}' Gs & 2 wv @ @ooepete—=e a eee “a o-eecey wee, tm Phe ib an th te in aah Ra Re te Fe AN sn yy sg Mi Ae dn A eae he Nh ki in RN lee te A EN lth ptt RA et tele AA ey eA Mel thednthe Aha lr teal Al gamit he A wg Oe dl A ath A A thant Hm it db hm hers SAD Annes yD Mee Biya actin pth ipso i o> an Bi rtm nf YB ti Fa a Pn nm me ha Bh nr ip Am te Ca Mtn mm te ee TI he nyt thy Me Ne Ne Ne eR ay tn pe ee ee en ach Rane ie Reortinn the Mn he Cape nes Mine eens = me ple the Ue ie one ng ty wie r AR cw thar Faw TN ROG ne a ah Grab Ne A Mp Pe hsm in He ma lH Math Mae Ny a RA eet SR BO-Ae ng A a eR A em Aa neh Nh A ins A pt te I AT tr eh Om a: HO te AeA AN Ae Rn tat tht tn te aA apg ee Boch tra a = AM He WO DA Ded St Me Re tert os ht AI dy setts 0M wih Me sn Ot 7 Pa Yi Br Ne fm omen OTH Ne Mn ny en a as Rag Ma Yat Hy tte Mery tr Winn Pei ht At My oy Hs Ho Atte a i ig lh Pe MP has NH lng ME NNN Rl et Le me Ol eM ll ate ah ly pete Rk A mane i nM cll, Her otal Ae Al ti My: Men th the Pedy ahetite dnd. Gah eet I RAR relihe Ne tion Helly Thal Ende Mente ahi ratte ptm tlt The Messen. rhe Mim «Amr i A I gl Te ee ee eee ene ee eee ee eee See a rey ey NN ee = Hse ee « a oS ae art A Hh Ag Ma Ae i Net an eh Pe el es OA a ag ATs Me A I A Ne a ee ite pee i He ele dem RAR RN the il AB ce mate MONG HR Total AB MMi Sa ps Le tRmety Methetibe Pama Raitin Mt ital tha 4 A rinse tate Pham yas Dam cveriin ® teotenth “ OP eS oe ee ee ee ree hag om 1 oe pt Bie it~ rete tray om: we Re rh MO i eet oe beh Ma Mme ee eT Se ee ee ee ee ee eee ee a oe ee ee ee Te elite ee i ee ee ee ee ee er a ee re a ee ee ee ee ee ee ee ee reeen re ee oer en thin oe tam im Hata ty shy tg eee, ut PTY Ne Hm eng a ip rie He en im een Vine deerme Bete Man ee Sie ee FONE he ne de cm Cmte tether ptm hae! Mele th My Nee Am elias i Fae n. Bathe acy He Anta YP a ts Ng hw Mew ww fe ihe ey the Py be it A Me ran ee Nyt: ine tl a Pa'oe © ee ee ae pe Pm i Ni Fp Ny ie le el Ne et Sot A hh et Nap agin th 6 thang ge ata ale et ce a ee ee We ee ee eee ee Ma Lenin Bie bee Mee Biren thet ty etre dre teen Mista ag Re fe Manet Serine a Pe eh Ve Ra Ale TR a Maye Hae He TI Me Mie Reh MR Ome rhe Re Re ag Re em a cm ee ee it Ret Me klk eth mek tal ate 8 te Me Rempel RG. Cagle IN IN tin HF the eine ote 7AM Pelle em jae ee ee ANN he tg te i ts 1 Sante! Le a sath Yo lente Te Naitty Bc Re Pywtotay d Me Mim Me yt Rein A eet Mace RIM) to A a Ne Nt te em, Rte MN Rt eM tly Neat Mh Mee AMM AI RP atin oh NA Metta es act 928 trent tele ite ie ete ANAL lle ty Matt Ma ner anlte tens tint ten ne phatnisnanimetanl as bevgyt Os ee re tee. oy molten te Borie en ew a i ok. A Si le Ne Re ty EI. tee ebm tien hy Bk i eR ee te et Beh te hteie alte Geelgtas tpiee yee ate tr 191 rte a de Nn ei ee He ae ie a ae Be Cele eer Me Re Be ety eee nthe Mey ee ee ee ee ee ee ee a ne or ee a a ee ee ee eee ee ee a oe et ae) BA he AM te te ern Sn MI mh Re MR Ae meted eae Pe Rett pe tte ie llamete Came Late Mmin” ahotatn tel ee page Ar atin me Bina thy a Kew Soave te Pa ee eg Sage = A Ss gm Fin MN age eve Mth en hy wna Re el a RN ME NE My MA Re A A ts lh ty ay eM Lg Ne ene MN A i ak A tes eM lO A ye Mth i Mente Hig Res etm aac le tee Ra Ath Se role Agel ytrtinn bingy 6 diy Am aptiedlhMenbee rabetyn Aw tl ant tdi te te Re Maite he Away — rtm fm fin he Bom te, ethan rte et ee ee Oe ea Bk lint mrt le OE ee el th hp et Nate My Te eh ete tel ee oe Ree ee Pathe gh Te ee TE ee ee ee a ee ee ee en ee ee eee ee ee ree ee ee ee ee ee ere ee ee ee re eS ete te ae Datnatlnthet mene “— eet ee wt RT Re ale te MN 0s Get tie ® | FeiTahecine tata cia Tal a A nl oe Soa a a a coe ee a ee Oe eee, eee oe ee TO ee | ee ee ee ee ee Ln ae at. fee Bs ta te te em mart; NurRbptisttih Abertiestierteanetierht>Uibrighapdibedis antinctintrtee dette etieentantinoke tanta ieee itn tania edie dieiialemiam ee ee ee ee ee ee ee ee ee ee ee es ne ee ee ee ee eee ee ee , RNS Nain OD He OT Rete te Rett a A a A it AL A ae Mate Ate Ma ie 1 at Hh MY eet ill ee ee ae i at Ae MD aes hall end Me Pie yy Aen a Petal Orit ea tn Re heen Tarte ety nO Mae eg Ra bm Nig Ae owt, a the the eee Ow mt ARMA elt AOE RD LP Ste Tbe i at, Re ee ee, oe nn ee ee ne eee an er ee ee em er Ts oe te Reh Mm teting Ha therngthe ain HD Og MF i Reign Mahe Me “ss n a ee ee a ee a ee Se ie el ee Ce ee are ~ Par eptetly th -atlarsartey CoP se tee oy Note - SH i > Mi Fc le Mh Bir Ay Se ct tg Da Mth Magli nt a Aa gs Mie tthe te Dt a tte Metal thst Mite Meena Siecl . nig fae Rolain We ear ee re Min Mer Pan Ma = evn ie hy eg ern A tg y's i Mi i ee" aay Shy Mt Ny Ma cie: Mo Tin FR aA i a a ee ee a al snc ty te Na lat tl Unt Mp le i Per 6 tere er dao=haiae dtiregtertepeiedinanaitl etek hacia ie ae ee eee, en ee ee ee ee en ee eee ee ee oe ee es Te etait Pn Fanos pehtihh sc oedibehaie Ta Oe eee ee ee —. eo aN hai ~e ee a ee eee an A Serre eer eee etait nai ao ae ee eee ee b natin a . ‘ ee ae ee ee | i ee ee te ae ee eee - ‘ - wr i ' Sista. ts. Be tues 7. - : eo See wee ‘ atl oth lth Pip, ain elie bea rer bathed “- . . - 7 - 7 “ Mahi es Mile etl —~ re . ape . »e ‘ ee tect . ae are , Allie = yt we ee _ ee er Ce eee ee Ey er weary “ vs . = . . » . ~ . ~ Pe % ee eee oe 1 oy Mat theayte Mat hate letonae eine tthe ao ‘ + ore jr oraar ty 2 Atal Re Rg yl Momo , - . 7 . . dandy « ‘ . , | Pan thinkin grt " “ a eget Me — » » . . e fim Mena. ' . . - " “* rn te . ere ees lag co lh i tig Diy te A Fe The it 7 _ ne * on . : . ‘ a aaner -e or . ‘ . ee ee ae te Mrccadltesti - ster i * - ’ ~~ - . ” . : ~ > vn ” ‘ . - » . 4 7 Xe . ~ es . . — . . “ _ - * * ‘ ‘ - - tiatieg - “a * “ we ow Ome © 5 “ a2 iM » . - a . . . , * ° oad ae oni ~ “ ‘ ° - - a — -- a . onan, - 7 a A tne ~ - - ‘ " oe w “ ve =e ’ = * “ ~~. P 7 — - a a - . 7 . n * om - - - . 7 - . = ‘ w ‘. “ tw . - ° ~ ‘ — " * “ - . . * ~ - ¥ * — ” = . ers . - * » ~ 4 le - . - . - ~ P . a . . - ‘ m ~ ” < . , ° - ts uf - ‘ - - * - * ° * » ’ * - . - ‘ . . or See ifs hs MEMOIRS AND PROCEEDINGS OF THE MANCHESTER LITERARY & PHILOSOPHICAL SOCIETY. 206.1 2 MeA-6 MEMOIRS AND PROCEEDINGS / ? OF Pre WAN C EE SD EK mee RARY & PHILOSOPHICAL SOC He ry FOURTH SERIES DiC rer oH VOL UMA MANCHESTER 36"GEORGE, STREET 1894 NOTE. The authors of the several papers contained in this volume are themselves accountable for all the statements and reasonings which they have offered. In these particulars the Society must not be considered as in any way responsible. CONTENTS: MEMOIRS. PAGE Experiments on the Relation between Uniform Stress and Permanent Strain in Wrought-iron and Steel. By T. E. Sranton, B.Sc., Demonstrator in the Whitworth Engineering Laboratory, Owens College, Manchester. Communicated by Professor OSBORNE REYNOLDS, F.R.S. Some Aspects of Town Air as contrasted with that of the Country. By Geo. Barry, D.Sc., Ph.D. eA ae Bats ce On a New Sporiferous Spike from the Lancashire Coal Measures. By THoMAs Hick, B.A., B.Sc., Assistant Lecturer in Botany, Owens College, Manchester, and JAMES LoMax Preliminary Experiments on the Latent Heat of Steam at 100°C. By P. J Harroec, B.Sc. (Lond. and Vict.), Assistant Lecturer and late Berkeley Fellow in the Owens College, and J. A. HARKER, D.Sc. (Tiibingen), Berkeley Fellow in the Owens College. Communicated by Professor ARTHUR SCHUSTER, Ph.D., F.R.S. General, Morphological, and Histological Index to the Author’s Collective Memoirs on the Fossil Plants of the Coal Measures. Part III. By WILLIAM CRAWFORD WILLIAMSON, LL.D., F.R.S., &c., Foreign Member of the Royal Swedish Academy ; Corresponding Member of the Royal Society of Gottingen Notes on Wudfenia Carinthiaca, Jacquin. By;JAMES'CosMO MELVILL, MOA., F.L.S. On the K-partitions of the R-gon. By the Rev. THos. P. KirKMAn, Mex, E.R.S. On the Osmotic Pressure of Solutions of Finite Concentration. By Tuomas Ewan, B.Sc., Ph.D.... On the Treatment of Sewage with Basic Per-Salts of Iron under varying conditions. By Harry GRIMSHAW, F.C.S. ... An Analysis of the Electro-Motive Force and Current Curves of a Wilde Alternator, under various conditions. By JuLIus FRITH, Hegin- bottom Physical Scholar of the Owens College. Communicated by ARTHUR SCHUSTER, Ph.D., F.R.S. II 22 Sy 54 75 . 109 S 1St V1. CONTENTS. PAGE On the Primary Structure of the Stem of Calamites. By THomas Hick, B.A., B.Sc., Assistant Lecturer in Botany, Owens College. Com- municated by F. E. Wertss, B.Sc., Professor of Botany in the Owens College On the Instantaneous Pressures produced in the Explosion-Wave. By H. B. Drxon, F.R.S., Professor of Chemistry, and J. C. Cain, Eiger 1851 Exhibition Scholar in the Owens College On the Influence of the Configuration and Direction of Coast Lines upon the Rate and Range of the Secular Magnetic Declination. By HENRY WILDE, F.R.S. A Sketch of the History of the Canal and River Navigations of England and Wales, and of their present condition, with suggestions for their future development. By LIONEL B. WELLS, M.Inst.C.E. Com- municated by RUPERT SWINDELLs, M.I.C.E. PROCEEDINGS. BaAILey, CHARLES, F.L.S.—On Gulls at Old Trafford... On a Map of Palestine ... a ee ae On the rapid advances in Keene of Fossil Botany BAILEY, G. H., D.Sc., Ph.D.—On some aspects of Town Air as contrasted with that of the Country BoTToMLEY, JAMES, D.Sc., B.A., Ph.D.—On the Completion of the Manchester Ship Canal... Cuay, Dr. CHARLES.—Presentation of Bust of ... Dixon, Pror. H. B., M.A., F.R.S.—On M. Moissan’s Isolation of Fluorine FARADAY, F. J., F.L.S., F.S.S.—On a deposit of dirt during foggy weather on Spinning Cops of Yarn in a Mill ... On the Canal Map by Mr. Swindells and Mr. Wells GRIMSHAW, Harry, F.C.S.—On the Treatment of Sewage with Basic Persulphate of Iron under varying conditions GwyTHErR, R. F., M.A.—On an Aurora seen at Buttermere ... Hartoec, P. J., B-Sc.—On the Latent Heat of Steam .. 9) 9) HODGKINSON, ALEXANDER, M.B., B.Sc.—On the Fertilization of an Orchid by Insect Agency Hoyt, W.E., M.A.—On Shells recently acquired by the Manchester Museum On the Luminous Organs of Cuttle Fish . 158 174 . 108 - I0O0 , OG CONTENTS. Vil. : PAGE Hypr, Hrenry.—On the Growth of Maize near Manchester 19 Jones, Francis, F.R.S.Ed.—On Specimens of Marble exposed to town air 30 Lamps, Horace, M.A., F.R.S.—On the mode of Propagation of Waves through Water by a Moving Object 34 Lancpon, M. J., Ph.D.—On Specimens of the Salt of Technical Chlorophyll 20 Lets, C. H.—On the determination of the Thermal Conductivity of Winter ... 34 MELVILL, J. Cosmo, M.A., F.L.8.—On Bulimus labeo and various New Zealand Insects 83 NICHOLSON FRANCIS, F.Z.S.—On Gulls at Old Trafford .. 86 REYNOLDS, OsBoRNE, LL.D., M.A., F.R.S., M.I.C.E.—On the Latent Heat of Steam 31 On Kitchen Boiler Explosions 90 On an Aurora seen at Fallowfield . 107 SMITHELLS, Prof.—On Flame and Flame Spectra ... 98 SCHUSTER, ARTHUR, Ph.D., F.R.S., F.R.A.S.—On an Oak Tree struck by Lightning .. QI On Apparatus for testing Clinical Thermometers . 100 WEIss, Prof. F. E.—On Second Crops of Fruit on Raspberry Cee on Second Flowers on Apple Trees, and on a Monstrous Wallflower... 20 On Pieces of Wood found in Gravel Beds near Stockport 35 WELLS, LIONEL B., M.I.C.E.—On the Inland Waterways of England and Wales 32 WILDE, Henry, F.R.S.—On the Philosophical Uses to which the Cor- poration Supply of Electricity may be put 93 General Meetings otk Ree ue ox me 305290, 96, 105 Annual General Meeting ee LT Report of the Council with Obituary Notices of Archibald Sandeman, John Tyndall, Heinrich Hertz, Arthur Milnes Marshall, Charles Clay, and Thomas Armstrong 205 Treasurer’s Accounts 220 Meetings of the Natural History and Microscopical Section: Annual 173 Ordinary’ .,. ue re er Be a 10.20, 63,92, 99, 106 Annual Report of the Microscopical and Natural History Section— 2273, List of the Council and Members 226 Vill. CONTENTS. PLATES, -&c. TO FACE PAGE I.—To illustrate Mr. Melvill’s paper on Walfenia Carinthiaca van) 2 II.—To illustrate Mr. Kirkman’s paper on the K-partitions of the R-gon 109 III, IV, V, VI, VII, VIII.—To illustrate Mr. Frith’s paper on the Wilde Alternator ... xe sais sit - ue ls yi sia: Ge IX.—To illustrate Mr. Hick’s paper on the Stem of Calamites ... nen”, IO X.—To illustrate Mr. Wilde’s paper on the influence of Coast-lines on Magnetic Declination... ey vee a vas sii + G0 XI.—To illustrate Mr. Wells’s paper on Canal and River Navigation ... 204 CORRECTIONS. On p. 40, line 11, the equation should be as follows :— a w | cdt im a t L W fea tf Page 98. Mr. Rospert Monp’s election recorded in ordinary Meeting, took place at a duly summoned General Meeting on the same evening. MEMOIRS AND PROCEEDINGS OF meee MANCHESTER LITERARY AND PmCOSOPEMICAL SOCIETY, = Ordinary Meeting, October 3rd, 1893. Professor OSBORNE REYNOLDS, M.A., LL.D., F.R.S., Vice-President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Mate STANTON, B.Sc, read a paper on “ Experi- ments on the relation between Permanent Strain and Uni- form Stress in Wrought Irons.” Dr. G. H. BAILEY read a paper entitled “ Some Aspects of Town Air as contrasted with that of the Country,” in which he dwelt on the importance of quantitative investi- gations as to the impurities other than carbonic acid present in air as a measure of pollution, and urged that, however minutethe quantities may be,they are sufficientto bringabout serious disorganization in plant lifeand in human beings. In illustration Dr. Bailey presented tables showing considerable variations in the quantity of sulphur compounds present in different localities in Manchester and London on clear days and on slightly or densely foggy days. Some surprise was caused by a table showing that during the dense fogs of December last in Manchester and London there was a much larger proportion of sulphur compounds present in the London than in the Manchester air, notwithstanding the fact that the coal consumed in Manchester is generally understood to be much more sulphurous than that burnt in London. Professor WEISS confirmed the statement made by Dr. BAILEY. 2 Mr. T. E. STANTON ox Experiments on the Relation between Uniform Stress and Permanent Strain in Wrought-iron and Steel. By T. E. Stanton, B.Sc., Demonstrator in the Whit- worth Engineering Laboratory, Owens College, Manchester. Communicated by Professor Osborne Reynolds, F.R.S. (Received October 21st, 1893.) When an iron bar is subjected to uniform longitudinal stress exceeding its elastic limit, the connection between the stress and the permanent set of the material is usually shown by means of a Stress-Strain diagram, in which the stresses are represented by the ordinates, and the corresponding elongations by the abscissae of the curve. ' These can be traced by an autographic apparatus attached to the machine, several types of which are in use. The objection to this method is, that the position of the curve, and also its form, are greatly influenced by the rate at which the load on the bar is increased. Prof. Ewing* has also shown that the effect of a pause in the loading has a hardening effect on the bar, which increases with the length of the time during which the load is. kept constant. On increasing the load after this interval, it is found that permanent set does not again take place until a considerable increase in the load has been made. This has been called the “hardening effect of the time.” An illustration of this is shown in zg, z, which is the Stress-Strain diagram for a wrought-iron bar tested in the following manner :— The bar was turned accurately parallel and fixed in the Testing Machine and a stress of 16 tons per square inch of Ency. Britt., *‘ Strength of Materials,” 30-33. a oa ee ‘NOILOAS IVILIN] 40 HON] AXYWNOS UGA SNOJ, NI ssaWLS Uniform Stress and Permanent Strazn. 3 Fre. 8 STRESS-STRAIN DIAGRAM. form of diagram for constant ratio of loading from zero. Diagram for bar subject to initial stress of 16 tons with increase of load every 20 minutes. eases agram for bar subject to initial stress of 16 tons kept constant for 18 hours. 2 (Sea See SecA en CR Diagram for bar subject to initial stress of 16°7 tons kept constant for 46 hours. PERMANENT SET IN A LENGTH OF I0 INCHES. 4 MR. TE STANTON 707 initial section applied, and kept constant for 20 minutes. The extension of the bar practically ceased after 15 minutes. On increasing the load it was found that no further per- manent set took place until the stress had reached 17°0 tons per square inch, at which point the bar commenced to draw out rapidly. At 17°40 tons another pause of 20 minutes was made, and the test continued as before. It is seen that the hardening effect of each load extends through practically the same range of stress. In the same figure the dotted curve is the diagram for a bar of the same length and sectional area as the first, but loaded in a different manner. The same initial stress of 16 tons per square inch was applied, producing the same permanent strain as before. The load was then kept constant for 18 hours. On increasing the load no permanent set took place until the stress had reached 18°8 tons per square inch. After this the drawing out took place rapidly. In this case, the hardening effect extended through an increase of stress of 2°5 tons. A similar test was made on a bar of the same material, the initial stress of 16°7 tons being kept constant for 46 hours. In this case the hardening effect extended through an increase of stress of 3°0 tons. It is seen from the diagrams that when permanent set again takes place, after the load being kept constant, that the amount of extension depends on the time during which the preceding stress has been kept constant. Thus, in the tests of bars A and C, which were of the same material and initial dimensions, the extension of A for a stress of 20 tons was 1°125” ina length of 10”. The extension of C for the same stress, after having been subject to a stress of 16°7 tons for 46 hours was °762”, or 327 less than that of A. In order to establish a relation betweeen the stress and permanent strain by experiment it was necessary that the elongation caused by any given stress should not be affected Uniform Stress and Permanent Strain. 5 by the time duration of previous stress. For this purpose a bar of wrought-iron of very soft quality was cut up into nine lengths of 20 inches, and the 20” bars turned parallel. Each bar was placed in the testing machine and a given stress applied, the full stress being attained in about one minute. The load was kept constant for thirty minutes, when the permanent strain was observed and the bar taken out of the machine. The results of these experiments when combined show that the relation between the stress and permanent strain is given by the formula p=Ce, | where g=stress in tons per square inch on the reduced section of the bar. ua) d é€= permanent strain = where /= initial length, 7’=stretched length, C = constant. (iis relation is clearly seen im zg: 2, in which Professor Reynolds’* method of logarithmic plotting is used. Thus, if points whose ordinates and abscissae are respectively the logarithms of ~ and e are plotted; these points are found to lie on a straight line, the inclination of which will give the value of £, which in the above experi- ments was found to be ‘25. Assuming the above formula, then in the case of the nine bars, the maximum variation in the value of the con- stant C is from 39°25 to 39°49, or o'6%. It seemed probable that the permanent strains produced _in testing a single bar by successive loads would not be given by the above formula ; but in /zg. 2, where the results for bars No. 2 and 7 are plotted, it is seen that the same relation holds approximately for cases where the time duration of the stress does not exceed 30 minutes. In the case of bar No. 13, where the second stress Phil. Trans. Roy. Soc., 1879—p. 753. Mr. T. E. STANTON on i i Fic: ‘NIVYLS LNANVWYAd SO SWHLIYVSO1 ‘NIVULG GNV SSHULS JO SHAUN OINHLIUVIOT LOGARITHMS OF STRESSES. Uniform Stress and Permanent Strain. 7 applied was kept constant for 66 hours, the hardening has a marked effect on the next permanent sets, which were each 20% less than that due to the loads. The above tests were all made on the same brand of B.B.B. iron, the only variation being in the value of the constant C. which, for bar No. 2, was 40°30, and for bar No. gavas. 40°35. Experiments were also made on some bars of mild steel, the results for which are shown in fzg. 2. In these, with a time duration of stress not exceeding 30 minutes, the value of £ was found to be }. The results are also plotted for the case of a mild steel bar,in which the initial stress was kept constant for 18 hours. Similar experiments made on nine specimens, cut from a long bar of Crown BB iron, did not give such consistent results. This was probably due to the varying hardness of the material in different portions of the bar. In the case of bars R1 and Rg taken from these, and tested with successive loads at intervals of 20 and I5 minutes respectively, the value of £ was found to be ‘275 for Ri, and ‘265 for Rg. The combined results for the other seven bars, each tested with one load only, gave ‘265 as the value of &. Several other experiments were made on iron of different qualities, including some on Low Moor iron. The results showed that in the case of iron of a soft and fibrous nature, the same law of permanent set held, but when the iron was harder and showed a crystalline fracture the relation could not be expressed in the above form. If p,=stress on initial section of bar, then, assuming the density to remain constant and taking the formula p=Ce" (1) we have (2) 8 Mr. T. E. STANTON oz the condition for a maximum value of #, being when e=4. Substituting this value for e in equation (2) we have, taking C = 39'4, Po= 22°45 tons per square inch. This corresponds very closely with the mean maximum stress of wrought-iron of this brand, experiments on rough bars giving values of g, ranging from 21°8 to 20°70 tons. TABLE “1. Tests of nine specimens cut from a bar of wrought-iron. Brand R.f., Crown BBB. No. of Initial Final Permanent Stress on Value Specm. Area. Area. Strain. Reduced Section. of C. sq. in. sq. in. Tons. Dis "7643 "7420 "03071 16°480 39°37 Line "7590 "7238 105,227 18°373 39°48 Z. 3 "7528 =. “6999 «08133, ; 20°077 ~~ )3Gme Z. 4 ‘FOASS) as 7e 03978 17°620 =. 39°46 Z. 5 "7905 "7158" 06585 | 10073) aaa Z,. 6 "7810 "7638 "02330 15°338 20°27 (LS "7600 FAL 702689 15893 39°25 ZO 7510 ries fe) 04668 18°306 39°38 Z. 9 ‘7560 - "7306 03490 «= 17068 = gaa Test of one specimen cut from a bar of wrought-iron. Brand R.H. Crown B.B.B. Duration of stress 15 minutes. No. of Initial Final Permanent Stress on Value Specm. Area. Area. Strain. Reduced Section. of C. sq. in. sq. in. Tons. Ze ‘7410 — —_— — — — —— a 2ae "02450 15888 40°17 a — ap ROBIN, 03547 17°45 = 40°20 — — *7066 "05005 19°06 40°40 — — "6940 (O7 150 20°893 40°40 a — “‘O725 "10875 23°050 40°15 ee ee > al a Uniform Stress and Permanent Stratn. 9 Test of one specimen cut from a bar of wrought-iron. Brand R.H, Crown B.B.B. Duration of stress 30 minutes. No. of Initial Final Permanent Stress on Value Specm. Area. Area. Strain. Reduced Section. Oi (Cs sq. in. sq. in. Tons. fa 7775 =e a i ris nas oe "7646 *OL7 3a 14°583 40°20 Br — "7594 "02462 <« 16°000 40°39 ae ... %16"4 Ao — sp < SC ind Deo we eve 2 eet fe . Se She a ee a nee trom the floor. ye! se.) fe.) 16'6 Thus, as a matter of fact, so enormous is the amount of sulphur in Manchester gas (it is usually at least double that allowed by the Metropolitan Act as a maximum) that the air of our rooms is liable to be as highly charged with sulphurous acid as the street air is in a moderately bad fog. It is probably only the relative dryness of the air which prevents it from becoming absolutely unbearable. Thus far, then, in the face of the numbers given, I contend that my remarks amount to a demonstration that, as a means of discriminating between polluted and unpolluted air, and as a means of forming some estimate of the extent of pollution, the determination of the sulphurous compounds and of organic matter are much to be preferred to that usually adopted, viz., an estimation of the carbonic acid. I may add that an equally convenient and, perhaps, even more valuable means lies in the direction of the estimation of the micro-organisms. Miquel, in Paris, has done much in this direction, but, though the difference between town and country air is very great, systematic experiments carried out at a sufficient number of stations in a town area have yet to be made before any general conclusions can be drawn. But any plea for such a method of examination of air should have greater weight if it can be shown that the matters which it is proposed to determine are themselves injurious to life and health. Unfortunately the amount of 16 Dr. G. H. BAILEY ox reliable information that we have is small; it is, however, increasing rapidly. Aitken (Proc. Roy. Soc. Edin. XX. 76) gives results which go far to show that the ferszstence of town fogs (and it is the persistence which lends them their virulence) is due to the presence of sulphur compounds and mineral matters; and Frankland years ago suggested that the condensible hydrocarbons in air possess a similar property. Oliver, in a report presented to the Royal Horticultural Society this year (“Effects of Urban Fog upon Cultivated Plants”), has shown that the presence of as little as 20 parts per million of sulphurous acid will, zf the laght 2s also cut off to the extent to which in towns zt zs cut off, bring about injuries to plants comparable to those which actually occur during fog. Also that mere traces of some hydrocarbons and of pyridine, impurities both found in polluted air and the deposits therefrom, are most injurious. Then with regard to human beings, hardly a winter passes without the death-rate from respiratory diseases at times running up to three or four fold the normal. That this is due, in some measure at least, to the abnormal pollu- tion of the air, such as has been indicated, is highly probable. It is a character practically confined to large towns; it is specially characteristic of densely populated districts where pollution of the air is most marked. Doubtless the preva- lence of such ailments is largely affected by climatic conditions, and their seriousness aggravated by a lowering of tone of bodily health already established. But this only raises the further question as to how far this very lowering of tone is the result of the constant inhaling of these minimal quantities of sulphur compounds, organic matter, and the like. Even though we are not yet in possession of sufficient information to enable us to speak decisively, there can be little room for doubt that the determination of these minor ve a ee eee ee = SS eee Some Aspects of Town Air. age constituents of town air and their further examination is worthy of more attention than has hitherto been given to it. And considering the importance of pure air, it would be well if such analyses of air were frequently and systemati- cally carried out by sanitary boards as a matter of routine under some such scheme as the following :— Azur of dwellings in special cases ; The estimation of carbonic acid gas, of organic matter and micro-organisms. Aur of streets ; The estimation of sulphur compounds, of suspended organic matter, micro-organisms, and noxious gases. 18 PROCEEDINGS. Ordinary Meeting, October 17th, 1893. ro fessor ARTHUR SCHUSTER, Ph.D., F.R.S...2eee President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. , Reference was made to the deaths, since the close of the previous session, of two of the Society’s members—Dr. CHARLES CLAY, elected in 1841, and Mr. ARCHIBALD SANDEMAN, M.A., formerly Professor of Mathematics in Owens College, elected in 1851. A bust of Dr. CLAY, executed in 1834, and presented to the Society by his executors in accordance with his wish, was exhibited. Mr. FARADAY alluded to a peculiar deposit of dirt during foggy weather on the spinning cops of yarn in a mill. According to his informant, the deposit increased when the gas was lighted, and so convinced were the firm in question of this that they were proposing to fit up the mill with the electric light on this account alone. A dis- cussion ensued in which Mr. N. BRADLEY, Mr. C. BAILEY, Mr. JOHN BoypD, Mr. ANGELL, Professor DIXON, and the PRESIDENT took part. It was variously suggested that the phenomenon might be due to the heat causing atmospheric currents, and thus bringing more of the polluted air into contact with the cops; to the fog preventing the escape of the products of combustion into the outer air; to the vaporisation of minute globules of water floating in the atmosphere, solid matter held by them being thus permitted to descend ; to the greater density of the fog, when it became necessary to light the gas; and finally to the lighting of the gas merely making the collection of dirt on - PROCEEDINGS. 