it is divided into two layers which pass over into the cutis layer and the muscle plate layer of the scleromyotome respectively. The mesectoderm can, therefore, be regarded as a fold arising from the above- * The anterior three mesodermic somites found in front of the auditory vesicle are not transformed into typical scleromyotomes ; nevertheless they are reckoned as such. The fifth scleromyotome is accordingly the second postotic. 464 Mr. 8S. Hatta. Mesodermic Origin and the Fate of the mentioned edge of the scleromyotome, and wedging itself in between the meso- and ectoderm. The manner in which the mesectoderm develops reminds us of the folding NNN, TEXT-FIG. 2. TEXT-FIG. 3. TExt-Fic. 2.—Diagrammatic representation of a transverse section through a post- branchial visceral arch, showing down-growth of the cutis-layer of scleromyotome to give rise to mesectoderm (left) and the established mesectoderm (right). The mesectoderm and its rudiments are shaded. ct., cutis layer ; sch., subchordal cells ; thy.g., thyroid groove. For other letters, see the explanation of the previous diagram. TEextT-FiG. 3.—Diagrammatic representation of a frontal section showing the position of the mesectoderm in relation to other layers. The posterior visceral pouches are in formation, and the lateral plates are being cut into the mesodermic visceral arches. The mesectoderm is shaded. 6., brain ; bcl., branchioccele ; ep., epidermis ; /., placode of lens; ma., mesodermal visceral arch; me., mesectoderm ; oc., optic cup; pt., pronephric tubule; v.va., rudiment of vascular arch; se., splanchnoccele ; spp., splanchnopleure ; sp., somatopleure ; s¢., stomodzum ; vp., visceral pouch. of the corresponding edge of the scleromyotomes in the postbranchial region, by means of which the scleromyotomes grow downwards so as to provide the ventral somatic walls of the body with the muscular and the cutis layers. The downward growth of the scleromyotomes in both cases is, I venture to i" 4 So-called Mesectoderm in Petromyzon. 465 assume, analogous, and the two structures thus brought about are serially homologous. The difference consists in that the mesectoderm is destitute of the muscle plate layer. Though to a very small extent, the mesectoderm is divided into two layers, and there can be no objection to our regarding the inner of these as homologous with the muscle plate. The causes of this modification in the branchial region are to be sought in the changes in that region produced by the respiratory mechanism and its skeletal framework. As will be stated later on, the mesectoderm is converted to a great extent into the cartilaginous branchial bar, which is a special skeletal arrangement for the respiratory mechanism. The remainder of the layer supplies the elements for the subcutaneous tissue, while, in the postbranchial region, the whole of the cutis layer is employed in the forma- tion of this tissue. The demand for the formative elements of this tissue has caused, as | believe, the cutis layer of the scleromyotomes in the branchial region to be developed so vigorously as to call the mesectoderm into existence, although the muscle plate in the branchial region is almost entirely sup- pressed. The consequence of this suppression is that the somatic walls of the branchial chamber are destitute of the segmental muscles, and have to fill this deficiency by the so-called hypoglossal muscles which with their cutis layers undergo an exceedingly modified mode of development,* as was pointed out by Neal(97) and was confirmed by Koltzoff(02),+ and by myself (14a, 140). The great modifications met with in the prootic section of the head cause the mesoderm to be modified to a still greater extent than in the postotic branchial region. Accordingly some peculiarities occur in the formation of the mesectoderm in this region. In this section of the head there are formed three mesodermic somites, the third being situated just in front of the auditory vesicle. Of these three somites the hindermost shows a structure very similar to a postotic somite ; the middle somite, under which the lateral platest develop into the enormous mesodermic mandibular arch, is represented by a narrow epithelial fold of the archenteric roof, which ascends along the lateral wall of the medullary canal and strikes with its distal extremity against the trigeminal placode of the * The hypoglossal muscles are produced by the forward bending and shifting of the ventral part of some postbranchial scleromyotomes. They are the only segmental muscles in the somatic walls of the branchial chamber. + As Koltzoff (02) remarks, Goette (90) gives incorrectly four somites in front of the auditory vesicle. { Koltzoff is the only author who gives a detailed account of his second somite. But he failed to detect the free somite, which is very small, whilst he regards the dorsal part of the colossal lateral plates as their somitic part. 466 Mr. 8S. Hatta. Mesodernuc Origin and the Fate of the ectoderm ; whilst the first somite is represented by the anterior blind end* of the epithelial archenteron, which is folded off from the rest as a median unpaired pocket. This somite is so small that it is practically destitute of the lateral plates. In the two posterior of the three somites the outer wall, which corresponds to the cutis layer of a postotic somite, gives off free cells which push their way between the ectoderm and the somitic layer of the lateral plates (text- fig. 4). The free cells are quickly increased to some extent by the cells Trext-Fic. 4.—Diagrammatic representation of a transverse section through the second somite, showing the proliferation of the mesectoderm cells from the lateral layer of the somite. me.c., mesectoderm cells (shaded) ; ne., nerve cells; sm.2, second somite standing still in connection with the pharynx ; ¢., placode for trigeminal ganglion. For other letters, see the explanation of the previous diagrams, budded out from the lateral layer of the somitic fold. The somites are cut off from the archenteric roof only on the fifth day, while the cell-proliferation begins at the early part of the fourth day, therefore earlier than the stage at which the ventro-lateral edge of the postotic scleromyotomes begins to be produced into the mesectoderm. The free cells soon make up a thick columnar epithelium which represents the mesectoderm of this preotic region. In contrast to that in the postotic region, the mesectoderm is here not confined to the lateral part, but is spread into the ventral wall of the body, which gives rise to the stomodzeum by invagination. At the same time, but on a smaller scale, the cells wander out of the lateral layer of the somites into the space between the medullary canal and the ectoderm opposite and above the ganglion placodes. They do not assume * The first somite may not be confounded with the anterior blind sac of the pharynx of Kupffer ; for the explanation of both the structures I refer to my paper above given (146). POL Ie : So-called Mesectoderm im Petromyzon. 467 _an epithelial character, but make up a simple network of cells with the nervous cells coming forth from the nerve ridge, by which the epidermic ectoderm is connected with the walls of the medullary canal. This network corresponds to that part of the mesectoderm which Koltzoff distinguishes as the dorsal division from the epithelial ventral part of it. The dorsal division . is accordingly confined to the preotic region, while the ventral division is to be traced uninterruptedly to the corresponding part in the postotic branchial region and constitutes one continuous structure from the mandibular arch to the hindmost visceral arch. The network forming the dorsal division becomes gradually more complicated, owing to further growth of both the nervous and mesectodermic cell-strings; so that the elements of both kinds are not easily distinguished from each other, as Koltzoff complains. The cells of the dorsal division ought not, therefore, to be overlooked at the earlier stage of their appearance, a” phase in which the nervous cells coming downwards from the medullary root and the mesectoderm cells arising from the lateral layer of the somites do not as yet meet with each other. Then, in the following stages, both kinds of cells are not very difficult to trace into the points from which they start respec- tively. At the dorsal corner of the lateral layer of the somites active cell- divisions can be observed which are repeated during the course of develop- ment for not less than 12 hours, and the course taken by the resulting cells in passing into the network is not difficult to make out. The nerve-cell strings passing downwards almost vertically can also be traced with certainty. The nerve fibres developed from these strings associate with those from the ganglion of epidermic origin and make their way between the mesectoderm and somitic layer of the lateral plates, as was obvious already in an embryo of the eighth day. The foremost somite gives off the mesectoderm cells not only from its outer wall, as in the following somites, but also from the anterior wall of the blind pocket by which the first somite on each side stands in connection with its fellow. The cells from both sources fill up the, space between the ectoderm and the somite and the space below the anterior extremity of the brain. The free cells lying in close contact with the ectoderm are trans- formed into the epithelial mesectoderm, which can by no means be distin- guished from the ventral division in the posterior region and is continuous to it; those inside are developed into the network of muscle fibres which occupies the interior of the upper lip in later stages. | While the outer walls of the first, second, and third somites and also the anterior wall of the first are broken up into the mesectoderm cells, the inner walls of these three somites, which correspond to the muscle plate of the 468 Mr. 8S. Hatta. Mesodermic Origin and the Fate of the postotic somites, give rise to the six muscles of the eye,* which are distinctly differentiated already at the close of the ninth day or at the commencement of the 10th day. The preotic section of the head is consequently totally destitute of the segmental muscles derived from the somites. This deficiency is, aS pointed out by Kupffer, made good by a few myotomes behind the auditory vesicle, the dorsal parts of which are bent and shifted forwards into the head. The mesectoderm is, therefore, the product of that part of the mesodermic somites which corresponds to the cutis layer. The dorsal division, which is distinguished by Koltzoff from the ventral division, is derived from the three preotic somites and confined to the preotic section of the head; it never assumes an epithelial structure, but'makes a simple network with the nerve- cell strings. On the contrary, the ventral division is a continuous layer of typical epithelium, extending from the anterior margin of the lateral plates to the posterior boundary of the branchial region and as high as the lateral plates. This division of the mesectoderm is cut into nine vertical epithelial bands, when the lateral plates are divided into nine visceral arches. That part of the mesectoderm which assumes the epithelial character in the snout may be regarded as the anterior continuation of the ventral division. If this assumption is correct, all the three preotic somites, just like the following somite in the postotic branchial region, contribute elements to the ventral division of the mesectoderm. 2. Fate of the Mesectoderm. While the nerve-cell strings in the pre-and postotic region are transformed into the nerves and ganglia, the mesectodermic elements of the dorsal division are converted mainly into the connective tissues standing in relation to the nerves and ganglia ; a small portion of them, which hes in contact with the medullary walls, gives rise to the most anterior section of the cranial skeleton, ze, the trabecule. The nine mesectodermic bands, into which the continuous layer of the ventral division is divided by the visceral invagination of the pharynx-walls, undergo the following differentiation. On frontal sections through a just- hatched larva, the first stage of the differentiation is very obvious. At the level of the visceral pouch, the entodermal wall of the pharynx is in close contact with the ectoderm, while at the level of the visceral arch, between the two layers, are contained the vascular cells, the mesodermic arch and the * Detailed accounts on the development of the ophthalmic muscles I have given in my above-mentioned paper (146). So-called Mesectoderm in Petromyzon. 469 mesectoderm. The proximal half of the last-named epithelium is thickened so as to be, raised inwards into a ridge, which shows on cross-sections a pyramidal outline and represents the first rudiment of the cartilaginous visceral arch. This stage of differentiation is represented in text-fig. 18 by Schalk, which is correct, except the connection of the mesectoderm with the: ectoderm. The cells forming the rudiment of the cartilage acquire a radial arrange- ment (text-fic. 1, A) and are soon constricted off from the remainder of the layer; this stage is followed by a stage in which they are wedged in between one another, so as to form one row (B). -On cross-sections through a larva of the 14th day this rudiment of the cartilage looks like a bar consisting of piled-up dises, in which three sections are distinguishable (C): a dorsal and a ventral section curved outwards and the middle section bowed inwards. While the ventral section touches with its distal extremity the lateral division of the vena jugularis impar, the proximal end of the dorsal section lies in the corner between the dorsal aorta and the chorda and under the anterior cardinal vein. In the course of the 15th day, the aorta together with the roof of the pharynx is separated from the chorda by enormous development of the reticular subchordal connective tissue. Accordingly the dorsal end of the rudiment of cartilage is also brought downwards, so as to be forced into the corner between the aorta and the roof of the pharynx which has been pressed down. It is interesting that the band of connective tissue which before and after this change connects the rudiment of cartilage with the chorda is drawn out into a string stretched between both the structures. In sharp contrast to other visceral arches, the hyoid arch does not undergo this dislocation of the cartilage bar, which is, on the contrary, shifted by stages a little upwards, and the cardinal vein passes into the mandibular vein under the cartilage bar. This peculiar feature is the first step towards the fulfilment of the function which the hyoid arch has secondarily acquired ; it enters into the formation of the primordial skull, leaving the service of respiration. The differentiation of the mesectoderm into the visceral arch takes place at first in the visceral arch behind the hyoid arch and proceeds backwards to the following arches, which undergo the same process one after another. For a long while the mesectoderm in the hyoid arch is not cut off from that in the mandibular arch. And it delays its differentiation into the rudiment of the cartilaginous hyoid arch, which is, however, obvious before the same process commences its work in the hindmost visceral arch. The rudiment of the cartilaginous visceral arch shifts inwards, when it is detached from the remainder of the mesectoderm, and presses and separates VOL. LXXXVIII.—B. 2 P 470 Mr. 8. Hatta. Mesodermic Origin and the Fate of the at last the lateral plates, the mesodermic visceral arch, into the inner adductor and the outer constrictor muscles, with which the cartilaginous arch is invested. The remainder of the mesectoderm constricted off from the rudiment of the cartilaginous visceral arch is stretched at the same time to the outside of the rudiment of arch so as to line the whole inner surface of the ectoderm ventral to the chorda level, and assumes the characteristic feature of the subcutaneous tissues which underlie the ectoderm, except the ventral part for the hypoglossal muscles. This differentiation of the mesectoderm is very obvious in larvee of the 13th to the 14th day. The mandibular arch, which we may now consider, is characterised particularly by the enormous mesodermic arch which it contains and which is folded upon itself, thrust inwards by the invaginating stomodzum, so that four layers of the folded lateral plates are obvious on a cross-section through this arch. While the ventral edge of the folded mesodermic arch, by which the somatic layer passes over into the splanchnic plate, is separated by the bottom of the stomodzeum from its counterpart on the opposite side, the dorsal edge, by which the two layers of the lateral plates also run into one another, 1s divided only by the carotid artery from the chorda. The ventral division of the mesectoderm, following this folding of the mesoderm, is brought into the same topographical relations to the stomodeum and to the chorda. The dorsal edge of this division of the mesectoderm is brought into contact with the lateral wall of the chorda, above the carotid artery, as is very clearly seen on the 9th to 10th day. On the 13th day the cells composing this edge are concentrated into a characteristic compact mass which-is soon constricted off from the remainder of the layer. This compact cell mass constitutes the earliest traces of what are known since Sewertzoff (97) as the anterior parachordals. The rudiment of the anterior parachordal is on cross-sections wedge-shaped and looks as if produced from the lateral wall of the chorda. On the 14th day the rudiment of the anterior parachordal can be traced as far as the branching of the facial artery from the carotid, which marks the boundary between the first and second somites, and it ends backwards rather suddenly in front of the roots of the vascular mandibular arches and of the earotids. The parachordal rudiment is most prominent at a little distance from the root of the vascular mandibular arch and grows gradually lower toward the snout, while it is decreased suddenly in height backwards. The parachordal rudiment cannot, therefore, be distinguished genetically from the rudiment of a cartilaginous visceral arch; both the structures are, So-called Mesectoderm in Petromyzon. 471 I believe, serially homologous with one another. The prominent point of the parachordal rudiment develops into a transverse bar of cartilage, which is, I assume, the rudimentary remnant of the equivalent of its corresponding visceral cartilage bar. This bar is,in the mandibular arch, reduced to its last remnant, because it has been shut off from the respiratory mechanism. The further fate of the cartilage bar interests us in developing into that important element of the primordial skull, which is known since Parker (83) as the palato-quadrate; the redevelopment of this rudimentary remnant into so conspicuous an element of the cranial skeleton is due to nothing but the law of “ Funktionswechsel” first enunciated by Dohrn. The remainder of the mesectoderm detached from the rudiment of the anterior parachordal is, in the mandibular arch as elsewhere, converted into the subcutaneous tissues, which develop in distinction to those in other arches not only beneath the ectoderm, the skin of the cheek, but also beneath the epidermis of the stomodzum, the cover of the mouth cavity. As Gaup (06) remarks, the single origin of the parachordal, which Koltzoff (02) maintains and Schalk (13) confirms, is incorrect. On the contrary, the posterior parachordal of Sewertzoff (97) is represented in reality by the medial horizontal process of the cartilaginous hyoid arch itself, and the transverse process, which, according to Sewertzoff, is very similar to an ordinary visceral bar in the following arches, is nothing more than the visceral bar itself in the hyoid arch. The anterior and posterior parachordals are separated for a long time by interposition of the large auditory capsule. Both the rudiments grow respectively backwards and forwards to meet and be fused together with each other only in a larva about thirty days old, in which the auditory capsule retreats and is detached from the chorda. But for a long time the transition is obvious, because both the rudiments are decreased in thickness towards their point of meeting. The trabecula is formed in front of the root of the facial artery and the rudimentary vascular arch in the premandibular segment; its centre lies close to this vascular root. From this centre it grows anteriorly along the basal wall of the brain and over the posterior cerebral artery, which isthe anterior prolongation of the carotid artery. The more it is prolonged, the more it diverges hand in hand with the artery from the median line, so that it lies opposite the optic cup rather on the lateral wall of the brain and embraces, with its counterpart on the opposite side from right and left, its infundibulum, and is finally lost on the lateral wall of the latter. The mesectoderm cells giving rise to the centre of the trabecula are doubtless those derived from the first somite, and are marked off from those 472 Mr. 8. Hatta. Mesodermice Origin and the Fate of the of the second somite by the facial artery, which appears much earlier than the mesectoderm. But the forward growth of the anterior parachordal seems to be carried on largely at the cost of the dorsal division of the mesectoderm from the second somite, viz., the cells which lie close to the chorda, and to be continued uninterruptedly to the formation of the trabecula in front; so that the trabecula is practically not formed separately from the anterior parachordal, as was believed by previous observers, but as the prolongation of the latter. The trabecula has, however, a special centre for itself, marked by a slight thickening in the rudiment. It is, however, still an open question, whether the trabecula is to be put in the series of the visceral arches or not. But it is obvious that the rudiment of the trabecula is genetically identical with the latter, because it is formed of material identical with that giving rise to the visceral arches and because it comes forth in a metamere identical with that which has a cartilaginous visceral arch for itself. To avoid misconception, a few words may be said about the relation of the mesectoderm to the metamery of the body. The epithelial bands into which the ventral division of the mesectoderm is divided are in numerical as well as topographical correspondence with the mesodermic visceral arches, that is to say, branchiomeric. The cartilaginous visceral arches are derived from these branchiomeric bands, also branchiomeric in arrangement. The mesodermic visceral arches arise from two unsegmented continuous layers, the lateral plates, which are divided by nothing but evagination of the visceral pouches, which is independent of the process dividing the mesoderm into the somites. This metameric repetition in the entodermal pouches has, therefore, no direct relation to the mesodermic somites at all, which represent the primary metamery of the body. In spite of its being the product of the segmented mesoderm, the mesectoderm is also an unsegmented continuous layer, until it is divided by the visceral invagination into branchiomeric bands. The branchiomery in this part of the meso- and mesectoderm, which is brought about in a passive way, 1s, therefore, not of the same value as the body-metamery. The branchiomery is furthermore in segmental accordance with the ventral series of tle ectodermic placodes for the ganglia.* But this series of ganglia is never in direct segmental connection with the mesodermic somites. Still another organic system, which is in segmental accordance with the * This series of ganglia consists of the facial and glossopharyngeal ganglia and of the series of the epibranchial ganglia. The placode for the lens may be put in this series, for it can by no means be distinguished from the ganglionic placodes, so far as its origin and mode of development are concerned. 1 . So-called Mesectoderm in Petromyzon. 473 visceral arches, is the vascular. In the first formed vascular system, in which the arterial and venous systems are not yet differentiated, we have before us a vascular system of the Annelid type (Hatta (07), Keiser (14) ), which consists of a dorsal and a ventral longitudinal vessel connected with each other by vascular rings which are repeated in each body somite (Hatta (07), (14a), Keiser (14) ). The anterior division of this ring vessel system is repre- sented by the vascular visceral arches (Goette (90) ), and is: followed by Mayer’s Quergefiisse in the pronephric region, while in the region still further posterior the segmental character of this series of vessels is obscure, because in this region there is no organ of segmental arrangement standing in direct relation with the vessels. The origin of the ring vessels, whether entodermic (Goette (90) ) ormesodermic (Hatta (14a), Keiser(14) ), 1s by no means mesomeric, because the vascular cells are derived, even in the strictly segmented branchial region, from an organ which is not segmented in accordance with the mesodermic somites.* The pronephros is the only organ in which the vasomery, 7.c. branchiomery, is connected with the mesomery. But there is an incontestable fact which shows that the segmental repetition of Mayer’s vessels in the pronephric region is not primary but secondary. The first pronephric artery, which is evident in a larva of the 10th day, corresponds with the 11th vasomere (counting the premandibular vascular arch as the first vasomere), and the first pronephric tubule with which the artery stands in relation, is the product of the 7th mesodermic somitet (Hatta (97), (00), (14a), (140)), while this somite is no longer found over the first pronephric tubule, when the tubule is cut off from it. There are two movements by which the segmental discordance between the nephromeres and the myomeres is brought about, viz., the pronephros is gradually pushed backwards by the outgrowth of the visceral pouches ; the somites, ze. myotomes, on the contrary, move forwards after their formation, so that a few anterior of them are pushed over the preotic section of the head to give rise to the capitis muscles. The ring vessels, i.e. Mayer’s vessels, occur in the visceral arches as well as in the pronephros only when the visceral pouches are formed and the backward movement of the pronephros has already been carried out. The vasomeres are thus put secondarily in relation to the pronephros, although the latter * According to Goette, the vascular cells (in branchial region) are derived from the enteric wall ; according to my results, which are confirmed by Keiser, the cells are derived from the mesodermic visceral arches. + In my papers formerly published (97, 00) the first pronephric tubule is regarded as the product of the fourth scleromyotome, which corresponds to the seventh, when the three preotic somites are counted in. Ye PD 474 Mr. 8. Hatta. Mesodermic Origin and the Fate of the is the descendant of the somites. In short, the branchiomery and mesomery are independent of each other. The cartilaginous visceral arches and their equivalents in ‘tha primordial skull are branchiomeric organs, and have no genetic relation to the mesomeres. Therefore, it seems extremely probable that the participation of the ~sclerotomes in the branchial elements and in their equivalents, which is maintained by Koltzoff, Schalk, and others, is, if such actually occurs, nothing but accidental. I wish to express my warmest thanks to Prof. MacBride, of the Imperial College of Science, London, and to Dr. Gadow, Prof. Gardiner, Mr. Doncaster, and Prof. Punnett, of Cambridge University, for the courtesy which they showed me during my stay in Cambridge, and especially in according to me the privilege of working there. LITERATURE. 80. Balfour, F., ‘‘ A Treatise on Comparative Embryology,” London, 1880-1881. 04. Brauer, A., “ Beitrage zur Kenntnis der Entwicklung und Anatomie der Gymmo- phionen. IV.—Die Entwicklung der beiden Trigeminusganglien” ‘ Zool. Jahr., Suppl. Bd. 7. 02. Dohrn, A., “Studien zur Urgeschichte des Wirbeltierkérpers. XXII.—Weitere Beitrage zur Beurteilung der Occipitalregion und der Ganglienleiste der Selachier,” ‘Mitteil. Zool. Stat. Neapel.,’ vol. 15. 06. Gaup, E., “ Die Entwicklung des Kopfskelettes,” ‘Hertwig’s Handb. vergl. exper. Entw.-lehre d. Wirbelt.,’ vol. 3. 90. Goette, A., “ Entwicklungsgeschichte des Flussneunauge (Petromyzon fluviatilis),” ‘Abh. entw. gesch. d. Tiere,’ Part 5. 91. Hatta, 8., “On the Formation of Germinal Layers in Petromyzon,” ‘ Journ. Coll. Sci. Imp. Univ. Tokyo,’ vol. 5 92. —— “Contributions to the Morphology of Cyclostomata. I.—On the Formation of the Heart in Petromyzon,” Lbid., vol. 8. 97. —— “Preliminary Note on the Development of the Pronephros in Petromyzon,” ‘Ann. Zool. Jap.,’ vol. 1. 00. —— “Contributions to the Morphology of Cyclostomata. IJ. On the Development of the Pronephros and Segmental Duct in Petromyzon,” ‘Journ. Coll. Sci. Imp. Univ. Tokyo,’ vol. 13. 01. —— “On the Relation of the Metameric Segmentation of Mesoblast in Petromyzon to that in Amphioxus and in the Higher Craniota,” ‘ Ann. Zool. Jap.,’ vol. 5. 07. —— “ Bemerkungen iiber die friiheren Entwicklungsstudien des Gefadssystems bei Ammocoeten,” ‘Journ. Agric. Coll., Imp. Univ., Sapporo,’ vol. 4. 14a. “Ueber die Entwicklung des Gefiissystems beim Neunauge (Lampetra mitsukurit, Hatta).” 14d. “Die Bildungsweise und die erste Differenzierung des Mesoderms beim Neunauge (Lompetra mitsukurti, Hatta).” 14. Keiser, W., “Untersuchungen iiber die erste Anlage des Herzens, der beiden Langsstaimme und des Blutes bei Embryonen von Petromyzon planeri,” ‘Jen. Zeitschr. Natur.,’ vol. 51. So-called Mesectoderm in Petromyzon. 475 13. Schalk, Alban, “ Die Entwicklung des Cranial- und Visceralskeletts von Petromyzon fluviatilis,” ‘Arch. Mikr. Anat.,’ vol. 83. 99. Koltzoff, N. K., “Metamerie des Kopfes von Petromyzon planeri,” ‘ Anat. Anz.,’ vol. 16. 02. —— “Entwicklungsgeschichte des Kopfes von Petromyzon planeri,” ‘ Bull. Soc. Imp. Natural.,’ Moscow, vol. 15. ‘ 90. v. Kupffer, C., “ Die Entwicklung von Petromyzon planeri,” ‘ Arch. Mikros. Anat.,’ vol. 35. 94. —— “Die Entwicklung des Kopfes von Ammocoetes,” ‘Stud. vergl. Entw. gesch. des Kopfes d. Cranioten,’ Part 2. 95. —— “Die Entwicklung der Kopfnerven von Ammocoetes Planeri,” /bid., Part 3. 85. “Ueber die Entwicklung des Kiemenskelettes von Ammocoetes und die organogene Bestimmung des Ektoderms,” ‘Verh. Anat. Ges., 9. Vers. in Basel.’ 06. Mollier, &., “Die erste Entstehung der Gefiisse und des Blutes bei Wirbeltieren,” ‘ Hertwig’s Handb. vergl. exp. Entw.-lehre d. Wirbelt.,’ vol. 1. 97. Neal, H. V., “The Development of the Hypoglossus Musculature,in Petromyzon and Squalus,” ‘ Anat. Anz.,’ vol. 13. 83. Parker, W. K., “On the Skeleton of the Marsipobranch Fishes,” ‘ Phil. Trans., London. 94. Platt, Julia B. “Ontogenetische Differenzierung des Ektoderms in Necturus,’ ‘Arch. mikr. Anat.,’ vol. 43. 98. -—- “The Development of the Cartilaginous Skull and of the Branchial and Hypoglossal Musculature in Necturus,” ‘ Morphol. Jahrb.,’ vol. 25. 