os a et PN OI ee arte fem ee et ee a ST tA item ty tae Bisa eandietiedh Go nk aa ae os doe CN PET e y = JOURNAL AND PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES FOR 1943 (INCORPORATED 1881) VOLUME LXXVII Parts I-IV EDITED BY THE HONORARY SECRETARIES THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN SYDNEY PUBLISHED BY THE SOCIETY, SCIENCE HOUSE GLOUCESTER AND ESSEX STREETS Issued as a complete volume, October 6, 1944 Al = 232 YRAROMOH ene \ Rt : aes ~s eA S , A. ? pee Ta ae Cie bel 7 a f£. ) eae P ig TT Pt ae eee > ¥% ; bs bi oe corn F re ry vey hohe 2M Maker Oe LEA Tee Le etek Pb ie ie “ mi ; enki’ ‘a. ae ata SURO BD | Pe i pe 8 OMUbO), aay 10 TEE DURE 3 R AG BEL CE EMV IRR LS? } Sp > nh Eek cet ine LAO Tey ee ipa 2 CONTENTS VOLUME LXXxXxVII Part I* 7 Page Art. I.—Presidential Address. By H. Priestley, M.D., Ch.M., B.Se. (Issued November 3, 1943) .. ne. i% a e wi ts a xs Me 1 Art. II:—Nova Puppis 1942. By H. W. Wood. (Issued October 22, 1943) | wits 17 Art. III.—A Polyhedral Model of the meee Plane. By F. A. Behrend. (Issued October 22, 1943) .. hes Pre bie fe ey, 20 Art. IV.—Preliminary Notes on Solution-Cracking Treatment of Torbanite. By J. A. Dulhunty, B.Sc. (Issued October 22, 1943) a a ue oe aie ve 24 Part II 7+ Art. V.—Tabulata and Heliolitida from the Wellington District, N.S.W. By O. A. Jones, M.Sc. (Communicated by Dr. Ida A. Brown.) (Issued February 9, 1944) ee 33 Art. VI.—The Etch Figures of Basal Sections of Quartz. Their Use in the Orientation of Water-worn Crystals. By F. N. Hanlon, B.Sc., Dip.Ed. (Issued February 9, 1944) 40 Art. VII.—Clarke Memorial Lecture. Australia’s Mineral Industry in the Present War. By H. G. Raggatt, D.Sc. (Issued February 9, 1944).. a on Be Sea sy Part III { Art. VIII.—Simple heh ie and Correlation. By D. T. Sawkins, M.A. (Issued February 9, 1944) .. ' A me ey ud ene ue 85 Art. IX.—The Erocuetion of Hyoscyamine from Duboisia Species. Part I. Methods of Quantitative Estimation. By J. A. Lean, M.P.S., and C. 8. Ralph, B.Sc. (Issued April 17, 1944) se 3 ay aie vie ae ae ae ae Bee LOO ArT. X.—The Production of Hyoscyamine from Duboisia Species. Part Il. Extraction of the Base. By C.S. Ralph, B.Sc., and J. L. Willis, B.Sc. (Issued April 17, 1944).. 99 Art. XI.—Kbonite as a Radiometer. The Distortion of Ebonite by Long Infra-Red Radiations. By G. G. Blake, F.Inst.P., M.I.E.E. (Issued April 17, 1944).. Stet ILOG Art. XII.—Studies on Colour Reactions for Sugars. Part I. The Identification and Determination of Monosaccharides with Thymol, Hydrochloric Acid and Ferric Chloride. By A. Bolliger, Ph.D. (Issued April 17, 1944) .. is er: sre OS Art. XITI.—The Chemistry of Bivalent and Trivalent Iridium. Part I. Compounds of Bivalent Iridium Halides with Tertiary Arsines. By F. P. Dwyer, M.Sc., and R. S. Nyholm, M.Sc. (Issued April 17, 1944) sie a id ae as ..) | 116 ArT. XIV.—Stringocephalid ogee | in Eastern Australia. By Ida A. Brown, D.Sc. (Issued April 17, 1944) ah ite sh ne aes ele) * Published January 19, 1944. 7 Published March 31, 1944. t Published May 16, 1944. AAI 1V CONTENTS. Part IV * ArT. XV.—The Vibrations of Square Molecules. Part I. The Normal Coordinates and Vibration Frequencies of Planar AB, Molecules. By Allan Maccoll, M.Sc. (Issued June 23, 1944) : aps A ar At aie = Ses cud a Art. XVI.—Further Determinations of Specialisation in Flax Rust Caused a Melampsora Inni (Pers.) Lév. By W. L. Waterhouse, M.Sc., D.Sc.Agr., D.I.C., F.L.S., and I. A. Watson, Ph.D., B.Sc.Agr. (Issued June 2a: 1944) 342 ae Art. XVII.—A Study of the Magnetic Behaviour of Complexes aap the Platinum Metals. By D. P. Mellor, M.Sc. (Issued June 23, 1944) ; Art. XVIII.—The Geology of the Cooma District, N.|S.W. Part II. The Country between Bunyan and Colinton. By W. R. Browne, D.Sc. (Issued June 23, 1944).. OspiTtuARY NOTICES TITLE PaGr, Contents, Notices, PUBLICATIONS OFFICERS FOR 1943-44 List oF MEMBERS, AWARDS OF MEDALS, ETC. ABSTRACT OF PROCEEDINGS PROCEEDINGS OF THE SECTION OF GEOLOGY INDEX TO VOLUME LXXVII *Published October 6, 1944. Page. 130 138 1X Xxi LP Seen XXXili NOTICES. iv) NOTICE. THe Roya Socrery of New South Wales originated in 1821 as the ‘ Philosophical Society of Australasia’; after an interval of inactivity, it was resuscitated in 1850, under the name of the ‘‘ Australian Philosophical Society ’’, by which title it was known until 1856, when the name was changed to the ‘“‘ Philosophical Society of New South Wales ”’ ; in 1866, by the sanction of Her Most Gracious Majesty Queen Victoria, it assumed its present title, and was incorporated by Act of the Parliament of New South Wales in 1881. TO AUTHORS. Particulars regarding the preparation of manuscripts of papers for publication in the Society’s Journal are to be found in the “ Guide to Authors,” which is obtainable on appli- cation to the Honorary Secretaries of the Society. FORM OF BEQUEST. I be yur ath the sum of £ to the Royat Sociery or NEw SovurH WaALEs, Incorporated by Act of the Parliament of New South Wales in 1881, and I declare that the receipt of the Treasurer for the time being of the said Corporation shall be an effectual discharge for the said Bequest, which I direct to be paid within calendar months after my decease, without any reduction whatsoever, whether on account of Legacy Duty thereon or otherwise, out of such part of my estate as may be lawfully applied for that purpose. [Those persons who feel disposed to benefit the Royal Society of New South Wales by Legacies, are recommended to instruct their Solicitors to adopt the above Form of Bequest.] vi PUBLICATIONS. The following publications of the Society, if in print, can be obtained at the Sociecyy s Rooms, Science House, Gloucester and Essex Streets, Sydney. Transactions of the Philosophical Society, N.S.W., 1862-5, pp. 374, out of print. Vols. I-xI Transactions of the Royal Society, N.S.W., 1867-1877 at Vol. xr Journal and Proceedings He Ae 1878, pp. 324, price 10s. 6d. Ay XIII a a me 4a Be 1879, ,, 255, a 99 XIV 99 be) be 99 99 1880, 99 391, 9? We XV 5 a Ni AA we 1881, ,, 440, ys on XVI Ae He es x AA 1882, ,, 3827, i is XVII a6 a 55 an 5 1883, ,, 324, ae ad XVIII an ve na ae 5 1884, ,, 224, ie <3 XIX ea Ae ate rs i 1885, ,, 240, na a xx a: at 4 Bs an 1886, ,, 396, oi as XXI A $5 fe cs a5 1887, ,, 296, a Ay XXII Ar Pa reer eae is Bs 1888, ,, 390, Pas an XXIII 5 i an a 35 1889, ,, 534, a a XXIV 5G ‘a a a ad 1890, ,, 290, st A XXV Aa ey 8 a a 1891, ,, 348, vt 3h XXVI ahs ne “ if an 1892, ,, 426, na AE XXVII sti ue aa we 3 1893, ,, 530, ef 55 XXVII “5 a a Ap 33 1894, ,, 368, a Ne XxIx An 5 55 at x 1895, ,, 600, ae 5s xxx a a 5 39 si 1896, ,, 568, 5 An XXXI ie “3 i a An 1897, ,, 626, ay BA XXXII A RA 5 <5 ar 1898, ,, 476, ‘i 50 XXXIII sts ae 33 59 a 1899, ,, 400, i ae XXXIV x a ae 5 iy 1900, ,, 484, a A XXKV 935 se ra 55 5 1901 Dole ae 6 XXXVI AS as AS ah as 19025" 5; ool. ae >> X&XXVII Ae Ae as ve Me 1903, ,, 663, ay >> XXXVIII ae a ce a aS 1904, ,, 604, a AMEBEND.@.0.4 b.8 A se a nA eS 1905, ,, 274,’ a 55 XL a af os - ae 1906, ,, 368, a >» XLI » 30 50 50 Bi 1907; ean ve A 2IHOE 9 oe af “ 3 on 1908, ,, 593, Be Ws XLII at Aa J Ad A 1909, ,, 466, Le 99 XLIV a An ae a a T9LO, -55. 79s ne 29 XLV 99 29 o6 $6 A 1911, ,, 611, aie 29 XLVI 9 N09 30 50 ts TPS as a7. Fe ss XLVII ae 35 35 55 . 19138, ,, 318, 5 ais XLVIII aN Bs a5 3 a 1914, ,, 584, 5 Ss XLIx an 35 ae ae 3 LOLS 45) LOST 53 2 I 99 ah Ap 5A a5 1916, ,, 362, 3 99 LI 9 29 29 a6 Lh 1917, ,, 786, 33 ” , LI » » 2 55 he 1918;..,, (624, a 99 LIII Op 39 of Bt aa 1919, ,, 414, Leb 9 LIV 9 39 35 e 35 1920, ,, 312, price £1 Is. 29 LV 29 99 na a a> 1921, ,, 418, ye 5 avi ia. 4 A be Sacu. tl O22 vie SMEs ie 29 LVII 99 be) 2° 29 99 1923, 99 421, 9 29 LVIII 29 99 29 96 55 1924, ,, 366, 55 oe) LIX 29 9 29 ” 9° 1925, Ar) 468, or) d9 Lx ” Q5 9 56 aA 1926, ,, 470, a 29 - LXI 99 99 96 90 5 1927, ,,. 492, ra Ss LXII 33 ae 3 bn 35 1928,.,, 458, oD 99 LXIII >» an 99 99 AO 1929, ,, 263, a a» LXIV = 5, 29 39 36 an 1930, ,, 484, Be 29 LXV » ”» 29 Ap An 1931, ,, 366, i 99 LXVI 99 99 33 50 aS 119325 {peenGols ue 99 LXVII 99 99 9. 30 an 1933, Ga) oul a 2s LXVIII 99 2 9 29 a 1934 ee oZe. 3 9 LXIX 9 99 7) 9 99 1935, % 288, 9 »» LXX oP) 99 99 90 BG 1936, ,, 528, aN I) LXXI 99 99 99 99 79 1937, 99 708, ry 9 LXXII 9 79 99 99 ) 1938, 9 396, ” 9 LXXTI ” 9 » 99 99 1939, ,, 344, ay 2» LXXIV ” 9 9 50 aS 1940, ,, 658, vs 3° LXXV 9 99 29 9? 9) 1941, 99 224, 99 9? LXXVI 9 39 99 99 9 1942, 99 432, 99 29 LXXVII oo) ” oe) 29 29 1943, 33. Sees 29 Royal Society of New South Wales OFFICERS FOR 1943-1944 Patrons: His EXcELLENCY THE GOVERNOR-GENERAL OF THE COMMONWEALTH OF AUSTRALIA, THE LORD GOWRIE, v.c., P.c., G.C.M.G., C.B., D.S.O. His EXcELLENCY THE GOVERNOR OF NEW SoutH WALES, THE LORD WAKEHURST, k.c.m.e. President : A. B. WALKOM, pD.sc. Vice-Presidents : A. BOLLIGER, Ph.D., A.A.C.1. Pror. H. PRIESTLEY, M.p., ch.m., B.sc.. IDA A. BROWN, pD.sc. H. 8S. HALCRO WARDLAW, D.sc., F.A.€.1. Honorary Secretaries : Pror. A. P. ELKIN, M.a., Ph.p. | D. P. MELLOR, m.se. Honorary Treasurer : A. CLUNIES ROSS, B.sc., F.c.a. (Awst.). Members of Council: R. L. ASTON, B.Sc., B.E., M.Sc., Ph.D., W. H. MAZE, M.sc. A.M.LE. (Aust.).f J. E. MILLS, M.se., Ph.p.* G. H. BRIGGS, B.Sc., Ph.D., F.Inst.P. F. R. MORRISON, 4.a.¢.1., F.c.S. J. A. DULHUNTY, B.sc. G. D. OSBORNE, D.sc., Ph.p. F. P. J. DWYER, M:.sc. H. H. THORNE, M.a<., B.Sc., F.R.A.S. F. LIONS, B.sSc., Ph.D., A.1.C. | H. W. WOOD, M:sc., A.Inst.P., F.R.A.S. * Died May 11, 1943. + Elected June 30, 1943. ~ na aalell— dtin@ ma RROL-£20) SOF 2H ney . tre ; marr en Fe ak Ute, PRCT RST Mie Pict eee yee ene Fi e 2ee me ae | 4 ; 2 hi c , ao 1 -.gaEeF { ? “fa $s #2 H Lae 4 r , 3 Me bed es 1 F « | iM s hf : 4% niet ae eos romait pe task), Ale : Hijatet) To” a aay i ae; nl ae as eee F j ’ Bis rE | “3 y . f ia «4 a as Te, ee - re : \ t * , i ; uit \ : we Ae 4 ; 1s e ae LIST OF THE MEMBERS OF THE Royal Society of New South Wales as at March 1, 1944 P Members who have contributed papers which have been published in the Society’s Journal. The numerals indicate the number of such contributions. { Life Members. Elected. 1944 Adamson, Colin Lachlan, Chemist, 22 Cremorne-street, Richmond, Vic. 1938 P 2 |fAlbert, Adrien, ph.p. Lond., B.sc. Syd., a.t.c. Gt. B., Commonwealth Research Fellow in Organic Chemistry, University of Sydney ; p.r. “ Green- knowe,’’ Greenknowe-avenue, Potts Point. 1935 tAlbert, Michel Francois, *‘ Boomerang,’’ Billyard-avenue, Elizabeth Bay. 1898 tAlexander, Frank Lee, Surveyor, 67 Ocean-street, Woollahra. 1941 tAlldis, Victor le Roy, L.s., Registered Surveyor, Young, N.S.W. 1905 P 3 |{Anderson, Charles, M.a., D.sc. Hdin., ©.M.z.S., 17 Towns-road, Vaucluse. (President, 1924.) 1909 P12 |{Andrews, Ernest C., B.A., Hon. Mem. Washington Academy of Sciences and of Royal Society of New Zealand, No. 4, “‘ Kuring-gai,”’ 241 Old South Head- road, Bondi. (President, 1921.) 1930 Aston, Ronald Leslie, B.Sc., B.E. Syd., M.Sc., Ph.D. Cantab., A.M.1.E.Aust., Lecturer in Civil Engineering and Surveying in the University of Sydney ; p-r. 24 Redmyre-road, Strathfield. 1919 By Aurousseau, Marcel, B.sc., 16 Woodland-street, Balgowlah. \ 1935 Back, Catherine Dorothy Jean, m.sc., The Women’s College, Newtown. 1924 iB yl Bailey, Victor Albert, M.A., D.Phil., F.Inst.p., Professor of Experimental Physics in the University of Sydney. 1934 jee Baker, Stanley Charles, M.sc., A.Inst.P., Head Teacher of Physics, Newcastle Technical College, Tighe’s Hill; p.r. 8 Hewison-street, Tighe’s Hill, N.S.W. 1937 Baldick, Kenric James, B.sc., 19 Beaconsfield-parade, Lindfield. 1919 Bardsley, John Ralph, 76 Wright’s-road, Drummoyne. 1939 P32 Basnett, Elizabeth Marie, mM.sc., 36 Cambridge-street, Epping. 1933 Bedwell, Arthur Johnson, Eucalyptus Oil Merchant, ‘‘ Kama,’ 10 Darling Point-road, Edgecliff. 1944 Bennett, Alwynne Drysdale, B.sc., 8 Courland-street, Randwick. 1926 Bentivoglio, Sydney Ernest, B.Sc.agr., 52 Crows Nest-road, Wollstonecraft. 1940 Betty, Robert Cecil, 67 Imperial-avenue, Bondi. 1937 P 6 Birch, Arthur John, m.sc., 15 Hilltop-road, Headington, Oxford, England. 1923 Birks, George Frederick, Wholesale Druggist, c/o Potter & Birks Ltd., 15 Grosvenor-street, Sydney; p.r. 42 Powell-street, Killara. 1916 Birrell, Septimus, 74 Edinburgh-road, Marrickville. 1920 | Bishop, Eldred George, Manufacturing and General Engineer, 37-45 Myrtle- street, Chippendale ; p.r. 17 Thompson-street, Clifton Gardens. 1938 Black, Una Annie Frazer (Mrs.), B.sc., 16 Melrose, Billyard-avenue, Elizabeth Bay. 1939 P.,3 Blake, George Gascoigne, M.I.E.E., F.Inst.p., “‘ Holmleigh,’”’ Cecil-avenue, Pennant Hills. 1933 P 22 Bolliger, Adolph, ph.p., Director of Research, Gordon Craig Urological Research Laboratory, Department of Surgery, University of Sydney. 1920 RP... Booth, Edgar Harold, M.c., D.sc., F.Inst.Pp.. New England University College, Armidale. (President, 1935.) 1939 Bk Bosworth, Richard Charles Leslie, m.sc., D.Sc. Adel., Ph.D. Camb., F.A.C.1., F.Inst.P., c/o C.S.R. Co., Pyrmont; p.r. 41 Spencer-road, Killara. 1938 Breckenridge, Marion, B.Sc., Department of Geology, University of Sydney ; p-r. 19 Handlev-avenue, Thornleigh. Pil P 4 P 22 124) D) 5 Pp 4 P 20 P 2 1 l P 5 7 Pea Pa P 4 P)3 | Brigden, Alan Charles, B.Sc., 22 Kelso-street, Enfield. Briggs, George Henry, D.Sc., Ph.D., F.Inst.P., Officer-in-Charge, Section of Physics, National Standards Laboratory of Australia, University Grounds, Sydney; p.r. 13 Findlay-avenue, Roseville. Brown, Desmond J., B.Sc., 9 Agnes-street, Strathfield. Brown, Ida Alison, p.sc., Lecturer in Paleontology, University of Sydney. Brown, Samuel Raymond, A.c.a. Aust., 87 Ashley-street, Chatswood. Browne, William Rowan, D.sc., Reader in Geology in the University of Sydney. (President, 1932.) Buckley, Daphne M. (Mrs.), B.sc., 4 Sharland-avenue, Chatswood. Buckley, Lindsay Arthur, B.sc., 4 Sharland-avenue, Chatswood. Burfitt, Barbara Joyce, M.B., B.S., Captain, A.I.F., 110 Elizabeth Bay-road, Elizabeth Bay, N.S.W. tBurfitt, W. Fitzmaurice, B.A., M.B., Ch.M., B.Sc. Syd., F.R.A.C.S., “* Radstoke,”’ Elizabeth Bay. Burkitt, Arthur Neville St. George, M.B., B.Sc., Professor of Anatomy in the University of Sydney. Burkitt, John Stanton, ‘‘ Moonbi,’’ 17 Cavell-street, West Hobart, Tas. Cane, Reginald Frank, m.sc., A.A.c.I., National Oil Pty. Ltd., Glen Davis, N.S.W. Callanan, Victor John, B.sc., 17 Wheatleigh-street, Naremburn. tCarey, Samuel Warren, D.sc., Practising Petroleum Geologist, c/o Australasian Petroleum Co., Melbourne. tCarslaw, Horatio Scott, sc.D., LL.D., F.R.S.E., Emeritus Professor of Mathe- matics, University of Sydney, Fellow of Emmanuel College, Cambridge ; Burradoo, N.S.W. Cavill, George William Kenneth, B.sc., Department of Chemistry, Technical College, Harris-street, Ultimo; p.r. 40 Chandos-street, Ashfield. Challinor, Richard Westman, F.I.C., A.A.C.I., A.S.T.C., F.C.S. ; p.r. 54 Drumalbyn- road, Bellevue Hill. (President, 1933.) Chalmers, Robert Oliver, a.s.T.c., Scientific Liaison Bureau, Box 4061, G.P.O., Sydney. Chambers, Maxwell Clark, B.sc., c/o J. and E. Atkinson Pty. Ltd., 469-75 Kent-street, Sydney. Cheel, Edwin, 40 Queen-street, Ashfield. (President, 1931.) Churchward, John Gordon, B.Sc.Agr., Ph.D., 1 Hunter-street, Woolwich. Clark, Sir Reginald Marcus, K.B.E., Central Square, Sydney. Clarke, Ronald Stuart, B.a., 28 Beecroft-road, Beecroft. Clune, Francis Patrick, Author and Accountant, 15 Prince’s-avenue, Vaucluse. Cohen, Max Charles, B.sc., A.I.F. Cohen, Samuel Bernard, M.sc., A.A.c.1., 24 Euroka-street, Northbridge. Colditz, Margaret Joyce, B.sc., 9 Beach-street, Kogarah. Cole, Edward Ritchie, B.sc., 14 Barwon-road, Lane Cove. Cole, Joyce Marie, B.Sc., 14 Barwon-road, Lane Cove. Collett, Gordon, B.sc., 20 Duchess-avenue, Fivedock. Cooke, Frederick, c/o Meggitt’s Limited, Asbestos House, York and Barrack- streets, Sydney. Coombs, F. A., Fr.c.s., Instructor of Leather Dressing and Tanning, Sydney Technical College; p.r. Bannerman-crescent, Rosebery. Corbett, Robert Lorimer, Managing Director of Robert Corbett & Co. Ltd., Manufacturing Chemists, Head Office, 379 Kent-street, Sydney. Cortis-Jones, Beverly, M.Sc., 62 William-street, Roseville. Cotton, Frank Stanley, p.sc., Research Professor in Physiology in the University of Sydney. Cotton, Leo Arthur, M.A., D.Sc., Professor of Geology in the University of Sydney. (President, 1929.) Craig, David Parker, Research Scholar, 62 Springdale-road, Killara. tCresswick, John Arthur, A.A.C.1., F.c.S., Production Superintendent and Chief Chemist, c/o The Metropolitan Meat Industry Commissioner, State Abattoir and Meat Works, Homebush Bay; p.r. 101 Villiers-street, Rockdale. | Crockford, Joan Marian, B.Sc., 219 Victoria-road, Gladesville. Culey, Alma Gertrude, m.sc., 37 Neirbo-avenue, Hurstville. Xl Elected. 1940 Dadour, Anthony, B.Sc., 25 Elizabeth-street, Waterloo. 1890 Dare, Henry Harvey, M.E., M.Inst.C.E., M.I.E.Aust., 14 Victoria-street, Roseville. 1919 P 2 de Beuzeville, Wilfred Alex. Watt, 3.p., ‘‘ Mélamere,’’ Welham-street, Beecroft. 1906 {Dixson, William, ‘‘ Merridong,’”’ Gordon-road, Killara. 1913 | Salta Doherty, William M., F.1.c., F.A.C.1., 36 George-street, Marrickville. 1928 Donegan, Henry Arthur James, A.S.T.c., A.A.C.1., Analyst, Department of Mines, Sydney ; p.r..18 Hillview-street, Sans Souci. 1943 Dudgeon, William, Manager, Commonwealth Drug Co., 50-54 Kippax-street, Sydney. 1937 Pee Dulhunty, John Allan, B.sc., Geology Department, University of Sydney. 1924 Dupain, George Zephirin, A.a.c.1., F.c.s., Director Dupain Institute of Physical Education and Medical Gymnastics, Manning Building, 449 Pitt-street, Sydney; p.r. “‘ Rose Bank,” 158 Parramatta-road, Ashfield. 1934 P 20 Dwyer, Francis P. J., m.sc., Lecturer in Chemistry, Technical College, Sydney. 1923 P 21 Earl, John Campbell, p.sc., Ph.p., Professor of Organic Chemistry in the University of Sydney. (President, 1938.) 1924 Eastaugh, Frederick Alldis, a.R.s.M., ¥F.1.c., Professor in Engineering Tech- nology and Metallurgy in the University of Sydney. 1934 Pea2 Elkin, Adolphus Peter, M.a., Ph.D., Professor of Anthropology in the University of Sydney. (President, 1940. Hon. Secretary.) 1940 | Emmerton, Henry James, B.sc., 41 Nelson-street, Gordon. 1937 English, James Roland, t.s. A.I.F. 1916 P 2 Enright, Walter John, B.a., Solicitor, High- street, West Maitland ; p.r. Regent- street, West Maitland. 1908 {Esdaile, Edward William, 42 Hunter-street, Sydney. 1935 Evans, Silvanus Gladstone, a.1.a.A. Lond., A.R.A.1.A., 6 Major-street, Coogee. 1939 Faull, Norman Augustus, B.Sc., A.Inst.P., c.o. National Standards Laboratory, University Grounds, City-road, Chippendale. 1909 P 7 |{Fawsitt, Charles Edward, D.sc., Ph.D., Professor of Chemistry in the University of Sydney. (President, 1919.) 1923 Fiaschi, Piero, 0.B.E., V.D., M.D. Columbia Univ., p.p.s. New York, M.R.C.S. fing., u.R.c.P. Lond., 178 Phillip-street, Sydney. 1940 Finch, Franklin Charles, B.sc., Kirby-street, Rydalmere, N.S.W. 1927 | egy Finnemore, Horace, B.Sc., F.1.c., Reader in Pharmacy in the University of Sydney. 1940 Fisher, Robert, B.Sc., 3 Sackville-street, Maroubra. 1920 Fisk, Sir Ernest Thomas, K.B., F.Inst.R.E., A.M.1.E. Aust., Chairman of Directors, Amalgamated Wireless (Australasia) Ltd., Wireless House, 47 York-street, Sydney ; p.r. 16 Beaconsfield-parade, Lindfield. 1940 Flack, Arthur Charles Allenby, B.sc., Agricultural High School, Yanco, N.S.W. 1933 Fletcher, Harold Oswald, Assistant Paleontologist, Australian Museum, College-street, Sydney. . 1879 ‘[Foreman, Joseph, M.R.c.s. Hng., u.R.c.P. Hdin., ‘“* The Astor,’ Macquarie-street, Sydney. 1932 Forman, Kenn. P., M.1.Refr.E., c/o Department of Aircraft Production, Box 20935, Melbourne, Vic. 1905 {Foy, Mark, c/o Geo. O. Bennett, 133 Pitt-street, Sydney. 1940 Franki, Robert James Anning, B.sc., 891 New South Head-road, Rose Bay. 1943 Frederick, Robert Desider Louis, B.z., 6 ‘“‘ Trinity Court,’ Telopea-street, Wollstonecraft. 1940 Freney, Martin Raphael, B.sc., Central Wool Testing House, 17 Randle-street, Sydney. 1935 1 ee 4 Garretty, Michael Duhan, m.sc., Chief Geologist, North Broken Hill Ltd., Broken Hill, N.S.W. 1939 1B Gascoigne, Robert Mortimer, 5 Werona-avenue, Killara. 1926 Gibson, Alexander James, M.E., M.Inst.Cc.E., M.I.E.Aust., Consulting Engineer, 906 Culwulla Chambers, 67 Castlereagh-street, Sydney; p.r. “ Wirruna,”’ Belmore-avenue, Wollstonecraft. 1942 Gibson, Neville Allan, 3B.sc., Industrial Chemist, 217 Parramatta-road, Haberfield. 1940 Gillis, Richard Galvin, 1 Dundee, 35 Adams-street, South Yarra, S.E.1, Vic. 1935 Goddard, Roy Hamilton, F.c.a. Aust., Royal Exchange, Bridge-street, Sydney. xu Elected. 1936 1940 1938 Pel P 6 P 4 P 4 Bins i ca | P15 Py 6 Pil | | Goulston, Edna Maude, B.sc., Third Officer, W.R.A.N.S., Navy socio Melbourne. Graves, John Nevil, B.sc., 96 Wentworth-street, Randwick. Griffiths, Edward L., B.Sc., A.A.C.I., A.I.c., Chief Chemist, Depa of Agriculture; p.r. 151 Wollongong-road, Arncliffe. Hall, Norman Frederick Blake, m.sc., Chemist, Council for Scientific and Industrial Research (Tobacco Section), Dept. of Organic Chemistry, University of Sydney; p.r. 154 Wharf-road, Longueville. tHalloran, Henry Ferdinand, L.s., A.M.1.E.Aust., F.S.I.Eng., M.T.P.1.Eng., 153 Elizabeth-street, Sydney ; p.r. 23 March-street, Bellevue Hill. Hanlon, Frederick Noel, B.sc., Geologist, Department of Mines, Sydney ; p-r. 14 Countess-street, Mosman. . tHarker, George, D.Sc., F.A.C.I.; p.r. 89 Homebush-road, Strathfield. Harper, Arthur Frederick Alan, M.sc., A.Inst.P., National Standards Laboratory, University Grounds, City-road, Chippendale. Harrington, Herbert Richard, Teacher of Physics and Electrical Engineering, Technical College, Harris-street, Ultimo. Hawley, J. William, 3.p., Financial Agent, 4 Castlereagh-street, Sydney ; p-r. 12 King’s-road, Vaucluse. Hayes, William Lyall, A.s.tT.c., a.A.c.1., Works Chemist, c/o Messrs. Wm. Cooper & Nephews (Aust.) Ltd., Phillip-street, Concord; p.r. 101 Essex-street, Epping. Henriques, Frederick Lester, 208 Clarence-street , Sydney. Hill, Dorothy, m.sc. Q’ld., Ph.D. Cantab., Geological Research Fellow, University of Queensland, Brisbane. Hindmarsh, Percival, M.A. B.Sc.agr., Principal, Hurlstone Agricultural High School, Glenfield. Hirst, Edward Eugene, A.M.1.£., Vice-Chairman and Joint Managing Director, British General Electric Co. Ltd.; p.r. “‘ Springmead,’’ Ingleburn. Hirst, George Walter Cansdell, B.sc., A.M.1.E. (Aust.), A.M.Inst.T.; p.r. “ St. Cloud,’’ Beaconsfield-road, Chatswood. Hoggan, Henry James, A.M.1.M.E. Lond., A.M.I.E. Aust., Consulting and Designing Engineer, “‘ Lincluden,”’ 81 Frederick-street, Rockdale. Howard, Harold Theodore Clyde, B.sc., Principal, Wollongong Technical High School, Wollongong. Howarth, Mark, F.R.A.s., Grange Mount Observatory, Bull-street, Mayfield, Newcastle, N.S.W. Hughes, Gordon Kingsley, B.sc., Lecturer in Chemistry, University of Sydney. tHynes, Harold John, D.sc., B.Sc.agr., Biologist, Department of Agriculture, Box 364, G.P.O., Sydney ; p.r. “‘ Belbooree,’’ 10 Wandella-avenue, Rose- ville. Iredale, Thomas, B.Sc., D.Sc., F.1.c., Chemistry Department, University of Sydney; p.r. 96 Roseville-avenue, Roseville. Jaeger, John Conrad, M.A., D.Sc., c/o Radiophysics Laboratory, The University, Sydney. Johns, Thomas Harley, 130 Smith-street, Summer Hill. Johnston, Thomas Harvey, M.A., D.Sc., C.M.Z.S., Professor of Zoology in the University of Adelaide. (Cor. Mem., 1912.) Joplin, Germaine Anne, B.Sc., Ph.D., Geological Department, University of Sydney ; p.r. 18 Wentworth-street, Eastwood. Judd, William Percy, 123 Wollongong-road, Arncliffe. Julius, Sir George A., Kt., B.Sc., B-E., M.I.Mech.E., M.I.E.Aust., Culwulla Chambers, Castlereagh-street, Sydney. Kelly, Caroline Tennant (Mrs.), ‘“ Hight Bells,’’ Castle Hill. Kennard, William Walter, 9 Bona Vista-avenue, Maroubra. Kenny, Edward Joseph, Geological Surveyor, Department of Mines, Sydney ; p-r. 17 Alma-street, Ashfield. | Kerslake, Richmond, 4.s.T.c., A.A.c.1., Industrial Chemist, c/o Australian Paper Mfrs. Ltd., Macauley-street, Matraville; p.r. 55 Harold-street, Matraville. Elected. 1943 1940 1920 1939 1936 1936 1920 1929 1942 1940 1940 1906 1927 1943 1940 1942 ie 3 P 55 Prt legate | cama! 2 Pee2 lego PALg Near Xill Kimble, Jean Annie, Research Chemist, B.sc., 383 Marrickville-road, Marrick- ville. King, Leonard Esmond, 161 Nelson Bay-road, Bronte. Kirchner, William John, B.Sc., A.A.Cc.I., Manufacturing Chemist, c/o Messrs. Burroughs Wellcome & Co. (Australia) Ltd., Victoria-street, Waterloo : p-r. 18 Lyne-road, Cheltenham. Lambeth, Arthur James, B.Sc., ‘‘ Naranje,’’ Sweethaven-road, Wetherill ParkaaNeSa . | Leach, Stephen Laurence, B.A., B.Sc., A.A.C.I., P.O. Box. No. 21, Concord. Lemberg, Max Rudolf, p-.phil., Biochemist, Royal North Shore Hospital ; p-r. 12 de Villiers-avenue, Chatswood. Le Souef, Albert Sherbourne, 3 Silex-road, Mosman. tLions, Francis, B.Sc., Ph.D., A.1.c., Department of Chemistry, University of Sydney. Lippmann, Arthur S., M.p., 175 Macquarie-street, Sydney. Lipson, Menzie, B.sc., Chemist, C.S.I.R., Central Wool Committee Testing House, 17 Randle-street, Sydney. Lockwood, William Hutton, B.sc., c.o. Department of Post-War Reconstruction, Hotel Acton, Canberra, A.C.T. tLoney, Charles Augustus Luxton, M.Am.soc.Refr.E., National Mutual Building, 350 George-street, Sydney. Love, William Henry, B.Sc., Ph.D., Lecturer in Physics, University of Sydney. Luber (Mrs.) Daphne, B.sc., 98 Lang-road, Centennial Park. Luciano, Albert Anthony, 16 Arthur-street, Bellevue Hill. Lyons, Raymond Norman Matthew, mM.sc., Biochemical Research Worker, 8 Boronia-avenue, Wollstonecraft. Maccoll, Allan, M.sc., 76 Springdale-road, Killara. McCoy, William Kevin, Analytical Chemist. McGrath, Brian James, 40 Mooramie-avenue, Kensington. McGregor, Gordon Howard, 4 Maple-avenue, Pennant Hills. McIntosh, Arthur Marshall, ‘* Moy Lodge,”’ Hill-street, Roseville. McKay, R. T., M.Inst.c.z., Eldon Chambers, 92 Pitt-street, Sydney. McKern, Howard Hamlet Gordon, 4.S.T.c., A.A.c.1., Chemist, Meggitt Ltd., Parramatta; p.r. 14 Orwell-street, Potts Point. , McKie, Rev. Ernest Norman, B.A. Syd., St. Columba’s Manse, Guyra. McMaster, Sir Frederick Duncan, kt., “* Dalkeith,’’ Cassilis, N.S.W. Malone, Edward E., No. 4, Astral, 10 Albert-street, Randwick. Mance, Frederick Stapleton, ‘‘ Binbah,’’ Lucretia-avenue, Longueville. Martin, Cyril Maxwell, Chemist, 22 Wattle-street, Haberfield. Maze, Wilson Harold, M.sc., Lecturer in Geography, University of Sydney. Melville, George Livingstone, Managing Director, Federal Machine Co. Ltd., Loftus-street, Arncliffe. Meldrum, Henry John, B.A., B.Ssc., Lecturer, The Teachers’ College, University Grounds, Newtown ; p.r. 98 Sydney-road, Manly. Mellor, David Paver, M.sc., Lecturer, Chemistry Department, University of Sydney; p.r. 137 Middle Harbour-road, Lindfield. (President, 1941-42. Joint Hon. Secretary.) Mercer, Edgar Howard, McMaster Laboratory, Parramatta-road, Glebe. Micheli, Louis Ivan Allan, M.sc., Ph.p., Research Chemist, c/o Colonial Sugar Refining Co., Pyrmont. Millership, William, m.sc., Chief Chemist, Davis Gelatine (Aust.) Pty. Ltd., 15 Shaw-avenue, Earlwood. : Molloy, Ernest Patrick, Assistant Sectional Manager, 129 Gibbes-street, Rockdale. Morrissey, Matthew John, B.A., 4.s.T.c., Auburn-street, Parramatta. Morrison, Frank Richard, 4A.A.c.1., F.c.S., Assistant Chemist, Technological Museum, Sydney. Mort, Francis George Arnot, A.A.c.1., Chemist, 16 Grafton-street, Woollahra. ‘Moye, Daniel George, Chemist, 89 Caroline-street, South Yarra, S.E.1., Vic. bh te X1V Elected. 1915 1923 1930 1943 1932 1943 1920 1940 1940 1935 1903 1921 1920 1940 1938 1935 1943 1919 1896 1921 1918 1938 1927 1918 1893 1935 1922 1940 1919 Po.2 P 6 P 4 Pe Ps P 74 Py 2 | ae | Pas 6 Ped P 3 Murphy, Robert Kenneth, pr.Ing., Chem.Eng., A.S.T.C., M.I.Chem.E., A.A.C.I., Lecturer in Charge of Chemistry and Head of Science Department, Sydney Technical College. Murray, Jack Keith, QX34748, B.a., B.Sc.agr., N.E. Reinf. Training Centre, Warwick, Queensland, and Professor of Agriculture in the University of Queensland. Naylor, George Francis King, M.A., M.Sc., Dip.Ed., A.A.I.I.P., Squadron Leader, R.A.A.F., Headquarters, Melbourne; p.r. “* Kingsleigh,’’ Ingleburn, N.S.W. Neuhaus, John William George, c.o. Meggitt and Co. Ltd., Parramatta. Newman, Ivor Vickery, M.Sc., Ph.D., F.R.M.S., F.L.S., Department of Biology, Victoria University College, Wellington, N.Z. Nicol, Alexander Campbell, a.s.T.c., A.A.c.1., Chief Chemist, Crown Crystal Glass Co.; p.r. No. 12 “ Florida,” 519 New South Head-road, Double Bay. tNoble, Robert Jackson, M.sc., B.Sc.agr., Ph.D., Under Secretary, Department of Agriculture, Box 36a, G.P.O., Sydney; p.r. 324 Middle Harbour-road, Lindfield. (President, 1934.) Norrie, Jack Campbell, B.sc., 28 Ray-road, Epping. Nyholm, Ronald Sydney, M.sc., 77 Bland-street, Ashfield. O’Connell, Rev. Daniel J. K., s.3., M.sc., F.R.A.S., Riverview College Observatory, Sydney. tOld, Richard, ‘‘ Waverton,’ Bay-road, North Sydney. Osborne, George Davenport, D.sc. Syd., Ph.D. Camb., Lecturer and Demonstrator in Geology in the University of Sydney. Penfold, Arthur Ramon, F.A.c.1., F.c.s., Curator and Economic Chemist, Technological Museum, MHarris-street, Ultimo; p.r. 67 Park-avenue, Roseville. (President, 1935.) | Pettingell, William Walter, B.Sc., 28 Conder-street, Burwood. Phillips, Marie Elizabeth, B.sc., 4 Morella-road, Clifton Gardens. Phillips, Orwell, 55 Darling Point-road, Edgecliffe. Plowman, Ronald Arthur, 4.s.T.c., A.A.c.1., Analytical Chemist, 78 Alt-street, Ashfield. Poate, Hugh Raymond Guy, M.B., ch.m. Syd., F.R.c.S. Hng., L.R.c.P. Lond., F.R.A.C.S., Surgeon, 225 Macquarie-street, Sydney ; p.r. 38 Victoria-road, Bellevue Hiuill. tPope, Roland James, B.A. Syd., M.D., Ch.M., F.R.c.S. Hdin., 185 Macquarie- street, Sydney. Powell, Charles Wilfrid Roberts, F.1.c., A.A.c.I.. Company Executive, c/o Colonial Sugar Refining Co., O’Connell-street, Sydney ; p.r. “‘ Wansfell,”’ Kirkoswald-avenue, Mosman. Powell, John, Director, Foster Clark (Aust.) Ltd., 17 Thurlow-street, Redfern ; p.r. “ Elgarth,’’ Ranger’s-road, Cremorne. Powell, John Wallis, A.s.T.c., A.a.c.1., Managing Director, Foster Clark (Aust.) Ltd., 17 Thurlow-street, Redfern. Price, William Lindsay, B.E., B.sc., Teacher of Physics, Sydney Technical College ; p.r. 8 Wattle-street, Killara. Priestley, Henry, M.D., Ch.M., B.Sc., Professor of Biochemistry, Faculty of Medicine, the University of Sydney. (President, 1942-43.) {Purser, Cecil, B.A., M.B., Chm. Syd., “‘ Ascot,” Grosvenor-road, Wahroonga. tQuodling, Florrie Mabel, B.sc., Demonstrator in Geology, University of Sydney. Raggatt, Harold George, D.sc., Director, Mineral Resources Survey, Depart- ment of Supply, Canberra, A.C.T. Ralph, Colin Sydney, B.Sc., 24 Canberra-street, Epping. Ranclaud, Archibald Boscawen Boyd, B.sc., B.E., Lecturer in Physics, Teachers’ College, The University, Sydney. pee P 6 P..2 Po 2 Pp, 2 Bol 1 eam | 1c | P 16 XV Randall, Harry, Buena Vista-avenue, Denistone. Rayner, Jack Maxwell, B.sc., F.1nst.p., Chief Geophysicist, Mineral Resources Survey, Department of Supply and Shipping, Census Building, Canberra, A.C. Reid, Cicero Augustus, 19 Newton-road, Strathfield. Ritchie, Ernest, B.sc., 6 Military-road, North Bondi. Roberts, Richard George Crafter, Electrical Engineer, c/o C. W. Stirling & Co., Asbestos House, York and Barrack-streets, Sydney. Robertson, Rutherford Ness, B.sc. Syd., Ph.D. Cantab., Flat 4, 43 Johnston- street, Annandale. Room, Thomas G., M.A., F.R.S., Professor of Mathematics in the University of Sydney. Rosenbaum, Sidney, 44 Gilderthorp-avenue, Randwick. Ross, Allan Clunies, B.Sc., F.c.A. Aust., Chartered Accountant Aust., 544 Pitt-street, Sydney ; p.r. The Grove, Woollahra. (Member from 1915 to 1924.) (Hon. Treasurer.) Ross, Jean Elizabeth, B.sc., Dip.Ed., 5 Stanton-road, Haberfield. Savage, Clarence Golding, Director of Fruit Culture, Department of Agriculture, Sydney. Sawkins, Dansie Thomas, m.a. Syd., B.A. Camb., Reader in Statistics, The University, Sydney ; p.r. 60 Boundary-street, Roseville. Scammell, Rupert Boswood, B.sc. Syd., A.A.C.1., F.c.S., c/o F. H. Faulding & Co. Ltd., 98 Castlereagh-street, Redfern; p.r. 10 Buena Vista-avenue, Clifton Gardens. Scott, Reginald Henry, B.sc., 154 Highfield-road, Camberwell, Vic. Selby, Esmond Jacob, Dip.com., Sales Manager, Box 175 D, G.P.O., Sydney. Sellenger, Brother Albertus, Marist Brothers’ College, Randwick, N.S.W. Sheahan, Thomas Henry Kennedy, B.sc., Chemist, 2 Edward-street, Gordon. Sherrard, Kathleen Margaret Maria (Mrs.), M.sc. Melb., 43 Robertson-road, Centennial Park. Sibley, Samuel Edward, Mount-street, Coogee. tSimpson, R. C., Lecturer in Electrical Engineering, Technical College, Sydney. Simpson, John Kenneth Moore, Industrial Chemist, ‘ Browie,’? Old Castle Hill-road, Castle Hill. Slade, George Hermon, B.Sc., Director, W. Hermon Slade & Co. Ltd., Manu- facturing Chemists, 23 Rosebery-avenue, Rosebery; p.r. “ Raiatea,”’ Oyama-avenue, Manly. Smith, Erie Brian Jeffcoat, 1 Rocklands-road, Wollstonecraft. Smith, Thomas Hodge, Australian Museum, College-street, Sydney. Southee, Ethelbert Ambrook, 0.B.E., M.A., B.Sc., B.Sc.Agr., Principal, Hawkes- bury Agricultural College, Richmond, N.S.W. Spencer-Watts, Arthur, ‘ Araboonoo,”’ Glebe-street, Randwick. Stephen, Alfred Ernest, F.c.s., c/o Box 1158 HH, G.P.O., Sydney. Stephens, Frederick G. N., F.R.C.S., M.B., Ch.M., 135 Macquarie-street, Sydney ; p.r. Captain Piper’s-road and New South Head-road, Vaucluse. tStewart, J. Douglas, B.v.sc., F.R.C.v.s., Emeritus Professor of Veterinary Science in the University of Sydney; p.r. “‘ Berelle,’> Homebush-road, Strathfield. (President, 1927.) Still, Jack Leslie, B.sc., Ph.D., Department of Biochemistry, The University, Sydney. i{Stokes, Edward Sutherland, m.B., chm. Syd., p.p.H. Irel., Medical Officer, Metropolitan Board of Water Supply and Sewerage, 341 Pitt-street, Sydney ; p.r. 15 Highfield-road, Lindfield. Stone, Walter George, F.S.T.C., A.A.C.I., Senior Analyst, Department of Mines, Sydney; p.r. 14 Rivers-street, Bellevue Huill. Stroud, Richard Harris, B.sc., “‘ Dalveen,’’ corner Chalmers and Barker-roads, Strathfield. tSullivan, Herbert Jay, Director in Charge of Research and Technical Depart- ment, c/o Lewis Berger & Sons (Australia) Ltd., Rhodes; Box 23, P.O., Burwood; p.r. “‘ Stonycroft,’ 10 Redmyre-road, Strathfield. tSussmilch, C. A., F.G.s., F.S.T.c., Consulting Geologist, 11 Appian Way, Burwood. (President, 1922.) tSutherland, George Fife, a.r.c.se. Lond., Assistant Professor of Mechanical Engineering in the University of Sydney. Xvl Elected. 1920 1941 1915 1944 1939 1919 1935 1923 1940 1932 1943 1940 1921 1935 1933 1903 1943 1919 1913 1921 1924 1919 1919 1941 1911 1936 1920 1921 1909 1940 1943 1928 . 1942 P Or | Sutton, Harvey, 0.B.E., M.D., D.P.H. Melb., B.Sc. Oxon., Professor of Preventive Medicine and Director, School of Public Health and Tropical Medicine, _ University of Sydney; p.r. “ Lynton,” 27 Kent-road, Rose Bay. Swanson, Thomas Baikie, m.sc. Adel., Lecturer in Chemistry, New England University College, Armidale. Taylor, Brigadier Harold B., M.c., D.Sc., F.1.C., F.A.C.I., Second Government Analyst, Department of Public Health, 93 Macquarie-street, Sydney ; p.r. 44 Kenneth-street, Longueville. Thomas, Andrew David, Flight-Lieutenant, R.A.A.S., M.Sc., A.Inst.P., 1 Valley- road, Lindfield. Thomas, Mrs. A. V. M., 1 Valley-road, Lindfield. Thorne, Harold Henry, m.a. Cantab., B.sc. Syd., F.R.A.S., Lecturer in Mathe- matics in the University of Sydney ; p.r. 55 Railway-crescent, Beecroft. Tommerup, Eric Christian, M.Sc., A.A.C.1. Toppin, Richmond Douglas, 4.1.c., 51 Crystal-street, Petersham. Tow, Aubrey James, m.Sc., No. 5, ‘* Werrington,’’ Manion-avenue, Rose Bay. Trikojus, Victor Martin, B.Sc., D.Phil., Professor of Biochemistry, The University, Melbourne. Turner, Ivan Stewart, M.A., M.Sc., Ph.D., Lecturer in Mathematics, University of Sydney; p.r. 120 Awaba-street, Mosman. Vernon, James, Ph.D., A.A.C.1., Chief Chemist, Colonial Sugar Refining Co., 1 O’Connell-street, Sydney. Vicars, Robert, Marrickville Woollen Mills, Marrickville. Vickery, Joyce Winifred, m.sc., Botanic Gardens, Sydney; p.r. 17 The Promenade, Cheltenham. Voisey, Alan Heywood, m.sc., Lecturer in Geology and Geography, New England University College, Armidale. {Vonwiller, Oscar U., B.Sc., F.Inst.P., Professor of Physics in the University of Sydney. (President, 1930.) Walker, James Foote, Company Secretary, 11 Brucedale-avenue, Epping. Walkom, Arthur Bache, D.sc., Director, Australian Museum, Sydney; p.r. 45 Nelson-road, Killara. (Member from 1910-1913. President, 1943-44.) Wardlaw, Hy. Sloane Halcro, pD.sc. Syd., F.Aa.c.1., Lecturer and Demonstrator in Biochemistry in the University of Sydney. (President, 1939.) {Waterhouse, Gustavus Athol, D.sc., B.E., F.R.E.S., F.R.Z.S., c.o. Australian Museum, College-street, Sydney. Waterhouse, Leslie Vickery, B.z. Syd., Mining Engineer, Shell House, Car- rington-street, Box 58 CC, G.P.O., Sydney; p.r. 4 Bertha-road, Neutral Bay. Waterhouse, Lionel Lawry, B.r. Syd., Lecturer and Demonstrator in Geology in the University of Sydney. Waterhouse, Walter L., M.c., D.Sc.Agr., D.I.C., F.L.S., Reader in Agriculture, University of Sydney; p.r. “‘ Hazelmere,’’ Chelmsford-avenue, Lindfield. (President, 1937.) Watson, Irvine Armstrong, Ph.D., B.Sc.Agr., Assistant Lecturer, Faculty of Agriculture, University of Sydney. Watt, Robert Dickie, M.a., B.Sc., Professor of Agriculture in the University of Sydney; p.r. 64 Wentworth-road, Vaucluse. (President, 1925.) Wearne, Harold Wallis, 22 Yarabah-avenue, Gordon. Wellish, Edward Montague, m.a., Associate-Professor of Applied Mathematics in the University of Sydney ; p.r. 15 Belgium-avenue, Roseville. Wenholz, Harold, B.sc.agr., Director of Plant Breeding, Department of Agri- culture, Sydney. ; tWhite, Charles Josiah, B.sc., Lecturer in Chemistry, Teachers’ College, Uni- versity Grounds, Newtown. White, Douglas Elwood, M.sc., D.Phil.. University of Western Australia, Nedlands, W.A. Whiteman, Reginald John Nelson, M.B., Ch.M., F.R.A.C.S., 143 Macquarie-street, Sydney. Wiesener, Frederick Abbey, M.B., Ch.M., D.oO.M.s., Ophthalmic Surgeon, 143 Macquarie-street, Sydney ; p.r. Jersey-road, Strathfield. Williams, Gordon Roy, B.sc., 45 Conder-street, Burwood. Elected. 1943 1940 1936 2 1906 PTZ 1916 1921 Elected. 1914 1915 1912 1922 Xvi Winch, Leonard, B.sc., Chief Chemist, Fielder’s General Products Ltd., P.O. Box 143, Tamworth, N.S.W. Wogan, Samuel James, Range-road, Sarina, North Queensland. Wood, Harley Weston, M.Sc., A.Inst.P., F.R.A.S., Assistant Astronomer, Sydney Observatory. | tWoolnough, Walter George, D.sc., F.G.S., 9 Lockerbie Court, East St. Kilda, Victoria. (President, 1926.) Wright, George, Company Director, c/o Hector Allen, Son & Morrison, 16 Barrack-street, Sydney ; p.r. ‘‘ Wanawong,”’ Castle Hill, N.S.W. Yates, Guy Carrington, Seedsman, c/o Arthur Yates & Co. Ltd., 184 Sussex- | street, Sydney ; p.r. Boomerang-street, Turramurra. HoNnNoRARY MEMBERS. Limited to Twenty. Hill, James P., D.sc., F.R.S., Professor of Zoology, University College, Gower- street, London, W.C.1, England. Maitland, Andrew Gibb, rF.a.s., ‘‘Bon Accord,” 28 Melville-terrace, South Perth, W.A. Martin, Sir Charles J., c.M.c., D.Sc., F.R.S., Roebuck House, Old Chesterton, Cambridge, England. Wilson, James T., M.B., Ch.m. Edin., F.R.S., Professor of Anatomy in the Uni- versity of Cambridge: p.r. 24 Millington-road, Cambridge, England. OBITUARY, 1943-1944. Elected. 1922. John Job Crew Bradfield. 1936 Archibald Howie. 1940 James Edward Mills 1939 Frederick Chapman. (Honorary Member.) THE REV. W. B. CLARKE MEMORIAL FUND. The Rev. W. B. Clarke Memorial Fund was inaugurated at a meeting of the Royal Society of N.S.W. in August, 1878, soon after the death of Mr. Clarke, who for nearly forty years rendered distinguished service to his adopted country, Australia, and to science in general. It was resolved to give an opportunity to the general public to express their appreciation of the character and services of the Rev. W. B. Clarke “ as a learned colonist, a faithful minister of religion, and an eminent scientific man.’’ It was proposed that the memorial should take the form of lectures on Geology (to be known as the Clarke Memorial Lectures), which were to be free to the public, and of a medal to be given from time to time for distinguished work in the Natural Sciences done - in or on the Australian Commonwealth and its territories; the person to whom the award is made may be resident in the Australian Commonwealth or its territories, or elsewhere. The Clarke Memorial Medal was established first, and later, as funds et the Clarke Memorial Lectures have been given at intervals. CLARKE MEMORIAL LECTURES. Delivered. 1906. ‘‘ The Volcanoes of Victoria,’ and “The Origin of Dolomite’”’ (two lectures). By Professor E. W. Skeats, D.Sc., F.G.S. 1907. ‘‘ Geography of Australia in the Permo-Carboniferous Period” (two lectures). By Professor ‘T. W. E. David, B.A., F.R.S. ** The Geological Relations of Oceania.’”’> By W.G. Woolnough, D.Sc. ‘* Problems of the Artesian Water Supply of Australia.”” By E. F. Pittman, A.R.S.M. “The Permo-Carboniferous Flora and Fauna and their Relations.’”” By W. 8S. Dun. 1918. ‘* Brain Growth, Education, and Social Inefficiency.’’ By Professor R. J. A. Berry, M.D., F.R.S.E. 1919. ‘* Geology at the Western Front,’ By Professor T. W. E. David, C.M.G., D.S.O., F.R.S. 1936. “‘ The Aeroplane in the Service of Geology.”” By W. G. Woolnough, D.Se. (TuHrs JouRN., 1936, 70, 39.) 1937. ‘‘ Some Problems of the Great Barrier Reef.’’ By Professor H. C. Richards, D.Sc. (Tuts JOURN., 1937, 71, 68.) 1938. ‘The Simpson Desert and its Borders.” By C. T. Madigan, M.A., B.Sc., B.E., D.Sc. (Oxon.). (THis Journ., 1938, 71, 503.) 1939. ‘* Pioneers of British Geology.” By Sir John S. Flett, K.B.E., D.Se., LL.D., F.R.S. (THis Journ., 1939, 73, 41.) ; 1940. “The Geologist and Sub-surface Water.” By E. J. Kenny, M.Aust.I.M.M. (Tus JouRN., 1940, 74, 283.) 1941. ‘‘ The Climate of Australia in Past Ages.” By C. A. Sussmilch, F.G.S. (THis Journ., 1941, 75, 47.) 1942. ‘* The Heroic Period of Geological Work in Australia.” By E. C. Andrews, B.Sc. 1943. ‘* Australia’s Mineral Industry in the Present War.” By H. G. Raggatt, D.Sc. AWARDS OF THE CLARKE MEDAL. Established in memory of The Revd. WILLIAM BRANWHITE CLARKE, M.a., F.R.s., F.G.S., etc. Vice-President from 1866 to 1878. The prefix * indicates the decease of the recipient. Awarded. 1878 *Professor Sir Richard Owen, K.C.B., F.R.S. 1879 *George Bentham, C.M.G., F.R.S. 1880 *Professor Thos. Huxley, F.R.s. 1881 *Professor F. M‘Coy, F.R.S., F.G.S. 1882 *Professor James Dwight Dana, LL.D. 1883 *Baron Ferdinand von Mueller, K.C.M.G., M.D., Ph.D., F.R.S., F.L.S. 1884 *Alfred R. C. Selwyn, LL.D., F.R.S., F.G.S. 1885 *Sir Joseph Dalton Hooker, 0.M., G.C.S.1., C.B., M.D., D.C.L., LL.D., F.R.S. 1886 *Professor L. G. De Koninck, m.p. 1887 *Sir James Hector, K.C.M.G., M.D., F.R.S. X1x Awarded. 1888 *Rev. Julian E. Tenison-Woods, F.G.S., F.L.S. 1889 *Robert Lewis John Ellery, F.R.S., F.R.A.S. 1890 *George Bennett, M.p., F.R.c.S. Eng., F.L.S., F.Z.S. 1891 *Captain Frederick Wollaston Hutton, F.R.S., F.G.8. 1892 *Sir William Turner Thiselton Dyer, K.C.M.G., C.I.E., M.A., LL.D., S¢.D., F.R.S., F.L.S. 1893 *Professor Ralph Tate, F.L.s., F.G.S. 1895 *Robert Logan Jack, LL.D., F.G.S., F.R.G.S. 1895 *Robert Etheridge, Jnr. 1896 *The Hon. Augustus Charles Gregory, C.M.G., F.R.G.S. 1900 *Sir John Murray, K.C.B., LL.D., Sc.D., F.R.S. 1901 *Edward John Eyre. 1902 *F. Manson Bailey, C.M.G., F.L.S. 1903 *Alfred William Howitt, p.sc., F.G.s. 1907 *Professor Walter Howchin, F.c.s., University of Adelaide. 1909 *Dr. Walter E. Roth, B.a. 1912 *W. H. Twelvetrees, F.G.s. 1914 Sir A. Smith Woodward, LuL.D., F.R.Ss., Keeper of Geology, British Museum (Natural History), London. 1915 *Professor W. A. Haswell, M.A., D.Sc., F.R.S. 1917 *Professor Sir Edgeworth David, K.B.E., C.M.G., D.S.0., M.A., SC.D., D.Sc., F.R.S., F.G.S. 1918 *Leonard Rodway, c.M.c., Honorary Government Botanist, Hobart, Tasmania. 1920 *Joseph Edmund Carne, F.G.S. 1921 *Joseph James Fletcher, M.A., B.Sc. 1922 *Richard Thomas Baker, The Crescent, Cheltenham. 1923 *Sir W. Baldwin Spencer, K.C.M.G., M.A., D.Sc., F.R.S. 1924 *Joseph Henry Maiden, I.S.0., F.R.S., F.L.S., J.P. 1925 *Charles Hedley, F.L.s. 1927. Andrew Gibb: Maitland, F.a.s., ‘‘ Bon Accord,’’ 28 Melville Terrace, South Perth, W.A. 1928 Ernest C. Andrews, B.A., F.G.S., 32 Benelong Crescent, Bellevue Hill. 1929 Professor Ernest Willington Skeats, D.Sc., A.R.C.S., F.G.S., University of Melbourne, Carlton, Victoria. 1930 ~=L. Keith Ward, B.A., B.E., D.Sc., Government Geologist, Geological Survey Office, Adelaide, 1931 *Robin John Tillyard, M. A, D.Sc., Sc.D., F.R.S., F.L.S., F.E.S., Canberra, F.C.T. 1932 Frederick Chapman, Aas, V.R.S.N.Z., Heese. Welbourne: 1933. Walter George Woolnough, D.sc., F.c.s., Department of the Interior, Canberra, F.C.T. 1934. *Edward Sydney Simpson, D.Sc., B.E., F.A.C.1., Carlingford, Mill Point, South Perth, W.A. 1935 George William Card, A.R.S.M., 16 Ramsay-street, Collaroy, N.S.W. 1936 Sir Douglas Mawson, Kt., 0.B.E., F.R.S., D.Sc., B.E., University of Adelaide. 1937 J. T. Jutson, B.sc., LL.B., 9 Ivanhoe-parade, Ivanhoe, Victoria. 1938 Professor H. C. Richards, D.sc., The University of Queensland, Brisbane. 1939 OC. A. Sussmilch, F.G.s., F.s.T.c., 11 Appian Way, Burwood, N.S.W. 194] Professor Frederic Wood Jones, M.B., B.S., D.Sc., F.R.S., Anatomy Department, University of Manchester, England. 1942 William Rowan Browne, D.se., Reader in Geology, The University of Sydney, N.S.W. 1943 Walter Lawry Waterhouse, M.c., D.Sc.Agric., D.I.C., F.L.S., Reader in Agriculture, University of Sydney. AWARDS OF THE SOCIETY’S MEDAL AND MONEY PRIZE. Money Prize of £25. Awarded. 1882 John Fraser, B.A., West Maitland, for paper entitled ‘‘ The Aborigines of New South Wales.” 1882 Andrew Ross, m.p., Molong, for paper entitled ‘* Influence of one Australian climate and: pastures upon "the growth of wool.” The Society’s Bronze Medal and £25. Awarded. 1884 W. E. Abbott, Wingen, for paper entitled ‘‘ Water supply in the Interior of New South Wales.” 1886 S. H. Cox, F.G.s., F.c.s., Sydney, for paper entitled ‘‘ The Tin deposits of New South Wales.” 1887 Jonathan Seaver, F.c.s., Sydney, for paper entitled ‘‘ Origin and mode of occurrence of ; gold-bearing veins and of the associated Minerals.” 1888 Rev. J. E. Tenison-Woods, F.G.s., F.L.S., Sydney, for paper entitled “‘ The Anatomy and Life-history of Mollusca peculiar to Australia.” XX Awarded. 1889 Thomas Whitelegge, F.R.mM.s., Sydney, for paper entitled ‘‘ List of the Marine and Fresh- water Invertebrate Fauna of Port Jackson and Neighbourhood.” 1889 Rev. John Mathew, m.a., Coburg, Victoria, for paper entitled ‘“‘ The Australian Aborigines.” 1891 Rev. J. Milne Curran, F.c.s., Sydney, for paper entitled ‘“‘ The Microscopic Structure of Australian Rocks.” 1892 Alexander G. Hamilton, Public School, Mount Kembla, for paper entitled “ The effect which settlement in Australia has produced upon Indigenous Vegetation.” 1894 J. V. De Coque, Sydney, for paper entitled the “‘ Timbers of New South Wales.” 1894 R. H. Mathews, t.s., Parramatta, for paper entitled ‘* The Aboriginal Rock Carvings and Paintings in New South Wales.” 1895 C. J. Martin, D.sc., M.B., F.R.S., Sydney, for paper entitled ‘‘ The physiological action of the venom of the Australian black snake (Pseudechis porphyriacus).” 1896 Rev. J. Milne Curran, Sydney, for paper entitled “‘ The occurrence of Precious Stones in New South Wales, with a description of the Deposits in which they are found.” , 1943 Edwin Cheel, Sydney, in recognition of his contributions in the field of botanical research and to the advancement of science in general. AWARDS OF THE WALTER BURFITT PRIZE. Bronze Medal and Money Prize of £50. Established as the result of a generous gift to the Society by Dr. W. F. Burrirt, B.A., M.B., Ch.M., B.Sc., of Sydney. Awarded at intervals of three years to the worker in pure and applied science, resident in Australia or New Zealand, whose papers and other contributions published during the past three years are deemed of the highest scientific merit, account being taken only of investigations described for the first time, and carried out by the author mainly in these Dominions. Awarded. 1929 Norman Dawson Royle, M.D., ch.m., 185 Macquarie Street, Sydney. 1932 Charles Halliby Kellaway, M.c., M.p., M.S., F.R.c.P., The Walter and Eliza Hall Institute of Research in Pathology and Medicine, Melbourne. 1935 Victor Albert Bailey, M.a., D.Phil., Associate-Professor of Physics, University of Sydney. 1938 Frank Macfarlane Burnet, m.p. (Melb.), ph.p. (Lond.), The Walter and Eliza Hall Institute of Research in Pathology and Medicine, Melbourne. 1941 Frederick William Whitehouse, D.sc., Ph.p., University of Queensland, Brisbane. AWARDS OF LIVERSIDGE RESEARCH LECTURESHIP. This Lectureship was established in accordance with the terms of a bequest to the Society by the late Professor Archibald Liversidge. Awarded at intervals of two years, for the purpose of encouragement of research in Chemistry. (THis Journnat, Vol. LXII, pp. x-xiu, 1928.) Awarded. 1931 Harry Hey, c/o The Electrolytic Zinc Company of Australasia, Ltd., Collins Street, Melbourne. 1933. W. J. Young, D.Sc., m.se., University of Melbourne. 1940 G. J. Burrows, B.Sc., University of Sydney. 1942 J.S. Anderson, B.sc., Ph.D. (Lond.), A.R.C.S., D.1.c., University of Melbourne. z orn : ISSUED JANUARY 19, 1944 : eeXVIl = PART I JOURNAL AND PROCEEDINGS me eo “OR THE Aor NEW SOUTH WALES ¥ : Bo : FOR = : 1943 (INCORPORATED 1881) : =) =. PART I (pp. 1 to 32) ee Ns OF es VOL. LXXVII Containing Papers read in April and May eas EDITED BY THE HONO RARY SECRETARI ES THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE STATEMENTS os AND THE OPINIONS EXPRESSED yes cae SYDNEY 4 PUBLISHED BY THE SOCIETY, SCIENCE HOUSE _ GLOUCESTER AND ESSEX STREETS 1944 CONTENTS Po i VOLUME LX XVID ae ; get Fass Part I ERS ® Arr, I—Presidential Address. By H. Priestley, M.D., Ch.M., B.Sc. November 3, 1943.) CRM Seria se alae oo a. tec) ee Oe ne Art. II.—Nova Puppis 1942. By H. W. Wood. (Issued October 22, 1943.) 4 ‘ age Arr. IIJ.—A Polyhedral Model of the Projective Plane. By F. A. Behrend. te ‘ Re pda ae’ Ootoher. a SOAR Ae, Re ya. , Sr conan Roce “ee fi ne eke iN pabotate B, Se. (Issued October 22, 1943.) . PRESIDENTIAL ADDRESS Bye A. -PRImStEny, M.D.’ Ch.M.. B.Sc. Delivered to the Royal Society of New South Wales, April 7, 1943. PART En EAST YEAR: The Society’s activities have been carried on very much as usual. The Council meetings and general monthly meetings have been held regularly. One of the few departures from the normal was an alteration in the times of the general monthly meetings. In an attempt to meet the difficulties which, it was anticipated, would be caused by ‘“ blackout ”’ conditions, afternoon and early evening meetings were held. The change was not a popular one, and it was decided to return to the usual time of 7.45. For the same reason no Popular Science Lectures were given during the winter months; instead, a series of Popular Science Talks were broadcast from the national stations during the month of July. This was in the nature of an experiment, but the possibility of continuing such broadcast talks, as well as arranging the usual series of Popular Science Lectures, is being considered. The Society’s thanks are due to those who gave considerable time and thought to the preparation and delivery of these talks. The membership of the Society stands at 292, a slight decrease from that of last year. Six members have resigned, seven new members have been elected, while seven members have died. We regret to announce the death of Marcus Baldwin Welch, who had been a member since 1920, and who contributed many papers to the Society. He was a member of the Council from 1932 to 1938, and again from 1939 to 1941, during which time he was Honorary Treasurer. Another death we regret to announce is that of Archibald Durrant Ollé, who was a member since 1913 and was chairman of the Section of Industry for several years. Other members lost by death are John Clifford Firth, elected in 1935, Gerald Harnett Halligan, elected in 1880, James Adam Dick, elected in 1894, Sir Kelso King, elected in 1896, and finally, Professor Sir J. J. Thomson, an honorary member since 1915. Despite the war, with its preoccupations and the extra duties that most workers in science have undertaken, either in the form of special researches or extra lectures owing to short staffs, the number of research papers presented before the Society was 28, an improvement on the number of papers read last year. In July a very successful symposium was held on the subject of Rubber. A large number of members and visitors attended, and a discussion took place after the addresses had been heard. A special celebration was that of the commemoration of the tercentenary of the death of Galileo and the birth of Newton, when an address was delivered at the meeting in October by Professor O. U. Vonwiller. The address, ‘‘ Galileo and Newton: Their Times and Ours ’’, has been printed in the Journal and Proceedings. A—April 7, 1943. 2 H. PRIESTLEY. Of the several foundations administered by the Royal Society of New South: Wales, the following activities may be reported : The Clarke Memorial Lecture was delivered by Mr. E. C. Andrews, a member of long standing and a former President of the Society, his subject being ‘‘ The Heroic Period of Geological Work in Australia ’’. The Clarke Memorial Medal was awarded to Dr. W. R. Browne, Reader in Geology of the University of Sydney. The Society invited Dr. J. 8S. Anderson, of the Chemistry School, University of Melbourne, to visit Sydney, in order to deliver two lectures under the Liversidge Research Grant. Dr. Anderson’s subjects were ‘‘ The Chemistry of the Earth ’” and ‘‘ The Imperfect Crystal ’’, and the lectures were given on October 27th and 29th respectively. A task to which a great deal of time has been devoted during the past year by the committee appointed for the purpose was the revision of the rules. A close examination and comparison of some of the rules revealed both contra- dictions and inconsistencies, which it has been the endeavour of the committee to remove. Provision has been made, among other alterations, for absentee voting for new office-bearers and members of Council. The period of years qualifying for life membership has been reduced from forty years to thirty-five,. thus bringing in a number of new life members whose loyalty to the Society throughout many years is thus rewarded. The finances of the Society are in a satisfactory state, as shown by the Annual Balance Sheet and Revenue Account. We have endeavoured to do our part in the war effort by subscribing or converting amounts totalling £1,400 to war loans during the past twelve months. The policy of inviting applications for grants for research, for which provision: is made in our rules, has been under consideration: a grant of £50 for 1943 has been made to the Australian National Research Couneil for the furtherance of scientific research. We acknowledge with grateful thanks the subsidy of me for the year 1943 from the Government “of New South Wales. The thanks of the members of the Society and myself are due to ke Honorary Secretaries, Professor Elkin and Mr. Mellor, for their work in keeping the wheels. running smoothly ; Mr. Mellor also has undertaken the onerous task of editing the Journal and Proceedings, a work so ably carried out for a number of years. by Dr. C. Anderson, who owing to important war work found himself obliged to resign from the post of Joint Secretary and Editor. Our thanks are due also to the Honorary Treasurer, Mr. Clunies Ross, to Mr. J. A. Dulhunty, Assistant: Editor, and to Mr. Maze and Dr. Bolliger, the Honorary Librarians, who have continued their efforts to increase the usefulness of the library. Part It. LIFE AND LIVING. Sir Gowland Hopkins has expressed very well the place of biochemistry in the study of life processes. ‘‘ Physiology as ordinarily understood is chiefly concerned in every case with the visible functioning of organs ; biochemistry rather with the molecular events which are associated with these visible activities. I venture to think that productive thought in biochemistry in particular calls for the widest possible survey of life’s manifestations. One of its ultimate tasks is to decide on what, from the chemical standpoint, is essential for these manifestations as distinct. from what is secondary and specific. For any such decisions the necessary harvest of contributory facts must come from many diverse fields.” In a very searching essay published in 1939 in “ Perspectives in Bio- chemistry ’’, N. W. Pirie discusses ‘‘ The Meaninglessness of the Terms, Life: PRESIDENTIAL ADDRESS. > and Living ’’. I hope to show that, incomplete as our knowledge is in most. directions, we can give some meaning to the terms. I shall develop the idea. that a living system is an everchanging and dynamic, but at the same time organised and coordinated assemblage of enzyme systems which, under suitable conditions, is able to synthesise, from simpler substances, the whole or part of each enzyme system of the assemblage. Death results from disorganisation of the assemblage beyond a certaim degree which may result from either physical means or the removal of the activity of one or more of the enzyme systems by the action of some foreign substance or the lack of substrates. There is involved here the idea that the cell protoplasm has a definite organised structure with a certain limited degree of freedom of alteration of structural pattern and is not simply a heterogeneous collection of substances in the colloidal and crystalloidal state. It is certain that almost all chemical changes taking place in living organisms are brought about by enzyme action. We may define an enzyme as an organic catalyst, probably always a protenr or containing a protein, which brings about in a smooth and orderly fashion, specific chemical changes in a substrate, the substance upon which it acts, such as the addition or removal of such groups as H, PO,, H,O and many others. Each enzyme is, so far as we know, specific, not only for the particular grouping it inmediately affects, but also the general chemical constitution of the molecule of the substrate or substrates. The product of the enzyme action may be modified by subsequent non-enzymic chemical change. For example, amino acid oxidase removes hydrogen converting the amino acid into an imino acid which subsequently by non-enzymic action loses nitrogen and hydrogen and. takes up oxygen to become a keto acid. Green (1941) expands the definition of an enzyme to *“‘ Any protein which uniquely performs some specialised physiological function’? One has then of course to define what is meant by specialised physiological function. Green would include visual purple, specific for photoreception in the retina, and hemoglobin, specific for the transfer of oxygen, aS enzymes. Enzymes such as pepsin and trypsin may be produced by living cells and passed out of the cells into the environment, but we shall not concern ourselves: with these here except to indicate that they are the products of cell activity. It is the intracellular enzyme systems that I wish to consider here. Very many enzyme systems have been isolated during the past few years from bacteria, yeast, plant and animal cells. Some of these have been obtained in crystalline and relatively pure form, others in an impure state only, but more or less separated from other enzymes. A great deal is known about the oxidative enzymes of cells which use as substrates carbohydrates or derivatives of carbohydrates such as pyruvie acid, citric acid, succinic acid and very many others. Very little is known yet about. the enzyme systems of cells which are concerned with the breakdown and: building up of lipids and proteins. All the oxidative enzyme systems seem to be made up of a speeific proteim and a non-protein or prosthetic group attached more or less firmly to the protein. The prosthetic groups vary greatly in composition and complexity from simple metals like copper to complex substances like flavin adenine dinucleotide and iron porphyrins. Each component of the enzyme is highly specific but the same prosthetic group may form a part of several different enzymes by being combined with different specific proteins. The prosthetic group acts, at least in many cases, aS a cyclic carrier being alternately reduced and oxidised. Thus consider the action of the d-amino acid oxidase isolated from liver. This is a specific protein combined with flavin adenine dinucleotide. 4 H. PRIESTLEY. The prosthetic group here is reduced by the alanine which is simultaneously oxidised and then oxidised by molecular oxygen so that it is available to repeat the cycle. ; d-Alanine +flavoprotein—pyruvic acid-++-NH,-+reduced flavoprotein Reduced flavoprotein +O,—-flavoprotein +H,O, In most cases the reaction is not as simple as this, there being 2, 3, 4 or more systems between the substrate and molecular oxygen. The only chemical change here is the transfer of 2 H atoms from the substrate to molecular oxygen and this takes place through the prosthetic group. The action of the specific protein part is essential but obscure. It is said to ‘‘ activate ’’ the substrate so that 2 H atoms can easily split off. This of course does not take us very far. There is evidence that the substrate in some way combines with the specific protein and the transfer of hydrogen takes place from the substrate to the prosthetic group in the complex substrate-specific protein-prosthetic group. Since water is essential for the process it is possible that the transfer is more complex than that indicated. As [ wish to consider the action of vitamins and drugs on cellular activities, it will be of interest to see what are the substances which are known to play the parts of prosthetic groups. The simplest enzyme systems from this point of view are the metallo- proteins where a simple metal acts as the prosthetic group. The best known of these are copper-proteins. The polyphenol oxidases of fungi and various higher plants as the potato, the monophenol oxidases of fungi and ascorbic acid oxidase of the cucumber belong here. Of more interest are the enzyme systems where one prosthetic group combined with different specific proteins gives different enzyme systems. Flavin- adenine dinucleotide is such a substance (Figure I). mf O Vv SO eee RIBOSE | Colon ON Cc’ | 4N7 CHC NC A C=O Xw-¢ CH Hee : : NH ee | O Fig. I. Here oxidation reduction changes take place on the nitrogens marked *. Enzyme systems bringing about the oxidation of d-amino-acids, purines, aldehydes, dihydrocoenzyme I and dihydrocoenzyme II are known. Two systems are known where the flavin phosphate is present without the adenine. Some of these are possibly artefacts and there is evidence that they may not exist within the living cell. Another important group contains as prosthetic group one of the pyridine dinucleotides, diphosphopyridin dinucleotide (Figure II) or triphosphopyridine dinucleotide which has another phosphoric acid group. These are specifically different. .A number of enzyme systems containing diphosphopyridine nucleotide PRESIDENTIAL ADDRESS. 5 (D.P.N.) (Coenzyme I), are known, including enzymes which oxidise, respectively, alcohol, aldehyde, lactate, malate, triose, glucose, glycerophosphate, 8 hydroxy- butyrate, glutamate and formate. Enzyme systems containing T.P.N. (Coenzyme II) are less numerous. Among them are systems oxidising hexose monophosphate, isocitrate, glutamate and glucose. Each of these dozen or more enzyme systems contains a different protein and is specific to the one substrate. H N=C—N CONH, ] C= pce a | = IN] | Nc" RIBOSE ao rt O=P O ape | OH O~ Fig. II. Systems containing an ironporphyrin are very important in most cells, plant and animal. Among these are the cytochromes, catalase, peroxidase and possibly cytochrome oxidase. Thiamine diphosphate combined with a specific protein and a metal usually ~ magnesium forms a number of enzyme systems which are of particular importance in the oxidation of ketoacids. Carboxylase of yeast is the best known of these enzymes but in animal tissues there is a diphosphothiamin system which oxidises ketoglutrate to succinate. Diphosphothiamin also catalyses carbohydrate synthesis from pyruvate or rather some step in this process, citrate synthesis from pyruvate and oxalacetate, synthesis of succinic acid and synthesis of aceto-acetate. N == C-NH_-HCI oo Ch; Mey eee N eos “N— S N— CH H thiamin chloride hydrochloride. Bigs TET, An indication of the many ways the same prosthetic group with different Specific proteins may act is given in the following table by Stern (1940)—the known biological reactions involving thiamindiphosphate with pyruvic acid (P.A.) as the substrate. A number of other similar enzyme systems are known in cells from various sources but nothing is known about their nature. There is evidence that ascorbic acid plays an important part in some enzyme systems particularly synthetic. There is much evidence to show that p-amino benzoic acid is essential for some enzymic processes in cells from many sources but how it acts is not 6 H. PRIESTLEY. Reaction. Material. Conditions. PA—Acetaldehyde + CO, Yeast. Aerobic, anaerobic. PA+5/2 O,>~3CO,+2H,0 Brain. Aerobic. PA+1/2 0,->Acetic acid+CO, Vigna aorone 2PA+H,O->Lactic acid-+Acetic acid-+CO, Cane rye’ Nels Jf Anaerobic. -PA+2H-Lactie acid ‘ ‘| Aerobic. a eee J Anaerobic. ; Aerobic. PA+Oxalacetic acid—Citric acid Muscle. Anaerobic. 2PA—-Acetic acid+ Formic acid Streptococcus. Anaerobic. PA+Glutamic acid > Alanine + «- -ketoglutaric acid : : Muscle. Anaerobic. 2PA—Suceinie acid _>Fumaric acid _>Malic acid —> Oxalacetic acid +PA-+CO, Kidney cortex. Aerobic. PA+H,0,—Acetie acid+CO, fs H,O Pneumococcus. Aerobic. 2PA wen toncene acid... Liver. Aerobic. 2PA—-6-Hydroxybutyric acid we Muscle. Anaerobic. 2PA—Succinic acid Me ae — Testis. Anaerobic. PA—Pyruvie acid . known. The importance of this will be seen later in the discussion on the effect of certain drugs on cell metabolism. The catalytic efficiencies of the different protein-prosthetic group enzymes vary very much as is indicated by their turn-over numbers, that is the number of oxidation-reduction cycles undergone per minute by the prosthetic group, or the number of substrate molecules oxidised per minute by one molecule of enzyme. Thus the flavin adenine dinucleotide protein which brings about oxidation of dihydrocoenzyme IT has a turn over number of 220, the diphospho- pyridine protein which oxidises alcohol to aldehyde has a turn over number of 20,000 at 20°C., while catalase which splits off oxygen from peroxides has a turn over number of 2,640,000 at 0° C. The firmness of attachment of the prosthetic group to the specific protein varies very greatly and there is considerable evidence that a particular prosthetic group may leave one protein and attach itself to another if various conditions are changed. The systems so far considered are oxidation-reduction systems concerned with the transfer of electrons finally to molecular oxygen and the simultaneous liberation of energy. In very few cases is there only one enzyme system between the substrate and molecular oxygen. Generally there are 3, 4 or more such systems, those nearest to the substrate having a more negative oxidation-reduction potential, those near oxygen having a more positive oxidation-reduction potential under the existing conditions. Thus Substrate +activating protein —_—_——1->F lavinadenine dinucleotide protein (Diaphorase I) —>Cytochrome oxidase Coenzyme I ——->(Cytochromes) +O, Oxidations of this type can be followed quite readily in isolated systems in vitro but they can be greatly accelerated by the simultaneous action of other systems not specifically concerned with oxidations and reduction. In particular PRESIDENTIAL ADDRESS. 7 the participation of phosphorylation processes has been much studied during the last few years and shown to be of very great importance. Ina recent review Fritz Lipmann (1941) has discussed the parts played by phosphorylation and dephospholylation in many enzymic processes which may take place in cells. Thus the conversion of glucose to glycogen and vice versa is always associated with phosphorylations, the oxidation of glucose and of trioses is probably always as their phosphorylated compounds. We have seen that most of the prosthetic groups considered contain phosphoric acid but in addition adenosine triphosphate, creatine phosphate and, in some animals, arginine phosphate are also necessary as sources of phosphoric acid radicles. Quite a considerable part of the energy exchanges in cells can and does come from changes in phosphoric acid bonding. Of other enzyme systems in cells not a great deal is known at present excepting the actions of certain of them in isolated systems. Thus several proteolytic enzymes and peptidases have been isolated from cells from various sources. The specificity and kinetics of these enzyme systems have been studied but not much information has been obtained which can be used in considering the reactions within the living cell. Claims have been made for the synthesis of proteins by proteolytic enzymes from simpler peptides but I have repeated most of the work and cannot satisfy myself that proteins are formed. A consideration of energy requirements shows the great unlikelihood of such Synthesis 7m vitro in isolated systems. Behrens and Bergmann (1939) have shown that the peptide linkage can be formed in homogeneous solution in vitro by papain. Many workers have shown the importance of the sulphydryl group in the proteolytic enzymes of cells. When this is oxidised to the disulphide the enzymes become inactive. Reduction and activation can be brought about in many cases by cyanides and cysteine. Fruton and Bergmann (1940) consider that the eyanide forms a prosthetic group and consider that glutathione may act in the same way within cells. The proteolytic enzymes are proteins, but excepting possibly for the foregoing no prosthetic groups have been found. Very little indeed is known about cell lipases and esterases and almost nothing about the enzymes concerned with the synthesis and breakdown of fatty acids. Many other enzyme systems have been demonstrated in cells from various - sources and it is certain that we are nowhere near knowing all the enzymes in any one cell. It is evident then that living cells contain a great number of different enzymes catalysing very many different chemical reactions. Within the cell there must be considerable interaction between different enzyme systems the product of one enzyme system being the substrate of another. Many of these are known in vitro with extracts of the cells, but whether any of these actually occur within cells is not certain. It will be observed that little has been said about synthetic processes. Synthesis of complex organic substances from simple substances is one of the most characteristic activities of living cells. For synthesis to occur we almost always require the intact living cell and so far it has not been possible to study in any detail particular synthetic processes among all the complex associations within the living cell. Not only do synthetic processes require the intact cell but so also do some oxidising processes. For example, surviving slices of liver with intact cells will oxidise fatty acids, but this property is lost so soon as the tissue is minced. The oxidation of the lI-series of amino-acids requires the intact cell. Further, most complex substances synthesised in cells have a higher energy content than the substances from which they were synthesised, so energy must 38 H. PRIESTLEY. be provided at the right time and in the right place to enable synthesis to go on, that is there must be coupled with the synthetic process other systems which provide energy by oxidation or in some other way. Thus Borsook and Jeffreys (1935) calculate that the net change of free energy in the synthesis of urea from ammonia, acid and carbon dioxide is +14,300 calories. Krebs (1935) has demonstrated that the synthesis of urea in the liver can go on only with the simultaneous oxidation of some substrate—glucose or lactic acid. Similarly Krebs (1935) has shown that the synthesis of glutamine is, under physiological conditions, an endothermic reaction. In the kidney and brain the energy required is derived from oxidation of some substrate. Extracts of tissues which synthesise glutamine contain an enzyme which splits glutamine into glutamic acid and ammonia, a reaction which does not take place with the living tissue. There is considerable evidence that this glutaminase is the same enzyme which, under the conditions existing in the living cell, brings about synthesis of glutamine. Glycogen and starch can be synthesised in small amounts from glucose-1-phosphate by extracts of suitable animal or vegetable tissues, which of course are mixtures of various enzyme systems and substrates, while in the intact cell the reaction may go almost wholly in the direction glucose->glycogen or starch, but can be reversed. The methods by which couplings of energy are accomplished in living cells is very obscure but has something to do with the structural disposition of the enzyme systems within the cell. It is evident that the structure of the living protoplasm is not a mere static condition of cell substance but an active, dynamie, condition necessary for the uninterrupted transfer of energy. The energy yielding system must be very close to, if not actually in contact with, the energy receiving system. A number of Russian biochemists, in particular Oparin and Kurssanov (1941), have advanced the thesis, and supported it by much evidence, that in the living cell a part of an enzyme exists in solution and exerts a hydrolytic action only, and a part is adsorbed to the water poor, plasmatic structure and works only by synthesis. The relative proportion between the enzyme distribu- tion in the watery phase and that adsorbed on the structural phase depends on certain physico-chemical and energetic relations within the cell. The ratio of speed of synthesis and hydrolysing action, or the prevailing direction of the enzymic process, in the cell, depends on the quantitative relations between the structurally bound and the free enzyme and also on the availability of adequate energy for the synthesis. Alterations in the prevailing direction of enzyme action in the cell can be caused by the influence of environmental factors on the inner condition of the cell as well as through inner causes. Various authors have discussed the internal structural organisation of cell protoplasm. KR. A. Peters (1939) suggests a cytoskeleton consisting of three parts, surface proteins, cytoplasmic proteins and nuclear proteins. He lkens it to the nervous system with nuclear protein corresponding to the central nervous system, to conducting units the cytoplasmic protein and to the surface protein the rédle of acceptor or receptor. He makes no suggestions as to the possible structure of such a cytoskeleton and concerns himself mainly with problems of transmission of effects from the surface to nuclear proteins or vice versa. A much more detailed theory as to the structure of cytoplasm in terms of native proteins has been developed by Dorothy Wrinch (1941). Briefly she summarises the idea as follows: ‘‘ Recent work on native proteins .. . yields a picture of the native protein as consisting of isolated or associated molecular units, having rigid globular structures, on the surface of which some or all the R groups are rooted in definite spatial patterns. On this basis a new picture is suggested for the structure of cytoplasm, consisting of interlacing PRESIDENTIAL ADDRESS. Q: native protein frameworks to which fats, carbohydrates, water and other foreign molecules are attached, located in an immense interpenetrating water phase, possibly with proteins in solution.’? She states that native proteins have, in general, very ‘ sticky ’’ surfaces, in the sense that they readily interlink with foreign molecules of a variety of types. They have on their surfaces, patches which have a “specific stickiness’? by means of which they can interlink preferentially with other individual proteins and with certain foreign molecules. These patches are apparently spaced on their surfaces so that other proteins are held in position within certain limits. Proteins so linked together can thus form rods or strands of some degree of stability. Bernal and Fankuchen (1937) as a result of X-ray examinations state that the molecule of tobacco virus is probably made of piles of submolecules of dimensions 22A x 20A « 20A, somewhat smaller than the normal protein molecule. The molecules of different strains of tobacco virus and of cucumber virus have substantially the same general shapes and size and are made up of similar subunits but these are arranged somewhat differently in the different tobacco virus strains and more markedly so in the cucumber virus. There is abundant evidence that proteins can form relatively stable associa-. tions with other proteins and with certain non-proteins, notably lipids. The specific nature of the associations, in other words the “ specific stickiness ’’, is well seen in antigen-antibody complexes. As regards the structural arrangement of the proteims in the cytoplasm we have to remember that an active cell may contain 95%, of water and yet have a definite form and a certain degree of rigidity. Wrinch (1941) suggests something of the nature of a diamond lattice of protein molecules joined together with a relatively immense interpenetrating water phase. She shows that such a structure will have some measure of rigidity and some tensile strength while at the same time showing plasticity and structural viscosity but allowing cyto- plasmic streaming. The Wrinch model is to some extent based on the supposed eyclol structure of the globular proteins but this is not necessary for any other type of globular or three dimensional structure for the protein molecules could be fitted into a similar pattern. Many considerations lead to the belief that what we might call the “ living ’”’ proteins are “ globular ’’ in nature and do not consist of simple peptide chains. The proteins which show the polypeptide chain structure are the fibrous proteins mostly extracellular and ‘ dead ”’. There are possibly exceptions here. The myosin of the muscle cell is fibrous but it is to serve a special function and is not an ordinary structural cytoplasmic protein. A point of some importance, which does not seem to have been made before, is that the individual protein units of such a network as has been postulated need not, indeed cannot, all be the same. How far we are to consider any of the protein of the cytoplasm, whether in the interior or on the surface of the cell as purely structural, it is not possible to say. When one considers the immense variety of enzyme proteins in the cytoplasm of any one cell and the fact that many enzyme activities are conditional on the integrity of the cell, it seems apparent that much, if not most, of the structural protein has enzymic properties or at least has enzyme proteins more or less firmly attached to it. One point must be stressed. There is no such thing as cytoplasmic protein as a single entity. The cytoplasm contains a great variety of different proteins many of which are enzyme proteins, others possibly purely structural or playing some part which is neither strictly structural nor strictly enzymic. In this connection it has been shown by Engelhardt and Lynbimova (1939) that myosin, or some substance at present indistinguishable from it, acts enzymatically in the dephosphorylation of adenosine triphosphate to diphosphate. Needham et al. (1941), in confirming this, have suggested a theory of muscle 10 H. PRIESTLEY. contraction in which adenosine triphosphate gives phosphorus to myosin—the phosphorylated myosin being the extended form. On stimulation the extended myosin phosphate liberates phosphate ions and contracts. The distribution of different proteins has been much more actively studied in the nucleus than in the cytoplasm. Caspersson (1940) has shown that nuclei contain protamines, histones and probably proteins of a globulin nature. These three types of protein are found in the chromosomes. Chromosomes show a banded structure or rather alternate areas which show respectively staining by various stains and absorption of ultra-violet rays on the one hand and absence of these properties on the other hand. The first areas have much nucleic acid, the second little or none. Caspersson has shown that the nucleoprotein bands have possibly all three of protamine, histone and globulin types ; in the interband spaces chiefly globulin and protamine types and chiefly histone type in the nucleoli and the related puff and chromocentral regions. The linear pattern of the chromosomes is most likely due to the protamine type. Protamines are of the polypeptide chain type, ideally suitable for a linear arrangement. Mazia (1941) has shown that some chromosomes contain a relatively large amount of ‘‘ matrix ’’ protein which is digestible by pepsin, while the ‘‘ skeletal’’ protein is not digested. Plant chromosomes, so far examined, show little of this ‘‘ matrix ’’ protein. The chromosome may be pictured as a protamine-like thread, with nucleo protein at intervals and a sheath of more complex protein related to the nucleolus and in some cases, if not all, a complex matrix protein. There are two types of nucleic acid in nucleoproteins—ribose nucleic acids and desoxyribose nucleic acids depending on the sugar they contain. Ribose nucleic acids contain as the pyrimidine portion uracil, while desoxyribose nucleic acids contain the methylated derivative thymine. The principal nucleic acid in the chromosomes is desoxyribose nucleic acid, but there is some evidence that the bands of chromatin contain also ribose nucleic acid. The nucleolus definitely contains ribose nucleic acid. The nucleic acids appear to play a large, and possibly essential, part in protein synthesis in the nucleus and in the cytoplasm. The nucleic acid in the cytoplasm is of the ribose nucleic acid type and in cells which are functionally active or which are dividing rapidly the concentration of ribose nucleic is high. Claude (1941) claims that the greater part, if not all, the ribose nucleic acid in the cytoplasm is in the mitochondria as a nucleo- protein-phospholipid complex. Now the mitochondria have long been thought to play an important part in the production of enzymes such as pepsin and trypsin, which are manufactured in cell cytoplasm and pass from the cell. These are, of course, proteins. It is of interest also that in segmentation of the ovum, where the cell divides a number of times without increase in volume and so without increase in cytoplasmic protein, the ribose nucleic acid is small in amount. The identification of the gene with a nucleoprotein seems to be generally accepted. Now during some phase of mitosis, probably the interphase, each gene duplicates itself, that is there is synthesis of the specific nucleoprotein of each gene. The gene nucleoprotein must then be looked upon as a synthesising enzyme. This is exactly paralleled by the power of virus proteins to reproduce themselves. A virus is a nucleoprotein with nucleic acid of the ribose type somewhat loosely attached to the protein. The protein part is not a protamine but is more complex, having a greater range of amino acids. Different strains of the tobacco virus have been shown by Stanley to differ in the relative amounts of the different amino acids, so the specificity of the different strains is evidently related to the make-up of the protein part. The whole virus can be looked upon as a protein synthesising enzyme. In the case of the gene there is simple PRESIDENTIAL ADDRESS. . 11 cduplication, one gene makes another exactly like it. The same must be the case with the virus. Now this idea of a protein synthesising itself and so being an enzyme must, it seems to me, be extended in a modified way to all the proteins in a cell, other- wise one is met with the difficulty of suggesting how the enzyme, itself a protein, which syntheses a protein is itself synthesised. Bergmann and Niemann (1937) have suggested that when an intracellular enzyme has at its disposal a number of protein fragments of different size and structure it subjects these fragments ‘to a series of transformations and thereby reconstructs one peptide bond after another until there is produced a protein pattern which is stable in the presence of the enzyme. Proteinases should exist which have the ability of synthesising replicas of their own structural patterns. It seems to me that this idea demands a separate self-synthesising proteinase for the synthesis of each cell protein. Delbruck (1941) has formulated a theory of short distance interaction, which does not attract the correct substrate, but which reduces its energy of oxidation, thus selecting the correct substrate for oxidation and thereby also for the synthesis. It would take too long to develop here the whole of his ideas which he applied to the duplication of chromosomes. The same process could, however, be applied to other cell proteins. Concerning the action of ribose nucleic acid and the relation of the genes to cell Synthesis, which of course must be very direct for the gene sets the pattern of the cell enzymes, it is of interest that the great increase in the cell cytoplasm pre- jiminary to cell division takes place after the nuclear membrane has ruptured and the cytoplasm comes into very close relation with the chromosomes. The action of viruses on the cell cytoplasm is also relevant here. The presence of virus protein in the cell cytoplasm is accompanied by changes in the enzymic activities of the cell in a variety of ways. Thus the varigations one so commonly sees in virus-infected plants must be due to the effect of the virus on the chlorophyll producing enzymic system. The importance of nucleoproteins in cell enzyme systems is seen in the fact that many of the important oxidising enzymes are of the nature of nucleo- proteins. Di- and tri-phosphopyridin nucleotides and flavin-adenin dinucleotide an be considered as particular types of nucleic acid. Adenosin triphosphate is a derivative of a nucleic acid. The work of Schoenheimer and his co-workers on the use of labelled atoms has introduced quite radical changes in our ideas as to the stability of the protein molecule. By feeding rats with amino acids containing heavy nitrogen N,; or heavy nitrogen and deuterium they have been able to show that there is going on a very active exchange in nitrogen and hydrogen atoms in proteins throughout the body. When leucine containing N,, was fed over 50% of the N,, was retained in the body in proteins, although the animals were in nitrogenous equilibrium, so that according to current views there should be little protein Synthesis proceeding. Further, the N,; was found not only in the leucine of the protein but alsc in most of the other amino acids, particularly glutamic and aspartic acids. This means that there is either a complete replacement of tissue protein by new synthesis every few days or that the peptide and other nitrogen linkages in the proteins can open, lose nitrogen, and have it replaced. The first seems impossible, the second means that proteins are exceedingly labile substances. These workers (Schoenheimer, Rattner and Rittenberg, 1939) consider that peptide links open, liberating amino acids which mix with others from whatever source. Some of these free amino acids re-enter directly into vacant positions left open by the rupture of peptide linkages, others transfer their nitrogen to deaminated molecules to form new amino acids, which in turn can enter the same chemical cycles. There does not seem any sound reason for postulating that amino acids as such are separated from the proteins and subse- 12 H. PRIESTLEY. quently replaced by others of the same kind. It seems to me more likely that the fabric structure of the ‘*‘ living ’”’ protein is very labile and can open and lose nitrogen and carbon atoms from time to time and at different places. Indeed it is probable that the activity of ‘‘ living ’’ protein depends on this lability for, according to Rittenberg (1941), tendon, a ‘‘ dead ”’ protein, does not show the isotope replacement. Frey-Wyssling (1938) suggests that the funetion of nucleic acid is to ‘* protect ’’ the active side chains of the protein molecule. In view of what has been said as to the probable nucleoprotein nature of the gene and the relation of cytoplasmic nucleic acid to cytoplasmic protein synthesis, it seems not improbable that the nucleic acid may act by ‘‘ protecting ”’ the protein in making it less labile in the sense just discussed. It might be suggested that the specific enzymic activity of proteins is associated with their lability, while their ability to synthesise themselves is associated with the change in this lability brought about by the nucleic acid. If, as has been postulated, the cell enzymes are largely arranged in a structural pattern, there should be restrictions within the living cell on various enzymic processes occurring in series in cell extracts where there is a large degree of freedom of movement of enzyme and substrate molecules. Small concentrations of cyanide will inhibit a large part, but not the whole of the respiration of living cells. Cyanide specifically poisons cytochrome oxidase, so there is evidently a part of the oxidation processes in the cell which does not go through this system. Ogston and Green (1935) find that the oxidation of hexose phosphate by intact yeast cells is cyanide-sensitive, but its oxidation by a system reconstructed from extracts of the same cell is relatively insensitive to cyanide. It is evident then that that part of the oxidation which, in the living cell, goes through cytochrome oxidase cannot be linked with the other system, probably a flavin protein system, so that there must be spatial separation between the systems. Commoner and Thimann (1941) have shown that there is a relationship between growth and respiration in plants. In the Avena coleoptile, they have shown that the effectiveness of auxin as a growth hormone is related to the activation of a specific and small part of the total respiratory system (the four carbon dicarboxylic acids). The respiratory activity of the four carbon acids, oxaloacetic, malic, fumaric and succinic acids, which act as hydrogen carriers, is necessary for all of growth, but is responsible for only a small part of total respiration ; when it is blocked, growth ceases entirely but the respiration rate falls only 10%. Here, too, there is evidence of specific orientation of the respiration systems. In a recent article, which was seen after the greater part of ihae address was completed, Commoner (1942) discusses this point and states ‘‘ It must be coneluded that the cellular protoplasm is characterised by an interenzyme structure which plays the predominant réle in determining the course of chemical events in the cell. The enzymes in the living cell are interrelated in such a manner as to constitute a relatively rigid structure which limits and orients the chemical activities which they carry out.’’ That the structural pattern can be altered and so bring about changes in cell metabolism is indicated by the changes which take place on fertilisation of ova. On fertilisation of marine ova one can see violent streaming and churning movements in the cytoplasm. Associated with these are usually marked alterations in the respiration rate. Many years ago Warburg (1908) observed that fertilisation of the sea urchin eggs was followed by a considerable increase in the respiration rate. Whitaker (1933) has demonstrated that if the respiration of the unfertilised egg of marine animals is low, then it rises on fertilisation, while if it is high it falls when the egg is fertilised. PRESIDENTIAL ADDRESS. is} So far we have considered only the internal constitution of a living cell and have said nothing about the factors which control the activity of the “large assemblage of enzymes within the cell so that it works as a coordinated whole. We can say nothing about the internal arrangements which compel the various enzyme systems to work in the right order and each to the extent necessary for the coordinated working of the whole. Some of this control is conditioned by the external environment. In the unicellular organism like an amceba or a bacterium, the external environment is the fluid surrounding it. Changes in the chemical composition of this fluid, the availability of suitable substrates, the effect of dissolved substances on the osmotic pressure, the surface tension, the hydrogen ion concentration and so on as well as temperature and other purely physical factors all are concerned. Increasing the temperature, raising the concentration of substrates within certain limits increase the activity and the rate of growth of unicellular organisms. Different kinds of unicellular organisms have different complements of enzymes so that the substances produced by enzyme action and excreted from the cell are very different and other activities of the cell are different. In the multicellular organisms the same considerations apply to each individual cell, but here there is an extraordinarily large measure of control of the nature of the environmental fluid. Each cell in the multi- cellular organism plays its part in this control, besides having its own specific activity. Thus in the higher animals the hydrogen ion concentration of the blood and tissue fluids and the content of many of the dissolved substances are kept remarkably constant and this constancy is essential for the continued coordinated activity of the cell. Destroy this constancy and the enzyme systems within some or all of the cells get out of step. The introduction of substances normally foreign to the cells into the environ- mental fluid may also affect some or all of the cells in a variety of ways. Here we have to consider the effects of drugs on cells‘and their activity. Certain substances are required in the diet of organisms, whether unicellular or multicellular, in very small quantities only, if the cells are to function properly. Others must be present in the environmental medium in small quantities but are substances which have not entered with the diet but have been produced by certain cells of the multicellular organism. These are the hormones. In a recent article, Green (1941) has developed the thesis that ‘‘ any substance which occurs in traces within the cell and which is necessary in traces in the diet or medium must either be an essential part of some enzyme system or the enzyme itself. The trace substance-enzyme thesis boils down to the view that enzyme catalysis is the only rational explanation of how a trace of some substance can produce profound biological effects.’’ Among these substances are the vitamins and hormones. We have seen already that certain of the vitamins form parts of essential enzyme systems. Thus nicotinic acid, the antipellagra vitamin, is a part of the pyridine dinucleotide enzymes, riboflavin is present in several different enzyme systems and so too, is thiamine, vitamin B,. The relationship of the other vitamins to enzyme systems we do not yet know, but there is some indirect evidence that they play similar parts. It is not unreasonable to consider the vitamin A- protein complex, visual purple, in the retina as part of an enzyme system. Cells vary greatly in their requirements for vitamins from outside. Some free- living cells seem to be able to synthesise every organic substance they require from inorganic materials. For example one of the sulphur bacteria grows on a strictly inorganic medium yet contains thiamine, pyridoxine, biotin, nicotinic acid, pantothenic acid and riboflavin, which it must have synthesised (O’ Kane, 1942). Some forms of proteus vulgaris require nicotinic acid or one of its derivatives as the sole growth factor—other strains grow well on a simple ammonium lactate medium. Streptococcus hemolyticus needs at least thiamin, 14 H. PRIESTLEY. nicotinic acid, adenine or related purines, and probably biotin. In the higher multicellular organisms the requirements for different vitamins is in general wider, but even here there are differences. So far as we know at present, ascorbic acid must be provided for only man, the higher apes and the guinea-pig ; other animals synthesise what they require. Hormones are chemical substances produced in one set of cells affecting cell activity in some other part of the organism. They have many parallels. with the vitamins. The concentrations at which hormones exert their activity are of the same trace order of magnitude as those of vitamins. The hormones. vary greatly in chemical complexity. Some, such as insulin, are highly complex proteins, apparently without any prosthetic group. Others, such as adrenalin, are quite simple chemical substances. Green (1941) suggests that we may consider hormones as either potential enzymes or as prosthetic groups. The protein insulin plays an important part in some phases of carbohydrate metabolism in the animal body. Its action strongly suggests a catalytic action,. but whether insulin can itself act as an enzyme or can form a part of an enzyme system, we do not know. Thyroxin occurs in the body in combination with protein-thyreoglobulin, and as such it leaves the thyroid gland. Thyroxine in some way controls the metabolic level of animal cells and it is not improbable that, in combination with some other protein within the cells, it acts as an. enzyme or part of an enzyme system. Such simple hormones as adrenalin and the sex hormones might act as prosthetic groups of enzyme systems, in the case: of adrenalin probably of short-lived duration. Certain drugs affect cells by interfering with enzymic activity, either by competing with the normal substrate of some enzyme system or by competing with the normal prosthetic group of some enzyme producing a pseudoenzyme which is inactive, or by combining with the enzyme and rendering it inactive. Malonic acid can compete with succinic acid as substrate for the enzymic system present in most cells which catalyses the reaction succinic acid = fumaric acid but is itself not attacked by the enzyme. Para-amino benzoic acid is essential for the metabolism of many eal or rather for the metabolism associated with reproduction. Some cells can synthesise it themselves, but others cannot, and it must be supplied from outside. Bacteria which cannot make it have their growth stopped by the closely related sulphanilamide (Figure IV), but it requires about 5,000 times as much sulphan- ilamide as the coincident p-amino benzoic acid to inhibit growth of streptococci in vitro. It is believed that the action is competitive, the p-amino-benzoic acid being either an essential metabolite for certain cell processes or more likely a part of some enzyme system. Sulphanilamide does not affect the normal oxidising systems of cells. NH, COOH N SO,.NH, Fig. IV. The idea that competition may occur between closely related substances: had led to the study of substances related to parts of known prosthetic groups such as nicotinic acid, and to other essential growth factors which probably play a part in enzyme action such as pantothenic acid. Thus pyridine-3- sulphonic acid can inhibit bacterial growth by competing with nicotinic acid PRESIDENTIAL ADDRESS. 15 (Mellivain, 1940), the sulphonic analogue of pantothenic acid competes with pantothenic acid (Snell, 1941). Some substances can combine with a particular part of an enzyme system and so put it out of action. For example, cyanides poison cells by combining with some component of the cytochrome system and so putting out of action the main path of oxidation in aerobic cells. There may be, and probably always are, several paths of oxidation in cells, but one of them, generally that connected with the cytochrome system, accounts for the predominant part of the oxidation. Yeast cells are very sensitive to cyanides, but Stier and Castor (1941) have developed a substrain of yeast by culturing in a medium containing cyanide which lacks cytochrome oxidase. Here there must have developed a considerable alteration in the enzyme systems of the yeast cells. Surface-active substances have marked effects on cells in many cases leading to death as in those which act as powerful bactericidal agents, in other cases interfering with cell activity to a less extent. Surface-active substances are effective protein denaturing agents and can dissociate conjugated proteins. In some cases the action of the surface-active substance may be removed by washing away the substance, so the change in the proteins cannot be great. Michaelis and Quastel (1941) have brought forward evidence that the narcotics, which form an extremely varied chemical group and are surface- active substances, produce their action by combining in some way with some flavo-protein or some component of the tissue respiratory system between flavo- protein and cytochrome oxidase. Some drugs appear to act by inhibiting enzyme systems in cells and, by providing unused substrate, apparently stimulate other enzyme systems. An example of this is dinitrophenol. This greatly stimulates oxygen uptake by certain bacteria and yeasts and also animal cells, but causes inhibition of cell growth. It appears that, by blocking some of the synthetic reactions, it allows complete oxidation of substrate. Many other examples could be given of the effects of drugs on cell enzyme systems, but sufficient has been said to indicate that this aspect is an important one, if not the most important one, in the action of drugs. Even with the limitations of existing knowledge, to develop the thesis which I have advanced fully would take a considerable time and I have had to make a selection of available evidence. I hope, however, that I have been able. to make a good case for the idea the life of a cell is dependent on, and is inherent in, the enzymes contained in the cell and the organisation within the cell of these enzymes. I shall conclude with a quotation from Claude Bernard, the father of modern experimental physiology, on the subject of scientific theories: ‘ they are only partial and provisional truths which are necessary to us, aS steps upon which we rest, So aS to go on with investigation ; they embody only the present state of knowledge and consequently they must change with the growth of science ’’. REFERENCES. Behrens, O. K., and Bergmann, M., 1939. J. Biol. Chem., 129, 587. Bergmann, M., and Niemann, C., 1937. Science, 118, 301. Bernal, J. D., and Fankuchen, I., 1937. Nature, 139, 923. Borsook, H., and Jeffreys, C. E. P., 1935. J. Biol. Chem., 110, 495. Caspersson, T., 1940. Chromosoma, 1, 562. Claude, A., 1941. Cold Spring Harbour Symposia, 9, 263. Commoner, B., and Thimann, K. V., 1941. J. General Physiology, 24, 279. Commoner, B., 1942. Quart. Review of Biology, 17, 46. 16 H. PRIESTLEY. Dellruck, M., 1941. Cold Spring Harbour Symposia, 9, 122. Engelhardt, V. A., and Lynbimova, 1939. Nature, 144, 668. Frey-Wyssling, A., 1938. Protoplasma Monographien, 15, 317. \Fruton, J. S., and Bergmann, M., 1940. J. Biol. Chem., 133, 153. ‘Green, D. E., 1941. Advances in Enzymology, 1, 177. Krebs, H. A., 19385a. Biochem. J., 29, 1620. Krebs, H. A., 19356. Biochem. J., 29, 1951. Kurssanov, A. L., 1941. Advances in Enzymology, 1, 329. Lipmann, F., 1941. Advances in Enzymology, 1, 99. MclIlwain, H., 1940. Brit. J. of Exper. Path., 21, 136. Mazia, D., 1941. Cold Spring Harbour Symposia, 9, 40. Michaelis, M., and Quastel, J. H., 1941. Biochem. J., 35, 518. Needham, J., Shen, S. C., Needham, D. M., and Lawrence, A. S., 1941. Nature, 147, 766. ‘Ogston, F. J., and Green, D. E., 1935. Buochem. J., 29, 1983. Kane, DD? J, 1942. I Bact. 43, 7. Peters, R. A., 1939. Perspectives in Biochemistry. Rittenberg, D., 1941. Cold Spring Harbour Symposia, 9, 283. ‘Schoenheimer, R., Rattner, S., and Rittenberg, D., 1939. J. Biol. Chem., 130, 703. Snell, E. E., 1941. J. Biol. Chem., 141, 121. Stern, K. G., 1940. Ann. Review Biochem., 9, 30. Stier, J. J. B., and Castor, J. G. B., 1941. J. Gen. Physiology, 25, 229. Warburg, O., 1908. Zeit. physiol. Chemie, 57, 1. Whitaker, D. M., 1933. J. Gen. Physiology, 16, 497. Wrinch, D., 1941. Cold Spring Harbour Symposia, 9, 218. NOVA PUPPIS 1942. Bye wv. WOOD. Manuscript received, February 13, 1943. Read (in title only), April 7, 1943. The announcement of the discovery of this Nova, in which it was stated to be blue in colour, first magnitude and still brightening, was passed on to us by Dr. J. M. Baldwin, Government Astronomer for Victoria, on November 13, 1942. When the Nova was seen here that evening at 11h. 30m. U.T. it was distinctly reddish and was noted to be a little brighter than ¢ Canis Majoris (n=1-63). The red colour seemed to show that the emission lines had begun to dominate the spectrum, the red being due to H« (e.g. Payne-Gaposchkin and Gaposchkin, 1938) and this, together with the failure of the magnitude to increase, seemed to indicate that maximum had passed. This note records some observations made at intervals during a period of frequently cloudy weather. Transit observations (with fundamental stars on the FK3 system) were made of the star’s position on five occasions with the result fee Sh. 7m. 588-33 Wet —a00 o 60-4 (1900 -0) This corresponds to galactic longitude 220°-6 and latitude +0°-1 with the position (1900-0) of the galactic pole R.A. 12h. 40m., Dec. +28°, so that the star conforms to the well-known tendency of nove towards galactic concentration. Using this position, available sources were examined to find if possible a record of the star before its outburst. The place which it would occupy in the Perth Astrographic Catalogue was computed, but the star was not recorded. Neither was it recorded in the ‘* Harvard Sky ”’ plates or our copy of the Franklin Adams Chart. This sets a limit to its brightness at various epochs as in Table 1. TABLE 1. Justification Date of Plate. Mag. Chart. for Limiting Limit. Mag, 1903, May 6... a 113 “Harvard Sky’ \ 1 1903, December 23... 114 “Harvard Sky” f ioeewpril 19... ae 16 Franklin Adams Chart. 2 and 3 1913, February 28 13 Perth Astrographic Catalogue } 4 1914, April 15 .. i 13 Perth Astrographic Catalogue 1927, February 6 aS 14 Union Observatory Chart. 1. Pickering, E. C., 1914. Cuorc. Harv. astr. Obs., 185. . Russell, H. N., Dugan, R. S., and Stewart, J. Q., 1927. Astronomy. Ginn and Co., New York, 600. 3. Chapman, S., and Melotte, P. J., 1914. Mem. R. astr. Soc., 60, Part IV. 4. Curlewis, H. B., 1916. Perth Astrographic Catalogue, 17, 3. bo The Union Observatory Chart shows, in the vicinity of the Nova, a number of stars intermediate between that of the Perth Astrographic Catalogue and the Franklin Adams Chart, so that 14 would probably be a fair estimate of its B—April 7, 1943. . 18 H. W. WOOD. limiting magnitude. From this table it may be seen that the star must have brightened by about (at least) fifteen magnitudes, or by a factor of a million times, a remarkable example of the spectacular character of nova outbursts. The spectrum was photographed on two occasions, November 17, mid exposure 17h. 32m. U.T., and November 18, mid exposure 16h. 36m. U.T, The continuous spectrum is faint by comparison with the emission which is dominated in the region recorded, from 4200A° to 5700A°, by the H@ line. The Hy is also quite prominent. A number of magnitude estimates has been made. The Henry Draper Catalogue (Cannon and Pickering, 1919) and the Revised Harvard Photometry volumes (Harvard Annals, 1908) were consulted for the magnitudes of the comparison stars which appear in Table 2. Table 3 records the observations of magnitude in which the decimal method of representing the comparisons is used TABLE 2. | FED: Star. ' Name. Mag. Star. Number. Mag. a e Puppis. 2:88 h 70002 5:66 b co Puppis. 3°27 k 69123 5-82 c € Puppis. 3°47 l 69511 6-20 d c Puppis. 3°72 m 67977 6-20 e k Puppis. 3-80 oO 68623 6-74 Pp 69280 6-76 H.D. Number. q 69066 7-00 tt 68980 4-77 g 70556 only TABLE 3. Deduced | | Deduced Julian Day. Comparison. Mag. Julian Day. | Comparison. | Mag. 2,430,000 + 2,430,000 + 676-98 Je CMa 733-96 m 5p 6-48 681-10 a 5e¢ 3:18 737-96 m 7p 6-59 689-14 De Yad 3-58 739-02 ‘golf 76) 6-59 a e ge 3°57 741-04 re aa 6-59 689-15 g 5-17 741-92 mv. 4p 6-59 fod) kk 4-98 742-97 mop 6-59 g 5:17 744-02 m 8p 6-65 See: f 4h 5-18 eevee m 8 p 6-65 691-22 fp One Stn Kk 8a 6-76 f : h 5:66 : m 7p 6-59 62g 01 f 8k 5-61 f fou k 7q eer 696-97 k 5:82 : m 6 p 6-54 Brae k 71 6-09 ieee? k 6 q 6-53 m 20 6-31 f - ae | 6-17 Observatiojns by W. H. Ro|bertson 708-00 m 50 6-47 689-96 fide 5-04 may, Gp 6-38 691-00 g 3h 5:32 708-95 o 3q 6-82 691-99 oe 5-51 715-03 oO 2. 6-70 698-98 k 5m 6-01 |. 721-96 o 4q 6-84 k 5] 60h p 6-76 700-98 m 6-20 730-97 m 8p 6-65 703-98 Is Oho 6-48 732-01 mG) 6-54 NOVA PUPPIS 1942. 19 whereby, for example, ¢3 e means a magnitude three-tenths of the way from ec to e. Comparisons, made by naked eye till November 18 and thereafter with a two-inch telescope, were sometimes not easy because of the differences in colour among the stars, the Nova being distinctly reddish. Some observations by W. H. Robertson are included at the end of the table. Since January 6 only stars selected by Father D. J. O’Connell, Director of Riverview College Observatory, for use by the Variable Star Section of the N.S.W. Branch of the British Astronomical Association, have been used. The star H.D. 67736 may be variable. Its magnitude according to Harvard photometric observations (Harvard Annals 1895 and 1901) is 7-36, but the residuals are slightly large and to me it has appeared (January 6 to January 17) brighter than either p or q. The rapidity of the magnitude development of the Nova seems to place it in the ‘* flashing ”’ type of Gerasimovie (1936) ; by December 14 it had fallen through five magnitudes but since then has been fairly constant. - REFERENCES. Ann. Harv. Coll. Obs., 1895, 34, and 1901, 45. Ann. Harv. Coll. Obs., 1908, 50 and 54. Cannon, A. J., and Pickering, E. C., 1919. Ann. Harv. Coll. Obs., 93. Gerasimovic, G. P., 1936. Pop. Astr., 44, 78. Payne-Gaposchkin, C., and Gaposchkin, S., 1938. Variable Stars. Harvard Observatory, . Cambridge, 252. OO A POLYHEDRAL MODEL.OF THE PROJECTIVE PLANE. By F. A. BEHREND. Communicated by Professor H. 8. CarsLaw. Manuscript received, February 13, 1943. Read (in title only), April 7, 1943. The first satisfactory topological model of the projective plane in 3-space was described by Boy? (1901): a bounded surface which crosses itself along a certain curve but otherwise is everywhere regular. The problem of an algebraic, or even only an analytic, representation of Boy’s surface is still unsolved. Recently Merz? (1942) and Humbert? (1942) obtained polyhedral varieties of Boy’s surface. Merz’s surface has 16 faces, 18 vertices and 33 edges; Humbert’s has 16 faces, 24 vertices and 39 edges. In the present paper a model is described which appears* to be of a much simpler type; it has 10 faces, 12 vertices and 21 edges; it is likely that it is the simplest polyhedron of this kind.® The surface is obtained as follows: it possesses the line 7,=—%,—a%, a8 an axis of order 3; there are four vertices in each coordinate plane ; those in the plane 7,=—0 are conveniently chosen as P,=(0, 0, 2), Q,=(0, Se aL), Rk, =(0, UW coast ©) S,=(0, 1, 1), and the others are obtained by cyclic permutation : P,=(2, 0, 0), G2=(— e 0, —3), R,=(—1, 0, 1), S,=(1, 9, 1), P,=(0 ’ 2, 0), V3=( ok 3, es 0), R,=(1, =, 0), S,=(1, 1, 0). The three (congruent) quadrangles P\Q; RS; («=1, 2, 3) (Fig. 1a) form three faces of the polyhedron lying in the coordinate planes and crossing each other along segments A,B, of the coordinate axes where Ay, By, are the unit points on thie x, axis (vj=0 for jk, x; —-L1 for 7—k) (Fig. 2a). The three (congruent) quadrangles P,Q,R3,8S,, P.Q.k,S;, P;Q,R.S, (Fig. 1b) form another triplet of faces (Fig. 2b; it is seen that P,, Q,, &;, S, lie in the plane x, —4#,+4,—2=—0, etc.). Again, Q,Fh,S;Rs, Q.RS,h,, O3h3S,h, (Fig. 1c) form a triplet of faces (Fig. 2c; Q,, #,, S;, R, are in the plane #,—2,—1=0); they interseet the coordinate planes, i.e. the first three faces, along A,B,, A,B,, A,B,. All nine faces are shown in Fig. 2d; the polyhedron is now closed by a last face, the hexagon S,P,S,P,S,P, (in the plane v7,+a%,+7,—2=0, Fig. 1d) which actually coincides with the regular triangle P,P,P;, the S; being the midpoints of its sides. 1See Hilbert and Cohn-Vossen, Anschauliche Geometrie, Springer, Berlin, 1932, pp. 266-283. 2 Kinseitige Polyeder nach Boy. I. Eimseitiges Hexadekaeder, Comment. Math. Helw., 14, 1942, pp. 134-137. 3 Hinseitige Polyeder nach Boy. II. Polyédre sans singularités topologiques homéomorphe au plan projectif dans l’espace a 3 dimensions, ibid. pp. 137-140. 4 The original papers of Merz and Humbert were not available to the author; a review is found in Mathematical Reviews, 3, 1942, p. 299. 5 As Boy’s surface possesses an axis of order 3, the possible models will probably have 3k+1 faces ; the one-sided heptahedron (viz. loc. cit. 1) has six singularities ; the next possible case is the dekahedron. A POLYHEDRAL MODEL OF THE PROJECTIVE PLANE. 21 R 4 Se Fig. la. Big. -2¢... Fig. 2d. F, A. BEHREND. Fig. 3c. Fig. 30. A POLYHEDRAL MODEL OF THE PROJECTIVE PLANE. 23 That the polyhedron thus constructed is a model of the projective plane is easily seen. Its net is shown in Fig. 3a. It is topologically equivalent to the triangle 8’,S’,S’, of Fig. 3b, ‘“‘ diametral ’’ points of whose boundary have to be identified, and this triangle, in turn, is topologically equivalent to the complete projective plane (Fig. 3c). The mapping can easily be studied in detail. A paper model can readily be constructed from the net which has to be cut along the lines A,B,, B,A, in the quadrangle P,Q,R,S, and the corresponding lines obtained by cyclic permutation. The model may actually be folded from a single piece passing the edges S,#,S,, S3;R,S,, S,R,S; simultaneously through the cuts 4,B,A,, 4,B,43, A;B,A,, but it may be found more convenient to detach the triangle P, P,P, and to use four separate pieces. Department of Mathematics, The University of Melbourne. February, 1943. PRELIMINARY NOTES ON SOLUTION-CRACKING TREATMENT OF TORBANITE. By J. A. DULHUNTY, B.Sc. Manuscript received, April 16, 1943. Read, May 5, 1943. INTRODUCTION. The organic matter of torbanite, when subjected to the usual method of retorting, first undergoes a change to a semi-solid material, or intermediate- product, which decomposes at higher temperatures into gas, oil vapour and carbon (Cane, 1942). The liquid products of this decomposition constitute a crude oil, which is used as a cracking stock for conversion to petrol. The original organic matter of torbanite is insoluble in organic solvents ; but the intermediate- product, formed by heat, is soluble, and can be extracted with common solvents (Cane, 1942 ; Dulhunty, 1942a, 1942b ; Diakova and Stepantzeva, 1940). This property makes possible a solvent extraction process in which the torbanite is first preheated and then extracted with solvents ; but the extracted products, when separated from the solvent, are of the nature of a heavy, black oil, which must be subjected to further treatment—such as cracking or hydrogenation— to convert it to petrol. Preheating and extraction may be carried out in one operation by using a solvent with a critical temperature higher than that required to render the organic matter soluble. Torbanite crude oil, obtained either by retorting or solvent extraction, can be used as the solvent at the required temperature. It undergoes cracking, however, but the solvent oil is replaced by oils extracted from the torbanite, and the products of cracking represent the final products desirable in the treatment of torbanite. This method of treat- ment, in which preheating, extraction and conversion of heavy oil to light products are effected simultaneously, may be referred to as solution-cracking ; and experiments in its manipulation are described here. | SOLUTION-CRACKING TREATMENT. Experimental Conditions. The conditions involved in solution-cracking include three variable factors— temperature, pressure and time. The temperature must be sufficient to convert the organic matter to the soluble intermediate-product, and at the same time to crack the extracted products as well as the solvent oil. The purpose of pressure is to keep sufficient of the solvent oil in the liquid phase to effect the solution of the intermediate-product, and, also, to keep the heavier cracked products, such as oil boiling from 200° C. to 300° C., in a suitable phase condition to enable them to be cracked to petrol. The extraction time depends on the extraction temperature for any particular solvent oil, but must be sufficient to effect the complete, or almost complete, conversion of the organic matter to the soluble intermediate-product. The main control problem, in connection with the conditions of operation, is the balancing of the amount of cracking and the rate of extraction. If the cracking is excessive during the time necessary to complete the extraction, the volume of the oil stock will decrease with successive extractions, and must be Oe SOLUTION-CRACKING TREATMENT OF TORBANITE. 25> replaced from external sources. If the cracking is insufficient, the volume of the oil stock will increase, or stocks of heavy, cracked products will be built up. The extraction temperature determines both the amount of cracking and. the rate of extraction ; but evidence obtained from experimental work suggests that, with increasing temperature, the amount of cracking increases more: rapidly than the rate of extraction, giving a limiting temperature of operation above which the volume of the oil stock will decrease. This is due, no doubt,. to the fact that the lighter cracked products, which are more refractory, crack very slowly at the lower temperatures, below 400° C., and tend to maintain the volume of the oil stock, but above 400° C. they commence to crack at appreciable rates, and rapidly increase the total amount of cracking. Solution-cracking, in laboratory equipment, has been accomplished by two methods. (i) By heating the solvent oil and the torbanite for short periods of time under the pressure of the vapours in the system at the temperature: of the extraction, and removing the light products (petrol or light fuel oil, as. required) by distillation between each heating stage. This method would be equivalent to recycling the reaction mixture several times through a heating unit, and removing the light products by flashing, or pressure-distillation,, between each cycle. (ii) By heating for the full extraction time in one stage,. and constantly removing the vapours of light products by releasing excess gas. so aS to maintain a constant pressure on the system. This method would approximate to passing the oil and torbanite through a heating unit, and then allowing the mixture to remain in a reaction chamber, at a given temperature and pressure for the required length of time to complete the extraction. The influence of pressure, in relation to the extraction temperature, is particularly important in the second method, as it directly determines the removal of certain products from the reaction almost as soon as they are formed.. The fractions possessing a critical temperature (under the existing conditions) lower than the extraction temperature, will be removed at any pressure, but those with higher critical temperatures will be allowed to remain in the reaction zone until their constitution is such that their vapour-pressures, under the conditions of the multiple component system, become equal to the pressure which is maintained. The pressure required to allow the removal of certain light products, such as petrol, and keep the heavier fractions in the reaction zone, depends on the phase equilibrium existing in the reaction mixture, and this is a function of the concentration on the various components, aS well as. temperature. Consequently the actual pressure required at a given extraction temperature departs widely from calculated figures based on the critical temperatures of hydrocarbons of various boiling points. In the following experiments, involving the continuous removal of gas and light products, a pressure was maintained, which would give a pressure distillate: of the required specific gravity. The relation between temperature and the required pressure tends to vary during the extraction owing to the changing concentration of the components in the system. In an extraction carried out at a constant temperature of 380°C., the required pressure may vary from 180 to 225 lb. as the extraction proceeds, and the concentration of heavy products decreases owing to cracking. In extractions in which the temperature is. gradually increased from 350° C. to 400° C., during the extraction, the required pressure may vary from 125 lb. at the commencement of the extraction to 300 lb.. at its conclusion, owing to the fact that both temperature, and the proportion of light components, increase as the extraction proceeds. EXPERIMENTAL DATA AND RESULTS. Early experiments in solution-cracking were based on the principle of heating the torbanite and solvent oil for short periods of time, or recycling, until ‘26 J. A. DULHUNTY. the extraction was complete, and, between each cycle, removing low-boiling products by distillation under atmospheric pressure. The number of cycles required to complete the extraction depended on the temperature and time for each cycle. Relatively low temperatures were used for the first cycles of each extraction, as most of the heavy, liquid intermediate-product is then being cracked, and low temperatures effect this cracking more efficiently. Higher temperatures were used in the later stages. The pressure was allowed to increase ‘during each cycle as gaseous and light, liquid hydrocarbons were formed. It was found that large numbers of cycles of short duration, involving low operating pressures, tend to produce high yields of liquid products and small volumes of gas, due to a minimum amount of cracking. The liquid products, however, contain relatively large quantities of heavy fractions, and may be expected to possess low anti-knock properties. Smaller numbers of cycles of longer duration, involving higher pressure, give greater volumes of gas ; but the liquid products contain larger proportions of light fractions, and would be expected to possess higher anti-knock properties. The following experiments, Nos. 1 and 2, illustrate the method adopted in carrying out multiple-cycle solution-cracking. Later work was based on the continuous removal of gas and light products ‘during the solution-cracking process, thus effecting complete extraction in one operation, and eliminating the necessity for recycling. In this method, illustrated by experiments Nos. 3, 4, 5 and 6, the time required to complete each extraction is determined by the temperature of operation, and a pressure is maintained on the system sufficient to allow light products, of required boiling range, to be removed as vapour with the gaseous hydrocarbons. The amount of cracking which occurs during the extraction is governed by both temperature and pressure. Preliminary experiments show that low temperatures and pressures, requiring long extraction times, give small volumes of gas and large yields of relatively low-quality liquid products. Higher temperatures and pressures involving short extraction times produce greater volumes of gas and somewhat smaller yields of better-quality liquid products. EXPERIMENTAL. Multiple-cycle Solution-cracking. Experiment No. 1. The apparatus consisted of a 2 1., cylindrical, high-pressure vessel, approxi- mately 16-5 by 3-5 in. (see Dulhunty, 1942a), completely closed except for a fine tube leading to a needle valve and pressure gauge. The body of the vessel was enclosed in, and heated by, an electric muffle. Temperatures were recorded ‘by means of a thermocouple located in a sheath which projected into the cylinder. One pound of Coolaway Mountain torbanite (see Table 5), crushed to pass ‘a + In. screen, was mixed with 1 1. of torbanite-crude oil distillate (B.P. 230°-250° C.) and placed in the pressure vessel. The mixture was subjected to successive heating-cycles until the volatile content of the solid residue was reduced to approximately 15%. This required seven cycles under the ‘temperature-pressure-time conditions for each cycle, given in Table 1. TaABnE: i; ; Conditions in Experiment No. 1 on Multiple-cycle Solution-cracking. Temperature, ° C. ve 375 390 400 400 400 400 400 Time, min. Ph 48 2 Max. pressure, lb. sq. in... 126 140 154 150 190 200 230 SOLUTION-CRACKING TREATMENT OF TORBANITE. 27 After each cycle the vessel was cooled to 240° C. and kept at that temperature while gas and vapours of light products were released through a condenser. The gas was measured, and the condensate fractionated into petrol (below 180° C.), light oil (180—230° C.) and residue (above 230° C.), which was returned to the pressure vessel. When the extraction was complete, the solid residue was filtered from the solvent oil stock and washed with benzene to remove the last of the oil. The oil was recovered from the benzene by heating in a distilling flask at 200° C. and then added to the solvent oil stock. This was made up to 1 1. with the original, heavy distillate, and used as the solvent oil for the next extraction. Three solution-cracking treatments were carried out in this way, the average results being set out in Table 2. TABLE. 2, Average Results in Hxperiment No. 1 on Multiple-cycle Solution-cracking. ‘Cycle No. EA 1 2 3 4 5 6 il Total. Gael «44 aN 0-44 ceria 2°5 1-5 2:5 2d oro Woe Ld Petrol below BSG. tal... 7 Hah 10 16 24 30 38 135 ‘Light oil, 180°— | Bovwc.. tal,’ . : 75 90 30 50 43 74 89 | 451 Final oil stock He mt = He ¥ Ae oe ene OHO ale 7 Total net yield of ight products .. at oe x 6 .. 456 ml. or 224 galls per ton Yield of gas .. ; 957 cu. ft. per ton Percentage by wt. conversion of organic matter to light products nt (S47, Fraction of light products distilling below 180° C. oe ts at 80/5 Fraction of petrol products distilling below 100°C... te: See OG Sp. gr. of petrol (below 180°C.) .. Hs ae ee wid a 0-74 Sp. gr. of light oil (180°-230°C.) .. ae of * ae Be OES Sp. gr. of final oil stock .. os a ni ie <6 ee Ooo Experiment No. 2. The apparatus was similar to that used for Experiment No.1. One pound of Coolaway Mountain torbanite (see Table 5), crushed to pass a i in. screen, was mixed with 1 1. of torbanite-crude oil distillate (200°-220° C.) and subjected to three heating-cycles under the temperature-pressure-time conditions given in Table 3. TABLE 3. Conditions in Hxperiment No. 2 on Multiple-cycle Solution-cracking. Cycle A? sh A il 2 3 Temperature, ° C. a8 375 ; 390 400 Max. pressure, lb. sq. in. 546 476 406 Time, min. ae Me: 60 90 120 After each cycle, gas and spirit vapour were removed by distillation from the vessel under atmospheric pressure and at 200° C. The crude condensate was fractionated into petrol (below 180° C.) and residue, the latter being returned to the pressure vessel. After the completion of the third cycle (the volatile content of the residue being reduced to 14%), and the removal of light products, the 28 J. A. DULHUNTY. contents of the vessel were removed, and the solid residue separated from the final oil stock by filtration and washing with benzene as described in Experiment No. 1. The oil stock, which amounted to approximately 1 1., was mixed with 1 lb. of fresh torbanite, and the next solution-cracking treatment carried out as. before. Ten batches of 1 lb. each were treated in this way. The volume of the: solvent oil stock remained constant—no additional solvent oil being used. The average yields of gas and petrol, obtained from each cycle, and the totals. for each batch are summarised in Table 4. TABLE 4. Average Results in Huperiment No. 2 on Multiple-cycle Solution-cracking. | Cycle .. ee Ape a3 I | 2 5 Total. Gas, l. ae oe alt 115) 10 6 31 Petrol (below 180° C.), ml... 82 80 76 239 Final oil stock sie ac sie ive as we uae ae Lok Yield of gas mh Ee: vie ae 2,453 cu. ft. per ton. Total net gain in petrol | 120 galls. per ton _ Percentage by wt. conversion of or ganic ‘matter to petr ol 40 Sp. gr. of petrol 0-73 Sp. gr. of oil stock 0-87 Fraction of petrol distilling below 100° C. 35% SOLUTION-CRACKING BY CONTINUOUS REMOVAL OF PRODUCTS. Experiment No. 3. The apparatus consisted of the pressure vessel already described, and. used in Experiments Nos. 1 and 2. A condensing system was connected in. place of the gauge, on the high-pressure side of the needle valve. The condenser,. in direct communication with the cylinder of the pressure vessel, carried a pressure: gauge and separate valves for drawing off condensate and releasing excess gas 3. thus providing for the condensation of vapours under pressure. A series of five solution-cracking treatments was commenced with 1-5 Ib.. of selected Glen Davis torbanite (see Table 5), crushed to pass a + in. sereen,. and 900 ml. of heavy oil (boiling above 250°C.) previously extracted from. torbanite. | TABLE 5. Torbanite Samples used in Huperimental Work on Solution-cracking. Proximate Analyses. Gray King Experiment Origin of Ash. Assay. Number. Sample. Hyero. Mol: Fixed Galls. Moist. Cont.c Carbon. per ton. 1, 2, 5, 6 | Coolaway Mt. run of mine : 0-08 19-02 17:05 3:85 195 3 Glen Davis selected sample ‘ , 0-25 66-45 11-60 21-70 141 4 Glen Davis run of mine 1-50 48-5 14:4 35-6 102 SOLUTION-CRACKING TREATMENT OF TORBANITE. 29 The temperature was raised rapidly to 360° C., then, slowly, over a period of 4hr., to 400°C. This was sufficient to complete the extraction. The pressure, controlled by releasing gas from the condensing system, was gradually increased with the temperature from 125 lb. sq. in. at 360° C. to 220 Ib. sq. in. at 400° C. After the 4-hr. heating period, the vessel was cooled, the contents removed and the solid residue separated from the solvent oil stock by filtration and washing with benzene (see Experiment No. 1). The pressure distillate, obtained during the treatment, was fractionated into petrol (below 180°C.) and residue, the latter being returned to the solvent oil stock. Each of the four following treatments involved the use of 1-5 lb. of torbanite, and the solvent oil stock from the previous batch. The yields of gas and petrol and the conditions of treatment are summarised in Table 6. Experiments Nos. 4, 5 and 6. The apparatus, general technique and weights of torbanite employed in these experiments were the same as in Experiment No. 3. The torbanite, used in Experiment No. 4, was an average sample from the Glen Davis deposit (see Table 5), and the solvent stock (750 ml.) was smaller than in Experiment No. 3, as the torbanite, under treatment, was of somewhat lower quality. Experiment No. 5 was made on Coolaway Mountain torbanite (see Table 5), using 1,000 ml. of solvent oil stock, and the solution-cracking was effected under constant temperature-pressure conditions—400° C. and 250 Ib. sq. in., respectively. In Experiment No. 6, the sample of torbanite and volume of solvent oil stock were similar to Experiment No. 5; but the temperature and pressure were gradually increased from 100 lb. sq. in. at 350° C. to 200 lb. sq. in. at 400° C. over a period of 5 hr. ' The experimental conditions and yields of products for Experiments Nos. 4, 5 and 6 are summarised in Table 6. TABLE. 6. Conditions and Yields in Experiments Nos. 3, 4, 5 and 6 on Solution-cracking with Continuous : Removal of Products. Temp.-press.-time Theor. Sol. Weight Conditions. Total | Petrol. | Max. Exp. No. and Oil of Gas. Galls. | Yield Sample. Stock, Torb. Cu. ft. | per ton. | Petrol. ml. No: Lb. sq. | per ton. | Galls. te: Fir. in. | per ton 3. Selected Glen 900 1:5 Increase | 4 Increase 3,826 89 101 Davis. 360-400 125-220 4, Glen Davis run 750 1-5 Increase | 4 Increase 3,189 74. 84 of mine. 360-400 125-220 5. Coolaway run | 1,000 1:5 400 2-5 250 4,464 100 PAS) of mine. 6. Coolaway run 1,000 1-5 Increase | 5 Increase 3,900 115 125 of mine. 350-400 100-200 SOLUTION-CRACKING IN ONE CYCLE WITHOUT THE REMOVAL OF PRODUCTS. Experiment No. 7. This experiment was made with the object of obtaining data relating to the special conditions of cracking and pressure, which obtain when solution cracking is carried out in one cycle without the removal of products. 30 Jc AS DULAUNIS: The apparatus consisted of the high-pressure vessel, as used in Experiments: Nos. 1 and 2, completely closed except for the fine tube leading to a high-pressure ga, ge and the et needle valve. One pound of Coolaway Mountain torbanite, crushed to pass a + in. screen, was mixed with 1 1. of solvent oil stock (boiling above 250° C.), a heated in the closed vessel at a temperature of 390° C. for a period of 6 hr. No products were withdrawn during the treatment ; thus the pressure was allowed to build up as a function of the gas and light, liquid products. formed during the total period. The high pressures, which existed during the latter stages of the treatment, probably kept all hydrocarbons with critical. temperatures above 390° C. (under the existing conditions) in the liquid phase, and so allow extensive cracking of some of the ight products. It is also probable that some polymerisation of the gaseous hydrocarbons may have resulted from the high pressures. The rate of increase of pressure with time, and the yields. and nature of the products are summarised in Table 7. TABLE 7. Time-Pressure Relations and Products Obtained in Experiment No. 7 on Solution-cracking in One Oycle without Removal of Products. | Pressure Continued. Continued. Time at | Pressure, | Increment ij 390°C. | Ib. sq. in. | each } hr. hr: Nor sq. am!) | “Time! Press.4| Ine nts Pinaet Press. | Inc’nt. = 200 — D4 a > 15628 108 44 2,228 62 4 295 95 24 1,722 94 43 2,289 61 3 400 105 23 DOOM ele Laie | eal ae) 2,352 63 3 598 198 3 1,855 DO cealeneakoee 2,420 68 1 803 205 [tae aa oo OG 6l | 5% 2,489 69 }} 1,013 210 hea Ws MS ORG." e680 53 2,558 69 13 1,228 215 Dee 12, 038\0 [0-62 6 2,623 65 13 1,400 | 172 4 | 2,101 63 2 1,520 120 4+ | 2,166 65 | Final stock . Ke Bi af fe a .. 650>mi- Yield of petrol (below 180° C. ) ee a Ne a 3 230 mak Yield of light oil es 230° ©.) AGS =$ ee a -. 220 mol Yield of gas ace fs vs a ae Sa 80-5 1. Net yield of petrol es fe Es wes Las ds .. 105ml, 1.e. 53 galls— per ton Fraction of light products distilling below 180°C. .. a Beamer i) Fraction of petrol distilling below 100°C... N. ae +. A Ree Specific gravity of petrol.. : He dk ae ae se 0-71 Specific gravity of light oil Be aie te a oy re 0-81 Specific gravity of oil stock an au ss ies “h 0-86 SUMMARY OF EXPERIMENTAL RESULTS. The results of Experiment No. 1 illustrate the remarkably high yields of spirit and light oil obtainable, without respect to quality, by multiple-cycle solution-cracking under comparatively low temperature-pressure conditions. The yield of 224 galls. per ton, representing light products formed from the torbanite, is equivalent to a conversion of 84% by weight of the organic matter. The amount of cracking occurring under these conditions is small, giving yields. equivalent to 4 cu. ft. of gas per gall. of the total ight products formed, and about. 957 cu. ft. from the conversion of each ton of torbanite. The quantity of petrol (below 180° C.) represents only 30% of the light products boiling below 230° C.. -_— ~see eet SOLUTION-CRACKING TREATMENT OF TORBANITE. 31 The specific gravity of the petrol (0-74) is relatively high, only 25% distilling over below 100° C. Solution-cracking in three cycles (Experiment No. 2) results in additional eracking, which produces a petrol with a specific gravity of 0-73 (35% distilling below 100° C.) and gives a net yield of 120 galls. per ton, although the gas yield is low at about 20 cu. ft. per gallon. It is interesting to compare these results with those obtained in Experiment No. 7, in which the treatment was effected in one cycle without removal of products. This gave more extensive cracking, and produced a petrol with a specific gravity of 0-71 (48% distilling below 100° C.), representing a net yield of only 53 galls. per ton. The gas yield in this. case was 55 cu. ft. per gall. of petrol formed. The results of Experiments Nos. 3, 4, 5 and 6 show that good yields of petrol can be obtained by solution-cracking with continuous removal of products. The gas yields are somewhat higher than those obtained in solution-cracking by recycling, being in the vicinity of 40 cu. ft. per gall. of petrol formed. The petrol, obtained from experiments on the continuous removal of products during solution-cracking of Coolaway Mountain torbanite, possessed the following properties : Crude petrol : Specific gravity ae a. ee 0-725 Fraction distilling below 100° Oe sw ae Sn ODA Sulphur (Lamp Method) .. a Mf ae 0-28% Unsaturated hydrocarbons .. ts ad ee eee, Aromatic hydrocarbons ae Sh) Gale ely Petrol refined by light alkali and acid aa (a: 5%, of 80° H,SO,) : Sulphur an ve a Se 0:26% Unsaturated hydrocarbons ae a. a re lao OG Aromatic hydrocarbons ae Seo OO, Colour: approximately water- -white, colour stable. ‘‘ Sweet ”’ to plumbite. Octane number We ne Ane us Po as, The conversion of the organic matter in the torbanite to petrol by solution- cracking is effected without distilling the heavy oils or allowing them to pass appreciably to the vapour phase until they have been cracked to the required final product. The oils, still heavier than the final products at the conclusion of a batch-treatment (i.e. the final oil stock), are used as the solvent oil for the next batch, where they are cracked again. The physical properties of the oil stock depend on the total amount of cracking involved in the treatment, but under given temperature-pressure-time conditions, the boiling range, specific gravity and viscosity of the oil stock remain reasonably constant, after equilibrium has been reached, during a series of batch-treatments. The oil stock is not black and opaque, as may be expected of a liquid residue from cracking as well as extraction, but it is dark-red and somewhat translucent with a strong, light-green fluorescence. This appears to be due to the absence of true coking at the relatively low operating temperatures, and the deposition of any poly- merisation products with the carbon formed by cracking. The green fluorescence would appear to be indicative of the presence of considerable quantities of paraffin hydrocarbons. The particle size of the torbanite (up to 1 in. cubes and possibly larger) does not influence the rate or efficiency of extraction (Dulhunty, 1942b). The solid residues, from the foregoing solution-cracking experiments, were fine, black powders together with some hard carbon deposited on the walls of the reaction vessel. These residues consisted of the inorganic material and small amounts of 32 J. A. DULHUNTY. vascular plant debris (humosite) originally present in the torbanite, together wit carbon formed by cracking. The proximate CO ee of the fresh torbanite and residue, after treatment in Experiment No. 2, are as follows : Fresh torbanite : Volatiles, 79-10%; fixed se 17-05°% sashes es Residue: Volatiles, 14% ; fixed carbon, 70%; ash, 16%. The pressure vessel used for experimental work on solvent extraction and solution-cracking treatment of torbanite was constructed entirely of mild steel. No evidence of corrosion was found on the inner surfaces on the cylinder and needle valve after the vessel had been used for a total running time of approxi- mately 600 hr. Hard carbon was removed from the walls and base of the cylinder ; and the surface of the steel, although dull, still carried the fine striz from the machining of the vessel when it was made. ‘There was no pitting, and the dull surface was made bright by light rubbing with fine emery paper. The thermocouple sheath, projecting into the cylinder to a central position, carried a light coat of soft carbon, which was removed by scraping with a piece of wood. This exposed bright steel showing the fine striz from the original turning of the sheath. The parts of the needle valve were examined, but no evidence of corrosion was found. Thus it was concluded that no appreciable corrosion of mild steel occurs during solution-cracking at temperatures in the vicinity of 400° C. On one occasion a copper strip, 4 in. in thickness, was placed in the vessel. After 16 hr. running, at temperatures between 350° and 400° C., much of the copper was changed to copper sulphide, and holes appeared in the strip, indicating vulnerability of copper to the action of sulphur ee during solution-cracking. a < 3 =; = ee eS a ACKNOWLEDGMENTS. The investigations outlined in this paper were made possible by a Common- wealth Research Fellowship, and the purchase of equipment by assistance from the Commonwealth Research Grant to the University of Sydney. The writer wishes to acknowledge valuable discussion with L. J. Rogers, M.Sce., B.E., members of the staff of National Oil Pty. Ltd. and the Department of Engineering Technology, University of Sydney ; also the technical assistance of A. J. Tow, M.Sc., University of Sydney. REFERENCES. ‘Cane, R. F., 1942. Proc. Roy. Soc. N.S.W., 76, 190. Diakova, M. K., and Stepantzeva, T. G., 1940. Comptes Rendus (Doklady) de V Academic des Sciences de 1 U.R.S.S., 26, No. 4. Dulhunty, J. A., 1942a. Proc. Linn. Soc. N.S.W., 67, 239. Dulhunty, J. A., 19426. Proc. Roy. Soc. N.S.W., 76, 268. ISSUED MARCH 31, 1944 fe vOL.LXxvll PART II ~ JOURNAL AND PROCEEDINGS OF THE | ROYAL SOCIETY FOR 1943 (INCORPORATED 1881) PART IT (pp. 33 to 84) OF VOL. LXXVII Containing Papers read in June and Clarke Memorial Lecture with Plates I-III EDITED BY THE HONORARY SECRETARIES THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN Z : SYDNEY PUBLISHED BY THE SOCIETY, SCIENCE HOUSE GLOUCESTER AND ESSEX STREETS 1944 Ki CONTENTS “i es VOLUME LXXVII rere ee Pare es a, Art. V.