19 the cops visible, and thus giving rise to the illusion that it was due to the lighting of the gas. Etoressor Hi. 6. DIXON, F.R.S:., gave an account of M. MOIssAN’s isolation of fluorine, as experimentally illustrated at the meeting of the British Association at Nottingham. The Rev. THOMAS P. KIRKMAN, M.A., F.R.S., read a paper on the “ K-partitions of the R-gon.” [Microscopical and Natural History Sectzon.| Ordinary Meeting, October goth, 1893. Mr. PETER CAMERON, F.E.S., in the Chair. Mr. HYDE referred to the past summer as having been highly favourable to the growth of maize in this country, and specially mentioned that grown in Alexandra Park, specimens of which were seven and eight feet in height with cobs in an almost ripe condition. Other plants noticed esrowing in the park were:—the castor oil plant, the tobacco plant, the mallow, the eucalyptus. 20 PROCEEDINGS. Ordinary Meeting, October 31st, 1893. Professor ARTHUR SCHUSTER, Ph.D., F.R.S., F.R, Ales President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Mr. FARADAY exhibited some specimens of yarn as soiled by fog in the process of spinning. Dr. LANGDON exhibited specimens of the salt of tech- nical chlorophyll, which were believed to be the copper salt. Professor WEISS exhibited raspberry canes bearing a second crop of fruit, and shoots of the apple bearing a second crop of flowers. He also exhibited specimens of a monstrous wallflower, which was considered by De CANDOLLE to be a separate variety, and named Chezranthus Chetri, var. gynantherus. The abnormal condition is due to the transformation of the stamens into carpels which are often fused round the true ovary. Mr. P. J. HARTOG read a paper, by himself and Dr. HARKER, describing a form of apparatus by means of which they have measured the latent heat of steam. The authors wish the results, which have so far been very con- cordant and give a value distinctly lower than that found by Regnault, to be considered preliminary only, until they have been able to extend their observations. A paper on anew sporiferous spike, apparently of the Calamarian type, from the Lancashire coal measures, by Messrs. T. HICK and JAMES LOMAS, was also read. The specimen was found near Oldham. PROCEEDINGS. 2H [ Microscopical and Natural History Sectzon.| Ordinary Meeting, November 3rd, 1893. mee LLis CUNLIFFE, President of the Section, in the Chair. Mr. T. A. COWARD and Mr. C. OLDHAM were elected Associates of the Section. Mr. ROGERS exhibited specimens of two rare mosses :— (A) Physcomitrium sphericum, found growing on the mud margin of the reservoir at Whalley Bridge, the second locality in England ; (B) Phycomztrella patens, found on the mud margin of the reservoir at Chapel-en-le-Frith. Both were found by Professor Barker, late of Owens College, October, 1893. | Mr. ROGERS also exhibited fine specimens of fruit of Pyrus japonica, grown at Bowdon. Mr. CAMERON made a short communication on galls, and exhibited a gall of the Lzorhzza aptera from the roots of the birch from Edenbridge ; the birch being a new food plant, the others, apart from the oak, being beech, pine and vine. Also an elongated spindle-shaped woody gall on Ulex nana from Warwickshire. This gall is undescribed, differing altogether from the other gall on Ulex (Asfondt- lyza uleces) which is a bud gall. : ire) COSMO MELVILL, M.A. F.LS., exhibited specimens of Waulfenia carinthiaca (Jacq.) from South Tyrol, perhaps the most local of European plants, and compared it with its nearest allies, both European and exotic, in the sub-order Digztalee of Serophulariacee, which were also exhibited. e2 Mr. T. HIcK AND MR. J. LOMAX ox On a New Sporiferous Spike from the Lancashire Coal Measures. By Thomas Hick, B.A., B.Sc., Assistant Lecturer in Botany, Owens College, Manchester, and James Lomax. (Recezved October 31st, 1893.) The fossil which forms the subject of this communica- tion was found by one of us at Moor Side, near Oldham, and has been derived from what is locally known as the Upper Foot Coal of the Lower Coal Measures, a layer which is practically identical with the Halifax Hard Bed in Yorkshire. Unfortunately only a‘single specimen was met with. From this two sections were prepared, both of which are longitudinal. One is nearly, but not exactly radial, except perhaps for a short distance at the base of the spike, while the other is so tangential as only to meet the axis at the upper extremity. The following description applies exclusively to the former section, unless the contrary is stated, but it contains nothing which is inconsistent with the other. GENERAL CHARACTERS. The section of the spike measures 4 centimetres in length by 8 or g millimetres in breadth. As if Gegm@e complete at either end, it was probably somewhat longer originally. The shape of the spike appears to have been cylindrical, but whether or not there was a narrowing at the ends it is impossible to say, nor is there anything to indicate its position on the parent plant. Like many other carboniferous fruits it is composed of an axis and numerous lateral appendages (/zg. 7). The latter are of two kinds, sterile bracts (Fzg. z, a, 6, f), and sporangiophores (g, 2), placed at the nodes of the axis in A New Sporiferous Sptke. 23 alternating whorls. In all 14 nodes, about 2°5 millimetres apart, can be counted, which originally bore whorls of sterile bracts, a few of which are present zz szfu in the section (/zg. z, a, 0). The oumber of bracts in each whorl cannot be made out with certainty, but it was probably small, and perhaps did not exceed 6 or 8. This estimate is based upon the appearance of the uppermost nodes, which are cut so tangentially that the anterior bracts are seen in transverse section (77g. 2, 2). STRUCTURE OF THE AXIS. The axis of the spike has a nearly uniform diameter of 1:17 millimetres throughout, and is obviously made up of.a central cylinder (or stele) (Fzg. 7, s), surrounded by a bette (772. 7,2), but the structure of the parts is very imperfectly shown. The stele, 05 millimetres in diameter, is, in part at least, composed of elongated elements which here and there bear faint traces of vascular markings. But the whole cylinder is so black, and the state of preservation such, as to preclude any decisive statement as to the nature of these elements and as to whether the centre of the stele was or was not parenchymatous. At the nodes which bear the bracts the cylinder widens out a little, but the section shows no such expansions opposite the sporangiophores, perhaps because at these points the section is not radial to these structures. Neither in the case of the bracts nor in that of the sporangiophores, has any vascular connection with the stele been met with, but this is no proof that such did not originally exist, as it may be due to the divergence of the section from the radial direction. THE CORTEX. The cortex, whose thickness is 0°33 millimetres, is made up chiefly of large cells, elongated longitudinally. In the hypodermal region, the cells have thick walls and appear 24 Mr. T. HIcK AND MR. J. LOMAx on to be prosenchymatous, so that their function was probably mechanical (zg. 2,a@). The inner cortex is made up of larger and thin-walled cells, and at certain points there are unmistakable evidences of the presence of canals or much elongated cells, in some of which are black carbonaceus contents similar to those met with in the young stems of Arthropitys and the sterile bracts of Calamostachys Binneyana (Fig. 2,¢e). At the sterile nodes the cortex is continued into the bracts in a way which will be described in dealing with the latter structures. Unfortunately these details cannot be illustrated by a figure, as they have been made out piece-meal from an examination of the cortex of both specimens, and are nowhere met with in combination. THE SEERILE SpRAGrs The members of the successive whorls of sterile bracts appear to have been superposed and not alternate. This is inferred from the fact that on one side of the spike the section has passed through five bracts in succession, and on the other side two successive bracts are superposed in two places. The bracts stand out from the axis at nearly a right angle—having an extremely slight inclination upwards— for about three millimetres, and then turn upwards so abruptly that the limb is approximately parallel to the axis. The only evidence: as to the presence or absence of cohesion at the base of the bracts is presented by the uppermost part of the section, which is tangential. It is not conclusive on the point, but it certainly proves that if cohesion does occur, it is restricted to the immediate neighbourhood of the axis, a conclusion, likewise, suggested by the second section. As to the structure and form of the base of the bracts, little that is definite can be made out. The upper part would seem to consist of narrow,elongated, sclerenchymatous : A New Sporiferous Sprke. 25 elements (/zg. 2, 0) continuous with the hypoderma of the internode above. The lower is composed of a large celled tissue with thin walls (/7zg. 2, Z), closely resembling that of the inner cortex and has no continuation of the hypoderma of the internode Jelow. In one or two instances the resemblance to the inner cortex is emphasised by the presence of black masses in the ceil cavities. From the appearance of the specimen, this softer tissue seems to have readily separated from the overlying harder part and to have been easily destroyed. The two layers together have a vertical thickness of o'5 millimetres at the point of insertion on the axis. The limb of the bract is a long slender body and probably reached to the second whorl of bracts above (Fig. z, 0). The structure of the limb is very imperfectly shown, but it seems to be made up of elongated, narrow, thick-walled elements, something like those met with in the upper part of the base. Whether or not it possessed a large-celled tissue like that of the lower part of the base, it is impossible to say, but it seems doubtful. THE SPORANGIOPHORES. The sporangiophores stood nearly, if not exactly, mid- way between the successive whorls of bracts, and projected at right angles from the axis. They are not found in the section however, and our knowledge of their position is based upon the short processes shown at g, which represent tangential sections of the basal portion. The section shows no trace of a peltate dilation of the distal end of the sporangiophore, an absence which is difficult to understand had such existed, unless it were remarkably small. For this reason we are inclined to think that the sporangiophores were simple columella-like structures. The histology cannot be made out with certainty, but there appears to have been a central strand of delicate tissue enclosed in an outer zone 26 Mr. T. HICK AND MR. J. LOMAX ox of harder and more lignified elements (Fzg. 7, 2). This strand may have been a small vascular bundle, but there is no proof of this. As the section in passing more or less radially through the sterile bracts has missed most of the sporangiophores, we infer that the latter have alternated with the former. We have no information as to the number of sporangiophores in each whorl, but it was probably small. THE SPORANGIA. The sporangia were arranged round the sporangiophores, but the number connected with each and the mode of attachment is not clearly shown. Where the section of the spike is radial two sporangia, one above the other, are seen between two successive whorls of bracts (Fzg. z, 5p.). Where it is tangential, the appearances point to the presence of four sporangia for each sporangiophore (/zg. z, #). The walls of the sporangia are composed of a single layer of cells, whose inner and radial walls are thickened (/7zg. 3, 0), In the surface view the cells are elongated and the longi- tudinal walls exhibit the projecting transverse processes so often met with in carboniferous sporangia (/zg. z). No layer of thin walled cells is observable lining the interior, though this may be due to disappearance. The size of the sporangia cannot be definitely stated as they seem to have been cut in all directions, save those which would enable their principle axes to be measured. The one shown in Fug. z at the base of the spike has a length of about 1°6 millimetres and a breadth of about o'8 millimetres. The same is shown enlarged in /7zg. 3, where the point of attach- ment is probably seen at (a). THE SPORES. The spores are all of one size, averaging 0'066 millimetres in diameter, and are rounded in shape (/zg¢ 3, c). No A New Sporiferous Spike. 27 external markings can be made out on the walls, nor are they grouped in tetrads. The wall appears to be thickened, but this is probably due to the mode of preservation, and, in most cases, the contents remain as a central black mass, in which there are occasionally indications of a nucleus. They were probably mature and not in process of develop- ment at the time of mineralisation. SYSTEMATIC POSITION. Imperfect as the preceding description is, it seems sufficient to enable us to refer the new spike to the Ca/ama- veeae rather than to the Lycopodineae, and hence its systematic position must be sought among the spikes of the former group. Unfortunately, the internal structure of these spikes is known in a few cases only, and the attempts to classify them by external characters alone, has not been very successful. Hence, any attempt to allocate the new spike to one of the groups into which the Calamarian fruits are divided can only be tentative and provisional, but this is no reason why the task should be avoided. A comparison of the new spike with the well-known Calamostachys Binneyana, Schr., which is now known to be the fruit of some form of Calamites, reveals the fact that there is a close general agreement between the two, accompanied by differences of some importance. They agree in the alternation of sterile and fertile whorls of appendages ; in the position of the sporangiophores, which stand midway between the successive whorls of bracts ; and the number of sporangia associated with each. But the new fruit differs from Calamostachys Binneyana in the form, length, and, perhaps, number, of sterile bracts in each whorl ; in the absence or great reduction of the sheath-like disk formed by the cohesion of the bracts; and probably in the absence of a peltate expansion at the distal ends of the sporangiophores. In addition, the spores are apparently 28 Mr. T. HICK AND MR. J. LOMAX on slightly larger, but this is, perhaps, of little importance, as exact measurements are, in most cases, difficult to make. The precise value of these agreements and differences, from the systematists’ point of view, is not easy to estimate, but we cannot be far wrong in regarding the former as of much greater value than the latter. To us, they seem sufficient to justify the inclusion of the new spike in the genus Calamostachys, which, as at present understood, is to some extent, a collective one. Acting upon this opinion, we propose that henceforth the fossil should be known as Calamostachys Oldhamia. RAN Wie A New Sporiferous Speke. 29 EXPLANATION OF THE FIGURES. PuaTE. Fic. 1. Longitudinal Section of the Spike, three times the natural size. s. Central Cylinder or Stele of the Axis. é. Cortex of the axis. a, b, f. Sterile bracts. a. do. cut transversely. g. Sporangiophores. h. do. cut transversely. sp. Sporangia with spores. Fic. 2. Lxlarged view of the base of the bract f Fic. 1 at its junction with the axts. Central Cylinder or Stele of the Axis. Lnner portion of Cortex of Axts. Outer do. do. Upper portion of base of bract. Lower do. do. VS & RX Fic. 3. Sporangia with Spores. a. Probable point of attachment to Sporangiophore. b. Wall of Sporangium. c. Spores. Fic. 4. Surface view of the Wall of a sporangium. 30 PROCEEDINGS. General Meeting, November 14th, 1893. Professor OSBORNE REYNOLDS, M.A., LL.D., F.RS., Vice-President, in the Chair. Mr. R. L. TAayvLor, Science Master, Central Schools, Manchester, and Mr. HORACE LAMB, M.A., F.R.S., Professor of Mathematics, Owens College, Manchester, were elected as ordinary members. Ordinary Meeting, November 14th, 1893. Professor OSBORNE REYNOLDS, M.A., LL.D., F.R.S., Vice-President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Dr. HODGKINSON exhibited a species of orchid (Catase- tum), illustrating the extraordinary means by which fertilization by insect agency is effected in this group of plants. Experimental demonstrations with the antennz were successfully given. Mr. FRANCIS JONES, F.R.S.Ed., exhibited specimens of polished white marble exposed on the top of the Grammar School and at Alexandra Park for one year to atmospheric influences. The first-named specimen showed a loss of I per cent of its original weight, and the other a loss of 02 per cent.; whereas a third specimen kept in a room showed no loss whatever. The utility of the experiment as a test of the purity of air and its bearing on the wasting of exposed marble statues were pointed out. PROCEEDINGS. 31 Professor OSBORNE REYNOLDS read the following note “On Mr. HARTOG and Dr. HARKER’S Experiments on the Latent Heat of Steam at 212° Fahr.” :— “ Since the publication of Regnault’s experiments in 1848 there has been a general agreement as to the value of this important constant. And no one, in the meantime, has pointed out any source of error in Regnault’s work. What- ever may be the true value of this latent heat an agreement as to the exact figure is of great importance. Otherwise by the use of different figures results into which this con- stant enters are thrown into discord. At the last meeting of this Society Mr. Hartog brought forward results of some very interesting experiments which show the latent heat to be something like 27% less than that obtained by Regnault. From the description, the experiments had evidently been ‘made with the greatest care and the results obtained from different experiments are fairly consistent. Any source of error must therefore be some general loss of heat which would exercise the same effect on all the experiments. After hearing the paper it occurred to me that such a loss of heat must necesarily take place in the experiments from a cause which appeared to have been overlooked by the author of the paper. Tothis I now direct the attention of Mr. Hartog, in the hope he may be able, by removing it, to bring his results into accordance with Regnault’s. The matter which seems to me to have escaped the attention of Mr. Hartog is the cooling effect on the interior tube of his apparatus, by which the steam passed into the calorimeter, of external radiation through the walls of his enclosing glass vessel. That this would cause a loss is certain; what this loss would be depends on the temperature of the room and on the constants of absorption of the surface of the interior tube and the glass envelope.” Messrs. HARTOG and HARKER replied to Professor REYNOLDS, pointing out that the loss by radiation could 32 PROCEEDINGS. only be very small, and nearly, if not quite, negligible, in their experiments. This loss, moreover, must in each case be proportional to the duration of the particular experiment. If, therefore, it were appreciable, other things remaining the same, the longer the experiment lasted, the lower would be the value found for L. But this was found not to be the case. A special experiment had been performed since the last meeting to test the validity of Professor REYNOLDS’ objection, of which notice had been privately given. The results obtained confirmed the view expressed by the speakers. Mr. LIONEL B. WELLS exhibited a map of the inland waterways of England and Wales prepared by himself and Mr. RUPERT SWINDELLS, and read a paper on “The Early History of the Inland Waterways of England and Wales; and their present condition, with suggestions for their Future Development.” He pointed out that the shipping entered at British ports has increased within fifty years from 10,000,000 to about 130,000,000 tons, and that not one of the old waterways has secured its due proportion of this enormous increase of traffic. Of the existing waterways about 1,222 miles are controlled by railway companies, and 2,468 miles are “independent ;” yet the former group carry only one-fifth the total tonnage which passes along the entire system. The more important river navigations are the Weaver, the Aire and Calder, and the Severn. The improved sections of these have an aggregate of only 112 miles, yet they carry about one-eighth of the total inland waterway traffic of England and Wales, and the waterways controlled by the railway companies, though nine times as long, carry only 65 per cent more traffic. There are about 126 different lengths of waterways in the hands of about 100 proprietors, and if many of these would amalgamate, through water routes from east to west and from north to south could be established. With reference to the map it PROCEEDINGS. 33 was pointed out that as no complete map of the navigable waterways of the country could be exhibited at the Inter- national Congress on Inland Navigation which met in Manchester in 1890, Mr. WELLS and Mr. SWINDELLS had privately prepared the one exhibited, which showed that there are 740 miles of navigable waterways in the country of the existence of which the compilers of the Government returns of canals seemed to be unaware. Omitting large estuaries, the total length of canals and navigable rivers in England and Wales is 3,790 miles. With scarcely an exception the sills of the locks are below the navigable draught of the existing waterways, proving that the founders of the system looked forward to the ultimate deepening of the canals. In the discussion which ensued Mr. FARADAY com- mented on the fact that it should have been left to the private enterprise of Mr. WELLS and Mr. SWINDELLS to pro- duce the first complete canal map of England and Wales, and contrasted the neglect and imperfections of our own Government departments in this matter with Continental Government work, as illustrated by the specimens of canal maps issued by the French, Belgian, and other Govern- ments exhibited by Mr. WELLS. 34 PROCEEDINGS. Ordinary Meeting, November 29th, 1893. Professor ARTHUR SCHUSTER, Ph.D., F-.R.S., Pigs - President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Professor LAMB, M.A., F.R.S., made a communication giving a mathematical explanation of the mode of propaga- tion of waves through water by a moving object. Mr. C. H. LEES read a note on the determination of the thermal conductivity of water. Mr. HARRY GRIMSHAW, F.C.S., read a paper “On the treatment of sewage with basic persulphate of iron under varying conditions, more especially with regard to results obtained in Salford.” From the experiments he concluded :— That the Salford sewage in June, 1893, was, in consequence of the long-continued drought, about 25 per cent more impure than the average. That while it is possible to vary the proportion of precipitants to such wide differences as exist between night and day sewage, it is not practicable on the large scale to meet the hourly fluctuations in the composition of the sewage of an industrial town ; during the working day, therefore, the maximum amount must be adhered to. That a too rapid flow through the tanks involves a greater expenditure of precipitants than is other- wise necessary, and that a flow of an average of about 10,000,000 gallons, or a maximum of 14,000,000 gallons, through tanks of a vertical area of Soft. by 7ft. renders proper subsidence impracticable. That in cases of this kind the only remedy is either the use of excessive amounts of precipitants or the subsequent passage of the tank effluent through a straining filter or through land; which a PROCEEDINGS. 35 of the two is most economical being determined by local circumstances. A discussion ensued, in which Mr. CorBETT, Mr. BARTON WORTHINGTON, Mr. WILLIAM THOMSON, and the PRESIDENT took part. Ordinary Meeting, December 12th, 1893. Pais bOLTrOMLEY, D.Sc., B.A., F.C.5., Vice-President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. The CHAIRMAN referred to the death, since the last meeting, of Professor TYNDALL, an honorary member of the Society, elected 1868. The SECRETARIES reported on the completion of the Joule Memorial statue by Mr. ALFRED GILBERT, R.A., originally promoted by the Council of the Society, and announced that the surplus funds,tothe amount of 4257.1Ts., had been handed over to the Society to be used for the commemoration of Joule’s name, and that the books and documents of the Memorial Committee had been given to the Society for safe keeping. Professor WEISS exhibited some pieces of wood taken from some gravel beds near Stockport. The wood was completely waterlogged and quite soft. Exposed to the atmosphere it became hard and black, lost its woody appearance, and showed a conchoidal fracture. It was 36 PROCEEDINGS. transformed into a substance like lignite,and burned like coal A discussion on the character of the change in the wood and on the presence and absence of wood in peat bogs was — participated in by Mr. C. BAILEY, Dr. BOTTOMLEY, and Dr. HODGKINSON. The third part of Dr. W. C. WILLIAMSON’S “ General, ”) Morphological, and Histological Index” to his collective memoirs on the fossil plants of the coal measures was read. A paper by Mr. J. C. MELVILL,.M.A., PS) one? fenia Carinthiaca, Jacquin,” was also vead ; anda paper “On the Osmotic Pressure of Solutions of Finite Concentration,” by Dr. THOMAS EWAN. Experiments on the Latent Heat of Steam. Ly Preliminary Experiments on the Latent Heat of Steam at 100°C. By P. J. Hartog, B.Sc. (Lond. and Vict.) Assistant Lecturer and late Berkeley Fellow in the Owens College, and J. A. Harker, D.Sc. (Tubingen) Berkeley Fellow in the Owens College. Com- municated by Professor Arthur Schuster, Ph.D., F.R.S. (Rececved December rath, 1893.) The latent heat of steam at 100°C. has been determined since the time of Black by a number of observers, Rumford (im), Ure (2), Watt (3), Despretz (4), Brix (5), Regnault (6) Favre and Silbermann (7), Andrews (8), Berthelot (9), and Schall (10). We quote in the following table the results obtained by Regnault and subsequent observers :— Observer. ae Extreme Values of L. Hts Ber: os Regnault... .. 44 533°3—538'4 (11)| 536°67 Favre and either. Tan... ° .. 3 532°59—541'77 GIEGWS ... ... 8 530°8 —543°4 Bermelot... ... 3 535°2 —537°2 530°2 Beall |... ...| no details, no details. 522 (1) Complete Works (Boston, 1875), Vol. If., p. 417. (2) Phzl. Trans., 1818, Part II., p. 385. (3) Robison’s Mechanical Philosophy, ed. Brewster, Vol. II., p. 5 (1822). (4) Ann. chim. et phys. [1] 24, ‘P- 323 (1823), and Zrazté elémentatre de physique, p. 94, et seq. (1825). (5) ogg. Ann., Vol. LV., p. 341 (1842). (6) Mémoires de P Academie des Sciences, Vol. XXI. (1847). (7) Comptes Rendus, Vol. XXIII., p. 411 (1846). a Se m “TI m “ i mr = ss = dS|/+ [oO o {ok In Fig. 1,if AB and CD represent the corrections for the temperatures t,.