93. —— ‘“ Ectodermic Origin of the Cartilages of the Head,” ‘ Anat. Anz.,’ vol. 8. 82. Scott, W. B., “ Beitrage zur Entwicklungsgeschichte der Petromyzonten,” ‘ Morphol. Jahrb.,’ vol. 7. . Sewertzoff, A., “ Die Entwicklung der Occipitalregion der niederen Vertebraten in Zusammenhang mit der Frage tiber die Metamerie des Kopfes,” ‘Bull. Soc. Imp. Natural.,’ Moscow, vol. 2. — “Beitrag zur Entwicklungsgeschichte des Wirbeltierschidels. Vorliiufige Mitteil.,” ‘Anat. Anz.,’ vol, 13. . Shipley, A., “On some Points in the Development of Petromyzon fluviatilis,’ ‘Quart. Journ. Mier. Se.,’ vol. 27. ’ iS) Po 476 On the Variation mm the Growth of Mammalhan Tissue in Vitro according to the Age of the Anwmal. By ALBEert J, WAuTOoN, M.S., F.R.C.S., M.B., L.R.C.P., B.Se. (Communicated by Prof. W. Bulloch, F.R.S. Received December 8, 1914.) (From the Bacteriological Department of the London Hospital.) [Puate 15.] In a previous communication* it was shown that there was considerable variation in the value, as a culture medium, of plasmata taken from different animals of the same species ; that these plasmata did not vary as to whether they were homogenous or autogenous, but that some plasmata were good media and some were bad. During this investigation certain evidence arose that this difference might in part be due to the age of the animal. In the present investigation a series of experiments was carried out to show what was the effect, if any, of the age of the animal upon the plasma as a culture medium, and upon the tissues as regards the power of growth. Carrel, Burrows, Harrison, and Ingebrigtsen have shown in several papers that embryonic tissue tends to grow more rapidly and more vigorously than adult tissue. There appears to have been, however, no work conducted on the characters of the plasma, although it has been frequently assumed that the plasma of the young or embryonic animals makes a more suitable medium than that of adults ; nevertheless it was permissible to believe that the reverse might in fact be true, and that the plasma of young animals is a less suitable medium. It would appear important that this point should be settled, that thereby evidence might be gained as to the controlling influences on the growth of young tissue i vivo. Technique. The following experiments were carried out entirely with the tissues of rabbits. As far as possible animals were used that had been bred in the laboratory, so that the exact age was known. This was the case with all the young animals. In certain cases, however, adult rabbits were bought of unknown age, but in such cases they were all fully grown and therefore could be used as adult animals. As far as could be judged, they were all over a year old. The technique of Carrel was rigorously adhered to, the tissues being grown in pure plasma so that the characters of the growth might be * ©Roy. Soc. Proc.,’ B, vol. 87, p. 452 (1914). Growth of Mammalian Tissue in Vitro. 477 unaffected by the presence of any stimulating or inhibiting substance. In the majority of cases two tissues were used, so as to lessen, as far as possible, any experimental errors. A few cultures were made of the spleen, but most of the experiments were carried out with thyroid and liver. These tissues as a rule grow well, and the growth is not obscured by the emigration of cells, © as so often happens when spleen is used. Young testicular tissue was not cultivated in the present series of experi- ments, for it was considered that, as this tissue only fully develops later in life, false conclusions might be arrived at if the immature testicle of the young rabbit was used. In the majority of cases fresh tissue was made use of, and in this case cross experiments were generally performed, the tissue of young and old animals being grown in plasmata of both animals. A certain number of experiments were conducted with stock cultures of adult testicle which had been growing for ten generations in a medium of plasma and tissue extract. (1) Cultivations of Splenic Tissue— EHaperiment 1.—An adult rabbit two years old was anesthetised. The fur on the ventral surface of the body was removed, the skin sterilised, the carotid artery exposed, and the blood collected in sterile paraffined tubes placed in ice. The blood was then collected from a young animal 10 days old and placed in ice-cold paraffined tubes. Both bloods were centrifugalised. The spleen was removed from the young and old animals and placed in Ringer’s fluid. Four cultures were made of each spleen in each plasma, so that there were four groups. At the end of 12 hours there was good emigration of round cells in all the preparations, but it was more marked in the case of the spleen of the young animals both in the young and old plasmata. At the end of 48 hours there was a well marked growth of retiform tissue, which formed mosaic-like masses in the case of the young spleen in the old plasma, but was present in a less marked degree in the case of the old spleen in the old plasma. In both cases such growth was apparently absent when the young plasma was used. ‘Owing to the amount of round cell emigration it was difficult to estimate accurately the extent of the growth, and hence experiments with this tissue were discontinued. (2) Cultivation of Thyroid and Liver Tissues— Eleven experiments, comprising 282 cultivations, were carried out in this group. In all cases both the thyroid and liver were cultivated at the same time. By this means experimental errors were less likely, for if the results 478 Mr. A. J. Walton. On the Variation in the were due to such errors they would be less likely to occur in both groups. The ages of the young animals varied from two days to two weeks, and during this limited period there seemed to be little, if any, variation in the nature of the tissues and plasmata as regards the capacity for growth. One experiment will be described in detail, the others being carried out on precisely the same lines. Experiment 2.—An adult rabbit over a year old was anesthetised. The fur on the ventral surface of the body was removed, the skin sterilised, the carotid artery laid bare, and the blood collected in sterile paraffined tubes kept in ice. This animal was kept anzsthetised. Blood was collected from a young animal five days old and also placed in iced paraffined tubes. Its thyroid and a piece of liver were removed and placed in sterile Ringer’s fluid. The young animal was killed. Similar tissues were removed from the old animal. The blood was centrifugalised. Cultures of the young animal were made in both plasmata, as were also those of the old animal, six cultures being made in each group. The nature of the growth was observed and at certain periods specimens were fixed and stained. After 48 hours the thyroid of the young animal showed marked growth in the old plasma, but that of the old animal in the old plasma, although showing considerable growth both of the con- nective and parenchymatous types of cell, was definitely less than that of the young animal. On the other hand, both specimens in the young plasma showed either no growth at all or only a few cells growing from the edge of the tissue. Similar results were obtained in the case of the liver. The above results are well shown by the following Table :— Young plasma. Moderate growth Very slight growth. | Old plasma. | | | Very good growth Slight growth, The above experiment was performed again 10 times. The old animals were in all cases fully developed, and the majority were known to be over one year of age. The young animals varied in age between two days and two weeks. The cultures were made in groups of four or six in each media and'were prepared under identical conditions. By taking groups of four or six specimens it was possible to estimate more accurately the changes in growth, for in the primary cultures it is unusual, excepting perhaps in the case of the testicle, to find that all the specimens have grown to an identical extent. Moreover, the percentage of successful growths in a series is in Growth of Mammalian Tissue in Vitro. 479 itself evidence of value of the suitability of the media and the power of the tissue to grow. The results of these experiments may be summarised as follows :— (a) Young Tissues Growing in Old Plasma.—It was found that growth was in all cases extensive and successful. Cultivations occurred in 100 per cent. of the cases. With the thyroid, masses of cuboidal cells were seen extending into the plasma, and between them were large numbers of branching connective tissue cells. In the case of the liver, the growth was also extensive, and after 48 hours large masses of the characteristic deeply staining cuboidal cells were visible together with large numbers of connective tissue cells (fig. 1). Successful results were again obtained in 100 per cent. of the cases. In both groups the growth continued for four or five days before any signs of degeneration occurred. (b) Young Tissues Growing in Young Plasma.—W ith this group there was a very marked difference. There is considerable difficulty in obtaining the blood of these young animals, owing to their small size, and it was at first thought possible that the difference in growth might be due to the fact that the plasma so obtained was not in good condition. It was found, however, that the results were practically constant even after these difficulties had been overcome. In the case of the thyroid only 8 per cent. of the specimens showed any growth, and they were thus sharply differentiated from the specimens of the same tissue growing in old plasma. Even in those cases in which the growth was present it was slight in amount. In no case were any cuboidal cells seen, and even after three or four days there were only present a few connective cells growing from scattered areas at the edge of the tissue. This result was constant apart from liquefaction of the plasma, which in other cases has been found to limit growth. That is to say, the decrease in growth did not appear to be due to mechanical causes. With the /iver similar results were obtained. In tis case a larger number of the preparations showed growth, the results being positive in 26 per cent. of the cases. In one specimen (Experiment 5) there was very good growth, but in three other specimens of the same series there was no growth, and such a result did not occur again in the other series. It is probable, there- fore, that this result was an experimental error. In the remaining specimens which showed growth the extent of the growth was slight. In some cases a few outgrowths of cuboidal deeply staining cells were seen, but in no case were they so extensive as when old plasma was used as a medium. 480 Mr. A. J. Walton. On the Variation in the In the majority of cases only a few branching connective tissue cells were visible, and they extended only for a short distance into the surrounding plasma, the difference between the amount of connective tissue growth in young and old plasma being very marked (fig. 2). (c) Old Tissues Growing in Old Plasma.—tIn this group there was a moderate amount of growth, about 80 per cent. of the cultivations being successful. In every experiment some of the specimens showed good growth. On comparison, however, it was always seen that this growth was less than when young tissues were used. In the case of the thyroid a few cuboidal cells were seen growing from the edge of the tissue after about 48 hours, growth being present in 70 per cent. of the specimens. After a further 24 hours connective tissue cells were visible. It was found in the parallel series that not only was growth more extensive when young tissue was used both as regards the parenchymatous and connective tissue cells, but that a larger percentage of specimens showed activity. With the liver good results were obtained, the characters of the growth corresponding with that described in previous communications.* In this series growth took place in 88 per cent. of the specimens, and when present there were always to be seen masses of deeply staining cuboidal cells, but these masses were always less marked than when young tissue was grown in the same plasma. Proliferation of connective tissue cells took place at a slightly later date, the cells soon growing beyond and between the masses of parenchymatous cells. Here again the zone of connective tissue growth was always less marked than was the case with young tissue (fig. 3). (d) Old Tissue Growing in Young Plasma.—This was found to be the worst combination. The tissues apparently were not so active and the medium was less suitable. In the majority of cases no growth took place. The tissues stained poorly and apparently died. In the case of the thyroid only 3 per cent. of the specimens showed any growth at all, and even in the successful eases this was extremely slight. In the majority, even at the end of three or four days, the edge of the tissue remained sharply cut. In some the plasma was liquefied, but in others this change had not taken place, so that here again the absence of the growth was not dependent upon a mechanical factor. Even in the few cases where there was any evidence of growth, no cuboidal cells could be seen. At most there were one or two elongated connective tissue cells to be found after a careful search, so that at first sight there was a tendency to believe that no growth had taken place. * ‘The Journal of Pathology and Bacteriology,’ vol. 18, p. 319 (1914). ne Growth of Mammalian Tissue in Vitro. 481 With the /iver there were positive results in 9 per cent. of the cases, but even in these growth was extremely slight. In no case were any cuboidal or parenchymatous cells seen. There was very slight growth of connective tissue, so that after three days there could be seen in a few cases several long connective tissue cells growing here and there from the edge of the tissue (fig. 4). Often a portion of the original tissue stained poorly and was mani festly dead. These experiments, therefore, strongly confirmed the observations of previous workers, namely, that the tissue of young or embryonic animals shows more active growth than similar tissues taken from adult animals. In addition to this, it appeared manifest that the plasma of such young animals was not nearly so suitable a medium as the plasma of adult animals. Such a condition has not previously been described, and is so opposed to what one would at first sight believe, that it seemed necessary to confirm these experiments. For this purpose further experiments were carried out with stock cultures of rabbit testicle. These tissues had been growing in vitro for 10 generations in a medium composed of two parts of plasma and one part of spleen extract. At this period they were growing vigorously, so that at the end of 48 hours there was a wide zone of newly formed cells surrounding the original tissue. Experiment 13.—Stock specimens of rabbit testicles as described above were cultivated in groups of four each—(qa) in plasma obtained from a young animal 10 days old; (4) in plasma obtained from an adult rabbit over a year old; (c) in a mixture of plasma from an old animal, two parts, and spleen extract one part. At the end of 48 hours, the specimens cultivated in the mixture of old plasma and spleen extract were growing vigorously and showed a very wide zone of active cells, mainly of the connective tissue type (see fig. 5). These specimens served as a control. Those growing in old plasma also showed a wide zone of cells of the connective tissue type, but the newly formed cells, as was to be expected, owing to the absence of a stimulating extract, were considerably fewer in number and did not extend so far into the surrounding media (see fig. 6). In the case of the tissues growing in young plasma, growth was very slight. There were only a few cells growing from the edge of the tissue (see fig. 7) and in these cells but few mitotic figures were seen. This experiment confirmed those carried out with fresh tissues, and there can be no doubt that young plasma is a less suitable medium for the growth of tissues than that of old animals. It has been previously shown* that with older animals some plasmata are not such good media as others, and * © Roy. Soc. Proc., B, vol. 87, p. 452 (1914). 482 Growth of Mammalian Tissue in Vitro. that in such cases the plasma is always improved if it be frozen for a period of about three days, which is very suggestive that the poor growth is due to the presence of an inhibiting substance. It is probable, therefore, that in young animals, also, the inhibiting substance is present in larger quantities, and hence the plasma makes a poor medium. Conclusions. 1. Growth of tissues in vitro affords a valuable means of investigation as to the effects of age upon growth. 2. The tissues of young animals grow more rapidly and vigorously than those of adult animals. 3. The plasma of young animals is a much less suitable medium for the growth of tissue im vitro than the plasma of old animals. 4, The unsuitability of the plasma of young animals as a medium is probably due to the presence of an increased amount of some inhibiting substance. EXPLANATION OF PLATE. Growth of Fresh Liver. Fig. 1.—Two days’ growth of young liver in old plasma. Fig. 2.— Two days’ growth of young liver in young plasma. Fig. 3.—Two days’ growth of old liver in old plasma. Fig. 4.—Two days’ growth of old liver in young plasma. t=) Growth of Stock Testicle. Fig. 5.—Two days’ growth of testicle in plasma plus spleen extract. Fig. 6.—Two days’ growth of testicle in old plasma. Fig. 7.—Two days’ growth of testicle in young plasma. * Walton. > r F : . - ” ene . 5 ’ » i P . + . E — “ x % Roy. Soc. Proc., B, vol. 88, Pl J te 15. 483 The Influence of Homodromous and Heterodromous Electric Currents on Transmssion of Excitation in Plant and Animal. By Prof. J. C. Bosr, M.A., D.Se., C.S.1L., C.LE., Presidency College, Calcutta. . (Communicated by Prof. S. H. Vines, F.R.S. Received June 2, 1914.) I have in a previous paper* described investigations on the conduction of excitation in Mimosa pudica. It was there shown that the various characteristics of the propagation of excitation in the conducting tissue of the plant are in every way similar to those in the animal nerve. Hence it appeared probable that any newly found phenomenon in the one case was likely to lead to the discovery of a similar phenomenon in the other. A problem of great interest which has attracted my attention for several years is the question whether, in a conducting tissue, excitation travels better with or against the direction of an electric current. The experimental difficulties presented in the prosecution of this enquiry are very numerous, the results being complicated by the joint effects of the direction of current on conductivity and of the poles ‘on excitability. As regards the latter, the changes of excitability in the animal nerve under electrotonus have been demonstrated by the well-known experiments of Pfliiger. In a nerve-and- M PS (a) (b) Fria. 1. muscle preparation, the presence of a pole P is shown to induce a variation of excitability of a neighbouring point S. When P is kathode, the excitability of the point S, near it, is enhanced ; stimulation of S, previously ineffective, now becomes effective, and the resulting excitation is transmitted. to M, * Bose, “ An Automatic Method for the Investigation of Velocity of Transmission of Excitation in Mimosa,” ‘ Phil. Trans.,’ B, vol. 204 (1913). VOL, LXXXVIII.—B, 2Q 484 Prof. J. C. Bose. Influence of Electric causing response of the muscle. Conversely, the application of anode at P causes a depression of excitability of S. Stimulus previously effective now becomes ineffective. In this manner the transmission of excitation may be indirectly modified by the polar variation of excitability of the stimulated point (fig. 1a). In the above experiment it will be noted that for inducing a variation of excitability, the tract of nerve SM, along which excitation is transmitted, need have no current passing through it. The presence of a given pole is enough to induce a definite variation of excitability in its neighbourhood. For convenience I shall call this the Znductive action of a pole. | The characteristic variations of excitability induced by polar action are :— (1) The enhancement of excitability at or near the kathode; (2) The depression of excitability near the anode. The boundary between the two polar extensions is reached at a point between the anode and kathode; this ‘point at which the excitability is unaffected is known as the indifference point. The question whether the inductive action of electric poles affects the rate of conduction has been investigated by von Bezold* and by Rutherford.t Von Bezold found that both descending and ascending currents at A (fig. 10) increased the propagation-time between B and © above the normal. Rutherford found on the other hand that the descending current diminished it. The results obtained are thus seen to be indefinite as regards the inductive effect of extrapolar current on conduction. Turning from the inductive effect of neighbouring poles, we have the definite object of enquiry: Does the direction of an electric current, as such, cause any selective variation in the propagation of excitation? In other words, will a homodromous current, z.c. one which flows in the direction of propagation, help or retard transmission of excitation ? Will a heterodromous current on the other hand give rise to an opposite effect ? The object of this particular enquiry is to determine the pure effect of direction of current on conduction of excitation in a tissue through which a current is flowing. We shall call this the Dynamic effect of a current on conductivity and distinguish it from the Inductive effect. The experimental difficulties in isolating the pure effect of current on the intensity and rate of propagation of excitation are very great. In the experiment where “ the whole polar region is interposed between the exciting electrode and the muscle, the conditions are (very) complex. I have been unable to find evidence of any marked alteration in propagation rate, unless * von Bezold, ‘ Elektr. Erreg. Nerven u. Muskeln,’ Leipzig, 1861. + Rutherford, ‘ Journ. Anat. and Physiol.,’ London, vol. 2, p. 87 (1868). Currents on Transmission of Excitation. 485 the polarising current is intense or of prolonged duration, in which case it is always retarded. The presence of two polar regions, a cathodic accelerating and an anodic retarding, causes the one change to counterbalance the other.”* The above would appear to indicate that a current has either no effect or a retarding action on conduction of excitation. These conflicting results are no ~ doubt due to the disturbing influence of the two poles. But this is not the only source of uncertainty in this investigation. Far more serious is the difficulty which arises, as we shall see, from the escape of the induction current employed as the test stimulus. In the course of this paper I shall show how these experimental difficulties have been overcome, and how definite is the characteristic variation of conductivity caused by the directive action of an electric current. The object of my present paper is primarily the demonstration of the selective conductivity induced in the conducting tissue of the plant by the passage of an electric current. After giving the results of this enquiry, I next deal with the question whether the various effects observed in the plant have their parallel in the case of the animal also. Method of Conductivity Balance.—I have previously carried out an electrical method of investigation dealing with the influence of electric current on conductivity. The method of Conductivity Balance which I devised for this purposet was found very suitable. Isolated conducting tissues of certain plants were found to exhibit transmitted effect of excitatory electric change of galvanometric negativity, which at the favourable season of the year was of sufficient intensity to be recorded by a sensitive galvanometer. A long strand of the conducting tissue was taken and two electric connections were made with a galvanometer, a few centimetres from the free ends. Thermal stimulus was applied at the middle, when two excitatory waves with their concomitant electric changes were transmitted outwards. By suitably moving the point of application of stimulus nearer or further away from one of the two electric contacts, an exact balance was obtained. This was the case when the resultant galvanometer deflection was reduced to zero. If now an electrical current be sent along the length of the conducting tissue, the two excitatory waves sent outwards from the central stimulated point will encounter the electric current in different ways; one of the excitatory waves will travel with and the other against the direction of the current. If the power of transmitting excitation is modified by the direction of an electric current, then the magnitudes of transmitted excitations will be different in the two cases, with the result of the upsetting of the conductivity balance. From the * Gotch, “Polar Excitation of Nerve” in ‘Text-Book of Physiology,’ edited by Schiifer, 1900, p. 502. + Bose, “ Comparative Electro-physiology ” (1907), Longmans, Green and Co. MeO) We 486 Prof. J. C. Bose. Influence of Electric results of experiments carried out by this method on the effect of feeble current on conductivity, the conclusion was arrived at that excitation is better conducted against the direction of the current than with it. In other words the influence of an electric current is to confer a preferential or selective direction of conductivity for excitation, the tissue becoming a better conductor in an electric uphill direction compared with a downhill. The result was so unexpected that I have been long desirous of testing the validity of this conclusion by independent methods of enquiry. In the paper already referred to, I have described an automatic method for recording the velocity of transmission of excitation in Mimosa where the excitatory fall of the motile leaf gave a signal for the arrival of the excitation initiated at a distant point. In this method the responding leaf is attached to a light lever, the writer being placed at right angles to it. The record is taken on a smoked glass plate which during its descent makes an instantaneous electric contact, in consequence of which a stimulating electric shock is applied at a given point E of the petiole (fig. 2). A mark in the recording plate indicates the moment of application of stimulus. After a definite interval, the excitation is conducted to the responding pulvinus P, when the excitatory fall of the leaf pulls the writer suddenly to the left. From the curve traced in this manner the time-interval between the application of stimulus and the initiation of response can be found and the normal rate of transmission of excitation through a given length of the conducting tissue deduced. The experiment is then repeated with an electrical current flowing along the petiole with or against the direction of transmission of excitation. The records thus obtained enable us to determine the influence of the direction of the current on the rate of transmission. I shall presently describe the various difficulties which have to be overcome before the method just indicated can be rendered practical. The scope of the investigation will be best described according to the following plan. Part I.—Influence of direction of electric current on conduction of excitation in plants. 1. The genera] method of experiment. 2. The effect of feeble heterodromous and homodromous currents on rate of conduction. 3. Determination of variation of conductivity by the method of minimal stimulus and response. . 4, The after-effects of heterodromous and homodromous currents. 5, Phenomenon of reversal under moderate current. Currents on Transmission of Excitation. 487 Part II.—Influence of direction of electric current on conduction of excitation in animal nerve. 1. The method of experiment. 2. Variation of velocity of transmission under heterodromous and homodromous currents. 3. Variation in the intensity of transmitted excitation under the action of heterodromous and homodromous currents. 4, After-effects of heterodromous and homodromous currents. 5. Phenomenon of reversal. Parr I.—INFLUENCE OF DIRECTION OF CURRENT ON TRANSMISSION OF EXCITATION IN PLANT. 1. The Method of Experiment. I may here say a few words of the manner in which the period of transmission can be found from the record given by my resonant recorder, fully described in my previous paper. The writer attached to the recording lever of this instrument is maintained by electromagnetic means in a state of vibration toand fro. The record thus consists of a series of dots made by the tapping writer, which is tuned to vibrate at a definite rate, say 10 times per second. This method of intermittent contact not only removes the error due to friction, but also enables time-relations to be deduced from the record. In a particular case whose record is given in Curve 1 (fig. 3) indirect stimulus of electric shock was applied at a distance of 15 mm. from the responding pulvinus. There are 15 intervening dots between the moment of application of stimulus and the beginning of response ; the time-interval is therefore 1-5 seconds. The latent period of the motile pulvinus is obtained from a record of direct stimulation; the average value of this in summer is 0-1 second. Hence the true period of transmission is 14 seconds for a distance of 15 mm. The velocity determined in this particular case is therefore 10°7 mm. per second. Disturbance caused by the Leakage of Current—LKmploying this method of record, an attempt was made to determine the changes of velocity under the action of heterodromous and homodromous currents. Buta serious difficulty encountered at the beginning of the investigation arose from the leakage of the induction current used as the testing stimulus. This will be understood from a concrete example. ‘The record in fig. 3, for example, shows 15 inter- vening dots between the moment of indirect application of stimulus (at a distance of 15 mm.) and the beginning of response. The recorded time-interval for transmission was thus 1:5 seconds. The latent period of the pulvinus ¢ 488 Prof. J. C. Bose. Influence of Electric obtained on direct stimulation was, as stated before, 0°1 second. Repetition of the experiment always gave a time-interval of 1°5 seconds for indirect and 0-1 second for direct stimulation. Now, on completing the circuit of the constant current, which for convenience I shall indicate as the polarising circuit, the time-interval for indirect stimulation was at once reduced to 0:1 second, which is the value of the latent period for direct stimulation. This happened on the mere completion of the polarising circuit, with current reduced even to zero. It is evident that this untoward result was due to the escape of the alternating induction current, which went not merely across the short path from E to E’, but also round by the circuitous path of the polarising circuit. It was the escaping current which caused the direct excitation of the pulvinus. This particular difficulty I was finally able to overcome by interposing the electromagnetic device of a choking coil, which effectively prevented the passage of the alternating induction current into the polarising circuit (fig. 2). Precaution has to be taken against another source of disturbance, namely, the excitation caused by the sudden commencement or the cessation of the constant current. I have shown elsewhere* that the sudden initiation or cessation of the current induces an excitatory reaction in the plant-tissue similar to that in the animal tissue. This difficulty is removed by the intro- duction of a sliding potentiometer, which allows the applied electromotive force to be gradually increased from zero to the maximum or decreased from the maximum to zero. The experimental arrangement is diagrammatically shown in fig. 2. After attaching the petiole to the recording lever, indirect stimulus is applied, generally speaking, at a distance of 15 mm. from the responding pulvinus. Stimulus of electric shock is applied in the usual manner, by means of a sliding induction coil. The intensity of the induction shock is adjusted by gradually changing the distance between the secondary and the primary, till a minimally effective stimulus is found. In the study of the effect of direction of constant current on conductivity, non-polarisable electrodes make suitable electric connections, one with the stem and the other with the tip of a sub-petiole, at a distance from each other of about 95mm. The point of stimulation and the responding pulvinus are thus situated at a consider- able distance from the anode or the kathode, in the indifferent region in which there is no polar variation of excitability. By means of a Pohl’s commutator or reverser, the constant current can be maintained either “with” or “against” the direction of transmission of excitation. The * Bose, ‘Plant Response’ (1906); ‘Irritability of Plants’ (1913), Longmans, Green and Co., London. y ore Currents on Transmission of Excitation. 