—Tabulata and Heliolitida from the Wellington District, N.S. w. By oO. AS M.Sc. (Communicated by Dr. Ida A. Brown.) - (Issued Pebranry 9, »~ Art. VI.—The Etch Figures of Basal Sections of Quartz. of Water-worn Crystals. By F. N. Hanlon, B.Sc., Dip.Ed. Arr. VII.—Clarke Meainwal Lecture. Australia’s Mineral Industry i in the Pi So BY oH, G. Raggatt, D.Sc. fey February 9, 1944).. rae . TABULATA AND HELIOLITIDA FROM THE WELLINGTON DISTRICT, N.S.W. By O. A. JONES, M.Sc., University of Queensland. (Communicated by Dr. Ipa A. Brown.) With Plate I. Manuscript received, May 10, 1943, Read, June 2, 1943. SUMMARY. The paper describes some corals collected in 1942 by Miss Basnett and Miss Colditz of Sydney University. Five species of Favosites, one of Heliolites and one of Propora are described, the genus Plewrodictyum is discussed and a new species described. The fossils are from six different localities, and while the collections are too small in some cases to give positive evidence of the age, some indication ean be given. The corals obtained from various localities are as follow: (A) Por. 82, PAR. MickmeTy MULGA. Favosites goldfussi d’Orbigny, F. basalticus (Goldfuss) var. moonbiensis Etheridge, Plewrodictyum bifidum sp. nov. The first of these ranges from the Lower to Middle Devonian, the second Middle Devonian, and is especially characteristic of the Nemingha and Moore Creek Limestones. Age: Middle _ Devonian. ~ (B) Por. 206, PAR. MIckETY MULGA. Pleurodictyum bifidum sp. nov. Age: Probably Middle Devonian. (C) DUBBO ROAD, 12 MILES FROM WELLINGTON. Favosites bryant Jones. This form is very characteristic of the lower Middle Devonian Murrumbidgee beds and is also found in the Lower Devonian Garra beds. (D) Por. 241, PAR. MICKETY MULGA, WELLINGTON, N.S.W. Propora sp. No deduction as to age. (E) Por. 50, PAR. CURRA. Favosties bryant. Age: Lower Middle Devonian or Lower Devonian. fh) Por: 119, Par. VEECH. Favosites sp. sp. nov.? No deduction as to age can be drawn. (G) WILLOWTREE CREEK, ATTUNGA. : Favosties ? goldfusst @Orbigny. Age: ? Devonian. (H) ONE QUARTER MILE N.E. OF APSLEY R.S. on the road to the dredge, Wellington. Favosiies richardsi Jones, Heliolites daintreet Nich. and Eth. (fourth group). The former is unknown outside the Upper Silurian and in fact this is the first C—June 2, 1943, 34 0. A. JONES. record outside the type area—Yass, N.S.W. The latter is known from the Upper Silurian of Yass, the Lower Devonian of Molong, N.S.W., and the Devonian of the Broken River, N. Queensland. The age may therefore be taken as Upper Silurian. SYSTEMATIC DESCRIPTIONS. MADREPORARIA TABULATA. Genus Favosites Lamarck. Favosites goldfussi d’Orbigny. Favosites goldfussi (partim) d’Orbigny, 1850, p. 107 (fig. 3b of Goldfuss, 1829 only). For synonymy, description and figures, see Jones, 1936 and 1937. Remarks : Favosites goldfussi as a member of the F’. gothlandicus-F. goldfussi eroup was discussed by Hill and Jones, 1940, pp. 191-3, where the similarity between F. gothlandicus forma multipora and F’. goldfussi was pointed out. The specimen from Portion 82, Parish Mickety Mulga lies between F'. gothlandica forma multipora and F'. gothlandica forma forbesi, but nearer to the former, and might be called either F'. goldfussi or F. gothlandica forma multipora. The adult corallites are 1-75 to 2-5 mm. in diameter, slightly smaller than usual ; the walls are thick and the septal spines long and numerous, though somewhat obscured by recrystallisation; the tabule are complete, usually horizontal, 3 or 4 in a space of 3 mm.; the mural pores are in at least three rows. A specimen from Willowtree Creek, Attunga, may be F. goldfussi. It is however excessively recrystalline, so that identification is very doubtful. Septa are numerous but apparently fairly short. The mural pores may be in two rows and there are 6 or 7 complete tabule in a space of 3 mm. Rhange: Lower and Middle Devonian (Garra beds, and Nemingha and Moore Creek limestone). Localities : Portion 82, Parish Mickety Mulga, Wellington District, N.S.W., and ? Willowtree Creek, Attunga, Tamworth District, N.S.W. (Univ. of Sydney Nos. 6252 and 5246.) Favosites basalticus (Goldfuss) var. moonbiensis Etheridge. Favosites basaltica (Goldfuss) var. moonbiensis Etheridge, 1899, pp. 164-5, pl. xxivatiiess 12: For synonymy and description, see Jones, 1937, p. 96. Remarks : The one specimen is poorly preserved but is typical of the variety moonbiensis in every way except that the spacing of the tabule is rather more variable than in the specimens from Tamworth—2 to 4in 1 mm. The mural pores are about 0-25 mm. in diameter in one row in the centre of the faces. Range: Middle Devonian. Locality: Portion 82, Parish Mickety Mulga, N.S.W. (University of Sydney, 6250.) | Favosites bryani Jones. Plate I, figures 1, 2. Favosites bryant Jones, 1937, pp. 96-7, pl. xv, figs. 3-6. Favosites bryant Hill and Jones, 1940, pp. 190-1, pl. v, fig. 2. Remarks: A specimen from portion 206, Parish Mickety Mulga and one from portion 50, Parish Curra, Wellington district, belong to this species. The TABULATA AND HELIOLITIDA FROM THE WELLINGTON DISTRICT. 30 corallites are 1 to 1:25 mm. in diameter, the walls moderately thick, and the corallite angles rounded. Both are recrystalline but the septal spines appear to be of the form typical of this species—long, slender, sharply pointed, upwardly directed spines. The arrangement of the mural pores is not apparent. The tabule are variable in number, 7 to 11 in a space of 3 mm. Range: Lower Devonian to lower Middle Devonian. Localities: Dubbo Road, 12 miles from Wellington, and Por. 50, Par. Curra, both in the Wellington district, N.S.W. (University of Sydney 6254 and 6253.) Favosites sp. Sp. Ov. ? Plate I, figures 3, 4. Remarks: A specimen from portion 119, Parish Veech (University of Sydney No. 5290) is probably a new species, but being poorly preserved and a single specimen, I refrain from creating a new name for it. The corallum is massive, the corallites regularly 1-5 mm. in diameter, the walls slightly dilated but the angles little rounded. No septa show in the transverse section but it is much recrystallised and there are indications in the longitudinal section that septa may be present. The mural pores are round, large—0:3 mm. in diameter—in a single row in the faces of the corallites. The tabule are complete, thin, regularly spaced, 7 or 8 in a space of 3 mm. Locality: Portion 119, Parish Veech, Wellington, N.S.W. (University of ‘Sydney 5290.) Favosites richardsi Jones. Plate I, figures 5, 6. Favosites richardsi Jones, 1937, pp. 89-90, pl. xu, figs. 2, 3. Remarks: The specimen (University of Sydney No. 7278) is completely typical of this species. The corallites are as usual of two orders of size, but this has been shown to be an environmental condition and not of specific value in F’. forbesa (Jones, 1936) and unpublished work on F. richardsi confirms this. The larger corallites have 8 to 11 sides, thus becoming nearly round, and are 3-5 to 4 mm. in diameter. The smaller corallites have 4 to 6 sides and are 2 to 2-5 mm. in diameter, but there are also many young, smaller corallites, triangular or four-sided. The corallite walls are thin or very slightly dilated. The septal spines are numerous, short, with a broad base but sharply pointed. The tabule are complete, thin, 3 to 6 in 3 mm. The mural pores are in two rows on the faces of the smaller corallites but the arrangement has not been observed on the larger faces of this specimen. Range: Upper Silurian. Locality : A quarter of a mile north-east of Apsley R.S. on the road to the dredge, Wellington, N.S.W. (University of Sydney 7278.) Genus Pleurodictyum Goldfuss. Pleurodictyum Goldfuss, 1829, p. 113. Genoholotype: P. problematicum, ibid., p. 113, pl. xxxviii, figs. 18 a-g. Lower Devonian, Eifel district and Nassau, Germany. ? Michelinia de Koninck, 1842, p. 29. Genolectotype (see Edwards and Haime, 1850, p. Ix) Calamopora tenuisepta Phillips, 1836, p. 201, pl. ii, fig. 30. Lower Carboniferous, Holland and the Mendips. Diagnosis : Cerioid Favositide, walls dilated, septa spinose, the spines sometimes arising from the free axial edges of very short lamelle, sometimes 36 oO. A. JONES. directly from the walls; tabule present, sometimes spinose; mural pores numerous. Topotypes usually (? always) have a worm ease in the base. Remarks: The above diagnosis is based on four topotypes of the genotype (from Oberstatfeld, near Gerolstein, Eifel) which are all internal moulds, as is the case with all topotypes known. The deduction of the complete structure of the coral from these internal moulds is difficult but there is no doubt concerning the polygonal shape of the corallites, the thick walls and the presence of numerous small mural pores, which in P. problematicum are usually in two, sometimes three, rows. The septal spines show as numerous round pits which sometimes are situated in a single row in a longitudinal groove, representing a short lamella ; many of the spines, however, arose directly from the walls, there being no trace of any groove; and the character varies from corallite to corallite and from specimen to specimen, some corallites having several grooves, most having only one groove in the centre, other pits to either side not being in grooves, one Specimen I examined having as far as I could see only one groove on one corallite, none on the remainder. In my opinion there were almost certainly tabule for, firstly, on many of the moulds of the corallites there are transverse striations which can only represent ridges left when tabule were broken away ; it may be thought remark- able that all the tabula were broken away before the corallites were filled and the other tissues dissolved, but the tabula were thin delicate structures while the walls were thick and the spines short ; perhaps the tabule were thinner and more delicate than usual in the Favositide ; further, the corallites were short and the corallum low and spreading, almost discoidal ; secondly, in the centre of each mould are a few corallites perpendicular to the bedding planes of the rock and in several instances the surface of these is covered with numerous small shallow pits which represent more or less vertical spines. The only structures on which these spines could have been based were tabule. Roemer (1883, p. 425), Hall (1876), C. L. and M. A. Fenton (1936, p. 23), Lang, Smith and Thomas (1940, pp. 84 and 102) and others consider that Micheinia de Konineck is a synonym of Pleurodictyum, although Roemer con- sidered that tabule were absent in the latter. This can only be finally decided by an examination of topotypes of M. tenwisepta (Phillips) de Koninck, the genotype of Michelinia. In the meantime, basing this opinion on published descriptions and figures, I agree that most species of Michelinia should be placed in Pleurodictyum. Nicholson (1879, p. 149) describes ‘‘ intramural canals ”’ in P. stylophorum (Eaton) but not in P. problematicum. I have not observed them in the latter species, nor so far as I know has any other writer. C. L. and M. A. Fenton do not mention them in P. stylophorum. Nicholson considers them to be of the same nature as similar structures which he described in Columnopora. Cox, 1936, refers the latter to Calapoecia Billings, and considers the ‘‘ intramural. canals ’’ to be the result of some boring organism. Pleurodictyum is not clearly distinct from the thick walled Favosites. Three features may be considered in this connection—lamellar septa, spinose tabule and strong holotheca. If the first be taken as diagnostic a number of forms, including that to be described below, without lamellar septa but with the other two features, must be removed from Plewrodictywm ; but the presence of spines on the tabule, a character which varies much from species to species, and the presence of a stronger holotheca than is usual in Favosites are not enough on their own to justify separation of the two genera. The whole group of forms stands in need of revision. TABULATA AND HELIOLITIDA FROM THE WELLINGTON DISTRICT. 37 Pleurodictyum bifidum sp. Nov. Plate I, figures 7, 8. Holotype: The specimen 6251 in the collection of the University of Sydney from Por. 82, Par. Mickety Mulga, Wellington, N.S.W. Age: Middle Devonian. Diagnosis: Pleurodictyum with numerous spinose septa, some of which are bifid, numerous irregular complete and incomplete tabule, mural pores large and rare. Description: The corallites are polygonal but the angles are rounded by the dilatation of the walls which makes them as much as 1-25 mm. in thickness. The diameter of the corallites is 4 to 6 mm. The septa are spinose and very numerous, arranged in longitudinal rows which number seven or more on each corallite face. The spines are stout but usually sharply pointed and about 5 per cent. divide near their axial ends into two (? sometimes three) branches, an unusual and important character, upon which I have based the trivial name. This appears to be a type of rhabdacanthine septa (Hill, 1936) but on a larger scale than any yet described, and unfortunately the coral is not sufficiently well preserved to observe the trabecule and confirm this suggestion. The tabule are thin, numerous, 10-13 in 5 mm., complete and incomplete in about equal numbers, horizontal or oblique, sometimes deeply invaginated. The mural pores are large, 0-25 mm. in diameter; their arrangement has not been definitely observed, but there is some evidence to suggest two rows, each row fairly close to the corallite angle. Remarks: I know of no species with which this is closely comparable, the very large number of septa and their frequent bifid nature are very striking characters. Localities: Por. 82 and Por. 206, Par. Mickety Mulga, Wellington, N.S.W. (6251 and 5287 University of Sydney collection). Middle Devonian. MADREPORARIA HELIOLITIDA Jones and Hill. Family Heliolitide. Genus Heliolites Dana. Heliolites daintreei Nicholson and Etheridge. Plate I, figures 9, 10. Heliolites daintreet Nicholson and Etheridge, 1879, p. 224, pl. xiv, figs. 3, 3a. Heliolites daintreet Jones and Hill, 1940, pp. 199-203, pl. vi, figs. 1-5; pl. vii, “isis. 1-5 ; pl. viii, figs.4-8 ; pl. ix, fic. 1. For synonymy, diagnosis, etc., see Jones and Hill, 1940. kemarks: This is a variable, long-ranged species, divided by Jones and Hill into four, ill-defined groups. The specimen under consideration falls into group four. The tabularia are 1-75 to 2-5 mm. in diameter, rather larger than usual, with none to six rows of tubuli, and 0 to 4 mm. between the tabularia. The tabularia are in contact in only one place in the section and usually there are 2-3 rows of tubuli between. The walls of the tabularia, the tabule and sola are typical of the group. The septa are largely obscured by recrystallisation but show in places in the transverse section, arising from the wall between the slight angles formed where two tubuli meet the wall. The walls are sometimes crenulate, and then the septa arise from the crenulations. That the septa are long spines upturned axially is shown in transverse section by their abrupt truncation and the occasional occurrence of apparently detached fragments towards the centre of the tabularia. 38 O. A. JONES. ange: Group four ranges in Australia from Upper Silurian to Middle Devonian. Locality: One quarter of a mile N.E. of Apsley on the road to dredge, Wellington, N.S.W. (University of Sydney 7272.) Genus Propora Edwards and Haime. Propora §p. Plate I, figure 11. Remarks: A single specimen is an undescribed species of Propora. The state of preservation is poor, but the transverse section shows the tabularia to be thick walled and crenulate with long septa, stout at the base but rapidly becoming thin, arising from the crenulations. The septa reach or nearly reach the centres of the tabularia and are prolonged in the other direction outside the tabularia into the reticulum. The longitudinal section is obscure so that the character of the septa cannot be determined. ‘The reticulum consists of test#, but beyond this the characters cannot be seen. Of described species it may be close to Propora tubulata Lonsdale but further comparison must await better specimens. It is quite unlike P. conferta Ed. & H., the only other species recorded from Australia (see Jones and Hill, 1940, p. 209). Locality: Por. 241, Par. Mickety Mulga, Wellington, N.S.W. (University of Sydney 7279.) REFERENCES. Cox, I., 1936. Revision of the Genus Calapoecia Billings. Nat. Mus. Canada, Bull. No. 80. Etheridge, R., 1899. On the Corals of the Tamworth District. Rec. Geol. Surv. N.S.W., 6, pt. 3, pp. 151-182, pls. XVI-XXXVITI. Fenton, C. L., and Fenton, M. A., 1936. The ‘‘ Tabulate’’ Corals of Hall’s ‘‘ Illustrations of Devonian Fossils’. Ann. Carnegie Museum, 14, pp. 17-58, pls. I-VIII. Goldfuss, G. A., 1829. Petrefacta Germanie, 1, pp. 77-164, pls. XXVI-L. Diuisseldorf. Hall, J., 1876 [? 1877]. Illustrations of Devonian Fossils ... Geol. Surv. State of New York, Paleont. Hill, D., and Jones, O. A., 1940. The Corals of the Garra Beds, Molong District, N.S.W. Proc. Roy. Soc. N.S.W., 74, pp. 175-208, pls. II-VI. Jones, O. A., 1936. The Controlling Effect of Environment upon the Corallum in Favosites ; with a Revision of Some Massive Species on this Basis. Ann. Mag. Nat. Hist., Ser. 10, 17, pp. 1-24, pls. I-III. — 1937. The Australian Massive Species of the Coral Genus Favosites. Rec. Aust. Mus., 17, pp. 1-24, pls. I-III. Jones, O. A., and Hill, D., 1940. The Heliolitide of Australia, with a discussion of the Morphology and Systematic Position of the Family. Proc. Roy. Soc. Q., 51, No. 12, pp. 183-215, pls. VI-XI. Koninck, L. G. de, 1842. Description des Animaux fossiles ... Leige, 1841-1844, 650 pp., pls. A-H and I-LV. Lang, W. D., Smith, S., and Thomas, H. D., 1940. Index of Paleozoic Coral Genera. Pp. 1-231. British Museum (Natural History). Nicholson, H. A., 1879. On the Structure and Affinities of the Tabulate Corals of the Palzozoic Period. Pp. x+342, pls. I-XV. Blackwood & Sons, London. Nicholson, H. A., and Etheridge, R., 1879. Description of Palzozoic Corals from Northern Queensland. Ann. Mag. Nat. Hist., (5), 4, pp. 216-226, 265-285, pl. XIV. d’Orbigny, A., 1850. Prodrome de Paleontologie Stratigraphique ... 1, pp. 294-+ix, 8vo, Paris. Phillips, J., 1836. Illustrations of the Geology of Yorkshire, Part II. Pp. xx+253, pls. I-XXV, London. Roemer, C. F., 1883. Lethea geognostica. I. Theil. Lethza paleozoica, 1, (2), pp. 113-544, Stuttgart. Journal Royal Society of N.S.W., Vol. LXXVII, 1943, Plate I TABULATA AND HELIOLITIDA FROM THE WELLINGTON DISTRICT. 39 EXPLANATION OF PLATE. All figures x 2 approximately. Figs. 1, 2.—Favosites bryant Jones. 1, transverse; 2, longitudinal section. Dubbo Road, 12 miles from Wellington. (S.U. 6254.) Figs. 3, 4.—Favosites sp. nov.? 3, transverse; 4, longitudinal section. Por. 119, Par. Veech, Wellington. (S.U. 5290.) Figs. 5, 6.—Favosites richardsi Jones. 5, transverse; 6, longitudinal section. One-quarter mile N.E. Apsley R.S. on road to dredge, Wellington. (S.U. 7278.) Figs. 7, 8.—Pleurodictyum bifidum sp. nov. Holotype. 7, transverse; 8, longitudinal section. Por. 82, Par. Mickety Mulga, Wellington. (S.U. 6251.) Figs. 9, 10.—Heliolites daintreii Nich. & Eth. 9, transverse; 10, longitudinal section. One- quarter mile N.E. Apsley R.S. on road to dredge, Wellington. (S.U. 7272.) Fig. 11.—Propora sp., transverse section. Por. 241, Par. Mickety Mulga, Wellington. (8.U. 7279.) THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. THEIR USE IN THE ORIENTATION OF WATERWORN CRYSTALS. By F. N. HANLON, B.Sc., Dip. Ed., Geological Survey of New South Wales. With Plates II-III and nine text-figures. (Presented by permission of the Under-Secretary for Mines.) Manuscript received, May 11, 1943. Read, June 2, 1943. INTRODUCTION. This paper sets out the results obtained by etching sections of quartz cut perpendicular to the optic axis. It also examines the possibility of orientating sections, cut in this plane, in which the positions of the prism faces are unknown. The main use of quartz is for the manufacture of quartz plate resonators. For this purpose the plates have to be cut in a definite relationship to the crystal axis, in order to obtain the optimum temperature-frequency coefficient and eliminate secondary resonances. The margin of error permitted in this operation is very small and in order that the cutting can be carried out with the required degree of accuracy, the positions occupied by the prism faces must be known fairly accurately. A certain quantity of waterworn quartz crystals of high quality is available, which cannot be used for piezo-electric purposes unless the positions of the prism faces are known reasonably accurately. The position of the optic axis can be obtained with an accuracy of +10 minutes by means of a polariscope (Booth and Sayers, 1939). The position of a plane perpendicular to all prism faces thus being known, it still remains, before the orientation is completely determined, to fix the positions these faces occupy in this plane. The order of accuracy required for this operation is not so great as that for the determination of the position of the optic axis. The work, the results of which are set out in this paper, was originally undertaken in order to see whether a way of solving this problem could be found, without recourse to X-ray methods. A summary of some of the properties of quartz has been included in order to make the results detailed easier to follow. Properties of Quartz. Quartz crystallises in the hexagonal system and occurs as two varieties, a-quartz and 6-quartz (Bragg and Gibbs, 1925; Gibbs, 1926). «-quartz, which forms at temperatures below 575° C., has only trigonal symmetry and possesses the property of piezo-electricity. Quartz is enantiomorphous, the form assumed depending on whether the spiral atomic structure is right- or left-handed. This variation in internal structure is reflected in a variation of the outward form by the disposition of the subsidiary # and s faces, as shown in Figure 1. The Bulletin of the Imperial Institute (1938) has given the following notation for piezo-electric quartz decided upon by the Department of Scientific and Industrial Research: The principal crystallographic axis, which is an axis of threefold THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. 41 symmetry and coincides with the optic axis, is referred to as the Z axis. At right angles to this axis there are the three crystallographic axes, which pass through the intersections of the prism faces. These axes of diagonal symmetry are known as the X axes. A further set of three Y axes, in the same plane but Left-handed Right-handed Quartz Crystals FIGURE | perpendicular to the X axes, is assumed for piezo-electric purposes. Under compression or tension along the X axes, the opposite ends take up opposite charges, the signs of these charges depending on whether the crystal is right- or left-handed. The positions of the X and Y axes are shown in Figure 2. The Z axis would be perpendicular to the plane of the paper through O. X at, X FIGURE 2 Quartz crystals are almost invariably twinned. The only types of twinned crystals considered in this paper are those which consist of complete inter- penetration twins. Twinning of this form can take place in three different ways. In all cases the Z axes of the two parts of the twin crystal are coincident. In D—dJune 2, 1943. 42 F. N. HANLON. type (a) a right- or left-handed crystal may be concerned. It is known as electrical twinning to distinguish it from types (b) and (c), which are known as optical twinning, because they can be recognised by. optical means. In types (b) and (c) both a right- and left-handed crystal are concerned. (a) One portion of the crystal may be considered as being rotated through 180° around the Z axis, with respect to the other portion. This can be detected by the appearance of x or s faces upon adjacent corners. The & of one individual coincides with the z of the other, so that the pyramid faces may consist of some parts which are bright and others dull. (6) This oecurs when a right-handed crystal is united with a left-handed crystal, so that the # faces of the two individuals coincide. It can be recognised by the appearance of both right- and left-handed x or s faces at the base of the same pyramid face. From the frequency of occurrence of this type of twin crystal in specimens from Brazil, it is often spoken of as a Brazil twin. (c) This occurs when a right- and left-handed crystal are so united that one individual is rotated through 180° around the Z axis with respect to the position taken up in (b) above, thus bringing the F face of one individual into coincidence with 2 face of the other. This type can be recognised by the appearance of the two types of x or s faces on the same corner. In the same crystal a combination of two forms can and very often does take place. It is only a very small proportion of the crystals found, in which optical and/or electrical twinning do not occur. Either type ruins the crystal for piezo-electric purposes as the X axes of the two portions are developed in opposite senses. Where a crystal consists of one major individual with only small portions which are twinned, the untwinned portion can be used and the remainder rejected. Etching of Basal Sections. The sections cut perpendicular to the optic axis were etched with hydro- fluoric acid. Total periods of etching varied from 10 minutes to 16 hours, with a reading taken on one crystal after 72 hours. Single specimens were examined after etching for periods of 10 minutes at a time, over several hours. Prior to etching the surfaces were prepared by being ground flat with carborundum powder. Three grades of powder were used, 4F, 3F and 220. Care had to be taken to remove all saw-cut marks, as these gave parallel bands, which influenced the reflections to some extent. The effects of the different surfaces on the results obtained are discussed below. The sections were placed on the rotating stage of a microscope, obliquely illuminated by a source of light and the reflected light examined through the microscope. The same effect could have been obtained by keeping the sections in a fixed position and rotating the beam of light. The light source used gave a reasonably concentrated beam, although it was not absolutely parallel. This source was set up about three feet from the microscope and directed through a slit in a piece of cardboard placed close to the microscope, the slit being just sufficiently large for the field of view to have full overall illumination. The light was so directed that when the beam was parallel to a prism face of the section, the scale reading of the stage of the microscope was zero. The angle of incidence for most of the readings was about 70° measured from the perpendicular. The effect of varying this angle is discussed below. The etch markings (see Plate II) commence as triangular pits and with deeper etching become triangular hillocks. Booth. and Sayers (1939) have postulated that the etch faces are curved. This is confirmed by the results given in this paper. The incident light is reflected from the faces which form the sides of these pits or hillocks, so that during a complete rotation of the stage there are THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. 43 three positions of maximum illumination, between which there is relative darkness. These three positions are not at exactly 120° with one another, nor do they bear a constant relationship to the X (or Y) axes as may have been expected. In the results that follow each position of maximum illumination has been measured as an angle, positive or negative, from the nearest 120° position, and the average of the three readings thus obtained taken as the angle in the horizontal plane at which maximum reflection is obtained for that particular section for the time of etching and strength of acid used. For instance, if maximum reflections were obtained at 8°, 131° and 251°, these positions would read as 8°, 11° and 11° respectively, the average figure or reading taken being 10°. Using this method of recording results, a zero (or +60°) reading would mean that maximum reflection occurred when the beam of light was parallel to a prism face, a +30° (or +90°) reading would correspond to the beam of light being perpendicular to a prism face, and so on. The sign of the reading so obtained would depend on the direction of rotation of the stage, being positive for anti-clockwise rotation and negative for clockwise rotation with the microscope used. It should be noted that a reading of —10 is equivalent to +110, —20 to +100, ete. If the light is parallel to one prism face and the section is turned around in a clockwise direction until the adjacent prism face is parallel to the light, still keeping the scale reading zero, a rotation of the stage until the scale reading is 60° will bring the section into its original position. It can be seen therefore, that a reading of 60+0 would be equivalent to a reading 0 depending on which prism face was originally parallel to the light, when the scale reading was zero. The angle at which maximum illumination occurs varies in a very com- plicated manner, depending on the length of time of etching and strength of acid used. Sections were taken and etched for periods as short as ten minutes at a time and the angles obtained plotted on a curve, with period of etching as the abscissa and angle at which maximum illumination occurred as the ordinate. Although the way in which the angle changes is complicated, the change seems to follow a definite order and is not haphazard. It is regretted that with the time available it was not possible to read several crystals through a complete cycle. A composite curve has been constructed by compounding the results from the etching of several sections which have been read at short intervals over different parts of this curve. The longest total period of etching during which readings were taken continuously was 16 hours. In Figure 3, several curves obtained are shown, together with the composite curve constructed from them. Cor- responding parts of the curves are similarly marked by the letters A, B, C, etc. The continuous parts of the curves are where the readings have been taken every ten or fifteen minutes and the broken curves are portions interpolated from analogy with other curves. Small circles on these broken sections show actual readings taken. The construction of a composite curve as being representative of any crystal has been rendered difficult by the fact that the individual curves are not directly comparable, and this even applies to different parts of the same curve, owing to the strength of the acid used varying. Large variations in the strength of the acid used were brought about mainly by lack of knowledge, in the early stages of the work, as to the period for which the acid would maintain a reasonably high rate of attack and secondly by one bottle of acid, although newly opened for the work, having apparently lost most of its strength during Storage. The fact that the angle was not changing could not be used as an indication that fresh acid was required, because there are apparently parts of the curve for which the rate of change is only small, irrespective of the strength of the acid used. Although differences in strength of acid may change the apparent Shape of the curve, it is considered that it would not affect the values of the maxima, minima and points of inflection. It should be pointed out at this DD—June 2, 1943. F. N. HANLON. 9608 H i 360 920 320 280 240 3 200 Bait PS ee Be, ee ‘VRE Ey: ae ea tek Ly deh ee ppd 480 FIGURE 440 160 320 360 handé 120 280 t tes. Zz =] g = e ; ee at eo ° 2 Cc © H > bcs is Re iw 5 UJ Ke) Zz Hs z i=2) 2 c o ae (aa Ripe % ue) tee D Pv | ie te a os Fagiallf iieadih | Mi Cie) g-}-Y s¢ : at i) (@) c® rc 7) £9 E we = 3 = > c) < FIGURE 4 ee THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. 45 stage that the composite curve shown is only intended to represent the most likely curve on which the positions of maximum illumination would fall. The variation is too complicated to be completely delineated from the amount of work done. It is hoped that, at a later date, further work may enable the details of the curve to be shown more exactly. It will be noted that the curves for different crystals do not commence at the same figure, but may begin at practically any point on the composite curve. Since the point of commencement is not fixed, the same point on the curve would correspond to different depths of etching for different crystals. This would probably have an effect on the rate of change of the angle of maximum reflection and also to a few degrees on the actual values of minor positions of maxima or minima and points of inflection. It is not considered that it would affect the values of the absolute maximum or minimum for the whole curve. The curves obtained for right- and left-handed crystals are similar in shape, but opposite in sign at corresponding parts of the curve. Right-handed crystals commenced with a positive reading and left-handed with a negative one in all cases examined. This difference will be further discussed under twinning. The composite curve for a left-handed crystal will now be considered in some detail. At A the curve is at its minimum value of —17°, a value which is repeated at G. From A the value changes relatively quickly to about —10° at B. Further on at C there is a maximum value of about —3°. This cor- responds to the values obtained with crystals Nos. 2 and 5. The value for crystal No. 4 was 0°, but as this point was only fixed by a single reading it may be in error. This maximum at C is —7° for crystal No. 1. This may be due to this erystal being much more deeply etched than either Nos. 2 or 5 by the time this point was reached. The value of the minimum at D varies from —7° to —9° for different crystals. The maximum value of 0° at # seems to be one of the most definite points on the curve, because it was obtained with all three erystals measured, which were considered to correspond to the early portion of the composite curve. After the maximum at #, there is a relatively flat portion of the curve which varies gradually from —6° to —10°. This portion was unduly prolonged in the curves for crystals Nos. 1 and 4, owing to the bottle of old acid, referred to previously, being used. From —10°, the curve changes suddenly to —17° at G, which is the absolute minimum value reached at any point of the curve. The suddenness of this change in crystals Nos. 1 and 4 may be partly due to the use of fresh acid, which was definitely stronger than the average acid used in the rest of the work. For this reason, the shape of this portion of the composite curve was taken from that for crystal No. 5, which was considered more representative of what would be obtained with medium strength acid. From G there is a normal rate of change to about —12° to —10° at H. Between H and J there are relatively few readings, on which to base the curve. The value of the maximum at J is about +4°. The values quickly change again to negative and are about —6° at J, which is the furthest point reached in any curve. The form taken by the composite curve in stages later than those Shown is unknown. One section which gave a reading of —10° after approxi- mately one hour’s etching read -+10° after a further 16 hours. From this it seems possible that a section which gave a reading of —17° at one stage may vary and give a reading of +17° at a later stage, in which case the composite curve would repeat itself on the opposite side of the zero position. The small difference obtained by varying the surface finish is well shown in Figure 4. This test was made on two portions of the same section of crystal. The section was first surfaced using 4F grade carborundum and then broken across the middle. One piece was then completely resurfaced using 220 grade carborundum. The similarity between the two curves obtained is striking. In actually taking the readings, the position of maximum illumination seemed 1G F, N. HANLON. somewhat easier to judge in the section having the coarser finish, but the curve obtained from the other section is somewhat smoother. The more uneven curve in the former case may be due to reflections from the rougher surface interfering to some extent. It is doubtful if there is anything to be gained by using a finer finish that that obtained with 220 grade carborundum. Variation in the angle of incidence of the light has no effect over quite a large range. This is shown in the graph, Figure 5. Below an angle of 52° the Angle of incidence of light 70° 60° 50° 40° 30° FIGURE 5 change is however very marked. This variation in the angle of maximum illumination as the angle of incidence changes would seem to be due to the curvature of the etch faces. The effect of varying the angle of incidence was only measured on one section. If the etching had been lighter or deeper, the critical angle of incidence, at which the angle of maximum illumination started to change, might have been different. Still the curve does seem to show that irrespective of the depth of etching, there would be a considerable range over which the angle of incidence could be varied without changing the readings obtained. In etching the sections of quartz, the acid tends to attack along lines of weakness in the crystal structure, and as a result the faces of the etch figures also bear a relationship to this structure. The fact that the relative positions of these etch faces is not constant may be due to the distorted spiral nature of the atomic structure of «-quartz. If a few layers only of the atomic structure be considered, the planes in which the packing of the atoms is closest may not be the same as when the neighbouring layers are considered too. If it were so it would mean that the relationship of the etch faces to the crystal faces would change with depth of etching. An interesting point is the relationship of the readings obtained on opposite sides of a slab. This relationship is best explained by reference to Figure 6. In the following it is assumed that the etching of both sides of the slab has reached the same stage of development and that the light is incident from the right. If the angle at which maximum reflection is obtained is taken as 10°, the readings obtained for the three positions of maximum illumination, with the slab in the position shown in Figure 6 (a), are approximately 10°, 130° and 250°, and an etch pit would be orientated as shown by the triangle abe. The same reading would have been obtained had CD or HF been in the position occupied by AB. When the slab is reversed as shown in Figure 6 (b), the readings obtained are 70°, 190° and 310°, and an etch pit would be as shown by the triangle def. THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. AT It will be noticed, however, that the light is now falling from the direction of corner F' instead of corner C. If allowance were made for this fact by deducting 180° from the above readings, they would become 10°, 130° and 250°, as on the other side of the slab. If the position of the pit abc be projected through the slab on to the other surface, it will occupy the position a,b,c,. From studying the relationship between the triangles a,b,c, and def, it will be seen that the bisectors of the angles between the positions for maximum illumination on one side of the slab and the projections of the positions of maximum illumination on the other side of the slab, that is, between the perpendiculars to the sides of the triangles, are parallel to the prism faces. With one slab examined in this way, it was found that after one hour’s etching the opposite sides of the slab gave the same reading. The slab was then etched for a further 16 hours and the two sides of the slab again gave equal readings, which, however, differed from those obtained after (a) (b) FIGURE 6. one hour. This may have been a coincidence, but if the readings on opposite sides of a slab were always equal, it would mean that the difference in the starting points for the curves for different sections were due, not to the relationship of the ground surface to the crystal structure, but to the conditions of etching, such as temperature and strength of acid and relative temperature of the slab itself, when placed in the acid. When a crystal is twinned, each portion of the section develops its own curve. No matter what type of twinning is present, each twinned portion of the section reaches the same degree of development after the same period of etching. If a crystal is electrically twinned (type (a)), one portion is rotated through 180° with reference to the other. In this case maximum illumination is obtained at approximately every 60°, the alternate readings being from the different twinned individuals. (See Plate II.) An interesting point is that the differences of each of these six readings from the average, taken in order, lie on a fairly regular curve. This fact was used as a check on the accuracy of the readings. Any reading to which this did not apply was always found to be inaccurate when checked. Portions of the curves obtained with one crystal are shown in Figure 7. Part (a) shows the curves for the six positions of maximum illumina- tion and part (b) the two curves for the mean position. When a crystal is optically twinned, the positions of maximum illumination of each portion vary by the same amount but in opposite directions from an intermediate position. This intermediate position depends on the type of 4c F. N. HANLON. Time of etching in minutes. —. 3390 = 370 410 450 490 H i fi a Q, HH au {A ALL A (a) (b) FIGURE 7 Incident Light /a FIGURE 8 optical twinning present, but the actual value is independent of the period of etching. } ; When the twinning is of type (b), the intermediate position is 30°. This is made clear in Figure 8. The right-hand side shows the shape of a triangular pit in a right-handed crystal, assuming the angle of maximum illumination is 10°. Similarly, the Se THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. 49 left-hand side represents a left-handed crystal, being a mirror-image of the right-hand side. Maximum illumination occurs when the sides of the triangle are perpendicular to the incident light and the light is shining across the triangle. It will be seen that maximum illumination occurs for the right-hand side at 10°, 130° and 250° and for the left-hand side at 50°, 170° and 290°. The intermediate positions are, therefore, 30°, 150°, 270°. The difference between two corres- ponding readings for maximum illumination of the two parts of the twin varies from 26° to 60°, depending on the period of etching. Practically all optical twinning is of this type. It usually occurs as a network crossing at 60°. (See Plate III.) This network is sometimes very fine (Plate IIT), and if this be the case it is masked by very deep etching. At the intermediate positions, the illumination from each twinned portion is equal. Figure 9 shows the three curves obtained from an optically twinned section, the two of maximum and the one of equal illumination. Time of etching in minutes. 1OO 140 180 220 2600 7 300 340 + 380 FIGURE <9 When type (ce) is present, the intermediate positions are 0°, 120°, 240°. The difference between two corresponding readings for maximum illumination of the two parts of the twin varies from 0° to 34°, depending on the period of etching. Provided etching is carried out over a sufficiently long period, types (bd) and (c) can always be differentiated, as a difference in corresponding readings greater than 34° can only occur with type (b), and less than 26° with type (c). These results are summarised in Table 1. The reading for maximum illumination, in each case, is taken as 10°, and for the sake of clearness, the TABLE f. Untwinned Crystal. Type of Twinning. Right-handed. Left-handed. (a) (b) (c) A A B B AA BB AB AB AB AB 10 10 10 10 50 50 50 50 70 70 70 70 110 110 110 110 130 130 130 130 170 170 170 170 190 190 190 190 230 230 230 230 250 250 250 250 290 290 290 290 310 310 310 310 350 350 350 350 50 F. N. HANLON. three positions of maximum illumination are taken as being at exactly 120° with one another. The first four columns show the readings in degrees for a right-handed and left-handed crystal and the same rotated through 180°. The other six columns show the sets of readings in degrees which would be obtained by combining any two of these. : : The order of accuracy of the readings is of importance. The brightness of the reflections shows large variations as the period of etching increases. At the commencement of etching the etch markings are only shallow and the faces formed small, resulting in relatively poor reflections. However, the brightness of the reflections does not continue to increase regularly as the period of etching increases, but passes through phases where the reflections are poor. ‘The brightest reflections would be expected when the perpendicular to the etch face bisected the angle between the incident and the reflected ray viewed through the microscope. With average strength of reflections, readings obtained for each particular maximum reflection vary about +2° from the mean position. This would give the average reading of the three positions of maximum illumination an order of accuracy of +2°, but as errors in the individual readings mostly compensate to some extent, the mean readings would usually vary by much less than --2°. When optical twinning is present, between the two corresponding readings for maximum illumination there is the position at which the illumination from the two portions of the twin is equal. This position can be judged much more accurately than the positions of maximum illumination, the order of accuracy in this case being better than -+1°. | Orientation of Unknown Sections of Quartz. T'wo cases arise, one being when the crystal is either untwinned or electrically twinned and the other when optical twinning is present. In the former case there may be two possible ways of orientating the section. One of these depends on etching the crystal over a sufficiently long period to obtain a curve which can be identified as some definite portion of the composite curve, so that either a maximum or minimum or point of inflection can be recognised. Whether a crystal were right- or left-handed would be known beforehand from examination in a polariscope. Unless further work were done in order to determine the composite curve more accurately, the order of accuracy to be expected from this method would not be closer than to within a few degrees. The main drawback is, however, the long period of etching which would be required, and this renders the method of very doubtful importance from an economic standpoint. The second method requires that readings be taken on both sides of an etched slab. The validity of this method depends on whether it can be assumed that the two etched faces would simultaneously reach the same stage of development. If it be so, any section can be orientated after one period of etching of any duration with an order of accuracy within +2°. When optical twinning is present, and practically all quartz crystals have some portion which is optically twinned, the positions of equal illumination can be read with an order of accuracy of +1°. This can be done after one period of etching and the duration of this period is immaterial. For the best results the section should be etched long enough to obtain good reflections from the etched surface. The positions of equal illumination can be checked by taking the mean of the two readings for maximum illumination. If the readings for maximum illumination are either wide, that is approaching 60°, or close, approaching 0° apart, it is advisable to re-etch for a further period, because it is much harder to judge the position of equal illumination under these circumstances. This method was tested on an unknown section. A slab of quartz cut perpendicular to the optic axis, containing the prism faces, was taken and a Journal Royal Society of N.S.W., Vol. LXXVII, 1943, Plate II Figure 2. Figure 3. BS a Lm re De wt kor i wie . = } s = o " res " = t LI iat a “ He sacs A i P ( " — _ as = . . oe ; a + i Js = : & ; ~oe “a ir rE ss nm ia 2 Mt . Bye “i! ; apt - : / ’ r a 3 he ° : T Dill it 7 i son yg y's | : { ¢ : =e as 1 Po 3 Journal Royal Society of N.S.W., Vol. LXXVII, 1943, Plate III Figure 5. Figure 7. THE ETCH FIGURES OF BASAL SECTIONS OF QUARTZ. 51 random cut made across the slab. The blank so obtained was kept for checking purposes and the other piece had all the remaining prism faces removed. Using this portion the relationship of the random cut to a prism face was determined and the result obtained checked from the blank. The error was only }°. SUMMARY. Basal sections of quartz, etched with hydrofluoric acid, were rotated on the stage of a microscope and obliquely illuminated. The position of the sections that gave the maximum reflected illumination through the microscope was determined. This position varied with the duration of etching and a curve showing this variation was constructed. While unknown sections of untwinned or electrically twinned crystals could be orientated by obtaining their curves, the process is long and tedious and the order of accuracy to be expected is not closer than to within a few degrees. If the etching on opposite sides of such a slab could be assumed to have simul- taneously reached the same stage of development, it could be orientated after one period of etching to within +2°. When optical twinning is present, and this is the ease with some portion of most crystals, an unknown section can be orientated to within +1°. ACKNOWLEDGMENTS. I wish to thank the Government Geologist, Mr. L. J. Jones, A.S.T.C., for permitting me to carry out this work; Mr. F. W. Booker, M.Sc., at whose suggestion the work was originally undertaken; and Mr. H. F. Whitworth, M.Sc., for taking the photographs accompanying this paper. I wish to sincerely thank Amalgamated Wireless (Aust.) Ltd. for the supply of all quartz sections and Miss V. Gazeley for typing the manuscript. REFERENCES. Booth, E. F., and Sayers, E. F., April, 1939. P.O. Hlectr. Eng. Jour., 32, 7. Bragg, Sir W., and Gibbs, R. E., 1925. ‘“‘ The Structure of «- and ®6-Quartz’”’, Proc. Roy. Soc., 109A, 405. | Gibbs, R. E., 1926. ‘‘ The Structure of «-Quartz’’, Proc. Roy. Soc., 110, 443. “** Piezo-electric Quartz’, 1938, Bull. Imp. Inst., 36, No. 2, 185. For a more complete bibliography, readers are referred to the last reference. DESCRIPTION OF PLATES. Puate II. Figure 1.—Etch markings by transmitted light after 72 hours’ etching. This photo shows a section which is electrically twinned. It will be noted that the triangular markings are pointing to the right in the upper portion of the photograph and to the left in the lower portion. x 200. Figure 2.—Hlectrically twinned section, showing one individual at position of maximum illumination. Figure 3.—Same rotated through 60°. Puate ITI. Figure 4.—Optically twinned section, showing left-handed individual at position of maximum illumination. Figure 5.—Same with light rotated through about 40°, bringing the right-handed individual into the position of maximum illumination. Figure 6.—Optically twinned section, showing fine mesh structure. Left-handed individual at position of maximum illumination. Figure 7.—Same with light rotated through about 40°, bringing the right-handed individual into the position of maximum illumination. - Figures 2 to 7 are all x 10 and were all taken from different parts of the same section ef one erystal. AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR.* By H. G. RAGGATT, D.Sc. I deeply appreciate the honour which the Society has done me by inviting me to deliver the Clarke Memorial Lecture for 1943, and it is with great pleasure that I welcome the opportunity which this occasion presents, to pay my tribute to the memory of the father of Australian geology, the Rev. W. B. Clarke. There are no rules laid down by the Society which Clarke Memorial Lecturers are expected to follow, and you will find that a great variety of topics has been selected in the past. It is clear, however, that past lecturers have set out to treat their subjects in a way which would make them of general interest and they have all tried to keep their discourses to a length of about 8,000 words. This lecture is a memorial to a great and versatile man whose observations covered a wide field, both in stratigraphy and economic geology. It would befit the occasion, therefore, to select a topic from one or other of those fields. Bearing in mind that stratigraphy is a difficult subject to make interesting to a non-specialist audience, and that we are engaged in the most terrible armed conflict in history, I thought it would be fitting to devote this lecture to some aspects of the Mineral Industry of Australia, and the effect of the war upon it. This audience will appreciate that there are many details which cannot be mentioned in discussing such a topic in wartime, and that this address must therefore keep mainly to generalities. Ready access to an adequate supply of most metals and many minerals is one of the most important essentials in the prosecution of war, but it would be wearisome merely to draw up a list of metals and minerals which are of special interest at the present time and to discuss them, one by one. I propose instead to deal with the whole field in very general terms but to place the accent upon the little known, new, and unusual rather than to divide my time according to the relative importance of the various metals and minerals. There are many ways in which the mineral industry of a country may be discussed. For the purposes of this address I propose to follow approximately the headings given hereunder : Precious metals. Fuels. Iron and Steel. Ferro-alloys ; Bismuth. Base metals other than Iron; Cadmium and Cobalt. Light metals. Antimony and Arsenic. Tantalum and Columbium. Beach Sand Minerals. Sulphur, Sulphides and Sulphuric Acid ; Phosphate Rock. Non-metallic Minerals. PRECIOUS METALS. We need only consider gold, silver, platinum and osmiridium. I put the precious metals first, because gold, normally the most important of the group, * The Clarke Memorial Lecture delivered to the Royal Society of New South Wales, June 23, 1943. - | be Shee ig ea ore AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 53 is the least important in this war, and thus may be dismissed in a few words. If one were seeking to compare our pre-war mineral industry with that of the present day, probably the most striking feature would be the relative unimportance of gold mining at the present time. In fact, but for special circumstances in individual localities there would have been a complete cessation of gold mining in Australia during the war. This has been brought about by the uselessness of gold as an industrial metal, the existence of a world-wide barter agreement among the anti-Axis nations which for the time being does not call for gold to balance trade deficiencies, and the need to employ aS many men as possible in the most essential work. It will be seen by reference to the table below that the decision drastically to curtail gold mining has had its most serious effects in Western Australia, where many communities have been established, entirely dependent upon gold mines. The gold mining districts of Victoria, e.g. Bendigo and Castlemaine, have also been hard hit. Other States have not suffered so much because only in very few places in those States are there mines solely devoted to winning gold. For instance 50 per cent. of the Queensland production is from Mount Morgan, which is also important as a source of copper. AUSTRALIAN GOLD PRODUCTION, 1939. Fine Oz. New South Wales. ee ae a oa ss 87,189 Victoria ne ae aes ae oe au a 156,522 Queensland .. ne At Be oe a a 147,248 South Australia .. se 5 a ih on 3,930 Western Australia Ls vey Es" a a 1,214,238 Tasmania .. Be he Bas ed re ae 19.984 Northern Territory is as 2 sie ae 16,586 Total ae ae oi ue ae oe 1,645,697 It is unnecessary to spend much time discussing silver. Ninety per cent. of the silver consumed in Australia is required for coins, 10°% is used industrially for such purposes as photographic materials, electroplating, surgical plates. However, no mining operations can now be regarded as being carried on primarily for the recovery of silver. Some silver is obtained with gold, but the bulk of it occurs associated with lead and zinc, both of which metals are required for war purposes. It is of interest to note that 60% of the world consumption of silver is used industrially. Australian resources of platinum and osmiridium are small and large pro- duction cannot be expected. Annual production of 200-300 ozs. of osmiridium can normally be obtained from alluvial sources in Tasmania. However, this is one kind of production which has been most seriously affected by manpower demands, as I shall indicate later in this address. FUELS. Australia possesses adequate supplies of bituminous coal for some consider- able time to come. These supplies however are not nearly as large as is popularly believed and certainly not large enough to permit our present wasteful methods. of mining to continue indefinitely. I speak of course in terms of a generation or two, not of the immediate future. Figure 1 shows the distribution of coal deposits in Australia. It will be noted that nearly all of them are close to the coast and it will be realised that this distribution has been an important factor in accentuating population distribution in the coastal areas. The principal deposits of bituminous coal are in Queensland and New South Wales. Reserves of coal available in these two States are probably approximately equal. B H. G. RAGGATT. 54 ess . NISWY THX) Wit) YIAIS/ NOSAPT UOZS OY OOS \* PSUY WENTIMM A Voaly WA ~AS Se re Ie OLANN DT (COVEY SUPA) FOMLY @, FIIMASNICT BYIASUAOL SUMO ] YOIIIY iff C//OIN(M t asvey egzlpiyy WITH) NOT 7 CUCLISNY YQNOD ASO7QISS2L VIDYZION V/INVWSEL 7/77 @ YaSed CVeSDSTIY U/SFQSOMN —————————s Oas oO SOISSTVOD OFAOTIANY WeoA/IN/IEA QN/MOAS | WWYLSTY JO SW Fig. 1. AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. tS % z CSa/ /6/ LA $+ — = | 5 0 XN 9 = s ~~ a fa \ ONE ORE al =o \ \ De % 8 i) ~ d ~N BLACK COAL AUSTRALIAN COAL AND IRON ORE PRODUCTION ern 2/6/ | = —-—|- — - == 2 = g “‘ D ss G/G/ hood P/G/ SUQY VOU} S 2/O Vor, 2 ib oO Sug SvOY/ pW > OF. \ 56 H. G. RAGGATT. Figure 2 shows graphically the production of coal in Australia for the period 1914-1942 and for each of the producing States in 1942. It can readily be seen from this figure that the production and distribution of coal won in New South Wales are the outstanding features of the industry and that transport of coal from New South Wales to the other States has presented a major problem to the responsible authorities. These difficulties have directed attention to the possibility of development of coal deposits considered relatively unimportant in pre-war years, but which, if developed now, would release transport for other purposes. Victoria is badly off for bituminous coal, but the sub-bituminous coal of Wonthaggi has come in useful in the present emergency. As the geologists present know, this coalfield is a maze of faults which makes mining costly and difficult. In the hope of releasing transport now engaged in carrying coal from New South Wales to the Riverina and Victoria, attention has recently been directed to the Coorabin-Oaklands coalfield. Supplementing the work done earlier by the New South Wales Mines Department and private interests, a considerable amount oi boring has been done on behalf of the Commonwealth Coal Com- mission. As a result it has been concluded that the field has interesting possibilities but that it could not be brought into production quickly on a scale large enough to be of much assistance in the present emergency. Tasmania has several areas of bituminous coal, and production has been augmented during the war to meet increased demands. This has been achieved with only slight addition to the labour force engaged. Production in 1942 was 134,442 tons. Importation of New South Wales coal amounting to about 100,000 tons is still required for gas making and special steam raising purposes. Difficulty in obtaining sufficient supplies of New South Wales coal has forced the South Australian authorities to give attention to the Leigh Creek (Copley) field. Leigh Creek is situated on the North Australian railway 165 miles north of Port Augusta in an area where large supplies of water are difficult to obtain. The coal from this field is a low ranking sub-bituminous type and its industrial application is going to be difficult. However, prospecting of the field has been pursued with great vigour by the South Australian Government and the field is now being developed. The State of Western Australia has very large areas occupied by sediments of the same geological age as those in which the major coalfields of Queensland and New South Wales occur, but unfortunately in their areas of maximum development, these sediments do not include coal of commercial quality and thickness. The coal produced from the Collie district in the south-west part of the State, however, is a very useful fuel and sufficient is produced to satisfy most of the State’s power and transportation requirements. Even were the resources of black and near-black coal in States other than Queensland and New South Wales fully developed, those States would still require a certain amount of coal from New South Wales or Queensland for gas making and some special steam-raising purposes. The lignite resources of Victoria are too well known to need redescribing. They have been very thoroughly investigated by the Victorian Mines Department and the State Electricity Commission. Total known reserves are very large but much of this is under deep cover. A great State-owned electricity generating station and briquetting factory are based on these deposits at Yallourn. Considerable reserves of lignite have also been proved in South Australia, but are undeveloped. AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 57 A fundamental weakness in the Australian economy is our lack of resources of liquid fuel or adequate substitutes therefor.. Parenthetically it may be remarked that this weakness goes side by side with lack of large suriace water resources for the generation of electric power. Australia has about the same area aS the United States (Alaska excluded) and it has been suggested that industrial development of this country comparable with that of the United States can be expected in the post-war years. Iam, I hope, a patriot, but when I consider the vast coal, oil and water resources of the U.S.A. and compare them with our own, I know that this cannot be, on the basis of known sources of fuel supply. Broadly our coal and water resources are known and the necessity for making an exhaustive survey of our petroleum possibilities is thus shown to be of paramount importance. Australia has never had a policy on petroleum exploration, but it will be forced to have one when the war ends. The American technical press is filled with articles pointing out that the discovery rate in that country is much below what is necessary to allow production rates to continue and maintain reserves. Exploration is going on at a rapid rate in South America, especially in Venezuela and Colombia, and after the war there is undoubtedly going to be a search for oil reserves on a seale previously unknown outside the U.S.A. Even areas previously considered, and regarded aS unpromising, will be reexamined. Australia will not go unnoticed in this world-wide exploration programme. Figure 3 shows very broadly the areas considered to offer the best promise for commercial accumulations of oil and gas in Australia and New Guinea. It must be admitted that, so far, the search for oil in Australia and the nearby islands has been disappointing. It is also true, however, that, with some notable exceptions, it has not been exhaustive. The most intensive studies have been made by private companies, e.g. the Australasian Petroleum Company in New Guinea and Papua, the Shell Company in Queensland, and Caltex (Aust.) Limited in Western Australia. The first of these had reached a stage where boring was in progress when the war began. Some commendable work, involving mapping of large areas, has also been done by the relatively small Australian companies, Oil Search Limited and Freney Kimberley Oil Company. The search for oil in Australia is now virtually suspended, partly because some of the most promising regions are within a combat area and partly because it is realised that in most of the potentially oil-bearing areas a long programme of survey, prospecting and development is required, and the results could have no bearing on the present conflict. An exception is the Lakes Entrance district in Victoria, where oil has been proved to be present in a glauconitic sandstone (of Middle Miocene age) extending over an area of 8 square miles at an average depth of about 1,200 feet. The proving of this area has been done mainly by small companies, but the State, and State and Commonwealth together have also assisted. Fortunately the area was brought under the control of one company, Austral Oil Drilling Syndicate, in 1940. This syndicate is to be congratulated for its efforts in collecting, collating and preserving all drilling data it was able to acquire. The central section of this area is to be developed by application of the method recommended by Mr. Leo Ranney (1941), which consists in drilling horizontal holes radially from a circular vertical shaft. The sinking of the shaft is in progress under the joint direction of the Commonwealth and Victorian Governments. The torbanites (kerosene shales) and oil shales of New South Wales, Queensland and Tasmania offer some possibilities of oil production. Interest at present centres in the Glen Davis project which recently formed the subject of a report by the Commonwealth Parliamentary Committee on Public Works. The 58 H. G. RAGGATT. following paragraph from that report expresses conclusions concerning this subject with which I agree: ‘‘ During the course of its enquiries, it was increasingly borne in on the Committee that this project could not be considered from the ordinary commercial standpoint, but must be regarded from a national point of view. The Committee has formed a definite conclusion that petrol of a satisfactory quality can be produced from Glen Davis shale, but the cost, having regard to the loss of Customs revenue, would be at least twice that of imported petrol. On economic grounds, the establishment of an industry is not warranted, and it could be justified only on the importance for national considerations of developing an Australian oil industry.’’ Here again Australia needs a national policy with regard to oil. The whole field of possibilities requires to be surveyed and the economics of various proposals contrasted—the different possible sources of oil cannot be properly considered by themselves. ENO RE ORD Great Synchine Nore/) - West Bosir I | foswich Basin | MIOST LAVOURABLE AREAS OL AND MATURAL Cas a ZN tackecyoce Wile. | =e YVI A thats seeceetel Caombrrean WTS4 ok AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 59 Australia never seems to have regarded her natural gas possibilities seriously. Preoccupation with the idea of finding petroleum has led many to dismiss gas showings with the phrase “it’s only methane ’’, yet natural gas, which is mostly methane, is almost the ideal fuel for many purposes. The main interest attaches to gas supplies near large industrial centres and probably this industry will follow the usual stages. Potential consumers will begin to realise that in natural gas they have a convenient fuel near at hand, and from this beginning its use will spread. It will be realised also, that with proper technique, large supplies of gas can be obtained over long periods from natural reservoirs, thus justifying capital outlay on pipe lines. Some industries will ultimately realise it may be to their advantage to establish themselves near gas supplies. Regional surveys and reconnaissance drilling for natural gas have been suspended for reasons similar to those mentioned under the heading of petroleum, but an interesting experiment is in progress at the Balmain Colliery, City of Sydney, which has reached a decisive stage. Here it is interesting to note how the chain ef events takes us back to the man we honour tonight, for it was the Rev. W. B. Clarke who first suggested, in 1847, when giving evidence before a Select Committee of the New South Wales Legislative Council, that coal might be found underlying the Sydney district at a considerable depth. Later Sir Edgeworth David became interested in the subject and in the light of the evidence then available forecast the approximate depth at which coal might be expected beneath Sydney. In 1891 the first Cremorne bore (on the north shore of Sydney Harbour) struck coal at approxi- mately 2,802 feet. The sinking of the Cremorne bore was followed by the putting down of the Birthday Shaft at Balmain on the southern shore of the harbour between the years 1897 and 1902. At the Balmain Colliery, coal has been developed from vertical shafts nearly 3,000 feet deep. From these shafts, workings extend underneath Sydney Harbour. When the colliery was working, the amount of gas liberated per ton of coal mined was 2,500 cubic feet. This is a very high yield. Recent investi- gation showed that despite the fact that seals placed in the workings when mining ceased in 1931 were broken, and that there was therefore free movement of air through the workings, they were full of gas and even under these conditions the gas had a calorific value of over 900 B.t.u. It was therefore demonstrated that under static conditions a coal face will exude gas for a long time. Seals have been erected in the colliery so that it will be completely air-tight and methane gas 1s being extracted from the workings by putting a vacuum on 2 pipe leading from the workings to the surface. The result of the experiment depends of course upon the length of time during which gas will continue to be given off by the coal face, and the rate of yield. At present 100,000 cubic feet of methane with a calorific value of 960 B.t.u. (equivalent to 1,000 gallons of petrol) is being marketed per week and used as a petrol substitute. The fact that the Balmain colliery is situated within the City of Sydney and that marketing of large quantities of gas will present no difficulty, made this experiment especially well worth doing. It is to be hoped that if the experiment is successful other similar prospects will be brought into production. At a later date it is hoped that Mr. Leo Ranney’s (1941) scheme of degasifying the coal in the Balmain colliery will be tried. This scheme has two objects : to produce methane gas for use as such, and to degasify the coal and so make working conditions safer. I should like to see our coal-mining engineers pay more attention to degasification as well as to gasification of coal seams in situ, a method apparently now widely practised in Russia. 60 H. G. RAGGATT. TRON AND STEEL. You all know that our iron and steel industry is based chiefly upon coal produced in central-eastern New South Wales and iron ore produced from the Middleback Ranges in South Australia. Due to the difficulties of maintaining shipments of coal and iron between these points it has been found necessary to draw upon the many somewhat small deposits in New South Wales. Cadia, Crookwell, Breadalbane, Tirranna and Michelago are some of the centres from which supplies are being drawn to help keep up Australia’s output of steel. There is a lesson to be learned here, namely that the value of a natural resource is not a static thing ; geographic position and changing circumstances may be, and often are, more important than any intrinsic worth the resource itself may possess. Some people were inclined to scoff at the re-surveys of New South Wales iron ore resources made in the years immediately preceding the war. But this audience would be surprised if I could say what proportion of Australia’s steel requirements are being met from sources in New South Wales. Surveys of mineral resources should therefore be comprehensive, thorough and continually subject to revision. Steel is the basic metal in war and, fortunately, far-sighted men have seen to it that Australia has a strong and efficient steel industry. Without this, our munitions effort could not have been made. It was only in 1915 that the B.H.P. Steel Works were opened. ‘* Today, Australia can produce steel more cheaply than either Great Britain or the U.S.A. Within the short period of twenty-five years Australia has become self-sufficient in steel. Her production per capita is roughly equal to that of Great Britain and more than double that of Japan. Only Germany and U.S.A. can claim a higher per capita production.”’ These words are taken from a recent statement by Mr. H. G. Darling (1943), Chairman of Directors of B.H.P., who himself has played no mean part in this modern miracle and who regards the future of the Australian steel industry with unshakeable confidence. We would do well to note the basic economic factor to which Mr. Darling attributes our success in steel making, namely, the ready availability of coal and iron ore at or near tidewater. The very close relationship which exists between coal and steel production is shown by Figure 2, which also gives the production of iron ore in Australia for the period 1914-1938. The publication of later figures is not permitted. FERRO-ALLOYS. In discussing coal, other fuels, and iron ore we have been dealing with commodities which, unless they occur in large quantities, are almost valueless. They are both also so essential to modern industry in times of peace, no less than in war, that the State and private interests have made it their business to investigate available resources. The same remarks do not apply to the ferro- alloys. The annual production of some ferro-alloy metals might reach only a hundred tons and still be a very valuable contribution to the war effort. The outstandingly important metals of the ferro-alloy group are tungsten, molybdenum, manganese, chromium. The principal localities for these metals in Australia are shown in Figure 4. Nearly all the deposits of the first three of these metals in Australia are of the vein or pipe type which do not lend themselves to large-scale production. Most of the chromium deposits are small and irregular. Hence production curves for these metals show sharp fluctuations, the peaks representing discoveries of easily worked deposits, soon depleted, plus a response to the stimulus of high prices ruling in the last war. (See Figures 5 and 6.) There are exceptions to the foregoing generalisation including the scheelite deposit at King Island, the deposits of manganese at Pernatty Lagoon, South Australia, and the well-known deposits of manganese at Horseshoe, and chromite at Coobina, Western Australia. Both the latter constitute a valuable reserve SS HEo Ace 61 AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. WO Shasons Bw YQI29Y WEY YY SOM % 4IIVSS YM adver yIe/G 6, “. g ov ; PL OYIA SITUM HLNOS MIN Beye gsuny vewda [+ Aegeweg -yZOomMwWe/ JaAly Ay¥DOy ‘C/AI/OG veapentdaag ex 1 $4 | sige one ae poamaysu/ —_— © mens *o —— ¢ ome YSP0IO0G/ JOM FWLISIG SUMED ONVISNIING ‘y “Sl SS Guy ] vewuyg ® | | ~ * | | ¥2AD » 1 soyqey" VW 200yaIeN/ WN/IWOLYZ asaueguey LIQSSUY wruapg hoy WIWOYH) 8 TSINVINGY NILSSON, ‘WANIDIA TOW OS SI/LITVIOJZ TWWADNIGS evooge7 Ayeway ~ augyw aw * WIT PLSD YY HL/NOS~ 20YSaS/Of ® WIIVALS = NATLST/Y CU/goo) + AYSOMYS DL NYSFHLYON | | | | | 4 | | | | | | | ! | H. G. RAGGATT. 62 Z/6/ O/8/ 006/ ™ rr) S G Qs % o 3 9 MSN apedssery ~ | vorgsdun P10 SweD WeusfoM . MSN \VOPOMTZOT | ape gissury Ql PPOLUEG 2 P10 { owen Wesf/OV ) 8/G/-/6/ JOEAL JEA/D SQUAD SUTIIGIACCD JOLIIUSD SUIMOYLS (“SW Y06) SILVALNIONQ) PLINZOGA TOY YO NO/LDNG04Y NYITVALSIY 9 is) ms 00 AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 63 puesy Suig ve { “yO SKEIS acoyorey YD S2Y7CH un} Wolfram a Scheelite Concentwtes (65 WoO, ) (Queensland production rretudes b/sinuth- wollramn concentwtas,) AUSTRALIAN TUNGSTEN FRODUCTION Showing Princ pal CONLILULING CONES, "YD Sul LN Sate plored WD S2YD7EH (MSN) DOL jy 9 VOZSUIIIO (se) Sy Buy 2 -¥D SkaIONS if P40) wwe WELJOM FP 2UIIVICD WV Great rar 19/4 -/8 vay 22/2517 aro Sjyipy { wrysg vozSi7s+/O (PLO) PVD IW 8 ASW wey LO owes WEL/OM SWILD BW ———<————_ [mm Ss SNVOL O/'VO7 ¥68/ IS /000 . 500 Fig. 6. 64 H. G. RAGGATT. but have not been worked, chiefly because of the high transport costs involved and because sufficient high-grade chrome ore has continued to be available from New Caledonia, and sufficient manganese ore for metallurgical purposes has been available until recently from Pernatty Lagoon. It is now believed the latter deposit is virtually exhausted and it has therefore been necessary recently to import ore for metallurgical purposes. When the war began it was thought unlikely that we would be able to obtain within Australia adequate supplies of high-grade manganese ore (pyrolusite) suitable for use in dry batteries. There are no large deposits of ore suitable for this purpose known to exist in Australia, but it now seems possible that sufficient ore will be available from a number of small deposits in Papua, Queensland and New South Wales to meet requirements. Recent work by the Ore-Dressing Laboratory of the New South Wales Mines Department suggests that simple methods of beneficiation can be applied to ores mined in the Tamworth district (which are known to be suitable for use in batteries but which are somewhat low in MnO, content) and perhaps also to similar ore from northern Queensland. Papuan ore is satisfactory without treatment but the ore bodies are believed to be small and mining and transport difficulties are of course difficult to overcome at the present time. High prices and a strong demand have had little effect upon production of molybdenite, the ore of molybdenum, and no really large sources of supply are known. Production is increasing, however, and the deposits now being developed at Wonbah in Queensland, Whipstick in New South Wales and Everton in Victoria should provide useful increments to output in the near future. Though very large supplies of molybdenite are available in the United States they are not sufficient to meet all requirements of the Allied Nations. By overrunning part of China, Thailand and Malaya the Japanese obtained access to abundant supplies of wolfram, and thus effectively cut off the allied nations from access to the principal sources of tungsten ore. In 1939 world production of tungsten concentrates was about 33,000 tons, of which 60% came from China and adjacent lands, 10°, each from Portugal, Bolivia and the United States, and 3% from Australia and Argentina. As tungsten is an absolutely essential metal for war purposes, a special effort on the part of all allied nations was obviously required. The maximum production of tungsten concentrates from Australia was obtained in 1905 when about 1,700 tons were produced, nearly all of it in Queensland. In 1917 and 1918 production fell just short of 1,100 tons. It will be apparent, therefore, that early in the war the prospects of really large production of tungsten concentrates from Australia were not hopeful but the outlook has been completely changed by the results of a diamond drilling campaign on King Island, Tasmania, completed a few months ago. This was done on the recommendation of the Minerals Committee following a report by Messrs. Mawby and Nye. The King Island deposit is now shown to be one of the largest of its kind in the world. It can, moreover, be developed by open cut methods and everything is now being done to increase output. It is greatly to be regretted that the size of this deposit was not realised earlier, but there is no blame attaching to anyone for that. The deposit is a disseminated type and its size could only be determined by a costly diamond drilling campaign. The planning of the development is the responsibility of the Controller of Minerals Production, acting as adviser to the King Island Scheelite Company. This seems to be a case where ‘“ Government interference ’’ has had useful results. Until full-scale development is reached at King Island, requirements must be met by existing production from that source and other smaller ones, chief of which are the Aberfoyle and Storey’s Creek mines in Tasmania, the Hatches Creek and Wauchope Creek fields in the Northern Territory and the Wolfram se AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 65 Camp area in Queensland. There are, of course, many other small producers, most of them in eastern Australia. As virtually all the bismuth currently produced in Australia is obtained as a by-product from the mining of tungsten and molybdenum, it may be conveniently mentioned here. By making a special effort Australia could just about attain self-sufficiency in bismuth, but it is impossible even to consider making such an effort whilst manpower requirements for other vital production are unobtainable. BASE METALS OTHER THAN [RON. The principal metals dealt with under this heading are lead, zinc, copper and tin. The principal producing centres are shown in Figure 7. As all current production of cadmium and cobalt in Australia is obtained as a by-product from the treatment of zinc concentrates those two metals are also dealt with in this section. Lead, zine and copper are all required in large quantities, the first three chiefly in direct munitions uses. Tin. The main uses of tin are in tinplate, solder and bearing alloys. Thus in normal times tin production is closely linked with that of steel. If necessary war could be waged without tin. Germany managed to do this in the first world war but only with special effort in other directions and at considerable inconvenience. One can imagine what difficulty we should have keeping up supplies of food and other commodities to the fighting forces if a substitute had to be developed for tinplate. The overrunning of Malaya and Burma by the Japanese and their blockade of China cut off the allied nations from large supplies of zinc and tin. Of the estimated tin content of ores produced in 1940 (231,700 tons) 70% (160,687 tons) came from the Malay States, N.E.I., Siam, China, Burma and Indo-China. It must also be remembered that nearly 60° of the world’s production of tin was smelted in this region. That this position might arise had been anticipated to some extent and stocks had been built up. Steps had also been taken to increase smelter capacity, especially in the U.S.A., where arrangements were made to handle Bolivian concentrates. In the period 1873-1882 (except for two years) Australia held first place among tin-producing countries. Her production was approximately 11,500 tons per annum, representing 25° of the world total. In 1939 Australia’s production was approximately 3,600 tons. Naturally everything possible has been done to increase output since that year and rationing has been applied to civil uses. Tin is produced in Australia from both veins and alluvial deposits and production is about equally divided between Queensland, New South Wales and Tasmania. Since much of the tinstone won in Queensland and New South Wales is obtained by a large number of small producers, factors to which I shall refer in closing this address, have had most serious effects on output. Fortunately these are gradually being overcome and good progress has been made in bringing new production into effect. The Tableland Tin Company’s dredge is now operating in the Mount Garnet district, Queensland and is expected to produce at the rate of 1,000 tons of metal per annum. A dredge is also being assembled on the Dorset Flats in north-east Tasmania and increases may also be expected from some of the regular producers. (It is interesting to note here that the dredge at Dorset Flats has been taken from a former gold-producing property.) Just as we have been very fortunate in having a scheelite deposit of the magnitude of King Island so it seems fortunate that the Mount Garnet area in northern Queensland is commencing production at a rate which will relieve any anxiety with regard to tin supplies in Australia and allow some H. G. RAGGATT. 66 SUL ep, 112A] WW OG2SOY -PE2Y NPLC4 WW IE/S SwieqeD @) SI7UM ™ ~y N Mines may | \\ = SSIU/OEYD IW ONVISNIFNG vedion aw ® JOVASIO MEDD, 2 E/ WW. AYOLISSY/ NeYFIHLYON Jadda)y O Du/izZ -pea7 ® PLA7 VY YISAOD B INIZ -OVI7 OVI) —-YO/ — SIYLNI) ON/DNOOYY TWd/IN ae biTvyisny Hinos — | V/ITVAHLSI YY NATLSTY AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 67 export. A most remarkable development has also occurred in the copper - industry to which reference will be made shortly. Copper. It may come as a surprise to many who remember the old copper mines of South Australia, northern Queensland and central New South Wales, that Australian production of copper is far below consumption. This, of course, is largely a reflex of our greatly expanded munitions projects. It will be of value to refer to a table giving production of copper in Australia in 1939: Tons. Queensland : Mount Morgan as “ie ee si oe ape 3,300 Cloncurry District ae che dit a Ne 2,083 New South Wales: Broken Hill.. wits abe ahs D: 4: ah 883 New Cobar Mine Sys ae Ale ui) as 515 Captain’s Flat Ae Sau ae ae oe ne 484 Tasmania: Mt. Lyell .. ie Ai Le We .. 13,483 All others (mainly Queensland) he ah i a 549 Rotaly) 2% ot: ee he oh aye ve 20718 From this table it will be seen that the number of mines which made note- worthy contributions to Australian copper production in 1939 were few and that Mt. Lyell was easily the largest source. It may be noted incidentally that Mt. Lyell reserves are nearly 15 million tons of 1-14% ore. This year the story will be very different, due to developments at Mount Isa. Production of copper (metal) commenced at this mine in April, 1943, and it is estimated that by the end of the year the monthly production will be approximately equal to that of Mount Lyell. Proved reserves at Mount Isa amount to 14 million tons of 3° ore, but the limits of the orebody have yet to be defined. The Mount Isa mine is known as a large producer of lead, zinc and silver and it must be considered very fortunate that two large orebodies should exist side by side in such a way as to enable them to be developed from the one set of main workings. It is further remarkable that the one orebody contains little copper and the other no lead or zinc. The copper-body is also free from other deleterious metals so that much of the copper produced will probably not require electrolytic refining, but can go into industry as ‘“ fire-refined ’’. The Mount Isa copper lode does not outcrop. It was discovered by diamond drilling into the hanging wall of the lead-zine orebody and has been outlined from level to level by thismeans. The relationship of the two orebodies is shown in Figure 8, which is slightly amended after Figure 6 of a paper by Blanchard and Hall (1942). This amendment has kindly been made by the authors. The production of copper at Mount Isa has only been achieved at the expense of that of lead and zine (because of manpower difficulties) and it has not been possible to man the other copper mines in the district. Another interesting wartime development relates to changes in smelter practice at Port Pirie. Before the war, the amount of lead present in copper matte received at Port Kembla from Port Pirie governed the amount of lead- bearing copper matte and concentrates Port Kembla could accept for treatment. One effect of this was, for instance, that copper produced from Captain’s Flat was lost to Australian consumers. Improved smelter procedure at Port Pirie however has changed all that. Port Kembla can now not only take the whole of Captain’s Flat’s production of copper concentrates, but copper-bearing ore from Mount Gunson, South Australia, has gone to Port Pirie in place of barren sandstone flux, and Read-Rosebery mines are enabled to mine ore containing a 68 H. G. RAGGATT. higher percentage of copper than was formerly possible. This must be acclaimed a noteworthy and opportune technical achievement. Lead and Zinc. Warlier I referred to the loss of.zine occasioned by the Japanese overrunning Burma. The extent of this loss may be realised from the aw et yeti Hoe ee ot Be ae a Ve) SMBS (ALAS 7 A Wie ri Bo fA Sa? AV CROSS SECTION ey aden LOOKING NORTH JHROUGH BLACK STAR MINE Section 17 pert projected CO Obt=117 Mat HEP | 200 ore. EID Soong Sreore of lead/- LC ore GE Copper ore bool. Fig...8. fact that Burma Corporation Limited, operating the Baldwin mines, produced 59,500 tons of concentrates in 1939, i.e. twice the normal annual production from Mount Isa mines. Fortunately, contributions of lead and zinc which Australia can make to the united nations war effort are considerable, but as production comes from a few well established mines, little discussion is required. The most important producing centres of lead and zine are Broken Hill district and Captain’s Flat in New South Wales, Mount Isa mines in Queensland and the Read-Rosebery | a AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 69 mines in Tasmania. In addition approximately 2,500 tons of lead are produced from Mount Farrell in Tasmania. The relative importance of these centres may be gauged from the following production figures for the year 1939 : Lead. Zine. Mount Isa ae ti ap oe 45,265 a 29,041 Broken Hill .. ae be M4, 204,692 ee 145,434 Captain’s Flat ike ha ey. 7,145 iy 11,425 Read-Rosebery A oh * 8,515 ae 25,021 Mount Farrell vie Uy. ne 2,365 a — It has already been pointed out that, because of the greater necessity to produce copper for home consumption it has been necessary to divert men from the production of lead and zinc at Mount Isa to copper, but the output of lead and zine from the other three main centres is still very considerable. In fact, despite all difficulties, production of both metals is above the previous record levels which were realised immediately prior to the first world war. If a special effort were needed to increase production of either or both of these metals, the only essential requirement is additional manpower, as reserves and all the necessary facilities for mining and treatment exist. The Broken Hill orebody besides yielding large quantities of silver, antimony, copper, lead, zine and gold, is also our only present source of cadmiwm and our main source of cobalt. Both these metals are recovered in the Electrolytic Zinc Company’s plant at Risdon, Tasmania. Cadmium is required mainly for electroplating and the manufacture of bearings. Cobalt has many uses, perhaps chiefly as metal, in cutting tools, dies, etc., as oxide in ceramics and in the form of cobalt salts as driers in paints and varnishes. LIGHT METALS. The most important of these metals are aluminium and magnesium, but beryl (the ore of beryllium) and lithium minerals may be conveniently referred to under this heading. Magnesium is produced by the B.H.P. at its Newcastle works. Magnesite mined in New South Wales is used and the metal is produced by thermal reduction of calcined magnesite with calcium carbide. Though production commenced only in 1941, for a time some metallic magnesium was exported to Great Britain, but this has now ceased due to the increase in our own consumption. Substantial quantities of calcined magnesite have, however, gone forward from New South Wales to the United Kingdom. Australian reserves of magnesite are fairly large, the principal deposits being situated in New South Wales, South Australia and Western Australia, and there are also large deposits of dolomite available in Queensland, New South Wales, Tasmania and South Australia. In Britain and U.S.A. substantial quantities of magnesium are made from sea water. It is most unfortunate that the manufacture of aluminium had not been commenced in Australia prior to the war. At that time, however, New South Wales was the only State which had made a survey of its resources of bausite, the principal ore of aluminium. At that time, also, the specification for bauxite suitable for use in the Bayer process, the most widely used method of producing alumina from bauxite was stated to be very rigid. Today it is admitted that bauxite containing not less than 50% alumina and slightly more than 7% silica can be classed as high-grade and bauxites containing only 30-45% alumina and 30-45% silica plus ferric oxide are included in the estimate of reserves given by the United States Bureau of Mines. Great stress was also laid upon the cheap electric current available in the principal producing countries. In those days the overall costs borne by the aluminium industry in Canada and U.S.A. were apt to be forgotten. It would be wise to remember them today. = 70 H. G. RAGGATT. Canada, one of the great aluminium producing countries of the world, has no bauxite deposits of any consequence and reserves of high-grade bauxite in the U.S.A. are insignificant. Nearly all the aluminium produced in both these countries is made from bauxite imported from British and Dutch Guiana. It is believed that all this ore occurs beneath overburden and that it is crushed, washed and dried before shipment. Much of the bauxite thus imported into U.S.A. is converted to alumina at New Orleans and Mobile—and from those’ points is sent to various parts of the United States where the larger supplies of electricity are available for reducing the alumina to metal. Some alumina is shipped via the Panama Canal as far as Vancouver. With this picture of the American industry in mind we may consider our own major bauxite resources. (See Figure 9.) In eastern New South Wales there are proved reserves of 15 million tons of bauxite containing from 34% to 45% alumina and from 1-76% to 5-47% silica. The deposits occur in two main groups, Emmaville-Inverell and Bundanoon-Wingello, the former being the higher grade. All the deposits are easily accessible and have little or no overburden. Though somewhat lower in grade, the Bundanoon deposits might be preferred to the others because of their nearness to black coal (which is required in the ratio of 1 ton to 1 ton of alumina), fabricating plants and shipping facilities. Victoria possesses the highest grade bauxite in Australia. Formerly believed to be very small, the reserves are now known to be considerable. For this development our thanks must go chiefly to Sulphates Pty. Limited. It is this company which has discovered most of the bauxite deposits and these discoveries are continuing. Useav Oo Pyrite Concentete Froduced Route ~~~ Used A Zine Cencentate Produced OC REGATTA 76 H. G. RAGGATT. unnecessary transport used by this arrangement and the transportation of pyrite concentrates, 50% only of which is sulphur, is also wasteful in itself. The ideal would be to manufacture elemental sulphur from sulphide concentrates in sufficient quantity to satisfy such requirements. Another alternative would be to transport acid, but that is very difficult. Australian producers of sulphides and especially those interested in the manufacture of sulphuric acid, are well aware of the advantages of transporting sulphur as compared with sulphides and it is believed that ultimately elemental sulphur will be produced in Australia in sufficient quantities to satisfy all domestic requirements. Phosphate Rock. From consideration of sulphur and sulphuric acid we pass naturally to phosphate rock. The Australian farmer has been encouraged to use the highest grade superphosphate fertilisers, and when war began con- sumption of such fertilisers had reached a high figure. For its phosphate this fertiliser industry depended upon high grade rock from Nauru and Ocean Island, both of which are now in Japanese hands. In the year ended 30th June, 1939, imports from these sources amounted to 800,422 tons, valued at £829,061. Loss of this phosphate is likely to have serious consequences upon certain phases of our pastoral and agricultural industries; hence the necessity for intensive prospecting and development of our indigenous resources. There seems little prospect of producing much phosphate rock in Australia, suitable for use in the manufacture of superphosphate, but considerable quantities of lower grade rock containing about 40% tricalcic phosphate and 10% ferric oxide and alumina are available in South Australia. Arrangements for using the small quantities of guano and phosphate rock still available on the Abrolhos Islands, W.A., are well advanced and a thorough investigation is being made of the deposits in the Dandaragan district in W.A. At this locality a reserve of 100,000 tons of rock with 25-35 tricalecic phosphate has been proved. The phosphate occurs as nodules in a zone up to four feet thick under relatively high overburden. Commercial use of this rock is obviously going to be difficult. Phosphate deposits are being investigated also at Holbourne Island, Queens- land, Molong-Canowindra, New South Wales, and Mansfield, Victoria, but the prospect of worthwhile production from them is not viewed very hopefully. Phosphate rock sufficient for the production of high-phosphorous pig iron is available in South Australia. NON-METALLICS. I pass now to a diverse group of minerals—the non-metallics. With a few notable exceptions this large group is apt to escape notice, especially in wartime, when attention is focused on the implements of war and thus on steel, aluminium and the metals generally. Growth in the consumption of the non- metallics is almost as true a guide to the degree of a country’s industrial develop- ment as its consumption of steel. This is particularly evident of course when we consider the refractories, the use of which increases side by side with steel. Compared with our metal industry our non-metallic industry is only a healthy baby. Some industries have been founded directly on domestic non-metallies consequent upon the local demand becoming large enough, but development has also come about in other ways. A company engaged chiefly in manufacture abroad may find it expedient to commence a branch factory in Australia. Commonly it does so on the basis of a raw material to which it is accustomed, but after a time it may find that supplies near at hand are quite as suitable as that which it has been importing. The white clay industry is an outstanding example of this. War or industrial trouble by causing a check to, or cessation of, a mineral supply may lead to the local article being tried. As an example { AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. TT we may take the tale industry which during this war has found Indian talc just as suitable as Italian and is now being foreed to go more carefully into examining the possibility of using Australian, in part or wholly. War demands and restrictions of imports have led to a greatly increased demand on domestic sources of supply of most non-metallic minerals. It is desirable that as much of this trade as possible should be retained after the war. The responsibility for this rests mainly with the producer and the processor (whose attitude to the problem is however governed to some extent by the volume of trade in individual minerals), but Government can and is doing much to establish production of many of these minerals on a sound basis. So far there is so little beneficiation practised that we are virtually trying to replace the carefully prepared article imported before the war with processed raw minerals. The Australian prospector has not helped himself as much as he might in fostering the use of Australian non-metallic minerals. He is apt not to send bulk parcels true to sample, and that is something industry will not stand. The consumer, for his part, could sometimes be a little more generous as to price. He is apt to ask the prospector at what price he can supply when he very well knows 1n many cases what he can afford to pay. From this you may conclude I am not a business man. That depends upon your viewpoint. To attempt a summary of the non-metallic industry of Australia in the course of this address would be quite impossible. It may indeed be doubted whether anyone is yet in a position to do this. There are men in the mining industry in Australia who could tell you all you want to know about silver, lead, zine, copper; others who could tell you all about tin, and a good deal about tungsten, antimony, bismuth, but I know of no person or company who could tell you all about clay, diatomite, barite, talc, to mention only a few non- metallics. Trade knows no State boundaries, and the number of companies which handle some non-metallics is very large compared with the few which deal with metals. Some members of this audience will realise that one could deal with minerals. as non-metallic in themselves or as ores of metals. In fact I have dealt with beryl as the ore of beryllium rather than as a non-metallic mineral constituent of pegmatites and as such perhaps to be described with felspar and mica. Magnesite also has been referred to as the principal ore of magnesium, and its use aS a refractory passed over. In looking through the list of non-metallic minerals it will be found that most of them are required for the same purposes in war as in peace. They are not required directly for arms and munitions but have some vital use in industrial production or in the manufacture of some article or articles the need for which is thrown into sharp relief by the war. Perhaps the micas and quartz crystals are outstanding examples. The micas are a well-known group of rock-forming minerals, which in a few places in the world occur in sufficiently large masses to become economically valuable. We are concerned only with the potassium mica, muscovite and the magnesium mica, phlogopite. Seventy-five per cent. of the world’s pre-war production of muscovite came from India and 10% from the U.S.A. Practically all the phlogopite came from Canada and Madagascar, which countries have been in keen competition, with Madagascar leading in recent years. Muscovite and phlogopite have rather different uses ; the former is chiefly now required for spark plug wrappings and radio condensers, the latter chiefly for heat insulating washers on spark plugs, though it is also used as a wrapper in some types of plugs (vide Figure 11). It will be realised that the enormously expanded aircraft industry calls for greatly increased quantities of high quality 78 H. G. RAGGATT. sheet mica. At the same time the prohibition placed on the manufacture of such articles as radiators, toasters and flat irons increases the cost of producing mica by taking away an outlet for somewhat inferior sheet. The principal muscovite deposits occur in the Harts Range and along the Plenty River, distant respectively 150 and 200 miles north-easterly from Alice Springs. Other deposits to which attention is being given are those at Morehead River, 100 miles north-west of Laura, Mount Kitchin, 40 miles west of Mungana, and Yinnietharra, on the Gascoyne River 150 miles east of Carnarvon. aA ITLL LLLDO MUSCOVITE br WRAPPING ¥- SPECIAL H.R. STEEL BRASS OR COPPER COPPER MUSCOVITE WASHERS TOTTI ee a — : Z| IS g BN VIN S/ ~_____— SPECIAL H.R. STEEL — SECTION - _— OF — — AIRCRAFT SPARK-PLUG — — TYPE B.G — Fig. 11. Phlogopite is being mined at Strangeways Range, 42 miles east-north-east of Alice Springs. The more important mica localities are shown in Figure 12. Nature certainly contrived to put all these deposits in out-of-the-way places. Muscovite has been produced from central Australia since 1890, but mining was only intermittent until 1928. The mining of phoglopite is a wartime development. A feature of mica pegmatites not generally recognised is their extreme irregularity. This is not invariably so, but it is true of a great number. When in addition it is remembered that the distribution of the mica within the dyke is sporadic and that there is a very large amount of waste it will be realised that high quality mica is a valuable commodity, especially in the larger sizes. Prices ruling in Australia have been fixed by the Prices Commissioner and range from 2s. 3d. per lb. for spotted mica and washer size (not less than 14 sq. inches) to 60s. per lb. for clear mica of special size (not less than 60 sq. inches). | ‘ ; j h AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 79 You have probably all seen the recent newspaper articles on the mica mining activities in the Northern Territory, which are now under the control of the Allied Works Council with the Controller of Minerals Production providing technical advice. The Department of Supply and Shipping is now the sole mica purchasing authority in Australia. SP ie A MrehearA Aver | §{COOK town | Me. Mikching Northern | Territory | | ' ME. /s2. [91 Ca? Pa | a” paiver | Stan Proys fe Yo Aer ls Ronge | RS. Q Wes) UCD S/I/ A Yirrietharra, | (ee Western Austrol2 louse ane sae, i OT ' South Austs/) ake? fill District - =e s = Oe ed I eS yer Gerarho’fa \\ . gstyor . PRINCIPAL LOCALITIES — FOR — NIICA . & Muscovile @ Shregopite © Damourite Pigid2: Quartz crystals are very much sought after these days. If any of you have crystals on your mantelpieces or in your offices as paper-weights, you should take them along to the Mines Department. They may be useful and valuable. Crystallised quartz, among a few other substances (e.g. tourmaline), possesses the property of generating an electrical potential when placed under stress. Conversely, if an electric charge is applied to a plate of the material in a certain manner, a change in its dimensions results. This phenomenon, known as the piezo-electric effect, is made use of in a variety of ways; notably in telephony and telegraphy, wireless transmission, ballistics, depth recording, direction finding and radio-location. In all these applications the quartz crystal is 80 H. G. RAGGATT. employed as a resonator or oscillator for the accurate frequency control of an electrical circuit operating at radio frequency. The frequency of operation is entirely a function of the thickness of the crystal plate used and its relationship to the crystallographic axes of the original crystal. Highly skilled technicians are needed for the preparation of such crystal plates. (See Figure 13.) Despite the fact that quartz is an extremely abundant constituent of the earth’s crust, flawless, untwinned specimens, suitable for use in resonator circuits, are rare. Prior to the outbreak of war practically the whole of the world’s supply was obtained from Brazil and Madagascar. In the cross sections four ways Of CUD) bites from the crysts! ere shown. (The & Y cuts are now a/mast obso/ete ) Gonventonized LDrowird of 2 Qurrls Cr ryst/ | Wie. oie: As a result of the war there is an unprecedented demand on the part of the Army, Air Force and Navy for equipment which depends entirely on the use of piezo-electric quartz for its proper functioning, and considerable difficulty is now being experienced in maintaining supplies. It may be deduced from the fact that specifications have been made less and less rigid that requirements are not being met satisfactorily. It may now be said that unflawed and untwinned crystals measuring about 1 inch by 1 inch in cross-section normal to the long axis and about 1 inch long can be used, though larger sizes are preferred. Some kinds of twinning are not objectionable in large crystals. The molybdenite-bearing pipes of the Glen Innes district, New South Wales, offer the best chances of production, but many other prospects are being explored. The remaining non-metallic minerals I must now pass over in rather rapid review. Australia’s requirements of amorphous and of some kinds of fine-grained flake graphite are being met chiefly by production from Collinsville in Queensland and Uley in South Australia, but it is still necessary to import most if not all our requirements of crystalline and of higher grade and coarser grained flake. 4000 AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 81 The domestic demand for lower grade barite and talc (including steatite and pyrophyllite) can be satisfied from sources in South Australia, Western Australia and New South Wales, but there is a great need for supplies of the higher grades of both minerals. If this need can be satisfied there is a good prospect of the trade being held after the war. Figures 14 and 15 show graphically how pro- duction of barite and the talc minerals has increased in recent years. These. may be taken as typical of most non-metallic minerals. In the ceramic industry most requirements of clay are being filled but there is a strong demand for a good plastic refractory clay. The greatly increased production of felspar is coming chiefly from Coolgardie, Broken Hill and Gumeracha. Austrahian Production of > Tate, Stestite @ Pyrophy lite. % SOO Australian Barite Productor Long Tons South Austre/ g g Me ee Long JODS a SS 4910 4920 {230 1940 /306 SHO 1920 1930 Fig. 14. Fig. 15. Some paint manufacturers (and paint is important in the war effort) are greatly in need of good yellow ochre. The fact that supplies of ochre are still being drawn from Rumbalara in the Northern Territory is both a tribute to the quality of that deposit and evidence of the need. Considerable reserves are believed to be available in this locality. Fluorite is available in sufficient quantities and research is proceeding in the preparation of synthetic cryolite therefrom, which will be required when the manufacture of ingot aluminium commences. In discussing asbestos it is necessary, because of their different uses, to distinguish clearly between the three main kinds which may be referred to as chrysotile, crocidolite (or blue amphibole) and white amphibole asbestos. There are large reserves of crocidolite in the Hamersley Range, Western Australia, and supplies are coming forward from this source in adequate quantities. There are also ample supplies of white amphibole asbestos, chiefly anthophyllite from Bindi Bindi in Western Australia and tremolite from the Lewis Ponds area in New South Wales. The amounts of blue and white amphi- bole asbestos produced in Australia in 1940 were considerably larger than in any previous year. Very large importations of chrysotile asbestos however are still necessary. In the year ended June, 1939, these amounted to 9,217 tons, valued at £178,494, and were obtained from Canada, Rhodesia and the Union of South Africa. Local production, though small, shows a rising tendency. A new development is 1940 NSW South Austrod//o 82 H. G. RAGGATT. the formation of a subsidiary company by Colonial Sugar Refining Co. Limited to work chrysotile deposits between Zeehan and Renison Bell, Tasmania. The aluminium silicates, chiefly sillimanite and kyanite, are in demand as refractories. Australia is self-sufficient in sillimanite, supplies of which are obtained from Broken Hill, New South Wales, and Mount Crawford, South Australia. Some kyanite deposits in Northern Territory and South Australia have been investigated lately but it seems as though we shall have to continue to rely on imports of this mineral from India unless sillimanite can be wholly substituted therefor. Trial parcels of sillimanite have been sent to the United States and a small export trade may result. Our known reserves. are not large enough to permit other than limited export. Diatomite, or diatomaceous earth, is a most interesting industrial mineral. It has two principal uses: filtration and insulation, and many minor ones including fillers and abrasives. Diatomite, as you know, is composed of the minute frustules of aquatic plants known as diatoms of which there are many different kinds. The assemblage of diatoms present in the diatomite determines its use. Generally if there is a mixture of long, thin, filiform and rounded honeycomb forms present, a diatomite is suitable for filtration, but if the forms present are ovate, cylindrical and boat-shaped it will be far less efficient as a filter medium though suitable for all other purposes. The extensive Californian deposits are the best known of the former type and filter aids based on this diatomite set the world standard. (Pre-war import of Californian filter aids into Australia amounted to approximately 3,000 tons.) In Queensland and New South Wales there are considerable reserves of the latter type in which Melosira is the principal form present. The Victorian diatomites are much more suitable than those of other States for filtration uses but, unfortunately, reserves are believed to be small. Something might also be done with the deposits of Recent Age in Western Australia and some interesting material has recently been received from northern Queensland and South Australia: but the outlook for large-scale substitution of Australian diatomite for imported Cali- fornian for the most exacting filtration requirements is not particularly promising. However, diatomite requirements for all uses other than filtration can be supplied without difficulty from New South Wales and Queensland. The amount of diatomite consumed in Australian industry is quite consider- able, being about 7,000 tons, of which somewhat less than half has been imported. The graph, Figure 16, shows how markedly production of diatomite has increased in the past ten years. SOME GENERAL CONSIDERATIONS. There is no doubt that in respect to knowledge concerning her resources of many metals and minerals Australia was caught at a great disadvantage compared with say Canada, India and U.S.A., all of which countries have national geological surveys and two of which have a Bureau of Mines as well. The most sketchy statements had to be compiled concerning these and other metals and minerals, which ought to have been ready before the war commenced. Take one example. It has happened that a demand has come for a mineral (beryl for instance), most of which could be met from one mine in one State. Before this could be found out, every State in the Commonwealth had to be written to and the replies examined comparatively. The essentials of this information ought to have been available in advance. I know I run the risk of being suspected of being a propagandist for my own ends, but I am prepared to run that risk because I am so certain that what I have said is true. And what is true in war should be true in peace-time. Demands usually do not arise so sharply in peace- time and the need may be less obvious, but they are there nevertheless. it f y ' iy ; - i ¢ ‘A AUSTRALIA’S MINERAL INDUSTRY IN THE PRESENT WAR. 83 In this war calls have been made upon Australia by the allied nations for production of certain minerals and metals. In some instances long delay has been caused because insufficient information was available upon which to base an answer or plan a production programme. Recently the outline of a plan to establish a United Nations Relief and Rehabilitation Administration has been released by the Australian Prime Minister. This plan clearly visualises pooling and sharing of resources along much the same lines as is being followed during the war. If the organisation there proposed comes into being it will be absolutely essential that information concerning all Australia’s resources shall be available through unified channels. Part of the job of the Branch which I direct has been, in conjunction with the Controller of Minerals Production and Minerals Committee, and of course the State Mines Departments, to make a survey of certain of Australia’s mineral resources. This work is full of absorbing interest and the most aggravating e000 Australien Dr2rtomite Froduction gS 5000 \ 4000 3000 EN 2000 < 4000 /900 1910 1920 1930 /940 Fig. 16. complexities. The Mines Departments of the States have a very good knowledge of mineral resources within their own borders, but something more than that is required for the planning of Australian industry. We want to know as much as possible about our mineral deposits on an Australian as well as a State basis. We also want to know what happens to the products of the mining industry. Until recently this has been nobody’s business. Take a common commodity such as clay. A clay produced in New South Wales is sold to an agent in Victoria. So far as New South Wales is concerned that is the end of the story ; Victoria isn’t interested and in fact is probably unaware of the transaction. By tracing what happened to the clay we find half of the production was used to make telephone insulators in Victoria and the other half found its way into a paper mill in Tasmania. Knowing that the clay is used for such purposes we are able to advise likely consumers. This is one way in which the Commonwealth supplements the work of the States and assists the development of their mineral industries. | Another most important aspect of mining in the present war may be conveniently stressed here and that is—manpower. In the war of 1914-1918 there was nothing like the call upon Australian manpower as in this. Hundreds of prospectors, gougers and small scale miners were left free to pursue their activities and under the stimulus of high prices these men contributed valuable 84 H. G. RAGGATT. quantities of ores of tin, tungsten, molybdenum, copper and other metals. This condition not only does not pertain today, but the mining industry has had to compete with others for men, machinery, transport and all essential supplies. These remarks are not made in a spirit of criticism but merely to state the facts. The effect of all this is that almost every operation—even a very small one—has to be planned and government is inevitably involved, if not in the actual mining, then in the supply of some essential requirement. Having regard to her small population, Australia is making a noteworthy contribution to the united nations’ need for mineral supplies. Mining is a hazardous business at any time and it is inevitable that in the hasty planning which has been demanded some mistakes will have occurred and money spent for little return. Critical attention is apt to be focused on such expenditure, and the favourable side of the story overlooked. An immense amount of information concerning our mineral resources has been gathered during the war which will be of permanent value ; a stimulus has been given to some kinds of production which should develop into a permanent feature of Australian industry ; in some directions we have a definite picture where before we had merely hazy ~ notions ; and some noteworthy discoveries have been made, outstanding among which is the size and value of the King Island scheelite deposit. It is a pleasure to record the assistance given in preparing this address by my colleagues in the Department of Supply and Shipping and to all those in the mining industry who have so freely made information available. REFERENCES. Ranney, Leo., 1941. Indiana Petroleum Association, North Petroleum Congress, Illinois. Darling, H. G., 1943. Ind. Australian and Min. Stand., 2512, 91. Blanchard, Roland, and Hall, Graham, 1942. Proc. Aust. Inst. Min. and Met., new series, 125, 39 cia g eo ee on. THE 7. OYAL SOCIET OF NEW SOUTH WALES a - ee nd Se : (INCORPORATED 1881) * PART III (pp. 85 to 129) |. VOL. LXXVII with Plates IV- ay, EDITED BY BS THE HONORARY SECRETARIES. _.-_-~+«*‘THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE _ STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN jek c za >, \ P - kK a ie ay ek §yYDNEY se tee PUBLISHED BY THE SOCIETY, SCIENCE HOUSE : GLOUCESTER AND ESSEX STREETS Pal 1944 t expt - Containing Papers read in August, September and October Part Ill ae ‘VIIL. —Simple Regression and Correlation, By D. T. Sawkins, M.A. ie oe aac’ 9, 1944) .. Loe ae a “ a eb i oS ArT. IX. iia: Production of Hyoscyamine from Dabenic’ Species. Part L Me hod: of Quantitative Estimation. By J. A. meee iY PS. sg é. 8. eas B. Se. a April 17, 1944). iw. ie : . weNG 5 ee ; (es Arr. X.—The Produetion of Hyoscyamine from ‘Daleiea gas _ of the Base. _ By C. S. Ralph, B.Se., and J. L. Willis, B.Se Arr, XI.—Ebonite as a Radiometer. _Radiations. By G. es Blake, F. Inst. ae M.I. E. KE. *tassued Apri 17, 1924). ; . he 4 ~ Ane. XIL ee on Colour ageiigne for Sugars. Part : The Identificatio 1 an _* Determination of Monosaccharides with Thymol, ht aii Acid | and ‘erric oe Chloride. Pe A. Bolliger, Ph.D. Monies April ig aes sis Baca Tilia Halides with Tertiary Apne ahicas M.Sc. (Issued April 17, Da a ue - rae Aer XIV _—Stringocephalid Brachiopoda i in Eastern Australia. ass nieces 17, 1944)... jens as SIMPLE REGRESSION AND CORRELATION. By D. T. SAWKINS, M.A., Reader in Statistics at Sydney University. Manuscript received, June 16, 1943. Read, August 4, 1943. This paper, which is partly expository, has been prepared with a view to making the mathematical theory of statistical regression and correlation more accessible to students. An attempt is made to clarify some important points in the theory of polynomial regression, and a simple proof of the distribution of the coefficient of normal correlation is appended. ‘The only published proofs of this distribution known to the author either require the notions of n-dimensional geometry or introduce heavy and involved multiple integration (Fisher, 1915 ; Uspensky, 1937; Plummer, 1940).* Many references are made herein to an earlier paper by the present writer which showed that the y?, ‘‘ Student’s ” ¢, Fisher’s 2, and Cauchy distributions are most easily described for practical statistical purposes in terms of Gamma and Beta functions in their simplest forms as used by Bayes and other skilful 18th century mathematicians. That paper is quoted briefly as ‘‘ E.P.’’* I. ESTIMATES OF COEFFICIENTS OF REGRESSION FROM SAMPLES. The theory of statistical covariation or ‘‘ correlation ’’ has been closely studied for a class of cases in which the mean value of one variate Y among those members of a collection of objects O which have any particular value of another variate X is a simple polynomial in X, that is M,(Y)=¢)+6,X +¢e,X7+... to a limited number of terms. This equation of mean values of Y is for historical reasons called the equation of ‘‘ regression ’’ of Y on X. Putting X=—M(X)+a, we have the alternative form M,(Y)=b,+6,0+0,07 +... When all the members of the bivariate collection are known and J}, b,,b,. . . are true coefficients of regression, we have the identities ne) -—0, +Botet... M(«Y) = [Dg es PUP at ge sg Oe Sa en ee (1) M(@?V)=boue+byu3+0ou4t--- J and so on indefinitely where the mean values now refer to the whole collection. These moment equations express the coefficients by, b,, bo, . . . in terms of the product moments of Y with successive powers of x, and the moments of X about the mean M(X). But in practice we have only a sample, large or small, of the members. The natural procedure then is to operate on the sample in the Same way aS we would on the whole of the members if they all were given. We shall thus derive estimates of the true coefficients of regression. Although there is no guarantee that an estimate of a coefficient derived from a sample of 40 members will be actually nearer to the true value than one * See references at end of paper. G—August 4, 1943. 86 D. T. SAWKINS. derived from 20 members, it will be exemplified presently in simple cases that among estimates from 40 members there is a smaller dispersion than among those derived from 20 members, and generally, that the standard error of an estimate tends to vary inversely as the square root of the number of members on en it is based, and so to follow the general statistical rule. In many practical cases of “‘ regression ’’, such as the progression of the length of a metal tape on increasing temperature, or of the calorific intake of children on increasing age, both the theory and the computations are much simplified by selecting a sample so that the values of the under-variate are symmetrically distributed, although this may be inconsistent with its actual distribution. Thus if the ages of the children under investigation are 7, 8,. . .13 years last birthday, we may take the same number of age 13 as of age 7, the same number of age 12 as of age 8, and so on. Supposing first that the equation of regression of Y on X is linear, namely M,(Y)=6,+6,x, and that the values of X are symmetrically distributed, we have from equations (1) MAY) 5-0, M(xY) =b,p., M (a? Y)=Dyu. M (a Y)=b,y,4 and so on indefinitely. When all the members are known, all these equations are satisfied by the true values of b) and b,. But with only a sample of the members available (in which X is still assumed to be symmetrically distributed), we have a choice of various estimates of 0), namely M(Y), M(a?Y)/u,, M(a*Y)/us,, ... and of b,, M(aY)/p., M(aY)/u,4, M(x ¥)/u,,. . . where all the quantities now refer to the sample. If the variance of Y about its mean value is the same for all values of X, and therefore of x, we may easily compare the variances of these alternative estimates of b, and of b,, by supposing the values of X fixed and only the values of Y variable. Taking M(ax?Y)/u, for example, this is a linear function of Y,, Ys, ...- Yn, there being » members in the sample, and writing it as es aval ‘yes — S(a?Y) its variance is seen to be Nie 1 Lee variance of the distributions of Y,, Y., ... Yn. Thus the above estimates Ya. 82 Bs g2 iy Mug 7 Mag? ng ta Me Pro Pia Mig” Mg” MULg” MULg” That each of these two series is divergent may be shown by Cauchy’s inequality (Hardy, Littlewood and Polya, 1931). For example, in the case of the second series we have the inequalities (i) wy?+be and determine a and b likewise, and so on. One advantage of this method of splitting up a regression equation of uncertain degree into separate orthogonal terms is that the coefficient of each additional orthogonal term and its variance are calculated without affecting the calculations relating to preceding terms. Of course the 3rd orthogonal term not only introduces a coefficient for x? but also alters the constant term, the 4th term introduces a a SS Teepe BERENS FES Sea SSeS SIMPLE REGRESSION AND CORRELATION. 89 coefficient for v? and also alters the coefficient of x, and so on; but when all coefficients of various powers of x are collected, we have the same regression equation as would be derived by using a single polynomial. It will be clear from the method of deriving (5) that, if by; is the coefficient of fx(x), the highest orthogonal polynomial used, we derive the standard normal variate 4/(nM[{ fe(a)})) Ps, and that the denominator of the ¢ variates is then | elas It follows from expressions (4) that when the sample is fairly numerous the standard errors of the estimated coefficients b vary inversely as the square root of the number of members. Further, if the equation of means is known to be linear, 4n members at each of two values of X as far apart as possible will give the largest value of yu, and therefore the smallest value of the standard error of b,; if the regression equation is known to be quadratic, 42 members at each of three equispaced values of X as far apart as possible will give the largest values of both y,—y,? and yw, and therefore the smallest standard errors of both b, and b,, and, generally, equispaced equifrequent values of X form the best arrangement for keeping the standard errors small. When the form of the regression equation is unknown, as it usually is in practice, it is obviously desirable to have a substantial number of values Y for each of a considerable number of values of X covering the whole range of interest. It is then possible by graphing the mean values of Y for the various values of X to get an idea of the form of the line of mean values, that is to say of the ‘line of regression ’’, and also to see, by applying tests for normality and tests for consistency of variance, whether there is reasonable justification for the usual postulate that for each value of X the values of Y are distributed normally with common standard deviation. The graph may suggest, for example, that the line of regression is of periodic, or of logarithmic, or of logistic type, and comparison of the standard deviations may indicate—as it often does in agri- cultural and biometric data—that the ratio of the standard deviation of Y to its mean value, that is the “ coefficient of variation ’’, rather than the standard deviation itself, is the same for the various values of X. When there is only one member, that is only one value of the variate Y, for each value of X, the fitting of a polynomial can hardly be expected to yield reliable conclusions if there is no prior knowledge of the scale of variation about their hypothetical mean values to which these values of Y are subject. Assuming that the unknown scale is the same for all values of XY, an estimate of this ‘* variance ’’ is S(r?)/(N —p), where N is the number of members of the sample, p the number of coefficients in the regression equation at any stage of the fitting, and S(r?) the sum of squares of residuals at this stage. Supposing the values of X equispaced, and therefore in this case symmetrically distributed, and denoting the orthogonal polynomials used above by 4, 9, . . .. we have, using the moment equations, S(7?) =S{(Y —)y —b, 0, —).9.— BAA Ch == 9 (2) —Nb,? —S(0,7) iy pee —8(@,?) i, Dee eer bie) ve; ie) (et ee g ele .ei7e lle) ce (6) So each rise in the degree of the regression equation means a further fall in the sum of squares of residuals. When this equation reaches the (N —1)th degree (involving WN coefficients), the moment equations of lowest degree produce a regression equation which is satisfied exactly by each of the N values of Y. Then S(r?)=0, and S(r?)/(N —p) takes the indeterminate form 0/0. Unless there is some knowledge (possibly from a parallel experiment) of the assumed common variance of the variates Y about their mean values, there is no means 90 D. T. SAWKINS. of gauging when S(r?) has reached a reasonable or an ‘‘ optimum ”’ value and so of judging when to stop the process of introducing further coefficients. We may of course, on the additional assumption of normal distribution of each Y variate, use the t variates (3) to test whether successive coefficients do or do not differ significantly from zero, but the answer to these tests at any stage, whether yes or no, apart from giving no information about higher coefficients, gives no information regarding the significance of S(r?) at this stage. We might try to meet this difficulty by choosing to stop the regression equation when, after the use of the pth coefficient, S(r?) first falls below aan —Y)*} where k is some fraction such as 0-2 or 0- 1 or 0-05; but as S(r?)/(N —p) is now our estimate of the common variance of the Y distributions, this neneed ane implies that we regard this latest estimate as being as small as or smaller than it should be, or, perhaps, as not worth further consideration, when it first falls below a = where s,” is the corresponding estimate of the variance on the basis of no regres- sion. Failing prior knowledge of the variance this, to say the least, would be an arbitrary rule, although in effect it is often applied. A more logical procedure would be to test at each stage whether the WN residuals could fairly be regarded aS members of some normal distribution (noting that they are bound by p relations. E.P., 230-238). On the other hand, if there are at least two values of Y for each value of X, there is available an estimate of the postulated common variance of the Y distributions which is quite independent of the degree of the regression equation. Let X,, X,,. . . Xn be the successive values of X, f,, f.,. . . fn the corresponding numbers of values of Y, and S(f)=N. If the experiment can be designed with the values of X equispaced and f,=f,= ... =fn, or at least f,=fn, fo=fn_1) and so on, so that X is symmetrically distributed and the orthogonal polynomials may be used, the progressive testing of S(r?) as the degree of the regression equation is raised is made much easier. But in any case, if the estimated M,(Y) is b)+0b,"-+),47+ .. . whether calculated from a single polynomial or a series of orthogonal polynomials, the sum of squares of residuals of the observed values of Y from their respectiveestimated means is S{(Y —b,—b,w— .. . )*}, and when the regression equation is of the (p—1)th degree, the estimated coefficients by, b,, . . . are derived from p moment equations, each linear in the values of Y. So the WN residuals are connected by these p linear relations. Assuming that the Y distributions are all normal with common variance s?, it can be shown that because of these connections S{(Y —b,—b,«— . . .)} is the sum of squares of (VY—p) normal variates with mean zero and variance s?. This proposition has been exemplified in deriving (3) for the case when orthogonal polynomials can be used, and the p normal variates were then separated from the rest. Performing the summation S(r?) first for individual values of « and then for all values of x, we have SSUY bb ee =SS{(¥s—Vx)9 SG (We be ey (7) where Y, is the observed mean. As there are n different values of X the first term on the right side is the sum of squares of (f,—-1+f,—1+.. .) or (NY —n) normal variates with mean zero and variance s*. Hence the second term is the sum of squares of (VY —p—N-+ 42) or (n—p) normal variates with mean zero and variance s*, and all these are statistically independent. SI ee eee : Fig a Beta paniots o(" a a When Hence si cv mame cea 01s ——. SIMPLE REGRESSION AND CORRELATION. 91 N —n is fairly large (as it should be if reliable results are demanded), us 1 kis Sf(Yx—b)—byn—. . af Waa S8((¥x—Ys)9} Perea wibl SUMICIONt APPYOXIMAUON 2... . 1. kee ce tee ees (8) n—p The computation naturally falls into two parts: first, S(Yx) for each value of x and SS{(Yx— Yx)*} are computed from the data ; then the moment equations are computed with f values Y, at the corresponding value «, which, of course, produces the same moment equations as those derived by using the individual values of Y. By using the Beta variate or, when NV —n is large, the y? variate, we may a then calculate the significance of the sum of weighted squares of residuals of observed means from their values as derived from the computed regression equation at any stage, in comparison with the common variance of the distribu- tions aS independently estimated from the deviations of the values of Y from their respective observed mean values. For example, in the case given in R. A. Fisher’s “ Statistics for Research Workers ’’, Sec. 44, Table 50, we have VN =823, n=9, p=2 as linear regression is being tested; SS{(Yx—Yx)*}=3,832, N—n=814; S{f(VYx—})—),x)*}=396, 3832 n—p=7T. Hence Ky? =396 (377 > 80. This value is far beyond the 0-001 significance level, which is reached at about the value 24. So linear regression is not a satisfactory hypothesis and the inclusion in the regression equation of a term or terms of higher degree is indicated. While the values of X are equispaced in Fisher’s example, opposite values of X have unequal frequencies (or numbers of values of Y), so the orthogonal polynomial of 2nd degree cannot be used to test whether quadratic regression would fit the data. The following example, in which XY is symmetrically distributed, illustrates the procedure with orthogonal polynomials. AD —4 —3 —2 —1 0 i 2 3 A f 5 5 5 5 5 5 5 5 5 2-0 2-0 3-2 5-9 4-3 5-4 6-5 5:7 6-3 0-2 3-2 3-1 6-2 4-1 5:8 4-9 6-3 4-2 Yx 2°9 3°8 3°4 4-2 4-2 5-9 5-9 5:8 5:3 1-7 2-9 2-0 4-8 4:5 5-0 5-9 7 | 5-1] 1-3 3:0 3°7 3°4 5-9 4-4 4-8 5-1 6-6 S(Yx) 8-1 14-9 15-4 24-5 23-0 26-5 28-0 30-0 27-5 S(2Y x) —32-4 —44-7 —30-8 —24:-5 +26°-5 56:0 90-9 110-0 x 16 9 4 1 1 4 9 16 S(2?Yx) 129-6 134-1 61-6 24-5 26-5 112:0 270-0 440-0 a 256 81 16 1 I 16 81 256 S(Y)=197-9 S(a¥)—150-1 300 S(a?Y)=1198 -3 S(fa*) =3540 Te 92 D. T. SAWKINS. COSCON ta a) (oer) rag rl Hence b5— W Toe aE == 4-40, pb) == i SO =0-50 . _ S(@?Y)—p8(¥) 1198-3 —6-667 x 197-9 2 N(uy—pe?) ———(«é LOO — 300 XE G7 2a = Fag oo: When orthogonal polynomials are used, as in this case S[F{Yx —09 —b1@ —bo(w? —pg)— . «PI —=iS(f¥ x2) —N by? —Nyub,? aie (ha Bao Computing S Fain from the data, this becomes 959 -3 —870:8 —75:0—9:°-6. As S{(Y¥x—Yx)}=S(Y,2)—fYx? we compute SS(Yx2)=981°8. Hence SS{( yt )*}=981-8—959 -3=22 -5 N —n=45 —9=36 i 7 av, 6 i SS{( y > CRED Yx)?