+ €, 5} and are 44 Mr. P. J. HARTOG AND Dr. J. A. HARKER om Regnault and Pfaundler assume that the correction for any intermediate temperature 7,, corresponding to the abscissa OE, is represented by the ordinate EF drawn to meet the straight line BD in F. For this assumption to be valid two things are. necessary :— (1) That the change of temperature of the calorimeter should not exceed 3° or 4°; this conditions satisfied both in M. Berthelot’s experiments and in those performed by us; and (2) That the temperature of the bodies radiating heat to the calorimeter should remain approximately constant. It is evident that we cannot possibly assume the correction to be merely a function of the temperature of the calori- meter, if we suddenly bring a hot body into its vicinity during the course of our observations. In M. Berthelot’s apparatus, we do the reverse ; we suddenly take away the hot burner which was radiating heat to the calorimeter during the actual observation. It is difficult to see how the observation of the march of the thermometer after this removal can serve as a datum for the calculation of the correction to be applied for the period of actual con- densation, during which the thermal conditions were so different. We proceed to describe the modified form of apparatus [Fig. 2] which we have adopted in order to be able to use either of the methods of correction just quoted. The boiler consists of a flask A, through which the tube BC passes centrally. The upper end of BC is ground coni- cally to fit into a hollow cap D, which is itself attached by a glass rod to the movable bell E. This bell fits loosely intoa » rim, which is filled with mercury so as to form a lute. The bell and cup may thus be raised or lowered at will, so as to open or close the valve at C, through which the steam passes Experiments on the Latent Heat of Steam. downwards through CB into the con- ity densing worm W. At Fa side tube is con- nected with a condenser, if desirable, by means of an indiarubber tube fitted with P\L aclip. Thetube at F is kept open during the preliminary period; it is shut just after C is opened, and opened again just before C is closed, so that at no period does the internal pressure exceed that of the atmosphere. The end B of the tube BC is ground into the upper end of the condensing worm, of which the construction differs slightly from that of Berthelot. The steam in our apparatus enters the condensing worm by the straight portion, and not by the spiral. We found this alteration neces- sary as, after the closing of the valve, the air entering the worm tends otherwise to drive the condensed water back into BC. The lower half of A was surrounded by a piece of copper _ gauze bound on with asbestos string, and the lower portion of the tube BC was surrounded in Expts. I. II., III. by asbestos, and in IV. and V. by a leaden steam-coil wrapped closely round it. The boiler was heated by a small ring burner, of which the flame was kept at a perfectly constant height from the moment of lighting till it was extinguished. The gas was passed first through a Moitessier glycerine regulator (which maintains the pressure constant to within a half millimetre of water), and then through a tap fitted with a long handle moving in front of a graduated circle. These precautions are necessary for accurate measure- ments, as variations in the height of the flame naturally A 46 Mr. P.J. HARTOG AND Dr. J. A. HARKER oz cause the radiation to the calorimeter to vary. We were able to regulate the amount of this radiation at will ; but, of course, too small a flame made the determination too slow, too large a flame the initial and final corrections too high. The calorimeter and its jacket, and the thermometer were protected from excessive radiation by means of screens of asbestos board. The calorimeter itself consisted of a copper vessel weighing 282 grammes. The general arrangement of the calorimeter and jacket was practically identical with that employed by Berthelot. The stirrer consisted of a ring of copper pierced with several holes, and moved up and down.on glass guides fixed into a light wire frame, which served to protect the glass worm from any accidental blow. It was moved by means of an electro-motor, and a wheel and crank mechanism. The thermometer used was one by Baudin, of Paris (No. 12,771, metastatic), divided into j,ths of a degree centigrade. It was compared with an instrument calibrated by Dr. Schuster, and compared by him with a thermometer standardized at the International Bureau of Weights and Measures at Sevres. It was read by means of a telescope, and the 2oth part of a division, ze, the zoooth part of a degree was estimated. The actual modus operandi was as follows :—The water was first boiled in the flask with the valve at C closed, and the steam escaping (not condensed). The thermometer was read from this time forward at the end of each half minute. When the march of the thermometer had become regular for some minutes, the valve was opened and the exit F closed. The steam then condensed in the worm. When it was thought that sufficient steam had been con- densed, the total rise being from 3 to 4 degrees, the exit was opened, and the valve closed, and thermometric obser- vations were only discontinued when the march of the Experiments on the Latent Heat of Steam. 47 thermometer again became regular. The flame was then extinguished, and the worm was detached, corked, carefully wiped, and allowed to remain in the balance-case for some time before weighing. The error on the determination of the weight of the water condensed in the worm could not exceed one part in 10,000. We made a special blank experiment without condensation, to ascertain if the Regnault-Pfaundler corrections were applicable, and found this to be the case. We give below the readings of the thermometer during one experiment (No. 1). [The first readings given are those taken after the water had been boiling for 20 minutes. ] Time. Reading. Notes. ih. =m. 4 6 a 14°524 A ae 14°532 4 7 a 14°540 , = a Kees Preliminary period. GR «ss 14°564 4 9 14°573 AROn <2. 14°582 aad 7S Ke, _— — na) Valve opened, 4h. 9 50% Mets ay. Ane ce om ACED 2s 157 o2 2s — 4 114 i, 20 Sie — A 12 ae -— ... Reading missed. 2 Be ge ee — Ay 13 se co Oy a — AS ws 767.22 “ih a A Td are 16° 46 bbe — ) A 16" Fo ee -— A a5 eS 16° 93 ace = a 7° 1c ae — 4 16 Seis r7 40 Jae = a) ere 07 OA ad — A 17 we Ly O6 ne — 48 Mr. P.J. HARTOG AND Dr. J. A. HARKER ox Time. Reading. Notes. i.) mn. OLE e a mt ft — :-» . Valve closed. 4 18 a 18°I40 ee — AAS S02 es — .«. Reading missed. 4 19 sna 18°170 sisi — AP BOs i! tee 18°176 _— as 4 20 =r 19 19a Hes — AOR. | ss. 18184 bie — A 2% iG 18187 sas — AT a 18192 oes — A722 ee 18°197 Ae eee |) Sa 18°201 4 23 Be 18°205 A 28R by os tee Final period. 4 24 ee 18°214 A BAe. 6s. 1O'217 4725 sists 18'220 A 25s. te Te228 Rise of thermometer per minute during initial period (4h. 6’ to 4h. 934’) ="0165°. Mean temperature of initial period = 14°°553. Rise of thermometer per minute during final period (4h. 22’ to 4h. 254’) = "0074". Initial temperature of calorimeter (4h. 9’ 50” =14°588° Final ie a (4h..22'= 18097 DPoas eect = 3°'609 From the above data we can calculate graphically, in the way described, the correction to be applied for each succes- sive half-minute of the middle period; and we thus find that the total correction is o'118°. The corrected temperature - difference is therefore (3°609 — O'118) = 3491”. The other data of the experiment were as follows:— i St a Experiments on the Latent Heat of Steam. 49 Sp. ht. Water eqt. Weight of copper calorimeter... 284°5 0°0933 26°54 ‘ RASS SUUTEL cae) pea) OF "086 7°50 Ss glasswormandguides 46°8 “O17 7°87 SMEIMIOMICtEr ... 0 65. wo. wee os: Ba E10 Water in calorimeter (corrected for GuGyancy in air) .:. ... Ue fe 1717°0O otal * =| E7007On Barometer-reading, 762°4mm. B.P. of water=100"07. Weight of water condensed = 10°122 grammes. We have then i pions Line — (100°07 — 18°20) IO°122 = 525°13 cals. We give in the following table the experimental details of five experiments (including the one just quoted). 1. Number of Ex- PELMMCHE ..5..:... I. i TEE: IV. WV; 2. Water- equivalent of calorimeter ameveontents .:. 1760°0 1723°2 1740'2 17434 1681°0 a) Vemperature of SECAUA, Wi5.2s 0's wre. £LOO.07) | LO0'07 (100°26 160"30 *. 100"00 4. Initial temp. of Calorimeter ...... 14-509) 13°027 12°968 12'526. 14°948 5. Final temp. of ealerimeter (7,)... 18°197 17°577 16°330 16°206 16°062 6. Observed _tem- perature - differ- CO nloiniaic ve ses 25000) 7 43550. 3302 iO 7S, 1 rr 7. Total correction. —*118 —'I134 —*I27 — °239 —"123 8. Duration of experi- ment in minutes. 12 15 8 13% 5% g. Weight of water condensed, in SRAMMIMCS 6... s0.00 forlez T2540, O270 | orS54.). 2°742 fon Value of L ...... Ban Gs 52407 52567 524-33 522"°65 Mean value of L deduced from experiments I.—V.=524'60 ” F) ” ” I.—IV. = 524°85 In the above calculations we have taken our heat unit as 50 Mr. P. J. HARTOG AND DR. J. A. HARKER on the amount of heat required to raise one gramme of water through 1° of the hard-glass mercury thermometer in the neighbourhood of 15° C. It will be observed that our results differ from those of Regnault by more than 2 per cent. We have failed so far to discover any explanation of this difference, either by ascertaining an appreciable error in our own work or in that of the great French physicist. Professor Osborne Reynolds has suggested that radiation from the inner surface of the central glass tube must cause a certain amount of conden- sation on its surface, and that the water so condensed would run down into the worm, and thereby cause an error of calculation possibly sufficient to account for the difference between our results and those of Regnault. The heat radiated into a vacuum by a square centimetre of glass at 100° has been determined by Gratz (Wzed. Ann., Vol. X1., p- 913), and the absorption by glass of heat radiated from a Leslie cube heated to 100° by Melloni (quoted by Wiillner, Lehrb. ad. Experimental Phystk, Vol. IIL. p. 197), so that we can form an estimate of the loss due to this source, and calculation shows that the loss falls in all. probability well within the error of experiment. This conclusion is borne out by the fact that the amount of water condensed by radiation must be proportional to the duration of the experi- ment; while a glance at the table shews that, other things being approximately the same (see ¢.¢. Expts. III. and IV.), the results of experiments, calculated without taking account of this correction, are independent of duration. We are, however, obliged to Professor Reynolds for his criticism, and shall meet it by making use of metal vessels in the experiments which are in progress. It may here be pointed out that the determinations of Regnault for the latent heat of steam at o° have not been confirmed by subsequent observers [see Winkelmann (Wied. Ann., Vol. IX., p. 208, 1880) and Dieterici (zbzd, Vol. Experiments on the Latent Heat of Steam. 51 XXXVII., p. 494, 1889)]; but on this point no agreement as yet exists. It occurred to us that our results might be controlled by the experiments made by Joly with his ingenious steam calorimeter. (Proc. Roy. Soc., Vol. XLI., p. 352 (1886), and Vol. XLVII, p. 219 (1889). By the use of this instrument we can calculate the specific heat of a body if we suppose the latent heat of steam to be known, and its author used it for this purpose ; and, inversely, we can also use it to calculate the latent heat of steam if we assume the specific heat of the bodies experimented on to be known. Unfortunately the specific heat even of bodies like silver, which are easily obtained in the pure state, is not known with the requisite accuracy. Thus Regnault gives for the sp. heat of silver 0'05701, Kopp gives 0'056, and Bunsen gives 070559. (The numbers are quoted from Joly, Proc. Roy. Soc., XLI., p. 358.) In order to use the Joly calorimeter for our purpose it is evidently necessary that we should make use of a par- ticular body of which the specific heat has been determined immediately beforehand with the water calorimeter. This we propose to do shortly. The other means at our disposal for controlling our numbers is less direct. A well-known equation in thermodynamics gives us a relation between L, the latent heat of steam at the absolute tempera- fire: a, (s‘—s), the difference between the specific volumes of saturated steam and water at T’, dp the differential coefficient of the vapour tension dT of water with regard to the temperature, at T°, and J, the mechanical ao of heat, namely :— d; L=3(¢— S) \o 52 Mr. P. J. HARTOG AND Dr. J. A. HARKER on The quantities J and s’ are by no means easy to measure, nor have we space to discuss the values obtained for them by different observers. There is, however, reason to believe that the value for J given by Griffiths (PA77. Trans., 1893, p. 493), viz., 47194 x 10’ C.G.S. units, is within zoo0 Of the truth. The experiments of Perot (Axx. Chim. Phys. {6| XIII, p. 159) were performed with extreme care, and we accept his value for the specific volume of steam at 99'60°C., namely, 1657cc. We have used two formulae to calculate dp/dT from Regnault’s experiments, which give results differing by less than 1 per thousand. That of Moritz gives ds/dT = 3°58574x 10* C.G.S. units; that of Roche gives as/dT = 3758318 C.GS. units, taking ¢=980'67.* If we calculate out L with the values given, we find that with Moritz’s number L=527°54 calories, with Roche’s number, L=527'16 calories, for the temperature 99°60 C. (which we choose because Perot’s determination was made at that temperature). At 100° the value of L would be somewhat less (about half a unit, if Regnault’s interpolation formula is approximately correct.) These numbers agree well with the one obtained by us, viz., 524°8, but at the same time, we should hesitate to regard this confirmation as conclusive. * Moritz’s formula, quoted by Wiillner, Lehrbuch der Experimental-Physik, Vol. IV., p. 683, is as follows :— log.49 P=at bat—cBt where p = pressure in 7 of mercury, ¢ = temperature centigrade. log.49 = ‘006864937 log. 19 P=1°996725536 log, 9 @=2°131990711 log. 49 ¢=0°611740767 &=4°7393707- Roche’s formula, quoted by Hirn, 7héorie Mécanique de la Chaleur, T. I., pp-'323, 325, 18 as follows.:— ip 0°090936948/ d aT [1+0°'0049528167(20+2)?] Experiments on the Latent Heat of Steam. 53 We are at present engaged in repeating our determina- tions with an apparatus made chiefly of metal. In the new model we have replaced the gas burner by a coil of wire placed within the boiler, and heated electrically. By this means, we hope to reduce the radiation to the calorimeter, and consequently the correction, very considerably, and thereby to increase the accuracy of our results. We have, in conclusion, to tender our thanks to Semessot Schuster and Mr. H. E. Hadley, B.Sc. for assistance given in the course of our work, and especially to Mr. S. H. Davies, B.Sc., who took part in a tedious series of preliminary rough experiments, of which no account is given here. 54 Dr. W. C. WILLIAMSON ox General, Morphological, and Histological Index to the Author’s Collective Memoirs on the Fossil Plants of the Coal Measures. Part III. By William Crawford Williamson, LL.D., F.R.S., &c., Foreign Member of the Royal Swedish Academy ; Corres- ponding Member of the Royal Society of Gottingen. (Recetved December 12th, 1893.) LIST OF WORKS ON THE ORGANISATION OF THE FOSSIL PLANTS OF THE COAL MEASURES, AND GENERAL INDEX TO THEIR CONTENTS. ROYAL SOCIETY SERIES, I. no XIX: Symbols. Parts. A. I. Calamites and suggested genus Calamopitus (not subsequently insisted upon). Figs. 16 and 17 do not belong to Calamites but to the subsequently adopted genus Astromyelon. Phil. Trans.5 187%. EB; II. Lepidodendron selaginoides, Diploxylon (Corda), Ulodendron, Favularia, Sigillaria, Stigmaria, Lepidodendroid Cone (?) ultimately Lepidodendron parvulum. (Memoir XVI.) Anabathra. hl. Trais., 1872. C. III. lLepidodendron brevifolium, (Burntisland form) and _ its Lepidostrobus. Restoration of Lepidodendron. iii. Trans., 1872. D. IV. Lyginodendron Oldhamium; Heterangium Grievii. /Phz/. Trans., 1873. E. V. Asterophyllites with Sphenophylloid axis. Sphenophyllum. Volkmannia (now Sphenophyllum) Dawsoni, Strobilus of Asterophyllites (subsequently Paracalamostachys Williamsoniana; Weiss) Asterophyllites fruit (subsequently Palzeostachya pedunculata. (See Weiss. Steinkohlen- Calamarien). Calamostachys Binneyana, Calamites verti- cillatus. Root of Asterophyllites (afterwards Amyelon). Phil, Trans., 1874. F. VI. Rachiopteris aspera (afterwards petiole of Lyginodendron Oldhamium) Rachiopteris Oldhamium, Rachiopteris duplex, Rachiopteris Lacattii, Rachiopteris bibractensis, Anacho- ropteris Decaisnii. Phz?. Trans., 1874. G. VII. Myelopteris (Medullosa of Cotta), Psaronius Renaultii, Kalo- xylon Hookeri (now known to be root of Lyginodendron). Phil. Trans., 1876. H. The Fossil Plants of the Coal Measures. 55 VIII. LX: XI. XII. AIT. XIV. XV. XVI. PV EL, XVIII. Rachiopteris corrugata, Fern Sporangia, Gymnospermous Dadoxylon, Gymnospermous Seeds, Lagenostoma ovoides, Lagenostoma physoides, Conostoma oblonga, Conostoma ovalis, Conostoma intermedia, Malacotesta oblonga, Trigonocarpon oliveforme, Hexapterospermun Noggerathi, Cardiocarpon anomalum, Cardiocarpon compressum, Car- diocarpon acutum, Cardiocarpon Butterworthii, Polyptero- spermum. /fzl. Trans., 1877. Astromyelon, subsequently A. Williamsonis (and now known to be the root of Calamites), Calamites, Asterophyllites, Lepidodendron selaginoides, Lepidostrobus, Macrospores, Rachiopteris cylindrica, Cordaites (?) epiderm, Lygino- dendron (?) anomalum, Lepidodendroid cortex, Oidospora anomala, Volkmannia (?) parvula (now Lepidodendron parvulum), Lepidodendron Spenceri. /z/. Zrans., 1878. Arran Lepidodendron, subsequently L. Wunschianum, Lepido- dendron Spenceri, Heterosporous Lepidostrobus, Calamo- stachys Binneyana, Rachiopteris insignis, Tylosis, Rachiopteris robusta, Sporocarpon elegans, Sporocarpon, pachyderma, Sporocarpon' asteroides, Sporocarpon ornatum, Traquaria, Zygosporites (subsequently shewn to be spores), Dadoxylon, Lagenostoma ovoides, Cardio- carpon anomalum, Calcisphera (Radiolariz of Judd) Rachiopteris di-upsilon. P&z/. Trazs., 1880. Lepidodendron selaginoides, Lepidodendron Harcourtii. (The plant so named here is now designated L. fuliginosum. See Proceedings Royal Society, Vol. XLII., p. 6). Stigmarian rootlets. Medullary rays of Lepidodendron selaginoides, Calamostachys Binneyana and C. Casheana, Fungi. hdl. Trans., 1881. Astromyelon Williamsonis (now root of Calamites), Psaronius Renaultii, Zygosporites (in a Sporangium), Calamites, Lepidodendron, Halonia, Sporocarpon ornatum, Salisburia Adiatifolia, Phzl. Trans., 1881. Heterangium Tilizoides, Kaloxylon Hookeri (now root of Lyginodendron). Phzl. Trans., 1887. True fructification of Calamites. Phz/, Trans., 1883. Rachiopteris Grayii, Rachiopteris Lacattii; Calamostachys Bin- neyana, Rachiopteris hirsuta, Rhizonium verticillatum, Rhizonium reticulatum, Rhizonium lacunosum. Lepidodendron fuliginosum, Lepidodendron mundum, Lepido- dendron Spenceri, Lepidodendron parvulum, Rachiopteris inequalis. Phzl. Trans., 1889. Lyginodendron Oldhamium, Bowmanites (Volkmannia) Daw- soni, now Sphenophyllum Dawsoni, Calamites. 1890. Bowmanites (now Sphenophyllum) Dawsoni. Rachiopteris ramosa, possibly R. hirsuta var. ramosa. ** On the structure of the woody Zone of an undescribed form of Calamite.” Memozrs of the Manchester Literary and Philosophical Society, 3rd Series, Vol. 1V., Session 1868-9. **On a new form of Calamitean Strobilus.” J/emozrs of the Manchester Literary and Philosophical Society, 3rd Series, Vol. IV., Session 1869-70. 56 Dr. W. C. WILLIAMSON ox W. ‘¢On some Anamolous Oolotic and Paleozoic forms of vegeta- tion.” Royal Institution of Great Britain, Weekly Evening Meeting, Feb. 16, 1883. 2. & ‘*On the relations of Calamites to Calamodendron,” with description of an intermediate form. Memoirs of the Manchester Literary and Philosophical Society, 3rd Series, Vol. X., 1886-7. v. A Monograph on ‘‘the Morphology and Histology of Stigmaria ficoides.” Palwontographical Society, Volume for 1886. Z. *‘On the Structure and Affinities of some Exogenous stems from the Coal measures.” Monthly Microscopical Journal, Aug. I, 1869. AA. ‘**On the Organisation of the Volkmannia (now Sphenophyllum Dawson). BB. XIX. Lepidodendron Harcourtii Bromenzart, Halonia, Ulodendron, Lepidophloios, Lepidostrobi, Lepidodendron Spenceri. Phil. Trans., 1893. INTRODUCTION. The Carboniferous Plants that I propose to deal with in Part III. of this Index are the Ferns. Seeing that fronds of this group are so extremely abundant in most of the shales and sandstones of the Coal measures, it might have been expected that their stems, branches, and petioles would be equally so in our calcareous nodules ; but, unfortunately, this is: not entirely the case; yet they are not wholly, wanting, but such as we do obtain are usually fragments of stems, petioles, and the secondary and ternary branches of fronds. It is extremely rare to find any of these accom- panied by their leaves or leaflets. Hence, it is often very difficuit to determine whether or not the objects we are studying belong to the Filicine group. There are certain well-known localities where fragments of stems are more abundant than elsewhere, which stems unmistakeably belong to the arborescent sections of the semi-tropical ferns. In these examples their internal organisation is too character- istic to leave much room for error respecting their primeval affinities, but there are many forms which leave abundant room for those differences of opinion respecting their true relationship that are so common in the writings of even our The Fossil Plants of the Coal Measures. By) most experienced observers. Under these circumstances I strongly object to the undue multiplication of generic and other names that are so common amongst us. Where we find considerable groups, the individuals composing which have certain very definite features existing throughout the entire group, as is the case, for instance, with that of the Zygopterids, it seems to me useful to give them a common name. But the cases are numerous in which this cannot be done. In such types each example would require a name of its own. This necessity would arise, partly from the imperfection of the fragments with which we have to deal, but also, in part, from the absence, in many such instances, of sufficiently individualised features to make their differences easy to define. In such cases as Lyginodendron and Heterangium these fundamental differ- ences are important and easily defined ; but in numerous other instances this is not the case. To these I have assigned, in my later Memozrs, the comprehensive term Rachiopterzs, which binds together a number of examples of which the general organisation is certainly fern-like, but which signifies nothing more. As more definite groups can be formed out of this very varied and comprehensive cluster, such groups can be differently dealt with. With the working of this method we have a good illustration in one of the earliest plants that came into my hands, which I had described in Part VI. of my Wemozrs under the name of Edroxylon, but to which I afterwards assigned the name of Rachiopteris aspera. In my Part IV., I had described a very distinct plant under its present name of Lyginodendron Oldhamium ; but in Part VI. I expressed my strong con- viction that the former plant would ultimately prove to be the petiole of the latter one. I never lost sight of this possibility, but I had to wait sixteen years before I obtained clear proof that my original surmise was absolutely correct. This determination was an important one, because 58 Dr. W. C. WILLIAMSON oz Lyginodendron was a plant with a magnificent zone of secondary exogenous wood, developed from a true cambium. But Dr. Scott and I have recently united with it a second | form viz.,the Kaloxylon Hookeri, which proves to be its root. “e No true fern previously discovered had exhibited such a cambium; but the Lygznodendron Oldhamium described now took its place, along with the Calamztes and the Lycopods, in both of which important groups of C7yptogams the possession of an active cambium was the normal condition. This family of the Lygznodendra is the first that I propose to deal with in this part of the Index. PIBICES: TYPE OF LYGINODENDRON OLDHAMIUM. Primary Branch prior to emergence through Cortex. Earliest State. Transverse. No Medullary Cavity occupied by primary Wood. R.—p. 92, Fig. 10, C.N. 1885A. a. Tracheids of Primary Xylem. b. Secondary Fascicular Xylem. Secondary State. R.—p. 92, Fig. 11, C.N. 1885H. Primary Xylem broken up into about five separate bundles, a’, a’. See also C.N. 1138. Medullary Rays. R.—p. 92, Fig. 10, C.N. 1885A. p. 92, Fig. 11,/C.N. 18S5H. BRANCH EXTENDED BEYOND THE CORTEX OF THE PARENT STEM, AND INVESTED BY ITS OWN CORTEX. R.—p. 93, Fig. 12, C.N. 1141. Medullary cavity further enlarged, and filled with medullary cells. SECONDARY XYLEM. R.—p. 92, Fig. 12, C.N. 1141. New Trachez to the periphery of each of the secondary Lamine. CoRTEX—YOUNG. R. 1141. Irregular Cambium at the innermost border of the Cortex. CORTEX MORE MATURED. Cambium not previously figured or described. Innermost Cortex of O.N. 1141, 1193, 1194, and 1195. The Fossil Plants of the Coal Measures. 59 CorRTEX—MIDDLE. R.—p. 90, Fig. 3c, C.N. 1138. Gum-Canals. R.—p: 90, Fig. 3b, C.N. 1138. Fig. 2A 2l. CoRTEX—EXTERNAL, R.—p. 90, Fig. 3b, C.N. 1138., fibrous Cortical Bands. Transverse. R.—Figs. 4a and 5a, C.N. 1138. Figs. 1g and 3g. Parenchymatous spaccs. Like R.—Fig 6f. R.—p. 90, Figs. 1f and 3f. p. 91, Fig. ghh’ and f. D.—p. 385, Fig. 10, C.N. 1113. LONGITUDINAL, RADIAL, AND TANGENTIAL. MEDULLA. Radial not Figured. See C.N. 1124 and 1982. PRIMARY XYLEM. Not distinguishable in long sections. ! SECONDARY XYLEM. Tangential, D.—p. 385, Fig. 15k’k”. See 1184. TRACHEIDS OF PRIMARY AND SECONDARY XYLEM. Tangential Surfaces. D.—p. 380, Fig. 4, C.N. 1167. Radial Surfaces. D.—Fig. 9, C.N. 1183. MEDULLARY Rays. Tangential Sections. D.—p. 382, Fig. 8. See C.N. 1184. Radial Section. D.—Fig. 9, C.N. 1183. CORTEX. Tangential of outer layer. D.—p. 385, Fig. 13, C.N. 1146. R.—p. 90, Fig. 6, C.N. 1144. Periphal Surface. See C.N. 1205 and 12072. Casé of the above Surface. See C.N. 1206 and 1207. PERIPHERAL APPENDAGES TO THE CORTEX. R.—p. 91, Fig. 6h, h’, h”, C.N. 1144. Fig. 8hh. Fig. ghh’. 60 Dr. W. C. WILLIAMSON oz LARGE CORTICAL TRACHEZEAL BUNDLES. Varied conditions.* Transverse, Double type, without secondary xylem. D.—p. 383, Fig. 17z, C.N. 1187. Single type, wzthout secondary xylem. oN. 211d. Double type, wzth secondary xylem. One wzth and one without secondary xylem. CAN. aia: Both Bundles, wzth secondary xylem. D.—p. 387, Figs. 19, 20, C.N, 1134. Single type, wth secondary xylem. pee CIN. 1293. FERN PETIOLES, PRIMARILY RACHIOPTERIS ASPERA. WILL. Of these I have sections from the broad bases, and from the ultimate twigs bearing the leaflets. BASE OF PETIOLE. Transverse. f.—p. 670, Wie. 1, C.N. 1575p. G82, Mic. by CN. ite. SMALLER BRANCHES. R.—p. 90, Fig 2, C N. 1854. Transverse. R.—p. 90, Fig. 1, C.N. 117. p. 91, Fig. 7, C.N. 1191, F.—p. 682, Fig. 7. See C.N. 119*, p. 682, Figs. 8and 9, See C.N. 135, 1191 .Q Longitudinal. F.—p. 680, Fig. 2. See C.N. 124, 125, 127’. R.—p. 91, Fig. 8, C.N. 1856. See also 1855. * The number, arrangements, and forms of these are most easily studied in fairly perfect transverse sections of the stems, in which we find seven modifications. I have noted their characteristic features in seventeen such sections. They are most commonly grouped in pairs, located in the innermost cortical zone, each pair being in more or less close contact. The above seventeen sections have furnished twenty-eight examples in this condition. Some- times we find solitary bundles, but such are otherwise undistinguishable from the twin ones. Of these I have recorded nine in the seventeen sections. We occasionally find a pair, one of which is in its normal condition, but where the peripheral surface of the second one is furnished with a variable number of secondary trachezal laminz arranged in a fan-shaped manner. In three instances I have found both the bundles so furnished, and in three examples I have seen the solitary bundles similarly supplied on their external borders. In nearly all the cases where the bundle of primary trachez has a zone of secondary xylem on its peripheral side I have found a zone of cambium investing its outer surface. In one instance the bundle must have been imbedded in the cambium, because the secondary laminz radiate equally in a star-like manner from the entire periphery of the primary bundle. We occasion- ally find the pair being pushed outwards through the outer cortex of the stem or branch. In such instances the two bundles are always imbedded in a considerable development of cortical parenchyma, which is obviously about to escape as a branch from the periphery of the parent stem. But this is a point that will require a more detailed examination later on—a point that involves the entire question of the position of the Filicinz during the Carboniferous age. The Fossil Plants of the Coal Measures. 