489 former, as stated before, will be designated as the homodromous, and the latter as the heterodromous current. Electrical connections are so arranged that when the commutator is tilted to the right, the current is homodromous when tilted to the left, heterodromous. For ‘the purpose of convenience I shall call the constant current the. Fie. 2.—Complete apparatus for investigation of the variation of conducting power in Mimosa. A, storage cell; S, potentiometer slide, which, by alternate movement to right or left, continuously increases or decreases the applied E.M.F.; K, switch key for putting current “on” and “ off” without variation of resistance ; E, E’, electrodes of induction coil for stimulation ; C, choking coil ; G, microammeter. polarising current, as it will be shown that its effect on conductivity is discriminative or polar, depending on the direction of the current. It should, however, be distinctly understood that under the particular experi- mental arrangement the possibility of polar variation of excitability at the points of stimulation and region of response is avoided. 490 Prof. J. C. Bose. Influence of Electric The electrical resistance offered by the 95-mm. length of stem and petiole will be from two to three million ohms. The intensity of the constant current flowing through the plant can be read by unplugging the key which short-circuits the microammeter G. The choking coil C prevents the alternating induction current from flowing into the polarising circuit and causing direct stimulation of the pulvinus. Before describing the experimental results, it is as well to enter briefly into the question of the external indication by which the conducting power may be gauged. Change of conductivity may be expected to give rise to a variation in the rate of propagation or to a variation in the magnitude of the excitatory impulse that is transmitted. Thus we have several methods at our disposal for determining the induced variation of conductivity. In the first place, the variation of conductivity may be measured by the induced change in the velocity of transmission of excitation. In the second place, the transmitted effect of a sub-maximal stimulus will give rise to enhanced or diminished amplitude of mechanical response, depending on the increase or decrease of conductivity brought about by the directive action of the current. And, finally, the enhancement or depression of conductivity may be demonstrated by the ineffectively transmitted stimulus becoming effective, or the effectively transmitted stimulus becoming ineffective. Exclusion of the Factor of Excitability.—The object of the enquiry being the pure effect of variation of conductivity, we have to assure ourselves that under the particular conditions of the experiment the complicating factor of polar variation of excitability is eliminated. It is to be remembered that excitatory transmission in Mimosa takes place by means of a certain conduct- ing strand of tissue which runs through the stem and the petiole. Ifa point on the petiole of a given leaf be subjected to strong stimulation, an excitatory impulse will not merely be transmitted across its own pulvinus, but will travel farther along the stem, inducing the fall of other leaves. Conversely, a strong stimulus applied on the stem gives rise to an impulse which passes through the pulvinus to the petiole and thence to the sub-petiole, as evidenced by the successive closure of its leaflets. The main pulvinus may thus be regarded as a contractile indicator of excitation, interposed in the path of the conducting strand which connects the stem with the petiole. In the experiment to be described, the polarising current enters by the tip of the petiole and leaves by the stem, or vice versd, the length of the intrapolar region being 95mm. ‘The point of application of stimulus on the petiole is 40 mm. from the electrode at the tip of the leaf. ‘The responding pulvinus is also at the same distance from the electrode on the stem. The point of stimulation and region of response are thus at the relatively great distance ry Currents on Transmission of Excitation. 491 of 40 mm. from either the anode or the kathode, and may therefore be regarded as situated in the indifferent region. This is found to be verified in actual experiments. 2. Effects of Direction of Current on Velocity of Transmission. A very convincing method of demonstrating the influence of electric current on conductivity consists in the determination of changes induced in the velocity of transmission by the directive action of the current, For this purpose we have to find out the true time required by the excitation to travel through a given length of the conducting tissue (1) in the absence of the current, (2) against and (3) with the direction of the current. The true time is obtained by subtracting the latent period of the pulvinus from the observed interval between the stimulus and response. Now the latent period may not remain constant, but undergo change under the action of the polarising current. It has been shown that the excitability of the pulvinus does not undergo any change when it is situated in the middle or indifferent region. The following results show that under parallel conditions the latent period also remains unaffected :— Table I.—Showing the Effect of Polarising Current on the Latent Period. 1 Specimens ............... Ne II. s sec. sec. Latent period under normal condition............... 0°10 0-09 if 3 heterodromous current ...... O-1l 0°10 pk » homodromous current ...... 0-09 | 0:09 The results of experiments with two different specimens given above show that a current applied under the given conditions has practically no effect on the latent period, the slight variation being of the order of one-hundredth part of asecond. This is quite negligible when the total period observed for transmission is, as in the following cases, equal to nearly 2 seconds. Induced Changes in the Velocity of Transmission—Having found that the average value of the latent period in summer is 0°1 second, we next proceed to determine the influence of the direction of current on velocity. Experiment 1—As a rule, stimulus of induction shock was applied in this and in the following experiments on the petiole at a distance of 15 mm. from the responding pulvinus. The recording writer was tuned to 10 vibrations per second; the space between two succeeding dots, therefore, represents a time-interval of 0°1 second. The middle record, N in fig. 3, is the normal. There are 17 spaces between the application of stimulus and the beginning 492. Prof. J. C. Bose. Influence of Electric of response. The total time is therefore 1°7 seconds, and by subtracting from it the latent period of 0:1 second we obtain the true time, 1°6 seconds. The normal velocity is found by dividing the distance 15 mm. by-the true interval 1:6 seconds. Thus V = 15/16 = 9-4 mm. per second. We shall next consider the effect of current in modifying the normal velocity. The upper- most record (1) in fig. 3 was taken under the action of a heterodromous Fie. 3.—Record showing enhancement of velocity of transmission induced by hetero- dromous (uppermost curve) and retardation of velocity induced by homodromous (lowest curve) currents. N, normal record in the absence of current. <— indicates heterodromous, and —- homodromous current. current of the intensity of 1:4 microamperes. It will be seen that the time- interval is reduced from 1:7 seconds to 1:4 seconds; making allowance for the latent period, the velocity of transmission under heterodromous current V, = 15/1°3 = 11°5 mm. per second. In the lowest record (3) we note the effect of homodromous current, the time-interval between stimulus and response being prolonged to 1-95 seoonds and the velocity reduced to 8:1 mm. per second. The conclusion arrived at from this mechanical mode of investigation is thus identical with that derived from the electric method of conductivity balance referred to previously. That is to say, the passage of a feeble current modifies conductivity for excitation in a selective manner. Conductivity is enhanced against, and diminished with the direction of the current. The minimum current which induces a perceptible change of conductivity varies somewhat in different specimens. The average value of this minimal current in autumn is 1°4 microampéres. The effect of even a feebler current may be detected by employing a test stimulus which is barely effective. Currents on Transmission of Excitation. Table II.—Showing Effects of Heterodromous and Homodromous Currents of Feeble Intensity on Period of Transmission through 15 mm. Namabex Intensity of current Period of hetero- Period of homo- i in microamperes. dromous transmission. dromous transmission. 1 1°4 14 tenths of a second 16 tenths of a second 2 1°4 13 » ” 15 2” ” 3 1°6 19 a, Ph Arrest. 4 7 12 55 5 14 tenths of a second. 493 Having demonstrated the effect of direction of current on the velocity of transmission, I shall next describe other methods by which induced variations of conductivity may be exhibited. 3. Determination of Variation of Conductivity by Method of Minimal Stimulus . and Response. In this method we employ a minimal stimulus, the transmitted effect of which under normal conditions gives rise to a feeble response. If the passage of a current in a given direction enhances conductivity, then the intensity of transmitted excitation will also be enhanced; the minimal response will tend to become maximal. Or excitation which had hitherto been ineffectively transmitted will now become effectively transmitted. Conversely, depression of conductivity will result in a diminution or abolition of response. We may use a single break-shock of sufficient intensity as the test stimulus. It is, however, better to employ the additive effect of a definite number of feeble make-and-break shocks. We may again employ additive effect of a definite number of induction shocks, the alternating elements of which are exactly equal and opposite. This is secured by causing rapid reversals of the primary current by means of a rotating commutator. The successive induction shocks in the secondary coil can thus be rendered exactly equal and opposite. Experiment 2.—Working in this way, it is found that the transmitted excitation becomes effective or enhanced under heterodromous current. The homodromous current, on the other hand, diminishes the intensity of excitation or blocks it altogether. 4. After-Effects of Homodromous and Heterodromous Currents. The passage of a current through a conducting tissue in a given direction causes, as we have seen, an enhanced conductivity in an opposite direction. We may suppose this to be brought about by a particular molecular arrangement induced by the current, which assisted the propagation of the 494 Prof. J. C. Bose. Influence of Electric excitatory disturbance in a selected direction. On the cessation of this inducing force, there may be a rebound and a temporary reversal of previous molecular arrangement, with concomitant reversal of the conductivity variation. The immediate after-effect of a current flowing in a particular direction on conductivity is likely to be a transient change, the sign of which would be opposite to that of the direct effect. The after-effect of a heterodromous current may thus be a temporary depression, that of a homodromous current a temporary enhancement, of conductivity. Experiment 3.—This inference will be found fully justified in the following experiment :—The first two responses are normal, after which the hetero- dromous current gave rise to an enhanced response. The depressing after- effect of a heterodromous current rendered the next response ineffective. The following record taken during the passage of the homodromous current exhibited an abolition of response due to induced depression of con- ductivity. Finally, the after-effect of the homodromous current is seen to be a response larger than the normal (fig. 4). These experiments show Fig. 4.—Direct and after effect of heterodromous and homodromous currents. First two Ye enhanced transmission under heterodromous current ; S arrest of conduction as an after-effect of heterodromous current. Next record records, N, N, normal. A i shows arrest under homodromous current. Last record shows enhancement of conduction greater than normal, as an after-effect of homodromous current. (Dotted 5 . . . A . arrow indicates the after-effect on cessation of a given current. ‘| homodromous and a heterodromous current.) that the after-effect of cessation of a current in a given direction is a transient conductivity variation, of which the sign is opposite to that induced by the continuation of the current. Currents on Transmission of Excitation. 495 5. Phenomenon of Reversal under Moderate Intensity of Current. In studying the effect of increasing intensity of polarising current, the concomitant variations of conductivity appeared at first sight to be very puzzling. The results obtained at different stages were, however, very definite. In the first stage with very feeble current there was no per- ceptible change of conductivity. At the second stage with increasing current the conductivity variation increased at a rapid rate, and soon attained a maximum. After this, at the third stage, the conductivity change underwent a decline, and then abolition. The effect outwardly resembled that induced at the first stage. There was, however, a difference, for a critical point had now been reached beyond which there was a complete reversal of normal conductivity variation. These different effects will be clearly understood from the following tabular statement :— Table II1[—Conductivity Variations at Different Stages. Heterodromous current. Homodromous current. WM StA LCs. esc tsct cae os No perceptible change ............... No perceptible change. 1 oe deo Sa Enhanced conductivity Reeb access Depression of conductivity culmi- nating in a block. 1110 Csi eee Hoe rae Conductivity change reduced to| Conductivity change reduced to zero at critica] point zero at critical point. Vigo Sac hee ese: Reversal: diminution of conduc- | Enhancement of conduction. tion culminating in a block I shall now describe a typical experiment which will clearly demonstrate the phenomenon of reversal. Experiment 4.—In this I was desirous of obtaining with an identical specimen alternate records showing (1) normal effect under feeble current, (2) reversed effect under moderate current, and (3) normal effect once more under the original feeble current. It took two hours to obtain these six records (fig. 5) at intervals of 20 minutes. The specimen was vigorous, and therefore was little affected by fatigue. The normal enhancement of conductivity under heterodromous current was observed at as low a current- intensity as 1 microampére. In the first of the pair of records (1) we find the interval between stimulus and response under heterodromous current to be 18 tenths of a second; this was prolonged to 21 tenths under homodromous current; the current was next raised to 3 microampéres, and we observe in the pair of records (2) the reversal of normal effect by a block of conduction under heterodromous, and transmission under homo- dromous current; the time-interval in the latter case is 20 tenths of a 496 Prof. J. C. Bose. Influence of Electric second. The pair of records (3) was taken once more under the original feeble current of 1 microampére (fig. 5). The plant had by this time become Fic. 5.—Records obtained with an identical specimen showing (1) normal action under feeble current ; (2) reversed action under strong current ; (3) normal effect once more under feeble current. —» represents homodromous and <— heterodromous current. slightly fatigued; the results, however, are similar to those of the first series. We have transmission once more under heterodromous current (the time-iuterval being 20 tenths instead of 18 tenths of a second), and retardation culminating in block under homodromous current. I give below a tabular statement of results of typical experiments on reversal under moderate current. Table IV.—Reversal under Moderate Intensity of Current. | Intercity OF Transmission period Transmission period ; Number. pants of under hetero- under homo- | ; dromous current. dromous current. microampéres 1 3 Block of transmission 19 tenths of a second. 2 3 3) bh} »” 12 ” ” 3 4 3) 2 16 ” ” A 4 ” » 22 ” ” 5 4 5 ” 3 18 bP) ” The action of a strong current induces a block of conduction under both heterodromous and homodromous currents. Currents on Transmission of Excitation. 497 Part II.—INFLUENCE OF DIRECTION OF ELECTRIC CURRENT ON CONDUCTION OF EXCITATION IN ANIMAL NERVE. I shall now take up the question whether an electrical current induced any selective variation of conductivity in the animal nerve, similar to that induced in the conducting tissue of the plant. . 1. The Method oj Experiment. In the experiments which I am about to describe, arrangements were specially made so that (1) the excitation had not to traverse the polar region, and (2) the point of stimulation was at a relatively great distance from either pole. The fulfilment of the latter condition ensured the point of stimulation being placed at the neutral region. In the choice of experimental specimens I was fortunate enough to secure frogs of unusually large size, locally known as “golden frogs” (Rana tigrina). A preparation was made of the spine, the attached nerve, the muscle and the tendon. The polarising electrodes were applied at the extreme ends, on the spine and on the tendon (fig. 6). The following are the measurements, in a typical case, of the different parts of the preparation. Fie. 6.—Experimental arrangement for study of variation of conductivity of nerve by the directive action of an electric current. 2 7’, nerve ; S, point of application of stimulus in the middle or indifferent region. 498 Prof. J. C. Bose. Influence of Electric Length of spine between the electrode and the nerve = 40 mm.; length of nerve = 90 mm.; length of muscle = 50 mm. ; length of tendon = 30 mm. Stimulus is applied in all cases on the nerve, midway between the two electrodes, this point being at a minimum distance of 100 mm. from either electrode. The point of stimulation is, therefore, situated at an indifferent region. Great precautions have to be taken to guard against the leakage of current. The general arrangement for the experiment on animal nerve is similar to that employed for the corresponding investigations on the plant. The choking coil is used to prevent the stimulating induction current from getting round the polarising circuit. The specimen is held on an ebonite support, and every part of the apparatus insulated with the utmost care. 2. Variation of Velocity of Transmission. In the case of the conducting tissue of the plant a very striking proof of the influence of the direction of current on conductivity was afforded by the induced variation of velocity of transmission. Equally striking is the result which I have obtained with the nerve of the frog. Experiment 5.—The experiments described below were carried out during the cold weather. The following records (fig. 7), obtained by means of the pendulum myograph, exhibit the effect of the direction of current on the BLL DPAPDRAARRADRAADAAADAIAS Fie. 7.—Effect of heterodromous and homodromous current in inducing variation in velocity of transmission through nerve. N, normal record; upper record shows enhancement and lower record retardation of velocity of transmission under heterodromous and homodromous currents respectively. period of transmission through a given length of nerve. The latent period of muscle being constant, the variations in the records exhibit changed rates of conduction. The middle record is the normal, in the absence of any current oe Currents on Transmission of Excitation. 499 The upper record, denoted by the left-handed arrow, shows the action of a heterodromous current in shortening the period of transmission and thus enhancing the velocity above the normal rate. The lower record, denoted by the right-handed arrow, exhibits the effect of a homodromous current in retard- ing the-velocity below the normal rate. I find thata very feeble heterodromous current is enough to induce a considerable increase of velocity, which soon reaches a limit. For inducing retardation of velocity, a relatively strong homodromous current is necessary. I give below a table showing the results of several experiments. Table V.—Effect of Heterodromous and Homodromous Currents of Feeble Intensity on Velocity of Transmission. Intensity of : | Intensity of : : Acceleration | Retardation Specimen. heterodromous above normal. _| homodromous Teas eae cane current. current. microampére. | er cent. | microampéres. er cent. P | P P P i 0°35 16 } 1 20 2 0° | 13 1-5 19 3 0°8 18 | 2°0 14 4 0°8 WL 2°0 13 5 1:0 18: PG 12 6 1°5 15 3:0 40 | 3. Variation of Intensity of Transmitted Excitation under Heterodromous and Homodromous Currents. In the next method of investigation, the induced variation of intensity of transmitted excitation is inferred from the varying amplitude of response of the terminal muscle. Testing stimulus of sub-maximal intensity is applied at the middle of the nerve, where the polarising current induces no variation of excitability. Stimulation is effected either by a single break-shock or by the summated effects of a definite number of equi-alternating shocks, or by chemical stimulation. Experiment 6—Under the action of feeble heterodromous current the transmitted excitation was always enhanced, whatever be the form of stimu- lation. Homodromous current on the other hand inhibited or blocked excitation (figs. 8, 9). Complication due to Variation of Excitability of Muscle.—In experiments with the plant, there was the unusual advantage in having both the point of stimulation and the responding motile organ in the middle or indifferent region. Unfortunately this ideally perfect condition cannot be secured in experiments with the nerve-and-muscle preparation of the frog. It is true VOL, LXXXVIII—B, 2k 500 Prof. J. C. Bose. Influence of Electric that the point of stimulation in this case is chosen to lie on the nerve at the ‘middle or indifferent region. But the responding muscle is at one end, not very distant from the electrode applied on the tendon. It is, therefore, Fic. 8.—Ineffectively transmitted salt-tetanus becoming effective under heterodromous current. Fie. 9.—Direct and after-effect of homodromous current. Transmitted excitation (salt-tetanus) arrested under homodromous current ; on cessation of current there is a transient enhancement above the normal. necessary to find out by separate experiments any variation of excitability that might be induced in the muscle by the proximity of either the anode or the kathode, and make allowance for such variation in interpreting the results obtained from investigations on variation of conductivity. In the experimental arrangement employed, the heterodromous current is obtained by making the electrode on the spine kathode and that on the tendon anode. The depressing influence of the anode in this case may be expected to lower, to a certain extent, the normal excitability of the responding muscle. Conversely, with homodromous current, the tendon is made the kathode and under its influence the muscle might have its excitability raised above Currents on Transmission of Eacitation. 501 the normal. These anticipations are fully supported by results of experi- ments. Sub-maximal stimulus of equi-alternating induction shock was directly applied to the muscle and records taken of (1) response under normal condition without any current, (2) response under heterodromous current, the tendon being the anode, and (3) response under homodromous current, the tendon being now made the kathode. It was thus found that under heterodromous current the excitability of the muscle was depressed, and under homodromous current the excitability was enhanced. The effect of current on response to direct stimulation is thus opposite to that on response to transmitted excitation, as will be seen in the following Table. Table VI.—Influence of Direction of Current on Direct and Transmitted Effects of Stimulation. Abe tah Transmitted : : : Direction of current. Bar Direct stimulation. excitation. Heterodromous current Enhanced response Depressed response. Homodromous current Depressed response Enhanced response. The passage of a current, therefore, induces opposing effects on the con- ductivity of the nerve and the excitability of the muscle, the resulting response being due to their differential actions. Under heterodromous current a more intense excitation is transmitted along the nerve, on account of induced enhancement of conductivity. But this intense excitation finds the responding muscle in a state of depressed excitability. In spite of this the resulting response is enhanced (fig. 8). The enhancement of conduction under heterodromous current 1s, in reality, much greater than is indicated in the record. Similarly under homodromous current the depression of conduction in the nerve may be so great as to cause even an abolition of response (fig. 9), in spite of the enhanced excitability of the muscle. The actual effects of current on conductivity are, thus, far in excess of what are indicated in the records. The two factors, namely, the induced variation of the conductivity of the nerve and the excitability of the muscle, being antagonistic, certain effects may be predicted when the relative values of the two are changed in definite ways. Let us first consider the effect of diminishing the factor of conductivity of nerve to zero, by bringing the stimulator near the muscle, this being tanta- mount to direct stimulation. The result is seen in the third column of the Table given above. We may next increase the value of the conductivity factor 2R 2 502 Prof. J. C. Bose. Influence of Electric by increasing the length of the conducting path, i.e. by taking greater length of nerve for transmission of excitation. The result is seen in the second column of Table VII. It will now be understood how, by shortening the length of nerve, the normal effect may undergo a reversal. I shall in the following Table denote the change of conductivity of nerve by C, that of the excitability of muscle, by E,. Table VII.—Reversal of Normal Effect by Shortening the Length of Nerve. Length of Direction of - Conductivity of nerve versus Resultiae eee nerve. Current. excitability of muscle. pipet gis i eelion'e eer Heterodromous ...... Enhanced C,, > Depressed E,, | Enhanced response. Homodromous......... Depressed C, > Enhanced E,, | Depressed response. 2. Short ...... Heterodromous ...... Enhanced C,, < Depressed H,, | Depressed response. Homodromous.........| Depressed C,, < Enhanced H,, | Enhanced response. i I shall now give experimental verification of the truth of the inferences that have been outlined above. Experiment 7—We have seen that, when the nerve is stimulated in the middle or indifferent region, the transmitted effect is normal. From the above we see that this normal effect will persist, as long as the nerve-tract is of sufficient length; and that the effect will undergo an apparent reversal when it is very much shortened. This is fully borne out by results of numerous experiments. For example, the length of nerve in a preparation was 90 mm. When stimulus was applied near the spine (length of trans- mission = 90 mm.), the transmitted effect was found to be normal, we. enhanced response under heterodromous, and depressed response under homodromous current. The transmitted effect remained normal as the stimulator was gradually moved towards the muscle, thus reducing the length of transmission. A critical length was now found below which the effects underwent a reversal. This was the case when the length of the nerve was reduced to 15 mm., the reversed effects being an enhanced response under homodromous, and a depressed response under heterodromous current. These are due, as explained before, to the induced variation of excitability of muscle, which now became the predominant factor. The very great influence exerted by the direction of current on conductivity of nerve is forcibly brought to our mind by the fact that under normal conditions it completely overpowers the opposing effect of change of excitability in the muscle. Currents on Transmission of Excitation. 503 4, After-Effects of Heterodromous and Homodromous Currents. On the cessation of a current there is induced in the plant-tissue a transient conductivity change of opposite sign to that induced by the direct current (cf. Experiment 3). The same I find to be the case as regards the after-effect of current on conductivity change in animal nerve. Of this I only give a typical experiment of the direct and after-effect of homodromous current on salt-tetanus. Experiment 8.—In this experiment sufficient length of time was allowed to elapse after the application of the salt on the nerve, so that the muscle, in response to the transmitted excitation, exhibited an incomplete tetanus. The homodromous current was next applied, with the result of inducing a complete block of conduction, with the concomitant disappearance of tetanus (fig. 10). Fie. 10.