} === (jis 625. On the hypothesis of linear regression, the sum of Mier squares of residuals of observed means is 13-5 and we have approximately y 2 = = == 21-6, 7 This is a very uncommon kind of value, for the total frequency of values of y 7 not more frequent than this value is of the order of 0:01. So linear regression 7 is not a satisfactory hypothesis. Trying quadratic regression, the sum of weighted squares of residuals is reduced to 3:9. To test this approximately we have x ? =; —_ 6-2. This is quite an ordinary value of 7 7, so we may stop 6 6 at quadratic regression. The estimated regression equation is M,(Y)=4-4+0-5x%—0-08(ax? —6 -67) ==3:°37 -+0;-5a—0-0807 20. ose. 6 oes bee (9) Lastly we note that, using expressions (3) after fitting quadratic regression, the standard errors of by, b,, b2, are found by dividing ih — Sr), where from | (7) S(r?) =S8{(Y¥x—YVx)%+8(f(Yx—b,—ba—.. .)9 =22:5+3:°-9, by the respective denominators +/N, s/(N Ha), een Us—p_")}, or 4/45, 4/300, 4/1540. The results are roundly 0-1, 0-05, The standard error of the constant term in (9), viz. 3-87, then is Die ee x 0-02)?} or 0-17 nearly. II. ESTIMATES OF NORMAL CORRELATION FROM SAMPLES. | Normal correlation is a particular case of linear regression. (Rietz, 1927.) — When two normal variates X, Y with means M(X), M(Y) denoted by — A, B and standard deviations ./M[{X —M(X)}], WME{Y—M(Y)}*] denoted by ox, oy are normally correlated, the coefficient of correlation e is defined as — M(X—A\Y—B)} M{(X—A)(Y —B)} : Ox0y Af MCX — AY 5) MY , the mean values being taken for all members or pairs of values of X, Y in the — doubly infinite universe. SIMPLE REGRESSION AND CORRELATION. 93 In practice we have only a finite sample of n random pairs. Operating on the sample as we would on all the members if they were known, we have an estimate 7 of the true coefficient o, namely “8((X Bee Td: st i Y\2 a a i se) ir —Ty| where XY, Y denote “s(x “8(¥). Change of origins aiid scales of measurement of the two variates does not affect r. So standardising the variates by putting x“, y for (X—A)/cx, (Y—B)/oy, we a S{(v—2£)(y —Y)} VJ [S{(x@—@)} - “Si (y—y)75] With these standardised variates the joint frequency is 1 —2 oxy ry") ——_—__—-;. ex — dad Peyi—o) t 2(1 —?) f i—— a ae (y — px)? nel Dae maeqi = 2) °XP : aay mee ai@e) ot and (y—px)/./(1—p?) and & are statistically independent standard normal variates. Writing z for (y— ex) /+/(1 , we have n pairs of standard normal variates (2, #), all of nich are natty independent. Hence S(27) =(1 — 97) XS8(y? Te ea poe) LSU a —Y¥)"} —2 pS{ (a —&)(y —Y)} + e?S{ (a —X)?} + n(y? —2 pxy + p7x?)| eee OPMR SME ATE em aiais el eiieedle eye ele alec a) lee e 60's ea ele sea dene anal (10) nh We form a set [&] orthogonal to [z] such that S(&?)=S(2?) in which 1 ORR ONE tee ere art cat - .. +%n), and so Sry aga ee) Hence from (10), writing S,?7=S{(#@—«)"}, S,2=S(y—y)?}, we have 1 Ee? +E,7 +... a eeran er sh ae it mm See SODT BTR a 1) (070) Sala Berea cme eb ails Vasia adel aay iieh'e-d Sl 6 letlehaben t/a aya (lids) n-1l We may note by the way that as oS e) and 4/n - & are statistically independent standard normal variates, therefore * and y are also normally correlated with coefficient o. Further, as [x] are eee independent of [z], we take a / SK ela. yay ta — Best... H(en—# )en}, which satisfies the orthogonal conditions (K.P., p. 215), whence au ue 4 1 bi Ne us bo= a/ S{(@ —#)3 a —2)(z mes a SA e2) S{(a% —2)(y —Y) — o(x@ —@)?} 255( 0s aera Cen gyal UiceayT eG) i, ia RS A ES Ur San ea AO en (12) meuee irom (11), 6.74. .). - on°— (1-17) 8,7/(1 — 6”), which’ is a x? AAS UI CAV INOSPEMOCIG OF Coe cies: oe cit) eie cles cia se sieieiet ee 6b vide lave « (3) 2-2 94 D. T. SAWKINS. We have also S,? as a y2y_, and statistically independent of both € and (Gat sess SER iy air ae AEE NIST Soar va an ee ee er (14) Thus there are three statistically independent variates : u=(rS,—08,)/1/(1—?), a standard normal variate ; 1 n —2\ eos ata oe 2a) 2 aay ‘ | pen Me ea \h oe iach (1—7r*), a Gamma variate ( 9 } : and Lele Sie a w=t8,?, a Gamma variate ¢ “=e Their joint frequency is Ly? —y A(m — —w 1in— We gee {ey ee pias oS *) audwdao = ag Co eae (15) Changing the variables from wu, v, w to S,, S,, 7, the jacobian is =e 0 Se 1 qe) 7 §8(1—r?) o | =8,8,2/0—py S, —S,?r O In terms of S,, S,, 7 the frequency element easily reduces to il vr0( Selle eet _ ayn \ exp saan oe) er S,S, +85] S,2-2,8,9-2(1 — p2) 4 n-0q8,dS8,dr, and as by writing out the factorials it appears that n—2 m—-1\ Wn | : )r( : )=frahe—2), 1 mI (n —2)(1 — 9?)2(a-)) This is the expression for the joint frequency derived by R. A. Fisher in 1915 the constant becomes In the important case when o=0, & takes a simple form and we derive directly from expressions (11) and (12) that r? is a Beta variate H( a ewe si _—_ 2 or Ape is a tp. variate (H.P., p. 223). 0... 4-2 246 REFERENCES. Fisher, R. A., 1915. Frequency Distribution of the Values of the Correlation Coefficient in Samples from an Indefinitely Large Population. Biometrika, 10, 507. | Fisher, R. A., 1926. Applications of “ Student’s ’’ Distribution. Metron, 5, Pt. 3, 90. : | es eT ee SIMPLE REGRESSION AND CORRELATION. 95 Fisher, R. A., 1932. Statistics for Research Workers. Oliver and Boyd, London and Edinburgh. Fisher, R. A., and Yates, F., 1938. Statistical Tables, Oliver and Boyd, London and Edinburgh, Introduction. Hardy, Littlewood and Polya, 1931. Inequalities, Cambridge. Jeffreys, H., 1939. Theory of Probability, Oxford. Plummer, H. C., 1940. Frequency and Probability, Macmillan, London. Rietz, H. L., 1927. Mathematical Statistics, Open Court, Chicago. Sawkins, D. T., 1940. Elementary Presentation of Frequency Distributions, THis JOURNAL, 74, 209 (quoted as E.P.). . Uspensky, J. V., 1937. Introduction to Mathematical Probability, McGraw Hill, New York and London. THE PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. Part I. METHODS OF QUANTITATIVE ESTIMATION. By J. A. LEAN, MP.S., and C. 8S. RALPH, B.Sc. Manuscript received, July 15, 1943. Read, August 4, 1943. During a survey of plant material made with a view to the production of hyoscyamine, a need has arisen for a reliable and accurate method for determining the total alkaloid content. Considerable attention has been paid in this regard to Duboisia species. Owing mainly to difficulties encountered with pigmented material, several of the published methods (British Pharmacopeeia, 1932, 1936) for other drugs of the Solanaceous group, when applied to Duboisia, have not yielded satis- factory results in our hands. The use of larger quantities of leaf, and decinormal solutions for titration, was found to provide a degree of accuracy unobtainable in the official process of the British Pharmacopeeia 1932 for Belladonne folium. The method submitted below has given consistent results and has been used as standard procedure in our laboratory for a considerable time. _ A number of analyses carried out on the dried material by the method described are set out in the table. Results have been expressed as hyoscyamine except in those cases where the major alkaloid present has been shown to be hyoscine, i.e. in Queensland D. myoporoides. (Annual Report C.S.1.R., 1941-42.) A. ESTIMATION OF TOTAL ALKALOID. The leaf is dried in a vacuum desiccator and ground to a No. 60 powder. About 40 gms. are accurately weighed and transferred to a flask containing 200 mls. of alcohol (95%). The mixture after the addition of 10 mls. of 10% ammonium hydroxide is shaken frequently during one hour. It is then transferred to a small percolator plugged with cotton wool, packed firmly and evenly, and percolated with 95% alcohol until complete extraction of the alkaloids is effected (Mayer’s test). The total time of percolation should not exceed 3 hours. The alcohol is recovered under reduced pressure at a temperature not exceeding 40° C. ‘To the residue is added a mixture of 50 mls. of 2% sulphuric acid and 100 mls. of water, the mixture being then completely trans- ferred to a flask. Any water-insoluble material remaining is dissolved in a small amount of alcohol (95%) and added. The acidulated solution in the flask is shaken frequently during 15 minutes and filtered, the residue on the filter being washed several times with a mixture of 10 mls. of 2% sulphuric acid and 20 mls. of water. The filtrate is transferred to a separator, 30 mls. of ether are added, the mixture shaken, allowed to separate, and the aqueous liquid reserved. The ether solution is transferred to a second separator containing 10 mls. of 2% sulphuric acid, shaken, allowed to separate, and the acid liquid added to the reserved portion. The ether is rejected. The mixed acid solutions are transferred to the first separator and the process repeated until no further colour is visible in the ether layer. To the mixed acid solutions in the first separator, 10% ammonium hydroxide is added until distinctly alkaline, and extraction is rapidly carried out using successive portions of chloro- form until complete extraction of the alkaloids is effected. The combined chloroform solutions are washed with 10 mls. of water, dried with anhydrous sodium sulphate, and filtered, the residue on the filter being washed several times with chloroform. The chloroform is then removed under Tok ee ee ie Fe = oes ae a a ot 3 en PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. 97 reduced pressure at a temperature not exceeding 40°C. The residue is treated with 3 mls. of absolute alcohol evaporated, and dried for 30 minutes at 100° C. Sulphuric acid (100 mls. N/10) is then added, any remaining insoluble pigmented matter being dissolved in a few drops of alcohol and added. The solution is finally made up to 250 mls. with distilled water. A measured volume (50 mls.) of the prepared solution, with the addition of 2 drops of methyl red indicator solution, is titrated with N/10 potassium hydroxide. Duplicate determinations are carried out. Each ml. of N/10 sulphuric acid used is equivalent to 0-0289 gm. hyoscyamine or to 0:0303 gm. hyoscine. The above method, slightly modified, was found suitable for Duboisia Liniment of the Australian War Pharmacopeia 1942. Where possible 200 mls. of the preparation are taken, the alcohol recovered in vacuo at a temperature not exceeding 40° C., and the assay completed as above, commencing with the words ‘‘ To the residue is added a mixture of 50 mls. of 2% sulphuric acid. . .”’ B. DECOMPOSITION OF HYOSCYAMINE. The process submitted below for estimation of tropic acid was utilised to determine the effect of different conditions on the decomposition of hyoscyamine in aqueous solution (cf. Part IT of this paper). ESTIMATION OF TROPIC ACID. The solution to be analysed is made acid to bromophenol blue with N sulphuric acid and extracted three times with 20% of its volume of ether. The mixed ether solutions are then washed twice with water, and the water washed twice with ether, complete separation being allowed to occur each time. The ether solutions are then combined, the ether recovered on the water bath and the residue dissolved in 10 mls. of water. Two drops of phenolphthalein solution are then added and the solution titrated with N/20 aqueous potassium hydroxide. One ml. N/20 KOH is equivalent to 0:0083 gm. tropic acid or 0:0145 gm. decomposed hyoscyamine. TABLE I, Total Alkaloid Locality. Species. Collected. Content w/w. Gosford D. myoporoides. August. 2-04% Grose River D. myoporoides. December. 1-96% Gros River D. myoporoides. January. 2-19% Grose River Bt 45 ae D. myoporoides. May. 1-82% Grose River (sucker growth) D. myoporoides. 2-59% John’s River i : D. myoporoides. December. 1-87% Macquarie Pass D. myoporoides. 1-50% Imbil (Q.) D. myoporoides. April. 2°71% Imbil (Q.) D. myoporoides. April. 2-91% Yarraman D,. Leichhardti. December. 2G Yarraman D, Leichhardtii. February. 1-72% Yarraman D. Leichhardti. March. 2°61% Yarraman D, Leichhardti. —— 2:96% Kingaroy. . D. Leichhardtii. February. 2°38% Yarraman D. Leichhardt. September. 1-82% Yarraman D. Leichhardti. November. 3°50% Yarraman D. Leichhardtii. July. Weegee Yarraman D. Leichhardtii. June. 2°63% Gosford D. myoporoides. 1-30% H—August 4, 1948. 98 LEAN AND RALPH. ACKNOWLEDGMENTS. The authors wish to express their thanks to the management of Cox, Findlayson & Company for permission to publish this work, and to Dr. A. Albert for helpful criticism. REFERENCES. British Pharmacopeeia 1932 and Addendum 1936. Council for Scientific and Industrial Research, Australia. 16th Annual Report, 1941-42, p. 11. Research Laboratory, Cox, Findlayson & Company, ° Sydney. THE PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. Part II. EXTRACTION OF THE BASE. By C. 8. RALPH, B.Sc., and J. L. WILLIS, B.Sc. Manuscript received, July 15, 1943. Read, August 4, 1943 The nature and quantity of the major alkaloids present in the leaves of Duboisia myoporoides R.Br. has been shown by Barger, Martin and Mitchell (1937), and others, to vary with the locality and other factors. Duboisia myoporoides growing in the southern and central coastal districts of New South Wales contains hyoscyamine as its major alkaloid, whilst hyoscine predominates in trees growing in northern localities. All samples of Dubowsia Leichhardt F. Muell. so far examined in this laboratory contained either hyoscyamine or a mixture of hyoscyamine and hyoscine with the former predominating. Whilst fewer difficulties were encountered during the extraction of hyoscyamine from Duboisia Letchhardiu than from Duboisia myoporoides, some stands of the latter trees were found to give excellent results and are suitable for this purpose. Duboisia myoporoides leaves from Richmond, New South Wales, for example, were found to be more satisfactory than those collected at Gosford or Macquarie Pass. The method adopted for the preparation of hyoscyamine is a modification of that used by Dunstan and Brown (1899) for the extraction of hyoscyamine from Hyoscyamus muticus Linn. The base is removed from a concentrated alcoholic extract of the leaf with water, shaken out with chloroform after the addition of ammonia, and recrystallised. Considerable variation in results was obtained during a year’s experiments with Duboisia myoporoides.* On occasions excellent yields of pure crystalline product were obtained, whilst often the product was contaminated with a large amount of syrupy base. It is thought significant that the cleanest product was obtained from leaf collected during the autumn and winter months, whereas that collected during late spring and summer was always found unsatisfactory. It seems most probable that a seasonal variation in the nature of the bases present is operative, but this result, based on only eighteen months’ work (February, 1942, to June, 1943), is subject to confirmation. The syrupy residue remaining after crystallisation of the hyoscyamine from Duboisia leaves collected during summer months consists largely of bases of lower molecular weight than that of hyoscyamine, and no further hyoscyamine could be obtained from the residue even on seeding. Insufficient data on Duboisia Leichhardt preclude any conclusions being drawn with regard to any such seasonal variation in this species. Drying. Consonant with general knowledge that solanaceous drugs lose hyoscyamine if dried without control, we have found that leaves dried in direct Sunshine retained only 70% to 75% of the alkaloid recoverable from leaves carefully dried in the shade until they became brittle. At best, only 70% to 75% of the alkaloids, indicated to be present by titration, could be recovered * Grown on Cox, Findlayson & Company’s Grose River farm. HH—August 4, 19438. 100 a ecomposition S or G Percentage DO bw RALPH AND WILLIS. ge Decomposition. fel Percenta 8 9 ph? Ce 13 Fig I Hyoscyamine concentration 27 Temp. ie. O ecomposition 197) Es) GH Percentage D tw 8 Ke) pu. 70 4/ 13 ps 10 13 Fig. 2 Fig. 3 Hyoscyamine concentration \%, Hyoscyamine concentration, o524 Temp. 18°C Temp. 18°C. Except in Fig. 4, time of reaction in each case was 5 hours. UE By extractions percentage in six tb Hyoscyam oq 3 ine removed bat) G S ~ aS Percentage, Decomposition. PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. 12 = 4 8 12 16 20 4 Time of Reaction Fig. 4. Ws Hyoscyamine concentration 27 ; p.4.9.5 , Temp. 18°C. Decomposition. Percentage 40 50 9 70 pH 7/ 72 73 72 70 20 30 Temp. in degrees C. Fig. 5 Fig. © Hyoseyamine concentration , 2 Ws 101 60 7O pt As 102 RALPH AND WILLIS. as hyoscyamine. The discrepancy is due mainly to the presence of minor alkaloids, and small losses during processing. Alcohol Recovery. It was found that up to 20° of the recoverable hyoscyamine was lost if the alcohol recovery was carried out at atmospheric pressure. .Not only is this loss avoided when reduced pressure is used, but a much cleaner product is obtained. The use of excessive steam pressures and high temperatures in the still jacket to speed up alcohol recovery leads to even greater losses of hyoscyamine. Extraction from Aqueous Solution. The ease of extraction of hyoscyamine by chloroform from aqueous solutions is increased with increase in the pH; however, saponification of the base also proceeds at an increasing rate when the pH is raised. Hxperiments were carried out to determine a satisfactory pH to adopt for this extraction, so that the number of operations would be kept to a minimum commensurate with a satisfactorily low rate of saponification. Figs. 1, 2 and 3 show the amount of decomposition in a given time of different concentra- tions of purified hyoscyamine in alkaline solution, whilst it may be seen from Fig. 4 that this decomposition is, within limits, proportional to the time of reaction. Fig. 5 shows the increasing efficiency of the chloroform extraction with increasing pH. These results indicate that a satisfactory pH value for chloroform extraction is from 9-0 to 9-5. If this value be not exceeded the rate of decomposition is not great, whilst the rapidity of extraction is sufficient for practical purposes. The amount of decomposition of hyoscyamine in alkaline solution also increases with increase in temperature of the solution. From Fig. 6 it may be seen that temperatures must not be allowed to rise above approximately 20° C. if it is desired to avoid excessive decomposition of hyoscyamine. On adding ammonia to the crude solution of hyoscyamine in water during the extraction of the base, there is often formed a finely divided brown basic precipitate. The production of this material seems to be largely dependent upon the locality from which the leaves were gathered. It has been found by experiment that no amount of rough treatment during drying will cause the formation of this brown material, but high processing temperatures seem to favour it. The removal is most troublesome and best carried out by the addition of a filter aid and filtration under vacuum. Isolation of Hyoscyamine. The hyoscyamine is extracted from the aqueous solution with chloroform and, after purification, is crystallised from its con- centrated solution in chloroform with the addition of petroleum ether. The solvent chloroform may be removed by distillation at atmospheric pressure provided that all traces of water are first removed by treatment with anhydrous sodium sulphate. The crude base obtained is a brown, fairly viscous material which crystallises on seeding with a little hyoscyamine. The brown colouring matter, if extensive, is removed by the procedure described for recrystallising the base. It is much less soluble in a mixture of chloroform and petroleum ether than is hyoscyamine, so that on addition of petroleum ether to a chloroform solution of the base the brown viscous pigment is precipitated, and on stirring adheres to the sides of the vessel. In many cases, however, when using a satisfactory source of raw material and careful methods of extraction, this procedure is not necessary. The crystallised hyoscyamine sometimes fulfils the requirements of the British Pharmaceutical Codex. If this is not so, however, treatment with animal charcoal or other activated charcoal in a mixture of chloroform and petroleum ether is generally sufficient to purify the product to this standard. If hyoscine as well as hyoscyamine is present in the leaf, separation may be effected by extraction with chloroform from the aqueous solution maintained PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. 103 at a pH of about 8; advantage is thus taken of the difference in basicity of the alkaloids, as under these circumstances very little hyoscyamine is extracted. The hyoscine may be worked up and crystallised as the hydrobromide by well- known methods. EXPERIMENTAL. Method of Extraction. Leaves from Duboisia Leichhardiu and selected Duboisia myoporoides trees were each successfully used for the extraction of hyoscyamine by this method. The following experiments were carried out on Duboisia myoporoides from near Richmond. Shade-dried leaves practically free from stalk (360 gms.) were percolated with 90% aqueous alcohol (3-5 1.) until subsequent washings gave no positive test with Mayer’s reagent. The alcohol was removed under a pressure of 60 mms. of mercury until the residue became semi- solid. The yield of soft extract was 40 gms. The alkaloid was repeatedly extracted from this material with water until the washings gave only a slight opalescence with Mayer’s reagent. The use of 2% acid gives a quicker and somewhat more complete extraction but also removes extraneous basic impurities. The combined aqueous solutions (600 mls.) were treated with ammonium hydroxide until sufficiently alkaline to turn a phenolphthalein paper decidedly pink (pH 9-3). This solution, after filtration, was repeatedly extracted with approximately 5% of its volume of chloroform until all the base was removed. The combined chloroform solutions (180 mls.) were washed with water and then twice extracted with 50 mls. of 4% sulphuric acid. The acid solution of hyoscyamine was next ammoniated and extracted to completion with chloroform. The chloroform solution of crude hyoscyamine was washed with water, carefully dried with anhydrous sodium sulphate and the solvent removed by distillation at atmospheric pressure. ‘The crude brown hyoscyamine was dissolved in a minimum of dry chloroform, light petroleum ether was added until a marked turbidity was formed, and the solution was vigorously stirred, when precipitated colouring matter adhered to the sides of the vessel. The remaining solution was decanted into a fresh beaker, a little more petroleum ether added and the whole let stand, when hyoscyamine crystallised in long needles. The crystals were filtered, washed with petroleum ether, and dried. The yield was 5 gms. or approximately 70% of the alkaloids indi- cated present by titration. Treatment of this product, in a mixture of chloroform and petroleum ether, with animal charcoal, followed by recrystallisation, gave a 90% yield of colourless crystals of melting point 107°-108° C., which gave an aurichloride of melting point 164°C. The specific rotation in 50% aqueous alcohol was —21°:8. When hyoscine was known to be present in the leaf, this was removed by adjusting the pH of the crude aqueous solution to approximately 8 with sodium bicarbonate or ammonium hydroxide, and extracted with chloroform. The pH was then adjusted to 9-3 in the usual fashion, and the hyoscyamine extracted. Drying the Leaf. Two parcels each of 10 lb. of leaf from one source (Grose River, Richmond) were dried, one in direct sunlight and the other under shelter, conditions approximating to those in use in large scale practice. In the latter case the loss in weight in ten days amounted to 70-6%, the leaves were green and brittle, and contained 1-8% alkaloid as hyoscyamine. Sun- dried leaves lost 74% of their weight in six days and were coloured a decided brown, whilst the alkaloid content, as hyoscyamine, was 1:3%. A portion of each sample (500 gms.) was worked up as described, the yields of crystallised hyoscyamine amounting to 6-8 gms. (m.p. 102°C.) from shade-dried leaf and 5-1 gms. (m.p. 102°C.) from sun-dried leaf. Recovery of Solvent Alcohol. The following experiments were carried out to decide between recovery of solvent alcohol at atmospheric or reduced pressure. Two parcels each of 500 gms. of shade-dried Duboisia myoporoides leaf were percolated with 90% alcohol until exhausted, and the solvent removed, in one case at atmospheric pressure, and in the other at a pressure of approximately 60 to 100 mms. of mercury. The extract was then worked up as already described. The hyoscyamine obtained from the experiments carried out at atmospheric pressure amounted to 5:7 gms. (m.p. 98° C.) and was contaminated with a quantity of sticky basic material precipitated on addition of petroleum ether to the chloroform solution ; the yield in the other instance was 6:8 gms. (m.p. 102° C.). In the former experiment a certain 104 RALPH AND WILLIS. amount of brown precipitated matter was formed on ammoniation of the aqueous extraction liquors, though this has been observed in greater quantity in leaf from other sources. No precipitate was formed when reduced pressure was used. Recovery of Hyoscyamine from Aqueous Solution. Four factors were considered as possibly contributing towards the decomposition of hyoscyamine in aqueous solution, viz. (a) the pH of the solution, (6b) the concentration of hyoscyamine in the solution, (c) the temperature of the solution, and (d) the time of reaction. To determine the effect of each of these variables on the rate of decomposition, solutions of hyoscyamine in dilute hydrochloric acid were prepared, and, whilst three of the factors were kept constant, the fourth was varied; the amount of decomposition was determined by estimating the amount of free tropic acid present at the end of the experiment (Lean and Ralph, 1943). The results are summarised in Table I and were used to draw the graphs 1 to 4 and 6; the discontinuity in Fig. 1 was due to the precipitation of hyoscyamine. TABLE I. Decomposition of Hyoscyamine in Aqueous Solution. (pH determined with “‘ Bayer ”’ indicator papers.) Concentration Time of Temperature pH. of Hyoscyamine. Reaction. Solution. Percentage (Per cent.) (Hours.) Se, Decomposition. 7 2 5 18 0-12 8 2 5 18 0-12 8-5 2 5 18 0-6; 0-4 9 2 5 18 0-8; 2-0 9-3 2 5 18 3°3 9-5 2 5 18 2:4; 2:7 9-7 2 5 18 4-7 10-0 2 5 18 2:5; 2-6 10-5. 2 5 18 2:8 12 2 5 18 3:8 14 2 5 18 27-0 9-5 2 2 18 1-1 9-5 2 4 18 1-9 9-5 2 6 18 3-2 9-5 2 24 18 10-9 10 2 2 18 1-1 10 2 4 18 2-2 10 2 6 18 2-6 10 2 24 18 A a 9 2 5 6 1-0 9 2 5 18 2-1 9 2 5 35 7:5 9 2 5 60 20-0 9°5 I 5 18 1-9; 2-6 10-5 1 5 18 3:7 10 1 5 18 3:2; 3:9 10 0:5 5 18 2°9 12 1 5 18 20-8 14 0:5 5 18 32-0 8 1 5 16 0-2 9 1 5 16 1-2 8 0:5 5 16 0-25 9 0-5 5 16 1-0 10 0-5 5 16 2-9 12-3 1 5 16 18-0 9-5 0-5 5 16 2-1 11-9 0-5 5 16 5:8 1 ih 100 16 0-4 PRODUCTION OF HYOSCYAMINE FROM DUBOISIA SPECIES. 105 To determine the efficiency of chloroform extraction from solutions of different pH values, solutions were prepared containing 4 gms. of hyoscyamine and sufficient ammonium sulphate, ammonium hydroxide and, if necessary, sodium hydroxide, to produce and maintain the required pH value, in 300 millilitres of water. Each solution was then extracted six times with 15 milli- litres of chloroform ; the base then was recovered and weighed directly. Percentages of hyo- scyamine recovered are given in Table 2. TABLE II. Extraction of Hyoscyamine from Aqueous Solution. (pH determined with ‘‘ Bayer ”’ indicator papers.) Hyoscyamine Extracted. pH. Weight (gms.). Percentage. a 0-1 2°5 7-5 0-1 2-5 8 D2) 0-8 30 ; 20 8-5 2-5 63 9 3°4; 3:2 85 ; 80 9-5 3:5; 3:8 88 95 10 3°6 90 For pH values above 9-5 it was found that the aqueous liquors after extraction gave a negative Mayer’s test even though 100% recovery of hyoscyamine was not obtained. Evaporation of Chloroform. Previous authors have noted that the chloroform solution of hyoscyamine must be dried prior to removal of the chloroform. Our results indicate that a-loss of up to 17% of the hyoscyamine is sustained if water be present in the chloroform during recovery. ACKNOWLEDGMENTS. The authors wish to express their thanks to the management of Cox, Findlayson & Company for permission to publish this work, and to Dr. A. Albert for much helpful criticism. REFERENCES. Barger, G., Martin, W., and Mitchell, W., 1937. J. Chem. Soc., 1821. Council for Scientific and Industrial Research, Australia. 16th Annual Report, 1941-42, p. 11. Dunstan, W. R., and Brown, H., 1899. J. Chem. Soc., 75, 73. —-———— 1901. J. Chem. Soc., 79, 71. Lean, J. A., and Ralph, C. 8. Tuts Journat, Part I of this series. Petrie, J. M., 1917. Proc. Linn. Soc. N.S.W., 42, 118. Research Laboratory, Cox, Findlayson & Company, Sydney. EBONITE AS A RADIOMETER. THE DISTORTION OF EBONITE BY LONG INFRA-RED RADIATIONS. By G. G. BLAKE, F- Inst.P., M.1.E.E. Manuscript received, June 11, 1943. Read, September 1, 1943, When one side of an ebonite strip is exposed to a greater amount of heat radiation than the other, the strip will be found to bend. Owing to its low thermal conductivity, thermal expansion will take place, mainly on the side which is irradiated, very much as is the case with a bi-metallic strip under Similar conditions. A simple and easily constructed radiometer for the detection and measure- ment of heat radiation has been developed by the author. This instrument is suitable for use in lieu of the usual thermo-pile and galvanometer when measuring and comparing heat radiations from different sides of a Leslie cube, for demon- strating absorption and reflection of heat radiation, and for experimental purposes, such as an experimental test of the Stefan-Boltzmann Law, ete. Two radiometers constructed by the author on these lines are now in use for instructional purposes in the Physics Department of Sydney University. The form of radiometer developed is illustrated in Fig. 1. The ebonite strip E (‘‘ unloaded ebonite ’’) is 2 ems. wide and 1:9 mm. thick. It is clamped at one end and projects 10 cms. beyond the clamp H. The surfaces of the ebonite may be rough or polished, but a rough surface when tested showed little or no advantage. The deflections of the free end of the strip (after magnification by a factor of approximately 1,200 obtained by means of an optical lever P, as shown) are read at a distance of 1 meter from the mirror M ; the latter is in the centre of a taut suspension system, between lightly twisted silk threads which provide a tortional restoring couple. O is an oil bath dipping into which are two small damping vanes B! and B*. The radiometer should be protected from stray radiant energy, including that from the observer. The sensitivity of the instrument described above is such that when one side of the ebonite strip is exposed to the radiation from the black surface of a 13-cm. Leslie cube, at a temperature of 90° C., and at a distance of 10 cms., a mirror scale deflection of approximately 5 cms. is obtained. Experiments with ‘“ unloaded ebonite ”’ strips of thicknesses varying from 0-7 to 7-5 mm. showed that within this range the thinner the ebonite the more sensitive and rapid was its response to long infra-red radiations. The thinner the strip the more is it subject to vibrational disturbance. The curves seen in Fig. 2 show the relationship between ebonite thickness and deflection on exposure to a fixed radiant energy source, e.g. a Leslie cube maintained at boiling point of water. The zero-drift curve, Fig. 2, represents the residual reading one minute after exposure to radiation. : The instrument may be employed as a differential radiometer by exposing simultaneously one side of the strip to each of the radiant energy sources to be compared. . MIRROR SCALE READINGS, EBONITE AS A RADIOMETER. 107 H Bigg i: THICKNESS OF EBONITE “7 4 Py 3 3-61 & ss) 6 "af CHYA | KwWRHR YE Hig. 2. MIRROR SCALE READINGS MINUTES Fig. 3. 108 G. G. BLAKE. In use the entire instrument is enclosed in a box fitted with openings on one side (or on both if it is desired to use it differentially) to admit radiations, and an opening at one end for mirror illumination. A sheet of asbestos is employed as a screen between the source of radiation and the radiometer; this is removed for each reading and then replaced. When plotting radiation curves for a gradually cooling hot body (either temperature or time against mirror scale readings), the zero drifts are deducted from the full scale readings. Fig. 3. Curve C is a radiation curve, after deduction of zero-drifts D from full scale readings R. From this figure, in which the mirror scale readings are plotted against time, it will be noted that the zero-drift, though progressive, slows down as the readings proceed. By zero-drift deduction, the radiation curve for any given radiator ig still recoverable even when an unscreened radiometer is employed, and when the latter is subjected to extraneous radiations in addition to those under measurement. It was also noted that each exposure of the ebonite to radiation produced a rapid deflection followed by a slow increase in deflection towards a maximum. On interrupting the radiation, the deflection at first rapidly decreased, and this was followed by a slow creep back to zero. It is advisable therefore to allow a short time interval between each successive reading when plotting radiation curves. In conclusion, reference to an article by Daynes (1942) is useful. It describes the various classes of ebonite now obtainable, of which ‘ unloaded ”’ has the largest coefficient of expansion, and is therefore the most suitable for use as a radiometer. This ebonite usually contains 68 parts of rubber hydrocarbon and 32 parts of sulphur, vulecanising at 140°C. ‘ For greater flexibility and deformation at higher temperatures less sulphur is required.’’? Therefore it is probable that a special ebonite could be made even more suitable for radiometric use. ACKNOWLEDGMENTS. The author wishes to thank Professor Vonwiller and members of the staff of the Department of Physics at the University of Sydney for the facilities so courteously afforded in the work associated with the development of the radiometer. REFERENCE. Daynes, H. A., 1942. Siemens Mag. Eng. Supp., 204, 1. STUDIES ON COLOUR REACTIONS FOR SUGARS. Part Il. THE IDENTIFICATION AND DETERMINATION OF MONOSACCHARIDES WITH THYMOL, HYDROCHLORIC ACID AND FERRIC CHLORIDE. By ADOLPH BOLLIGER, Ph.D. (From the Gordon Craig Research Laboratory, Department of Surgery, University of Sydney.) Manuscript received, August 14, 1943, Read September 1, 1943, The colour tests generally in use for the identification of sugars are tests for chemical groups of sugars such as Selivanoff’s test for ketohexoses or Bial’s test for pentoses. In order to identify fully an unknown sugar one has to resort to some other tests, such as the formation of crystalline derivatives, e.g. an osazone. This application of multiple tests requires a considerable amount of material which frequently is not available and sometimes is also not capable of supplying a satisfactory answer for other reasons. In contrast to these procedures of multiple tests, the possibility was studied of using one colour test only in order to identify at least any of the commonly occurring monosaccharides. The first colour test examined may be considered to be a modification of one of the original colour tests described by Molisch in 1886 for the identification of carbohydrates in general. Principle. On heating, monosaccharides form with thymol in the presence of ferric chloride and hydrochloric acid deeply coloured compounds which are insoluble in water but soluble in chloroform. The colour of the aqueous suspension and of the chloroform extract and its fluorescence are characteristic of the different chemical groups of monosaccharides. In addition, individual monosaccharides are characterised by the velocity of reaction with thymol in the presence of ferric chloride and hydrochloric acid. This velocity of reaction is indicated by the varying depth of colour of the chloroform extract of the reaction mixture alter different periods of heating. Reagents. | 5% solution of thymol in alcohol (95%). 0-5% solution of ferric chloride in concentrated hydrochloric acid. Technique. 2 ml. of an aqueous solution of the monosaccharide or the substance con- taining it are placed in a test tube of about 1 cm. diameter together with 0-5 ml. of a 5% solution of thymol in alcohol and 5 ml. of concentrated hydrochloric acid containing ferric chloride (0-5%%). After mixing the contents the tube is placed in boiling water. The water in the bath should be sufficiently deep to cover at least the height of the contents of the test tube and the bath should be sufficiently large and sufficiently heated so that the transfer of the cold test tube into it should not interrupt the boiling for more than a few seconds. During the heating the contents of the tube are observed for colour changes and the time at which they occur is noted. The changing of colour of the reaction I—September 1, 1943. 110 ADOLPH BOLLIGER. mixture is due to the formation of deeply coloured oils which become dispersed in the solution. Jf after heating for 10 minutes a definite colour change has occurred, the test tube is rapidly cooled in running cold water or in ice water. If no colour change has occurred during the first ten minutes the tube is heated further until a definite change is seen but usually for not more than 40 minutes. After cooling, a known amount of chloroform, usually 5 ml., or correspondingly less for amounts of the unknown solution under 2 ml., is added. After thorough mixing, the coloured reaction product formed dissolves readily in the chloroform. By transferring the reaction mixture and the chloroform into a larger vessel thorough mixing is facilitated. After the two layers have separated it is in most instances possible to recognize the chemical group to which the unknown monosaccharide belongs if one examines the chloroform extract in daylight and correlates this observation with the changes observed during the heating of the reaction mixture. Con- firmation and further information is gained if one examines the chloroform extract in ultra-violet light also and if one makes it alkaline with ammonia. This applies to a saccharide content of the solution tested of not less than approximately 0:05%. If it exceeds 1% observation is frequently made difficult on account of very deep colours and precipitates formed. Results obtained by different methods of examination with pure known monosaccharides are described in the following paragraphs. OBSERVATIONS ON PURE KNOWN MONOSACCHARIDES. The following biologically important monosaccharides were tested with regard to their reaction with thymol, hydrochloric acid and ferric chloride : Methyl pentose: Rhamnose. Pentoses: Arabinose, xylose. Ketohexoses: Fructose, sorbose. Aldohexoses: Galactose, glucose, mannose. A summary of the results is presented in Table I. Further details are as follows : (a) Observations during the Heating of the Reaction Mixture. If a ketohexose is present a quick and intense reaction takes place. If it is present in amounts of about 0:5 mg. or more the reaction mixture takes on a deep blackish purple colour within two minutes. If present in smaller amounts an intense green TABLE I. Chloroform Extract. Observations during Heating. Ultra-violet Daylight. Ultra-violet Light and Light. Ammonia. Rhamnose .. aS No reaction. No colour. Blue. Blue. — snes or } Olive green. Purple blue. Dull slate blue. tar tac 5" Galactose m Glucose > Green. Sky blue. Greenish blue. Greenish blue. Mannose 4] | Fructose Quick reaction. Pale whitish blue. Sorbose Dark green or Dark purple. | Brilliant greenish | Pale purple blue. purple. blue. STUDIES ON COLOUR REACTIONS FOR SUGARS. tel colour is produced within three minutes. One of the pentoses, xylose, reacts nearly as quickly as the ketohexoses but the colour produced is more of an olive green. The pentose arabinose, on the other hand, reacts only about half as fast as xylose. The aldohexoses galactose, glucose and mannose react approximately equally fast as arabinose which, however, produces more of an olive green colour compared with the green of the aldohexoses. If present in amounts of 0:2 mg. approximately, the aldohexoses begin to change towards green after about ten minutes of heating. With larger amounts they may change after five minutes. Generally speaking, with all monosaccharides mentioned the velocity of colour change is influenced considerably by the concentration of the monosaccharide present, but notwithstanding this it is possible to obtain considerable information with regard to which of the three groups mentioned the monosaccharide in question belongs. No change in colour occurs even after 40 minutes of heating if the methyl pentose rhamnose is present. The reaction mixture remains yellow throughout the whole period of boiling, though it will turn cloudy as is the case with all monosaccharides and the blank. (b) Observations on the Chloroform Extract in Daylight. Rhamnose does not yield a markedly coloured chloroform extract, it being practically colourless as is the blank, or only slightly yellow brown or bluish after boiling for 30 minutes or more. The aldopentoses, xylose and arabinose furnish a chloroform extract which has a typically cobalt or almost purplish blue appearance. The colour produced by xylose is about twice as intense as that produced by a similar amount of arabinose after a similar period of heating. The aldohexoses produce in small concentration a bluish green chloroform extract. In higher concentrations (0:1°% or more) it is of a sky blue colour. The chloroform extract from mannose cannot be distinguished from that of glucose by simple visual examination in any concentration, but that of galactose produces a slightly more dull blue and less intense colour than the other two aldohexoses concentrations being about equal. The ketohexoses, fructose and sorbose, if present in amounts of 1:0 mg. or more form a deep purple chloroform extract, which can be observed readily by drawing up some of the chloroform extract into a narrow pipette. The colour of this chloroform extract, however, is not permanent and on standing for ten minutes or more it may change towards a deep blue somewhat similar to that produced by aldohexoses in high concentration. If, however, the concentration of the ketohexoses present in the reaction mixture is considerably less than 1-0 mg. then the purple colour will not be observed and a prussian blue chloroform extract will be obtained which sometimes is somewhat difficult to differentiate from that obtained from aldohexoses in higher concentration. But in conjunc- tion with the rapid colour change in the process of heating it is possible to arrive at a definite conclusion as to whether one is dealing with a ketohexose or an aldohexose. (c) Observations on the Chloroform Extract in Ultra-violet Inght. At this Stage of the examination, the chloroform extract need not be separated from the aqueous layer, but only the chloroform extract will give a typical fluorescence when examined in ultra-violet ight in a dark room. The fluorescence obtained from a chloroform extract of a blank is a faint blue which does not materially interfere with the observations on monosaccharides. The chloroform extract from the methyl pentose rhamnose though practically colourless when examined in daylight, fluoresces with a bright blue colour of greater intensity than that of the blank and of a darker tint. The aldopentoses, xylose and arabinose, exhibit a dull slate blue. This is in marked contrast to the much more brilliant blue fluorescence of keto- and aldohexoses. The ketc- 112 ADOLPH BOLLIGER. hexoses give a strong sky blue (turquoise) fluorescence while the aldohexoses if present in about equal or even twice or three times the concentration of the ketohexoses are characterized by a less intense and more greenish blue fluorescence. If, however, the aldohexoses are present in a concentration approximately ten times stronger or more, then they show a fluorescence like that of ketohexoses. Besides these major differences between the chemical groups some of the individual monosaccharides are characterized by variations in intensity and even in colour. For example, sorbose gives a somewhat darker blue and slightly 6 YVanIT Ss Color } Intensity oO! 10 ao 340 GO MINS Time Fig. 1—A.—Arabinose. §S.=—Sorbose. F.=—Fructose. X.=Xylose. Intensity of colour-time curves of some aldopentoses and ketohexoses. The results were obtained with 0-01% solutions of the four mono- saccharides. The measurements were made with a “ Spekker”’ absorptiometer. more intense fluorescence than fructose. On the other hand, when the chloroform extract resulting from these two sugars is examined in daylight, that of sorbose is distinctly lighter in colour than that of fructose, concentration and period of heating being equal. Xylose and arabinose are again distinguished by a marked ~ difference in intensity, arabinose as also found on examination in daylight being of a lighter colour than xylose. (d) Observations on the Alkalinized Chloroform Ezxtract. The separated chloroform extract containing the reaction product is acid in reaction and if it is washed with water to remove the hydrochloric acid present it becomes pink in colour. If it is made alkaline by the addition of dilute sodium hydroxide STUDIES ON COLOUR REACTIONS FOR SUGARS. 113 the aqueous layer becomes strongly green in colour while the chloroform layer assumes a slightly pink tint. These colours, however, are not stable and within a Short time they begin to fade. However, on the addition of ammonia (e.g. 1 ml. 3% ammonia for 1 ml. chloroform extract) the aqueous layer does not become coloured. The chloroform extract changes its original colour to a pink or a red, depending on the type of monosaccharide present, its concentration and the period of heating of the original reaction mixture. These differences are not very distinct, but when examined under ultra-violet light they become marked. For example, rhamnose and the aldohexoses exhibit a fluorescence similar to that seen in the acid medium. Xylose and arabinose show a dull greenish brown fluorescence which distin- guishes these aldopentoses against the brilliant greenish blue fluorescences of the aldohexoses more markedly than that seen in the original acid medium. =! UNITS - Intensity. of Color LO MINS Time Fig. 2.—GA.=Galactose. GL.=Glucose. M.—Mannose. Intensity of colour-time curves of three aldohexoses. The results were obtained with 0:01% solutions. The measurements were made with a ‘‘ Spekker ”? absorptiometer. Finally, the ketohexose, fructose, in concentrations higher than 0-05° exhibits a typically pale whitish blue fluorescence after the addition of an excess of ammonia to the deep purple freshly prepared chloroform extract. Under similar conditions sorbose exhibits a pale purple blue fluorescence. These pale fluorescences, however, are not stable and begin to change towards a deep blue on standing for ten minutes or more. (e) Observations on Reaction Velocity. It has already been pointed out in this paper that the velocity of reaction varies for different monosaccharides. To determine the variation of the colour produced as a function of time, four or more test tubes containing equal amounts of the same monosaccharide and reagent are heated in a boiling water bath for ten minutes, another for twenty minutes, a third one for thirty minutes and a fourth one for forty minutes, and so on. After cooling the reaction mixtures adequate amounts of chloroform (usually 10 ml.) are added to each test tube to ensure complete extraction. After separating the chloroform layer 1 ml. of acid alcohol (prepared by the 114 ADOLPH BOLLIGER. addition of 5 ml. of concentrated hydrochloric acid to 95 ml. of absolute alcohol) are added to 9 ml. of chloroform extract in order to clarify it. Then the colour present in the extract is measured with a photo-electric colorimeter using chloroform as a blank. The values obtained for each monosaccharide using 2 ml. of a 0:01% solution for each test are shown in Fig. 1 and 2 in the form of time curves. They demonstrate that some of the individual monosaccharides are characterized by typical curves. For example, xylose forms a much steeper curve than arabinose. Fructose and sorbose also yield curves which are typical for each of the two ketohexoses, sorbose reacting slower and giving a less intense colour than fructose (Fig. I). In the case of the aldohexoses the picture is less favourable, glucose and mannose forming very similar curves. Galactose, however, furnishes a definitely different colour-time curve (Fig. 2). Rhamnose, of course, is characterized by the fact that practically no reaction takes place. Sorensen and Haugaard (1933) studied the velocity of reaction between a number of monosaccharides and orcinol and sulphuric acid. In their case the reaction has to be carried out at exactly 85°C. The resulting time curves from some of the monosaccharides tested (xylose and arabinose, fructose and sorbose) lie very close together. Moreover, the blank is not colourless and there seems to be no possibility of removing interfering colours from the aqueous reaction mixture. THE DETERMINATION OF AN UNKNOWN MONOSACCHARIDE. When dealing with an unknown monosaccharide it is advisable to determine first the chemical group to which it belongs. This can readily be accomplished by performing the reaction as described and by observing the colour of the chloroform extract before and after alkalinization in daylight and in ultra-violet light. The colours produced by the unknown should, if possible, be compared with those obtained under similar conditions from known monosaccharides in order to facilitate their recognition. If using similar amounts for standard and unknown, some individual monosaccharides such as fructose and sorbose may readily be recognized also at this stage, while in other instances it will be necessary to determine the velocity of reaction for this purpose. In this instance again the concentration of the unknown should preferably be about the same aS that of the known. The correct concentration which should not differ by. more than 50% from that of the known can be determined by preliminary experiments, for example, when determining the chemical group to which the unknown belongs. The velocity curves shown in this paper were obtained with concentrations of 0:01% or 0-2 mg. of monosaccharide for each test tube. Such a concentration will be found to be satisfactory though for aldohexoses a concentration of about 0:02% may be found to be preferable. While performing this qualitative test it is also possible to determine quantitatively the amount of monosaccharide present simply by comparing the colour of the chloroform extract with that of a standard of the identical monosaccharide similarly treated. SUMMARY. A simple colour reaction of monosaccharides with a reagent consisting of thymol, ferric chloride and hydrochloric acid, has been described. By observing the water-insoluble reaction product and its solution in chloroform in daylight and in ultra-violet light, it is possible to recognize the chemical group such as aldopentose, methyl pentose, aldohexose and ketohexose ¢ ay i mi he ier STUDIES ON COLOUR REACTIONS FOR SUGARS. JUIU5) to which the monosaccharide belongs. Individual monosaccharides such as fructose and sorbose may also be recognized at this stage. Individual monosaccharides differ in velocity of reaction and utilization of this easily ascertained gradient enables one to differentiate between the members of a chemical group of monosaccharides. REFERENCE. Sorensen, M., and Haugaard, G., 1933. Biochem. Z., 260, 247. THE CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. Part I. COMPOUNDS OF BIVALENT IRIDIUM HALIDES WITH TERTIARY ARSINES. By F. P. DWYER, M.Sc., and R. S. NYHOLM, M.Sc. Manuscript received, September 8, 1943. Read, October 6, 1943, Compounds of bivalent iridium appear to be as rare as those of bivalent rhodium, the only simple salt described being apparently the chloride, obtained by the pyrolysis of the trichloride, which at higher temperatures is claimed to yield the monochloride. (Wohler and Streicher, 1913.) A series of alleged ammines prepared by Skoblikoff (1885) have been shown by Palmaer (1895) to be in reality compounds of trivalent iridium, whilst the single compound 21rCl, . 3(C,H;),.S,, prepared by Ray and Adikari (1932) by heating iridium trichloride with dimethyl sulphide in alcoholic solution, cannot be formulated according to the usual coordination formule, and merits further study. From preliminary tests by the present authors, it does not appear that alkyl sulphides are capable of effecting the necessary reduction from the trivalent to the bivalent state. As with the corresponding rhodium compounds (Dwyer and Nyholm, 1941) the coordinated bivalent iridium complexes described in this paper have been prepared by reduction with hypophosphorous acid of the trivalent complexes prepared in situ. Like rhodium, simple trivalent or tetravalent iridium salts were readily reduced in acid solution to a very pale orange yellow solution in which the metal was present in the bivalent state, as a hypophosphite of consider- able stability, since no coordination compounds could be prepared from such prereduced solutions, nor could the metal be precipitated even on prolonged boiling. A series of well-defined tris arsine complexes IrX, . 3AsR, and two poorly defined tetrakis complexes IrX, . 4AsR, have been isolated. The tris complexes, although of lower stability than the corresponding rhodium compounds, were weakly coloured solids, poorly crystalline except the bromide, with relatively low melting points, insoluble in water, but easily soluble in organic solvents. In alcoholic solution they reduced silver nitrate to the metal even at room temperature but gave no precipitate of the silver halide except on prolonged boiling. The superior reducing power of the bivalent iridium complexes over the bivalent rhodium is to be expected. Although the tris compounds are probably dimeric in the solid state (I), unlike the rhodium compounds, they are dissociated in boiling chloroform or freezing benzene. In the rhodium series (Dwyer and Nyholm, 1942) stable monomeric covalent tetrakis arsine complexes were prepared with the bromide and the iodide, but with iridium the tetrakis compounds always had an odour of arsine, and they could be transformed easily to the tris compounds by shaking with petroleum ether. However, from their extreme solubility in organic solvents and their failure to precipitate the silver halide except on prolonged boiling with alcoholic silver nitrate, it is concluded that they are to be formulated IrX, . 4AsR3. DE ke eS, ye ag 5 THE CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. 117 As a result of the experiments with iridous bromide, which gave the most stable complexes, and by varying the proportions of arsine to iridium, it is concluded that only the tris and tetrakis compounds are formed, and hence that bivalent iridium, in the arsine series at least, has a coordination number of six. EXPERIMENTAL. Hexakis-diphenylmethylarsine-u-dichlorodi-iridium. JI. Sodium hexachloriridate solution (10 mls.) containing 0-123 g. iridium was reduced to the trivalent state by refluxing with alcohol (10 mls.) and concentrated hydrochloric acid (10 mls.) until the colour changed to olive green. Diphenylmethylarsine (0:47 g.) was then added, followed by alcohol sufficient to produce one phase on heating. The mixture was refluxed for a few minutes to form the trivalent compound and then 30% hypophosphorous acid (3 mls.) was added and the mixture refluxed for 50 minutes. The resulting pale yellow solution was cooled, precipitated with water, and allowed to stand, when a yellow oil separated, which was purified by solution in alcohol, reprecipitation with water, and finally addition of a little petroleum ether. The resulting very pale yellow microcrystalline powder, m.p. 115—116° C., was very soluble in organic solvents such as benzene and chloroform. An alcoholic solution reduced silver nitrate to the metal at room temperature. The substance was analysed for iridium by ignition, followed by heating in a reducing atmosphere, and for the halogen as described previously for the rhodium compounds. In this compound high results were always obtained for the halogen owing to oxidation during purification. Found: Ir=19:3%; Cl=8:9%; calculated for (IrCl, . 3(C,H;),AsCH,),, Ir=19-38% ; ei—7- 13%. Hexakis-diphenylmethylarsine-u-dibromo-di-iridium. I. This was prepared as with the chloride above, substituting hydrobromic acid (49%) for the hydrochloric acid. After about 30 minutes’ refluxing a bright yellow precipitate came down, and the reaction was stopped when the bumping had become extremely violent. The resulting compound after washing with aqueous alcohol was dried in a desiccator over sulphuric acid. The substance crystallised in short thick rods m.p. 245° C. with decomposition. It was soluble in hot acetone, sparingly soluble in hot alcohol, but soluble in cold chloroform or benzene, and precipitated from the latter solvent by petroleum ether. Found: Ir=17-6%; Br=14:7%. Mol. wt. (ebullioscopic in chloroform), 864, 816; (cryoscopie in benzene) 811. ; Calculated for (IrBr, . 3(C,H,;),As(CH,),: Ir=17-79; Br=14-74; mol. wt.=2,170. Hexakis-diphenylmethylarsine-u-dviodo-di-iridium. I. The iridium solution (10 mls.) as before was treated with potassium iodide (2 g.) and boiled until iodine ceased to be evolved. The arsine and alcohol was added to the light brown solution of iridium triiodide and finally the reducing agent, as before. After refluxing for 40 minutes, the solution was treated with water and cooled, when it gave a brownish yellow microcrystalline powder, m.p. 80-82°C. The compound reduced silver nitrate to the metal and was soluble in organic solvents. Found; Ir=16-25%; I=21-8%; calculated for (IrI, . 3(C,H;),As(CH,),: Ir=16-33% ; =21-54%. J—October 6, 1943. 118 DWYER AND NYHOLM. Hexakis-dimethyl-p-tolylarsine-u.-dichloro-di-iridium. I. This compound, prepared in the same manner as the chloride above, gave an oil which was induced to crystallise by treatment with petroleum ether at low temperatures. It was fairly soluble in petroleum ether, and easily soluble in benzene and chloroform, and reduced silver nitrate to the metal. The very pale yellow powder melted at 97°C. Found: Ir=21-9%; calculated for (IrCl, . 3(CH,),As(C,H,),; Ir=22-66%. Hexakis-dimethyl-p-tolylarsine-u.-dibromo-di-iridium. I. Prepared as before, this gave a yellow microcrystalline powder, difficult to obtain solid, m.p. 96°C. The substance was easily soluble in alcohol, acetone and benzene. Found: Ir=20:5%; Br=17-2%; calculated for (IrBr, . 3(CH;),As(C,H,), ; Ir=20-53% ; Br=17-00%. Hexakis-dimethyl-p-tolylarsine-.-diiodo-di-iridium. JI. Prepared as with the iodide above, the complex with this arsine was precipitated with water, coagulated with a little petroleum ether, washed with aqueous alcohol and dried over concentrated sulphuric acid. The substance melted at 86—88° C, was easily soluble in alcohol, acetone or benzene and reduced silver nitrate to the metal at room temperature. Found: Ir=18-7%; I=24:9%; calculated for (IrI, . 3(CH,),As(C,H,),; Ir=18:-65% ; I= 24-54%. Tetrakis Compounds. A number of attempts were made to prepare these compounds using dimethylphenylarsine and dimethyl-p-tolyl-arsine, which readily gave this type of compound with rhodium. Using iridium bromide and four moles of the arsine under the usual conditions, two oily yellow substances were obtained, which became solid after trituration with petroleum ether. The yellow solids had a powerful odour of arsine, were very readily soluble in organic solvents, and on treatment with alcoholic silver nitrate gave no precipitate of the halide except on prolonged boiling. On analysis they appeared to be mixtures of the tris and tetrakis com- pounds, and on further treatment with petroleum ether the percentage of tris compound increased, indicating loss of arsine from the tetrakis compound. Found: Ir—16-5%; calculated for IrBr, . 4(CH,),As(C,H,); Ir=16-97%. Found: Ir=19-53%; calculated for IrBr, . 4(CH;),As(C,H,;); Ir=17-87%. (The tris | compound required Ir=21-47%.) SUMMARY. The preparation of a number of bivalent iridium halides stabilised with tertiary arsines is described. The compounds are probably dimeric, with halogen bridges, and act as powerful reducing agents towards silver nitrate solution. ACKNOWLEDGMENT. The authors are indebted to Mr. D. P. Mellor for suggesting the problem of preparing bivalent iridium compounds and also for specimens of potassium chloro-iridate. REFERENCES. Dwyer and Nyholm, 1941. THis Journat, 75, 127. Dwyer and Nyholm, 1942. Tuis JouRNAL, 76, 133. Palmaer, 1895. Zest. fur anorg. Chemie, 10, 320. Ray and Adikari, 1932. Jour. Indian Chem. Soe., 9, 251. Skoblikoff, 1853. Bull. Acad. St. Petersburg, 11, 25. Wohler and Streicher, 1915. Ber., 46, 1577, 1720. Department of Chemistry, Sydney Technical College. STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. By IpA A. Brown, D.Sc., Depariment of Geology, University of Sydney. With Plates IV, V and three text-figures. Manuscript received, September 22, 1943. Read, October 6, 1943. 1. INTRODUCTION AND SUMMARY. Brachiopods of the Middle Devonian family Stringocephalide are known to occur in two provinces in eastern Australia, one in north Queensland, in the Burdekin River District south of Townsville; the other in New South Wales, near Attunga, in the Tamworth District. The genus Stringocephalus was first recorded from Australia by R. Etheridge, Junr. (1892, p. 67), who doubtfully referred a single specimen from the Fanning Limestone, Queensland, to this genus. Later, W. 8S. Dun (1900, p. 195) recorded from Crawney, N.S.W., ‘‘ Brachiopoda, indet.’’, which W. N. Benson (1922, p. 178) listed as ** Stringocephalus (2) sp. indet.’? The latter specimens (G.S. 4845, 4862, 4866) are in the Geological Survey Collection, Sydney, but are not available at present. The present paper is based on two collections: (1) from north Queensland, from Fanning River and from Reid Gap, 31 miles south of Townsville, and (2) from north of Sulcor Quarry, 4 miles north of Attunga and 17 miles north of Tamworth, N.S.W. The Queensland specimens prove to be indistinguishable from Stringocephalus burtint Defrance of the upper Middle Devonian (Givetian) beds of Paffrath, Germany. In Queensland, the associated fossils are Atrypa spp. (possibly new species), Gypidula sp. (=Pentamerus brevirostris of Etheridge, 1892, p. 67, Pl. 37, figs. 9-11), Amphipora sp., Favistella rhenana (Frech) and other Rugose corals. The presence of Stringocephalus has not been proved in New South Wales. Large brachiopods from Attunga, showing some external resemblances to Stringocephalus and erroneously referred to that genus (Brown, 1942), prove to be Bornhardtina. This form is closely related to Stringocephalus both morphologically and stratigraphically. Near the north-eastern corner of Portion 115, Parish of Burdekin, in the Attunga district, Bornhardtina forms a shell-bank or bioherm, which occurs between biostromes of Amphipora sp. (? A. ramosa). A small Rugose coral (Favistella sp.) occurs in the Brachiopod limestone. The writer is indebted to Mr. and Mrs. E. D. Coulter of Attunga for kindness and hospitality during the course of field-work, and gratefully acknowledges a grant covering field-expenses from the Commonwealth Research Fund administered by the University of Sydney. She has to thank Dr. Dorothy Hill for making possible the loan of all the material from the Queensland localities, and Mr. L. L. Waterhouse, of the Geology Department, University, for the loan of photographic apparatus. 2. STRATIGRAPHICAL CONSIDERATIONS. In the standard Devonian sequence near Cologne, in the Rhine Valley, Germany, the Middle Devonian is divided into the Calceola Stage (Eifelian or JJ—October 6, 1943. 120 IDA A. BROWN. Couvinian) below, and the Stringocephalus Stage (Givetian) above. (Tilmann et al., 1938.) The maximum development of Stringocephalus appears to be in the upper part of the Middle Devonian, where it is accompanied by Rensselandia (Newberria) caiqua (d’Archiae and Verneuil) and Amphipora ramosa (Phillips). Stringocephalus also occurs in the Middle Devonian limestones of the south of England, especially near Torquay and Plymouth. (Davidson, 1853, 1864.) Bornhardtina occurs in the lower part of the Stringocephalus Stage (the Upper Honsel or quadrigeminus beds) of the Paffrath Basin at Gerolstein, Germany. In North America, Stringocephalus was first recognised by Whiteaves (1890), 1891, as occurring in western Canada. Kirk (1927) recorded its occurrence in Prince of Wales Island, Alaska, and also in Nevada and Utah. Other occurrences are recorded by Cooper et al. (1942). Dr. Cooper discusses the evidence of the stratigraphical horizon of the Stringocephalus beds and concludes : ‘‘ This evidence suggests that the American and Canadian occurrences represent a single zone. The writer knows no contradictory evidence and has therefore placed all the Stringocephalus occurrences as equivalent and low in the Givetian.”’ A. W. Grabau (1931) and C. C. Tien (1938) have described Stringocephalus as occurring abundantly in the provinces of Hunan and Yunnan in southern China. Several mutations of Stringocephalus burtini are present, and these are shown by Tien (1938, p. 13) to occur on two horizons, S. burtini at the top of the Chitzechiao Limestone, and S. burtini mut. alpha Grabau and S. burtini mut. beta Tien in the underlying Ichiawan Shale. These beds represent the upper part of the Middle Devonian Series as developed in southern China, and are equivalent to the Givetian of Europe. (Tien, 1938, p. 6.) The associated brachiopods are Meristella spp., Atrypa spp., Athyris aspera and Ambocoelia sinensis. In Australia, the occurrence of Stringocephalus in the Fanning River and Reid Gap Districts, Queensland, proves the existence of rocks of Givetian age, and this is confirmed by the presence of Rugose corals, shown by Dr. Dorothy Hill (1942) to be either identical with or closely related to species in the quadrigeminus and Buchel beds of the Paffrath Basin. The writer’s work in the Tamworth District, New South Wales (Brown, 1942) was originally undertaken with the object of correlating the limestones there with extra-Australian occurrences. The age of the Sanidophyllum fauna in, the Moore Creek limestone has been one of the problems of Devonian stratigraphy ever since the fauna, largely endemic, was described by Etheridge in 1899. The discovery of a clear sequence of fossiliferous limestones in the hills north of Sulcor Quarry established the relative positions of the Murrumbidgee (=Sulcor) and Moore Creek faunas. The identification of Bornhardtina, one of the Stringo- cephalidz, in the lower part of the Moore Creek Limestone, below the Sanido- phyllum zone, now definitely fixes the age of the Brachiopod Limestone as Lower Givetian (—Upper Honsel of Paffrath Basin), and by inference, places the underlying Sulcor Limestone (—Murrumbidgee) in the Couvinian or lower part of the Middle Devonian. The Tamworth sequence is shown diagrammatically in the accompanying columnar section. 3. SYSTEMATIC DESCRIPTIONS. The terminology used in the following descriptions is that of Allan Thomson (1927), Muir-Wood (1936) and Ulrich and Cooper (1938), where applicable. Definitions of the terms are tabulated by Cloud (1942). STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. 121 European Stages Attunga, N.S.W. = Radiolarian Cherts Fame een Moore Creek Limestone Zone of Sanidophyllum davidis, Spongophyllum giganteum, Tamworth Series, Middle Devonian Givetian Moore Creek Stage ae igh wed Disphyllum robustum, Phacellophylium porteri, fe ee i aT Mesophyllum cornubovis, Syringopora porteri, Peat 24 ea bie sa Heliolites porosa. fee ia] 2008 ot ath Stringophyllum bipartitum, Sanidophyllum colligatum. Zone of Bornhardtina coultert , Favistella cf. rhenana,| Amphipora sp. Cherts and Tuffs Couvinian Sulcor Stage Sulcor Limestone Eridophyllum bartrumi , Receptaculites australis , ) Phillipsastraa maculosa A Prismatophyilum browne , Pseudamplexus princeps, Trapezophylium couliert, Xystriphyllum mitchelii, Syringopora spéleana, Syringopora spp., Stromatoporelia loomberensts Ht Lower Devonian Nemingha Stage Cherts and Tutis with Coblenzian band of limestone-breccia. Nemingha Limestone Acaathophylium cf. mansfieldense, Xystriphyllum insigne, Lyriclasma floriforme, Iitophyllum konineki, Syringopora spp. Cherts and Tuffs Ged innian Fig. 1.—Diagrammatic columnar section of the Tamworth Series near Attunga, N.S.W., to show the position of the Bornhardtina zone. Vertical scale approximately 100 ft.=1 in. 122 IDA A. BROWN. Class BRACHIOPODA Duméril. Order TELOTREMATA Beecher, 1891. Super-family Terebratulacea Waagen, 1883. Family Stringocephalidz King, 1850. 1850. King, W., Paleontog. Soc. Monogr., The Permian Fossils of England, p. 141. 1942. Cloud, P. E., Geol. Soc. Amer., Special Paper, No. 38, p. 91. ‘¢ Large, thick shelled, generally unornamented terebratuloids with discrete hinge plates and a long marginal loop with modified crural points. Dental plates obsolete or obsolescent. Median septa and cardinal process present or absent. Deltidial plates discrete or conjunct. Foramen hypothyrid to permesothyrid. “Geologic Range: Middle Devonian of the Northern Hemisphere. This family is so characteristic of the Middle Devonian that the presence of any of its members is presumptive evidence of that age for the beds involved.”’ (Cloud, ip.) Ou:) Sub-family Stringocephalinze Cloud, 1942. Genus Stringocephalus Defrance, 1825. 1824. Defrance, Tabl. Corps Org. Foss., p. 110 (no description). 1825. Defrance, in Blainville, Manuel de malacologie et de donchyliologie, text (1825), p. 511; atlas (1827), Pl. 53, f. 1-1c. 1827. Defrance, Dict. Sci. Nat., Vol. 51, p. 102, pl.,75. 1842. Sandberger, Leonhard und Bronn’s Jahrb. ftir Mineralogie, p. 386. 1942. Cloud, P. E., Geol. Soc. America, Special Paper, No. 38, p. 104 and refs. For further discussion of the genus see Davison (1853, 1864), Hall and Clarke (1893), Torley (1908), R. Wederkind (1917, 1925), A. W. Grabau (1931), T. H. Ting (1936) and C. C. Tien (1938). Genotype (by original designation): Terebratula burtini=Stringocephale burtint Defrance, 1825. Diagnosis (after Cloud, 1942): ‘* Large, subglobular to transversely sub- lenticular Stringocephalide. Beak prominent, deltidial plates conjunct in adults, foramen hypothyrid, ventral interarea or planareas present and ordinarily well developed. Having prominent median septa in both valves and a long rod-like, terminally bifid cardinal process in the dorsal valve. Hinge plates discrete, not supported by crural plates.’’ Geological Range: Upper Middle Devonian of the Northern Hemisphere. Stringocephalus burtini Defrance, 1825. Plate IV, Figs. 1-6, text-figure 1. Holotype (Chosen Cloud, 1942, p. 108): Specimen represented by Fig. la in Plate 53 in original illustrations by Defrance (in Blainville, 1827, loc. cit.). Refigured by Cloud, 1942, Plate 18, fig. 6. Locality: Paffrath, Germany. Diagnosis: As for the genus. Australian Material and its Preparation: The fossils from Queensland are replaced by white, crystalline calcite and are embedded in massive, black lime- stone, from which it is extremely difficult to obtain complete specimens. One specimen (Q.U. Coll. F.6977) from Ryan’s Quarry, Calcium, Reid Gap (Plate IV, figs. 1-3) was naturally weathered to reveal an internal mould showing the pallial markings and septa. Both beaks were preserved as replacements in crystalline calcite and black limestone matrix occurred between the beaks. STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. 123 Careful removal of the matrix by means of an automatic percussion chisel revealed on the ventral valve the interarea, the concave deltidial plates and the foramen. Another incomplete specimen (Q.U. Coll. F.6976) embedded in black limestone from the same locality was sliced vertically along the median plane of symmetry; one-half of this specimen was ground parallel to this plane, and a series of camera lucida drawings is reproduced in text-figure 2: a wax scale- model, constructed from these drawings in the manner described by St. Joseph (1938) (Plate IV, figs. 4-5) revealed several important internal characters. Description: Exterior. Shell large, sub-globose ; lateral profile bi-convex ; immature specimens are relatively long, but older specimens are as wide as long ; measurements of three specimens from Queensland are compared with those of typical specimens from Paffrath, Germany, and from China in the accompanying table. Measurements of Stringocephalus burtint (in mm.). Column ss a a, I. a III. IV. Vie VI. VII. Height of ventral valve .. SOMO mcs ee 67 690"), 69-5 40 85 Height of dorsal valve .. 65 | 56 61 ayeien | 55:5 30 72 Width zi ae ui 72 83 62 61-0 74-7 35 80 Thickness .. ar oe 60 | 465 47 43-0 — 25 56 Umbonal angle V.V. a =) as 100° 113° 103° — — Umbonal angle D.V. ops — | 160° 160° HOOK) Lobe — — Length of hinge line a — 36 22 USriols in roll — —- I. Queensland Univ. Coll. F.6976, sectioned specimen. Loc. Ryan’s Quarry, Por. 62v, Par. Wyoming, Q. (Calcium, Reid Gap.) II. Queensland Univ. Coll. F.6977, specimen figured Pl. A, figs. 1-3. Loc. as for I. III. Queensland Univ. Coll. F.6978. Loc. as for I. IV. Topotype in Sydney Univ. Coll. 2480. Loc. Paffrath, Germany. V. Mansuy’s Specimen, quoted Grabau, 1931, p. 525. Loc. Yunnan, China. VI. Original type of Defrance (Chosen Cloud, 1942). Approximate measurements of immature specimen, Paffrath, Germany. (Cloud, p. 108.) VII. Dimensions of average adult, quoted Cloud, p. 108. Anterior commissure rectimarginate; cardinal margin submegathyrid, hinge slightly less than half the width of the shell. Beak of the ventral valve high, erect in immature specimens, incurved in older specimens, with an umbonal angle of 115 degrees. The ventral interarea is relatively high, being about 15 mm. high and about 50 mm. wide, orthocline to anacline, dividéd medially into two triangular regions by concave, conjunct deltidial plates, which together are 30 mm. wide along the hinge-line. The foramen is marginally hypothyrid in position (Plate IV, fig. 1). The beak of the dorsal valve is inconspicuous ; the dorsal palintropes are not visible from the outside but appear in the scale model aS narrow areas, radiating from the beak. Surface smooth, without trace of fold or sulcus. Shell substance thick posteriorly (ca. 10 mm.), becoming thinner anteriorly ; the structure of the shell is obscured by recrystallisation. Interior: In the ventral valve dental lamelle are absent and the hinge- teeth, about 40 mm. apart, are free, short and curved upwards and inwards in the manner characteristic of the genus and of the Terebratulacea generally. A double median septum is present (Q.U. Coll. F.6978) extending two-thirds of the distance from the beak to the anterior margin. Muscle-scars (?) and radially arranged vascular markings appear in the cast of specimen Q.U. Coll. F.6977. (Plate IV, fig. 1.) 124 IDA A. BROWN. 50 mm oat ee eata ] cae ary ATARSE een. ~—< ~~ Fig. 2.—Serial sections through Stringocephalus burtina from Calcium, Reid Gap, Queensland. Natural size. i [ Explanatory note continued on next page.] | ; STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. 125 In the dorsal valve there is a shallow, broad, median septum. The hinge- sockets receive the teeth in their deepest parts, and the serial sections suggest the presence of an accessory socket for the reception of a denticulum, as in more recent Terebratuloids (Muir-Wood, 1934, p. 518). Inner socket-ridges are well-defined, and are supported by the hinge-plates, which curve downwards and backwards towards the floor of the valve. The cardinal process is very prominent, being about 2-5 cm. in length, and is bifid. The direction appears to vary somewhat in different specimens; in one sectioned specimen it is directed at right-angles to the plane of the com- missure, in others it is directed more anteriorly. The base is much thickened by secondary shelly material and almost fills the space between the crural bases. The cardinal process branches into two wedge-shaped apophyses, which ‘go towards the sides of the septum of the ventral valve, the flat edge of the wedge being parallel to the hinge-line. Traces of the brachial loop may be seen in a few specimens, but unfortunately the material available is too poor for this structure to be properly studied. - The adductor muscle-scars are large and well separated, and there is a broad shallow median septum, which in the anterior region is bounded by thinner parts of the shell covering the vascular sinuses (Plate IV, fig. 1). Remarks: The genus Stringocephalus Defrance resembles Bornhardtina Schulz externally, but Bornhardtina has no dorsal or ventral median septa and no large cardinal process (Ting, 1936). Three species of Stringocephalus from Europe and two species from China are recognised as valid by Cloud (1942). Of these, the Queensland specimens most closely resemble the genotype, S. burtina, and on the available evidence I can find no reason for distinguishing them from this European species. Sub-family Bornhardtininz Cloud, 1942. Genus Bornhardtina Schulz, 1914. 1914. Schulz, E., Naturh. Vereins preuss. Rheinlande und Westphalens, Verhandl., Jahrg. DO (1923), pp. 363-366; Pl. 7, f. 6; Pl. 8, f. 1-3. 1936. Ting, T. H., Geol. Soc. China, Bull., Vol. 15, No. 3, pp. 343-359. 1942. Cloud, P. E., Geol. Soc. Amer. Special Paper No. 38, pp. 100-104. Genotype (designated Cloud, 1942): Bornhardtina uncitoides Schulz, 1914. Diagnosis (after Cloud, p. 101): ‘‘ Characteristically asymmetrical Stringo- cephalidz with deltidial plates conjunct at maturity, foramen hypothyrid, beak large and conspicuous, delthyrium occupying most of the posterior border. ‘‘In the ventral valve the dental plates are obsolete and the muscle impressions relatively weak. ‘‘In the dorsal valve the ventrally convex hinge plates are discrete and crural plates are absent. The thickening of the posterior border is ordinarily not pronounced but in some shells it is conspicuous. Full details of the loop structure are not known but such information as is available is consistent with the assignment to the Stringocephalide.”’ Geological Range: Middle Devonian of Europe, generally coexistent with Stringocephalus. [ Continuation of explanatory note to Fig. 2.] Distances of sections from the median vertical plane (in millimetres) : Bae, | 2— No oO, 3-12-10) 4-11-60, 510-70, 6==9-75, Toe; o— Salone 9) 740, 110 6-85; Li 6-10; 12—5+30, e400, 4 — 4430) 15— 13-60, ) b6—- 3-15, Lsj— 1-80, 18=-0-0. V=ventral valve, D=—dorsal valve, a=interarea, s=septum, d=deltidial plate, f=foramen, t—tooth, hp=—hinge plate, cp=cardinal process, cb=crural base. 126 . IDA A. BROWN. Bornhardtina coulteri 0.Sp. Plate IV, fig. 7; Plate V, figs. 1-5. Holotype: Sydney Univ. Coll. 7569. Loc. north-east corner of Por. 115, Par. Burdekin, 4 miles north of Attunga, N.S.W. Horizon: 100 feet above the base of the Moore Creek Limestone, upper Middle Devonian (Givetian). Diagnosis: Large, symmetrical Bornhardtina. The black limestone contains abundant remains of shattered shells, but so far no complete specimen has been isolated from the rock. Plate V, fig. 1 shows the typical appearance of the rock. The specimen chosen as holotype is unfortunately incomplete, but the essential features are unmistakable. Nearly twenty specimens were sliced in attempting to work out the internal characters. Haterior : The shells are large and globose, similar in shape to Stringocephalus burtini, the valves being usually symmetrical. Specimens, which I take to be- immature shells, closely resemble the illustrations of the genoholotype (Schulz, 1914, Taf. VIII, figs. 1-3), and have a relatively high, erect beak in the ventral valve. These specimens show also very fine concentric and radiating ornamenta- tion of the surface of the shell. In the larger, mature specimens the ventral beaks are very much incurved, and overhang the posterior portion of the dorsal valve (Plate V, fig. 5; text-fig. 34). The dimensions of the shell are approxi- mately as follows: Height of ventral valve: 70 mm. Height of dorsal valve: 55 mm. Width: 60 mm. Thickness: 55 mm. Distance between teeth: 30 mm. The thickness of the shelly material in the region of the umbones is about 4 mm., but it is very much less than this in the anterior region, so that usually the lower part of the shellis crushed. The shell is fibrous and distinctly punctate. The cardinal margin is subterebratulid. Deltidial plates are large, apparently rather thin in some specimens, and form an outwardly concave surface (text- fig. 3B). The foramen is medially hypothyrid, and a pedicle-collar is formed by the infolding of the deltidial plates around the former position of the peduncle. (See text-fig. 3B and Plate V, fig. 2.) This structure has been observed by Jackson (1916, p. 24), Thomson (1927, p. 75), Muir-Wood (1935, p. 521) and others in Mesozoic and Cainozoic Terebratulids, but does not seem to have been noticed hitherto in Palzozoic shells. Ventral palintropes are very narrow. Beak of the dorsal valve gently convex and almost covered by the ventral umbo. There is no trace of a fold or sulcus; anterior commissure probably rectimarginate. In the ventral valve there is no trace of a septum and there are no dental plates. The teeth are typically terebratuloid and about 3 cm. apart; sections (cf. text-fig. 344) indicate the presence of denticula. The muscle scars are relatively large and vertically striated (Plate IV, fig. 7; Plate V, fig. 3). In the dorsal valve there is no cardinal process, but there is a very small, concave area of muscle attachment immediately below the beak. The hinge plates are discrete and there are no crural plates. The crural bases are situated between the hinge plates, and support the delicate crura, which are about 3 mm. apart. Several specimens reveal crura as shown in text-fig. 3A, extending anteriorly and ventrally for about 1-5 cm., but the further extension of the loop is not known; a few specimens showed curved shelly plates lying within the valves, suggesting a structure comparable with the cross-plates of the loop of Enantiosphen (Cloud, 1942, Plate 26, fig. 6), but no organic connection with the valves could be proved. The muscle scars of the dorsal valve are not known. STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. 127 Fig. 3.—Sections through Bornhardtina coulteri n.sp. from limestone north of Sulcor, near Attunga, N.S.W. Natural size. Fig. A.—Restored median vertical section showing beaks of ventral and dorsal valves, deltidial plate and crus. The specimen was sliced in the directions indicated 1-12. (x1.) Figs. 1-6, 11, 12 illustrate the appearance along the corresponding directions indicated in Fig. A. V=ventral valve, D=dorsal valve, hp=—hinge-plate, cb=crural-base, cr—crura. Fig. B.—Sections through the umbo of ventral valve, normal to the plane of symmetry. Distances from the tip of beak: 14-3 mm., 28-6 mm., 3=11-:4 mm., 4—14-0 mm. dp=deltidial plate, f=-foramen, pc=pedicle collar, uc==umbonal cavity. (x 1.) 128 IDA A. BROWN. Remarks : Bornhardtina is distinguished from Stringocephalus by the absence of septa and a cardinal process. It appears to be related to Rensselandia, one or two species of which it may resemble in external shape, but the deltidial plates are conjunct and the form of the brachial support is distinctive. In Rensselandia the crura are short, divergent, and widely separated, and the loop follows the margin of the shell, whereas the crura of Bornhardtina uncitoides (figured Cloud, p- 103) are long, parallel and close together. The specimens under consideration thus appear to belong to the genus Bornhardtina. As already stated (p. 126) certain small specimens from Attunga resemble the outward form of the geno- type, B. uncitoides, and it may be that this species is present, but, in the absence of comparative material, it seems preferable to erect a new species, Bornhardtina coultert. 4. BIBLIOGRAPHY. Benson, W. N., 1922. Materials for the Study of the Devonian Paleontology of Australia. Rec. geol. Surv. N.S.W., 10 (IL), 178. Brown, Ida A., 1942. The Tamworth Series (Lower and Middle Devonian) near Attunga, N.S.W. J. Roy. Soe NiS:W.,.. 76," 165: Cloud, P. E., 1942. Terebratuloid Brachiopoda of the Silurian and Devonian. Geol. Soc. Amer., Special Paper, 38. Cooper, G. A., et al., 1942. Correlation of the Devonian Sedimentary Formations of North America. Bull. geol. Soc. Amer., 53, 1784. Davidson, T., 1853. British Fossil Brachiopoda. Palaeontogr. Soc. Monogr., 73-76. — 1864. British Fossil Brachiopoda. Palaeontogr. Soc. Monogr., 11. Dun, W. S., 1900. A.R. Dept. Mines, N.S.W., 195. Etheridge, R., Junr., 1892. Geology and Paleontology of Queensland and New Guinea. 67. Govt. Printer, Brisbane; Dulau & Co., London. 1899. On the Corals of the Tamworth District, etc. Rec. geol. Surv. N.S W50(8), T5ks Grabau, A. W., 1931. Devonian Brachiopoda of China (Yunnan). Geol. Surv. China, Palaeontol. Sica, BL, 3, 222. Hall, J., and Clarke, J. M., 1893. An Introduction to the Study of the Genera of the Palzozoic Brachiopoda. Palaeont. New York, 8 (2), 282. Hill, Dorothy, 1942. The Middle Devonian Rugose Corals of Queensland. Proc. Roy. Soc. Queensland, 53 (14), 229. Jackson, J. W., 1916. Brachiopod Morphology. Geol. Mag. Lond., 3, 24. Kirk, E., 1927. New American Occurrences of Stringocephalus. Amer. J. Sci., (5), 13, 219-222. Muir-Wood, Helen M., 1934. On the Internal Structures of Some Mesozoic Brachiopoda. Philos. Trans. Roy. Soc. Lond., B, 223, 516. —_——_—_—___—__-—_—-— 1936. Brachiopoda of the British Great Oolite Series. Palaeontogr. Soc. Monogr. 5. Schulz, E., 1914. Ueber einige Leitfossilien der Stringocephalenschichten der Eifel. Naturh. ver. preuss. Rheinlande u. Westphalens. Verh. Jahrg. 70, 363. St. Joseph, J. K. 8., 1938. The Pentameracea of the Oslo Region. Norsk. geologisk tideskrift, 17, 225. Thomson, J. Allan, 1927. Brachiopod Morphology and Genera. New Zealand Board of Studies, Wellington, N.Z. Tien, C. C., 1938. Devonian Brachiopoda of Hunan. Nat. Geol. Surv. China, Palaeontol. Sinica, n.s., B, 4, 114. Tilmann, N., e¢ al., 1938. Geology of the Rhenish Schiefergebirge. Proc. Geol. Assoc. Lond., 49, 1. Ting, T. H., 1936. Zur Kenntnis der Gattungen Bornhardtina Schulz und Stringocephalus Defrance, Bull. geol. Soc. China, 15 (3), 343. Torley, K., 1908. Die Fauna des Schleddenhofes bei Iserlohn, Abh. d. Kéniglich Preuss. Geolog. Landesanst., N.F., H.53, 10. i Ulrich, E. O., and Cooper, G. A., 1938. Ozarkian and Canadian Brachiopoda. Geol. Soc. Amer., Special Paper, 13. Wederkind, R., 1917. Ueber Stringocephalus burtint und verwandte Foramen, Nachr. K. Gesellschaft d. Wissensch. zu Géttingen. Mathem.-physik. Klasse. ——_—_—_——-- 1925. Das Mitteldevon der Eifel. Teil II. Schriften ges. Beférd. ges. Naturw. Marburg. Whiteaves, J. F., 1891. Devonian Fossils of Manitoba. Trans. Roy. Soc. Canada, 7 (for 1890), 93-110. Journal Royal Society of N.S.W., Vol. LXXVII, 1943, Plate IV =- Ora Journal Royal Society of N.S.W., Vol. EXXVII, 1943, Plate V STRINGOCEPHALID BRACHIOPODA IN EASTERN AUSTRALIA. 129 EXPLANATION OF PLATES. Pirate IV. Photo. I.A.B. Figs. 1-3.—Stringocephalus burtini Defrance. Queensland Univ. Coll. F.6977. Loc. Ryan’s Quarry, Por. 62v, Par. Wyoming, Calcium, Reid Gap, Queensland. (xX 1.) Fig. 1.—View showing pallial markings in dorsal valve, interarea, deltidial plates and foramen in ventral valve. Fig. 2.—Side view showing both valves (internal mould). Fig. 3.—View of ventral valve showing median septum. Figs. 4-5.—Scale model of the umbonal region of dorsal valve of Stringocephalus burtini based on serial sections of specimen Queensland Univ. Coll. F.6976. (x 1.) Fig. 4.—Posterior view, showing large, bifid cardinal process and dental sockets. Fig. 5.—Oblique lateral view, showing cardinal process, dental sockets, hinge plate and crural bases. Fig. 6.—Median vertical section through Stringocephalus burtini specimen Queensland Univ. Coll. F.6976, showing foramen in ventral valve (6a) and cardinal process in dorsal valve (6b). Fig. 7.—Bornhardtina coulterit n.sp. Sydney Univ. Coll. Loc. north-east part of Por. 115, Par. Burdekin, Attunga District, N.S.W., showing muscle scars in the ventral valve. (x 1.) PuaTE V. Photo H. G. Gooch. Bornhardtina coulteri n.sp. All figured specimens are from north-east corner of Por. 115, Par. Burdekin, north of Attunga, New South Wales. Fig. 1.—Slab of limestone showing typical preservation of shells. Specimen in upper right- hand corner shows trace of crus (cr.) below the beak of the dorsal valve. (x 1 approx.) Fig. 2.—Vertical section through beak of ventral valve, showing foramen (f.) and trace of pedicle-collar (p.c.). Figs. 3-5.—Bornhardtina coulteri n.sp. Holotype. Syd. Univ. Coll. No. 7569. (x1 approx.) Fig. 3.—View of ventral valve. Shelly material worn away from part of right-hand side, exposing impression of muscle scar. Fig. 4.—Side view of same specimen, showing worn beak of ventral valve and section of dorsal valve. Fig. 5.—Vertical median section of same specimen, showing overhanging beak of ventral valve, with trace of deltidial plate (d.p.) and dorsal valve with trace of descending crus (cr.). Note absence of cardinal process. a K—October 6, 1943. hw rge'e ae a Tid tye fie oN a. ed fs, a ‘ it : ra ; Ph i “~*y DY wt eee Bd + aad ons Large 10 Demme ot eee it {i a i wed” Teas. ‘7 reas 2 5 a"F 3 awed iTOA CA these) 4) _ 50 j = ry at ; 1; ‘ Y . f ' 1 t : = x = 7 k - j 4 OF THE FOR | | 1943 (INCORPORATED 1881) PART IV (pp. 130 to 175) | ‘VOL. LXXVIL EDITED BY oH E HONO RARY SECRETARI ES hs ‘THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE re By: SETS. MADE AND THE OPINIONS EXPRESSED THEREIN xt j nS Re a Tessa es gypnry ~ ‘PUBLISHED BY THE SOCIETY, SCIENCE HOUSE SOCIETY -—— ‘CONTENTS. <* . = - Part iV oe oo ART. XV. —The Vibrations of Square Molecules. Part: e The Nereal Coon Vibration Frequencies of Planar AB, ‘Molecules. By Allan Maccoll, M _ June 23, 1944) Nate mi oa ssa iene ial URGES ane ; Ae ft Arr. XVI. —Further Deiornnenons of Specialisation i in Flax Rust’ Cue aoe Lini (Pers. ) Lév. By W. L. Waterhouse, M.Sc., D. Se.Agr., D.I. C., F.L 8. Watson, Ph. ie B. Se. Agr. (Issued inte 23, 1944)... .: . “Ann, XVII.—A Sindy of the Maguotie Ree e bf Genes Containing the % oe Metals. aes IB ewes 1 Mellor, M. Se. (Issued ape 23, Des) BY ain A, fe Ant. “XVIII. —The Geology of the ocak District, N. 8. W.. ‘Part II. The Country Mat Bunyan and Colinton. By W.. R. Browne, D.Se. — Meoake June 23 , 1944). Osiruary Notices ... Re eee Mend ee BE ss ee Bhs Tite Pace, Conrents, Notices, PUBLICATIONS .. °° .. ... )Orpicnrs FOR 1043-44 63m ge ee ee Sey ae - List oF MemBers, Awarps oF MEpALsS, Etc. 7 Anstmaor or PRocerpines <04/0 6 yc Get a es ee ce ee * - ~ Ska ‘PRocEEDINGS OF THE SECTION OF GEOLOGY INDEX TO Votume LXXVII- | < Journal and Proceedings of the Royal Society of New South Wales VOLUME LXXVII PART IV K—November 3, 1943. THE VIBRATIONS OF SQUARE MOLECULES. ParT [. THE NORMAL COORDINATES AND VIBRATION FREQUENCIES OF PLANAR AB, MOLECULES. By ALLAN MACCOLL, M.Sc. Manuscript received, October 14, 1943. Read, November 3, 19438. INTRODUCTION. One of the final stages in the assignment of a structure to a molecule is the calculation of the force constants of the normal vibrations in terms of the observed Raman and Infra-red frequencies. That these constants are invariable for molecules containing similar bonds has been shown by Glockler and Wall (1937), Crawford and Brinkley (1941) and others. Two procedures may be adopted. First the force constants may be taken from similar molecules and the frequencies of the given molecule calculated. Secondly, the force constants can be determined from the vibration frequencies, and since to a valence force approximation for symmetrical molecules there are usually more frequencies than force constants, relationships of self consistency can be examined. Agreement in either case provides a verification of the structure. In the case of unsymmetrical molecules the calculation of the frequencies may be very tedious, as an n-atomic molecule has 3n—6 normal vibrations, whose frequencies are given as the roots of an algebraic equation of the (3n—6)th degree. However, Wigner (1930) has shown how the theory of groups may be employed to factorise the secular equation in the case of symmetrical molecules. In this case there are certain symmetry operations which transform the molecule into itself. These symmetry operations form a group. The kinetic (T) and potential (V) energies expressed in terms of the mass-reduced normal coordinates are 3n— 6 oT— > Q? a wg eee (1) Vi oO a where );=47?y;2, y, being the frequency. Since T and V are invariant under the operations of the group, a restriction is placed on the form of the normal coordinates. If R is an operation of the group, then RQ:=Q:i symmetric \ : : RQ: = —Q; asymmetric gan degenerate coordinate. nl RQi:= X RijQ; degenerate coordinate. jal J= n is the degree of degeneracy, that is the number of vibrations having the same frequency. Every group has a certain number of irreducible representations each of which represents a type of symmetry permitted to the normal coordinates. The latter will in general form a reducible representation of the group and by THE VIBRATIONS OF SQUARE MOLECULES. 131 reduction of this it is possible to determine the number of vibrations belonging to each symmetry type. The theory of groups can also be used to predict which of the irreducible representations of the symmetry group of the molecule will be active in the Raman or Infra-red spectrum. The depolarisation factor of the Raman lines can also be calculated. These considerations are of the utmost importance in correlating the observed frequencies with the normal vibrations. Early work in this field was carried out by Duncan and Murray (1934), who investigated the normal vibrations of AB, molecules in connection with the Raman spectrum of nickel carbonyl, using Andrews’ mechanical method (1930, 1934). Wilson (1935) has given the relationships between the frequencies and force constants to the valence force approximation, while Kohlrausch (1938) considered a more general model to the same approximation. Bernstein (1938) calculated the frequencies of the vibrations symmetrical to the fourfold axis. In connection with his investigation of the structure of inorganic com- plexes, Mathieu (1939) has described the normal vibrations of AB, molecules. However, no attempt has been made to determine the frequencies in terms of the force constants of the general potential function. Since six frequencies are obtainable from the Raman and Infra-red spectra of this type of molecule, the valence force model which uses only three constants is rather restrictive. THE SYMMETRY COORDINATES. Planar AB, molecules possess the symmetry D,n. The classes of symmetry operations of the group D,n are: (a) The identity operation E. (6) Twofold rotations around the fourfold axis C,. (c) Fourfold rotations around the fourfold axis C,. (d) Twofold rotations around the x and y axes C’,. (e) Twofold rotations around the diagonal axes O©",. (f) Inversion in the centre of symmetry I. (g) Fourfold rotations around the rotation-reflection axis 8,. (h) Reflection in the xy plane op. (t) Reflections in the xz and yz planes o’y. (7) Reflections in the diagonal planes o’y. These symmetry elements are shown in Fig. 1. Ce Cp Fig. 1.—Coordinate axes of the AB, moiecule. 132 ALLAN MACCOLL. Knowing the characters of the irreducible representations of D,n and using the method of Wigner (zbid.), Wilson (1934) and Rosenthal and Murphy (1936), the reducible representation of the molecular vibrations can be expressed in terms of the irreducible representations of Dyp. D=A,e+AoutBieg+Biu+Bogt2Eu ....-.......04.0.. (4) In (4) A and B refer to symmetry and asymmetry of the irreducible repre- sentations with respect to C,, E represents a doubly degenerate irreducible representation, while g and u refer to symmetry and asymmetry with respect to inversion in the centre. Sufficient information to determine the normal vibrations is given in Table I (cf. Wilson (1934a)). The symmetry type of each symmetry coordinate is set out together with the number of normal vibrations of that type and their properties with respect to the Raman and Infra-red spectra. TABLE 1. Symmetry with respect to: Active in No. of Polar- Class. Vibs. Coord. | | isation. E. Cy. Cy. OKA: C",. 1p Oh- we O'v: O’y. Raman. I R Aig 1 Q: ae ae tt See eee Wola We se ih ae lh ae eh oats Yes No. 18 Aou 1 Q. + + + = x rir ie: —P Si + No Yes bans Big af Q; + ++ ae i st == aoe Peay. ren + Yes No. D Byu al Q, Te = ere = = FF a =i + aa No No. imag Bog 1 Q; ap hare eam se ose pose oS ae boss Yes No. D aaa) Sn a ca i ls Pals AR EY IR a iba Eu 2 Sea, Sra | Seb, Seb Transform like x + iy. No. Yes. — In Table I a +ve sign indicates that the symmetry coordinate considered transforms into itself under the given operation, while a —ve sign indicates that it transforms into its negative. The symmetry coordinates corresponding to the degenerate irreducible representation transform under the operations of the group like translations along the x and y axes. It can be shown that the symmetry coordinates -have the following properties : on—6. (a) ig Meena 2) Si? i=1 Ree. i (5) (b) 2h DijOi9}, with ajj—aAji i, j=1 aij 1s zero unless S; and §S; belong to the same non-degenerate repre- sentation of the group or unless they transform identically under all the operations of the group in the case of a degenerate irreducible representation. From these considerations it can be shown that the secular equation is factored as follows:: (a) If there is only one symmetry coordinate belonging to a given non- degenerate irreducible representation, it will be given as the root of a linear equation and will be identical with the normal coordinate. THE VIBRATIONS OF SQUARE MOLECULES. 113335 (b) If there are n symmetry coordinates belonging to a non-degenerate irreducible representation, then the frequencies will be given as the roots of an algebraic equation of the nth degree. The n normal coordinates will be given by linear combinations of the n symmetry coordinates. (c) If there are n symmetry coordinates belonging to an m-fold degenerate irreducible representation, then the equation of the nth degree will be repeated m times. So with the information shown in Table I five symmetry coordinates (Q,-Q;) can be set up which are also normal coordinates and two sets Sga, Sra ; Seb, Szb, linear combinations of which give the corresponding normal coordinates. These are shown in Fig. 2. t f a = ad @ <«e® e+ @ +e A ® ? e+ © +e e> @® <@ mane @ x e—> e eo ] eo i <—@ Soa Séb Ey S7a Sop Fig. 2.—Vibrational symmetry coordinates of the AB, molecule. KINETIC AND POTENTIAL ENERGIES AND VIBRATION FREQUENCIES. The kinetic and potential energies in this system of coordinates can be obtained by the method of Wilson (19346) also used by Silver and Shaffer (1941) in their study of planar AB, molecules. (covalent com- plexes). Owing to the pairing of electrons which takes place in the formation of a covalent bond there is a profound difference in the magnetic moments of Fel and Co! in each of the two types of complexes. In the course of developing 1 In the iron series the magnetic effect of orbital angular momentum is blotted out by the action of asymmetric electric fields of surrounding atoms with the result that the magnetic moment is given by the formula us—4/S(S+1), where S=the spin quantum number (ds). A full discussion and the first satisfactory explanation of this quenching phenomenon has been given by Van Vleck (1932). 2 Explanations alternative to Pauling’s are (1) the crystalline potential theory of Penny and Schlapp, (2) Mulliken’s theory of molecular orbitals. Van Vleck (1935) has demonstrated ‘‘ the similarity of the predictions with all three theories ’’. M—December 1, 1943. 146 D. P. MELLOR. his theory of the nature of the chemical bond Pauling predicted the magnetic moments of atoms of the Fe, Pd and Pt transition series as set out in Table I. TABLE I. Predicted Magnetic Moments of Complexes Containing Transition Elements. No. of |For ionic electrons| or sp* For 4 For 6 The Iron The Palladium The Platinum in d shell] (tetra- dsp? d’sp*. Series. Series. Series. (3d, 4d | hedral) | (square) | (octahedral) or 5d). | bonds. | bonds. bonds. KiCallgciNTilv RbISrHYUIZrIVNbVMoVI | CslBal—HfIVTaVWVI 0 0:00 0-00 0:00 viv NbIVMoV wy 1 1-73 1:73 1-73 villcriv MolVRuVI WIVOsvVI 2 2°83 2°83 2°83 VilcrHIMniv Molll 3 3°88 3°88 3°88 CrilIMniliFelV Moll Rulv OsIV 4 4-90 4-90 2-83 MnUFelIColV Rulll OsllI[rIV 5 5-91 3°88 1:73 FellColll RullRhUIPdiv TrllIptiv 6 4-90 2°83 0-00 ColINiIIl Rhu rll 7 3-88 | 1:73 er Nill RhiPpdtAgiI Pt Aull 8 2°83 0-00 This table is taken from the paper by L. Pauling and M. L. Huggins, Zeit. ftir Krist., 87, 1934, 214. It will be seen that the moments (given in Bohr magnetons) depend on (1) The valence or oxidation state of the metal atom. (2) The nature of the bonds linking other atoms or groups to the metal (ionic or covalent). (3) The number and configuration of these bonds (when they happen to be covalent). The iron series has been extensively examined in the light of Table I (Pauling, 1940) and from the experimental results it is clear that when allowance is made for small orbital contributions due to incomplete quenching in atoms not in S states, there is reasonably good agreement between theory and experi- ment for both ionic and covalent complexes. It is a matter of some interest to enquire how far the predictions embodied in Table I hold good for the heavier and more complex atoms of the two later transition series. So far as the author is aware, no systematic study, such as is attempted here, has yet been made for the purpose of testing the extent of the validity of Pauling’s predictions for the Pd and Pt series. It is true that a good many measurements on compounds of the platinum metals have already been reported but there are inadequacies in the data in the sense that a number of valence states have not received attention and a number of investigations have been largely confined to the sometimes poorly defined anhydrous chlorides whose magnetic dilution is low and whose identity and purity may be in doubt since analytical data were in these instances, not always reported. From the experimental data already available, more especially that of Bose and Bhar (1928), Guthrie and Bourland (1931), Christiansen and Asmussen (1934, 1935), Janes (1935) and Cabrera and Duperier (1939) and the present work it is clear that the platinum metals do not form ionic complexes,’ a fact which is 3 It is possible that fluorides like RhF,, PdF, and PdF,, etc., may prove exceptions to this statement. The crystal structure of PdF, (Ebert, 1931) does indeed strongly suggest that Pd-F bonds are ionic. The author’s attempts to prepare PdF, without the use of elementary fluorine were unsuccessful. : MAGNETIC BEHAVIOUR OF COMPLEXES. 147 responsible for the diamagnetic or generally low paramagnetic susceptibilities of the compounds of these metals. Octahedral (dsp?) complexes of Ptr, Pdtv, Rh™, Rutt and Ir! and square complexes of Pt" and Pd" are in point of fact all diamagnetic, which accords with Pauling’s predictions for complexes of these types. There are, however, instances where theory predicts paramagnetism as for example in the octahedral (d?sp*) complexes of Rulv, Ost, Irv, Rum, Os", Ir, and Rh" (see Table I). It is the purpose of the present communication to report results of measurements on complexes containing metal atoms in the above valence states. As far as practicable, attention has been centred on well defined, magnetically dilute compounds whose structure is known from either physical or chemical investigations. EXPERIMENTAL. Method of Measurement. Susceptibilities were measured by the Gouy method, full details of which have already been given by Sugden (1932). The small tube used was calibrated for fields ranging up to 6,000 gauss by means of CuSO,5H,0 whose relatively small paramagnetic susceptibility was in the neighbour- hood of the susceptibilities of the compounds studied. The following values of the specific susceptibility (x) for CuSO,5H,O have been reported : Neo Ks Ley Wetnpss a. 4 Author. 6-14 19-0 | Sugden (1932). 5+ 82 19-2 | Reekie (1939). 5:95 iO de Haas and Gorter (1930). 5:9 18-9 Feytis (1911). 5°85 | 18-9 Honda and Ishiwara (1915). The average value corrected to 19° C. is 5-92 x 10-8, and this was the value adopted for the present work. Effective magnetic moments were calculated on the assumption that the com- pounds, being magnetically dilute, obey Curie’s Law: verr=2-839\/bmT. This assumption is undoubtedly only an approximation, but it is sufficiently close to the truth to test the validity of the predictions of Table 1.4 All compounds were made from the purest starting materials available and analysed for at least one constituent to check their purity and to ensure their correct identification. Some identifications were also checked by means of X-ray powder photographs. SUBSTANCES STUDIED. COMPOUNDS OF RUTHENIUM. Quadrivalent Ruthenium. Potassium pentachlorohydroxyruthenate (IV) K,[RuCl,OH]. Relatively few compounds of quadrivalent ruthenium have been described, and of these two at least have been the subject of a good deal of investigation and not a little controversy. As a result of the work of Aoyama (1924), Howe (1904, 1927) and Charonnat (1931), the constitution of the brown and black complex chloro salts of ruthenium have been established as K,[RuCl,;OH] and K,[RuCl,] respectively. Of the two, K,[RuCl;OH_] is the more easily prepared, and it forms a convenient starting material for the other chloro salts mentioned here. For the present work K,[RuCl;0H] was made by the 4 It is possible that the deviations from Curie’s law are considerable with the compounds (NH,),OsCl, and (NH,),OsBr,. It is hoped to carry out low temperature measurements on these substances to check this point. MM—December 1, 1948. 148 D. P. MELLOR. method of Antony and Lucchesi (1899) starting with the C.P. metal. The dark brown relatively insoluble crystals were analysed for chlorine; found Cl, 47-1%, 47-05%: calculated for K,[RuCl;OH]: Cl, 47-4%. As a rough check on the oxidation state of ruthenium in this compound, solutions were reduced with 0:0196N SnCl, (in excess) and then back titrated with standard iodine solution in the manner described by Howe (1927). For example, 25 ml. of K,[RuCl,OH] solution required 3-4 ml. of 0-0196N SnCl, for its reduction to the trivalent state, whereas the amount required by theory was 3-5 ml. The reduction of K,{[RuCl,OH] with KI and Ti,(SO,), has been studied by Crowell and Yost (1928) and there can be no doubt that in this compound the ruthenium is truly quadrivalent. The valency of ruthenium in this compound is a matter of some interest in connection with later discussion of its magnetic behaviour which differs so radically from that of the closely related compound K,RuCl,. Potassium Hexachlororuthenate (IV) K,[RuCl,]. The starting material used for the prepara- tion of this compound was the red chloro salt, K,[RuCl;H,O], formed as a result of the reduction of K,[RuCl,OH] by prolonged boiling with alcohol under the conditions given by Charonnat (1931). The red chloro salt was dissolved in a minimum of 10N HCl and the solution saturated with chlorine: The hexachloro salt separated as fine, jet black cubes and octahedra which when crushed on white paper gave a deep wine red smear very similar to that produced by K,[IrCl, ]. The compound was analysed for chlorine; found: Cl, 54:2%; calculated for K,[RuCl,]: Cl, 54-18%. The X-ray powder photograph of K,RuCl, corresponded to a cubic lattice and was similar to the photographs produced by K,[IrCl,] and K,[OsCl, ]. The hexachloro compound readily hydrolysed to the hydroxy compound, and several specimens of the hydroxy compound were prepared in this way. The extraordinary ease with which K,[RuCl,] hydrolyses was noted by Charonnat, who states that even in 3N HCl the con- version to K,{RuCl,OH] is almost complete. Tervalent Ruthenium. Potassium aquopentachlororuthenate (III) K,[RuCl,H,O]. Since Ru"! complexes have already been extensively examined by Gleu and Cuntze (1938), one compound only was studied in the present work, and this mainly for the purpose of checking its constitution. K,[RuCl,H,O] was prepared following the method given by Charonnat (1931). When precipitated in a finely divided condition by means of alcohol, this substance possesses a light pinkish brown colour. The compound was analysed for chlorine ; found: Cl, 47-3%. Calculated for K,[RuCl,;H,O] : Cl, 47:-3%. The anhydrous compound K,[RuCl,] was prepared by heating the aquo compound to a temperature of 300° C. in a stream of dry HCl. (Charonnat, 1931.) RHODIUM COMPOUNDS. Bivalent Rhodium. As Rh! complexes have been fully studied by Christiansen and Asmussen (1934), attention has been confined to Rh", the only other oxidation state in which well defined rhodium complexes appear to exist. Measurements were carried out on specimens described and analysed by Dwyer and Nyholm. They were as follows : RhCl1,SnCl,(AsMePh,),. (Dwyer and Nyholm, 1942a.) Rh(AsMePh,),Cl,. (Dwyer and Nyholm, 1941a.) Rh(AsMePh,);Br,. [Rh(AsMe,Ph),Br,]. (Dwyer and Nyholm, 19420.) [RhPy ,Br]Br. PygH¢|RhBr,Py,]. (Dwyer and Nyholm, 1942c.) The evidence for the state of oxidation of Rh in these complexes rests mainly on the analytical data but partly on their reducing action on silver nitrate solution. This reducing action was of — course absent in the Rh™! analogues which have been described by Dwyer and Nyholm (19410). — From the analyses of numerous compounds there can be little doubt that, in the compounds under discussion, the Rh exists in the bivalent state. Work carried out recently in this laboratory ; on the polarography of Rh complexes clearly demonstrates the existence of Rh'! complexes at_ MAGNETIC BEHAVIOUR OF COMPLEXES. 149 the dropping mercury electrode. ° Although this work does not prove that such compounds can be easily isolated, it does show that they can be formed in solution by the action of moderately powerful reducing agents. OSMIUM COMPOUNDS. Sexavalent Osmium. Potassium osmyloxynitrite trihydrate K,[OsO,(NO,),]3H,O. This substance, first described by Wintrebert (1903) was prepared by the action of saturated KNO, solution on OsO,. The dark olive green needle-shaped crystals were analysed for nitrogen. Found: N, 6:2%; cal- culated for K,[OsO,(NO,),|3H,O: N, 6-1%. Quadrivalent Osmium. Ammonium hexachloro-osmate (IV) (NH,),[OsCl,]. Among the most readily prepared and well defined compounds of Os!V are the hexachloro and hexabromo salts, both of which have been deemed sufficiently stable for atomic weight determinations (Gilchrist, 1932). (NH,),[OsCl, ] and (NH,),[OsBr,] were both prepared by Gilchrist’s method, viz. digestion of OsO, with HC! (or HBr) and subsequent addition of NH,Cl (or NH,Br), except that recrystallzation from weakly acid solutions was avoided owing to the possibility of hydrolysis to hydroxy compounds. Analyses : (I) Found: N, 6:4%, 6-5%; Os, 42:8%. Calculated for (NH,),OsCl,: N, 6-4%; Os, 43°3%. (II) Found: Os, 26-9%. Calculated for (NH,),[OsBr,]: Os, 26:95%.° Osmium in each case was determined by igniting the salt in a stream of pure hydrogen. The identity of (NH,),[OsCl, ] was checked by means of an X-ray powder photograph. K,[OsCl,] was prepared in a similar manner and also by treating K,[OsO,(NO,),] with concentrated HCl (Wintrebert, 1903). No analysis of the potassium salt was made. Tervalent Osmium. Potassium pentanitro-osmate (III) K,[Os(NO,);]. Attempts to prepare this salt by Wintre- bert’s (1903) method were not very successful owing mainly to the fact that K,[Os(NO,);] was invariably accompanied by large amounts of K,[OsO,(NO,),|3H,O. However, approximately 50 mg. of a substance corresponding to Wintrebert’s description of the pentanitro complex were isolated. The amber yellow crystals were very soluble and quite stable in water. They were analysed for nitrogen. Found: N, 13-8%. Calculated for K,[Os(NO,);]: N, 14:0%. The pentanitro complex would seem to be the only” stable Os!" complex so far described. Crowell, Brinton and Evenson (1938) studied K,[OsCl, | in solution but succeeded in isolating only impure specimens of the solid. Bivalent Osmium. Potassium hexacyano-osmate (I1) trihydrate K,[Os(CN),]3H,O. This compound was prepared according to the method given by Krauss and Schrader (1928), viz. by the interaction of K,OsO, and KCN. The colourless crystals (isomorphous with K,[Fe(CN),]3H,O, Dufet, 1895) were analysed for nitrogen. Found: N, 15:2%. Calculated for K,[Os(CN),]3H,O: N, 15-1%. A NOTE ON THE ATOMIC WEIGHT OF OSMIUM. There exists at the present a curious discrepancy between the atomic weight of Os as deter- © “mined chemically by Gilchrist (1932) and that determined physically by Nier (1937). Although the former’s value of 191-6 was accepted from 1935 to 1938, Nier’s value of 190-2 is the one now adopted by the International Committee on Atomic Weights and the one used in the present work. In the past, when chemical and physical measurements have disagreed, subsequent chemical work has harmonized the values. (Cf. Be, B, Sb, etc., Aston, 1942.) This may well prove true with ® Experiments carried out by J. B. Willis at the author’s suggestion. 6 These analyses were made by Dr. G. Burger of the University of Adelaide. ” A dipyridyl complex [OsC],2dipy]Cl . 3H,O is mentioned by Burstall (1936), but no details have been published. 150 D. P. MELLOR. regard to osmium. There are remarkably close analogies between the chemistry of osmium and that of ruthenium (Wintrebert, 1903). As already mentioned, K,[RuCl,] shows an extra- ordinary tendency to undergo hydrolysis to K,[RuCl,OH] (Charonnat, 1931) which occurs even in 3N HCl. It seems very probable that Gilchrist’s high value of 191-6 may be due to the contamination of the hexachloro (and hexabromo) salts with the corresponding hydroxy com- pounds, especially when it is recalled that the salts were purified by recrystallization from solutions containing as little as 7% HCl (or HBr). The author has noted subtle differences in the colour of various specimens of K,OsCl, which did not seem to depend on the state of subdivision of the specimen but which may well have been due to the presence of the hydroxy salt, in amounts too small to be easily detected by analysis. COMPOUNDS OF IRIDIUM. Quadrivalent Iridium. Ammonium hexachloro-iridate (IV) (NH,).[IrCl,]. Very few, if any, complex cations con- taining Ir are reported in the literature ; on the other hand complex anions are well known and easily prepared in the form of the hexachloro salts. The ammonium salt was prepared simply by adding NH,Cl to H,IrCl,, the latter having been prepared from iridium which had been purified according to hydrolytic method of Gilchrist (1932). Ammonium chloro-iridate (IV) crystallizes in jet black octahedra. Analysis for the metal was made by simply igniting the salt. Found : Ir, 48:77%. Calculated for (NH,).IrCl,: lr, 43-70%. The identity of the salt was also checked by an X-ray powder photograph. In preparing the potassium salt, the more direct method of passing chlorine over a heated mixture of Ir and KCl was employed. Since C.P. Ir and KCl were used for this preparation, no analysis of K,[{IrCl,] was made. Bivalent Iridium. From a search of the literature it would seem that with the possible exception of IrCl, (Wohler and Streicher, 1913), no compounds of bivalent iridium have been described. The very early work of Skobolikoff (1853) dealing with alleged Ir! ammines could not be confirmed by Palmaer (1895), who after a very extensive investigation stated that he was unable to prepare the ammines of either Ir!'Y or Ir The complex measured (Ir(AsMePh,),Br,) was prepared and described by Dwyer and Nyholm (1943). The results of the susceptibility measurements are set out in Tables IIa and IIn. TABLE ITA. Paramagnetic Compounds. Specific Dia- Atomic eff Number Molecular | Suscept- | Temper- Uy magnetic* | Suscept- ohr of Compound. | Weight. | ibility. ature. M Correc- ibility. Mag- Unpaired | x x 10°. i tion. Va. netons. | Electrons. (NH,)2[IrCl, | 442 2-30 293 1,017 167 1,184 1-67 a K.[IrCl,} 484 2:02 295 978 ih 1,149 1:65 1 K.[RuCl, ] 393 9-71 293 3,816 ial 3,987 3:07 2 K.[OsC], |] 481 1:47 294 707 Zl 878 1:44 —_ (NH,).[OsCl, } 439 1-63 290 716 167 883 1:44 — (NH,).[OsBre] 706 1:01 291 713 238 951 1:49 — K,[RuCl,H,O Jt 375 4°31 293 1,616 160 1,776 2-04 i * The diamagnetic corrections applied are those of Trew (1941), which are the means of the most reliable results in the literature. Ti For K, {Ru Cl;], {Z=1-8 B.M. MAGNETIC BEHAVIOUR OF COMPLEXES. 151 TABLE IIB. Diamagnetic Compounds. Specitic Specific Compound. Susceptibility. Compound. Susceptibility. x xX108. “xx 108. K,[RuCl;0H | sy ie be —0-49 (RhPy;Br]Br as Ke gis —0-31 K,[OS(CN),|J3H,O .. as si —0-10 Py.H.«[RhBr,Py,] ms ne —0:12 K,[Os(NO,); ]* .- o. OUD Fi 0) ‘55 [Rh(PhMe,As),Br, } cr S56 —0- 38 [Rh(Ph,MeAs),Brz lo Ae ee —0-4 Ir(Ph,MeAs),Br, .. m: te —0-2 [Rh(Ph,MeAs),Cl, J» ay nes —0°5 * T am indebted to Mr. W. A. Rawlinson of the Walter and Eliza Hall Institute of Research, Melbourne, for this measurement, which was made on 13:9 mgm. in water solution. The measurement was made by means of a Curie- Cheneveau Balance. DISCUSSION. COMPOUNDS WHICH CONFORM TO THEORY. Diamagnetic Compounds. It will be seen that Pauling’s theory of magnetic moments accounts for the behaviour of a fairly wide range of platinum metal compounds, both para- and diamagnetic. Previous workers have shown that octahedral complexes of Pty, Pdvy, Ri, Ir, Rut and square complexes of Pt™ and Pd" are all diamagnetic, which is to be expected (see Table I). To this list must now be added octahedrally coordinated Os", whose complexes are also diamagnetic. Paramagnetic Compounds. Notwithstanding the fact that some paramagnetic platinum metal compounds have been previously reported and the fact that the present work adds several more to the list, compounds in this category form a minority group among the platinum metals. One of the most interesting cases brought to light is that of K,[RuCl,], the effective moment (3-07 Bohr magnetons) of which Shows that Rwv contains two unpaired electrons. This compound is notable as being the most strongly paramagnetic compound of the platinum metals described to date. Indeed so pronounced is its paramagnetism that the first preparation of the compound was mistakenly set aside as accidentally contaminated. On repeating the work there was no doubt about the high paramagnetic susceptibility which proved to be independent of field strength. To make sure that no accidental contamination had occurred, the K,[ RuCl, | was hydrolysed to K,[RuCl,OH | and the susceptibility again measured. It was then found that the paramagnetism had entirely disappeared. Although this 8 From Table I it appears that a substance like PtI, should have a moment of 4:9 Bohr magnetons, if the platinum is quadricovalent. Actually the substance is diamagnetic. (Unpub- lished work with J. B. Willis.) It is almost certain that the configuration of the platinum is not tetrahedral but octahedral : If this is so, the diamagnetism of PtI, accords with Table I. 152 D. P. MELLOR. procedure proved beyond doubt that K,[RuCl,|] was not contaminated with aby paramagnetic (or ferromagnetic) impurity, the diamagnetism of K,[RuCl,;OH] was a quite unexpected phenomenon. Several compounds exhibited effective magnetic moments corresponding to the presence of one unpaired electron spin. To this class belong K,[IrCl,] and (NH,),[IrCl,], the moments for both of which correspond fairly closely to the theoretical value (1:73). In this respect the hexachloro-iridates are different from K,[RuCl,], whose moment (3-07 Bohr magnetons) is somewhat greater than the theoretical value for two unpaired spins owing, no doubt, to some unquenched orbital contribution. This difference is interesting in view of the practically identical atomic environment of the iridium and ruthenium atoms in the potassium salts, a point which will be referred to again in the discussion of the hexachloro-osmates. One unpaired electron spin is found for Ru! in K,[RuCl,H,O], although here again the moment (2:04 Bohr magnetons) is larger than the theoretical value owing to unquenched orbital contribution. The effective moment agrees reasonably well with that found by Gleu eé al. (1936)? for trivalent ruthenium ammines and also with the values reported by Malatesta (1938) and Guthrie and Bourland (1931) for other Ru! complexes. The main interest of the present measurement is that it confirms the constitution proposed by Howe (1927) and Charonnat (1931) for the red chloro salt, viz. K,[RuCl,H,O]. The moment of the anhydrous salt K,[RuCl;] can be taken to mean that in preserving its coordination number of six, Ru!!' does not form a Ru-Ru bond in this complex but instead forms a halogen bridge : Hi Ne None COMPOUNDS WHICH DO NOT CONFORM TO THEORY. Paramagnetic Compounds. . A moment in the neighbourhood of 2-8-3 -0 Bohr magnetons was anticipated for K,[OsCl,] and related compounds but the measured value is far short of this and is in fact even less than the value required for one unpaired electron spin. To determine the extent of the deviation from theory it is obvious that low temperature measurements are required for the hexachloro and hexabromo- osmates. One point is, however, worthy of comment. Since these osmium compounds form part of the isomorphous series K,[PtCl,], K,[IrCl,], K,[ RuCl, ], etc., it would seem that departure from theory is not closely related to the effects produced by interatomic forces since, except for a small variation in the lattice constant, the environment of the platinum metal atoms is the same throughout the series. Diamagnetic Compounds. Among the elements of iron transition series, manganese has been examined in the widest range of oxidation states (from -++-2 to +7) and even in such a high oxidation state as +6 (in K,MnO,) conformity with Pauling’s theory has been found (Goldenberg, 1940). It was therefore a matter of interest to see whether this was true of a heavy metal atom like osmium in a high oxidation state as in K,[OsO,(NO,),.] for which w should be in the neighbourhood of 2-83 Bohr magnetons. The observed value of zero would seem to show that Hund’s rule requiring a maximum number of unpaired electrons (in the 5d sub-shell) breaks down for osmium. ® The values given by Gleu and Rehm (1936) range from 2-00 to 2-07 Bohr magnetons. MAGNETIC BEHAVIOUR OF COMPLEXES. 153 Perhaps the most interesting complexes in this group are those in which the metal atoms should have an odd electron and a moment corresponding to ohne unpaired spin. Complexes of Rb" and Ir" fall into this group. A wide search was made for paramagnetic bivalent rhodium complexes, but entirely without success. The evidence for state of oxidation of Rh" in the complexes studied has already been reviewed and there seems little doubt that one is here really dealing with Rh™. In the polynuclear compounds there is the possibility that Rh-Rh bonds are formed by means of which the odd electron on each metal atom becomes paired as it does in Hg,Cl,. X-ray crystal analysis (Wells, 1938) has shown that the bridge in the arsine complexes of palladium consists of two halogen (bromine) atoms : and in all probability the same kind of bridge functions in the Rh" complexes. Since it is theoretically possible to formulate the binuclear Rh?! complexes in terms of Rh-Rh bonds, at the same time retaining octahedral coordination for the metal atom, efforts were made to secure compounds in which the possibility of metal to metal bonds was ruled out. Two such compounds, [RhPy,Br]Br and [Rh(AsMe,Ph),Br,], were examined and found to be diamagnetic. It appears therefore that we have in the Rh complexes an instance of the quenching of spin moment, the mechanism of quenching being similar to that responsible for the disappearance of orbital moment in the transition elements other than the rare earths. A similar quenching of spin moment was observed in the one Ir" compound measured. In regard to the quenching of spin, the electronic configuration of the two atoms (Rh" and Ir@) in d?sp? complexes may possess some special significance ; in each instance the odd electron must occupy an outer unstable orbital (cf. Cot,?° Pauling, 1940) and in this situation it may be peculiarly liable to the quenching effect. It is doubtful whether these are the only circumstances in which the electron Spin is quenched because Os™! compounds!! which should behave similarly to those of Ru™ and have moments in the neighbourhood of 1-73 Bohr magnetons are in point of fact diamagnetic (u=0). It must be admitted that the number of cases so far examined is very small owing to the fact that very few compounds of tervalent osmium have been described. One only, namely K,[Os(NO,);], was studied in the present work and for this .=0, an observation in agreement 10 Unfortunately the data for Col are none too clear on this point ; K,[Co(CN),] has been reported to be diamagnetic, but the substance is so unstable that its empirical composition cannot be regarded as established with any great certainty. There is evidence that in solution the cobaltocyanide ion has the composition [Co(CN),|— (or Co,(CN,,) (Glasstone and Speakman, 1930). The diamagnetism of the cobaltocyanide ion may be explained in terms of a structure such as: CN CN CN | -CN CN =CO ~ Co_= CN ON ee GN CN CN This formulation finds some support from the measurements of Cambi and Ferrari (1935) on K,Ca[Co(NO,),] and closely related compounds whose moments show that Col has one unpaired spin. On the other hand, if we accept the results of Cambi and Ferrari it must mean that there is no quenching of the spin of the electron occupying the unstable 4p orbital. 11 The unpaired electron in Os!¥ and RuH! occupies a stable orbital, viz. one of the 5d. 154 D. P. MELLOR. with Bose and Bhar’s value of OsCl,.17, With K,[Os(NO,),;] it is difficult to be sure whether we have a real departure from theory or not. The probability is that Os is not penta-coordinated here and that the complex is [Os,(NO,),9]. In that event, there are two possible alternative structures, one involving a metal-metal bond, the other a bridge formed by two nitro groups : : That the nitro group can so bridge has been demonstrated by Mann and Purdie (1936), and it would seem that such a bridge is the more probable explana- tion of the structure of K,[Os(NO,),]. On the other hand, if the complex should contain the Os-Os bond its diamagnetism is not inconsistent with theory. Apart from crystal structure analysis there appears to be little prospect of distinguishing between these alternatives unless it can be shown that Os'! complexes which are definitely mononuclear have a moment of 1-73 Bohr magnetons. One further curious anomaly remains to be noted, namely the difference between the magnetic moment of Ru'v in K,[RuCl,] (u=3-07 Bohr magnetons) and the moment of Ru'v in K,[RuCl,OH] (u=0). The quenching of the moment to be expected from two unpaired electron spins in the latter compound is extremely puzzling and the only circumstance of significance seems to be the difference in the symmetry of the crystalline field about the ruthenium atom in the two compounds. X-ray powder photographs of K,[RuCl,;OH] show that its symmetry is lower than cubic and it follows that there must be a corres- pondingly lower symmetry in the distribution of interatomic forces about the ruthenium atom; indeed the very nature of the complex ion itself suggests this. If this is the correct explanation of the phenomenon, it is not clear why the quenching of spin moment does not occur in K,/ RuCl,H,O] (u=2-04 Bohr magnetons). Investigations of compounds of the type [IruA,]X,, [RhA,|X, and K,[IrX,Y |] may throw more light on this question. There seems to be little doubt that covalent bond formation alone does not explain the abnormally low magnetism of complexes of the palladium and platinum metals since the moments found in certain instances are lower than those required by Pauling’s theory. In concluding his chapter on the palladium and platinum transition groups, Van Vleck (1932) wrote: ‘‘ Further experimental data on the different salts of the Pd and Pt groups are greatly to be desired. Without them further discussion would be too speculative.’’ The main purpose of this investigation has been to supply some of the previously missing data for the purpose of testing as far as possible the existing theories, with the hope that the discrepancies between theory and experiment which have been brought to light may lead to further — theoretical investigation. ACKNOWLEDGMENTS. The author is indebted to Messrs. F. P. Dwyer and R. 8S. Nyholm for their generous cooperation in making available compounds of rhodium and iridium ; for their assistance in micro-analyses for chlorine and for the loan of. ruthenium ; to Mr. J. L. Sullivan for taking powder photographs of some of the materials used in this investigation ; to Mrs. L. Buckley for micro-analyses for 12 In discussing the moment of Os!!! (in OsCl,) Van Vleck (1932) has considered the possibility that strong internal magnetic coupling leading to quenching of both spin and orbital moment is responsible for the abnormally low magnetism, but he points out that a possibly are a objection to this explanation is the fact that Os!II is in a 8S state. MAGNETIC BEHAVIOUR OF COMPLEXES. 155 nitrogen ; and finally to the Commonwealth Research Fund for a grant to purchase osmium and iridium. SUMMARY. Complexes of the platinum metals in different valence states have been systematically examined with the object of testing the table of magnetic moments drawn up by Pauling for the transition elements. Over a wide range the pre- dictions have been verified, but instances of breakdown have been observed, notably for Osv, Ostv, Os, Rulv, Rh™ and Iru Interesting examples of the quenching of spin moment have been noted among the complexes of Rhu, Ir! and Rulv. REFERENCES. Antony, V., and Lucchesi, A., 1899. Gazz. Chim. Ital., 29, 312. Aoyama, S., 1924. Z. anorg. Chem., 138, 249. Aston, F. W., 1942. Mass Spectra and Isotopes. Edward Arnold & Co., London, p. 214. Bose, D. M., and Bhar, H., 1928. Z. Physik., 48, 716. Burstall, F. H., 1936. Unpublished data. Cabrera, B., and Duperier, A., 1939. Phys. Soc. Lond., 51, 845. Cambi, L., and Ferrari, A., 1935. Gazz. Chim. Ital., 65, 1162. Charonnat, R., 1931. Ann. Chim., 16 (Ser. 10), 5. Christiansen, J. A., and Asmussen, R. W., 1934. Kong. Danske Videns Math-fys., 12, 1. ——— 1935. Kong. Danske Videns Math-fys., 13, 1. Crowell, W. R., and Yost, D., 1928. J. Amer. Chem. Soc., 50, 374. Crowell, W. R., Benton, R. K., and Evenson, R. F., 1938. J. Amer. Chem. Soc., 60, 1105. de Haas, W. J., and Gorter, C. J., 1930. Leiden Comm., 210d. Dufet, H., 1895. C.R. Acad. Sc. Paris, 120, 377. Dwyer, F. P., and Nyholm, R. S8., 194la. Turis Journat, 75, 127. ————— 19416. TuHis Journatr, 75, 140. 1942a. Tuis JouRNAL, 76, 129. 19426. THis JouRNAL, 76, 133. 1942c. THis JoURNAL, 76, 275. 1943. Tuts Journnat, 77, 116. Ebert, F., 1931. Z. anorg. allg. Chem., 196, 395. Feytis, E., 1911. C.R. Acad. Sc., 153, 668. Gilchrist, R., 1932. Bur. Stand. J. Res., 9, 279. 1932. Bur. Stand. J. Res., 9, 547. Glasstone, S., and Speakman, J. C., 1930. Analyst, 55, 93. Gleu, K., and Rehm, K., 1936. Z. anorg. Chem., 227, 237. Gleu, K., and Cuntze, W., 1938. Z. anorg. Chem., 237, 187. Gleu, K., and Grenel, W., 1938. Z. anorg. Chem., 237, 197, 326. Goldenberg, N., 1940. Trans. Faraday Soc., 36, 847. Guthrie, A. N., and Bourland, L. T., 1931. Phys. Rev., 37, 303. Honda, K., and Ishiwara, T., 1915. Sc. R. Tohoku Imp. Univ., 4, 215. Howe, J., 1904. J. Amer. Chem. Soc., 26, 543, 942. 1927. J. Amer. Chem. Soc., 49, 2381, 2393. Janes, R. B., 1935. J. Amer. Chem. Soc., 57, 47. Krauss, F., and Schrader, G., 1928. J. prakt. Chem., 119, 279. Malatesta, L., 1938. . Gazz. Chim. Ital., 68, 195. Mann, F. G., and Purdie, D., 1936. J. Chem. Soc., 873. Nier, A. O., 1937. Phys. Rev., 52, 885. Pauling, L., 1931. J. Amer. Chem. Soc., 53, 1367. ————— 1940. The Nature of the Chemical Bond. Cornell Univ, Press, Ithaca, N.Y. Palmaer, W., 1895. Z. anorg. Chem., 10, 320. Reekie, J., 1939. Proc. Roy. Soc., 173A, 367. Skobolikoff, 1853. Bull. Acad. St. Petersburg, 11, 25. Sugden, S., 1932. J. Chem. Soc., 246. Trew, V. C. G., 1941. Trans. Faraday Soc., 37, 476. Van Vleck, J. H., 1932. The Theory of Electric and Magnetic Susceptibilities. Clarendon Press, Oxford, p. 313. ——_—_—_—_—_—_———— 1935. J. Chem. Phys., 3, 807. Wells, F. A., 1938. Proc. Roy. Soc., 167A, 169. Wintrebert, M. L., 1903. Ann. Chim. Phys., 28, 15. Wohler, L., and Streicher, S., 1913. Ber. dtsch. Gess., 46, 1577, 1720. Department of Chemistry, University of Sydney. THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. Part II. THE COUNTRY BETWEEN BUNYAN AND COLINTON. By W. R&R. BROWNE, D8c., Depariment of Geology, The University of Sydney. With Plate VI. Manuscript received, X uvember 10, 1943, Read, December 1, 1943, CONTENTS. Page Introduction is ae 7 yA oe “als bare 9 a thee be bore Ordovician .. te als ae wig aN fe ia we ae a sini Lc Silurian =e in - Se Sie a 3s oe aa ae it oe VSS Cainozoic .. nie si ‘ae a ee a He My the Sa Ht (a) Basalilogs, at bbe ie ae Bs Ane Aa ws a ey dee (6) Sediments ie He ie A he a will oie bil -. 164 Igneous Intrusions. : Ad Be ae ate an ae Ay ae .. 164 (A) The Cosran Gace We : ve an ae ee Se ay .. 164 (B) The Murrumbidgee Buthylith, i ae 5 Be an ah vie ua sob DS (C) Porphyries and Porphyrites .. an ate Ae si wh i army 41°, (a) Granite-porphyries ait, ar a ay aw ie Lid Ow (6) Other Porphyries and Borphanies sks sid ae te nis) ee (D) Genetic Relations among the care Rocks at ie he a oe (EK) Quartz-Dolerites i is a ia a Sie ae .. 169 Geological Structure of the Palmiente ‘Rooke Ae a se By La Aes Uy Structural Relations of Ordovician and Silurian Rocks .. ea se Bhd es Summary .. sie af ue Hu oe Ae she a ee ie a ia References .. ne af axe lt a ah a ee se nt PN INTRODUCTION. Nearly thirty years ago an account was given of the geology of the country around Cooma (Browne, 1914); this was to have been followed by detailed petrological studies, but it was considered desirable first to map the country to the north, and the field-survey on which the present paper is based was begun as far back as 1917. Various circumstances interfered with the completion of the work, though references to some aspects of it were included in other papers (Browne, 1929, 1931a, 1933). Detailed field- and laboratory-studies of the petrology of the metamorphic rocks of the Cooma area have since been undertaken by Dr. Germaine A. Joplin, who has already published two valuable papers on the subject (Joplin, 1939 and 1942). The field-work was made possible by the unbounded kindness and hospitality of many local residents, including the late Mr. T. J. Sherlock and family of ‘¢ Kiaora ’? near Cooma, the late Mr. and Mrs. V. Fraser of Murrumbucka, Mr. and Mrs. J. J. Harnett, formerly of ‘‘ Gurrabeal’’, Mr. and Mrs. H. Fane de Salis, late of Chakola, the late Mrs. Cosgrove of ‘ Oakvale ””, and the late Mr. A. Cosgrove and Mrs. Cosgrove of Billilingra House. — SS THE GEOLOGY OF THE COOMA DISTRICT, N.S8.W. 157 I am much indebted to Dr. Joplin for drawing the maps, for the loan of microslides and for discussion. The area mapped and investigated extends to the Colinton Gorge, some 30 miles north of Cooma. For the most part only the field-relations and megascopic characters of the rocks have been studied, but in some instances microslides have been examined. The investigation has served to confirm some of the conclusions and interpretations put forward in the 1914 paper and has proved others to be in error. Reference will be made to these in the appropriate places. The principal formations whose northerly extensions have been studied are : (1) An Upper Ordovician series, chiefly of sediments, metamorphosed to varying degrees ; (2) An Upper Silurian series, also mainly sedimentary ; (3) A late Ordovician or epi-Ordovician intrusion, known as the Cooma eneiss, intimately penetrating the most altered parts of (1); (4) A late Silurian intrusion with a considerable extension to the north, which may be called the Murrumbidgee bathylith. Its component rocks are gneissic—the blue, white and pink gneisses of the earlier paper ; (5) Granite-porphyries, thought to be of late Silurian age ; (6) Porphyries, felsites and porphyrites mostly of post-Silurian age. In addition to these there are minor dolerite intrusions of unknown age, Tertiary basalt-flows and dykes and Tertiary and post-Tertiary gravels and other alluvium. The area examined is roughly divisible both geologically and physio- graphically into two parts. In general the Silurian and less altered Ordovician rocks form relatively low, undulating country, most of which lies to the east of the Murrumbidgee ; west of the river the Ordovician schists and granites and the late Silurian gneiss, which are more resistant to erosion, rise in the south to heights of more than 3,000, and farther north to 4,400 feet above the sea, forming country in places rugged and deeply dissected. A distinct relation can be observed between geology and physiography not only in the broad way but often also in matters of detail. ORDOVICIAN. In the 1914 paper it was pointed out that west of Cooma a gradual eastward transition could be traced from graptolite-bearing Ordovician slates through phyllites and schists into rocks granitized by the Cooma gneiss. In 1931 on Cottage Creek, some 15 miles south of Cooma, graptolites, including Diplograptus bicornis, were found in black and grey shales or slates on the strike of the schists (Browne, 19316); this discovery suggested that the latter were of Upper Ordovician age,* and showed that their metamorphism was not continued along the strike but was in all probability peripheral to the Cooma gneiss. Dr. Joplin has divided the Ordovician sequence west of Cooma into the Coolringdon beds, chiefly of siliceous slates, and the Binjura beds lying to the east of them, and composed of alternating shaly and sandy beds now metamorphosed. She has also demonstrated conclusively the causal relation existing between the Cooma gneiss and the metamorphism, and has been able to plot zones of progressive alteration marked by the entry of certain index-minerals. * The statement in the 1914 paper that Didymograptus and Tetragraptus were associated with Diplograptus at Geygedzerick Hill near Berridale was an unfortunate and unaccountable error. N—December 1, 1943. 