61 TRACHEIDS. F,.—Fig. 3, A, B,C. C.N. 128, CORTEX. Dp Ool, big. 5. See C.N. 149, 150, 151. me 662, Fig. 11, C.N. 142. p- 681, Fig. 4, 5 (erroneously numbered Fig. 3 in the text). TERMINAL TWIGS AND FOLIAGE OF PETIOLE. F.—p. 683, Fig. 13.* See a similar example in C.N. 143. STRUCTURE OF INDIVIDUAL LEAVES. See 193a and 1856. DOUBLE BUNDLES ESCAPING THROUGH THE CORTEX, TO BECOME VASCULAR BUNDLES OF LEAF PETIOLES (RACHIOPTERIS ASPERA) OF LYGI- NODENDRON OLDHAMIUM. R.—p. 89, Fig. 1, k. C.N. 1880. See also C.N. 1890 and I150. C.N. 1980 (another section). C.N. 1981. A second section from the specimen 1980, but in which the pair of bundles and their appropriate investments have become almost completely detached from the parent Lyginoden- dron, and become an ordinary example of the Rachiopteris aspera. Thus, since the latter condition is a true fern frond, we now know that at least one of the carboniferous ferns possessed a true cambium by which was developed an elaborate zone of exogenous secondary xylem possessing conspicuous medullary rays. HETERANGIUM GRIEVII. Will. This plant approximates so closely to Lygznodendron in most features of its structure as to convince me that they belong to the same division of the fern family. Their distinctions are chiefly seen in the arrangements of the tissues which occupy the interior of the medullary cavity. Instead of finding the primary xylem, in its young state, entirely filling that cavity, and ultimately breaking up into about five very distinct masses, each of which adheres closely to the inner margin of the secondary xylem, as is the case with Lygznodendron, that central area of the stem is partially filled with very numerous small bundles of primary xylem, the intervals between which are firmly occupied by a network of what apparently ought to have been true medullary cells ; notwithstanding the peculiarity of their position and arrangement, I venture, as I did in the * The original of this figure is in the Cabinet of my old friend, Mr. J. Butterworth, of Shaw, near Oldham. 62 Dr. W. C. WILLIAMSON oz very similar condition seen in the axial centre of Lepzdo- dendron selaginoides, to apply to these cells the term medullary. PRIMARY XYLEM AND MEDULLARY PARENCHYMA. Transverse. D.—p. 395, Fig. 30a, C.N. 1250. p.' 395, Fig. 31b andic, ‘CIN 12co. Longitudinal. D.—p. 395, Figs. 32 and 33b and c, C.N. 1266, 1268, 1270, 1276, 1278, 1284. STRUCTURE OF TRACHEIDS. ; D.—p. 395, Fig. 24, C.N. any of the above longitudinal sections. SECONDARY XYLEM. Transverse. D.—p. 395, Fig. 30d, C.N. 1250. Longitudinal, D.—p. 395, Fig. 32d, d. C.N. 1250. MEDULLARY RAYS. Tangential. D.—p. 396, Fig. 332,.C.N. 1265, Fic: 33, C.IN: 1268. Radial. p. 396, Fig. 33f, C.N. 1268. CORTEX—INNERMOST ZONE. Transverse. D.—p. 396, Fig. 30g and 35g, C.N. 1250. Longitudinal, D.—p. 397, Fig. 32g, C.N. 1270. MIDDLE ZONE. Transverse. D.—p. 397, Fig. 3oh, C.N. 1250. Fig. 35h’. Longitudinal, D.—p. 397, Fig. 45, C.N. 1270. Fig. 32hh, C.N. 1278. LARGE VASCULAR BUNDLES IN INNER AND MIDDLE CORTEX. Transverse. D.—p. 399, Figs. 30 and35 m.m’. Figs. 37 to 44, C.N. 1240 to 1247- Longitudinal. D.—p. 399, Fig. 32m, C.N. 12849. See also 1248 (barred). ORIGIN OF BUNDLES. D.—p. 4o1, Fig. 30m’, C.N. 1250. ANOMALOUS BUNDLE WITH SHORT TRACHEIDS. D.—p. 401, Fig. 36, C.N. 1260. a The Fossil Plants of the Coal Measures. 63 OUTERMOST CORTEX. Transverse. D.—p. 398, Fig. 35. See C.N. 1244k.” Longitudinal. Dp. 208. Fig. 32k”, C.N. 1278k”. Youne Twices. Transverse. D.—p. 402, Fig. 46, O.N. 1244, 1280, 1283, 1295, and 1296. Longitudinal. D.—p. 42, Fig. 47. See numerous sections in C.N. 1287 and in C.N. 12964. HETERANGIUM TILIAZOIDES. Will. This plant approaches so closely to WH. Grzeviz, not only in the general type of its structure, but even in many of the details of its organisation, that I see no reasonable grounds for placing them in separate typical groups. At the same time, as I have shown in my Memoir XIX., notwithstanding its typical resemblances to H. Grzevzz, and though the differ- ences between the two are those of detail, and not of type, the beautiful structures of A. Tzlzewoides show a distinct advance to a higher order of exogenous organisation than we find in the former plant. ‘So far as its central vasculo- medullary axis is concerned, it is a true Heterangzum in every detail characteristic of the genus; but when we turn to the aspects of its secondary xylem, and its investments of highly developed Phloem, we discover the differences between the two forms. This is important. Lygznodendron, now clearly proved to bea true fern, carries inseparably along with it Heterangium Grievzz, and in like manner the latter cannot be disjoined from H. Tz/wozdes. If all this is incontrovertible, it results that the fern must now be regarded, not only as ranking amongst the exogenous Cryptogams, but as taking a high position in that well- characterised group. STEM OR BRANCH. Transverse. Nic —p. 289,) Fig, 1, C.N. 1302. 64 Dr. W. C. WILLIAMSON on MEDULLARY AXIS AND ITS PRIMARY XYLEM. N.—p. 289, Fig. 1A. Fig. 3a Medullary Cells, 3b Primary Xylem. p. 289. Fig. th. - Fig. 2B, CoN. 1303. N:—p. 289, Fig. 5, b.c and b.c.,* CN. 1303: N.—p. 389, Fig. 4d secondary vascular laminz ; 4h secondary medullary rays, C.N. 1303. N.—p. 389, Fig. 4g,g extensions of primary medullary rays. See also Fig. tg, C.N. 1302. Longttudinal. Radial. N.—p. 291, Fig. 9, including secondary xylem and cortex, C.N. 1628. p- 291, Fig. 9A, vasculo-medullary axis. a, medullary cells; 4, tracheids of primary xylem. p- 291, Fig. 9B; d,d, secondary xylem; h,h, medullary rays. PHLOEM ZONE C. Transverse. N.—p. 290, Fig. 1C, C.N. 1302. p. 290, Fig. 1k, defined Phloems of separate Vascular bundles. C.N. 1302. See also Fig. 5k, C.N. 1303, and 4k. PHLOEM RAYS. N.—p. 290, Fig. mn, C.N. 1302; p. 290, Fig. 2n, C.N. 1303; p. 292, Fig. 13n, C.N. 1619; p. 290; Fie.4n,n, CN. 53035 CORTEX, INNER. Transverse. N.—p. 290, Fig. Ip, C.N. 1302. p. 290, Hig. 2p, C.N: 1303 = ps 290, igs 3p, CANeer- PHLOEM ZONE AND INNER CORTEX. Longitudinal. Radial. SECONDARY MEDULLARY RAYS. N.—p. 291, Fig. 9hh, C.N. 1628. PHLOEM RAYS. Transverse. N.—p. 291, Fig 4n,n. Longitudinal. N.—p. 291. Fig. 9cn, C.N. 1628. PHLOEM TUBES—SIEVE TUBES OR CAMBIFORM CELLS. N.—p. 291, Fig. ol, C.N. 1628. PHLOEM. Tangential. N.—p. 291, Fig. tol (Sieve tubes ?), C.N. 1622. * Two bundles pushed outwards from their normal position as a regular portion of the secondary xylem cylinder. The Fossil Plants of the Coal Measures, 65 _ INNER CoRTEX. Radial, N.—p. 291, Fig. 9D, C.N. 1622. OUTER CORTEX. Transverse. N.—>p. 290, 291, Fig. 6r, C.N. 1303. Fig ir, C.N. 1302. Radial. N.—p. 292, Fig. 11, C.N. 1304. MEDULLARY CYLINDER. Transverse. P.—p. 156, Fig. 2b, C.N. 1833 ; p. 156, Fig. rbb.* AERIAL ROOTLET BUNDLES. Transverse. P.—p. 157, Figs 1 and 3. Seen in most of the transverse sections. EPIDERMAL HAIRS. P,—p. 158 and the Longitudinal section, C.N. 1857. TRACHEIDS OF MEDULLARY CYLINDER. All barred. See 1842 x and 1843 x. CORTEX. Mixture of Parenchyma, C.N. 1840, and Prosenchyma, C.N. 1841. ZYGOPTERIS.—Petioles only preserved. ZYGOPTERIS CORDA. During the last twenty years numerous organisms have been described under the name of Zygopteris. Most of these have been fern-like petioles. In 1889, Professor Stenzel, of Breslau, published a Wemozr “On the Stem ofa Carboniferous Plant,” to which he gave the name of Zygop- teris scandens, but which he placed in a secondary division of the Zygopteroid group (Ankyopterzs). Ihad previously (in 1888), figured one under my type-name of Rachiopteris Grayiz. Dr. Stenzel having sent me a copy of his WZemozrs, I arrived at the conclusion that our two plants were identical. In order to obtain his opinion on the matter, I sent him one of my sections of Rachiopterzs Grayzz for comparison with his * There is still some obscurity in the relations of rb’ to 2b’. Is the latter a modified con- dition of the former, or is it identical with the axil-sprosse of Stenzel, Fig. 3b, the Zygop- teriod petiole being wanting ? 66 Dr. W. C. WILLIAMSON om own plant. Owing to some differences in the dimensions of the two forms, he was unable to conclude that they were identical. The petioles of both examples being alike fur- nished with the characteristic Zygopterozd vascular bundle, I shall, for the present, continue to follow his example, and recognise my form under the more definite type-name of Zygopterts. The first specimen of this type which came into my hands I published in Wemoir VI., under Corda’s name of Axachoropteris Decatsnzi, its peculiar Zygopteroid petiolar bundle not having been discovered at that time. This, however, has now been done; and since has given the name of Zygopterzs to a characteristic example of this group of stems. I have elected to follow his example, and to apply the same type-name to the entire Zygopterozd group. RACHIOPTERIS. Will. Z. BIBRACTIENSTS. Renault. Transverse. F.—p. 697, Fig. 49, C.N. 195. For more perfect specimens see O.N. 196A and 197. Longitudinal. F. p. 697, Fig. 50, C.N. 108. See also C.N. 1815. ISOLATED CLUSTERS OF SEMI-SCLEROUS CELLS, APPARENTLY CORRE- SPONDING TO FIG. 32H OF HETERANGIUM GRIEVII. N.—p. 291, Fig. 11,t,t,t. C.N. 1304. VASCULAR BUNDLES, PASSING OUTWARDS THROUGH THE CORTEX, RESEMBLING THOSE OF D, Fic. 17Z, IN LYGINODENDRON, AND D, Fic. 35M’M, IN HETERANGIUM GRIEVII. N.—p. 290. Fig. tu’,u.’ C.N. 1302. Fig. 7u,u, C.N. 1623. Fig. 8u,u, 1302. SPECIAL BUNDLES, RESEMBLING D, FIG. 7, AND X, Fic. 24, IN LYGINO- DENDRON, AND D, FIG. 36, IN HETERANGIUM GRIEVII. N.—p. 292, Fig. 12w, C.N., 1622, and Fig. 13eee’. Fig. 13,e,e,w, C.N. 1619. STRUCTURE OF TRACHEIDS. N.—Barred, p. 293, Fig. 14, C.N. 1301. Reticulate, p. 291, Fig. 9, 1628. : \ Bordered Pits, p. 293, Fig. 16, C.N. 1621, | and i = | The Fossil Plants of the Coal Measures. 67 ZYGOPTERIS GRA YII.* MEDULLA. Transverse. P.—p. 156, Fig. 12, C.N. 1832. The medullary ce//s have disappeared. See also 264a. Medullary cells preserved in C.N. 1919D. ZYVGOPTERIS LACATTI. Renault. Transverse. F.—p. 696, Fig. 45, C.N. 201. p. 296, Fig. 47, C.N. 214. Longitudinal, F.—p. 696, Fig. 43, C.N. 212. Tracheids barred. For secondary branches passing off from the primary axis consult C.N. 1808 to 1812; for barred and recticulate tracheids in the same bundle see C.N. 1812. ZYGOPTERIS DI-UPSILON. Wil. Transverse. K.—p. 537, Fig. 90, C.N. 216. Longitudinal. K.—p.538, Fig. 91, C.N. 218. For longitudinal sections of Fig. 90c see C.N. 220¢. Ditto Fig. 9of see C.N. 210f. Ditto Fig. 90h see C.N. 218, 220, and 221th. RACHIOPTERIS. Will. This group comprehends a number of apparent fern structures, from amongst which no very definite sub- divisions can be established worthy of having assigned to them distinctive names. They will, therefore, retain their present provisional name of Rachzopterzs, until more can be ascertained respecting their several mutual relations. RACHIOPTERIS INSIGNIS. Will. A rare type, the Tracheids of which are very liable to be filled with Tylosa. Transverse. K.—p. 506, Fig. 19, C.N. 265. Fig. 20, the central vascular bundle of Fig. 19 further enlarged. p. 506, Fig. 22, Tracheids devoid of Tylose, C.N. 267. Longitudinal. K.—p. 506, Fig. 21, C.N. 265. See also C.N. 266B. Transverse. p. 507, Fig. 23. Section of the bundle of a secondary branch issuing through the Cortex of Fig. 21. * In memory of my old and distinguished friend, Asa Gray. 68 Dr. W. C. WILLIAMSON on RACHIOPTERIS ROBUSTA. Will. A rare form, of which I have only two sections taken from the same specimen. K.—p. 505, Fig. 23A, C.N. 269-270. RACHIOPTERIS ING@QUALIS. Will. Accidentally called A. zrregularzs in the text of Q. Q.—p. 206, Fig. 28. See C.N. 265b, 320 & 1814. RACHIOPTERIS CYLINDRICA. Will. Transverse. I.—p. 350, Fig. 80, C.N. 179. p- 351, Fig. $7, C.N. 179; p. 351, Fig. 38, C_N. 179; Longitudinal. Dp. 351, Fig. 86, C.N. 182, RACHIOPTERIS ROTUNDATA. Anachoroptis rotundata Corda. I.—p. 350, Fig. 79, C.N. 271. See also 272 and 273. RACHIOPTERIS GONIOCENTRA. Will. C.N. 274 and 275. Not yet figured. RACHIOPL ERTS DOPLLILXG A very distinct form, only obtained hitherto from the Burntisland or Petticur district. VASCULAR AXIS. Transverse. F.—p. 688, Fig. 28, C.N. 223, > a,a, Fig. 30, C.N. 237. Fig. 35, ete CO Ne NG 227 Loneztudinal. p- 688, Fig. 29, C.N. 234. See also 232 and 244. SECONDARY PETIOLES. Transverse. See Figs. 35A to 35K. P.—p. 693, Fig. 39, C.N. 240. Fig. 41. Figs. 35D, 35E. Longitudinal. See C.N. 239, 242, and 243. Tracheids reticulated. The Fossil Plants of the Coal Measures. 69 CORTEX. Transverse. Most of the specimens. Longitudinal. See C.N. 229 to 236. The specimens from which Figs. 35A to 35K were drawn are in the Cabinet of Wm. Carruthers, Esq., of the Natural History Museum, Cromwell Road, London. ZYGOPTEROID PETIOLE, VASCULAR BUNDLE. P.—p. 157, Fig. 3f, C.N. 1831.* AXIL-SPROSSE OF STENZEL.T P.—p. 156, Fig. 3e and Fig. 5a, C.N. 1831. RACHIOPTERIS OLDHAMIUM. Transverse. Matured. F.—p. 685, Fig. 20. See C.N. 160. Very young twigs. Hades, Mics. 22,23. 24. See 160 and 167, p. 685, Fig. 21, C.N. 160. Rather more advanced. p- 686, Fig. 25A, see C.N. 150, ‘4° C.N. 156, with a triangular bundie. p. 686, Fig. 26. With two secondary bundles becoming detached from the primary one. Longitudinal. p. 686, Fig. 27, C.N. 162. p- 220, Fig. 30, C.N. 1471. © * It is possible that this plant may only be a variety of Rachiopteris hirsuta. t+ When Memoir VIII. was published I was under the impression that each of these sporangia possessed an annulus. This is probably true of the rare form, Fig. 25, though I have no certain evidence that this was the case; but it is not true of the more common York- shire and Lancashire form, Figs. 27—30. 72 Dr. W. C. WILLIAMSON ox SPORANGIA; AN APPARENTLY DISTINCT FORM. p. 220, like Fig. 25, C.N. 1879. Cells of Sporangial wall much smaller. See C.N. 319a and 318. Several sporangia like Fig. 25, C.N. 318. Several sporangia, one evidently pendunculate, C.N.3109. MYVELOPTERIS. Renault. MYELOXYLON. Brongniart. Few fossil plants have been the subjects of more con- troversy than those figured in my Memozr under the first of the above names. The Medullosa of Cotta, the Palmacites of Corda, the Stenzelia of Goeppert, to the two names at the head of this paragraph, including the new Rachiopteris Williamsont of Seward,—this group has not only received a confusing number of names, but the question of its position in the vegetable kingdom has led to its being tossed to and fro between the Cycads and the Ferns. In my Memoir referred to above, I described a series of specimens in my Cabinet, numbered from C.N. 276 to 305. M. Renault at the same time was, unknown to me, studying similar objects ; we finally and independently arrived at the same conclusion, viz., that they were Carboniferous representa- tions of the living group of Marattiaceous ferns. At a later period my old pupil, Mr. Seward, undertook a re-examina- tion of my specimens, and found amongst them what appeared to be examples of two different forms. One of these he regarded as being true J/yelorylons, and the others as belonging to a more ordinary type of ferns, which he determined to publish under the name of Rachiopteris Williamsont. ‘The first result was the publica- tion, in the Azzals of Botany, for March, 1893, of a memoir on the Myeloxylons, which he regarded as constituting an independent type of plants intermediate between the Ferns and the Cycads, but apparently having a nearer relation to the latter than to the former family. Under these circumstances, in reply to my request, he The Fosstl Plants of the Coal Measures. 7% has kindly given me the following outline of his views on the structure of his R. W2llzamsonz :— “The structure of the petiole of Rachiopleris Willtamsont “resembles in many respects that of AZyeloxylon ; but the vascular ‘bundles show certain well-marked peculiarities, and a divergence “from those of Brongniart’s genus Mye/oxylon, which seems to “justify a specific separation. Mdyeloxylon agrees with Cycads “rather than with Ferns. acheopterts Williamsont approaches “much more closely to the typical fern-bundle, and is, therefore, ‘regarded as a fern petiole. “The two plants agree in (1) the nature of the hypodermal “tissue, consisting of alternating bands of sclerenchyma and paren- ““chyma; (2) in the possession of larger gum (?) canals in the ‘“sround tissue. ‘Their most important differences may be briefly “stated as follows :— “In Myeloxylon the bundles of vascular tissue are collateral, ‘Cand the protoxylem is placed next tothe phloem ; in Rachiopteris ““ Williamsont the bundles are concentric, and they agree in “position with that of the ferns. In Myeloxylon there are no *‘ parenchymatous elements associated with the xylem vessels. In “R. Williamsont there is much xylem parenchyma; another “marked difference consists in the occurrence of regularly disposed “canals surrounding the xylem in &. Williamsont. These do not “occur in Myeloxylon. The specimens which have been examined “of the new species show these canals in various stages of “development. They are quite distinct from the larger canals *‘scattered in the ground tissue, and are regularly arranged ‘towards the periphery of the phloem in each bundle.” Some of the differences here recorded are easily seen. Others are not so clear in my specimens. Of course the most conspicuous one is the existence of collateral bundles in one case, and of circumferential ones in the other. But even here we must remember that we have collateral bundles in ferns (¢g., Osmunda), and De Bary has found circumferential ones ina Cycad. Hence, the question arises, did these differences always possess the same distinctive value that they may do now? Schenck and Solms-Laubach 7A The Fossil Plants of the Coal Measures. differ even now on this point. Hence, I cannot conclude that these distinctive features settle the question of the boundary lines between the Ferns and the Cycads. For the present, however, I have arranged my specimens in Seward’s two groups, to facilitate their further investigation and study. MVELOPTERTS. PRIMARY PETIOLL. Transverse. G.—>p. 3, Fig. 1, C.N. 276. See also C.N. 286, 286b)> 26Genmeeus very large Petiole from Autun. SECONDARY DIVISIONS OF PETIOLE. Pe 3, Figs. 3, 4, and 4*. See 286a and others from C.N. 286 to 202. Longitudinal. Primary. p- 3, Fig. 2, C.N. 298—303. Secondary. p- 3, Figs. 5and6. See C.N. 286, 293 and 4. Oblique. G.—See C.N. 286e. SUB-EPIDERMAL SCLERENCHYMA. Transverse. C.N. 276. See also 305. Longitudinal, C.N. 276f, ¢. RACHIOPTERIS WILLIAMSONI. Seward. Transverse. C.N. 277 to 282. Five transverse sections from the same entire Petiole. Longitudinal. C.N. 283, 284, 285. Three sections from the same Petiole as the transverse ones. Memoir VII. Figure 7 is a transverse section of a vascular bundle of this plant. Wulfenta Carinthiaca, 75 Notes on Wulfenia Carinthiaca, Jacquin. By James Cosmo Melvill, M.A., F.L.S. (Recetved December 12th, 1893. or many years this beautiful member of the natural order Scrophularinee has maintained its prestige as the most local, perhaps, of European plants, if we except the Dioscorea pyrenacca, Bubani, from the P. de Gavarni, Eastern Pyrenees. To those who have read what may be called the ‘ Pioneer Guide-book’ to the Dolomites and Tyrol, the name of Wuz- fenia will be familiar, for Mr. G. C. Churchill, who collaborated with the late Mr. Gilbert in the production of this work,* based upon three successive visits in 1861-63, to what was till then a ‘terra incognita’ almost, to the tourist or botanist, made the search after this plant one of the chief aims of his journeys. Twice were the travellers disappointed at finding it past flower, but the third time they were fully rewarded. The following notes, written by my brother, the Rev. A. H. Melvill, and my sister, Miss Evelyn Melvill, who spent four months in the Tyrol this summer (1893), may be interesting : We left Lienz on Tuesday, July 25, by an early train for Greifenburg. With very great difficulty we succeeded in procuring a carriage to drive us to Hermagor, a distance of about 15 miles. Arrived there we had also great trouble in finding rooms, as the Post Hotel was full. At length we found some in the small hotel opposite the Post, where we were made exceedingly comfortable. We made enquiries first thing about the Wudfenza, and found the landlady knew all about it, in fact had dried specimens, and showed us in the garden some living plants, but out of flower. She said she would engage a guide for us, and afterwards, while we were talking to him and making arrangements for an early start the next morning, a German gentleman came * The Dolomite Mountains, by Josiah Gilbert and F. C. Churchill, F.G.S. London: Longmans. 1864. 76 Mr. JAMES COSMO MELVILL ox up who knew all about the subject, the best localities, &c., and said he thought we might find some late specimens, but that the bulk of it was over, the proper flowering time being May or June, and not July and August, as stated in most botanical works, and that this year was an earlier one than usual, owing to the small amount of snow which had fallen in the winter. He advised us to try the Watschiger Alp, and not the Kuhweger, though the latter is easiest to get at. So we settled to start at 5 the next morning, driving as far as possible, and engaged Josef Gobendorfer as guide. July 26. A perfect summer’s day. At 5.10 we were off in an einspanner, and in about three-quarters of an hour arrived at the village of Watschig. Here we alighted, and, crossing the Gail by a wooden bridge, found ourselves in shady pinewoods, which we traversed for a mile or so till we came toa brawling stream, where the ascent soon began in earnest. We had to mount along the bed of this stream, crossing it many times. Then up a very steep and rocky mule track till we came to a small lake with wonderfully transparent water (the Watschigersee). It was full of pine trunks. Then up again till we came in sight of the chalets on the Alps, and our guide pointed out to us the first plants of the Wu/fenza, but, alas! utterly over. We began to doubt whether we should find any flower at all. However, we found some other plants quite new to us and very pretty, especially one white flower of the order Caryophyllee. Waving rested awhile, we mounted yet higher up the slopes of the Gartnerkofel, and now we came upon the Wz/fenza in extraordinary abundance, covering in places every atom of the ground, young plants growing almost on the top of old ones, and seeming to struggle with them for existence ; but nearly all were out of flower, the tall seed-spikes rising in every direction, and showing what a splendid display there must have been earlier in the year. Our guide told us the whole mountain side here appeared dark blue ; but higher up than this (we were about 6,000 feet) the plant is entirely absent. The whole appearance of the Alp is like one vast garden ; lower down, where the Wzuifenia does not occur, there are great beds of Alpine roses, and by the stream many saxifrages and other flowers. It is appropriately called ‘¢‘ The Garden Mountain.” The Alpine roses were in places covered with curious | galls, looking like small peaches, and some bright scarlet, like tomatoes, of great size. To return to the Wu/fenza—after a long search we found about a dozen specimens, some of them with flowers still perfect, and some good enough for a sketch in our a7. Hist. Journal. Waulfenza may be thus characterized. A glabrous herb, with perennial stalk. Leaves nearly all radical, stalked, crenulate. Flowers in a one-sided cyme, blue, calyx 5-partite, sepals narrow. Corolla with cylin- drical tube, narrowed. Lobes four, the upper bifid, the lower ones either undivided or crenate. Stamens two, exserted ; inserted between the upper lobes; anther cells divergent, Wulfenia Carinthiaca. | 77 but confluent at the tips. Stigma capitate. Capsule acute, septiferous ; scepes leafless. Fl. end of May and June. The species are as follow: W. Carinthiaca, Jacq. Carinthian Alps. W. orzentalzs, Boissier. Seleucia, N. Syria, (Aucher Eloy), Antioch (Montbret.). Cf. Boissier, F7. Orzentalis, IV.. pp. 430, 431. W. Amherstzana, Bentham, Scroph. [nd., 46. Western Himalaya,nr. Kumaon,and Afghanistan, W.rentformis, Douglas ? It is uncertain whether this belongs to the genus. Wulfenta Carinthzaca, Jacquin (Mzscellanea 2, p. 62, t. 8). Leaves oblong, crenated, somewhat narrowed at the base, radical ; tube of the corolla swollen above the base, segments of the limb rounded, upper bifid, lower crenate, lateral often undivided, blue, whitish within: 1% to 2 ft. Syn. :—Pedarota Wulfenia, Lamk. Introduced to England, 1817. Named by Jacquin in honour of the Rev. Francois Xavier Wulfen, author of the Plante rariores Carinthtace. The localities affected by Wulfenza are most circum- scribed : Nyman, Conspectus Flore Europea, p. 543, cites “Carinthia meridionalis (Kiihweger Alp et secus Grise- bachium in 1872 detecta a cl. Schenk in alia alpe huic proxima) Carniola (Auernick Kofel el Ball 1865). Friul (pr. Ponteba sec Pir. Syll.) alp.” To which in the Supplement to the above, 1889, p. 235, he adds: “In Carinthia loca speciei natalia sunt (ex Pucher ; Gail- thal, circa montem Gartner Kofel in Watschiger-Kihweger-, Granitzer-, Zichel-, et Auernigalm: Hab. inter rupes; in pascuis et silvaticis apertis, 1500-1900 metr. s.m.; loca italica, 78 Mr. JAMES COSMO MELVILL oz ut in Conspectu indicantur, duo, sec. Caruel (1886) unum solum sistunt ; Friul, et quidem in limite extremo bor.- or alpium Italicarum in latere meridionali montis Auernick Kogel, orientem versus a monte Nassfeld (1500 metr.) qui versus boreali-occidentem a pago Pontebba (Pontafel) situs est.”. Comment. 193. In my Herbarium are specimens collected by Mr. Chas. Packe from the ‘ Gartner Kofel, supra Hermagor Carinthiz,’ July 9, 1870. Alpen de Trdpolaz by Dr. Lagger. “ Vallich, Carinthia, with no collector’s name, from the Boswell (Syme) collection; and the specimens collected by Rev. A. E- Melvill and Miss E. H. Melvill at the Gartner Kofel, July 27th, 1503. These places, Gartner Kogel, Kiihweger Alp, Vallich Trépolaz Alp, Granitzer Alp, Friul, Pontebba, are all within a very appreciable distance of each other, say 10 square miles, and may almost be called at most two localities, both in the same neighbourhood. The Italian boundary line is not far S. of Hermagor, and Friul and Pontebba are just below it. Bentham and Hooker, Genera Plantarum, Vol.I1., p. 913, divide the large order Scrophularcnee into three series, Pseudosolanea, Antirrhinide, Rhinanthidee, these being again subdivided into twelve tribes. The tenth in sequence, and the first of the series R/zxanthidea, is that of the Digztalee, thus well characterized, the following being a translation from the Latin :— RHINANTHIDE. Leavesvarious. Inflorescence simply centripetal. Lower lip or lateral lobes of the corolla external in the bud. Stamens very rarely more than four, often only two. Tribus X. Digztalee. Wulfenza Carinthiaca. 