—Normal transmitted salt-tetanus without current at 0. Enhancement under heterodromous current of 3 microampéres. Reversal at 10 microampéres. The homodromous current was gradually reduced to zero by the appropriate movement of the potentiometer slide. The after-effect of homodromous current is now seen in the transient enhancement of transmitted excitation, which lasted for nearly 40 seconds. After this the normal conductivity was restored. Repetition of the experiment gave similar results. 5. Phenomenon of Reversal. In experiments with Mimosa it was shown that an increase of polarising current above the critical value gave rise to a reversal of the normal conductivity variation (cf. Experiment 4). Even in this matter of reversal I find a very remarkable parallelism between the reactions of the conducting tissue of the plant and the nerve of the animal. The reversal was obtained 504 Prof. J. C. Bose. Influence of Electric both with heterodromous and homodromous currents, the testing stimulus being either electrical or chemical. Reversal under Increased Heterodromous Current. Experiment 9.—In this the transmitted excitation due to the application of salt, gave rise to an incomplete tetanus of moderate intensity. After securing this condition, heterodromous current was applied and continuously increased. It will be seen (fig. 10) that a great enhancement of conduction took place when the current attained a value of 3 microampéres. A reversal was, however, induced as soon as the current reached a value of 10 microampéeres, and we observe a complete cessation of normal incomplete tetanus. From this we see that under reversal the conductivity is depressed below the normal. Table VIII.—Showing Normal and Reversed Effects under Heterodromous Current. Intensity of current for Intensity of current for No. normal enhancement reversed effect of depression of conductivity. of conductivity. microamperes. microampeéres. 1 1°65 2 1°5 8 3 2 10 4 2 10 5 3 10 6 3 10°65 7 3 11 8 3 12 | Reversal under Increased Homodromous Current. Experiment 10.—On the appearance of incomplete tetanus T brought on by the application of salt at the middle of the nerve, homodromous current was continuously increased from zero to 10 microampeéres and then gradually brought back to zero. This was accomplished, as stated before, simply by the forward and backward movement of the potentiometer slide. The resulting record taken on the revolving drum shows the cycle of effects. It is seen that the conduction in this case is arrested as soon as the polarising current attains a value of 2 microampéres; at 10 micro- amperes there is induced a reversal with an enhanced conductivity above the normal. Under a continuous diminution of current there is an arrest of conduction once more at 2 microampéres, and restoration of normal con- duction at zero intensity of current. A continuous increase of current gives Currents on Transmission of Excitation. | 505 rise once more to an arrest of conduction at 2 microamperes and a reversal at 10 microampéres. It is a curious fact that the reversal under hetero- dromous and homodromous currents takes place, generally speaking, at the same intensity, namely, 10 microamperes. Before passing under review the characteristic results obtained under varying conditions of the experiment, I shall discuss briefly the question whether it is possible to explain the observed results merely by considering the induced variation of excitability as the sole cause. We shall take, then, the simple case of arrest of conduction by homodromous current; I find that the arrest takes place just the same, whether the anodic electrode is placed on the spine or on an adjacent point n, on the nerve itself (see fig. 6). Discarding from our consideration the possibility of an induced variation of conductivity, we may assume that the arrest was due to the depression of excitability of the stimulated point of nerve on account of the proximity of the anode. But the point of stimulation was, in general, placed not near the anode, but in the middle or indifferent region. In fact the diminution or arrest of transmitted excitation was observed even in the case where the stimulus was applied at the far end of the nerve, at a distance of about 70 mm. from the anode at one end of the nerve, and only 20 mm. from the muscle at the other end. Against this it might be urged that under the action of strong currents the anodic depression might extend to a considerable distance. It has, however, been shown that for causing a depression or arrest of transmitted excitation a strong current was not at all necessary, such a depression sometimes taking place under an intensity of current as feeble as 0°3 microampere, the applied E.M.F. being less than one third of a volt. The difficulty of explaining the observed results by an assumption of induced variation of excitability would thus appear to be insurmountable. This difficulty is greatly intensified—indeed borders on the impossible— when we follow the same reasoning as regards the action of increasing intensity of homodromous current beyond the critical point. With stronger current, not only will the indifferent point be pushed towards the kathode, but the depression induced by the anode will be so great as to render the stimulated point of the nerve inexcitable. There being no excitation to be transmitted, the response should then undergo an extinction. Instead of this we find that the response shows an actual enhancement, on account of the reversal of the induced variation of conductivity which has already been described. This shows conclusively that the phenomenon we have studied is due not to a variation of excitability, but to that of conductivity. 506 Prof. J. C. Bose. Influence of Electric We have seen further that a perfect parallelism exists in the conductivity variation induced in the plant and in the animal by the directive action of the current. No explanation: could be regarded as satisfactory which is not applicable to both cases. Now with the plant we are able to arrange the experimental conditions in such a way that the factor of variation of excita- bility is completely eliminated. The various effects described about the plant-tissue are, therefore, due entirely to variation of conductivity. The parallel phenomena observed in the case of transmission of excitation in the animal nerve must, therefore, be due to the induced change of conductivity. I may now briefly recapitulate some of the principal facts established in this paper. The variation of conductivity induced by the directive action of current has been investigated by two different methods :— (1) The method in which the normal speed and its induced variation are automatically recorded ; (2) That in which the variation in the intensity of transmitted excitations is gauged by the varying amplitudes of resulting responses. The great difficulty arising from leakage of the exciting induction current into the polarising circuit was successfully overcome by the interposition of a choking coil. The following summarises the effects of direction and intensity of an electric current, on transmission of excitation through the conducting tissue of the plant :— The velocity of transmission is found to be enhanced against the direction of a feeble current, and retarded in the direction of the current. Feeble heterodromous current enhances conductivity, homodromous current, on the other hand, depresses it. Ineffectively transmitted excitation becomes effectively transmitted under heterodromous current. Effectively transmitted excitation, on the other hand, becomes ineffectively transmitted under the action of homodromous current. The after-effect of a current is a transient conductivity change, the sign of which is opposite to that induced during the passage of current. The after- effect of a heterodromous current is, thus, a transient depression, that of homodromous current a transient enhancement of conductivity. When the intensity of current is gradually increased, the characteristic conductivity variation is also increased, at first slowly, then rapidly. There is a critical intensity of current above which the conductivity variation under- goes a decline, culminating in an actual reversal. The effect of heterodromous Currents on Transmission of Excitation. 507 current is then a diminution, and that of a homodromous current an enhance- ment of conductivity. The characteristic variations of conductivity induced in animal nerve by the direction and intensity of current are in every way similar to those induced in the conducting tissue of the plant. These various effects are demonstrated by the employment of not one but various kinds of testing stimulus. Excitation may thus be caused (1) by a single break-induction shock, or (2) a series of equi-alternating tetanising shocks, or (3) by chemical stimulation, The results that have been given are only typical of a very large number, which invariably supported the characteristic phenomena that have been described. The records given in this paper are photographic reproductions of the original. Conclusion. The action of an electrical current in inducing variation of conductivity may be enunciated under the following laws, which are equally applicable to the conducting tissue of the plant and the nerve of the animal. 1. The passage of a current induces a variation of conductivity, the effect depending on the direction and intensity of current. 2. Under feeble intensity, heterodromous current enhances and homo- dromous current depresses the conduction of excitation. 3. The after-effect of a feeble current is a transient conductivity variation, the sign of which is opposite to that induced during the continuation of current. 4. The normal conductivity variation undergoes a reversal under a strength of current above the critical value. The heterodromous current then induces a depression, while the homodromous current induces an enhancement of conductivity. In my ‘ Researches on Irritability of Plants’ I have shown how intimately connected are the various physiological reactions in the plant and in the animal. And I ventured to predict that the recognition of this unity of response in plant and animal will lead to further discoveries in physiology in general. This surmise has been justified, for it was by the study of effect of current on the conducting tissue of the plant that I was led to the discovery of the characteristic effects of the direction of an electric current on the conductivity of the animal nerve. My research assistants, Messrs. Guruprasanna Das, L.M.S., and Surendra Chandra Das, M.A., rendered me very valuable help in this long investi- gation. 508 The Measurement of Arterial Pressure in Man. L—The Auditory Method. By Martin Fuack, Leonarp Hitt, F.R.S., and James McQuEEN. (Received December 3, 1914). From the Physiological Laboratory, London Hospital Medical College (London Hospital Research Fund), and the Pathological Laboratory, Aberdeen University.) In a previous communication* we showed that when an artery, exposed in a living animal, is compressed in a glass compression tube full of water (Ringer’s solution), the pulse, distal to the compression, is not obliterated until the pressure of the water is raised just above the systolic pressure of the blood in the artery, whereas when the same artery, placed on bone, wood or glass, is compressed by the hag of Leonard Hill’s pocket sphygmometer, or by the armlet of the sphygmometer, so arranged that it does not embrace the surrounding pulsating tissues, the pulse is abolished by pressures under, and even much under, the diastolic pressure of the blood stream. These facts are correlated with the manner in which the artery is com- pressed in each case. Enclosed in the compression tube the artery is compressed by the water equally in a circular fashion so that the rise of external pressure, up to the diastolic pressure, has no effect in producing deformation of the artery. Ultimately, when the compression becomes greater than the diastolic pressure, the artery flattens and changes to the oval shape during diastole. It is flattened during systole when the external pressure rises above the systolic pressure. When the carotid artery of the living animal is freed from the surrounding tissues and placed on a watch- glass and compressed by the bag of the pocket sphygmometer, or by the armlet so arranged as not to embrace the pulsing tissues of the neck, the oval deformation sets in at far lower pressures and is complete in relatively thin-walled labile arteries at pressures much under diastolic pressure. Consequently the blood flow is cut down by an external pressure less than diastolic to a mere ineffective trickle of blood, and the pulse is completely damped out. Ifa branch of the carotid artery distal to the bag were con- nected with a C-spring manometer, the record would show a progressive lowering of both the systolic and the diastolic pressure almost to zero, till with an external pressure much less than the diastolic pressure in the aorta the pulse was damped out and the blood flow became a trickle. * “Roy. Soc. Proc.,’ B, vol. 87, p. 344 (1914). ” The Measurement of Arterial Pressure in Man. 509 Now MacWilliam and Melvin* have studied the behaviour of an excised artery compressed in a tube containing water with external pressures from zero up to diastolic pressure. They find that with pressures, say, from zero up to 50 mm. Hg, the distal manometers, systolic and diastolic, give practically unchanged readings. It is only when the external pressures become equal - to the diastolic, or above it, that any great alteration in the distal manometer reading occurs. We find that this is so in the case of the carotid of a living animal enclosed in a glass compression tube and compressed by water. On the other hand, in the case of the same artery placed on a watch-glass, a much lower external pressure applied with the bag of the pocket sphyg- mometer is sufficient to obliterate the pulse. The question now arises, how far do the conditions of an artery embedded in the tissues and surrounded with the tissue vessels resemble those of an excised artery enclosed in a compression chamber full of water? Were the facts demonstrated by this simple schema applicable, pressures from zero to 50 mm. Hg should have no effect upon the diastolic and systolic pressures in the brachial or radial artery distal to the point of compression. But such a conclusion is contrary to every-day clinical experience. When an armlet is placed on the upper arm and the radial artery is felt at the wrist, and the pressure in the armlet is raised from zero to 50 mm. Hg, at some stage, varying under different conditions to be detailed later, the pulse beat at the wrist increases perceptibly in force. Accordingly we have here a seeming paradox, augmentation of the pulse produced by an external pressure rising from 0 to 50 mm. Hg—a pressure sufficient to deform an exposed artery lying upon bone and damp out the pulse, but which has no effect on an artery in the simple schema of MacWilliam and Melvin. The phenomenon in question is obtained with a varying degree of facility in different people. In a patient whose systolic pressure was 130 and diastolic 80 mm. Hg, the increase in the force of the pulse beat became perceptible at the wrist when the external pressure in the armlet on the upper arm was raised to 35-40 mm. Hg. After taking exercise for three or four minutes, when the heart had been made to beat violently and the pulse- pressuret range increased in extent, a condition approaching to the findings in aortic disease, an increase in the force of the pulse beat became apparent with 10 mm. of external pressure in the armlet. In a case of aortic disease with a large pulse-pressure range, the increase in the force of the pulse beat became apparent with 5-10 mm. of external pressure in the armlet. * ‘Heart,’ vol. 5, p. 153 (1914). + The difference between the diastolic and systolic pressures. SO Messrs. M. Flack, L. Hill, and J. McQueen. From a case of aortic disease having an aberrant radial artery we have taken tracings, using the Dudgeon sphygmograph with weight extension. The absence of the plethysmograph effect and of disturbance from ven comites—there are none accompanying the aberrant artery—-was thus secured. In the tracing (fig. 1) the external pressure in the armlet at the upper arm was, to begin with, 0 mm., then raised to 30 mm. The size of the initial pulse beat was regulated by the weight in the pan of the extension apparatus; some little skill is needed to secure the proper weight to show the phenomenon in a marked fashion. Fic. 1. The tracing demonstrates that there is an increase in the systolic pressure in the radial artery as a result of compressing the upper arm; there is consequently an increase in the pulse-pressure range. Corroboration of these tracings has been obtained by placing one armlet on the calf and another on the thigh of a boy. In normal boys we find the thigh reading to be higher than that of the calf.* Systolic pressure (horizontal posture). Boy. Thigh. Calf. mm. of Hg. | mm. of Hg. I 165 135 140 when 60 mm. Hg was maintained in the thigh armlet. 130 LP) 0 th) 99 ” Il. 155 120 135 ,, 40—50 mm. Hg a 9 III. 105 80 90 bed ” ” ” * In the case of two patients we have observed a systolic pressure considerably higher in the calf than in the thigh ; e.g., 175 thigh, 300 calf, upper part, and 225 mm. Hg lower part of leg. These differences of pressure obtained in both legs. It has been suggested to us that the higher reading in the leg may be due to protection of the artery from compression by fascine. Itis difficult to see how this can come about, and we are at present unable to offer an explanation other than one based on the observations detailed in the following paper. The difference was certainly not due to resistance to compression on the part of the arterial wall, for the dorgalis pedis artery in these patients offered no unusual The Measurement of Arterial Pressure in Man. bal The measurements of systolic pressure thus show an increase in the calf reading brought about by raising the pressure in the thigh armlet to 40-50 mm. He. We find that the increase in the force of the pulse beat in the artery distal to an armlet coincident with the rise of armlet pressure from 0 to 50 mm. is . correlated with the appearance of certain sounds audible with a stethoscope (Bowles’ tambour form was used) placed at the elbow. An armlet is placed on the upper arm and the stethoscope laid without pressure at the bend of the elbow; on raising the pressure in the armlet above systolic pressure and then lowering it gradually, clear sounds became audible, gradually getting louder; these are followed first by murmurs and then by loud clear sounds. Ata certain level, which is taken as the index of the diastolic pressure, the loud clear sounds suffer a sudden diminution in volume and tone, to be succeeded by dull sounds; these may be audible much below diastolic pressure, e.g. 28-35 mm. or so. These dull sounds, we believe, are due to the sudden tension of the artery and its branches produced by the impact of the systolic wave. This sets the mass of tissue enclosed in the armlet into vibraticn, the vibrations becoming audible to the ear as sounds. In the case noted previously these dull sounds appeared when the pressure in the armlet enclosing the upper arm reached 35-40 mm. Hg. At precisely the same level the force of the pulse beat could be felt increased while palpating with the fingers the radial artery at the wrist. When the subject took violent exercise the dull sounds became audible when the pressure in the armlet reached 10 mm. Hg, and increased in loudness with successive increments of this pressure. In aortic disease, where the pulse-pressure range is high and the artery with each systole suffers considerable distension, the sounds are audible at the bend of the elbow under ordinary conditions. These sounds are at once increased in volume when the pressure in the armlet is raised 5-10 mm.; at precisely the same levels of external pressure the finger feels the pulse grow stronger, and the sphygmographic tracing confirms the sensory impressions from the fingers. Therefore we conclude that a sound is given forth dependent on the pulse-pressure range of the blood stream which enters the armlet. If that range is high, ¢.g. in aortic disease, the systolic wave is big enough to cause the sounds to be produced under ordinary conditions. If the pressure within the armlet is raised 5-10-20-30-40 the systolic wave is reinforced by the resistance to compression by the bag of the pocket sphygmometer. Further, itis difficult to suppose that the arteries in either leg should offer the same local resistance to compression and a resistance wholly different to that of arteries in other parts of the body. 512 Messrs. M. Flack, L. Hill, and J. McQueen. increased tension of the arteries under the armlet and the sound becomes louder. The bigger the pulse wave passing to the armlet, the ampler will swing the vessels under the armlet and the less the compression required to increase the force of the pulse wave in the radial artery at the wrist, and produce the dull sounds. We believe these sounds are produced by the impact of the systolic wave vibrating the tense artery and its branches, big and small. Consequently they should be independent of the blood flow. An artery ligated beats up to the point where the ligature is applied. Normally we feel pulses by closing the artery with the ball of the finger or thumb and feeling the impact of the systolic wave on the tip of the finger or thumb. To deform the artery with the finger a lower pressure than the diastolic pressure suffices. The finger deforms it just as the bag of the sphygmometer deforms an artery placed on bone. It is possible to arrange an armlet on the upper arm and a second armlet immediately below the tambour at the bend of the elbow. Suppose the pressure is raised in the armlet to 30-40 mm. Hg, sufficient to call forth the dull sounds, and that then the armlet pressure below the bend of the elbow is raised far above the ascertained systolic pressure of the blood. This stops all effective flow in the artery at the bend of the elbow; the subsidiary branches of the main arterial trunk rapidly fill up, as all exit for the blood is blocked in the distal area by the compressing armlet placed below. Yet the dull sounds persist. Further, these sounds become more audible as the blood flow is stopped. That is exactly what one would expect if the sounds are due to sudden tension. The artery and all its branches become tenser above the block. The whole kinetic energy of the pulse spends itself in striking the tense labile artery and its branches. The whole mass of tissue under the armlet—permeated with blood vessels—is struck by the pulse. We find that the phenomenon of an increase in the pulse force in the radial artery at the wrist is often felt best at the first examination of the patient with the sphygmometer. The excitement produced by examination increases the force of the heart, and the high crest of the pulse wave reaching the tissues compressed by the armlet becomes reinforced : an increase in the force of the pulse beat is thus felt easily with a low pressure within the armlet. As the excitement subsides, and the pulse beat becomes normal in its range, the level of pressure at which the increase is felt becomes higher. As regards the production of the loud sounds and murmurs heard on passing from diastolic level to systolic level, it is possible to separate the element of sound due to pure tension of the arterial ‘wall and the element of sound which requires a flow of blood. The Measurement of Arterial Pressure in Man. 513 Suppose one raises the pressure in the armlet on the upper arm to 100 mm. and so develops at the elbow the loud characteristic murmur. If one then raise the pressure in another armlet placed below the auscultating tambour, to well above the known systolic pressure in the artery, this murmur disappears but a sound synchronous with each pulse beat appears in its place. . This is the dull sound due to the sudden tension of the arterial wall, a sound independent altogether of those vibrations which are set up in the arterial wall by that inrush and outrush of the blood which is synchronous with the crest of the systolic wave. On lowering the armlet pressure to the level at which the outflow of blood regains sufficient velocity the characteristic murmur returns. We conclude, therefore, that stoppage of the blood flow by the lower armlet while the pressure in the upper armlet ranges up to systolic pressure, cannot prevent the occurrence of sounds, though it leaves their quality changed. The sounds due to vibrations set up by the sudden in and outrush of blood disappear; the sounds due to the periodic sudden tension of the arterial walls persist. As we have said before,in an artery placed on bone or glass and compressed with an armlet (the armlet not embracing pulsing tissues), or with the bag of the pocket sphygmometer, the pulse is obliterated by pressures below diastolic pressure. By the reinforcement of the pulse in the vessels of the tissues which are enclosed by the armlet or bag, used in the ordinary way, these critical pressures are successfully passed, and the normal process of arterial deforma- tion proceeds at the proper level. So accurate systolic blood-pressure measurements, both auditory, tactile, and visual, become possible. On this mechanism—the conserving effect of the tissue vessels on the pulse—depends the accuracy of the auditory method of estimating the diastolic pressure. Without this mechanism the diastolic level would be too low. Thus we have found it so when the bag of the pocket sphygmometer, or armlet, used so as not to embrace pulsing tissues (as described in a previous communication, loc. cit.) is applied to the aberrant radial. But when the armlet is applied round the arm then the diastolic auditory index, as heard in the aberrant radial, comes much closer to the truth. It has been noted by MacWilliam and Melvin, and others, that sounds can be produced which are audible at the brachial artery at the elbow when finger pressure is applied to the brachial artery in the arm. Hill, McQueen, and Flack (loc. cit.) have shown that finger pressure applied discretely to any artery, brachial or radial, deforms the artery in a precisely similar manner as does the bag, or armlet (used with the box so as not to embrace pulsing tissues), when applied to the aberrant radial. The pulse is damped out below 514 Messrs. M. Flack, L. Hill, and J. McQueen. diastolic pressure. Consequently the sounds produced by finger pressure on the brachial artery, while they somewhat resemble in quality the normal sounds produced by armlet pressure, do not show a perfect similarity to these. Thus, suppose we obliterate the brachial artery with the finger, on releasing it gently we hear for a very short period clear sounds followed by murmurs. These murmurs are not in our experience followed by clear sounds and then by dull sounds, as is the case when the armlet embraces the upper arm. It is obvious that on slightly releasing the occluding pressure the blood flows in jets into the artery, which is relaxed below the seat of compression. Hence the first clear tension sounds. When the artery becomes oval in shape, the clear sounds are dulled by the murmurs. When the artery becomes circular the pulse wave does not suffice to make tense the now patent artery and produce the dull sounds. The finger does not obstruct the peripheral outflow and bring into play the reinforcement due to the vessels of the tissues. If we place an armlet distal to the position of the stethoscope, and by the pressure in this armlet obstruct the blood flow, then on compressing the brachial artery with the finger, only clear sounds are heard for a short period as the artery is compressed and released. The murmurs vanish, clear sounds take their place, and the range of sounds is short. As we have pointed out, the whole range of sound is dependent on the resonating effect of the vessels in the tissues surrounding the artery. This resonating effect is absent when an artery is discretely occluded by the finger so that surrounding tissues are not included. If an armlet is placed on the upper arm and the external pressure raised till the clear sounds are produced, just below the systolic level, then if the artery be occluded by the finger between the armlet above and the tambour of the stethoscope below, all sounds vanish. Supposing we place the tambour under the lower part of the armlet, then compression of the artery imme- diately distal to the edge of the armlet does not abolish the sounds. They become clearer, because the energy of the pulse spend itself on the mass under the armlet; the obstruction further tightens up the vibrating drum. The sounds are abolished when the finger occeludes the artery between the armlet and the tambour; first, because the artery is not distended by the systolic phase of the pulse wave; secondly, because the sounds produced under the armlet are not now conducted by the fluid column of blood in the artery. Suppose we place the stethoscope under the lower part of the armlet, that is just above the occluding finger, then the sounds are audible just up to the systolic pressure of the blood, eg.,110 mm. Hg. But if the stethoscope is placed exactly under the upper edge of the armlet, sounds are audible up to, The Measurement of Arterial Pressure in Man. 515 say, 200 mm. Hg. in the armlet, a pressure far above the systolic pressure of the blood. It is clear that when an armlet is placed on the upper arm and the pressure raised above systolic pressure, the systolic wave must meet a sudden check at the upper part of the armlet, hence the arteries are here made tense and the sounds produced by their sudden tension are heard. It’ is possible under these conditions (MacWilliam and Melvin* have observed it in one case) that the sound, if exceptionally loud, may be conducted by bone below the armlet. Their case was one of aortic regurgitation in a healthy student—a good athlete. Conclusions. _ 1. In the measurement of arterial pressure by means of the armlet and sphygmometer the auditory method gives clear indices of systolic and diastolic pressure. The auditory indices are (1) loud throbs heard in the artery below the armlet when the compression is lowered just below the systolic pressure ; (2) a sudden diminution in the sounds when the pressure falls just below the diastolic pressure. We find that these indices depend on the pulsatile flow of blood in and out of the part compressed by the armlet; the artery is deformed by the compressive force and its wall swings out and in when the pulse wave strikes the deformed part. When the flow of blood into the limb peripheral to the armlet is obstructed the throb is no longer heard, but is replaced by dull sounds caused by the pulse striking the tense arterial wall. 2. Dull sounds are heard, under ordinary conditions, when the compression is reduced below the diastolic pressure. Such slight compression gives occasion to the pulse to produce the dull sound, by obstructing the venous outflow, and thus raising the diastolic pressure in the arteries and the tension of their walls. . 