158 WwW. R. BROWNE. The northern extensions of these zones have been traced by Dr. Joplin and are described by her in a paper shortly to be published in the Proceedings of the Linnean Society of New South Wales. The rocks of most of the zones are easily identifiable in the field ; for example the blue-black slates cropping out near McCarthy’s Crossing, 7 miles north-west of Cooma, may be observed again near the mouth of Bulga Creek, 4 miles to the north, and have been noted at the head of Long Creek outside the limits of the map. The belt of andalusite-schist (Slack’s Creek phyllites) passes to the west of Muddah Lake, reappears from beneath Tertiary basalt in the headwaters of Barkgunyah Creek, and has been picked up again in Long Creek about two miles up from ‘ Oakvale ’’, north of which it is cut off by the Murrumbidgee bathylith. Good exposures of the schists of the Binjura beds may be seen in the western tributaries of Pilot Creek and along Barkgunyah and Long Creeks. The beds are more or less granitized, and include two rather distinctive types. One of these, which for field-purposes has been designated the corduroy granulite, is really a fine-textured paragneiss in which bands of light-coloured material, about one-tenth of an inch wide, alternate with still narrower layers rich in biotite ; the other, the mottled gneiss, was originally thought from its trans- gressive relations to be a much contaminated igneous type, but Dr. Joplin has shown that it is a granitized sedimentary schist. Its position has been noted on the map only where it has been encountered in the course of the survey. Along the ridge which runs from Mittagang Bridge to Murrumbucka Gap on the left bank of the Murrumbidgee the granitized rocks make prominent outcrops, flanked east and west by bars of Cooma gneiss and the blue gneiss of the Murrumbidgee bathylith. Further north inclusions of the granitized rocks are seen at intervals in the blue gneiss, especially in its eastern half, as in Sandy and Spring Vale Creeks, at the head of Colinton Gorge, and elsewhere. The mottled gneiss has not been observed north of Sandy Creek. Good sections showing intimate penetration of granulites by Cooma gneiss are to be seen along the Murrumbidgee above and below Mittagang Bridge. In the zones of higher grade metamorphism only slight indications of contemporaneous Ordovician igneous activity are visible. Dr. Joplin (1942) has shown that in Cooma certain granulites containing hornblende and some- times pyroxene, which are enclosed in the Cooma gneiss, are basic in composition and probably represent former basalt-flows or dolerite-sills. Further north lenses of similar rocks are to be seen at several points enclosed in the blue gneiss, mostly at less than three-quarters of a mile from its eastern margin. Along the road from Cooma to Mittagang two miles past the cemetery such inclusions were formerly to be seen, intimately penetrated by blue gneiss, and several have been noted in the parishes of York and Bullanamang near ‘‘ Dromore ” and ‘* Riversdale ”’ homesteads, in Sandy Creek, and in Spring Vale and Long Flat Creeks. Some of the rocks have been completely recrystallized but others retain relics of former minerals and structures. Grainsize is very variable and may be quite coarse, and field-evidence is not lacking of interaction between host and inclusion. It is, of course, possible that some of these amphibole- rocks are intrusions comagmatic with though antecedent to the blue gneiss, and therefore of Silurian age. The Ordovician rocks lying to the east of the Murrumbidgee bathylith offer a most remarkable contrast to those just described, both in composition and in grade of metamorphism. The black slates and chloritic phyllites cropping out — at Bunyan and to the south of it continue to the north but with decreasing importance, being apparently encroached upon gradually by the blue gneiss. The most abundant rock-types are schistose acid crystal-tuffs and schistose rhyolites. Rocks of these types can be traced from Bushy Hill east of Cooma at intervals to Bunyan and through the parish of Woolumla, but attain greater THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 159 extent and importance in the parish of Callaghan, and form the bulk of the Ordovician rocks in Billilingra and Bransby. The degree of crushing and recrystallization may be such as to obliterate all traces of the original rocks, but the alteration is always of low grade. Just south of the Colinton gorge there seems to be an almost unsheared core of the rhyolite grading outwards into the sheared or crushed phase. Other sheared rocks appear to have been felspathic grits, and the coarsest type, found in the parish of Callaghan, is a quartz-grit which becomes conglomeratic, with pebbles up to 14 inches in length. A little limestone is seen, always in small lenses, always recrystallized, and with no traces of fussils. The most southerly, overlooked in the 1914 survey, is on the railway line 34 miles north of Cooma; this is in part silicified and silicated by a tongue of acid blue gneiss. Another small patch is on the roadside at Pearman’s Hill and three others have been noted in the parish of Bransby at or near the Ordovician-Silurian boundary. A very constant feature of the Ordovician terrain is a series of sill-like intrusions of schistose quartz-porphyry or porphyroid. Though recognized at Bushy Hill and south of Bunyan, it becomes conspicuous first in the parish of Woolumla, where its dark green colour contrasts with that of the schistose rhyolites and tuffs. It makes strong outcrops on the right bank of the Murrum- bidgee near its junction with the Bredbo, and further north in the parishes of Bullanamang and Bransby, where it is in contact with the blue gneiss. Particu- larly good sections may be studied up Spring Vale Creek, and in the Colinton gorge, where the Ordovician belt has narrowed considerably and about two- thirds of its width is of porphyroid dipping to the west, the remainder being schistose rhyolite, of which also many lenses are enclosed in the intrusive rock. The porphyroid is essentially a product of low-grade metamorphism ; chlorite and sericite are abundant in it and crushing and straining of the quartz phenocrysts are characteristic. The rock is in places banded parallel to the planes of schistosity, which are remarkably even and regular and appear to conform with the boundaries of the associated phyllites and slates. The geological age of these porphyroids is not known with certainty, but in their degree of alteration they are more closely allied to the Ordovician than to the Silurian rocks described below, and provisionally they are assigned a late Ordovician age. The proximity of the slates, phyllites, porphyroids and other low-grade rocks of the eastern belt to the Cooma gneiss and granitized schists is a matter that calls for comment. Owing to the interposition of blue gneiss, nowhere can a gradation be traced between them, but in places less than half a mile separates their outcrops, and near the entrance to Colinton gorge lenses of porphyroid enclosed in the blue gneiss are very close to and on the same strike as inclusions of granitized schist and Cooma gneiss. It may be, of course, that, as suggested in the 1914 paper, the metamorphic effect of the Cooma gneiss was much greater on the west than on the east, due to the shape of the mass and the attitude of the enclosing rocks, and perhaps also to the fact that pressure during crystal- lization came from the east. It may be also that the inclusions in the blue gneiss changed their relative positions as the result of compression during the crystal- lization of the latter. Nevertheless the possibility of overthrusting having occurred and brought high- and low-grade rocks into apposition before the intrusion of the blue gneiss, though incapable of proof, cannot be lightly dismissed. SILURIAN. The Silurian rocks were not mapped or studied for any great distance from their western boundary. They lie entirely to the east of the Ordovician belt with a submeridional strike and variable dips. Clay-slates imperfectly cleaved and in places differing but little from shales are perhaps the dominant rocks, especially near the Ordovician boundary. Massive and flaggy sandstone and 160 W. R. BROWNE. quartzite appear with them west of and along the railway in the parish of Callaghan. Undoubtedly, however, the most interesting rock-types are the acid crystal-tuffs which form an important proportion of the sediments. These are characterized by abundant grains of quartz and felspar, generally with hornblende and biotite and a variable proportion of cryptocrystalline matrix. Angular rock-fragments up to a few inches in length may be locally abundant, and in Por. 22, Par. Billilingra rounded pebbles of quartz-porphyry convert the rock into a tuffaceous conglomerate. Alteration of the constituent minerals to albite, chlorite, epidote, etc., is common and the rocks often show jointing and other signs of strain. The distribution of the tuffs is rather capricious, the belts being somewhat discontinuous, and for many of the exposures accurate mapping is difficult or impossible. The behaviour of the rocks is rather like that of intrusives, outcrops making and ceasing abruptly, or tapering out and transgressing the strike of the associated slates. Elsewhere they are distinctly bedded and show gradations and variations in grainsize like ordinary clastic sediments. Some of the rocks, plentifully studded with grains of quartz and felspar, are clearly recognizable in hand-specimen as fragmental, but others, in which the matrix is more prominent, have the appearance of porphyries. Moreover the tuffs themselves may be intimately invaded by quartz-porphyry, and it is almost or quite impossible to distinguish between the two types in hand-specimen and sometimes even in thin section. Not improbably some of the doubtful types are really porphyries which have picked up mineral fragments on their way through the tufts. Good sections of these rocks are few, but on the railway-line between 2 and 3 miles north of Bredbo two low cuttings afford opportunities to study their characters and their relations to the slates. In these the suggestion of intrusion is produced by marginal hardening of the associated slates, by inclusions of hardened, cherty shale in the tuff and by long tongues of tuff transgressing the slate. The tuffs also enclose ellipsoids of recrystallized limestone in some of which fossil corals and brachiopods may be observed, recalling the somewhat similar occurrences at Yass and Jenolan. In the 1914. paper the tuffaceous character of these rocks was not fully recognized, largely owing to their close association in the field with true igneous rocks, and they were all grouped as porphyries. The tuffs attain probably their greatest development in the parish of Bransby (Fig. 1): Two or three miles north of Bredbo they extend from the railway line east to the road and beyond it for at least another three-quarters of a mile. In the parish of Callaghan (Fig. 2) tuffs exactly similar to those just described are found among the Ordovician beds not far from the Silurian boundary. Though marginally somewhat sheared, the outcrops are in the main perfectly massive and in marked contrast to the highly schistose rocks around them. Similar rocks appear in Por. 48, Par. Billilingra and elsewhere. The occurrence of these tuffs among the older rocks heightens the impression that they are intrusive, but does nothing to elucidate the mystery of their emplacement. A little impure fossiliferous limestone crops out in Gungoandra Creek a mile south of the Colinton gorge and some has been observed in the ridge north of Bredbo outside the limits of the map. Apart from the crystal-tuffs contem- poraneous voleanic activity is indicated by a basic lava-flow, much carbonated and possibly spilitic, on a ridge in Por. 96, Par. Bransby, and by a well-banded felsitic rhyolite, possibly 150 feet thick and three-quarters of a mile long, in Pors. 23 and 41, Par. Billilingra. A smaller outcrop of fluidal rhyolite at the very base of the Silurian beds in the north-east corner of Por. 24 of ~ same parish is also probably a flow. THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 161 It seems possible that only the western parts of the Silurian sequence are eharacterized by the presence of igneous material. A reconnaissance trip along the road for 20 miles from Cooma in a north-east direction to Umaralla and beyond revealed that east of Rock Flat Creek there are only shales, slates, quartzites and impure limestones. Fossils have been found at various points. The shales at Billilingra Siding have yielded Encrinurus mitchelli Eth. fil., Brachiopoda indet., Alveolites sp., Favosites sp., Heliolites sp., (?) Hercophyllum sp. An impure limestone in Gungoandra Creek one mile south of Colinton gorge contained Hercophyllum shearsbyi (Sussmilch), “ Cystephyllum ”’ sp., Mucophyllum crateriodes Eth. fil., and Favosites allani Jones. Favosites and a small Orthid have been got from limestone ellipsoids in tuff in a railway cutting 24 miles north of Bredbo. These fossils, for whose identification I am indebted to Dr. Ida A. Brown, all came from approximately the same horizon and indicate a correlation with the lower part of the Upper Silurian Hume Series of the Yass district. It is very probable that more careful collecting would enable closer and more detailed correlations with the type-area at Yass and help in the elucidation of the geological structure. CAINOZOIC. (a) Basalt-Flows. These crop out extensively in the parish of Murrumbucka. Remnants of basalt are seen at about 20 feet above the river near Mittagang Bridge, and more extensive relics are found on the sides and floor of the valley of Pilot Creek. Three or four miles up this creek the basalt rises abruptly from the valley through about 400 feet to form a flat-topped plateau area which extends north for a couple of miles and then westward almost to the Murrumbidgee. Along the _ track from Muddah Lake south to McCarthy’s Crossing further thin patches of basalt are encountered on the plateau surface. A westward view from Reeve’s Point, one of these remnants, itself some 600 feet above the river, makes it clear that the basalt crops out extensively in the river valley, and there are evidences that it once filled the valley at a time when its floor was 20 or 30 feet above its present level. In the eastern part of the area examined basalt is almost entirely lacking. A couple of small patches survive in the parish of Woolumla within a radius of a mile from Chakola railway station. It is hard to understand why these should have escaped erosion, for neither is favourably situated for preservation. The basalt of the flows is the usual olivine-bearing type with augite that appears slightly titaniferous. Physiographic considerations suggest that the flows are of Pliocene age for they occupy parts of the broad lowland area eroded by the Murrumbidgee system out of the Miocene surface forming the plateau. It is evident that they filled river valleys to a depth of at least 600 feet and spread over the higher ground, smoothing out its inequalities. A succession of flows is indicated by terracing or trap-featuring, and there has evidently been consider- able erosion since the eruptions. A former extension of the basalt to Pearman’s Hill is inferred from the presence there of remnants of silicified gravel, and there is a possibility that the flows may have continued much farther north, for kaolinized dykes are seen in railway-cuttings in the parish of Callaghan and a dyke of fresh olivine-basalt has been found in a cutting just north of Bredbo River and another in the parish of Bullanamang near the mouth of Long Flat Creek. Though outside the area surveyed, mention should be made of an occurrence of Tertiary basalt, believed to be older than Pliocene, which caps the ridge forming the watershed between Toll Bar Creek and Umaralla River just south of the Cooma-Umaralla road. It is underlain by alluvial sediments which form a kind of elevated deep lead some 400 feet above the level of the Pliocene basalt in the valley of Toll Bar Creek (Browne, 1933). W. R. BROWNE. 162 uartz -Porphyry, dgee Bathylith. , Schists, etc. Q (a urrumbi LEGEND Alluvium . Cooma Gneiss. AAW Slates , elc ERISA (FLTL]] states JHigh Level Gravels. acne Basalt. eke Tuffs and Breccia. Granite -Porphyry LAG] Quartz-P, orphyrite and = SEs yy > = — —Felsite~ Dyke ae es MANG i a . PARISH \ OF, va ee oe yy | BULLANA NORTH SCALE Mile Fig. 1. 163 THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. aie i) S \ Qa a Sy See S wv NORTH tr} = ee (oy as = 1s a 3 Sait eA A & ay : ei ener Ny A BB 4 tee - aN . Az ; vat: (@) . y Miieieaiiee ea OI HEAY. a a a i ea oll Dod Dad - - -- = ISH| | OF| |! | ee f : < f - n, ea oe a ¢. Lie LAs ste op Me etal es = se $ Peat big! ER? Bs Se ea! baa. ZAAAAAAAAALIZIZIA Det ae $5 FAAIVIA VAVAWA Fig. 2. Figs. 1 and 2.—Detailed geological map of the eastern half of the Bunyan-Colinton area. 164 WwW. R. BROWNE. (b) Sediments. | (i) About 100 feet of gravels and ferruginous shales and sandstones underlie the basalts on the ridge east of Toll Bar Creek. In the shales have been found remains of leaves including those of Cinnamomum sp. (ii) Sub-basaltic Pliocene sediments are seen on the Murrumbucka plateau in a tributary of Bulga Creek, where some two or three feet of micaceous grit overlie a small thickness of lignitic or peaty material. Water-bearing sands between basalt-flows have been encountered in sinking wells. In the eastern part of the area large blocks of cemented gravels or ‘ grey billy ’’ are exposed by the erosion of overlying basalt on the roadside at Pearman’s Hill about 170 feet above the Umaralla River. More extensive outcrops appear farther south, north of the road from Cooma to Umaralla and at a height of 100 feet above Toll Bar (Rock Flat) Creek. These are the rock-types referred to in the 1914 paper as quartzitic conglomerate, whose origin was not then understood. (iii) At many points along the Murrumbidgee and Umaralla Rivers there are banks and terraces of high-level gravels. These are particularly abundant along the Murrumbidgee in the parishes of Billilingra and York and in the village of Bredbo. They are at varying heights above the river, but most of them at about 240, 150 and 70 or 80 feet. The exact age of these is uncertain; the highest and oldest as well as the lower deposits contain pebbles of basalt and grey billy, which would suggest that they postdate the Plocene basalt, but in view of the existence of older Tertiary basalt-flows referred to above such an inference is not altogether justified. (iv) Gravels and flood-plain silts, clearly of Recent age, are found at 20 or 25 feet above river-level. They are extensively developed in the Umaralla River valley and along the mature stretches of the Murrumbidgee. Alluvium is found also along many of the tributary creeks in the eastern half of the area, in places as much as 20 feet deep, and considerable quantities of detritus have been spread along the bases of the ridges, partly alluvial, partly colluvial. Gullying, which is unfortunately very widespread, reveals sections showing that these deposits have been to some extent cemented. IGNEOUS INTRUSIONS. (a) The Cooma Gneiss. In the south the chief outcrop of this intrusion is in and around the town of Cooma; over to the west there are numbers of lenses and sills intimately penetrating the Ordovician schists. Northward the intrusion is continued as two more or less well defined parallel zones of granitic penetration, one on either side of the ridge that stretches from Mittagang Bridge to Murrumbucka Gap. The zones are separated by a belt of granitized rock—chiefly mottled gneiss and granulite—that caps the ridge and forms the watershed between the Murrum- bidgee on the east and Pilot and Murrumbucka Creeks on the west. The igneous rock igs massive aS a rule and devoid of directional structures, so that it is a granite rather than a gneiss, and in many places it forms conspicuous tors. | Where it is in narrow lenses, however, it has been crushed and granulated and has developed a rude submeridional cleavage which is thought to have been imposed on it at the time of the epi-Silurian diastrophism. On account of the intimate nature of the penetration it is impossible to lay down definite boundaries to the igneous zones, and the outcrops have been marked on the map (Plate VI) only wherever they have been observed in the course of the survey. The petrography of the rock—a contaminated granite—has been fully described by Dr. Joplin. Both zones of granite are cut off just south of Murrumbucka Gap by the Murrumbidgee bathylith, but remnants of it enclosed in the latter and generally = THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 165 accompanied by granitized schist can be seen at intervals far to the north, as in Sandy (Reedy) Creek and in Gap Creek at the head of the Colinton gorge. The most westerly observed occurrence forms an isolated outcrop in Barkersdale Creek about a furlong up from where it is crossed by the blue gneiss. Along the western margin there are occasional dykes and lenses of an acid or aplitic modification ; these crop out on the Murrumbucka Road near Mittagang Bridge, and at various points among the schists to the north. A few large masses are enclosed in the blue gneiss near Oakvale homestead and there is a little on Barkersdale Creek. It is noteworthy that lit-par-lit injection of the adjacent schists, so conspicuous about Cooma, is virtually absent to the north, though it has been observed to a slight extent in association with the more easterly belt of Cooma gneiss. As to the shape of the intrusion the impression gained is that the numerous lenses and sills are upward projections from an underlying solid mass forming a synchronous bathylith with a very marked submeridional trend and a cross- section roughly approximating that suggested in the 1914 paper, or perhaps a tabular mass with a general dip to the east and many projections into the roof. The extension of the metamorphic effects on the western side of the intrusive axis and their sharp decline on the east, together with the presence of acid modifications on the west, suggest that the progress of the magma was influenced by pressure from the east. The age of the intrusion is considered to be probably closing Ordovician because of its close relation to the folding and metamorphism of the Ordovician rocks, which was clearly accomplished before the deposition of the Silurian strata. (b) The Murrumbidgee Bathylith (see Plate VI). Reconnaissance has shown that this extends for about 60 miles north from Cooma, so that only its southern half comes into the present survey. It is the dominating formation in the western half of the area mapped, but since much of it, in the parishes of York and Bullanamang, is in a region uninhabited, untraversed by roads or tracks, and in places mountainous, uncleared and rising to more than 2,000 feet above the river, its detailed study is a matter of some difficulty. To the south the intrusion frays out into a number of tongues, but near Murrumbucka Gap these coalesce, and in the area mapped the bathylith has a known width of nearly five miles ; farther north it is considerably wider. The eastern margin of the mass is marked by a considerable development of acid rock, mapped separately in 1914 but not differentiated in the present survey. ‘Two types are recognized, one with much pink felspar and some mica and the other aplitic in constitution and devoid of mica. The former seems in places to grade westward into the normal blue gneiss, and the latter, occasionally veined with bright green epidote, appears as sheets or sills arranged subparallel to the margin of the intrusion. Immediately west of the acid phases the prevailing rock-type is a well-foliated primary gneiss with the constitution of a quartz-mica-diorite ; westward the foliation becomes less marked and the rock acquires a little potash-felspar in addition to plagioclase. Dykes and veins of tourmaline-pegmatite and quartz-tourmaline rock, some of which contain a graphic quartz-tourmaline intergrowth, are not uncommon. Along both its margins the gneiss contains numerous elongated and lenticular inclusions, principally of schist on the west and of schist and granulite, Cooma gneiss, and amphibolite and hornblende-granulite on the east. The margins are in general smooth, but modified by concordant tongues and apophyses, and there are intrusive marginal lenses of gneiss amid the invaded rocks. Indeed in some places, as to the east of the river in the parish of Woolumla 166 W. R. BROWNE. and again farther north along the river in the parish of York, the igneous mass might almost be regarded as consisting of great numbers of lenses or sills separated by screens of country-rock. The Murrumbidgee bathylith is clearly unrelated to the regional meta- morphism of the Ordovician rocks and was evidently injected long after this was accomplished ; nevertheless it has induced a limited amount of alteration in the invaded rocks. This is most evident along the eastern margin, where it seems to have consisted chiefly of partial or complete metasomatism. Granitiza- tion of slates and phyllites is particularly well seen in the parishes of Woolumla and Bransby, and where the Ordovician porphyroids are in contact with the gneiss they have been locally hardened by introduction of quartz and have had their felspars albitized, though here, as elsewhere, the zone of evident alteration is surprisingly narrow. Along the western boundary of the mass from Barkersdale Creek into the high country lying to the north some contact-hardening of the invaded schists seems to be indicated, for these form the ridge that has determined the course of the creek, and the country occupied by the gneiss falls away to the east (Fig. 3). The primary gneissic foliation of the rock and the eastern selvage of acid gneiss and metasomatized country-rock suggest strongly the effects of magmatic differentiation determined by horizontal pressure directed from the west, and acting OR @ magma already partially crystallized. The resemblance to the occurrences described by Barrow from the Scottish Highlands (Barrow, 1892 and 1893) is heightened by the rounded appearance of some of the plagioclase in the quartz-diorite gneiss. That compression was renewed after consolidation is indicated by cataclastic structure, particularly in the acid phases. In places the shearing of the gneiss has been very intense. The characteristics of the intrusion enumerated above reveal it as a bathylith of the synchronous type, though differing in a number of respects from the Cooma bathylith. In detail, however, its structure is rather puzzling, and in particular the shape of its cross-section as deduced from surface-indications. Thus we find that along the western limb the foliation dips consistently to the east, in places at angles as low as 20° and even less. That these dips are those of the mass as a whole is indicated between Barkgunyah and Barkersdale Creeks where the dipping boundary shows well the characteristic relation of outcrop to contour. On the other hand the eastern limb and its extensions dip steeply to the west. The western limb forms relatively low ground and passes beneath the main ridge of granitized schist and Cooma gneiss that extends north from Mittagang Bridge, but. near Murrumbucka Gap it rises to overtop the ridge of older rocks. The narrow middle tongue of blue gneiss projecting south from the Gap forms a valley-depression, and its foliation is vertical or dips steeply under another ridge of Cooma gneiss lying to the east. This is seen to be underlain on the other side by the eastern limb of the intrusion. Farther north where the various tongues have coalesced the foliation on the east dips west with increasing steepness till it becomes vertical, and then takes on an easterly dip with a value gradually declining towards the western margin, the vertical position being attained about three-quarters of a mile from the eastern edge. An attempt is made in Fig. 3 to represent these facts semi-diagrammatically. It would appear on a survey of the available evidence that the intrusion is trough-shaped, and there is a suggestion that the trough is pitching to the south. However these conjectures—for they are little else—relate only to the southern part of the intrusion, and to the surface-outcrops, and no conclusions are justified as to the shape of the mags in cross-section at depth, though it should be pointed out that the easterly dip of the western margin has been observed over a vertical range of nearly 1,500 feet. THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 167 The general structural similarity between the bathylith and the Ordovocian rocks it invades is noted below. It seems certain that the shape assumed by the intrusion was conditioned by either folding or faulting, and it may well be that the fanning out or deployment of the magma as it made its way up was directed by structural features of the enclosing Ordovician strata, which at that time probably formed the floor of the Silurian geosyncline. Within the area examined the bathylith is enclosed entirely by Ordovician rocks, but many miles further north it invades the Silurian beds as well. From the structural peculiarities which distinguish it from the massive granites of Devonian and Carboniferous times it is considered to have been injected during the epoch of folding that closed the Silurian period. Fig. 3.—Sections across the Murrumbidgee bathylith (Plate VI) to show its structure. a=bathylith ; b=—Cooma gneiss and granulite with undetermined dip ; c==slates, phyllites, etc., of the eastern belt ; d—schists and granulites ; e==-Tertiary. basalt. Datum line about 2,000 ft. above sea-level. (c) Porphyries and Porphyrites. (See Figs. 1 and 2.) A bewildering variety of these rocks has invaded both Ordovician and Silurian beds, and their distinction is not always easy. Reference has already been made to the oldest of them, the sheared porphyry or porphyroid which is prominent as sills and lenses among the Ordovician rocks and which is believed _ to be of Ordovician age. Others of the intrusions are thought to be Silurian and others again are post-Silurian. Granite-Porphyries. Along the right bank of the Murrumbidgee are two large lenticular masses of porphyry enclosed by Ordovician strata, the more southerly upwards of six miles and the other three miles long. The rock is distinguished by containing abundant large bipyramidal phenocrysts of quartz up to three-eighths of an inch in length and rather smaller idiomorphic white felspars, with biotite and occasional hornblende. Much of the very altered phenocrystal felspar appears to be plagioclase, but the finely granular groundmass contains orthoclase, and the rocks, though possibly more closely allied to tonalites, may be provisionally called by the general name of granite-porphyry. The intrusive character of the lenses is obvious in the field from their relations to the Ordovician rocks. The outcrop in the parish of Bransby has 168 WwW. R. BROWNE. been mapped as a solid mass, but it really appears to be to some extent a congeries of thin lenses or sills. The rocks have in general suffered shearing or crushing; in the larger intrusions this is mainly peripheral, though in thin section all the rocks are seen to have suffered some strain. Small intrusions of aplite and of coarse pink acid granite or alaskite are closely associated and evidently comagmatic with the porphyries. Apart from the two large intrusions numbers of small ones have been observed, particularly in the parishes of Billilingra and Woolumla, and it has been found that certain of the porphyries occurring at intervals from Bunyan south to Bushy Hill really belong to the same suite. Some of them are massive but others are crushed and may be confused with the porphyroids. It is noteworthy that these intrusions are arranged parallel to and never very far away (less than a mile) from the Murrumbidgee bathylith. The parallelism is so close as to suggest that the porphyry is comagmatic with the blue gneiss, and this feeling is strengthened by the fact that near its northern end in the Capital Territory the granite of the bathylith merges into granite- porphyry with aplite. Moreover, in the country north of Colinton the porphyry invades Silurian strata. That the porphyry was injected and crushed before the intrusion of the bathylith is shown by the occurrence of a small lens of the latter among the crushed porphyry about a mile north of Billilingra House. The granite-porphyries may thus perhaps be regarded as forerunners of the main bathylith during the epoch of intrusion associated with the late Silurian folding. Other Porphyries and Porphyrites. The Silurian and Ordovician strata are invaded by a series of minor intrusions of porphyritic rocks. In general they are somewhat dyke-like but so irregular in habit that for the most part it has been possible to map only zones of intrusion and not individual occurrences. Though transgressive in plan, these are seen to be grouped with a rough parallelism to the trend of the invaded terrain. One well-defined dyke of felsite north of Por. 93, Par. Bransby has been traced in a direct line for upwards of half a mile. Certain porphyries associated with crystal-tuffs, like some of those in Pors. 23 and 44, Par. Billilingra, are much fractured and show signs of epi- dotization, silicification, etc., and these may be of Silurian age, contemporaneous with the tuffs ; others, perfectly massive to outward showing, would appear to postdate the late Silurian orogenic movements: Some of the porphyries are readily recognizable as such in hand-specimen, but, as stated above, there are crystal-tuffs which are easily mistaken for igneous rocks, and it would appear that some of these have been mapped as porphyries. In particular parts of the large bifurcate mass depicted west of the railway in parish Billilingra and shown as intersecting both Ordovician and Silurian rocks, have proved to be of crystal-tuff though other parts are definitely of porphyry, and some at least of the fresh, blue rock at Billilingra Siding and north of it is likewise tuff. Again, there are certain types about which, even when examined in thin section, one cannot be quite sure; these have the large resorbed quartz- grains characteristic of porphyries, but their cryptocrystalline groundmass is thickly bestrewn with smaller angular fragments of quartz and felspar, possibly resulting from the invasion of a tuff by a porphyry. A rock of this type constitutes the mass in the Travelling Stock Reserve three-quarters of a mile south of Bredbo and has been mapped as tuff. Of the true porphyries a number of types may be distinguished in the field. Most common perhaps are those containing abundant phenocrysts of quartz and felspar, with hornblende and sometimes biotite, in a stony groundmass which may vary in colour—bluish, brown or greenish generally. In some types there is little or no megascopic felspar and the quartz phenocrysts are small, and in a few others quartz is missing. Ss THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 169 Light-coloured felsitic rocks, some showing flow-banding, are not uncommon. An intrusion of this type forms the rough, bare and prominent Cosgrove Hill in parish Billilingra and is continued in the larger mass immediately to the north. Quartz-felsites and porphyries also seam the Silurian tuff in Por. 23, Par. Billi- lingra, and these continue north at intervals in the parish of Bransby in the low ground and on the ridge east of the road. Some two miles north of Bredbo is an intrusion of porphyry and felsite between road and railway and a small inter- formational lens of white felsite shows in a railway-cutting. Narrow sills of the same rock traverse the Silurian tuffs two miles farther north. The Silurian strata in the parishes of Callaghan and Woolumla are invaded by a series of irregular intrusions, some of which show a brecciform structure, angular fragments of dark blue porphyry up to about 4 inches in length being embedded in a lighter-coloured felsitic matrix, evidently a later cognate intrusion. No petrographical study of these porphyritic rocks has been made beyond the examination of a few microslides. Felspars in many of them have been altered almost beyond recognition, but plagioclase and orthoclase have both been detected among the phenocrysts. Some of the rocks are probably to be classed as porphyrites but the most acid or felsitic types are true porphyries with dominant alkali-felspar and little or nothing in the way of dark minerals. In certain felsites phenocrystal quartz is almost absent and this mineral is contained mainly in the cryptocrystalline groundmass. Around the intrusions occasional very local hardening and silicification of the country-rock is to be seen, and the igneous rocks themselves may be traversed by quartz-veins. Indeed it appears probable that quartz as veins was introduced with all or nearly all of the igneous intrusions, both plutonic and hypabyssal, so that Ordovician and Silurian rocks are intersected with them and the suriace in places is littered with quartz-fragments. It is noteworthy that the porphyries and porphyrites appear only in the eastern half of the area. No sign of them has been found west of the eastern margin of the Murrumbidgee bathylith. How far eastward they extend has not been ascertained. As to the age of these porphyritic rocks all one can say definitely is that they are post-Silurian. Very similar porphyries and porphyrites are known elsewhere which were injected during a diastrophic epoch at the close of Middle Devonian time, and to this the rocks just described may well belong. (da) Genetic Relations among the Igneous Rocks. It is interesting to examine the possibilities of consanguinity between the Paleozoic rocks of the area. The Ordovician amphibolites and sheared rhyolites and tufis are probably complementary, likewise the rhyolites and basic lavas of the Silurian, and the Ordovician porphyroids may be hypabyssal equivalents of the sheared rhyolites. There is very probably a close genetic relation between the Silurian tuffs, granite-porphyries and intrusive gneisses, and the kinship may even extend to the post-Silurian porphyries, porphyrites and felsites. Indeed the region may have been a petrological province from Ordovician to Devonian time, but in the absence of detailed chemical and petrological studies such a suggestion is little more than speculation. (e) Quartz-Dolerites. A few minor dolerite intrusions of indeterminate age have been noted. One of these, possibly a small volcanic neck, breaks through the Ordovician rocks in Por. 34, Par. Billilingra and a smaller one cuts the Silurian beds in the adjoining Por. 23. An east-west dyke cutting through blue gneiss near its eastern Margin occurs in a tributary of Spring Vale Creek, and a number of other 170 W. R. BROWNE. similar dykes were found in the course of traverses across the Murrumbidgee bathylith west of Bredbo, one of which, just west of the south-west corner of Por. 40, Par. Bullanamang, has a width of 12 yards and extends for at least half a mile in a direction about W. 15° 8S. Some of the dykes have basaltic margins and it is possible that a few small dykes of basalt without olivine, noted in the north-west of Par. Billilingra, may belong to the same series. Under the microscope the rocks are seen to belong to the quartz-dolerite kindred. Some of them have uralitized augite and rather plentiful biotite with interstitial quartz and micropegmatite ; others though devoid of quartz, betray their affinities by the presence of pigeonitic pyroxene. | These rocks are of quite a different petrological type from the Tertiary basic rocks, which they probably antedate, but apart from this and the fact that they are post-Silurian nothing is known as to their geological age. They may possibly be related to other similar rocks which have been found seattered through the highlands (Browne, 1933). GEOLOGICAL STRUCTURE OF THE PALZOZOIC ROCKS. The elucidation of the structure of both Ordovician and Silurian rocks is somewhat difficult, largely because there are few good natural sections across the strike. Moreover the Ordovician rocks have all acquired schistosity or cleavage, and the lavas and tuffs of the eastern zone have no bedding-planes, while the Silurian rocks are largely massive tuffs, and shales which have taken on an imperfect cleavage. No beds have been detected whose repetition might provide a key to the Silurian structure, and it is not even quite certain if the beds in contact with the Ordovician are at the base of the sequence. It is for these reasons that no attempt has been made to estimate thicknesses of strata. In the eastern half of the area many of the dips are high and many more are vertical, hence it is not surprising that with the same strike changes in dip-direction are relatively common ; in the circumstances these changes have little or no significance. The Ordovician belt in its extreme western parts has a predominance of easterly dips, which may be as low as 40°. On the east high westerly dips are the rule, though they may be down to 65°. In the south of the parish of Murrum- bucka along Pilot Creek persistent westerly dips occur, but the beds turn over again very quickly and slope to the east near Mittagang Bridge and along the Murrumbidgee downstream from it. This minor anticline is on the line of that observed south of the Murrumbidgee along Spring Creek, and what is possibly its northern continuation is seen in the schists up along Barkersdale Creek west of the blue gneiss. 7 It would thus appear that the Ordovician strata have been thrown into a rather asymmetrical syncline modified by the subsidiary Pilot Creek anticline. The coming in of the Murrumbidgee bathylith to some extent obliterates and disturbs the folding of the beds, but the structure indicated by the foliation of the blue gneiss, as noted above, is broadly synclinal. Actually the folding of the Ordovician rocks was probably synclinorial rather than synclinal, with smaller crumplings superimposed on the main folds. This interpretation of the structure, if correct, would make the Coolringdon beds lying to the west and east older than the more metamorphosed Binjura beds in the middle. This does not accord with the views of Dr. Joplin, who places the beds in the reverse relation and considers that the Cooma gneiss was injected not into a syncline but into an anticline. Of course the present structure is to be regarded as the result of both the late Ordovician and the late Silurian diastrophism, but how far the latter was effective in modifying the attitude of the rocks it would be hard to say. The slates and chloritic phyllites may well have suffered further deformation along with the Silurian sediments, and indeed THE GEOLOGY OF THE COOMA DISTRICT, N.S.W. 171 such a condition is suggested by the apparent dipping of Silurian under Ordovician beds in the parish of Bransby, but it is most unlikely that any new folding could have been imposed on the recrystallized and rigid granulites and granitized schists, especially when stiffened by the intrusive Cooma gneiss, and the broadly synclinal structure within which the latter is contained at Cooma is therefore probably to be regarded as original—that is to say, Ordovician. On the other hand, since the Ordovician rocks presumably formed the floor of the geosyncline in which the Silurian beds were deposited, it is reasonable to suppose that the folding of the latter by pressure from the west was accompanied by and related to disruption and faulting of the older rocks, and this may be in part responsible for their present attitude, though it must be said that no traces of such dislocation have been observed. The Silurian rocks have not been examined for any great distance across their strike, so that little is known about their structure. In the most westerly beds there is perhaps a preponderance of easterly dips, but in the parish of Bransby this is reversed. Dips are as a rule fairly high and may be vertical, due in part to the fact that the beds in contact with the Ordovician rocks are prac- tically everywhere soft and easily deformed shales, though indeed the tuffs are also in places on end. Low angles were noted only in sandstones some distance away from the boundary in the parish of Callaghan. It is possible that the more resistant units of the series are broadly and gently folded while the weaker shales have been thrown into minor isoclinal folds. This view appears to be supported by the result of a reconnaissance trip along the road from Cooma to Umaralla; about two miles east of Rock Flat Creek some open folding of ripple-marked shales and sandstones was observed and farther on some smaller folds in shale, while the repeated occurrence of bars of brown quartzite suggested folding of these beds on a major scale. STRUCTURAL RELATIONS OF ORDOVICIAN AND SILURIAN ROCKS. This is another question whose solution is hampered by the absence of clear-cut sections across the strike and by the fact that the Silurian slates have been folded along approximately the Ordovician axes. Moreover, in a number of places the actual boundary cannot be exactly defined owing to a covering of soil, or where, as in the parish of Woolumla, there are weathered Ordovician and Silurian slates in apposition. Nevertheless it is indisputable that a big chrono- logical hiatus exists between the two series, representing Lower and probably Middle Silurian time, and the difference in degree of alteration between the Silurian slates and tuffs and the sheared Ordovician tuffs and lavas points to a major tectonic break, but whether overthrust or angular unconformity is not easy to say. Along the boundary the Silurian in places dip east off the Ordovician strata, elsewhere both sets of beds dip east, while again the Silurian may be seen dipping west and the Ordovician east. The fact that many of the dips are high discounts the value of the observations, but in the parish of Bransby _ the westerly dip of both Ordovician and Silurian rocks is rather constant and probably significant. There is a small but definite discordance in strike between the two formations which is particularly noticeable in the parish of Bransby, where the Silurian trend is in general about 15° or 20° west of north and parallel to the boundary, while that of the Ordovician is almost meridional. As a result the Ordovician beds are very gradually truncated northwards, so that beyond the limits of the map the Silurian slates are in contact with the porphyroids, the sheared rhyolites and tuffs having disappeared. The available evidence, though inconclusive, would seem to point to an angular unconformity rather than a faulted junction, with the possibility that 172 W. BR. BROWNE. overturning of the plane of unconformity was accomplished locally through thrusting or folding from the west during the late Silurian diastrophism. It is of interest that in the Yass district, some 80 miles north of Bredbo, Sherrard (1939) and Brown (1941) have figured an angular unconformity between highly cleaved and steeply dipping Upper Ordovician strata and overlying Middle (?) Silurian beds, which are there only very moderately folded. SUMMARY. In continuation of Part I, published in 1914, the geology of a belt of country some 20 miles in meridional extent has been mapped and examined. The Paleozoic formations are Upper Ordovician and Upper Silurian, both containing igneous as well as sedimentary material. There are also remains of Tertiary basalt-flows and Tertiary and Recent sediments. The Ordovician beds are invaded by late Ordovician granite, to which is due in part their metamorphism, and by a late Silurian bathylith with satellitic granite-porphyries, and both Ordovician and Silurian strata are cut by numerous intrusions of porphyry and porphyrite, and by quartz-dolerites. The geological structure of the Palzozoic rocks is discussed and the structural relations of the Silurian and Ordovician formations. It is thought probable that these are separated by an angular unconformity. REFERENCES. Barrow, G., 1892. Geol. Mag., 9 (n.s.), 64. 1893. Quart. J. geol. Soc., 49, 343. Brown, Ida A., 1941. THis Journat, 74, 312. Browne, W. R., 1914. The Geology of the Cooma District. Pt. I. Tuis Journat, 48, 172. 1929. Pres. Add. Proc. Linn. Soc. N.S.W., 54, xii-xvili. 193la. Tuts JoURNAL, 65, 112. 19316. JIbid., 65, xliv. 1933. Pres. Add. THis JouRNAL, 67, 17 and 21. Joplin, Germaine A., 1939. THis JoURNAL, 73, 86. 1942. Proc. Linn. Soc. N.S.W., 67, 156. Sherrard, Kathleen, 1939. Proc. Linn. Soc. N.S.W., 64, 577. EXPLANATION OF PLATE. Geological Map of the Country between Bunyan and Colinton. Note.—Dips shown within the Murrumbidgee bathylith are those of the foliation of the gneiss. Journal Royal Society of N.S.W., Vol. LXX' ra Journal Royal Society of N.8.W., Vol. LXXVII, 1943, Plate VI GEOLOGICAL MAP OF THE COUNTRY BETWEEN 5 ; : & J ; 71) 22 \ yale SINE lis oe eee = BN [a ‘| } \ v x oe =A Ri es mi . ALINGRA f- | \ | \ | Ih | e€ \F \\ ins \ ~ x CANDY I! | [Ak Congreve & L SN hl! ‘ ANG a | yijt \ ‘ WATT | \ ng 4 Ole J ae x 4 i un 2. \"o- TERTIARY F-—1Basalt SILURIAN AAV Siac ele. UMLA‘ RAY geen A — Murrumbidgee ae Ly Bathylith. b—\ L = (ys schists, porphyries, etc, (EAT Mottted Gneiss EESEE8] Cooma Gneiss. » not wee eee OBITUARY NOTICES. JAMES ADAM Dick was born at Windsor, N.S.W., and died at the age of 76 in December, 1942. He graduated in Arts at the University of Sydney in 1886, and then proceeded to Edinburgh, where he obtained the degrees of Bachelor of Medicine and Master of Surgery in 1891, his Doctorate of Medicine in 1892, and Fellowship of the Royal College of Surgeons, Edinburgh, in 1901. In 1893 Dr. Dick returned to Australia, and began practice at Randwick, where he remained until he died, and where he had a large practice and a high reputation as a medical man. He was always actively concerned in advancing the interests of the British Medical Association, and held office in various capacities, including that of President. He was for a number of years a member of the Medical Board of N.S.W., and a Councillor of St. Andrew’s College, University of Sydney, from 1925 to 1937. He was Honorary Medical Officer of the Home for Aged and Infirm, Randwick, and of the Asylum for Children. Dr. Dick was on active service in the Boer War, from 1899 to 1902, and in the Great War from 1914 to 1919. He was mentioned in dispatches, and was made a Companion of the Most Distinguished Order of St. Michael and St. George. As commanding officer of a base hospital in France he enjoyed the respect and confidence of those serving under. him. The records of the Royal Society show that he always had a keen interest in its affairs, and attended its meetings and functions regularly. On several occasions he even sent greetings to the President and members from his base hospital in France. He joined the Royal Society in 1894, and from 1898 to 1901 was Joint Honorary Secretary of its Medical Section. Dr. Dick was known and respected by medical men throughout Australia, and it has been said of him that ‘‘ he showed in his everyday life those qualities of heart and hand which are credited to the true follower of A sculapius and he was prepared to spend much of his life and energy in maintaining at a high standard the corporate life of the profession which he adorned ’’. GERALD HARNETT HALLIGAN was born on 21st April, 1856, and died at Killara on 23rd November, 1942. He was for many years a supervising engineer in the Department of Public Works of New South Wales, and was concerned with harbour and river construction work. He was especially an authority on. tides and currents, of which he made a life study, not only round the coasts of Australia, but of the world. A number of maps were compiled by him in connection with the Oceanography Section of the Pacific Science Congresses. For some years Mr. Halligan was Government Hydrographer of New South Wales. He accompanied the third Expedition organized by the Royal Society of London, to Funafuti, and was in charge of the boring operations, which proved very successful. The results of the expedition were published in 1904 by the Royal Society of London, in the volume “ The Atoll of Funafuti’’. Mr. Halligan joined the Royal Society in 1880, and at the time of his death was the oldest member but one, having had an unbroken membership of sixty-two years. KELSO KING, who died on 7th February, 1943, in his ninetieth year, was born in Sydney on 30th December, 1853. His first training was on a Queensland Station as jackeroo, but he soon returned to the city, where he entered the banking business as a junior in the Bank of New South Wales. He later joined O 174 OBITUARY NOTICES. the Commercial Banking Company of Sydney, where he later became a manager, and an inspector. He also became a director of the Bank of New South Wales. He was managing director of the Mercantile Mutual Insurance Company, of which he had been chief executive officer since its inception in 1877 ; and of the Australian General Insurance Company ; Chairman of Directors of Mort’s Dock and Engineering Company ; a director of the Colonial Mutual Life Assurance Society, and of the Australian Fertiliser Company, and of many other companies. His public and philanthropic activities were legion, and he was associated, in the capacity of chairman and president, with such organizations as the Walter and Eliza Hall Trust, the St. John Ambulance Association, and Brigades, and the Royal Life Saving Society. He always had a keen and sympathetic interest in the youth of the State, and was closely associated with the Boy Scouts’ Association—one of the oldest Scout Troops in the State bears the designation ‘‘Sir Kelso King’s Own ’’—and the Navy League. He was a member of the Council of The King’s School, the Canberra Grammar School, and Trinity Grammar School. Sir Kelso was a member of the Australasian Pioneers’ Club. In recognition of his many services of a public and philanthropic nature, he was created a Knight Bachelor in 1929. ARCHIBALD DURRANT OLLE was born at Beyton, Suffolk, England, on 23rd November, 1868, and died on 9th September, 1942, at Ashfield, N.S.W. Ollé joined the British Navy as a signalman in 1883, and left England in the brig Nautilus in 1886. In Sydney he joined H.M.S. Nelson and later H.M.S. Calliope, the only ship which escaped from Apia Harbour, Samoa, on the historic occasion of the hurricane which burst so suddenly that all other ships, including several warships of other nations, were wrecked and swept ashore. Ollé contracted malaria which resulted in his discharge from the Navy in 1888. He entered the service of the Hamilton Hospital, Victoria, and later entered the Sydney Hospital under Dr. Muskett. His hospital experience enabled him to undertake private work, and he travelled with patients to England on many occasions. He visited Vienna and worked in Germany for two years, returning to Sydney, where he set up a practice in massage, electrotherapy and radiology in 1899, in which he continued until his death. MARCUS BALDWIN WELCH, who died at the age of 47 on 29th September, 1942, was born at Palmerston North, New Zealand, and came to Sydney in 1906. He was educated at Fort Street and Sydney Boys’ High Schools, and in 1916 eraduated at the University of Sydney with first-class Honours in Chemistry, and the University Medal in Botany. On leaving the University he joined the A.I.F., but was discharged by order of the Minister for Defence for special work in explosives in Great Britain. On returning to Australia he was appointed Demonstrator in Botany at the University of Sydney. In August, 1919, he joined the staff of the Sydney Technological Museum, as Assistant Economic Botanist, remaining there until he was transferred to the Forestry Commission in 1936. During 17 years at the Technological Museum, Mr. Welch carried out — very valuable researches in the physical and mechanical properties of Australian timbers, and the results were published in the Journal and Proceedings of the Royal Society of N.S.W. Thirty-nine papers in all were contributed by Mr. Welch to the Journal. He was also author and co-author of several Bulletins of the Sydney Technological Museum. On joining the Forestry Commission Mr. Welch was appointed Senior Research Officer in the Division of Wood Technology, where he rendered excellent service to the timber industry of the State. The rapid development of the OBITUARY NOTICES. 175 Division in a comparatively short time was largely due to the energy, initiative and organizing ability of Mr. Welch. He also rendered useful service to the Standards Association of Australia in the preparation of standard specifications for timber. During the present war Mr. Welch’s work was chiefly connected with defence matters, such as testing timber for aeroplanes and rifle stocks, and experimenting with timbers for producer gas units. He became Chairman of the Producer Gas Committee, and of the Charcoal Research Committee. Mr. Welch was a member of the Royal Society from 1920 until his death, and held office on the Council from 1931 to 1941, and as Honorary Treasurer for 1939 and 1940. He could not be persuaded to accept the office of President, believing that his official post would prevent him from giving the necessary time to the work entailed. His sudden death while on a holiday in the Blue Mountains came as a great shock to his friends and colleagues, who will remember him for his friendly nature, his simplicity, integrity and conscientious attention to his work, in which he never spared himself. In fact his assiduous attention to his work since the war began probably contributed in part to his early death. eae ik a ABSTRACT OF PROCEEDINGS | OF THE Royal Society of New South Wales April 7th, 1943. The Annual Meeting, being the six hundred and third General Monthly Meeting of the Society, was held in the Hall of Science House, Gloucester and Essex Streets, Sydney, at 7.45 p.m. The President, Professor Priestley, was in the chair. Sixty-one members and eleven visitors were present. ‘The minutes of the previous meeting were read and confirmed. The following were elected officers and members of the Council for the coming year : President : A. B. WALKOM, D.sc. Vice-Presidents : Pror. H. PRIESTLEY, ™.p., ch.m., B.Sc. | A. BOLLIGER, Ph.D., A.A.C.1. IDA A. BROWN, D.sc. | H. S. HALCRO WARDLAW, D.sc., F.A.C.1. Hon. Secretaries : Pror. A. P. ELKIN, m.a., Ph.p. | D. P. MELLOR, mse. Hon. Treasurer : A. CLUNIES ROSS, B.sc., F.c.a. (Aust.). Members of Council: G. H. BRIGGS, pD.sc., Ph.p., F.1nst.P. J. E. MILLS, M.sc., Ph.p. J. A. DULHUNTY, B.sc. F. R. MORRISON, 4.4.¢.1., F.c.8. F. P. J. DWYER, M.sc. G. D. OSBORNE, pD.sc., Ph.p. F. LIONS, B.sSc., Ph.p., A.1.C. H. H. THORNE, .a., n.Se., F.R.A.S. W. H. MAZE, .sc. H. W. WOOD, M.sc., A.tnst.P., F.R.A.S. Xxil ABSTRACT OF PROCEEDINGS. The Annual Balance Sheet and Revenue Account were submitted to members by the Honorary Treasurer, and on the motion of Mr. A. Clunies Ross, seconded by Dr. A. Bolliger, were adopted. THE ROYAL SOCIETY OF NEW SOUTH WALES. BALANCE SHEET AS AT 28th FEBRUARY, 1943. LIABILITIES. 1942. £ Trust Funds— Clarke Memorial Fund— Balance as at 28th February, 1942 Add Interest for year ended 28th February, 1943 a Ay ‘ Less Expenses— Lecture Fee .. ih £26 5 0 Printing Lecture Bi ao LOW AO Printing Tickets and Cir- culars and Advertising 310 1 Engraving... se Ob" 6 Walter Burfitt Prize Fund— Balance as at 28th February, 1942 .. Add Interest for year ended 28th February, 1943 aa is Liversidge Bequest— Balance as at 28th February, 1942 .. Add Interest for year ended 28th February, 1943 a Me si Less Expenses— Lecture Fee .. ta £31 10 0 Travelling Expenses. . sa aes Printing Tickets and Cir- culars and Advertising 4 14 6 3,154 14 Subscriptions Paid in ‘Advance £1,861 1943. ane £o) eed. 3 0 3 7 1,798 15 8 5 690 7 5 8 0 8 5 680 16 3 94 Provision for Unexpired Proportion ‘of Life ‘Membership Subscriptions 27,820 Accumulated Fund Contingent Liability—In connection with perpetual Leases granted to the Australian National Research Council and the Pharmaceutical Society of New South Wales. £31,082 28,284 £31,552 280 8,839 14,650 15 £31,082 ABSTRACT OF PROCEEDINGS. ASSETS. Cash at Bank and on Hand— The Union Bank of Australia Ltd. Commonwealth Savings Bank of Australia Petty Cash Aye as Bonds and Inscribed Stock— Bonds (Face Value £1,000) Stock (Face Value £8,060) Science House Management Committee— ' Payments to date Sundry Debtors— Subscriptions Unpaid Less Reserve ae Library Furniture ae os ae Less Depreciation written off Pictures ; a it ai Less Depreciation written off Lantern ‘ Me ne a Less Depreciation written off 410 19 54 18 4 1 LOL 5 8,027 11 0 3 Se) He He Xxlil 469 19 5 9,038 16 3 14,756 0 0 65800" 0)".0 436 0 0 37 10 0 14 0 O £31,552 5 8 The above Balance Sheet has been prepared from the books of account, accounts and vouchers of The Royal Society of New South Wales, and is a correct statement of the position of the Society’s affairs on the 28th February, 1943, as disclosed thereby. We have obtained certificates showing that the whole of the Bonds and Inscribed Stock are held by the Society’s bankers for safe keeping. Prudential Building, HORLEY & HORLEY, Chartered Accountants (Aust.). 39 Martin Place, Sydney, 15th March, 1943. (Sgd.) A. CLUNIES ROSS, Hon. Treasurer. d. XXIV ABSTRACT OF PROCEEDINGS. REVENUE ACCOUNT FOR THE YEAR ENDED 28th FEBRUARY, 1943. Year ended 28th Feb., Year ended 1942. 28th February, 1943. £ £ sides Ss. 4 To Advertising se Me a ve sets dea —— 36 ,, Cleaning.” ... He ue aN nd sik sie 3607.0 25 ,, Depreciation ou ain ay ae ed ve 25 13 9 1l_,, Electric Light cot es ne Ks Les 3S 1 GS 15 ,, Insurance ys oe ay ie wad Mes 1412 6 36 =~, Library Maintenance Bs ys ah ad aa 56 14 9 63 ,, Miscellaneous Expenses .. ie ne oa ne 45 8 0O 292 =«,, Office Salaries and Audit Fees oe ips i 305 9 6 29 ~~,, Office Sundries and Stationery a es ie 21° 9°"8 164, .,, Bunting ce Hi a ae 82 4 3 453 ~=—,, Printing and Publishing J Journal we oi) Ok ae Oar 6 ,, Repairs ta — 55 ,, Stamps and Telegrams Yo ae © Equine 16 72.) Delephoue (- 20 14 9 1,205 ———— 941 16 ,» Balance, being Net Revenue for the Year, transferred 683 to Accumulated Fund 481 5 £1,888 £1,423 2 Year ended 28th Feb., Year ended 1942. 28th February, tee: 5 £ Ee eA aah 23 504 By Members’ Subscriptions 485 3 800 ,, Government Subsidy 400 0O 325 », Science House Receipts 315 0 O 4% Less Rent Paid 30, ETS — 283 279 18 74 ,, Miscellaneous Receipts 38 13 324 ,, Interest Received one) tan a Less— Clarke Memorial Fund .. £7 a2 0 Walter Burfitt Prize Fund 26: 1b VO Liversidge Bequest 28 0 0 127 126 3 0 a 197 —— 196 8 30 ,, Proportion of Life Members’ Subscriptions 23 (0 £1,888 fi Ags 2 ACCUMULATED FUND ACCOUNT FOR THE YEAR ENDED 28th FEBRUARY, 1943. 1943—February 28— £ S. To Arrears of Subscriptions, written off .. 21 O , Balance, carried down 28,284 0 £28,305 0O 1942—February 28— £ S. By Balance from last Account 27,819 15 1943—February 28— By Amount Transferred from Bad Debts Reserve Account .. 3 19 ,», Net Revenue for the Year 481 5 £28,305 0 1943—February 28— By Balance, Brought Down £28,284 0 ABSTRACT OF PROCEEDINGS. xXxXV The Annual Report of the Council (1942-43) was read, and on the motion of Professor Elkin, seconded by Mr. R. W. Challinor, the report was adopted. REPoRT OF THE CounciL, 1942-1943 (RULE X XVI). We regret to report the loss by death of seven members since April Ist, 1942: James Adam Dick (1894), John Clifford Firth (1935), Gerald Harnett Halligan (1880), Sir Kelso King (1896). Archibald Durrant Ollé (1913), Marcus Baldwin Welch (1920), and Sir Joseph J. Thomson, an Honorary Member since 1915. By resignation the Society has lost six members: Harry Williams, Francis W. Firth, Thomas H. Tennant, Stanley W. E. Parsons, Harold Tindale and Robert T. Wade. During the year, fifteen members with a continuous membership of thirty-five years were elected to life membership, without further payment of fees, namely Frank Lee Alexander, Charles Anderson, Horatio Scott Carslaw, Henry Harvey Dare, James Adam Dick, Edward William Esdaile, Mark Foy, Henry Ferdinand Halloran, George Harker, Sir Kelso King, Charles A. L. Loney, Arthur Marshall McIntosh, Cecil Purser, Oscar Ulric Vonwiller, and Walter George Woolnough. Seven new members have been elected during the year, and the membership now stands at 292. The new members are Jack Leslie Still, Desmond J. Brown, John Conrad Jaeger, Neville Allan Gibson, Raymond Norman Matthew Lyons, Gordon Roy Williams and Arthur Lippman. Several members are on active service abroad, others are with the home forces. Brigadier H. B. Taylor has been reported prisoner of war in Malaya. Many of our members are engaged in important war work and on research connected with the war effort. In conformity with the resolution passed during 1941, no Annual Dinner was held during 1942. Eleven ordinary meetings of Council and one special meeting have been held during the year beginning April Ist, 1942, at which the average attendance was 14. Attendances of individual members of Council have been as follows: Professor H. Priestley, 11; Dr. Ida Brown, 11; Mr. J. A. Dulhunty, 11; Dr. F. Lions, 11; Professor A. P. Elkin, 10 ; Mr. W. H. Maze, 10; Mr. F. R. Morrison, 10; Dr. A. Bolliger, 9; Mr. F. P. J. Dwyer, 9; Dr. J. E. Mills, 9; Mr. D. P. Mellor, 8; Dr. G. D. Osborne, 8; Mr. H. H. Thorne, 8; Dr. H.S. H. Wardlaw, 7; Mr. E.J. Kenny, 6; Mr. A. Clunies Ross, 6; Dr. C. H. Briggs, 5; Dr. A. B. Walkom (elected August), 5; Dr. C. Anderson (resigned July), 1. Number of meetings: 11. During the same period nine general meetings have been held, with an average attendance of 33. Twenty-eight papers were accepted for reading and publication, and a short talk on “ The Native Peoples of the Australian Territories ’’ was given by Professor A. P. Elkin. Sympostum.—At the monthly meeting in July, a symposium on Rubber was held, the following being the speakers and subjects : ‘“Some Important Natural Rubber Resources’”’, D. H. Priestley (Manager, Dunlop Rubber Company’s Factory). ‘“Some Aspects of the Chemistry of Natural and Synthetic Rubber ”’, F. Lions, B.Sc., PhD. ‘* Production and International Control of Rubber ’”’, W. H. Maze, M.Sc. ‘“* Possible Plant Sources of Rubber in Australia ’’, F. R. Milthorp. Great interest was shown in the subject, and over sixty members and visitors were present, a number of whom took part in the discussion which followed the addresses. Popular Science Broadcasts—Owing to war-time conditions, including the very restricted lighting in the city and suburbs, it was thought advisable to dispense with the usual series of Popular Science Lectures, as it was considered that the attendance would be limited, and would not justify invitations to busy scientists to deliver lectures. Instead it was proposed to have a series of broadcast talks from the Australian Broadcasting Commission. The Commission responded cordially to the suggestion, and four broadcasts were given during July, as follows : July 7th.—* Vitamins and the Loaf of Bread ”’, Professor H. Priestley, M.B., Ch.M. July 14th.—" Chemical Wonders: Glass from Coal, Rubber from Petroleum’”’, A. R. Penfold. July 21st.—‘‘ Minerals in Peace and War ’’, G. D. Osborne, D.Sc. July 28th.—** Chemical Harvest of the Sea’’, D. P. Mellor, M.Sc. Clarke Memorial Lecture.—The lecture for 1942 was delivered by Mr. E. C. Andrews on May 21st, and the subject was “‘ The Heroic Period of Geological Work in Australia ”’. Clarke Memorial Medal for 1942.—The Medal was awarded to Dr. William Rowan Browne, of the University of Sydney, for his research work during past years, and also for his work in preparing for publication a large amount of material left by the late Sir Edgeworth David for the book on the Geology of the Commonwealth, on which Sir Edgeworth had been working for some years before his death. XXVl ABSTRACT OF PROCEEDINGS. Liversidge Lecture.—Two lectures were delivered under the Liversidge Bequest, by Dr. J. 8. Anderson, Senior Lecturer in Inorganic Chemistry in the University of Melbourne, as follows : October 27th.—** The Chemistry of the Earth.” October 29th.—‘‘ The Imperfect Crystal.” Galileo-Newton Tercentenary.—This double historical anniversary was commemorated at the meeting of October 7th, and an address entitled “‘ Galileo and Newton: Their Times and Ours ”’ was given by Professor O. U. Vonwiller. There was an attendance of fifty-seven. Finance.—The audit of the Society’s accounts discloses that the finances are in a satisfactory condition. Government Grant.—A government grant from the Government of New South Wales of £400 for the year 1943 has been made. Science House.—The Royal Society’s share of the profits on Science House during the period April Ist, 1942-March 3lst, 1943, has been £315. From this, however, must be deducted the cost of the air-raid shelter which was built in the Small Hall in order to comply with National Security Regulations. The Royal Society’s share of the cost was £106. In accordance with the regulations it was necessary for the glass to be removed from the doors of the Council Room and of the office, and to be replaced with masonite. The windows in the Society’s rooms facing Gloucester Street were treated by covering them with cellophane to prevent splintering in case of air raids. Time of General Meetings——Owing to the “ black-out’’ conditions in city and suburbs, it was decided to hold the general meetings at an earlier hour, and the hours of 4.30 and 6.30 were tried at successive meetings. This alteration did not prove popular, and resulted in a great decrease in attendance. It was resolved to revert to the hour of 7.45. Science House Management Commiitee—The Royal Society has been represented at meetings of the Science House Management Committee by Mr. M. B. Welch and Mr. A. R. Penfold, with Dr. G. D. Osborne and Dr. C. Anderson as substitute representatives. On the death of Mr. Welch, Dr. G. D. Osborne was appointed as representative and Mr. Clunies Roos as substitute representative. On the resignation of Dr. C. Anderson, Dr. F. Lions was appointed as substitute representative in his place. Alteration of Rules.—The committee appointed to consider the rules and make recom- mendations as to their revision met on a number of occasions and considered the question of revision of each rule very thoroughly. The revised rules were finally approved, and have now been printed and distributed to members. It is hoped that by this revision the rules have been made clearer and that a number of inconsistencies which existed have been removed. Purchase of Periodicals and Binding.—The amount of £29 13s. 9d. has been spent on pur- chasing periodicals during the year under review, and the amount of £27 ls. has been spent on binding. The total amount expended on the library was thus £56 14s. 9d. A number of volumes are still at the binders, and others are prepared and waiting to go. Exchanges.—The number of exchanges sent has been still further reduced, and now stands at 221, chiefly within the British Empire. It is possible still to send to South America, Mexico, Hawaii, the United States, Palestine, Russia, Spain and Switzerland. A few societies have asked that our Journal be held for them and sent after the war, and such societies are holding their publications back also until such time as they can be safely forwarded. I¢ is noteworthy, however, that only a small number of periodicals seem to have been lost in transit since the war began. Accessions.—For the twelve months ended February the number of accessions entered in the catalogue is 1,741 parts of periodicals and 88 whole volumes. Borrowers and Readers.—Members and visitors reading in the library during the year numbered. thirty-five. The number of books and periodicals borrowed by members, institutions and accredited readers was eighty-one. Among the institutions which made use of the arrangements for inter-library borrowing were: Standards Association of Australia, National Standards Laboratory, National Museum, Melbourne, the Universities of Sydney, Queensland and Mel- bourne, Division of Economic Entomology, Canberra, Food Preservation Laboratory, Homebush, MacMaster Laboratory, Sydney, The Colonial Sugar Refining Co., the Australian Gas Light Company, Amalgamated Wireless Limited, the Public Works Department, Sydney, the Depart- ment of Home Security, and the Sydney County Council. General.—In accordance with decisions taken at the end of 1941, further numbers of old text-books were given to the Medical School, Botany Department and Geology Department in the University of Sydney. Some extra shelf accommodation is thus available, but it will be necessary to take still further action to dispose of duplicates and books which are not needed in a periodical library, in order to make room for the increase of periodicals year by year. Recommendations.—It is recommended that the Council proceed with the proposed scheme of making the libraries of the Royal Society and the Linnean Society of New South Wales comple- mentary. i i, ) : F ‘ Se ee, ee Ee ee ABSTRACT OF PROCEEDINGS. xxvii The deaths were announced of the following members: Dr. James Adam Dick, a member since 1894, and Sir Kelso King, a member since 1896. The certificates of three candidates for admission as ordinary members of the Society were read for the first time. The certificate of a candidate for admission as an ordinary member of the Society was read for the second time. The following person was duly elected an ordinary member of the Society : William Kevin McCoy. Clarke Memorial Medal.—The announcement was made of the award of the Clarke Memorial Medal for 1943 to Dr. W. L. Waterhouse. Election of Auditors.—On the motion of Professor Elkin, seconded by Mr. Challinor, Messrs. Horley and Horley were re-elected auditors to the Society for 1943-44. Library.—The following donations were received: parts of periodicals, 557; 35 whole volumes, and 202 back numbers. The President, Professor H. Priestley, delivered his address, entitled *‘ Life and Living ”’. Professor Priestley then installed Dr. A. B. Walkom as President for the year 1943-44. Dr. Walkom thanked the members for the honour they had done him in electing him President. He then called upon Dr. Lions to propose a vote of thanks to the retiring President for his address and his work for the Society during his term of office. This was carried by acclamation. The following papers were read by title only: “Nova Puppis 1942”, by H. W. Wood, M.Sc., A.Inst.P., F.R.A.S. ““A Polyhedral Model of the Projection Plane’’, by F. A. Behrend. (Communicated by. Professor H. 8. Carslaw.) May 5, 1943. The six hundred and fourth General Monthly Meeting of the Royal Society of New South Wales was held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Twenty-three members and two visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of three candidates for admission as ordinary members were read for the first time. The certificates of three candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : William Dudgeon, Ronald Arthur Plowman, and Leonard Winch. Inbrary.—The following donations were received : 88 parts of periodicals, six whole volumes and eight back numbers. Correspondence.—A letter was received from Buckingham Palace, conveying the sincere thanks of His Majesty the King and the Duchess of Kent, for sympathy in the death of the Duke of Kent. The following paper was read : ** Preliminary Notes on Solution-cracking Treatment of Torbanite ’’, by J. A. Dulhunty, B.Sc. > Lecturette.—A lecturette on “ Reflections of Light from Film-Covered Glass Surfaces ”’ was given by Mr. J. Bannon, B.Sc. June 2, 1943. The six hundred and fifth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Thirty-three members and one visitor were present. The minutes of the previous meeting were read and confirmed. The death was announced of Dr. James Edward Mills, a member since 1940. The certificates of three candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : Thomas Iredale, Ivan Stewart Turner and Jean Annie Kimble. Clarke Memorial Lecture—It was announced that the Clarke Memorial Lecture for 1943 would be delivered at Science House on June 23rd, by Dr. H. G. Raggatt, the title being “ Australia’s Mineral Industry in the Present War ”’. Inbrary.—The following donations were received: 85 parts of periodicals and seven whole volumes. ; The following papers were read : ““Tabulata and Heliolitida from the Wellington District, N.S.W.”, by O. A. Jones, M.Sc. (Communicated by Dr. Ida A. Brown.) “The Etch Figures of Basal Sections of Quartz : Their Use in the Orientation of Water- worn Crystals’”’, by F. N. Hanlon, B.Sc., Dip.Ed. XXVili ABSTRACT OF PROCEEDINGS. Lecturettes.—The following lecturettes were given : ‘‘ Evaporated Metal Films’, by F. P. J. Dwyer, M.Sc. ** Biotin ’’, by Professor H. Priestley, M.D., Ch.M., B.Sc. Popular Science Lectures.—The Popular Science Lectures for 1943 were announced, as follows : Thursday, June 17th.—‘‘ How We Came to Stand Upright ’’, by Professor Harvey Sutton, O.B.E., M.D., Ch.B., D.P.H., B.Sc. Thursday, July 15th.—* Veterinary Science and the Community ”’, by H. Parry, B.A. Thursday, August 19th.—** Exploring the Inside of the Earth ’’, by Professor L. A. Cotton, M.A., D.Sc. Thursday, September 16th.—*‘ Soviet Research on Applied Botany ”’, by Professor E. Ashby, D.Se., A.R.C.S., D.I.C. Thursday, October 21st.—‘‘ Architecture : The Setting for Life ’’, by Professor L. Wilkinson, F.R.I.B.A., F.R.A.LA. July 7, 1943. The six hundred and sixth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Twenty-seven members and two visitors were present. The minutes of the previous meeting were read and confirmed. The certificate of a candidate for admission as an ordinary member was read for the first time. Inbrary.—The following donations were received: 87 parts of periodicals and three whole volumes. Commemoration of the Bi-centenary of the Birth of Sir Joseph Banks.—The meeting was devoted to commemoration of the bi-centenary of the birth of Sir Joseph Banks, and an address, entitled ‘Sir Joseph Banks and Australia ’’, was given by Dr. G. Mackaness, M.A. August 4, 1943. The six hundred and seventh General Monthly Meeting of the Royal Society of New South Wales, held in the Hal! of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Fifty members and four visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of four candidates for admission as ordinary members of the Society were read for the first time. The certificate of a candidate for admission as an ordinary member of the Society was read for the second time. The following person was duly elected an ordinary member of the Society : ‘Reginald John Nelson Whiteman. Library.—The following donations were received: 180 parts of periodicals and 44 whole volumes. The following papers were read : ‘““An Elementary Proof of R. A. Fisher’s Distribution of the Coefficient of Normal Correlation with Some Introductory Notes on Correlation and Regression ’’, by D. T. Sawkins, M.A., B.A. (Read by title only.) “The Production of Hyoscyamine from Duboisia Species— Part I. ‘*‘ Methods of Quantitative Estimation ”’, by J. A. Lean, M.P.S., and C. S. Ralph, B.Sc. Part II. ‘“‘ Extraction of the Base’’, by C. 8. Ralph, B.Sc., and J. L. Willis, B.Sc. Exhibit.—Mr. D. P. Mellor showed an exhibit: A New Source of Light: the Fluorescent Lamp. Lecturette.—Dr. F. Lions gave a lecturette, entitled ‘‘ Penicillin and Gramicidin ” September 1, 1943. The six hundred and eighth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, 157 Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Twenty-four members and two visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of three members for admission as OUIMRDEY members of the Society were read for the first time. The certificates of four members for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : Robert Desider Louis Frederick, Ernest Patrick Molloy, John William George Neuhaus and James Foote Walker. ’ i s hf 4 : 3 : i ABSTRACT OF PROCEEDINGS. XxX1x Library.—The following donations were received: 81 parts of periodicals and four whole volumes. The following papers were read : ‘* Studies on Colour Reactions for Sugars. Part I. The Identification and Deter- mination of Monosaccharides with Thymol, Hydrochloric Acid and Ferric Chloride ”’, by A. Bolliger, Ph.D. ‘‘Ebonite as a Radiometer. Part I. The Distortion of Ebonite by Long Infra-red Radiations and New Methods of Radiation Measurement ’’, by G. G. Blake, F Inst.P., M.I.E.E. Lecturette.—A lecturette was given by Mr. H. H. Thorne on “ Edmund Halley ”’. October 6, 1943. The six hundred and ninth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Thirty-five members and visitors were present. The minutes of the previous meeting were read and confirmed. The death was announced of Dr. John Job Crew Bradfield, a member since 1922. A certificate of a candidate for admission as an ordinary member of the Society was read for the first time. The certificates of three candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : Mrs. Daphne Luber, Howard Hamlet Gordon McKern and John Kenneth Moore Simpson. Inbrary.—-The following donations were received: 88 parts of periodicals and four whole volumes. Presentation of the Clarke Memorial Medal.—The President presented the Clarke Memorial Medal for 1943 to Dr. Walter Lawry Waterhouse, for his great services to science, especially his studies of rust in wheat. The following papers were read : “The Chemistry of Bivalent and Trivalent Iridium. Part I. Compounds of Bivalent Iridium Halides with Tertiary Arsines ’’, by F. P. Dwyer, M.Sc., and R. 8. Nyholm, M.Sc. ‘““ Stringocephalid Brachiopoda in Eastern Australia’, by Ida A. Brown, D.Sc. Lecturette—A lecturette on “‘ The Future of the Native Peoples of the South-west Pacific ”’ was given by Professor A. P. Elkin. November 3, 1948. The six hundred and tenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. . The President, Dr. A. B. Walkom, was in the chair. Thirty-eight members, four visitors, and a guest speaker, Dr. Cowlishaw, were present. The minutes of the previous meeting were read and confirmed. The death was announced of Sir Archibald Howie, a member since 1936. The certificates of two candidates for admission as ordinary members of the Society were read for the first time. The certificate of a candidate for admission as an ordinary member of the Society was read for the second time. The following person was duly elected an ordinary member of the Society : Alexander Campbell Nicol. Royal Society’s Medal.—It was announced that the Royal Society’s Medal had been awarded to Mr. Edwin Cheel for his contributions in the field of botanical research, and to the advancement of science in general. Inbrary.—The following donations were received: 101 parts of periodicals, six whole volumes and 18 back numbers. The following paper was read : “The Vibrations of Square Molecules. Part I. The Normal Coordinates and Vibration Frequencies of Planar AB, Molecules ’”’, by Allan Maccoll, M.Sc. Celebration of 400th Anniversaries.—The meeting was chiefly devoted to the celebration of the 400th anniversary of the publication of two works : “De Revolutionibus Orbium Coelestium ’’, Libri VI, by Copernicus and ‘““De Humani Corporis Fabrica ”’ (Libri Septem), by Vesalius. An address on Copernicus was given by the Rev. Father D. J. K. O’Connell and one on Vesalius was given by Dr. Leslie Cowlishaw. Both addresses were illustrated with lantern slides. XXX ABSTRACT OF PROCEEDINGS. December 1, 1943. The six hundred and eleventh General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Thirty-two members and one visitor were present. The minutes of the previous meeting were read and confirmed. The certificates of two candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected as ordinary members of the Society : Barbara Joyce Burfitt and John Stanton Burkitt. LIibrary.—The following donations were received: 108 parts of periodicals and nine whole volumes. Presentation of Royal Society’s Medal.—The President presented the Royal Society’s Medal to Mr. Edwin Cheel. The following papers were read : ‘* Further Determination of Specialisation in Flax Rust caused by Melampsora lini (Pers.) Lév.”’, by W. L. Waterhouse, D.Sc.Agr., and I. A. Watson, Ph.D., D.Se.Agr. This paper was read by title only in the unavoidable absence of both the authors. ‘“ A Study of the Magnetic Behaviour of Complexes Containing the Platinum Metals ”’, by D. P. Mellor, M.Sc. ‘“'The Geology of the Cooma District, N.S.W. Part II. The Country between Bunyan and Colinton ’”’, by W. R. Browne, D.Sc. ABSTRACT OF PROCEEDINGS GeO) ONE: ) ¢ Chairman: Dr. W. R. Browne. Honorary Secretary: Mr. J. A. Dulhunty. Hight meetings were held during the year 1943, the average attendance being twelve members and three visitors. Meetings. April 16th.—Annual Meeting. Election of Office-bearers for 1943 : Chairman, Dr. W. R. Browne, and Honorary Secretary, Mr. J. A. Dulhunty. Business: Exhibits: By Mr. Lambeth: Defects in glass due to formation of metamorphic minerals during processing. By Mr. Dulhunty: Rich torbanite from Marangaroo, assaying 236 gallons of oil per ton. Address by Dr. W. R. Browne, “‘ Origin and History of the Tasman Geosyncline’’. The structure persisted from Cambrian to Permian time, after which it lacked geographical continuity and moved westward. May 21st.—Address by Mr. C. A. Sussmilch, “‘ The Physiographic Criteria of Faulting, with Special Reference to Eastern Australia’. Distinction was made between monaclinal fold searps and true fault scarps. Evidence of faulting was detailed, and the application of physiographic criteria to problems in Eastern Australia was discussed. June 18th.—Address by Mr. J. A. Dulhunty, “‘ Recent Research on the Nature and Origin of Coal, with some applications to New South Wales’. Theories for the differentiation of organic débris into coals of different type and rank. The influence of environment of accumulation. Constitution of coals. Metamorphism of coal-forming débris. Applications to New South Wales coals. July 16th.—Address by Mr. W. H. Maze, “ Landform Analysis from Topographic Maps ’’, Frequency of occurrence of surfaces at different levels, generalised contours, true profiles, and applications to the Bathurst district were discussed. August 20th.—Address by Mr. F. N. Hanlon, “‘ Piezo-electric Quartz’. The address dealt with properties, uses, sources of supply, technique in processing, and orientation by means of etched surfaces. September 17th.—Exhibits and Notes: By Dr. Brown: Devonian brachiopods from eastern Australia. By Miss Crockford: A mid-paleozoic ostracod fauna from the Yass district. By Dr. Osborne: An illustration of spiral garnets. By Miss Quodling: Surface oxidation and tarnish of some minerals. By Mr. Kenny: Oil shale from Mount Kembla. By Mr. Lambeth: Note on currents in tanks of molten glass and similar phenomena in nature. By Dr. Browne: Gneissic granites from Cooma, Tharwa, Wheeo, Adelong and Kosciusko. October 15th.—Address by Dr. G. D. Osborne, ‘“* Recent Researches in Experimental Geology and their Significance’. An account was given of the work of Bridgman on the influence of high temperatures and pressures on minerals, also the researches of Griggs, and the contributions by Daley to geophysics. November 19th.—Exhibit by Dr. Osborne: Quartzite pebble from Upper Kuttung tillite of Dunn’s Creek. Address by Dr. I. A. Brown, “‘ The Permian Problem : Recent Researches affecting Australian Correlations’. Reference was made to results of the Permian Conference during the XVIIth International Geological Congress, Moscow, U.S.S.R., 1937, and recent research on Permian problems in Australia and other parts of the world. INDEX. A Page. Abstract of Proceedings. . LP Pe >< Section of Geology . {ox Annual Report and Balance Sheet .. xxii Australia’s Mineral Industry in the Present War .. - te soe Awards of— Clarke Memorial Medal for 1942 .. xxv Clarke Memorial Medal for 1943 XXV1i Royal Society’s Medal sits Hiie.® 40,< B Balance Sheet... ee Ee TSAI Behrend, F. A.— A Polyhedral Model of the Pee oe Plane <. 20 Bequest, Form of a bie ee UN Blake, G. G.— Ebonite as Radiometer. The Dis- tortion of Ebonite by pig Infra- red Radiations S% . 106 Bolliger, A.— Studies on Colour Reactions for Sugar. Part I a LOS Brachiopoda in_ Eastern Australia, Stringocephalid ais e ee LO Brown, I. A.— Stringocephalid on ae in Eastern Australia .. se LI) Browne, W. R.— The Geology of the Cooma District, Nese Part It .. ats 156 Burfitt Prize oe i iy ee. ©. CG Chemistry of Bivalent and Trivalent Iridium. Part I. Compounds of Bivalent Iridium Halides’ with Tertiary Arsines.. oe «116 Clarke Memorial Lecture— Australia’s Mineral aera in the Present War as ag Oe Clarke Memorial Medal . .- XXV Colour Reactions for Sugars, Studies on 109 Complexes Containing the Platinum Metals, A Study of the Magnetic Behaviour of 145 Compounds of Bivalent Iridium Halides with Tertiary Arsines ; 116 Cooma District, N.S.W., The Geology of. Part II. The Country Between Bunyan and Colinton vie pen baoO P D Page. Duboisia Species, Production of Hyo- scyamine from— Part I. Methods of Quantitative Estimation .. ss, 196 Part II. Extraction of the Base bo BE Dulhunty, J. A.— Preliminary Notes on Solution Crack- ing Treatment of Torbanite. . eis ee: Dwyer, F. P., and Nyholm, R.S.— Chemistry of Bivalent and Trivalent Iridium. Part I. Compounds of Bivalent Iridium MHalides’ with Tertiary Arsines... LLG E Ebonite as Radiometer .. 106 Etch Figures of the Basal Sections of Quartz: Their Use in the Orienta- tion of Water-worn Crystals ant 40 Exhibits .. te a ay XXVIii F Flax Rust, Further Determinations of Specialisation in, Caused by Melampsora Lint (Pers.) Lév. .. 138 Further Determinations of Specialisation in Flax Rust Caused by na geile Innit (Pers.) Lév. .. 138 G Geology of the Cooma District, N.S.W. Part II. The Country Between Bunyan and Colinton hae .. 156 Government Grant ee AMA ..XXVi1 Guide to Authors Hah oe ae Vv H Hanlon, F. N.— The Etch Figures of Basal Sections of Quartz: Their Use in the Orienta- tion of Water-worn Crystals .. 40 I Infra-red Radiations, Distortion of Ebonite by Long .. . 106 Iridium, Chemistry of Bivalent and Trivalent Iridium. Part I .. 116 XXXIV J Page. Jones, O. A.— Tabulata and MHeliolitida from the Wellington District, N.S.W. -. 3d L Lean, J. A., and Ralph, C. S.— Production of Hyoscyamine from Duboisia Species. Part I. Methods of Quantitative Estimation oie /290 Part II. Extraction of the Base 2, 99 Lecturettes ; : XXVIli Lafe and Lavine 3.05), 0-5. iy: Bs 2 List of Members .. ak Ais oe ER M Maccoll, Alan— Vibration of Square Molecules. Part I 130 Magnetic Behaviour of Complexes Con- taining the Platinum Metals .. 145 Melampsora Lint (Pers.) Lév., Further Determinations of Specialisation Caused by .. a a - el doo Mellor, D. P.— A Study of the Magnetic Behaviour of Complexes containing’ the Platinum Metals hee i pg W859 Members, List of ix Mineral Industry in the “Present War, Australia’s .. er ite Monosaccharides, Tleteaa en ane De ‘ termination of with Thymol, Hydro- chloric Acid and Ferric Chloride.. 109 N Nova Puppis, 1942 . SEN Nyholm, R. 8.—See Dwyer, Fy PR. 5,116 O Obituaries .. > ms Ae eon Officers for 1943- 44 Aes vy 2 ioe P Platinum Metals, A Study of the Magnetic Behaviour of Complexes containing .. Eu BQUiO on iets eee Polyhedral Model of the Projective Plane .. ‘ ay le hh po) Popular Science Broadcasts Hy XKV Preliminary Notes of Solution Cracking Treatment of Torbanite .. “(oi 2A: Presidential Address. H. Priestley .. 1 Priestly, H.—-Presidential Address... 1 Production of Hyoscyamine from Duboisia Species. Part I So oe Production of Hyoscyamine from Duboisia Species. Part I 99 Projective Plane, che em Lae Model of the... 20 AUSTRALASIAN MEDICAL PUBLISHING COMPANY LIMITED INDEX. Q Page. Quartz, Etch Figures of Basal Sections of 40 R Radiometer, Ebonite as My .. 106 Raggatt, H. G.— Australia’s Mineral Industry in the Present War. Clarke Memorial Lecture We Lg yang Ralph, C. $.—See Lean, ‘ig ee sae wie Ralph, C. S., and Willis, J. L— Production of Hyoscyamine from Duboisia Species. Part II. Ex- traction of the Base ee mi) OO Regression and Correlation, Simple .. 85 Report of Council, 1942-1943 .. 4 Revenue Account ni Ms vie RRAV Ss Sawkins, D. T.— Simple Regression and Correlation... 85 Simple Regression and Correlation .. 85 Society’s Medal and Money Prize > 9? 99 9? 1879, >? 255, >> 99 XIV > a 99 99 99 1880, 99 391, 99 9? XV 29 be 99 99 9? 1881, 9 440, 9? 9” XVI > 99 99 be 99 1882, 9 327, 99 99 XVII 99 be be] 73 99 1883, »”? 324, 39 9° XVIII PB) 9? 9) 99 99 1884, 9 224, 99> 9) xIx >» >”? 99 9 9? 1885, °,, 240, bed 99 xx 99 be 9? x” be) 1886, ” 396, 99 ” XXI ” ? 29 ” De) 1887, ” 296, ”? 99 XXII 29 ” ” ” ” 1888, i) 390, 9 99 XXIII ” a9 ” ? a? 1889, ” 534, ” 9? XXIV 7” 9? 9? 99 3) 1890, bt 290, 3:9. Bi] xXXV +B] be) 99 29 be 1891, 99 348, > \ a 9 XXVI be 9? 99 9? ” 1892, be) 426, 99 +] XXVIII 37> ? 9? J'9. 99 1893, 9) 530, ‘7 i) XXVIT >? be) be] 9? 9” 1894, + 368, be) ” XXIX ”” »”? ? be) 9 1895, ” 600, 27 ” XXX a) 9 e) ” 9 1896, 9 568, ” 9 XXXI 2? 9 ” ” 9 1897, ” 626, ” 2? XXXIT > 9°. be “ED 99 1898, ” 476, 9? 7 XXXUI ” 2? 29 ” ” 1899, ” 400, ” 9? XXXIV 99 be) +] 9? 23, 1900, be 484, 9” 99 XXXV >” > >? >B] ”” 1901, >” 581, ” + XXXVI > be 99 9°) 9 1902, 9%? 531, 99 be XXXVIT 9? 9? > >P] BP: 1903, ” 663, “7 >) XXVIII a Er We we Be 1904, ,, 604, fy 9? XXXIX 9? 99 > be] ””> 1905, 9”? 274, 99 9 XL > > >? ”> > 1906, ” 368, bi] 39 XxLlI 9 %? ”” 3) > 1907, ” 377, 9? 99 XLII 99 9) bed > > 1908, 9 593, > J ) XLII be > eI 9° ” 1909, >”? 466, ””? > XLIV ”” 9) be) 9? 9 1910, ” 719, 99 9° XLV 99 9? +e] 9? ” 1911, ” 611, a +7” XLVI 9 99 %> - he 39) 1912, ”? 275, 99 ” XLVII 9 9, 9? 9? ” 1913, de} 318, 99 ” XLVIII be 9? x9? > 9 1914, ”? 584, 9? 9 XLIX »”> > 9 99 99 1915, 99 587, a9 2» i >? ” > 9 ” 1916, ” 362, 9? rs LI a ae Aa a a TOV ERTS: at a AC Lil Pe) ” ” ” ” 1918, ”” 624, ” 99 LI bd 9) bed >> 9? 1919, %”? 414, ” a LIV ae a5 we 1 cies Pr 1920, ,, 312; price £1 ls. Lard LV + x9 99 ” St, 9) 1921, bi) 418, be] 9) LVI ’? be > 9° ” 1922, ” 372, 99 9 LVIl 9” 9> 9? 99 ”” 1923, 9? 421, be) 34 LVI sel 5 “8 ws 53 AOZE, | BOO; 2 9? LIX 9 99 9 9 %” 1925, 29 468, ” P ) LX 2? be) raps > 9 1926, + 470, > 9) LXI > >»? 99 be] be) 1927, %? 492, 9° 99 LXII 99 29 > be) 99 1928, > 458, 9 99 LXIII ” 2? 29 ” a9 1929, ” 263, ”» 99 LXIV 99 be 99 9 99 1930, %” 434, ” ” LXV ” 99 ” ” 2 1931, ” 366, ” yy LXVI 99 ” 99 9 > 1932, rh) 601, be] 9? LXVII > 99 99 9? b>) 1933, ”? 511, bP ti) LXVIIl be 39 29 be] tH} 1934, tp} 328, 99 ” LXIX > a) 29 ” ey 1935, 9 288, ” a” LXxX > 99 9? >? ” 1936, bi 528, 3? x” LXxI be be) 99 > dp 9 1937, x” 708, »? ” LXXII be) +9 29 ””° 9 1938, ay: 396, x” ”? LXXTI >” a? 9? be tae J 1939, >”) 344, 7? 9? LXXIV + bi 99 be > 1940, ” 658, a ’ LXXV be >” 9? be Me) 1941, 99 224, 7 a” LXXVI » 9 ” ba) ’9 1942, ” 432, ” iy LXXVII i e x ie Ai LOSS eae ty “4 ” LXXVUI as ” ” ” ” 1944, ” 338, ” * Royal Society of New South Wales OFFICERS FOR 1944-1945 Patrons : His EXcELLENCY THE GOVERNOR-GENERAL OF THE COMMONWEALTH OF AUSTRALIA, THE LORD GOWRIHE, v.c., P.c., G.C.M.G., C.B., D.S.O. His EXcELLENCY THE GOVERNOR OF NEW SoutH WALES, THE LORD WAKEHURST, x.c.m.a. President : G. D. OSBORNE, D.sc., Ph.p. Vice-Presidents : IDA A. BROWN, D.sc. Pror. H. PRIESTLEY, M.p., ch.M., B.se. A. BOLLIGER, Ph.D., A.A.C.I1. A. B. WALKOM, D.Sc. Honorary Secretaries : Pror. A. P. ELKIN, m.a., pPh.p. | D. P. MELLOR, m.sc. Honorary Treasurer : A. CLUNIES ROSS, B.sc., F.c.a. (Awust.). Members of Council: R. L. ASTON, B.Sc., B.E., M.Sc., Ph.D., W. H. MAZE, oM.sc. A.M.I.E. (Aust.). F. R. MORRISON, 4.a.c.1., F.C.S. J. A. DULHUNTY, B.sc. R. S. NYHOLM, m:.sc. F. P. J. DWYER, M.Sc. H. H. THORNE, M.a., B.Sc., F.R.A.S. N. A. FAULL, M.sc. H. W. WOOD, M.sc., A.Inst.P., F.R.A.S. F. LIONS, B.se., Ph.D., A.1.C. LIST OF THE MEMBERS OF THE Royal Society of New South Wales as at March 1, 1945 P Members who have contributed papers which have been published in the Society’s Journal. The numerals indicate the number of such contributions. { Life Members. Elected. 1944 Adamson, Colin Lachlan, Chemist, 22 Cremorne-street, Richmond, Vic. 1938 P 2 jtAlbert, Adrien, ph.p. Lond., B.Sc. Syd., A.R.1.c. Gt. B.. Commonwealth Research Fellow in Organic Chemistry, University of Sydney ; p.r. “* Greenknowe,”’ Greenknowe-avenue, Potts Point. 1935 tAlbert, Michel Francois, “‘ Boomerang,’’ Billyard-avenue, Elizabeth Bay. 1898 tAlexander, Frank Lee, Surveyor, 67 Ocean-street, Woollahra. 1941 tAlldis, Victor le Roy, u.s., Registered Surveyor, Young, N.S.W. 1909 P12 |tAndrews, Ernest C., B.a., Hon. Mem. Washington Academy of Sciences and of Royal Society of New Zealand, No. 4, “ Kuring-gai,”’ 241 Old South Head- road, Bondi. (President, 1921.) 1930 Aston, Ronald Leslie, B.sc., B.E. Syd., M.Sc., Ph.D. Cantab., A.M.I.E.Aust., Lecturer in Civil Engineering and Surveying in the University of Sydney ; p.r. 24 Redmyre-road, Strathfield. 1919 1 eu Aurousseau, Marcel, B.sc., 16 Woodland-street, Balgowlah. 1935 Back, Catherine Dorothy Jean, m.se., The Women’s College, Newtown. Woe ie Bed Bailey, Victor Albert, M.A., D.Phil., F.Inst.p., Professor of Experimental Physics in the University of Sydney. 1934 Polk Baker, Stanley Charles, m.Sc., A.Inst.Pp., Head Teacher of Physics, Newcastle Technical College, Tighe’s Hill; p.r. 8 Hewison-street, Tighe’s Hill, N.S.W. 1937 Baldick, Kenric James, B.sc., 19 Beaconsfield-parade, Lindfield. 1919 Bardsley, John Ralph, 76 Wright’s-road, Drummoyne. 1939 pe 2 Basnett, Elizabeth Marie, m.sc., 36 Cambridge-street, Epping. 1933 Bedwell, Arthur Johnson, Eucalyptus Oil Merchant, ‘“ Kama,’ 10 Darling Point-road, Edgecliff. 1944 Bennett, Alwynne Drysdale, B.sc., 8 Courland-street, Randwick. 1926 Bentivoglio, Sydney Ernest, B.Sc.Agr., Rosebery-road, Killara. 1940 Betty, Robert Cecil, 67 Imperial-avenue, Bondi. 1937 P 6 Birch, Arthur John, M.sc., D.Phil. (Oxon.), 37 Museum-road, Oxford, England. 1923 Birks, George Frederick, Wholesale Druggist, c/o Potter & Birks Ltd., 15 Grosvenor-street, Sydney; p.r. 42 Powell-street, Killara. 1916 Birrell, Septimus, 74 Edinburgh-road, Marrickville. 1920 Bishop, Eldred George, Manufacturing and General Engineer, 37-45 Myrtle- street, Chippendale ; p.r. 17 Thompson-street, Clifton Gardens. 1938 | eae | Black, Una Annie Frazer (Mrs.), B.sc., Flat 2, 4 Clowes-street, South Yarra, Melbourne, Victoria. 1939 came Blake, George Gascoigne, M.I.E.E., F.Inst.P., “‘ Holmleigh,’’ Cecil-avenue, Pennant Hills. 1933 P 24 Bolliger, Adolph, Ph.p., Director of Research, Gordon Craig Urological Research Laboratory, Department of Surgery, University of Sydney. 1920 P 9 Booth, Edgar Harold, M.c., D.sc., F.Inst.P.. New England University College, Armidale. (President, 1935.) 1939 Ps Bosworth, Richard Charles Leslie, m.sc., p.sc. Adel., Ph.p. Camb., F.A.C.1., F.Inst.p., c/o C.S.R. Co., Pyrmont; p.r. 41 Spencer-road, Killara. 1938 Breckenridge, Marion, B.sc., Department of Geology, University of Sydney ; p.r. 19 Handley-avenue, Thornleigh. 1940 Brigden, Alan Charles, B.sc., 22 Kelso-street, Enfield. PS ot P 4 P22 d eset | 2 5 P 4 P 21 Puce 1 1 Pb P 8 7 P 3 Brod P 4 Pig Pod ~ Briggs, George Henry, D.Sc., Ph.D., F.Inst.p., Officer-in-Charge, Section of Physics, National Standards Laboratory of Australia, University Grounds, Sydney; p.r. 13 Findlay-avenue, Roseville. Brown, Desmond J., B.Sc., 9 Agnes-street, Strathfield. Brown, Ida Alison, p.sc., Lecturer in Paleontology, University of Sydney. Brown, Samuel Raymond, 4.c.a. Aust., 87 Ashley-street, Chatswood. Browne, William Rowan, D.sc., Reader in Geology in the University of Sydney. (President, 1932.) Buckley, Daphne M. (Mrs.), B.sc., 29 Abingdon-road, Roseville. Buckley, Lindsay Arthur, B.sc., 29 Abingdon-road, Roseville. Burfitt, Barbara Joyce, M.B., B.S., Captain, A.I.F., 110 Elizabeth Bay-road, Elizabeth Bay, N.S.W. {Burfitt, W. Fitzmaurice, B.a., M.B., Ch.M., B.Sc. Syd., F.R.A.C.S., “‘ Radstoke,”’ Elizabeth Bay. Burkitt, Arthur Neville St. George, M.B., B.sc., Professor of Anatomy in the University of Sydney. Burkitt, John Stanton, ‘‘ Moonbi,”’ 17 Cavell-street, West Hobart, Tas. Cane, Reginald Frank, M.sc., A.A.c.1., General Motors-Holdens Ltd., Fishermen’s Bend, Melbourne, Victoria ; p.r. 2 David-street, St. Kilda, 8.2. Callanan, Victor John, B.sc., 17 Wheatleigh-street, Naremburn. {Carey, Samuel Warren, D.Sc., Government Geologist, Department of Mines, Hobart, Tasmania. {Carslaw, Horatio Scott, sc.D., LL.D., F.R.S.E., Emeritus Professor of Mathe- matics, University of Sydney, Fellow of Emmanuel College, Cambridge ; Burradoo, N.S.W. Cavill, George William Kenneth, B.sc., Department of Chemistry, Technical College, Harris-street, Ultimo; p.r. 40 Chandos-street, Ashfield. Challinor, Richard Westman, F.R.I.C., A.A.C.I., A.S.T.C., F.C.S.; p.r. 54 Drum- albyn-road, Bellevue Hill. (President, 1933.) Chalmers, Robert Oliver, A.s.tT.c., Scientific Liaison Bureau, Box 4061, G.P.O., Sydney. Chambers, Maxwell Clark, B.sc., c/o J. and E. Atkinson Pty. Ltd., 469-75 Kent-street, Sydney. Cheel, Edwin, 40 Queen-street, Ashfield. (President, 1931.) Churchward, John Gordon, B.Sc.Agr., Ph.D., 1 Hunter-street, Woolwich. Clark, Sir Reginald Marcus, K.B.E., Central Square, Sydney. Clarke, Ronald Stuart, B.a., 28 Beecroft-road, Beecroft. Clune, Francis Patrick, Author and Accountant, 15 Prince’s-avenue, Vaucluse. Cohen, Max Charles, B.sc., A.I.F., 80 “ St. James,”’ Stanley-street, Sydney. Cohen, Samuel Bernard, M.sSc., A.A.c.1., 34 Euroka-street, Northbridge. Colditz, Margaret Joyce, B.sc., 9 Beach-street, Kogarah. Cole, Edward Ritchie, B.sc., 14 Barwon-road, Lane Cove. Cole, Joyce Marie, B.sc., 14 Barwon-road, Lane Cove. Collett, Gordon, B.Sc., 20 Duchess-avenue, Fivedock. Cooke, Frederick, c/o Meggitt’s Limited, Asbestos House, York and Barrack- streets, Sydney. Coombs, F. A., F.c.s., Instructor of Leather Dressing and Tanning, Sydney Technical College; p.r. Bannerman-crescent, Rosebery. Corbett, Robert Lorimer, Managing Director of Robert Corbett & Co. Ltd., Manufacturing Chemists, Head Office, 379 Kent-street, Sydney. Cornforth, Rita Harriet, D.Phil. (Oxon.), M.sc. (Syd.), ec 0 Dyson Perrin’s Laboratory, South Parks-road, Oxford, England. Cortis-Jones, Beverly, M.sSc., 62 William-street, Roseville. Cotton, Frank Stanley, D.sc., Research Professor in Physiology in the University of Sydney. tCotton, Leo Arthur, M.A., D.Sc., Professor of Geology in the University of Sydney. (President, 1929.) Craig, David Parker, Research Scholar, 62 Springdale-road, Killara. tCresswick, John Arthur, A.A.C.I., F.c.S., Production Superintendent and Chief Chemist, c/o The Metropolitan Meat Industry Commissioner, State Abattoir and Meat Works, Homebush Bay; p.r. 101 Villiers-street, Rockdale. Crockford, Joan Marian, B.Sc., 219 Victoria-road, Gladesville. Culey, Alma Gertrude, m.sc., 37 Neirbo-avenue, Hurstville. Dadour, Anthony, B.sc., 25 Elizabeth-street, Waterloo. tDare, Henry Harvey, M.E., M.Inst.C.E., M.1.E.Aust., 14 Victoria-street, Roseville. de Beuzeville, Wilfred Alex. Watt, 3.p., ‘‘ Mélamere,’’ Welham-street, Beecroft. Elected. 1906 | 1913 1928 1943 1937 1924 1934 1923 1924 1934 1940 1937 1916 1944 1908 1935 1921 1939 1909 1940 1927 1940 1920 1940 1933 1879 1932 1905 1940 1943 1940 1944 1935 1939 1926 1942 1940 1935 1936 Pie Pi 7 P 25 Pat te Pr 2 1 caine | Pet Pp P 2 | any’ X] {Dixson, William, ‘‘ Merridong,’”’ Gordon-road, Killara. Doherty, William M., F.R.1.C., F.A.C.1., 36 George-street, Marrickville. Donegan, Henry Arthur James, A.S.T.c., A.A.C.1., Analyst, Department of Mines, Sydney; p.r. 18 Hillview-street, Sans Souci. Dudgeon, William, Manager, Commonwealth Drug Co., 50-54 Kippax-street, Sydney. Dulhunty, John Allan, B.sc., Geology Department, University of Sydney. Dupain, George Zephirin, A.A.C.1., F.c.s., Director Dupain Institute of Physical Education and Medical Gymnastics, Manning Building, 449 Pitt-street, Sydney; p.r. “‘ Rose Bank,” 158 Parramatta-road, Ashfield. Dwyer, Francis P. J., m.sc., Lecturer in Chemistry, Technical College, Sydney. Earl, John Campbell, p.sc., Ph.p., Professor of Organic Chemistry in the University of Sydney. (President, 1938.) Eastaugh, Frederick Alldis, a.R.s.M., F.R.1.c., Professor in Engineering Tech- nology and Metallurgy in the University of Sydney. Elkin, Adolphus Peter, M.A., Ph.p., Professor of Anthropology in the University of Sydney. (President, 1940. Hon. Secretary.) Emmerton, Henry James, B.Sc., 41 Nelson-street, Gordon. English, James Roland, us. A.LF. Enright, Walter John, B.A., Solicitor, High-street, West Maitland ; p.r. Regent- street, West Maitland. Erhart, John Charles, Chemical Engineer, 33 Beaconsfield-parade, Lindfield. tEsdaile, Edward William, 42 Hunter-street, Sydney. Evans, Silvanus Gladstone, 4.1.4.4. Lond., A.R.A.I.A., 6 Major-street, Coogee. Farnsworth, Henry Gordon, “ Rothsay,” 90 Alt-street, Ashfield. Faull, Norman Augustus, B.Sc., A.Inst.P., c.o. National Standards Laboratory, University Grounds, City-road, Chippendale. tFawsitt, Charles Edward, D.Sc., Ph.D., F.A.C.I., Professor of Chemistry in the University of Sydney. (President, 1919.) Finch, Franklin Charles, B.sc., Kirby-street, Rydalmere, N.S.W. Finnemore, Horace, B.Sc., F.R.1.c., Reader in Pharmacy in the University of Sydney. Fisher, Robert, B.sc., 3 Sackville-street, Maroubra. Fisk, Sir Ernest Thomas, K.B., F.Inst.R.E., A.M.1.E. Aust., Managing Director, Electric and Musical Industries Ltd., Blyth-road, Hayes, Middlesex, England. Flack, Arthur Charles Allenby, B.sc., Agricultural High School, Yanco, N.S.W. Fletcher, Harold Oswald, Assistant Paleontologist, Australian Museum, College-street, Sydney. {Foreman, Joseph, m.R.¢.s. Eng., L.R.c.P. Hdin., ‘* The Astor,’’ Macquarie-street, Sydney. Forman, Kenn. P., M.1.Refr.z., c/o Department of Aircraft Production, Box 20935, Melbourne, Vic. {tFoy, Mark, c/o Geo. O. Bennett, 133 Pitt-street, Sydney. Franki, Robert James Anning, B.sc., 891 New South Head-road, Rose Bay. Frederick, Robert Desider Louis, B.E., 6 “‘ Trinity Court,’’ Telopea-street, Wollstonecraft. Freney, Martin Raphael, B.sc., Central Wool Testing House, 17 Randle-street, Sydney. Friend, James Alan, 16 Kelburn-road, Roseville. Garretty, Michael Duhan, m.sc., Chief Geologist, North Broken Hill Ltd., Broken Hill, N.S.W. Gascoigne, Robert Mortimer, 5 Werona-avenue, Killara. Gibson, Alexander James, M.E., M.Inst.C.E., M.I.E.Aust., Consulting Engineer, 906 Culwulla Chambers, 67 Castlereagh-street, Sydney ; p.r. “‘ Wirruna,” Belmore-avenue, Wollstonecraft. Gibson, Neville Allan, sB.sc., Industrial Chemist, 217 Parramatta-road, Haberfield. Gillis, Richard Galvin, 1 Dundee, 35 Adams-street, South Yarra, S.E.1, Vic. Goddard, Roy Hamilton, F.c.a. Aust., Royal Exchange, Bridge-street, Sydney. Goulston, Edna Maude, B.sc., Third Officer, W.R.A.N.S., Navy Office, Melbourne. xi Elected. 1940 1938 1943 PZ P 6 P 4 P 4 Pos Pil P 15 P' 6 Pid Graves, John Nevil, B.sc., 96 Wentworth-street, Randwick. Griffiths, Edward L., B.Sc., A.A.C.1., A.R.1.C., Chief Chemist, Department of Agriculture ; p.r. 151 Wollongong-road, Arncliffe. Hall, Norman Frederick Blake, m.sc., Chemist, Council for Scientific and Industrial Research (Tobacco Section), Dept. of Organic Chemistry, University of Sydney ; p.r. 154 Wharf-road, Longueville. tHalloran, Henry Ferdinand, L.s., A.M.I.E.Aust., F.S.I.Eng., M.T.P.I.Eng., 153 Elizabeth-street, Sydney ; p.r. 23 March-street, Bellevue Hill. Hanlon, Frederick Noel, B.sc., Geologist, Department of Mines, Sydney ; p-r. 12 Countess-street, Mosman. tHarker, George, D.Sc., F.A.C.I.; p.r. 89 Homebush-road, Strathfield. Harper, Arthur Frederick Alan, M.sc., A.Inst.P., National Standards Laboratory, University Grounds, City-road, Chippendale. Harrington, Herbert Richard, Teacher of Physics and Electrical Engineering, Technical College, Harris-street, Ultimo. Hawley, J. William, J.p., Kinancial Agent, 4 Castlereagh-street, Sydney ; p-r. 12 King’s-road, Vaucluse. Hayes, William Lyall, a.s.tT.c., a.a.c.1., Works Chemist, c/o Messrs. Wm. Cooper & Nep hews (Aust.) Ltd., Phillip-street, Concord; p.r. 101 Essex-street, Epping. Henriques, Frederick Lester, 208 Clarence-street, Sydney. Hill, Dorothy, m.sce. Q’ld., pPh.p. Cantab., Geological Research Fellow, University of Queensland, Brisbane. Hindmarsh, Percival, M.a. B.Sc.agr., Principal, Hurlstone Agricultural High School, Glenfield. Hirst, Edward Eugene, 4.M.1.E., Vice-Chairman and Joint Managing Director, British General Electric Co. Ltd.; p.r. “ Springmead,”’ Ingleburn. Hirst, George Walter Cansdell, B.sc., A.M.1.E. (Aust.), “‘ St. Cloud,”’ Beaconsfield- road, Chatswood. Hoggan, Henry James, A.M.I.M.E. Lond., a.M.1.E. Aust., Consulting and Designing Engineer, N.R.M.A. House, 3 Spring-street, Sydney. Howard, Harold Theodore Clyde, B.sc., Principal, Wollongong Technical High School, Wollongong. Howarth, Mark, F.R.A.s., Grange Mount Observatory, Bull-street, Mayfield, Newcastle, N.S.W. ; Hughes, Gordon Kingsley, B.sc., Lecturer in Chemistry, University of Sydney. tHynes, Harold John, D.sc., B.Sc.Agr., Biologist, Department of Agriculture, Box 36a, G.P.O., Sydney; p.r. “ Belbooree,’”’ 10 Wandella-avenue, Rose- ville. Iredale, Thomas, B.Sc., D.Sc., F.R.1.c., Chemistry Department, University of Sydney; p.r. 96 Roseville-avenue, Roseville. Jaeger, John Conrad, M.A., D.Sc., University of Tasmania, Hobart, Tasmania. Johns, Thomas Harley, 130 Smith-street, Summer Hill. Johnston, Thomas Harvey, M.A., D.Sc., C.M.Z.S., Professor of Zoology in the University of Adelaide. (Cor. Mem., 1912.) Joplin, Germaine Anne, B.Sc., Ph.D., Geological Department, University of Sydney; p.r. 18 Wentworth-street, Eastwood. Judd, William Percy, 123 Wollongong-road, Arncliffe. Julius, Sir George A., Kt., B.Sc., B.E., M.I.Mech.E., M.I.E.Aust., Culwulla Chambers, Castlereagh-street, Sydney. Kelly, Caroline Tennant (Mrs.), “ Hight Bells,’’ Castle Hill. Kennard, William Walter, 9 Bona Vista-avenue, Maroubra. Kenny, Edward Joseph, Geological Surveyor, Department of Mines, Sydney ; p-r. 17 Alma-street, Ashfield. Kimble, Jean Annie, Research Chemist, B.Sc., 383 Marrickville-road, Marrick- ville. Kirchner, William John, B.sc., A.A.c.1., Manufacturing Chemist, c/o Messrs. Burroughs Wellcome & Co. (Australia) Ltd., Victoria-street, Waterloo : p-r. 18 Lyne-road, Cheltenham. xi Elected. 1939 Lambeth, Arthur James, B.Sc., ‘“‘ Naranje,’’ Sweethaven-road, Wetherill Park, N.S.W. 1936 Leach, Stephen Laurence, B.A., B.Sc., A.A.C.I., Major, Scientific Liaison Officer, Victoria Barracks, Melbourne. 1920 Le Souef, Albert Sherbourne, 3 Silex-road, Mosman. 1929 P 55 |tLions, Francis, B.Sc., Ph.D., A.R.1.C., Department of Chemistry, University of Sydney. 1942 Lippmann, Arthur S., mM.p., 175 Macquarie-street, Sydney. 1940 P 2 Lipson, Menzie, B.Sc., A.A.C.I., Chemist, C.S.I.R., Central Wool Committee Testing House, 17 Randle-street, Sydney. 1940 3 Sane | Lockwood, William Hutton, B.sc., c.o. Department of Post-War Reconstruction, ; Hotel Acton, Canberra, A.C.T. 1906 tLoney, Charles Augustus Luxton, M.Am.soc.Refr.E., National Mutual Building, 350 George-street, Sydney. 1943 Luber (Mrs.) Daphne, B.sc., 98 Lang-road, Centennial Park. 1942 Lyons, Raymond Norman Matthew, m.sc., Biochemical Research Worker, 8 Boronia-avenue, Wollstonecraft. 1939 Ping Maccoll, Allan, m.sc., Senior Lecturer in Chemistry, University of Sydney, Sydney. 1943 McCoy, William Kevin, Analytical Chemist, R.A.A.F.; p.r. 16 Bishop’s- avenue, Randwick. 1940 McGrath, Brian James, 40 Mooramie-avenue, Kensington. 1940 McGregor, Gordon Howard, 4 Maple-avenue, Pennant Hills. 1906 2 |tMcIntosh, Arthur Marshall, “* Moy Lodge,’ Hill-street, Roseville. solace) — 1891 t{McKay, R. T., M.inst.c.z., Eldon Chambers, 92 Pitt-street, Sydney. 1944 Pw 2 McKenzie, Hugh Albert, B.sc., Assistant Research Officer, C.S.I.R.; p.r. 52 Bolton-street, Guildford. 1943 McKern, Howard Hamlet Gordon, 4.s.T.c., A.A.c.1., Chemist, Meggitt Ltd., Parramatta; p.r. 14 Orwell-street, Potts Point. 1932 McKie, Rev. Ernest Norman, B.a. Syd., St. Columba’s Manse, Guyra. 1927 McMaster, Sir Frederick Duncan, kt., ‘‘ Dalkeith,’’ Cassilis, N.S.W. 1940 Malone, Edward E., No. 4, Astral, 10 Albert-street, Randwick. 1924 Mance, Frederick Stapleton, ‘‘ Binbah,”’ Lucretia-avenue, Longueville. 1944 Martin, Cyril Maxwell, Chemist, 22 Wattle-street, Haberfield. 1935 Pe Maze, Wilson Harold, m.sc., Lecturer in Geography, University of Sydney. 1912 Meldrum, Henry John, B.a., B.Sc., Lecturer, The Teachers’ College, University Grounds, Newtown; p.r. 98 Sydney-road, Manly. 1929 P22 Mellor, David Paver, m.sc., Lecturer, Chemistry Department, University of Sydney; p.r. 1837 Middle Harbour-road, Lindfield. (President, 1941-42. Joint Hon. Secretary.) 1941 | Melville, George Livingstone, Managing Director, Federal Machine Co. Ltd., Loftus-street, Arncliffe. 1940 Mercer, Edgar Howard, McMaster Laboratory, Parramatta-road, Glebe. 1928 Micheli, Louis Ivan Allan, M.sc., Ph.D., Research Chemist, c/o Colonial Sugar Refining Co., Pyrmont. 1940 Millership, William, m.sc., Chief Chemist, Davis Gelatine (Aust.) Pty. Ltd., 15 Shaw-avenue, Earlwood. 1943 Molloy, Ernest Patrick, Assistant Sectional Manager, 129 Gibbes-street, Rockdale. 1941 Morrissey, Matthew John, B.A., a.s.t.c., Auburn-street, Parramatta. 1922 P27 Morrison, Frank Richard, 4.A.c.1., F.c.s., Assistant Chemist, Technological Museum, Sydney. 1934 Mort, Francis George Arnot, A.A.c.1., Chemist, 16 Grafton-street, Woollahra. 1944 Moye, Daniel George, Chemist, ‘‘ Whiporie,’’ Holland-street, Cronulla. 1915 Murphy, Robert Kenneth, Dr.ing., Chem.Eng., A.S.T.C., M.I.Chem.E., F.A.C.L., Lecturer in Charge of Chemistry and Head of Science Department, Sydney Technical College. 1923 Br YD, Murray, Jack Keith, QX34748, B.aA., B.Sc.Agr., c/o H.Q., Angau, New Guinea Force, and Professor of Agriculture in the University of Queensland. 4 1930 P 6 Naylor, George Francis King, M.A., M.Sc., Dip.Ed., A.A.I.I.P., Squadron Leader, R.A.A.F., Headquarters, Melbourne; p.r. ‘* Kingsleigh,’’ Ingleburn, N.S.W. XIV Elected. 1943 1932 1943 1938 1920 1940 1940 1935 1903 1921 1920 1938 1935 1943 1919 1896 1921 1918 1938 1927 1918 1893 1935 1922 1940 1919 1936 1931 1935 1939 1933 1940 1935 1 P 4 P13 | Soa, P 74 Pi Poa Pius P 6 1 | ee) Peo P14 Pi 2 Neuhaus, John William George, c/o Meggitt Ltd., Parramatta. Newman, Ivor Vickery, M.Sc., Ph.D., F.R.M.S., F.L.S., Department of Biology, Victoria University College, Wellington, N.Z. Nicol, Alexander Campbell, a.s.T.c., a.a.c.1., Chief Chemist, Crown Crystal Glass Co.; p.r. No. 12 “ Florida,” 519 New South Head-road, Double Bay. Noble, Norman Scott, D.sc.agr., M.Sc., D.1.c., Secretary, Linnean Society of N.S.W., Science House, Gloucester-street, Sydney. tNoble, Robert Jackson, M.Sc., B.Sc.Agr., Ph.D., Under Secretary, Department of Agriculture, Box 364, G.P.O., Sydney; p.r. 32a Middle Harbour-road, Lindfield. (President, 1934.) Norrie, Jack Campbell, B.sc., 28 Ray-road, Epping. Nyholm, Ronald Sydney, m.sc., 77 Bland-street, Ashfield. O’Connell, Rev. Daniel J. K., s.3., M.Sc., F.R.A.S., Riverview College Observatory, Sydney. tOld, Richard, ‘‘ Waverton,’’ Bay-road, North Sydney. Osborne, George Davenport, D.sc. Syd., Ph.D. Camb., Lecturer and Demonstrator in Geology in the University of Sydney. Penfold, Arthur Ramon, F.A.C.1., F.c.s., Curator and Economic Chemist, Technological Museum, MHarris-street, Ultimo; p.r. 67 Park-avenue, Roseville. (President, 1935.) Phillips, Marie Elizabeth, B.sc., 4 Morella-road, Clifton Gardens. Phillips, Orwell, 55 Darling Point-road, Edgecliffe. Plowman, Ronald Arthur, A.s.1T.c., A.A.c.1., Analytical Chemist, 78 Alt-street, Ashfield. Poate, Hugh Raymond Guy, M.B., ch.m. Syd., F.R.C.S. Hng., L.R.c.P. Lond., F.R.A.C.S., Surgeon, 225 Macquarie-street, Sydney; p.r. 38 Victoria-road, Bellevue Hill. tPope, Roland James, B.A. Syd., M.D., Ch.M., F.R.c.8. Hdin., 185 Macquarie- street, Sydney. Powell, Charles Wilfrid Roberts, F.R.I.c., A.A.c.I., Company Executive, c/o Colonial Sugar Refining Co., O’Connell-street, Sydney; p.r. “* Wansfell,”’ Kirkoswald-avenue, Mosman. Powell, John, Director, Foster Clark (Aust.) Ltd., 17 Thurlow-street, Redfern ; p.r. “ Elgarth,”’ Ranger’s-road, Cremorne. Powell, John Wallis, A.s.T.c., A.A.c.1., Managing Director, Foster Clark (Aust.) Ltd., 17 Thurlow-street, Redfern. Price, William Lindsay, B.E., B.Sc., Teacher of Physics, Sydney Technical College ; p.r. 8 Wattle-street, Kuillara. Priestley, Henry, M.D., Ch.M., B.Sc., Professor of Biochemistry, Faculty of Medicine, the University of Sydney. (President, 1942-43.) tPurser, Cecil, B.A., M.B., Chm. Syd., “‘ Ascot,’’ Grosvenor-road, Wahroonga. {tQuodling, Florrie Mabel, p.sc., Demonstrator in Geology, University of Sydney. Raggatt, Harold George, p.sc., Director, Mineral Resources Survey, Depart- ment of Supply, Canberra, A.C.T. Ralph, Colin Sydney, B.sc., 24 Canberra-street, Epping. Ranclaud, Archibald Boscawen Boyd, B.Sc., B.E., Lecturer in Physics, Teachers’ College, The University, Sydney. Randall, Harry, Buena Vista-avenue, Denistone. Rayner, Jack Maxwell, B.Sc., F.mst.p., Chief Geophysicist, Mineral Resources Survey, Department of Supply and Shipping, Census Building, Canberra, A.C.T. Reid, Cicero Augustus, 19 Newton-road, Strathfield. Ritchie, Ernest, B.Sc., 6 Military-road, North Bondi. Roberts, Richard George Crafter, Electrical Engineer, c/o C. W. Stirling & Co., Asbestos House, York and Barrack-streets, Sydney. Robertson, Rutherford Ness, B.sc. Syd., Ph.D. Cantab., Flat 4, 43 Johnston- street, Annandale. Room, Thomas G., M.A., F.R.S., Professor of Mathematics in the University of Sydney. ——— ae Elected. 1940 1928 1940 1935 1941 1920 1940 1933 1936 1938 1936 1917 1900 1943 1933 1940 1922 1919 1921 1916 1914 1900 1942 1916 1940 1918 1901 T919 1920 1941 1915 1944 1939 1919 1935 Bi 2 PZ Pod Pot Pd P 16 PZ P 3 xXV Rosenbaum, Sidney, 44 Gilderthorp-avenue, Randwick. Ross, Allan Clunies, B.Sc., F.c.A. Aust., Chartered Accountant Aust., 544 Pitt-street, Sydney ; p.r. The Grove, Woollahra. (Member from 1915 to 1924.) (Hon. Treasurer.) Ross, Jean Elizabeth, B.Sc., Dip.Ed., 5 Stanton-road, Haberfield. Savage, Clarence Golding, Director of Fruit Culture, Department of Agriculture, Sydney. Sawkins, Dansie Thomas, m.Aa. Syd., B.A. Camb., Reader in Statistics, The University, Sydney ; p.r. 60 Boundary-street, Roseville. Scammell, Rupert Boswood, B.Sc. Syd., A.A.C.1., F.C.S., c/o F. H. Faulding & Co. Ltd., 98 Castlereagh-street, Redfern; p.r. 10 Buena Vista-avenue, Clifton Gardens. Scott, Reginald Henry, B.sc., 3 Walbundry-avenue, East Kew, Victoria. Selby, Esmond Jacob, Dip.com., Sales Manager, Box 175 D, G.P.O., Sydney. Sellenger, Brother Albertus, Marist Brothers’ College, Randwick, N.S.W. Sheahan, Thomas Henry Kennedy, B.Sc., Chemist, 2 Edward-street, Gordon. Sherrard, Kathleen Margaret Maria (Mrs.), m.sc. Melb., 43 Robertson-road, Centennial Park. Sibley, Samuel Edward, Mount-street, Coogee. {tSimpson, R. C., Park-avenue, Roseville. Simpson, John Kenneth Moore, Industrial Chemist, “‘ Browie,’’ Old Castle Hill-road, Castle Hill. Slade, George Hermon, B.sc., Director, W. Hermon Slade & Co. Pty. Ltd., Manufacturing Chemists, Mandemar-avenue, Homebush; p.r. “ Raiatea,”’ Oyama-avenue, Manly. Smith, Eric Brian Jeffcoat, 1 Rocklands-road, Wollstonecraft. Smith, Thomas Hodge, Australian Museum, College-street, Sydney. Southee, Ethelbert Ambrook, 0.B.E., M.A., B.Sc., B.Sc.Agr., Principal, Hawkes- bury Agricultural College, Richmond, N.S.W. Spencer-Watts, Arthur, “ Araboonoo,”’ Glebe-street, Randwick. Stephen, Alfred Ernest, ¥.c.s., c/o Box 1158 HH, G.P.O., Sydney. Stephens, Frederick G. N., F.R.c.S., M.B., Ch.M., 135 Macquarie-street, Sydney ; p-r. Captain Piper’s-road and New South Head-road, Vaucluse. tStewart, J. Douglas, B.v.sc., F.R.C.v.S., Emeritus Professor of Veterinary Science in the University of Sydney; p.r. “ Berelle,’” Homebush-road, Strathfield. (President, 1927.) Still, Jack Leslie, B.sc., Ph.D., Department of Biochemistry, The University, Sydney. Stone, Walter George, F.S.T.C., F.A.C.1., Chief Analyst, Department of Mines, Sydney. Stroud, Richard Harris, B.sc., “‘ Dalveen,’’ corner Chalmers and Barker-roads, Strathfield. tSullivan, Herbert Jay, Director in Charge of Research and Technical Depart- ment, c/o Lewis Berger & Sons (Australia) Ltd., Rhodes; Box 23, P.O., Burwood; p.r. *‘ Stonycroft,’” 10 Redmyre-road, Strathfield. tSussmilch, C. A., F.G.s., F.S.T.c., Consulting Geologist, 11 Appian Way, Burwood. (President, 1922.) tSutherland, George Fife, a.r.c.sc. Lond., Assistant Professor of Mechanical Engineering in the University of Sydney. Sutton, Harvey, 0.B.E., M.D., D.p.H. Melb., B.sc. Oxon., Professor of Preventive Medicine and Director, School of Public Health and Tropical Medicine, University of Sydney ; p.r. “ Lynton,”’ 27 Kent-road, Rose Bay. Swanson, Thomas Baikie, M.sc. Adel., c/o Technical Service Department, Icianz, Box 1911, G.P.O., Melbourne, Victoria. Taylor, Brigadier Harold B., M.c., D.Sc., F.R.I.C., F.A.C.I., Second Government Analyst, Department of Public Health, 93 Macquarie-street, Sydney ; p-r. 44 Kenneth-street, Longueville. Thomas, Andrew David, Flight-Lieutenant, R.A.A.F., M.sc., A.Inst.p., 1 Valley- road, Lindfield. Thomas, Mrs. A. V. M., 1 Valley-road, Lindfield. Thorne, Harold Henry, m.a. Cantab., B.sc. Syd., ¥.R.A.S., Lecturer in Mathe- matics in the University of Sydney ; p.r. 55 Railway-crescent, Beecroft. Tommerup, Eric Christian, M.Sc., A.A.C.1., Queensland Agricultural College, Lawes, via Brisbane, Queensland. XVi1 Elected 1923 1940 1932 1943 1940 1921 1935 1933 1903 1943 1919 1913 1944 1921 1924 1919 1919 1944 1941 1911 1936 1920 1921 1909 1940 1943 1928 1942 1944 1943 1940 1936 1906 1916 1921 P18 ayo P 10 Pr? 5 1 6 1 1 | Fac | 3 1 Pe? P12 Toppin, Richmond Douglas, A.R.1.c., 51 Crystal-street, Petersham. Tow, Aubrey James, m.sc., No. 5, *‘ Werrington,’’ Manion-avenue, Rose Bay. Trikojus, Victor Martin, B.Sc., D.Phil., Professor of Biochemistry, The University, Melbourne. Turner, Ivan Stewart, M.A., M.Sc., Ph.D., Lecturer in Mathematics, University of Sydney; p.r. 120 Awaba-street, Mosman. Vernon, James, Ph.D., A.A.C.1., Chief Chemist, Colonial Sugar Refining Co., 1 O’Connell-street, Sydney. Vicars, Robert, Marrickville Woollen Mills, Marrickville. Vickery, Joyce Winifred, m.sc., Botanic Gardens, Sydney; p.r. 17 The Promenade, Cheltenham. Voisey, Alan Heywood, m.sc., Lecturer in Geology and Geography, New England University College, Armidale. tVonwiller, Oscar U., B.Sc., F.Inst.P., Professor of Physics in the University of Sydney. (President, 1930.) Walker, James Foote, Company Secretary, 11 Brucedale-avenue, Epping. Walkom, Arthur Bache, D.sc., Director, Australian Museum, Sydney; p.r. 45 Nelson-road, Killara. (Member from 1910-1913. President, 1943-44.) Wardlaw, Hy. Sloane Halcro, p.sc. Syd., F.A.c.1., Lecturer and Demonstrator in Biochemistry in the University of Sydney. (President, 1939.) Warner, Harry, A.S.T.c., Chemist, 15 Belmore-road, Penshurst. tWaterhouse, Gustavus Athol, D.sc., B.H., F.R.E.S., F.R.Z.S., ¢c.o. Australian Museum, College-street, Sydney. Waterhouse, Leslie Vickery, B.x. Syd., Mining Engineer, Shell House, Car- rington-street, Box 58 CC, G.P.O., Sydney; p.r. 4 Bertha-road, Neutral Bay. Waterhouse, Lionel Lawry, B.E. Syd., Lecturer and Demonstrator in Geology in the University of Sydney. Waterhouse, Walter L., M.c., D.Sc.Agr., D.I.C., F.L.S., Reader in Agriculture, University of Sydney; p.r. “‘ Hazelmere,’’ Chelmsford-avenue, Lindfield. (President, 1937.) Watkins, William Hamilton, B.sc., Industrial Chemist, 57 Bellevue-street, North Sydney. Watson, Irvine Armstrong, Ph.D., B.Sc.Agr., Assistant Lecturer, Faculty of Agriculture, University of Sydney. Watt, Robert Dickie, M.a., B.sc., Professor of Agriculture in the University of Sydney ; p.r. 64 Wentworth-road, Vaucluse. (President, 1925.) Wearne, Harold Wallis, 22 Yarabah-avenue, Gordon. Wellish, Edward Montague, m.a., Associate-Professor of Applied Mathematics in the University of Sydney ; p.r. 15 Belgium-avenue, Roseville. Wenholz, Harold, B.sc.agr., Director of Plant Breeding, Department of Agri- culture, Sydney. tWhite, Charles Josiah, B.sc., Lecturer in Chemistry, Teachers’ College, Uni- versity Grounds, Newtown. White, Douglas Elwood, M.sc., D.Phil.. University of Western Australia, Nedlands, W.A. Whiteman, Reginald John Nelson, M.B., Ch.M., F.R.A.C.S., 143 Macquarie-street, Sydney. Wieccnen Fuedaate Abbey, M.B., Ch.M., D.o.M.s., Ophthalmic Surgeon, 143 Macquarie-street, Sydney; p.r. Jersey-road, Strathfield. Williams, Gordon Roy, B.sSc., 45 Conder-street, Burwood. Willis, John Bryan, B.sc., Demonstrator in Chemistry, University of Sydney ; p-r. Flat 2, Russell Hall, 17 Mount-street, Coogee. Winch, Leonard, B.sc., Chief Chemist, Fielder’s General Products Ltd., P.O. Box 148, Tamworth, N.S.W. Wogan, Samuel James, Range-road, Sarina, North Queensland. Wood, Harley Weston, M.Sc., A.Inst.P., F.R.A.S., Assistant Astronomer, Sydney Observatory. tWoolnough, Walter George, D.sc., F.G.s., 9 Lockerbie Court, East St. Kilda, Victoria. (President, 1926.) Wright, George, Company Director, c/o Hector Allen, Son & Morrison, 16 Barrack-street, Sydney; p.r. ‘‘ Wanawong,”’ Castle Hill, N.S.W. Yates, Guy Carrington, Seedsman, c/o Arthur Yates & Co. Ltd., 184 Sussex- street, Sydney ; p.r. Boomerang-street, Turramurra. Xvi HonorAaRY MEMBERS. LInmaited to Twenty. Elected. ! 1914 Hill, James P., D.sc., F.R.S., Professor of Zoology, University College, Gower- street, London, W.C.1, England. 1915 Maitland, Andrew Gibb, F.a.s., ‘“‘ Bon Accord,’ 28 Melville-terrace, South Perth, W.A. 1912 Martin, Sir Charles J., C.M.G., D.Sc., F.R.S., Roebuck House, Old Chesterton, Cambridge, England. 1922 Wilson, James T., M.B., Ch.M. Hdin., F.R.S., Professor of Anatomy in the Uni- versity of Cambridge; p.r. 24 Millington-road, Cambridge, England. OBITUARY, 1944-1945. 1905 Charles Anderson. 1929 Norman Dawson Royle. 1909 Edward Sutherland Stokes. THE REV. W. B. CLARKE MEMORIAL FUND. 4 The Rev. W. B. Clarke Memorial Fund was inaugurated at a meeting of the Royal Society of N.S.W. in August, 1878, soon after the death of Mr. Clarke, who for nearly forty years rendered distinguished service to his adopted country, Australia, and to science in general. It was resolved to give an opportunity to the general public to express their appreciation of the character and services of the Rev. W. B. Clarke “ as a learned colonist, a faithful minister of religion, and an eminent scientific man.” It was proposed that the memorial should take the form of lectures on Geology (to be known as the Clarke Memorial Lectures), which were to be free to the public, and of a medal to be given from time to time for distinguished work in the Natural Sciences done in or on the Australian Commonwealth and its territories; the person to whom the award is made may be resident in the Australian Commonwealth or its territories, or elsewhere. The Clarke Memorial Medal was established first, and later, as funds permitted, the Clarke Memorial Lectures have been given at intervals. CLARKE MEMORIAL LECTURES. Delivered. 1906. ‘‘ The Volcanoes of Victoria,’ and ‘“‘The Origin of Dolomite’’ (two lectures). By Professor E. W. Skeats, D.Sc., F.G.S. 1907. ‘“‘ Geography of Australia in the Permo-Carboniferous Period’’ (two lectures). By Professor T. W. E. David, B.A., F.R.S. ““The Geological Relations of Oceania.”” By W. G. Woolnough, D.Sc. ‘“* Problems of the Artesian Water Supply of Australia.” By E. F. Pittman, A.R.S.M. “The Permo-Carboniferous Flora and Fauna and their Relations.””» By W. 8. Dun. 1918. ‘‘ Brain Growth, Education, and Social Inefficiency.”” By Professor R. J. A. Berry, M.D.) FRESE: 1919. ‘“* Geology at the Western Front,” By Professor T. W. E. David, C.M.G., D.S.O., F.R.S. 1936. ‘The Aeroplane in the Service of Geology.” By W. G. Woolnough, D.Sc. (THIs JouRN., 1936, 70, 39.) 1937. ‘‘ Some Problems of the Great Barrier Reef.’ By Professor H. C. Richards, D.Se. (Tus JOURN., 1937, 71, 68.) 1938. ‘“‘The Simpson Desert and its Borders.” By C. T. Madigan, M.A., B.Sc., B.E., D.Sc. (Oxon.). (THis Journ., 1938, 71, 503.) 1939. ‘‘ Pioneers of British Geology.” By Sir John §S. Flett, K.B.E., D.Sc., LL.D., F.B.S. (THis JourNn., 1939, 73, 41.) 1940. ‘‘ The Geologist and Sub-surface Water.’”’ By E. J. Kenny, M.Aust.1.M.M. (Tus Journ., 1940, 74, 283.) 1941. ‘‘ The Climate of Australia in Past Ages.” By C. A. Sussmilch, F.G.S. (THis JouRN., 1941, 75, 47.) 1942. ‘‘ The Heroic Period of Geological Work in Australia.” By E. C. Andrews, B.Sc. 1943. ‘‘ Australia’s Mineral Industry in the Present War.” By H. G. Raggatt, D.Sc. 1944. ‘An Australian Geologist Looks at the Pacific.” By W. H. Bryan, M.C., D.Sc. AWARDS OF THE CLARKE MEDAL. Established in memory of The Revd. WILLIAM BRANWHITE CLARKE, m.a., F.R.S., F.G.S., etc. Vice-President from 1866 to 1878. The prefix * indicates the decease of the recipient. Awarded. 1878 *Professor Sir Richard Owen, K.C.B., F.R.S. 1879 *George Bentham, C.M.G., F.R.S. 1880 *Professor Thos. Huxley, F.RB.s. 1881 *Professor F. M‘Coy, F.R.S., F.G.S. 1882 *Professor James Dwight Dana, LL.D. 1883 *Baron Ferdinand von Mueller, K.C.M.G., M.D., Ph.D., F.R.S., F.L.S. 1884 *Alfred R. C. Selwyn, LL.D., F.R.S., F.G.S. 1885 *Sir Joseph Dalton Hooker, 0.M., G.C.S.1., C.B., M.D., D.C.L., LL.D., F.R.S. 1886 *Professor L. G. De Koninck, M.D. xix Awarded. : 1887 *Sir James Hector, K.C.M.G., M.D., F.R.S. 1888 *Rev. Julian EH. Tenison-Woods, F.G.S., F.L:s. 1889 *Robert Lewis John Ellery, F.R.S., F.R.A.S. 1890 *George Bennett, M.D., F.R.c.s. Hing., F.L.S., F.Z.S. 1891 *Captain Frederick Wollaston Hutton, F.R.S., F.G.S. 1892 *Sir William Turner Thiselton Dyer, K.C.M.G., C.I.E., M.A., LL.D. Sc.D., F.R.S., F.L.9. 1893 *Professor Ralph Tate, F.L.S., F.G.S. 1895 *Robert Logan Jack, LL.D., F.G.S., F.R.G.S. 1895 *Robert Etheridge, Jnr. 1896 *The Hon. Augustus Charles Gregory, C.M.G., F.R.G.S. 1900 *Sir John Murray, K.c.B., LL.D., Sc.D., F.R.S. 1901 *Edward John Eyre. 1902 *F. Manson Bailey, o.M.G., F.L.s. 1903 *Alfred William Howitt, D.sc., F.G.s. 1907 *Professor Walter Howchin, F.a.s., University of Adelaide. 1909 *Dr. Walter E. Roth, B.a. 1912 *W. H. Twelvetrees, F.a.s. 1914 Sir A. Smith Woodward, LL.D., F.R.s., Keeper of Geology, British Museum (Natural History), London. 1915 *Professor W. A. Haswell, M.A., D.Sc., F.R.S. 1917 *Professor Sir Edgeworth David, K.B.E., C.M.G., D.S.0., M.A., Sc.D., D.Sc., F.R.S., F.G.S. 1918 *Leonard Rodway, c.m.c., Honorary Government Botanist, Hobart, Tasmania. 1920 *Joseph Edmund Carne, F.«.s. 1921 *Joseph James Fletcher, M.a., B.Sc. 1922 *Richard Thomas Baker, The Crescent, Cheltenham. 1923 *Sir W. Baldwin Spencer, K.C.M.G., M.A., D.Sc., F.RB.S. 1924 *Joseph Henry Maiden, 1.8.0., F.R.S., F.L.S., J.P. 1925 *Charles Hedley, F.u.s. 1927 Andrew Gibb Maitland, F.a.s., ‘‘ Bon Accord,’ 28 Melville Terrace, South Perth, W.A. 1928 Ernest C. Andrews, B.A., F.G.S., 32 Benelong Crescent, Bellevue Hill. 1929 Professor Ernest Willington Skeats, D.Sc., A.R.c.S., F.G.S., University of Melbourne, Carlton, Victoria. 1930 = L. Keith Ward, B.A., B.E., D.Sc., Government Geologist, Geological Survey Office, Adelaide. 1931 *Robin John Tillyard, M.A., D.Sc., Sc.D., F.R.S., F.L.S., F.E.S., Canberra, F.C.T. 1932 *Frederick Chapman, A.L.S., F.R.S.N.Z., F.G.S., Melbourne. 1933. Walter George Woolnough, D.sc., F.c.s., Department of the Interior, Canberra, F.C.T. 1934 *Edward Sydney Simpson, D.sc., B.E., F.A.C.1., Carlingford, Mill Point, South Perth, W.A. 1935 George William Card, A.R.S.M., 16 Ramsay-street, Collaroy, N.S.W. 1936 Sir Douglas Mawson, Kt., 0.B.E., F.R.S., D.Sc., B.E., University of Adelaide. 1937. J. T. Jutson, B.sc., Lu.B., 9 Ivanhoe-parade, Ivanhoe, Victoria. 1938 Professor H. C. Richards, D.sc., The University of Queensland, Brisbane. 1939 OC. A. Sussmilch, F.G.s., F.s.T.c., 11 Appian Way, Burwood, N.S.W. 1941 Professor Frederic Wood Jones, M.B., B.S., D.Sc., F.R.S., Anatomy Department, University of Manchester, England. 1942 William Rowan Browne, D.sc., Reader in Geology, The University of Sydney, N.S.W. 1943. Walter Lawry Waterhouse, M.cC., D.Sc.Agric., D.I.c., F.L.S., Reader in Agriculture, University of Sydney. 1944 Professor Wilfred Eade Agar, 0.B.E., M.A., D.Sc., F.R.S., University of Melbourne, Carlton, Victoria. AWARDS OF THE SOCIETY’S MEDAL AND MONEY PRIZE. Money Prize of £25. Awarded. 1882 John Fraser, B.a., West Maitland, for paper entitled “The Aborigines of New South Wales.”’ 1882 Andrew Ross, m.p., Molong, for paper entitled ‘‘ Influence of the Australian climate and pastures upon the growth of wool.” The Society’s Bronze Medal and £25. Awarded. 1884 W. E. Abbott, Wingen, for paper entitled ‘‘ Water supply in the Interior of New South Wales.”’ 1886 8S. H. Cox, F.a.s., F.c.s., Sydney, for paper entitled “‘ The Tin deposits of New South Wales.”’ xx Awarded. . 1887 Jonathan Seaver, r.c.s., Sydney, for paper entitled ‘ " Origin and mode of occurrence of gold- bearing. veins and of the associated Minerals.” 1888 Rev. J. E. Tenison-Woods, F.G.s., F.L.s., Sydney, for paper entitled ‘‘ The Anatomy and Life-history of Mollusca peculiar to Australia.” 1889 Thomas Whitelegge, F.R.M.s., Sydney, for paper entitled ‘‘ List of the Marine and Fresh- water Invertebrate Fauna of Port Jackson and Neighbourhood.” 1889 Rev. John Mathew, m.a., Coburg, Victoria, for paper entitled “‘ The Australian Aborigines.” 1891 Rev. J. Milne Curran, F.c.s., Sydney, for paper entitled “The Microscopic Structure of Australian Rocks.”’ 1892 Alexander G. Hamilton, Public School, Mount Kembla, for paper entitled ‘“ The effect which settlement in Australia has produced upon Indigenous Vegetation.” 1894 J.V. De Coque, Sydney, for paper entitled the “‘ Timbers of New South Wales.” 1894 R. H. Mathews, L.s., Parramatta, for paper entitled *‘ The Aboriginal Rock Carvings and Paintings in New South Wales.”’ 1895 C. J. Martin, D.sc., M.B., F.R.S., Sydney, for paper entitled ‘‘ The physiological action of the venom of the Australian black snake (Pseudechis porphyriacus).”’ 1896 Rev. J. Milne Curran, Sydney, for paper entitled ‘‘ The occurrence of Precious Stones in New South Wales, with a description of the Deposits in which they are found.” 1943. Edwin Cheel, Sydney, in recognition of his contributions in the field of botanical research and to the advancement of science in general. AWARDS OF THE WALTER BURFITT PRIZE. Bronze Medal and Money Prize of £50. Established as the result of a generous gift to the Society by Dr. W. F. Burrirt, B.A., M.B., Ch.M., B.Sc., of Sydney. Awarded at intervals of three years to the worker in pure and applied science, resident in Australia or New Zealand, whose papers and other contributions published during the past three years are deemed of the highest scientific merit, account being taken only of investigations described for the first time, and carried out by the author mainly in these Dominions. Awarded. 1929 Norman Dawson Royle, M.D., ch.m., 185 Macquarie Sivest, Sydney. 1932 Charles Halliby Kellaway, M. C., M.D., M.S., F.R.C.P., The Walter anid Eliza Hall Institute of Research in Pathology and Medicine, Melbourne. 1935 Victor Albert Bailey, M.a., D.Phil., Associate-Professor of Physics, University of Sydney. 1938 Frank Macfarlane Burnet, m.p. (Melb.), ph.p. (Lond.), The Walter and Eliza Hall Institute of Research in Pathology and Medicine, Melbourne., 1941 Frederick William Whitehouse, D.sc., Ph.p., University of Queensland, Brisbane. 1944 Hereward Leighton Kesteven, D.sc., M.D., c/o Allied Works Council, Melbourne. AWARDS OF LIVERSIDGE RESEARCH LECTURESHIP. This Lectureship was established in accordance with the terms of a bequest to the Society by the late Professor Archibald Liversidge. Awarded at intervals of two years, for the purpose of encouragement of research in Chemistry. (THis JoURNAL, Vol. LXII, pp. x-xii, 1928.) Awarded. 1931 Harry Hey, c/o The Electrolytic Zinc Company of Australasia, Ltd., Collins Street . Melbourne. 1933. W. J. Young, D.sc., m.se., University of Melbourne. 1940 G. J. Burrows, B.Sc., University of Sydney. 1942 J. 8S. Anderson, B.Sc., Ph.D. (Lond.), A.B.C.S., D.I.c., University of Melbourne. 1944 F. P. Bowden, ph.D., Sc.p., University of Cambridge, Cambridge, England. PARTS I-II ate ; FOR pe Setar aeees ee 1944 . : tet Ge (INCORPORATED 1881) | I een = — PARTS I-II (pp. 1 to 69) oe s i - Containing Papers read in April, May and July = — cee eS With piste bo es oe EDITED BY | Ls THE HONORARY SECRETARIES — "THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE nee _ STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN s fe ee ae SYDNEY aie: ae 7 LDN PUBLISHED BY THE SOCIETY, SCIENCE HOUSE gee _ GLOUCESTER AND ESSEX STREETS : Pe ee Ries 1048 - CONTENTS VOLUME -LXXVIII | 4 ae Parts I-II Arr. COREE lug wrt ce! Address. By A. B. Walkom, D. Se.. (Issued August 3 230, ‘ oe a ee II —Extension of Maxwell’s Tiaweerns. By P. ‘Foulkes, BSc. 1944) .. oe es iy vat es rae outs Same Vs “s > « Northern N ew South Wales. By Irene Crespin, B. A. (Communicated by Dr 7 Brown.) (Issued August 30, 1944) ae Lat hg Cec me ea a i — . Arr. IV.—A Note on the Role of the Nitrosyl Group in Metal Conpleeee By. Mellor, M.Sec., and D. P. Craig, M.Se. (Issued August . 30, 1944) a ART. V:—The SRS N 3800 i 3400 , ta T MACQUARIE . } ; wa 3200. 5 ? Ay GROVE ROCKS 3800 0 . OBERON = L sh j 95 —~Sao0 yl 1a 9" 20° Map 2. Reconstructed or Generalised Contours for the Orange-Bathurst District. q WILSON H. MAZE. YE;|THOLME e 4000, e Orange-Bathurst District. 38060 e OBERON ‘3800 abeivee SeIND THE SUMMIT GEOMORPHOLOGY OF CENTRAL EASTERN NEW SOUTH WALES. oo 149° 00’ E., and on the north and south by the parallels 35° 15’ S. and 33° 45'S. respectively. (See Maps 1 and 2.) This area, of about 2,000 square miles, is drained by the Macquarie River and the Belubula River and their tributaries. (1) Altimetric Frequency Curves for the Orange-Bathurst District. In Figure 1, the frequency of occurrence of the heights of the ‘* high points ”’ in each 1,000 yards square on the topographic maps covering this area has been recorded. The frequencies have been plotted with the heights arranged as having an interval or range of 50 feet. For instance, heights recorded on the sheets of tracing paper as 2,800 are assumed to range between 2,800 and 2,850 feet. (20) 0) rare FREQUENCY OF OCCURRENCE Oo [e) oO °o xv ” DE 2 800- re) To} UNS. ° ° ° ° 2° 9° ° (2) 8 8 s S ° SiS ° ° 9 3 8 ° 8 q 2 8 & x 3 3 3 a 8 S esi eRe BB 8 S om RON NOR Si ey reas es ALTITU Figure 1.—Altimetric frequency curve for the Orange-Bathurst District. Heights are grouped into ranges of 50 feet. On this graph the range of 3,000 to 3,050 occurs most frequently in the area. Sharp drops in the curve indicating a preponderance of steep slopes occur at 2,800, 3,000 and 3,350 feet. Attention may be drawn to several features of the curve. (i) Between the values of 3,000 and 3,250 feet there exists a block of maximum frequencies which indicates an extensive gently sloping surface between these elevations, bounded particularly on its lower levels by steep slopes. (Maximum break in the curve occurs at the 3,000 feet frequency.) (ii) Minor frequency maxima occurring at the 2,400-2,450, 2,600-2,650, 3,600-3,650 and 3,800-3,850 feet levels indicate local preponderance of surfaces at these levels. (iii) Maximum breaks in the curve, in order of importance, occurring at the 3,000, 3,350, 2,800 and 3,250 feet levels may indicate that breaks of slope occur at these levels. (iv) The curve decreases rapidly from a maximum at 3,000 feet to 2,650 feet ; then it rises and the solid block of frequencies occurring between the 2,650 and 2,400 levels, bounded at each extreme by a maximum, indicates an extensive gently sloping surface between these elevations. The drop in the curve between 2,650 and 2,800 may indicate a steepening of the surface up to the 2,800 level. (v) No attention can be paid to the frequencies for heights below 2,400 feet owing to the limited area in the Orange-Bathurst District at these lower elevations. It is felt that the grouping of the frequencies for an altitudinal range of 50 feet is inadequate when dealing with any extensive erosion surface. On an ancient erosion surface it is almost certain that the range of height from its lowest point to its highest point and its local relative relief would in general exceed this. DD—May 3, 1944. 