79 Corolla usually little if at all bilabiate, the lobes all plane, at the apex and often confluent. parasitic. the lateral or one of them external. Anther-cells contiguous Herbs, or shrubs, not The following genera belong to this section :* Stbthorpia, Linn. - - Hemiphragma, Wallich - Scoparia, L. - - - Capraria, L. - - Camptoloma, Bentham - Digitalis, L. - - - Lsoplexis, Lindley - - Erinus,L. - - - Campylanthus, Roth. - Lafuentea, Lag. - - Ourista, Comm, - - Pucrorhiza, Royle - - Synthyris, Bentham - Waulfenia, Jacquin - - | Calorhabdos, Bentham - Fedarota, 1. - - - Veronica, L. - Aragoa,H.B.andK. - 6 sp E Sp. I Sp. 4 sp. E Sp: 18 sp. 2 Sp. 1 Sp: 4 Sp. LSp. 18 sp. I Sp. 6 sp. 4 Sp. Sle: 2 Sp. circa 200 sp. 3 5P W. Europe, Africa, India, S. America. Himalayas. Tropics of both hemispheres. W. Indies and S. America, Mexico and Florida. W. Africa. Europe, Asia. Madeira, Canary Isles. Europe. Canares, Cape de Verde, and Arabia. Spain. N. Zealand, Andes of S. A. Himalayas. N. America. Carinthia, W. Asia, Hima- layas. Himalayas, Japan, China. Europe. Europe, Asia, America, Aus- tralia, N. Zealand. 5S. America. This list of Durand’s entirely agrees with the arrange- ment in Bentham and Hooker excepting in the removal of the Chinese and Japanese genus of 2 species, Rehmannia, Lib. and Fisch., to the Cyrtandreous section of the order * Index Gen. Phanerogam, Durand, p. 296, being a revision to date (1889) of Benth. and Hook. Gen. Plantarum, as approved by Sir J. D. Hooker. + Forbes and Hemsley in the Enumeration of Chinese Plants, Journ. Lin, Soc., Vol. XXVI., p. 195, enumerate 5 sp, of Calorhabdos, but they allude to the one celled ovary being more Gesneraceous than Scrophularious. fe) Mr. JAMES COSMO MELVILL on Gesneree, from which it had, with apparent reason, been removed (Ge. Plant, I1., p. 960), as agreeing with Ourzsza in several important details. I have brought here to exhibit with the specimens of Wulfenta from my herbarium, examples of all these genera excepting two, viz.: Camptoloma, of which only one specimen has ever been gathered, and Ca/orhabdos, which, as being so near an ally of Wu/fenza, | am sorry to have been unable to procure. As a substitute, however, I exhibit a plate of two species of the genus. We here in Lancashire can boast of, perhaps, the hand- somest of the whole series, as it is the type, viz., Dzgztalzs purpurea, to the Purple Foxglove, being one of the most plentiful plants in the neighbourhood, often, as at Prestwich, monopolizing everything else, self-sown, in a shrubbery, or open space, and ornamenting many a woodside in the summer. Many species of Veronica likewise abound around us. The genera have been placed with much care and circumspection by the learned authors of Gen. Plantarum: and there can be no doubt but that the nearest allies of Wulfenia are Ourisza, - - Stamens 4 Picrorhiza, - “ * 4 Synthyris, - - . 2 Calorhabdos, - : 2 and Pedarota, - - - 4 this last showing a decided link between this plant, and the spicate Veronice (Pseudolystmachia, Bentham, Leptandra, North), the first mostly natives of Europe and Asia, the latter of North America. Pedarota, Linn, in its’ two species P. Agevza, Te and Sonarota, L., with the hybrid between the two, named by Huter P. Churchillz in honour of Mr. G. C. Waulfenia Carinthiaca. 81 Churchill, of Clifton, Bristol, the well-known European botanist, shows much affinity, as already said, with Wudfenza, especially in the species Boxarota, the flowers of which are purple, while those of Agevza are yellow. But the nearest approach to the genus in formation of its corolla and other particulars is undoubtedly the N. American genus Syuthyris,* Bentham. Here the flowers are small, purplish for the most part, in a simple spike, the stamens (two exserted) are situate close to the sinuses of the corolla, which is 4 cleft, somewhat irregular. Style filiform, with capitate stigma. The main difference between the two genera is that the anther-cells are in Syuthyris not confluent. Wulfenta is, however, a much more showy plant. The genus Gymwnandra, Pallas, a small Oriental and Arctic group, now placed in the WV. O. Selagznee, and allied to Globularia, L., has several points in common both with Pedarota, Synthyris, and Wulfenza,; indeed by George Don, in Dichlamydeous Plants, Vol. 1V., p. 581, it was placed in Scrophularinee, next to Wulfenza. Here the corolla is bilabiate, upper lip either emarginate or bifid, lower one 2-4 cleft. Stamens 2. The order Selaginee has many points in common with Scrophularinee ,; all (nearly) the species of both orders turn black in drying, the Se/agznew are, how- ever, as a ruleofa different habit; many assume an eriliform appearance, and they differ mainly from the Scrophularinee by the cells of the ovary being 1-2 ovulate, and even Ben- tham and Hooker confess this character is not always to be relied upon. Lastly, the genus Ourzsza, Comm., native of New Zealand, Tasmania, and the Andes of S. America, may be compared with Wulfenia, as possessing many attributes in * In May, 1872, I had the privilege of spendinga short time with the late Dr. Asa Gray, at Cambridge, Mass., U.S.A., and, showing me Syuthyris Houghtoniana growing in the Botanical Garden, he pointed out its characteristics and touched upon its afflnity to Wz- Jenia, Digitalis, and Veronica, and, if I remember aright, mentioned that he considered the genus one of the most interesting in North America. 82 Wulfenta Carinthiaca. common, but the stamens are 4, and not exserted. The habit of such species as the New Zealand O. macrophylla would be, I should imagine, the same. In a Flora like the European, in which are found very large assemblages of certain genera like Wzeraczum,Centaurea, Linaria Ranunculus, Saxifraga, and Carex, it is all the more interesting to note a few isolated types which have just put in an appearance, as it were, and only just impinge upon the Flora. How the Waulfenza first became established near Hermagor we cannot divine, but it is evidently of Eastern origin. The Dzoscorea, to which we have already alluded, is even more interesting as being a member of a subtropical genus, not otherwise known in Europe except in one Pyrenean station, and the Ramondia Pyrenaica, Lam., with its two allies Haberlea Rhodopensis, Frivaldsky, and /ankwa fLeldreichiz, Boissier, of the natural order Cyrtandree, a section of Gesneracee, otherwise tropical or subtropical, are parallel instances of localization. These three are found, one in the EK. Pyrenees only, the next in the Balkan Mountains, Thrace, and the third, and rarest, on the Thessalian Olympus. Other instances might be adduced: all one can do is to note the facts, and attempt to draw conclusions. The question of the geographical distribution of plants is most fascinating, and some of the data are quite without the possibility of solution. Our own islands afford plenty of material; many of our rarest plants are confined to one spot, and two, Sperantheus Romanzoffiana (gemmupara, Linn.) and Evriocaulon septangulare, L., natives also of the Neartic region, are unknown in Europe excepting in Ireland, and as regards the latter the I. of Skye, in addition. JACQUIN. WULFENIA CARINTHIACA: : of Plate I. MEMOIRS AND PROCEEDINGS MANCHESTER LIT. AND PHIL. 800. PROCEEDINGS. 83 | Wecroscopical and Natural History Section.| Ordinary Meeting, December 18th, 1893. Mr. R. E. CUNLIFFE, President of the Section, in the Chair. Mr. J. F. ALLEN and Mr. J. WATSON were elected Associates of the Section. The PRESIDENT moved :—“ That the Section notes with great regret the loss to science caused by the death of ieeceevnadall, oD. M.D; DIC.L., Ph.D. F.R-S., F.C.” Dr. BROADBENT gave a microscopical demonstration of Infusoria found in water obtained from manure heaps. Pee, MELVILL, F.L.S., exhibited a specimen of Bulimus (Porphyrobaphe) labeo (Broderip) from Peru, a very scarce land mollusk, conspicuous for the swollen, almost diseased appearance of the marvellously incrassate and reflected outer lip, which has deep pittings and crenula- tions all over its swollen and tumid surface, not dissimilar to the appearance of cooled lava. Mr. Melvill also exhibited eleven of the thirteen or fourteen known Rhopalocera of New Zealand, which country is the poorest in the world for its size for not only these insects, but also those of most other orders, although the bulk of the Coleoptera and Hymenoptera which do occur are peculiar, and show the extreme antiquity of this land, formerly, according to Wallace, a large continent embracing the Macquaries, Lord Howe Island, The Auckland and Campbell Isles, and Norfolk Island. That it has been dissociated from Australia from the earliest times, is evident by the differences in the Flora as well as in the Fauna. 84 PROCEEDINGS. “The Butterflies” remarked Mr. Melvill, “are as follow : DANAIDA. Hlamadryas Zotlus (Fabr.). Also occurs in Australia. Danais Evrippus (Cramer). A North American species, very nearly cosmopolitan. NYMPHALID~. Pyramets [tea (Fabr.). Also in Australia. P. Gonerilla (Fabr.). Endemic. A handsome species. P. Carduz (1). Quite cosmopolitan. Known in England as the ‘Painted Lady. ing the boundaries of human knowledge in the future as it has been in the past.” In answer to one of the members, Mr. WILDE said that the electricity from the Corporation mains could be rendered suitable by means of an induction coil for the ozonizing of oxygen for bleaching purposes. Special interest was manifested by the members in Mr. Wilde’s exhibition of a new line which he has observed in the spectrum of thallium, and in experiments with his “Magnetarium,” to illustrate his theory of terrestrial magnetism. 98 PROCEEDINGS. Ordinary Meeting, February 6th, 1894. Professor ARTHUR SCHUSTER, PH.D., F.R.S., Foie President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Mr. RoBERT MOND, M.A., F.C-S., of Winnington Hall, near Northwich, was elected an ordinary member of the Society. Messrs. H. GRIMSHAW and R. E. CUNLIFFE were appointed auditors of the accounts for the current year. Professor SMITHELLS, of the Yorkshire College, read a paper on “Flame and Flame Spectra,” and showed a series of experiments. After separating the inner and outer cones of a coal gas flame, he showed that lithium burnt in the inner cone and copper oxide or chloride in the outer cone. The paper contributed considerably to the question of the temperature of flame and the causes of its luminosity. Professor DIxon, Dr. BAILEY, Dr. BOTTOMLEY, Dit HARTOG, Mr. JONES, and Dr. SCHUSTER Joined in a long discussion on the interpretations of Professor SMITHELL’S experiments. PROCEEDINGS. 99 [Wecroscopical and Natural History Sectzon.]| Ordinary Meeting, February 12th, 1894. Mr. R. E. CUNLIFFE, President of the Section, in the Chair. Mr. ROGERS exhibited specimens of Egyptian cloth from the Fayoum, twelve hundred years old. Mr. BAILEY drew attention to the weaving and the pattern, and suggested that it could only have been pro- duced by means of a loom constructed on the Jacquard principle. Mr. ROGERS also exhibited specimens of cotton silicate. Dr. BROADBENT exhibited large specimens of cloth prepared from the bark of trees, decorated with geometric coloured patterns by the natives of the Samoa islands. Mr. HYDE exhibited several cockroaches found near the Ship Canal at Ellesmere Port, supposed to have been conveyed from the United States in cargo. es SS 100 PROCEEDINGS. Ordinary Meeting, February 20th, 1894. Professor ARTHUR .SCHUSTER, Ph.D., F.R-S, Faas President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Mr. W. E. HOYLE, M.A., exhibited the following shells, recently acquired by the Manchester Museum :— (1) Bathybembix argenteonztens,from Japan. A character- istic deep-sea form, with delicate sculpture and pale irides- cent colouration. (2) Columbarium pagoda. A rare marine shell from Japan. (3) Columbarium distaphanotis. A beautiful shell, of which the type-specimen from an unknown locality is unique. This example is from the Cholmondeley collection. (4) Opzsthostoma mtrabile. A land shell from Borneo, in which, after a certain number of spiral turns, the shell bends upwards and the mouth comes to lie close to the apex. (5) Palaina Quadrasz, An exquisitely sculptured oper- culate land shell from Manila, in which the first whorls form a right-handed and the last a left-handed spiral. Professor SCHUSTER exhibited an apparatus in use at Owens College for testing clinical thermometers, and read the following note :— “The Owens College has, during the last few years, undertaken the testing of clinical thermometers for medical men and others. It is the object of this note to describe the apparatus by means of which a number of these thermo- PROCEEDINGS. IOI meters can be conveniently tested at the same time. The apparatus consists essentially of two cylindrical vessels, one being placed inside the other. Both are filled with water. The thermometers are placed in a carrier inside the inner vessel. The water in the outer vessel is maintained at any desired temperature by an electric current passing through a platinum wire in the water. The water in the inner vessel is kept stirred by means of a revolving screw turned by an electric motor. “The details of the different parts are shewn in (Figs. I to 3). Fig. 1 shews the carrier made of brass which holds an Gee ees = He —\—__I\ \ = AV} — \==\ Xi as el f | Zea (iil ide ! He {\a —t = the thermometers. They are all kept in place by an india- rubber band pressing them against strips of brass bent so as to form a triangular groove. Fig. 2 shews the inner vessel round which a framework is fixed, carrying platinum wire. This wire is insulated from the vessel by means of indiarubber tubing which is placed over the brass supports. Fig. 3 shews the whole apparatus when put together. The brass rod passing into the vessel carries a pulley at one end and a stirrer at the other. “The thermometers are compared at three temperatures, Viz., 98°, 103°, 108°. The vessel, without the carrier, having G 102 PROCEEDINGS. been filled with water at some temperature below 98°, a current is sent through the platinum wire of such strength Pisce 2: that the temperature ascends fairly rapidly to the required temperature, this current strrength is then suddenly reduced to an amount previously determined and chosen so as to give a very slow and steady increase of temperature in the inner vessel. This rise should not exceed o* IF. per minute, but it is essential that the temperature should con- tinuously rise during the time of testing. As soon as the temperature has thus been regulated the carrier is plunged into the water. There is at first a cooling of the water due to the introduction of the carrier, but the rise in temperature soon begins to reassert itself. The cooling may, if desired, be almost reduced to nothing, by keeping the carrier before insertion into the testing vessel in water of about go°, but there is no particular object in thus complicating the ne Ege PROCEEDINGS. 103 method of procedure. The thermometers are kept in the water for a sufficient time to allow them to acquire its temperature. The standards are then read off, the carrier fem out, and the thermometers may be read off aecicure: che operation is repeated’ at the other temperatures for which the testing is to be carried out. As the ultimate standard of temperature, I use at the College a thermometer calibrated carefully at the Bureau Interna- tional des Poids et Mesures at Paris. With proper precau- tions a temperature can be read off to three or four thousandth of a degree Centigrade. The College also possesses a thermometer divided into tenths of a degree Centigrade, compared with the Standards of the Technische Reichs Anstalt at Berlin. Both thermometers agree in their indications; they are made of glass, the composition of which in each case is definitely known, and their indications may without trouble be reduced to the air thermometer if desired. “T have, further, two thermometers compared at Kew. One of them (A) is divided into fifths of a degree Fahren- heit, and ranges from goto 115. Unfortunately it does not contain the freezing point, so that its changes cannot be followed. The other (B) is divided into tenths of degrees, and ranges from 90 to I10. Another part of the stem, separated from the rest by a bulb, is divided from 30° to 30°, so that its freezing point may be tested at any time. Finally, three clinical thermometers are also used for com- parison, two of them having been standardised at Berlin and one at Kew. “There is some doubt as to what the scale of temperature used at Kew really is; but the difference between the Kew temperatures, and the scale used on the continent, being probably about =4th Fahrenheit near 100° F., is of no im- portance as regards clinical thermometers. “The following comparison shows the agreement between 104 PROCEEDINGS. the different thermometers used as intermediate standards in an experiment carried on exactly as in an actual test :— Kew, B reading, 101°4. Clinical thermometer, found at Kew to be COFFECE sie oor Sete Sti. oo!) Oa (1) Clinical thermometer, found correct at Berlin Mee ae oy ae +s) SOT (2) Clinical thermometer, taking account of corrections supplied by the Technische Reichs Anstalt ... i % i eR “The thermometers agree, therefore, to the limits of accuracy which can be attained. “The College has tested about 300 thermometers in the last two years, and, as a general rule, it may be said that the corrections have been small; but it has occasionally happened that thermometers were found to be wrong by o'4 and 0°'5, which shows that no thermometer can be trusted to be sufficiently accurate which has not been compared with some standard. “The result of testing also has shown that the more expensive kinds of thermometers have errors as great as the cheaper ones. The advantage which the more expensive thermometers claim, of being more rapid in their indications is often illusory. When a clinical thermometer is plunged into water of 100°F. it takes up the temperature almost immediately, and as to the time required when the ther- mometer is placed into the mouth of the patient it is the bad conductivity of the tissues of the mucous membrane which causes the lag in the rise. The skin or tongue is, under ordinary conditions, below the blood temperature, and is further chilled by the introduction of the cold thermometer. By reducing the mass of the thermometer the first chilling effect may be diminished and the instrument would indicate more quickly the correct blood temperature. But the gain is not as great as is generally supposed.” PROCEEDINGS. 105 Mr. THOMAS Hick, B.A., B.Sc., read a paper “On the primary structure of the stem of Calamites.” A discussion ensued, in which Professor WEISS and Mr. CHARLES BAILEY took part, the latter commenting on the rapid advances which are being made in the knowledge of fossil botany, which threaten, in his opinion, to make it necessary to revise all the classificatory work which has previously been done. General Meeting, March 6th, 1894. Pretessor ARTHUR SCHUSTER, PH.D., F.R.S., F.R.A.S., President, in the Chair. Professor A. S. DELEPINE and Dr. G. H. BROADBENT were elected ordinary members of the Society. Ordinary Meeting, March 6th, 1894. Proiessor ARTHUR SCHUSTER, Ph.D., F.R.S., F.R.A.S., President in the Chair. The thanks of the members were voted to the donors of the books upon the table. Reference was made to a recent display of Aurora Borealis visible in Manchester. Mr. W. E. Hove, M.A., gave a demonstration of the luminous organs of cuttle fish, exhibiting sections with the aid of the lantern and under the microscope. A conversation on the causes and purposes of apparent self-luminosity in the eyes of carnivorous animals ensued. es 106 PROCEEDINGS. Ordinary Meeting, March 2oth, 1894. EDWARD SCHUNCK, PH.D., F.R.S., F.C.S., Vice-President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. A discussion on meteorites took place. Mr. JULIUS FRITH read a paper “On an analysis of the Electro- Motive Force and Current Curves of a Wilde alternator under various conditions.” The object of the ex- periments described was to determine how far the behaviour of an alternator containing iron in the armature agrees with that of the theoretical alternator without iron. It was stated, as an approximate result, that the agreement was fairly good. [Wicroscopzcal and Natural Hestory Section.| Ordinary Meeting. March 12th, 1894. Mr. CHARLES BAILEY, F.L.S., in the Chair. Professor F. E. WEISS, of Owens College, was elected a member of the section. Mr. HYDE drew the attention of members to the flowers of the alder, poplar, willow, hazel, and birch, which are unusually large, numerous, and early this year. Mr. ALLEN exhibited specimens of natural asbestos, and of silicate wool, produced by steam blown through slag when in a molten condition. Mr. BROADBENT exhibited additional specimens of genatoo (tapa,) prepared from the bark tissue of trees by the natives of the Samoa Islands. r PROCEEDINGS. 107 Ordinary Meeting, April 3rd, 1894. Professor ARTHUR SCHUSTER, Ph.D., F.R.S., F.R.AS., President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. Professor OSBORNE REYNOLDS read the following note “On the Aurora Seen at Fallowfield on March 30th, 1894.” “ At 10.20 p.m. on March the 30th I observed, from the Ladybarn Road, immediately in front of my house, which runs east and west, that there was an unusual amount of light, for the time of year, all over the sky on the northern side of the zenith, from east to west. At first it was only the amount of diffused light on the northern side, as com- pared with the southern, that caught my attention. The sky was perfectly clear at the time, and the stars were bright over the south, while over the north only the larger stars were visible. There was no moon. “Tt soon became evident to me that the light was that of the aurora; but at first it was only remark- able for the amount of diffused light, of a pale green for the most part, but passing into red towards the south. After observing it for some 15 minutes aie appearance became much more remarkable Streamers rose towards the zenith from all parts of the northern horizon with great rapidity and vanished again as quickly, and following these up to the zenith I saw what I have never seen before. The sky was in a state of fluttering light, wave following wave three or four a second, the waves moving in the direction of the streaks of light which suggested showers of luminous hail. The most remarkable thing was, however, that the appearance of 108 PROCEEDINGS. waves was owing to the fluctuation of light in set places— more or less a series of broad bands across the direction of motion of the waves. These broad bands of misty white, fluctuating light, with more or less well defined dark between, preserved a set shape. The light ap- parently ended in a very bright arc like “ay ‘one bright cloud running in an irregular line east and west through the zenith. The line had a decided wriggle in it near the zenith, and on the north was a dark space with another bright band with a corresponding wriggle, so as to create the appearance of a dark river between two bright banks. This shape lasted some time, disappearing and reappearing with the light. Watching this phenomenon, and looking towards the zenith, it became clear that the waves of light were moving nearly vertically, a little towards the south, and that they only took effect over a portion ofthe sky. Thus the motion seemed to diminish as it neared the zenith, and in the bright arc, exactly as though there was an illuminated vertical hail storm. I watched it about half-an-hour, when the zenith effects seemed to me to be diminishing.” Mr. GWYTHER gave an account of the appearance of the phenomenon at Buttermere, where it presented a some- what different aspect. Mr. BROTHERS and Professor SCHUSTER also took part in the discussion. Mr. HENRY WILDE, F.R:S., read a paper “Onpeme Influence of the Configuration and Direction of Coast Lines upon the Rate and Range of the Secular Magnetic Declination.” be , eit ae ce eae na Pe ee : P. ’ voy i r 1" 1 Abn ta eee ¥ + ts a rt Ay a = AY Sertes Vol VILL K Larthitions of. the B-gor Plate ll 756" =-1'2 -J¢ 213 MEMOIRS AND PROCEEDINGS, MANCHESTER LIT. AND PHIL. SOC. The K-partitions of the R-gon. 109 On the k-partitions of the R-gon. By the Rev. Thos. P. Kirkman, M.A., F.R.S. (Recezved, October 17th, 1893.) A tape-face in a partitioned R-gon is either a triangle having only one edge, or a quadrilateral having two opposite edges, in the contour of the R-gon. No tape-face carries a marginal triangle. The first step in the reduction of the general k-par- titioned R-gon is to drop out all its tape-faces. The 14-gon, Fig. 9, has four tape-faces. By dropping them out, we make it the 10-gon, Fig. 10, which has no tape-face. Fig. 9 has thus lost 4 contour edges, and 3 diagonals, and therefore three faces. The last operation in the construction of a definite partitioned R-gon, is the insertion of the tape-faces that it is intended to contain. Our definitions and reasonings, until we come to handle the tape, Fig. 8, apply to partitions in which there is no tape-face. Definitions: The bases of all marginal triangles of a partition are marginal diagonals. All other diagonals are non-marginal. A prime partition is one whose diagonals are all marginal. iies) 2. 5S, 2, 0, c, d, e, f, 6, 7, are primes. A sub-marginal face has for an edge one, and only one, non-marginal diagonal, and carries 251 marginal triangles. A belt is a row of primes only, which have each more than two marginal triangles, and cohere each with the next by the united bases of two marginal triangles, that are H iG THE REv. THOS. P. KIRKMAN oz hidden, being both creased under. J; is a belt of 5, Fig. 10 isa Delt of 2, primes. We are to conceive two marginal triangles creased under every non-marginal diagonal, in every partition. When a belt falls asunder into its component primes, the undercreased marginal triangles are seen in their right position. Compare ABCiD,D: in 9§8,, with ABadbd out ‘ of JB, in the figures. In 96:1, A and D2 are submarginal, but not in Figs. 1, 2, 3, 4) 5. The belt JB, has 41 summits, besides the two terminals crossed, and has 19 faces. 2. Every face of a partitioned R-gon, which has two and only two non-marginal diagonals dd’ for edges, may be and will be here considered, as a loose pane, and may fall out with its complete contour and fringe of marginal triangles, after which the two da’, whether they have or not a common point, can become one; so that the number of non-marginal diagonals in the R-gon is diminished by one. When a face has for edges 3+2 non-marginal diagonals it cannot fall out, so that the 3+z non-marginal diagonals can become one, diminishing thus by one the number of non-marginal diagonalsin R. In Fig. His no face that can drop out. IniJ8 any one of A BC: D, D, may so fall out ; and any two, or all the five, can disappear with their mar- ginal triangles. If all the primes that have for edges only two non-marginal diagonals were so to fall out of Figs. (1, 2, 3,4, 5), nothing would be left of all the five but exactly our first figure H. 3. This H is zvreduczble, because it has no face, having two, and only two, non-marginal diagonals for edges. Definition: A k-partitioned R-gon in which is a face that has for edges 3+z non-marginal diagonals, but no The K-partztions of the R-gon. III face having two, and only two, non-marginal diagonals, is an irreducible partition. | H has 23 summits and 14 diagonals, Its 15 faces are 9 marginal triangles, 4 sub-marginal faces, and 2 faces, a triangle and a pentagon, each of which has for edges 3 -non-marginal diagonals that are the 5 dark edges of the figure. An irreducible H may have any number of faces that have each for edges 3+2 non-marginal diagonals, z being different or not in any two such faces; and such edges will be called dark edges in H. It is equally true that triangles and squares having two,and two only, non-marginal diagonals for edges, and carrying no marginal triangle, may fall out and leave only an irreducible H. Thus, if in Fig. H the dark line 65 be completed into either the triangle 655’ or the 4-gon 6'655’. carrying no marginal triangle, either 655’ or ‘6655’ would have for edges two non-marginal diagonals da’, and both could fall out, leaving, after union of each @ and d@’, exactly the irreducible H. We shall have to consider both the insertion and the dropping out of such simple faces. They form, when out of the R-gon, not a belt, but a tape, like Fig. 8. The dark and dotted lines in Figs. (1, 2, 3, 4, 5), viz., the pairs 65, 6'5’, 31, 3/1, &c., are the diagonals dd’ that could unite if the primes between them were to fall out. The dark line on the right in A, Fig. 1 is an error—it should be not dark. This brings out a novel and useful notion—that H can be completed into a k-partitioned R-gon by simply splitting its dark diagonals to receive one or more primes out of a given belt. And it is evident to the reader that there lies the secret of the construction upon H of Figs. (1, 2, 3, 4, 5). 