3. The bigger the systolic wave the less compression is required to make audible the dull sound. 4, The accuracy of the auditory method depends on the conserving effect which the tissue vessels have on the arterial pulse when the arm is com- pressed. 5. The method cannot, therefore, be used to give accurate measurements in the case of an artery lying on bone and unsupported by tissue vessels such as the aberrant radial or dorsalis pedis. 6. Accurate readings can be obtained from these arteries where they lie embedded in tissues, and the reinforcing effect of the tissue vessels comes into play. * ‘Brit. Med. Journ.,’ 1914, A, p. 697. VOL, LXXXVIII—B, 258 516 Messrs. M. Flack, L. Hill, and J. McQueen. 7. Clinicians know that the pulse in the radial artery becomes more forcible when they begin to compress the arm. At the beginning of com- pression of the arm, the armlet, by obstructing the venous outflow and making tenser the arteries in diastole, improves. the conduction of the systolic wave. The pulse in the radial artery, therefore, becomes reinforced. The dull sound and the reinforcement of the pulse are due to the same cause. 8. Evidence has been obtained then, by experiments on man, of the effect of increased tension of the arterial wall (lessened lability) on the conduction of the crest of the systolic wave. The peripheral conditions affect the lability and the pressure readings. The Measurement of Arterial Pressure in Man. I1.—A Schematic Tiwestigation. By Martin Friack, Leonarp Hitt, F.R.S., and JAMES McQUEEN. (Received December 3, 1914.) (From the Physiological Laboratory, London Hospital Medical College (London Hospital Research Fund), and the Pathological Laboratory, Aberdeen University.) MacWilliam and Melvin* have demonstrated in the case of the excised artery—compressed in their schema—that a compressing force which was not sufficient to obliterate the pulse caused a great fall in the manometer, which they placed distally to the compression tube. To cite an example, the entering pressures in the proximal manometers were: systolic 178 mm. Hg, diastolic 118 mm. Hg. A compressing force of 140 mm. Hg caused a great fall in the distal manometers—systolic became 42 mm. Hg, diastolic 22 mm. Hg. We find that the artery, under these conditions, is flattened during diastole, and the inflow during systole is not of sufficient duration to maintain the distal pressure, supposing the resistance to outflow is unchanged. If the resistance to outflow is increased, no such distal fall of pressure occurs. Their schema differs in essential points from the conditions which pertain to an artery embedded in living tissues and encircled by an armlet. The pressure within the armlet at first does not deform the artery, but expresses blood from, and increases the peripheral resistance in, the mass of tissue it * ‘Heart,’ vol. 5, p. 153 (1914). The Measurement of Arterial Pressure in Man. a7, encloses, by compressing the capillaries and obstructing the peripheral exits. It thus converts the compressed area into a resonating mass ; the pulse is not damped down in the labile arteries, but strikes the blood which fills to distension not only the main artery, but every patent arteriole throughout the mass, and causes the whole tense mass to vibrate. Thus we find in the case of the living subject, if the systolic pressure be 115 mm. Hg and the armlet pressure be kept at 110 mm. Hg, the venous pressure in the limb rises distal to the armlet. If another armlet be put below the first, and the pressure raised within this, the pulse at the radial will not be obliterated until the pressure in this armlet reaches 115 mm. Hg. If the conditions found in the schema of MacWilliam and Melvin held good in the arm, a far lower pressure in this lower armlet would suffice to obliterate the pulse, for in their schema, under similar conditions, the distal manometers show a great diminution in pressure. In the case of the arm, as the pressure is raised in the upper armlet the venous outflow becomes obstructed, and the pulse then strikes a mass of blood congested within the vessels which permeate the tissues; no pressure less than systolic in the lower armlet can prevent the vibration of the mass reaching the radial artery. It is true that the pulse felt in the radial becomes feeble as the pressure is raised in the upper armlet to 110 mm., but the pressure in the radial does not sink, because the blood still flows in and cannot escape from the veins. The pulse in the radial is enfeebled by the resistance which arises from the deformation of the brachial artery brought about by a pressure in the upper armlet of 110 mm. Its force is partly spent in the labile artery above this armlet. The range of pulse pressure below the upper armlet is greatly diminished too, because the diastolic pressure is raised owing to the venous obstruction. There is in consequence a much smaller swing, but this swing cannot be stopped until the pressure in the lower armlet is raised to the full systolic pressure, 115 mm. Hg. The facts we have detailed above show that the simple schema, in which an artery is compressed in a chamber full of water, does not represent the conditions which pertain in the arm. We have attempted to imitate these conditions in the schema represented in fig. 1. Two glass compression chambers are filled with water and connected with each other and to a compression bottle. In one is placed a piece of human carotid artery, in the other a schematic representation of the tissue vessels. This consists of a condom (thin-walled, wide rubber tube) filled to distension with chopped rubber sponge. The expansion of the condom is limited by 282 518 Messrs. M. Flack, L. Hill, and J. McQueen. N N N Fia. 1. The Measurement of Arterial Pressure in Man. 519 an external coat of muslin. The condom is tied at either end on to a thistle funnel; the tube of each funnel passes through a rubber cork. The corks close the ends of the compression chamber. A pulsatile flow of water is maintained through (A) the artery, (B) the tissue schema. The water finally escapes through a glass nozzle into a pail. The resistance of the tissue schema is such as to give a continuous flow from the nozzle, marked at each systole by a slight pulsatile increase. The pulsatile flow is secured in this wise. Water flows through the tap through a length of rubber tubing to the schema. Close to the tap a mercury valve is inserted so that the pressure in the tube is kept con- stant. The rubber tube is pulsed rhythmi- cally between two wooden discs (cotton reels), one of which is fixed to the support, and the other to the piston rod itself, of Brodie’s respiration pump. At each stroke of the pump the tube is compressed and the flow interrupted. The rate of the pulse can be varied. T-pieces are inserted in the schema so that the pulse and pressure can be recorded in turn from (i) one compres- sion chamber or both chambers, (ii) the tube connecting artery and tissue schema, Qii) the outflow nozzle. An alternative pathway is arranged in the compression chamber which contains the artery, so that the flow can be directed either through the artery or through a piece of rubber tube, which acts as a rigid tube. Or two pieces of artery, one acting as artery and the other as vein, can be placed in this compression chamber; the flow is then made to pass through (1) the artery, (2) the tissue schema, (3) the vein, and so to the outlet nozzle. Or the flow can be made to go through Fia. 2. 520 Messrs. M. Flack, L. Hill, and J. McQueen. the artery and vein alone, excluding the tissue schema, or through the artery alone. For the tissue schema we have substituted a human kidney in some of our experiments. We finally modified the above schema and made a still closer imitation of the conditions which pertain to the circulation in the arm. The artery passes through the tissue schema and is surrounded by it. The inflow tube branches and the water flows through both the tissue schema and the artery; the outflow tubes from artery and tissue schema join and pass to another length of artery placed in the same compression chamber ; this acts as the vein. We have used two such complete schemata joined in series in some of our experiments, one representing the upper arm, the other the forearm (fig. 2). Eapervment TI, We first observed the effect of circulating water through two lengths of artery—in place of one—both being placed in the same compression chamber. The water flowed through (1) the first length of artery, (2) a connecting length of rubber tubing, (3) the second length of artery, and so to the outlet (fig. 3). When the compression chamber was connected with the recording spring TO PRESSURE BOTTLE S —» TO MANOMETER TO MANOMETER Fia. 3. manometer the record showed that the maximal pulsation occurred at a lower level when the flow was through two lengths of artery (fig. 4) than when it was through one length (fig. 5). Owing to the frictional resistance in the length of tube through which the water flowed, the systolic pressure was partly spent in distending the labile first length of artery and in overcoming the frictional resistance during diastole; the second piece of artery, in consequence, had the lower diastolic The Measurement of Arterial Pressure in Man. 521 pressure, and was thus the first to be deformed and give its maximal swing. The first length of artery became taut, owing to the rise of diastolic pressure, Fig. 4. Off. On. Fie. 5. as the second length of artery was flattened. Finally, the first piece of artery was deformed by the increase of compression and gave its maximal swing. The excursion of this maximal swing, owing to the increased diastolic pressure, was smaller, and the pressure at which it occurred higher, than was the case when this length of artery was compressed by itself. 522 Messrs. M. Flack, L. Hill, and J. McQueen. In one such experiment the following were recorded :— Compression. | Deformation. | Outflow per minute. One length of artery — Ce 0 cm. H,O Nil 210 Gye 5, Nil 45 5 Begins to flatten visibly in diastole 160 85 3 Flat in systole Nil Two lengths of artery— 0 cm. H,O Nil 182 Wor 333 Second length of artery begins to deform 186 45, Second length flat in diastole 106 67 a First length of artery flat in diastole Drops 85 3 | First length of artery flat in systole Nil In Experiment I the conditions, of course, are not the same as those which pertain in the arm, for in the arm there is the capillary field with its resist- ance which precludes the pulse from reaching the veins. However, the experiment shows that the behaviour of the artery is notably influenced by the compression of a vessel placed distally to it, and therefore that the study of the compression of a length of artery placed in a simple schema does not suffice to elucidate the compression of the brachial artery in the arm. This conclusion is confirmed by Experiment II. Experiment II. We repeated Experiment I, but recorded the pressure in the tube which joined the first and second length of artery. On raising the compression to 5 em. H.O the second length of artery began to flatten in diastole, the first length became distended in diastole, and the record then showed a rise in pressure. A maximal pulse developed as the compression increased. i Off. On. Off. On. Fic. 6. The Measurement of Arterial Pressure in Man. 523 Finally the first length of artery flattened (fig. 6, A). On repeating the experiment upon the first length of artery by itself we found it began to flatten and give a maximal pulse at 38cm. H.O; the pressure then fell in the manometer and reached zero as the compression was increased (fig. 6, B). Experiment ILI. We varied Experiment I by arranging the outlet in a U-tube containing mercury, so that the resistance to outflow and diastolic pressure increased pari passu with the compression (fig. 7). The diastolic pressure being thus TO PRESSURE BOTTLE —— TO MANOMETER Dorie — fe ee Mia sa Ge Fic. 7. raised part passu with the compression, the maximal pulsation appeared just before the point at which the artery was flattened in systole. The result under these conditions was, of course, the same whether one length or two lengths of artery (or artery, tissue schema, and vein) were used. Their walls were equally stretched and made more and more rigid by the rising diastolic pressure. There was thus little loss of systolic force as the pulse passed along the tubes. On the other hand, the pulse transmitted to the compression chamber and recorded by the manometer became less as the arterial wall became more rigid. When the diastolic pressure which pertained in the schema, was over-topped, the length or lengths of artery began to flatten, and a maximal pulse resulted. MacWilliam and Melvin* conclude from their study of the simple schema that the diastolic pressure and maximal pulse do not always coincide. By our experiments we bring into play the effect on the artery of obstructed venous outflow, and find the diastolic pressure and maximal pulse in agreement. * ‘Heart,’ vol. 5, p. 153 (1914). 524 Messrs. M. Flack, L. Hill, and J. McQueen. For let us consider the arm when it is compressed. The veins and capillaries under the armlet are first flattened and some are emptied, the pressure rises pari passu with the compression in all the remaining patent blood-vessels enclosed by the armlet. This must be so, for their outlet is obstructed. In the forearm, peripheral to the armlet, the venous pressure will steadily rise, and the veins become more and more swollen and tense if the compression is maintained just below the arterial pressure. Under these conditions the arterial blood can flow into the forearm, so long as capillary fields, hitherto empty, open out and the veins swell, the forearm becoming more and more congested. If a second armlet be put just below the first, and the pressure raised equally in the two armlets, the conditions are made the same as in Experi- ment II, in so far as the peripheral resistance in the arm is concerned. But, be it noted, as the blood has other arterial pathways open to it in the rest of the body, the diastolic pressure in the arteries enclosed by the upper armlet is not raised nearly up to the systolic pressure. To make the conditions in that the same as in Experiment II, the resistance to outflow in all the arteries would have to increase part passu with the compression of the arm. In the case of the vessels enclosed by the lower armlet under these condi- tions the diastolic pressure is raised nearly up to the systolic pressure. Now, we have determined experimentally that the reading of systolic pressure taken with an armlet round the calf is raised by placing a second armlet round the thigh and raising the pressure therein to, and keeping it at, say, 50 mm. Hg. This correspondingly increases the diastolic pressure in the veins and arteries of the leg, and the arteries, being made more rigid thereby, conserve better the crest of the systolic wave in its passage from thigh to calf. Similarly, the pulse in the radial artery increases in amplitude at first when the compression is raised in an armlet placed round the upper arm, because the compression by obstructing the outflow and making tenser the arteries aids the conduction of the systolic wave. Hxpervment IV. A single length of artery was compressed and the outflow measured. The compression chamber was connected with the manometer.- When the com- pression reached 25 cm. H.O the artery began to flatten. At 34 cm. H:0 the pulse became maximal, and the water then issued in strong pulses; 212 cc. flowed out in one minute. At 47 cm. H.O the water issued in shorter pulses, for the artery remained deformed for a longer period during each diastole; the outflow was reduced to 166 c.c. At 70 cm. H,O the outflow was reduced to feeble spurts synchronous with the systoles ; while at 77 cm. The Measurement of Arterial Pressure in Man. 525 H,0 the outflow ceased to pulse and was reduced to fast drops. The artery then appeared flattened all along its length, but its end proximal to the pump was slightly expanded by each systolic wave. At 87 cm. H2O drops still escaped from the outflow nozzle. Experiments on excised arteries have been undertaken* to test the correctness of the obliteration method of measuring the systolic blood pressure, and the complete cessation of outflow has been taken as the index of obliteration. Wrong conclusions have thus been drawn as to the power of the arterial wall to resist compression. The disappearance of the pulse at the distal end must be taken as the index of obliteration, not the absolute cessation of outflow. Experiment V. The flow was through (1) a length of artery, (2) tissue schema, (3) a second length of artery acting as vein. All these were placed in the same compres- sion chainber, and this connected to the recording manometer (fig. 8). The = J _— Fie. 8. tissue schema was not tightly packed with chopped sponge, and the pulse travelled through it to the vein. On compression the vein first flattened and gave a maximal pulsation, while the artery became taut, then the tissue schema shrank. The outflow at this period was partly due to the water expelled from it. The recorded pulse now became very small as the whole system (artery, tissue schema, vein) was raised up to the diastolic pressure and approximated to a rigid system. Finally, the diastolic pressure was overtopped in the artery, and this gave a maximal pulse (fig. 9). * Herringham and Womack, ‘ Brit. Med. Journ.,’ 1908, B, p. 1614. 526 Messrs. M. Flack, L. Hill, and J. McQueen. 65 60 50 40 10 5 0 cm. H,0. Let tpsaaiea Serer “hn Hy Hy} f 4 HEA oe Artery Vein On. maximal pulse. maximal pulse. Fre. 9. The following were the outflows at each stage :— Compression. Outflow. Compression. cm. H,O. - c.c. per min. 0) 146 5 71 Vein beginning to deform during each diastole. 10 63 Maxima! pulse of vein. 40 33 Tissue schema shrinking. Manometer scarcely pulses at all. 60 aL Artery beginning to deform during each diastole. 65 2 | Maximal pulsation of artery. Experiment V elucidates the behaviour of the brain when compressed. When the brain is compressed by fluid forced into the subdural cavity the capillaries, venules, etc., similarly shrink, the pressure rises in these vessels, and the whole cerebral vascular system approximates to a rigid system and gives a small cerebral pulse. Similarly, when the armlet compresses the arm, part of the blood contained in the tissue vessels is expelled and the remaining patent vessels approximate to a rigid system, in which arterial pressure pertains and through which a diminished flow continues until these are emptied; the artery itself is then flattened; that is, when the systolic pressure is overtopped. Experiment VI. In Experiment VI the flow was arranged through the artery and the tissue schema placed in separate compression chambers. These chambers were connected with each other and the manometer (fig. 1). A. On compression the tissue schema first shrank, then the artery began to flatten and the maximal pulse resulted. On decompression the maximal pulse was more ample and occurred at a lower level than on compression (fig. 10). Like results were obtained when the tissue schema was replaced by the 927 The Measurement of Arterial Pressure in Man. 10 ‘TL SM ‘OL “Sl vO 528 Messrs. M. Flack, L. Hill, and J. McQueen. kidney, the length of artery being connected to the renal artery and the renal vein to the outflow. The compression chambers being connected with each other, a part of the pulsatile force transmitted through the artery to its chamber is conveyed to the chamber of the tissue schema and helps to pulse fluid out of the tissue schema. When decompression is begun the tissue schema is shrunken, and it takes time to fill out. The outflow is in drops while the expansion is going on. The pulsatile force transmitted through the artery to its chamber is now less spent on the shrunken schema, for this is a more rigid structure. On the other hand, the pulse transmitted directly along the artery to the tissue schema spends part of its force in expanding the shrunken tissue schema. So long as the tissue schema acts as a rigid structure and stores little of the systolic force, the diastolic pressure in the artery will fall to lower level, and, in consequence, the pulsatile swing will be bigger. The recorded pulse is the summation of that from either chamber, the pulse of the artery, and of the tissue schema. There is a certain degree of expansion of the tissue schema, which favours the development of a maximal pulse, the stage when the arterial pulse is spent least on, and reinforced most by, the tissue schema. If the compression and decompression be done in stages, and time be given between each stage for the tissue schema to shrink or expand, then the maximal pulse occurs at the same height on decompression as on compression. B. The share which the tissue schema takes in the phenomena is shown in fig. 11. The artery was replaced by a rubber tube (rigid). On com- pression the tissue schema shrank and gave a maximal pulse. On decom- pression the tissue expanded, and the recorded pulse in this case became smaller because the pulsatile force was spent largely on the expansion of the tissue schema. Phenomena of the same order happen when the tissues of the arm are compressed by the armlet, and hence arise those differences between com- pression and decompression readings of systolic pressure which are so often recorded. Hapervment VII. The flow was through a length of artery and the tissue schema placed in series (fig. 1). Each was in a separate chamber, and these were joined together and to the compression bottle. The tube connecting the artery and tissue schema was joined to the recording manometer. The Measurement of Arterial Pressure in Man. 529 A, Compression of both artery and tissue schema (fig. 12). The pressure first rose and the pulse amplitude diminished owing to the shrinkage of the Off. Fig. 12. On. tissue schema, and the greater resistance thus developed in it, and higher diastolic pressure consequent in the artery. Then a fall of pressure accom- panied by maximal pulses resulted owing to the artery flattening during diastole. Finally the artery shut up and the pulse ceased to reach the mano- meter. On taking off the compression the pressure and outflow did not return to their previous amounts until the shrunken tissue schema had expanded. B. The compression tube leading to the tissue schema chamber was closed. On compressing the artery it began to flatten, and gave maximal swings, and then shut up. No rise of pressure occurred as in “ A,” because the tissue schema was not compressed (fig. 13). mu Off. Fie. 13. On. C. The compression tube leading to the artery chamber was closed (fig. 14). On compressing the tissue schema the pressure rose and the pulse yt Hui Pa Off. On. Fig, 14. 530 . Messrs. M. Flack, L. Hill, and J. McQueen. amplitude diminished; the tissue schema shrank and became more rigid as the resistance to flow was increased. The artery was not itself compressed in this experiment, but acted as a rigid tube in its closed chamber. Experiment VILL. The flow was through two lengths of artery joined in series. Each length was placed in a separate compression chamber. Tubes connected the two chambers with each other and with the manometer. A. The tube leading from the compression chamber of the vein was closed. The artery alone affected the manometer. Aleta stateless <~— 24 14:1 15-4 = | == a — | 25 16-0 17°9 = = = “= ee 8°6 | 9-4 = — — | 27 10°8 12°5 = | = — — Mm i 2 550 Messrs. J. F. Twort and L. Hill. Pulmonary Considering the slight physiological differences which may arise, ¢.g., from posture, exposure to cold during the collection of samples, etc., we cannot expect to get closer results. To carry out work the subject grasped a spring ergograph and squeezed it 20 times a minute; the sample was collected immediately at the end of two minutes period of work. The average of 22 analyses taken when resting and breathing air quietly is 9°5 per cent., breathing oxygen quietly 11:8 per cent. The average of analyses when working and breathing air quietly is 57 per cent., while that when working and breathing oxygen quietly is 5°6 per cent. The corresponding resting analyses in the same subjects gave 9°4 per cent. when breathing air quietly, 10:0 per cent. when breathing oxygen quietly. Table I1.—Effect of Warming One Arm. Subject. Arm not warmed. | Arm warmed with bath. | Remarks. per cent. per cent. 1 9-4 14°8 Air quietly breathed. 2 4:3 11°7 =| 7 3 10-0 11°6 4 ss 4 10 °3 12-2 Oxygen quietly breathed. 5 12 6 14-2 Air forcibly breathed. 6 15-2 16 4 Oxygen breathed quietly. The average of the six experiments is 10°3 per cent. for the unwarmed and 15°5 per cent. for the warmed arm. Table I11.—Hffect of Forcibly Breathing Air or Oxygen. Subject. | Forcibly breathing air. | Forcibly breathing oxygen. | per cent. per cent. 1 142 14°8 | 2 11°3 11°8 3 13 °5 131 4 12°9 12°4 | 5 9-2 9°8 6 facil 16°8 i 141 14 °1 8 14:2 16°0 9 11°6 a 10 15 2 = 11 130 = | 13 | 146 = The average of 13 analyses of samples taken when forcibly breathing air is 13°5 per cent., and of eight breathing oxygen forcibly 13°6 per cent. To sum up, then, the average results of the analyses are :— Ventilation and Oxygen mm the Venous Blood of Man. 551 | | | Dae S| Cubic Centimetres Oxygen. | analyses. | | | 22 9°5 resting, breathing air quietly. 22 -11°8 resting, breathing oxygen quietly. 7 5°7 working, breathing air quietly. 7 5°6 working, breathing oxygen quietly. 3 7°9 resting, breathing air quietly, arm not warmed. 3 12-7 resting, breathing air quietly, arm warmed (to produce vaso-dilatation). 3 12°7 resting, breathing oxygen quietly (2), air forcibly (1); arm not warmed. 3 14°2 resting, breathing oxygen quietly (2), air forcibly (1) ; arm warmed. 13 13 °5 resting, breathing air forcibly. 8 13°6 resting, breathing oxygen forcibly. Our subjects were medical students and laboratory servants. Some of them, when lying on the couch and breathing quietly, gave us low, and some high readings. The difference is an individual one, and cannot be ascribed to errors in technique, for he who gives a low reading when resting gives a low reading when working. Moreover the deep breathing readings are uniformly high. Some of our subjects were emotionally affected by the operative procedure and breathed about 10 litres a minute, while others breathed only 5-6 litres, while resting on the couch. We cannot, however, ascribe the higher reading in all cases to the ampler breathing. All we can affirm is that quiet breathing gives us a certain proportion of low readings in the given number of subjects, while deep breathing gives us uniformly high readings. Looking at the difference between the average figures for resting and breathing air quietly and breathing oxygen quietly it might be assumed that this was due to the oxygen simply dissolved in the blood according to the law of partial pressures. If pure oxygen were breathed we might expect a little over 2 per cent. O2 to be simply dissolved, and the tissues, using this oxygen first, would dissociate the hemoglobin less by the same amount. The subjects were breathing not 100 per cent. but about 80 per cent. of oxygen, so the amount simply dissolved would not be quite as much as 2 per cent. When, however, we compare the figures obtained during forcible breathing of air or oxygen, we see no evidence of any excess of oxygen due to simple solution under the increased partial pressure of this gas. Similarly in a very careful series of analyses of cat’s blood recorded by Buckmaster and Gardner and obtained by means of the Topler pump we see no evidence of any increase in oxygen of the arterial blood due to the breathing of oxygen in place of air. The average of 13 analyses, the cats breathing air, was 14:2 per cent., breathing oxygen 14°9 per cent. 552 Pulmonary Ventilation and Oxygen in Venous Blood of Man. The theoretical oxygen capacity determined from the hemoglobin value was about 17 per cent. These authors say :— “From the experiments it is a fair conclusion that during its passage through the pulmonary capillaries the blood is rarely fully saturated with oxygen even when oxygen is inhaled. For an explanation, it is probable that parts of the lung, for example the apices, are imperfectly ventilated, and also, since the circulation time in the lung is only about five or six seconds, that complete equilibrium is not attained between the blood and alveolar air.”* We know that anything over 75 per cent. of an atmosphere of oxygen when continuously breathed produces pneumonia, and that exposure to two or three atmospheres of oxygen causes convulsions. A high partial pressure and concentration of oxygen on the blood acts asa poison. It may be that there is at work some mechanism which prevents, within certain limits of oxygen partial pressure, the over-concentration of free oxygen in the blood, and therefore we find no more oxygen in the venous blood on forcibly breathing air than on forcibly breathing oxygen. Our figures show that forcible breathing of air, or oxygen, equally and notably increases the oxygen in the venous blood above the average result obtained when breathing air quietly. We cannot ascribe this result to vaso- dilatation and accelerated flow through the arm produced by the forced breathing, for G. N. Stewart has shown that forcible breathing diminishes the velocity of flow in the hand by about 40 per cent.t Forcible breathing mechanically interferes with the circulation and the hand tends to become pale and cold when such is continued. We conclude that the arterial blood is not always saturated with oxygen during the passage through the lungs when the breathing is quiet. Some parts of the lung may remain unexpanded, and the blood passing through these parts is not oxygenated. Forcible breathing ensures the expansion of all parts and the better saturation of the arterial blood. In one case Caske and Barcroftt obtained a sample of arterial blood and found it 94 per cent. saturated with oxygen. The sample was obtained from a young woman acting as donor in a direct transfusion of blood. Her artery was opened under local anesthesia. The emotional conditions probably ensured in her a good pulmonary ventilation. If it be true that the person engaged in sedentary occupation does not expand the lungs sufficiently to arterialise the blood in all their parts, this * ©Roy. Soc. Proc.,’ B, vol. 85, p. 56 (1912). + ‘Amer. Journ. Physiol.’ vol. 28, p. 190 (1911). { ‘Proc. Physiol. Soc.,’ ‘Journ. Physiol.,’ vol. 47, p. xxxv (1914). Intracranial Ganglion on Oculomotor Nerve in 8. canicula. 