36 WILSON H. MAZE. — On the assumption therefore that 100, 200 and 300 feet are more likely figures for the local relief of an erosion surface, the data were regrouped and frequency curves drawn. The 200 feet grouping, which is here presented in Figure 2, gave results which are more in accordance with the results from other methods and with field observations. eG z | fis eee ee 1000 @ uJ O Gi w &00 @ [od =) O O : a O RG 600 : nae O z uJ >) : cE uJ t ee u 400 200 ie 0 wee 5 tot oP own oS Or ig jaan ) Os Dig (OL Oa VG UN SW. AGH ais enc, LC MG Oh ee ae ne! ee) vt N () Co) oO + fal) ro) oO o i N Oo vt vt Tt Tt ica) fap) (a9) (a9) in) N N N N lav) ALTITUDE) TN PEE Figure 2.—Altimetric frequency curve for the Orange-Bathurst District. Heights are grouped into ranges of 200 feet. With this greater range, the frequency curve is naturally more striking. The range of heights from 3,000 to 3,200 feet occurs most frequently and is bounded on each side by a sharp drop in the frequency curve. Thus the dominant erosion surface in the region is one varying from a height of 3,000 to 3,200 feet above sea level. This surface stretches from Orange southwards through Blayney. Inspection of the topographic map shows that a considerable area of this surface is dissected to a depth of 1,000 feet. Other parts are reduced to a series of closely spaced, isolated hills, which diminish in altitude from south to north. This surface is portion of the Central Tableland of New South Wales and has been called the Orange Tableland. It would be more correct to refer to it as the Orange Plateau. The next most important group on the graph is that of the frequencies for the range of heights from 2,400 to 2,800 feet. These two columns must be considered together for they are of almost equal value and they are bounded by Sharp drops in the graph at 2,800 and 2,400 feet. The latter drop is accidental, GEOMORPHOLOGY OF CENTRAL EASTERN NEW SOUTH WALES. 37 due to the artificial limits set down in defining the area. This 2,400 to 2,800 feet surface represents what is usually called the Bathurst Plains. The graph suggests that a surface exists (Bathurst Plains) which has high points with an altitudinal range of 400 feet, while the preceding higher and more extensive surface (Orange Plateau) has a range of 200 feet only. This is borne out in Map 2, where these surfaces have been reconstructed. The “ high-point ”’ surface of the Orange Plateau is thus a smoother and more graded surface than the ‘‘ high-point ’’ surface of the Bathurst Plains. The compact group of columns of approximately the same frequency value in the range from 3,400 to 4,000 feet may indicate a gently sloping surface within these altitudinal limits. The sharp break at the 4,000 feet value indicates the smaller number of areas over that height and is due to a number of residuals rising abruptly from the 4,000 feet surface to 4,400 feet. (2) Strip ‘‘ High-Point ’’ Profiles for the Orange-Bathurst District. In Figure 3, six strip high-point profiles running from west to east in the Orange-Bathurst District are presented. Map 2 shows the positions of these strips, each 3,000 yards wide and 10,000 yards apart from centre to centre. The profile JJ, the most northern section, traverses two main surfaces, an extensive one at from 3,000 to 3,200 feet and another between 3,800 to 4,000 feet in the vicinity of Eskdale Trigonometrical Station. The break between the two surfaces is so abrupt as to suggest faulting or sharp warping. The 3,000 feet surface is cut by the gorge of the Macquarie River and by the wider valley of the Windburndale River. The level persists in the area between the two rivers and remnants of it are observed at the base of the steep rise to Eskdale. In profile KK four surfaces are indicated. The lowest is the Macquarie Valley Plain, which is traversed by the channel and narrow flood plain of the Macquarie River. It has an elevation of 2,200 feet and is entrenched some 200 feet in the Bathurst Plains, which in the latitude of this profile have an elevation of from 2,400 to 2,500 feet. The most extensive surface is that at an elevation of 3,200 feet to the west of the Bathurst Plains. In the east the rise to the highest surface of 3,800 to 4,000 feet is very steep in the profile and on the ground surface. On this surface, residuals such as Sunny Corner Trigonometrical Station rise some 200 feet above its general level. At its western end the profile cuts the Canobolas Mountains, high residuals rising 800 feet above the surrounding district. In profile LL the Orange Plateau is traversed at 3,200 feet and the remaining profiles show that this surface rises gradually to the south, where it attains a maximum elevation of 3,400 feet along the line OO. Some residuals such as North Brother, Middle Brother, Three Brothers and Mt. Macquarie stand some 200 to 300 feet above the general level. Along LL the Bathurst Plains are quite wide, and Panorama Hills, basalt-capped residuals to the west of the city of Bathurst, rise above the level of the surface. In the region of profile MM the Bathurst Plains reach their widest extent and have a general elevation of 2,800 feet. Campbell’s River and Fish River are entrenched some 300 feet into the surface; this corresponds with the entrenchment of the Macquarie Valley farther north. (See profile KK.) Profile NN runs to the south of the Bathurst Plains, where the valleys of Campbell’s River and its tributaries have notched the 3,200 feet surface. Profile OO does not exhibit the distinct breaks of slope indicated in the others. Instead the main feature is the long continuous upward slope of the upper surface from Eagle Hawk Hill through the Oberon Plateau to the Summit. This surface is broken by the valley of Campbell’s River and between Oberon and the Summit by the Fish River Creek and Duckmaloi River. 38 WILSON H) MAZE. The western border of the high eastern surface shows a remarkable change from north to south. In profiles JJ and KK it is very abrupt, in profile LL it is still steep but less abrupt. In profile MM a modification is introduced by the Fish River valley, but in profiles NN and OO the abrupt change has disappeared. This is particularly noticeable in profile OO, where the lower western surface, some 3,400 feet high, merges gradually into the higher surface at Oberon (3,800 feet) and the Summit (4,000). aise Orange. Macquarie R. WindburndaleR, Es kdale. Tyldesley Sunny Corner. 4.00 iaptuen TolTen's Hill Crackajack Rock. Portland, untley | | Macquarie R 500q os a om SET ON ae as 2009 K K Cadia Millfhorpe. Rocks. Bathursl. Yet holme. Lambe. Blayney Wimbledon. Campbells R - Jerry's Mt | FishR. FishR J M M Carcoar. Moorilda. CampbellsR Sugar Loaf Fish R 409 BelubulaR North Brother | 3009 Lo OOS — = a WY Sinet a 2°00 ‘ N | N Eagle Hawk Grove Rocks. Rockley Oberon. The Summ. se) cLeary Hill. Campbells R. (@}uie) 50,000 YDS. 100,000 YosO) Fig. 3.—Strip “ high-point”’ profiles for the Orange-Bathurst District. The location of the ae is indicated in Map 2 The strip high-point profiles thus illustrate the occurrence of the following surfaces : (1) The Eastern Surface or Yetholme Plateau, 3,800 to 4,000 feet, with © some residuals (Sunny Corner, Mt. Lambie, etc.) rising to 4,200 feet. (2) The Orange Plateau, 3,000 to 3,400 feet, with some residuals (The Brothers, Canobolas Mts., etc.) rising some 200 feet above its general level. To the south this plateau seems to merge into the iia Plateau, at 3,800 feet. a, GEOMORPHOLOGY OF CENTRAL EASTERN NEW SOUTH WALES. 39 (3) The Bathurst Plains, with an elevation of 2,400 to 2,800 feet. (4) The Macquarie Valley Plain, elevation 2,200 feet. The Yetholme Plateau and the Orange Plateau may be part of the same surface, the Oberon Plateau being the connecting link between them. (3) Reconstructed or Generalised Contours for the Orange-Bathurst District. In Map 2 the high-point surface of the Orange-Bathurst District has been reconstructed by using the high points in each of the 1,000 yards grid-squares as spot levels and drawing the contours at 200-feet vertical intervals. The vast extent and even surface of the Orange Plateau extending from Orange southwards through Blayney is immediately apparent. The detailed contour maps show that the amount of minor valley dissection has, particularly in the south-east corner, partly reduced this area to a number of isolated hills and the extent of the original summit surface is not nearly so apparent. This surface clearly extends to the east along the southern portion of the map at least as far as longitude 140° 35’. To the east of this the deep valleys of Campbell’s River and Fish River isolate the Oberon Plateau (3,600- 3,800 feet) from the Orange Plateau on the west and from the Yetholme Plateau surface on the east. But the spacing of the contours taken in conjunction with the profile OO in Figure 3 suggests that this was originally one continuous surface rising gradually from 3,200 to 3,800 feet and trenched by Campbell’s River and the Fish River. On this basis then it may be again argued that. the Orange Plateau, the Oberon Plateau and the Yetholme Plateau were formerly one and the same surface. This surface was recognised by Andrews (1910) and is considered by him and other writers (Colditz, 1942) to be of Miocene age. The strip ‘ high- point ’’ profile and the generalised contours thus add further confirmation to the existence and continuity of the Miocene surface. In the south beyond the confines of the map the surface appears to be continuous and is clearly part of an east-to-west doming which produced the Blue Mountain Plateau surface. Passing north there may be faulting or a local increase in warping as suggested in profile JJ or KK. The Bathurst Plains are also depicted in extent, and from the crowding of the contours at their margins it is obvious that the 2,800-foot line marks the upper limit in their development. They have apparently been excavated out of the Miocene surface and are probably of Pliocene age. This surface has been described as a senkungsfeld by Sussmilch (1931), who suggests that the boundaries are determined by fault-planes. An examina- tion of the contour map does not support the view that the northern, southern and western boundaries are fault-scarps, since, aS shown by the contour lines, a number of long spurs project out from the upper surface. The western boundary is complicated by the basalt capped Panorama Hills which form a long spur between the Macquarie River and Evans Plains Creek. The latter and its numerous tributaries are actively besieging the Orange Plateau along a steep Slope which is almost continuous with and similar to the slopes forming the deep entrenchment of the lower course of the Macquarie River. The reasons for the steep breaks of slope which occur in various places bounding the Bathurst Plains may be assigned either to faulting or to changes in rock character. No detailed geological maps of this area have been published, but David (1931) and Sussmilch (1931) both indicate that there is a considerable variation in rock character within the area. The Bathurst Plains have been mainly cut in granite. The Orange Plateau is in part covered with Tertiary basalt overlying highly inclined Ordovician and Silurian strata. In the Yetholme Plateau there are areas of granite and Lower Palexozoic strata of Silurian and Devonian age. In 40 WILSON H. MAZE. conversation, Mr. L. L. Waterhouse, who has carried out geological surveys in this area, stresses the fact that the varying resistance to erosion of the rock types has played a very important part in determining the extent and shape of the major and minor landform features. It also is necessary (Bryan, 1940) to bear in mind the possibility that during retreat caused by erosion, the slopes bounding the Bathurst Plains have maintained their steepness. In regard to this Davis (1932) has laid it down that slopes flatten during the progress of the cycle of erosion, whereas Penck (1924) maintains that slopes retreat without loss of their inclination and that steepness disappears only because the land above the grade of the gentle slopes has been consumed. If Penck’s view is correct, and experience would indicate that it is more feasible and acceptable than that of Davis, then once a surface experiences an initial entrenchment, when steep slopes are formed, these will persist in their inclination as they retreat, exposing the new and lower erosion surface. Normal retreat through erosion, then, rather than faulting, may be the cause of the steep slopes bounding the Bathurst Plains. Thus the Bathurst Plains, if eroded out of a higher surface which has not been completely consumed, should be bounded by slopes which in general would exhibit the degree of steepness of the original entrenchment. The slopes to the north, west and south of the Bathurst Plains are only a slightly more gentle version of the slopes of the forge of the lower part of the Macquarie River shown onthemap. These latter slopes have been oversteepened with the new phase of erosion which initiated the cutting of the Macquarie Valley Plain out of the Bathurst Plains. Even in this case the break from the Bathurst Plains to the Macquarie Valley Plain is quite steep, though of small magnitude. The eastern boundary of the Bathurst Plains has been referred to in the previous section. The upper steepness which may be due to faulting, warping or differential erosion has been accentuated by the retreat of the erosion slope | almost to its base. The map and the strip profiles show that there are a few remnants of the 3,200 feet surface along the bottom of the steep upper scarp. The Eastern or Yetholme Plateau surface can be seen in Map 2 to vary in height from 3,800 to 4,000 feet with small residuals of an older surface rising up to 4,200 feet. It is also interesting to note that in both Map 2 and in Figure 3, the Main Divide, separating the eastward from the westward flowing rivers, does not correspond with the ‘ high-point ’’ line; it is invariably to the east and is much lower. SUMMARY. Three methods of landform analysis which can be rigidly and consistently applied to topographic maps have been described. Using these methods, one can set down with some precision facts about the various summit planes which are not at all obvious from mere inspection of the contour map. These methods have been applied to the topographic maps of the Orange-Bathurst District, New South Wales. It has been shown that at least four separate erosion surfaces exist and the map depicts their nature and extent. REFERENCES. Andrews, E. C., 1910. The eas Unity of Eastern Australia in Late abe Post-Tertiary Time. Tuts J OURNAL, 44, 420. Barrell, J., 1920. ‘“* The Piedmont Terraces of the Appalachians.” A.J.S., 49, 228, 327. Baulig, rab 1928. Le Plateau Central de la France. Baulig, H., 1935. The Changing Sea Level. Inst. of Brit. Geographer, Pub. No. 3. Bryan, K., 1940. The Retreat of Slopes. Ann. Assoc. Amer. Geog., 30, 1940. Colditz, Margaret J., 1942. Physiography of the Wellington District, N.S.W. THis JOURNAL, 76, 235. GEOMORPHOLOGY OF CENTRAL EASTERN NEW SOUTH WALES. 4] David, T. W. E., 1931. Geological Map of the Commonwealth of Australia. Davis, W. M., 1932. Piedmont Bench Lands and Primaerruempfe. Geol. Soc. Amer. Bull. 42. Hollingworth, 8. E., 1938. ‘‘ The Recognition and Correlation of High-level Erosion Surfaces in Britain: A Statistical Study.” Quart. Jour. Geol. Soc., 94, 55. Miller, A. A., 1937. ‘‘ The 600-foot Plateau in Pembrokenshire and Carmarthenshire.’ Geog. Jour., 90, 148-159. Miller, A. A., 1939. Attainable Standards of Accuracy in Determination of Preglacial Sea Levels by Physiographic Methods. Journal of Geomorphology, 2, 95. Penck, W., 1924. Die Morphologische Analyse. Sussmilch, C. A., 1931. Physiography of the Bathurst District of New South Wales. Tuts JOURNAL, 65, 80. Vance, T. A., 1940. Mapping a Continent. The Australian Surveyor, 8, No. 3, 151. a ak ae and Morgan, R. 8., 1937. ‘“‘ The Physical Basis of Geography.’ London, pp. -261. THE RELATIONSHIP OF THE AUSTRALIAN CONTINENT TO THE PACIFIC OCEAN—NOW AND IN THE PAST.* By W. H. BRYAN, M.C., D.Sc. INTRODUCTION. I feel singularly honoured at having been invited to give the Clarke Memorial ‘Lecture for 1944. It is an honour to be chosen for the task of commemorating such a worthy scientist and pioneer as William Branwhite Clarke, and it is an added honour to be admitted to the select company of those who have given the lecture in the past. Although Clarke is best known for his geological work in the State of New South Wales, he also played a very important part in laying the foundations of the geology of Queensland. Indeed during the earlier part of his career, that is until 1859, he was employed by the Government of New South Wales in what was then called the Northern District, which included the present State of Queensland. Clarke did not make many visits to this, the northernmost part of his field, but he made most valuable use of his occasional excursions. The extent of his influence on our northern State is best seen not in his own writings, but in the well-known book on The Geology and Paleontology of Queensland as set out by R. L. Jack and R. Etheridge, Jr., in 1892. This great and enduring work, which was published in 1892, is dedicated ‘‘ to the memory of three worthy pioneers in Australian Geology, Samuel Stutchbury, William Branwhite Clarke and Richard Daintree’’. The book itself shows clearly why Clarke was one of the three men so honoured. The text will be found to contain nearly sixty references to his work, and these are not clustered about a few points, but find a place in almost every one of the forty chapters that constitute the book. As a Queenslander and as a geologist, I am glad to acknowledge the extent of our indebtedness to the great man whom we commemorate this evening. THE PACIFIC AT PRESENT. The most striking feature of the Pacific Ocean is, of course, its immensity. It covers more than one-third of the globe; it occupies a greater area than all the continents put together ; it is far larger than any of the other oceans. But although the absolute size of the Pacific is impressive, and may be of the first importance as bearing on its mode of origin, and although its size relative to the other major geographical units is of undoubted interest, these quantitative considerations may not be as significant as those qualitative features that are distinctive of and, in part, peculiar to this ocean. The differences of degree that separate the Pacific from the other major units are important, the differences of kind may well be fundamental. Certain obvious geographical features of the Pacific Ocean as depicted on a terrestrial globe at once attract the attention of even the casual observer. Perhaps the most striking of these is that it occupies a basin almost completely surrounded by land. This is clearly seen on any globe, and on those maps based. * Clarke Memorial Lecture delivered before the Royal Society of New South Wales, May 30, 1944. RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 43 on appropriate projections. Unfortunately, the world maps on Mercator’s and similar projections so commonly in use distort the higher latitudes so much that the principal gap in the encircling land masses (that of 1,500 miles or so which separates New Zealand from Antarctica) looks far greater than it is. This misleading effect is further emphasised owing to the custom of curtailing in many of these maps the higher latitudes of the southern hemisphere (although the same maps include the corresponding latitudes in the northern hemisphere). In this way the large continent of Antarctica is frequently omitted altogether and the Pacific basin wrongly appears to have a huge opening in the south. A second distinctive feature of the Pacific is the close parallelism between the coast line and the adjacent mountain ranges of its eastern margin, which persists almost unbroken along the whole western coast of the Americas. Another striking pattern is to be seen on the western side of the Pacific where a series of outlying island arcs is linked together to form a chain extending down the whole length of the Asiatic coast. Simple observations such as these have probably been made by innumerable school children, generation after generation, but their obviousness does not detract from their significance, and they form the basis of the more elaborate generalisations that have been put forward by trained geographers from time to time. Of these, one of the earliest and most important appears to have been that of von Drasche (1879, p. 265), who recognised that the island ares of Asia, in spite of their distinctive pattern and outlying position, were to be regarded as homologous with the mountainous coasts of the American mainland, and who therefore suggested that the western boundary of the Pacific Ocean proper should be drawn, not along the coast of the Asiatic mainland, but from Kamschatka through Japan, the Philippines, to New Guinea, on to New Caledonia and New Zealand and thence to South Victoria Land in Antarctica. Six years later Suess (1885, Vol. 1, p. 5) argued that the various coasts of the Pacific Ocean, although they might differ in detail, all conformed to the one geographical plan in that they all exhibited a close parallelism between the Shorelines and the direction of the neighbouring mountain chains. Although he did not suggest that coasts of this description were restricted to the Pacific, Suess pointed out that they were so characteristic of it and so different from the coasts of the other oceans as to warrant the term Pacific Type. In contrast, the coasts which exhibited no parallelism between the shore lines and the nearest mountain chains he referred to as the Atlantic Type. The geographical differences between these two coast types, it is important to note, are differences of kind not merely differences of degree, for whereas the Pacific Type is defined on the basis of a positive criterion the Atlantic Type is based only on negative or at best neutral criteria. The geographical elements of the Pacific margin are indeed in such close harmony as fully to merit the term ‘ Concordant Coasts ”’ afterwards introduced by Supan (1911, p. 793). On the other hand the lack of relationship between the geographical and geological features of the Atlantic coasts equally merits the term ‘ discordant ”’. Suess was the first to emphasise that the geographical homogeneity of the Pacific coasts has a geological basis. Earlier workers doubtless anticipated his reference to the important and significant facts that the Pacific margin was paralleled by a volcanic zone—the so-called ‘‘ Girdle of Fire ’’ and a similarly placed seismic zone, where earthquakes were frequent and severe, but Suess made important additions to these well-known generalisations. Thus he pointed out that the bordering mountain ranges are all folded towards the ocean and that a Mesozoic marine series can be traced around the Pacific. J. W. Gregory (1913, p. 41), while in general agreement with Suess’ general- isations, elaborated and strengthened them by his constructive criticism. In 44 W. H. BRYAN. particular he showed that even where the Pacific type of coast was represented by outlying island arcs, the coast of the mainland itself, although very different, © was in harmony with Suess’ general plan. Gregory demonstrated this by subdividing the Pacific coast types into Primary Pacific Coasts which ‘ are determined by the proximity of long lines of fold mountains, to the trend of which the coast is in general parallel. The mountains bounding these coasts are of comparatively young geological age. Over-thrusts and over-folds, if present, are mostly directed toward the ocean ’’, and Secondary Pacific coasts that ‘‘ are due to the subsidence of basins on the inner side of mountain chains along coasts of the Primary Pacific type. They are frequently bordered by active or recent volcanoes. The oldest rocks usually occur along the coast and are followed by younger rocks further inland. Except where horsts project into the sea, the coast-line is approximately parallel to the average grain of the adjacent country.’’ In his Primary Pacific coasts Gregory included the Asiatic arcs together with Taiwan, the Philippines, New Guinea, New Caledonia and New Zealand, while the eastern coast of Asia proper and that of Australia were referred to his Secondary Pacific coasts. Gregory was further of the opinion that both of his Pacific coastal types were represented in Antarctica. As a result of this work Gregory went far to establish the geographical and geological unity of the Pacific as it at present exists, although, as we shall see, he was strongly opposed to the idea of the Pacific Ocean as a permanent feature of the earth’s crust. Most subsequent workers have been in general agreement with Suess’ views as interpreted by Gregory, for they are based simply on facts of distribution, but from time to time modifications have been suggested as new knowledge has been gained or aS some new avenue of approach to the problem has been discovered. Of the more strictly geographical additions two may be selected for reference—Marshall (1911, p. 99) urged the replacement of the New Guinea, New Caledonia, New Zealand margin by another that included the Fiji, Tonga and Kermadec islands; and Taylor (1928, p. 995), while agreeing as to the importance of the Asiatic island ares as usually defined, pointed out that the less conspicuous but very large Mariana arc although ‘‘ new and weak in its develop- ment ’’ must be taken into consideration as a likely alternative for the true edge of the Pacific basin south of Japan. Of the geological contributions that have provided supplementary evidence of the unity of the Pacific, the most interesting was perhaps that of Becke (1903, p. 125), who pointed out that, whereas the volcanoes circling the Pacific were predominantly andesitic—we may remind ourselves that the very name of the rock andesite is based on its common occurrence in the Andes—the volcanoes of the extra-Pacific regions are characterised by the alkaline basalt, tephrite. Of this important generalisation Suess (Vol. IV, p. 588) wrote: ‘* Thus there is a tephritic or Atlantic series, and an andesitic or Pacific series. The Atlantic series is characterised by the greater quantity of alkalis, especially sodium, while in the Pacific series the alkalis diminish and calcium and magnesium occur in greater quantity.’”’ He added that ‘‘ All the coasts of the Pacific Ocean— from New Zealand to Java, Alaska and all the western coast of America belong to the andesitic series’’. Becke had also pointed out that the Pacific type volcanic rocks were associated with folding movements while those of the Atlantic type were related to faults. | Harker (1909, p. 90), who had been working along similar lines for many years, brought forward a comparable but even more comprehensive scheme of distribution, which was not confined to volcanic rocks or to recent geological times. Jn this he divided the igneous rocks of the world into two great groups, which he named the ‘“ Atlantic suite’’ and the “ Pacific suite ’’ respectively on account of their segregation into two great petrographical provinces arranged ne hale ba piahite, r *, ¥. He 3 fd 7 RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 45 in and about these oceans. The Atlantic suite was characterised by alkaline rocks and the Pacific by cale-alkaline. Furthermore the Atlantic and Pacific suites were associated with structures due to tension and to compression respectively. The next advance along these lines was made by Born (1933, p. 759), who concentrated on the geographical relationship of the andesitic lavas characteristic anne a \ sean | 4. | “Bi iia os Z OS yay — “ee WER, CSS AUSTRALIA'S PLACE IN THE PACIFIC. a (he Marshall) Lire. o—o— Borr7's Andesite Line. ---- The Limits of the Sersmic Belt. of the circum-Pacific lands to that of the basalts characteristic of the Pacifie itself. He argued that the line of demarcation, ‘ the Andesite Line’’, was, in the western Pacific, only in partial accordance with the edge of the basin as defined by von Drasche, Suess and Gregory, for it followed the Mariana are in the north and the Marshall Line in the south, thus giving strong support to these important modifications of the original plan. 1D) 46 W. H. BRYAN. The most recent contributions to our knowledge of the nature and limits of the Pacific basin are based on seismological observations. These have recently been summarised by Gutenberg and Richter (1941) in their most informative ‘‘ Seismicity of the Earth ’’. Although the earlier views with regard to the distribution of earthquakes gave strong support to the claim of Suess that the Pacific basin was a structural unit, the support was only of a general kind. Thus it was known that the great majority of earthquakes, especially the large ones, occurred within a circum-Pacific belt, but the belt was not clearly defined and appeared to include some quite large areas within the Pacific basin proper. The result of Gutenberg and Richter’s work has been to give far more specific and even more convincing support to the hypothesis that the Pacific is not only a homogeneous unit but differs qualitatively from all the other major geographical units, continents and oceans alike. Their principal conclusions with regard to the circum-Pacific belt may be summarised as follows: (1) it includes a large majority of the (common) shallow shocks with epicentral depths of less than 30 kilometres. (2) It contains a still larger proportion of the inter- mediate shocks from depths between 30 and 300 kilometres. (3) It contains all the shocks from a depth greater than 300 kilometres. (4) It separates areas of continental structure from those of Pacific structure. (5) It surrounds an area (the Pacific basin proper) which is ‘‘ conspicuously inactive ”’. The circum-Pacific seismic belt as now defined is of sufficient precision to afford definite information on two debatable sections of the margin of the basin. Thus to the south of Japan the belt is quite clearly split into two branches, one of which follows the inner island arcs adjacent to the Asiatic mainland (the edge of the basin as suggested by von Drasche), the other following the festoon of the Mariana are and agreeing nicely with the position of the Andesite line as drawn by Born. But the belt to the south-east of New Guinea follows only the outer of the two suggested margins and thus lies beyond New Caledonia and follows the Marshall Line very closely. fy Attention may be directed particularly to the deep earthquakes listed by Gutenberg and Richter, which total no fewer than 212. Although the origin of these is still in dispute, there is general agreement as to their significance as indicators of deep-seated instability. Writing on the “ Structure of the Pacific Basin as indicated by Earthquakes ’’ Gutenberg (1939) stated: ‘‘ The fact that the foci of all earthquakes originating deeper than 200 miles have been found close to and on the continental side of this [cireum-Pacific] boundary indicates that the Pacific Basin has a unique structure.’’ This all-important conclusion Gutenberg supports by another line of evidence, namely the peculiar nature of the sea floor under the Pacific. Reviewing Suess’ well-known suggestion that the floors of all the oceans are composed of sima (as contrasted with the sial of the continents), he has concluded from a study of earthquake travel times under the several continents and oceans that ‘ the surface structures of the Atlantic and Indian Oceans are similar to those under the continents but thinner. . .”’ whereas ‘‘ All evidence agrees with the conclusion that the layers which form the uppermost crust in the continents are lacking in the Pacific Basin. . .” This restriction of sima to the floor of the Pacific, while it introduces an important modification to what was probably the most famous of Suess’ general- isations, is nevertheless quite in accord with the spirit of that genius who never tired of emphasising the essential differences of origin, history and structure that separated the Pacific and Atlantic oceans. Gutenberg’s contribution represents an advance rather than a conflict, and one feels sure that Suess would have welcomed it as such. The evidence converging from many points has now reached such impressive proportions and is so clearly interlocked and mutually confirmatory that the RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 47 present unity of the Pacific basin is now accepted by the great majority of geologists, although not all of these may yet be convinced as to its uniqueness. THE PACIFIC IN THE PAST. If we agree to accept the unity of the Pacific as sufficiently well established, our next step should be to determine, in so far as this can be done, how long it has existed as a unit. It is patent that its unity was not achieved in recent geological times. The ring of volcanoes which circumscribe the basin are not all active. Many are dormant and others appear to have been extinct for very considerable periods. The mountains that circle the Pacific are due to folding movements which, although confined to the Cainozoic, may well have been initiated very early in that era. Further, many geologists holding quite diverse views with regard to the structure and history of the Pacific basin have been impressed by the evidence on which Haug based his generalisation for a continuous Mesozoic sea, the position of which was in close harmony with the present margins of the basin, although few of them follow Haug in his conclusion that the site of the Pacific was then oceupied by a great continental mass. The most recent reference to Haug’s Mesozoic “ geosyncline ’”’ is that of Wade (1941), who in his paper on ‘‘ The Geology of the Antarctic Continent and its Relationship to Neighbouring Land Areas” wrote that ‘ there can be little doubt that the Andean geosyncline is continued by way of the great loop to Graham Land and crosses the Antarctic Continent through James W. Ellsworth and Maria Byrd Lands. Nor can there be any doubt that the geosyncline passing through New Zealand also connects on the opposite side of the continent to Graham Land and is part of the same circum-Pacifiec geosyncline.”’ Back beyond the Mesozoic era there seems to be little direct evidence either for or against the existence of the Pacific basin as such. It is not the intention here to canvass the detailed evidence for and against the permanence of ocean basins, but the following points may be emphasised. Dana, the original protagonist of the theory, placed all the oceans in the one category and argued that they were all essentially permanent features of the earth’s crust. Suess adopted a different attitude and consistently emphasised the contrasts between the Atlantic and Pacific Oceans, claiming that they were different in structure, history and origin, and that the Pacific was the older and more permanent feature. Suess’ claim has been strongly fortified by recent seismological advances which indicate that the Pacific is even more fundamentally different from the Atlantic than he himself had supposed. ° J. W. Gregory, although he was largely responsible for demonstrating the present unity of the Pacific, remained unchanged in his early belief that all the ocean basins belonged to the same category and that none was permanent. He argued as vigorously and confidently against the permanence of the Pacific in 1930 as he had against that of the Atlantic in 1929. Further he expressed the opinion that the Pacific might even be the youngest of the oceans. The permanence of the Pacific throughout geological time is far from established, but its assumption as a working hypothesis is, I think, warranted for the following reasons : if, as the recent evidence suggests, the Pacific basin is a unique structure, its origin may well have been due to some special event or circumstance early in the earth’s history, that may not have been directly relevant to the origin of the other oceans. But even if this be not admitted and the Pacific be retained in the same category as the other oceans, every argument of a general nature in support of the permanence of ocean basins can be applied _ with special cogency to the Pacific, while general arguments against permanence are weaker when applied to that ocean than to others. 48 W. H. BRYAN. If, then, we assume that the enormous Pacific basin has been a permanent feature on the earth’s surface, its effects, both direct and indirect, on the history of the earth must have been prodigious. For, in the first place, its birth may well have been due to some earth-shaking event, the repercussions of which have come echoing down the ages. But even if its origin were unaccompanied by any such dramatic circumstance, the very presence of such a large and permanent structure would doubtless have had far-reaching and long-lasting effects upon the course of geological history. But, this evening, we are concerned not so much with these general considera- tions aS more particularly with the implications on the structural and stratigraphical history of the Australian continent, of the permanent presence of such a powerful neighbour. Let us then change our point of view and now look at the Pacific Ocean from our Australian vantage point. AUSTRALIA AT PRESENT. The Australian continent as it exists at present is by far the smallest of the major land areas and appears almost insignificant beside the mighty Eurasian continent to the north. It is small even compared with Antarctica, which is half as large again. And if we consider relative masses instead of areas Australia becomes even less impressive on account of its low average height—far smaller in bulk than any of the other continents, and only one-twelfth that of Antarctica. Indeed, as a continent, whether we consider its area or mass, Australia seems quite out of scale. On every side Australia appears to be geographically incomplete, but our present interest is with that part fronting the Pacific. Here, quite clearly, the minor topographical features of the land surface can be traced beneath the Pacific. This may be simply explained as due to a submergence of the coast of only two hundred feet or so. But many major geographical features also appear to be truncated at the coast line and something far more important than a relatively small change in sea level is indicated. Geologically, the incompleteness of Australia is even more marked. It reminds one of a damaged picture in which so much has been torn away that the composition of the remainder, incomplete and out of balance, leaves us wondering what the picture as a whole looked like. It is of especial interest that the first scientist to realise the geological incompleteness of Australia was the man in whose honour we are assembled this evening and whose discernment in this as in other geological matters was unusually acute. Clarke (1878, p. 7), basing his conclusions on the complete absence of marine strata of Tertiary age from the eastern coast of Australia, suggested that a former extension has been lost by subsidence into the Pacific. He added that ‘‘ this has some support in the fact that there is a repetition of the Australian formations in the Louisiade Archipelago, New Caledonia and New Zealand—in the latter of which occur abundant Tertiary deposits ”’. The present coast line has, at most, only a secondary significance, in that, instead of forming the natural limit to the geological structure, it merely reflects the general geological grain. We have seen that Gregory could not agree with Suess that the coast of Australia proper was comparable with the more typical coasts of the Pacific, and placed it with his Secondary Pacific coasts which in no case indicate the true limit of the land. They are essentially coasts developed within a coast. Some geographers, including Sir John Murray, emphasise the importance of the 100-fathom line as probably representing the edge of the continent proper. In Queensland the outer edge of the Great Barrier Reefs virtually coincides with this line for a good part of its length and thus gives it added emphasis. Some years ago (1928), using as my basal assumption that the 100-fathom line She: a wes RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 49 represents the common limit of continental masses, I endeavoured to explain the variable breadth and strangely convex projection of the Queensland continental shelf as a natural extension to the east. In view of the fact that the eastern edge of the continental shelf and the western edge of the highlands are symmetrical about the coastal ranges of Queensland, I suggested the hypothesis that the continental shelf represents the drowned half of the Queensland highlands of which the coast ranges formed the dominant axis. A consideration of the structural geology of Queensland appeared to support the hypothesis, for the known highlands are composed of Paleozoic rocks laid down in a geosyncline,, the supposed eastern edge of which roughly coincided with the edge of the continental shelf. Spender (1930, p. 279), commenting on this suggestion, expressed the opinion that I had not gone far enough, and that at least for a considerable part of the Queensland coast the 100-fathom line does not mark the eastward limit of the continental mass since ‘*‘ from Saumaurez Reefs in lat. 22°8. ... to . Osprey Reef in lat. 14°8. ... there is a great bank within the 1,000-fathom contour ”’. Extending our observations still further to the east we note that the now isolated island of New Caledonia has all the appearance of being an outlier of the main Australian mass. It fits perfectly into the geological picture. In 1925, after pointing out that in addition to the more conspicuous geological trend to the north-north-west, Queensland rocks also exhibited an older trend to the north-east, I wrote that ‘‘ it is a notable and probably significant fact that the north-east trend is most strongly developed in that part of Queensland adjacent to the abnormally wide portion of the continental shelf which is here projected in a north-easterly direction. It is also known that parallel north-easterly trend lines actually pass through New Caledonia. The presence of these north- easterly trends in the old metamorphies of this island (in which all the newer rocks trend N.N.W.), which Suess regarded as one of the three most puzzling features of the Melanesian ares, is thus simply explained if New Caledonia be regarded as part of an ancient Australian continent.”’ Most Australian geologists would, I think, be in agreement with Jensen’s (1936) statement that ‘‘ the island of New Caledonia is a remnant of a once continuous continent, the Melanesian plateau, which extended westwards to eastern Australia and New Guinea, and south perhaps to New Zealand ”’. But New Caledonia does not necessarily mark or even approximate the eastern edge of the Australian structure, as Woolnough (1903) showed us over forty years ago when he described from Fiji rocks of ancient aspect which he concluded might well be Archean in age, and which indicated that the Fiji Islands instead of being a cluster of oceanic islands were remnants of a continental mass that once extended eastwards from Australia. . With regard to the true limit of the Australasian mass, the most satisfactory Statement and one for which Australian geologists are particularly indebted is that of Marshall (1911), who in his reasoned and critical address to the Australasian Association for the Advancement of Science, concluded that : “1. Bathymetrical, structural and petrographic characteristics support the: idea that the real boundary of the south-west Pacific passes through New Zealand, Kermadec, Tonga, Fiji, New Hebrides, Solomon and on to the Admiralty Islands. . That this practically coincides with biological knowledge as to plant: and animal distribution within the area. 3. That the land connection or approximation took place in the late Mesozoic or in the Pleistocene, probably in both. 4. That the eastern Pacific islands are different in structure, nature and origin from the lands on the line of islands mentioned, and that they have derived their fauna and flora by chance migrants from them.”’ Lo 50 W. H. BRYAN. Marshall added that ‘it is perhaps necessary to explain what one means by the statement that the island line so often referred to is the real margin of the Pacific basin. By this statement it is intended to convey the meaning that any movement of elevation or depression, or any rock movements that may have affected that portion of the earth’s crust that lies to the west of this line, may have left all that which lies to the east unaffected. In other words, the structural margin is supposed to mark the limit of that portion of the area that according to geological ideas may in the past have formed an eastern extension of the Australian continent.”’ The marginal line as thus defined has frequently been mentioned with approval in geological literature. Indeed it is most striking to observe how geologists of almost every creed are unanimous in accepting the fact of the Marshall Line, although they may disagree fundamentally as to its implications. Thus while Schuchert regarded it as marking the limit of a vigorous continent at the acme of its development, T. C. Chamberlin (1916) thought of it as the boundary of an abortive mass that had failed to achieve the status of a continent proper. | ~~ Let us consider the significance of the Marshall Line as it affects Australian geology. There is a tendency, seldom stated explicitly but frequently implied, to regard the present Australian continent as intact, complete and sufficient to itself—a perfect example of geological autarchy! If the submerged outlying portions are taken into consideration at all they are regarded merely as the unstable margin of a stable continent—the ricketty front porch of an otherwise well constructed house. Let us make an inspection of this front porch. First, let us measure its dimensions. It reaches to Tonga 2,000 miles from the nearest point of our present coast. Its area is approximately three million square miles, that is about equal to that part of the Australian mass now above sea level. It is certainly a large porch. Now let us consider its stability as compared with the rest of the house. The only satisfactory criterion for comparing the stability of neighbouring areas as they exist today is that of seismology, and this shows quite clearly that at the present time there is a seismic zone developed on either side of the Marshall Line within which seismic disturbances are frequent and severe, but the area between the inner edge of this seismic belt and the present coast of Australia, far from being ricketty, is every bit as stable as Australia itself. To take a specific example, New Caledonia is, judged by the criterion of seismicity, as stable as central Australia. During the past seven years I have carefully scrutinised every record obtained by the Seismological Station of the University of Queensland in my search for evidence of earthquakes between the Australian coast and the Marshall Line. Earthquakes on or near the Marshall Line itself are of almost daily occurrence, but they never occur within the magic circle of the seismic zone. I have watched particularly for disturbances, even small ones, from the edge of the Great Barrier Reef, which especially might be expected to yield evidence of mobility, but although our station is in an excellent position for receiving such shocks, not one has been recorded. One cannot avoid the conclusion that one large continuous stable mass now extends almost to the edge of the Pacific basin proper, which is also an essentially stable area. The Marshall Line thus shows up all the more vividly as the mobile hinge between two great stable masses. Recently, investigations have begun in our seismological station with a view to determining the nature and thickness of the material composing the 1 As a tribute to an eminent geologist I have referred to it in this lecture as the Marshall Line. = RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. ol Australasian mass. The results of this are far from complete, but the first indications, as calculated by my assistant, Mr. N. de Jersey, are that Australia proper consists of sial to a depth of approximately 40 kilometres, and is in this respect closely comparable with western North America, while the area between the Kermadee and the present Australian coast also consists of sial, but in a much thinner layer of about 25 kilometres, and thus comparable with Gutenberg’s findings for the floor of the Atlantic. This suggested difference in thickness as between the emergent and submerged portions of the continent, important though it may be, does not appear to have any effect on the present stability of the submerged part whatever it may suggest with regard to its mobility in the past. We may, I think, reasonably conclude that the alleged ricketty front porch is in fact a large and important part of the house proper and that it is as firm as the remainder. The Australian continent of the geographers is, we may conclude, not, as commonly believed, a geological unit. It is not an integer, it is a fraction. With the frank recognition of this we should abandon the geological isolationism that characterises so much of our work, and seek to understand the position of Australia in the general scheme of things and, more particularly, of the part it plays as one of the circum-Pacific continental masses. AUSTRALIA IN THE PAST. We might well begin with a consideration of the most appropriate carto- graphical basis on which to place the various interpretations of the growth of our eontinent. It is customary at present to study the past history and geography of Australia with reference to its present geographical outline. This has certain obvious advantages, the chief of which is that we are referring that which is doubtful to that which is certain. But these are offset by equally marked disadvantages attached to the method. The uniformitarian dictum that ‘ the present is the key to the past ”’ is irrefutable in as far as it means that the known ig the key to the unknown, but in as much as it suggests that the present is to be regarded as the normal state of affairs it is highly dangerous. Dr. Woolnough has consistently and with good reason warned Australian geologists against using present day conditions as a measuring stick for the events of the past. The present geographical outline of Australia does not represent the norm, and therefore should not be used as our standard of reference. To a geologist, looking at a map of Australia is like viewing the last picture of an incomplete cinematograph film. Viewed as a *‘ still ’’ it has all the appearance of permanency, but seen as an episode in the geological story of Australia it is merely a transitory flick. The present geographical outline is neither normal nor permanent ; nor does it represent a culmination. It is a point arrived at, not a goal achieved. The question thus arises, is there any better basis for a paleogeographic map than the present continental outline? I suggest that the most logical back- ground available for all but the Upper Tertiary maps is the area limited by the Marshall Line. It is not likely that this line has been rigidly fixed throughout our geological history, but even so, it is likely to be a more significant datum than the present transient coast line. The suggestion is not strictly original for, thirty years ago, my old friend and teacher, Dr. A. B. Walkom (1919) set us all a good example by publishing with his Mesozoic maps of the Australasian region a reference map showing the ‘ approximate limits of the Continental Mass, of which Australia, New Zealand, New Caledonia, etc., are remnants ’’. Perhaps a satisfactory compromise with present practice could be effected if our paleogeographic maps showed both the Marshall Line and the present coastal outline, the former to indicate the distribution of land and water within the 52 W. H. BRYAN. Australasian mass, the latter for ease of geographical reference. In this way our convenience might be served and at the same time our perspective retained. I propose that we now consider the history of our continent in terms of the great Australasian mass noting, at the same time, how this history fits into the picture and story of a permanent Pacific. It would be convenient for our purpose if we could niche use Of some generally approved summary of the geological evolution of Australia, in which controversial details disappeared into a background of generalisations acceptable to all. Unfortunately, no such general statement is possible, for Australian geologists are as strongly at variance in their interpretations of the major aspects of our geological history as they are about the significance of stratigraphical details. If, instead of completely accepting any particular viewpoint, we endeavour to amalgamate the principal features of the several schools of thought, we are confronted at the very outset by an impasse, for Australian geologists have long been in sharp disagreement as to the nature of the pre-Cambrian ‘ foundation block ”’ on which all our geological history is based. On the one hand are those who regard it aS representing the minimum development of our continent, and on the other are those who picture it as marking the maximum! And even if we evade this initial obstacle by beginning our history with the Paleozoic. era, we are left with the impossible task of reconciling the opinions of those who hold that the history of Australia has been essentially a triumphal march to the east, with those others who see it as a dogged retreat to the west. It is patent that divergences such as these cannot be coaxed into parallelism and we are perforce left with the task of examining seriatim the several interpretations of our continental evolution. Although many geological papers deal in part or by implication with the development of Australia, three principal hypotheses have been formally propounded. These, in the order in which they appeared, may be called, for convenience, the hypotheses of Marginal Growth, of Continental Fracturing, and of Progressive Welding respectively. The first and third of these are clearly related and both may be regarded as hypotheses of constructive advance. The second, by contrast, is an hypothesis of destructive retreat. The hypothesis of marginal growth was advanced by an early and powerful school whose influence is still strongly felt and who held that the Australian continent was originally only half its present size, extending no farther to the east than a line joining Darwin and Melbourne. From this small beginning the continent was supposed to have advanced step by step towards the east, the continental shelf of one period becoming the coastal strip of the next. In this way, lifting itself by its own boot-straps as it were, the little continent marched into the Pacific. Jensen (1911), although he does not claim to be the originator of the hypothesis, is the author who has stated it most explicitly and in greatest detail. Several geologists, including EH. C. Andrews in particular, have pointed out that other aspects of our geological history were in keeping with this simple programme, and that the history of diastrophic events and of the major plutonic intrusions, together with the spatial distribution of ore-bodies, were all consistent with the one general plan. So that, in the end, Australian geologists had visions of an aggressive and completely organised army, horse, foot and artillery, advancing successfully upon the Pacific in a veritable ‘‘ Drang nach Osten ”’. The result of this advance was claimed to be the consolidation of the Australian continent, much as we know. it now, and the (temporary) acquisition of a large mobile area to the east. Even as late as 1932, Sir Edgeworth David (p. 49) remarked in his ‘‘ Explanatory Notes ”’ that ‘“‘ as regards the evolution of the Australian continent, the point has frequently been stressed that after the disappearance of the Proterozoic Nullagine seas, deposition was mostly restricted RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. Dd: to what is now the eastern portion of the continent, and that there was a steady gain of land eastwards at the expense of the sea ”’. Although this attractive theory was based largely upon interpretations of events restricted to the Paleozoic era it came in time to be regarded as an epitome of our complete geological history. Indeed it developed ultimately into what was virtually a philosophical background for Australian geology. The Australian continent was seen as a stable unit, the result of a number of victorious conflicts, followed by final consolidation of the positions successively won. These ideas were seldom explicitly stated, but their implications may be detected intricately interwoven through many a writing on the history and growth of our continent. In particular they are clearly visible in the most recent statements of Andrews, who, however, as we shall see in the sequel, no longer agrees with the theory of marginal growth in its simple original form. The second hypothesis, that of continental fracturing, was introduced not by an Australian but by the eminent American geologist, Schuchert (1916, p. 94), who challenged the commonly accepted thesis of marginal growth and launched another that was radically different. Schuchert’s hypothesis was built upon the assumption of an originally great continental mass that became progressively smaller by loss to the Pacific. It replaced the very small pre- Cambrian continent of the earlier hypothesis by a very large one; the marginal seas of the Palzozoic by a central Mediterranean (the Tasman geosyncline) ; a migrating locus of deposition by a relatively fixed one; a continent growmg at the expense of an ocean, by an ocean growing at the expense of a continent ! Schuchert saw Australia not as a stable unit achieved by constructive effort, but as the shrunken remnant of a disintegrated continent. Such a startling statement from one who was already recognised as a world authority on stratigraphy and paleogeography would, one might expect, have caused a sensation in geological circles in Australia. It did not, as far as I can gather, create even aripple. The basal philosophy underlying the interpretation of the growth of a continent had been turned topsy-turvy—and apparently nobody cared. At least there was no immediate reaction expressed in the printed word. The third hypothesis, that of progressive welding, is clearly a lineal descendant of the hypothesis of marginal growth to which it bears a strong family resemblance, and although they differ in important respects they are both imbued with the same spirit. The chief sponsor of progressive welding is Andrews, who has stated his case clearly and explicitly in two recent papers (1937, 1938). In his own words, ‘ the nuclei of the existing continent appear to have been formed, step by step, far back in pre-Cambrian time, by a process of marginal growth, by the addition, or accretion, of mountain chains. These mountain systems are composed in the main of rocks which have been subjected to severe compression or folding. In the late pre-Cambrian, these blocks, once coextensive as surface features, appear to have been separated by the formation, within them, of structural depressions. A reasonably safe point for the start of our studies, however, appears to be the closing pre-Cambrian period.”’ ‘* Beginning then, at this stage, it may be shown that the stable block known today as Australia (with Tasmania) has resulted from the welding, or knitting, together of these ancient nuclei. This welding, fusion or stabilisation, took place, almost wholly, during the Paleozoic era.”’ The relationship of this to the first hypothesis is clearly exhibited in these quotations and is also shown in passages such as this (p. 182): ‘ great arcuate belts of Silurian, Devonian, Carboniferous and Permian rocks in this region [New England and eastern Queensland] were attached successively, as dry land, to the growing continent of Australia ”’. D4 W. H. BRYAN. On the other hand the important differences which separate this variant from the original hypothesis and which show the belated influence of Schuchert’s hypothesis are (1) the recognition that the pre-Cambrian basement extended at least as far as the limits of the present continent; (2) the acceptance of the © Tasman geosyncline a8 a major geological structure; (3) the addition of the Palgwozoic borderland of Tasmantis. But these appear as admissions almost grudgingly conceded. Thus Andrews represents the Tasman geosyncline as an extraordinarily narrow, ribbon-like feature, confined within the present continental limits, while the borderland of Tasmantis as depicted by him appears quite inadequate for its important geological roéle. In these and other points the hypothesis of BECoEoseiye welding exhibits the awkward misfits of unhappy compromise. . Although the three hypotheses differ in many minor details, they are most widely at variance in their opinions as to the former extension of the Australian continent into what is now the Pacific Ocean. In searching for evidence bearing on this most controversial question, the critical region for investigation is obviously the easternmost strip of Australia as we now know it. Consideration of the structure and history of this area may well supply us with that link between the known and the unknown which we are endeavouring to find. This critical region, we should note was, according to Schuchert, occupied by the western half of the major structural and geographical feature that he called the Tasman geosyncline, the eastern half lying beneath the Tasman Sea, from which it got its name. Schuchert showed the geosyncline as a persistent trough “ which began to appear as a seaway in the Ordovician, had its greatest development in the Devonian’ and “ continued with some interruptions throughout the Carboniferous and Permian’. It was completely enclosed at its northern end which, in the Permian, is shown as near Cape York, and had a large opening at its southern — between Tasmania and the southern end of New Zealand. Much detailed information about the geology of this part of Austraha that was not available when Schuchert published his paper is now at hand, and it is pertinent to enquire as to the bearing of this later knowledge on the question at issue. In particular our information concerning south-eastern Queensland is now far less incomplete as a result of Denmead’s (1927) valuable investigations. We now know that the oldest rocks of this easternmost part of Australia are the Greenstones, which form the basal series of that large assemblage of sedi- mentary and volcanic rocks, metamorphosed to varying degrees, which we know collectively as the Brisbane Schists. The Greenstone Series is made up almost entirely of metamorphosed sub-basie and basic volcanic rocks which, though they display some variety, quite obviously all belong to the one suite. The base of the series is not visible, but Denmead now estimates their thickness from the lowest known horizon to the top of the series as 10,000 feet. The age of the Greenstones is unknown but David (1932, p. 36) suggested an age ‘‘ between Proterozoic and Ordovician ’’ and assigned them tentatively to the Newer Proterozoic. Conformably above the Greenstones and merging into them by interbedding lie the basal members of the Bunya Series. The series consists essentially of grey and green mica phyllites (locally represented by quartz mica schists), with phosphatic cherts, slates and quartzites in the upper portion. The series as a whole presents remarkable uniformity over a great thickness, estimated by Denmead at 18,000 feet. Near the top of the series there has been found (and lost) one specimen diagnosed as a Diplograptid, on the strength of which the Bunya Series has been assigned to the Ordovician. The 28,000 feet of phyllites RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 55 and greenstone below this horizon may, as David indicates, well cover the Ordovician, Cambrian and part of the Proterozoic. The phyllites of the Bunya Series pass conformably and without any sudden lithological break into the Neranleigh Series. This is composed essentially of greywackes (non-calcareous) and thin-bedded quartzites with a total thickness of 15,000 feet. In the lowest part of the series one sees an alternation of sedi- ments of the Bunya and Neranleigh types, but once the new sedimentary type (greywacke) is well established it persists with remarkable uniformity, showing @ purity of facies comparable with that of the Bunya Series. Both the Bunya and Neranleigh Series are essentially siliceous and the Fernvale Series which follows is extraordinarily so, being made up for the most part of grey and green banded cherts and red radiolarian jaspers. It is perhaps not so strikingly uniform as the two preceding series and differs from them in the more pronounced development of interbedded andesites. The top of the series is unknown, but Denmead estimates the visible thickness at approximately 10,000 feet. The age of the series cannot be directly determined, but latest estimates suggest that it is pre-Devonian. At some time subsequent to the deposition of the Fernvale Series, the Brisbane Schists were folded into a large but simple anticlinal structure with a moderate dip to the east and an almost vertical dip to the west. Denmead, who first described this important feature, modestly referred to it as the “* Indooroopilly Anticline ’’, but David was so seized with its importance that he promoted it to the ‘ Brisbane Geanticline ’’ and showed it as one of the few text figures in his Explanatory Notes, describing it as a “‘ vast denuded geanticline perhaps originally 30,000 feet high ”’. The evidence presented by the several series of the Brisbane schists is cumulative, mutually-confirmatory and free of ambiguity. It points to the presence of a geological structure, the many features of which are so consistently those to be expected in a geosyncline that it may be regarded not merely as typical but as exemplary of its kind. In particular the Greenstones exhibit just those chemical, mineralogical and petrological features that we might expect as the result of vulcanicity related to the establishment or previous existence of a geosyncline. The Bunya Series is composed of a great thickness of sediments all of which indicate deposition in deep quiet waters possibly far from land. The greywackes of the Neranleigh Series are also characteristic of geosynclinal sediments, although suggestive of deposition nearer one of the geosynclinal margins. The radiolarian jaspers and banded cherts of the Fernvale Series are again characteristic geosynclinal deposits and indicate a reversion of conditions to deeper and clearer water further from land, while the accompanying interbedded andesites are the very volcanics to be expected under these conditions. Throughout the whole range of the Brisbane Schists totalling well over 50,000 feet, there is a remarkable absence of normal conglomerates, sandstones and limestones. Hxcept for radiolaria, one graptolite and certain indefinite carbonaceous markings are the only signs of life. These negative criteria added to the positive criteria based on the uniformity and persistence through great thicknesses of the appropriate lithological types fully satisfy the conditions for geosynclinal sedimentation (as opposed to that of shelf seas) as recently set out by O. T. Jones (1938) in his presidential address to the Geological Society of London, ‘ On the Evolution of a Geosyncline ’’. The evidence and conclusions outlined above have all been collected from one small area, deliberately chosen, first for its critical geographical position and second because it has been studied in considerable detail. But the Brisbane Schists are by no means confined to this small area. The Greenstones, it is true, seem to be somewhat localised, but I have examined rocks DG W. H. BRYAN. typical of the Bunya phyllites from Nambucca Heads (N.S.W.) in the south to Emu Park (Queensland) in the north, a distance of over 600 miles. The Neran- leigh greywackes also have their counterparts in north-eastern New South Wales, while Professor Richards and I have seen them typically developed on the Keppel Islands. The Fernvale jaspers have been confidently correlated with the Woolomin Series of New South Wales, and they too have been traced as far north as Rockhampton. It may well be that these series are represented even farther to the south and north. Indeed my colleague, Captain F. W. Whitehouse, informs me that he has recognised the principal types of the Brisbane Schists from Narooma, on the southern coast of New South Wales, to the highlands of New Guinea. At the very least, there is ample evidence that closely similar lithological types were deposited in the same sequence over an established length of 700 miles. With regard to the width of the area in which these geosynclinal deposits. were laid down, there is little direct evidence. To the west the same bathyal types persist with no suggestion of change until they disappear beneath a cover of later rocks. To the east, the typical deep-water deposits vanish under the sea, but a clue as to a possibly restricted extension in this direction during the deposition of the Neranleigh Series is provided by the greywackes themselves, which are not so typical of the median portion of a geosyncline, and more particularly by certain gritty beds in the valley of the Nerang River, which contain rounded boulders to two feet in diameter and which as Denmead (1927, p. 84) points out are of a rock unknown in southern Queensland. In general it would seem likely that the original width of the geosyncline must have been considerably greater than the width of the deep-water sediments that now represent it, and in particular that (except possibly in Neranleigh times) the eartern margin was well out under the Tasman Sea. The igneous activity associated with the Brisbane Schists was restricted to that characteristic of geosynclinal conditions, namely contemporaneous flows and tuffs of a sub-basic and basic character. These, after providing an impressive prelude, as the Greenstone Series, form a recurrent theme, albeit with some minor variations, that is still dominant at the top of the Fernvale Series. There was a notable absence of folding movements during the deposition of the Brisbane Schists. The several] series are all conformable and grade one into another and the regional metamorphism of similar lithological types becomes gradually less intense with each successive series. The simple but enormous structure of the Brisbane geanticline, the formation of which closed the history of the Brisbane Schists, has all the features to be expected of a median geanticline produced during the uplifting of bathyal deposits that had been accumulated in the middle of a trough. On the evidence provided by the Brisbane Schists alone (and it must be remembered that this evidence was not available to Schuchert) one may deduce that a large geosyncline, at least 700 miles in length, extended to the north beyond Rockhampton and to the east well beyond the present coast line ; further, that the geosyncline was already fully established at the beginning of the Paleozoic era. The essential simplicity of this major geographical feature during a large part of geological history as shown by persistence of facies, constancy of position and absence of overfolding and overthrusting, suggests further that, at least in this stage of its evolution, the Tasman geosyncline was sheltered to the east, not by a narrow mobile borderland, but by a broad stable buttress of almost continental dimensions ; that the Brisbane Schists were laid down in a mediterranean sea in the heart of a great Australasian continent. If we wish to continue our inquiry into the evolution of the Tasman geosyncline we must leave the Brisbane area and move, not to the east and RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 57 north as the hypothesis of marginal growth would suggest, but to the west and south. There at Silverwood Professor Richards and I (1924) have investigated a series of jaspers very similar to those of the Fernvale (and Woolomin) Series and presumably coeval with them, which is followed by the 6,000 feet of andesitic and spilitic volcanics which constitute the lower half of the Silverwood Series. Almost at the top of these very uniform volcanic rocks are the lenses of coralline reef-like material known as the Silverwood Limestone. ‘This horizon has several points of interest. It is the first strongly developed calcareous type met with in the history of the geosyncline as developed in southern Queensland ;_ it presents us with the first really satisfactory evidence of shallow water conditions ; it is the first horizon the age of which (Couvinian) can be definitely determined. But the Silverwood Limestone did not herald a permanent change from the bathyal facies, for it was very shortly followed by banded radiolarian cherts and shales showing that conditions of deposition were again essentially similar to those of the Fernvale Series. The top of the Silverwood Series is unknown but there are at least 5,000 feet of banded shales becoming progressively less chertified as the series is ascended. Since the hypothesis of progressive welding agrees with that of continental fracturing in admitting the existence of the Tasman Geosyncline from the Devonian to the end of the Paleozoic era, there is no need for the further review of the evidence bearing on the continued existence of that structure. Suffice it to say that the first definite change from sedimentation of the bathyal type (and the dominance of siliceous rock types), to typical neritic deposits (of dominantly calcareous type), appears within the Carboniferous strata of the Rockhampton Series. After that change there is no further reversion to deep- water sedimentation but, on the other hand, there is, following the development of Permian paralic deposits in many parts of eastern Australia, a transition to lacustrine and terrestrial conditions. Perhaps the outpourings of andesitic lavas and tuffs that preceded the deposition of the lowest of the Mesozoic fresh- water deposits in southern Queensland (the Andesitic Boulder Beds at the base of the Esk Series) mark the final manifestation of the influence so long exerted in eastern Australia by the Tasman Geosyncline. A suggested interpretation of the development of the Tasman Geosyncline in southern Queensland is shown in the accompanying table. When we assemble all the available data, including that of the Brisbane Schists, we have, I suggest, adequate evidence of the existence of one of the major geographical features of the world, a feature comparable in size with the Mediterranean Sea which now separates the continents of Europe and Africa and a feature which remained virtually constant in position throughout the whole of the Paleozoic era. In its size and persistence, although not in the minor details of its development, it is comparable with the great Appalachian Geosyncline of North America. The rival hypotheses of Schuchert and Andrews, although they agree as to the existence of the Tasman Geosyncline in Upper Paleozoic times, do not, however, agree as to the nature and extent of the continental mass beyond the geosyncline. Schuchert pictures it as ‘‘ the eastern half of the Australasian continent, a land about 1,800 miles east and west and 2,200 miles north and south ’’. Andrews, although he points out that the limits are unknown, pictures it as of far smaller dimensions. Sussmilech and David (1920, p. 277) evidently regarded it as of considerable size and referred to it as ‘‘ a separate land area [in the Carboniferous] which existed to the east of the Australian continent at least as far back as the beginning of the Devonian period and probably as far back as the beginning of the Palexozoic era’. They stated further that this land, which they suggested should be | 4 WoOryEryTUT “OSB [VAYQV “UOMISUBLI, W. H. BRYAN, “OSVUd JILIN “UOTPSUBL FE euljsnoery “IUTPILLASOV¥) JO A10481 7 “SUOTIIPUOY) Bee | a2 3 | | { | | { | “ONSO[OIg dMBdjOA. "a[BOs pusis uo posUuocy -o1d pue snonuyuop ‘[BIOT PUL [VUOTSRIDQ) "9994 1ULL05UT “g[eos vpqerepIsmoo uo pue snonuULD | | | i es | “(orduAy st) 4199) TUL ye | -19JUT puke pasI[eoo'y * post voor qveyMouos yng snonuiyuo0o 4ysowLy ‘onsovIdy = OTUBOTOA ‘AUAIQOV O[sepuy jo syoodq | { | | | | | | i | | *S9U0}S -pues pue sya ‘QU0JSOUTT 91400 "spaq Japfnog = aI4Isapuy ‘sorBys pue souoys | -pUuvs ‘SozVLONLO[SUOD ‘aUlIqSNIV T | "SOUOJSPUBS ‘SJTI) “SOUOISPUBS DULICUL uyIM = pappeqiequr soreys 1078 M-Ysely ‘oT[eled *[81094IT a Se Ea ae ‘NOLLVINGANIGHS 40 CAOLYN ‘ANVISNHGNO x TVULNAO ANY) NUAALNIOS NI GHdOTHATC SV “SaTesBq ue soquAydiod soqids ‘soyisopuy i——— 04UI Sutssed —— soqizqienb yqTAr ‘soqe[s pus soqAyd | uoals pues Aaly —— 0ojJUI sulssed —— soqizzIend | EA SosOVMADLY ojul SuIssed | SozIsopue | pues szleyo popueq ua015 ‘srodsel UvMVvpOIpea poy ‘saqtyids SOJISIPUYW [e109 [jeurg ‘soyvys pure sz1oyo UBLIVJOIPel papueg pue *syjoorl "sqzotfo pure sopeys 4 | ‘TEUI[OAASO+) *)]0Z010}01q “QUOJSUIIIF) : ee “eAUNG “YSIOTUBLO N “OTBAUIOT “TIMOT “pOOMIATIS ‘toddgq ‘pooMmIoalig ‘1OMO'T ‘uoy dw eyyooy ~ ‘roddy ‘Woyd ule yyyI0Y ‘IOMOT ‘SYOOTG yneg ‘eddy ‘syoorg anew “IOMOT SHS “todd SHSuT ee ANTITONASOWS NVWSVIL GAL | ‘OIOZOW[Cd >} “OlOZOSOW ‘SBIG RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. 59 known as “‘ Tasmantis ’’, ‘‘ became permanently united tu the mainland towards the close of the Permo-Carboniferous Period ’’. When we leave the Paleozoic history of Australia and enter the Mesozoic, the three hypotheses which deal formally with the evolution of the continent exhibit interesting developments. Andrews’ hypothesis of progressive welding realigns itself with the original hypothesis of marginal growth, from which it had diverged, in affirming that the stabilisation of the small Australian continent was virtually complete by the end of the Paleozoic era. As a necessary corollary of this conclusion, the Mesozoic extension to the east, although admitted as such, was not recognised as constituting a positive step in the evolution of Australia. Rather it was considered spatially as no more than an excrescence of the true continent, and temporally as but a postscript to the story proper. This attitude of depreciation is in strong contrast with that adopted by Schuchert, for whom the events of the Mesozoic were of the first importance as providing the basis for his thesis that the progressive disintegration of the eastern half of the great Australian continent occurred principally during that era. Schuchert suggested that this supposed continental fracturing actually - began before the close of the Palzozoic era. He advanced no specific reasons based on local evidence for this opinion, but apparently selected this time of widespread diastrophism in other parts of the world as an appropriate date for the initiation of the destructive forces in Australasia. But there is much local evidence to suggest that the Hercynian orogenies were not nearly so pronounced in Australia as elsewhere. Further, although the disappearance of the great Tasman geosyncline in late Paleozoic times marked a geographical and geological transformation of the first importance, it may be. explained, without recourse to violent orogeny, as due to simple filling accompanied by gentle, if intermittent, uplift. If this explanation be accepted it is unnecessary to assume any marked mobility in the complementary borderland at this time, still less the severe fragmentation that Schuchert shows in his late Palzozoic maps. In conformity with his main thesis, Schuchert’s Mesozoic maps depict the eastern half of Australasia as an archipelago, changing in the shape and distribu- tion of its constituent parts from period to period. But, strangely enough they show a minimum of change between the Cretaceous and the Eocene, when the effects of the widespread Laramide revolution might have been expected to have brought about important geographical modifications, especially as this orogeny is well represented in Queensland, New Caledonia and New Zealand. At this stage attention should be drawn to the work of Walkom (1918) who, having specialised in the study of the Australian Mesozoic deposits, published an instructive series of maps which were radically different from those of Schuchert in that they showed the eastern half of Australasia, not aS an ever-changing archipelago, but as a great continuous mass that persisted throughout the Mesozoic with no more marked geographical modifications than those exhibited by the other continents during that era. Such changes as he shows, and more especially the impressive redistribution of land and water in the Lower Cretaceous, may well be regarded as no more than local examples of the widespread transgressions which flooded all the continents, including Australia proper, at that time and which thereby provided maximum geographical changes accompanied by only a minimum of structural deformation. Benson (1923, 1924) gave strong support to Walkom’s main conclusions as to the Mesozoic extension of the continent and introduced modifications and elaborations of his own based on numerous geological details collected from the scattered literature. These he marshalled with meticulous care in his two important papers dealing respectively with the paleogeography of Australasia and the nature of its structural margin. 60 W. H. BRYAN. A most important item in the several attempted reconstructions of the Australasian region in Mesozoic times concerns the relationship of New Zealand to the main Australian mass. Schuchert, in accordance with his scheme of early fragmentation showed the two regions as definitely divorced early in the Triassic. Marshall (1933), in an address which reads almost like a declaration of independence, emphasised the many differences between New Zealand and Australia and expressed the opinion that the former has been a separate and distinct unit at least since the Middle Mesozoic. Walkom, on the other hand, shows New Zealand as part of the one continuous land area as late as the Lower Cretaceous, while Benson’s map pictures the two regions as still securely linked in Upper Cretaceous times. The possibility of such diverse interpretations on this matter as on other aspects of our Mesozoic paleogeography is, of course, due to the fragmentary nature of the evidence on which, in common, they are based, and which consists of the incomplete geological records of a few widely scattered localities. The only points of general agreement in the several interpretations of the Mesozoic era appear to be that (1) an eastern extension of the Australian continent (either in the shape of a continuous mass or a detached archipelago) reached far out into the Pacific, and (2) at least by the end of the era ‘ the various portions of Australasia ceased to have any striking unity of geological history ”’ (Benson, 1923, p. 53). At the beginning of the Tertiary era there were then, in the Australasian region, a number of independent geographical units which, by that time, appear to have achieved some geological independence as well. But, if the seismological evidence is to be accepted, these units are all nevertheless part of one continuous sheet of sial extending eastward from Australia and clearly circumscribed by a well defined and active seismic zone. Tt is difficult, if not impossible, to reconcile this unity and disunity as both due to the same evolutionary process. In particular it is difficult to explain them both in terms of circum-Pacific relationships. For these latter, although they exhibit some variety of expression, are always and everywhere consistent among themselves and all are variations of the same compelling process. It would seem that the origin of the Pacific, however it may have been brought about, produced, as an immediate and direct consequence, the setting up of an essentially simple structural relationship between the ocean basin and the surrounding continental masses. This expressed itself as movements of compres- sion concentrated in a circum-Pacific mobile belt or hinge area in which the stresses as between ocean and continent were accommodated. This compression manifested itself most clearly as overfolding and overthrusting of the circum- Pacific continental rim towards the Pacific basin, but complementary underthrusting outward from the basin may have been equally pronounced although more difficult to prove. Such an essentially simple structural relationship would, I suggest, tend to perpetuate the pattern originally imposed—not to destroy it. There might and probably would result changes in the area of the Pacific from time to time, but the primal pattern would persist with little modification. But in the south-western sector of the Pacific the pattern is now only weakly displayed. Indeed the geographical evidence of its very existence is incomplete and ambiguous. Fortunately, the seismic evidence is clear and direct, but the continuity of sial from Australia to the Marshall Line and the well-defined seismic zone are almost the only Australasian witnesses to the unity and persistence of the Pacific structure. Between Australasia and Antarctica even the seismic evidence is incomplete and the pattern disappears, and in place of the Pacific rim there is only a region of subsidence. RELATIONSHIP OF AUSTRALIAN CONTINENT TO PACIFIC OCEAN. - ‘61 It is difficult to see how this foundering in the south-west quadrant can be reconciled with the inherited circum-Pacific forces in which tension is only an occasional and incidental factor. And why should forces which elsewhere tend to perpetuate a well established plan seek, in Australasia, to destroy it ? Is it not more probable that this movement had a completely different and independent origin ? May it not have been due to the imposition of an extraneous force on an area previously dominated by the circum-Pacifie control, with the result that the effects of the latter were almost neutralised in Australasia and quite negatived near Antarctica ? Only a force of world-wide dimensions would, I submit, be capable of disturbing the balance of the great Pacific structure, and of interfering with its directive mechanism. Schuchert, we have seen, invoked just such a world tendency to explain the fragmentation of the continents and especially of Australasia, but he believed it to have been initiated even before the Mesozoic, not as is here suggested, at the close of that era. If we are right in thus regarding the foundering of Australasia as independent of and in actual conflict with the circum-Pacifie control, then the expansion of that ocean in the Tertiary era no longer appears as a successful advance, for it has been made only at the expense of its own unity. Geographical extension in such circumstances is as much an indication of structural failure as is geological collapse. Consequently phrases such as ‘‘ the Pacific has grown at the expense of Australia ’’ are misleading, and untrue except in the strictest geographical sense. Emphasis should, I think, be laid on the concept of Australia and the Pacific as partners equally involved in the failure of an apparently well-estab- lished mechanism which previously had preserved their mutual relationships. CONCLUSION. We have now covered the ground we set out to traverse and, although the survey has necessarily been a hurried one, I submit that the evidence examined, although by no means conclusive, is consistent with the following propositions : (1) That the basin of the Pacific Ocean is the greatest and most important of terrestrial structures. (2) That as a structure it is homogeneous, permanent and unique. (3) That the structural boundary of the Pacific basin, in the south-western quadrant, is represented by the Marshall Line. (4) That a great Australasian continent existed immediately to the west of this Line from as early as pre-Cambrian times and persisted throughout the Paleozoic and Mesozoic eras. (5) That forces of extra-Pacific origin brought about the dismemberment of the Australasian continent in early Tertiary times. (6) That nevertheless the structural limit of the original Australasian mass is still clearly indicated by a well-defined seismic zone. It will be seen that the adoption of this interpretation of the relationship of the Australian continent to the Pacific Ocean necessarily involves the denial alike of hypotheses of Palzeozoic expansion and of Mesozoic decline, and definitely precludes the acceptance of the present Australian continent as a complete geological unit. LITERATURE CITED. Andrews, E. C., 1937. Structural History of Australia during the courant (The Stabiliza- tion of a Continent.) Proc. Roy. Soc. N.S.W., 71, pp. 118, et. seq. —-——_—————— 1938. Some Major Problems in Structural Geology. Proc. Linn. Soc. N.S.W.. 73, pp. xi-xiii. Becke, F., 1903. Verh. Ges. deutsch. Naturf: und Aerzte zu Karlsbad. (Cited in Suess, Vol. 4, p. 588.) F 62 : W. H. BRYAN. Benson, W. N., 1923. Palzozoic and Mesozoic Seas in Australasia. Trans. N.Z. Inst., 54, pp. 1-62. 1924. The Structural Features of the Margin of Australasia. Trans. N.Z. Inst., 55, pp. 99-137. Born, A., 1933. Der geologische Aufban der Erde. Handbuch der Geophysik, Vol. 2, p. 759 and Fig. 306. (Cited in Gutenberg and Richter, 1941, p. 303, Fig. 20.) Bryan, W. H. 1925. Earth Movements in Queensland. Proc. Roy. Soc. Qld., 37, pp. 1 et seq. 1928. The Queensland Continental Shelf. Reports Great Barrier Reef Cte., 2, p. 58. Chamberlain, T. C., 1916. The Origin of the Earth, p. 210. Clarke, W. B., 1878. Remarks on the Sedimentary Formations of New South Wales, fourth edition, p. 7. David, T. W. E., 1932. Explanatory Notes to Accompany a New Geological Map of the Com- monwealth of Australia. Denmead, A. K., 1927. A Survey of the Brisbane Schists. Proc. Roy. Soc. Qld., 39, pp. 71 et seq. von Drasche, R. N., 1879. R.N. Jahrbuch Min., pp. 265-269. (Cited in Suess, Vol. 2, p. 207.) Gregory, J. W., 1913. The Nature and Origin of Fiords, pp. 26 et seq. -—-———. 1929. The Geological History of the Atlantic Ocean. Q.J.G.S., 85, Pres. Add., pp. Ixvii et seq. —-—_-_—_—_——— 1930. The Geological History of the Pacific Ocean. Q.J.G.S., 86, Pres. Add., pp. lxxii et seq. Gutenberg, B., 1939. The Structure of the Pacific Basin as Indicated by Earthquakes. Science, 90, pp. 456-8. Gutenberg, B., and Richter, C. F., 1941. Seismicity of the Earth. Geol. Soc. Am., Special Papers No. 34. ; Harker, A., 1909. The Natural History of Igneous Rocks, p. 90. Jack, R. L., and Etheridge, R., 1892. Geology and Paleontology of Queensland. Jensen, H. I., 1911. The Building of Eastern Australia. Proc. Roy. Soc. Qld., 23, p. 154. —-——_—_————— 1936. Problems in the Geology of New Caledonia. Proc. Linn. Soc. N.S.W., 61, p. 263. Marshall, P., 1911. A.A.A.S., 13, Pres. Add., p. 99. 1933. Stability of Lands in the South-West Pacific. A.N.Z.A.A.S., 21, Pres. Add., pp. 410-411. Richards, H. C., and Bryan, W. H., 1924. The Geology of the Silverwood-Lucky Valley Area. Proc. Roy. Soc. Qld., 36, pp. 55 et seq. Schuchert, C., 1916. The Problem of Continental Fracturing and Diastrophism in Oceanica. Am. Jour. Sci., 42, pp. 94 et seq. Spender, M., 1930. Island-Reefs of the Queensland Coast. Geog. Jour., 76, p. 279. Suess, E., 1885. Das Anlitz der Erde. (Authorised English translation.) Supan, A., 1911. Grundzitige der Physischen Erdkunde. (Cited in Gregory, 1913, p. 30.) Sussmilch, C. A., and David, T. W. E., 1920. Sequence, Glaciation and Correlation of the Carboniferous Rocks of the Hunter River District, New South Wales. Proc. Roy. Soc. N.S.W., 53, p. 277. Taylor, F. B., 1928. North America and Asia: A Comparison in Tertiary Diastrophism. Bull. Geol. Soc. Am., 39, p. 995. . Wade, A., 1940. The Geology of the Antarctic Continent and its Relationship to Neighbouring Land Areas. Proc. Roy. Soc. Qld., 52, p. 28. Walkom, A. B., 1919. The Geology of the Lower Mesozoic Rocks of Queensland. Proc. Linn. Soc. N.S.W., 43, p. 105. Woolnough, W. G., 1903. The Continental Origin of Fiji. Proc. Linn. Soc. N.S.W., 16, pp. 457-495. ERRATA. Page 63. Line 1. Delete ‘‘ The Status of Melaleuca Fasciculiflora Benth.”’ Line 16. Delete ‘‘ and M. decora (Salisbury) Britten, q.v.”’ Line 17. For 1822, read 1852. { j ‘A s , (a ; Se ‘ 7" 2, SoS H + ai { NOTES ON THE NOMENCLATURE AND TAXONOMY OF CERTAIN SPECIES OF MHELALEUCA. By EDWIN CHEEL. Manuscript received, June 19, 1944. Read, July 5, 1944. THE STATUS OF MELALEUCA FASCICULIFLORA Benth. In working out a plant from South Australia, which is apparently Melaleuca fasciculiflora Bentham (1866), certain difficulties arose in connection with the Spelling of specific names and citation of papers published by Turczaninow, on the Myrtacee collected by Drummond in Australia. As no copy of Turezaninow’s paper was available in any of our lbraries I wrote to Sir Arthur Hill, Director of the Royal Botanic Gardens, Kew, England, who very kindly submitted my queries to Mr. T. A. Sprague for report, with the result that I am now able to clear up some of the discrepancies between the citations given in Mueller’s ‘“‘ Fragmenta ’’ and those given in ‘ Index Kewensis ’’. The following notes deal with certain species of Melaleuca, of which there is an abundant collection of herbarium material in the National Herbarium of Sydney, collected by various collectors in different parts of South and Western Australia. Examination of the material of W. fasciculiflora Bentham leads to the conclusion that Bentham included three distinct species under that name, viz. WM. bracteosa Turez., M. brevifolia Turez., and M. decora (Salisbury) Britten, q.v. Melaleuca bracteosa Turez. (1822), vide “ Index Kewensis ”’. In Mueller’s ‘‘ Fragmenta ”’, VIII (1872-1874), 184, the spelling is given as ‘‘ bracteata’, but according to Mr. Sprague, who has kindly looked up the original work of Turezaninow at my request, the spelling and citation given in ‘‘ Index Kewensis ’’ are correct. This is particularly important since there is a M. bracteata F.v.M. in ‘‘ Fragmenta ’’, I (1858-1859), 15, which is united with M. genistifolia Sm. by Bentham (1866), but since regarded as a valid species by Baker and Smith (1910), with which I am entirely in accord. Turezaninow’s paper, ‘‘ Myrtaceze Xerocarpice, ete.’’, was originally published in Bull. Phys. Math. Acad. Sc. St. Petersbourg, X, Col. 321-346 (1852), as stated in the Royal Society’s Catalogue of Scientific Papers, Vol. VI, 66, No. 17. There are two columns per page, and the columns, not the pages, are numbered. Turezaninow’s paper was reprinted in Melanges Biologiques Bull. Acad. Petersb., I, 394-428, and it was this reprint that was cited by Mueller in his “ Fragmenta ’’, VITI, 182 (1872-1874), hence the discrepancy in the pagination. Alike in the Bulletin and the Melanges, the name of the species appeared as Melaleuca bracteosa, not bracteata as cited by Mueller. Melaleuca brevifolia Turcz. Syn. M. fasciculiflora Benth. in part. Leaves alternate, crowded, 4-6 mm. long, distinctly petiolate (the petiole although small is more clearly defined than in Drummond’s No. 159, which Bentham included with WM. fasciculiflora—which is M. bacteosa Turczaninow), somewhat terete in general appearance but slightly flattened on the upper surface, sub-acute at the apex, or scarcely obtuse. Flowers sessile in lateral G—July 5, 1944. 64 EDWIN CHEEL. head-like clusters, frequently coalescing so as to appear to be in lateral spikes. Petals very deciduous, jagged along the margin. Stamens in bundles, the claw slightly shorter than the petals. Capsules 3-5 mm. diam., the rim comparatively thick, valves somewhat sunken. Drummond’s No. 164 quoted by Bentham under M. fasciculiflora belongs to this species, as also the Gordon River specimen collected by Maxwell. This species is quite distinct from M. brevifolia of Mueller, which is quoted as a synonym of M. microphylla by Bentham. Melaleuca decora (Salisb.) Britten (1901). Syn. Metrosideros decora Salisb. Melaleuca genistifolia Sm. The original specimens described by Salisbury (1796) were collected by David Burton in the Port Jackson district. It was placed in the genus Metrosideros but afterwards transferred to Melaleuca by Smith (1797), who evidently had not seen Salisbury’s specimens, hence the specific name “‘ decora ”’, which has a prior claim, was not taken up. According to Bentham (1866), who records it under the name WM. genistifolia, it has a very wide distribution in Australia, but, as he included M. lanceolata Otto (1822) and M. bracteata Mueller (1858-59) as synonyms, which are now recognised as distinct species, the localities mentioned by Bentham for M. genistifolia (=decora) and copied by various botanists may have to be considerably reduced. Specimens in the National Herbarium of Sydney are represented from numerous localities along the coast from Sydney to Nowra in the south, and as far north as Newcastle, extending to Bullahdelah, Stroud, in New South Wales, and Jimboomba and Sunnybank in Queensland. (Burton states that the plant was commonly known as ‘‘ White Tea Tree’’. In Queensland it is called ‘‘ Ironwood ” or *‘ Ridge Myrtle ’’. Melaleuca lanceolata Otto (1822). Syn. M. genistifolia Baker and Smith (1910) non Smith (1797) ; M. parviflora Moore and Betche (1893) ; M. parviflora var. latifolia Maiden and Betche (1916). ~ This is @ tall shrub or small tree from twenty to thirty feet tall with white papery bark and small flat acute lance-shaped leaves, usually about one-half to three-quarters of an inch long, and not revolute or tapering as in M. decora and M. bracteata. It has been confused with M. parviflora Lindley! (1839) and M. decora. It is quite common in the Gosford district, extending to the Crawford River near Bullahdelah and Wallace Island. Melaleuca bracteata I’.v.M. (1858-59). The original specimens described by Mueller (1858-59) were collected at Moreton Bay in Queensland, and regarded as an intermediate species between M. styphelioides and M. curvifolia. Bentham linked it as a synonym under . M. decora (syn. M. genistifolia), but as shown by Baker and Smith (1910), it is quite distinct. It is a small tree from twenty to forty feet tall, with a rough brown bark and somewhat umbrageous habit, usually found hugging the sides of creeks and rivers, very plentiful between the Gwydir and McIntyre Rivers in the Boggabri and Narrabri districts, extending to Bingarra and Warialda and thence to Lismore and as far north as Rockhampton in Queensland. The specimens included under M. genistifolia by W. V. Fitzgerald (1918) — and by Ewart and Davies (1917) from the Kimberleys in north-west Australia and Northern Territory respectively, as well as those from the Torrens River, 1 See Cheel (1926), where M. parviflora of Lindley is treated as a glabrous form of M. pubescens. NOMENCLATURE AND TAXONOMY OF CERTAIN SPECIES OF MELALEUCA. 65 Mueller (1856), and Ord River, Mueller (1880), are very probably sufficiently distinctive to be regarded as a variety or new species. | MELALEUCA NERVOSA (LINDL.) CHEEL N. COMB. Syn. Callistemon nervosus Lindl. (1848). This species was described by Lindley from specimens collected at Mitchell’s Camp of 16th July, 1846, which is quite close to Mantua Downs on the Claude and Nogoa Rivers, south of Springsure, north Queensland. At p. 235 of Mitchell’s work the following remarks are published: ‘‘. . . and we found a magnificent new crimson Callistemon, with its young flowers and leaves wrapped in wool ”’. In a footnote of the same work a description in Latin by Lindley is published as follows: ‘‘ ramis pallidis foliis ovatolanceolatis, quinque-nerviis mucronatis Junioribus tomentosis, rachi calycibusque lanatis’’. It will be noted that no mention is made of the cohering of the filaments by Lindley, hence there has been much confusion concerning this species as to whether it should be placed in Callistemon or Melaleuca. Callistemon nervosus Lindley is included by Bentham as a synonym under M. Leucadendron var. ? parvifolia (rock, Balmy Creek, in the interior of Queens- land, Mitchell). In a previous paper (Cheel, 1917, p. 298) I included M. Crosslandiana W.V.Fitzg. as a form under MM. leucadendron var. coriacea, and stated that it differed only ‘“‘ in the flowers being crimson ’’, whereas in var. coriacea they were greenish-yellow. Since the publication of the above, a communication received from Mr. Fitzgerald contained the following remarks: ‘‘I do not agree with Mr. Cheel in assigning M. Crosslandiana to M. leucadendron. Besides the crimson filaments mentioned by him, there are other details, including the rough, fibrous, persistent grey bark and hard reddish-brown timber of the former. It is certainly not a ‘‘ Paper-bark’’. In view of this further information, together with a fuller description published by Fitzgerald (1918)—which I had not seen previously, I am now of the opinion that M. Crosslandiana W. V. Fitz- gerald should be included as a synonym under M. nervosa (Lindl.) Cheel. On the subject of the generic distinctions between the two genera Callistemon and Melaleuca the remarks of Mueller (1863-64), which are in Latin, and may be translated as follows, are of interest: ‘“‘ The genus Callistemon is in my opinion, entirely artificial, not safely distinguished from Melaleuca by a single character and better united with it. Even the species which is the type of the genus Melaleuca i.e. M. Leucadendron, which abounds from the vicinity of Port Jackson through East and North Australia has a doubtful position between the Melaleucas with fasicicled stamens and the true Callistemon as originally observed by Brown. Thus Callistemon nervosus Lindley belongs to Melaleuca Leucadendron var. minor. This species varies extremely in habit; in moist valleys and on river banks, occasionally subject to inundation, the tall arborescent form is found ; in arid, stony and sandy tracts the low tree or shrubby form is found. The filaments appear sometimes whitish-yellow and sometimes citron-yellow, the height is also very variable; the staminal bundles adhere occasionally at the base in a ring as happens also in Callistemon lanceolatus, and often they are all free ; the filaments of the staminal bundles are sometimes only very slightly united and sometimes considerably. The bark of several species is paper-like and flakey.’’ Notwithstanding the remarks of Mueller, Bentham (1866, p. 124) regarded the genus Melaleuca as a ‘“ well-defined group readily distinguished from Cal- listemon, by the 5-adelphous stamens . . . The only exceptions are one or two Species in which the claws of the staminal bundles are so short as to connect the genus with Callistemon of which one species C. speciosus has the stamens almost 66 EDWIN CHEEL. or quite 5-adelphous, but single transitionary species appear scarcely to justify the union of very large groups otherwise well characterised.’’ The species C. speciosus referred to by Bentham was originally described by R. Brown (1812) as Melaleuca paludosa (and is retained in the genus Callistemon, as the venation of the leaves is characteristic of all the species of the latter). The species Callistemon lanceolatus mentioned by Mueller above was also noted by Bentham (1866) as having the ‘‘ stamens united at the base’’; a compound description was drawn up for both species by Bentham. Britten (1901, p. 37) published a Latin description copied from Solander’s Mss. together with a drawing (Fig. 109) of a plant collected by Banks and Solander in 1770, on the Endeavour River, which he erroneously classified as Callistemon rigidus. 'This is Metrosideros viminalis Gaertner, I, 171 (1788), tab. 34, Fig. 4, which includes C. lanceolatus Mueller and Bentham in part, mentioned as having been collected by Fitzalan on the Pine River in Queensland.’ REFERENCES. Baker, R. T., and Smith, H. G., 1910. Jour. and Proc. Roy. Soc. N. S. Wales, xiv, 601. Bentham, G., 1866. Flora Australiensis, 1, 144. Betche, E. See Maiden. Betche, E. See Moore. Britten, J., 1901. Botany of Cook’s Voyage (Banks and Solander), edited by Britten, ii, 37, fig. 109. 1919. Journ. Bot., 71. Brown, Robert, 1812. In Aiton’s Hort. Kew., ed. II, iv, 410. Cheel, E., 1916. In Maiden’s Forest Flora of N.S. Wales, vi, 15. ———-—— 1917. In Ewart abd Davies, Flora Northern Territory, 298. —-——_—— 1926. Proc. Linn. Soc. N.S. Wales, li, 408. Ewart, A. J., and Davies, O. B., 1917. The Flora of the Northern Territory, 208. Fitzgerald, W. V., 1918. The Botany of the Kimberleys, Jowrn. and Proc. Roy. Soc. of Western Australia, 85. Lindley, J., 1839. In Edward’s Bot. Register, xxv, App. 8. — 1848. In Mitchell’s Journ. Trop. Australia, 235. Moore, C., and Betche, E., 1893. Handbook of the Flora of N.S. Wales. Maiden, J. H., and Betche, E., 1916. Census of N.S. Wales Plants. Mueller, F. von., 1858-59. Fragmenta Phytographie, i, 15. -—-— 1863-64. Ibid., iv, 55. 1872-74. Ibid., viii, 184. —_-___—_-—_-—— 1869. Plants of North-West Australia, collected by John Forrest. Otto, F., 1822., Iin-Link, Hnwm. Pl. Hort. Berol.. un, 272. Salisbury, R. A., 1796. Prodromus Stirpium in horti. ad Chapel Allerton Londonie. Smith, James Edward, 1797. Trans. Linnean Soc. London, ii, 273. Turezaninow, 1852. Myrtacee Xerocarpicer, Bul. Phys. Math. Acad. Sc., St. Petersbourg, x, col. 321-346. Reprinted in Melanges Biologiques Bull. Acad. Petersb., i, 394-428. 2 See Cheel (1917) in Maiden’s F. Fl., VII, 15, for further particulars concerning these two species under C. viminalis (Sol.) Cheel. Pe ies, THE CHEMISTRY OF BIVALENT AND TRIVALENT RHODIUM. Part VII. COMPLEXES WITH DIETHYL SULPHIDE. By F. P. DWYER, M.Sc., and R. 8S. NYHOLM, M.Sc. Manuscript received, June 7, 1944. Read, July 5, 1944. Coordination compounds of the dialkyl sulphides with the elements of Group 8 have been extensively studied in the cases of platinum, palladium and iridium, the former two elements of which yield well defined compounds of the general type MX,.2R.S (Jensen, 1935; Angell, Drew and Wardlaw, 1930 ; Mann and Purdie, 1935), while tervalent iridium compounds IrX,.3R,S have been described by Ray and Adikari (1934). In the present investigation, the compounds of rhodic halides with diethyl sulphide have been studied, and attempts have been made to reduce these compounds in accordance with the usual procedure, to the corresponding bivalent compounds. In all cases the reduction has failed, owing, it is suggested, to the weakness of the bond between sulphur and rhodium in the trivalent compounds and the probability that in the bivalent compounds this bond would be even weaker. This latter suggestion is supported by the fact that diethyl sulphide fails to coordinate with the more basic members of Group 8—iron, cobalt and nickel—and in the cobalt sub-group, reacts only slowly and incompletely with trivalent iridium which is markedly more acidic than trivalent cobalt or rhodium. Bivalent rhodium has been shown to be distinctly basic—the oxide showing no amphoteric properties (Dwyer and Nyholm, 1941). The rhodic compounds possessed the general formula RhX,. 3(Et.S) (1) and were usually prepared by heating the rhodic halide with an excess of the coordinating group in aqueous alcoholic solution. Coordination took place slowly, and incompletely, and at room temperature was incomplete even after a day. The resulting compounds had a strong odour of diethyl sulphide both in the solid state and in solution, while the molecular weight determinations gave evidence of dissociation. The solubility in organic solvents was very high, and alcoholic silver nitrate precipitated the silver halide only on boiling. In hot acid alcoholic solution dissociation occurred, followed by destruction of the diethyl sulphide until ultimately the pure rhodic halide was left behind, according to the following scheme : H+ PPLOM ash cp mmm (ROL. 2E.S), (lk) + 2ht.s S SS RhCl, + 2Et,S oHt,8 + H++ HO —— ((Bt,S8))+ + EtOH + 4,8 Although pure specimens of compound II were not isolated, impure Specimens were obtained from time to time. 68 _ DWYER AND NYHOLM. Further evidence of the weakness of the bond between the rhodium and the sulphur in these compounds was found in the iodide complexes. Diethyl sulphide failed to react with pure precipitated rhodic iodide, while by the addition of potassium iodide to the chloride complex, nearly 90% of the rhodium was precipitated as rhodic iodide. Yields of the required complex were obtained by this latter reaction only in the presence of excess of diethyl sulphide. Finally, it was noted that diethyl sulphide failed to react at all with potassium rhodium thiocyanate, in which the metal is present as the stable red (Rh(CNS),)™ ion. Reduction experiments on the rhodic complexes either formed in situ as in the cases of the arsine or pyridine complexes (Dwyer and Nyholm, 1942, 43), or on pure isolated specimens, were fruitless. The usual reducing agent, hypo- phosphorus acid, in halogen acid solution failed owing to destruction of the diethyl sulphide by the acid, and in the absence of the acid by the preferential formation of a complex hypophosphite, noted previously. Sodium formate and formaldehyde both slowly reduced the complex, but much more rapidly reduced the dissociated rhodium halide to the metal. Sodium sulphite gave an insoluble basic rhodie sulphite, while hydroxylamine failed to react. Diethyl sulphide itself, which is a mild reducing agent in acid solution, yielded black precipitates which were slowly soluble in concentrated hydrochloric acid and were apparently rhodous sulphide. Finally, it was noted that ethyl alcohol in hot neutral or faintly acid solution, while it failed to reduce the complex, reduced the dissociated rhodic halide to the dark red rhodous state, and then to the metal. C| C/ C/ C/ Gh Li Main ti SCAN (Ft), S St), Gi /SEt), — (E0),S citi), Gi Ci C/ 4 vi EXPERIMENTAL. Tris-diethylsulphide trichlororhodium. Rhodium trichloride solution, 30 mls., containing 0-285 g. of rhodium was neutralised with caustic soda, and then made just acid with hydrochloric acid. Diethyl sulphide, 0:75 g. with alcohol 15 mls., and water 10 mls., was added and the solution boiled for one to two minutes. After cooling the mixture was precipitated with water, and the oily precipitate coagulated with petroleum ether. The yellow rods and plates melted at 126° C., and were easily soluble in organic solvents, even hot petroleum ether, from which it crystallised on cooling. Found: Rh=-21°5% 3; Cl==22-39%, Calewlated ” for* hei 3(C.H;)o8 : Rh=21-46% ; Cl 22-207. Tris-diethylsulphide tribromorhodium. SYDNEY: ny Sang | _- PUBLISHED BY THE SOCIETY, SCIENCE HOUSE | Arata’ || Bee Gera, GLOUCESTER 4 AND ESSEX STREETS 1945 f Arr. IX.—The Light Absorption and Mawuctiar Properties of Nickel Cont pes : H. A. McKenzie, D, P. Mellor, and the late J. E. Mills aed LL N. _ Shor oe oe 1944) e a . bs Se hoes Bio ae = Giiall Number of Observations. By R. C. LE; Bosworth. Ph; D1 F A. c. I. alienate 1944) .. a aie a es 6h is on cat ei 7 j ae tae XI. .—Review of Analyses of Some Australian Fleece Wools. By M. Lipson, B. BS, _ A.A.C.I., and Una A. F. Black, B.Sc. (Issued June 15,1945)... .. ws we as ; ee XII.—The Bauxites of New South Wales. Their Distribution, Composition ona i poe B R Probable Origin. By F. N. Hanlon, B.Sc., Dip.Ed. (Issued June 15, pace fey _ Arg. XIII.—The Determination of Calcite and ciesneee! in Invettebsate hella: By om Ne D. M. Bray. (Issued June 15, 1945) .. ; ois ‘ tie, eee eae 113 Need, XIV [ae teranes: Arsonium Salts and their Metal Co- ‘rdtaplaee daeeeuees cae Part I. Bismuth. By F. P. Dwyer, M.Sc., N. A. Gibson, B. Sc,, and R. S. Be ahs ob Mi M.Se. (Issued June 15, 1945)... ah ia x a abe ae he Art. XV.—The Sternal Integument of Trichosurus vulpecula. By Adolph Bolliger : and Margaret H. Hardy. (Issued June 15, 1945) .. ; a sat peta oe Arr. XVI.—Studies in the Phenanthridine Sories, Part I, The Cyclisation ok a _2-Formamido-Diphenyls. By E. Ritchie, M.Sc. (Issued August 10, (1945) Me 1st 3 Art, XVII.—Studies in the Phenanthridine Series. Part II. The Gyclisaueae of Sore” eset 4’-Bromo and 4’-Dimethylamino Acyl-2 bese asics By E. Taleha M. Se. E¢a9 (Issued August 10, 1945)... Say aes . a Sei ee i aaa heen et | Aer. XVIII.—Studies in the Phivaninmatis ec Part III. The Tetuaten of the AL | Group in the Morgan-Walls Reaction and the Mechanism of the sara By ee aa _. Ritchie, M.Sc. (Issued August 10, 1945) .. — .- ae eee te Arr. XIX.—Studies in the Phenanthridine Series. Part IV. ae : 10 Dimethyl Aree Phenanthridines. By E. Ritchie, M. Se. (Issued ene 10, 1945) 52 hs o ae ‘Art. XX.—Studies in the Phadcnphedion Sie Part v. Phenanthyidine-@- Aldehyde : ei aay Related Compounds. By E. Ritchie, M.Sc. (Issued oe 10, 1945). oe ye an a a pore are. XXI.—Studies in the Phenanthridine . Series. Part VI. A oe of. B-Mothyl Phenanthridine, By i. Ritchie, M.Se. (tasted. August 10, hace 7 parse hi imp Arr. XXII.—Studies in the Phenanthridine Seri: Part VII. A Stes of Benz0(0)- ; | phenanthridine. By E. Ritchie, M.Sc. a ganigh August 10, ieee v | Ann. XXIII,—Studics in the’ Phonanthridige- Genes. ‘Part “VI[L 9: cea a aa Phenanthridine and Related Substances. By E. Ritchie, Bee Se. oan. Aan 10, Pies Wht Arr. XXIV.—The Physis of Rubbing Surfaces. By ae a Ra ANT 7, 1945) . Neen eo a ea eS < et ; 'y ¢ a G ete A SUNTAY re ES ay db ote bb m MA bya A Pea eae ; tae “ MUSED Bee re Cetera a ! af Ne by Ch x ti 4s Journal and Proceedings of the Royal Society of New South Wales VOLUME EXXYVI!I PART Il THE LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. By H. A. McKenziz, D. P. MrELtor, J. E. MILLS AND L. N. SHORT. Manuscript received, July 12, 1944. Read, August 2, 1944, INTRODUCTION. It has long been known that the colour of internal complexes of nickel sueh as bis-dimethylglyoxime nickel, is strikingly different from that of simple nickel salts and for a time there appeared to be a simple correlation between the colour of nickel compounds and their magnetic behaviour. Many para- magnetic nickel complexes are some shade of green or blue, while diamagnetic complexes have, as a rule, colours ranging from bright red through reddish brown to various shades of yellow. In the course of a survey of the magnetic properties of nickel compounds carried out in this laboratory it has become apparent that little reliance can be placed on the above correlation (Mellor and Craig, 1940). It has been found, for example, that vermilion bis-1-hydroxy- acridine nickel and deep wine red bis-thiolquinoline nickel are both paramagnetic, with moments of 3:2 Bohr magnetons in each case ; that green formylcamphor- ethylenediamine nickel and green bis-salicylaldoxime nickel are diamagnetic. It is clear that colour, concerned as it is with absorption in the visible spectrum, is too superficial a basis for studying the correlation and that the absorption spectrum extending over the visible and ultra-violet region offers a more satisfactory basis for making comparisons. Furthermore, it is obvious that the light absorption of the organic addendum (or chelate group) in the compounds studied, must play a part in determining colour and that if we are to discover what influence the nature of the binding and the magnetic condition of the nickel atom has on its absorption of light, proper allowance must be made for absorption by the remainder of the molecule to which the nickel atom is bound. In order to make such allowance, the procedure adopted in the present work has been to compare the absorption spectrum of the nickel complex with that of the chelate molecule (or molecules) from which it was derived, both compounds being studied in the same solvent wherever possible. In the main, the substances which have been used as chelate groups, produce, under the dispersion used, relatively simple absorption spectra which enable the effect of introducing the nickel atom into the molecule (or molecules) to be more readily determined. It has been found convenient to classify the nickel compounds themselves according to the nature of the four atoms bound to the metal. Five classes have been considered : N Oi YA. 0 MO oN Ve aes (1) Nl. (22) NA (00) NY) Oo x0 aan ye Wes LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. 71 An example from each class is shown in I-V. H ona ay N 0 ae. es eH paw ae. a9 ty es Onn: N [ | CH CHs He 1 1l. 5 ele lV aa ae OCoH eo wee eee 2°5 Ve EXPERIMENTAL. The absorption curves were obtained by using a Hilger quartz spectrograph in conjunction with a rotating sector photometer and a condensed tungsten steel spark as light source. Spectra were photographed mainly on Kodak Orthochromatic plates. After the sector had been set at a predetermined value, match points were determined for two adjacent spectra, the intensity of one of which was controlled by the sector, the other by absorption in the solution under examina- tion. Wave lengths of the match points were computed by means of the Hartmann interpolation formula. Solvents: Whenever the compounds were sufficiently soluble, absolute alcohol was used, but occasionally it was found necessary to use chloroform or dioxane. The chloroform used was of B.P. quality and the dioxane was purified by the method of Weissberger and Prosskauer. Compounds : These were principally specimens prepared and used for investigations already published (Mellor and Craig, 1940; Mellor, 1941). Any compounds made specially for the work were analysed for the metal content as a check on their nurity. Magnetic Measurements: Data for most of the compounds studied have been already published. Measurements for two compounds not previously given are set out below : Bis-1-hydroxyacridine nickel : “x=8-27x10-§; Yu=3,700; ba—3,924 (294°); y—3-04 B.M. Bis-thiolquinoline nickel : X=11-57; bu=4,380; ba=—4,570 (293°); w=3-17 B.M. Except where indicated, magnetic measurements were made exclusively on solids. H—August 2, 1944. t2 MCKENZIE, MELLOR, MILLS AND SHORT. EXPERIMENTAL RESULTS. The results of an examination of a limited number of compounds of Classes I and II have already been briefly reported (Mills and Mellor, 1941) and the extension of the survey to the remaining classes has served to confirm and to enable elaboration of the earlier findings. The principal features which emerge from the full survey are these: in paramagnetic complexes the nickel atom, as a rule! makes very little contribution to the absorption spectrum; in other words, there is little difference between the absorption spectrum of the metal complex and that of the chelate molecule from which it is derived. The curves for acetyl acetone and bis-acetyl acetone nickel (Fig. 1) illustrate this point. (See also Fig. 2.) On the other hand, with diamagnetic complexes it has been found that, in general, the presence of the nickel atom can be associated with the appearance of a broad band of considerable intensity, the molecular extinction coefficient at the maximum of which ranges from 3,000 to 12,000 or more. This band, referred to in the sequel as the nickel band, is to be clearly seen in Figs. 3-7, which show curves for some of the compounds examined. Data concerning the position of the maximum of the nickel band (Amax) and the molecular extinction coefficient at the maximum (emax) for the diamagnetic nickel complexes are summarised in Table I. DISCUSSION. Diamagnetic Complexes. It will be seen from Table I that the position and intensity of the nickel band maximum varies from compound to compound, a point which is perhaps more clearly brought out in Fig. 8, showing the band as it appears in a number of different compounds. There is little evidence that the electronegativity? of the four atoms attached to nickel exercises any marked influence on the position? of the band maximum (see curves II, ITI, IV and V, Fig. 8) since the variation in position for compounds within a class (Class II, curves I and II, Fig. 8) is greater than the variations between the classes ; nor does it seem that the electronegativity of these four atoms produces any regular effect on the intensities of the bands as judged, either by the extinction coefficient at the band maximum or more appropriately, by the band areas. Since the nickel band becomes vanishingly weak in ionic complexes where electronegativity differences between nickel and the atoms attached to it are greatest, it was at first believed that a study of the effect of electronegativity differences might be a promising line of investigation. The electronegativity differences between Shot His the attached atoms and nickel are least in Class V Ni but it will be seen Se Ss from Fig. 8 that, for the one Class V compound studied, the intensity, instead of being greatest, is the least of all. In varying the kind of atom bound to nickel, large changes are simultaneously brought about in the remainder of the molecule and there seems little hope of disentangling the effects produced by these variations on the absorption of light, from absorption effects produced by concomitant changes in the remainder of the molecule. The introduction of a nickel atom into a chelate molecule causes the bands of the chelate molecule (or molecules) to undergo some change, usually a bathochromic shift. The curves for salicylaldehyde-propylene-diamine and its nickel derivative (Fig. 5) may be taken as an illustration of what happens to the chelate bands. The 1'The one exception to this rule will be referred to in the discussion. 2 The electronegativities of O, N and S as defined by Pauling are 3-5, 3-0 and 2-5 respectively. The bands are sufficiently symmetrical to enable the use of the position of the band maximum to describe the position of the band itself. LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. 73 TABLE I. The Position and Intensity of the Nickel Band.* Compound. Leet Formula. | Solvent. | Emax X 1075. | Amax- ww a eee fe: | te Class II. ; | | Bis-o-aminophenol-nickel .. | (0.C,H,NH,).Ni | Absolute alcohol. 2, 432 Bis-iminopaonol-nickel .. —... | (O-CH,.CH».0.CH:NH),Ni | Absolute alcohol eyed) sae and dioxane. Bis R salicylaldehyde - propylene | (C,,H,,0.N,)Ni Absolute alcohol. 6 405 diamine nickel. Bis - salicylaldehyde - ethylene- | (C,¢Hi,N.0,)Ni _ Absolute alcohol. | 6 | 405 diamine nickel. | cr a eS ee (Ee Bis-salicylaldimine-nickel .. | (O-C,H,.CH:NH).Ni | Absolute alcohol. | 4 405 Bis -N : methyl - salicylaldimine | (O.C,H,.CH.N.CH;).Ni | Absolute alcohol. | a | 400 nickel. | | Bis - o - hydroxyacetophenone | (-O—-C,H,.C.CH;:NH).Ni Dioxane. | 5 411 imine nickel. | | Bis - nicotinylacetone - ethylene- | (C.)>)H.»O02N,)NiH,O | Absolute alcohol. | 7 | 405 diamine nickel. | | | Bis - acetylacetone - ethylene- | (C,.H,,0,N.)Ni | Absolute alcohol. | 6 | 365 diamine nickel. | | | Bis - benzene - azo B naphthol | (-—O.C,,H,.N:N.C,H;).Ni | Chloroform | ipl ae 1~400 nickel. | | (by analysis) | Bis-salicylaldoxime nickel .. | (-O-C,H,.CH:NOH),Ni Absolute alcohol. | 3 | 385 | (Very weak) | 620 Bis - formyleamphorethylene - | (CoH,40.N.)Ni:3H20 "Absolute alcohol. | 385 diamine nickel. | pt OF O12 T1620 Class III. | Bis-c-benzildioxime nickel .. | (C.H,;.C:NO.C.NOH.C,H;).Ni | Absolute alcohol. 12:5 | 405 Bis - 44’ - dimethyldiazo - amino - | (CH,.C,H,.NH.N:N.C,H,.CH;).Ni | Dioxane. Ni | 400 benzene nickel. | 1:8 naphthalene diamine nickel | [Ni(C,,H,.NH..NH.).|SO, | Dioxane. | 1-8 390 sulphate. | | Class IV. | | Bis-thiosemicarbazide nickel .. | (NH>.CS.N.NH,).Ni | Propylene glycol. | 4 _ 405 7 "(3G cee | ekg eet ae in De aM ane a pier ' Class V. | Bis-xanthic acid nickel .. | (8.C8.0C.H;).Ni | Absolute alcohol. | 3 | 415 * The concentration of the solutions used varied from around 0-001 M to 0:0001 M. Some of the absorption curves (Figs. 1-12) have been constructed from data for solutions at two or even three different concentrations. + This second weaker band has been found only in green diamagnetic complexes. i t Owing to the empirical nature of the process used to analyse these curves these results are approximate only. two strong bands shown thus —.-—.-— in Fig. 5 are presumably the x and y bands (Lewis and Calvin, 1940); both are shifted to the red in the nickel derivative. Although a bathochromic shift is what generally happens, two examples of the opposite (hypsochromic) shift have been observed with nicotinylacetone-ethylene- diamine and acetylacetone-ethylenediamine nickel. As a rule, the nickel band stands out clearly, but there are occasions when it is partly or wholly obscured by strong absorption of the chelate molecule itself in the same region of the spectrum. This is to be seen in the complexes of benzene-azo-8-naphthol and diazoaminobenzene. Let us first consider the case of benzene-azo-8-naphthol whose absorption in chloroform solution is shown in Fig.9. This curve exhibits two maxima at 485 and 311 my respectively. 74 oO A o ~o N .£N319¢33309 NOILd4OSAV YYINIITOW Jo WHLIYYION MCKENZIE, MELLOR, MILLS AND SHORT. Oo 2 Ss N oe —— oO oO faa) e Ti) Bo So = . 2 Te ae ra a= G2h Ey =} = = : IGN se . 35 bb S d el S2t w > a «i =~ oO ive) bh OQ Oo wo wo = : ¢-O8 ¥ 1N319144309 Notidyosa¥Y uv INIIION e-Ol X IN319145309 NO!idyNOSaY Yv¥INDIION (SS) wo N Oo = 350 IN MILLIMICRONS ° re ze or . oo on ara — tas = Fy Ww > = = Oo wo hs o o io i ne? =) oy o” ow = e-Ol X LN919145309 NOtLdyosa¥Y Yv1NI3Z10N 400 350 300 250 WAVELENGTH IN MILLIMICRONS 450 500 400 350 300 250 WAVELENGTH IN MILLIMICRONS 450 500 Fig. 4. 20 oO LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. 250 350 IN MILLIMICRONS 400 WAVELENGTH 450 500 w ° ey o re) ei he SO % } e-Ol X 1N319144309 NOILayOSeV YvINDITON e ENGTIT13309 NOT Ld8OSEY By 1099 10N 400 350 300 250 WAVELENGTH I!N MILLIMICRONS Je 450 500 oO wn Oo wn N ¢-01 ¥ 1N319135309 NONdHOSeY ¥¥INDZION wn wm Nm 7 ¢-O1 X¥ 1N319153309 NOI1dYOSEV YVINDII0ON 75 400 350 300 250 WAVELENGTH IN MELLIMICRONS Fig: 8: 450 500 350 300 250 IN MILLIMICRONS WAVELENGTH 509 76 o eo p-01 ¥ 4N319144309 NOILdNOSAaY YvIND370N MCKENZIE, MELLOR, MILLS AND SHORT. _~>—S o £2 )~— oe ? x a ae —~, =. ta! oO Si > x wn = d = { B= \ Loe] 3 \ = ine =) Eb oOo Oo? om eg om oa = iu ee oa Ze —_—— = oOo wo vs ee ae o ae 3 Ve Se he set ° ? Ss = = e S e-Ob X 1N3191443900 NOILd¥OSaY Y¥1NI3710W g-08 KX IN319144109 NONLayOSeV YvINIIION Oo ve) N oOo = 350 IN MILLIMICRONS : (=r) > = oO fF e oo 22 oo Ww — fe w > a = Oo wo wv =! oO wo o. mo o wo N ~ ¢-94 K 1N319144909 NOILGYOSaY YvINDITIOW 350 308 250 IN MILLIMICRONS Fig. 12. 400 WAVELENGTH 450 500 300 250 350 IN MILLIMICRONS Fig. 11. x oF oo +2 us 4 w > x = °o w +t Q o o hoe: LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. 77 The chloroform solution of the corresponding nickel complex (Fig. 9) shows maxima at 486 and 307 my; a significant feature of this last absorption curve is the inflexion between 400 and 450 mu. If the 486 my band of the nickel complex is continued down parallel to the corresponding band of the chelate (this is identified on the assumption of a hypochromic shift) and the ordinates of this curve are subtracted from those of that part of the curve of the nickel complex lying between 350 my, a curve is obtained which suggests that the inflection is due to the super-position of two bands in this region. There is little doubt that the curve shown thus —---—--- represents the nickel band. On attempting to apply similar treatment to the curves of diazoaminobenzene, 44’-dimethyl-diazoaminobenzene and the corresponding nickel complexes, difficulties are met with owing to the complexity of the effects produced in the bands of the chelates themselves which, as a result of the formation of the metal complexes, are considerably changed in contour as well as in position. The curves for diazoaminobenzene and its nickel derivative are shown in Fig. 10. Paramagnetic Complexes. The following paramagnetic complexes were examined: Class I (a) bis- acetylacetone nickel (Fig. 1), (b) bis-salicylaldehyde nickel, (¢) bis-formylcamphor nickel; Class II (d) bis-1-hydroxyacridine nickel, (e) bis-8-hydroxyquinoline EXPLANATION OF FIGURES. Fig. 1.—Acetylacetone (2c) in 0-01 alcoholic NaOH: —-— Bis-acetylacetone nickel in absolute alcohol : —— Fig. 2.—a-Phenylenediamine (2¢) in chloroform: —-— Bis-phenylenediamine nickel sulphate in water: ——— Fig. 3.—Thiosemicarbazide (2c) in water: —--— Bis-thiosemicarbazide nickel in propylene glycol: ——— Fig. 4.—Potassium ethyl xanthate (2c) in absolute alcohol: ——— Bis-xanthic acid nickel in absolute alcohol : Fig. 5.—Bis-salicylaldehyde-propylenediamine in absolute alcohol: —-— Bis-salicylaldehyde propylenediamine nickel in absolute alcohol : ——— Fig. 6.—a-Benzildioxime (2c) in chloroform: —-— Bis-a«-benzildioxime nickel in absolute alcohol : ——— Fig. 7.—Bis-salicylaldehyde-ethylenediamine in absolute alcohol: ——— Bis-salicylaldehyde-ethylenediamine nickel in absolute alcohol : ——— Fig. 8.—The nickel band in different complexes. I. Bis-acetylacetone-ethylenediamine nickel (alcohol) : ——— II. Bis-salicylaldehyde-propylenediamine nickel Scare —--- III. Bis-«-benzildioxime nickel (chloroform): —---—---— IV. Bis-thiosemicarbazide nickel (propylene aoa oe V. Bis-xanthic acid nickel (alcohol): --:-:: Fig. 9.—Benzene-azo-8-naphthol (2e) in absolute alcohol : — — Benzene-azo-8-naphthol (2c) in absolute chloroform : —-— Bis-benzene-azo-§-naphthol nickel in chloroform : ——— The nickel band (see text): —---—--:-— Fig. 10.—Diazoaminobenzene (4c) in dioxane : —--—-— Tetrakis-diazoaminobenzene nickel in dioxane: ——— Fig. 11.—Thioquinoline (2¢) in chloroform : Pee aol) Bis-thiolquinoline nickel in chloroform : ——— The nickel band: ---:--- Fig. 12.—33’ : 55’ Tetramethyl 44’ dicarbethoxy-dipyrromethene (2c) in alcohol: —--—--— Bis 33’ : 55’ tetramethyl 44’ dicarbethoxy-dipyrromethene nickel in alcohol : —— 78 MCKENZIE, MELLOR, MILLS AND SHORT. nickel,* (f) salazine nickel, (g) bis-glycine-nickel, (h) bis-orthophenylenediamine nickel sulphate (Fig. 2), and (7) 33’:55’ tetramethyl 4:4’ dicarbethoxy- dipyrromethene nickel.* The absorption curves of these complexes appear to be essentially those of the chelate molecules with the bands displaced to longer wave-lengths and broadened. An interesting feature of the curves is that, in the two instances examined, (a) and (e), the absorption bands of the nickel complexes in alcohol solution bear a strong resemblance to the bands of the chelate molecule in alcoholic sodium hydroxide. (See Fig. 1.) Although no strong nickel band was observed in any of the above complexes, there was some evidence of the presence, in the more soluble complexes, of a very weak band in the 400 my region. Such a band (2e¢max~10) was found at 375 my with nickel glycine whose solubility in water is sufficiently great to permit the investigation of the band. Similar very weak bands have been observed with aqueous solutions of simple nickel salts which are also paramagnetic. Datta and Deb (1935), for example, have reported that nickel chloride shows a weak band (emax~10) with a maximum at 407 my. In ethyl alcohol, the band maximum is shifted to 420 my and in concentrated hydrochloric acid to 430 my. More recently, Tréhin (1943) has studied the absorption of nickel sulphate, both in the solid state and in solution ; for the former he finds a maximum at 389 my and for the latter a maximum at 395 my. Extinction coefficients are not stated in the abstract of this paper which is the only form in which it is available at present. Bjerrum (1941) reports a very weak band (¢max~1) at 680 my for the hexaquo nickel ion. The fact that these maxima are located in the same regions of the spectrum as the stronger nickel bands of the diamagnetic complexes suggests that the bands, both strong and weak, have similar origins. In other words, if this assumption is correct, the effect of lowering the magnetic moment of the nickel atom is to enhance enormously its absorption of light in the 400 my region of the spectrum. As to why reduction in moment should produce this effect, no explanation can, as yet, be advanced. There are, however, certain departures from the general rule which suggest the course of further investigation. ANOMALIES. Paramagnetic Compounds. Perhaps the most puzzling anomaly is offered by bis-thiolquinoline nickel, whose moment (3:17 Bohr magnetons) has been checked on several occasions. ® Reexamination of the absorption spectrum which is shown in Fig. 11 also failed to remove the anomaly. Analysis of the absorption curve obtained with a chloroform solution, along the lines used with benzene azo-6-naphthol nickel, indicates the presence of a band with a maximum at about 390 mu. Diamagnetic Compounds. Although the enhanced nickel band is produced by bis-thiosemicarbazide nickel, it is absent from the absorption of aqueous solutions of bis-thiosemi- 4The absorption spectrum of 8-hydroxyquinoline in alcoholic sodium hydroxide has been described by Fox (1910). 5 This compound is of special interest since it is considered to owe its paramagnetism to steric effects which force normally co-planar Ni-N bonds to adopt a tetrahedral configuration. The absorption spectra of the metal complex and the chelate from which it is derived are shown in Fig. 12. Although there is an increase in the absorption in the 400 my region, there seems to be no new maximum which can be attributed to the nickel atom. 6 We are indebted to Mr. J. B. Willis for checking the measurements on this compound. Jensen (1936) states that the compound is diamagnetic, but we are unable to confirm this. LIGHT ABSORPTION AND MAGNETIC PROPERTIES OF NICKEL COMPLEXES. 79 carbazide nickel nitrate (Fig. 3) which, in the solid state, is diamagnetic.’ There are reasons for believing (Jensen, Rancke-Madsen, 1934) that in aqueous solution there is an equilibrium between square and octahedral complexes formed by the coordination of two water molecules or a third semicarbazide molecule : 3[Ni(thio),]++ + 6H,O = 3[Ni(thio),(H,O),]** = 2[Ni(thio),]}++ + [Ni(H,0),]++ It was found that Beer’s Law does not hold for aqueous solutions of bis-thio- semicarbazide nickel nitrate, which fact is consistent with Jensen’s interpretation of the behaviour of these solutions. The absence of the nickel band from the absorption of these solutions must mean that the equilibrium is very largely in favour of the hexaquo and tris-thiosemicarbazide nickel ions. A consideration of the anomalous behaviour of solutions of bis-thiosemicarbazide nickel nitrate naturally raises the question as to whether the magnetic condition of the metal complexes should not have been studied in solution as well as in the solid state. Owing to the low solubility of many of the compounds and also to the fact that no microbalance was available for the magnetic measurements, attention was confined, in the present work, to solids. It is clear, however, that susceptibility measurements should be extended to solutions with aid of a Curie-Chéneveau or a more sensitive Gouy balance. Even with the more sensitive equipment, the low solubility of many internal nickel complexes may considerably restrict the field of investigation. The need for further work on solutions is obvious from the work of French, Magee and Sheffield (1942), who report that in methyl alcohol, at a concentration not stated, ethylenediamine-formylecamphor nickel is weakly paramagnetic (1-9 Bohr magnetons). In the solid state, this compound is diamagnetic. Mr. W. A. Rawlinson, of the Walter and Eliza Hall Institute of Research, Melbourne, has confirmed the observation of French et al (1942), although the observed moment was somewhat lower at the particular concentration he studied. He has also extended measurements to chloroform solutions and has found similar behaviour. On the assumption that the moment of Ni++ must be either zero or in the neighbourhood of 2-8 Bohr magnetons, there is, in these solutions, an equilibrium between dia- and paramagnetic molecules with the equilibrium in favour of the former.’ Under these circumstances it may be questioned whether Beer’s Law is applicable to solutions like those of ethylenediamine-formy]- camphor nickel. Tests for the validity of the law for solutions of a few dia- magnetic nickel complexes have been made over a rather limited range of concentrations and the law has been found to hold approximately (within the limits of experimental error of ~2%). Plans are in hand for testing the applic- ability of Beer’s Law to ethylenediamineformyleamphor nickel and similar substances over aS wide a range as solubility restrictions permit; it is also hoped to discover whether there is any variation of paramagnetic susceptibility over the same concentration ranges. SUMMARY. It has been found that with paramagnetic nickel complexes the metal atom, as a rule, makes very little contribution to the absorption spectrum. On the other hand, nickel complexes which, in the solid state are diamagnetic, form solutions whose absorption spectra are characterised by the appearance of a ? Moniminobiuret nickel, which is diamagnetic in the solid state, also fails to show any evidence of the nickel band when dissolved either in alcohol or propylene-glycol. No explanation can, as yet, be given for the anomalous behaviour of this substance. 8 Variations in the position of this equilibrium from compound to compound may have some influence on the intensity (area) of the nickel band. On the other hand, the above assumption may not be correct—all the nickel atoms may have moments of 1:9 Bohr magnetons. 