4. If 4 be the number of the primes in the belt, and 4 the number of the dark diagonals of H, we have only to form all the k-partitions of +, as a,, a,,... a,, Whose sum of 112 THE REv. THos. P. KIRKMAN oz k parts is x, any number $0 of the parts beginning the partition being zero (as ¢.g. a,=a,=a3=0, the rest being >o) ; next, having named clearly our £ dark diagonals as 1%, 274, 374, ... £™, we have to insert into the 1** split diagonal, the first a, ($0) of the x primes in the belt; next to insert into the 2™¢ split diagonal the second number a, So of the remaining +#—a, primes in the belt, and so on, lifting and dropping the primes in their order till all the & dark diago- nals have been, by the guidance of the same partition, a,, 4,...a,, Charged each with a given number a, >0, of primes, leaving none in the belt. We have to make the like use of the next k-partition of the number x to empty by its guidance the same full belt, by charges >0 placed in every split diagonal, till every k-partition has been so used to empty the same full belt. If the number of the k-partitions of x so used is 7, we shall have turned the same irreducible H into m different partitioned R-gons. We shall presently see what number m is, and be able to describe them. But our task is only begun by this handling of all the 7 k-partitions of our number x of primes in our belt, each in dictionary order. We have to handle in the same way, for the charging of our & dark diagonals of H, every permutation of the parts of each of those 7 k-partitions, emptying our full belt into the split diagonals M times; where M is the number of all the permutations of the k-partitions of +; which M includes the number 7 of the k-partitions above handled unpermuted. It is plain, that each of these M permutations bids us distribute in a different way our x primes into our £-split and unsplit dark diagonals, of which the z™ will be unsplit, when the z™ place of the guiding permutation is zero. But all this fuss of distribution is mere wind. M is all we really want, and that theorem Q gives at once. 5. It is then evident that the number of distributions of The K-partitions of the R-gon. 113 the z=5 primes of the belt JB: into the 2=5 split and unsplit dark diagonals of H, is not less than M, that of all the permutations of the 5-partitions of the number 5, zeros and repetitions being allowed in the partitions. By my theorem Q (vzde the coming volume of Reprint of Educational Times), this number is the £ co-efficient in (1+1)"""", ze, the 5 coefficient in (1+1)**’, which is So7o0:;1234—120. This M=126 is all the answer we can get if our +=5 primes can be exhibited in no other belt besides one JB:. This can be the case only when the primes are 5 squares in the belt, which out of it are 5 6-gons, that are capable of only one order and posture in a belt JB, having their marginal triangles all creased under and forming not a belt, but a tape, like figure 8. Such primes, having only two marginal triangles, are excluded by defini- tion of a belt in Art, I. There are many belts. There is no second 6-gon A,, nor second 9g-gon Bz, having 3 marginal triangles ; but there are three 14-gons (C) and three 15-gons (D) that all have 6 marginal triangles. These are the 2-zoned C, and the monozones- C2 C;, Figs. a,c,¢; also the 3-zoned D,, the asymmetric D. and the monozone Ds, Fig. 4,a,7 To secure the construction of all our partitioned R-gons» these eight primes must all alike contribute to form a belt of five, the belts so formed being equivalents. 6. We have to use the 18 equivalent belts following: JB, ABCiD,D; ; 3B,, ABC,D.D; ; JB, ABCiDSD; ; B:, ABC,D,.D; ; B:, ABC,D5Ds ; ‘Bs, ABG,D;Dy; JB,, ABC;D:1D.; JBe, ABC,D.D,;; JB, ABC;D,D; ; nine belts in which is no repeated prime ; and JB, ABC,D,D,; JBi3, ABC: D2D2; YBic, ABCiD;D; 5 Bu, ABC,D,D, ; Bu, ABC,D2Dsz ; Bu, ABC,Ds3Ds 5 Bu, ABC3D,D; ; JBis, ABC;D2Dz ; JBis, ABC;D3Ds3 ; nine in each of which is a repeated prime. 114 THE Rev. THos. P. KIRKMAN on We have no z-ple: prime in our belt. The least 2-ple is Fig 6; the least 3-ple is Fig 7. Observe that by z-ple we mean always zoneless z-ple If we had used a prime having 12 marginal triangles, our belts, of += five primes each, would be numbered not by half dozens, but probably by hundreds, all as much alike in features as are the above 18 equivalents of one selected belt. We shall speak in Art. 8 of the uw permutations of their order in the belt of the #=5 primes, which are gure different from the above M permutations of the parts of all the # integral k-partitions of the mere number x The numbers £ and # may differ in any way or degree. Here 4—=%5 is convenient: Each of the 18 belts will add 39 of its 41 not crossed (Art. 1) summits to the 23 of H (Art. 3); and 10 fagesyed which 14 are marginal triangles, to the 15 faces of H. At this point it is requisite that we clearly state the complete problem that we intend to solve. It is this—to enuinerate the number of partitioned R-gons, that all alike fulfil the conditions following :— ist. That they be all reducible to the same irreducible H, or to an irreducible identical with H, in the names of the polygons that compose H, two of which shall be exactly the 3-gon and the 5-gon that have for edges the dark diagonals in H. 2nd. That they each contain one of the 18 equivalent belts above described. 3rd. That they all contain the same tape, Fig. 8. Observe, that no partitioned R-gon can contain more than one selected belt or tape, still less two irreducibles H. Our problem resembles a famous old one: In how many ways can you put all of f things into z fixed places, leaving any tS of the z places vacant? In that, when e things are wanted to put in places chosen, any e of the / things, if they are all units, will serve as well as any other e of The K-partitions of the R-gon. II5 the unplaced. But it is otherwise with our belt units. I call them units because each has to contribute a unit to the occupation of a place. To prevent confusion in our final account, it is necessary, and fortunately it is sufficient, to insist that if, directed by the first term of your guiding permutation (any one of the above M), you begin by putting (c) of my units in the place (the split diagonal) that you have first chosen, you shall take the first cin the belt before you; and that if you take @d more to put into your second place, you shall take the next d in the belt ; and, moreover, that my units in their new place shall sit in order, and wear their marginal triangles, exactly as they did in the belt, Mie diagonal split has a name, 12, 13, or ac; where a1, having, out of a belt, # mar- ginal triangles, can take in a belt, by using all its base-ties, mu(m — 1 )2~* postures, and no more. 2. An 7z-ple prime, 72, having, out of a belt, 7 marginal triangles, can take in a belt, by using all its base-ties, 27(7—1)z" postures, and no more. 3. An asymmetric prime that has, out of a belt, #z mar- ginal triangles, can take in a belt, face-up and face-down, 21(# — 1) postures, and no more. This gives us, if ~ be the posture-number of a prime, and # the number of its marginal triangles, For the 3-zoned A, m=3, 7=2; 2-zoned B, m= 3, m7—25 33 r 2-zoned Ci, m=6, 7=15; 3-zoned D,, #=6, t=10; ‘ t-zoned Ci, 7% =—6, 3 — 205 " 1-zoned C,, 7=6, 7=30; 1-zoned D,, 2=6, 7=30; 5 asymmetric De, m= 6, r= 60. The table following gives the posture-factor of every belt, as the product of five posture-numbers :— a ai a et The K-partitions of the R-gon. I2I 38.1, ABC, D, Dy JB, ABC, D, D, 2°2°15°10°60; II,= 36000 2°2°15°10°10; Iy=6,000 B., ABC, D, Dy | Bu, ABC, Di Di 2°2°30°10°60; Il,= 72000 2-220; 10: GOr tli £2,000 ‘B:, AB Cs D, Dz Bu, AB C; Di D, 2;2°30'10°60; II; = 72000 22720, TOLEO + ka — 52,000 §B., ABC, D. D; JBis, AB C, Dz Dy 22°15°00°30; Il;— 108000 2°2°15°60°60 ; Il; = 216,000 §Bs, ABC; D: Ds JBu, ABC, D; Dz 212720°00:320; 1I;= 216000 22-20, 00100); lhe 4.22,000 JB:, AB C; D, D; JBis, ABC; Dz Dz 2°2°30°60°30; IIs = 216000 2°2°30°60°60; Ils = 432,000 B:, AB Cy D; D, Bic, AB Cy Ds; IDS 2°2°15°30°10; II;= 18000 2°2°15°30°30; Ilg=54,000 Bs, AB C, D; D, Bu, AB C, D, Ds 2°2520°30:10; Il, 20000 2°2°30°30°30 ; Ih710 = 8,000 Bo, AB (Fe D; D, Bis, AB C; D; D; 22:20;30;10; Il>— 26000 2°2520:20:20 ; lihg— 103,000 810,000 1,380,000 By our subsequent changes of postures in 126 times distributed primes in the 120 times or 60 times permuted belts, we multiply every posture-factor in this table either by 126°120 or by 126°60. It follows that the sum S of all the partitions of the R-gon thus far constructed is 126°120°810000+ 126°60°1380000=S or S=22,680,000,000. We need not distress ourselves about the undercreased triangles in this hurly-burly of changing base-ties. The pretty primes are nimble and well drilled. There is no fear of damage to their wings in these thousands of millions of evolutions. 12. In Fig. 4 and 5 are seen different postures of the five distributed primes of the unpermuted ‘Bi. In 4, D2 uses its base-tie 13 (Fig. Z); in 5 it uses 12 of (7); the third triangle in (@) being hid in 4 and seen in 5,. while the second triangle in (@) is seen in 4, but hid in 5. 122 THE REv. THos. P. KIRKMAN oz In 4, D, uses its base-tie 14 of (6); in 5, °D, uses fen (4). In 4, C, uses 52 of (2); in 5, C, uses 54 of (2), Tis A and B have each made a revolution about its base-tie, used in 4. B,in 5, should have a small marginal triangle on the left at the fracture between 1 and I’. The above S results would be nearly all that is required, had we not to give an account of a third datum (Art. 6) which is the tape, Fig. 8, of three triangles and three rect angles that carry no marginal triangle. All the S partitions above made have each (Art. 3) 33 diagonals, z2., places to receive, after splitting the proper diagonals, the 6 units of the tape, which, after each distribu- tion of them, will have added six faces to the 34 in each of the S partitions. Instead of 5, aByds, we have now 33 places to name, to fix, and to split; in these we include the bases of the 23 marginal triangles in H, and in each of the 18 distributed belts ; for by splitting such bases we cannot introduce a new submarginal, as the tape carries no marginal triangle. 13. Our distribution of the tape can be exactly effected in gIR, different ways, which is the number of ‘permuta- tions of the 33-partitions of 6, without altering the order or the posture of any prime in the tape. Such prime, out of the tape, has two marginal triangles that in the tape are hid. This 3IR, is by theorem Q, Art. 5, the 33'4, which is also fae 7*~. COciicient Ol( 1-1) = =,1eu 38°37°30°35°34'33 = 2760681. [-2°3'4°56 This is the partition-factor of the tape. Its permutation and posture-factors are 20 and 8, so that we have 160 tapes to distribute, which are all one (Art. 6). The product of the three factors is 20°8'2760681 = 441708960= T. The K-partitions of the R-gon. 123 Since each of these T variations of our distributed tape will be combined with all the preceding S configurations, without changing anything in the latter, we shall obtain TS =441708960 X 22680000000, = 1001795921 2800000000 different partitions of our R-gon. The (19+15)-partitioned (39+23)-gon (Art. 3) has become by addition of the 6 faces and 9 summits of the tape, a 40-partitioned 71-gon, and we have formed T.S of these asymmetrical propyramidal 71-gons, no one of which is either the repetition or the reflected image of another. That is, if M be a million, we have formed of them 10M?+ 17959M?2+ 212800M. If the submarginals in H, standing on the dark edges 43 ana 21, be detached, the first is seen to be the prime.A, and the second is a square under four marginal triangles; and evidently neither of these could change the figure H by undercreasing another triangle. The marginal diagonal of that A should begin at 4. The submarginals on the dark edges 65 and 31 are monozones, each of which can, by undercreasing its other two triangles, place its zonal trace in three positions, and thus the two together can, still occupying 65 and 31, give to H 3°3=9 configurations. Further, these four submarginals can occupy the edges 43, 21,65, and 31 in 24 different ways, and by the same changes of two concealed marginal triangles turn H into 24°90 = 216 ,equivalent H’s, all of which have been virtually handled by us, and have each given us the same number T.S of 40-partitioned 71-gons. Thus we have constructed more than 2,000 millions of billions of them, namely, -216.T.S =2163M*+ 879189M?+964800M. 124 THE .REv. THos. P. KIRKMAN oz Each of them could be crowned upon its 23 marginal triangles by a different asymmetric propyramidal 23-ace, and the similar 216TS 23-aces could be by one entry registered as a small fraction of all the asymmetrical pro- pyramidal 23-aces, built by other belts on other irreducibles, that are to be found among the asymmetric summits of the 72-acral 63-edra. But since, in the solution of the problem of the Polyedra, no asymmetric summits are obtained by coronation, none of these 40-partitioned 71-gons are required in that solution. All asymmetric summits, the a@-aces, d-aces, &c., are given in that theory by their reciprocal a-gons, d-gons, &c., in the 63-acral 72-edra, which faces are obtained by their edges, constructed in vast numbers by crowning (or imagin- ing so crowned) penesolids with those edges. The only asymmetric reticulations of use in the study of Polyedra are quite elementary ones, by the zoned and zoneless repetition of which round a circle are formed the sym- metricals that alone are crowned, and give the z-zoned and z-ple summits. Those small ones are readily obtained by inspection of previous tables, in which they have been again and again used. This general problem of the k-partitions of the R-gon is outside the theory of Polyedra. It may yet find its use in analysis. 15. We have not above solved this general problem in terms of & and R; that, I fear, is impossible. But it will be seen that when with & and R are given the list of faces in the irreducible H, with the number of its non-marginal diagonals of which each is in a face that has more than one other non-marginal diagonal among its edges, and when the faces in R that have each two, and only two, non-mar- ginal diagonals for edges are exactly given, whether they carry or do not carry one or more marginal triangles, that The K-partctions of the R-gon. 125 the problem is completely solved. And it is evident that whatever be our data, as irreducible, as belt, and: as tape, it is impossible that 2 of our constructions can be alike, unless the primes are twice distributed in the same way, by the guidance of the same permutation of the diagonals that may be split to receive the primes. And this, fwzce, is clearly impossible, because the number M of those permuta- tions is exactly given by theorem Q, and none of them is twice used ; vzde Art. 7. It is impossible, also, that any one of our constructions should be the reflected image in any position of another ; because the reflected image of H has never been used. If in a partitioned R-gon there is no irreducible H, the R-gon is either a belt in which a tape is or is not distributed, orit is a pair of collateral submarginals in which a tape is or is not distributed, or it is a tape. The tape in every case can be dropped out, and the belt that remains can have all its possible configurations enumer- ated by the above method, after which the tape can be inserted again in every way possible. Thus all the possible different k-partitions can be deter- mined, both symmetric and asymmetric. But all the faces must be exactly given in the sense in which the faces of our 18 belts were given in any single belt of them. Hence it is quite correct (Art. 6) to say that no k-par- titioned R-gon can have more than one irreducible H, one belt, or one tape, although each of the three may have many equivalents, due to permutation and altered posture of the primes, and to the equal right, which like primes, CiC2C;.. . . D,D,D,... (Art. 5) have to every possible admission into the equivalent belts. The number of distinct equivalent belts that we have used in the pages preceding (Art. 11) is 9'120°x (h+Mk + .. +1])s)+9°60° x (Tio +Tit+ .. -+Tis); and these are all given with any one of them that may be first handled J 126 THE REv. THOS. P. KIRKMAN oz and called JB. The number of equivalent tapes used is 1'20'°8 = 160, 16. There are yet questions about primes, belts, and irreducibles that might be easily raised and disposed of in reasonable limits; but they are forbidden ground. They are too closely connected with the elementary theory of the polyedra, upon the teaching and learning of which, over 30 years ago, was imposed, by the highest scientific tribunal, a solemn and dire taboo. This is on record in p. 165, Vol. CLII., 1862, of the PAzlosophical Transactions, in the very last printed sentence of my complete Theory of the Polyedra, thus: “Thus we have demonstrated, in this second section, that the data of article xxxvi. are sufficient for the entire completion of the tables A, B, C, D(xxxi.) faa for faees and for summits. All that remains for the com- plete solution of our problem, of classification and enumera- tion of the P-edra Q-acra and of the P-acra Q-edra, is that we show how these data can be obtained and registered, without ambiguity or repetition. We shall consider first the reciprocals of the faces (d) (f) xxxvi., and the edges (¢) XK. The taboo is the sudden, loud, and long silence of that close. My first two sections are very summary statements of things to be distinguished and well-arranged in large groups» before handling in detail, and not quite easy to a reader of less than a De Morgan’s power. De Morgan read them easily, and very early, without a complaint of my obscurity, and, simply and only from the little just printed, so far shaped to himself what was coming, that he could write the letter now before me, dated “Adelaide Road, N.W., April 18, 1862,” expressing his satisfaction with it, with acute and kind remarks on the success which he foresaw. But then, I am here bound by candour to own that Professor De Morgan, The K-partitions of the R-gon. 127 whom I never saw three times, was not of the Council, nor even a Fellow, of the Royal Society. That letter is the only evidence, direct or indirect, that has yet reached me, that any competent judge, dead or alive, ever tried to read six pages of mine on this. subject, printed or in my MS. Of my definite teaching ad zuztzo, to which a student at the beginning would gladly turn, the first lessons are all in my third section, which is evident in the above quotation. Not a line of the third section was permitted to see the light in 1862. And i have been informed by the Secretary of the Royal Society that they have no intention to print more of my JZemozr. I have to confess that, a few years ago, I was tempted to a violation—a very little violation—of this dread taboo. Of that impiety I hope to die sufficiently penitent ; and I am confident that I am much too virtuous to repeat the sin. It is this—in Vol. XLIII., 1888-9, of the Proceedings of the Literary and Philosophical Society of Liverpool, there is, plentifully illustrated by plates containing 70 figures, an analysis and synthesis of four autopolar solids, three that have each six, and one that has nine, different edges. Of this taboo, for my very brief time, I make no complaint. My two first sections printed contain, not a production—our planet is yet but young—but a sufficient protection, of my theorems. And I am very far from denying that what was done with them was perfectly regular and in order, and simply what, when a like somewhat rare case recurs, will with equal propriety be done again. Wherefore I sing lustily, and shall sing to the end, the song they have taught me—Procul este, profani! Floreat Taboo! It is a genuine and an effectual taboo. For it has a droll side, which I leave to the reader who knows a little about the Grand Prize Question that, early in 1858, was 128 THE REv. THos. P. KIRKMAN ox published by the French Academy for their world com- petition in 1861, was in 1862 kept open for another year, and was apparently closed without result in 1864: wzde Comptes Rendus, 1858, Vol. XLVI., p. 301; and 1862, Vol) LV., p. ‘989. There is often amusement in a contrast, and sometimes in agreement. At the moment when the Academy were recording, in that page (989), their decision to repeat, for a fourth successive year, their offer to the circles of latitude of their gold and honour for one who should work out “en quelque point important la théorie géométrique des polyedres,” that and every other important point of the completed theory (completed before 1858) had, months agone, been presented in London, and had there been flung aside as a troublesome cumbrance, to be mentioned never more for that generation. That was perfectly in order, and has been as such accepted by all, for 33 years. The reader will none the less enjoy what he finds of droll in this taboo, if he has detected in that grave page (989) a half-hidden twinkle or two of harmless fun. It has also a serious side, that alone concerns me, in the duty which'my reverence for the Royal Society lays on me, of shunning in future all breach of their taboo. The bold student, who in another lifetime or two may have the valour to smash it, will see that my good fortune in these k-partitions of the R-gon is due, first, to my theorem Q, and next, to an old and very successful device. In the Polyedra the diagonals under my propyramidal edges and symmetrical summits were split for the insertion of tapes of pyramidal bases, with vertices downwards, which were all, by an easy routine of inspection of completed tables, after- wards registered as higher and still higher metapyramidals,. in groups ever larger and larger, each with its symmetry and deletes to be read in one entry, soon enormous, without ambiguity in class or number. The K-partitions of the R-gon. 129 Of anything like a reason for the taboo I know nothing fiatis not to be read in lines 4,5... of page 72 (the last but three of the paper) of the Liverpool volume above named. Of my paper in that volume I can give a copy to the mathematician who thinks it worth while to inform me of his wish and intention to read it. 130 Dr. THOMAS EWAN ox On the Osmotic Pressure of Solutions of finite Con- centration. By Thomas Ewan, B.Sc., Ph.D., 1851 Exhibition Scholar in the Owens College. (Recezved December 12th, 1893.) In 1885, van’t Hoff* showed that the equation PV=RT> which expresses the connection between the pressure, tem- perature, and volume of a perfect gas, is also true for the osmotic pressure of a solution. The behaviour of most solutions is not in strict accord with this equation, and my object in this note is to take into account certain factors which are omitted in the simple equation PV=RT, and so obtain an expression which shall approximate more nearly to the truth. The most important of these factors is the heat of dilu- tion, as van’t Hoff pointed out. He says, after showing that the equation PV=RT applies both to gases and solutions : “‘ Seulement la méme réserve nécessaire dans l’application en “cas des corps gazeux convient encore ici, et l’analogie qu’offrent ‘ces deux états de la matiére est telle que l’origine de la restriction “est. absolument la méme dans les deux cas. Aussitot que la * concentration, soit dans les gaz soit dans les corps dissous, est “telle que action mutuelle des particules n’est plus négligeable ‘on sait que dans le premier cas les deviations se font sentir et ‘de méme le raisonnement sur lequel se basent, pour la solution ‘‘les lois déduites ne peut plus étre accepté dans ces circonstances. *¢ Ajoutons que pour les solutions, un phénoméne facile a produire “‘trahit existence de l’action mutuelle des particules dissoutes ; “ces actions donnent lieu 4 la production de traveaux intérieurs ** dans l’acte de dilution, qui se manifestent dans leur équivalent * K. Svenska Vet: Ak: Handlingar 21. No. 17. 1885. The Osmotic Pressure of Solutions. 131 “thermique ; par consequent les lois exposées s’appliquent a des “solutions tellement diluées que la chaleur de dilution devient *“‘négligeable.” I have obtained the equations connecting osmotic pressure, temperature, volume, and heat of dilution of a solution, and on giving the value zero to the heat of dilution the equations become identical with van’t Hoff’s. Connection between Osmotic Pressure and Temperature. The osmotic pressure of a solution cannot be directly measured with any great accuracy, but it can be calculated from the vapour pressure of the solution or from its freezing point. The relation between the osmotic pressure and the vapour pressure of a solution was first given by van’t Hoff (loc. cit.) and afterwards reproduced by other authors, first by Gouy and Chaperon.* The form given by the latter is :— Mee RTlog”2 + Pp ) BR: where P= the osmotic pressure at T. K is a coefficient which depends on the contraction which takes place when the solution is diluted. M, is the molecular weight of the solvent (as gas). A, is the density of liquid solvent. p, and # the vapour pressures of the solvent and solution respectively, and R the gas constant for 1 gram molecule. The coefficient K is defined by Gouy and Chaperon+ by the following equation :— A being the density of the solution and zw the quantity of * Ann. Chim. Phys. (6), XIII, p. 120. 1888. + Ann. Chim. Phys. (6) XII., p. 384. 1887. 132 Dr. THOMAS EWAN oz solvent it contains to a given quantity of dissolved substance. 0 We may therefore consider K— as the increase of volume ce) which an infinitely large quantity of the solution would experience if M, grams (=1 gr. molecule) of the solvent were added to it. Call K caret and write the equation : Po, = RTiog™ a This equation is true for solutions of any concentration, and is only subject to the restriction that the vapour of the solvent over the solution may be considered as a perfect gas. The connection between the osmotic pressure and the freezing point of a solution may be obtained as follows. According to Kirchhoff’s}+ well-known equation connecting the heat of dilution of a solution with its vapour pressure we have: dQ RT? d, p se a ee! ee ] ye dw JM, av © p , : , l where J is the mechanical equivalent of heat, and = heat of dilution—is taken positive when heat is evolved on —the Ae dQ . ; dilution. - is very nearly independent of temperature. As a first approximation its variability may be neglected, and the equation integrated gives me : Ww The constant £ is of considerable importance, it is inde- pendent of temperature, but varies with the concentration of the solution. Its value may be obtained as follows :— The heat required to melt 1gr. molecule of the solid solvent at any temperature T (call it w,) is given by the expression RT? dp = cae l oO fl I ate. log =k+ Wr t+ Ann. Phys. Chem. 103, p. 177. 1858. The Osmotic Pressure of Solutions. 