553 may be a contributory cause of a lessened immunity to the organism of disease such as phthisis. Our results too confirm the need for caisson workers not to rest during - decompression but to take exercise and to breathe deeply so as to secure the escape of nitrogen, which has been dissolved in their body fluid during their work in compressed air. On the Occurrence of an Intracranial Ganglion upon the Oculo- motor Nerve in Scyllium canicula, with a Suggestion as to us Bearing upon the Question of the Segmental Value of Certain of the Cranial Nerves. By Gero. E. NicHouis, D.Sc., Beit Memorial Fellow (Zoological Department, King’s College, London). (Communicated by Prof. A. Dendy, F.R.S. Received January 21,—Revised March 1, 1915.) During the study of a number of elasmobranch brains made in connection with my work on Reissner’s fibre, I noticed, in a specimen of Sceylliwm canicula, a collection of ganglion cells upon a length of nerve lying freely beneath the mid-brain. This particular brain had been sectioned in the longitudinal vertical plane and the ganglionic mass occurred at a place which corresponded with the level of the third cranial nerve. Further examination showed that these cells were undoubtedly related to the oculomotor nerve. They are situated upon it ina scattered group which, beginning at a point about 1'4 mm. from the superficial origin of the nerve, stretches to its severed end (roughly 1:6 mm. from its origin). The cells, though only about 15 in number, are moderately large (averaging 20u~x18y) and are apparently unipolar or bipolar. Their distribution suggested that other cells of the group must have existed distally to the point of severance of the nerve. Upon the opposite side of the brain the corresponding nerve had been cut away quite close to its superficial origin, when the brain was removed from the cranium. A second specimen of S. canicula in which some 2 mm. of the third nerve had been left attached to the brain, on either side, showed the ganglion well on both nerves. 554 Dr. G. E. Nicholls. Intracranial Ganglion upon the This specimen had been sectioned in the transverse plane and is the one from which the text-figure has been drawn. The cells are seen on either side, Ag. Syl. vA fas.Ing. m. ganglion N. Oculomotor Lower part of a Transverse Section through the Brain of Scylliwm canicula, taken at the level of the origin of the oculomotor nerves, and showing some of the ganglion cells slightly diagrammatically. (Outline with camera lucida. x 14 approx.) at about the same distance from the superficial origin of the nerve, in a group extending between points roughly 15 mm. and 2 mm. from that origin. On the one side some 50 odd ganglion cells were counted, while upon the opposite side there were about 70. The exact numbers cannot be certainly stated, for it is probable that in some cases a cell may have been reckoned twice, appear- ing, as they do, each in several adjacent sections. They were accompanied by smaller cells, the nuclei of which stained much more deeply, the whole forming a quite obvious ganglion. My remaining series of S. canicula brains, in every case, showed the third nerves cut at a point nearer to the brain than that at which the ganglia might be expected. Another series, therefore, was cut transversely specially for this investiga- tion, care being taken to remove the brain with as much as possible of the third nerves attached. The ganglionic masses were found well marked and were composed of cells of the same character and in about the same number as in the case described and figured above. The discovery of even a smal] ganglionic mass related to the third cranial Oculomotor Nerve in Scyllium ecanicula. 555 nerve was so unexpected that I was led to examine other elasmobranch brains of which I had sections, with a view to determining whether the condition in S. canicula was peculiar to that species or whether the ganglion was of normal: occurrence but had escaped observation. The examination was without result, however, for in a number of specimens of Acanthias, Raia, and Rhina I found no trace of ganglion cells upon the third nerve root, but it cannot be asserted that in these species the ganglia are absent, for in all of these brains the third nerve had been severed, it was found, comparatively close to its origin. A search for these cells upon the third nerve of Amphibia was somewhat more fruitful. In Rana esculenta a cluster of half a dozen large ganglion cells was found upon the nerve on one side of my single specimen. In &. temporaria and Molge sp. an odd cell or two appeared in the sections of the nerve. In each case the cells were found at or very near to the severed end, and it is highly probable that still other cells had existed in the distal part of the nerve. The existence of these ganglia in S. canicula has not, I believe, been recorded hitherto. This is, perhaps, not surprising, for the head of the adult animal is far too large to be readily cut in its entirety, and, as my own experience illustrates, where a study of the central nervous system is the prime object, brains are likely to be removed from the head by the cutting of the third nerves fairly closely to their superficial origin. And, in sections of the entire heads of embryos of this species, of several different sizes, which I examined, I was unable to distinguish the future ganglion cells among the numerous Cellular elements present in the developing third nerve, although some of my specimens were sufficiently advanced to show ‘the elements of the trigeminal ganglion already fairly well differentiated. Although I have seen these ganglia in but three specimens of S. canicula I believe that they will prove to be actually of invariable occurrence in this species. That “degenerated” ganglion cells occur upon the oculomotorius in man has long been known, having been recorded by Thomsen in 1887.* They were said to occur in the adult human subject as patches of altered tissue which, Thomsen stated, were the remains of the encapsulated ganglion cells observed by him in the root of the oculomotor in the new-born child. Gaskell (89), two years later, confirmed the discovery of these degenerated cells in the third nerve of man and recorded the finding of similar cell masses in the root of the fourth and sixth nerves also.. He seems to have assumed that these cells were always merely vestigial structures, but he nevertheless attached considerable importance to the discovery. On the strength of the transient * Vide Tozer, 12. 556 Dr. G. E. Nicholls. Intracranial Ganglion upon the existence of these cells he put forward the hypothesis that “both third and fourth nerves are in themselves complete segmental nerves of the type which Balfour supposes to have been the original type ‘when mixed motor and sensory posterior roots were the only roots present’; that then owing to some change which occurred during the past history of the vertebrate the sensory parts of these two nerves degenerated and their place was taken by the sensory elements of the fifth nerve.” Two other early observers, Reissner and Rosenthal,* are also said to have recorded the existence of cells in the root of the third cranial nerve. Accord- ing to Sherrington, the former suggested that these cells were connected with the sympathetic system. Of the occurrence of (presumably) actively functional ganglion cells upon the oculomotor root, in numbers comparable to those which I have found in S. canicula, the only record, so far as I can discover, is contained in the paper by Miss Tozer (12). This author found ganglion cells in (or near) the roots of the third, fourth, and sixth cranial nerves of young Macacus rhesus and of a teleost (Gadus virens) and in the third and sixth roots of the common pigeon. ‘ It would appear, then, that oculomotor gangliat are present, either as functional structures or as vestiges, in widely separated vertebrate classes. Since ganglia, other than sympathetic, are known to occur normally only upon “sensory ” nerves, or if upon mixed (motor and sensory) nerves, then upon the dorsal (sensory) root only, a question at once presents itseli— What is the significance of the occurrence of ganglia upon the oculomotor root ? In seeking an answer to this question it will be necessary to discuss the generally accepted belief in the homology of the oculomotor nerve with a ventral spinal root in the light of our present knowledge of («) the composition of this nerve and (0) its relation, in development, to the ciliary ganglion. The Comparison of the Oculomotor uith a Ventral Spinal Nerve Root. That the oculomotor has, of recent years, been accepted by the great majority of authors as a purely motor nerve, composed only of somatic motor components and equivalent simply to the ventral root of a typical segmental nerve is common knowledge. Neal (14), the most recent contributor to this question, has insisted upon the correctness of this interpretation. He * Vide Sherrington, 94a, p. 254. + These ganglia (which occur on the roots of the oculomotor just within the brain case) must not be confused with the ciliary ganglia which have sometimes been referred to as oculomotor ganglia. The ciliary ganglia, of course, lie in the orbit and, although connected with the oculomotor (and ophthalmicus profundus) nerves, are generally recognised as being ganglia properly referred to the autonomic (sympathetic) system. Oculomotor Nerve in Scyllium canicula. 557 concludes* “the evidence of its histogenesis and its central and peripheral relations so strongly support the supposition that it is a somatic motor nerve, as the majority of morphologists have believed, that the acceptance of the latter seems unavoidable.” While Neal is undoubtedly justified in claiming that this view is that held at the present time by the majority of observers, yet, as he points out, this has not always been the case. In the past, many morphologists have main- tained that the oculomotor is serially homologous with the segmental nerves. Comparatively recently this earlier view has been supported by Gast (’09) upon embryological grounds. In arriving at his conclusion, however, Neal appears to be altogether unaware of the hght which has been shed upon this question by the results of the experimental work of Sherrington and Tozer. The Composition of the OculomotorNerve.—Sherrington (’94), as the result of certain experiments upon the eye-muscle nerves, came to the conclusion that the oculomotor might prove to be “sensori-motor” (afferent-efferent). He repeated and laid stress upon this suggestion in a later paper (’97). Herrick (’99, p. 230) noted that medullated nerve fibres of two kinds were to be recognised in the oculomotor nerve of a bony fish (Mendia). He remarked that muscle-spindles were said not to occur in the eye-muscles but suggested that the more slender nerve fibres, whose origin he failed to determine, might, nevertheless, form part of a sensory mechanism analogous with muscle-spindles. Recently the existence of muscle-spindles in the extrinsic eye-muscles has been demonstrated by Sherrington and Tozer (10). Not only so, but the related spindle nerves were stated to pass into the central nervous system by way of the oculomotor. This, upon the generally accepted interpretation of the oculomotor nerve as merely a ventral root, is altogether anomalous, for the nerve fibres from muscle-spindles in all other muscles are derived from ganglion cells of the dorsal spinal ganglia and thus are connected with the central nervous system only by way of the dorsal (posterior) nerve roots. The experimental work carried out by Miss Tozer (12) upon Macacus showed that lesion of the third nerve peripheral to the ganglion cells (7.¢., at a point between the ganglion cells and the muscle-spindles) resulted in the alteration of the ganglion cells and the complete degeneration of the muscle- spindles; whereas, when the lesion was effected centrally to the ganglion, somey at least of the muscle-spindles persisted. * Op. cit., p. 105. t It should be noted that while only a few muscle-spindles are said to have persisted, in Macacus only a few ganglion cells are normally present. 558 Dr, G. E. Nicholls. Intracranial Ganglion upon the This is not only definite confirmation of Sherrington’s suggestion that the third nerve is sensori-motor, but it also points to a connection between these ganglionic cells and afferent fibres from sensory end-organs (the muscle- spindles). It thus affords evidence that these cells are of a type normally occurring elsewhere only upon the dorsal roots of segmental nerves. In summing up, Miss Tozer remarks* “The great variation in the number of these cells renders an explanation of their nature and function at present impossible. The cellsin Gadus are possibly sufficiently numerous to represent the source of the afferent fibres .... but in Macacus the number of these cells seems to be insufficient to do this.” If the latter part of the statement is correct, some of the afferent fibres presumably have an intracerebral origin. Since the oculomotor nuclei are in close relation to several sensory centres through the mediation of the fasciculus longitudinalis, there is nothing improbable in the suggestion that some of the afferent fibres may pass directly, by the fasciculus longitudinalis, to one of these centres. The Relation, in Development, of the Oiliary Ganglion to the Oculomotor.— Carpenter (06, p. 192) has demonstrated that while, in the embryo chick, the ciliary ganglion arises, in part, from cells which have migrated from the ophthalmicus profundus ganglion, it is also, to some extent, derived from neuroblasts which pass peripherally along the fibres of the oculomotor nerve to the ciliary ganglion. In view of this statement I was, at first, inclined to believe that the oculo- motor ganglia which I had observed were merely cell masses destined primarily for the sympathetic system but which had been arrested at this place in their transit to the ciliary ganglion. Such an explanation, however, merely introduces another difficulty, viz., the apparently anomalous origin of the ciliary ganglion. In a typical segmental nerve, sympathetic ganglia are generally believed to arise by the migration of neuroblasts from the spinal ganglion upon the dorsal root.t The alternative view that sympathetic ganglia arise, either wholly or in part, by the migration of medullary cells along a ventral nerve root has met with little acceptance. The former view is that favoured by Neal. In an earlier work, that author maintained (03) that, although the cells in the anlagen of the spinal ventral roots had a medullary origin, yet these had nothing todo with the formation of neuraxons. They are differentiated solely into neurilemma elements. In his recent work he notes (14, p. 54) that this fact has never been called in question but that, on the contrary, it has been confirmed by several subsequent workers. Concerning Carpenter’s statement that neuroblasts pass along the oculo- * Op. cit., p. XV1. + Cf. Johnston, ’07, p. 206. Oculomotor Nerve vn Scyllium canicula. 559 motor to the ciliary ganglion, Neal says (14, p. 73) “If this conclusion be confirmed—and this has been done by Belogolowy—it appears that at least some of the medullary cells in the oculomotorius anlage are neuroblastic.” Belogolowy (710) apparently only infers, however, that the cells of medullary origin in the anlage of the oculomotorius are those which later enter the ciliary ganglion. Gast (09) believes that, in the embryo, there exists a “root-ganglion” upon the oculomotor nerve, the cells of which are of medullary origin. Concerning this, Neal points out the difficulty of recognising with certainty a nerve cell among the mass of cellular elements in the anlage of the nerve, a difficulty to which I have already alluded.* Moreover, many of these cells of medullary origin are known to become altered later into neurilemma elements, as do the similar cells in spinal nerves. Carpenter has observed all stages in the differentiation of such medullary cells into neurilemma, in the chick. Neal records that he, too, has seen such a transformation, in embryo Squalus. His statement (14, p. 74) is, “. . . in Squalus, it is possible to demonstrate that a large number, if not all, of the cells present in the anlage of the oculomotor become differentiated as neurilemma cells.” In this connection Neal reviews the evidence bearing upon the question: To what extent, if at all, do emigrating medullary elements contribute to the formation of the sympathetic ganglia related to spinal nerves? The work of Kuntz(11) (which Neal remarks is “the latest presentation of the case in favour of the medullary origin of some of the elements of the sympathetic”) comes in for somewhat severe criticism. Neal concludes that convincing evidence “that cells of somatic motor nerve anlagen in Squalus... migrate into the anlagen of the sympathetic is wanting. The assertions ot Kuntz in this connection appear quite unconvincing.” As the result of his own observations Neal expresses the opinion (14, p. 57) that “the evidence . . . seems rather to favour the view that the sym- pathetic anlagen receive their cellular elements largely, if not exclusively, from the sensory ganglia, as inferred by investigators upon all classes of vertebrates from Schenck and Birdsall to Held and Marcus.” Thus, in contributing to the formation of the sympathetic (ciliary) anlage the oculomotor would, as a ventral root, appear to be altogether exceptional. In the anlage of the oculomotorius, however, there are present cells which are derived from the neural crest. The statement that free cell migration takes place from the mesocephalic ganglionic mass towards the oculomotor anlage has been made by numerous * Vide p. 555, supra. 560 Dr. G. E. Nicholls. Intracramal Ganglion upon the observers. Recently Carpenter (’06), Gast (09), and Belogolowy (10) have confirmed this, but the two latter authors suppose that these migrant neural crest cells simply become neurilemma and supporting cells in the oculo- motor root. Johnston (05, p. 244) believed that the cells of the ciliary ganglion arose from a part of the neural crest distinct from that which gave rise to the profundus ganglion. Neal denies that this is true for Squalus, but makes a very interesting statement concerning the origin of the ciliary ganglion in that form. “... the first clusters of cells associated with the anlagen of these nerves (the oculomotor and the trochlear) are derived from the neural crest. These cell clusters, in their relations and—in the case of the ciliary — in their adult structure, appear to be sympathetic. Their derivation from the neural crest favours the inference that the sympathetic anlagen of the trunk have a similar origin” (14, pp. 58-59). From all of which, indefinite and conflicting as it appears, three facts emerge :— (i) that, in the anlage of the oculomotor, cells are found derived by migra- tion (a) from the medulla, and (0) from the neural crest; (11) that certain of these cells in the oculomotor anlage migrate into the anlage of the ciliary ganglion, precisely as do cells from a typical dorsal ganglion into a typical sympathetic ganglion ; Qu) that the weight of evidence appears to be against the belief that cells of medullary origin contribute to the formation of sympathetic ganglia. The inference is that the cells which pass along the oculomotor to the ciliary ganglion must have been derived, in the first instance, from the neural crest. The Comparison of the Oculomotor with a Complete Segmental Nerve. In view of what has been said it will be obvious that, in certain particulars, the peripheral relations of the oculomotor are not exactly those of a ventral root. Nor does it appear that the histogenesis of this root (in its relation to the neural crest and the ciliary ganglion) corresponds, in every particular, with that of a spinal ventral root. In attempting to arrive at a correct interpretation of the homology of the oculomotor we must take into account the following facts :— (i) that this nerve is undoubtedly “ afferent-efferent ”; some or all of its afferent fibres being related to sensory end-organs (muscle-spindles) ; (ii) that, upon its root, in some forms, occurs a ganglion or scattered ganglion cells; Oculomotor Nerve in Scyllium canicula, 561 (iii) that, in such forms it is prohable that the afferent fibres are wholly, or in part, derived from these ganglion ceils ; (iv) that, in development, ganglion cells from the neural crest enter into association with the anlage of the oculomotor, apparently establishing a transient dorsal root ; (v) that, in development, certain ganglion cells migrate peripherally along the fibres of this nerve to the (sympathetic) ciliary ganglion, such cells having probably arisen primarily from the neural crest. Now in a typical segmental nerve (in the general acceptation of the term) we find a dorsal root bearing a ganglion and a ventral root without ganglion cells. These roots unite at, or just peripheral to, the ganglion. From the ganglion, or from the common nerve distal to the ganglion, there arises a branch which enters a ganglion of the autonomic nervous system. The cells of this latter ganglion are derived predominantly, if not exclusively, in embryonic life, from cells which migrate from the spinal ganglion situated upon the dorsal root of the related segmental nerve. ‘The nerve fibres of the dorsal root are, in general, derived from the cells of its dorsal ganglion and are described as afferent (somatic and visceral sensory) fibres, transmitting centripetal impulses. The nerve fibres of the ventral root are of intraspinal origin (arising from the ventral and lateral cornua of the spinal cord), and are efferent (motor), transmitting only centrifugal impulses. In a comparison of the condition observed in the oculomotor with that defined as typical for a segmental nerve we cannot but be struck with the fact that the oculomotor appears to combine most of the features of dorsal and ventral roots of the typical segmental nerve. That it may lack certain nerve components, normally present in a typical segmental nerve, is not denied, nor is a distinct dorsal root recognisable. On the other hand it presents features altogether unknown in any typical segmental ventral nerve root. The Absence of Certain Components from the Oculomotor.—The suggestion that there has occurred in this nerve a cenogenetic atrophy of certain com- ponents is neither new nor improbable. Indeed, Johnston has pointed out (07, pp. 151-153) that, in the nerves of that region of the head occupied by the lateral eyes, the absence of the general cutaneous component of the sensory system might reasonably be expected. Similar reasoning might connect the absence of visceral sensory components with a possible dis- appearance of visceral tissue in the region of the more anterior somites. With the disappearance of these nerve elements from the nerves of the segments occupied (or encroached upon) by the lateral eyes and the serially 562 Dr. G.E. Nicholls. Intracranial Ganglion upon the following “accessory optic vesicles,” the dorsal roots would undergo con- siderable diminution in size and importance. The dorsal ganglia, from which most of the remaining afferent fibres would have arisen, would consist of comparatively few cells (the large contingents of cells, normally present in ganglia upon nerves supplying regions of less specialisation, being altogether wanting). Gast has stated that a transient dorsal root, related to the oculomotor, appears in development, and considers that this and other eye-muscle nerves represent segmental nerves from which the sensory elements have altogether disappeared. Neal, dissenting entirely from Gast’s speculation, remarks (14, p. 105): “The supposed demonstration of the participation of sensory elements in the genesis of the oculomotor is one that would satisfy only on the basis of a strong presumption in its favour.” I would suggest that the actual occurrence, in Macacus rhesus, Columba livia, Gadus virens, and Sceyllium canicula, of numerous ganglion cells in the oculomotor roots, apparently related to afferent nerve fibres connected with sensory end-organs, supplies the “strong presumption” which Neal requires. Neal continues, “The position of the nidulus of the oculomotor and its peripheral distribution create a strong presumption against the assumption. Spindle-shaped cells lying in the mesenchyma between the profundus ganglion and the oculo- motor nerve are not necessarily neuroblasts. Spindle-shaped cells may be found almost anywhere in the mesenchyma. Even if it be admitted that the evidence that these cells are in the process of migration towards the oculomotor anlage is convincing, Gast does not know their fate. They may form neurilemma or they may enter the sympathetic or what-not.” Is it not reasonable to suppose that the cells of the oculomotor ganglion may represent some of the sensory elements which take part, according to Gast, in the formation of the oculomotor anlage ? If others of these cells migrating to the oculomotor should subsequently “enter the sympathetic” as Neal suggests, it would strengthen rather than weaken the case for regarding the oculomotor as a segmental nerve. Neal himself states (14, p. 57) that neural crest cells do come into connection with the oculomotor anlage and do apparently, amongst other destinations, arrive ultimately at the sympathetic (ciliary) anlage. Indeed, Neal makes use of this fact to support his contention that spinal sympathetic ganglia have an origin from dorsal spinal ganglia, by analogy with this development of the ciliary ganglion. The Displacement of Dorsal Roots——The dorsal roots of cranial nerves practically always shift ventrally from their primary position, during the Oculomotor Nerve mm Seyllium canicula. 563 course of development. This ventral dislocation may easily have been carried to a greater length in the case of a greatly reduced oculomotor dorsal root and have resulted in a confluence of both dorsal and ventral roots into a single structure. The Structure of the Oculomotor unlike that of Ventral Spinal Loots—The existence in the oculomotor of sensory fibres (related to muscle-spindles) derived in part, if not entirely, from ganglion cells upon the root of the nerve is absolutely incompatible with the interpretation of the oculomotor as homologous with a ventral spinal nerve root. That some of the afferent fibres may have (as Tozer implies) an intracerebral origin does not affect the ease. Sherrington (97a, p. 210) states that, in mammals, none of the fibres of dorsal (posterior) spinal roots have an intraspinal origin. For many vertebrates, other than mammals, however, it appears to be established that afferent fibres with an intraspinal origin do exist. Thus the possible occurrence of afferent fibres of intracerebral origin in the oculomotor does not lessen the resemblance of this nerve to a true segmental nerve. What is of the utmost importance is the indisputable fact that afferent fibres enter the brain by the root of the third nerve. Of the spinal or typical segmental nerves Tooth concludes (92, p. 783): “The posterior roots are the only points of entrance of sensory, or, more broadly, centripetal impressions.” The explanation which appears to be most in agreement with the facts is that the oculomotor is not correctly viewed as the equivalent of a ventral root only. Rather we must accept it as the homologue of a complete segmental nerve, containing elements of both dorsal and ventral roots, although some of these components have apparently become obsolete, and the distinction of the originally separate dorsal and ventral roots has disappeared. , Against this hypothesis, which, it should be noted, is not quite identical with that put forward by Gaskell and earlier observers, three objections may be raised. Equally with that earlier view it is opposed to the generally accepted interpretation of the ramus ophthalmicus profundus as the dorsal nerve root of the oculomotor neuromere. In the first place, then, if may be urged that, in ontogeny, the oculomotor and profundus nerves are very intimately related, and that the connection of the latter nerve and the neuromere of the oculomotor (v, according to Johnston) is only lost relatively late in development, when the profundus acquires a new connection with the brain (im neuromere vii) through the mediation of the trigeminal root. Indeed, the ramus profundus retains its relation to the third nerve, throughout life, by the ciliary nerve (radix ciliaris longus) and ganglion. VOL, LXXXVIIIL—B, 2X 564 Dr. G. E. Nicholls. Intracranial Ganglion upon the It seems to me that the only inference which can safely be drawn from the observed shifting of the ophthalmicus profundus is that its change of relations from the third nerve to the fifth may recapitulate a change which took place comparatively late in the history of the development of the vertebrate head. It does not, however, justify the assertion that the relation of the ophthalmicus profundus to the third nerve was necessarily primary. The mesocephalic neural crest has a considerable antero-posterior extension and other connections between the neural crest and the oculomotor have been observed, anterior to that existing between this nerve and the ophthalmicus profundus. That the trigeminus, the ophthalmicus profundus, the eye-muscle nerves, and the nervus thalamicus, should be found related and more or less fused, perhaps shifted, or even become obsolete, is little to be wondered at, occurring as they do in a region where shifting and obliteration of myotomes has, admittedly, been such a marked feature. Nor is it surprising that there remains little or no evidence, in ontogeny, of the primary arrangement in serial independence of the nerves of this region, for the changes which took place in connection with the development of the eye must have been some of the very earliest to disturb the serial arrangement of the nerves. In the nerves of the branchial region we have, apparently, a nearly parallel case. There, although we find the several branchial nerves connected from the earliest developmental stages, yet it 1s generally accepted that these nerves were primarily independent and serially distinct. It is assumed that their displacement and fusion was a feature acquired so early in the develop- ment of the vertebrate head that the prior condition no longer occurs in an abbreviated ontogeny. Bearing upon this hypothesis that the ophthalmicus profundus does not represent the dorsal root of a segmental nerve to which the oculomotor would stand merely in the relation of a ventral root, an interesting point may be noted. In certain elasmobranchs, of which Scyl/iuwm is one, the ophthalmicus profundus is said to be little developed* and in the adult to be absent.t It is precisely in Scylliwm, where the oculomotor is found retaining its ganglion, that the encroachment of the ophthalmicus profundus is thus least in evidence. As a further objection, the connection of the ciliary ganglion with both the oculomotor and the ophthalmicus profundus nerves might be adduced, on the assumption that this is homologous with the relation, in the spinal region, of * Sedgwick, 05, p. 135. + Parker and Haswell, ’10, vol. 2, p. 161, footnote. Oculomotor Nerve m Scyllium canicula. 