30 MCKENZIE, MELLOR, MILLS AND SHORT. broad band of considerable intensity. This band can be associated with the presence of the nickel atom in the complex. Several deviations from these rules have been noted, but in only one instance has the exceptional behaviour been explained. ACKNOWLEDGMENTS. We are indebted to Mr. W. A. Rawlinson for magnetic measurements on solutions of nickel-ethylenediamine-formylcamphor ; to Mr. F. P. Dwyer for specimens of nickel triazene compounds and finally to Dr. R. Lemberg for making available facilities for visual work, with the Hilger-Nutting spectro- photometer. REFERENCES. Bjerrum, J., 1941. Metal Ammine Formation in Aqueous Solution. Haase, Copenhagen, 195. Datta, S., and Deb, M., 1935. Phil. Mag., 20 (Ser. 7), 1121. Fox, H., 1910. J. Chem. Soc., 97, 1119. French, H. 8., Magee, M. Z., and Sheffield, E., 1942. J. Amer. Chem. Soc., 64, 1924. Jensen, K., and Rancke-Madsen, 1934. Z. anorg. Chem., 219, 243; 221, 11. Jensen, K., 1936. Z. anorg. Chem., 229, 252. Lewis, G. N., and Calvin, M., 1939. Chem. Rev., 25, 273. Mellor, D. P., and Craig, D. P., 1940. Tuis Joupnan, 74, 475. Mellor, D. P., 1941. THis JouRNAL, 75, 157. Mellor, D. P., and Lockwood, W. H. THis JourNAL, 74, 141. Mills, J. E., and Mellor, D. P., 1942. J. Amer. Chem. Soc., 64, 181. Tréhin, R., 1943. C.R. Acad. Sc. Paris, 216, 558. Physical Chemistry Laboratories, Department of Chemistry, University of Sydney. ‘ BESSEL’S FORMULA IN RELATION TO THE CALCULATION OF THE PROBABLE ERROR FROM A SMALL NUMBER OF OBSERVATIONS. By R. C. L. BOSWORTH, Ph.D., F.A.C.I. Manuscript received, July 4, 1944. Read (in title), August 2, 1944. The probable error (s) from a group of observations 4, » in number with a mean at «# calculated from Bessel’s formula, viz. a4 =@—ay s=0°6745 Ss ee ae oes (1) is known to be an underestimate. Jeffreys (1932), basing his theory on the lemma that the prior probability for the precision factor lying in the range h to h+dh is proportional to 1 jah has shown that the probable error can be represented by an expression of the type BE ae eee) DE aed Ly | ioe (2) where the coefficient U varies with n. Some values of U given by Jeffreys are reproduced in Table [. TABLE I. Coefficients given by Jeffreys. n U -000 -816 - 766 -740 -728 -718 -703 -674 a SAS SP & bo eooqo°o°oooc-= If now we assume that equation (2) can approximately be replaced by 2 (% —a) n—1l—a which is Bessel’s equation with a small correcting factor a, we must have approxi- mate equality between s=0-6745 (0-674)? Vint CA HE 0-455 | a or Sie pity) whee U? n—l1 82 R. C. L. BOSWORTH. as 1 if Figure 1 shows Tr plotted against Pee ae The figure shows that with the exception of the point referring to the sample of two observations only, the points fall very satisfactorily on the straight line 0-455 He 0-63 U2 n—1 Consequently we have as an improved form of Bessel’s equation —1-63 The error incurred in using this approximation as compared with Bessel’s equation is given in Table II for a number of different sample sizes. BESSEL’S FORMULA. 83 TABLE II. A Comparison of the Errors involved in Equation (4) with Bessel’s Equation. Number of Error in Error in Observations. | Equation (4). Bessel’s Equation. 2 | ale-S7, 48-0% 3 | Dai boy 21:0% 4 0-:9% 14-:0% 6 0-8% | 8-0% 10 | 0-5% 4-0% oc | 0:0% 0:0% It will be noted that the trifling modification involved in the change from equation (1) to equation (4) results in a great increase in the accuracy of the expression when applied to small numbers of observations in all cases except that of two observations only, in which case the significance of estimated error is very low anyhow. Applied to large numbers of observations equation (4) is not significantly different from equation (1). For these reasons it is suggested that equation (4) is a more suitable expression for checking the accuracy of experi- mental results than the original Bessel’s equation (equation (1)). REFERENCE. Jeffreys, H., 1932. Proc. Roy. Soc. (London), A, 138, 48-55. REVIEW OF ANALYSES OF SOME AUSTRALIAN FLEECE WOOLS. By M. LIPSON, B.Sc., A.A.C.L, and UNA A. F. BLACK, B.Sc. Manuscript received, July 22, 1944. Read, September 6, 1944. INTRODUCTION. The presence in the fleece of constituents other than wool is realised by scientific workers and wool valuers. Methods for the quantitative division of the fleece into six major constituents, namely wool wax? (lanolin), suint, dirt, moisture, vegetable material and wool fibre, have been previously described. In the commercial evaluation of any greasy wool sample, one essential is a knowledge of its content of pure wool fibre, which naturally depends upon the amounts of the other five constituents present. The work outlined in Section A of the paper was undertaken to gain some idea of the variations that can occur in the fleece as it comes to the market and to determine the effect each may have on yield. The results cover about 200 analyses of different types of wools appraised during 1941 to 1944. Most of the samples were from the N.S.W. clip, but some were from South Australia, Western Australia, Queensland and Victoria. In biological studies concerned with the secretion of wax and suint by the sheep, it is desirable to express the results for each constituent as percentages on the oven-dry clean wool weight. This procedure presents a clearer picture of the relative amounts of wax, suint and wool produced by the sheep than is shown by percentages based upon fleece weights which would be affected by extraneous dirt and vegetable material. In the present work, however, we are mainly concerned with the actual percentage composition of the fleece, irrespective of whether the constituents are natural or extraneous materials. The results are therefore in general expressed as percentages on the fleece weight, although ratios for wax and suint ipercentaers on oven-dry clean wool weight) are included as a matter of interest. The bulk of the N.S.W. clip is merino fleece wool, and this constitutes the major group studied. Cross-bred fleece wools have also been examined as well as black wool, fellmongered wool and ‘‘ canary stained ’’ wool from the Kimberley district of Western Australia. In addition to the investigations outlined above, experiments have been carried out to find the basis of certain terms used by wool valuers in their descrip- tion of the greasy fleece. A discussion of this work appears in Section B of this paper and includes results of observations on greasy wools described by the practical man as “oily ”’, “‘ dry ’’, °° gritty ”’, “ earthy 7’, ete. SECTION A. QUANTITATIVE VARIATIONS OF FLEECE CONSTITUENTS. 1. Experimental Procedure. The samples for analysis usually weighed about 15-25 gm., and were obtained by carefully sub-sampling larger samples of 6-8 Ib., which had been drawn from the wools as they appeared 1The term “fat” is often applied to the ether-soluble fraction of the fleece, but this is incorrect, as there are no glycerol esters present. REVIEW OF ANALYSES OF SOME AUSTRALIAN FLEECE WOOLS. 85 for appraisement in the wool stores.” By comparing the oven-dry yields of pure wool in these small samples with those obtained from larger samples scoured in the testing house it was shown that with careful drawing they were representative of the lots from which they were selected. The method of fleece analysis was essentially that described by Freney (1940). Wax was removed by four hours’ extraction with redistilled petroleum ether, and suint by extraction with distilled water at room temperature. The dirt remaining in the wool after wax and suint extraction was estimated by weighing the oven-dry wool before and after a mild soap and soda ash scour. The loss in weight added to the weight of the dirt which was collected by filtration of the suint extract gave the total dirt content. Moisture content of the fleece was found by drawing another small sample at the same time as the first and finding the loss in weight on drying at 100-105° C. for four hours. The results of this work have been set out in Tables I to VIII. To facilitate the interpretation of these results, Table IX contains the range and average value for each constituent in the different groups of fleeces. TABLE |. Fleece Constituents of Merino Types 70’s and Up. Wax. Suint. ; fuer Oven- Sample (see | Dirt oisture Dry | | No. District. | | Per Per Wool | Quality. Length. Per Per Cent. Cent. Per . | | Cent. | Ratio.) Cent. | Ratio. Cent | | | 2706 Armidale. . 20-4 | 35:2 39915) 6-0 10°5 | 58-0 | 90’s and up | 2715 Armidale. . 19-7 | 33-3 4:9 8-3 6-9 59-1 90’s and up 6073 Yass 22:4 | 39-7 3:3 5:9 9-9 8°3 56:5 | 90’s and up Good. 6074 Yass 15:2 | 25-1 2-0 3°3 11-6 8:1 | 60:6 | 70’sandup | (Minimum 6423 Yass a ALPPOG SY PAPE) Ze BOG) 8-0 9-0 63-5 | 70’s and up 22”) 1539 Barraba 19:9 | 39:0 9-4 | 18-4 8-5 51-0 80’s and up 1288 Wanaaring 17-5 | 42-5 8:1 | 19:7 25-6 41-1 70’s and up 124 13-8 \040-3° 19 3-8) \\ 8-1) one 46-7 | 70’s and up | 125 FAlisore he eo 4-9 | 10-9 Dy my 44°8 70’s and up | 5821 Longreach, Q. .. | 14:2 | 28-2 je 2 a ASS ite} 9:4 | 50:3 70’s and up 5890 Julia Creek, Q. 13-2 | 26-0 7-6 | 15-0 19-5 9-1 50°8 70’s and up |_ Irregular. 3320 Wanaaring 16:0 | 44:0 (AD PAS? 34-8 | oO ae 70’s and up 1360 Brewarrina 14:2 | 35-8 ow |) altssedt 29-0 38°6 74’s and up | 1361 Brewarrina | 14-8 | 40:3 5-5 | 15-0 35-2 | 36-7 74’s and up | 1341 Bourke 14:0 | 47:6 CON Z3e8 43-8 | 29-4 70’s and up | 6172 Inverell .. 17-5 | 29-6 4-2 (or 8-0 10-74 59+ 1 70’s and up | 1239 Bourke 17-4 | 42-0 4-65\ elie d 28-1 | 41-5 74’s and up : 1401 | 18:8 | 36-0 5-5 | 10-5 12:8 Wi D2 2 70’s and up Short. 107 14°8 | 28:4 3°8 oe 19-2 1 By4 ou? 70’s and up 2645 Braidwood 25:4 | 59-8 Aro) se wilt 21-4 | 42-5 70’s and up | i i TABLE II. Fleece Constituents of Merino Type 64/70’s. Wax Suint. | Ware Sample wr Pe Dirt | Moisture | OV@n-Dry No. District. | Per | Per Per Length. Per Per Cent Cent. | Gent Cent Ratio Cent. Ratio. | eee i | ec ae ee | | | | 2377 Uralla .. Dio, 41:1 5-1 9-5 |} 9-4 | ye OE Ol nl 5266 Manilla 19-9 | 33:8 3:9 6-6 ta ee | le SO OLrs. 1334 Bourke 13-9 31-0 4-9 11-0 PAE ed. AA Se 652 Warialda a 17-0 32:7 6-0 11:5 17:9 ly cro2)0 Good. 6196 Stockinbingal 18:4 Byers) 4-7 9-0 13°8 1023 | 52-1 | (Minimum 3826 Bourke : 16-6 44-6 5-6 15:1 33°8 ena te2ahient om) 3429 Bourke 13-4 30°3 6:4 14-5 28-0 44-2 3319 Wanaaring 1487 27°8 5-4 11:8 28-4 45-7 1675 18:5 42-0 5:4 Pop 25-0 9-4 44-0 2'The samples represented quantities of wool which varied from one to 62 bales, averaging 10 bales. 56 Sample No. Sample . No. 3834 1130 1242 6429 3493 3109 3496 3497 3425 446 453 3822 5626 123 3120 LIPSON AND BLACK. TABLE L[I.—Continued. Fleece Constituents of Merino Tupe 64/70’s.—Continued. } | { Wat. Suint. Oven-Dry | Dirt Moisture Wool District. Per Per Per Length. Per Per Cent Cent. Cent. Cent. Ratio. Cent. Ratio. Isisford, Q. .. 14-6 27-7 a3 13-8 16-1 9-3 52-8 Dirranbandi, Q. 13-1 22-8 4:9 8:5 15-5 8-7 57-6 Tambo, Q. .. 12-8 24-2 4:3 8-1 20-8 8-7 53-0 Blackall, Q. 13-3 26-5 7-1 14-1 20:8 8-3 50-2 Morven, Q. Igo 32-0 5-8 10-6 13-9 8°3 54:6 Morven, Q. 16-7 3l+1 4-9 9-1 15-0 9-1 53-6 Wee Waa 15-3 28:7 8-8 16-5 12-7 53-3 Bourke 16-5 38-3 6:9 16-0 25-9 43:1 Narrabri 14:3 26-5 (0583 13-5 13:9 54-0 Irregular. Bundarra 15-7 25-2 4-2 6-7 8-2 62-3 Brewarrina 21-0 52-4 4-3 10-7 27-8 40-1 Bourke 14-0 40:5 3-8 11-0 38°9 34°6 Wanaaring 15-6 38-9 4-8 12-0 31:6 40-1 Bourke 13-3 28-2 6:3 13-4 24-2 47-1 Bourke 14-1 29-7 5-1 10-7 21-0 47°5 Murrurundi 12-6 19-9 6-7 10-6 11-9 63-4 Wanaaring B35 ¢/ 30:7 8-6 19-3 24-8 44-6 Wanaaring 18:5 42°8 6:3 14-6 28:1 43:2 Bourke 11-4 27-0 6-6 15:6 31:3 42-3 Bourke 15:2 31:8 6-6 13°8 22-0 47-9 Upper Murray iCiyere i 14-6 23-1 4-1 6-5 7:2 10°3 63-2 Wee Waa 20:5 45-8 7°83 17-4 17-1 44-8 Bourke 15-2 40-0 6:7 17:6 32°4 38:0 , Wanaaring 16-2 37-0 7:1 16-2 24-9 43°8 Short, ——, W.A. 12:5 28:3 1ILo| 25-1 21-8 11:8 44-2 | Wanaaring 15-0 32:6 6:3 13-7 25-6 46-1 | Wanaaring 14-0 30-4 7-6 | W625 25°3 46-1 | Wanaaring 21:0 70-0 6:4 21°3 29-5 30-0 | TABLE III. Fleece Constituents of Merino Type 64's. | Wax. Suint. Dirt Moisture | Oven-Dry District. Per Per Wool Length. Per Per Cent Cent. Per Cent. Ratio. Cent. Ratio. Cent. Brewarrina abs) 39-6 4-9 11-1 PAS Tf 44-3 . Good Wee Waa 18-2 33°8 5284) S028 12°3 53:9 (Minimum Bourke 16:3 39°8 9-4 23-0 25°6 41-0 33”) Bigga .. i 20:5 34-8 4-6 7°83 7-4 9:0 59-0 Bookaloo, S.A. 14-0 33°8 6:5 15-7 30-1 41-5 Inverell ae 15-5 25-4 B04 5-2 9-8 61-2 Sth. Hills, 8.A. 14°5 34°83 6-0 14-4 24-8 41:7 Irregular a SAR 15-2 40-0 4-0 10°5 34-4 38:0 Wanaaring 13-9 31-0 5-3 11:8 27-6 44-8 Wanaaring 15-3 34-2 Wed 17-2 24-1 44-8 Wanaaring 15:8 34-0 8:0 17°3 25-5 46:4 North East, S.A. 17-9 46-4 8-4 21-8 23:4 38-6 Barraba ee 14-0 22:1 5-5 8:7 7-6 9-7 63-4 16:1 28-8 5-0 8-9 12-6 *56-0 Short Berridale 17-4 31-4 7°8 14-1 7:5 55-5 ‘ 13-4 28-0 10-7 22-4 17-3 1PACIL 47°83 5229 * Lincoln wools. I—September 6, 1944. REVIEW OF ANALYSES OF SOME AUSTRALIAN FLEECE WOOLS. 37 TABLE IV. Fleeee Constituents of Merino Types 64/60’s and 60’s and Under. | Wax. Suint. Oven- | Sample Dirt Moisture} Dry No. District. Per Per Wool Quality. Length. Per Per Cent. Cent. Per Cent Ratio.) Cent. | Ratio. Cent 6221 Nundle 16°8 | 27-8 5-0 8-3 to 10°8 60-4 64/60’s 3500 ——, S.A. pez | Blow 3°8 8:9 29-6 43-0 64/60’s 5602 Tenterfield ee OAS eS c6 4-1 6-1 6-3 10-7 66°8 60’s Good 3498 Nth. East, 8.A. 1 13-5 | 30-1 Daal 210) 256 44-9 60/64’s (Minimum 441 Wanaaring ap Be |) Bes} 5-6) 12-3 24-4 45:4 60/64’s 34’) 3499 Nth. East, S.A. Oo) 1) Beak 4:5 | 10-5 32:6 42-7 60/58’s 5232 ——, W.A. 10-8 | 19:9 | 10°8 | 19-9 12-6 12-6 54-2 60’s and up 6200 Singleton 14-5 | 23-5 3-9 6°3 8:9 | 9-7 61:8 60/64’s 3811 , 9.A. .. | 16:0 | 36°6 af alesot0) 25:3 43°38 64/60’s 3819 iRateNith., SA... j-19°3 7) 52-5 Gyatchy lie tants} 28-9 36°8 64/60’s Irregular 6174 Inverell .. 15:2 | 25-0 5-4 8-9 (974 11:0 60°8 60/64's 3115 Inverell .. Bevel 2 aieotiay, 3°8 6-2 8:8 61:5 60’s 3494 Bookaloo, S.A... | 15°7 | 33:2 522) dit <0 20:8 47-4 60’s TABLE V. Fleece Constituents of Crossbred Types 58’s to 60/64’s., Wax. Suint. Oven- Sample EM Diath A chin pede gs | Dirt Moisture! Dry | No. District. | Per Per Wool Quality. | Style. Per eer: | Cent Cent Per Cent. | Ratio. Cent. | Ratio.' Gents =| | | = 672 Holbrook GRAZ O Mas eS Ouiy tac Desa Oss BPG: 60/64’s 1705 Bathurst PRO Se SoS “ScONl Te Sry eG 54-0 | 60/64’s | 1706 Bathurst NGS PO Sr ese al slp dae () 72g) 58°01] 60’s Super 5837 Wheeo .. 15:0 | 25-6 6:6 | 11-2 | 7-5 12°3 58-7 60/58’s 1699 Breadalbane 97 NW be. Slag’ | lets oak 63-0 58/60’s_ 5853 | Crookwell Qs i470) elon 5-8 12-6 66-2 58’s | 1564 | Blayney .. 134 | 24-4) 8.4 15-3 1113-3 55-0 60/64’s 406 Albury TPO aL Ih WoW. SOs Ged 13-4 10-7 53°5 60’s Good 4645 Narrandera ilePs Aye alke)are} 5:8 | 10:3 12-0 56°6 58's 1546 3074 || Masog) aaa s ye aby ate) | Get (e133 58's 5840 | Goulburn Polar Onna dae i bei7 else ll See akes etal sy 60-4 60's 6212 Uralla Pea Selec leo) ved 2a Same = bl il a Eteh 58’s Average 662 Warialda. . so, We A7ede WP TES ey de ai Es al io (G) 6:2 62-2 58’s 5787 West Wyalong .. | 10:6 | 21-5 Sie le GRA 223 '7 9-5 49-3 60/58’s TABLE VI. Fleece Constituents of Crossbred Types 32/36’s to 56/58’s. | | | We: Suint. | Oven- | Sample Pee ieee nals ee ee Dino Moisture | ery | | No. District. Per Per Wool | Quality. | Style. Bers! Per Cent. | Cent Per | | | Cent. | Ratio.| Cent. | Ratio.| | Cent. | Goq i Breadalbane .. | .7-9 | 12-2} 8:9 |) 13-7] 7-0 65-0 | 58/56’s 357 | West Buckley | | | | (Vic.) a edibotey i sos = G\eG) || 0) aC}. | 5-6 10-6 61-4 58/56’s 407 Albury .. 11-0 eI og) 13°7 NOK Lk os ates 56/58’s | 5838 Wheeo .. 9-9 5:3 730) 1 al@ats | Oral: i, 64:9 | 56/58’s 1625 Galong. .. DO Fer Oe CUO li Cds ei Gi6 GUE ou| 56's Super 646 West. Mortlake | | | Vic.) Sor alsioik yo bor 6-3 | 5:6 115 69-5 | 56/50’s 5801 Galong 6:8 | 10-2 8:6 | 12-9 7:2 ile 66-6 50/46’s 5854 | Crookwell he Deo alee RO Tene Or Srily 43 11°6 72,2 46’s 902* | Wangaratta (Vic.) | 11:4 | 19-0 9-0 | 15:0 | 5:4 12°8 60-0 32/3678 903* | Wangaratta (Vic.) ado ty alle @ 55 8-0 5-4 12-2 69-0 32/36’s 1563 9-2 | 15-3 | 9-4 | 15-6 9-2 60:1 | 58/56’s 1565 Blayney .. 9-3 | 14-5 Sev 13-6 | 6°5 64-0 58/56’s 4646 | Narrandera 10-69 ae Olen 9-2, 59-9 56/58’s | Good 5698 Young Qed tbr 2 8:4 |] 14-1 |} 11-8 11-4 59-8 56’s | 5836 VeetacSe lil conse O sali Oren. | 6:5 14°2 61-8 56/50’s si 5839 Wheeo .. 5 O18 | 20 dale: 9-6) \Pdibe 2) Hy a2, 63-4 50/56’s 5844 | Bungendore era WILo0) Seon elo 6 6°5 Hien 65-6 56’s 5695 Mudgee 6-9 | 10-9 | 10-1 | 16-0 Gae2 13-0 63°3 50/46’s 5786 Binalong erore® 8:7 8:9 | 13-2 6°5 22 OW OR || 46/50’s Average 4647 Narrandera PQ a 23 4: 7°4 | 14-0 13-1 53-0 56/58’s 6527 | Exeter ORS Orseian (ode 1152 7:4 12-6 66-0 | 56’s 838 LIPSON AND BLACK. TABLE VII. Fleece Constituents of Fellmongered Wools. | Wax. suint. | Sample : : RA SSRs ee Dirt Moisture | Oven-Dry Minimum No. District. : Per Per Wool Quality. Length. Hg te Pers, Cent. Cent Per | Cent. |Ratio. |Cent. | Ratio. Cent. Merino 3563 | Queensland .. | 20:0 | 39:5 0-1 0:2 19-0 7:9 50:6 64’s up 34” 3564 | Queensland .. | 16:0 | 25:8 0-1 0:2 10:6 9:5 62:1 64/70’s Die 3566 | Queensland .. | 17:3 | 28-0 0-1 0:2 9-9 9-0 61:6 64’s up 14” 3853 | N.S.W. ve | L626) | 2720 0:8 1:3 14-2 7:2 61:5 60/64’s 30 5441 | N.S.W. on Akos Wh PAsySts} 0-8 a, 9-7 8-4 64:3 60/64’s on 3220 | N.S.W. Hobe |p Alora. | By ors 0-6 0:9 6-9 8:1 68-0 60/64’s 13” *3565 | Queensland .. | 14:6 | 21-9 0-1 0:2 GAG 9:6 66:6 58/60’s yg *3222 | Queensland .. 9:9 | 14-1 0-4 0:6 10:3 70-0 58/56’s 3h’ 9289 | N.S.W. Ho |p aebo(ty. |) Bazin 1:4 2:6 21:5 9:6 53°6 64’s up 22” 9259 | Queensland .. | 12-0 | 21-2 0-6 ied 21-5 icp 56:6 64/70’s rad 9260 | Queensland .. | 17:0 | 26-4 0-9 1:4 8-2 8-6 64°5 64/70’s 2 9261 | Queensland .. | 16-0 | 25-2 0-8 1:3 10-0 8:6 63:7 64/70’s 13” 9262 | Queensland .. | 18:5 | 30:0 0:9 Iho) 10°6 9-0 61:5 64/70’s 14” 9263 | Queensland .. | 15:5 | 23-0 0-7 1-0 (hoy? 9-2 67-4 60/64’s 24” 9264 | Queensland .. | 14-4 | 21:3 0:5 0:7 8:6 9:2 67:6 60/64’s 24” 9265 | Queensland .. | 16:5 | 24-6 One 1-0 6:4 9:2 67-0 60/64’s De 9267 | Queensland .. | 16:5 | 25°5 0:7 ies] 8:6 9-4 64:7 64/70’s ie * Cross-bred type. TABLE VIII. Fleece Constituents of Some Black Wools. | Wax. Suint. ; Sample | iat oa : Dirt Moisture | Oven-Dry No. Per Cent. | Per Cent. Wool Type. Per Cent. Ratio. Per Cent. Ratio. Per Cent. 5672 19:7 36-6 6:7 12 Ae 10-2 10°5 53:8 Merino 5676 WZ 31-9 6-9 12-4 11-4 9-5 55°5 Rs 5677 16-0 35-6 8-2 18°3 Dore 8:7 44-9 a 5685 8:8 i528) 11-0 19-3 14:2 10-2 57-0 Crossbred TABLE IX. Average Values and Variations in Fleece Constituents. Wool Wax | — Suint Dirt Moisture Per Cent. Per Cent. | Per Cent. Per Cent. Per Cent. P Max. Min. | AY. | Max. | Min. yAcvi. 1 Viaas. |) Vinim): Av. | Max. | Min. Av. ‘| Max: |) Mim") Avy: x Me lrino. 66-8 29-4 48-9 25:4 | 10-0) | 16-4 14000) 2-0 6-1 | 438-8 6-3 ) 19°76 (1 2e6 Salt 9:6 | 4 Cross bred. (222, 49-3 61-0 19°3 5-3 | 10:6 | 18-4 4-4 SCZ dl 2 oe ines 8:4 | 14-2 9:5 | 12:0 F ellmon \gered |Wool. 70-0 50:6 63-0 20-0 9:9 | 15:8 1:4 0-1 0:6 | 21-5 6:4 | 11:2 9-6 ee? 8:8 | i It can be seen that the dirt content is the most variable factor affecting the yield of the fleece. It is interesting to note that the samples with the highest dirt content are those giving the lowest yield of wool in both merino and crossbred types. These are sample No. 1341, a merino fleece wool from the Bourke District of N.S.W. which contained 43-8 per cent. of red dust and gave an oven-dry wool yield of only 29-4 per cent., and No. 5787, a crossbred sample from the West Wyalong district of N.S.W. which contained 23-7 per cent. of dirt and 49-3 per cent. of oven-dry wool, which is very low for this type of fleece. REVIEW OF ANALYSES OF SOME AUSTRALIAN FLEECE WOOLS. 89 Wax after dirt is the constituent having the greatest effect upon the yield. In the cross- bred types, sample No. 5854 has the lowest wax content of 5:3 per cent. and the highest oven- dry yield of 72:2 per cent. — The suint contents show a variation of about 10 per cent. for both merino and crossbred types. The least variable of the fleece constituents is the moisture content, the variations of which are much less marked than those of the other constituents. In unusual circumstances, such as when the sheep are shorn wet, the moisture contents can be higher than these recorded. The maximum and minimum figures given in Table [X refer to representative samples of commercial lots of wool. It has been found that small portions chosen from some lots can give ‘figures for fleece constituents outside the above ranges, but these samples are not representative of the lots from which they are drawn. Thus one small sample taken from a bale contained 78 per cent. of dirt in the form of caked mud and only 11-2 per cent. of oven-dry wool, 4-3 per cent. of wax and 0:6 per cent. of suint. Another small sample selected on account of its high wax content was found to contain 40 per cent. of wax. A suint content of 15-0 per cent. was recorded for a small sample of wool from the Kimberley district of W.A. These samples, however, are not representative of any one clip and simply illustrate extreme values of the fleece constituents. In comparing the different types of fleece wools, it can be seen that the crossbreds contain in general less wax and more suint than the merinos. Furthermore, the merino types usually have higher dirt contents than the crossbred types, probably due to the greater amount of wax in the merino fleece, which facilitates the adherence of dirt. The moisture contents of the crossbred fleeces are in general higher than those of the merino fleeces, which is to be expected as the former contain greater percentages of the two fleece constituents which attract moisture, namely wool and suint. There are no outstanding differences in fleece constitution between the various merino types. Owing to their lower wax and dirt contents, the crossbred types below 58’s quality (Table VI) give significantly greater yields of oven-dry wool than those of higher quality (Table V). Wools fellmongered by soaking in water and then removed from the skins after “‘ sweating ” are characterised by extremely low suint contents, as can be seen from an examination of Table VIT. The suint contents in this table have varied from 0-1 to 1-4 per cent., showing that this constituent is removed from the fleece during the soaking process. The dirt contents range from 6:4 to 21-5 per cent., and 1t seems probable that a certain amount of the dirt is removed in suspension during the fellmongering process. This was investigated by analysing samples taken from fleeces before and after soaking in cold water as in fellmongering. In these experiments a crossbred and merino fleece sample were each immersed in cold water for 24-hours and then removed and squeezed by hand. They were then allowed to dry in the air and analysed. Samples from the original products were also analysed. Before soaking, the crossbred sample had suint and dirt contents of 10-1 per cent. and 4-4 per cent. respectively. During soaking it was found that 72 per cent. of the original suint and 62 per cent. of the original dirt was removed. The merino sample had initial suint and dirt contents of 2:7 per cent. and 14-2 per cent. respectively, and of these constituents 72 per cent. of the suint and 22 per cent. of the dirt were removed during soaking. Changes in wax contents were negligible for both crossbred and merino types. It can be seen therefore that dirt as well as suint would be removed during fellmongering. A considerable amount of data has been obtained concerning the wax and moisture contents of slipe wools. These are in addition to the results set out in Table VII. Of 143 tests for wax content only, it was found that the maximum was 22-3 per cent. and the minimum 11-7 per cent., the average being 14-5 per cent. The average value for moisture was 8-4 per cent. in 98 tests, the range being from 12-3 per cent. to 5-1 per cent. In Table VIII some results are set out showing the fleece composition of certain samples of black wool. It can be seen that there are no significant differences between the amounts of fleece constituents present in these samples and those of ordinary wool. 90 LIPSON AND BLACK. Section B. Some PROPERTIES OF THE FLEECE COMPONENTS. The work in this section was planned to find the scientific basis for views on the properties of the fleece components held by certain wool valuers, and each of the fleece constituents will be discussed separately. 1. Wax. The opinion is sometimes expressed that the wax in the fleece is attractive to moisture. This is no doubt based upon the fact that lanolin will readily mix with water to form emulsions without the addition of any emulsifying agent. The experiments of Wright (1909) and Hill (1922) also indicate that wool wax can absorb appreciable quantities of moisture from the atmos- phere. Conversely, Lennox (1938) showed that the ether-soluble fraction plays a negligible part in the ability of the fleece to absorb moisture from the atmosphere. Three experiments have been carried out to investigate the above points. Firstly a sample of pure commercial lanolin (3-4 gm.) was weighed and mixed with about one-third of its weight of distilled water to produce a homogeneous emulsion. This was spread thinly on a watch glass, exposed to the atnmiosphere and protected from dust and the loss in weight due to moisture evaporation determined by daily weighing. The second experiment was undertaken using similar conditions but with the initial addition of 0-1 gm. of powdered suint. In the third experiment a sample of wool wax previously dried and weighed was spread thinly on a watch glass and exposed to the atmosphere with the other two samples, daily weighings being carried out to determine the moisture uptake. In each experiment a small stirring rod was weighed together with the wax, which was thoroughly mixed after each weighing. The experiments were continued for one month, during which room temperature and relative humidity approxi- mated to 75° F. and 50 per cent. respectively without extreme variations. Tt was found that the samples to which water had been added lost moisture and came to equilibrium with the atmosphere after a period of 17 days. They then started to take up small amounts of moisture. The sample containing suint did not show any increase in its capacity to retain moisture when compared with the sample containing no suint. This was apparently due to the suint being embedded in the wax emulsion and not exposed to the atmosphere. In the fleece, the suint occurs outside the wax layer, and being unprotected from the atmosphere by the latter can markedly exert its hygroscopic properties. The dry wax sample gradually absorbed small amounts of moisture from the atmosphere and appeared to reach equilibrium after 14 days, when the moisture regain was 0:82 per cent. After that there was a slight loss in weight and then a gradual increase. An outline of the results is given in Table X below. TABLE X. Moisture Regains of Wool Wax. | Regain* Per Cent. Experiment. | After After After | Initial. 3 Days. 17 Days. 31 Days. (a) Lanolin emulsion ree ee ae tl 30:1 10:9 1:26 1:49 (b) Lanolin emulsion and suint te Zone 9-1 0:67 0-80 (c) Lanolin a ae ae ail 0 0:27 0-89 1-44 * The moisture content expressed on the original weight of wax. These results show that lanolin has no marked ability to attract moisture from the atmos- phere. On the contrary, added water tends to dry out on exposure and the equilibrium moisture content is of the order of 1-14 per cent. 2. Suint. The most hygroscopic constituent of the fleece is the suint, and when present in unusually high amounts it can exert a marked effect on the total moisture content. Graphs given by REVIEW OF ANALYSES OF SOME AUSTRALIAN FLEECE WOOLS. 91 Lennox (loc. cit.) show that the moisture regain of suint is of the order of 40 per cent. at 30° C. and a relative humidity of 70 per cent. In the present work, observations have been carried out on two portions of the same lot of greasy wool (No. 5229), one of which was described as “‘ yellow and oily’ and the other as “‘ pale and dry’’. Both portions were analysed as described in Section A of this paper and the following results were obtained : 6 Percentage Composition. | Sample. ie eer Ser se eb cs ie original rock texture and removal of the bulk of the impurities may lead to tke develop- ment, at a later stage, of concretionary masses similar to those from Trundle. At Lorne, near Emmaville, clay derived by the weathering of basalt has developed a vermicular texture (see Plate II). How this structure was first developed is only a matter of conjecture. J. W. Gruner (1922) has shown that in sections of olivine gabbro, peat solutions attacked the olivine preferentially and it had either been changed to an almost isotropic substance with a refractive index of 1-552 or had been eaten out entirely in a few places. It may be that the olivine phenocrysts in the basalt are preferentially leached and that later the pore spaces so formed become joined to form tubes. Many of our Tertiary basalts are amygdaloidal and the amygdules too, may be preferentially leached. Once tubular openings, however small, were formed, the relatively greater percolation along them would tend to increase their size, eventually giving rise to typical vermicular structure. Stronger leaching of the iron oxide around the tubes is well shown in material from the Sutton Forest deposit (see Plate IT). SUMMARY. 1. The New South Wales bauxite deposits can mainly be classed as ferruginous bauxites, although high-grade deposits occur, and are character- istically pisolitic, vermicular, nodular or earthy. 2. The mode of origin postulated is as follows : (a) They were formed by the weathering of Oligocene basalts (except the Trundle deposits) during the Miocene period, on areas protected from mechanical erosion, and are true residual deposits, not replacements. Journal Royal Society of N.S.W., Vol. LXXVII, 1944, Plate I See eee ROW ON ass ie Journal Royal Society of N.S.W., Vol. LX XVIII, 1944, Plate ILI — THE BAUXITES OF NEW SOUTH WALES. nia (b) The lateritisation consisted in the removal of the alkalis and alkaline earths, with almost complete de-silication, but, in most cases, practically no removal of the iron. (c) The climate was probably warm, tropical to sub-tropical, and the rainfall moderate with possibly variable incidence, such incidence, however, not necessarily being either seasonal or divided into definite wet and dry periods. REFERENCES. Andrews, E. C., 1903. Rec. Geol. Survey N.S.W., VII, Pt. 3, 140. -————. 1904. Rec. Geol. Survey N.S.W., VII, Pt. 4, 281. —-—__—_————— 1910. THis Journat, 44, 420. —-—_____—_—— 1933. Tuis JourRNat, 67, 251. Booker, F. W., 1941. Unpublished report, Mines Department, N.S.W. Booker, F. W., and Hanlon, F. N., 1944. Unpublished report, Mines Department, N.S.W. Browne, W. R., 1933. THis JoURNAL, 67, 9. Bucher, W. H., 1918. J. Geol., 26, 593. Campbell, J. M., 1909-10. Trans. Inst. Min. and Met., 19, 432. Card, G. W., 1903. Rec. Geol. Survey, N.S.W., VII, Pt. 3, 226. Carne, J. E., 1911. Min. Res. Geol. Survey, N.S.W., No. 14, 75. Cleland, H. F., 1916. ‘‘ Geology, Physical and Historical’, American Book Co., New York. Crespin, I., 1943. Bull. Mineral Res. Survey, No. 9. David, Sir T. W. E., 1887. Mem. Geol. Survey N.S.W., No. 1. Evans, J. W., 1909- 10. Discussion of paper by J. M. Campbell, 1909-10. Fermor, L. ic, 1915. Geol. Mag. Lond., N.S.W., Dec., VI, 2, 28. Fox, C. S., 1927. ** Bauxite ’’, Crosby, Lockwood and Son, London. Gruner, J. W., 1922. Hc. Geol., 17, 407. Harper, L. F., 1924. Bull. Geol. Survey N.S.W., No. 8. Holland, Sir T. H., 1903. Geol. Mag. Lond, N.S., Dec., IV, 10, 59. Holmes, A., 1914. Geol. Mag. Lond., N.S., Dec., VI, 1, 529. Jaquet, J. B., 1899. Ann. Rept. Mines Department N.S.W., 176. Lacroix, A., 1913. See L. L. Fermor, 1915. MacLaren, M., 1906. Geol. Mag. Lond., N.S., Dec., V, 3, 536. Mead, W. J., 1915. Hc. Geol., 10, 28. Mohr, E. C. J., 1930. ‘“‘ The Soils of Java and Sumatra ’’, translation by R. L. Pendleton Peiping (1933). Moore, E. S., and Maynard, J. H., 1929. Hc. Geol., 24, 272. Mulholland, C. St. J., 1941. Unpublished report, Mines Dept. N.S.W. Raggatt, H. G., 1939. Unpublished report, Mines Dept. N.S.W. Rogers, A. F., and Kerr, P. F., 1942. ‘* Optical Mineralogy ”’, McGraw-Hill Book Co., New York. Schade, H., 1909-10. See W. H. Bucher, 1918. Simpson, EH. S., 1912. Geol. Mag. Lond., N.S., Dec., V, 9, 399. Vageler, P., 1933. ‘‘ An Introduction to Tropical Soils ’’, MacMillan & Co. Van Bemmelen, R. W., 1941. Hc. Geol., 36, 630. Voisey, A. H., 1942. Tuts JouRNAL, 76, 82 Weiss, F., 1910. See E. S. Moore and J. E. Maynard, 1929. Whitworth, H. F., 1939. Unpublished report, Mines Dept. N.S.W. Wilkinson, C. S., 1875. Mines Min. Stat. N.S.W., 77. Woolnough, W. G., 1927. Tuis JourNAL, 61, 17. DESCRIPTION OF PLATES. Prare. EL, Fig. 1.—View looking across the valley of the MacIntyre River towards the road to Copeton, taken from the Lookout near Inverell. Basalt outcrops as a continuous sheet from the high ground down over the cleared patch in the centre of the photo, to the valley floor. Granite outcrops in the trees to the left of the cleared patch. Fig. 2.—Vermicular basalt in small gully below base of bauxite deposit at Lorne, near Emma- ville. The tubes are filled with iron-stained earthy material. Fig. 3.—Photograph of sample of vermicular bauxite from Sutton Forest. The association of the lighter coloured material with the tubular openings is well shown. x }. Fig. 4.—Micro-section of sample of vermicular basalt shown in Fig. 3, illustrating what appears to be incipient pisolitic structure. x 250. Fig. 5.—Micro-section of bauxite nodule from Four-Mile near Inverell. The original outlines of the felspar laths can be seen. At present they consist of aggregates of gibbsite crystals. x12. K—September 6, 1944. how, F. N. HANLON. Pruate III. Fig. 6.—Photograph of quarry between Stannifer and Tingha. The shaft sampled is shown in the foreground, covered by wood. Photograph taken about midday in winter. Fig. 7.—Photograph of quarry at Four-Mile, near Inverell. The shaft sampled is shown in the floor of the quarry covered by wood. The sampling cuts on the northern side and eastern end of the quarry are also shown. The surveyor’s staff shown is 10 feet long. Fig. 8.—Close-up of the face of the quarry shown in Fig. 6. The variation in the size and the banding of the pisolites can be seen. The Brunton compass shown at the bottom of the photograph was included as a scale. Fig. 9.—Close-up of the face of the quarry shown in Fig. 7. The nodular structure is well shown. THE DETERMINATION OF CALCITE AND ARAGONITE IN INVERTEBRATE SHELLS. By D. M. BRAY, Geological and Mining Museum, New South Wales. Communicated by F. N. Hanton, B.S8Sce., Dip.Ed. (Published by permission of the Under Secretary for Mines.) Manuscript received, August 10, 1944. Read, September 6, 1944. INTRODUCTION. It has been generally accepted for many years that certain invertebrate shells are essentially composed of calcite, whilst others are essentially aragonite. Small amounts of conchiolin and other calcium salts are of course always present, and also, in certain cases, magnesium salts. Calcite is the hexagonal form of CaCO,, and aragonite the orthorhombic form. The presence of calcite or aragonite in invertebrate shells has some geological significance in the formation of certain limestone beds (Clarke, 1924) and is also of interest to zoologists in the study of shell deposition, environment, etc. Workers in these fields have used various methods for the differentiation of the two crystal identities, but in view of the metastability of aragonite, solubility differences due to varying particle size and the presence of organic matter and other salts, the purely chemical tests are considered unreliable under certain circumstances. One of the most frequently used methods is that due to Meigen (1901, 1905). The test, which depends on the differences in solubility of calcite and aragonite, consists of boiling samples of the powdered material in dilute solutions of cobalt nitrate, when aragonite develops a reddish-violet colour, calcite remaining colourless, or becoming pale blue. This procedure is well known and has lately been used by Manigault (1939). While some results obtained by the method are confirmed by X-ray powder analysis, experiments will be described in this paper which show that calcite of sufficiently small particle size will yield the same result as aragonite. The effect of small traces of impurities, as well as variations in boiling time and strength of reagent, have also been investigated. Another purely chemical test which has been investigated consists in treating the specimens with a solution of manganous sulphate containing silver sulphate, aragonite producing a black precipitate in from one to two minutes, the colour with calcite developing only after one to two hours (Feigl, 1939). Of the other tests which have been used, little need be said. The differences in specific gravity and solubility in H,CO, have been employed, but many of the results obtained thereby contain serious discrepancies (Cornish and Kendall, 1888 ; Cole and Little, 1911). Optical methods are available in many cases, and have been used for the determination of invertebrate shells (Trueman, 1942). It should be pointed out, however, that the inversion of aragonite to calcite does not necessarily involve actual morphological changes, and it is insufficient to classify crystals as aragonite on the basis of mere external Symmetry. 114 D. M. BRAY. EXPERIMENTAL. Reference to literature published since Meigen’s original papers on the subject reveals a singular lack of uniformity in the procedures recommended by various authors with regard to the cobalt nitrate reaction. The time of boiling varies from three to twenty minutes, and in most cases, no indication is given as to the strength of the cobalt nitrate solution which should be used. Moreover, the staining colours recorded by different authors cannot be reconciled, and in many cases are quite contradictory. In the early stages of the work described herein, Meigen’s original paper was not available to the author, and it was decided to study the effects of variations in reagent strength and boiling time. Accordingly, the following solutions of cobalt nitrate were prepared: 0-31 M, 0-062 M, 0-031 M and 0-016 M. As the pH values of the solutions are important, these were determined, and found to be 5:3, 5-8, 6-2 and 6-3 respectively. The technique adopted was as follows: Approximately 0-1 gm. of crushed specimen was placed in each of six test tubes and covered with 5 ml. of one of the cobalt nitrate solutions. They were then placed in a boiling water bath, and one removed every five minutes, so that the immersion time varied from five to thirty minutes. Immediately on removal the test tube was filled with water, the stained specimen washed several times by careful decantation, and then transferred with the minimum amount of water into a small porcelain crucible. After several minutes’ settling, the water was carefully removed with a pipette. The colour was then observed against the white crucible, all colours being observed by daylight. In nearly all cases a small amount of the specimen floated on the reagent, and after boiling either a blue or violet ring was observed round the test tube at this point, according as the mineral was calcite or aragonite. The bulk of this was removed in the decantation, the remainder adhering quite firmly to the tube, however, so that it did not vitiate the test. These coloured rings appear to be related to the concentration of CO, at the surface of the liquid. Meigen’s researches, subsequently referred to, indicated that in his experiments he added 1 g. powdered specimen with 100 ml. boiling water to boiling solutions of cobalt nitrate in Erlen- meyer flasks. The resulting solutions varied from 0-66 M to 0-16 M, and the boiling time ranged up to fifteen hours. This was to convert all the calcite or aragonite to cobalt carbonate, for chemical analyses. In publishing the results of this work, he recommended that the samples should be boiled from 10 to 20 minutes (Meigen, 1905). The tests were first carried out on the powdered shells of Saxostrea commercialis. The results obtained are best presented in tabular form, and these are set out below. TABLE [. Shows staining produced by boiling powdered shells of Saxostrea commercialis in Co(NO3;). of different strengths, with variations in time of boiling. Time of Boiling and Colour Produced. Concentration | of Solution. (1) (3) (5) (6) 5 mins. 10 mins. | 15 mins. 20 mins. 25 mins. 30 mins. 0:31 M Pinkish lilac. Lilac + light | Pale mauve+ | Mauve +blue | Mauve+blue | Deeper mauve blue specks. blue specks. specks. specks. + blue specks. 0-062 M Pink. light mauve. | Mauve +blue | Mauve, etc. Deeper}! Same as (5). specks. mauve, etc. 0:031 M Faint lilac | Same as (1). light mauve. light mauve. Deeper/| Same as (5). pink. mauve. 0-016 M Pinkish | Mauve. Mauve. Deeper! Same as (4). Same as (4). mauve. mauve. On boiling powdered specimens of aragonite (pseudo-hexagonal from Molina, Spain) in the same set of solutions, a reddish-violet colour was developed in every case. It will be seen from Table I that the staining produced could be taken as indicative of aragonite, but comparison with the above results shows that the colours were somewhat lighter DETERMINATION OF CALCITE AND ARAGONITE. | AS in shade than those developed by the genuine aragonite. Moreover, the results of the X-ray powder analyses indicate as shown below that the shell of this organism is composed exclusively of calcite : Saxostrea commercialis. (1) Porcellaneous layers from near base of shell—thickest cross section of porcellaneous layers. Pt. Hacking he me a as .. Calcite. (2) Nacreous layers from near base of shell-—thickest cross section of nacre. Pt. Hacking. . ays Ae br is bs ss ye ... | Calcite. (3) Composite sample. Pt. Hacking Ne he oe oe .. Calcite. (4) Porcellanous layers. George’s River... ae By eae .. Calcite. (5) Nacreous layers. George’s River og He ie ee .. Calcite. (6) Composite sample. George’s River he ee he ae .. Calcite. Cobalt nitrate tests on the shell of Pecten medius resulted in the development of a pale mauve colour, with numerous small blue spots, which rather suggested the presence of both aragonite and calcite. The powder photograph, however, indicated the presence of calcite only. Since less than 5% of impurity is detectable by the X-ray method, the likelihood of sufficient aragonite being present to materially affect the test was remote. Certain of the species tested with cobalt nitrate by Meigen and Manigault were examined by the former’s method, and also by X-ray analysis. ‘The powdered specimens were boiled in a 0-062 M solution of reagent for fifteen minutes. The specimens examined were Distichopora, Pocillopora, Heliopora, Millepora, Helix aspersa, Spirula peronii and Terebratula caput serpentis. All of these gave a violet colour with cobalt nitrate excepting the last mentioned, which assumed only a faint greenish tinge. On X-ray analysis, all were found to be aragonite except 7’. caput serpentis, which was calcite. These results were in complete agreement with Meigen’s determinations. Of the above species H. aspersa was examined by Manigault (Manigault, 1939). ‘In order to ascertain the effect, if any, of the time which elapsed between the crushing of the specimens and their analyses, an additional sample of nacre of S. commercialis was crushed and immediately X-rayed. This sample, however, was also calcite. In addition, a portion of the genuine aragonite specimen from Spain under the same treatment revealed the true aragonite structure, thus indicating that the action of crushing and grinding could not have effected the inversion to calcite. From this it is evident that the cobalt nitrate tests on the shells of S. com- mercialis and P. medius were invalid. Tests carried out with Feigl’s reagent also erroneously indicated the presence of aragonite. Although chemical analysis indicates that the shells of S. conymercialis contain 97-55% CaCO,, 1-48% organic matter and only 0:97% MgCOs,, it was decided to study the effects of small amounts of MgCO,, Ca,(PO,), and CaSO,. Accordingly, samples of pure calcite were crushed, and 2% of the above salts added. These were then treated with the cobalt ritrate solutions, and boiled for fifteen minutes. TABLE II. Showing the Effect on Meigen’s Reaction of Small Impurities in Calcite. Sample. 0-062 M. 0-031 M. | 0-016 M. Calcite (clear rhomb) .. ne uy 3 .. | Pale blue. | Pale blue. | Very pale blue. rss (pseudomorph after aragonite) he .. | Pale blue. Pale blue. | Pale blue. cs +2% MegCO, (chemically prep.) uy .. | Pink+blue specks. | Pink+blue specks. | Pink-+-blue specks. ” +2% magnesite (MgCO,) .. x .. | Pale mauve-+blue | Mauvish pink. Very pale pink. sp. | Ue +2% apatite (CaF)Ca,(PO,)s; ny Brin -- Very pale blue. Bluish discolora- tion. ey +2% gypsum (CaSO,.2H,O) oe, .. | Pale buff pink. Very pale pink, — very pale blue. 116 D. M. BRAY. By carrying out similar experiments with Feigl’s reagent, it was similarly observed that calcite with 2% of MgCO, also gave the aragonite reaction. At this stage of the work further experiments with Feigl’s test were abandoned, as the reaction is regarded as too sensitive for use on shells, and requires also special precautions to ensure the absence of CO, from the reagent and atmosphere. No deeper staining was produced by cobalt nitrate when the content of MgCO, was increased to 15%. The sheils of S. commercialis, however, exhibit greater reactivity than the pure calcite, and this may be explained by the extreme smallness of ultimate particle size, which increases the rate of solution. Moreover, although the organic matter is thought to have no chemical effect, it is present as a matrix which is an integral part of the microscopic shell structure, and this undoubtedly increases its porosity. It was observed that the shells of S. commercialis and P. medius contained considerable areas of chalky material, whereas the shells of S. peronit, H. aspersa and T'. caput serpentis contained practically none. The nacreous layers of S. com- mercialis are not quite as hard as even that of the calcite shell 7’. captus serpentis, and are much less resistant to attrition and comminution. This is attributed to the fact that the calcite crystals in S. commercialis are not firmly coherent, particularly in the chalky areas, which are very soft, and may result from colloidal precipitation, random orientation of particles perhaps being involved. The above facts were demonstrated by immersing crushed samples of shell, aragonite and calcite for one hour in solutions containing 0:01% and 0-001% methylene blue, with subsequent washing. The calcite and aragonite were only very slightly coloured, whilst the shell was considerably stained. Approximately 0-1 g. sample in 10 mls. solution :— Calcite— Methylene blue 0-:01% at oes ie .- No colour at yl) OR001S, Ms BY: us .. Just discoloured Aragonite— Methylene blue 00-01%, se Ye “ft .. Pale lavender blue as cen) “OOS. oe a es .. Very faint blue Shell— Methylene blue 0:01% fe ve ile .. Royal blue a oo O-00L, ay i ng .. Lavender blue Some of the intense colour of the shell is due to the dye attaching itself to the organic matter, but it must be remembered that this only amounts to 1-48%, and cannot be entirely responsible. Most of the colour is undoubtedly caused by the porosity of the shell, the dye being trapped in the cementation cracks of the somewhat loosely cohering calcite crystals. Meigen reported (1905) that the degree of comminution of samples did not materially affect his test, except where large quantities of sample were taken. It is pointed out, however, that extreme fineness of particle size not only enhances solubility considerably, but also greatly increases surface area. Solution takes place most readily at particle corners and edges, and fragments may become more angular as the fineness of crushing increases (Gaudin, 1939). It is considered also that reaction takes place between the solid-liquid phases by surface absorption of cobalt nitrate and deposition of cobalt carbonates on the solid material. Thus, fineness of grinding will have a very big effect on the reaction. In any case, the samples used in the present work (0-1 g.) could hardly be called large, and the effect of smallness of crystal and fragment sizes if shown below. Calcium carbonate was precipitated in gelatinous form from hot concentrated solutions of Na,CO, and CaCl,. This was then filtered and carefully washed at the pump to remove all traces of soluble chlorides and carbonates. It was then transferred to a beaker, washed with boiling water several times by the decantation method (which caused complete precipitation) and dried at 50°-60° C. Two such preparations were made, one containing a trace of SO,’, which was added to ascertain whether its presence would stabilise the aragonite structure if formed. When dry, samples were boiled in 5 ml. of 0-062 M cobalt nitrate for fifteen minutes. Both samples DETERMINATION OF CALCITE AND ARAGONITE. AN, gave a violet colour, identical with that produced by the aragonitic shells. In order to ensure that these samples were calcite and not aragonite, portion of one was left in its mother liquor exposed to the atmosphere overnight, and washed by repeated boilings and decantations next morning. It was then boiled in water for fifteen minutes, decanted and portion treated as above with cobalt nitrate. The same violet colour was produced, with darker specks. X-ray powder analysis four days later showed that both preparations were calcite. Further Meigen reactions carried out on the same samples showed, if anything, a slightly deeper colour, thus indicating that no falling off in reactivity or inversion had occurred. By pulverising a shell of S. commercialis to a fragment size of —200 mesh and boiling in cobalt nitrate under the above conditions, a colour approaching amethyst was obtained. Pure calcite under the same conditions was coloured blue with a considerable amount of pink. SUMMARY AND CONCLUSIONS. It has been shown that Meigen’s reaction and also that of Feigl for the differentiation of calcite and aragonite give erroneous results when applied to the shells of Saxostrea commercialis and Pecten medius taken from Sydney waters, although seven determinations by the former method have been verified by X-ray powder analysis. Variations in reagent strength, boiling time and the presence of MgCO, all cause small cumulative errors in the cobalt nitrate method, while the presence of organic matter and MgCO, vitiates Feigl’s test. With loosely cohering bodies of CaCO, of a cryptocrystalline or colloidal nature, or extremely comminuted fragments, both of these tests are definitely not specific for either aragonite or calcite. If these tests are to be used, they should be applied with extreme caution, and not to samples of the abovementioned nature. It is probable that many invertebrate shells contain large quantities of chalky material, and the test should definitely not be applied to them. In any case, for shells which are nacreous and homogeneous throughout, the more certain physical tests are always available. Finally, the author desires to emphasize the extreme metastability of aragonite, and the probability of variations from place to place even in a single Species. So many factors can be involved in the aragonite-calcite system that the classification of whole genera and species of invertebrates on the examinations of a few specimens is unsafe. Without a comprehensive investigation of many Specimens from all localities, it is considered that little can be done other than to record the crystal forms found in the different species at various places. ACKNOWLEDGMENTS. The author desires to express his thanks to Mr. F. P. J. Dwyer, M.Sc., for his interest and helpful criticism, and to the Chemistry Department, Sydney Technical College, for the use of the X-ray apparatus ; also to Mr. J. L. Sullivan, A.S.T.C., who performed most of the analyses. Thanks are due to Mr. T. Hodge- Smith for his criticism of the manuscript, and to Messrs. H. F. Conaghan, B.Sc., and A. C. Brigden, B.Sc., for partial shell analyses. Assistance in the matter of specimens by officers of the Australian Museum is also acknowledged. REFERENCES. Clarke, F. W., 1924. Data of Geochemistry, U.S. Geol. Surv., Bull. 770. Clarke, F. W., and Wheeler, 1917. U.S. Geol. Surv., Prof. Paper 102. Cole and Little, 1911. Geol. Mag. Lond. Cornish and Kendall, 1888. Geol. Mag. Lond., 5. Feigl, F., 1939. ‘‘ Spot Tests’”’, Nordemann, New York. Gaudin, H. M., 1939. ‘* Principles of Mineral Dressing ”, McGraw-Hill, New York. Manigault, P., 1939. Ann. de VInst. Ocean., XVIII, 331. Meigen, W., 1901. Centralb. Min., 577. 1905. Ber. Naturforsch. Ges. Freiburg, (15), 55. Trueman, E. R., 1942. J. Roy. Mic. Soc., LXII, 3 and 4, 69. QUATERNARY ARSONIUM SALTS AND THEIR METAL CO-ORDINATION COMPOUNDS. Part I. BISMUTH. By F. P. DWYER, M.Sc., N. A. GIBSON, B.Sc., and R. 8. NYHOLM, M.Sc. Manuscript received, August 8, 1944. Read, September 6, 1944. A number of metal compounds with quaternary arsonium iodides have been described previously, but no attention has yet been paid to the possibility of using them as reagents in the micro-detection or estimation of the metals con- cerned, for which, by reason of their large molecular weights they should be particularly suitable. The work described in this paper was carried out to determine their suitability for the detection and estimation of the metal bismuth. Burrows and Turner (1921) have described addition products of the type (PhMe,As)MI, with iodides of phosphorus, antimony, bismuth and stannic tin ; also similar products from PhMe,AsI and iodides of cadmium, lead and mercury. Later (1926) they prepared PhAsHMe,. Bil, by heating together the arsine, bismuth iodide, and excess of HI, and obtained a scarlet compound soluble with difficulty in hot concentrated hydrochloric acid, from which it recrystallised in scarlet plates. In the present work, two series of methyl aryl tetra-iodo bismuthites have been prepared, and the effect of the substituent groups in the arsonium radical on the properties of the resultant tetra-iodo bismuthites studied, with special reference to the sensitivity of the reaction. The compounds were insoluble in water, soluble in hot N/2 HCl, the solubility decreasing rapidly as the molecular weight increased, being less than one part in a thousand in the case of p-tol,MeAsBil,. The compounds were quite soluble in hot alcohol. The colours changed progressively from brick-red to yellow as the molecular weight increased, while the p-tolyl derivatives in each case were lighter than the corresponding phenyl derivatives. The melting points also decreased with increasing molecular weights, with the exception of ((C,H,)3(CH,)As)(Bil,), which had a melting point higher than ((C;H,).(CH;),A8)(Bil,) and ((CgH5;)3(CH3)As)(Bil,). The sensitivities of the quaternary arsonium iodides for the detection of bismuth reached the maximum near the middle of the series, owing to the fact that as the molecular weight increased, decreasing solubility was compensated by decreasing intensity of colour. Contrary to expectation, the (C,H ,)(CH;),AsI is the most sensitive, being probably due to the fact that, although it is almost as deeply coloured as the compound with (C,H;)(CH;),AsI, the para methyl group depresses the solubility out of proportion to its contribution to increase to molecular weight. The reagents at present used for the detection of bismuth — include thiourea (Sensi and Seghezzo, 1929) which will detect 10y/ml. and cinchonine iodide (Feigl and Neuber, 1923) capable of detecting 3y/ml. All of the arsonium iodides used in this study were superior to both of these reagents, whilst the best, (C,H,)(CH;),AsI, is twenty times as sensitive as thiourea, and six times a8 cinchonine iodide. — QUATERNARY ARSONIUM SALTS. 119 These R,AsBil, compounds readily underwent hydrolysis, partially in cold water, and completely in boiling water, to the compound R,AsBiOI, (III), which could be readily reconverted to the compound R,AsBil, by boiling with dilute hydrochloric acid and a small amount of R,AsI. The partially hydrolysed compound can be further hydrolysed, by boiling with N/2 ammonium hydroxide, to the final possible hydrolysis compound, (R,AS8),(Bi,O;I,). When, however, the compound R,AsBil, was boiled with N/2 ammonium hydroxide, instead of the final hydrolysis compound being formed, complete breakdown to bismuth oxide occurred, whilst a smell of tertiary arsine could be detected. It was considered that these reactions proceeded as follows : IESG) wh H 120, \y/ \ | aon vg L ii 7, Kp) wey wiser | > 4 end /t,As bc 100 heAs Bel 100% AAs aber HO HON 0 a = aa = | The intermediate compounds (II) and (IV) were too unstable to be isolated, and their formation was probably only transitory. It was considered that the inability of the compound (I) to be transformed directly to (V) by boiling with N/2 ammonium hydroxide is due to the fact that the required intermediate (R,As)(Bi(OH),I) (VI) either cannot be formed or is extremely unstable, and hence the hydrolysis must be carried out in two stages as shown above. With the exception of the well-known BKiOI,, these hydrolysed compounds have not been previously described. EXPERIMENTAL. The quaternary arsonium iodides were made from the corresponding tertiary arsines by refluxing ‘with a slight excess of methyl iodide in alcoholic solution. They were precipitated by the addition of a large excess of ether, and recrystallised by dissolving in the minimum amount of alcohol, and reprecipitating with ether. The bismuth nitrate solutions were prepared by dissolving C.P. bismuth tri-oxide in the minimum amount of nitric acid with warming. Phenyltrimethylarsonium tetra-iodo bismuthite. (C,H;)(CH3),AsI (1 gm.), potassium iodide (1 gm.) and sulphuric acid (0:5 ml.) were dissolved in water (400 ml.) with warming. (More water is needed in the case of the higher quaternary arsonium iodides.) To this solution was added in a thin stream with rapid stirring a solution of bismuth nitrate (equivalent to 0-4 gm. Bi,O;). This is approximately equivalent to 0:5 gm. of the arsonium iodide ; the excess of the latter was considered necessary owing to the fact that any excess of bismuth would lead to the formation of the highly insoluble KBiOI,. The fact that the compound R,AsBil, was formed in preference is obviously due to its greater insolubility. The use of the quaternary arsonium iodides for the quantitative estimation of bismuth, and other metals, for example cadmium, will be the subject of a subsequent communication. The arsonium tetra-iodo bismuthite separated as a fine red precipitate, which coagulated on continued stirring, leaving the supernatant liquid 120 DWYER, GIBSON AND NYHOLM. quite clear and colourless. The crystalline compound was readily filtered with suction, recrystal- lised from N/2 hydrochloric acid, to prevent hydrolysis, in the presence of (C,H,;)(CH;),AsI (0:2 gm.), washed with N/2 hydrochloric acid, and dried in a desiccator over calcium chloride and calcium hydroxide. The compound was analysed as follows: The iodine was estimated by the method of Dwyer and Nyholm (1942). Arsenic was estimated by the chromic acid method of Anderson and Burrows (1936). It was noted that in this estimation, when the chromium is precipitated, the solution should be made definitely alkaline with sodium hydroxide, and boiled to decompose the sodium chromite to chromic hydroxide, otherwise the arsenic is precipitated as the basic chromium arsenate, and a low result is obtained. Bismuth was determined by fuming the compound with sulphuric acid and potassium nitrate, precipitating the resultant bismuth salt as the cupferron complex, and igniting to the oxide. This gave a high value, owing to co-precipitation of the arsenic, so the ignited precipitate was dissolved in nitric acid, repre- cipitated as the oxalate and the latter titrated with standard potassium permanganate. Found: As=8-17%; Bi=22-5%; I=56-1%. .Caleulated for ((C,H;)(CH),As)(Bil,) : As——8- 2197): Bi-— 229°) ipo 6.97. Diphenyldimethylarsonium di - iodo - oxy - bismuthite. This was produced by boiling ((C,H;),(CH,).As)(Bil,) three times in distilled water, grinding up any lumps formed in the process. The light chocolate powder was filtered and dried over calcium chloride. Found: IJ=34:2%. Calculated for ((C,H;).(CH 3),As)(BiOI,): I-=34-4%. Bis-diphenyldimethylarsonium di-iodo dioxy-4-oxy dibismuth (V). This was prepared from the previous compound (III) by grinding to a fine powder, suspending in N/2 NH,OH, boiling for half an hour, and filtering. The pale red-brown powder was dried over calcium chloride. Found: I=20-6%. Calculated for ((C,H),(CH,),As),(Bi,0;1,): t=_ 20859, Sensitivity. The sensitivity tests were carried out as follows : The standard bismuth nitrate solution was prepared by dissolving bismuth oxide (0-1117 gm.) in 15N nitric acid (approximately 18 ml.), and diluting to 100 ml. This solution contained 1-0 mgm. or 1,000 y of bismuth per ml., and was 2:5 N with respect to nitric acid. It was diluted 100 times to give a stock solution containing 10 y Bi/ml. and N/40 with respect to nitric acid. The quaternary arsonium iodide solutions were prepared by dissolving 0:5 gm. R,AsI, 0-5 gm. KI, 0-2 ml. H,SO, (36N) and 0-2 ml. H,SO, (5% free SO.) in 100 ml. of aqueous solution. Owing to low solubility, a saturated solution of (C,;H,),(CH,)AsI was used. The sulphurous acid was introduced to suppress aerial oxidation of the iodide to iodine, which gives a precipitate of the triiodide on standing. The sensitivity tests were carried out in half-inch test tubes, 1 ml. of the bismuth nitrate solution being pipetted in first, then 1 ml. of the arsonium iodide solution. The resultant test was viewed in artificial light directed at right angles to the line of vision. The R,AsBil, compounds at the concentrations observed appeared as an orange opalescence, with a faint green fluorescence, which was not observable at higher concentrations. The sensitivities are summarised in Table I. TABLE I. M.P. of M.P. of ; Compound. Arsonium Bismuth Colour. Iodine. -Todine. Sensitivity Iodides. Compounds. | (Theor.) (Found.) y/ml. GEG RSG cap tats Swe i i Aen ok B See ((C,H;)(CH,)3;As)(Bil,) La 244° C, 171° C. Deep red. 55-6% 56-1% 1-0 ((C,H;)2(CH;)2As)( Bil.) a 191° C. 161° C. Reddish orange. 52-0% 52-0% 0:7 ((C,H;)3(CH;)As)(Bil,) i 173° C. 80° C. Light orange. 49-1% 49-2% 0:8 (OG 7H e (CH): A2)\ BEL) s 269° O.* 124° C Red. 54-°8% 54:7% 0:5 ((C,H,).(CH;).As)(Bil,) ada 77° C Orange. 50°6% 50:7% 0:6 ((C,H,)s( CHAS CRLS Oe, 175° C 103° C Orange yellow. 47-1% 47:6% 0:8 ((CsH5)2(CHs)2As)(BiOI,) .. — — Light chocolate. 84-4% 84-2% — | | Where C,H; =phenyl. C 7=p-tolyl. * Rayziss and Gavron, “‘ Organic Arsenical Compounds ’’, gives M.P.=247-5° C. QUATERNARY ARSONIUM SALTS. 121 SUMMARY. The use of quaternary aryl alkyl arsonium iodides for the micro-detection of bismuth has been investigated, and it has been shown that these compounds can be used to detect bismuth in concentrations of less than ly of bismuth per millilitre of solution, p-tolyltrimethylarsonium iodide giving a reaction down to 0-5 y/ml. REFERENCES. Anderson and Burrows, 1936. Tuts JouRNAL, 70, 63. Bartholomew and Burrows, 1926. THis JourNAL, 60, 208. Burrows and Turner, 1921. J.C.S. Transactions, 117, 1373-83 and 426-37. Dwyer and Nyholm, 1942. TxHis JourNAL, 76, 129. Feig] and Neuber, 1923. Z. anal. Chem., 62, 273. Rayziss and Gavron. ‘“‘ Organic Arsenical Compounds ’’, First Edition, 90, 92. Sensi and Seghezzo, 1929. Annali Chim. Appl., 19, 392. Department of Chemistry, Sydney Technical College. THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. By ADOLPH BOLLIGER AND MARGARET H. HARDY,* — From the Gordon Craig Research Laboratory, Department of Surgery, Unwwersity of Sydney, and The F. D. McMaster Animal Health Laboratory, Sydney. Council for Scientific and Industrial Research. With Plates IV-VII. Manuscript received, September 13, 1944. Read, October 4, 1944. The skin of the newly born young of Trichosurus vulpecula, the common or grey Australian possum, presents a homogeneous unpigmented appearance up till two months of age. In the third month of its life in the pouch a grey pig- mentation appears on the skin of its dorsum which within a few weeks deepens and frequently appears as an almost black discoloration. This is in marked contrast to the ventral aspect, which is unpigmented with the exception of a narrow median band of grey situated over the sternum. Towards the end of the third month it becomes evident that the pigmented skin on both the dorsal and ventral aspects of the body is covered by grey or black hairs and that the unpigmented areas are covered by white hairs alone. The pigmented sternal strip at this stage is approximately of equal width in the two sexes (4-8 mm.). In the female it is about 25 mm. long, while in the male it extends further up the neck and is about 35 mm. long (Figure 1). The grey hairs on this area are not obvious during the fifth and sixth months of life, partly because the light hairs on the ventral surface pass through a period of yellow coloration (Bolliger and Carrodus, 1940). After the replacement of these yellow hairs by white ones during the seventh or eighth month of life the sternal hair fibres are distinguished from those elsewhere on the venter by their grey colour and coarser texture. Between the eighth and twelfth months these grey sternal hairs in both sexes begin to assume a brown colour along the whole length of the shaft, a process more pronounced and rapid in the male than in the female. It is mainly this sternal skin area, covered by its distinctive hair fibres, which forms the subject of the present investigation. In many animals, notably the fully grown male, this region is further characterised by a moist and frequently copious secretion with a distinct but not disagreeable odour. It seemed reason- able to suppose that some special glandular activity in the skin was responsible for this, and consequently not only the hairs but also the underlying skin were examined in detail. MATERIAL AND METHODS. Over a hundred specimens of Trichosurus vulpecula of all ages obtained from many different localities of eastern New South Wales have been examined. Skin from the sternal region and other parts of the body of four adult males and four adult females, obtained immediately after the animals were killed, was fixed in 10°, formalin. Serial sections parallel to the surface of the skin were * Walter and Eliza Hall Fellow in Economic Biology, University of Queensland. THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 123 prepared by the method described by Carter (1939). These were stained with hematoxylin, eosin and picric acid. In a limited number of cases sections perpendicular to the skin surface were also made. Skin from three pouch young was also obtained and treated in a similar manner. Whole mounts of fibres were made for the detailed study of fibre structure. MORPHOLOGY. In order to discuss the peculiarities of the sternal integument it will be necessary to compare it with that on other parts of the body. It is convenient to describe first the general external appearance of the skin and hairs as seen in the living animal, and then the histological features of the underlying skin structures as studied in serial sections. Haternal Appearance. (1) Integument in General. The skin of the adult is unpigmented. The body is covered with a dense, rather woolly coat of fur, there being from 75 to 300 hairs to the square millimetre, varying with the individual and the body region examined. The length of the individual hairs varies considerably, but the average length of hairs in most animals is between 2 cm. and 3 cm. Two types of hair may be distinguished: (a) long, thick and straight hairs which project beyond the rest of the fur, and (bd) shorter, fine and wavy hairs which are much more numerous and form a dense underfur. The first type are grey or white in the proximal half. The thicker distal half is generally dark brown or black in hairs from the back, and white in hairs from the belly. A medulla is generally continuous along the major portion of the fibre, but is absent from the extreme tip and is discontinuous at the base. Hairs of the second type show considerable variation in length, but there is a continuous gradation from the shortest to the longest fibres. The longest ones have a thickened tip which is bent at an angle to their main axis. All hairs of type (b) have a more or less crimped appearance similar to that of sheep’s wool. They are all grey or white in the proximal half and those of the back are brown or black in the distal half, while those on the belly are white. On the back the longest hairs of this type generally have a white band in the thickened portion near the tip. The hairs are usually medullated over most of their length, but the attenuated tips and frequently the bases at epidermal level are without a medulla. In older males on limited and varying areas over the thoracic part of the back the bases of all types of hair may have a light yellow-brown colour. In a small number of possums, pigment may be entirely lacking from the hairs of all parts except the sternum. These animals though white are not however true albinos. (2) Sternal Arca. The skin is unpigmented here as elsewhere. In juvenile and fully grown possums all the hairs on the sternal area are short, thick and fairly straight. Their average length is 5-10 mm. They are more sparse than those on the surrounding areas, and their slope caudally may be clearly seen. In this region it is not easy to discern the two main hair types which are found on other parts of the body, although histological studies indicate that the same two types of hair follicles are present here as elsewhere. In the adult all the hairs are a reddish-brown due to the presence of pigment in both cortex and medulla. The extreme tips of the fibres are without medulla or pigment but elsewhere within the shafts the pigmentation is very uniform. Even specimens of the white variety of Trichosurus vulpecula which lack pigmentation on the rest of the body except the eyes have a typically pigmented sternal area. 124 BOLLIGER AND HARDY. The area of skin covered by these hairs differs in the two sexes after reaching maturity. In the female it may measure about 5-7 cm. in length and 1-3 em. in width. In the fully grown male it may measure 8-10 cm. in length with a maximum width of 3-5 cm. Though there are fairly wide variations in the size of the sternal area of different possums the average dimensions in the male are definitely greater than those in the female. The male sternal patch is also characterised by its shape, which, due to lateral extension at the level of the forelegs, frequently takes on a triangular or diamond shape. Cephalad it usually extends along the ventral neck in a broad band while the female sternal patch extends cephalad only as far as the end of the sternum (Figures 2 and 3). Skin Histology. Unfortunately no detailed description of skin structure in the possum has been published. De Meijere (1894) devoted one paragraph to the arrangement of hairs in groups and bundles, and Gibbs (1938) gave a full deseription of the arrangement of hairs in the pouch young but did not describe the adult appearance. Neither author mentioned the modified sternal area. It will therefore be necessary to describe general skin structure in some detail. (1) General Integument. The epidermis of the adult possum consists of a thin stratum corneum and a thin stratum malpighi, the latter having about four layers of cells. No trace of pigmentation could be seen in the epidermis. The dermis is about one millimetre or less in thickness and consists mainly of a network of collagen fibres which pass between and below the groups of hair follicles and associated structures extending into the dermis. There are no papille projecting into the epidermis, and pigment was not seen. The hairs are arranged in groups (Figure 4), each group censisting of a central, more or less thick hair and two or more lateral clusters of finer hairs. Occasionally a central hair follicle stands alone or is accompanied by only one lateral cluster. In the deeper portions of the dermis each fine hair of the lateral clusters has its own follicle but near the skin surface the adjacent follicles of each cluster unite to form a common follicle. The follicle of the central hair stands alone and generally has a separate opening on the skin surface, but occasionally one or two of the finer hair follicles unite with it near the surface. Each central follicle has a sebaceous gland and a sudoriferous gland associated with it, while each lateral cluster has a sebaceous gland but no sudoriferous gland. The absolute and relative sizes of the two types of hair follicles, the number of lateral clusters per central follicle and of hairs in each cluster, and the total density of follicles in the skin may vary considerably between individuals and between body regions. Each central follicle extends deep into the dermis and forms a more or less acute angle with the skin surface.1 A follicle papilla of dermal origin and inner and outer root sheaths are present. The inner root sheath is keratinised and continuous with the stratum corneum superficially but in the deeper layers its nucleated cells closely surround the fibre. The outer root sheath has only about two layers of cells. The hair fibres are round or oval in cross-section, and possess a cuticle and cortex with or without a medulla at skin level. Pigment, which is confined to the cortex and medulla of the hair, varies according to the age of the specimen and the body region examined. ‘There is also considerable variation even between adjacent fibres in the occurrence of pigment. Some central follicles may be involuted, the fibres being absent. 1 Structures within or close to the acute angle formed by the follicle and its perpendicular projection on the skin surface are referred to as lying on the acute aspect of the follicle, while those within or close to the obtuse angle so formed are referred to as lying on the obtuse aspect of the follicle. THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 125 Each central follicle has a sebaceous gland and a sudoriferous gland opening into it. The lateral clusters of follicles lhe parallel to and on either side of the central follicles. Individual lateral follicles extend to various depths in the dermis but generally are not so deeply implanted as the central follicle of the group to which they belong. In the deeper levels the inner root sheath is similar to that of the central follicle. In the common follicle opening the nucleated inner root Sheaths may continue to surround each fibre, or they may be replaced by keratinised epithelial fragments scattered between the fibres. The outer wall of the common follicle is similar in structure to the stratum malpighi and con- tinuous with it. Directly below this each fibre has its own follicle sheath, in which the outer root sheath is very thin, the follicles being packed tightly together. Still deeper, the individual follicles become quite distinct, although still close together, and each has its own root sheaths and papilla. The fibres are fine, and round or oval in cross section. Pigment may or may not be present in the cortex, as in the central fibres. A medulla, which is less frequently present than in central fibres, may contain pigment. At the level where the follicles of a cluster are most closely packed together, they are surrounded by a mass of sebaceous gland tissue. No sudoriferous glands are associated with the lateral clusters. Fragments of smooth muscle which might represent rudimentary arrector pili muscles are seen in the dermis, but their attachment to follicles or connective tissue sheaths has not been seen. Gibbs (1938) described an arrector pili muscle attached to the connective tissue sheath around each follicle group in pouch young. The sebaceous gland accompanying each central follicle is of the simple acinous type and forms a collar of tissue surrounding the upper part of the follicle. The most superficial part of the gland opens directly into the follicle from the acute aspect. This opening is deeper in the skin and on the side of the follicle opposed to the opening of the sudoriferous duct. Slightly deeper in the skin there may be a second sebaceous gland opening from the obtuse aspect (i.e. on the same side of the follicle as the sudoriferous duct opening). The sebaceous glands show the characteristic appearance of active holocrine glands, with cells undergoing decomposition in the formation of secretion. Masses of eosinophilic secreted material may be seen at the gland opening and in the hair follicle mouth. The collar of sebaceous gland tissue surrounding each lateral follicle cluster is slightly deeper in the skin than that of the central follicle. The opening into the common follicle mouth is at the most superficial end of the gland and on the acute aspect of the follicle. This gland shows the same signs of activity as the central follicle sebaceous gland. The sudoriferous gland accompanying each central follicle is a simple coiled tube and its activity is essentially of the apocrine type (Schiefferdecker, 1917). The gland hes beside a follicle group in the dermis and its duct opens into the upper portion of the central follicle just below the opening of the latter on, to the skin surface, and on the obtuse aspect. The duct is a slender tube passing through the dermis and its course is straight or only slightly sinuous. It lies beside the central follicle on the obtuse aspect. The wall of the duct is of uniform thickness and consists of one layer of a connective tissue sheath and two layers of undifferentiated epithelial cells with flattened nuclei. The lumen is very narrow, and no secreted material has been seen in it. The transition between the duct and the secretory portion of the gland is effected by a gradual widening of the lumen and a marked change in the nature of the lining epithelial cells. The secretory portion takes the form of a wide unbranched tube ending blindly and arranged in four or five loose coils. It lies beside the central follicle papille. 126 BOLLIGER AND HARDY. The wall of the gland consists of three layers: (i) a single layer of connective tissue cells surrounding the gland, (ii) a layer of myo-epithelial cells in various stages of contraction, (Ui) a single layer of secretory epithelium whose cells vary in shape according to the stage in the cycle of apocrine gland secretion (Schieffer- decker, 1917). They may be flattened, cubical, subcylindrical or cylindrical with irregular cytoplasmic processes extending into the lumen. The flattened type is the most common. The lumen is generally empty but eosinophilic colloidal or granular material is occasionally present. When the myo-epithelial layer 1s contracted the lumen decreases in size. Histological examination suggests that at least some of these sudoriferous glands are active. (2) Sternal Integument. An examination of skin sections from the sternal region shows that here the position and arrangement of the various structures is essentially similar to that on other parts of the body, but that there are clear differences in both absolute and relative sizes of certain parts, and in the apparent degree of activity of the glands. Some of the differences will now be described. A summary of the main quantitative differences in a comparison between the back and sternal regions of a single male specimen is given in Table 1. On the sternum the skin is 50 per cent. to 100 per cent. thicker than else- where on the body. This is due to a slightly thicker stratum malpighi and a very much thicker dermis. The epidermis is again unpigmented. The dermis is similar in structure to that of the general skin. The hair follicles are grouped in the same manner as elsewhere on the skin (Figure 5) but the number of lateral clusters per central follicle and the number of fibres per lateral cluster is appreciably lower than on any other body region examined. The number of groups per unit area is also low. A small lateral cluster often has a common follicle opening with the central follicle. The follicles are more deeply implanted on the average than in other regions and the average fibre thickness is greater. The follicles extend obliquely into the skin as on other regions. — The outer root sheath of the central follicles is somewhat thicker than on other regions, but the inner root sheath is very similar. The hairs of the central follicles are almost invariably medullated. Yellow or brown pigment is always present in the medulla and sometimes in the cortex. The inner and outer root sheaths of a lateral cluster follicle are more distinct from those of adjacent follicles at sebaceous gland level than they are on other parts of the body, probably due to the fact that on the sternum the follicles of a cluster are not so tightly packed together in the skin. Medullation is more common on the sternal region than elsewhere. Ags in other regions, no definite arrector pili muscle could be distinguished. The sebaceous glands accompanying both central follicles and lateral follicle clusters are extremely large and apparently very active. That sur- rounding the central follicle has one or two openings in the usual positions. Large areas within the body of the gland are filled with secretory material, the cells having broken down. Eosinophilic secreted material is also seen in quantities at the gland openings and in the hair follicles superficial to this level. The sebaceous glands of the lateral follicle clusters open widely into the common follicle near their superficial end in the acute dermal area. Deeper there are one or more openings on the obtuse aspect of the follicle. These openings are either into individual follicles or into the common follicle. The glands show the same signs of activity as do the central follicle sebaceous glands. The sudoriferous gland accompanying each central follicle is very much larger and apparently more active than those on the general body surface. The duct is much thicker and varies in diameter along its length. At its opening into the central follicle is a dilatation lined with epithelium similar to and continuous with the stratum corneum and stratum malpighi. Below this the THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 1 TABLE I. bo ~) Comparison of Quantitative Characters of the Skin of Dorsal and Sternal Regions in Possum No. 400 (Adult Male). Character. Thickness of epidermis Thickness of stratum corneum Thickness of stratum malpighi Number of cell layers in stratum malpighi | Thickness of dermis .. Number of lateral clusters ae hae Follielos | about each central follicle. Number of follicles per lateral cluster Number of lateral follicles per central follicle Number of central follicles per square millimetre Total number of follicles per square milli- metre : Average diameter of hair at levels Sebaceous glands of central follicles : ~ Depth : Width Sebaceous glands of lateral ‘clusters : Depth : Width ~=s.. Sudoriferous gland duct, external diameter (major axis). Internal diameter (major axis) Thickness of duct wall Content of lumen Secretory portion of sudoriferous gland : Size of gland mass 8 Width of lumen Wall thickness Activity judged by secretory epithelium. . Content of lumen L—October 4, 1944. Dorsum. Sternum. 24 u 28 u 12p 12u. l2u l6u 3 4 \Niaanaays 2 mm. 2—4 0-4 Average 3 Average 2 1-11 2.-8 Average 8 Average 6 Average 17 Average 5-1] Average 85 1a Max 80u Max. 70 u. Max. 90u Max. 70u About 20 Very narrow. One layer nective Two flattened — epi- dermal cells. con- To 500 u long. To 170u wide. To 150u. Po pl2cw: Some glands appear to be active. Eosinophilic material oc- easionally pre- sent. Average 12 Average 2:5 Average 30 32 Max. Max. 650 uw 300 u. Max. 700 u Max. 250 180 u superficially (dilatation). 70u ut base of dilatation, decreasing gradually to 50 UL, then increasing to 70 u | Up to 150 Ww in dilatation, 24 decreasing to 10 u, increasing to 35u in duct. Dilatation lined by stratum tissue. | layers | corneum and stratum mal- pighi. Superficially, duct proper has 6-7 layers of epi- dermal cells, the innermost partially keratinised. De- creases to 4-5 layers. De- creases to 2 layers epidermal cells connective tissue _ sheath. gs ra) reservoir and duct, occasionally eosinophilic secretory material. Occasion- ally crystals in male. To 1,100 long. To 700 u wide. To 500 u. To l4u Most glands appear to be very active. Eosinophilic material fre- quently present. Crystals in male. en 128 BOLLIGER AND HARDY. width of the lumen and of the duct decrease gradually ; here the wall consists of six or seven layers of epithelial cells similar to those of the stratum malpighi, the innermost layer of cells being flattened and partially keratinised. At the middle of the sebaceous gland level in the skin the lumen is at its narrowest and the flattened cells become replaced by a layer of more cubical cells whose cytoplasm is strongly eosinophilic and bears fine inwardly directed projections. Below this the thickness of the gland wall decreases to four or five cell layers. In the deepest part of the sebaceous gland stratum the thickness of the epidermal part of the wall decreases to two cell layers, and a connective tissue sheath becomes clearly distinguishable. Eosinophilic secretcry material has been found both in the dilated opening and in the duct proper. The lumen of the duct widens rather suddenly into the secretory portion of the gland, and at the same time the character of the innermost cells changes. The secretory portion (Figures 6, 7) is a wide unbranched tube ending blindly and forming a large mass of loose coils. It is situated deep in the dermis at and well below the level of the hair follicle papillz, and lies beside central and lateral hair follicles. The three typical cell layers are present in the wall. The myo-epithelial cells are generally relaxed. The secretory epithelial cells frequently assume the cubical and cylindrical rather than the squamous form, and cytoplasmic processes into the lumen are very numerous. The tips of these processes are sometimes basophilic. The iumen is sometimes empty, but eosinophilic colloidal or granular material is frequently present. Some of this material shows traces of cellular origin. The general picture appears to be one of marked activity. Perhaps the most significant result of the examination of sternal skin was the discovery of crystals in the lumen of the sweat glands of the sternal region. These were yellow, rod-shaped crystals with: oblique or square ends; they varied considerably in size. They have been found in the sweat glands of the sternal region of four male adult possums but were not seen in the corresponding region of four adult females. Nor have any such crystals been seen in any other body region either in males or in females. Crystals were seep most commonly in the lumen of the secretory portion of the sudoriferous glands (Figures 6, 8), where the rods tended to form rosettes, and were frequently associated with eosinophilic secretory material. Clumps or rosettes of crystals were also seen in the lumen of many gland ducts, both in the terminal reservoir and in the duct proper (Figure 9). Here again the crystals were frequently associated with eosinophilic material. The only other site of crystal formation which has been found is the wall of the secretory portion of the glands, where scattered rods were sometimes to be seen, most frequently in the connective tissue layer (Figure 10). A blood capillary was often seen near the gland wall not far from a crystal deposit. The crystals seen were yellow even in unstained sections, but their colour seemed to be enhanced by picric acid staining. They were sparingly soluble and birefringent. From this and other evidence to be published later, it is pes ok that they were urates. It seems definitely established that the crystals have been i in the course of secretion by the sudoriferous glands, but it is likely that this particular substance is not synthesised within the secretory cells. (3) The Histology of Sternal Skin in Pouch Young. The general features of the histology of the skin in pouch young which were described by Gibbs (1938) were confirmed in the present study. In addition to our observation on adult Trichosurus vulpecula, a comparison was made between the sternal region and other parts of the body in three male pouch young. The first of these measured 16:0 cm. from the tip of the snout to the root of the tail, and was about two months old, i.e. in the unpigmented stage mentioned at the beginning of this paper. The sternal area was at about the same stage of development as the back and belly, with groups of three, four and five follicles of which the THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 129 central one contained a keratinised fibre. No pigmentation of skin or fibres was seen in any of these regions. There was a sudoriferous gland accompanying each central follicle, but the terminal expansion had not yet developed. The sudoriferous glands on the sternal area extended as deep into the skin as the central follicles, while those on the back and belly did not extend so far. A sebaceous gland had begun to develop beside each central follicle on all areas, but only on the sternal area could sebaceous glands accompanying lateral follicles be distinguished. The second specimen was 17:9 cm. long, and was about 100 days old, i.e. in the stage where dark hairs cover the back and sternum. Here fibre develop- ment on the sternum was slightly ahead of that on the back, and most of the fibres on both areas were pigmented. Sudoriferous glands were slightly further developed on the sternum than the back and the terminal expansion which later gives rise to the secretory portion was just visible in the former region. Sebaceous glands had developed in association with the central and most lateral follicles and these also were slightly further developed on the sternum. The third specimen was 27-6 cm. long, and its age was estimated at seven months. Its ventral fur was in the travsitory stage of yellow coloration referred to earlier. In this specimen skin development had reached a more advanced stage than in any of the animals examined by Gibbs, there being about nine fibres developed to each group on the back. In this region the terminal expansion of the sudoriferous glands had begun, but on the sternum the development was further advanced, and the secretory portion was forming coils in the dermis. By comparison with those on the back, the sebaceous glands in the sternal region were very large and well developed. Thus it is clear from these few observations that while the general course of skin development in the sternal region is similar to that on other parts of the body, the main features which distinguish the sternal skin in the adult are already apparent in large pouch young. PHYSIOLOGY. From the morphological findings one may conclude that the sternal colour patch represents a male character which has its homologue in the female though on a lesser scale. This assumption is supported by the following observations. With the approach of sexual maturity the brown colour patch on the sternum becomes very obvious in the male. An oily fluid in noticeable amounts appears to be secreted on the sternal region which, particularly during the breeding Season, may be actually wet from this secretion. In captivity these males have been observed to rub the sternal region against protruding parts of the wall of the enclosure, and as a result a brown pigment is deposited in such places. Females, in which the development of the sternal patch is slower and less marked, have not so far been observed doing this. In fact in this sex the sternal hairs appear to be almost dry during the breeding season. However, increased secretory activity, as indicated by the wetness of the sternal region and the staining of surrounding hairs, is noted in the mother two or three months after a young possum is born ; and the young, after leaving the pouch, have frequently been seen licking the sternal area of the mother. In order to obtain some information on the nature of the sternal secretion pilocarpine was administered to male and female possums. The administration of 0-1-0-2 grain of pilocarpine, in all animals injected caused profuse sweating on the soles of the feet, and small beads of sweat were also seen on the clipped sternal area in both sexes, but to a greater degree in males than in females. The secretions throughout were alkaline in reaction. In all cases the fluid, which was collected by blotting with filter paper, was colourless in daylight. Fluid from the sternal region of the male turned to vellow and then to reddish- 130 : BOLLIGER AND HARDY. brown within a few days of exposure to air on the filter paper. In the female this change was less marked but samples of sweat containing considerable amounts of this or a similar chromogen were obtained from the pouch. Sweat from the foot pads of both males and females contained very little or no chromogen. However, judging from experiments in which hair or skin or both have been extracted with ether and water in the soxhlet apparatus for a consider- able time it may be concluded that the chromogen, though in much smaller amounts, is secreted by the skin in general, though here, too, more markedly in the male than in the female. From experiments with male sex hormones, which will be reported in detail on another occasion, there is further supporting evidence that in the male the sternal region undoubtedly is the preferential site for the excretion of the chromogen but that smaller amounts are secreted by the skin in general. It may be of interest to recall that Bolliger and Whitten (1940) reported the appearance of a chromogen in the urine of male and female T'richo- surus vulpecula which on standing in an alkaline medium produced a brown colour ; it seems possible that these two substances, the one from the skin and the other from the urine, are similar. If the hairs of the sternal patch as well as some of those surrounding it which may also be stained brown, are cut off, the new hairs regenerating will be brown again if they belong to the typical sternal area, while the hairs surrounding the sternal region and belonging to the integument in general will appear as white hairs and remain so until stained again from secretions derived from the specific sternal skin. ~ Compared with the rest of the fur, the sternal hairs show only little fluorescence when examined with ultra-violet light in a darkened room, and practically none when covered with skin secretion (Bolliger, 1944). DISCUSSION. So far only secant attention has been given in the literature to the distinctively coloured patch of fur over the sternal area of Trichosurus vulpecula. Le Souef and Burrell (1926) and Troughton (1941) in their recent books on the animals of Australia do not mention it. Wood-Jones (1924) in his description of the bushy-tailed possum (Z'richosurus vulpecula) as occurring in southern Australia states that on the ventral aspect the hairs are of a dirty white colour with a slight yellowish red tinge. ‘‘ This yellowish colour is accentuated on the chest in the males and around the pouch in the females.’’ Bolliger and Carrodus (1938) mentioned that in young females a brown streak over the sternum may or may not occur. From observations on over 100 individual possums from eastern N.S.W., it may now be stated that in fully grown sexually mature animals a brown sternal streak is always present, though varying considerably in size and being much more prominent in males. In comparing the brown sternal area with the yellow-orange or brown hairs present in the pouch of females, it may be said that the latter is a female character while the former may be considered a rudimentary male character when occurring in females. The orange or brown hairs which frequently surround the opening of the pouch simply represent temporarily stained abdominal hairs and cannot be compared with the straighter and sparser sternal hairs which are permanently stained through the whole fibre from the follicle papilla to the tip, and which are different from the remaining pelage of the possum. The pelage has few outstanding features to distinguish it from that of other mammals in various families possessing a dense woolly coat. The integument, in its histological structure, is fairly typical of the marsupials so far examined. All the sweat glands present in the furry skin are of the apocrine type only. Theskin of the sternal region differs from the rest of the integument Journal Royal Society of N.S.W., Vol. DXXVIIT, 1944, Plate IV eA ; ty. Pe Z hey { , Se ae oy) e mn i t 4 ' \ AS \ ‘ \ i N ‘ \ J . . an ~ ” t \ ¢ : = ’ . Journal Royal Society of N.S.W., Vol. LXXVIIT, 1944, Plate V ” Journal Royal Society of N.S.W., Vol. UXXVIII, 1944, Plate VI SARRETE rece rr mre entrgrenenmee cre ERS . ‘ Bee re. 7. Journal Royal Society of N.S.W., Vol. LX XVIII, 1944, Plate VII THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 131 only in the degree of development of various structures. There seems to be no justification for the term ‘“ sternal gland ”’ as it has been applied in other animals, and which implies something structurally discrete. The region is merely a limited but diffuse area of skin where certain structures such as the sebaceous and sudoriferous glands found normally over the general integument are locally enhanced in size and activity. In the fully grown male the sternal area attracts immediate attention due to the rich colouring and shiny “ lacquered ”’ appearance of the sternal hairs. As has been shown, this is due to a copious local skin secretion containing a chromogen which, within a few days, is converted into an orange-brown pigment, heightening the colouring of the sternal hairs proper and temporarily staining the surrounding hairs. In the normal female all these manifestations are on a smaller scale and more confined to the region of the pouch, and this may explain why well-formed and fairly large crystals, tentatively considered to be urates, were only found in the sternal sweat glands of the male. The biological significance of the sternal region is very probably that of an organ attracting the other sex by its colour and by the odour of its copious secretion. In the male both functions become very noticeable when approaching sexual maturity. These functions may also be employed by the male to mark a tree or a building which it inhabits in order to guide the prospective partner, an expedient probably necessary for nocturnal animals living widely dispersed. Similar behaviour, such as marking trees, has been noticed in the male of the koala (Phascolarctos cinereus). The odour, and probably the stain, from the sternal region of the mother may assist the young to find its parent after it has left the pouch, because in this period the female sternal secretion seems to be at its maximum. A sternal area characterised by large and active glands, as well as vividly coloured hairs, has so far only been met with in Trichosurus vulpecula. However, Beddard (1887) described an almost hairless glandular patch ‘‘ just anterior to the sternum ’”’ in another marsupial, the banded ant-eater (Myrmecobius fasciatus). This was confirmed by Bourne (1934) and Ford (1934) and the latter referred to the frequent occurrence of sternal or pectoral glands in marsupials, however, without giving any specific instances besides Myrmecobius. On superficial examination, no such sternal areas could be observed in the bandicoot (Perameles nasuta), grey kangaroo (Macropus major) or one species of wallaby, but the male of the native bear (Phascolarctos cinereus) seems to possess such an area, rich in active skin glands (MacKenzie, 1934). Bourne (1934) described glandular gular areas or pouches in some Australian Jerboa mice (Notomys cervinus, Notomys mitchelli and Podanomalus sp.) which probably are of a similar nature as those found in marsupials. He also observed a gular pouch in a bat (Nyctinomus australis). Amongst other mammals, sternal areas characterised by marked glandular secretion were described in primates such as the orang-utang (Simia sp.) (Wislocki and Schultz, 1925) and the spider monkey (Ateles sp.) (Schwarz, 1937). In all these mammals, including T'richosurus vulpecula, the same anatomical features appear to be present, i.e. an enlargement of the sebaceous and sudoriferous glands in a small area in the neighbourhood of the sternum. In Trichosurus vulpecula the connection between this accumula- tion of glands and the sex life of the animal seems fairly well established. SUMMARY. On studying the skin and hairs of Trichosurus vulpecula in pouch young and in fully grown specimens, it was found that at the age of three months a pigmented area appeared over the sternal region which subsequently developed into the distinctive sternal area. | 132 BOLLIGER AND HARDY. The morphology of the skin over the sternum was similar to that of the rest of the body but there were differences in the relative size and activity of several structures. The most important difference noted was that in the sternal area both sebaceous and sudoriferous glands were very large and active. For example, in one specimen sebaceous glands on the sternum had about 15 times the volume of those on the dorsum, and sudoriferous glands had about 25 times the volume of those on the back. The skin secretions were found to be as copious as the size of the glands suggested, giving the hairs, particularly in the male, a wet and oily appearance. In addition, the secretion contained a chromogen which stained the sternal hairs, as well as the surrounding hairs, a vivid brown. The hairs on, the sternal area of the fully grown animal were notable for their distinctive brown coloration, which was present throughout the shaft, and for their straightness and sparseness. These hairs may be considered to be a secondary sex character. They cover a larger area and are more brilliant in the male than in the female. The associated skin glands also were apparently more actively secreting in the male than in the female, and well formed crystals believed to be urates were found in the lumen of the sweat glands of the male only. ACKNOWLEDGMENTS. We wish to thank Mr. H. B. Carter for his encouragement and valuable criticism. We are indebted to Mr. E. Parrish for the microphotography, Mr. S. Woodward Smith for the macrophotography, and Mr. W. Clarke for the histological preparations. We also wish to thank the Council for Scientific and Industrial Research, and particularly Mr. D. A. Gill, Officer-in-charge, for accommodation given to one of us (M.H.H.) at the McMaster Laboratory. REFERENCES. Beddard, F. E., 1887. Proc. Zool. Soc. Lond., 527. Bolliger, A., 1944. Awst. J. Sc., 7, 35. Bolliger, A., and Carrodus, A., 1938. THis JouRNAL, 71, 615. —— 1940. Med. J. Aust.. No. 2, 368: Bolliger, A., and Whitten, W. K., 1940. Aust. J. Scz., 2, 178. Bourne, G., 1934. Mem. Nat. Mus., Melbourne, 8, 90. Carter, H. B., 1939. J. Coun. Sei. Ind. Res. (Aust.), 12, 250. Ford, E., 1934. J. Anat., Lond., 68, Part 3, 346. Gibbs, Helena F., 1938. Proc. Zool. Soc. Lond., 108B, 611. Jones, F. W., 1924. The Mammals of South Australia, Part II. Handbooks of the Flora and Fauna of South Australia. Issued by the British Science Guild (South Australian Branch). Govt. Printer, Adelaide. MacKenzie, W. C., 1934. Vict. Nat., 51, 58. Meijere, J. C. H. de, 1894. Morph. Jb., 21, 312. Schiefferdecker, P., 1917. Biol. Zbl., 37, 534. Schultz, A. H., 1921. J. Mammal., 2, 194. Schwarz, W., 1937. Morph. Jb., 79, 600. Souef, A. S. le, and Burrell, H., 1926. The Wild Animals of Australasia, Harrap and Co., Ltd., London. Troughton, E., 1941. Furred Animals of Australia, Angus and Robertson Ltd., Sydney. Wislocki, G. B., and Schultz, A. H., 1925. J. Mammal., 6, 236. THE STERNAL INTEGUMENT OF TRICHOSURUS VULPECULA. 133 EXPLANATION OF PLATES. PLATE IV. Fig. 1.—Sternal area in male pouch young of about 34 months of age. Note the grey hairs over this area. Fig. 2.—Sternal area of fully grown female, Trichosurus vulpecula. Fig. 3.—Sternal area of a fully grown male. Note the characteristic lateral expansion of the deeply coloured sternal hairs. Puate V. Fig. 4.—Dorsal skin region of male adult possum. Section cut parallel to the skin surface and showing two groups of hair follicles. Note the sebaceous tissue surrounding each lateral cluster of follicles, and the two sudoriferous ducts, linked to the central follicles by connective tissue traversing the middle of the group. The fibres are not pigmented. x 150. Fig. 5.—Sternal skin region of male adult possum. Section cut parallel to the skin surface and showing one follicle group. Compare with Fig. 4 and note the large and actively secreting sebaceous glands, the large sudoriferous duct, and the small number of fibres, most of which are pigmented. x 150. PiatTe VI. Fig. 6.—Sternal skin region of male adult possum at the level of the sudoriferous glands. Section cut parallel to skin surface and showing two portions of the secretory tube of a sudoriferous gland lying beside a group of hair follicles. The secretory epithelium varies in height on different parts of the gland wall, and a mass of crystalline material is seen in the lumen. x 150. Fig. 7.—Section through the secretory portion of a sudoriferous gland in the sternal region of an adult female possum. Note the high secretory epithelium with processes projecting into the lumen, and the granular secretory material within the lumen. x 150. PLATE VII. Fig. 8.—Crystals in the lumen of a sudoriferous gland in the sternal region of an adult male possum. Section cut parallel to skin surface. The rosette of rod-shaped crystals is in the lumen of the secretory portion of a sudoriferous gland. At the upper right the transition from the secretory portion of the gland to the duct is seen. In the upper left a pigmented hair fibre is cut obliquely. x 500. Fig. 9.—Crystals in the duct of a sudoriferous gland in the sternal region of an adult male possum. Section cut parallel to skin surface. Note the layers of cells forming the wall of the duct, and the two clumps of crystals in the lumen. Above is part of a sebaceous gland, and to the left is portion of a hair follicle. 500. Fig. 10.—Crystals in the wall of a sudoriferous gland in the sternal region of an adult male possum. Section cut parallel to skin surface. Note the crystalline rods mainly in the connective tissue sheath of the gland, and the blood capillary in the lower centre of the photograph. x 500. STUDIES IN THE PHENANTHRIDINE SERIES. PART I. THE CYCLISATION OF 2-FORMAMIDO-DIPHENYLS. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. The chemistry of phenanthridine has probably been studied less than that of any of the simple ring systems. The lack of interest in the past was due no doubt to the comparative inaccessibility of phenanthridine derivatives and to the fact that no natural products derived from phenanthridine were known. In recent years, however, interest in this series has been stimulated by the discovery that the lycorine and the chelidonine-sanguinarine alkaloids are derivatives of phenanthridine and by the possibility of obtaining therapeutic agents. A number of phenanthridine syntheses are known, but most of them use high temperatures or other drastic conditions, which greatly restricts their usefulness. The most attractive synthesis is that of Morgan and Walls, who found that the acyl derivatives of 2-amino-diphenyls are cyclised to phenan- thridines by boiling with phosphorus oxychloride. These starting materials have lately been rendered much more accessible by improvements effected in the Gomberg reaction and by the commercial availability of certain diphenyl derivatives. Apart from the publications of Morgan and Walls there is no information in the literature on the scope and applicability of this reaction or on its mechanism. Certain aspects of these questions have been studied in the work which follows, in the description of which all ring systems are named and numbered according to ‘“‘ The Ring Index ”’ (A. M. Patterson and L. T. Capell, 1940, the Reinhold Publishing Corporation, New York). Ring Index numbers (R.I. No.) are also quoted for the less common systems. By fusion with anhydrous zinc chloride at 250-300°, Pictet and Hubert (1896) cyclised acyl-o-amino-diphenyls to the corresponding phenanthridines. Because of the high temperature used, extensive charring occurred, purification was tedious and wasteful, and the yields were low. Moreover the reaction could not be applied to substances containing reactive groups such as nitroxyl. Later Morgan and Walls (1931) found that these defects were largely eliminated by using boiling phosphorus oxychloride to effect the cyclisation, and now this reaction is the most generally useful method for the synthesis of phenanthridines. Although 2-formamido-diphenyl is converted by zine chloride to phenan- thridine, the cyclisation cannot be accomplished by phosphorus oxychloride. This unexpected result is unfortunate, because it would appear to render phenanthridines unsubstituted in the 6 position not directly accessible by this reaction. In the well-known Bischler-Napieralski reaction, which is very similar to the Morgan-Walls reaction, formyl-B-phenylethylamine gives a small yield of dihydro-isoquinoline (Decker, Kropp, Hoyer and Becker, 1913) and Bamberger and Goldschmidt (1894) also found that isoquinoline was obtained in low yield by the action of phosphorus pentoxide on cinnamaldoxime, which presumably first underwent a Beckmann transformation to w-formamido- STUDIES IN THE PHENANTHRIDINE SERIES. 135 CHO CHO CHO CHO NO, CHO ; | CH,O CHO CHO CHO CHO | T T CHO CHO CHO 3 3 3 CHO CHO CHO CHO | NO, NH, NHCOR NO, CHO CHO CHO CHO CHO CHO CHO CHO 3 3 3 1V V vl vil CHO CHO CHO CHO CHO CHO 3 3 3 a, ! : | C-CH(CHOH) C-COOH CH a CHO CHO CHO CHO CHO CHO =) 3 3 Vill 1X x M—October 4, 1944, 136 E. RITCHIE. styrene. However, when a strongly para directing group is para to the position of ring closure, cyclisation occurs easily (Callow, Gulland and Haworth, 1929) and dihydro-isoquinolines are obtained in good yields from such substituted 6-phenylethylamines (e.g. Richardson, Robinson and Seijo, 1937). In this connection para methoxyl and 3:4 dimethoxyl are particularly effective. This suggested that it would be of interest to apply the Morgan-Walls reaction to a 2-formamido-diphenyl appropriately substituted by such an activating group, since it could be safely concluded that if cyclisation did not occur or if a low yield were obtained, then cyclisation would not take place to any useful extent in other cases. A substance suitable for this purpose was synthesised through the inter- mediates shown in the chart. 4-Iodo-veratrole (I), which was readily obtained by a modification of the method of Seer and Karl (1913), was converted to 3:3':4: 4'-tetramethoxy-diphenyl (Il) by heating with copper powder (Spath and Gibian, 1930). When nitrated under mild conditions, this yielded a mono- nitro compound (IIT), the constitution of which was proved by further nitration to the known 2: 2’-dinitro-4 : 4’: 5 : 5’-tetra-methoxy-diphenyl (IV) (Hughes, Lions, Maunsell and Wright, 1937). This substance was also directly obtainable by dinitration. By catalytic reduction of the mono-nitro compound using Raney nickel, an almost quantitative yield of the corresponding amine (V) was obtained, which was readily characterised by its piperonylidene and salicylidene derivatives. The formyl, acetyl, propionyl and benzoyl derivatives of the base (VI, R,=H, R,=CH,;, Rz;,=C,H;, R,=C,H;, respectively) were prepared by the usual methods and then subjected to the action of boiling phosphorus oxy- chloride. As anticipated, the three latter substances were rapidly and almost quantitatively cyclised to the corresponding phenanthridines (VII, R,—CH,, R,=—C,H,;, R,=C,H,; respectively). The formyl derivative, however, was converted to an intractable tar and a similar result was obtained when the reaction was carried out in boiling toluene. At lower temperatures either a tar formed or there was no reaction. When the cyclisation was attempted by phosphorus pentoxide in boiling xylene the results were more hopeful and there was eventually obtained a product from which a pure crystalline methiodide could be isolated in very low yield. This substance analysed correctly for a tetramethoxy-phenanthridine methiodide and behaved as expected, but to establish its identity with certainty it was essential to make a direct comparison with an authentic specimen. The synthesis of the required 2:3: 8: 9-tetramethoxy-phenanthridine methiodide was achieved by the degradation method of Walls (1934). By condensation with formaldehyde, the methyl-tetramethoxy-phenanthridine (VII, R,=CH;) yielded 6-(86’-dihydroxy-isopropy])-2 : 3:8 : 9-tetramethoxy- phenanthridine (VIII), which on oxidation afforded the carboxylic acid (IX). On heating, this decarboxylated smoothly to 2:3: 8: 9-tetramethoxy-phenan- thridine (VII, R, =H), which readily formed a methiodide (X). This substance proved to be identical with that obtained from the cyclisation of the formyl derivative (VI, R,=H). Hence it must be concluded that even in the most favourable case ring closure occurs to a very small extent only and the cyclisation of 2-formamido-diphenyls is not a reaction of preparative value. The phenanthridine derivatives prepared above are all colourless crystalline substances which show a marked blue fluorescence in neutral solution. In glacial acetic acid the fluorescence is stronger, but the yellow salts with mineral acids are too insoluble in water or alcohol to show this phenomenon. However, the yellow methiodides are strongly fluorescent in aqueous-alcoholic solution. STUDIES IN THE PHENANTHRIDINE SERIES. 137 EXPERIMENTAL. 4-Todoveratrole (I). A solution of veratrole (56 g.) in alcohol (150 c.c.) was vigorously stirred and treated. alternately with iodine (100 g.) and mercuric oxide (60 g.) during three hours. After another hour the reaction mixture was filtered and the alcohol distilled from the filtrate. The residue was dissolved in ether and the ethereal solution washed successively with sodium thiosulphate solution, dilute sodium hydroxide solution and water. After drying, the ether was removed and the residue distilled under reduced pressure. The product (60 g.) boiled at 163—4°/26 mm. The forerun (21 g.) consisting of veratrole and iodoveratrole, was suitable for another batch. 3:3':4:4'-Tetramethoxy-diphenyl (II). The method of Spaéth and Gibian was modified as follows: A mixture of 4-iodoveratrole (25 g.) and copper powder (‘‘ Naturkupfer C’’, 25 g.) was heated in a stream of carbon dioxide, with intermittent stirring, in an oil bath. When the temperature of the bath reached about 235° a vigorous reaction set in. Heating was interrupted until this had abated and then continued for an hour longer. After cooling, the reaction mixture was powdered and extracted with methyl] alcohol. Evaporation of the extract to a small bulk yielded the nearly pure product (10 g.), which, recrystallised from the same solvent, had m.p. 133°, as reported by Spath and Gibian. 2-Nitro-4: 4’: 5: 5’-tetramethoxy-diphenyl (ITT). A solution of tetramethoxydipheny! (5-5 g.) in glacial acetic acid (60 c.c.) at room temper- ature was stirred and treated with nitric acid (1:2 ¢.c., sp. gr. 1-4) in glacial acetic acid (3 c.c.) during fifteen minutes. After an additional thirty minutes, the reaction mixture was warmed on the water bath for fifteen minutes, cooled and diluted with ice water. The precipitate was collected, washed, dried and recrystallised from alcohol from which it (5 g.) separated as yellow rhombs, m.p. 149°. Foend = C, 60-1; H, 5:2; N, 4:6%: Caleulated for C,,H,,0,N: C, 60-2; H, 5:3; N, 4:4%. 2: 2'’-Dinitro-4: 4’: 5: 5’-tetramethoxy-diphenyl (IV). This substance was prepared from both tetramethoxydiphenyl and nitro-tetramethoxy- diphenyl by the same method. A solution (1-5 g.) in glacial acetic acid (15 c.c.) was treated with. nitric acid (1-5 c.c., sp. gr. 1-4) at room temperature. The product soon began to crystallise. After ten minutes, water was added and the product collected. It crystallised from alcohol in. fine yellow needles, m.p. 218°, identical with an authentic specimen (Hughes e¢ al., loc. cit.). 2-Amino-4 : 4’: 5: 5’-tetramethoxy-diphenyl (V). When a solution of the mono-nitro compound (III, 5 g.) in hot alcohol (200 c.c.) was shaken at normal pressure in an atmosphere of hydrogen in the presence of Raney nickel, slightly more than the theoretical amount of hydrogen was absorbed. After filtering off the catalyst, the alcohol was removed and the residue crystallised from benzene, from which it separated in nearly theoretical yield as colourless prisms melting at 151°. Found: C, 66:6; H, 6:8%. Calculated for C,,H,,0O,N: C, 66-4; H, 6-6%. The piperonylidene derivative prepared in alcohol from the base and piperonal as usual, separated from this solvent as yellow plates, m.p. 155°. Found: C, 68:3; H, 5:9%. Calculated for C,,H,,0,N: C, 68-4; H, 5-7%. The salicylidene derivative formed orange needles from alcohol, m.p. 144°. Peon 3) ©. 69-9; H; 5-895) Calculated for C,,H,,0,N: C, 70:2; H; 6:0%. 2-Acetamido-4 : 4’: 5 : 5’-tetramethoxy-diphenyl (VI, R,=CH,). The base was readily acetylated by warming on the water bath for a short time with excess. acetic anhydride. The product crystallised from aqueous alcohol as small colourless prismatic: needles melting at 164°. Found :'°C, 65:1; H, 6-2%. Caleulated for C,,H,,0;N: C, 65:3;-H, 6-3%. 138 E. RITCHIE. 2-Propionamido-4 : 4’: 5: 5’-tetramethoxy-diphenyl (VI, R,=C,H;). This substance, prepared in a similar fashion from propionic anhydride, separated from aqueous alcohol as small colourless plates, m.p. 138°. Found: C, 65:7; H, 6:9%. Calculated for C,,H,,0;N: C, 66-1; H, 6-6%. 2-Benzamido-4 : 4’: 5; 5'-tetramethoxy-diphenyl (VI, R,=C,H;). A solution of the base (5-8 g.) in dry pyridine (10 c.c.) was treated with benzoyl! chloride (2-8 g.) and heated on the water bath for one hour. After cooling, the reaction mixture was poured into dilute hydrochloric acid, and the product which soon solidified, collected, washed and recrystallised from aqueous alcohol. It separated in 85% yield as colourless glistening needles, m.p. 128°. Found: C, 69-8; H, 6-0%. Calculated for C,,H,,0;N : C, 70-2; Hy 6-0%: Cyelisation of the Amides (VI, R,=CH,, R,=C.H,;, R,=C,H;). The following procedure was used in each case: The amide (4 g.) was warmed with freshly distilled phosphorus oxychloride (10 c.c.) under reflux in a dry atmosphere. Almost immediately the solid turned bright yellow and clouds of hydrogen chloride were evolved. These soon diminished, but gentle refluxing was continued for one hour longer. Excess phosphorus oxy- chloride was removed under reduced pressure and the residue was then warmed with dilute sodium hydroxide solution. The liberated base was collected, washed, dried and recrystallised from benzene. 6-Methyl-2 : 3:8: 9-tetramethoxy-phenanthridine (VII, R,=CH,). This substance (85% yield) crystallised as colourless rods, m.p. 212°. Found: C, 68-8; H, 6:3%. Calculated for C,,H,,O,N: C, 69-0; H, 6-1%: The methiodide, obtained by heating the base with excess methyl iodide in a sealed tube at 100° for two hours, separated from water to which a little alcohol had been added, in yellow needles, which begin to decompose at about 260° but do not melt completely until 284°. Found: C, 49:8; H, 4:9%. Calculated for C,,H,,0,NI]: C, 50-1; H, 5:0%. 6-Ethyl-2 : 3: 8: 9-tetramethoxy-phenanthridine (VII, R,=C,.H;). The pure compound (85% yield) formed colourless prisms melting at 202°. Found: C, 69:9; H, 6-4%. Calculated for C,,H,,O,N: C, 70:0%; H, 6:4%. The methiodide prepared as above, crystallised from very dilute alcohol in yellow needles, melting and decomposing at 286°. Found: C, 51:3; H, 5-2%. Calculated for C,,H,,0,Nl: C, 51-2; H, 5°1%. 6-Phenyl-2 : 3: 8: 9-tetramethoxy-phenanthridine (VII, R,=C,H;). The pure substance (90% yield) crystallised from benzene as small colourless blades, m.p. 20K. Found: C, 74:0; H, 5-7%. Calculated for C,,H,,0,N: C, 73:6; H, 5:6%. The methiodide, prepared as usual, separated from dilute alcohol in yellow needles melting with decomposition at 273°. Found: C, 55:7; H, 4:8%. Calculated for C,,H,,0,NI1: C, 55-7; H, 4-6%. 6-(BB’-dihydroxy-isopropyl)-2 : 3: 8 : 9-tetramethoxy-phenanthridine (VIII). A solution of methyl-tetramethoxy-phenanthridine (VII, Rz=CH;; 10g.) in alcohol (50 c.c.) and formalin (20 c.c. 40%) was refluxed for three hours and then more formalin (20 c.c.) added. STUDIES IN THE PHENANTHRIDINE SERIES. 139 After heating for another ten hours an additional amount of formalin (10 e.c.) was added and refluxing continued for a further three hours. The syrup, obtained by evaporation, was taken up in water, made strongly ammoniacal, boiled and cooled. The solid which separated was collected, and recrystallised successively from benzene and benzene-aleohol (1:1) mixture. It (6-5 g.) separated as minute colourless prisms which melted at 214°. Found: C, 64:5; H, 6-3%. Calculated for C,,H,,0,N: C, 64-3; H, 6°2%. 2:3:8: 9-Tetramethoxy-phenanthridine-6-carboxylic acid (IX). A solution of the glycol (VIII, 4-5 g.) in water (50 c.c.) and concentrated sulphuric acid (1-2 c.c.) was heated on the water bath and stirred whilst it was treated during thirty minutes with a solution of potassium dichromate (7-5 g.) in water (35 c.c.) and concentrated sulphuric acid (5-6 c.c.). Oxidation proceeded gradually and after heating and stirring for an additional three hours, the reaction mixture was cooled and diluted. The crude acid which separated was purified by solution in hot dilute sodium hydroxide and reprecipitation by acetic acid. The colourless product (3-9 g.) then decomposed with evolution of carbon dioxide at about 240°. It was not purified further. but was used directly in the next step. 2:3:8: 9-Tetramethoxy-phenanthridine (VIII, R,=H). When maintained at 245°, the acid was rapidly decarboxylated. The dark residue was ‘taken up in alcohol, charcoaled, filtered and the sparingly soluble hydrochloride precipitated by the addition of a little concentrated hydrochloric acid. The regenerated base was extracted with hot benzene and the extract evaporated until crystallisation commenced. After cooling, the product (2-5 g.) was collected and recrystallised from methyl] alcohol, separating as colourless needles which partially melted at 135°, resolidified and then melted at 185°. This behaviour is due to solvent of crystallisation, which is very tenaciously retained. Found: C, 67:0; H, 6-0%. Calculated for C,,H,,0,N .}(CH,OH): C, 66-5; H, 6-0%. The methiodide (X), prepared as usual, crystallised from aqueous alcohol in small pale yellow needles which darkened slightly at 265° and melted at 295°. Found: C, 48-5; H, 4:4%. Calculated for C,,H.,O,NI: C, 49-0; H, 4-5%. 2-Formamido-4 : 4’: 5: 5’-tetramethoxy-diphenyl (VI, R,==H). A solution of the amine (V, 2 g.) in glacial formic acid (15 c.c.) was refluxed for three hours, then cooled and diluted. The precipitated product was recrystallised from aqueous alcohol from which it (1-7 g.) separated as small colourless diamond-shaped plates m.p. 168°. Found: C, 64-2; H, 6-1%. Calculated for C,,H,,O;N: C, 64-4; H, 6-1%. Cyclisation of (VI, R,=H). A solution of the formy! derivative (7-5 g.) in dry xylene (80 c.c.) was refluxed for three hours with phosphorus pentoxide (10 g.) which gradually became yellow. After cooling a little, the xylene was decanted and the residue washed in turn with xylene and ether. The bases, liberated by the addition of dilute ammonia, were extracted with benzene and the extract evaporated to dryness. The residue was dissolved in alcohol and then treated with a little concentrated hydrochloric acid, which precipitated a mixture of sparingly soluble hydrochlorides. The regenerated bases, isolated with the aid of benzene, were methylated by heating with excess methyl iodide at 100° in a sealed tube for one hour. The product was washed with hot alcohol, then with hot water and finally recrystallised several times from aqueous alcohol, being much more soluble in this mixture than in either of its components. The substance (0-3 g.) thus obtained (X) formed small pale yellow needles, identical with that obtained above, as proved by identity of crystalline habits, solubilities, melting and mixed melting points. Found: C, 48:6: H, 4:7%. Calculated for C,;H..0,NI: C, 49:0; H, 4:5%. 140 E. RITCHIE. REFERENCES. Bamberger, E., and Goldschmidt, C., 1894. Ber. dtsch. chem. Ges., 27, 1954. Callow, R. K., Gulland, J. M., and Haworth, R. D., 1929. J. chem. Soc., 1444. Decker, H., Kropp, W., Hoyer, H., and Becker, P., 1913. Liebig’s Ann., 395, 299. Hughes, G. K., Lions, F., Maunsell, J., and Wright, L. E. A., 1937. J. Roy. Soc. N. S. Wales, 71, 428. Morgan, G. T., and Walls, L. P., 1931. J. chem. Soc., 2447. Pictet, A., and Hubert, A., 1896. Ber. dtsch. chem. Ges., 29, 1182. Richardson, T., Robinson, R., and Seijo, E., 1937. J. chem. Soc., 835. Seer, C., and Karl, E., 1913. Mh. Chem., 34, 647. Spath, E., and Gibian, K., 1930. S.B. Akad. Wiss. Wien., 139, 2b, 234. Walls, L. P., 1934. J. chem. Soc., 104. STUDIES IN THE PHENANTHRIDINE SERIES. Part II. THE CYCLISATION OF SOME 4’-BROMO AND 4’-DIMETHYLAMINO ACYL-2-AMINODIPHENYLS. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. There is no information in the literature on the cyclisation of halogen substituted acyl-6-phenylethylamines to dihydro-iso-quinolines or on the influence of a heteronuclear halogen on the formation of phenanthridines from acyl-2-aminodiphenyls. From considerations of the inductive and resonance effects exerted by halogens it was anticipated that if the halogen was meta or para to the position of ring closure, then cyclisation would be rather more difficult than in an unsubstituted acyl-2-amino-diphenyl, but not so difficult as in the case of a meta nitro group (Morgan and Walls, 1932). The case when bromine is meta to the position of ring closure was examined in this work. It has been shown (Morgan and Walls, 1932; Walls, 1935) that homo- nuclear substituents do not affect the ease of cyclisation, and consequently the acyl derivatives of the more readily accessible 2-amino-4 : 4’-dibromo-dipheny] (1) were studied rather than those 2-amino-4'-bromodiphenyl. The substance (1) was prepared by the bromination of diphenyl (Scholl and Neovius, 1911), followed by nitration to 2-nitro-4 : 4’-dibromo-diphenyl (Shaw and Turner, 1932) and reduction by stannous chloride in alcohol. The nitro compound was recovered unchanged when reduction by West’s method (1925) was attempted. The acetyl (II, R,—CH;) and benzoyl (II, R.=C,H;) derivatives were prepared by conventional methods. When the acetyl derivative was boiled with phosphorus oxychloride the reaction was slow, as evidenced by the rate of evolution of hydrogen chloride, and extensive resinification occurred. However, from the reaction mixture 3 : 8-dibromo-6-methyl-phenanthridine (III, R, =CH,) was isolated in about 7% yield, together with a neutral substance, not further studied in about 3% yield. When heated to 180° with phosphorus oxychloride in nitrobenzene solution, under the ‘‘ forcing ’’ conditions of Morgan and Walls (1939, 1940) resinification of (II, R,=—CH,) was rapid and complete. The benzoyl derivative, although eyclised slowly by boiling phosphorus oxychloride, was not resinified. After three hours, about 15% had cyclised and the remainder could be recovered. Under ‘‘ forcing ’’ conditions, however, cyclisation was complete in six hours and 3: 8-dibromo-6-phenyl-phenanthridine (III, R,=C,H;) was obtained in’ nearly theoretical yield. The cyclisation in this case is effected in less than half the time (15 hours) required for 4’-nitro-2-benzamido-diphenyl under the same conditions. These results are in general agreement with the predictions made above. A similar study was made of the acetyl (V, R,—CH,) and benzoyl (V, R,=C,H,) derivatives of 2-amino-tetramethyl-benzidine (IV). In these cases it was expected that cyclisation would be difficult to accomplish. These expecta- tions were more than fulfilled, since all attempts to cyclise these substances produced uninviting tars from which no crystalline products could be isolated. In the course of these experiments it was noticed that the acyl derivatives dissolved in pure nitrobenzene to deep red solutions, from which they could be 142 E. RITCHIE. recovered unchanged. The same behaviour was shown by the parent base (IV), which was used for further experiments in this connection. It dissolved to a bright yellow solution in nitromethane, but in all other solvents examined, including acetonitrile and benzyl cyanide, no colour was observed. The Br Br Br NH, NHCOR N Il C—R Br Br Br | il tI N(CH,), N(CH;), NHo NHCOR N(CH3), | N(CH3),, IV V phenomenon is not due to salt formation arising from traces of acids in the nitro compounds since the mineral acid salts of the base are colourless. The base readily formed a bright yellow picrate, and an addition compound with trinitro- toluene which crystallised from alcohol in beautiful black shining needles. The picrate is undoubtedly a true salt and the black substance a true molecular compound (Pfeiffer, 1927). All attempts to prepare addition compounds with o-, m- and p-nitro-toluenes and with m-dinitrobenzene were fruitless. According to Pfeiffer this is due to the formation in solution of coloured addition com- STUDIES IN THE PHENANTHRIDINE SERIES. 143 pounds, which are too unstable to exist in the solid state, but later work by Gibson and Loeffler (1940) indicates that in these cases, postulation of unstable addition compounds is unnecessary and that the formation of coloured solutions should be attributed to the reciprocal polarisations of solvent and solute molecules. The base (IV) condensed readily with salicylaldehyde to yield an orange Schiff’s base, which dissolved in alcohol to an orange-yellow solution, but in nitrobenzene to a red solution. From p-dimethylamino-benzaldehyde there was obtained a yellow derivative which showed a similar behaviour in these solvents. However, the Schiff’s bases derived from o-, m- and p-nitrobenzal- dehydes and from 3- and 5-nitro salicylaldehydes were coloured deep red, brownish red, purple, brownish red and red respectively, and gave deep red solutions in both alcohol and nitrobenzene. These results, although not con- clusive, would appear to support the conclusions of Gibson and Loeffler. EXPERIMENTAL. 2-Nitro-4 : 4’-dibromo-diphenyl. The crude product obtained by the method of Shaw and Turner (loc. cit.) was best purified by crystallisation from glacial acetic acid. 2-Amino-4 : 4’-dibromo-diphenyl (1). A hot solution of the nitro compound (47 g.) in absolute alcohol (1,300 c.c.) was treated with a solution of stannous chloride (95 g.) in absolute alcohol (100 c.c.) and the mixture refluxed for twelve hours. After distilling off the alcohol the cooled residue was shaken with ether and dilute sodium hydroxide solution (1,000 c.c. of 10%). The ethereal extract was washed, dried and then the ether removed. The residue which crystallised spontaneously, was purified by recrystallisation from methanol yielding 27 g. Another recrystallisation from the same solvent gave the pure substance as colourless needles which melted at 132°. Found: C, 44-3; H, 2-8%. Calculated for C,,H,NBr,: C, 44-0; H, 2-8%. The substance was very sparingly soluble in light petroleum, moderately in methanol and easily soluble in other organic solvents. 2-Acetamido-4 : 4’-dibromo-diphenyl (II, R,=CH,). Acetylation with acetic anhydride on the water bath in the usual manner afforded the acety] derivative which crystallised from alcohol in colourless needles, m.p. 192°. Found: C, 45-3; H, 3:0%. Calculated for C,,H,,ONBr,: C, 45-5; H, 3-0%. 2-Benzamido-4 : 4’-dibromo-diphenyl (II, R,==C,H;). A solution of the amine (I, 9-8 g.) in dry pyridine (15 c.c.) was treated with benzoyl] chloride (4-2 g.). After warming on the water bath for an hour in a loosely stoppered flask, the mixture was poured into dilute hydrochloric acid. On rubbing, the oil which separated was converted to a solid, which crystallised from alcohol as colourless needles (12 g.) which melted at 176°. Found: C, 52-4; H, 3-1%. Calculated for C,,H,,ONBr,: C, 52:9; H, 3-0%. Cyclisation of the Acetyl Derivative (II, R,=CH,). When a mixture of the acetyl derivative (7 g.) and phosphorus oxychloride (10 c.c.) was gently refluxed, hydrogen chloride was slowly evolved. After three hours this had almost ceased and the reaction mixture was dark reddish brown. It was cooled, basified with ammonia and extracted with benzene. After removing the benzene, the residue was dissolved in boiling alcohol (200 ¢.c.), charcoaled and filtered. On evaporation, the filtrate gradually deposited crystals A. which were collected from the hot solution when the volume had been reduced to about 30 c.c. 144 E. RITCHIE. The hot filtrate was treated with alcoholic picric acid, and the crude picrate which separated on cooling and rubbing was collected and washed with alcohol. The regenerated base was taken up in benzene, and the solution washed, dried, and then evaporated to a small bulk. The crystalline mass which separated on cooling was collected, washed with a little benzene and recrystallised several times from ethyl acetate with the aid of charcoal. Eventually pure 3 : 8- dibromo-6-methyl-phenanthridine (III), (R,=CH,; 0-5 g.) was obtained as small colourless needles which melted at 186°. Found: C, 48-1; H, 2-6%. Calculated for C,,H,NBr,: C, 47:8; H, 2°5%. The substance was easily soluble in benzene, less so in ethyl acetate, and only slightly in ethyl alcohol. The picrate separated as small yellow needles when hot alcoholic solutions of its components were mixed. It decomposed at 235°. Found: N, 9-9%. Calculated for C,,.H,,O,N,Br,: N, 9-7%. The substance A after several recrystallisations from alcohol-benzene formed small colourless needles (0-2 g.), which melted and decomposed at 244°. Beyond noting that it was non-basic and did not form a picrate, it was not further examined. 3: 8-Dibromo-6-phenyl-phenanthridine (III, R,=C,H;). A mixture of the benzoyl] derivative (II, R,—C,H,; 8 g.) nitrobenzene (20 c.c.) and phos- phorus oxychloride (10 ¢.c.) was heated in an oil bath at 180° under reflux. After six hours the slow evolution of hydrogen chloride had ceased. The mixture was cooled, made strongly alkaline with dilute ammonia, and the nitrobenzene removed by steam distillation. The nearly pure product (7-5 g.) which had separated in light brown needles during the course of the steam distillation was collected and dried. Recrystallisation from benzene afforded the pure substance as long colourless needles which melted at 201°. Found: C, 54:8; H, 2:7%. Calculated for C,,H,,NBr,: C, 55:2; H, 2-7%. It was fairly readily soluble in hot benzene, much less in cold benzene, and only slightly soluble in hot alcohol. Like (III, R,=CH,), it formed a yellow insoluble salt with concentrated hydrochloric acid, which was completely hydrolysed by the addition of water. The picrate prepared in alcohol-benzene solution, separated as. bright yellow needles which melted at 223-4° with decomposition. Found: N, 8:6%. Calculated for C,,H,,0O,N,Br,: N, 8°8%. 2-Nitro-tetramethyl-benzidine. Tetramethyl-benzidine (Ullmann and Dieterle, 1904) was nitrated according to Bell and Kenyon (1926). The product crystallised from ethyl acetate containing a little benzene in beautiful deep red rhombs which melted at 165° (lit. 164°). 2-Amino-tetramethyl-benzidine (IV). This substance was briefly described by Bell and Robinson (1927). The details of its prepara- tion are as follows : The powdered nitro compound (24 g.) was added during fifteen minutes to a stirred solution of stannous chloride (65 g.) in concentrated hydrochloric acid (100 c.c.), which had been warmed to 65°. It rapidly dissolved and the temperature rose to about 80°. The orange-red solution was then warmed on the water bath for fifteen minutes, cooled and poured into excess dilute sodium hydroxide solution. The precipitated product was extracted with benzene, the extract washed, dried and evaporated to a small bulk, when the nearly pure amine (20 g.) separated. One recrystallisation from alcohol yielded the pure substance as colourless needles which melted at 144°. Bell and Robinson (loc. cit.) report 138°. Found: C, 74-9; H, 8-2%. Calculated for C,,H.,N,: C, 75:3; H, 8-2%. STUDIES IN THE PHENANTHRIDINE SERIES. 145 The picrate separated as sparingly soluble yellow needles when alcoholic solutions of its components were mixed. It began to darken at about 150° and decomposed at 167°. Found: N, 17-:1%. Calculated for C,,H,,N,0O,: N, 17-3%. The trinitrotoluene addition compound prepared in a similar fashion, crystallised as beautiful black shining needles, which began to melt at 87°, but were not completely liquefied until 112°. Found: N, 17-5%. Calculated for C,,H,,O,N,: N, 17:4%. 2-Acetamido-tetramethyl-benzidine (V, R,=CHs). The base (IV) was warmed on the water bath with excess acetic anhydride for five minutes and then poured into dilute ammonia. The product crystallised from aqueous alcohol in slender colourless needles, m.p. 157°. Found: C, 72:4; H, 7-4%. Calculated for C,,H,,ON, : C, 72-1; H, 7-7%. 2-Benzamido-tetramethyl-benzidine (V, R,—C,H;). Benzoylation with benzoyl chloride in pyridine solution was carried out as usual. The product crystallised from ethyl] alcohol in faintly yellow prismatic needles which melted at 145°. Found: N, 11-:9%. Calculated for C,,H,,ON,: N, 11-7%. Condensation of the Base (IV) with Aldehydes. The various Schiff’s bases listed below were prepared from equimolecular proportions of the reactants in concentrated alcoholic solution as usual, and also recrystallised from alcohol. Salicylidene-(2-amino-tetramethyl-benzidine). Orange prisms, m.p. 119°. Alcoholic solution orange yellow, nitrobenzene solution red. Found: C, 77-1; H, 7-3%. Calculated for C,,H.,ON,: C, 77-3; H, 7-0%. p-Dimethylamino-benzylidene-(base). Yellow prisms, m.p. 140°. Alcoholic solution pale yellow, nitrobenzene solution red. Wound':€, 17-9; HH, 7-7%. -Caleulated for'C,,H,,N,: C, 77-7; H, 7-8%. o-Nitrobenzylidene-(base). Small deep red needles, m.p. 175°. Alcoholic solution red, nitrobenzene solution red. Found : C, 71 . 2 3 H, 6: Do Calculated for C.3H.,0.N, : C, 71 ; ] 3 H, 6- 2. m-Nitrobenzylidene-(base). Deep brownish red needles, m.p. 171°. Alcoholic solution red, nitrobenzene solution red. Meuner, ©, 11-3; H, 6-1%. Calculated for C,,H,.,0,N,: C, 71:1; H, 6-2%. p-Nitrobenzylidene-(base). Purple needles, m.p. 212°. Alcoholic solution red, nitrobenzene solution red. Found: C, 70-8; H, 6-3%. Calculated for C,,H,,O.N,: C, 71-1; H, 6-2%. 2-Hydroxy-3-nitro-benzylidene-(base). Small brownish red prisms, m.p. 168°. Alcoholic solution red, nitrobenzene solution red. Found: C, 68-2; H, 6:1%.. Calculated for C,,H,,0,N,: C, 68-3; H, 6-0%. 2-Hydrozy-5-nitro-benzylidene-(base). Red needles, m.p. 234°. Alcoholic solution red, nitrobenzene solution red. Found: C, 68-1; H, 5:9%. Calculated for C,,H,,0,N,: C, 68:3; H, 6:0%. 146 E. RITCHIE. REFERENCES. Bell, F., and Kenyon, J., 1926. J. chem. Soc., 2705. Bell, F., and Robinson, P. H., 1927. J. chem. Soc., 1695. Gibson, R. E., and Loeffler, O. H., 1940. J. Amer. chem. Soc., 62, 1324. Morgan, G. T., and Walls, L. P., 1932. J. chem. Soc., 2225. Morgan, G. T., and Walls, L. P., 1939. Brit. Pat., No. 511, 353. Morgan, G. T., and Walls, L. P., 1940. Brit. Pat., No. 520, 273. Pfeiffer, P., 1927. Organische Molekulverbindungen, Enke, Stuttgart. Scholl, R., and Neovius, W., 1911. Ber. dtsch. chem. Ges., 44, 1075. Shaw, F. R., and Turner, E. E., 1932. J. chem. Soc., 285. Ullmann, F., and Dieterle, P., 1904. Ber. dtsch. chem. Ges., 37, 23. Walls, L. P., 1935. J. chem. Soc., 1405. West, R. W., 1925. J. chem. Soc., 494. Sa STUDIES IN THE PHENANTHRIDINE SERIES. Part III]. THE INFLUENCE OF THE ACYL GROUP IN THE MORGAN- WALLS REACTION AND THE MECHANISM OF THE REACTION. By E. RITCHIE, MSc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. The cyclisation of acyl derivatives of 2-amino-diphenyls by the Morgan- Walls reaction depends not only upon the presence and nature of nuclear substituents but also upon the nature of the acyl group. On the evidence available at the time relating to the latter effect, Walls (1934) summarised the position when he stated that ‘‘ The course of the reaction must depend on the nature of R and of nuclear substituents, particularly in the heteronucleus. Although the reaction proceeds smoothly when R=CH,ClI it fails completely for R=CHCl, and CCl,, which suggests that as the negative character of R is increased, enolisation, the most probable preliminary to ring closure, is inhibited. The electronic character of R has however no consistent influence, for CH,Cl gives better results than CH, but COOC,H,; gives a very poor yield and CHCl, fails; o- exceeds m- and p- O,H,NO, and equals C,H;.’”’ In order to gain additional information on this effect and with the hope of ultimately providing an adequate explanation for it, the cyclisation of a number of other acyl derivatives of 2-amino-diphenyl has been examined. At the outset it was realised that it would not be possible to examine the o-xenylamides of all types of acids and so it was decided to extend the investigations on some of those types already begun by Walls. The results of these experiments will be given first and a discussion of them will follow. The observation that a poor yield was obtained when R=COOC,H, led to the preparation and study of the o-xenylamides of the lower dicarboxylic acid half esters. A number of methods were employed to prepare the individual members of this series. The first member (I, R=CH,COOC,H,) was obtained by refluxing the amine with excess malonic ester. On boiling with phosphorus oxychloride it was rapidly converted to a tar. The succinic acid derivative (I, K=(CH,),COOCH;) prepared by the action of 6-carbomethoxy-propionyl] chloride on the amine, was also resinified by the cyclising agent. The related acid (I, R=(CH,),COOH), formed by mixing benzene solutions of succinic anhydride and the amine, gave a similar result. The third member of the series was synthesised in several steps. Glutaric anhydride and o-xenylamine reacted readily to form N-(o-xenyl)-glutaramic acid (I, R=(CH,);COOH), which also failed to form a phenanthridine. On pyrolysis, this substance was dehydrated to N-(o-xenyl)-glutarimide (III) which was readily converted to 2-(y-carbo- methoxy-butyramido)-diphenyl (I, R=(CH,),COOCH,) by boiling with 1% sulphuric acid in anhydrous methyl alcohol. This substance was cyclised smoothly by phosphorus oxychloride, and the product, methyl y-(6-phenan- thridyl)-butyrate (II, R=(CH,),COOCH,) obtained in good yield. Hydrolysis by alcoholic potassium hydroxide in the usual manner afforded the cor- Tesponding acid. The last member of this series, which was examined (I, R=(CH,),COOC,H,) was prepared from 6-carbethoxy-valeryl chloride and 0-xenylamine. This also was cyclised in good yield by phosphorus oxychloride 148 E. RITCHIE. to the corresponding ethy] $-(6-phenanthridyl)-valerate (II, R =(CH,),COOC,H a) from which the related acid was obtained by the usual procedure. These results indicated that the study of the NN’ di-o-xenylamides of the dicarboxylic acids would also yield interesting information. The study of this NHCOR © Frm N N cH C—R ‘oe CHY /NHCO(CH,)C OHN N N ut C-(cH,)=C SOH Vi. series was begun in reverse order. Adipyl chloride and o-xenylamine gave a nearly theoretical yield of NN’ di-o-xenyl-adipamide (IV, n—4), which gave two isomeric products on boiling with phosphorus oxy chloride. The one formed in very small yield was weakly basic, fluoresced in acid solution, and formed a - STUDIES IN THE PHENANTHRIDINE SERIES. 149 picrate. It is therefore regarded as 1: 4-di-(6’-phenanthridyl)-butane (V). The other, a neutral substance obtained in nearly 50% yield, was not further investigated. The substituted glutaramide (IV, n=3), prepared from glutaryl chloride and o-xenylamine was slowly resinified by boiling with phosphorus oxychloride. After two hours approximately 50% of it was recoverable but no other product could be isolated. Further investigations in this series did not appear worth while. The o-xenylamides of the third series examined were of the w-phenyl-alkyl type, of which 2-benzamido-diphenyl may be regarded as the first member. The second member of the series (I, R=CH,C,H;) was obtained as a low melting solid by the action of phenyl-acetyl chloride on o-xenylamine in dry pyridine. When boiled with the cyclising agent it was converted to 6-benzyl-phenanthridine (II, R=CH,C,H;) in only 20% yield and no starting material could be recovered. A much better result was obtained from the next member of the series (I, R=(CH,),C,H;), which also was prepared from the base and the corres- ponding acid chloride. In this case cyclisation proceeded smoothly and 6-(6-phenyl-ethyl)-phenanthridine (II, R=(CH,),C,H,;) was isolated in nearly 70% yield. An almost equally good result was obtained from the somewhat similarly constituted 2-(phenoxy-acetamido)-diphenyl (I, R=CH,OC,H;), which gave 6-(phenoxy-methyl)-phenanthridine (II, R=CH,OC,H;) in 65% yield. It appeared to be of some importance to determine whether the principle of vinylogy (Fuson, 1935) held in this reaction. Accordingly 2-crotonamido- diphenyl (I, R=CH=—CHCH,), 2-cinnamamido-diphenyl (I, R=CH=CHC,H;) and N-(o-xenyl)-maleamic acid (I, R=CH=CHCOOH), the vinylogues of 2-acetamido-diphenyl, 2-benzamido-diphenyl and _ o-xenyloxamic acid, respectively, were prepared and then treated with phosphorus oxychloride. The first and last of these vinylogues were rapidly converted to tars from which no crystalline products could be isolated. But from the cyclisation of the second substance, 6-styryl-phenanthridine (II, R=CH=CHC,H,;) was obtained in about 12% yield. The cyclisation of 2-acetoacetamido-diphenyl (I, R=CH,COCHS), presented an interesting case since theoretically it could take place in two ways. Either 6-phenanthridyl-acetone could be formed by a Morgan-Walls reaction, or 2-hydroxy-4-methyl-8-phenyl-quinoline by a Knorr reaction. The amide was prepared in moderate yield by heating o-xenylamine with ethyl acetoacetate at 160° for a short time. When refluxed with phosphorus oxychloride it was rapidly converted to a tar, but cyclisation with concentrated sulphuric acid proceeded easily and cleanly. The product which was a sulphonic acid, gave no tests for a carbonyl group and therefore the cyclisation had taken the second course. It is probably 2-hydroxy-4-methyl-8-(phenyl-4’-sulphonic acid)-quinoline (V1). One other aspect of the influence of the acyl group was examined. It seemed possible that cyclisation would be prevented or retarded if the acyl residue was bulky, and a few experiments were therefore carried out to ascertain if this steric factor was operative. It was found that 2-(a-naphthamido)- diphenyl was cyclised as rapidly as 2-benzamido-diphenyl (as far as could be judged from qualitative experiments) and the product 6-(«-naphthyl)-phenan- thridine was isolated in 70% yield. In agreement with Pinck and Hilbert (1937), who had prepared it by another method, it was found to melt at 125°. This base also forms a methiodide without difficulty. It is noteworthy that the latter substance is dimorphous and both forms crystallise together from the same solution. On standing or more rapidly on rubbing, the yellow form changes to the stable orange form. The cyclisation of the mesitoyl derivative of o-xenylamine, which was very slowly formed when mesitoyl chloride (1 mol.) and o-xenylamine (2 mol.) were allowed to react in dry ether at room temperature, was then examined and found to proceed with equal facility. In this case also, 150 E. RITCHIE. the reaction with methyl iodide did not appear to be hindered. It must be concluded then that the size of the acyl group is not an important factor in the Morgan-Walls reaction. The yields given in the cyclisation experiments described above refer to the amount of product isolated in a reasonable state of purity (i.e. melting within 5° of the correct m.p.). In those cases where cyclisation is reported to have failed, it is unlikely that even very small yields would have escaped detection since the isolation of phenanthridine derivatives is usually facilitated by the slight solubility of their picrates and salts with mineral acids. THE MECHANISM OF THE REACTION. The suggestion by Walls (loc. cit.) that the first and in some eases critical step in the reaction is enolisation, raises the question of the structure of amides. Amides of the type under discussion are associated and according to recent work (reviewed by Hunter, 1941) this association is due to hydrogen bond resonance, the nitrogen atom of one molecule being united through a hydrogen atom to the oxygen atom of another molecule. An amide molecule then has neither an amide nor an imidol structure but may be regarded as a resonance hybrid of these two forms. The postulation of an enolisation step therefore appears to be unnecessary. In the mechanism now proposed, the first step is the formation of a carbonium ion under the influence of the strong acid, phosphorus oxychloride. The carbonium ion has the general formula [X —N—C—R]+ (where X represents o-xenyl) in which the carbon atom originally has a share in six electrons only. The formation of this ion possibly occurs through the substitution of a chloride ion or a phosphate ion for a hydroxyl ion, but like other carbonium ions it is probably never entirely free from the influence of the anion. In the next step the positive and hence electrophilic carbon atom attacks a carbon atom at the cyclisation position where the electron density is comparatively high. Simul- taneously a proton is released and immediately attaches itself to the lone pair of ‘electrons of the nitrogen atom, thus forming a phenanthridinium ion. According to this mechanism, cyclisation will depend on at least three factors : (a) the formation of the carbonium ion, (b) the stability of the carbonium ion, and (c) the electron density at the heteronuclear ortho carbon atoms. By a consideration of these three factors it should therefore be possible to account in a large measure for the experimental results. The slight effect of a homonuclear substituent is readily understandable since it would affect (a) and (b) very slightly and (c) hardly at all. A hetero- nuclear substituent, on the other hand, largely controls the electron density at the position of ring closure by its inductive and resonance effects. If it increases the electron density at this point it should therefore facilitate cyclisation, but if it decreases it, cyclisation should be hindered. Experimentally this is found to be the case; for example, when a methoxyl group is para, the cyclisation proceeds very easily (Part I), but when a nitro group is meta, ‘ forcing ’’ con- ditions are required (Morgan and Walls, 1932, 1939, 1940). It should be noted that when there is no substituent in the heteronucleus, the electron density at the strategic carbon atom is still high, because it is ortho to an electron releasing group, i.e. the homonucleus. Both the formation of the carbonium ion and its stability are affected by the nature of the acyl residue. If the carbonium ion cannot be formed, cyclisa- tion cannot occur, and if the ion is formed very slowly, cyclisation is slow. In these cases the amide may be recovered unchanged or it may undergo side STUDIES IN THE PHENANTHRIDINE SERIES. 151 reactions with the cyclising agent. Side reactions may also occur when the ion is easily formed but is unstable. The yield of phenanthridine then depends on the relative rates of cyclisation and side reaction, 1.e. it is determined by the stability of the ion. But if ion formation is easy and if it is stable, then cyclisa- tion should be the predominant reaction. In the following brief discussions of the individual cases, the course of the reaction to be expected is given first and then the experimental results are stated. These of course are qualitative only. 1. R=H. The ion is formed but simultaneously loses a proton, the overall effect being a dehydration, to an isonitrile, which then undergoes side reactions. No phenanthridine is formed (M. and W., 1931). 2. R=CH,. The ion is readily formed and is stabilised by an electron drift from the donor methyl group. The same remarks apply to homologues. Good yields are obtained (M. and W., 1931). 3. R=CH,Cl. The strongly electrophilic nature of the chlorine makes ion formation more difficult, but once formed it is stable. The reaction should be slower than in 2. The yield actually obtained is shghtly better than in 2 (M. and W., 1931), but this is probably not significant and may be due to greater ease of isolation of the product. . 4. R=CHCl,, CCl,. The accumulation of electrophilic chlorine atoms completely prevents ion formation and since the amides are stable, they should be recoverable. m The amides are recovered uncharged (W., 1934). 5. R=COOC,H;. This is similar to 4, but the carbethoxy group is not so strongly electrophilic. Ion formation is difficult and hence cyclisation is slow. A low yield is obtained but unchanged amide is recoverable (W., 1934). 6. R=CH,COOC,H,;,. The ior is formed readily but is unstable, immediately losing a proton from the reactive methylere group. The overall effect is dehydration to the azomethine X —N=C=CHCOOC,H;, which is then resinified by the cyclising agert. The amide is rapidly converted to a tar. 7. R=(CH,),;COOCH,. The case is probably similar to 6. The amide is resinified. 8. R=(CH,),COOCH,;, (CH.),COOC,H,;. The inductive effect of the ester group is not important at the y (or 5) carbor atom and the case is similar to 2. Good yields are obtained. 9. R=(CH,),COOH. Stable ions are not formed, but the carboxyl group is converted to COCI1 with the consequent formation of chain polymers. Similar remarks apply to the higher homologue. No phenanthridine is formed and no amide is recoverable. 10. R=CH,COCH;. This case is similar to 6. The amide is converted to a tar. | 11. R=C,H,;. Ion formation is easy and the ion is stabilised by resonance with the benzene nucleus. Good yields are obtained (M. and W., 1931). 12. R=o-C,H,NO,. This strongly electrophilic group should render ion formation very difficult and hence retard cyclisation. The good yield actually obtained (M. and W., 1931) may be due to a so- called ‘‘ ortho effect ’’ such as a high stabilisation of the carbonium ion by the positive carbon atom gaining a share of two of the electrons of a nitroxyl oxygen, which would in fact lie close to it. N—October 4, 1944. 152 E. RITCHIE. 13. R=p-C,H,NO,. The strong electron withdrawing action of this group is undisturbed by any “ ortho effect’’. Ion formation is therefore difficult, but the ion is stable. Cyclisation is slow, but unchanged amide is recoverable (M. and W., 1931). 14. R=m-C,H,NO,. The case is similar to 13, but since the group is not so electrophilic, ion formation is easier and cyclisation faster. The yield is good but not as good as in 11 (M. and W., 1931). 15. R=CH,C,H;. The case is similar to 6, but the ion is more stable. The yield should be small and no amide recoverable. The yield is small and no amide is recoverable. -16. R=(CH,),C,H;, CH,OC,H,;. Electron displacements are now slight and unimportant, hence the ion should be stable and cyclisation should be straightforward as in 2. Good yields are obtained. 17. Diamides (IV, n=3, 4). Good yields would be expected in both cases. The slow resinification of (IV, n=3) and the very poor yield obtained from (IV, n=4) cannot be accounted for. 18. R=CH=CHCH,. Ion formation would be easy but since the ion would have a conjugated structure, the positive charge would not reside on any one carbon atom. MHence‘cyclisation would be very slow and polymerisation, a most probable side reaction, would predominate. The amide is converted to a tar. 19. R=CH=CHCOOH. Ion formation would be difficult as in 5 and side reactions aS in 9 would occur. The amide is converted to a tar. 20. R=CH=CHO,H;. This case is similar to 18, but the ion is sufficiently stabilised by the benzene nucleus acting as an electron donor to allow the formation of a small yield of phenanthridine. The yield is small and no amide is recoverable. The mechanism suggested for the reaction thus accounts for the influence of homo- and heteronuclear substituents and the effects of most acyl groups can also be explained. It is realised, however, that only a study of the kinetics of the reaction can establish its mechanism with certainty. The Morgan-Walls reaction has many features in common with other eyclodehydrations brought about by strong acids and leading to the formation of heterocyclic compounds, such as the Bischler-Napieralski reaction, and the Combes quinoline synthesis. If the mechanism proposed above is the correct one, it will probably apply in its essentials to these reactions also. HXPERIMENTAL. Hihyl N-(o-xenyl)-malonamate, (I, R=CH,COOC,H;). A solution of o-xenylamine (15 g.) in ethyl malonate (70 g.) was gently refluxed for eight hours, during which time alcohol was slowly evolved. After removing unchanged reactants by steam distillation, the oily residue was taken up in ether, washed with dilute acid, then water and dried. The thick light brown oil (20 g.) left after removing the ether could not be induced to solidify. . When warmed with phosphorus oxychloride, there was a copious evolution of hydrogen chloride and the reaction mixture rapidly darkened, but no crystalline substance could be isolated. Methyl N-(o-xenyl)-succinamate, (I, R=-(CH,),COOCH)). A solution of o-xenylamine (26 g.) in dry ether (500 c.c.) was stirred and treated rapidly with a solution of 8-carbomethoxy-propionyl chloride (11-5 g.) in dry ether (100 c.c.). The amine- hydrochloride soon began to separate and after thirty minutes it was collected and washed well STUDIES IN THE PHENANTHRIDINE SERIES. 153 with ether. The combined filtrates were washed successively with water, dilute acid and water and then dried. Removal of the ether left a colourless oil (22 g.) which solidified completely on cooling. The substance was easily soluble in the usual organic solvents, but only slightly soluble in light petroleum, and was best recrystallized from ether light petroleum mixture, from which it separated as long colourless needles m.p. 73°. Found: C, 72:0; H, 5:9%. Calculated for C,,H,,0O,N: C, 72-1; H, 6-0%. The substance was rapidly resinified when boiled with phosphorus oxychloride. N-(o-Xenyl)-succinamic Acid, (1, R=(CH,),COOH). Succinie anhydride (5 g.) dissolved in hot dry benzene (75 c.c.) was added in one lot to a solution of o-xenylamine (8-4 g.) in the same solvent (25 ¢c.c.). The product began to crystallize almost immediately but refluxing was continued for one hour. After cooling it was collected and washed with benzene. The pure substance (12-4 g.) crystallised from aqueous alcohol in colour- less needles melting at 126°. Found: C, 71:2; H, 5-6%. Calculated for C,,H,;0,N: C, 71-4; H, 5-6%. Hydrogen chloride was evolved, when the substance was boiled with the cyclising agent, but no phenanthridine derivative could be isolated from the reaction mixture. N-(o-Xenyl)-glutaramic Acid, (I, R=(CH,),COOH). A solution of glutaric anhydride (25 g.) in hot dry benzene (500 c.c.) was treated with a solution of o-xenylamine (37 g.) in the same solvent (100 ¢.c.) and the mixture boiled for one hour. On cooling, the product which had separated in nearly quantitative yield was collected, washed with benzene and recrystallised from ethyl acetate. It formed colourless needles which melted at 137°. Wound: C, 72:2; H,6-0%. Calculated for C,,H,,O,N: C, 72-1; H, 6-0%. It was easily soluble in alcohol, but very slightly soluble in benzene. Boiling with phosphorus oxychloride converted it to a tar. N-(0-Xenyl)-glutarumde (III). The above acid (25 g.) was melted and then refluxed gently over a free flame for ten minutes. Water was evolved and a little of the acid sublimed. The dark gum obtained on cooling was readily rendered crystalline by rubbing with solvents. The product (15 g.) was finally obtained colourless by several recrystallisations from alcohol with the aid of charcoal. By rapid crystal- lisation small prisms are formed, but by slow crystallisation large irregular rectangular plates showing the hour glass structure are obtained. It melted at 158°. Found: C, 76:2; H, 5-6%. Calculated for C,,H,,O,.N: C, 76:4; H, 5:6%. The substance was easily soluble in benzene and in hot alcohol, but slightly soluble in cold alcohol. Methyl N-(o-xenyl)-glutaramate, (I, R=(CH,),COOCH,). A solution of the imide (III; 13 g.) in dry methyl alcohol (100 ¢c.c.) containing concentrated sulphuric acid (0:8 g.; i.e. 1%) was refluxed for three hours. Then solid sodium carbonate slightly in excess of the theoretical amount required to neutralise the acid, was added and the mixture evaporated to a small bulk. After the addition of water, the product was extracted with ether, the extract washed thoroughly with dilute sodium carbonate solution and water and then dried. Removal of the ether left a residue (11 g.) which rapidly solidified. The pure substance obtained by recrystallisation from acetone light petroleum mixture formed long colourless needles melting at 85°. Wounds: C,: 72-1; H, 6-4%.. Caleulated for. O,,.H,.O,N: C, 72:7; H, 6:4%. It was easily soluble in the usual organic solvents excepting light petroleum. Methyl y-(6-phenanthridyl)-butyrate, (II, R==(CH,),COOCH,). The above half-ester amide (6 g.) and boiling phosphorus oxychloride (12 ¢.c.) reacted vigorously. After refluxing gently for one hour, excess cyclising agent was distilled off under 154 , __ E. RITCHIE. reduced pressure, the residue dissolved in alcohol and the solution poured into excess dilute ammonia. The precipitated oil was extracted with ether, the solution dried, charcoaled, and evaporated. The residual oil solidified on rubbing and the solid obtained was recrystallised from light petroleum. The pure substance (3-6 g.) separated as colourless rods melting at 71°. Found: C, 77:1; H, 6-0%.. Calculated for C,,H,,O,.N:C; 77>4:5.Hp Geta It was easily soluble in the usual organic solvents excepting light petroleum. Its solutions in glacial acetic acid and dilute sulphuric acid showed a blue fluorescence. The picrate crystallised in sparingly soluble microscopic yellow plates, m.p. 198° with decomposition, when hot alcoholic solutions of its components were mixed. Found: N, 11-1%. Calculated for C,,H,,O,N,: N, 11-0%. y-(6-Phenanthridyl)-butyric Acid, (II, R==(CH,),COOH). Hydrolysis of the above ester with alcoholic potassium hydroxide in the usual manner gave the acid. It crystallised from alcohol in rosettes of slender colourless needles melting at 158°. Found: C, 76-7; H, 5-7%. Calculated for C,,H,,O.N: C, 76-4; H, 5:6%. Solutions of the substance in glacial acetic acid, dilute sulphuric acid, and in water show a blue fluorescence, but solutions in alcohol, benzene and dilute alkali do not. The fluorescence is obviously connected with phenanthridinium-salt formation, and this must oceur also to some extent in aqueous solution. Kthyl N-(o-xenyl)-adipamate, (I, R=(CH,),COOC,H ;). This substance was prepared in dry ethereal solution from $-carbethoxy-valeryl chloride (16 g.) and o-xenylamine (28-2 g.) by the usual method. It was isolated in nearly theoretical yield as a colourless syrup which could not be induced to crystallise. Ethyl 8-(6-phenanthridyl)-valerate, (II, R=(CH,),COOC,.H;). When the above half-ester amide (25 g.) was boiled with phosphorus oxychloride (35 c.c.) there was a vigorous evolution of hydrogen chloride. This soon diminished but refluxing was continued for one hour. Then excess phosphorus oxychloride was removed under reduced pressure, the residual gum dissolved in alcohol and the solution poured into excess dilute ammonia. The precipitated oil was shaken out with ether and after washing and drying, the ether was distilled off. The residue, which soon solidified, yielded the pure product (16 g.) after several recrystallisations from light petroleum. It formed long colourless needles which melted at.54°. Found: C, 77-8; H, 6:7%. -Calculated for C,,H,,O,N : C, 78-2; H, 6-9%. The picrate prepared in and recrystallised from alcohol separated as yellow needles m.p. 131°. Found: N, 10:6%. Calculated for C,,H,,0,N,: N, 10-4%. 0-(6-Phenanthridyl)-valeric Acid (II, R=(CH,),COOH). The acid, obtained by hydrolysis of the ester, crystallised from aqueous methyl] alcohol in colourless needles. On heating it melted at 108°, but a clear melt was not obtained until 115°. Found :, C, 77:2; -H, 6:3%. , Caleulated for C,,H,,O,N: C, 77:4; Ei, oto Solutions of this substance and of the related ester show the same fluorescence phenomena as those of their respective lower homologues. N : N’-di-(o-xenyl)-adipamide, (IV, n=4). When a stirred solution of o-xenylamine (33-8 g.) in dry ether (1,000 c.c.) was treated with a solution of adipyl chloride (9-2 g.) in ether (100 c.c.), precipitation of a white solid began almost immediately. After stirring for thirty minutes, this was collected and washed well with ether. The combined filtrates were washed with dilute acid and water and then dried. Evaporation of the ether left only a small amount of material and it was evident that the main bulk of the. product remained with the amine-hydrochloride. This was extracted with hot benzene, and the extract evaporated to a small volume and then cooled, when the nearly pure product (21 g.) STUDIES IN THE PHENANTHRIDINE SERIES. 155 separated. The pure substance crystallised from alcohol as colourless needles melting at 171°. It was fairly soluble in hot benzene, less so in hot alcohol, and sparimgly in cold alcohol. Found: C, 80:6; H, 6:0%. Calculated for Cy3)H,,0.N.: C, 80-4; H, 6-2%. Cyclisation of (IV, n=4). The diamide (19 g.) and phosphorus oxychloride (50 c.c.) were heated under reflux until evolution of hydrogen chloride had almost ceased (about two hours). After distilling off excess phosphorus oxychloride under reduced pressure, the dark residue was dissolved in hot alcohol and the resulting solution poured into excess dilute ammonia. The product was extracted with benzene, washed with water and dried over solid sodium hydroxide. The benzene solution was evaporated to a small bulk and then an equal volume of hot alcohol added. The substance A (7 g.) which separated on cooling was nearly pure. The filtrate was taken to dryness and the residue dissolved in the minimum of hot ether. On standing a substance B (0-5 g.) separated. The filtrate from this was evaporated, the residue dissolved in the minimum of hot acetone and three volumes of alcohol added. After some time an additional amount of A (1-5 g.) crystallised out. From this last filtrate no crystalline product could be isolated. The substance B crystallised from benzene in small elongated hexagons melting at 214°. It was sparingly soluble in the usual organic solvents. Its solutions in glacial acetic acid and in hot sulphuric acid showed a blue fluorescence, and its benzene solution gave an immediate precipitate with alcoholic picric acid. It is therefore regarded as I : 4-di-(6’-phenanthridyl)- butane (V). Found: C, 87:0; H, 5:6%. Calculated for C,,H,,N,: C, 87-4; H, 5-8%. The substance A separated from alcohol benzene mixture in colourless needles, m.p. 167°. It is moderately soluble in hot benzene, but less so in alcohol. Found: C, 87-1; H,;, 5-7%. Calculated for C,,H,,N.: C, 78-4; H, 5:8%. Solutions of the substance in glacial acetic acid and concentrated sulphuric acid were non- fluorescent. It did net form a picrate nor did it react with methyl sulphate in boiling toluene. N : N’-di-(0-xenyl)-glutaramide, (IV, n==3). The action of glutaryl chloride (9-8 g.) on o-xenylamine (39 g.) in dry benzene (700 c.c.) eventually yielded the crude diamide in over 90% yield. The pure substance crystallised from alcohol in long colourless needles melting at 162°. Found: C, 80:4; H, 6-1%. Calculated for C,,H,,O,.N,: C, 80:2; H, 6:0%. The product of the reaction with boiling phosphorus oxychloride for two hours was a dark red gum from which only starting material (about 50%) could be isolated. 2-Phenylacetamido-diphenyl, (I, R=CH,C,H;). Phenylacetyl chloride (31 g.) and o-xenylamine (34 g.) reacted in dry pyridine (50 c.c.) to form the amide (55 g.). Purification of this was difficult but eventually by crystallisation from light petroleum containing a little ether it was obtained as a colourless wax-like solid m.p. 37°. It is easily soluble in the usual organic solvents. Found: C, 83-2; H, 6:0%. Calculated for C,,H,,ON: C, 83-6; H, 5-9%. 6-Benzyl-phenanthridine, (II, R=CH,C,H,). The amide (45 g.) was gently refluxed with phosphorus oxychloride (100 ¢c.c.) for one hour, during which time much hydrogen chloride was evolved and the reaction mixture darkened. After removing excess phosphorus oxychloride under reduced pressure, the residue was shaken with benzene and dilute ammonia. After washing and drying, the benzene solution was distilled, finally under reduced pressure. The fraction boiling at 270-5°/18 m.m. solidified on rubbing. Several recrystallisations from methyl] alcohol afforded the pure product (8 g.) as colourless prismatic needles m.p. 112°. Found: C, 88:8; H, 5:3%. Calculated for C,,H,,N: C, 89:2; H, 5-5%. It was very easily soluble in benzene, but much less so in ethyl and methy] alcohols. 156 E. RITCHIE. The picrate crystallised rapidly when hot alcoholic solutions of its components were mixed, as fine yellow needles. It sintered at about 190°, and then gradually decomposed. Found: N, 11-4%. Calculated for C,,H,,0,N,: N, 11-2%. 2-(B-Phenyl-propronamido)-diphenyl, (I, R=(CH,).C,H;). This substance, prepared by the action of 8-phenyl-propiony] chloride (11 g.) with o-xenyl- amine (22 g.) in dry ether (600 c.c.) in the usual manner, was purified by two recrystallisations from light petroleum containing a little acetone. It (15 g.) formed colourless needles m.p. 81°. Found: C, 80-1; H, 6-2%. Calculated for C,,H,ON: C, 80:4; H, 6:3%. It was easily soluble in the usual organic solvents, excepting light petroleum. 6-(8-Phenyl-ethyl)-phenanthridine, (II, R=(CH,),C.H;). The above amide (14 g.) was cyclised by boiling with phosphorus oxychloride (25 c.c.) for one hour. The pure substance (9 g.) separated from methyl] alcohol in colourless tablets which melted at 93°. Found: C, 88-8; H, 6-0%. Calculated for C,,H,,N: C, 89:0; H, 6-0%. The picrate prepared in alcohol crystallised in glistening yellow plates which began to decompose at about 170° but did not melt completely until 198°. Found: N, 10:8%. Calculated for C,,H,,O0,N,: N, 10-9%. 2-(Phenoxy-acetamido)-diphenyl, (I, R—=CH,OC,H;). A solution of o-xenylamine (10 g.) in dry pyridine (20 c.c.) was treated with phenoxy-acetyl chloride (10 g.) and heated on the water bath for one hour. The material precipitated by pouring the reaction mixture into dilute hydrochloric acid soon solidified. By recrystallisation from methyl] alcohol it was obtained in long colourless needles (15 g.) which melted at 91°. Frequently the needles were flattened and showed the hour glass structure. Found: C, 79:4; H, 5-8%. Calculated for C,,H,,O,.N: C, 79-2’; H, 5:6%. 6-(Phenoxy-methyl)-phenanthridine, (II, R=CH,OC,H;). When the amide (8 g.) was heated with phosphorus oxychloride (15 c.c.) it rapidly dissolved and hydrogen chloride was evolved. After one hour, the reaction mixture was cooled and poured into excess dilute ammonia. The product was extracted with benzene and the benzene solution washed and dried over solid sodium hydroxide. Evaporation of the benzene left an oil which easily solidified, but the product was contaminated with a small amount of amorphous material which could not be removed by recrystallisation. Purification was best effected by conversion to the picrate in alcoholic solution followed by regeneration. The substance (5 g.) then recrystal- lised from alcohol to which a little benzene had been added, in colourless need m.p. 142°. It was easily soluble in benzene but less so in alcohol. Found: C, 84:1; H, 5-5%. Calculated for C,.H,,ON: C, 84:2; H, 5-3%. The sparingly soluble picrate separated from hot alcohol in small bright yellow needles which begin to decompose at about 170°. Found: IN, lL. Calculated for Cot .O,N,; WN, 10299,: 2-Crotonamido-diphenyl, (I, R=CH=CHCH,). This substance was prepared in 90% yield from crotonyl chloride (1 mol.) and o-xenylamine (2 mol.) in dry ether in the usual manner. It crysta!lised from acetone light petroleum mixture in colourless needles which melted at 96°. Found: C, 81:3; H, 6-2%. Calculated for C,,H,,ON: C, 81:0; H, 6-3%. It was rapidly resinified by boiling phosphorus oxychloride. STUDIES IN THE PHENANTHRIDINE SERIES. 157 2-Cinrnamamido-diphenyl, (I, R=CH=—CHC,H;). The action of cinnamoyl chloride (15 g.) on o-xenylamine (15 g.) in dry pyridine (25 c.c.) in the usual manner yielded the cinnamoy]l derivative. The pure substance (24 g.) separated from alcohol in fine colourless needles, m.p. 141°. Found: C, 84:5; H, 5:9%. Calculated for C,,H,,ON: C, 84:3; H, 5-7%. 6-Styryl-phenanthridine, (II, R—=CH=—CHC,fHH;). The above amide (8 g.) was refluxed gently with phosphorus oxychloride (15 c¢.c.), for one hour. Evolution of hydrogen chloride had then ceased and the reaction mixture was dark red. After cooling, it was poured into excess dilute ammonia and extracted with benzene. The benzene solution was evaporated to a small bulk and then treated with excess alcoholic picric acid. The picrate was collected and washed with warm alcohol and benzene. The base, regenerated in the usual manner, was nearly pure (0-8 g.). It crystallised from alcohol in long colourless slender needles melting at 133°. It was easily soluble in benzene, but less so in alcohol. Found : C, 89-5; H, 5-4%. Calculated for C,,H,,N: C, 89-7; H, 5-3%. The picrate crystallised from hot alcohol in very small bright yellow needles which decomposed at 237°. Pound«) N, 11:1. Caleulated for C,,H,,0,N,: N, 11-0%. N-(0-Xenyl)-maleamic Acid, (I, R=CH=CHCOO8H). When a solution of o-xenylamine (8-4 g.) in a little warm dry benzene was added to a solution of maleic anhydride (4-9 g.) in hot dry benzene, an orange colour developed. This rapidly faded to pale yellow and the product began to separate. After refluxing for thirty minutes, the reaction mixture was cooled and the nearly theoretical yield collected. The substance crystallised from alcohol in rosettes of small colourless needles m.p. 167°. Found: C, 71:7; H, 4:9%. Calculated for C,,H,,0,N: C, 71:9; H, 4:9%. It is almost completely insoluble in hot benzene, but fairly soluble in hot alcohol. Attempts to cyclise it led to the formation of a dark green tar. 2-Acetoacetamido-diphenyl (I, R=CH,COCEH,). o-Xenylamine (10 g.) was melted and added in one lot to ethyl acetoacetate (40 g.) maintained at 160°. After ten minutes, when the brisk evolution of alcohol had ceased, the liquid was allowed to cool somewhat and the excess ester removed under reduced pressure. The oily residue solidified overnight and the material, after draining on a porous tile, was recrystallised from light petroleum containing a small amount of benzene. It separated as long colourless needles (9 g.) which melted at 84°. Found: N, 5:6%. Calculated for C,,H,;0,.N: N, 5-5%. When heated with phosphorus oxychloride, hydrogen chloride was evolved, but the product was a tar. 2-Hydroxy-4-methyl-8-(phenyl-4’-sulphonic acid)-quinoline (V1). A solution of the amide (20 g.) in ice cold contentrated sulphuric acid (100 c.c.) was allowed to come to room temperature during thirty minutes and then heated on the water bath for twenty minutes. After cooling, the solution was poured into ice water (500 c.c.). The product (15 g.) which slowly separated was recrystallised from dilute sulphuric acid, forming colourless glistening leaflets, which did not melt below 300°. It was insoluble in organic solvents, but easily soluble in water, from which it was partially precipitated by the addition of strong acids. On the addition of sodium carbonate to its aqueous solution, carbon dioxide was evolved. The substance also gave a, test for sulphur when fused with sodium. Hound: C, 56-9; H, 4°8%.' E.W. 332. Calculated for C,,H,,0,NS .H,O: C, 57-6; H, 4:5%. E.W. 333. Attempts to cyclise the amide without concomitant sulphonation were unsuccessful. 158 E. RITCHIE. 2-(a-Naphthamido)-diphenyl, (I, R==a-C,,H.). The amide, prepared in good yield from o-xenylamine (11-3 g.) and «-naphthoyl chloride (13-7 g.) in dry pyridine (20 ¢c.c.) by the usual method, crystallised from alcohol in colourless needles m.p. 142°. Found: C, 85-2; H, 5-2%. Calculated for C,;H,,ON: C, 85:4; H, 5-3%. 6-(a-Naphthyl)-phenanthridine, (II, R=a-Cy)H,;). Cyclisation of the «-naphthoyl derivative (18 g.) with phosphorus oxychloride (35 c.c.) proceeded easily and in five minutes evolution of hydrogen chloride had nearly ceased. The pure substance (12 g.) formed small colourless plates, m.p. 125°, from methyl alcohol. It was easily soluble in the usual solvents, excepting light petroleum. Found: C, 90:6; H, 4-8%. Calculated for C,,H,;N: C, 90°5; H, 4:9%. The picrate crystallised from alcohol in sparingly soluble yellow ferns, melting at 245° with decomposition. Found: N, 10:3%. Calculated for C,,H,,0,N, N, 10-5%. The methiodide formed rapidly when the base was heated with excess methyl iodide at 100° in a sealed tube. After thirty minutes, the reaction mixture was diluted with ether and the product collected. It was dimorphous appearing as yellow rods or orange prisms. Both forms separated from alcohol at the same time, irrespective of the seed used, but on standing or more rapidly on rubbing the yellow form disappeared. The orange prisms melted with decomposition at 211°. The pure yellow form was not isolated. Found: C, 64:6; H, 3-9%. Calculated for C,,H,,NI: C, 64:4; H, 4:0%. 2-Mesitamido-diphenyl. Mesitoyl chloride (9 g.) reacted slowly with o-xenylamine (16-7 g.) in dry ether (600 c.c.) at room temperature. After standing for two weeks the mixture was refluxed for fifteen hours and then filtered. The filtrate, after washing thoroughly with dilute acid, dilute alkali and water, and drying, was evaporated. Crystallisation of the residue from light petroleum to which a little acetone had been added, yielded the pure substance (4 g.) as colourless needles m.p. 125°. Found: C, 84:0; H, 6:8%. Calculated for C,,H,,ON: C, 83-8; H, 6-7%. 6-(2’ : 4’ : 6’-Trimethyl-phenyl)-phenanthridine. The amide (3-5 g.) was rapidly cyclised by boiling phosphorus oxychloride (7 c.c.), the reaction being substantially complete in five minutes. By working up the reaction mixture in the usual way, a crude product (3 g.) was obtained. Recrystallisation from methyl alcohol afforded large colourless rhombs melting at 157°. Found: C, 88-3; H, 6-7%. Calculated for C,.H,,N: C, 88-9; H, 6-4%. The picrate separated from alcohol in microscopic yellow irregular plates melting at 231° with decomposition. Found: N, 10-8%. Calculated for C,,H,.O,N,: N, 10-6%. The methiodide prepared without difficulty in the usual manner crystallised from acetone ethyl acetate mixture as yellow prisms which began to decompose at 202°. Found: C, 63:0; H, 4:9%. Calculated for C,,H,.NI: C, 62-9; H, 5:0%. It was easily soluble in alcohol and acetone, but only slightly soluble in water and ethyl acetate. REFERENCES. Fuson, R. C., 1935. Chem. Rev., 16, 1. Hunter, L., 1941. J. Soe. chem. Ind., Lond., 60, 32. Morgan, G. T., and Walls, L. P., 1931. J. chem. Soc., 2447. Morgan, G. T., and Walls, L. P., 1932. J. chem. Soc., 2225. Morgan, G. T., and Walls, L. P., 1939. Brit. Pat., No. 511, 353. Morgan, G. T., and Walls, L. P., 1940. Brit. Pat., No. 520, 273. Pinck, L. A., and Hilbert, G. E.. 1937. J. Amer. chem. Soc., 59, 8. Walls, L. P., 1934. J. chem. Soc., 104. . STUDIES IN THE PHENANTHRIDINE SERIES. Part IV. 1:10-DIMETHYL PHENANTHRIDINES. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. In spite of several attempts to synthesise it, 4 : 5-dimethyl-phenanthrene (1) has not yet been prepared. These failures are undoubtedly due to the consider- able strain which would exist in this molecule, for a scale drawing in which the molecule is planar, shows that the methyl groups interfere with each other to a large extent. However, it has been proved by Newman (1940), who has briefly reviewed this subject, that more complex substances of this type do exist. He succeeded in preparing 4: 5-dimethyl-chrysene for example, and in @ more recent article Cook (1942) cites additional examples. The successful synthesis of these more complex molecules is probably made possible by their greater flexibility, but it must be pointed out that there is still great resistance to their formation as evidenced by the low yields obtained in their preparation. Sub- stances of this type also have abnormal properties. . The question of the spatial disposition of the methyl groups is an interesting one, and Newman has suggested three alternatives (or a combination thereof) : ‘‘ (1) the methyl groups lie bent away from each other but in the same plane as the aromatic rings; (2) the aromatic rings are distorted in some way; (3) the methyl groups are bent out of the plane of the aromatic rings. Of these, the latter seems more likely. Should this indeed be the case, optical activity would be possible.”” A completely satisfactory answer to this question could probably be obtained only by an X-ray crystallographic examination of the substances in question. In view of the resistance to the formation of hydrocarbons of the 4 ; 5-dimethyl-phenanthrene type, it was surprising to find in the literature that certain analogously constituted heterocyclic compounds are easily prepared in good yield. Thus 2: 2’-dinitro-6: 6’-dimethyl-diphenyl readily yields 1 : 10-dimethyl-benzo(c)cinnoline oxide by sodium sulphide reduction (Sako, 1934), and 1: 10-dimethyl-benzo(c)cinnoline (R.I. No. 1909) (II) by sodium amalgam reduction (Kenner and Stubbings, 1921) or by electrolytic reduction (Wittig and Stichnoth, 1935). The latter workers also attempted the optical resolution of this substance but without success, which suggests but does not prove a planar configuration for this molecule. Sako (loc. cit.) found that optically active 6-nitro-6’-acetamido-diphenic acid was slowly transformed by hot concentrated sulphuric acid to inactive 10-nitro-phenanthridone-1-carboxylic acid (III). No attempt was made to resolve this substance and since the diphenic acid is readily racemised by mineral acids, no conclusions regarding the configuration of (III) can be drawn from this experiment, as realised by Sako. These results suggested that a study of the cyclisation of the acyl derivatives of 2-amino-6 : 6’-dimethyl-diphenyl (V) would contribute useful information to this question. The required base was prepared as follows: By partial reduction (Sako, loc. cit.) 2: 2’-dimethyl-6 : 6’-dinitro-diphenyl was converted to 2; 2'’-dimethyl-6-amino-6’-nitro-diphenyl, which was deaminated by diazotisa- tion and treatment with hypophosphorus acid to 2-nitro-6 : 6’-dimethyl-dipheny] (IV), a pale yellow solid m.p. 45°. The properties of this substance agreed with 160 E. RITCHIE. those reported for it by Mascarelli and Angeletti (1938), who obtained it by a slightly different method. Reduction by stannous chloride in acetic-hydrochloric acid yielded the base (V) as a colourless oil boiling at 169-170°/20 m.m. which could not be induced to crystallise, although Mascarelli and Angeletti (loc. cit.) CH, HOOC cH.L JNO, ie CO H, ON cH 2-2 ( i iii IV CH, NHCOR CH CH, N N i] i dg inna Ch, CH, CH, V Vi Vil Vill 1X x describe it as a pale yellow crystalline powder m.p. 105°. These authors did not characterise their product by the preparation of solid derivatives so further checks were not possible. The base (V) condensed readily with 5-nitro-salicylaldehyde to a yellow crystalline Schiff’s base and also yielded a pale yellow p-nitro-benzoyl derivative (VI, R,=p-C,H,NO,), but the benzoyl derivative (VI, R,=C,H;) could not be induced to crystallise. STUDIES IN THE PHENANTHRIDINE SERIES. 161 The cyclisation of these acyl derivatives was carried out with phosphorus oxychloride in nitrobenzene solution at 180°, since it was thought that the 6’ methyl group and the p-nitro-phenyl group would retard cyclisation by virtue of their inductive effects. Under these ‘ forcing’’ conditions cyclisation was rapid and good yields (about 70%) of 1: 10-dimethy1-9-phenyl-phenanthridine (VII, R,=C,H;) and 1: 10-dimethyl-9-(4’-nitro-phenyl)-phenanthridine (VII, R,=pC,H,NO,) were obtained. The former substance was isolated as a gum which could not be crystallised even after regeneration from its picrate and sulphate, both of which were crystalline. However it showed the fluorescence phenomena typical of phenanthridines and its picrate had the correct analytical figures, so there can be no doubt as to its identity. The substance (VII, R,=pC,H,NO,) erystallised from ethyl acetate in pale yellow prisms. When heated, these partially melted at 148°, resolidified and then melted again at 163°. Repeated recrystallisations from a variety of solvents failed to alter this behaviour, and in each case the crystals appeared to be homogeneous. The same double melting point was also observed after the substance had been converted to its picrate (which had a single m.p.) and regenerated. A conclusive proof of its homogeneity was provided by the fact that no bands were formed when it was adsorbed on Brockmann alumina and that the various fractions obtained by elution were identical. There can be no doubt then that 1: 10- dimethyl-phenanthridines are capable of existence and that they are readily formed by the application of the Morgan-Walls reaction to appropriately substituted diphenyls. It was originally intended to reduce the nitro compound to the base (VIII, R,=H) and attempt an optical resolution, but the amine proved to be unsuitable for this purpose. On catalytic reduction the nitro compound took up the theoretical amount of hydrogen (further evidence for its structure and homo- geneity) but the amine thus obtained was a glass. From this a crystalline acetyl derivative (VIII, R,—COCH,) was prepared which was hydrolysed back to the amine, which was then obtained in crystalline form. Purification of this was difficult because of its high solubility in all solvents excepting water and light petroleum, its tendency to separate from mixed solvents as an oil and its discoloration on exposure to the air. The best specimen obtained was light tan. It dissolved in dilute hydrochloric acid to an orange-red solution, changed to yellow by addition of concentrated acid. According to Morgan and Walls (1931, 1938) these colour changes are due to the formation of the resonating ion (LX) by the attachment of a proton to the ring nitrogen in dilute acid and its conversion to the non-resonating ion (X) in concentrated acid by the addition of another proton to the other nitrogen. Now resonance could occur only if the phenanthridine nucleus were planar or nearly so, hence it appears certain that this is the case in the ion (IX). It does not necessarily follow, however, that the phenanthridine nucleus in the free base (VIII, R,=H) and in the other 1: 10-dimethyl-phenanthridines is planar or nearly so, since there is probably considerable resonance energy associated with the ion (IX). EXPERIMENTAL. Copper Powder. Wittig and Stichnoth (loc. cit.) prepared 2: 2’-dimethyl-6 : 6’-dinitro-diphenyl by heating 2-iodo-3 nitro-toluene with “‘ Naturkupfer C’’. This grade of copper powder was not available in sufficient quantity, but it was found that copper powder prepared by the following method was equally good. A solution of copper sulphate (300 g.) in water (1,000 c.c.) was stirred and cooled in running water whilst it was treated gradually with zinc dust (about 90 g.) until decolorised. Addition of the zinc dust was regulated so that the temperature did not rise above 60°. After settling, the supernatant liquid was decanted and the copper washed by decantation successively with 162 : E. RITCHIE. water (twice, 2,000 c.c. each time), dilute hydrochloric acid (twice 2,000 ¢c.c. of N), water (thrice, 2,000 c.c.) and acetone (twice, 1,000 ¢c.c.). It was then transferred to a Buchner funnel with acetone, pressed well and dried for three hours over concentrated sulphuric acid in an evacuated desiccator. The product, which weighed 70-75 g., should be used shortly after its preparation. 2-Nitro-6 : 6’-dimethyl-diphenyl (IV). The method described by Mascarelli and Angeletti (loc. cit.) is much inferior to the one described below. _ The ice cold diazo solution prepared from 2 : 2’-dimethyl-6-amino-6’-nitro-diphenyl (69 g.) concentrated hydrochloric acid (207 c.c.), water (420 ¢.c.) and sodium nitrite (20 g.) in water (40 c.c.) in the usual manner was added in one lot to a solution of potassium hypophosphite (250 g.) in concentrated hydrochloric acid (100 c.c.) and water (625 ¢.c.) at 0°. The reaction mixture was maintained at this temperature for ten hours and then allowed to come to room temperature during twelve hours. After heating to boiling, the mixture was cooled and extracted with ether. The ethereal extract was washed thoroughly with dilute sodium hydroxide solution and water, dried and distilled. The pure product (40 g.) distilled at 191-2°/20 m.m. as a yellow oil which eventually solidified to a pale yellow solid mp. 465°. 2-Amino-6 ; 6’-dimethyl-diphenyl (V). ‘To a solution of the nitro compound (36 g.) in glacial acetic acid (400 c.c.) warmed to 70° was added rapidly a solution of stannous chloride (160 g.) in concentrated hydrochloric acid (180 c.c.) warmed to the same temperature. - The temperature rose to about 85° and the yellow colour began to fade. After fifteen minutes the reaction mixture was heated on the water’bath for one hour, then cooled, diluted and poured into excess dilute sodium hydroxide solution. The product was extracted with ether, dried over solid sodium hydroxide, and then distilled. The pure amine (27 g.) was isolated as colourless oil boiling at 169-170°/20 m.m. which could not be crystallised. Found: C, 85:1; H, 7-4%. Calculated for C,,H,,N.: C, 85:3; H, 7-6%. The 5-nitro-salicylidene derivative prepared in the usual manner, crystallised from alcohol im small yellow needles m.p. 108°. Found: C, 72:4; H, 5:4%. Calculated for C,,H,,0,N,:C, 72:8; H, 5-2%. 2-Benzamido-6 : 6’-dimethyl-diphenyl (VI, R,==C,H;). A solution of the amine (5 g.) in dry pyridine (10 c.c.) was treated with benzoyl chloride (3-5 g.) and heated on the steam bath for one hour. The product precipitated on pouring the reaction mixture into dilute hydrochloric acid was extracted with ether and the extract washed thoroughly with dilute acid, dilute alkali and water and dried. Evaporation of the ether left a thick colourless syrup (7:5 g.) which could not be crystallised. 2-(p-Nurobenzamido)-6 : 6’-dimethyl-diphenyl (VI, R,=pC,H,NO,). This derivative prepared as above, crystallised from a little alcohol in very pale yellow rectangular plates melting at 122°. Found: C, 72:5; H, 5-4%. Calculated for C,,H,,0,N,: C, 72:8; H, 5-2%. 1 : 10-Dimethyl-6-(4'-nitro-phenyl)-phenanthridine (VII, R,==pC,H,NO,). When a solution of the p-nitrobenzoyl derivative (11-5 g.) in nitrobenzene (30 c.c.) and phosphorus oxychloride (15 c¢.c.) was heated to 180°, a vigorous reaction set in, and hydrogen chloride was copiously evolved. The vigour of the reaction soon diminished but heating was continued for five hours. After cooling, the reaction mixture was poured into excess dilute ammonia and the nitrobenzene removed by steam distillation. The product which formed a glass on cooling crystallised readily when rubbed in contact with ethyl acetate. Two recrystal- lisations from alcohol-ethyl acetate yielded the pure product (8 g.) as pale yellow prisms which melted at 148° and then at 163°. Found: C, 76-5; H, 4-9%. Calculated for C,,H,,O,.N,: C, 76:8; H, 4:9%. STUDIES IN THE PHENANTHRIDINE SERIES. 163 The substance was easily soluble in benzene, moderately soluble in ethyl acetate, but only slightly soluble in alcohol. _ Various tests of the homogeneity of this substance were made as described in the theoretical section above. The picrate separated as bundles of small yellow rods when hot alcoholic picric acid was added to a benzene solution of the base. It melted at 235° with slight darkening. Found: C, 58-1; H, 3-6%. Calculated for C,,H,O.N,;: C, 58-2; H, 3-4%. 1: 10-Dimethyl-6-phenyl-phenanthridine (VII, R,=C,H;). , Cyclisation of the benzoyl derivative (7-5 g.) carried out as above, yielded a colourless gum (5 g.) which could not be crystallised even after regeneration from its picrate or sulphate. In glacial acetic acid or concentrated or dilute sulphuric acid it formed a yellow solution with a green fluorescence. The picrate prepared in alcohol, separated as small yellow elongated plates. Found: C, 63:2; H, 3-9%. Calculated for C,,H,,O,N,: C, 63:3; H, 3°-9%. This substance also has a double melting point. When heated it sinters at 170°, melts at 177°, then solidifies and melts at 199°. 1 : 10-Dimethyl-6-(4’ -acetamido-phenyl)-phenanthridine (VIII, R,—=COCH,). A solution of the nitro compound (VII, R,=p-C,H,NO,; 12 g.) in the minimum of hot alcohol was shaken in an atmosphere of hydrogen in the presence of Raney nickel. The theoretical amount of hydrogen was taken up without difficulty. After filtering off the catalyst, the filtrate was evaporated leaving a pale yellow glass, which could not be crystallised. However, a crystal- line acetyl derivative was readily obtained by the usual methods. The crude substance crystal- lised once from benzene yielded a nearly pure product (10 g.). The pure substance separates from much aqueous methyl! alcohol in small colourless plates, m.p. 199°. Found: C, 80-5; H, 5-8%. Calculated for C,,H,,ON,: C, 81-2; H, 5-9%. It is easily soluble in alcohol and in ethyl acetate but less so in benzene. 1 : 10-Dimethyl-6-(4’-amino-phenyl)-phenanthridine (VIII, R,=H). The acetyl derivative (3 g.) was hydrolysed by boiling for three hours with alcoholic potassium hydroxide (100 c.c. of 10%). After removing the alcohol, the mixture was diluted and extracted with benzene. Distillation of the benzene left an orange gum which eventually solidified. Purification of this was difficult but eventually by crystallisation from ether containing a little light petroleum the product was obtained as small light tan needles which melted at 134°. Found: C, 84-5; H, 6-0%. Calculated for C,,H,,N,: C, 84-6; H, 6-0%. It was easily soluble in the usual organic solvents, excepting light petroleum. It dissolved in very dilute hydrochloric acid to an orange-red solution. On the addition of concentrated hydrochloric acid the colour changed to yellow. REFERENCES. Cook, J. W., 1942. Rep. Progr. Chem., 171. Kenner, J., and Stubbings, W. V., 1921. J. chem. Soc., 600. Mascarelli, L., and Angeletti, A., 1938. Gazz. chim. Ital., 68, 29. Morgan, G. T., and Walls, L. P., 1931. J. chem. Soc., 2447. Morgan, G. T., and Walls, L. P., 1938. J. chem. Soc., 389. Newman, M. S., 1940. J. Amer. chem. Soc., 62, 2295. Sako, S., 1934. Bull. chem. Soc. Japan, 9, 393. STUDIES IN THE PHENANTHRIDINE SERIES. PART V. PHENANTHRIDINE-6-ALDEHYDE AND RELATED COMPOUNDS. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. The experiments recorded in Part III of this series show that the application of the Morgan-Walls reaction to acyl derivatives of o-xenylamine cannot always be relied upon to yield the corresponding phenanthridines, particularly when the acyl residue carries reactive atoms or groups. Accordingly another synthesis was sought which would furnish phenanthridines substituted in the 6 position even when the substituents bore reactive groups. The most suitable starting material for such a synthesis appeared to be phenanthridine-6-aldehyde (1), since it has been shown that the similarly constituted quinoline-2-aldehyde and isoquinoline-1-aldehyde possess the normal reactivity of aldehydes (Kwartler and Lindwall, 1937 ; Burger and Modlin, 1940; Johnson and Hamilton, 1941 ; Barrows and Lindwall, 1942). It was decided therefore to prepare this substance and examine its behaviour in some common reactions typical of aldehydes. The «- and y-aldehydes of the pyridine, quinoline and isoquinoline series have been prepared in poor to moderate yields by oxidising the corresponding methyl] derivatives with selenium dioxide (Henze, 1934; Glenn and Bailey, 1941 ; Kaplan, 1941; Borsche and Hartmann, 1940; and references quoted above). When 6-methyl-phenanthridine was refluxed with selenium dioxide in alcoholic solution the reaction did not proceed smoothly. But when the oxidation was carried out in ethyl acetate, the aldehyde (I) was obtained in about 70% yield. It dissolved easily in dilute hydrochloric acid to a yellow solution which showed a faint blue fluorescence, but all attempts to obtain quaternary ammonium salts from it failed. It was recovered unchanged, except for slight resinification, after heating with excess methyl iodide at 100° for two hours, and was gradually resinified when refluxed with methyl sulphate in toluene. This unreactivity cannot be caused by steric hindrance by the formyl group, since phenanthridines substituted in the 6 position by much bulkier groups react readily under these conditions (Part III), but must be due to the lowered availability of the lone electron pair of the nitrogen for quaternary ammonium salt formation, by conjugation with the strongly electrophilic oxygen of the carbonyl group. This effect is shown diagrammatically in the resonance form (1a). The aldehyde readily yielded an oxime, semicarbazone and phenylhydrazone by the usual methods and also condensed with p-toluidine to a Schiff’s base. When oxidised in dilute sulphuric acid solution with the theoretical amount of potassium dichromate, no carboxylic acid was formed; instead the reaction mixture consisted of unchanged aldehyde and phenanthridone. A _ nearly quantitative yield of the latter substance was obtained when excess of the oxidising agent was used. Under the same conditions phenanthridine and 6-methyl-phenanthridine are unaffected and the facile conversion of the aldehyde to phenanthridone is remarkable. Phenanthridone was also obtained exclusively by oxidation with permanganate in acid solution at 60°, but at 40° in alkaline aqueous acetone, phenanthridine-6-carboxylic acid was the main product. This was easily decarboxylated to phenanthridine as reported by Walls (1934), and henee the sequence, selenium dioxide oxidation, permanganate oxidation, STUDIES IN THE PHENANTHRIDINE SERIES. 165 decarboxylation, provides an alternative method of degrading 6-methyl- phenanthridines to their parent bases. — Although quinoline-2-aldehyde is converted to quinaldoin under the conditions of the benzoin condensation (Henze, loc. cit.; Kaplan, loc. cit.), io UI , yt " UW C—CH(CH,C OCHg) C—CCH,NO, | OH tii IV V VI phenanthridine-6-aldehyde was immediately resinified by potassium cyanide and no crystalline product could be isolated from the reaction mixture. Reduction by formaldehyde in alkaline solution also yielded tars and a similar result attended efforts to apply D6ebner’s cinchoninic acid synthesis. Some condens- tions of the aldehyde with reactive methylene groups were then examined. When the aldehyde was heated with malonic acid in pyridine solution in the presence of piperidine, carbon dioxide was evolved but the product was a tar. 166 E. RITCHIE. However reaction with ethyl malonate proceeded smoothly, yielding crystalline ethyl a-carbethoxy-6-(6-phenanthridyl)-acrylate (II). Only amorphous products were obtained when the aldehyde was condensed with acetone, but from aceto phenone there was obtained 6-(diphenacyl-methyl)-phenanthridine (III). Nitromethane also reacted readily with the aldehyde, producing §-hydroxy-8- (6-phenanthridyl)-«-nitroethane (IV), and from trinitrotoluene, w-(6’-phenan- thridyl)-2 : 4 : 6-trinitro-styrene (V) was obtained. All attempts to oxidise 6-ethyl-phenanthridine by selenium dioxide to 6-acetyl-phenanthridine were unsuccessful, but 6-benzyl-phenanthridine (Part IIT) was easily converted to 6-benzoyl- phenanthridine (VI) characterised by its oxime, semicarbazone and phenylhydrazone. Note.—The substances, ethyl a-carbethoxy-(-(6-phenanthridyl)-acrylate (II), y-(6-phenanthridyl)-butyric acid and 4-(6-phenanthridyl)-valeric acid (Part III) were tested for plant hormone activity by the kindness of Dr. R. N. Robertson of the Botany Department of this University, who reported complete inactivity in each case. EXPERIMENTAL. Phenanthridine-6-aldehyde (1). A solution of 6-methyl-phenanthridine (12 g.) in ethyl acetate (250 c.c.) was treated with finely powdered selenium dioxide (7-6 g.) and the mixture refluxed for ten hours. The selenium which had gradually separated during this time, was filtered from the hot solution and the red filtrate evaporated on the water bath. The dark residue was then extracted with hot dilute hydrochloric acid (250 c.c. of N) and the extract charcoaled, filtered and cooled. By the careful addition of solid sodium carbonate, dark seleniferous compounds were precipitated. These were filtered off and the product recovered by further addition of sodium carbonate. After washing with hot water, it was recrystallised twice from alcohol yielding pale yellow opaque needles (9 g.) which melted at 139°. This melting point was not raised by further recrystallisation. Found: C, 80-9; H, 4-2%. Calculated for C,,H,ON: C, 81:1; H, 4-3%. It was insoluble in light petroleum, but easily soluble in other organic solvents and in dilute mineral acids. The oxime prepared in the usual manner, crystallised from a large volume of alcohol in small pale yellow plates which melted with decomposition at 227°. It was sparingly soluble in the usual organic solvents. Found: N, 12-8%. Calculated for C,,H,,ON,: N, 12-6%. The semicarbazone which is also sparingly soluble in the usual solvents, separated from much alcohol as colourless shining elongated plates, m.p. 238° with decomposition. Found: N, 21-4%. Calculated for C,,;H,,ON,: N, 21:2%. The phenylhydrazone is moderately soluble in hot alcohol and separated from this solvent in sheaves of golden needles melting at 166°. Found: N, 14:4%. Calculated for C,,)H,;N,: N, 14:1%. 6-Phenanthridyl-formylidene-p-toluidine, prepared from the aldehyde and p-toluidine in alcohol crystallised in lemon yellow needles, m.p. 87°. Found: N, 9:5%. Calculated for C,,H,,N.: N, 9-4%. Oxidation Experiments. (1) A solution of the aldehyde (6-2 g.) in water (200 c.c.) and concentrated sulphuric acid (13 c.c.) was stirred and heated on the water bath whilst it was treated with a solution of potassium dichromate (6 g.) in water (100 c.c.) during thirty minutes. Stirring and heating were continued for two hours, then the reaction mixture was diluted and the product (5-8 g.) collected. After purification it was identified as phenanthridone. (2) A solution of the aldehyde (2:1 g.) in water (50 ¢.c.) and concentrated sulphuric acid 3 c.c.) was stirred and maintained at 60° whilst 1% potassium permanganate solution was slowly added. The reaction appeared to be completed when about 125 c.c. had been added. After the STUDIES IN THE PHENANTHRIDINE SERIES. 167 addition of a little sulphurous acid to dissolve some manganese dioxide which had separated, the product (1-9 g.) was collected. It was identified as phenanthridone. (3) A solution of potassium permanganate (1:5 g.) in water (30 c.c.) was added during thirty minutes to a solution of the aldehyde (3 g.) in acetone (150 c.c.) stirred and kept at 40-45°. After stirring and warming for another thirty minutes sufficient sulphurous acid was added to dissolve the precipitated manganese dioxide, followed by dilute hydrochloric acid (40 c.c. of N). Then the acetone was boiled off and the product collected. Extraction of this with hot dilute sodium hydroxide solution (100 c.c. 4-50 ¢.c. of 5%) left a residue of phenanthridone (0:8 g.). Acidifica- tion of the alkaline extract afforded phenanthridine-6-carboxylic acid (1-8 g.), which evolved carbon dioxide at 155°, leaving phenanthridine. Ethyl x-carbethoxy-3-(6-phenanthridyl)-acrylate (II). A solution of the aldehyde (2 g.) and diethyl malonate (1-5 g.) in absolute alcohol (40 e.c.) was treated with a few drops of piperidine and allowed to stand at room temperature. After standing for about three weeks, the solvent was evaporated and the residual oil rubbed with a little acetone. The product which slowly solidified was drained on a porous tile. The pure substance (0-8 g.) obtained by two crystallisations from methanol, formed very pale yellow flat needles m.p. 91°. It was easily soluble in the usual organic solvents excepting light petroleum. Found: C, 72-4; H, 5-6%. Calculated for C,,H,,0,N : C, 72-2; H, 5-4%. 6-(Diphenacyl-methyl)-phenanthridine (IIT). When a solution of the aldehyde (2 g.) and acetophenone (1-8 g.) in alcohol (40 c.c.) was treated at room temperature with aqueous sodium hydroxide (20 c.c. of 10%) it immediately darkened in colour and in a short time the product began to separate. After half an hour it was collected, and washed with alcohol and water. Recrystallised from alcohol it (2 g.) formed dull yellow needles with a faint green tinge, which melted at 157°. Found: C, 84:1; H, 5-3%. Calculated for C,,H,,0,N : C, 83:9; H, 5-3%. 8-Hydroxy-8-(6-phenanthridyl)-x-nitroethane (IV). A solution of the aldehyde (2 g.) and nitromethane (2 c.c. excess) in alcohol (50 c.c.) was treated at room temperature with piperidine (0-5 c.c.). After standing for a few hours the product began to crystallise in rosettes of slender needles, which were collected after two days and washed with a little alcohol. The pure product (1-2 g.) crystallised from this solvent in nearly colourless needles which began to decompose at 132°. Found: N, 10-:7%. Calculated for C,;H,,.O,;N.: N, 10-5%. «-(6’-Phenanthridyl)-2 : 4 : 6 :-trinitro-styrene (V). Trinitrotoluene (1-1 g.) and the aldehyde (1 g.) were dissolved in hot alcohol (30 c.c.) and a few drops of piperidine added. The solution immediately became red and in a short time the product began to crystallise out. After cooling the nearly theoretical yield was collected and washed with alcohol. The substance crystallised from benzene in almost colourless fibrous needles which began to decompose at 180°. Found: N, 13:4%. Calculated for C,,H,.O,N,: N, 13-5%. It was only slightly soluble in the usual organic solvents. 6-Benzoyl-phenanthridine (V1). When a solution of 6-benzyl-phenanthridine (5-2 g.) and selenium dioxide (2-3 g.) in alcohol (200 c.c.) was refluxed, selenium began to separate almost immediately. After ten hours, the hot solution was filtered and the filtrate evaporated on the water bath. The residue was extracted with ethyl acetate, the extract charcoaled filtered and evaporated. The product which was now almost free of selenium compounds, was recrystallised three times from alcohol with the aid of charcoal, yielding finally the pure substance (3-5 g.) as colourless hair-like crystals which melted O—October 4, 1944. 168 E. RITCHIE. at 152°. It was easily soluble in benzene, acetone and ethyl acetate, moderately in hot alcohol, but sparingly in cold alcohol. Found: C, 84:2; H, 4-9%. Calculated for C,,H,,ON: C, 84-8; H, 4°6%. The oxime crystallised from ethyl alcohol in colourless prismatic needles which began to darken at 210° and then decomposed at 217°. Found : |.N,. 9:09. Galeulated.for\C,,H,,ON p04 Nao bo The semicarbazone separated from ethyl alcohol in colourless plates melting at 175°. Found: N, 16-8. Calculated for C,,H,,ON,: N, 16-5%. The phenylhydrazone crystallised from alcohol benzene in small yellow leaflets which melt at 92°. | Found :: N, 11-69,: : “Calculated: for 'C, HN, : 7 Ny Liss: REFERENCES. Barrows, R. 8., and Lindwail, H. G., 1942. J. Amer. chem. Soc., 64, 2430. Borsche, W., and Hartmann, H., 1940. Ber. dtsch. chem. Ges., 73, 839. Burger, A., and Modlin, L. R., 1940. J. Amer. chem. Soc. 62. 1081. Glenn, R. A., and Bailey, J. R., 1941. J. Amer. chem. Soc., 63, 639. Henze, M., 1934. Ber. dtsch. chem. Ges., 67, 750. Johnson, O. H., and Hamilton, C. S., 1941. J. Amer. chem. Soc., 63, 2864. Kaplan, H., 1941. J. Amer. chem. Soc., 63, 2654. Kwartler, C. E., and Lindwall, H. G., 1937. J. Amer. chem. Soc., 59, 524. Walls, L. P., 1934... J. chem. Soc., 104. Be: ee STUDIES IN THE PHENANTHRIDINE SERIES. Part VI. A SYNTHESIS OF 3-METHYL PHENANTHRIDINE. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. Of the nine possible mono-methyl phenanthridines, only three are described in the literature, and all of these were first synthesised by Pictet and his collaborators. The 6 isomer was obtained by fusing 2-acetamido-diphenyl with zine chloride (Pictet and Hubert, 1896), a reaction later improved by using phosphorus oxychloride to effect the cyclisation (Morgan and Walls, 1931). The 2 and 4 isomers were prepared by pyrolysing benzylidene p- and o-toluidines respectively (Pictet and Ehrlich, 1891) but the method was troublesome and gave poor yields. These two substances were later synthesised by a different method by Kenner, Ritchie and Statham (1937), who condensed p- and o-toluidines, respectively with 2-hydroxymethyl-cyclohexanone and dehydro- genated the resulting tetrahydro-phenanthridines. The synthesis of 3-methyl- phenanthridine has now been achieved by a method which could also be readily adapted to the synthesis of 1-, 2-, or 4-methyl-phenanthridine. Although Grieve and Hey (1932) failed to obtain 2-nitro-4-methyl-diphenyl (I) by the action of diazotised 3-nitro-4-amino-toluene on benzene in the presence of alkali, this reaction was found to proceed smoothly and afforded (I) in 40% yield, when the general conditions for the Gomberg reaction, recommended by Elks, Haworth and Hey (1940), were adopted. The nitro compound was catalytically reduced in good yield to the amine (II), from which an acetyl derivative (III, R,—CH,), melting at 148°, and a benzoyl derivative (III, R,=C,H;), melting at 92°, were prepared by the usual methods. By oxidation with permanganate in neutral solution (III, R,=CH;) was converted to 2-acetamido-diphenyl-4-carboxylic acid (IV) melting at 225°. These three substances (III, R,—CH;, R,=C,H; and IV) have been prepared by Grieve and Hey (loc. cit.), who obtained them from a by-product of the nitration of 4-methyl-diphenyl, and who reported for them, melting points of 145°, 221° and 222° respectively. The wide divergence in the values of the melting point of the benzoyl derivative is possibly due to these authors confusing it with the melting point of the acid (IV) during transcription. By boiling with phosphorus oxychloride the acetyl derivative was cyclised to 3: 6-dimethyl-phenanthridine (V, R,=—CH;) and the benzoyl derivative likewise yielded 3-methyl-6-phenyl-phenanthridine (V, R,—C,H,;), a further proof of its constitution. On oxidation with potassium dichromate in glacial acetic acid solution, the dimethyl-phenanthridine formed 3-methyl-phenanth- ridone (VI), the 3-methyl group being unattacked under these conditions. When distilled with zinc dust, the latter substance afforded 3-methyl-phenanthridine (VII), which was readily characterised by its picrate. . EXPERIMENTAL. 2-Nitro-4-methyl-diphenyl (I). The diazo solution prepared from 3-nitro-4-amino-toluene (71 g.) concentrated hydrochloric _ acid (180 c.c.), water (100 c.c.), sodium nitrite (35 g.) and water (70 c.c.) was added to benzene OO—October 4, 1944. 170 E. RITCHIE. (1,000 c.c.), vigorously stirred and cooled in an ice bath. Then a solution of sodium acetate (180 g.) in water (350 c.c.) was added at such a rate that the temperature of the mixture did not exceed 10°. The reaction was allowed to proceed at 5°-10° for three hours, then the cooling bath was removed and vigorous stirring continued at room temperature for forty-eight hours. CH; CH; CH, NO, NH, NHCOR | T in COOH CH, NHCOCH, ty —R IV v CH, CH, NH N | il CO CH vi vil The organic layer was separated, the benzene distilled off and the residue distilled under reduced pressure. The fraction boiling at 200-230°/30 m.m. was taken up in ether, and the ethereal solution washed thoroughly with hydrochloric acid, sodium hydroxide solution and water. The residue left after removing the ether was fractionated twice under reduced pressure, giving finally a pale orange oil (40% yield), which boiled at 207-9°/28 m.m. - nee Found: C, 73-6; H,5-0%. Calculated for C,;H,,0.N: C, 73:2; H,5-1%. i STUDIES IN THE PHENANTHRIDINE SERIES. Pik 2-Amino-4-methyl-diphenyl (II). When a warm solution of the nitro compound (I; 10 g.) in alcohol (100 ¢.c.) was shaken in an atmosphere of hydrogen at normal pressure in the presence of Raney nickel catalyst, slightly more than the theoretical amount of hydrogen was rapidly absorbed. After filtering otf the catalyst, the filtrate was distilled, finally under reduced pressure, giving the amine in nearly quantitative yield as a clear colourless oil, boiling at 193-4°/29 m.m. Found: C, 84:7; H, 7;3%. Calculated for C,,H,,N: C, 85:2; H, 7-1%. The picrate crystallised from alcohol in old gold needles which melted and decomposed at 161°. Found: N, 13:4%. Calculated for C,,H,,0;,N,: N, 13:6%. The p-nitrosalicylidene derivative separated from much alcohol in fine yellow needles, m.p. 216°. Found: N, 8-3%. Calculated for C,,H,,O,N,: N, 8-4%. 2-Acetamido-4-methyl-diphenyl (III, R,==CH,). Acetylation of the base with acetic anhydride in the usual manner yielded the acetyl derivative which crystallised from aqueous alcohol in colourless glistening flattened needles, m.p. 148°. Found: N, 6:2%. Calculated for C,;H,,ON: N, 6-2%. 2-Benzamido-4-methyl-diphenyl (III, R,—=C,H;). A solution of the amine (II; 9 g.) in dry pyridine (25 c.c.) was treated with benzoyl chloride (7 g.) and heated on the water bath for one hour.The oil which separated on pouring the reaction mixture into dilute hydrochloric acid soon solidified. The product, recrystallised from a little alcohol, formed colourless needles which melted at 92°. Found : C, 83-9; H, 6-0%. Calculated for C,,H,,ON: C, 83-6; H, 6-0%. 2-Acetamido-diphenyl-4-carboxylic Acid (IV). This substance was prepared according to the directions of Grieve and Hey (loc. cit.). The pure material crystallised from aqueous alcohol in colourless needles, m.p. 225° (lit. 222°). Mound: C, 71-1; H,5-3%. Calculated for C,,H,,0,N: C,.70-6; H, 5-1%. 3: 6-Dimethyl-phenanthridine (V, R,==CH;). When the acetyl derivative (III, R,=CH;; 9 g.) was refluxed with freshly distilled phosphorus oxychloride (15 ¢.c.) in a dry atmosphere, clouds of hydrogen chloride were evolved and after about five minutes a yellow solid separated from the reaction mixture. After heating for one hour, the excess phosphorus oxychloride was distilled off under reduced pressure. The residue was extracted with hot hydrochloric acid (250 c.c. of N), the extract filtered and basified with ammonia. The precipitated oil was taken up in ether and the ethereal solution washed and dried. The solid (6-5 g.) remaining after removal of the ether, crystallised from light petroleum (60-80°) in large prisms which melted at 101°. Found: C, 87:3; H, 6-4%. Calculated for C,,H,,N: C, 87-0; H, 6-2%. The picrate separated as small yellow needles when hot alcoholic solutions of its components were mixed. It melted with decomposition at 240°. Found: N, 13:0%. Calculated for C,,H,,O,N,: N, 12:9%. . The methosulphate was prepared by treating a solution of the base (2 g.) in dry nitrobenzene (15 ¢.c.) at 180° with methyl sulphate (1-3 g.). The product which had begun to separate immediately, was collected on cooling, washed with ether and recrystallised from alcohol. It formed pale yellow needles which melted and decomposed at 227°. Solutions in water or alcohol had an intense blue fluorescence. Found: C, 61:6; H, 5:9%. Calculated for C,,H,,O,NS: C, 61-3; H, 5-7%. 172 E. RITCHIE. The methiodide obtained by heating the base with excess methyl iodide in a sealed tube at 100° for thirty minutes, crystallised from water in fine yellow needles m.p. 278° with decom- position. Solutions of this substance in alcohol or water also had a strong blue fluorescence. — Found: C, 54:8; H, 4-7%. Calculated for C,,H,,NI: C, 55:0; H, 4-6%. 3-Methyl-6-phenyl-phenanthridine (V, R,==C,H;). Cyclisation of the benzamido-methyl-diphenyl (III, R,—C,H,;; 19 g.) with phosphorus oxychloride (25 c.c.) proceeded smoothly. The residue remaining after removal of the excess phosphorus oxychloride was warmed with ammonia and the liberated base (17-5 g.) isolated with the aid of ether. The substance crystallised from methyl] alcohol in colourless needles m.p. 120°. Found: C, 89:3; H, 5-8%. Calculated for C..H,;N: C, 89-2; H, 5-6%. The picrate prepared in alcohol formed slender lemon yellow needles decomposing at 243°. Found: N, 11-4%. Calculated for C,,H,,0,N,: N, 11-2%. The methosulphate prepared in nitrobenzene as above, crystallised from alcohol in rosettes of long pale greenish yellow needles, melting at 259°. Found: C, 67:0; H, 5-3%. Calculated for C,.H,,0,NS: C, 66:8; H, 5:3%. Solutions in alcohol had an intense blue fluorescence. 3-Methyl-phenanthridone (VI). A solution of 3 : 6-dimethyl-phenanthridine (10 g.) in glacial acetic acid (75 c.c.) was heated on the water bath and stirred whilst it was treated during thirty minutes with powdered potassium dichromate (23 g.). After refluxing for ninety minutes, the reaction mixture was poured into warm dilute hydrochloric acid. The product was collected, washed well with water and then with a little warm alcohol. After drying, the light brown material (6-2 g.) was recrystallised from nitrobenzene, affording long colourless needles which melted at 251°. It was sparingly soluble in the usual organic solvents. Found: C, 80:0; H, 5-5%. Calculated for C,,H,,ON: C, 80-4; H, 5-4%. 3-Methyl-phenanthridine (VII). An intimate mixture of 3-methyl-phenanthridone (3 g.) and zine dust (50 g.) was loosely packed into a hard glass tube sealed at one end, and then heated in a furnace. The greenish yellow fluorescent distillate was dissolved in ether, the solution washed with water, and dried. The oil (1-2 g.) remaining after removal of the ether solidified on rubbing. | After several crystal- lisations from light petroleum (60-80°) the product was obtained as rosettes of colourless needles melting at 81°. Found: C, 86:7; H, 5-7%. Calculated for C,,H,,N: C, 87-0;.H, 5°7%. It was easily soluble in the usual organic solvents. Solutions in dilute acids showed a marked blue fluorescence. The picrate separated as minute yellow needles m.p. 251° with decomposition, when hot alcoholic solutions of its components were mixed. Found: N, 13:6%. Calculated for C..H,,0,N,: N, 13-3%. REFERENCES. Elks, J., Haworth, J. W., and Hey, D. H., 1940. J. chem. Soc., 1284. Grieve, W. S. M., and Hey, D. H., 1932. J. chem. Soc., 1888. Kenner, J., Ritchie, W. H., and Statham, F. 8., 1937. J. chem. Soc., 1169. Morgan, G. T., and Walls, L. P., 1931. J. chem. Soc., 2447. Pictet, A., and Hubert, A., 1896. Ber. dtsch. chem. Ges., 29, 1182. Pictet, A., and Ehrlich, 8., 1891. Liebig’s Ann., 266, 153. STUDIES IN THE PHENANTHRIDINE SERIES. Part VII. A SYNTHESIS OF BENZO(C)PHENANTHRIDINE. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944. By distilling the alkaloids, chelerythrine chloride, sanguinarine and chelidonine with zinc dust, Spith and Kuffner (1931 (a), 1931 (6)) isolated a substance C,,H,,N which they identified with the «-naphthaphenanthridine (benzo(c)phenanthridine, R.I. No. 2740) of Graebe (1904), who had obtained it by a Series of reactions from chrysene quinone. The constitution of his product was not strictly proven by Graebe, there being a remote possibility that it was 6-naphthaphenanthridine (benzo(i)phenanthridine, R.I. No. 2741). However the structures assigned to these two substances by Graebe were confirmed later by the work of Kenner, Ritchie and Statham (1937), who synthesised the a-naphthaphenanthridine by condensing «-naphthylamine with 2-hydroxy- methyl-cyclohexanone and dehydrogenating the resulting benzotetra-hydro- phenanthridine. A further synthesis of benzo(c)phenanthridine is described below. 2-Phenyl-naphthalene, which is now readily accessible (Elks, Haworth and Hey, 1940) was nitrated to 1-nitro-2-phenyl-naphthalene (Hey and Lawton, 1940) which was catalytically reduced to the amine (I). The acetyl derivative (IT) of this (Hey et al., loc. cit.) was readily cyclised by boiling phosphorus oxychloride to 6-methyl-benzo(c)phenanthridine (III), which was converted to benzo(e)- phenanthridine by two routes. Firstly, oxidation by potassium dichromate in acetic acid gave benzo(c)phenanthridone (IV), which on distillation with zinc dust yielded the desired product (VII), as reported by Graebe (loc. cit.). Secondly, condensation with formaldehyde to 6-(88’-dihydroxy-isopropyl)- benzo(c)phenanthridine (V), followed by oxidation with Kiliani’s dichromate mixture, gave the carboxylic acid (VI). This acid rapidly split out carbon dioxide when heated slightly above its melting point, leaving (VII). The behaviour of this substance with methylating agents is noteworthy. When heated with excess methyl iodide at 100° in a sealed tube reaction was not complete even after eight hours. Methylation by methyl sulphate in boiling toluene proceeded more rapidly and appeared to be complete in about five hours. This unreactivity must be attributed to steric hindrance by carbon atom 4 and its attached hydrogen atom. When heated, the yellow methiodide gradually became red and at 195° it decomposed, evolving methyl! iodide and leaving benzo(c)phenanthridine, identified by a mixed melting point after purification. EXPERIMENTAL. 1-Amino-2-phenyl-naphthalene (1). Catalytic reduction of 1-nitro-2-phenyl-naphthalene (10 g.) in hot alcohol (250 c.c.) in the presence of Raney nickel proceeded slowly at atmospheric pressure, but eventually the theoretical amount of hydrogen was taken up. The product purified from methyl! alcohol melted at 105° (lit. 104°) and formed colourless needles. The acety] derivative (II), prepared in the usual way, was found to melt at 234° in agreement with Hey and Lawton (loc. cit.). 174 E. RITCHIE. 6-Methyl-benzo(c)phenanthridine (ITI). When the acetyl derivative (5 g.) was heated with phosphorus oxychloride (15 c.c.) it rapidly dissolved and hydrogen chloride was evolved. After five minutes’ refluxing a solid began to separate from the reaction mixture, but heating was continued for one hour. Then the excess ct ve N NH, NHC OCH, ChHs C,Hs li it CH(cH,OH), of § HN N IV Vv COOH C HC 4 4 N N vi vil phosphorus oxychloride was removed under reduced pressure and the residue shaken with ammonia and ether until no solid remained. After drying, the ether was distilled off, leaving a pale yellow solid (4-2 g.). The product obtained after recrystallisation from light petroleum (60-80°) formed colourless prisms melting at 117°. Found: C, 88-2; H, 5-6%. Calculated for C,,H,,N: C, 88-9; H, 5:3%. —— STUDIES IN THE PHENANTHRIDINE SERIES. 175 The picrate crystallised from much alcohol in small bright yellow needles, melting at 233° with decomposition. Found: N, 12-1%. Calculated for C,,H,,O,N,: N, 11-9%. Benzo(c)phenanthridone (IV). A stirred solution of methyl-benzophenanthridine (1-8 g.) in glacial acetic acid (50 c.c.) was heated on the water bath and treated during thirty minutes with powdered potassium dichromate (3:4 g.). Then after refluxing for two hours, the reaction mixture was diluted with water, and the product collected. It was washed successively with water, hot dilute sulphuric acid, boiling alcohol and hot glacial acetic acid. The crude product (1-1 g.) after two crystallisations from nitrobenzene afforded light brown needles m.p. 314° (corrected 329°) not changed by further recrystallisation. Graebe (loc. cit.) reported m.p. 332-5°. ; Found: C, 83-5; H, 4-3%. Calculated for C,,H,,ON: C, 83-2; H, 4-6%. Distillation of this substance with zinc dust in the manner described by Graebe (loc. cit.) eventually yielded benzo(c)phenanthridine m.p. 135°, identical with the substance obtained below. 6-(BB’-dihydroxy-isopropyl)-benzo(c)phenanthridine (V). A solution of (III ; 7 g.) in alcohol (150 c.c.) and aqueous formaldehyde (50 c.c. of 40%) was refluxed for eight hours and then evaporated to a small bulk. The residue was then dissolved in a mixture of alcohol (50 c.c.) and aqueous formaldehyde (50 c.c. of 40%) and refluxing continued for thirty hours. The thick syrup obtained on evaporation of the reaction mixture was boiled down several times with benzene to free it of formaldehyde and yielded finally a colourless crystalline solid. Recrystallisation of this from alcohol-benzene (1:10) afforded the pure substance (6-4 g.) as colourless needles which melted at 169°. Found: C, 79-0-; H, 5-7%. Calculated for C,.H,,O,.N: C, 79:2; H, 5-6%. The substance was very slightly soluble in light petroleum, but fairly easily soluble in other organic solvents. Its solutions in hot dilute mineral acids were yellow with a blue fluorescence. Benzo(c)phenanthridine-6-carboxylic Acid (VI). The glycol (V; 10-1 g.) was refluxed gently with dilute sulphuric acid (100 c.c. of 2N) and the suspension so obtained treated during thirty minutes with Kiliani’s dichromate mixture (135 g.). After refluxing for another three hours the reaction mixture was diluted, the product collected and washed well with water. The crude acid was purified by dissolving in hot dilute ammonia, charcoaling, and reprecipitating with acetic acid. It (6:5 g.) then formed a pale yellow powder, m.p. 198°, with evolution of carbon dioxide, which could not be satisfactorily recrystallised. Benzo(c)phenanthridine (VII). The crude acid (6 g.) was heated to 210° in an oil bath. In five minutes evolution of carbon dioxide had almost ceased, but heating was continued for another five minutes. After cooling, the crystalline solid was dissolved in ether and the ethereal solution charcoaled, washed with dilute sodium hydroxide solution and water, dried and evaporated. The pale yellow residue (4-5 g.) was recrystallised from alcohol with the aid of charcoal, yielding small colourless plates melting at 135° (lit. 135°). Found: C, 88-7; H, 4:6%. Calculated for C,,H,,N: C, 89:0; H, 4:8%. The picrate prepared in alcohol formed very sparingly soluble yellow microscopic crystals which melted and decomposed at 255-6° (lit. 256°). Found: N, 12-4%. Calculated for C,,H,,0,N,: N, 12:2%. j-Methyl-benzo(c) phenanthridinium iodide. When the base (VII) was heated with excess methyl] iodide in a sealed tube at 100°, reaction took place very slowly and was not complete even after eight hours. The product was collected, 176 E. RITCHIE. washed well with dry ether and recrystallised from water, separating as small bright yellow plates. When heated it gradually became red until at 195° it decomposed, evolving a gas. The residue was identified as benzo(c)phenanthridine. Found: C, 58:0; H, 3:9%. Calculated for C,,H,,NI: C, 58-2; H, 3-8%. 5-Methyl-benzo(c)phenanthridinium metho-sulphate. A solution of the base (2 g.) in dry toluene (25 c.c.) and methyl sulphate (1-1 g.) was refluxed for five hours. The product, which had gradually separated during this period, was collected, washed with ether and recrystallised from alcohol, forming small colourless needles which melted at 217° with decomposition. Found: C, 64:4; H,4:9%. Calculated for C,,H,,O,NS: C, 64-2; H, 4-8%. The substance was easily soluble in cold water and in hot alcohol, in which solvents it had a greenish blue fluorescence. Treatment of the aqueous solution with potassium iodide solution precipitated the methiodide. REFERENCES. Elks, J., Haworth, J. W., and Hey, D. H., 1940. J. chem. Soc., 1284. Graebe, C., 1904. Liebig’s Ann., 335, 122. Hey, D. H., and Lawton, 8S. E., 1940. J. chem. Soc., 374. Kenner, J., Ritchie, W. H., and Statham, F. S., 1937. J. chem. Soc., 1169. Spath, E., and Kuffner, P., 193la. Ber. dtsch. chem. Ges., 64, 370. Spath, E., and Kuffner, P., 1931b. Ber. dtsch. chem. Ges., 64, 1123. STUDIES IN THE PHENANTHRIDINE SERIES. ParT VIII. 3: 8-DIAMINO-PHENANTHRIDINE AND RELATED SUBSTANCES. By E. RITCHIE, M.Sc. Manuscript received, September 21, 1944. Read (in title only), October 4, 1944, A number of phenanthridine derivatives prepared by Morgan and Walls (1938) were tested for antiseptic activity by Browning, Gulbransen and Kobb (1938), who found that two of them, 2-amino-6-(4’-amino-pheny]1)-phenanthridine methochloride and 8-amino-6-(4’-amino-phenyl)-phenanthridine methochloride showed some promise. It was also found that the latter substance and the diacetyl derivative of the former had some trypanocidal value (Browning, Morgan, Robb and Walls, 1938). However no derivatives in which both amino groups were directly attached to the phenanthridine nucleus were prepared and tested. In view of the antiseptic activity of the well-known “ proflavine ”’ (3 : 6-diamino-acridine) it was decided to attempt the preparation of its formal phenanthridine analogue, 3 : 8-diamino-phenanthridine. Although this object was not accomplished, it is felt that the results obtained possess sufficient interest to warrant recording them. Kondo and Uyeo (1935, 1937) have synthesised a number of derivatives of phenanthridine by condensing substituted o-bromobenzaldehydes with substituted o-bromoanilines and heating the resulting Schiff’s bases with copper powder at 200°. The yields obtained in the latter reaction are very low, being usually about 1%, but it was thought that much better results would be obtained if both halogen atoms were activated by nitro-groups para to them. Accordingly 2; 2’-dichloro-5 : 5’-dinitro-benzylidene-aniline (I), prepared from 2-chloro- 5-nitro-aniline and 2-chloro-5-nitro-benzaldehyde, was heated with copper powder. But in spite of numerous attempts made under a large variety of conditions and with several grades of copper powder, none of the desired product could be isolated. The starting material was either recovered unchanged or converted to amorphous materials. However, an explanation can be given for this result and for the poor yields obtained by Kondo and Uyeo. It has been shown by dipole moment measure- ments (De Gaouck and Le Fevre, 1938, 1939) that Schiff’s bases have a trans, not @ cis configuration, and hence the two aromatic nuclei of such molecules cannot easily approach one another sufficiently closely at the temperature of the reaction to allow the formation of a bond between them, even after the removal of the halogen atoms by the copper. In the second attempt to synthesise 3 : 8-diamino-phenanthridine, dihydro- phenanthridine (II, R=H) served as the starting material. By reducing phenanthridine with tin and hydrochloric acid Pictet and Ankersmit (1891) obtained this substance, which they described as crystallising from dilute alcohol in fine colourless needles melting at 90° and giving blue fluorescent solutions in neutral solvents. On repeating the preparation it was found that dihydro- phenanthridine crystallised from aqueous alcohol in lustrous irregular plates melting at 123°. The solid itself as well as its solutions in neutral solvents showed a striking blue fluorescence, which disappeared on the addition of acid. 178 E. RITCHIE. There can be no ddubt that the material described by the above authors was very impure. This is understandable since the substance is unstable and a specimen exposed to the air or even kept in a tightly stoppered tube soon becomes C1 | NH NCOCH, tn CHR CHR N NCOCH, N | CH CH, CH NO, NO, NO, Vil Vill 1X yellow and its melting point falls. However the acetyl derivative (III, R=H) was quite stable and in agreement with Pictet and Ankersmit it was found to melt at 108°. Dihydrophenanthridine may be regarded as a secondary amine of the diphenyl series. It. was therefore anticipated that it would be dinitrated. 10, excess concentrated sulphuric acid to 3 : 8-dinitro-dihydrophenanthridine, STUDIES IN THE PHENANTHRIDINE SERIES. 179 which it was then intended to oxidise to 3: 8-dinitro-phenanthridine, since it had been found that dihydrophenanthridine was easily and quantitatively oxidised to phenanthridine by potassium permanganate. But in all the nitration experiments which were carried out under a variety of conditions the dihydro- phenanthridine was converted to dark amorphous products. Occasionally it was possible to isolate a crystalline substance from this, but the yield was so small that further experiments with it were not justified. According to its analytical figures, this substance is a dinitro-phenanthridine, possibly the required one (IV), but in view of the unsatisfactory nature of the reaction by which it was formed it may well be an isomer of this. While the substance was at hand some nitration experiments were also carried out on 5-acetyl-dihydrophenanthridine (III, R=H). When nitrated in sulphuric and acetic acid solution under the conditions which convert 2-acetamido-diphenyl to its 4’-nitro derivative (Scarborough and Waters, 1927), it yielded a mixture which appeared to consist cf a large amount of one substance and much smaller amounts of two others. Purification was tedious and involved considerable loss of material, but there was finally isolated a pure product, which is almost certainly 8-nitro-5-acetyl-dihydrophenanthridine (V). On hydrolysis with alcoholic hydrochloric acid it was converted in poor yield to a substance which was evidently the related nitro-phenanthridine (VI), since (a) it was stable to oxidation by permanganate in boiling acetone or by dichromate in boiling dilute sulphuric acid, and (b) it could not be acetylated even under the most vigorous conditions. The poor yield obtained suggested that the nitro- group and not atmospheric oxygen was responsible for the oxidation accompanying hydrolysis. This was confirmed by the observation that hydrolysis of 5-acetyl-dihydrophenanthridine under the same conditions yielded dihydrophenanthridine. When a larger proportion of nitric acid was used in the nitration, the product was a thick dark oil. However a dinitro compound, possibly 2 : 8-dinitro-5-acetyl-dihydrophenanthridine (VII), was isolated from this but the yield was so small that it was not further investigated. Much better results were obtained when the nitration was carried out in acetic acid solution, and the product was then readily isolated in good yield. Under similar conditions Bell (1928) found that 2-acetamido-diphenyl was nitrated in the 5 position and hence the substance obtained in the present experiments was formulated as 2-nitro-5-acetyl-dihydro-phenanthridine (VIII, R=H), a supposition later proved to be correct. On boiling with alcoholic hydrochloric acid, it hydrolysed and oxidised to 2-nitro-phenanthridine (IX, R=H). This substance which melts at 263° is possibly identical with the nitro-phenanthridine m.p. 260-2° of unknown constitution obtained by Morgan and Walls (1932) as one of the products of the direct nitration of phenanthridine. The constitution of the nitro compound (IX, R=H) was established by submitting 6-methyl-phenanthridine to the same series of reactions as those used in the preparation of (IX, R=H) from phenanthridine. Reduction by tin and hydrochloric acid yielded 6-methyl-dihydrophenanthridine (II, R =CHs) which had properties similar to those of its lower homologue but was rather more stable. Nitration of the acetyl derivative (III, R=CH;) gave (VIII, R=CH,) which as expected was converted by alcoholic hydrochloric acid to 2-nitro-6-methyl-phenanthridine (IX, R=CH,), identical with an authentic specimen prepared by the method cf Morgan and Walls (1932). Since it is highly improbable that (III, R—H) and (III, R—CH,) nitrate in different positions when the reactions are conducted under the same conditions, it follows that (IX, R=H) is 2-nitro-phenanthridine. It was difficult to devise any other straightforward synthesis of 3 : 8-diamino- phenanthridine, which had reasonable hope of success. The apparently obvious route through the preparation and degradation of 3 : 8-dinitro-6-methyl- 180 B. RITCHIE. phenanthridine was in fact unattractive since the cyclisation of 2-acetamido- 4’-nitro-diphenyl gives impracticable yields of the corresponding phenanthridine (Morgan and Walls, 1932, 1938). However, 3: 8-diamino-6-phenyl- NO, NO, NO, NH, - H, N(COCH,}, ox KI ya NO, NO, NO, NHCOC,Hs N N hi tt Cae C-C,H, XU XIV XVI + NH, NH. NH> ng (} N N-H N-H r aatl eae No siahe NH, NH, XVII XVII phenanthridine appeared to be accessible and its synthesis was undertaken, although it was realised that the presence of the aryl substituent would not be advantageous to antiseptic activity. Under carefully controlled conditions 2-amino-diphenyl was nitrated in excess concentrated sulphuric acid by nitric STUDIES IN THE PHENANTHRIDINE SERIES. 181 acid to 2-amino-4: 4’-dinitro-diphenyl (X). This method was much more convenient than that of Finzi and Bellavita (1938), who used ethyl nitrate to effect the nitration. These authors stated that by short boiling with acetic anhydride an acetyl derivative melting at 168-9° was formed. On repeating the experiment a product which began to melt at approximately this temperature was obtained, but it was obviously a mixture. By fractional crystallisation from alcohol and aqueous pyridine it was separated into a substance m.p. 175° and another m.p.195°. Both of these analysed correctly for carbon and hydrogen for an acetamido-dinitro-diphenyl and hence it was thought that the original nitration product was a mixture, in spite of its sharp melting point and homo- geneous appearance. But on hydrolysis both yielded the same dinitro-amino- diphenyl, which as a further check was deaminated in each case to 4: 4’-dinitro- diphenyl. The matter was finally cleared up when it was realised that the mono-acetyl and di-acetyl derivatives of 2-amino-4 : 4’-dinitro-diphenyl happened to have very nearly the same percentages of both carbon and hydrogen. Analyses for nitrogen showed that the substance m.p. 195° was the di-acetyl (XII) and the substance m.p. 175° the mono-acetyl (XI) derivative. The facility with which the diacetyl derivative is formed and its stability are remarkable. The benzoyl derivative (XIII) of the amine, prepared by the usual method, was cyclised by phosphorus oxychloride in nitro-benzene solution at 180° (Morgan and Walls, 1939, 1940) to 3 : 8-dinitro-6-phenyl-phenanthridine (XIV) in good yield. Because of its slight solubility, the reduction of this substance was troublesome. The usual methods did not give promising results, the products being red intractable gums. From the gum produced by reduction with phenyl-hydrazine in boiling xylene, there was isolated in small yield a bright red substance, which proved to be a nitro amine. This is probably 3-nitro-6-phenyl-8-amino-phenanthridine (XVI) rather than its isomer, since Muller and Zimmermann (1925) have shown that the reduction of a nitro group by phenyl-hydrazine is more difficult when an amino group is attached to the same benzene nucleus. Attempts to effect complete reduction by working at higher temperatures were unsuccessful. Recourse was then had to catalytic reduction. When a solution of (XIV) in hot dioxane was shaken in an atmosphere of hydrogen in the presence of Raney nickel at atmospheric pressure, the reaction ceased when about 85°, of the theoretical amount of hydrogen had been absorbed. The addition of more catalyst did not cause a resumption of hydrogen uptake. The product of the reduction was a thick dark red gum, but by acetylation this war converted to a solid. The purification of this was tedious. Several recrystallisations removed amorphous by-products, but the product was still a mixture. The fractional crystallisation of it from alcohol was rendered long and difficult by the fact that each substance greatly increased the solubility of the other. Eventually a less soluble diacetyl derivative m.p. 296°, and a more soluble triacetyl derivative m.p. 193°, were isolated in a state of purity. On hydrolysis with the concentrated hydrochloric acid both yielded 3 : 8-diamino-6-phenyl-phenanthridine (X VIT). Dilute solutions of this substance in dioxane and ethyl acetate had a striking green fluorescence, but this was much weaker in alcohol. In concentrated | sulphuric acid solution the fluorescence was blue. A 1° solution of the substance in the minimum of N/15 hydrochloric acid wes very deep red. On the gradual addition of concentrated hydrochloric acid the colour faded to very pale yellow, and in 10N solution it was barely perceptible. From more concentrated solutions of the substance in 10 N hydrochloric acid a white hydrochloride was deposited. The intense red colour of the dilute acid solution is due to the formation of a resonating ior, the chief forms cf which are shown in (XVIII), by the attachment of a proton to the ring nitrogen atom. In concentrated acid, 182 E. RITCHIE. protons become attached to the other nitrogen atoms also, eliminating this resonance and thus destroying the colour. A preliminary test of the antiseptic activity of this substance was made by Dr. H. L. Jensen in the Bacteriology Department of this University, who reported as follows : ‘‘ Bacteriostatic effect of the compound was tested towards a strain of Staphylococcus aureus used for assay of penicillin. The compound was added from solution in minimum quantity of hydrochloric acid to ordinary meat-extract peptone broth (sterilized separately) in concentrations ranging from 0-1 to 0-001. Duplicate test tubes were inoculated with one drop of young Staphylococcus aureus culture dil. 1/10,000 and incubated 48 hours at 37° C. Result : Growth at Concentration Percentage of Compound. 0-1 0-05 0-02 0-01 0-005 0-002 0-001 0 (Control) After 24-hr. Precipitations (—) ey — “ + + After 48-hr. Precipitations (—) — as -j os ++ ‘‘ At concentration 0-02% the medium became slightly cloudy and at the two higher concentrations a heavy yellow precipitate was formed. “3 : 8-Diamino-6-phenyl-phenanthridine thus possesses antiseptic properties, but does not appear highly active, and its apparent protein- coagulating properties would probably militate against its use for many medical purposes.”’ EXPERIMENTAL. 2: 2’-Dichloro-5 : 5’-dinitro-benzylidene-aniline (I). A mixture of 2-chloro-5-nitro-benzaldehyde (1:85 g.) and 2-chloro-5-nitro-aniline (1-7 g.) was kept molten until water was no longer evolved. The product which solidified on cooling was sparingly soluble in the usual solvents but could be recrystallised from cineol or nitrobenzene, from which it separated as pale yellow glistening needles, m.p. 244°. Foufid: N, 12-6%. Calculated for C,,H,O,N,CI,: N, 12:4%. ) . 6-Dihydrophenanthridine (II, R=H). A solution of phenanthridine (80 g.) in concentrated hydrochloric acid (800 c.c.) was heated with tin (200 g.) under reflux until the tin had almost completely dissolved (about four hours). After cooling, the reaction mixture was made strongly alkaline with sodium hydroxide solution and extracted with ether. Removal of the ether left an oil which rapidly solidified. After washing with a little light petroleum the crude product (76 g.) melted at 122°. The pure substance erystallised from aqueous alcohol or benzene light petroleum mixture in lustrous irregular plates m.p. 123°, which had a beautiful blue fluorescence. It was easily soluble in the usual organic solvents and the solutions showed the same strong blue fluorescence. Solutions in dilute mineral acids were colourless and non-fluorescent. Found: C, 86-4; H, 5-9%. Calculated for C,,H,,N: C, 86-2; H, 6-1%. When shaken with warm 3% potassium permanganate it was rapidly and quantitatively oxidised to phenanthridine. The picrate, prepared in alcohol, crystallised in orange-red needles, which began to darken at 180° but did not melt until 238°. Found: N, 13-8%. Calculated for C,,H,,0,N,: N, 13-7%. Nitration of (II, R=H). The following was a typical experiment. A solution of the base (7-2 g.) in concentrated sulphuric acid (100 c.c.) was nitrated at —3° to —5° by the addition of a mixture of nitric acid (4:4 ¢.c., sp. gr. 1-5) and concentrated sulphuric acid (10 ¢.c.). After one hour, the mixture was ‘STUDIES IN THE PHENANTHRIDINE SERIES. 183 poured on to ice when a chocolate coloured product was precipitated. This was mainly amorphous but by fractionation of an alcoholic extract there was finally isolated a bright orange substance (0-2 g.) which crystallised from alcohol in small plates, m.p. 209°. Found: N, 15-6%. Calculated for C,,H,O,N;: N, 15-6%. d-Acetyl-5 : 6-dihydrophenanthridine (III, R=H). _A solution of dihydrophenanthridine in excess acetic anhydride was warmed on the steam bath for twenty minutes and then poured into water. The product crystallised from aqueous alcohol in colourless prismatic needles, m.p. 108°. Found: C, 80:4; H, 5:8%. Calculated for C,;H,,ON: C, 80-7; H, 5-8%. When hydrolysed by boiling with alcoholic hydrochloric acid, it was converted to dihydro- phenanthridine. 8-Nitro-5 -acetyl-5 : 6-dihydrophenanthridine (V). A solution of the acetyl derivative (5:5 g.) in acetic acid (7 c.c.) and concentrated sulphuric acid (12 ¢.c.) was stirred and kept at 0° to 5° whilst it was treated with a mixture of nitric acid (1-3 ¢.c., sp. gr. 1-5) and acetic acid (3 c.c.). After standing at 0° for one hour longer it was poured into ice water, and the precipitate collected. After many recrystallisations from alcohol, benzene and ethyl acetate in turn, there was finally obtained a pure substance (1-5 g.). It crystallised in orange prisms which melted at 176°. Found: C, 67:8; H, 4:3%. Calculated for C,;H,,O,N,: C, 67-2; H, 4:4%. When twice the proportion of nitric acid was used in the above nitration, the product was a thick dark oil. When this was boiled with alcohol (100 c.c.) a yellow crystalline substance (0-7 g.) was gradually deposited. This, which may be (VII), crystallised from benzene alcohol mixture in glistening yellow plates m.p. 218°. Found: C, 57:2; H,3-4%. Calculated for C,,H,,0;N,: C, 57-5; H, 3-5%. 8-Nitro-phenanthridine (VI). A solution of (V; 1-3 g.) in alcohol (100 c.c.) and concentrated hydrochloric acid (30 c.e.) was refluxed for six hours and then distilled to a small bulk. The residue was basified with - ammonia and then extracted with benzene. The oil remaining after evaporating the benzene crystallised when stirred with a little warm methyl alcohol. The pure substance (0:2 g.) crystal- lised from ethyl alcohol ethyl acetate mixture in glistening orange leaflets m.p. 178°. Bound: ©,,69-2; H, 3:7%. Calculated for C),H,O,N,: C, 69:7; H, 3-6%. It was easily soluble in benzene, moderately in ethyl acetate, but slightly soluble in alcohol. 2-Nitro-5-acetyl-5 : 6-dihydrophenanthridine (VIII, R=AH). Fuming nitric acid (10 c.c., sp. gr. 1-5) was added gradually to a solution of (III, R—=H ; 5 g.) in glacial acetic acid (10 c.c.) stirred and cooled in an ice bath. Since no reaction appeared to take place, the reaction mixture was removed from the cooling bath. In a short time, the temperature began to rise rapidly and the product commenced to separate. By suitable cooling the temperature was kept between 20° and 25°. Then after standing at room temperature for thirty minutes ice water was added, and the product collected. The pure substance (4 g.) crystallised from alcohol in pale yellow needles m.p. 181°. Found: C, 67-1; H, 4-5%.. ; Caleulated for C,,H;,0,N,:.C, 67-2; H, 4-4%. _ 2-Nitro-phenanthridine (IX, R=H). A suspension of (VIII, R=H ; 5 g.) in alcohol (200 c.c.) and concentrated hydrochloric acid (100 c.c.) was refluxed for six hours and then the alcohol was distilled off. The residue was basified with ammonia and the precipitated product collected. After washing well with water, 184 BE. RITCHIE. it was boiled with alcohol (100 c.c.) and filtered from the hot solvent. Recrystallisation from pyridine alcohol mixture afforded pale yellow needles (2 g.) which melted at 263°. Found: C, 69-8; H, 3-4%. Calculated for C,,H,O,.N,: C, 69-7; H, 3-6%. It was very sparingly soluble in alcohol, moderately soluble in benzene and easily soluble in pyridine. 6-Methyl-5 : 6-dihydrophenanthridine (II, R—CH,). Reduction of 6-methyl-phenanthridine was accomplished by the method used for the reduction of phenanthridine. The product crystallised from light petroleum in clumps of prisms m.p. 89°, which had a blue fluorescence. Found: C, 85:9; H, 6-5%. Calculated for C,,H,,N: C, 86-1; H, 6-7%. Oxidation to 6-methyl-phenanthridine by potassium permanganate was somewhat slower than in the case of dihydrophenanthridine. The picrate separated as sparingly soluble bright red needles when hot alcoholic solutions of its components were mixed. It began to decompose at about 220° but was not completely melted until 240°. Found: N, 13-4%. Calculated for C,,H,,O,N,: N, 13-22%. 5-Acetyl-6-Methyl-5 : 6-dihydrophenanthridine (III, R=CH,). Acetylation in the usual manner gave an acetyl derivative which crystallised from aqueous alcohol in colourless needles m.p. 102°. Found: C, 80-7; H, 6-2%. Calculated for C,,H,,ON: C, 81:0; H, 6:3%. 2-Nitro-5-acetyl-6-methyl-5 : 6-dihydrophenanthridine (VIII, R=CH;). Nitration of (III, R=CH,; 9 g.) in acetic acid under the conditions used in the nitration of ({1I, R =H) gave a product which crystallised from much alcohol in pale yellow plates (9 g.) melting at 204°. Found: C, 68-3; H, 5:0%. Calculated for C,,H,,0,N,: C, 68-1; H, 5-0%. 2-Natro-6-methyl-phenanthridine (IX, R=CH,). A suspension of (VIII, R=CH,; 8-5 g.) in alcohol (500 c.c.) and concentrated hydrochloric acid (200 c.c.) was refluxed for six hours and then evaporated to a small bulk. After basifying with ammonia it was extracted with benzene. The residue left after removing the benzene was washed with a little alcohol and crystallised from ethyl acetate. It (3-5 g.) separated in long yellow needles which melted at 201°, undepressed by admixture with an authentic specimen, with which it was identical in all respects. Found: C, 70:4; H, 4:5%. Calculated for C,,H,,O,N,: C, 70-6; H, 4-2%. 2-Amino-4 : 4’-dimitro-diphenyl (X). The following procedure was developed as the result of many experiments. Departure from the conditions specified leads to poorer yields of less pure product. A solution of o-xenylamine (8-5 g.) in concentrated sulphuric acid (100 c.c.) was cooled to —3° to —5° and vigorously stirred whilst a mixture of nitric acid (5-5 c.c., sp. gr. 1-5) and concentrated sulphuric acid (10 c.c.) was gradually added. The addition, which took ahout forty-five minutes, was regulated so that the temperature did not rise above —3°. After standing at this temperature for one hour longer the reaction mixture was poured into a large volume of ice and water. The precipitate was collected, washed well with water and then washed by stirring with alcohol (twice, 40 c.c. each time). One recrystallisation from alcohol pyridine mixture gave a pure product (6 g.) as small orange needles m.p. 208°. Found: C, 55:6; H, 3°:4%. Calculated for C,,H,O,N,: C, 55-6; H, 3-5%. It was sparingly soluble in the usual organic solvents and insoluble in dilute mineral acids. a Se ¥ = 3TUDIES IN THE PHENANTHRIDINE SERIES. 185 4:4'-Dimtro-diphenyl. A solution of (X ; 1-3 g.) in concentrated sulphuric acid (5 c.c.) was poured into ice water (60 c.c.) and the finely divided precipitate thus obtained diazotised by the addition of sodium nitrite (0-35 g.) in water (5 ¢.c.). The reaction was very slow and after eight hours about half of the amine had reacted. The mixture was filtered and the filtrate added to a solution of sodium hypophosphite (3 g.) in water (10 ¢.c.) at 0°. After standing overnight it was heated to boiling and the product collected. Crystallisation from alcohol and then acetic acid gave pale yellow needles m.p. 231°, undepressed by admixture with an authentic specimen. i Acetylation of (X). 2-Amino-4 : 4’-dinitro-diphenyl (10 g.) was refluxed for five minutes with just sufficient acetic anhydride to dissolve it at the boiling point. After pouring into water, the product was collected, washed, dried and dissolved in boiling alcohol (500 ¢.c.). The substance which crystal- lised on cooling was recrystallised from the same volume of alcohol and was then nearly pure (6 g.). When pure it separated from alcohol in nearly colourless irregular flattened needles m.p. 173°. Analysis showed it to be 2-acetamido-4 ; 4’-dinitro-diphenyl (XJ). Pound), bo°6; HH, 3:5; N, 14:3%. Calculated for C,,H,,0,N,: C, 55:8; H, 3-6; N, 14-0%. The first filtrate was evaporated to a small bulk (about 70 c.c.), cooled, and the product which separated collected and dissolved in the second filtrate, which was then evaporated to 50 c.c. The fraction which then crystallised was repeatedly recrystallised from aqueous pyridine, yielding finally the pure dzacetyl derivative (XIJ) in glistening plates m.p. 195° (2 g.). Round aor.) Ho 3-1, |N, 12-5%. Calculated for C,,H,,0,N,: C, 55:9; H, 3-8; N,. 12-3%. ; Both substances yielded 2-amino-4 : 4’-dinitro-diphenyl on hydrolysis. 2-Benzamido-4 : 4’-dinitro-diphenyl (XIII). A solution of the amine (15-6 g.) in dry pyridine (50 ¢.c.) was treated with benzoyl] chloride (7 c.c.) and heated on the steam bath for one hour. Then alcohol (150 ¢.c.) was added to the hot reaction mixture. The nearly pure product (20 g.) began to crystallise immediately. Recrystal- lised from alcohol pyridine mixture, it formed very pale yellow needles which melted at 236°. It was sparingly soluble in the usual solvents. Founda, 62-8); TH, 3-64. Calculated for C,,H,,0,N,:. C, 62-8; H, 3-6%. 3: 8-Dinitro-6-phenyl-phenanthridine (XIV). A solution of the amide (16 g.) in pure dry nitrobenzene (20 c.c.) and phosphorus oxychloride (10 c.c.) was refluxed in an oil bath at 180°. After twenty hours the slow evolution of hydrogen chloride had ceased. The reaction mixture was poured into excess dilute ammonia and steam distilled to remove nitrobenzene. The product was collected, washed with warm alcohol to remove a soluble amorphous by-product, and then recrystallised from alcohol pyridine mixture. The pure substance (10 g.) separated in very pale yellow needles m.p. 261°. It was sparingly soluble in the usual solvents. Hound +, 66+2; H, 3-2%.. Calculated for C,,H,,0,N,: C,, 66-1; H,.3-2%. Reduction by Phenylhydrazine. - When the above phenanthridine (2 g.) and phenylhydrazine (4 g.) in xylene solution (40 c.c.) were refluxed, water soon commenced to separate. After four hours the xylene was distilled off . under reduced pressure and the dark red residual oil washed several times with light petroleum. When warmed with alcohol it slowly became partially crystalline and eventually a product (0-3 g.) forming bright red needles m.p. 281-3° was isolated. It is probably (XVI). Meend)-: C, 72:1; HH, 4:2; N, 13-5%.. Caleulated for C,,H,,0,.N,: C, 72-4; H, 4-2; N, 13-3%. 186 E. RITCHIE. Catalytic Reduction of (XIV). A solution of (XIV; 25 g.) in hot dioxane (1,000 c.c.) was shaken with Raney. nickel in an atmosphere of hydrogen at normal pressure. Absorption of hydrogen was rapid at first and when a little more than half of the theoretical amount had been taken up, the solution acquired a strong bright green fluorescence. From this point absorption was slow and stopped completely when about 85% of the theoretical amount had been absorbed. Addition of more catalyst did not result in further action. After filtering off the catalyst, the solvent was removed under reduced pressure and the dark red gum which remained acetylated by warming with excess acetic anhydride on the water bath. The light yellow solid was recrystallised three times from a large volume of aqueous pyridine, twice from much aqueous alcohol, and then fractionally crystallised from alcohol. After a prolonged fractionation two pure substances were isolated. The less soluble fraction (14 g.), which formed a pale yellow crystalline powder m.p. 296°, was evidently a diacetyl derivative. It was sparingly soluble in the usual solvents. Found: (C; 74:8; H, 5-0; N, 11-4%. Calculated for’ C,.H,,O,N,: @, 147650 aera N31 449%. The more soluble fraction (3-5 g.), a triacetyl derivative, crystallised in pale yellow prisms m.p. 193° with frothing. Kound: N, 10:5%. . Calculated for C,;H,,0O,N,: N; 10°3%. 3 : 8-Diamino-6-phenyl-phenanthridine (X VIZ). When either acetyl derivative was heated with excess concentrated hydrochloric acid, it rapidly dissolved to a yellow solution. Ina short time a white hydrochloride began to crystallise out, but refluxing was continued for three hours. On dilution, the precipitate dissolved readily to an intensely red solution from which the product was precipitated by the addition of ammonia (80% yield). Recrystallisation from aqueous alcohol atforded pale yellow irregular prisms melting at 198°. Found: C, 79:5; H, 5-3:; N, 14:6%. Calculated for C,,H,.N.; Cy 60-0 - ee: IN, 14 oe Dilute solutions of the substance in dioxane or ethyl acetate had a strong bright green fluorescence. In alcohol the fluorescence was much weaker. It dissolved readily in very dilute sulphuric or hydrochloric acid to an intensely red solution, the colour of which faded to pale yellow on the addition of concentrated acid. REFERENCES. Bell, F., 1928. J. chem. Soc., 2770. Browning, C. H., Gulbransen, R., and Robb, J. V. M., 1938. J. chem. Soc., 389. Browning, C. H. , Morgan, G. T., Robb. Ji. Vs M.. and Walls, L. P., 1938. J.Path. Bact., 46, 203. De Gaouck, V., and Le Howe BR. ds. Ws 1938. J. chem. Soc., 741. ——— - —— 1939. J. chem. Soc., 1392. Finzi, C., and Bellavita, V., 1938. Gazz. chim. Ital., 68, 77. Kondo, H., and Uyeo, S., 1935. Ber. dtsch. chem. Ges., 68, 1756. —______-________—. 1937. Ber. dtsch. chem. Ges., 70, 1087, 1094. Morgan, G. T., and Walls, L. P., 1932. J. chem. Soc., 2225. ——_—_—_— 1938. J. chem. Soc. 389. 1939. Brit. Pat.,: No. 511, 353. eo —— 1940. Brit. Pat., No. 520, 273. Muller, E., and Zimmermann, G., 1925. J. prakt. Chem., 3, 277. Pictet, A., and Ankersmit, H. J., 1891. Lvebig’s Ann., 266, 138. Scarborough, Hi. “As; and “Waters, “Wir A inS2 7.” f. chem.Soce,, 89: ACKNOWLEDGEMENTS. The author gratefully acknowledges the assistance of Mrs. D. M. Buckley, B.Se., and Miss J. Fildes, B.Sc., who carried out most of the analyses recorded. He is also indebted to the Commonwealth Research Grant Committee of this University for a grant which helped to defray the cost of some of the chemicals. Department of Chemistry, The University of Sydney. THE PHYSICS OF RUBBING SURFACES.* By F. P. BOWDEN, Sc.D., Council for Scientific and Industrial Research, The University, Melbourne. With Plates VIII-XIII and twenty-one text-figures. PART I. INTRODUCTION. In this lecture we shall be dealing with a very old and very unfashionable branch of natural science—friction—and we wish to discuss some of the physical processes that occur when two solids are rubbed together. The title is a compre- hensive one but this is in no sense a general summary of the field, and it does not do justice to other workers. There are many obvious gaps, for example no mention is made of frictional electricity. I thought, however, that the most satisfactory way would be to discuss certain aspects of the subject from my own point of view, and to deal with some of the work of my associates and colleagues. The more recent work was done under the Council for Scientific and Industrial Research in the Chemistry Department of the University of Melbourne, and the earlier work in Cambridge. At the outset I must offer an apology. For the past five years we have been concerned almost entirely with war problems and the basic work we have been able to do has been fragmentary and fitted into odd corners. For the same reason the pressure of immediate work has made it impossible to give the proper thought and preparation that the Liversidge Lecture warrants. It would be interesting to attempt a lecture on the chemistry of rubbing surfaces, but unfortunately our information about this is very scanty. Although we know a great deal about the influence of heat, of light, and of electricity in stimulating molecules to react, we still know comparatively little about the influence of mechanical forces on reactivity and of the effects produced when we seize a molecule by the head and the heels and tear it in half. We shall confine our attention to some of the physical processes that occur when one solid slides over another. There is a resistance to motion which we call friction. What is the mechanism of that frictional force, and from the point of view of a molecule sitting on the surface, what is really happening? We may ask a few simple questions and then endeavour to answer them by direct experiment. To begin with we wish to ask three questions: (i) what is the real area of contact between the solids? (ii) What is the surface temperature of the rubbing solids ? (iii) What is the nature of the surface damage ? THE REAL AREA OF CONTACT BETWEEN SOLID SURFACES. It is, of course, a very difficult matter to prepare surfaces which are really flat. Even on carefully polished surfaces hills and valleys which are large — compared with the dimensions of a molecule will still be present. The upper * Two Liversidge Research Lectures delivered on October 17 and 18, 1944, at the Chemistry Department, University of Sydney, arranged by the Royal Society under the terms of the Liversidge Bequest. ls 188 F. P. BOWDEN. surface will be supported on these irregularities and large areas of the surfaces will be separated by a distance which is great compared with the dimensions of a molecule. We do not know very much about the size of these small irregularities nor the degree of flatness of the surfaces. Optical methods cannot reveal irregularities much smaller than one-half to one-tenth of a wave length of light. Although the techniques of grinding and polishing have advanced in the past few years, it is still a difficult matter to prepare surfaces of appreciable size which are flat to within one or two thousand angstroms. Since the range of molecular attraction is only a few angstroms, we may expect that the area of intimate contact, that is the area over which the surfaces are within molecular range, will, even for very carefully prepared surfaces, be quite small. A new and powerful method for studying surface irregularities and surface structure is the electron microscope, and recent stereoscopic pictures taken by Heidernich and Matheson (1944) show in some detail small scratches ca. 250 A. deep on polished metal surfaces. In general the surfaces used in engineering practice are less flat and the surface irregularities present on them are, in terms of molecular dimensions, enormous. ‘These irregularities may be seen by cutting a section at right angles to the surface and examining it with high power microscope, but a more revealing method is to cut a section at an oblique angle to the surface. This has the advantage that it magnifies the irregularities in the vertical direction and leaves the horizontal magnification unchanged. Taper sections prepared in this way by Mr. Moore which show some characteristic contours of surfaces finely machined and also ground with abrasives of varying degrees of fineness are given in Plate VIII, Figs. 1-5. It is seen (Fig. 5) that even with the finest abrasive the surface irregularities are of the order of 0-1 yu, i.e. 10-° cm. If the surfaces are polished the effect is to cause the summits of the peaks to flow into the valleys so that the contour resembles rolling downs rather than rugged alpine peaks. But again the surfaces will touch on the summits of the hills and the area of intimate contact will be small. Some knowledge of the real area of contact between solid surfaces is essential for our purpose and its determination is a matter of some experimental difficulty. Fortunately, an approximate estimate of it in the case of metals can be made by measuring the electrical conductance across the surface of the metals when they are in contact. Fig. 1.—The current flow through a con- Fig. 2.—The deformation of a spherical surface striction of radius a@ in a metal conductor. resting on a hard (undeformable) plane The behaviour is similar if two metal surfaces surface. The current flow and _ electrical make electrical contact over an irregularity of conductance are similar to those described radius a. The electrical conductance A of an Hig) ht such a junction is given by A=2ah. THE PHYSICS OF RUBBING SURFACES. 189 In the simple case of metal surfaces which make electrical contact over a surface irregularity of radius a (see Fig. 1) it can be shown (Maxwell, 1873 ; Holm, 1929; Bowden and Tabor, 1939) that the conductance A is given by Oe RN ee, (1) when A=the specific conductivity of the metal. A measurement of the con- ductanee will therefore enable us to estimate the area of contact. Variation of Area of Contact with Load. Hertz (1881), in his classical paper on the elastic deformation of solid surfaces, calculated how the area of contact between curved surfaces should depend upon the load. He showed that, if elastic deformation is assumed, the theoretical radius of contact a between a spherical surface resting on a plane surface of the same material (see Fig. 2) is given by $ a=const. (= Bae. Bie a Cee Dr oto a Uae (2) where W is the load between the surfaces, 7 is the radius of curvature of the Spherical surface and H is Young’s Modulus. From equations (1) and (2) INE COMG Ue Wade eet He es hr oe anh (3) so that if we assume elastic deformation of the metals in the region of contact the conductance should be proportional to the cube root of the load. There is,. however, another possibility. We may consider that the metal in the region of contact will flow plastically under the applied load until the cross section of the contact is sufficient to support the load. In this case Rees 7 Fol eR ee a RL ec ee eee (4) where f is the maximum pressure the metal can withstand without flowing. Combining this with equation (1) it follows that Ne COMMS MV ery ahah RG Aish tie ees hese ot wd 6 (5) i.e., the electrical conductance should be proportional to the square root of the load. Some experiments were carried out by Dr. Tabor to investigate this point and results obtained for a number: of different metals in contact are shown in Figs. 3 and 4. In these experiments two crossed cylinders were used, since geometrically this is the same as a spherical surface resting on a flat one and it is more convenient experimentally. It is clear that the conductance varies in an orderly manner with the applied load. Moreover, an examination of the slope of the lines (see Fig. 4) shows that the conductance is proportional to the square root of the load, 1.e. equation (5) is obeyed, and hence the deformation of the surfaces in the region of contact is mainly plastic. Some actual values of the area of contact between cylinders of steel and between cylinders of silver at different loads are given in Table 1. TABLE 1. Area of Contact between Cylinders of Steel and Silver at Various Loads. Area of Contact cm.? oe (From Electrical Conductance.) s Steel. Silver. 0-5 — 0-0002 1 0-00013 — 5 0-00061 0-002 50 0-0045 0-018 500 0-042 0-15 190 F. P. BOWDEN. 1O 10 COPPER 1 rove io STEEL ANTIMONY io” BISMUTH ELECTRICAL CONDUCTANCE (Mnos) ( w> souW) SAILIAILINGNOD D141353dS 4O 319V1 CARBON CARBON 103 tOu LOAD (Gms Wt) Fig. 3.—The variation of electrical conductance with load for crossed cylinders of various materials. The conductance varies in an orderly manner with load and the position of the curve for each material is in the same order as the specific conductivity of the material. ELECTRICAL CONDUCTANCE MPOS | Observed Im — —-assum:ng Plastic Deformation Il —-—-Assuming Elastic Deformation 10° Tow ne} LOAD (Gms wr.) Fig. 4.—Variation of electrical conductance with load for crossed cylinders of copper and steel. The broken lines are calculated assuming plastic and elastic deformation. The observed values show that the deformation is essentially plastic. THE PHYSICS OF RUBBING SURFACES. 191 Area of Contact between ‘‘ Flat ’’ Surfaces. It is interesting to compare the conductance across flat surfaces with that across curved surfaces at the same load, and Fig. 5 shows such a comparison. Mhos, ELECTRICAL CONDUCTANCE . 10 10 [oO 10 LOAD (Gms.wet.) Fig. 5.—Variation of electrical conductance with load for steel surfaces. Curve I refers to curved surfaces, Curve IT (a) to flats 0-8 sq. cm. in area, Curve IT (b) to flats 20 sq. cm. in area. Curve I is for curved surfaces of steel, Curve II (a) for flat steel surfaces 0-8 Sq. cm. in area, Curve II (b) for similar surfaces 20 sq. cm. in area. The surfaces were flat to a few fringes. The values for flat surfaces do not lie on a straight line but it is clear that the values of the conductance are not very different from those observed with curved surfaces. The possible area of contact between the flat surfaces is of course enormously greater than that of the curved, but the conductance is of the same order of magnitude. At a load of 5 kg., for example, the area of contact between the curved surfaces is ca. 6 x10-4 sq. em. If nearly all the surface of the 21 cm. flat were in contact the area of contact would be ca. 30,000 times as great and the conductance would be correspondingly increased. Experiment shows, in fact, that it is only twice as great. It is thus clear that only a very small fraction of the flat surfaces can be in intimate contact. It will also be noted that the conductance of both sizes of flats is almost the same although their apparent areas stand in the ratio of 30:1. We can conclude that the conductance is independent of the apparent area of the surfaces. It 192 F. P. BOWDEN. depends mainly on the load. Hxperiments also showed that the conductance was little influenced by the degree of roughness of the surfaces. The surface of the steel cylinders was rubbed with a coarse file so that it was as rough as possible and the conductance was compared with that of a finely ground surface. No appreciable difference between the two was observed. It is clear from these simple experiments that the area of intimate contact between. solid surfaces is very small indeed. It varies with the pressure, but for flat steel surfaces it may be less than one-ten thousandth of the apparent area. The real area of contact is not greatly affected by the size of the surfaces nor by the shape and the degree of roughness of the surface. It depends mainly on the pressure. The general behaviour is consistent with the view that the surfaces are held apart by small irregularities. This means that even with lightly loaded surfaces the local pressure at these small points of contact is very high and may be sufficiently great to cause steel to flow plastically. Although the stresses will cause elastic deformation of the metal in the vicinity of the points of contact the experiments suggest that the summits of irregularities on which the bodies are supported flow plastically and are crushed down until their cross section is sufficient to enable them to support the applied load. The fact that the real area of contact is so small has important practical implications. Even when loads of only a few hundred grammes are applied to the surfaces the local pressure between them may be sufficiently great to cause the flow of metal. When large flat surfaces are used it does not mean that the real pressure is much less, but merely that the points of contact are more widely distributed. As we Shall see later, this intense pressure may cause an actual welding together at the tiny points of contact and so produce small metallic junctions between the surfaces. The pressure between the surfaces in the regions where contact occurs is determined primarily by the flow pressure of the metal itself and is only influenced to a secondary degree by the shape and size of the surfaces and by the load which is applied. It is interesting to note that even when the surfaces are lubricated with a boundary film of a good lubricant, similar conditions hold. The load will again be carried by the surface irregularities and even for light loads the local pressure on the lubricant film will be very high. In the case of mild steel, for example, it will be ~104 kg. cm.?. Unless the lubricant molecule possesses an active group capable of attaching it firmly to the surface the local high pressure will force it from underneath the points of contact. In this sense all good lubricants must act as extreme pressure lubricants. We would now like to consider in some detail the physical processes that occur when we set these solids in motion and cause one to slide over the other. We find that an appreciable tangential force is necessary to start them moving (static friction) or to keep them sliding (kinetic friction). This frictional resistance occurs, of course, over the small localized regions where the solids are actually in contact, and we will enquire into its mechanism. THE SURFACE TEMPERATURE OF RUBBING SOLIDS. The energy lost during sliding is dissipated mainly in the form of heat and the first query is, what will be the temperature of the surfaces? Quite primitive calculations of the amount of heat which is liberated and of the rate at which it is conducted away suggest that the temperature rise of the surface layers may be high. If we endeavour to measure this by embedding thermocouples in the solids near the surface we find that the rise is very small, but this is mainly because we cannot get close enough to the surface. An obvious method is to use the surfaces themselves as a thermometer. This can be done by making = eee ae a en, a THE PHYSICS OF RUBBING SURFACES. 193 them of two different metals, and using the rubbing contact itself as a thermo- couple. A measurement of the electromotive force generated on sliding them provides a record of the surface temperature. It is apparent that the electrical contact and the friction occur at the same points where the surfaces touch so that the measurement gives information about the temperature of the surface layer of the metal where they are rubbing. Such measurements confirm in a very striking way the existence of local high temperatures at the points of contact of the rubbing surfaces. As would be expected, the temperatures reached depend upon the load, the speed of sliding and the thermal conductivity of the metals. Also the temperatures fluctuate very rapidly during sliding and it is necessary to use an instrument of high frequency such as a cathode ray tube or high frequency galvanometer in order to record them. Fig. 6 shows the maximum temperature reached when small eylinders of gallium, Wood’s alloy, lead, and constantan are slid on a steel surface. It will be seen that with the lower melting metals we readily reach their melting point and the temperature does not rise above this. With constantan, temperatures of the order of 1000°C. are reached. These temperatures are | GALLIUM 1IOO 200 SOO 400 380) 600 700 SPEED | IN CM/SEC. Fig. 6.—Maximum temperatures reached when small cylinders of gallium, Wood’s metal, lead and constantan are slid on a steel surface. The temperature cannot exceed the melting point of the metal. 194 F. P. BOWDEN. confined to the very thin surface layer and the mass of the metal appears to be quite cool. The rapid and intense nature of the temperature fluctuations is shown strikingly in the cathode ray trace recently taken by Mr. Stone and Mr. Tudor (Fig. 7). It will be seen that the very high temperature flashes of 1000° C. may last only for a few ten-thousandths of a second. Each div. ee Constanten slider on lapped steel surface. Load 500 gms Speed of sliding 300cmsf6ec. Fig. 7.—Cathode ray trace of thermal emf. developed between a constantan slider on a lapped steel surface. The temperature flashes are extremely high and of very short duration. It is interesting to note that if the surfaces are flooded with water the local high temperatures may still occur. If the surfaces are lubricated with a boundary film of good lubricant the surface temperatures are greatly reduced but the experiments show that localized metallic contact may still occur through the film and the surface temperature at the summits of the surface irregularities may Still be sufficiently high to cause volatilization and decomposition of the lubricant film. / The influence of the thermal conductivity of the metals on the surface temperature is shown in Fig. 8. It will be seen that, with the metals of lower thermal conductivity such as bismuth, the temperature rise is considerably greater. Constantan — 4 Wood's metal Bi (T—T,)°C 0 2, + 6 8 }/Ki Fig. 8.—Temperature rise of rubbing surfaces as a function of their thermal conductivity K (load 32 g., velocity 20 cm./sec.). Visual Observation of Hot Spots. We should expect that these high temperatures would be reached very much more readily on glass and other non-conducting solids. Unfortunately, ‘OO! sec. s THE PHYSICS OF RUBBING SURFACES. 195 the thermocouple method cannot be used with these solids but recently Mr. Stone has shown the existence of the hot spots by visual means. If polished surfaces of glass or quartz are used and the apparatus so arranged that a clear image of the rubbing surfaces can be seen it is found that when sliding starts a number of tiny stars of light appear at the interface between the rubbing surfaces. The points of light are reddish in colour at low speeds and become whiter and brighter as the speed or the load is increased. It is clear that they correspond to small hot spots on the surface and their position shifts continuously as sliding proceeds. They can be recorded photographically. The method is not quanti- tative but, by making one of the surfaces of metal and using alloys of different melting points, it is possible to fix approximately the temperature at which hot spots first become visible. Experiments suggest that this temperature is about 500°C. These hot spots may occur at very low sliding speeds. For example, with constantan sliding on glass under a load of 1 kg. hot spots appear when the sliding speed is as low as 30 cm. per sec. An example of the photo- graphic method of recording the light from the hot spots is shown in Piate IX, Fig. 1. The surfaces (steel on glass) are allowed to run continuously for two minutes in a circular track so that a cumulative exposure of the hot spots occurs. The circles of decreasing radius mean decreasing sliding speed. The lowest speed at which the hot spots are recorded under the conditions of the experiment is 70 cm./sec. The fact that these high temperatures occur so readily is of some interest and it is suggested that they may play an important part in a number of processes associated with the rubbing of solids. Before considering its bearing on friction we may enquire what part it plays in the polishing of solids. POLISHING AND THE SURFACE FLOW OF SOLIDS. The usual method of polishing surfaces is to rub them together with a fine powder between them. By this process a rough surface having visible surface irregularities is changed into one where the irregularities are invisible. If the surface gives specular reflection the height of these irregularities will be less than half a wave length of visible ight. The classical work on polishing is that of Sir George Beilby (1921), who showed that the top layer of the polished solid is different in structure from that of the underlying material. It has lost its obvious crystalline properties and has apparently flowed over the surface, bridging the chasms and filling up the irregularities in it. The mechanism of the process has been a subject of discussion for many years. Newton, Herschel and Rayleigh considered that polishing was essentially due to abrasion. Beilby’s view is that it 18 a surface tension effect—that when the polisher tears off the surface atoms the layer below this ‘‘ retains its mobility for an instant and before solidification is smoothed over by the action of surface tension forces ’’. As we have seen, however, the frictional heat generated at the rubbing surfaces may easily raise the temperature to a high value and this suggests that the local thermal softening or actual melting may play an important part in the polishing process. The action of a typical polisher may be represented diagramatically in Fig. 9. The polishing particles of rouge or alumina are embedded in a block of pitch and rubbed on the specimen in the presence of a liquid such as water. At the points of rubbing contact between the polishing powder and the Specimen, hot spots will occur which will cause a local surface softening or melting of the specimen. The melted or softened solid would be smeared over the surface and would quickly solidify to form the Beilby layer. We may perform a simple experiment to test this hypothesis. If polishing is due primarily to a mechanical abrasion and wearing away of the specimen we may expect the relative hardness of the specimen and of the polisher to be of major importance. If, however, it is due to surface melting it is the 196 F. P. BOWDEN. relative melting points which will be the determining factor. If the polisher melts or softens at a lower temperature than the specimen it will melt or flow first and will have comparatively little effect on the specimen. The result of rubbing Wood’s metal (M.P. 75°) with a camphor block (M.P. 178° C.) is shown in Plate IX, Fig. 2. Although the Wood’s metal (hard- ness ca. 2) is very much harder than the camphor, it melts at a lower temperature and it will be seen that surface flow and polishing of the alloy occurs. On the other hand camphor will not polish tin (M.P. 232° C.), lead, white metal or zinc, which melt at a higher temperature. A polisher using a powder of oxamide (M.P. 417°) will readily cause flow of all these metals but does not produce any effect on speculum metal (M.P. 745°) (see Plate IX, Fig. 3) or copper (M.P. 1083°) which melt at temperatures well above 417°C. Lead oxide (M.P. 888°) will polish speculum metal and all metals melting below it, but has little effect on nickel and molybdenum, which melt above it. These in turn are readily polished by the high-melting oxides such as chromic oxide, stannic oxide, ete. POLISHING ~ BLOCK Fig. 9.