133 where 7, is the vapour pressure of the solid at T, and the other letters have their former signification. We also have Wy = W, — (¢, — )(T, — T), , where w,=latent heat of fusion for I gr. mol. solvent at T,. T,=melting point of solvent. ¢, and ¢,=capacity for heat for I gr. mol. liquid and solid solvent respectively. Call. (¢,-4)=c and we get RT? 4s, Dr dT Sn, Da Integrating this equation between the limits T, and T, and remembring that at T, W, +e(T,- T)=->- P1i=Po we get ti Gud he T-T, loss) Sar ont wr. oT : ih : after expanding log 7 and neglecting terms after the second this becomes: arte)... . © Now at the ale point of a solution its vapour pressure is the same as that of the solid solvent at the same temperature. Call the freezing point of the solution F. At F, therefore, 2=hy and therefore P1 De Po —lo Al = log”) = loo— ep, "71 ° At F, therefore, we have from equations (2) and (3), by putting T=F and writing the quantities on the right hand side of (2) equal to those on the right of (3) with negative “sign, and after making all reductions, b= a] omit aa -5("a* M, dQ t TRE oe o(—F-) - F dw ean) 134 Dr. THOMAS EWAN ox Substitute this value of £ in equation 2, and it becomes a er ee and finally equation (1), gives which value substituted in 5 gives —F cT/T,-F\?_. dQT-F Pv, =J] wT tae g(a =) -¥ ae In the following table the vapour pressures of some copper chloride solutions calculated from their freezing points, by means of equation (5), are compared with those found. The determinations used were made by Mr. Ormandy and myself.* Concentration : re To-F = (howisen) p. cal. p. found. Diff. 05 T°9Q05 "22 15086 15°033 "053 ala 4°12 78 14°788 | 14°706 082 "2 9°63 2.48 I4°I04 | 14'035 069 The differences are rather larger than the error in the measurements of the vapour pressures, but may be due to some extent to errors in the freezing points. Equation (5) also shows that v. Babo’s law, according to Po. which Ae independent of temperature is only true when d : : sa olncehi = =0 ae will evidently decrease with rising temperature when dy iS Positive, and vice versa. This conclusion has. already been reached by Dieterici,} though in a different way, * Chem. Soc. Fourn., 1892, p. 769. + Wied. Ann., 45, p. 207, 1892. ee The Osmotic Pressure of Solutions. 135 A number of interesting conclusions may be obtained from equation 6. (a) At the freezing point of a solution the external work done when solvent is added to it ina reversible way is inde- pendent of the nature of the dissolved substance, and it is the same for all solutions in the same solvent which have the same freezing point. The equation for the osmotic work becomes at F— ike oan 2 Fe ee ey T F This equation was first obtained by Arrhenius,* in 1892, ina slightly different form, and in a different way. (0) Again dividing equation 6 by T and bringing all the terms which are independent of T together into one con- stant, we get Po, JM, dQ 7 = const + Wee re It is easy to see from equations (1) and (2) that this may also be written dQ oe eM ek (0) Differentiating this with respect to T at constant volume (or concentration )— o(Pe) _ Rp or For any given solution v, may be regarded as indepen- dent of temperature, and we get (ae Rt _ Gonsiamibsty if tPA Pd hole Pex eet (8) Vo This result is quite analogous to the result obtained by * Zein Ph, Chen. 103, p. 92; 1892. 136 Dr. THOMAS EWAN ox Ramsay and Young,* that for a gas or liquid at constant volume the pressure may be represented by an equation of the form =4T —a where 6 and a are constants. From this follows (a) =b= constant. The same result follows from Van der Waals’ equation. Equation 8 shows that the osmotic pressure may decrease when the temperature rises, or (°") may be negative. The sign of = is the same as that of 2, which depends chiefly, as equation 4 shows, on the sign and magnitude of ae k : dQ . : will be — when = is +, and the sum of the terms contain- dQ ing ¢ and on in equation 4 is greater than the term con- taining w,. (c) If we put from equation 8 | bP Rio ae in equation 7. it becomes (dP dQ Pw, => TH oT), + JW ea from which oP\ P JM, dQ (a i ee ae This last equation becomes, when —=—=0. es ae roy APRS That is at constant volume the osmotic pressure of a solution is proportional to the absolute temperature when the heat of dilution of the solution is zero. This result was obtained by Van’t Hoff. (loc. czt. p. 11). * Phil. Mag. (5) 23, 435, 1887. The Osmotic Pressure of Solutions. 137 Connection between Osmotic Pressure and the Concentration of the Solutzon. _ The equations obtained so far are quite general, and apply to solutions of any substance in any solvent, and show how the freezing point, vapour pressure, heat of dilution and osmotic pressure of a solution are connected among . themselves and how they are affected by changes of tem- perature. The effect of a change of concentration (that is of the volume of solution which contains a gram molecule of the dissolved substance) has not yet been considered. Consider the simple case of a dissolved body consisting of only one kind of molecules, the nature of which is not affected by dilution. That is no dissociation of more complex molecules into more simple ones, and no chemical action is to take place on diluting the solution. In these circumstances suppose the solution consists of z gr. mols. of the solvent to I gr. mol. of the dissolved body. From equation 7 we have Pnv, = RTnk + J ie dw This equation is very similar to Van der Waals’ well-know equation connecting pressure, temperature, and volume of a gas or liquid. To make this clear, consider a quantity of a solution (or of a gas or vapour) in a cylinder which is closed by a piston, the pressure on which just balances the osmotic pressure of the solution (or the pressure of the gas). In case of a solution the piston must be permeable for the solvent, but not for the dissolved substance. The whole arrangement is kept at temperature T. Allow the piston to rise so that dw gram. of solvent is dw added to the solution, increasing its volume by K xR = UNG e) In the case of the gas let the volume simply increase by aV. 138 Dr. THOMAS EWAN oz The osmotic pressure of the solution being P, and the pres- sure of the gas f. Then in both cases, in order to keep the temperature constant, a certain amount of heat must be added, which is the equivalent of the external and internal work done by the solution (or oul in expanding. For the solution the “JQ where dQ is the heat evolved on adding dw er. on to the solution without doing external work. For the gas or vapour the work done is according to Van der Waals pdV + [dV The heat of dilution evidently represents the internal work done in the expansion. According to Van der Waals we have a RT Assuming that a similar equation is true for the sum of external and internal work done when a solution expands, we get dw RT Kdw ar qs or Ar? ‘Or P(V.-8)- JE AV -B) = BE. As R is the gas constant for 1 gr. mol. of substance, V must be taken also as the volume in which a gram ‘mol. is contained. Compare equation (10) with (9), viz. :— dQ Pav, — JZ Mn = RI nk | nv, is the volume by which the solution diminishes when n gr. mols. of solvent are withdrawn from it without changing its concentration, it may, therefore, be regarded as the The Osmotic Pressure of Solutions. 139 volume occupied in the solution by the solvent. V is the volume of z gr. mols. solvent+1 gr. mol. dissolved body. We have, accordingly, V—-nv,=8, where 0 is the volume in the solution of the dissolved substance (for 1 gram molecule). We get, therefore, (V — 4) =xv, We have also =D a: Lea Se putting these values into equation (10), it becomes Pn, —J Mn =f). dw By comparing this equation (which is obtained on the assumption that an equation of the same form as that of Van der Waals holds good for solutions) with (9), it is evident that if the equation connecting osmotic pressure, volume, and temperature of a solution is really of the same form as Van der Waals’ equation, we must have kn= +1. The sign will depend on the sign of &. The equation for the osmotic pressure of a solution may, then, be written P(V—8)=4RT+JnM,5° sia te ett) The meaning of RT having the —- sign is, that when the solution is diluted in a reversible way, the maximum of external (osmotic) work being done, there is still heat evolved by the system. This is a case which, so far as I know, never occurs with gases. The experimental material necessary to test the truth : ies. f i of the expression ogee is unfortunately not in existence. The only substance for which I have been able to find sufficient determinations of the freezing point and of the 14G Dr. THOMAS EWAN on heat of dilution is alcohol, and even in this case the determinations of the heat of dilution (Dupré and Page™*) , d are not sufficiently numerous to allow of the values of A: being calculated with any approach to accuracy. The following table contains, however, the numbers which I have obtained. ‘The determinations of the freezing point are Raoult’s.t Grams alcohol F _k I iS to 100 gr. H,O. highest. lowest. nN 15°19 20752 + '0065 += Oreg "0595 19°56 265°2 —°0765 —"O147 0765 24°70 2024 25a | 670454 "0967 29°15 260°2 or ye "0890 yee me bri 40°68 254'1 Tost "I241 "1591 Rr o2 248°7 "1985 "1545 "1996 59°66 244°8 "2405 "1365 "2335 7O'LS 240°9 ‘s5o6 "1788 ‘2744 I could find no interpolation formula which would represent Dupre and Page’s numbers for the heat of Q solution of alcohol. I, therefore, calculated a from three different curves. Owing to the small number of deter- minations (there were only five which I could use) the curves could not be drawn accurately, and, therefore, the values of = obtained do not agree with each other. The value of & has, therefore, been calculated by equation (4), d using the highest and lowest values of ~ found, and as will Aa: Pe, be seen the value of | lies between these limits, except for the most dilute solution. It may be remarked that in this case (viz., alcohol dissolved in water) & has the negative sign. * Phil. Trans., 1869, p. 501. + Ann. Chim. Phys. (5) 20, p. 220. 1880. The Osmotic Pressure of Solutions. I4I The equation (11) includes the equations which have 7 2 dQ_ already been given by Van’t Hoff for the case that 7 =o. In this case it is easy to see from (4) that & is positive, and therefore, equation (11) becomes P(V —2)=RT. If the solutions considered are dilute,—that is V large compared with ,—this may be written— Ev a Again (11) may be written in the form Pny,= {RT +I mo” dw or for a = 0, this becomes - = 4 And from (1) Po 7 p iy ee accordingly Po _ °F =: ” or for dilute solutions PoP = i Po nN which is the well-known equation of Raoult. It was also obtained in this form by Planck,* in 1887. The expression for the molecular depression of the freez- ing point of a solution follows from equation 4. Putting ie Bt DA Rae aon Yee i We TO Pee doe “M is called the molecular depression of the freezing point of a solution, when ¢= the depression produced by g grams * Wied. Ann. 32, p. 502. 1887. 142 The Osmotic Pressure of Solutions. dissolved substance in 100 gr. solvent, and M = molecular weight of the dissolved substance. 100 M Oe We have accordingly = ye and writing SS , : i " Pa sci, Ss ‘7 Gy MEMOIRS ano PROCEEDINGS. , , E 4 4 = 49 f a) “ 4 . ps Oe ee are te ee ee) seals a 5 > : = _ ree - eee | - gh) bela ‘ ices aig _ ul apn ft ret tee eo OO SMe Seg ieee LP eat Wilde Ae y 4% Series Vol Vi (Jtt4 bh) CLS, Ad ‘A a aan oe a Cen a pe MEMOIRS ano PROCEEDINGS, | OT PP ensayo PA | | | Fat un ap YI202. SANLTED [ “$7124 MANCHESTER LIT ano PHIL. SOC. | > . 3 ; =a} a a ; ; = ee Sal (Eo PG e = eh . a - = = gem a * ee Sa (oe - ae = ee ; ; sea : —— ~~ = ess c % Lee 4 ont) As ——— — — : : : 3 ; des |i = re 23S se - a : 2 . | - —_—— 3 = a 1 « = ( 2 : : ‘y - de an ey NY MANCHESTER LIT ann PHIL. SOC a2 MEMOIRS Ano PROCEED/NCS, Electro-Motive Force and Current Curves. I51 An Analysis of the Electro-Motive Force and Current Curves of a Wilde Alternator, under various con- ditions. By Julius Frith, Heginbottom Physical Scholar of the Owens College. Communicated by moet Schuster, Ph.D., F.R.S. (Recezcved March 20th, 1894.) These experiments were undertaken with a view to find- ing out how far the actual behaviour of an alternating- current dynamo follows the laws deduced for it from the theory of the alternator ; and if the electro-motive force and current deviate from the theoretical sine curve, how many terms of the Fourier’s expression should be taken into account. They have led to the conclusion that, for the case of an alternator whose armature contains iron, at least three terms of the Fourier’s expression must be considered, —- 152 . MR. JULIUS FRITH on - but most especially the third, the second being in most cases comparatively small. Description of Machine.—The Wilde alternator used con- sists of two crowns of cast-iron facing each other; from the internal surfaces of these crowns project the fixed field coils, six in number, on each side. These are bobbins ot wire wound on iron cores. The armature revolves between these, and consists of six similar bobbins on tubular iron cores, held ‘in position between two discs of brass which are keyed on to the shaft. The six coils of the armature are connected in series to the commutator. Hig: 1. D= Wilde alternator. R= Resistance, without self-induction. C= Ammeter. E= Electrometer. B= Two insulated brushes bearing on the ebonite disc which carries the copper contact piece. Principle of Intermittent Contact—On the end of the shaft beyond the commutator is keyed an ebonite disc ; in a slot cut in this, about ;; inch wide, a piece of copper is fixed, and turned down flush with the ebonite. On this disc bear two insulated brushes, side by side; it will be seen that at one instant in every revolution these two brushes are connected together, while remaining insulated from the rest of the machine. . If one of these brushes is connected to one pole of the dynamo, and wires are Electro-Motive Force and Current Curves. 13 taken’ from the other brush and the other pole: of the machine to an electrometer, the electrometer becomes connected to the poles of the dynamo, at one definite point in the revolution of the armature. This point is known, and can be altered by the arm which carries the two insu-. lated brushes moving round a fixed divided circle. Method of obtaining Current Curves.—This arrangement would give the E.M.F. curves at the terminals of the machine. To obtain the current curves, the fact is made use of that the current in a non-inductive resistance is in phase with, and proportional to, the E.M.F. at the terminals of the resistance. Therefore, if the electrometer can be connected through the same intermittent contact apparatus to the terminals of an ohmic resistance, the form of the current curve can be obtained. The Electrometer—The Electrometer feed was cena S| quadrant, and consisted of an aluminium needle suspended in the quadrant by a silver wire ; to the needle was attached the mirror and a damping vane dipping into oil. Each opposite pair of quadrants was connected to the ends of a water battery of 48, 96, or 144 cells; the middle of the battery was always connected to the frame of the instrument. The wires from the intermittent contact apparatus were connected respectively to the frame, and to the needle by means of the silver wire suspension. | With this arrangement of the electrometer, the deflection is directly proportional to the difference of potential between the needle and the frame. | The constant of the instrument was mand by means of a battery of Clark’s cells. With 24 water cells 1 scale div.='87 valte > 48 ” ” = A359 » f2 = ” ="29 5 Magnetization Curves of Field Magnets —To obtain the magnetization curve of the iron of the field magnets, a flat coil of 12 turns of wire was wound, and arranged so that it 154 Mr. JULIUS FRITH ox could be suddenly withdrawn from between the field coils and the armature, when the latter was at rest with its coils in a line with the field coils. The ends of this coil were connected to a ballistic galvanometer and the kick observed for different values of the magnetizing current on suddenly withdrawing the coil. Curve I. (Plate III.) represents the results of these experiments. E.M.F, Curves Varying Speed and Exciting Current.— Curve II. (Plate IV.) shows the E.M.F. curves at the terminals of the dynamo on open circuit, (1) keeping the speed constant and varying the exciting current, and (2) keeping the exciting current constant and varying the speed of the dynamo. Current Curves.—Curve III. (Plate V.) shows the effect of taking current from the machine, the resistance coils through which the current passed being nearly without self-induc- tion ; the lag recorded being due to the self-induction of the armature. If E, the impressed E.M.F., be of the form E,sinpé, then in a circuit of self-induction L and resistance R the current is given by Tyee a where sec Lp. ana = R From the observations shown in Plate V. the equations to the curves are found by the method of least squares to be: For the E.M.F. curve, R=co = — 229sin(0 — 2°) — 16°4sin(26 — 3°) + 36sin(30 + 1°). For R=21'5, C= — 8:7sin(@ — 20°) — ‘14sin(20 + 84°) + °34sin(36 + 33°). For R=9, C= — 16-4sin(@ — 31°) — ‘4sin(26 — 61°) — r°6sin(30 + 51°) For R=; C= —20'7sin(0 — 64°) — *7sin(20 — 20°) — 1'9sin(36 + 12 ). Electro-Motive Force and Current Curves. 155 The lag in the first term of last three curves is 22°, 34°, and 66° respectively. From the formula Tan 06=L# R where 0 = angle of lag p=2rn mu =alternations per second L=self-induction R =told resistance in circuit the self-induction may be calculated as follows :— tan22°= "4 “ L="4 x 21°93 _ ‘0174 secohms. 2°75 tan34 = °67 wou "67 x 943 2°6°75 ia F tano6" = 2°25 jodi = 225.4 2°43. 116 27775 - These show that the self-induction of the armature decreased with the increase of magnetization of the iron in the cores. This agrees with the results of measurements of the self-induction carried on in the ordinary way with the machine at rest. These give, with the fields not excited, L=:02 secohms ; with the fields excited, L=:013 secohms. On taking power from the machine the irregularities die out of the curve, and it becomes first nearly straight from the maximum in one direction to the maximum in the other, and then gradually approaches the sine curve. Electrolyte.—Next a copper-plating bath, consisting of two plates of copper 30cms. x 4ocms., placed Ioocms. apart in an acid solution of copper sulphate, was put in circuit, and a current of 10 amperes was passed through. 156 . Mr. JuLius FritH on Se D : ae Oc =e iiig. 2. I Intermittent contact. ' ‘E Electrometer.. D Alternator. B Copper bath. C Ammeter. R Resistance without self-induction. A resistance of 4 ohms was placed in series with the bath, and the curve taken at the terminals of the dynamo. On Plate VI. are also drawn the curve for 10 amperes passed through resistance, and the curve of the machine on open circuit. Next an arc lamp was substituted. for the copper bath. The potential at the terminals of the lamp was kept constant and equal to 40 volts. Fig. 3.- Electro-Motive Force and Current Curves. us 7) The Electrometer can be alternatively connected to the terminals of the lamp, for the E.M.F. curve, or to the terminals of a resistance of 5 ohms for the current curves. This is done by the key &. The Lamp used was a simple hand regulating one, and therefore had no series coils and no self-induction. Two curves were taken at the poles of the alternator, one with the lamp direct to the dynamo, the other with a resistance of 1 ohm in series. On the same sheet, the current and E.M.F. curves of the lamp with 5 ohms in series are shown. Fig. 4. Blondil—lt is interesting to compare these curves with some obtained by the French electrician Blondil—7he Electrician, December 15, 1893. Some of these curves almost exactly agree with the ones drawn on Plate VII. Surging of Lines——Plate VIII. shows the surging of the lines of force of the field magnets. This was measured by fitting over one of the field coils a light wooden frame which carried 155 wires stretched radially across the face of the pole, in the air gap between the fields and the armature. The surging of the lines past these induced in them an E.M.F., which was measured in the usual way by the electro- meter through the intermittent contact apparatus. The induced _E.M.F. is proportional to the rate of motion of the magnetic field. This motion must in a large degree account for the deviation of the electro-motive force and current curves from simple sine curves, L 158 Mr. THOMAS HICK on ve the primary structure of The Stem of Calamites. By Thomas Hick, B.A., B.Sc., Assistant Lecturer in Botany, Owens betece Manchester. Communi- cated by F. E. Weiss, B.Sc., Professor of Botany in the Owens College. , (Received February 20th, 1894.) Though a great deal has been written on the anatomy of the Stem of Calamztes, the references in the literature to its primary structure, that is, the structure previous to the commencement of secondary thickening, are extremely few. Binney, who was struck with the fact that the size of the specimens met with varies within wide limits, refers! to small stems which were not more than 3/50 of an inch in diameter, but even in these secondary thickening had begun, for his description shows that a zone of secondary xylem, at least two elements in thickness, had already been developed. Solms-Laubach remarks that “almost all the petrified specimens which have been examined show the presence of secondary wood,” and gives no description of the stage where such wood is absent; while Schenk? candidly confesses that in all the sections seen by him the formation of secondary wood had already begun. It is only in Williamson’s fine series of JZemozrs that any account of the early condition of a Calamitean stem is to be found, and this is given in the ninth J/emozr, which was published in 1878. In that MZemozr* Williamson describes several stems which were still in an early stage of development, but in 1 Observations on The Structure of Fossil Plants, Part I., p. 16. Lalaeon- tographical Society, 1868. 2 Fossil Botany, p. 295. 3 Die fosstlen Phlanzenreste, p. 107. 4 Philosophical Transactions, 1878, p. 322. The Primary Structure of the Stem of Calamites. 159 all of which the pith, carinal canals, and cortex were distinguishable. One of them was “not more than 0'033 inch in diameter,” and the stage it had reached may be inferred from the following description which Williamson gives of it :— “The medullary cells are here unruptured, the medullary fistular cavity having as yet no existence. Nine longitudinal internodal canals are seen, and these form the only recognisable line of demarcation between the pith and the bark [cortex]. There is little difference between the cells of these two structures.” ? Another, which was slightly more advanced, he describes thus :— “We still discover the bark [cortex], the internodal canals, again nine in number, and the medullary parenchyma; but the bark [cortex] in this example is a thick layer of parenchyma of coarser tissue than that composing the medulla, and the latter now displays a central fissure, which obviously indicates the commence- ment of the medullary fistular cavity. We have but still very slight indications of the formation of woody wedges external to each or the internodal canals.” ” From these descriptions and the figures which accom- pany them, these two stems seem to have been so young that even the primary structure had not received its full development. Summarising the facts obtained from a study of the whole of the specimens Williamson describes this early condition and some of the subsequent changes in the following terms :— “One thing is clear, viz., that the bark [cortex] as we see it in Figs. 8, 10 and 13, is a primitive generalised parenchyma ; but as the stems become arborescent this generalised tissue developed within its interior the thick layer of prosenchyma, which resembles so closely the cork layer of living phanerogams.” ® PTAC. Citas) Pa gees = 2btG.g Ds 322: 3 Jbid., p. 324. The figures referred to are on Plates 19 and 20 of the Memoir. +: : 160 Mr. THOMAS HICK ox Since the publication of this account of the primary condition of the stem of Calamztes, I cannot discover that anything has been added to it, a result which is doubtless due to the fact that stems of a suitable age and in a proper state of preservation are so rarely met with. In now attempting to carry our knowledge somewhat in advance of this, I shall base my statements upon an exquisite series of sections prepared some time ago by Mr. James Binns, of Halifax. One of the specimens is a transverse section of a young stem, which is represented on Plate [X., Fig. 1. It is roughly elliptical in shape, but broader at one end than the other. The length of the major axis is +’; inch, and the breadth of the broader end, at the margin of the pith cavity, is 5 inch. Except the central part of the pith, which has disappeared, all the tissues are preserved, and that in a degree of per- fection and clearness which is rarely met with in the petrifactions of Carboniferous plants. The peripheral part of the pith forms a zone of parenchyma, a, on the inside of the primary vascular bundles, which has a breadth of s$5 inch. If complete, the pith would form a nearly elliptical mass of tissue, the longer diameter of which would be 34 inch and the shorter 335 inch. The periphery of the section shows a number of irregular projections, which indicate that the stem was not smooth but marked by longitudinal ridges and bands. Some of them were merely narrow wing-like extensions, 0, but others, c, were broader. The latter were not rounded, however, but flattened, and had more or less angular edges. How far accident has entered into the formation of these ridges it is impossible to say, but the normal appearance of the tissue beneath them proves that, to a large extent, if not wholly, they are natural. Surrounding the zone of pith are the primary vascular bundles, @, Fig. 1, which are here 16 innumber, Like those of Equtsetum, they are imperfect, the xylem consisting of The Primary Structure of the Stem of Calamites. 161 little more than the carinal canal, formed by the breaking down of the initial strand of vessels. A striking and remarkable feature of some of these canals is the presence, at the margin, of projecting elements, which I have no hesitation in interpreting as the remnants of the vessels, v. The presence of these elements gives the canals an appearance which is perfectly identical with that of the homologous canals of Eguzsetum, but the lateral xylem elements found in the latter plant are not distinguishable. External to each canal is a mass of small elements, Z, which, from their position and their distinct character when compared with the ground tissue on either hand and between the canals, must be regarded as the phloem of the primary bundles. It is true that some of the histological characters of phloem cannot be recognised in these groups of elements, but this is most probably due to the fact that in the process of fossilisation, their contents have all disappeared. If, however, they be compared with the phloem of Lguzsetum after the protoplasm, &c., has been removed, it will be found that they are in close agreement therewith not only in position, but in the size and general arrangement of the constituent elements. If a pericycle ever existed outside this phloem, it is no longer recognisable. Still moving outward, we next come to a sharply defined line, s, which is traceable nearly all round the stem, just outside the-ring of primary vascular bundles. The line is slightly undulated, the parts opposite the bundles being convex, and those opposite the medullary rays concave, outwardly. I regard it as marking the boundary between the stele, or vascular-bundle cylinder, and the cortex. In most of my preparations, it is apparently a simple but thick black “line,” but in the one under description, and one or two others, there are vague indications of a single layer of narrow cells in place of the “line” at some points. If this could be proved to be the normal structure, few would 162 Mr. THOMAS HICK ox hesitate to call the layer of narrow cells the endodermis, and to regard the axis as monostelic. But at present this has not been done. Nevertheless there are good reasons for regarding this “line” as the boundary between the stele and the cortex. In the first place, it is strongly suggested by the typical species of Eguzsetum, eg., E. arvense and E. maximum, which, it will be allowed, are something more than analogous. In the second it is supported by the mode of origin and the development of the secondary xylem, as will be shown later. Outside the “line” just dealt with, we have the cortical tissues, which are here seen to present a considerable amount of differentiation. At the first glance, indeed, it is obvious that the cortex of this specimen has a remarkably complex structure. It is made up of two layers or zones, an outer and an inner, 9, z, between which runs a dividing line, which is undulated and roughly parallel to the surface ‘of the stem. The inner zone is the broader of the two—having a breadth of 745 inch—and is generally much better preserved. In the middle of it the elements, though of ‘different sizes, are for the most part large and angular, and in shape and arrangement are not unlike the xylem elements of the vascular bundle of a fern. But the walls are not specially thickened, and the cavities frequently contain black car- bonaceous masses, #. Whether these represent special substances, such as resin, tannin, or latex, or merely an unusual accumulation of ordinary cell contents, it is impossible to say. Between the black masses, which are usually eccentric, and the distant cell wall, a faint concentric line is often discernible, recalling the appearance of the primordial utricle of recent plants. This, and the whole appearance of the zone, seem to show that the contents of these elements were introduced in the living state, and are not mere infiltrations into empty cavities during the fossil- The Primary Structure of the Stem of Calamites. 