565 a sympathetic ganglion to both the ventral and dorsal roots of its related spinal nerve. Neal, discussing the question of the relation of the oculomotorius to the ramus profundus, decides (14, p. 102), “there appears to be no insuperable objection to the view that the ophthalmicus profundus is serially homologous with spinal somatic sensory nerves.” While this is readily conceded, it falls short of establishing that the ramus profundus necessarily represents the sensory root in the oculomotor neuromere. Neal continues: “The comparison of the profundus nerve with spinal somatic sensory nerves is still further strengthened by the evidence of the relations with the ciliary ganglion, which have been found above to be those of a somatic motor nerve to a sympathetic ganglion. The facts which prove the sympathetic character of the ciliary anlage have already been stated above and need no restatement. The ciliary ganglion of Squalus is to be regarded as partly, if not exclusively, a sympathetic ganglion. So that in its relations with a sympathetic ganglion the oculo- motor forms no exception in the series of morphologically similar somatic motor nerves.” From what we know of the development of the ciliary ganglion, however, this relation may, quite as reasonably, be interpreted upon the hypothesis of the segmental distinctness of the two nerves. In accordance with which, I suggest that the ciliary ganglion is to be regarded as the product of the fusion of sympathetic ganglia related to at least two segmentally distinct cranial nerves. Such a condition is paralleled in the spinal region in the cervical ganglia, for example. Even if we admit the correctness of Krause’s view that the ciliary ganglion has a double nature (containing, in addition to its sympathetic elements, the bipolar cells of a cerebro-spinal ganglion), the comparison of the oculomotor nerve with a typical segmental nerve would not be affected. We should merely recognise that certain cells, migrating to the ciliary ganglion along the fibres of the oculomotor and ophthalmicus nerves, which were hitherto supposed to be simply sympathetic cells, were, in fact, sensory cells. It may well be that the ciliary ganglion is not strictly homologous in all species. The oculomotor ganglion has been observed in but comparatively few species, and, while its presence will probably be revealed by further investigation in many other forms, yet it is scarcely credible that it can have been completely overlooked in many types which have been carefully studied. In such forms, then, the cells of the oculomotor ganglion may have migrated into the proximity of,-or even into actual fusion with, the ciliary ganglion, which would thus have the double character claimed for it by Krause. On the other hand, in those species in which a distinct oculomotor 2x 2 566 Dr. G. E. Nicholls. Intracranial Ganglion upon the. ganglion persists, the ciliary ganglion may prove to be composed strictly of sympathetic elements. Herein, perhaps, lies an explanation of the contradictory observations which have been recorded, and the diverse opinions expressed in the controversies concerning both the nature of the ciliary ganglion and the segmental value of the oculomotor nerve. The fact that ganglion cells are known to occur upon ventral spinal roots, constantly in the cat,* and occasionally in man and monkey, must not be overlooked. These cells are not, however, associated with afferent fibresT nor related to sensorial end-organs. What their nature and function may be has not yet been explained, but they are clearly not comparable with the cells found in the roots of the eye-muscle nerves. Neal (14, p. 58) remarks, in connection with the migration of medullary cells into the oculomotor and trochlear nerves, that a similar migration is observed in the case of the abducens which has no related sympathetic anlage, and that in this case these cells can have no destination other than the neurilemma. In the specimens which I have examined, ganglion cells were apparently absent from the roots of the fourth and sixth nerves. From the results obtained by Gaskell, Sherrington and Tozer, however, it would seem that both of these nerves also are sensori-motor, and have ganglion cells upon their roots in some species. In the case of the trochlear nerve, Neal points out that, in development, it is closely associated with fragments of the neural crest and is related to a transient sympathetic ganglion. The abducens undergoes considerabie dislocation, and has lost all trace of any relation to a sympathetic ganglion anlage, if such ever existed. Nor has any connection between this nerve and the neural crest been recorded. Nevertheless, it would seem that the arguments adduced in favour of the segmental value of the oculomotor would apply, in the main, to all the eye-muscle nerves, making allowance for the progressively greater reduction and displacement which has occurred in the two more posterior nerves. In conclusion, I would submit that the occurrence in the oculomotor of afferent nerve fibres (conducting centripetally impulses arising in sensorial end-organs), and of ganglion cells upon the root of the nerve almost certainly related to these afferent fibres, taken in conjunction with the part which this nerve plays in the development of the ciliary ganglion, constitutes evidence in favour of the complete segmental character of the nerve too important to be ignored. Neal, who upholds a view opposed to this, says * Schafer, ’80, p. 348. + Sherrington, ’94a. Oculomotor Nerve m Scyllium canicula. 567 that, in his opinion, “ the demonstration of the serial homology of head and trunk metameres depends largely upon the proof of the resemblance of eye-muscle and spinal somatic motor nerves.’ It seems to me that the demonstration ot the resemblance of the oculomotor (and other eye- muscle nerves) to a complete spinal nerve (including sensory as well as motor roots) would have even greater value in establishing the segmental character of the head metameres. Thus, while I dissent from Neal’s state- ment just quoted, I am altogether in accord with his further statement concerning the eye-muscle nerves that “failure to convince morphologists of their meristic homology with spinal nerves would tend to undermine the foundations of the traditional conception of the head.” I desire to take this opportunity to acknowledge my indebtedness to Prof. Dendy for valuable advice and criticism, and to Profs. Elliot Smith and J. P. Hill for kindly directing my attention to the literature of the subject. LITERATURE. ‘10, Belogolowy, G., ‘Zur Entwicklung der Kopfnerven der Vogel,’ Moscow, 1910. 06, Carpenter, F. W., “ The Development of the Oculomotor Nerve, the Ciliary Ganglion, and the Abducent Nerve in the Chick,” ‘ Bull. Mus. Comp. Zool.,’ Harvard, 1906, No: 172: *89, Gaskell, W. H., “On the Relation between the Structure, Function, Distribution, and Origin of the Cranial Nerves; together with a Theory of the Origin of the Nervous System of Vertebrates,” ‘ Journ. Physiol.,’ vol. 10 (1889). 09, Gast, R., ‘Die Entwicklung des Oculomotorius und seiner Ganglien bei Selachier- embryonen,” ‘ Mitth. Zool. Stat. Neap.,’ vol. 19 (1969). 99, Herrick, C. J., “The Cranial and First Spinal Nerves in Menidia,” ‘ Journ. Comp. Neur.,’ vol. 9 (1899). 05, Johnston, J. B., “The Morphology of the Vertebrate Head from the Viewpoint of the Functional Divisions of the Nervous System,” ‘Journ. Comp. Neur.,’ vol. 15 (1905). ‘07, Johnston, J. B., ‘The Nervous System of Vertebrates,’ London, 1907. ‘11, Kuntz, A., “ The Development of the Sympathetic Nervous System in Certain Fishes, ‘ Journ. Comp. Neur.,’ vol. 21 (1911). 03, Neal, H. V., ‘The Development of Ventral Nerves in Selachii,” ‘Mark Anniv. Vol., 1903. 14, Neal, H. V., “The Morphology of the Eye Muscle Nerves,” ‘Journ. Morph.,’ vol. 25 (1914). ’80, Schafer, E. A., “‘ Note on the Occurrence of Ganglion Cells in the Anterior Roots of the Cat’s Spinal Nerves,” ‘ Roy. Soc. Proc.,’ vol. 31 (1880). 05, Sedgwick, A., ‘Student’s Text-book of Zoology,’ vol. 2, London, 1905. 94, Sherrington, C. S., “On the Anatomical Constitution of the Nerves of Muscles,” ‘Physiol. Soc. Proc.,’ June 23, 1894, ‘ Journ. Physiol.,’ vol. 17. 94a, Sherrington, C. 8., “On the Anatomical Constitution of the Nerves of Skeletal Muscles ; with Remarks on Recurrent Nerves in the Ventral Spinal Nerve Root,” ‘Journ. Physiol., vol. 17 (1894). 568 Miss D. J. Lloyd. ‘97, Sherrington, C. S., “ Further Note on the Sensory Nerves of Muscles,” ‘Roy. Soe. Proc.,’ vo]. 61 (1897). ‘97a, Sherrington, C. 8., “ On the Question whether any Fibres of the Mammalian Dorsal (Afferent) Spinal Root are of Intraspinal Origin,” ‘Journ. Physiol.,’ vol. 21 (1897). 10, Sherrington, C. S., and Tozer, F. M., “Receptors and Afferents of the IIIrd, IVth, and VIth Cranial Nerves,” ‘ Roy. Soc. Proc.,’ B, vol. 82 (1910). ‘92, Tooth, H. H., “On the Relation of the Posterior Root to the Posterior Horn in the Medulla and Cord,” ‘ Journ. Physiol.,’ vol. 13 (1892). ‘10, Tozer, F. M. See Sherrington and Tozer. 12, Tozer, F. M., “On the Presence of Ganglion Cells in the Roots of Third, Fourth, and Sixth Cranial Nerves,” ‘Physiol. Soc. Proc.,’ July 27, 1912, ‘ Journ. Physiol.,’ vol. 45. The Osmotic Balance of Skeletal Muscle. By Dorotuy JorDAN LLoyp. (Communicated by W. B. Hardy, F.R.S. Received February 24, 1915.} Fletcher was the first to follow continuously, for any considerable length of time, the change in weight of a muscle immersed in a hypotonic solution.* He found that the muscle at first increased in weight and then decreased. In isotonic solution the muscle “neither gains nor loses weight.” This amounts to a definition of an isotonic solution. The changes in weight of the gastrocnemius or sartorius, the muscles used by Fletcher, are slow, owing to the low value of the ratio of surface to volume. Very early in this work therefore it was decided to use a thin flat muscle sheet. The sternocutaneous muscle of the frog was fixed upon. It reaches its maximal intake from a hypotonic solution in from 5 to 20 minutes according to the concentration of the solution and the state of the muscle. In the case of so small a muscle it is possible that all the fibres are nearly in the same state at the same time. This cannot be the case with larger muscles. The central fibres of, for instance, the sartorius may be irritable whilst the external fibres are in water rigor. Complex physical and physiological reactions between the fibres must occur and complicate the problem. An obvious disadvantage of a muscle with a large surface is the magnitude of the error in weight due to variation in the quantity of moisture adherent to the surface. The surface was always dried quickly by filter- paper before weighing, and the smoothness of the curves of variation of * * Journ. Physiol.,’ vol. 30, p. 414 (1904). The Osmotic Balance of Skeletal Muscle. 569 weight with time is, I think, sufficient proof that the error was reduced to about 1 per cent. of the total weight. According to the definition of an isotonic solution given above, such a solution does not, strictly speaking, exist. A muscle may remain steady within the limit of error of weighing for periods up to half-an-hour, but sooner or later measurable variations of weight appear. In other words the muscle removed from the body is a changing system. It is a question whether the apparently steady periods are not really periods of very slow change. It is noteworthy that different workers have fixed upon solutions of sodium chloride over such a wide range as from 0°6 to 0°8 per cent. as being isotonic with frog’s muscle; and the only curve given by Fletcher of a muscle in an isotonic solution shows a steady rise in weight. A muscle simply removed from the body is called by Fletcher a resting muscle. The use of the term is inadvisable. Such a muscle has suffered a certain amount of mechanical disturbance, and in the process of pithing the frog it has been thrown into tetanus lasting a minute or more, at a time when the circulation of blood is defective. Such a muscle is best indicated by the term untreated muscle. The weight changes characteristic of an untreated muscle in solutions of the sugars biose, dextrose, sucrose, and raffinose between the concentrations zero to 0:27 molecular are shown in fig. 1, which shews a curve for . 1605 : 140 120 100 /4M Sucrose SIEM SuCTOSe 80 Fic. 1.—Abscissz = hours from beginning of experiment ; Ordinates = weight of muscle expressed in percentage of initial weight. 570 Miss D. J. Lloyd. 0:14 molecular sucrose.* There is a rapid intake of water followed by loss, the weight often falling much below the initial weight. This curve should be compared with the curve of change of weight in hypotonic (0:10 molecular) Ringer, given in fig. 2. The curve characteristic of a concentration higher than 0:27 molecular is 140 120 5 Ars. 00 = Fic. 2. also given in fig. 1. The initial intake not only is not present but is not even represented by a variation in the rate of loss. The relation throughout is simply lineay. The fact that the loss of weight in the more concentrated solutions is in linear relation to time is of interest. The osmotic equivalent of an untreated muscle exposed to a solution such that it takes in water, clearly undergoes a change since the intake gives place to a loss in weight usually greater than the previous gain. Does the osmotic equivalent of the surviving muscle spontaneously change or is the change just mentioned due to the exposure to a hypotonic solution? The lnear form of the curve of loss in a hypertonic solution suggests that the variation of state is due to the influence of the solution. The linear form of the curve also would imply that the loss is due to a change in the state of the muscle, for if it were merely the establishment of an osmotic balance with a fixed effective mass of solute within the muscle. the rate would diminish as the effective concentration within the muscle approached that outside it. Both above and below a certain concen- tration the progress of change bears the character, not of the simple establish- ment of osmotic equilibrium between two solutions initially of different con- centration, but of the response of a labile system to an external change of state. * It must be remembered that a 07125 molecular solution of sodium chloride (taken by most writers as isotonic) is osmotically equivalent to a 0°23 molecular solution of a sugar. The Osmotic Balance of Skeletal Muscle. 571 The region between 0-21 molecular and 0°27 molecuiar for the sugars is one in which the initial intake may or may not appear. The muscle either at once loses weight after a short period of slight change, or shows a typical - initial intake followed by a typical loss. The variation over the region may provisionally be attributed to a variation in the state of the muscle due to mechanical disturbance. For instance, in dissecting the muscle out it 1s subjected to a varying amount of tension, and this will tend to produce passage of fluid from the interior of the fibre to the lymph space or vice versa, and pithing the frog causes fairly prolonged twitching when the blood flow is poor. If it were possible to secure muscles in a definite physiological state this diffuse zone would probably narrow to a critical concentration, below which the initial intake would occur and above which it would vanish. Some part of the initial intake of water from hypotonic solution is unquestionably due to the mechanical disturbance of the muscle. Liffect of Oxygen.—F letcher found that the osmotic changes induced in a muscle by activity were removed by exposure to oxygen. The muscle was ? put back into the “resting” state, which was characterised by a large intake of water from hypotonic solutions. My results with the sterno- cutaneous do not readily harmonise with those of Fletcher. The effect of previous exposure to oxygen is to reduce, and finally to obliterate, the initial intake of fluid even from distilled water. In fig. 3 are two curves, (a) from a muscle put directly after removal from the body into distilled water, (>) from a muscle placed in distilled water after three hours’ exposure to moist oxygen. 120 ee | 2 3 Ars 18ie, BE It might be urged that the intake has already taken place from the water vapour during exposure to oxygen. This is not so. Muscles in oxygen and water vapour always tend to lose weight. If any intake does occur it must be so transitory and slight as to have escaped detection. Exposure to Water Vapour.—The behaviour of a muscle immersed in a Si, Miss D. J. Lloyd. solution must always be open toa variety of interpretations. The course of events depends not only upon the initial state of the muscle but also upon properties of the surfaces of muscle and muscle fibre considered as semi- permeable membranes. Further, osmotic relations are complicated by change” in the muscle due to the chemical nature of the solution. Thus, to take salts as examples, the sternocutaneous muscle in 4 molecular solution of sodium chloride usuaily remains of constant weight for some time, and then loses weight ; in a similar solution of potassium chloride there is an immediate and prolonged rise in weight followed by a fall; in an isosmotic solution of calcium chloride there is a very small and fleeting initial rise in weight followed by a long and steady fall which may reduce the muscle to 60 per cent. of its original weight. It is possible, however, to examine the osmotic balance of a muscle by exposing it to water vapour of varying pressure. The gas space round the muscle then acts as a theoretically perfect semi-permeable membrane so far as non-volatile solutes are concerned. The method adopted was to suspend the muscle in a flask over a solution of known concentration, the gas in the flask first having been shaken thoroughly with the solution and then left at the desired temperature for some hours in order to attain equilibrium. The muscle was removed for each weighing, and results therefore are affected by an error due to loss of water during weighing, and loss of vapour from the flask during removal and replacement. Control experiments with fine plates of agar saturated with water gave a maximal loss of 3 per cent. in four hours. In fig. 4 are shown curves of the weight changes of muscles suspended in oxygen above the plane surface of (0) distilled water, (¢) C06 molecular 120 7 100 80 601 IDives, 4b The Osmotic Balance of Skeletal Muscle. 573 Ringer’s fluid, (d) 0°125 molecular Ringer’s fluid, and (¢) in air above 0:13 molecular Ringer’s fluid, and («) in hydrogen over distilled water. In oxygen saturated with aqueous vapour over distilled water the weight: may either remain constant for a variable period and then fall, or start falling at once. That is to say, oxygen obliterates the intake from vapour which occurs in air or hydrogen. There is here at first sight a contradiction. If the untreated muscle has a vapour pressure less than that of water—as would appear from the curve 2, fig. 4—why does it not at first condense vapour when in oxygen? Oxygen cannot instantly remove the condition which leads to the intake. A reason may probably be found in the nature of the diffusion column which must be formed at the free surface of the muscle and of each fibre. At the free surface are water vapour and oxygen. Consider a superficial shell at the surface. If the presence of free oxygen within this shell either raises its vapour pressure to that of water, or maintains its vapour pressure at that level, the quantity of water vapour taken up by the shell will depend upon the ratio of the rate of diffusion of water vapour to that of oxygen. The diffusion column of oxygen progressively retards the diffusion of water vapour, so that, even if their diffusion rates initially were equal, that of the water vapour would rapidly tend to vanish. An analogous retardation is that seen when a retardation in the dissipation of heat or diffusion of impurities from the face of the solid arrests the solidification of over-cooled liquids.* In hydrogen saturated with water vapour the muscle gains in weight. Strictly speaking, the gas was air very much diluted with hydrogen and saturated with water vapour. In this the sternocutaneous maintains its irritability for about six hours. | What change in the muscle is it which is caused by excess oxygen and which raises the vapour pressure? The work of Fletcherf and of Fletcher and Hopkins suggests that exposure to oxygen reduces the concentration of metabolites (such as, for instance, lactic acid), and so raises the vapour pressure. This would accord with the view of Ranke and of all who have followed him, that the intake of water by a fatigued muscle is due to the production during activity of chemical substances of low molecular weight. The secondary fall in weight of muscles due to loss of water was ascribed by Fletcher to a “loss of the semi-permeable character of the fibres,” just as an ordinary osmometer whose membrane is not completely impermeabie * Wilson, ‘ Phil. Mag.,’ (5), vol. 50, p. 238. + ‘Journ. Phys.,’ vols. 23 and 28. t ‘Journ. Phys.,’ vol. 35, p. 247 (1906-7). 574 The Osmotic Balance of Skeletal Muscle. to a solute shows first an intake of solvent and later, as the solute escapes, a loss. If this were the sole cause, then the secondary loss of water would be due, either directly or indirectly, to loss of carbonic acid, for this is the only known solute which can escape into the vapour. It must be pointed out that these experiments cannot be used to calculate the vapour pressure of the muscle substance because of the error in weighing mentioned above, and because the rise of the vapour pressure of the muscle due to curvature of the surfaces is included. Summary. 1. The sternocutaneous muscle of the frog, immersed in a hypotonic solution of Ringer’s fluid, of biose, dextrose, sucrose, raffinose, or of NaCl undergoes first a gain in weight and later a loss. In a hypertonic solution the weight falls from the start. 2. The initial gain in weight in hypotonic solutions or even in distilled water can be reduced and finally suppressed by previously exposing the muscle to wet oxygen. 3. Muscles absorb water from an atmosphere of hydrogen and water vapour, but not from one of oxygen and water vapour. In the latter a fall in weight was observed. DAS Surface Tension and Ferment Action. By E. BEARD and W. CRAMER. (Communicated by Sir Edward Schifer, F.R.S. Received March 17, 1915.) (From the Physiology Department, Edinburgh University.) The following investigations were carried out with the object of determining whether the action of a ferment on a substrate is affected by surface tension. Since in the living organism the action of ferments proceeds in a system in which surface development has reached a maximum, the problem is one of considerable theoretical importance. So far as we are aware, it has not been studied. The experimental difficulty is, of course, to allow the factor of surface tension to operate on the action of ferments in such a way that a sufficient amount of the digest can be obtained at the end of the experiment in which the progress of the ferment action could be determined. Various devices were used to attain this object. In some preliminary experiments the reaction was allowed to proceed in a capillary tube, in others in test-tubes filled with glass wool, with thin short capillary glass tubes or with glass beads. The reaction as it proceeded in these tubes was then compared with that of a control in an ordinary test-tube. A distinct effect was observed in these preliminary experiments with lipase, diastase, and yeast invertase, all of which showed a retardation. The effect was then studied in some detail in the case of invertase. The experiments with invertase were carried out as follows :—Solutions of sucrose and invertase were mixed in definite proportions. Part of the mixture was placed in a test-tube and served as a control. The rest was put into test-tubes filled with glass beads, 3 to 4 mm. in diameter. Care was taken that the level of the fluid was always well below the top level of the glass beads. The tubes were then closed with a rubber stopper and incubated at a given temperature. After a given number of hours, readings were taken with a polarimeter. In every series of experiments the control tubes were read first, so that the slightly prolonged period of incubation in the tubes filled with beads would tend to diminish any retardation that might occur. Special control experiments showed that the presence of glass beads did not affect the readings obtained with pure sucrose solutions. Similarly the readings obtained with solution of invertase alone remained constant. That the effect of the mutarotation of the glucose formed by the action of invertase could be neglected will be pointed out below. The beads were washed after each experiment for several hours, first in hot running tap water, then with distilled water, and dried in an oven at about 180° C. 576 Messrs. E. Beard and W. Cramer. The invertase was prepared from yeast obtained from Distillers Co., Ltd. Edinburgh, by two different methods :-— (1) Preparation of Invertase from Fresh Yeast (called “ Invertase F” in the following Tables)—A weighed quantity of yeast was added to a measured volume of chloroform water. The mixture was kept at a temperature of 38° for 27 hours or longer. It was then filtered. The proteins were removed from the filtrate with kaolin. The second filtrate was a clear yellow liquid, which was used as the solution of the ferment. (2) Preparation of Invertase from Dried Yeast (“Invertase D” in the following Tables).—Fresh yeast was pounded with distilled water and washed three times by decantation. It was then collected in a Buchner funnel, washed with alcohol and ether and spread in a thin layer on a glass plate. It was completely dried in a vacuum desiccator. When dried the material was ground in a mortar, placed in an oven at 40° and gradually warmed during half an hour up to 100°. The brown powder then obtained was kept in a stoppered bottle. When a solution of invertase was required, weighed amounts of the brown powder were added to chloroform water and kept for about 24 hours at a temperature of 38°. After filtration a clear yellow liquid was obtained, which represented a solution of invertase. Eight different invertase preparations, all of which were strongly active, were used in these experiments. In the Tables the various ferment preparations are designated (1) by the letter D or F, indicating whether they have been prepared from dried or fresh yeast respectively, (2) by the date of their prepara- tion, and (3) by the percentage of yeast (fresh or dried) used with reference to chloroform water. It was found that increasing the surface led to a distinct retardation in the inversion of cane sugar by invertase if the concentration of the substrate was . relatively high, and that of the ferment relatively low. With a high ferment concentration the retarding effect was not noticeable. The following experiments are given as examples :— Experiment A. 23.7.14. Invertase F, 22.7.14, 100 per cent. Sucrose solution 19'6 per cent. Temperature 18°. 50 c.c. sucrose: +5 ¢.c. invertase. +1 c.c. invertase. +0 ‘2 c.c. invertase. Hours. = = Control. With beads. Control. With beads. Control. With beads. | 0 +11 °88 +11°88 +1293 +12 93 +13 :02 +1302 22 — 3°21 — 3°21 + 4:93 + 5:13 +11 -06 +11°78 AT — 0°25 — 0°03 + 9:03 + 9°71 96 — 3:16 > — 3°08 + 5:80 + 7:09 Surface Tension and Ferment Action. 577 Since in all the following experiments relatively small concentrations of invertase were used, so that the process of inversion was slow, the effect of the mutarotation of glucose did not affect the readings. This was verified by © special control experiments in which dilute sodium carbonate was added to the solution before the readings were taken. The retarding effect of surface tension was confirmed in a number of other experiments. It is not necessary to quote these experiments in detail, as the effect is evident in the experiments given below which deal with the further analysis of the phenomenon. In the action of an enzyme on a substrate one may distinguish two separate phases: firstly, the combination of the enzyme and the substrate, and, secondly, the chemical change which proceeds in this compound of enzyme and substrate. A priori surface tension may have an effect either on the first phase or on the second phase or on both. We will consider first the possibility of surface tension action on the second phase of enzyme action. Just as the catalytic action of an enzyme has been likened to a lubricant producing its action by diminishing friction, so the effect of surface tension might be like that of a brake, retarding the process by increasing the friction. If that were so, one would expect to find that retardation disappears when the brake is removed, 7.c. when surface energy is reduced to the dimensions which obtain in the control. To test this point a number of test-tubes filled with beads were charged with the mixture of substrate and ferment. After a given number of hours, when a distinct inhibition had become noticeable, the fluid (or part of it) was removed from one of the test-tubes filled with beads, and inversion allowed to proceed as usual in a test-tube without beads. An example will make the arrangement clear. Experiment B. 22.6.14. Invertase D, 18.5.17, 10 per cent. 2°5 ¢.c. invertase and 100 ¢.c. sucrose, 20 per cent. Temperature of digestion 27°. Without Beads. With Beads. ~ Hours. ———-——_____+ ———> SS s Control tube. Tube la. Tube 2a. Tube 3a. Tube 1. Tube 2. Tube 3. Tube 4. 0 +1290 +12°90 +12°90 +12°90 +12-90 10 +12°14 j +12 ‘59 = — — 24, +11°37 +11°80 i + 11°88 — — 49 +9°26 +10°25 +10 °42 j + 10°57 — 98 +580 +7 °66 +7°75 +808 +847 In this particular experiment the inhibition disappeared only partially after reducing the surface energy. It is still evident after the removal of beads, 578 Messrs. E. Beard and W. Cramer. but not so marked as in the test-tubes in which the surface remained extended. Similar results were obtained in a number of other experiments, with other preparations of invertase. A very different result was obtained, however, in the following experiment carried out with the same preparation of invertase but at a higher tempera- ture and with a slightly lower ferment concentration. EHzpervment C. 15.6.14. Lnvertase D, 18.65.14, 10 per cent. 1 cc. invertase and 100 c.c. sucrose solution, 19°5 per cent. Temperature 42°. Without beads. With beads. ge > Cage BS aN Hours. Control tube. Tube 1a. Tube 1. Tube 2. (9) +12 63 + 12°63 +12 -63 18 — + 12°47 — 43 +10°12 SS +12 °45 + 66 + 8°98 +12 °22 +12 °29 90 + 7°80 +12 :09 +12 °30 Here the inhibition in the system in which the surface has been extended retains its full strength after the surface has again been reduced to the dimensions of the control. The opposite condition was realised in only one experiment. Experiment D. 20.5.14. Invertase F, 23.4.14, 25 per cent. 10 c.c. invertase and 100 c.c. sucrose, 9°8 per cent. Temperature 18°. Without beads. With beads. ie AS = Hours. Control. Tube 1a. Tube 1. 