—Diagrammatic representation of a typical polisher. Similar results are obtained with glasses, quartz and some non-metallic crystals ; calcite, for example (Plate IX, Fig. 4), shows no flow on cuprous oxide which melts a little below it, but is readily polished by zine oxide which melts above it. It is well known that the mechanical strength of many metals and solids falls to a very low value at temperatures well below the melting point. The rounding of sharp metal crystals and the low value of tensile strength, hardness, etc., at these temperatures show that metals may lose their rigidity and resistance to shear at comparatively low temperatures. For such solids, surface flow would be expected to take place at temperatures well below the melting point and experiment shows that, in many cases (e.g. gold) this can occur. The rate of flow and polish is, however, very much less and may take hours, instead of the few minutes required by a high melting polisher. The experiments provide strong evidence, not only that high local temper- atures occur, but that they play a large part in the process of polishing. In many cases the frictional heat will raise the temperature to a sufficiently high value to cause a real melting of the solid at the points of sliding contact. The molten solid will flow or will be smeared on to cooler areas, and will very quickly solidify to form the Beilby layer. Polishing under these conditions is rapid. THE PHYSICS OF RUBBING SURFACES. 197 If the sliding is gentle or the melting point of the polisher is low, the surface of the solid may not reach the temperature of melting. Polish and surface flow may still occur under these conditions, provided the temperature reaches a point at which the mechanical strength of the solid is sufficiently low for it to yield under the applied stress or under surface tension forces. Polishing under these conditions is a slow process. The relative hardness of solid and polisher as normally measured at room temperature is comparatively unimportant. This is shown clearly in the case of Wood’s alloy and tin on camphor, or speculum metal and nickel on lead oxide. The harder metal of low melting point is polished, while the softer metal of higher melting point hardly flows at all. Similarly zine oxide which is comparatively soft (Mohs’ hardness 4) readily polishes quartz (Mohs’ hardness 7). The amount of surface flow is governed, not by the properties of the solids at room temperature, but by their relative mechanical properties at the high temperature of the sliding surfaces. THE MECHANISM OF SLIDING ON ICE AND SNOW. Another phenomenon where surface melting may play a part is in the sliding of solids on ice and snow—in skating or skiing. It is well known that the friction under these conditions may be remarkably low (u=ca. 0-03). The suggestion has often been made that in skating or skiing the surfaces are lubricated by a layer of water formed by pressure-melting (e.g., Reynolds, 1901), but few experi- ments have been made to support or to disprove the suggestion. Experience shows that skis slide quite readily on snow at —20°C. Caleula- tions of the pressure necessary to cause melting at this low temperature suggest that it is unlikely that they would be attained. On the other hand calculations of the amount of heat liberated by frictional heating as the ski moves forward a small distance show that it is sufficient to warm up the snow and to melt an appreciable water layer. Some experiments were carried out at the Jungfraujoch Research Station in Switzerland, to determine whether a water layer is formed at all, and if so, whether it is due to pressure melting or to frictional heating. Measurements of the electrical conductivity between metallic electrodes on the bottom surface of miniature ski sliding on salty ice indicated that at low temper- atures the surface melting occurred only at localized areas but at temperatures near 0° C. a continuous water film was formed. The effect of temperature on the friction of different solids sliding on ice surfaces js shown in Fig. 10. It will be seen that the friction increases markedly as the temperature falls and at a temperature of —140° C. it is some five or six times as great as itis at 0° C. The value for the coefficient of friction (ux, =0-1) at these low temperatures is of the same order of magnitude as that observed on other crystalline solids such as calcite. The large influence of temperature on the friction of ice is in marked contrast to the behaviour of most other solids where temperature has only a small influence. It emphasizes the anomalous behaviour of ice and supports the view that the low friction is due to a lubricating water layer. As the temperature falls it becomes increasingly difficult for a water layer to be formed and the friction rises. It is of particular interest to determine the influence thatthe thermal con- ductivity of the ski has on the friction at low temperatures. If sufficient pressure is applied to the ice to lower the melting point to the actual temperature of the ice, it is, of course, capable of melting. An appreciable quantity cannot melt, however, unless heat is supplied from some source at a temperature higher than the pressure melting point equilibrium. Both the heat capacity of the ice and its thermal conductivity are small and this heat can most readily be supplied from some outside source. If the temperature of the atmosphere is higher than either of the ice surfaces, it could be supplied by conduction from the air. 198 F. P. BOWDEN. Under these conditions we should expect that the friction of a good thermal conductor would be less than that of a bad one. The friction of a brass ski on cold ice should be less than that of an ebonite one. If, however, the lubricating film is formed by frictional heating, the converse will be true. The frictional heat is liberated at the interface between the sliding surfaces, and if the ski is a good thermal conductor, the heat will be carried away rapidly and less will be available for surface melting. On this view the friction of a brass ski on cold ice should be greater than that of an ebonite one. Fig. 10 shows the results obtained using a miniature ski of brass and of ebonite. : At temperatures near 0° C. the frictions of both skis were the same. At lower temperatures, however, the results showed that the friction of the brass was considerably greater than that of the ebonite. The lower the temperature the more pronounced this difference usually became. These results provide evidence that the frictional heating plays an important part in the formation of the water film. 0:04 fe) o°c - 20° - 40° -60° -80° MEAN TEMPERATURE —» Fig. 10.—Effect of temperature on the friction of brass, ebonite and ice sliding on ice. The friction increases markedly as the temperature falls and is lower for the slider possessing the lower thermal conductivity. The observations have an interesting bearing on sledging and skiing. No quantitative measurements of the friction of sledges seem to have been published, but there is general agreement that the friction increases at low temperatures. Many arctic explorers (Wright, 1924, p. 44; J. M. Scott, 1933, p. 273; Cherry- Garrard, 1922, pp. 456-7) have recorded that at very low temperatures, —30 to —40° C., the friction between the snow and the runners became so great that the sensation was that of pulling a sledge over sand. Wright, summarizing the conclusions of the Scott Polar Expedition of 1911-13, says: ‘¢ Quite apart from any question of the hardness of the snow, however, the surface temperature has an important influence. Our opinion was that the friction decreased steadily as the temperature rose above zero Fahrenheit (—18° C.), the presence of brilliant sunlight having an effect, which was more than a psychological one, on the speed of advance. Below zero Fahrenheit (—18° C.) the friction seemed to increase progressively as the temperature fell, as if a greater and greater proportion of the friction were due to relative move- THE PHYSICS OF RUBBING SURFACES. 199 ment between the snow grains and less to sliding friction between the runner and snow.’’ This steady increase in friction as the temperature of the ice or snow falls is clearly shown in Fig. 10. The effect is less marked at very low temperatures, and it is probable that below —40° C. very little surface melting occurs under these conditions. The influence of the thermal conductivity of the sliding body on the friction, shown in Fig. 10, is also borne out by practical experience. It will be seen from this figure that at low temperatures the friction of a good thermal conductor is considerably greater than that of a poor one. Nansen (1898, pp. 445-6) compared two sledges, one having nickel plated runners and the other maple runners. The temperature was low—the actual value is not given but the mean temperature during that month was —36-8°C. (—34-2° F.). He found that the friction of the metal was higher: ‘‘ The difference was so great that it was at least half as hard again to draw a cee on the nickel runners as on the tarred maple runners.”’ The thermal conductivity is also era in skiing. Nowadays most skis are fitted with brass or steel edges, although sometimes vulcanite or composition edges are used. The friction measurements show that the latter should be faster at low temperatures. If metal must be used, one of low thermal conductivity such as German silver or constantan should be better. THE SURFACE DAMAGE OF SLIDING METALS. Returning again to the sliding of metallic surfaces, there are, as we have seen, two major experimental observations. First that the area of contact between them is very small so that the pressure in the local regions of contact is very high and is sufficient to cause plastic flow of the metal, and secondly that at sliding speeds frequently used in practice the surface temperatures may rise to very high values. The third point we need to investigate is the type of inter- action between the moving surfaces and the physical changes which occur in them during sliding. Careful examination shows that some surface damage always occurs even with lightly loaded, well lubricated surfaces. The nature of the damage depends upon the combination of metals which is used. Plate IX, Fig. 5, shows the tracks formed (i) on a softer metal when a hard one slides over it, (ii) on a harder metal when a softer one slides over it, (iii) when similar metals are used. In these experiments a heavy load was used and the surfaces were unlubricated so that the effect could be seen more clearly. It will be seen that in (i) a ploughing out and tearing of the softer metal has occurred ; in (ii) the harder surface is comparatively little damaged but the softer metal has welded on to it and remains adhering to the surface. With similar metals (iii), which are homogeneous, the damage is more profound and it is this combination which always gives the highest friction. (It should be emphasized that in these experiments and in. the ones described below the sliding speed was very slow (a few millimetres per second), so that the temperature rise is inappreciable.) Highly magnified taper sections made by Mr. Moore which represent a cross section of these three types of tracks are shown in Plate X, Figs. 1-4. Plate X, Fig. 1 shows a track characteristic of type (i) made on a steel surface. The localized nature of the damage is at once apparent. |500) wv 2 Te ENOOO! © ie vu a oo O@ Track Width dmm. Fig. 13.—Graph showing the frictional force as a function of the track width for a cylindrical slider (Curve 1) and for a spade (Curve 2) sliding on indium. The difference between Curves 1 and 2 gives the shearing term S. 204 F. P. BOWDEN. Curve 2 (P) the results for a spade, and Curve 3 the difference between them, gives the shearing term S. It is then possible to compare these experimental values of P and S with the theoretical ones calculated from the flow pressure and the shear strength of indium, and such a comparison shows that the results are in reasonable agreement with the theory, for indium, lead and other metals. It will be noted that the ploughing term P is small compared with the shearing term S. Elastic and Plastic Deformation. Frictional Hysteresis. The reality of the welding process is clearly demonstrated when frictional measurements are made with soft plastic metals. With lead or indium, for example, sliding on steel, the value of the friction is very variable and depends on the previous history. If the load is first applied and then decreased or removed altogether, the surfaces continue to ‘‘ stick ’’ together and the tangential force necessary to cause sliding remains high, although there is no normal load applied between the surfaces. This is not observed on harder, more elastic, metals. For such metals, the frictional force decreases when the load is decreased and the effects are reversible. It is suggested that the elastic deformation of the metals is responsible for this. At the actual region of contact where the local pressures are very high, the deformation of the metals will be plastic. Near this region, however, where the pressures are smaller, the metals will be elastically deformed. When the load is decreased, these stresses will be released and the small movement that results will serve to break the metallic junctions. The area of contact at any given time will, therefore, be determined by the actual load between the surfaces. In the case of a soft plastic metal like indium, however, which shows little elastic recovery, this will not occur, and once the clean surfaces are pressed together they will continue to adhere even when the load is reduced or removed. Similar effects have been observed with gold when its temperature was raised to its softening point. For this reason, measurements of the ‘ coefficient of friction ’’ of an indium slider on a flat steel plate may have little meaning. If, however, the friction is measured as a function of the real area of contact, this difficulty does not arise and consistent results are obtained. In the experiments described in this paper, the speed of sliding was very low, so that the rise in the surface temperature due to frictional heating was negligible ; the melting between the metals is brought about by the intense pressures in the regions of contact. If higher speeds are used, the rise in temper- ature may be considerable, and a surface softening or local melting may occur. Under these conditions, it is clear that the controlling factor in the frictional behaviour of the metals is not their mechanical properties at room temperature, but their properties at the high temperature of sliding. The analysis outlined has been applied to metals which differ in hardness. It is clear, however, that the same general considerations apply if the surfaces both consist of the same metal, but since both surfaces are torn, it is not easy to evaluate the ploughing and shearing terms separately. Also the discussion has been confined to metals, but we may expect similar conclusions to hold for many types of non-metallic solids. Lubrication by Thin Metallic Films. As we have seen, the frictional resistance between solids may (when P is small) be represented approximately by the expression #=As. If then we wish to get a low friction we should make both A and s as small as possible. Unfortunately, with most solids this is not possible. If we choose a solid of low Shear strength it usually means that it has a low flow pressure so that when we apply a load to the surfaces the area of contact A becomes correspondingly larger. An obvious exception to this is a solid with a plate-like structure such — THE PHYSICS OF RUBBING SURFACES. 205 as graphite. This is able to withstand a pressure normal to the plates but will shear readily when a tangential force is applied. The friction of graphite is notoriously low. It is difficult to achieve this condition for metals since they are less anisotropic. If we choose a soft metal of low s, A will be large (Fig. 14 (a)). If we choose a hard metal A will be small but s will be great (Fig. 14 (b)). For this reason the friction of most metals is of the same order of magnitude and u~0-6-1-0. We may, however, achieve this condition efficiently by depositing a very thin layer of a soft metal on to the surface of a hard one (Fig. 14 (¢)). Provided. the metallic film does not break down the shear strength will be that of the soft "ts small but "1S large Sse asirall but Hard Metal | 2 1S) Ietge Herd Metal Both Aas are small Herd Metel Thin film of soft metal Fig. 14.—The friction between metal surfaces is not greatly dependent on their hardness. A low friction may be obtained by depositing a thin film of a soft metal on a hard metal substrate. metal. At the same time A will remain small even for heavy loads since it is. not greatly affected by the plastic deformation of the hard metals; it is deter- mined essentially by the thickness of the film and the geometry of the surfaces. A series of experiments was carried out by Dr. Tabor with films of indium, lead and copper of varying thicknesses which were plated on to flat metallic surfaces. The substrate was steel, nickel, copper or silver. The upper surface was @ hemispherical steel slider. It was found that the friction is almost independent of the underlying metal and of the load. The main effect of increasing the load is to cause a slight increase in the deformation of the under- lying metal, and so cause a slight increase in the track width d. Since silver is softer than steel the same load will produce a greater deformation. The track width d, and hence the frictional force F' for any given load and film thickness will, therefore, be greater on silver than on steel. This effect was observed. In Fig. 15, F is plotted against d for lead films deposited on various substrates ; the different values of the track width being obtained by varying the thickness of the deposited layer and the radius of curvature of the upper surface. It will be seen that in all cases the frictional resistance is determined, primarily, by-the track width, whether the friction is measured on the surface film or on the 206 F. P. BOWDEN. bulk metal. Hssentially similar results were obtained for indium and copper films. It was also found, as we should expect, that the value of the friction was proportional to the shear strength of the respective metals and for films of similar thickness the friction of copper >lead >indium. 150071 Lead films 0 On Steel oOn Silver x On Copper ® Solid Lead Frictional Force gm. SOO ON 2 32 AN SG 7 ee oe Track width mm. Fig. 15.—Friction of steel slider on lead films deposited on steel, silver and copper substrates, as a function of track width. It is seen that the friction is determined primarily by the track width, whether the friction is determined on the surface film or on the bulk lead. The Limiting Film Thickness. Experiments were carried out to determine the minimum thickness of the deposited film that would influence the friction. The results for indium on tool steel are shown in Fig. 16. The film thickness was calculated from the quantity of electricity used to deposit it. The upper curved slider was of hard steel (radius 0-3 cm.) and the load was 4,000 g. As one might expect, the friction decreases as thinner films are used because the area of contact A becomes smaller. There is, however, a limit to this, and a minimum friction is reached when the thickness is of the order of 10-5 cm. With thickness less than this, e.g. 10—® em. (or 50 atomic layers) the film ceases to be effective. The underlying surface of the tool steel was not highly polished nor was it homogeneous. These factors might easily cause the indium to be deposited preferentially in certain regions, leaving patches of the surface uncovered or only very thinly covered. It is possible that if highly polished surfaces of a uniform metal were used as a substrate, very much thinner layers, even perhaps one or two molecular layers, might prove effective. Breakdown and Wear of Films. If very heavy loads are used, the values of F deviate from the curves shown in Fig.16. After a certain load has been exceeded the friction increases markedly with load, even though the track width remains essentially unaltered. An THE PHYSICS OF RUBBING SURFACES. Coett of Friction: O -6 —-5 -4 -3 -3 1210" oO lO 3xtO Film Thickness (cm) (Arbitrary Scale) Fig. 16.—The effect of film thickness on the frictional properties of thin films of indium deposited on tool steel. A minimum friction is obtained for a film thickness approximately 10-4 to 10-° cm. Mlndiamy filmae4clOh em’ thick on | sted. I. Stearic acid film (Qmolec.layers) on steel. Ol rection ls Coefficient O ©) lO iS 20 ZO Novot (Rumson Same Track. Fig. 17.—Rise in friction with wear of an indium film 4 x 10-4 cm. thick deposited on tool steel. A stearic acid film two hundred times thinner has similar wear resisting properties. 208 F. P. BOWDEN. examination of the track shows that this is connected with the progressive breakdown of the film, as the heavier loads are applied. The load at which this breakdown occurs depends upon the film thickness and upon its strength of adhesion to the underlying surface. It is also influenced by the shape of the slider and takes place more readily if this has a small radius of curvature. The film may also be worn away if the slider traverses the same track a sufficient number of times. With an indium film 4 x10-4 em. thick on tool steel the sliding during the first run was smooth and the coefficient of friction was about ~=0-08. With successive sliding over the same track the friction gradually rose and after the seventh run small ‘‘ stick slips ’’ setin. The friction and the size of the stick slips increased and after 20 runs yu had risen to about 0-4. Microscopie examination of the surface after the seventh run showed that the indium film had been partially worn away and portions of the steel surface exposed. The rise in friction with wear of an indium film 4 x10-* em. thick is shown in Fig. 17. With thicker films (not shown in Fig. 17) the rise in friction is less rapid, showing that the thicker film is more resistant to wear. This behaviour may be compared with the dotted curve which represents the rate at which stearic acid layers are worn off a steel surface. We have found that with nine molecular layers of the acid an appreciable rise of the friction occurs after 20 successive runs. With 53 molecular layers, i.e. about 10-° cm., no appreciable change was observed after 100 successive runs. It is clear that the metallic films resist wear quite well, but they are more easily worn off the surface than a much thinner film of the fatty acid. One factor that may be of importance here is the surface mobility of the film. The fatty acid molecules have a certain mobility and are able to move over the surface and repair the damaged film. Many metals also possess this ability to wander over solid surfaces, but it is probable that the rate at which this process occurs is less rapid for solid metallic films than it is for the fatty acids. Metallic Films as Lubricants. Some experiments were carried out to determine the extent to which the coefficient of friction p is independent of the load (Amontons’ law). The results for tool-steel surfaces are shown in Fig. 18. Curve IJ is for the unlubricated steel ; Curve II for steel lubricated with a film of mineral oil; and Curve III for steel lubricated with a thin film of indium approximately 4 x10-4 em. thick. The comparative behaviour is striking. The coefficient of friction for the unlubricated steel, and for the steel lubricated with the mineral oil, is independent of the load. With the indium film, however, » decreases markedly as the load increases. In the case of unlubricated steel, or of steel lubricated with the mineral oil, the area of contact should be proportional to the applied load, since it depends on the amount of plastic deformation that occurs. For this reason # should be directly proportional to the load, i.e., Amontons’ law should hold and the coefficient of friction should be constant. This is in fact observed. For the unlubricated steel surfaces, the coefficient of friction is constant and w=—0-34, while for the surfaces lubricated with the mineral oil, the coefficient of friction is again constant and w=ca. 0-14. If a thin film of a soft metal is used as a lubricant, however, A will no longer be proportional to the load. The real area of metallic contact now increases only to a slight extent when the load is increased, since the increase in A due to the increased deformation of the underlying steel is small compared with its actual value A. The shear strength s remains constant and is approximately that of metallic indium, so that # should be almost independent of the load. Amontons’ law will no longer hold, and the coefficient of friction » will not be constant but will decrease as the load increases. It will be seen from Fig. 18 THE PHYSICS OF RUBBING SURFACES. 209 that this occurs. The coefficient of friction falls from »=0-2 at the light load to 7=0-04 when the heaviest load is used. If it were possible to lubricate the surfaces with metallic films of molecular dimensions, we should expect a closer agreement with Amontons’ law. In some ways the behaviour of the metallic films closely resembles that of ordinary lubricant films. They produce a substantial reduction in the friction, they can cause smooth sliding, and they protect the underlying metal surfaces. In addition, metallic films are worn off the surface by successive sliding over the same track in a manner similar to that of a lubricant film, except that the metallic films are worn away at a greater rate than the hydrocarbon films. A further point of striking similarity is the effect of melting. The transition from smooth sliding to stick slips when the metallic films are melted, is closely analogous to the change observed when solid hydrocarbon films are heated - through their melting point. Coeff of Friction 4 6 8 iO Load ‘Kgm) Fig. 18.—Effect of load on the coefficient of friction for unlubricated steel surfaces (Curve I) steel surfaces lubricated with mineral oil (Curve II), and for steel surfaces lubricated with a film of indium 4 x 10~4 em. thick (Curve III). The coefficient of friction is independent of load for Curves I and II, but rapidly decreases with increasing load for Curve III. There are, however, several marked differences between metallic films and lubricant films. The earlier experiments showed that even on rough surfaces a lubricant film need only be one molecule thick to be effective as a boundary lubricant (see also Langmuir). A metallic film must be appreciably thicker, of the order of 10-° cm., if it is to be effective. A further striking and fundamental difference is that lubricant films obey Amontons’ law and metallic films do not. With metallic films the coefficient of friction decreases as the load is increased, and at high loads the coefficient of friction may be extremely low. With the indium films described in this paper the value of w under heavy load (u—0-04) is considerably less than that observed with even the best boundary lubricants. This value is similar to that obtained on ice surfaces. The Action of Bearing Alloys. The fact that values of friction as low as this may be obtained with unlubricated metals is of obvious practical interest. For a long time it has been customary to use special metallic alloys which have more desirable frictional 210 F. P. BOWDEN. properties than pure metals. The nature and composition of these alloys— bearing alloys they are usually termed—varies widely. The ‘‘ white metal ”’ alloys are divided roughly into two classes: tin-base alloys, i.e. alloys which consist primarily of tin with various additions such as Sb, Pb, Cu; and lead-base alloys which consist mainly of lead with additions such as Sb and Sn. In general these alloys are complex in structure and contain hard particles dispersed throughout a softer matrix. Another bearing alloy which was developed later and which has found a wide application in aircraft and other engines when the conditions of operation are very severe, is the copper-lead alloy. This is not a true alloy at all but consists of a fine dispersion of lead distributed throughout copper. In one type of the bearing the lead is distributed as minute isolated droplets throughout the copper; in the other, the dendritic type, the lead forms the continuous phase round the copper dendrites. There is no doubt that the friction of these bearing metals is very much lower than that of the pure metals. Typical values obtained by Dr. Tabor are given in Table 2. TABLE 2, Friction of Steel Sliding on Pure Metals and on Bearing Alloys. Coefficient Metal. of Friction. f Lia See * 0-9 Tin base alloy 0-7 Lead wee 1-0 Lead base alloy 0-35 Copper a3 Ew 0-9 Copper lead alloy .. 0-18 The mechanism of this reduction in friction is, however, by no means certain. It has been maintained in the past that an essential characteristic of a bearing alloy is that it should possess a duplex structure consisting of hard crystals embedded in a relatively soft matrix. It is suggested (see, for example, Greaves and Wrighton, 1939) that the function of the hard crystals is to resist wear and that of the softer constituents to permit a more uniform distribution of the load, by allowing any of the hard crystals that are heavily loaded to sink so that the load is spread over a greater area. It is also suggested that the hollows worn in the softer material serve as reservoirs for the lubricating oil. There is probably something in these suggestions, and it is certainly true that many suecessful bearing alloys do possess a structure of this type. It is clear, however, that it is not a complete explanation and in many cases is certainly not true at all. Many bearing alloys, e.g. copper-lead alloys, may consist of a matrix of the harder metal with a small amount of the softer metal finely dispersed through it. In the case of the non-dendritic copper-lead alloy, for example, the hard copper forms the continuous phase so that it is not possible for it to “ sink” into the lead. Why then is the copper-lead alloy so effective in reducing the friction? An obvious suggestion is that it is due to the smearing of a lead film over the surface of the copper. In order to test this a series of experiments was carried out with thin films of lead electrodeposited on to the copper. Journal Royal Society of N.S.W., Vol. LX XVIII, 1944, Plate VIII oid ee aly us fitags hy we i oh : = eat " 4 ne Journal Royal Society of N.S.W., Vol. LXXVIL, 1944, Plate LX Journal Royal Society of N.S.W., Vol. LXXVIII, 1944, Plate X Journal Royal Society of N.S.W., Vol. LXXVIII, 1944, Plate X1 Journal Royal Society of N.S.W., Vol. LXX VIII, 1944, Plate XII de Journal Royal Society of N.S.W., Vol. LXX VIII, 1944, Plate XIII THE PHYSICS OF RUBBING SURFACES. 211 Thin Fiims of Lead on Copper. The relation between the track width and the friction of a curved steel contact sliding on a copper surface which has been coated with lead films of varying thickness is shown in Fig. 19. A film of lead 10-® em. thick (Curve II) causes very little reduction in the friction. As the thickness of the film is increased, the friction for any given track width decreases, and reaches a minimum when the film is 10-3 em. thick (Curve V). Further increase in film thickness produces no further change. The frictional force is now governed by the shear strength of the lead and the width of the track and is not influenced by the substrate (except in so far as this fe) fe) fe) 500 Frictional Force gm. © Slay Brae Ar Say ee 8 ) SOE) Track width mm. Fig. 19.—Friction of a curved steel contact sliding on a copper surface coated with lead films of varying thickness. I, clean copper; II, lead film 10-* cm. thick; III, lead film 107° cm. thick ; IV, lead film 10-4 em. thick ; V, lead film 10-% em. thick and more and solid lead; VI, lead-copper alloy. The behaviour of the alloy suggests that a thin film of lead is extruded and lubricates the copper matrix. may affect the track width). When similar measurements are made on solid lead, the points lie on Curve V (Fig. 19). It is clear that after the film has reached a thickness of about 10-3 cm. the friction is due to the interaction between the steel and lead. This minimum film thickness of 10-* cm. for lead films on copper is greater than that observed when the films are deposited on steel. It is also greater than that observed when indium films on steel. As pointed out earlier, the minimum film thickness for complete “ lubrication ’”’ is influenced by the hardness of the metal substrate. With softer metal substrates, such as copper, which are more readily deformed, thicker films are necessary. It is also influenced by the fineness of the original surface finish and by the state of cleanliness of the substrate since this affects the strength of adhesion of the deposited layer. Copper-lead Alloys. The frictional behaviour of both dendritic and non-dendritic alloys at room temperature was similar. Steel on the alloys gave smooth sliding and a co- ZAP F. P. BOWDEN. efficient of friction of about u.=0-18. The tracks were smooth grooves showing little tearing. Some signs of smearing of extruded lead could be detected. The alloys on steel gave stick-slip motion, the maximum value of the friction being about w=0-3. The tracks showed that some metal from the upper contact had been welded on to the steel surface but the extent of this was considerably less than the welding observed with pure copper or lead. It is at once apparent that the friction of these alloys does not lie between the values of their constituents. The friction of steel on copper, for example, is about ~=0-9, and for steel on lead w=1-0. With steel on the copper-lead alloys, however, the friction is about .=0-18. The alloys and the pure copper have approximately the same flow pressures at room temperature, so that the area of contact A is nearly the same for a given load. Nevertheless the friction of the alloys is less than one-quarter that of their pure constituents. The relation between the friction / and the track width d for the alloy is plotted on the broken curve (Curve VI) in Fig. 19. It will be seen that these results he very close to those obtained when a lead film 10-4 em. thick is artificially deposited on to copper. This at once suggests that the lead in the alloys is extruded during sliding and forms a thin “ lubricating ”’ film of effective thickness between 10-4 cm. and 10-3 cm. on the copper. This is borne out by a microscopic examination of the track which showed traces of lead smeared over the surface. The Role of Thin Films in the Action of Bearing Metals. We see, therefore, that the frictional behaviour of a copper-lead alloy resembles very closely that of a copper surface on which a very thin film of lead has been deposited. The actual value of the friction of the alloy is the same as that of a copper surface on which a lead film 10-4 em. thick has been artificially deposited. Additional experiments showed that the temperature coefficient of friction was the same in each case. The increase in friction with wear was also very similar except that, in the case of the alloy there is evidence that the potential supply of lead is greater since it may continue to be expressed from the alloy during the sliding process. With a thin film of lead deposited on copper the supply is, of course, limited to the amount which is actually present on the surface at the beginning of sliding. These results show the important part which may be played by thin surface films of metal in reducing the friction and wear of bearing alloys and support the view that in certain bearing alloys the anti-frictional and anti-seizure properties may be due, primarily, to the spreading of thin films of the soft low melting constituent, over the surface of the harder constituent. Certain solids are able to provide their own surface films. This is the case, as we have seen, with ice. Under the frictional heating local melting of the surface layers may occur and may provide a lubricating film, while the bulk of the solid remains hard. With most pure metals it is difficult to achieve this, because of the progressive softening of the metal as the temperature is raised which leads to a corresponding increase in the area of contact, so that the friction is little affected. The problem is to keep both the area of contact A and the shear strength s as small as possible. We see, however, that the condition of small s and small A may be achieved artificially by putting a very thin film of a soft metal on to the surface of a hard one. This may be done by deposition beforehand, or it may be achieved by selecting an alloy of suitable composition and structure so that the film is spread during the sliding process. In this connection the importance of surface tension is apparent. The soft low melting constituent must spread readily on the harder one to form a thin film. This is borne out by experiment. It was shown, for example, with lead, indium and mercury films on steel, copper and silver that unless the soft metal THE PHYSICS OF RUBBING SURFACES. . 213 readily wetted the harder one, it was, under severe conditions of sliding, relatively ineffective as a lubricant. Although it is desirable that these metallic films should adhere to and wet the substrate, the converse should be true about the other moving surface, e.g., the journal of a bearing. Provided the metallic film does not break down, the smaller the adhesion and the less the tendency of the metallic film to spread on to this surface the better. As has been pointed out in Part I, one important function of a lubricant is to reduce the adhesion between the two moving surfaces and help to reduce the effective shear strengths of the metallic junctions. Apart from these surface tension and spreading effects the availability of the soft constituent is important. The dendritic copper-lead alloy, for example, may under certain severe conditions of sliding be more resistant to seizure than the non-dendritic alloy. In the former alloy the pockets of lead are all inter- connected so that potentially a greater supply of lead is available to any particular area of the surface which may be in need of it. With the non-dendritic alloy the pockets of lead are isolated so that a local exhaustion of the lead may occur and a seizure on to the copper surface may take place. This difference is shown by the “ sweating ’’ out of the lead (see Plate XI, Fig. 1). When the non- dendritic alloy is heated, the lead appears as a fine mist of droplets over the surface. When the dendritic alloy is heated the lead collects in one or two large drops, by a surface tension effect (see Plate XI, Fig. 2). The dendritic alloy thus has a three-dimensional supply of lead to the surface while with the non-dendritic alloy it is restricted to the surface layers. Lubrication by these thin metallic films may be used effectively in a number of practical problems. One of the newer type of bearings which is used in modern aircraft is the silver-lead bearing. The bearing surface consists of pure silver on which is deposited a thin layer of lead or of lead and indium. The silver itself has very desirable mechanical and thermal properties but it gives a high friction and readily seizes on steel. It may, however, be effectively lubricated by a thin film of a soft metal such as lead and indium. It is interesting to note that Atlee, Wilson and Filmer (1940) find that steel ball bearings in a vacuum tube may be effectively lubricated with thin sputtered films of barium and other metals. It is probable that thin metallic film lubrication will find an increasing application in a number of practical operations such as the pressing, drawing and heavy working of metals. THE EFFECT OF TEMPERATURE CHANGES ON BEARING ALLOYS. : As we have seen, the frictional heat liberated during running may raise the temperature of the surface layer of a bearing to high values, and this may have a marked effect on the mechanical and frictional properties of the bearing. In addition, however, to this localized surface heating the bearing as a whole may be gradually warmed up and cooled down as the engine is started or stopped or as the temperature of the oil changes, and we may enquire what effect these thermal cycles may have on the bearing alloy. If, as is frequently the case, the alloy is bonded on to another metal, we may expect that the difference in the thermal expansion of the two metals may set up thermal stresses near the boundary. In the case of a typical lead-base alloy the coefficient of thermal expansion is 24 x10~—§/° C., while for steel it is ca. 12 x10-§/° C. If the bearing is heated and cooled through 100° C. or so, the difference in expansion may cause a deformation and cracking of the softer metal. A characteristic example of this is shown in Plate XI, Fig. 3. This effect, which might be expected, is of practical importance, but it is of comparatively small scientific interest. Some recent experiments by Mr. Honeycombe have, however, revealed a new effect. He found that if he heated and cooled a tin-base bearing metal, wnatiached to any steel, through a temperature range of ca. 100° C., deformation and cracking 214 F. P. BOWDEN. ~ occurred throughout the mass of the metal (see Plate XI, Fig. 4). A lead base alloy treated in the same way showed no such effect. The rate of heating and cooling did not affect the phenomena. In some cases the heating and cooling for a single thermal cycle took as long as seven hours. This shows that the effect is not due to a thermal gradient in the specimen, i.e., it is not due to a heating of the outside (with consequent thermal expansion) while the inside is cool. To what, then, is it due, and why does it occur with tin-base and not with lead-base alloys? Crtt and Criu—Cr respectively. The green sulphate undergoes irreversible reduction to a substance which seems to decompose in some complex way which is not satisfactorily recorded by the polarograph. The potential of the first step, however, does not change with dilution, which indicates that the ionic species remains the same throughout. The solution of the violet alum when heated changes to green and gives a polarogram similar to that of the green sulphate. ACKNOWLEDGMENTS. The author wishes to thank Mr. D. P. Mellor and Dr. B. Breyer for suggesting this investigation, and for much valuable discussion. This work was carried out during the author’s tenure of the Masson Memorial Scholarship, for which grateful acknowledgment is made to the Australian Chemical Institute. POLAROGRAPHIC STUDY OF ISOMERIC CHROMIUM SULPHATES. 245 REFERENFES. Demassieux and Heyrovsky, 1929. J. Chim. Phys., 26, 219. Graham, 1912. Am. Chem. J., 48, 145. Grube and Schlecht, 1926. Z. Hlektrochem., 32, 178. Jones and Mackay, 1897. Am. Chem. J., 19, 83. Kolthoff and Lingane, 1941. ‘‘ Polarography ’’, Interscience Publishers, New York. Weinland and Krebs, 1906. Z. anorg. u. allgem. Chem., 49, 157. Werner and Gubser, 1901. JBer., 34, 1579. Willis, 1944. J. Am. Chem. Soc., 66, 1067. THE VIBRATIONS OF SQUARE MOLECULES. PART Il. THE VIBRATION FREQUENCIES OF PLANAR AB,C, (TRANS) MOLECULES. By ALLAN MACCOLL, M.Sc. Manuscript received, November 16, 1944. Read (in title only), December 6, 1944. INTRODUCTION. In Part I (Maccoll, 1943) expressions for the symmetry co-ordinates and - vibration frequencies of the planar AB, molecule (D,n) were obtained. It is of interest to examine the case of the planar trans AB,C, molecule in which the symmetry has been reduced to Vy. Expressions for the frequencies of the vibrations symmetrical to C,2 have been obtained to a valence force approxi- mation by Bernstein (1938). THE SYMMETRY CO-ORDINATES. The classes of symmetry operations of the group Vn are: (1) The identity operation E. (6) Twofold rotations around the z axis, C,?. (c) Twofold rotations around the y axis, ©,Y. (dq) Twofold rotations around the x axis, C,*. (€) Inversion in the centre of symmetry, I. (f) Reflection in the xy plane, oxy. (g) Reflection in the xz plane, oxz. (h) Reflection in the yz plane, cyz. The symmetry elements are shown in Fig. 1. As in Part J, the reducible representation of the molecular vibrations can be expressed in terms of the irreducible representations of Vp. y . Cn y Fig. 1—Coordinate axes of the AB,C, molecule. THE VIBRATIONS OF SQUARE MOLECULES. 247 The transformation properties of the symmetry co-ordinates are shown in Table I (cp. Wilson, 1934). TABLE 1. | | Symmetry with respect to | Active in No. Co- | | Polar- Class. | Vibra- ord. | l l ;. | isation. | tions. E CyZ Gaye Lx te Oxy. Oxz | Oyz | Raman. Ree Aig 2 BE + SF + + + a a + yes no P Ba he| ee Rei Fase Sas) ave RTS ol na Bsu 2 Ss “iF = = ot = als ci = no yes — So | Big i 8; eas “+ — = F aF = — yes no D Buu 2 Sa ae ere = Ses = or a no yes aaa Ss |: Bou 2 at tg = =F Ss an = at no yes a 7 It is seen that from symmetry considerations the secular equation of the ninth degree can be factored into one linear and four quadratic factors. So one symmetry co-ordinate S, can be set up which is also a normal co-ordinate, while S,, S.; S, S;; S., S,; and 8,, S, have to be linearly combined to give the corresponding normal co-ordinates. The symmetry co-ordinates are shown in Fig. 2. ; = : : e e ° aed @ —e t ® i e+ @ +e e+ ® +e t ° o> ; am $1 Aue So Big Se S4 Bau Ss [ o> Dd * 82+ Se te Mead ENCE See Aer aie ee 2Ke s.8 — i=1 Bi (Ute)? (Yates)? (ele7)? 67 (ueun)h where Y4=Mm, =n u3—=(mm’)? 2 © 0 0 © © ele eel elteiaerete nent 2M(m+m’) re Bes 2m +2’ +-M m+m’ 5 = hes Ug = 9 The solution of the secular equation then gives rie) Woe ie 27C \ 3 and 1 Ol Ke Ke Ke) 4a 2122 Vio I=] ae a0 Fees. =p : TC ( Ui BL Mi pj bis | J where i,J==1,2 $:134,0:5,10 854.5) OF A839. THE POTENTIAL FUNCTION. As in Part I, a valence force potential function, modified by the addition of cross terms, will be used. THE VIBRATIONS OF SQUARE MOLECULES. 249 This is given by: 2V =k,(dr,?+ dr,”7) +ky’(dr,?-+ dr,”) +2k,yr’(dr,+ dr3)(8r,+5r,)+2k'rr Sr, Sry +2k’'y’r’ Org Sry+kea?(da,?+ da5”) ++ Koy’ a’2(So9?+ Soy?) + 2k ae’ (day + dag)(Sa%g+ daxq) Se OC OMe OK: oy NOC MOM Ea halide choise e's es ere ae bales (6) +k ga?(3B,2+ 585%) +kg.a’2(3B.?-+ 584") + 2k gq aa’(3B, + 583)(3B2+ 984) +2k' gga” 68, dB, T2k'g ga” dB. dy. On expressing (6) in terms of the S;, the following relations are obtained between the force constants of the general potential function (4) and those of the valence force function (6): K, =k;+k'rr K, =ky’+k’r’r’ mt , ! / O et / I. / K, mcmama 22ema/) 2m"?(Ke+K'ax) + a*m?(ke’+k’y’a’) —4aamm Kee} K; =2(k, +k apt Kg +k pg’ + 4Kga) 1 ate "2 / 2 ’ a Bee. K, own (k, +k gg) +m (kak a’p’) 4mm‘) aes K: —=2(ky—k'yy+ Ka’ —k'g’a’) winaize| © 0] © © jer 60 © 0. (se 0) 10 © 0 © 0.0 0 « 0.6 © 6 6) 0 « (Gs) i ; ; / Ke = Sean’ 2(ky —k rr) +m 2( Kg’ —k on’cx’) Ky =2(kr’—k’p’y’ + Ka —k' ac) 1 f ' / K, = 5 (Ke —k'y’r’) +m "(Ko —K'aa)} Kyo =2krr ee ema’ )s —m (k, +k gg) TM(k, +k gg) +2(m -——m )K gg J / , { nm k (ky —k'ypr) —m(Ke’ —k'e’«’)} i i Ko ast Hata ‘r’r’) —m’ (Ky —k'aa«)} The expressions (7) when substituted into (5) give the frequencies in terms of the generalised valence force potential constants. By making the substitution m’=m, k,’=ky, etc., these expressions reduce to those of Part I. DISCUSSION OF THE RESULTS. It is of interest to compare the frequencies of the AB,C, (trans) molecule with those of the AB, and AC, molecules. This will be done to a valence force approximation. For AB,C, (trans) the frequencies are given by - _kr woke Ay Ba eer 7 __(a’m’)?ka + (am)*Ke’ , = mm’(ma?-+m’‘a’?) 250 ALLAN MACCOLL. meth (14 Gr Jeet (A wee oy = (Oe fs heb a= (Lt ap ee derby =(14 gp + (1+ are Aghg =(1450 tar aun where A; =477c?,;?. fe AB;C; se Vg : we Me Bou 1s. Eu Vi,Arg Vs, Bau Va Agu V7,Euw V3, Big Va Bre Fig. 3.—The correlation of the frequencies of the series AB,-AB,C,-AC,. On the left-hand side is shown the correlation of the non-degenerate frequencies and on the right-hand side that of the degenerate frequencies. THE VIBRATIONS OF SQUARE MOLECULES. 251 If m=m’, ky=ky’, ky Ky’ and kg =kg’, these relations reduce to the valence force frequencies of the AB, molecule given by Wilson (1935). If reasonable values are assumed for the constants occurring in the expres- sions (8), the frequencies of the three molecules AB,, AB,C, (trans) and AC, can be determined. This is done in Fig. 3. In the Raman spectrum of AB, two lines appear, while there are three in the infrared. For AB,C, (trans), however, three lines appear in the Raman spectrum and six in the infra red. In either case, to determire the ‘‘ out of the plane ”’ force constants requires an investigation of the infra red spectrum. SUMMARY. Analytical expressions have been obtained for the normal co-ordinates of the planar AB,C, (trans) molecule, together with the relationship between the constants of the general potential function and those of a valence force function, modified by the introduction of cross terms. A diagrammatic representation of the correlation between the frequencies of the series AB,—AB,C, (trans)—AC, to the valence force approximation is given. REFERENCES. Bernstein, 1938. J. Chem. Phys., 6, 718. Maccoll, A., 1943. Tuts JouRNAL, 77, 130. Wilson, 1934. J. Chem. Phys., 2, 432. 1935. J. Chem. Phys., 3, 59. Department of Chemistry, University of Sydney. ERRATUM. In Part I (THis JournaL, 77, 136) the right hand sides of the fourth and fifth equations of (21) should be interchanged. U—December 6, 1944. SOME INTERFERENCE EFFECTS WITH MICA. By O. U. VONWILLER, F.Inst.P. With Plate XV. Manuscript received, November 14, 1944. Read (in title only), December 6, 1944. When white light is incident on a thin sheet of transparent material the spectrum of the transmitted light consists of a number of bright bands, the wave- lengths of maximum intensity being given by A in the equation Qta/ NR? —SIN7I== PA... ahs eo ee (1) where /p is an integer, ¢ the thickness of the sheet, n the index of refraction and i the angle of incidence. As is well known, the sharpness of the bands increases with the coefficient of reflexion and therefore is greatest for grazing incidence. The spectrum of the reflected light gives the complementary effect ; at grazing incidence we have a continuous spectrum broken by sharp dark bands in the positions occupied by the bright bands in transmission. Several years ago photographs showing these effects were prepared for class demonstrations ; on these some observations were made that appear of sufficient interest to be put on record. In Plate XV, (1) is the transmission spectrum through a thin sheet of mica, at large angle of incidence (about 88°). Two sets of bands are present as the material is doubly refracting, and from the band spacing it is seen that the dispersion differs for the two. One set of bands is distinctly sharper than the other, indicating a difference in the coefficient of reflexion. All the spectra shown were obtained with the mica placed so that the directions of vibration were respectively in the plane of incidence and at right angles to it. The incident light was plane polarised, the relative intensities of the two systems of bands being controlled by altering the plane of polarisation. Spectrum (2) was obtained by turning the polariser so that one set of bands, the less sharp in (1), was completely suppressed. Numbers (3) and (4) are reflexion spectra corresponding with (1) and (2). Numbers (9) to (14) are also transmission spectra, obtained with a different spectrograph. In (9), (10) and (11) the thickness was about 0-04 mm.; (9) shows the double refraction, in (10), the polariser was placed so as to cut out one of the rays; in both of these the light was at grazing incidence; in (11) the light was incident normally ; the bands are much less sharp corresponding with the smaller coefficient of reflexion. In (12) and (13) we have spectra corresponding with (9) and (10) but for a much thinner sheet, ¢ being about 0-006 mm., and in (14) we have a spectrum corresponding with (10), for a very thin sheet, ¢ being only about 0-003 mm. Elsewhere it is shown how from observations on the bands the value of p for any band, the thickness of the mica, and the three principal indices of refraction for various wave-lengths can be determined.' ime O. U. Vonwiller and F. L. Arnot. ‘‘ Interference Phenomena with Thin Films,”’ Proc. Third Pan Pacific Science Congress, Tokyo, 1926, Vol. II, pp. 1314-1322. SOME INTERFERENCE EFFECTS WITH MICA. 253 COMPARISON OF THICKNESS OF SHEETS OF MICA. In (5) we have a transmission spectrum for a piece of mica that had been torn in two in such a way that different thicknesses were traversed by different portions of the light. Corresponding with these we see several distinct sets of bands. The sharpness of some of the bands suggested that observations on them might provide an accurate means of comparing the thicknesses of the corresponding portions of the mica. The positions of the bands of each of three systems were read by means of a travelling microscope ; in the case of the sharpest set it was found that the error of the mean was not more than one or two-thousandths of a millimetre ; the means for the others were less accurate. The mean readings for the three sets, measured from an arbitrary zero, are given in Table 1; the number in brackets after each reading is the estimated error of the mean, in thousandths of a millimetre. The value of p, determined as described below, is given for each band. TABLE 1. Series 1. Series 2. | Series 3. p Reading. Dp Reading. p” Reading. 30 10-828 (1) 18 10-553 (5) 17 31 12-167 (1) 19 12-807 (3) 18 11-937 (5) 32 13-575 (1) 20 15-208 (7) 19 14-351 (5) 33 15-037 (2) 21 17-781 (3) |. 20 16-973 (6) 34 16-551 (2) pApe 20-513 (14) 21 19-782 (2) 315) 18-139 (2) 23 23-442 (5) 22 22-719 (3) 36 19-782 (2) 24 26-573 (2) De 25-904 (5) 37 21-502 (1) 25 29-914 (4) | 24 29-339 (9) 38 23-285 (2) 26 33-535 Peay) 33:012 (4) 39 25-147 (2) 26 36-976 (7) 40 27-084 (2) 41 29-101 (1) 42 31-209 (2) 43 33°398 (2) 44 35-690 (2) 45 38-084 If band p’ of system 2 comes between bands p and p+1 of system 1 band, p’ corresponds with p+a, where x is less than unity; by the application of methods described later, x may be determined to a high degree of accuracy. If ¢ and ?t’ are the thicknesses of the two sheets of mica we have rash from equation (1), so that if p and p’ are known and «x determined we obtain the ratio of the thicknesses. The value of p for a particular band can be found as follows: Let A, and A, be the wave-lengths of bands p and p+q where q is known by counting from p. The wave-lengths can be determined by the usual method of a comparison line spectrum. If the index of refraction were constant we should have PA=(P+Q)Az ov Ne and p TG ema re Oe athe rhea (2) The index of refraction increases as the wave-length decreases so that the value of p, obtained from equation (2), is too great; with spectra such as those con- sidered here, where p is relatively small, the excess is usually less than one and 254 O. U. VONWILLER. neglecting it we have the value of p. Similarly p’ is determined. Should a mistake be made in estimating p or p’ it will be revealed by inconsistent values for the ratio of the thicknesses, obtained from readings with the different bands ; by trial the correct numbers are easily found. The value of « was determined by one or other of the methods of inverse interpolation described in an article by K. Mader, in the Handbuch der Physik.? These are methods of successive approximation involving difference of the second, third and higher orders. In a table giving these differences for the lines of series one, apparently the most accurate of the three, it is found that the third order differences are fairly large, up to 0-020 mm., and that several are positive and several negative, the changes of sign alternating over part of the series, with the result that differences of the fourth order are also large, in some instances too large to be neglected, although the multiplying factor is generally very small. This is probably because the screw of the travelling microscope is faulty, instru- mental errors being introduced in the readings considerably greater than the observational errors in series one, if not in the others. However the results given in Table 2 show that very good agreement is obtained from observations with the several bands of a series; possibly that would be improved to some extent by the use of a better screw or by the application of calibration correc- tions, when the final limitation would be due to the observational errors, that is to the lack of sharpness of the bands. _ In those instances in which the value of x, the fractional part in column 2 or 5 of Table 2, was greater than 0-3, the value was determined by two modifications of the general formula given by Mader, and the result quoted is a mean value; the differences were generally small, the largest being 0-0004. TABLE 2. p a) t/t’ p” 7? t/t” 19 31-45998 1-65579 18 30- 83215 1-71290 20 33-11482 1-65574 19 32-53557 1-71240 21 34-77757 1-65610 20 34-27023 1-71351 ®& 22 36: 43024 1-65592 yA | 36 23 38-08603 1-65591 22 37-68674 1-71303 24 39-74109 1-65585 20 39 - 39540 1-71284 25 41-39070 1-65563 24 41-11510 1-71313 26 43-06109 1-65619 25 42 - 82685 1-71312 Mean Las 1-65589 | Mean .. 1-71299 Table 2 gives the results obtained, comparing the bands of series 2 and 3 with those of series 1. Column 1 gives the values of p’, column 2 the corres- ponding values of p, and column 3 the ratio t/t’, obtained by dividing p by p’. In columns 4, 5 and 6 we have the corresponding results for the bands of series 3. The overlapping of band 36 of series 1 and band 21 of series 3 made precise position determination of band 21 impossible; the corresponding ratios are omitted in Tables 2 and 3. Before reference was made to the article mentioned a method of approxi- mations was worked out that gave two values of « that might be, one slightly below, and the other slightly above, the correct value ; taking the mean of these, in effect, is applying a third order correction. Suppose the position of the band examined to be between bands p and p+1 of the comparison series ; let f(p+2), 2 Handbuch der Physik, Vol. 3, p. 615 et seq. SOME INTERFERENCE EFFECTS WITH MICA. 255 f(p), ete., represent the readings of the various bands concerned; the two formule derived for x are f(p +%) —f(p) i — ipa ipl) t(p) fp a reps) ape 5 x Uf il) ee NG Oe (dea) a f(p+1) ae fet) —f(p) PET) 56 (p42) —2f(p +1)-44(0)] Fp +2) Tu) These formulz were applied to a Pai of the bands of series 1 and 2 with those of series 3, the values of f(p), f(p +1), ete., being the readings of p” in Table 1 and the values of f(p+a) view of p or p’. The results are given in Table 3 where the values obtained from the two formule are indicated by the pairs of figures for the last two or three decimal places. TABLE 3. p’ p” b” [t’ 6) p” Wit 20 19-33522 0-96676 33 19-27030 0-58392 346 67 6805 88 21 = 20 29532 96644 34 19-84453 58366 218 29 359 63 22 = 21-25163 96598 | 35 20-42369 58353 658 621 | | 045 44 23 =: 2.2. 23436 96671 | 3621 0 cE 97 -91-59106 58354 24 23- 20090 96670 528 66 an ot | 38 22-18382 58378 25 24-16111 96644 | 10 76 of a | 39 22-77006 58385 26 25- 13644 96679 6900 B29 ae ue | 40 23+35224 58381 | 108 78 41. | 23-9332 58374 28 73 42 24-5137 58366 84 17 43 25-1008 58374 10 74 44 25-6841 58373 33 71 Mean OL 8e654 | Wend D0. 58372 The ratio of the two mean values given in Table 3 is 1-65583, in good agreement with the value of t/t’ given in Table 2, 1-65589, while the reciprocal of the value of t’’/t in Table 3 is 1-71315 as compared with 1-71299 for ¢/t’’ in Table 2. Calculations by the above formule are made much more quickly than by the methods described in the Handbuch der Physik, and in a case such as this are at least as trustworthy. These results, as well as an examination of the individual values leading to the means, indicate that the comparison of thickness can be made to an accuracy of the order of one in ten thousand. The thicknesses of the three portions of mica were about 0-0080, 0-0048 and 0:0047 mm. respectively, so that the error 256 - O. U. VONWILLER. in the value for the ratio of thicknesses corresponds with but a few Angstroms, that is less than the grating space for mica, about 9A. REFLEXION AT MICA-ANTIMONY SURFACE. On half of one side of a thin piece of mica a layer of antimony was deposited by an evaporation process. Light incident at grazing angle (t=88°) on the uncoated side was reflected, and the reflexion spectra (6), (7), (8) were obtained. In these the upper half corresponds with the uncoated half of the mica, and the lower half with the antimony-coated portion. In the upper part of (6) we see the usual two systems of bands, but apparently only one system in the lower half. In (7) and (8) the polariser was rotated so as to suppress each of the beams in turn. In (7) the change of phase at reflexion results in a considerable displacement of the band, while in (8) the displacement is much less, so that in (6) the two bands in the lower half are nearly coincident. The phase change could be measured with considerable accuracy by applying the methods described earlier for the determination of the band displacement, though for other than grazing incidence there would be a lack of sharpness with corresponding decrease in accuracy. It was intended to explore this field further, as it might provide a useful method in the investigation of the optical properties of metals. Demands arising from war conditions made this impossible, but it is hoped that opportunity for further work in this field may be obtained in the future. Spectra (9) to (14) on the plate were portion of a plate prepared for the paper by O. U. Vonwiller and F. L. Arnot at the Pan-Pacifie Science Congress ; it was not included in the volume probably because detail in some of the other spectra on the plate was too fine for reproduction. I am indebted to Miss D. Tarrant, Commonwealth Research Assistant, and Mr. E. Warner for the preparation of an enlarged print, for the plate, of spectra selected from a number of original plates. SUMMARY. Measurements of the wave-lengths of the sharp bright bands in the spectrum of white light transmitted, at grazing incidence, through thin ‘sheets of mica, ranging in thickness from 4 » to 8 yu, and the application of suitable interpolation formule, enable the ratio of the thicknesses to be determined with an accuracy of one in 10,000. The fine dark lines in the spectrum of white light reflected at grazing incidence from a thin sheet of mica, half the back surface of which is coated with a metal, occupy different positions for the two halves, enabling the phase change at reflexion at the metal surface to be determined with considerable accuracy, suggesting a method for investigating optical properties of metals. Journal Royal Society of N.S.W., Vol. LXXVIII, 1944, Plate XV BRR Ea TE Le. ieeeeaneaecepaae ee ee SOME INTERFERENCE EFFECTS WITH MICA. EXPLANATION OF PLATE. (1) Transmission spectrum, showing double refraction. (2) Transmission spectrum, one set bands suppressed. (3) Reflexion spectrum corresponding with (1). (4) Reflexion spectrum corresponding with (2). (5) Transmission spectrum, mica of irregular thickness. (6) to (8) Reflexion spectra; lower half of back surface coated with antimony. (6) showing double refraction. (7) electric vector in plane of incidence. (8) electric vector perpendicular to plane of incidence. (9) Transmission spectrum, showing double refraction. 10) Transmission spectrum, one beam suppressed. 11) Transmission spectrum, as (10) but for normal incidence. 12) and (13) Transmission spectrum, corresponding with (9) and (10). (14) Transmission spectrum, corresponding with (10) and (13). In all except (11) the angle of incidence was about 88°. The approximate thicknesses were : ton v4) 72 te. oye te aA UI 9 to ll: 38-0n. 12 and 13: 5:8u. 14: 3-Ou. Magnification, Nos. 1-8, x2-4; Nos. 9-14, «2-3. A NOTE ON THE MAGNETIC BEHAVIOUR OF VERDOHAMOCHROMOGEN. By D. P. CRAaiG, a8e., and D. P. MELLOR, M.Sc. Manuscript received, November 2, 1944. Read (in title only), December 6, 1944, One of the oxidation products of hemin is a green pigment to which the name verdohemochromogen has been given. It has been assigned the following structure by Lemberg (1935, 1938) : This structure implies that (a) the porphyrin retains its general ring-like configuration after fission, and (b) the iron is ferrous and octahedrally linked to six nitrogen atoms. If the second statement is correct, then the molecule should be diamagnetic. Susceptibility measurements have been made on samples of verdohemochromogen in order to determine whether iron in this compound is bivalent and whether it is ionically or covalently bound. EXPERIMENTAL. Measurements of susceptibility were made by the Gouy method on three different preparations. Specimen 1 was an old specimen much of which was insoluble in chloroform. It proved to be paramagnetic with a moment of 4-0 Bohr magnetons (calculated on Lemberg’s formula FeC,3;H,.N,0,). The insolubility of this sample in pyridine indicated that it was largely an alteration product of verdohemochromogen. Specimen 2 was freshly prepared, but was handled without any precautions against oxidation. Its magnetic moment was 3-7 Bohr magnetons. Micro-analysis of the sample indicates the composition FeC,,H;,N,0,, instead of the expected FeC,,H,,.N,O,. The magnetic moment was calculated using the analytical molecular weight. Specimen 3 was freshly prepared and kept in an atmosphere of nitrogen until the time of measurement. Its moment was 2:16 Bohr magnetons. Analysis of this sample corresponds to the following formula, on the assumption that there is one atom of iron per molecule : BOC, ag ilgacg Ws pO aan A NOTE ON THE MAGNETIC BEHAVIOUR OF VERDOH HMOCHROMOGEN. 259 DISCUSSION. The fact that the most reliable sample measured did not correspond very closely to Lemberg’s formula and the fact that verdohemochromogen undergoes a change on standing, makes any interpretation of these results necessarily provisional. A small amount of diamagnetic material would not seriously affect the results, but the evidence is that the alteration products of verdo- hemochromogen are paramagnetic. On the assumption that specimen 3 contained but little alteration product, its magnetic behaviour is consistent with the view that the iron is in the covalent ferric condition as K;Fe(CN), (u=2-33 Bohr magnetons). This view about the valency of iron in verdohemochromogen is held by other workers on chemical grounds (Fischer and Libowitzky, 1938; Libowitzky, 1940; Stier, 1942). Because of some uncertainty regarding the purity of even the best specimen of the compound (3), it is perhaps safest to conclude that, whatever the valency of the iron in verdohzemochromogen, the iron is bound to the remainder of the molecule by covalent bonds. ACKNOWLEDGEMENT. : We are indebted to Dr. R. Lemberg, who very kindly prepared the specimens of verdohemochromogen used in this work. REFERENCES. Fischer, H., and Libowitzky, H., 1938. Z. Physiol. Chem., 251, 198. Lemberg, R., 1935. Biochem. J., 29, 1322. Lemberg, R., Cortis-Jones, B., and Norrie, M., 1938. Brochem. J., 32, 149. Libowitsky, H., 1940. Z. Physiol. Chem., 265, 135. Stier, E., 1942. Z. Physiol. Chem., 273, 47. V—December 6, 1944. THE CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. Parr IJ. THE STANDARD OXIDATION REDUCTION POTENTIAL FOR THE CHLORIRIDITE-CHLORIRIDATE SYSTEM IN HYDROCHLORIC ACID SOLUTION. By F..P. DWYEBR,-MSe., H. A. MCKENZIE, B.Sc., and R. 8. NYHOLM, MSc. Manuscript received, November 8, 1944. Read, December 6, 1944. It has been pointed out by Kolthoff and Tomsicek (1935) that many accepted values for standard oxidation reduction potentials are in error by as much as 0-1 volt because measurements were not carried out at sufficiently low ionic strengths. The oxidation reduction potential for the equilibrium between trivalent and quadrivalent iridium has been studied by Ogawa (1929) and more © recently by Sho-Chow Woo (1931). Inconsistencies in the results of Ogawa (loc. cit.) lead Sho-Chow Woo (loc. cit.) to determine the potential for the system IrCl,= —e2IrCl,- in 1 Normal hydrochloric acid, but no attempt was made to obtain the true standard oxidation reduction potential at infinite dilution (zero ionic strength). In this paper measurements are described in which the potential has been determined in hydrochloric acid of decreasing concentration down to that point where hydrolysis of the complex ion becomes Serious. The usual method of mixing analysed solutions of ammonium hexachlor- iridite and ammonium hexachloriridate with various amounts of hydrochloric acid was adopted, the solution being stirred in contact with an inert electrode until equilibrium was attained. Rhodium plated on glass was found to be the most satisfactory electrode, stable equilibrium being reached rapidly. Much trouble was experienced with other electrodes such as platinum or iridium plated platinum, a difficulty also encountered by Sho-Chow Woo (loc. cit.). The system IrCl = —e2IrC1,= is of particular interest in that the oxidation- reduction potential diminishes with decreasing ionic strength instead of rising as with cationic systems such as ferrous-ferric iron. A similar case is the ferro- cyanide-ferricyanide system investigated by Kolthoff and Tomsicek (loc. cit.). This behaviour is in accord with the Debye-Huckel theory, the following caleu- lation showing that for very low ionic strengths the oxidation-reduction potential, . E, should be a linear function of +/I. eee ae CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. 261 For any reversible oxidation-reduction system For the system IrCl,=—ezIrCl,- B=Eo+- In “l_, (n=1) me) AirCl,— Circi, X firci,~ Cre

77e. Z =Valency of ion, n=Zox —ZRea. C =Molar concentration for a given ion. k =Universal gas constant. fF =Faraday’s number. f =Molar activity coefficient. dox =Activity of oxidant. Area =Activity of reductant. From the above calculation it is clear that under ideal conditions the oxidation-reduction potential of the system should decrease by 14-5 millivolts for every 0-1 unit decrease in 4/ionic strength. This limiting Debye-Huckel law holds only below an ionic strength of approximately 0-1. In the absence of much hydrochloric acid, at low ionic strengths, hydrolysis of the hexachlor- iridate ion becomes serious and hence the experimental values obtained cannot be expected to give a true value of H for the reversible system IrCl,=—e2IvC1,~. Since the standard oxidation-reduction potential is in any case a limiting value at infinite dilution, it is necessary to extrapolate from experimental data to obtain it. Extrapolation through experimental points obtained at low ionic strengths is clearly not satisfactory since hydrolysis of the complex ions makes it probable that at this dilution we are no longer dealing with the reversible system IrCl=—ezIrCl,=. Under the circumstances, it was decided to obtain the limiting value of Hy by assuming that the Debye-Huckel limiting law holds below the point where ,/J=0-2, i.e. where [=0-04, the point A in Fig. 1. This is justifiable on theoretical grounds and also because a smooth curve is obtained which would not be the case if experimental points below A were used. It is of interest to note that the experimental points below A all lie to the right of the Debye-Huckel slope. This is to be expected since hydrolysis would tend to 262 DWYER, MCKENZIE AND NYHOLM. 1°07 ! iy / Calculated Debye-Huckel Slope. 1-03 1°02 1°00 0-0 0-5 1°0 15 / Ionic Strength. Big. 1. POTENTIOMETRIC TITRATION OF 10 m1 (NH, )oIrCle with H3TiCle. 4/2 6 3 6 0-9 Potential v. Saturated Calomel. otential. Ml of Titanous Chioride. Fig. 2. ere SS CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. 263 lower the activity of the oxidant more than the reductant and hence give a lower value for H. In the determination of EH, for the stannous-stannic system by Huey and Tartar (1935) it was found that, for small values of J, the curve fell very steeply as J decreased. This behaviour was attributed to the greater hydrolysis of the stannic ion decreasing its activity more rapidly than that of the stannous ion. From the graph (Fig. I) it may be seen that the standard oxidation-reduction potential for the system IrCl,_=—ezIrCl,= is 1-017 volts at 20°C. on the standard hydrogen activity scale. The value for Hy) obtained by Sho-Chow Woo (loc. cit.) in 1 Normal hydrochloric acid was 1-021 volts ; this result is lower than that obtained by the authors in this paper for that particular acid concentration. Two factors are suggested in explanation of the low result obtained by Sho-Chow Woo (loc. cit.)—the use of an iridium electrode, which the present authors have found is more readily attacked by chlorine and chloriridate ion than rhodium, and hence decrease the concentration of the oxidant at the electrode surface ; and the failure to take the precaution of chlorinating the iridate solution so that the concentration of the oxidant ion had not been reduced by traces of impurity during the preparation of the solution. EXPERIMENTAL. Preparation of Solutions. Ammonium Hexachloriridate. A warm concentrated solution of sodium hexachloriridate, made by heating iridium and sodium chloride in a current of chlorine, dissolving in water and filtering off the residue of undissolved iridium, was treated with excess ammonium chloride and allowed to crystallise overnight. The precipitated ammonium hexachloriridate was filtered through sintered glass, washed with hydrochloric acid at —10° C. and finally with ice cold water and dried at room temperature. 0-5 g. was dissolved in 100 ml. water (0-0113 molar), made approximately 0-063N with respect to hydrochloric acid, and divided into two parts. One part was treated with chlorine to oxidise any chloriridite which might be present and the excess chlorine removed with a current of air purified by passage through alkaline permanganate, acid permanganate and finally distilled water. The solution was then made up to the original volume with distilled water. Ammonium Hexachloriridite. ‘The second half of the above solution was refluxed with A.R. silver wire until reduction to trivalent iridium was complete, and the excess silver and silver chloride removed by filtration. Contrary to the impression gained from Sho-Chow Woo (loc. cit.), a small loss of iridium occurred during reduction, partly as insoluble silver hexachloriridite and partly as iridium metal. This was confirmed by dissolving the silver chloride out with ammonia when the greenish silver hexachloriridite was left insoluble; the latter on treatment with potassium cyanide left a grey deposit of iridium. The loss of iridium amounted to about 8%. The concentration of the chloriridite solution was determined by treating 5 ml. of the solution with an equal volume of hydrochloric acid then chlorinating and diluting to 100 ml. ; the latter solution was then compared colorimetrically, using a Klett colorimeter, with 5 ml. of the chloriridate solution of known concentration treated similarly. The molecular ratio of chloriridate to chloriridite was found to be 1-083: 1, for which a slight correction was necessary in the E.M.F. figures owing to the ratio of the concentrations of oxidant to reductant not being exactly 1:1. Titanium Solution. Approximately 10 g. pure redistilled titanium tetrachloride were dissolved in hydrochloric acid (5 normal) and reduced electrolytically with a lead cathode. The resultant solution was filtered and diluted with boiled out, distilled water and hydrochloric acid until the final concen- tration was 0-025 N with respect to titanous chloride and 1-113 N with respect to hydrochloric acid. Because of ease of aerial oxidation the titanous chloride solution was stored under carbon dioxide. 264 DWYER, MCKENZIE AND NYHOLM. APPARATUS. E.M.F. measurements were carried out by the usual Poggendorff compensation method using a Leeds and Northrup potentiometer capable of an accuracy of 0-1 millivolt. The galvano- meter was a Cambridge moving coil type with lamp and scale, sensitive to 10-° amp. A Weston standard cell was used. The electrode vessel consisted of a 250 ml. pyrex, round bottom, five neck flask. The various openings allowed the ingress of the salt bridge, rhodium electrode, stirrer, burette and carbon dioxide when required. A semi-micro burette graduated to 0-05 ml. was used. Electrodes. Drawn out, sealed pyrex tubes were thoroughly cleaned in chromic acid and dipped in a 4% solution of rhodium trichloride deep enough to cover a fine platinum wire wound around the top of the tube to make contact. After drying, the rhodium trichloride was ignited, leaving a fine silvery deposit of rhodium. This was repeated three times and the rhodium plate washed with hot aqua regia and water many times and the platinum contact covered with Apiezon wax. The resistance of the electrode was about 50 ohms for three inches length. The reference electrode consisted of a saturated calomel electrode connected through a saturated potassium chloride, agar-agar bridge to the electrode vessel. The potential of the calomel electrode was taken as 0: 2476 volt at 20° C. as recommended by Dole (1940). Other electrodes tried included pure platinum, rhodium-plated platinum and iridium-plated platinum. The deviations between the E.M.F. obtained whilst stationary and whilst stirring were initially of a much higher order than those obtained with the electrode finally adopted. Moreover, true equilibrium could not be reached with these electrodes irrespective of the time of stirring. Titrations of ammonium hexachloriridate with titanous chloride were carried out in carbon dioxide following an early suggestion that the oxidation-reduction potential might be obtained by this method by measuring H at 50% reduction. Owing to slowness in attaining equilibrium, results were too low. Potential mediators such as ceric sulphate and potassium iodide also failed to bring about equilibrium. Figure III shows the curve obtained in a typical titration. It will be observed that the fall in potential during the addition of the first 25% of reducing agent is much greater than might be expected. This is no doubt due to a false equilibrium being reached. It is of interest to note that the values for H# obtained by this method also diminished with decreasing ionic strength, a curve of the same shape as Fig. I being obtained but all values were approximately 0-1 volt lower. Experimental Procedure. Equal volumes (5 ml.) of the hexachloriridite and hexachloriridate together with the appropriate amount of hydrochloric acid were made up to 30 ml. with distilled water. The mixture was stirred in contact with the electrode for many hours, a stable potential being reached after approximately one hour; this potential remained constant for periods up to two days. Initial potentials were approximately 15 millivolts too high and before equilibrium was reached there was a considerable difference between the potential when at rest and when stirring. How- ever when equilibrium was established there was only a very small difference of the order of 0-2 to 0:3 millivolt. Although thermodynamic theory does not take into account a mechanical movement of the liquid relative to the electrode, which may interfere with the establishment of the electrical double layer, without thorough mixing there is no guarantee of definite equilibrium throughout the whole system, and in particular in the surface layer adjacent to the electrode. The effect of stirring is, however, seen to be negligible. Over protracted periods the salt bridge was removed to prevent reduction, although tests showed this to be negligible. Stirring with carbon dioxide was found to give no detectable difference to that shown by mechanical stirring. Experimental results are shown in Table I, whilst the relation between 4/f and EH is shown on the graph (Fig. I). / Ee a ee 5 (to eg oi Die CHEMISTRY OF BIVALENT AND TRIVALENT IRIDIUM. 265 TABLE’ I. Table of Results. | | HCl | | E.M.F. Normality. /I HCl. | /f Total. | (Corrected). 2-037 1-428 1-433 1-065° 1-029 | 1-014 | 1-023 1-0612 0: 4890 0-699 | 0-711 1-0562 0-092 0-303 | 0-330 1-050° 0-04.94 0-222 | 0-257 1-046 0-0423 | 0-206 0-244 1-0467 0-0281 | 0-167 0-212 1-044? 0-0210 | 0-145 0-195 1-0443 0-0126 0-112 0-152 1-033} 0-0063 0-079 0-107 1-0293 SUMMARY. The standard oxidation-reduction potential for the hexachloriridite-hexa- chloriridate system has been determined by extrapolation from values obtained in hydrochloric acid of various concentrations. At 20°C. this has been found to be 1-017 volt. REFERENCES. Dole, 1941. ‘“‘ The Glass Electrode’, Wiley, New York. Huey and Tartar, 1935. J. Amer. Chem. Soc., 56, 2585. Kolthoff and Tomsicek, 1935. J. Phys. Chem., 39, 945. Ogawa, 1929. J. Chem. Soc. Japan. 50, 123. Sho-Chow Woo, 1931. J. Amer. Chem. Soc., 53, 469. Chemistry Department, Sydney Technical College. THE CHEMISTRY OF BIVALENT AND TRIVALENT RHODIUM. Part VIII. RwHopIc AND RHODOUS COMPLEXES WITH DIMETHYLGLYOXIME. By PSP. DWYER, iUSe. and R. 8. NYHOLM, MSc. Manuscript received, November 22, 1944. Read, December 6, 1944. During the study of bivalent rhodium complexes, and in particular during the search for square complexes, the rhodic and rhodous compounds with dimethylglyoxime have been investigated. Rhodium in the bivalent state might be expected to react analogously to bivalent cobalt with this grouping and yield tetracovalent as well as hexacovalent compounds, similar to the simple bis compound Co(DH),* (Mellor and Craig, 1940) and the halide complexes [CoCl,.(DH,)] and [CoCl,(CH,),] (Thilo, 1931). The rhodic compounds M.[Rh(NO,).(DH),.].2H,O have been described by Lebedenski and Federov (1938), whilst Malatesta and Turner (1942), during an investigation of the colour reaction between dimethylglyoxime, cobalt chloride and sodium polysulphide, prepared the analogous rhodium compound Na[{Rh(DH).8,]. There appears to be no record, however, of the reaction between the simple rhodic halides and dimethylglyoxime. Solutions of the rhodic halides in varying acid concentrations from approxi- mately five normal down to just sufficient acid to prevent hydrolysis, have been found to react with dimethylglyoxime to yield sparingly soluble bis-halide, bis-oxime complexes of the general formula H[{RhX,.(DH),], which function as strong monobasic acids, giving stable, insoluble silver salts, and stable, water- soluble alkali metal, barium and ammonium salts. Treatment of these com- pounds with excess dimethylglyoxime and sodium acetate or caustic soda failed, even after many hours’ boiling, to replace the halogen groups. The complex acids were soluble in ammonium oxalate, demonstrating their pronounced acidity, but many hours’ refluxing failed to yield an oxalato compound, and the ammonium salt of the original acid crystallised out on evaporation. The halogen eroups, thus, probably occupy trans positions in the octahedral complex—a conclusion reached by Lebedenski and Federov (loc. cit.). The halogen could not be precipitated by boiling with silver nitrate, which merely yielded the silver salt—itself easily soluble in dilute nitric acid. The halogen atoms are thus bonded covalently. Crystalline pyridine salts could be isolated, but they easily lost pyridine during drying. From titration studies, and conductivity determinations on the chloride, which was the most soluble, it has been found that although three hydrogen atoms are present which might function as acidic, the substance acts exclusively aS a monobasic acid, comparable in strength with hydrochloric acid. This observation has an interesting bearing on the structure of the well-known nickel dimethylglyoxime (I), which, although it has two hydrogen atoms that might be acidic, has been shown to fail in the reaction with phenylisocyanate (Tschugaeif, 1905) or with acetic anhydride (Barker, 1925) and yields no methane with methyl magnesium iodide (Brady and Muers, 1930). This lack of reactivity has been * DH, = Dimethylglyoxime, H, . C,H;N,O,. ee ae CHEMISTRY OF BIVALENT AND TRIVALENT RHODIUM. 267 suggested (Brady and Muers, Jbid.), as due to hydrogen bonding and a similar explanation seems feasible for the pronounced monobasic character of the rhodium compounds (II). The rhodic compound [Rh(DH),] (III) was ultimately prepared in small yield from rhodic sulphate as a brown, highly insoluble powder. It was insoluble in alkalis, but dissolved in dilute hydrochloric acid to a reddish solution, con- taining probably the cis-dihalide-bis-glyoxime compound. The reddish solution lightened gradually and deposited the usual yellow trans compound (II). “ . ‘Be sf C-CH3 eh N—> 0 O—=N N—>0 d a eye ; \ H Na H L H k——Rh—x ' eae C-CH3 CHa-C Wee Tage Seen et II The rhodic halide complexes on boiling with hypophosphorous acid in halogen acid solution gradually dissolved, but gave only a complex with hypo- phosphorous acid, the oxime apparently being destroyed by this reducing agent. In neutral or slightly acid solution in the presence of dimethylglyoxime, sodium formate, however, gave a black substance which very rapidly changed to the brown tris-glyoxime rhodic complex (III). This compound was soluble in caustic soda to a dark brown solution, which gave the original black substance on acidification, and reduced silver nitrate in the cold to the metal. Although, owing to oxidation, it was not possible to obtain specimens pure enough for analysis, it is probable that this is the tris-glyoxime rhodous compound H[Rh(DH),] (IV). The black compound (IV) dissolved in warm dilute hydrochloric acid to an intense purplish-red solution, which very rapidly became lighter in colour due to oxidation. The purplish-red substance would appear to be the rhodous halide complex H,[RhCl,(DH),] (V), but all attempts at isolation failed. The reddish oxidised solution on evaporation at room temperature gave a mixture of red and yellow crystals. The yellow substance was readily identified as the usual rhodic halide complex (II), and the reddish crystals from tests applied under the microscope with individual crystals separated by hand were found to react strongly acid to litmus, were soluble in sodium acetate and caustic soda, and gave an orange-yellow silver salt. An aqueous solution on evaporation deposited yellow crystals of the usual complex (II). As the mode of preparation and reactions suggest the red compound is probably the cis complex H[RhC1,(DH), | (VI). W-—December 6, 1944. 268 DWYER AND NYHOLM. In the structural formulation of compounds (IIT), (IV), (V), (VI), hydrogen bonding in the manner of (II) presents some difficulties since the chelate rings are skewed with respect to each other and sufficient distortion might then arise to prevent the formation of the relatively weak hydrogen bond. On the other hand if the normal hydroxyl oxime form is postulated, the tris-oxime complex (IIT), for instance, might reasonably be expected to be at least as acidic as the free oxime. Actually, it is quite insoluble in alkali. For this reason it is felt that some form of hydrogen bonding is present in these compounds. Titration of 25 m/l Sally rotéed Solution H[RhD, Cy] wilh 0.0214N: Baryla. Glass Electrode.) On then 2 uae Aue ae clue a ee PU of BalOH),~- (0-0214N) Curve 1]. CHEMISTRY OF BIVALENT AND TRIVALENT RHODIUM. 269 EXPERIMENTAL. Monohydrogen dichloro-bisdimethylglyorime rhodatc™™. A slightly acid solution of rhodium trichloride (20 ml.), containing 0-19 g. rhodium was treated with dimethylglyoxime (0:8 g.), in alcohol (15 mls.), and refluxed for ten minutes. On cooling golden yellow rod-like crystals were precipitated. The crystals showed pronounced twinning, with characteristic V-shaped indentations at each end. A further yield was obtained by concentration of the mother liquor. The compound was washed with alcohol and finally ice water. On heating, the substance decomposed violently. It was slightly soluble in water, yielding an approximately 0-005 molar solution, which reacted strongly acid having a pH of 2:3. The compound was easily soluble in ammonia, caustic soda, sodium acetate and ammonium oxalate solutions, yielding salts, which regenerated the original substance by treatment with hydrochloric acid, but was only sparingly soluble in organic solvents. For the conductimetric measurements, the pure silver salt was decomposed with the theoretical weight of sodium chloride by shaking at room temperature for ten hours, and the determinations made on the sodium salt so obtained. Aj .,—65:5 and A,.=56:7. Hence difference=8:-8. From the expression Aj9,--A32==10 x basicity, the acid is monobasic. The pH curve of the titration of a saturated solution of the substance with barium hydroxide using a glass electrode is shown in titration curve 1, and shows the substance to be a strong monobasic acid. Found: Rh=25-:7%; Cl=17:4%. Calculated for H[RhCl,(C,N,H,O,).]: Rh=25-4% Cle=17+58%. Silver dichloro-bisdimethylglyoxime rhodate The acid compound above (0-2 g.) was dissolved in sodium acetate solution, and treated in warm solution with a warm solution of silver nitrate (0:28 g.). The initially amorphous yellow precipitate rapidly crystallised on shaking and was filtered off and washed with warm water until the washings were free of silver ion. The compound was soluble in dilute nitric acid to a clear solution, and precipitated the halogen only on heating with an acid sufficiently concentrated to destroy the oxime group. Found: Rh=20-:3%; Cl=13-9%. Calculated for Ag[RhCl,(C,N,H,O,),]: Rh—20:0%: Cl 1s 37%. Monohydrogen dibromo-bisdimethylglyoxime rhodate. This compound was prepared as for the chloro compound above, using rhodic bromide in hydrobromic acid solution. The bright yellow rod-like crystals were sparingly soluble in water to an acid solution, and gave an orange- silver salt. Hound: Rh=—20:8%; Br=—32-4%. Calculated for H[RhBr,(C,N,H,O,).]: Rh=20:8% ; Br=-32-39%. Til Monohydrogen di-iodo-bisdimethylglyoxime rhodate™!, Rhodium trichloride solution (20 mls.) was treated with potassium iodide (5 g.) and heated to 80°C. Dimethylglyoxime in alcohol as before was now added and the mixture refluxed until a brown crystalline precipitate appeared. The compound was very sparingly soluble in water, but the solution reacted acid, and gave a reddish brown silver salt. The silver salt on solution in nitric acid and warming gradually deposited silver iodide. Found: Rh=17-58%. Calculated for H[RhI,(C,N,H,O,),]: Rh=17-50%. Tris-dimethylglyoxime rhodium™. A weakly acid solution of rhodic sulphate (20 mls.) containing 0-19 g. rhodium was treated with dimethylglyoxime 0-6 g. in alcohol (15 ml.) and refluxed for some time. A solution of sodium acetate 3 N. (20 ml.) was then added and finally caustic soda, until the solution was very faintly alkaline. A dark brown precipitate gradually came down, and the solution was then made acid with acetic acid. The brown precipitate was filtered, and washed many times with alcohol. It was insoluble in all organic solvents, but dissolved in hydrochloric acid to a reddish solution, which gradually lightened and deposited yellow crystals of the chloro complex. Found ; Rh—22-9%. Calculated for [Rh(C,N,O,H,),]: Rh==22-9%. Mono-hydrogen-tris-dimethlyglyoxime rhodium™. The chloro rhodic compound above (0°5)g. was dissolved in aqueous sodium acetate (20 mls.), heated to boiling with dimethylglyoxime X—December 6, 1944. 270 ' DWYER AND NYHOLM. (0:5 g.) in 15 mls. of alcohol, and a 20% solution of sodium formate added drop by drop until a black precipitate appeared. The solution was then rapidly cooled to prevent complete reduction to rhodium metal, and filtered. The black precipitate almost instant]y commenced to become brown due to aerial oxidation. It proved impossible to dry the substance without oxidation occurring. ‘The substance was soluble in caustic soda to a dark brown solution. A solution in alcohol, in which it was shghtly soluble, instantly reduced silver nitrate to the metal, but even on complete decomposition with nitric acid no halogen could be found. | Mono-hydrogen-cis-dichloro-bis-dimethylglyoxime rhodate“!. The rhodous compound above was dissolved in warm hydrochloric acid 2 N., when it gave an intense purplish-red solution of (probably) the cis-dichloro rhodous compound, and rapidly commenced to lighten in colour. By shaking in air for a minute this oxidation was completed to a lighter reddish solution, which was then evaporated rapidly in a current of cold air. A mixture of yellow rods of the usual trans- dichloro compound and orange-red prisms was obtained. By recrystallisation much of the red compound was transformed to the yellow compound, but the reactions of the red substances were determined (q.v. swpra) on individual crystals under the microscope. Even from a number of . preparations, insufficient material was obtained for analysis. SUMMARY. Rhodic halides and dimethylglyoxime yield in addition to the tris-glyoxime complex, a series of strong, complex, dihalogen bis-glyoxime acids, in which the metal is always hexacovalent. The rhodous compounds were also hexacovalent, and oxidised very rapidly in the air. REFERENCES. Barker, 1925. Chem. News, 130, 99. Brady and Muers, 1930. J. Chem. Soc., 99. Lebedenski and Federov, 1938. Ann. sec. plat. Inst. Chem. gen. U.S.S.R., 15, 19. Malatesta and Turner, 1942. Gazz. chim. ital., 72, 489. Mellor and Craig, 1940. THis JouRNAL, 74, 475. Thilo, 1931. Sammlung. chemischer-technischer, 13. Tschugaeff, 1905. Z. anorg. Chem., 46, 144. Department of Chemistry, Sydney Technical College. CORRECTION OF PAPER “ON CONTACT TRANSFORMATIONS ASSOCIATED WITH THE SYMPLECTIC GROUP ”’. (THIS JOURNAL, Vol. LX XVI, 1942, 177-181) By H. SCHWERDTFEGER. It has been pointed out by M. 8. Knebelman (Math. Reviews 4, 1943, 184) that the simple substitute for the lemma of Radon used in my proof of the theorem on the symplectic determinant (6=-+1) is incorrect. Hitherto I have not been able to provide for Radon’s original lemma (mentioned in footnote 5, l.c. 180) a purely algebraic proof such that it would be possible to save the simple formal structure of the proof of the above theorem which, indeed, was the main purpose of my paper. While I still hope to find an algebraic proof of Radon’s lemma, I have now reasons to believe that such a proof is not going to be simple. Therefore the proof of the theorem based upon the lemma will not excel by simplicity all the existing proofs for which references have been given in my paper. xX Ra i ytd ’ -—- a. hinted ate ole Pe bn ABSTRACT OF PROCEEDINGS OF THE Royal Society of New South Wales April 5th, 1944. The Annual Meeting, being the six hundred and twelfth General Monthly Meeting of the Society, was held in the Hall of Science House, Gloucester and Essex Streets, Sydney, at 7.45 p.m. The President, Dr. A. B. Walkom, was in the chair. Forty-three members were present. The minutes of the previous meeting were read and confirmed. The following were elected officers and members of the Council for the coming year ! President : G. D. OSBORNE, D.sc., Ph.p. Vice- Presidents : IDA A. BROWN, D.Sc. | Pror. H. PRIESTLEY, m.p., ch.m., B.Sc. A. BOLLIGER, Ph.p., A.A.C.1. | A. B. WALKOM, p.se. Hon. Secretaries : Pror. A. P. ELKIN, .a., Ph.p. | D. P. MELLOR, M.sc. Hon. Treasurer : A. CLUNIES ROSS, B.sc., F.c.A. (Aust.). Members of Council: R. L. ASTON, B.sec., B.E., M.Sc., Ph.D., W. H. MAZE, .sc. A.M.1.E. (Aust.). F. R. MORRISON, 4A.4A.C.1., F.C.S. J. A. DULHUNTY, B.se. R. 8S. NYHOLM, M.sc. F. P. J. DWYER, M.Sc. H. H. THORNE, o.a., B.Sc., F.R.A.S. N. A. FAULL, M.sc. H. W. WOOD, M.Sc., A.Inst.P., F.R.A.S. F. LIONS, B.sc., Ph.p., A.R.I.C. XXli ABSTRACT OF PROCEEDINGS. The Annual Balance Sheet and Revenue Account were submitted to members by the Honorary Treasurer, and on the motion of Mr. A. Clunies Ross, seconded by Mr. A. E. Stephen, were adopted. 1943. £31,552 9,038 14,756 6,800 436 38 14 £31,552 THE ROYAL SOCIETY OF NEW SOUTH WALES. BALANCE SHEET AS AT 29th FEBRUARY, LIABILITIES. Accrued Expense : Subscriptions Paid in Advance Life Members’ Subscriptions—Amount Carried Forward Trust and Research Funds (detailed below)— Clarke Memorial . BPs Pe at Walter Burfitt Prize.. Liversidge Bedien fe Research shee Accumulated Funds (detailed below) Contingent Liability—In connection with perpetual leases granted to the Australian National Research Council and the Pharmaceutical Society of N.S.W. Maximum liability, £901 16s. 8d. ASSETS. Cash at Bank and in Hand .. : Investments—Commonwealth Bonds ‘gad indented Stock—at Face Value— Held for Clarke Memorial Fund Held for Walter Burfitt Prize Fund Held for Liversidge Bequest Held for Research Fund Held for General Purposes Prepayment ‘ Debtors for Subscriptions Deduct Reserve for Bad Debts Science House—one-third Capital Cost Library (at Valuation) ‘ Furniture (at Cost—less Depreciation) Pictures se ae ay Lantern... oe oe 1944, 1944. f° £ s. d 35 0 0 9 9 O ——— 44 9 O 81 0 0 1,884.94 oF 725 15 O 725 15 0O 3,049 5 O —— 6,334 19 7 25,397 11 11 £31,858 0 6 1944. £ ssid: £ s. d. 450 18 6 1,800 0 O 700 0 O 700 0 0 3,000 0 O 3,160 0 0 =— — 9,360 0 0O 28 12 0 101 1 =O LOD fk 14,756 0 0 6,800 0 0 414 0 0 35 10 0O £3.) Oo £31,858 0 6 ABSTRACT OF PROCEEDINGS. TRUST AND RESEARCH FUNDS. Walter Clarke Burfitt Memorial. Prize. Ee ee ale Ser Sa. Cd. Balance at 28th February, 1943 TOSS 2690. 5) 6 Transfer from Accumulated Funds bt’ 4s 4 ONT AT Capital at 29th February, 1944 £1800" 010 S700" 0" 0 Revenue—Interest on Investments... 65 10 2 Zoo oO Deduct Expenditure SH, Dunia — Balance at 29th February, 1944 .... £34. 4 (57 £25 15 0 ACCUMULATED FUNDS. Balance at 28th February, 1943 Deduct Transfer to other Funds Add ialyastmont of Investments Reduction of Reserve for Bad Dobte” Surplus for Year (see Income and Expenditure Account) Balance at 29th February, 1944 Liversidge Bequest. case 2] d. 680 16 3 De oo £700: (0° 0 Xxil Research. £ s. d. 3,000 0 O £3,000 0 0O aA 0 O 99 50 £49 5 O A. CLUNIES ROSS, Hon. Treasurer. The above Balance Sheet has been prepared from the books of account, accounts and vouchers of The Royal Society of New South Wales, and is a correct statement of the position of the Society’s affairs on the 29th February, 1944, as disclosed thereby. showing that the whole of the Bonds and Inscribed Stock are held by the Society’s Bankers for safe keeping. Prudential Building, 39 Martin Place, Sydney, 15th March, 1944, We have obtained certificates HORLEY & HORLEY, Chartered Accountants (Aust.). Xxiv ABSTRACT OF PROCEEDINGS. INCOME AND EXPENDITURE ACCOUNT. ist March, 1943, to 29th February, 1944. 1942-3 1943-4 £ £0 4g. 5. eae 268 To Printing and Binding Journal—Vol. 76 on a5 412 ire 293 +~«,, Salaries a “ Ay tee 302 12 O Ot fs, Library—Purchases: ‘and Binding $e *g ay 117 8 6 50 ~,, Printing—General . sy es or 72 17 11 67 ,, Miscellaneous : a ah ss ‘ ie 55 8.4 58 ,, Postage and Telegrams ae ee 53 14 11 35 ,, Rent—Science House Management Cammuntie! a 43 3 8 36 ~=,,_~Cleaning Ae he eo bye a, ae ee 36 0 0O 26 = ,, Depreciation on Me : an rate ee 25 0 0 21 ,, Telephone... Be ufis ne f Be Bie 17 11 11 14. ,, Insurance .. se a ps a Ras hs 1412 2 13 —,, Audit oe aie “yn 5s - a. Ei 12 12 0 7 4, Electricity se he fs ee es sas LI 2 Qies », Repairs he he Bo ee ree: ie a2 510 0 945 ———— 1,180 2 3 481 ,, Surplus for Twelve Months Ry Fp as sas 112 19 6 £1,426 £1,293 - 4.9 1942-3 1943-4 £ £ SB £ ss 485 By Membership Subscriptions .. a ope we 470 8 0 400 ,, Government Subsidy a or ane 400 0 0O 315 ,, Science House—Share of Surplus sis rae as ; 300 0 O 196 ,, Interest on General Investments .. an fs 105 4° 8 30 ,, Receipts from Reprints wks we =A bie 21) a 32 Less Expenditure. . a Bs ae As 17 6 O = —2 — 2 aoa Meany | 9 ,, Other Receipts. . iG rages Reni} 23 +4, Proportion of Life Members’ Subscriptions bee Lk 1050 £1,426 : £1,293. 59 The Annual Report of the Council (1943-44) was read, and on the motion of Professor Elkin, seconded by Mr. EK. Cheel, the report was adopted. REPORT OF THE COUNCIL, 1943-1944 (RuLE XXVI). We regret to report the loss by death of three members since April Ist, 1943: James Edward Mills (1940), John Job Crew Bradfield (1922) and Archibald Howie (1936); also of an honorary member, Frederick Chapman (1939). By resignation the Society has lost five members: Mrs. Margaret Maccoll, Miss V. Suvoroff, Professor J. Macdonald Holmes, Alban J. M. Murray and Alan M. Willison. The names of thirteen members have been removed from the register and their arrears have been written off. The membership now stands at 287, eighteen new members having been elected during the year, namely William Kevin McCoy, William Dudgeon, Ronald Arthur Plowman, Leonard Winch, Jean Annie Kimble, Thomas Iredale, Ivan Stewart Turner, Reginald John Nelson Whiteman, Ernest Patrick Molloy, John William George Neuhaus, James Foote Walker, Mrs. Daphne Luber, Howard Hamlet Gordon McKern, John Kenneth Moore Simpson, Alexander Campbell Nicol, Barbara Joyce Burfitt and John Stanton Burkitt. Eleven ordinary meetings of the Council were held during the year beginning Ist April, 1943, at which the average attendance was 14. During the same period nine general monthly meetings were held, with an average attendance of 35 members. Kighteen papers were accepted for reading and publication during the year, and the following short talks were given: ‘* Reflection of Light from Film-covered Glass’, by Mr. J. Bannon, B.Sc. ** Evaporated Metal Films’’, by Mr. F. P. J. Dwyer, B.Sc. ** Biotin ’’, by Professor H. Priestley, M.D., Ch.M., B.Sc. “Penicillin and Gramicidin’’, by Dr. F. Lions. ‘‘The Future of the Native Peoples of the South-west Pacific’, by Professor A. P. Elkin. ABSTRACT OF PROCEEDINGS. XXV An exhibit—‘‘ A New Source of Light—the Fluorescent Lamp ’’—was given by Mr. D. P. Mellor, M.Sc. Popular Science Lectures.—On account of the lifting of lighting restrictions it was decided to revert to the holding of Popular Science Lectures during the winter months of 1943, and the following lectures were given : ‘**How We Came to Stand Upright ’’, by Professor Harvey Sutton, O.B.E., M.D., Ch.B., DeP.H., BSc. ‘* Veterinary Science and the Community ’’, by Mr. H. Parry, B.A. ‘** Exploring the Inside of the Earth ’’, by Professor L. A. Cotton, M.A., D.Sc. “Soviet Research on Applied Botany ”’, by Professor Eric Ashby, D.Sc. ** Architecture: the Setting for Life ’’, by Professor L. Wilkinson, F.R.I.B.A., F.R.A.I.A. The attendances at most of the lectures were very good. Clarke Memorial Lecture.—The Clarke Memorial Lecture for 1943 was given by Dr. H. G. Raggatt, Director of the Commonwealth Mineral Resources Survey, and was entitled “‘ Australia’s Mineral Industry in the Present War ”’. Clarke Memorial Medal for 1943.—The Medal was awarded to Dr. W. L. Waterhouse, in recognition of his contributions to plant pathology, in particular, his researches on rust in wheat. The Royal Society’s Medal was awarded to Mr. Edwin Cheel for his contributions to botanical science and for his work for the advancement of science in general. This was the first award of the Society’s medal since 1896. The Bi-centenary of the Birth of Sir Joseph Banks was commemorated at a general meeting of the Society, when a lecture on “Sir Joseph Banks and Australia’? was given by Dr. G. Mackaness. Commemoration of Other Events.—At the November monthly meeting the 400th anniversary of the publication of two important works was celebrated, namely ‘‘ De Revolutionibus Orbium Coelestium ’’, by Copernicus, and *“‘De Humani Corporis Fabrica’’, by Vesalius. Lantern lectures were given by the Rev. Father D. J. O’Connell, on Copernicus, and by Dr. Leslie Cowlishaw, on Vesalius. Professor Shellshear, in proposing a vote of thanks to the lecturers, pointed out how Copernicus and Vesalius expressed in their works the urge for truth which was abroad in their period. Centenary of the Foundation of the Royal Society of Tasmania.—The greetings of the Royal Society of New South Wales were sent to the Royal Society of Tasmania, which celebrated its centenary on October 12th, 13th and 14th, 1943. The occasion was marked by the striking of special medals, to be presented to the Royal Society of London, and the Royal Societies in each State of the Commonwealth, and to Professor Ashby and Dr. Mackaness, who had been invited to deliver lectures during the centenary celebrations. Supply of Paper for Scientific Periodicals.—Attention having been drawn to a possible shortage of paper for the printing of scientific periodicals, it was resolved to write to the Chairman of the Book Publication Committee, pointing out the necessity for continuing the publication of research work, and asking that paper be made available for the purposes of the scientific journals. A reply was received intimating that careful consideration was being given to technical and scientific publications, and that when possible to afford some relief to the scientific journals the Royal Society would be advised. Government Grant.—A grant of £400 was received from the Government of New South Wales. The Government’s continued interest in the Society’s work is much appreciated. Science House.—The Royal Society’s share of the profits on Science House during the period from March Ist, 1943, to February 29th, 1944, has been £300. Science House Management Committee.—The Royal Society has been represented at meetings of the Management Committee of Science House by Mr. A. R. Penfold and Dr. G. D. Osborne, with Mr. Alan Clunies Ross and Dr. F. Lions as substitute representatives. On the resignation of Dr. Lions, Mr. H. H. Thorne was appointed in his place. Medals.—A gift was received from Mr. Henry F. Halloran, one of our senior members, for the purpose of providing two medals, one of which was to be awarded for distinguished services to science, and human welfare in the Southern Hemisphere. After consultation with the donor, it was decided to call this medal ‘‘ The James Cook Medal ”’, and the other one ‘‘ The Edgeworth David Medal’’. A special committee has consulted with Mr. H. E. Dadswell concerning the designs, and finality will be reached in the near future. With these two medals the Society will be responsible for the award of five medals : the other three are the Society’s own medal (revived after a lapse of 48 years), the Clarke Memorial Medal and the medal which goes with the Walter Burfitt Prize. xxvil ABSTRACT OF PROCEEDINGS. The Council decided that the scope of these awards should be as follows : The Edgeworth David Medal, donated by Mr. Henry Halloran, to encourage the younger research worker, is to be presented annually for distinguished contributions in one of the following fields, namely biological science, physical science, or social science, such fields of research to be considered in turn in successive years. Recipients shall be under thirty-five years of age, and the work done shall be in or on Australia and New Zealand. The Clarke Memorial Medal may be awarded from time to time for distinguished work in natural science, done in or on the Australian Commonwealth or its territories; the person to whom the award is made may be resident in the Australian Commonwealth or its territories, or elsewhere. Natural science, for the purpose of the award, has been defined as geology, zoology and botany. The Walter Burfitt Prize and Medal consists of a bronze medal and a money prize of £50, awarded at intervals of three years to the worker in pure and applied science resident in Australia or New Zealand whose papers or other contributions published during the past three years are deemed to be of the highest scientific merit, account being taken only of investigations described for the first time, and carried out mainly in these Dominions. The Royal Society’s Medal may be awarded at any time at the discretion of the Council to any member of the Society who has made meritorious contributions to the advancement of science in Australia. The award will be considered annually, but not necessarily awarded annually. The James Cook Medal, donated by Mr. Henry Halloran, is to be awarded not more than once annually for outstanding contributions to science and human welfare in and for the Southern Hemisphere. For the purposes of this award, there shall be no limitations of the field of science or its applications, and Southern Hemisphere shall include South Africa, South America, Australia, New Zealand, all the Dutch East Indies and the islands of the southern seas. Research Grants.—It was resolved to set aside a portion of the funds of the Royal Society of New South Wales to form the nucleus of a research fund, the income from which is to be made available to assist research workers. Finance.—The audit of the Society’s accounts shows that the finances are in a satisfactory condition. During 1943, £300 has been invested in war loans. The Royal Society’s Library.—Librarian.—Mr. W. H. Maze, who had done much good work for the library, tendered his resignation owing to pressure of other duties, and Mr. H. W. Wood was appointed in his place. Dr. A. Bolliger acted as assistant librarian. Purchase of Periodicals and Binding.—The amount of £30 12s. 6d. has been expended on the purchase of periodicals, and the amount of £88 16s. has been spent on binding. The total amount expended on the library was thus £117 8s. 6d. Of this, £16 7s. 6d. was for binding done in 1942, so the total for 1943 is actually £101 ls. Exchanges.—The number of volumes now being sent to other societies is 225. Accessions.—For the twelve months ended in February, the number of accessions entered in the catalogue was 1,585 parts of periodicals and 136 whole volumes. Borrowers and Readers.—Members and visitors reading in the library numbered thirty-seven. The number of books and periodicals borrowed by members, institutions and accredited readers was seventy-one. Among institutions which made use of the arrangements for inter-lbrary borrowing were : The University of Sydney (Fisher Library and Botany School), the University of Western Australia, the C.S.I.R. Food Preservation Laboratory, the McMaster Laboratory, the Veterinary Research Station, Glenfield, the National Standards Laboratory, the National Herbarium of Melbourne, the Commonwealth Forestry Bureau, Canberra, the Sydney Technical College, Department of Public Works, Sydney, the Standards Association of Australia, the Colonial Sugar Refining Company, the Sydney County Council, and Amalgamated Wireless Research Laboratory. Disposal of Old Books.—A number of out-of-date printed lists of members of several societies abroad, such as the Iron and Steel Institute, Institution of Mechanical Engineers, England, and the American Society of Civil Engineers, have been disposed of. Text books on mathematics and astronomy were presented to the University of Sydney, the Sydney Observatory and Riverview Observatory. Supplement to Pitt’s Catalogue.—At the request of the Institute of Librarians, a list to date of periodicals received in the Royal Society’s library was prepared and forwarded for reference in the Public Library. This action was suggested by the Institute of Librarians because, owing to wartime conditions, it has not been possible to issue a third supplement to Pitt’s Catalogue. It was resolved to make available to other libraries, on request from the Public Library, bound volumes containing specific articles. It was decided that unbound parts could be referred to in the Royal Society’s library on a request being received from the Public Library. 1 a } ABSTRACT OF PROCEEDINGS. XXVil Duplication in the Libraries of Science House.—The review of the list of periodical sets common to the libraries of the Royal Society of New South Wales and the Linnean Society of N.S.W. is now complete. It has been divided into three main categories, namely sets which should be retained by both libraries, sets which should be retained by the Linnean Society only, and sets which should be retained by the Royal Society only. The acceptance of such division would serve to relieve the congestion in the library very considerably. The matter has still to be considered by the Council. The certificates of two candidates for admission as ordinary members of the Society were read for the first time. Clarke Memorial Medal.—The announcement was made of the award of the Clarke Memorial Medal for 1944 to Professor W. E. Agar. Election of Auditors ——On the motion of Professor Elkin, seconded by Mr. R. W. Challinor, Messrs. Horley and Horley were re-elected auditors of the Society for 1944-45. Library.—The following donations were received : 350 parts of periodicals, 20 whole volumes. The President, Dr. A. B. Walkom, delivered his address, entitled ‘‘ The Succession of Car- boniferous and Permian Floras in Australia ’’. Dr. A. B. Walkom then installed Dr. G. D. Osborne as President for the year 1944-45. Dr. Osborne thanked the members for the honour they had done him in electing him President. He then called upon Mr. C. A: Sussmilch to propose a vote of thanks to the retiring President for his interesting address, and for the work he had done for the Society during his presidential term. This was carried by acclamation. The following papers were read by title only : ‘* Generalisation of Maxwell’s Equations ’’, by P. Foulkes. (Communicated by Prof. E. M. Wellish.) ‘“Some Lower Cretaceous Foraminifera from Bores in the Artesian Basin, Northern New South Wales ’’, by Irene Crespin, B.A. (Communicated by Dr. Ida A. Brown.) ‘‘ A Note on the Role of the Nitrosyl Group in Metal Complexes ”’, by D. P. Mellor and D. P. Craig. May 3rd, 1944. The six hundred and thirteenth General Monthly Meeting of the Royal Society of New South Wales was held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Twenty-two members were present. The minutes of the previous meeting were read and confirmed. The death was announced of the following member: Dr. Norman Dawson Royle, a member since 1929. The certificates of four candidates for admission as ordinary members of the Society were read for the first time. The certificates of two candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : Alwynne Drysdale Bennett and Daniel George Moye. Clarke Memorial Lecture.—It was announced that the Clarke Memoria! Lecture for 1944 would be delivered at Science House on May 30th, at 8 p.m., by Dr. W. H. Bryan, M.C., the title being “An Australian Geologist Looks at the Pacific ”’. Inbrary.—The following donations were received: 75 parts of periodicals and one whole volume. Correspondence.—A letter calling attention to the agenda paper for the A.A.S.W. Conference on “‘ The Planning of Science’ was read. The following papers were read : ‘“Geomorphology of the Central Eastern Area of New South Wales’, by W. H. Maze, M.Sc. Papers read by title at the April meeting were open for discussion. Miss Irene Crespin, who was present as a guest, discussed her paper on ‘‘ Some Lower 3 Cretaceous Foraminifera from Bores in the Artesian Basin, Northern New South Wales ”’. Lecturette.—A lecturette on ‘“‘ Potential Topography in Electronics ’’ was given by Dr. A. L. Reimann. June 7th, 1944. The six hundred and fourteenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Thirty-two members and six visitors were present. The minutes of the previous meeting were read and confirmed. XXVili ABSTRACT OF PROCEEDINGS. The certificates of four candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : George William Kenneth Cavill, Cyril Maxwell Martin, Andrew David Thomas and Colin Lachlan Adamson. Popular Science Lecture.—It was announced that the first Popular Science Lecture for 1944 would be given by A. N. Colefax, B.Sc., on Thursday, 15th June, at 8 p.m., the title being “‘ Man and Heredity ”’. Correspondence.—A letter was received from Dr. J. Pearson, Honorary Secretary of the Royal Society of Tasmania, advising that a replica of the medal which was struck on the occasion of the centenary of the Royal Society of Tasmania was being forwarded to the Royal Society of N.S.W. The medal had arrived safely, and was on view at the meeting. Inbrary.—The following donations were received: 85 parts of periodicals and one whole volume. Lecturettes.—The following lecturettes were given, and were illustrated by lantern slides and special apparatus : ‘* Physical Aspects of Vision’’, by R. G. Giovanelli, M.Sc. ““Some Aspects of Physiological Black-out in Aviation’, by Professor F. 8. Cotton, D.Sc. July 5th, 1944. The six hundred and fifteenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Thirty members and two visitors were present. The minutes of the previous meeting were read and confirmed. The death was announced of the following member: Sir Thomas Ranken Lyle, an honorary member since 1931. Popular Science Lecture.—It was announced that the second Popular Science Lecture for 1944 would be given by J. A. Dulhunty, B.Sc., on Thursday, 20th July, at 8 p.m., the title being ““ Fuels in the Service of Man’’. Library.—The following donations were received: 167 parts of periodicals and nine whole volumes. The following papers were read : ‘* Notes on the Nomenclature and Taxonomy of Certain Species of Melaleuca’’, by Edwin Cheel. ‘“ The Chemistry of Bivalent and Trivalent Rhodium. Part IV. Complexes with Diethyl Sulphide ”’, by F. P. Dwyer, M.Sc., and R. S. Nyholm, M.Sc. Lecturette—Mr. H. H. Thorne gave a lecturette, entitled ‘“* Approximations ”’. Exhibit.—Dr. A. Bolliger showed an exhibit showing the fluorescence of animal skins, with particular reference to Trichosurus vulpecula. August 2nd, 1944. The six hundred and sixteenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. One hundred and twelve members and visitors were present. The minutes of the previous meeting were read and confirmed. The certificate of one candidate for admission as an ordinary member of the Society was read for the first time. Popular Science Lecture.—It was announced that the third Popular Science Lecture for 1944 would be given by F. P. J. Dwyer, M.Sc., on Thursday, 17th August, at 8 p.m., the title being *“Man and Metals ”’. Inversidge Lectures.—It was announced that the Liversidge Lectures for 1944 would be given by Dr. F. P. Bowden on August 31st and September lst, and would be entitled “‘ The Physics of Rubbing Surfaces’, Parts I and II. LInbrary.—The following donations were received: 67 parts of periodicals and one whole volume. The following papers were read by title only : ‘* Bessel’s Formula in Relation to the Calculation of the Probable Error from a Small Number of Observations’’, by R. C. L. Bosworth, M.Sc., D.Se., Ph.D. “The Light Absorption and Magnetic Properties of Nickel Complexes ”’, by H. A. McKenzie, B.Se., D. P. Mellor, M.Sc., the late J. E. Mills, M.Sc., Ph.D., and L. Short, B.Sc. ABSTRACT OF PROCEEDINGS. Xxx Symposium.—A Symposium was held on ‘“‘ Trace Elements Essential to Life ’’. The President, Dr. G. D. Osborne, opened the Symposium, and referred to trace elements in minerals, which eventually found their way into soils. Dr. E. G. Hallsowth spoke on “ Trace Elements in the Soil’’. Professor E. Ashby’s subject was ‘‘ Trace Elements in Plants ’’, and Dr. J. L. Still spoke on “Trace Elements in Biochemistry ”’. The following took part in the discussion: Professor R. D. Watt, Drs. A. Albert, F. Lions, R. Breyer, Messrs. F. P. J. Dwyer and D. P. Mellor. September 6th, 1944. The six hundred and seventeenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, 157 Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Thirty-eight members and twelve visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of two candidates for admission as ordinary members of the Society were read for the first time. The certificate of one member for admission as an ordinary member of the Society was read. for the second time. The following person was duly elected an ordinary member of the Society : Hugh Albert McKenzie. Liversidge Lectures.—It was announced that, owing to the indisposition of Dr. Bowden, the Liversidge Lectures had been postponed for a few weeks. Popular Science Lecture.—It was announced that the fourth Popular Science Lecture would be given by Rev. Father D. J. K. O’Connell, S.J., M.Se., F.R.A.S., on Thursday, 21st September, at 8 p.m., and would be entitled “‘ Man and the Expanding Universe ”’. Library.—The following donations were received: 104 parts of periodicals and 10 whole volumes. The following papers were read : ‘Review of Analyses of Some Australian Fleece Wools’’, by M. Lipson, B.Sc., A.A.C.I., and Una A. F. Black, B.Sc. ‘“The Bauxites of New South Wales. Their Distribution, Composition and Probable ‘Origin ’’, by F. N. Hanlon, B.Sc., Dip.Ed. ‘“The Determination of Calcite and Aragonite in Invertebrate Shells’, by D. M. Bray. (Communicated by F. N. Hanlon, B.Sc.) ‘“Quaternary Arsonium Salts and their Metallic Co-ordination Compounds. Part I. Bismuth ”’’, by F. P. Dwyer, M.Sc., N. A. Gibson, B.Sc., and R. S. Nyholm, M.Sc. 99 Lecture.—A lecture entitled ‘‘ The Evolution of Surveying Instruments ’’ was given by Dr. R. L. Aston. The lecture was illustrated by lantern slides and many instruments shown by Dr. Aston to illustrate the evolution of surveying instruments. Exhibit —An exhibit was made by Mr. D. P. Mellor of “‘ A Powerful Permanent Magnet in Alnico ”’. October 4th, 1944. The six hundred and eighteenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, 157 Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Forty members and ten visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of three candidates for admission as ordinary members of the Society were read for the first time. i The certificates of two candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : William Hamilton Watkins and Harry Warner. Liversidge Lectures.—It was announced that the Liversidge Lectures would be given by Dr. F. P. Bowden on Tuesday and Wednesday, 17th and 18th October, 1944. Popular Science Lecture.—It was announced that the fifth Popular Science Lecture would be given by F. R. Morrison, A.A.C.I., F.C.S., on Thursday, 19th October, at 8 p.m., the title being ‘“My Lady’s Perfume ”’’. Inbrary.—The following donations were received : 180 parts of periodicals and five whole volumes. The following paper was read : ‘‘ The Sternal Integument of T'richosurus vulpecula’’, by A. Bolliger and Margaret H. Hardy. The following papers were read by title only : “* Studies in the Phenanthridine Series ’’, Parts I-VIII, by E. Ritchie, M.Sc. XXX ABSTRACT OF PROCEEDINGS. Film.—A sound colour film, “‘ Crystals go to War’, was shown. Lecturette.—A lecturette entitled ‘‘Some Aspects of Settlement in the Ord River District, North-West Australia ’’, was given by Mr. W. H. Maze. November Ist, 1944. The six hundred and nineteenth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Twenty-seven members were present. The minutes of the previous meeting were read and confirmed. The death was announced of the following member: Dr. Charles Anderson, a member since 1905. The certificates of three candidates for admission as ordinary members of the Society were read for the second time. The following persons were duly elected ordinary members of the Society : John Charles Erhart, James Alan Friend and John Bryan Willis. Exhibition of Photographs.—It was announced that an Exhibition of Photographs of Scientific Interest and a President’s “‘ At Home’”’ would be held on Saturday, 25th November, 1944, from 10 a.m. to 12.15 p.m., at the Small Art Gallery, David Jones. Library.—The following donations were received : 63 parts of periodicals. The following paper was read by title only : ‘‘ Thermal Conductivity from Measurements of Convection ’’, by R. C. L. Bosworth, D.Sc., F.A.C.I. The following papers were read : “Quaternary Arsonium Salts and their Metallic Coordination Compounds. Part II. Cadmium ’’, by F. P. Dwyer, M.Sc., N. A. Gibson, B.Sc., and R. 8. Nyholm, M.Sc. ‘“ Complexes of Ferric Chloride with Tertiary Arsines’’, by R. 8S. Nyholm, M.Sc. ‘“'The Response of Sternal Integument of T’richosurus vulpecula to Castration and to Sex Hormones ’’, by A. Bolliger, Ph.D. Exhibit.—An exhibit was shown by Mr. D. P. Mellor: “‘ The Streaming Double Refraction of Tobacco Mosaic Virus ”’. 7; Lecturette.—A lecturette on ‘*‘ The Chemical Attack of Tuberculosis ’ F. Lions. was given by Dr. December 6th, 1944. The six hundred and twentieth General Monthly Meeting of the Royal Society of New South Wales, held in the Hall of Science House, Gloucester Street, Sydney, at 7.45 p.m. The President, Dr. G. D. Osborne, was in the chair. Forty members and five visitors were present. The minutes of the previous meeting were read and confirmed. The certificates of three candidates for admission as ordinary members of the Society were read for the first time. Library.—The following donations were received: 90 parts of periodicals and four whole volumes. The following papers were read by title only : ‘“A Polarographie Study of the Isomeric Chromium Sulphate ’’, by J. B. Willis, B.Sc. ‘The Vibrations of Square Molecules. Part II. The Vibration Frequencies of Planar AB,C, (trans) Molecules’, by Allan Maccoll, M.Sc. ‘* Some Interference Effects with Mica’’, by O. U. Vonwiller, F.Inst.P. ‘“A Note on the Magnetic Behaviour of Verdohemochromogen’’, by D. P. Craig, M.Sc., and D. P. Mellor, M.Sc. The following papers were read : ‘“'The Chemistry of Bivalent and Trivalent Iridium. Part II. The Standard Oxidation Reduction Potential for the Chloriridite-Chloriridate System in Hydrochloric Acid Solution ’’, by F. P. Dwyer, M.Sc., H. A. McKenzie, B.Sc., and R. S. Nyholm, M.Se. “The Chemistry of Bivalent and Trivalent Rhodium. Part VIII. Rhodic and Rhodous Complexes with Dimethylglyoxime ’’, by F. P. Dwyer, M.Sc., and R. S. Nyholm, M.Sc. Lecturettes.—Two short addresses on ‘*‘Some Post-War Problems of New Guinea’’, were given by Brig.-Gen. Sir Walter McNicoll, K.B.E., C.B., D.S.0O., C.M.G., V.D., and Professor. A. P. Elkin, M.A., Ph.D. Walter Burfitt Prize.—It was announced that the Walter Burfitt Prize for 1944 had been awarded to Dr. Hereward Leighton Kesteven, D.Sc., M.D., of Sydney, for his outstanding published work in related fields of osteology, embryology and anatomy of vertebrates. A. B. WALKOM, President. ABSTRACT OF PROCEEDINGS OF THE SECTION OF GEOLOGY Chairman: Mr. T. Hodge Smith. Honorary Secretary: Mr. J. A. Dulhunty. Seven meetings were held during the year 1944, the average attendance being thirteen members and five visitors. Meetings. April 2lst.—Annual Meeting. Election of Office-bearers for 1944: Chairman, Mr. T. Hodge Smith, and Honorary Secretary, Mr. J. A. Dulhunty. Business : Address by Mr. H. F. Whitworth, ** The Commercial Utilization of Rock-Forming Minerals ’’. May 19th.—Exhibits and Notes: By Mr. H. O. Fletcher: Aulosteges from Little Gorge Creek, Queensland. By Mr. H. F. Whitworth: An electro magnet for the separation of iron- bearing minerals. By Miss F. M. Quodling: Illustrated note on X-ray methods of locating axes in quartz. By Dr. I. A. Brown: Fossil wood (Pittus Sussmilchi) from Babbonboon. By Mr. J. Lambeth: Refractory material consisting of Mullite. By Dr. W. R. Browne : Two types of tributary creeks of the Warragamba River. By Dr. C. A. Anderson: The silicification of wood. By Mr. J. A. Dulhunty : Recent peats from Kosciusko district. June 16th.—Address by Dr. G. D. Osborne, ‘‘ Clay Mineralogy ”’. July 2lst.—(i) Address by Miss F. M. Quodling, ** Fluorescence in Minerals’. (ii) Address by Mr. T. Hodge Smith, ** Use of the Ultra-Violet Lamp in Prospecting ”’. (ii) Address by Dr. G. A. Joplin, ** The Origin of Certain Ordovician Gneisses ”’. September 15th.—Discussion of Dr. A. B. Walkom’s Presidential Address entitled “* The Suc- cession of Carboniferous and Permian Floras in Australia ’’, delivered to the Linnean Society of N.S.W., 1944. Exhibit by Mrs. Sherard : Coloured sketch of ammonite from Tibooburra, and silicified wood from Milparinka. October 20th.—Address by Mr. C. A. Sussmilch, ** Tertiary Voleanic Rocks of Western Australia ’’. Exhibit by Mr. C. A. Sussmilch: Charts showing submarine topography along west coast of North America. November 17th.—The death of Dr. C. A. Anderson was announced, with regret. Address by C. Mulholland, ‘‘ Geological Problems Related to the Southern Intake of the Great Artesian Basin ”’, Ph Me sb wack INDEX. A Page. Abstract of Proceedings. . ahs Ree. o-4 Section of Geology .. a XXxXl Analyses of .Australian Fleece Wools, Review of we .. 84 Annual Report and Ballance Sheet Rip 0.41 Authors, Guide to ¥, i Big a Awards of— Clarke Memorial Medal for 1944 XXvli Walter Burfitt Prize for 1944.. bo Se B Balance Sheet... Wesel Bauxites of New South eles, Their Distribution, Composition and Prob- able Origin bys - yA ssh AOE Bequest, Form of Be: os 5 Vv Bessel’s Formula in Relation to the Calculation of the Probable Error from a Small Number of Observations .. 81] Bismuth, Quaternary Arsonium Salts and their Metal Co-ordination Com- pounds .. A ails oe te ELS Black, Una A. F. — Review of Analyses of Some Australian Fleece Wools aR ante ee Oe Bolliger, A.— The Sternal ee of T'richosurus vulpecula id 3 122 The Response of the Stordal inked ment of Trichosurus vulpecula to Castration and to Sex Hormones .. 234 Bosworth, R. C. L.— Bessel’s Formula in Relation to the Calculation of the Probable Error from a Small Number of Observa- tions .. Sac On Thermal erhductinies: Rote Mieaehrs: ments of Convection iis needy) Bowden, F. P.— Liversidge Lecture—The ava of Rubbing Surfaces... Sat tat] Bray, D. M.— The Determination of Calcite and Aragonite in Invertebrate Shells .. 113 Bryan, W. H.— Clarke Memorial Lecture—The Re- lationship of the Australian Con- tinent to the Pacific Ocean—Now and in the Past ee a fee D Cc Page. Cheel, E.— Notes on the Nomenclature and Taxonomy of Certain Bee of Melaleuca... 63 Chemistry of Bivalent an Reivalon Rhodium— Part VII. Complexes with Diethyl Sulphide : 67 Part VII. Rhodic and Rhodous Complexes with Dimethylglyoxime 266 Chemistry of Bivalent and Trivalent Iridium. Part It. The Standard Oxidation Reduction Potential of the Chloriridite-Chloriridate System in Hydrochloric Acid me te .. 260 Clarke Memorial Leeture— The Relationship of the Australian Continent to the Pacific Ocean— Now and in the Past aah ws “AZ Clarke Memorial Medal . a OY, Complexes of Ferric Chloride with Tertiary Arsines : -. 229 Craig, D. P.—See Mellor, S5y Pp. PAR WAT }c} Crespin, Irene— Some Lower Cretaceous Foraminifera from Bores in the Great Artesian Basin, Northern New South Wales.. 17 D Determination of Calcite and Aragonite in Invertebrate Shells oa So “LTS Dwyer, F. P. J., and Nyholm, R. S.— The Chemistry of Bivalent and Tri- valent Rhodium— Part VII ae aiid Be ee OF Party; ViEn 22. t 266 Quaternary Arsonium Salts and their Metal Co-ordination Compounds— Part I. Bismuth .. a ot TES Part Il. Cadmium 226 The Chemistry of Bivalent and Tri- valent Iridium. Part IT .. .. 260 E Exhibits... eV Extension of Maxwell’s Equations Beem i F Foulkes, P.— Extension of Maxwell’s Equations .. 14 XXXIV G Page. Geomorphology of the Central Eastern Area of New South Wales. Parts I and II anaes! Gibson, N. See Dwyer, F. P. dela. ATS Government Grant : ANGE Guide to Authors ba Be ee Vv H Hanlon, F. H.— The Bauxites of New South Wales. Their Distribution, Come annie and Probable Origin : 94 Hardy, Margaret H.—See polices. Ase 122 I Tridium, The Chemistry of Bivalent and Trivalent. Part II. The Standard Oxidation Reduction Potential of the Chloriridite-Chloriridate ihe in Hydrochloric Acid De js 260 Tsomeric Chromium Sulphates, A hsb graphic Study of sd Ae Bh. L Landform Analysis, Some Methods of.. 28 Landform Analysis of the Orange District 28 Lecturettes >: ‘ XXVil Light Absorption and Magnatil Properties of Nickel Complexes ere Lipson, M.— Review of Analyses of Some Australian Fleece Wools a a .. 84 List of Members .. cile bie Oe M Maccoll, A.— The Vibrations of Square Molecules. Part II. The Vibration Frequencies of Planar AB,C, (Trans) Molecules 246 Maxwell’s Equations, Extensions of .. 14 Maze, W. H.— The Geomorphology of the Central Eastern Area of New South Wales. Parts I and II he : pa he Mckenzie, A. H.— See Mellor, D. P. a aa Bra.) See Dwyer, F. P.. J a 260 Melaleuca, Notes on the N dpmorplaites and Taxonomy of Certain Species of.. 63 Mellor, D. P.— A Note on the Réle of the Nitrosyl Group in Metal Complexes. . 25 The Light Absorption and Magnetic Properties of Nickel Complexes .. 70 _A Note on the Magnetic Bebaviour of Verdohzmochromogen ee 258 INDEX Page. Members, List of ix Metal Complexes, Light Absorption and Magnetic Properties of ne nO Mica, Some Interference Effects with .. 252 Mills, J. E.—See Mellor, D. P. .. 70 N Nitrosyl Group—A Note on the Role of, in Metal Complexes .. 25 Note on the Role of the Nitrosy] Group in Metal Complexes 25 Note on the Magnetic Behaviour of | Verdohemochromogen 258 © Notes on the Nomenclature and Taxonomy of Certain Species of Melaleuca : 63 Nyholm, R. S. —See Dwyer, rep. 67, 118, 260, 266. Complexes of Ferric Chloride ‘with Tertiary Arsines : «'s aoe O Obituaries .. 2 Officers vii P Pacific Ocean, Relationship of the Australian Continent to—Now and in the Past wt 42 Phenanthridine Series, Studies in— Part I. The Cychsation of 2-Form- amido-Diphenyls ¥ 134 Part II. The Cyclisation of Some 4’-Bromo and _ 4’-Dimethylamino Acyl-2-Aminodiphenyls Big 141 Part II. The Influence of the Acyl Group in the Morgan-Wells Re- action and the Mechanism of the Reaction ne if me .. 147 Part IV. 1: 10-Dimethyl Phenanth- ridines 159 Part V. Phenanthridine- 6- Aldehyde and Related Compounds 164 Part VI. A Synthesis of 3- Methyl Phenanthridine ; 169 Part VII. A Synthesis of Benzo(c “ phenanthridine : 173 Part VIII. 3: 8- Diamino-Phenanth- ridine and Related Substances 177 Physics of Rubbing Surfaces, The 187 Polarographic Study of the Isomeric Chromium Sulphates, A as eae Presidential Address. A. B. Walkom, D.Se. a i fe i: Q Quaternary Arsonium Salts and the Metal Co-ordination Compounds— Part I. Bismuth : Je LTS Part II. Cadmium .. 226 INDEX. R Page. Report of Council, 1943-44 . XXIV Response of the Sternal Integument of Trichosurus vulpecula to Castration and to Sex Hormones, The . . 234 Revenue Account . XXIV Review of Analyses of Some Australian Fleece Wools .. 84 Rhodium, The Chemistry of Bivalent and Trivalent— Part VII. Complexes with Diethyl Sulphide a se oe pe OMe Part VIII. Rhodic and Rhodous Complexes with Dimethylglyoxime 266 Ritchie, E.— Phenanthridine Series, Studies in— Part I. The Cyclisation of 2- Formamido-Diphenyls 134 Part II. The Cyclisation of Some 4’-Bromo and 4’-Dimethylamino Acyl-2-Aminodiphenyls .. 14] Part III. The Influence of the Acyl Group in the Morgan-Wells Re- action and the Mechanism of the Reaction .. a =e Sipe he Part IV. 1: 10-Dimethyl Phenan- thridines. .. at Bi 4 159 Part V. Phenanthridine-6-Alde- hyde and Related Compounds... 164 Part VI. A Synthesis of 3- Methyl Phenanthridine .. 169 Part VII. A Synthesis of Benzo(c)- phenanthridine .. 173 Part VIII. 2: 8-Diamino- phen. thridine and Related Substances 177 Rubbing Surfaces, The Physics of He Goll S Short, L. N.—See Mellor, D. P. Pawel AU Some Interference Effects with Mica .. 252 Some Lower Cretaceous Foraminifera from Bores in the Great Artesian Basin, Northern New South Wales .. 17 Square Molecules, The Vibrations of. Part II. The Vibration Frequencies of Planar AB,C, (Trans) Molecules.. 246 Sternal Integument of Trichosurus vul- pecula .. oe agi as som 22 | | XXXV Page. Studies in the Phenanthridine Series— Part I. The Cyclisation of 2-Form- amido-Diphenyls _... . 134 Part II. The Cyclisation of Some 4’-Bromo and _ 4’-Dimethylamino Acyl-2-Aminodiphenyls .. 14] Part III. The Influence of the Acyl Group in the Morgan-Wells Re- action and the Mechanism of the Reaction bbe wre a .. 147 Part IV. 1:10-Dimethyl Phenan- thridines : 159 Part V. Phenanthridine-6- Aldehyde and Related Compounds .. 164 Part VI. A Synthesis of 3- Methyl Phenanthridine : 169 Part VII. A Synthesis of Benzo(c c)- phenanthridine ‘ 173 Part VIII. 3: 8-Diamino- Phenanth- ridine and Related Substances .. 177 Surfaces, The Physics of Rubbing .. 187 Symposium on Trace Elements XXix T Tertiary Arsines, Complexes of Ferric Chloride with 2. 229 Thermal Conductivity from Wessute: ments of Convection .. : 220 Vv Verdohemochromogen, A Note on the Magnetic Behaviour of 258 Vibrations of Square Molecules. “Part IL. The Vibration Frequencies of Planar AB,C, (Trans) Molecules 246 Vonwiller, O. U.— Some Interference Effects with Mica.. 252 Ww Walter Burfitt Prize a Ses 0h Willis, J. B.— A Polarographic Study of the Isomeric Chromium Sulphates Seedpeer ne ere nae AUSTRALASIAN MEDICAL PUBLISHING COMPANY LIMITED Lon 0 di ah Me ats, Se ae | > D ia A) 0) baa Rh 2s et ra & GS Genie ee, a +i) mats Sete SYDNEY ae AUSTRALASIAN MEDICAL PUBLISHING COMPANY ‘. ee. 1946. "3 (9g CAL ii 0 f ; Wi , ‘s ; \ ¢ v a Z ; iM i “if \ \ . . ‘ a P| 0 i ’ , ‘ eas 8 01308 4678