0 +5°77 +5°77 13 +4 °44 +493 17 +396 : +471 37 +212 +340 43 +162 +3 “03 +3°25 48 +1°21 +267 as 64 +0 °32 +1°89 +247 Here the inhibition is almost completely removed when surface energy is reduced. The conclusion to be drawn from these experiments is that the retarding action of surface tension on the inversion of cane sugar by invertase is made up of two components. One component leaves the system cane sugar-invertase Surface Tension and Ferment Action. bio unchanged when surface tension ceases to operate. The action of this component would be in accordance with the first alternative suggested above, namely, that surface tension acts asa brake on the chemical process which proceeds in the substrate under the influence of the ferment. ‘The action of the second component involves a permanent alteration in the system cane sugar-invertase. How is the action of this component to be interpreted ? We have considered hitherto only the possibility that surface tension inhibits the second phase of an enzyme action, namely, the chemical change which proceeds in the substrate after it has become combined with the enzyme. We may now examine what effect surface tension could have on the first phase of an enzyme action—the combination of enzyme and substrate. The question is then: Is it a priori possible that surface tension could inhibit the combination of enzyme and substrate, and if so, how can this possibly be tested experimentally ? It is a well known fact that ferments tend to go into the surface layer, so that the concentration of the ferments is higher on the surface layer than in the remainder of the solution. This property of ferments is expressed by the statement that ferments are “surface active.” By increasing the surface, therefore, more ferment will be driven into the surface layer. Cane sugar, on the other hand, is practically not surface active, and its distribution in the solution will, therefore, remain unaltered when the surface is increased. Theoretically, therefore, it would seem possible that by extending the surface a certain amount of invertase is driven into the surface and thus prevented from combining with the cane sugar. The point is capable of being tested experimentally. If new surfaces are created the ferment is, as just stated, driven into the surface layer. If, for instance, a ferment solution is shaken so that foam is formed, the concen- tration of the ferment in the foam is greater than that in the rest of the solution. If now the foam is allowed to subside, the concentration of the ferment in the solution rises again, although the original value may not be reached. But if new surfaces are created by the introduction of solid substances, a second phenomenon may come into appearance. The ferment not only goes into the surface layer bounding the solid, but may become adherent to the solid, so that when the new surfaces are destroyed by with- drawing the solid, the ferment remains adherent to the solid and is withdrawn with it. This is the phenomenon of adsorption, which is exhibited in a marked degree by a number of ferments brought into contact with solids such as charcoal, collodion membranes, suspensions of mastix. It is not necessary to VOL. LXXXVIII.—B. 2a 580 Messrs. E. Beard and W. Cramer. discuss here the conditions on which adsorption depends. In connection with the present problem the phenomenon of adsorption is of interest only in so far as ib presupposes as a preliminary condition a concentration of the adsorbed material at the surface of the adsorbing material. It must also be borne in mind that such a surface concentration may occur without adsorption taking place, as in the case of foam, for instance, where the surface separates a gas and not a solid from the liquid. We shall, therefore, use the general term of “surface concentration” to describe the alteration of concentration produced in a system by altering the surface energy, and the term “adsorption” in order to designate the special result which is caused under certain conditions by surface concentration. If it can be shown then that under the conditions of our experiments invertase is adsorbed by glass beads, we would have evidence that the factor of surface concentration accounts, at any rate partially, for the retardation of the ferment action observed in our experiments. If, on the other hand, no adsorption is observed, no conclusions can be drawn either for or against this possibility, as one may conceive of surface concentration occurring without adsorption. To test this point, experiments were carried out, in which invertase was kept in contact with glass beads. Another portion of the same ferment preparation was allowed to stand in a test-tube by itself at the same temperature as a control. After a definite number of hours the ferment solution was removed from the glass beads and its activity compared with that of the control. The former preparation will be described in the Tables as “contact invertase,” the latter preparation as “ control invertase.” The following experiments may be given as examples of the results obtained :— Experiment B. 14.514. Invertase D, 6.5.14, 5 per cent. Kept for 93 hours at 38°, (1) in contact with beads ; (2) alone as control. 5 c.c. of each of these two preparations mixed with 5 ¢.c. sucrose solution, 5 per cent. Digested at 38°. Hours. Contact invertase. Control invertase. (0) +3 °26 +3°25 5 +3:°25 +223 27 +198 —0°86 | (Note that in this experiment the relative amount of ferment used for inversion is very large.) Surface Tension and Ferment Action. 581 Huperiment F. 16.514. Invertase Preparation same as in Previous ELapervment. Kept for 25 hours at 41°, (1) in contact with beads; (2) alone as control.” 1 c.c. of each added to 10 c.c. sucrose solution, 9°6 per cent. Digested at 18°. Hours. Contact invertase. Control invertase. | (0) +5°79 +5°70 19 +2 92 +1°99 42 +0 °44, —0°52 | oo It will be noted that there is a marked effect in both cases, and that with the longer exposure of the ferment to the beads, the disappearance of the ferment becomes more marked. The effect of temperature will be seen from the following experiment :— Experiment G. 7.7.14. Invertase D, 30.6.14. Kept for 22 hours, (1) in contact with beads at 18°, 29°, and 40°, and (2) alone as control at the same temperature. 1 c.c. of each of the six ferment solutions added to 10 c.c. of sucrose solution, 9 per cent. Digested at 30°. | 18°. 29°. 40°. | | | Hours. Cont Be Control. 9 Contact Control. Contact Control. invertase. invertase. invertase. ; | | (0) +599 +599 | +599 +599 +5°99 | +599 264 +2°73 +1°79 +2 °37* +1°67* +4445 | +3 °74t * Examined half an hour later than tubes kept at 18°. + Examined one hour later than tubes kept at 18°. This experiment is of interest because it shows, in addition to the dis- appearance of invertase owing to adsorption, a destruction of the ferment even in the control at higher temperatures (40°). This destructive effect is apparently enhanced by increasing the surface. This effect would explain the results obtained in Experiment C (15.6.14), where there was not only an almost complete inhibition of the action of the ferment, but where this inhibition persisted even after the factor of surface energy was again removed. The experiments just quoted, which were carried out with two different preparations of invertase, gave a distinctly positive result with reference to 582 Messrs. E. Beard and W. Cramer. adsorption. With two other preparations no evidence of adsorption was obtained. The following experiment is given as an example :— Experiment H. 20.514. Invertase F, 23.4.14, 25 per cent. Kept for 20 hours at room temperature, (1) in contact with beads, and (2) alone as control. 1 c.c. of each added to 10 cc. sucrose solution, 10 per cent. Digested at room temperature. Hours. Contact invertase. Control. | | (8) | +5°77 +5°77 18 +3:°98 | + 3°84 23 + 3°53 + 3°35 29 +291 — +2°79 45 +1°76 +1 °68 90 —0°31 —0°36 This experiment should be compared with the results obtained in Experi- ment D, in which the same preparation of invertase was used. In this experiment the inhibition produced by increasing the surface disappeared again when the surface was reduced to the dimensions of the control. The same absence of adsorption was observed with a second preparation of invertase. This preparation showed the usual inhibition of its action by surface tension, and asin the previous case this inhibition disappeared in two experiments when the factor of surface tension was removed. Ina third such experiment in which less invertase was used the inhibition persisted partially. General Remarks. The observations demonstrate that the action of invertase on cane sugar is retarded by increasing the surface of the system. They show further that this retardation is due partly to a surface concentration effect: the surface- active ferment is driven into the surface and thus prevented from combining with the surface-inactive cane sugar. If one looks upon the combination of substrate and enzyme as a surface concentration effect, as Bayliss does, one can readily understand that this combination can be inhibited by the same force acting in the opposite direction, so that these observations . form incidentally a confirmation of the conception formulated by Bayliss. A question which cannot yet be answered with certainty is whether the retardation observed in these experiments can be explained entirely as a surface concentration effect, or whether surface tension acts also by retarding the chemical process faking place in the substrate, that is, the second phase of ferment action. That surface tension retards certain chemical processes, for Surface Tension and Ferment Action. 583 instance, the formation of chloroform from chloral by alkali, isa well known fact. In the present case it has been found that the action of invertase on cane sugar may be retarded even although the effect of adsorption is not . noticeable under the conditions of the experiment. It has also been found that in some cases the retardation disappears almost completely when the influence of surface tension is removed after it has been operative. Observations such as these would be most readily explained by assuming that the second phase of ferment action-has been inhibited by surface tension and not the first phase. But even when there is no evidence of adsorption we cannot exclude the possibility of surface concentration occurring without adsorption. In other words it seems possible that the ferment goes into the surface layer of the liquid without becoming adherent to the solid and without being removed with latter. Such a condition would also explain the observations referred to above. We have convinced ourselves by a number of preliminary experiments that the retardation by surface tension is a phenomenon shown by other ferments _ besides invertase. The conditionsin the case of other ferments have not been studied by usin detail. But it is evident that the effect obtained may differ with the nature of the ferment, of the substrate, of the products of ferment action, and of the surface. If for instance the substrate itself is surface active. the conditions will differ markedly from those which obtain when the substrate is surface inactive. 584 Surface Tension as a Factor Controlling Cell Metabolism. By W. CRAMER. (Communicated by Sir Edward Schafer, F.R.S. Received March 17, 1915.) Although it is known that many chemical processes taking place within the cell are due to the actions of ferments, and although we can in many cases separate these ferments from living protoplasm and study their action in vitro, there still remain considerable discrepancies between the processes as we see them occur in vivo and as we study them in vitro. One of the most characteristic features of the processes taking place within the living cell is what, for want of a better term, may be called their “adaptability,” that is the delicate sensitiveness with which they respond to very slight changes in the surrounding medium by being retarded, accelerated, or reversed. It is known, of course, that the action of ferments is influenced by changes in — temperature or in the alkalinity or acidity of the surrounding medium. But in the case of the living cell these factors remain practically constant, so that their influence can be excluded. We know, too, that many reactions brought about by the action of ferments are reversible, and that the direction in which the ferments act depends upon the concentration of the various substances entering into the reaction. But here again these differences are of an order of magnitude far greater than the variations which exist in the living organism. It is noteworthy, too, that the equilibrium of a reaction brought about by a ferment separated from living protoplasm lies almost always near the point of complete hydrolysis, and contrasts in that respect markedly with the behaviour of the same ferment when its reaction is studied in the living cell, where the reverse process may be found to occur or where very minute changes in the surrounding medium are sufficient to transfer the point of equilibrium from hydrolysis to synthesis. If we take the liver cell as an example we find that the living cell can, with equal readiness, transform glycogen into glucose and glucose into glycogen. If the liver is removed from the body a marked glycogenolysis occurs, so that a marked amount of sugar is formed, while the power to synthesise glycogen appears to be almost completely inhibited. In other words, the equilibrium point les now near the point of complete hydrolysis. The same takes place if an extract of the liver is allowed to act on glycogen or on glucose in the concentrations found in the blood. It is known that the formation of glycogen im vivo occurs when the percentage of blood-sugar is Surface Tension as a Factor Controlling Cell Metabolism. 585 relatively high, and the reverse process when the blood-sugar concentration falls. But the differences in the concentration of the blood-sugar which accompany these processes are too slight to be an adequate explanation in - themselves, especially if we compare them with the large difference of con- centration necessary to effect the reversal of a ferment action in vitro. Moreover, when the liver is removed from the body the formation of such a large amount of sugar takes place that the hydrolysis of glycogen should be inhibited if the concentration of the sugar was the main factor. Nevertheless the glycogen under these conditions completely disappears. The dis- appearance of glycogen from the liver in the living animal as the result of “sugar puncture ” or under the influence of the thyroid hormone cannot be satisfactorily explained on the ground of changes in concentrations of the reacting substances. In order to explain the predominance im vivo of the synthetic power of ferments which in vitro act almost entirely as hydrolytic agents, the assump- tion has been made that im vivo the products of synthetic action by a ferment. are withdrawn, as they are formed, from the sphere of action, so that the equilibrium is always being disturbed in favour of the synthetic process. If we take the liver again as an example, we find that with the ferment acting in vitro the equilibrium point lies near the point of complete hydrolysis. That means that a very small amount of glycogen can be synthesised by the ferment, even in vitro. But since the glycogen thus formed remains in solution the reaction stops when once this point of equilibrium is reached. In the cell the slight amount of glycogen synthesised by the ferment is deposited in an insoluble form as it is formed. The equilibrium is thus disturbed and ancther slight amount of glycogen is formed. In other cases. it is assumed that the product of synthesis is removed by diffusion or excretion or carried away by the tissue fluids. Now this consideration may account for the fact that ferments which show only a slight synthetic power im vitro have a marked synthetic action in vivo. It is doubtful whether this explanation can be applied in every case. It certainly does not explain the “adaptability ” of the cell, the readiness with which the cell metabolism responds to shght changes in the environment. It does not explain, for instance, the ease with which the liver cell regulates its glycogenic function in the one or other direction, why a difference of less than Q-l per cent. in the concentration of the blood-sugar determines whether synthesis or hydrolysis of glycogen is to take place, or why the “piquure” and the thyroid hormone produce a “ mobilisation ” of glycogen, It is clear, therefore, that there are factors conditioning the actions of ferments within the cell which do not come into play when we study the 586 Mr. W. Cramer. Surface Tension as a actions of these ferments im vitro under the usual conditions, and which have therefore escaped observation. If one compares the conditions under aia chemical processes proceed am vivo with the experimental conditions under which the same processes are usually studied in vitro, one finds as the most obvious difference that in the latter case the influence of surface energy is reduced to a minimum, while in the former it is developed to a maximum. Jn vivo there are interfaces between cytoplasm and the surrounding medium, cytoplasm and nucleus, cytoplasm and deposits in the cytoplasm, besides the interfaces presented by the various colloids constituting the cytoplasm. In the conditions usually obtaining in experiments im vitro all these sources of surface energy are non- existent ; only the colloidal nature of ferment or substrate may give rise to surface-tension effects, as they do, of course, also within the cell. The question is therefore whether surface-tension effects (apart from those possibly caused by the colloidal nature of ferment or substrate) are factors conditioning the action of ferments. It was found that when this factor of surface tension was introduced by allowing the reaction to proceed in a test- tube filled with glass beads or in a capillary glass tube, so that the surface was increased, the action of invertase, diastase, and lipase was distinctly retarded. A detailed account of the results observed with invertase is given in the preceding paper. Further analysis of the phenomenon showed that the two phases which can be distinguished in the action of a ferment are probably both subject to the influence of surface tension, firstly the combination of substrate and ferment, and secondly the chemical reaction which takes place in the substrate and in which the ferment acts as a catalyst. While these observations establish the principle that ferment action is conditioned by surface tension, they can only give a faint and incomplete idea of the degree to which this factor controls the action of ferments within the cell. For, compared with the immense development of surface which obtains in the living cell and the living organism, the increase in surface energy pro- duced by the presence of glass beads in the mixture of ferment and substrate is very small. It must also be borne in mind that the experimental condi- tions deal with surface tension between glass and a watery solution, while im vivo surface-tension effects are produced between colloidal solutions of different composition, membranes, colloids in the form of gels, and so forth. Lastly, the effect of surface tension may vary with the nature of the ferment, of the substrate and of the products of ferment action, according to the “surface activity” of these substances, 7.e. their property to lower surface tension. Factor Controlling Cell Metabolism. 587 The first general conclusion which may be drawn from these considerations is that the great surface development in the cell or the organism produces conditions which markedly affect the action of ferments 7 vivo when compared with their action in vitro. We may now consider in some detail in what way this surface development exercises its effect with regard to the metabolism of the cell. We find then, firstly, that different parts of the cell present different conditions for the action of ferments. Surface tension is operative at the periphery of the cell, at the interface between cell and the surrounding medium. There the con- ditions for ferment action will differ from the conditions presented by the interior of the cell. But even in the interior of the cell, conditions in the protoplasm surrounding the nucleus, vacuoles, granules, etc., will differ from those presented by the rest of the cytoplasm. We must then conclude that ferment action will not proceed evenly throughout the cell. It may be retarded or inhibited in one part of the cell, while it is proceeding actively in another. Whether it can even be reversed as the result of surface tension is not yet clear from the experimental evidence before us. In this connection reference may be made to the work of Warburg, who has demonstrated that the oxidative processes in the cell are dependent on the structural parts of the cell and not on the fluid cell contents. We have hitherto assumed, for the sake of simplicity, that the surface tension of protoplasm is constant. But we know that it is always changing as the result of chemical processes leading to the formation or disappearance of surface-active substances. In cells with free surfaces, such as unicellular organisms for instance, these fluctuations in surface tension result in the decrease or increase of surface: they become manifest in the form of movement. Ameceboid movement and ciliary movement have long been recognised as surface-tension effects. The existence of such changes in different physiological conditions of a unicellular organism has also been demonstrated recently by MacCallum, by a micro-chemical study of the distribution of potassium salts within the cell. Similar changes must occur also in cells aggregated in cell masses or organs. But here, where the surfaces are not free, and possibly less elastic, and where the cell cannot extend or diminish its surface except to a limited extent, the result is, not movement, but alterations in the concentration and composition of the substances constituting the surface layer of the cell, and this leads to further alterations of the cell metabolism. Thus surface tension conditions cell metabolism, and is, in turn, also con- ditioned by the metabolism of the cell. Chemical changes within the cytoplasm may lead to the formation or disappearance of surface-active substances VOL. LXXXVIII.—B. 2 2Z 588 Mr. W. Cramer. Surface Tension as a or to variations in their concentrations. And it may be pointed out here that surface tension is affected by very slight quantities or changes in concentra- tion of surface-active substances, particularly when the initial concentration is low. Or interfaces may be formed, or disappear, as the result of chemical changes, or become extended or shortened, and thus affect again chemical changes, just as in the experiments with invertase the reaction was influenced by extending the surface “glass” through the watery solution. We thus get a conception how the cell regulates and controls its metabolism, how a chemical change may accelerate or retard other chemical processes which may have no chemical relation to it and which i vitro would remain unaffected by it. It is thus possible to explain the chemical organisation of the cell without having to postulate, as has hitherto been done (Hofmeister, Hans Meyer), the existence of hypothetical membranes in the cytoplasm which are supposed to separate the different chemical systems. The validity of this conception of surface tension and surface energy as factors controlling cell metabolism is capable of being tested experimentally. For, according to this idea, substances which do not affect the protoplasm chemically (that is to say are neither bases nor acids, neither reducing nor oxidising agents, etc.), but which are strongly surface active, should markedly affect the metabolism of the cell. That is actually the case. The condition of narcosis in which the metabolism of the cell is profoundly affected is brought about by substances which fulfil these postulates. They do not affect the protoplasm chemically and they are all strongly surface-active; there is even direct evidence that these substances influence the action of cell ferments. For Chiari has shown that in the excised liver of anesthetised animals the process of autolysis proceeds more rapidly than in the liver of normal animals. And Guignard has demonstrated that the effect of anzesthetics on laurel leaves is torelease the action of ferments, as evidenced by the production of hydro- cyanic acid.* * In order to avoid misunderstandings, it may be pointed out here that these substances, in addition, must be capable of penetrating into the cell. They do so by virtue of their solubility in lipoids, according to the Meyer-Overton theory, or by virtue of their “ Haftdruck,” according to J. Traube. These theories of narcosis enable us to understand quantitative differences in the strength of different narcotics or in their action on different tissues. But they do not enable us to understand how they act upon cell metabolism in such a way as to bring about the state of narcosis when they have penetrated into the cell, and it is this question which is being considered in this paper. Hans Meyer himself recognised this, and put forward as a further explanation the assumption that narcotics soften the hypothetical lipoid membranes which surround each molecule of ferment and substrate and keep them apart under normal conditions. Others (Hober, Mansfeld, Giirber) have put forward different hypotheses to explain the state of narcosis. Factor Controlling Cell Metabolism. 589 Wearrive ata similar confirmation of the relation of surface-active substances to cell metabolism, when we examine the substances which, as Loeb has shown, affect the metabolism of unfertilised eggs in such a way that they either produce cytolysis or, under certain conditions, incite artificial partheno- venesis. The substances which belong to this group—narcotics, saponin, soap, bile salts—are all substances which are strongly surface active. UE: = eats tala Sion gnoidibetyg: vit nt ian sinodes ohuignhe TO PRODUCE | ce RARE GASES BY ELECTRIC DISCHARGE. By Tuo sas R. Merton, B.Sc. (O 549 NOTE ON THE PAPER BY ice R. MERTON ON 8 ATTEMPTS * To PRODUCE THE PARE GASES BY ELECTRIC DISCHARGE.” By J. Norman Covute, 554 pe sei a Sk FOR HEAT OF METALS AT LOW TEMPERATURES. GriFFituHs, Sc.D., F.R.S, and Ezer Grirrirus, M.Sc. (Abstract) 557 DILUTE SOLUTIONS OF ALUMINIUM ant ones By a Ts Heycock, F. R S., and F. H. Nevitiz, F.R.S. (Abstract) i 560 PHILOSOPHICAL ‘—,TRANSACTIONS 1913 — 1914. SERIES A. CONTAINING PAPERS OF A MATHEMATICAL OR PHYSICAL CHARACTER. Vol. 213 (1914). Pp. 494. Price £1 6s. Vol. 214 (1914). Pp. 448. With 10 Plates. Price £1 IIs. 6d. SERIES B. CONTAINING PAPERS OF A BIOLOGICAL CHARACTER. Vol. 204 (1914). Pp. 456. With 29 Plates. Price £1 12s. 6d. Vol. 205 (1914). Pp. 492. With 43 Plates. Price £2 3s. 6d. SOLD BY HARRISON & SONS, 45, ST. MARTIN’S LANE, W.C. RECENT PAPERS IN THE ‘PHILOSOPHICAL TRANSACTIONS.’ Sold separately by Dulau & Co., Ltd., 37, Soho Square, London, W. A. 515. DILUTE SOLUTIONS OF ALUMINIUM IN GOLD. By C. T. HEYCOCK, A. 5I6. A. 517. A. 5I8. A. 519. A. 520. A. 521. A. 522. B. 319. B. 520. B. 321. B. 322. ‘'B. 323. B. 324. B. 325. F.R.S., and F. H. NEVILLE, F.R.S. (from the Goldsmiths Metallurgical Laboratory, University of Cambridge). Pp. 10; 2 Plates. Price Is. 6d. ON THE ORIGIN OF THE ELECTRON EMISSION FROM GLOWING SOLIDS. By FRANK HORTON, Sc.D. Pp. 18. Price Is. ON THE LUNAR DIURNAL VARIATION OF THE EARTH’S MAGNETISM AT PAVLOVSK AND POLA (1897-1903). By S. CHAPMAN, B.A., D.Sc., Fellow of Trinity College, Cambridge, and Chief Assistant at the Royal Observatory, Greenwich. Pp. 23. Price Is. 6d. ON THE CAPACITY FOR HEAT OF METALS AT LOW TEMPERATURES. By E. H. GRIFFITHS, Sc.D., F.R.S., and EZER GRIFFITHS, M.Sc., Fellow of the University of Wales. Pp. 39. Price 2s. 6d. ; THE SPARK SPECTRUM OF NICKEL UNDER MODERATE PRESSURES. By E. G. BILHAM, A.R.C.Sc., B.Sc., D.I.C., Research Student at the Imperial College of Science and Technology. Pp. 13; 2 Plates. Price Is. 6d. ON THE CONVECTION OF HEAT FROM SMALL CYLINDERS IN A STREAM OF FLUID: DETERMINATION OF THE CONVECTION CONSTANTS OF SMALL PLATINUM WIRES, WITH APPLICATIONS TO HOT-WIRE ANEMOMETRY. By LOUIS VESSOT KING, B.A. (Cantab.), Assistant Professor of Physics, McGill University, Montreal. Pp.60; | Plate. Price 4s. ON ACQUIRED RADIO-ACTIVITY. By Sir WILLIAM CROOKES, O.M., LL.D., D.Sc., Pres.R.S. Pp. 13; 2 Plates. Price Is. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 214. Pp. 10. Price 6d. THE DEVELOPMENT AND CERTAIN POINTS IN THE ADULT STRUCTURE OF THE STARFISH, ASTERIAS RUBENS, L. By JAMES F. GEMMILL, M.A., D.Sc., Lecturer in Embryology at Glasgow University, and in Zoology at Glasgow Provincial Training College. Pp. 82 ; 7 Plates. Price 5s. 6d. ON THE NATURE OF THE TUBES IN MARSUPIAL ENAMEL, AND ITS BEARING UPON ENAMEL DEVELOPMENT. By J. HOWARD MUMMERY. Pp. 19; 2 Plates. Price 2s. 6d. ON FOSSIL PLANTS, SHOWING STRUCTURE, FROM THE BASE OF THE WAVERLEY SHALE OF KENTUCKY. By D. H. SCOTT, For. Sec. R.S., and E. C. JEFFREY, Ph.D., Professor of Plant Morphology in Harvard University. Pp. 59; 13 Plates. Price 7s. 6d. THE EXCITATORY PROCESS IN THE DOG’S HEART. PART I.—THE AURICLES. By THOMAS LEWIS, JOHN MEAKINS, and PAUL D. WHITE. Pp. 46; 4 Plates. Price 3s, 6d. EXPERIMENTS ON INHERITANCE IN PARTHENOGENESIS. By W. E. AGAR, University of Glasgow. Pp. 69. Price 3s. 6d. TITLE, CONTENTS, INDEX, &c., Vol. 205. Pp. 10. Price 6d. CROONIAN LECTURE: ON THE ORIGIN OF MAMMALS. By ROBERT BROOM, M.D., D.Sc. Pp. 48; 7 Plates. Price 5s. Pe PUBLISHED BY THE ROYAL SOCIETY, i BURLINGTON HOUSE, LONDON. 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 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. Crown 4to, 128 pages. Price 55. THE CELEBRATION OF THE 250th ANNIVERSARY OF THE ROYAL SOCIETY OF LONDON, JULY 15-19, 1912. A reduction of price to Fellows of the Society. Sotp BY Harrison & Sons, 45, St. MARTIN’s LANE, W.C. A GENERAL INDEX (under Authors’ Names) to the ‘ PROCEEDINGS OF THE ROYAL SOCIETY.’ Vols. 1—715. Covering the Period 1800—1905. Price 5s. Half price to Fellows of the Society REPORT ON THE PEARL OYSTER FISHERIES OF THE GULF OF MANAAR. By W. A. Herpman, F.R.S. With Supplementary Reports by otiet Naturalis upon the Marine Biology. ato, OT boards. 5 Parts, 1903—1906. Price per Part, 20s.; to Fellows, 1 REPORTS OF THE SLEEPING SICKNESS COMMISSION OF THE ROYAL one Demy 8vo. ee I., 1903, Price 7s. 62. No. II., 1903, Price Ts. 6¢. No. III., 1903, Price 6s. Nez IV., 1903, Price 6s. 6d. No. , 1905, Price 3s. No. VI., 1905, Price 4s, 6d. No. VII., 1906, Price 4s. 6a. No. VII, 1907, Price 2s. 6d. No. x, 1908, Price ls. 62. No. x, 1910, Price 3s. No. XL, 1911, Is. No. XII., 1912, 5s. 6d. No. XIII., 1913, 2s. 6d. No. XIV., 1913, 2s. 6a.* * Nos. VI.—XIV., Published by Wyman anp Sons, Lrp., Ferrer Lane, E.C. REPORTS OF THE COMMISSION ON MEDITERRANEAN FEVER. 8yo. Part I., 1905, Price 3s. Part II., 1905, Price 2s. Part III., 1905, Price 2s. 6d, Part IV., 1906, Price 3s. 6d. Part v., 1907, Price 2s. 62. Part V1, 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. ne F.R. ‘s., Miss Saunpers, R. C. PUNNETT, M.A., and C. C. Hurst, F.t.s. Pp. 154. 1905. Price Ts. 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 pi Conclusive) By M. WueEtpate, D. C. E. Marrvart, aea LB. J. Sorvas. 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 comple in one volume, Price 20s. CATALOGUE OF SCIENTIFIC PAPERS, 1884—1900. Now Ready. Vol. XIII. By C. -