163 ising process. On the inner, and more especially on the outer, side of these larger and more central elements, which are practically continuous all round the stem, are smaller elements of a different character, z. In transverse section, they have a more circular outline, and there are distinct evidences of thickening deposits having been laid down upon the original walls. On the outer side of the zone under description they form two or three layers, and at some points they are found penetrating in triangular masses between the larger elements of the middle. On the inner side they chiefly fill up the angles between the larger middle elements, so that the entire zone has a tolerably uniform width, with more or less even and uniform margins. The outer zone of the cortex, 0,is seldom well preserved, but it appears to have been composed of a thin-walled tissue, in which thicker-walled elements were imbedded. The latter have very thick walls, with clear rounded lumina, and are somewhat irregularly distributed. A curious point is, that they vary much in size. At the periphery of the section is the epidermis, but in this, as in most specimens, its structure is for the most part obliterated. At a few isolated points, however, we can make out that it originally consisted of a single layer of cells. My efforts to obtain longitudinal sections of this type of stem in its primary condition have not yet been as suc- cessful as could be desired. Numerous fragments have been met with, but no one large and complete enough to give a connected view of the primary tissues in their longi- tudinal aspect. Nevertheless, by putting together the items of information picked up from a large number of these fragments, we may obtain a fairly reliable idea of the longitudinal structure, at least in its main outlines. The following description is based upon knowledge obtained in this way. | | 164 Mr. THOMAS HICK on The pith, so far as it is preserved, is made up of thin- walled cells, elongated longitudinally, which are usually narrower at the periphery than towards the centre. In some cases a few larger cells, with carbonaceous contents, are intermingled with the smaller peripheral ones, but these are not present in the transverse sections figured. They have a close resemblance to the cells which occupy the middle of the inner zone of the cortex. The pith cells at the nodes are rounded, and may or may not contain accumulations of carbonaceous matter. The vascular elements which cling to the sides of the carinal canals are not all of one kind. Some of them are clearly annular, and others are spiral; but occasionally reticulated ones are also present, a state of things which may occur in Eguzsetum. Longitudinal views of the phloem are much rarer and still more fragmentary than those through the xylem, and at present I can only say that the phloem elements apieer to be narrow elongated structures. Coming to the cortex, it may be said with some degree of confidence that the larger elements of the inner zone, though often considerably elongated, are nevertheless cellular. They are, in fact, several times as long as broad, they have oblique or square ends, and stand in vertical rows. But there are no signs of thickening or sculpturing of the walls. The carbonaceous contents are usually retracted from the side walls, and at the ends sometimes take the expanded, trumpet-like form, characteristic of the contents of some sieve-tubes. This, and the arrangement in vertical rows, suggests that they formed conducting channels, but the nature of the conducted materials cannot at present be determined. With respect to the outer zone of the rita little has been made out in the longitudinal view beyond the fact that the thick-walled elements seem to be more or less fibrous The Primary Structure of the Stem of Calamites. 165 in form, and probably belong to the category of scleren- chymatous fibres. From this account of the primary structure of this type of Calamitean stem, it will be seen that the specimens now described differ in many respects from those described and figured by Williamson in 1878. As already stated, the tissues are much more differentiated, and that in nearly every part of the stem. In the pith, we have the elements at the periphery smaller than those in the centre, and the occasional occurrence of larger elements with black con- tents, may be indicative of other differences. In the stele, ‘we have phloem strands accompanying the carinal canals, to the walls of which the torn vessels still adhere, and there is a sharp distinction between the stele and the cortical tissues. The latter again are distinguishable into two zones, and within each there are considerable histological differences, which add to the complexity of the whole, and make it a very different structure from the “primitive generalised parenchyma” of Williamson’s specimens. My own impression is, that these differences are due to the fact that the latter appear to be in an earlier stage of develop- ment than those under treatment, which seem to present the primary structure fully matured and ready for the initiation of secondary thickening. It is possible, however, that the two sets of specimens do not belong to the same type of Calamztes, and that this is the explanation of the want of agreement between them. | An interesting question in connection with the fossil plants of the Coal Measures is the degree of correspondence between the size of a stem and the extent of the develop- ment.it has undergone. The preparations under considera- tion appear to throw a little light upon it. The transverse section of the stem which has been described in detail, measures, as already. stated, 7; inch by sy inch. But my collection includes others smaller than this, in which 166 Mr. THOMAS HICK on practically the same structure obtains, and that in equal perfection. Of these one measures +; inch by 35 inch, and another, which is circular, has a diameter which is not more than the latter figure. Thus in stems which range in dia meter from =5 to 7s inch, we have the same differentiation into stele and cortex, and within these an equal complexity of structure. Another subsidiary point of some interest receives fresh elucidation from these specimens, viz., the nature of the lacunae, which are almost constantly present in the primary vascular bundles of Calamztes. Most palzobotanists now accept the interpretation of Solms-Laubach that “in the lacunae, or the tissue that fills them, we are dealing with the tracheal initial strand of the primary bundle”! This interpretation, however, has hitherto been based entirely upon transverse sections, the author quoted pointing out, that longitudinal sections bearing upon the point are precarious, and “are of value only when the sculpture of the walls is preserved, which is seldom the case.” No such sections appear to have been described hitherto, and hence it seems worth while to note that the sections here dealt with are exactly of the kind required, and fully confirm the interpretation of the lacunae suggested by Solms-Laubach. These matters disposed of, we may now turn to one of much greater importance, viz., the place of origin of the secondary thickening and the first changes brought about by the same. Fig. 2 represents a stem of the same type of Calamites as those previously described, but it is much older, and has developed a zone of secondary xylem which is nearly 35 inch in breadth. It was partially described by Mr. W. Cash and myself many years ago? and has its tissues much more 1 Fossil Botany, p. 298. ; . Proceedings of the Yorkshire Geological and Polytechnic Society, 1883. The Primary Structure of the Stent of Calamites. 167 complete and in a much finer state of preservation than any other section I have yet come across. In it will be seen the pith, a, surrounded by the carinal canals, d, seven- teen in number, arranged as in the younger specimens. But outside each carinal canal is a wedge-shaped mass of secondary xylem, x, and between these masses are the somewhat broad medullary rays,7. It will be noticed that, as has been pointed out by several observers,’ the first formed elements of the secondary xylem stand near or abut upon the carinal canals, and the rest are developed centrifugally in radiating rows. Hence, as the young stems described in this paper show phloem strands in immediate proximity to the canals,? it seems a warrantable inference that the secondary thickening begins in the position usual for open collateral bundles, ze., between the phloem and the xylem. ? As the development of the secondary xylem would necessitate the displacement of the phloem and the ‘line’ of demarcation between the stele and the cortex, one naturally looks for traces of these in the older stems; but so far I have not been able to detect them. It is otherwise, however, with the cortical tissues. At z we have the inner cortical zone of the older stem, and it needs little examina- tion to see that it is identical with that of the primary stem. (Fig. 1, z.) The arrangement and general appearance of the elements are the same in both cases, and the same may be said of their histological structure. The breadth in the older stem, as in the younger, is ;45 inch, so that there has been no growth in the radial direction. Obviously, however, there must have been growth in the tangential direction, as the layer still completely encircles the stem, though not quite so uninterruptedly as in the earlier stage. 1 See especially Binney /oc. cz#., and Williamson, Philosophical Trans., 1871. * Ante, Pi 168 Mr. THOMAS HICK oz In the outer cortical zone, we again recognise the thin- walled tissue at 0, but its bulk is still small. In it area few lacunae, Z, but whether they are natural air-canals, or due to accidental rupture, there is nothing to show. The thick- walled elements, on the other hand, have increased con- siderably, and now form a dense thick hypodermal layer, s, which appears to have been sclerenchymatous. The elements of this layer are not arranged in radiating series, nor are they grouped in triangular bundles, thus differing from the corresponding tissues described by Williamson." On the whole then, it would seem that the structure of the cortex, as seen in the primary stem of this type of Ca/a- mutes, retains its characteristic features for some time after secondary thickening has set in, the chief modification, apart from the doubtful lacunae, being the increase in the mechanical tissue. | | Reviewing the facts as set forth in what has gone before, botanists will probably be most struck with the remarkable features of the tissue which makes up the inner zone of the cortex. From what I have seen of it, in many preparations, I am convinced that it is an important tissue both in a morphological and a physiological sense, though I cannot as yet specify in what its importance consists. It is not confined entirely to the stem, but is found also in the leaves, where it forms a conspicuous layer, which extends from one edge to the other, and runs from base to apex beneath the epidermis on the convex side. Finally, an identical layer is present in a similar position in the sterile bracts of Calamostachys Binneyana, as 1 have shown else- where,” awakening the suspicion that in the type of Calamites here considered we have the plant that bore Calamostachys Binneyana as its fruit-spike. 1 Phil. Trans., 1878, p. 3243; Lbzd, 1881, p. 465. ® Proceedings of the Yorkshire Geological and Polytechnic Soctety, 1893, p. 287- The Primary Structure of the Stem of Calamites. 169 Not the least perplexing fact about this tissue is that it appears to belong to the cortex, as has been pointed out in the earlier portion of this communication. Insome respects it is not unlike a well-developed phloem tissue, and when the older stem was first described by Mr, Cash and myself in 1884, we called attention to this, and tentatively suggested that it might be the phloem of the secondary bundles. Now however that we know it to be a constituent of the cortex of the primary stem, and find it as a broad longi- tudinal band in the leaves, it is clear that this view of its nature can no longer be put forward. In 1876 M, Renault described’ a form of Calamiztes, which he named Arthropztys /ineata, in which he found a layer of cortical tissue in which were elements that somewhat resembled those of the layer before us, so far as one can judge in the absence of figures, He speaks of it asa cellular layer, enclosing groups of resin canals, placed in front of the secondary xylem bundles. Solms-Laubach questions this interpretation,” but whether true or not for Renault’s specimens, it is scarcely applicable to the case before us, For here we have a zone of ézssue which is practically uniform and continuous round the whole stem, and not merely a number of isolated “ groups” of elements, distinct from, but imbedded in, an ordinary cellular layer. Moreover, the elements of this tissue are themselves mainly cellular, and the carbonaceous masses they contain are apparently derived from the normal cell contents. They might be secretory reservoirs of some kind, but their arrangement as a special tissue is not in favour of this view, and their longitudinal course, both in the stem and the leaves, is more suggestive of the function of con- duction, as already stated. Turning to Eguzsetum, as the living representative of 1 Comptes rendus, Vol. 83 (1876) p, 574. 2 Fossil Botany, p. 301, - 170. The Primary Structure of the Stem of Calamites. Calamites, we find little in its structure to elucidate the nature of the tissue under consideration. Strasburger mentions’ that in LAguzsetum maximum tannin bearing elements are scattered in the ground tissue, both of the stele and the cortex, and that they are elongated structures arranged in longitudinal series. These, however, are scarcely comparable with the elements of the inner cortical zone of Calamites, though they are not unlike the groups of cells with black contents, sometimes seen in the pith. For the present, then, we may leave the interpretation of this peculiar tissue an open question, in the hope that further specimens may soon be forthcoming to throw additional light upon it. EXPLANATION OF THE FIGURES. fig. 1. Transverse section of the Stem of a young Ca/lamites. Peripheral portion of the pith. . Narrow and broad ridges respectively at the surface of the stem. @d. Carinal canals of primary vascular bundles. v. Carinal canals with projecting vascular elements. P 5 SSS ~ Phloem of primary vascular bundles. Boundary between the stele and the cortex. z, o. Inner and outer zones respectively of the cortex. m. Elements of inner cortical zone with black contents. m. Smaller elements of inner cortical zone. Fig. 2. Transverse section of Stem of Calamztes with secondary thickening. i, Pith. d@ Carinal canals of primary vascular bundles. x. Secondary xylem. | vy. Medullary rays. z, 0. Inner and outer cortical zones respectively. 7. lLacunae in the outer cortical zone. s. Sclerenchyma of do. do. 1 Histologische Bettrdge, Heft IIl., p. 433. 4 Series Vol W. Plate TK Catamites. MEMOIRS AND PROCEEDINGS, MANCHESTER LIT.AND PHIL.SOC. ' -. . ‘ . ‘ a . _) ’ <-> 7 r . . - on 7 ‘ 2 , ‘ . . 4 ‘ . . * : : RS ’ : a ‘ 3 od , f ‘ . = ‘ 4 \; - 7 = ‘ . ‘ ‘ * ‘ -_ PROCEEDINGS. 171 Annual General Meeting, April 17th, 1894. Professor ARTHUR SCHUSTER, PH.D., F.R.S., F.R:A.S., President, in the Chair. The following gentlemen were elected Honorary Members :— Eig eAun APPELL, Paris; J. W. L. GLAISHER;, D.Sc., F:R:S:, Cambridge; Prof. L. KOn1IGSBERGER, Leipsic; Prof. M. Sopuus Lig, Copenhagen ; Prof. A. Gouy, Paris; Prof. E. Warsur«, Freiburg ; Dr. G. NEUMEvER, Director of the See Warte, Hamburg ; Prof. Eee oToONne, F.R.S., Radcliffe Observer, Oxford ; Prof. H. A. Row.tanp, For. Mem. R.S., Baltimore; OLIVER HEAVISIDE, F.R.S., Paignton, Devon; Prof. W. OstwaLp, Leipsic; A. G. VERNON Harcourt, F.R.S., Oxford ; Dr. H. Desus, F.R.S., Cassel ; Prof. T. E. THorps, F.R.S., London ; Prof. Prerrer, Jena; Dr. JoHN Murray, Edinburgh ; Prof. Sir Wma. Turner, F.R.S., Edinburgh ; Prof. A. WEISMANN, Freiburg ; Prof. SipNey Vings, F.R.S., Oxford ; Prof H. M. Warp, F.R.S, Cooper’s Hill; Prof. C. M. GuLpBere, Christiania; Pro. P. Waacer, Christiania; Prof. J. S. Burpon SANDERSON, F.R.S., Oxford. The Annual Report of the Council was presented and .amended, and it was moved by Mr. J. B. MILLAR, M.E., seconded by Mr. S. C. Trapp, and resolved :—“ That the Annual Report as amended be adopted, and printed in the Society’s Memozrs and Proceedings.” It was moved by Mr. W. E. HOYLE, M.A., seconded by Mr. S. C. TRAPP, and resolved :—“ That the system of electing Associates of the Sections be continued during the ensuing session.” The following gentlemen were elected Officers of the Society and members of the Council for the ensuing year :— President—HENRY WILDE, F.R.S. Vice-Prestdents EDWARD SCHUNCK, Ph.D., F.R.S., 172 PROCEEDINGS. F.C.S.; ‘(OSBORNE REYNOLDS, M.A., LL.D, ERS) eae ARTHUR SCHUSTER, Ph.D. F.RS, PR ASe ee CosMoO MELVILL, M.A., F.L.S, Secretaries—FREDERICK JAMES FARADAY, F.LS., | F.S.S.; REGINALD F, GWYTHER, M,A. Treasurer—CHARLES BAILEY, F.L.S. Librarian.— FRANCIS NICHOLSON, F.Z.S. Other Members of the Council—HAROLD B, DIXON, M.A., F.R.S.; ALEXANDER HODGKINSON, MUBe oc JAMES BOTTOMLEY, B.A., D.Sc, F.C.S.>~ All@enaaem JOSEPH THOMPSON ; FRANCIS JONES, F.R.S, Ed., F.C. ; W, 2. Hoven wi. Ordinary Meeting, April 17th, 1894. Professor ARTHUS SCHUSTER, Ph.D.,. F.R.S;5 eRe President, in the Chair. The thanks of the members were voted to the donors of the books upon the table. | Professor DIXON, M;A., F.R.S., read a paper “On the Instantaneous Pressures produced in the Explosion-Wave.” PROCEEDINGS. 173 [Microscopical and Natural Hzstory Sectzon.| Annual Meeting, 9th April, 1894. Mr. R. E. CUNLIFFE, President of the Section, in the Chair. Mr. BROADBENT described some observations on fission in the Infusoria. Mr. OLDHAM exhibited some bird snares used by the inhabitants of the island of St. Kilda for capturing the Fulmar Petrels and the Puffins. Also the oil vomited by the Fulmar when captured, and the receptacle made from the crop or stomach of the guillemot, in which the islanders collect the oil. This oil was formerly used as a specific for rheumatism and for dipping sheep. Mr. ALLEN showed a specimen of alloy made from 60 per cent of copper and 4o per cent of manganese, and described a number of observations on the injurious effects of noxious vapours brought by prevalent winds from manufacturing towns. The Annual Report of the Council, and the Treasurer’s Financial Statement, were presented and adopted. The following gentlemen were elected Officers and Council for the ensuing Session :— President.—JOHN BOYD. Vice-Prestdents—PETER CAMERON, F.E.S.; ROBERT BELLIS: CUNLIFFE ; JAMES CosMO MELVILL, M.A., F.LS. - Treasurer—MARK STIRRUP, F.G.S. Secretary THEODORE SINGTON. Council—_CHARLES BAILEY, F.L.S.; GEORGE HARRY BROADBENT; .M.R-C.S.; °. HERBERT C. ,CHADWICK, F.R.M.S.; ROBERT DUKINFIELD DARBISHIRE, B.A., m.G.s, F.S.A..; ALEXANDER. HODGKINSON, M.B., B.Sc. ; HENRY HYDE; FRANCIS NICHOLSON, F.Z.S.; .THOMAS ROGERS. sy M 174. Mr. H. B. DIXON AND MR. J. C. CAIN on the On the instantaneous pressures produced in the Explosion-Wave. By H. B. Dixon, F.R.S., Pro- fessor of Chemistry, and J. C. Cain, B.Sc., 1851 Exhibition Scholar in the Owens College. (Rececved May 22nd, 1894). The problem of directly measuring the pressure pro- duced in the ‘ explosion-wave’ of a mixture of gases is one of great difficulty. The movement of the wave is so rapid, and the zone of high temperature so thin, that the high pressure over any given area lasts an exceedingly short time. Nevertheless the problem is one of great importance for the elucidation of the phenomena of explosion. To take one instance: If the pressure produced in the explosion-wave could be accurately measured it would decide between the two theories of gaseous explosions that have been put forward. According to M. Berthelot* the velocity of explosion is equal to the mean rate of translation of the products of combustion heated at constant pressure. In the alternative hypothesist the velocity of explosion is equal to the velocity of sound in the burnt and burning gas at a temperature double that due to the combustion of the gases at constant volume. The calculated temperatures and pressures of the explosion-wave are very different in the two hypotheses. For example, in the explosion of cyanogen with its own volume of oxygen the pressure in the wave is calculated to be 35 atmospheres according to the first view; it is calculated to be 117 atmospheres according to the other. In the elaborate investigation made by M. Berthelot, in conjunction with M. Vieille, on - the pressures produced in the explosion of gases, the *Sur la force des matiéres explosives. + H. B. Dixon, Phil. Trans., vol. 184., p. 134 (1893). Instantaneous Pressures in the Explosion-Wave. 175 pressure produced in this reaction is given as 25 atmospheres, a number more in accordance with the first than the second hypothesis. But the measurements made by MM. Berthelot and Vieille* do not, we think, apply to the pressures in the wave. They fired mixtures of gases in a bomb and observed the movement of a piston working against a spring in a tube attached to the bomb. From the accelera- tzon of the piston they calculated the pressure in the bomb. The pressures so measured are called by Berthelot the “effective pressures.” Now, since the .explosion-wave travels faster than sound in the unburnt gas, the explosion- wave is the first impulse which reaches the piston. It follows that when the piston feels this impulse and begins to answer to it, the explosion-wave has traversed the whole of the gas and the true explosion is over. The piston receives the blow of the wave and then the thrust of the expanding gases, no doubt still combining, to a greater or less extent, behind the wavefront. In Berthelot’s experiment, therefore, the movement of the piston gives, in the main, the rate of expansion of the heated gases after the explosion-wave has passed through them: it does not give the instantaneous pressure in the wave front itself. That higher pressures are produced for a moment in the explosion of gases has been proved by Mallard and Le Chatelier by the use of the delicate indicator designed by Deprez. MM. Mallard and Le Chatelier have also suggested a method of measuring these pressures by the fracture of glass tubes of known strength. This method we believe to give approximately correct results: it depends on the principle that if a pressure is produced in a glass tube greater than it can stand, the glass will be broken although the pressure may only last for a very small interval of time. SS * Ann. Chim. et Phys. [vi.] 4. p. 14 (1885). 176 Mr. H. B. DIXON AND MR. J. C. CAIN on the In 1893 one of us published some preliminary experi- ments by this method.* Tubes which stood a steady hydraulic pressure of 25 atmospheres, were broken into small fragments by the explosion-wave of carbonic oxide and oxygen ; whereas, stronger tubes which stood a pressure of 50 atmospheres were not broken by the explosion of oxygen, either with carbonic oxide or with hydrogen; on the other hand, the stronger tubes which had withstood 50 atmospheres on the hydraulic press, were broken by the explosion of cyanogen and oxygen in equal volumes, and one of these tubes was broken at 78 atmospheres on the press. It seemed desirable to repeat these experiments and to find, if possible, narrow limits within which the pressure of the explosion-wave must lie. Cyanogen was chosen as the combustible gas for most of the experiments, because the carbonic oxide and nitrogen yielded by its explosion with oxygen are simple in composition and approximate in physical properties to a perfect gas. Equal volumes of cyanogen and oxygen were mixed in an iron gas-holder over mercury. The explosion vessel consisted of a firing-piece with platinum wires, and two metal tubes between which the glass tube to be tested could be inserted by means of Faraday’s cement. After the apparatus had been filled with gas from the holder, the taps were closed at each end, and a spark was passed. The explosion-wave was generated in the first metal tube and traversed the glass tube. If the latter held it was removed and labelled, and another tube inserted in its place. The glass tubes were about 20 cm. in length; they were cut from long tubes of fairly uniform bore and thickness of wall. When a tube was broken it was our endeavour to gauge its strength by testing hydraulically the strength of the pieces cut on either side of it from the parent tube. « H. B. Dixon, Phzl. Trans., Vol. 184, p. 150. Instantaneous Pressures in the Exploston-Wave. 177 With equal volumes of cyanogen and oxygen, a very high pressure is produced in the explosion-wave. Soda-lime glass tubing of 18 mm. external diameter, and 2°5 mm. thickness, was fractured by the explosion. Green glass tubing of 28 mm. in thickness held. Experiments with the hydraulic press showed a very considerable difference in the strength of these tubes. Three pieces of the first glass broke when submitted to the following pressures :— 1. 8g0 lbs. on the square inch. Ze 950 99 >] 3: 1220 rp) bP) mean 1020 ,, a = 70 atmospheres. We think it safer to take the mean breaking strain of the three pieces as representing the strength of the tubes broken by the explosion, than to take the highest figure as the minimum force exerted by the explosion. We thus come to the conclusion that the pressure exerted in the explosion-wave exceeded 70 atmospheres. Pieces of the green glass tubing which withstood the explosion gave very unequal results on the press :— 2050 lbs. on square inch. 2. 450 5, ” mcan £750 ,, - = 120 atmospheres. Unfortunately we had no other specimens of the same kind to test. Our result, therefore, is that in the explosion of equal volumes of cyanogen and oxygen, the pressure produced falls between the limits of 60 to 140 atmospheres, and more probably between 70 and 120 atmospheres. In the next experiments, the mixture of equal volumes of cyanogen and oxygen was diluted with its own volume of nitrogen. The reaction occurring may be written :-— C.N, + Og + 2N2= 2CO + 3Nz. The “ effective pressure” produced on firing this mixture 178 Mr H..B. DIXON AND MR: J. C. CAIN on the in a bomb has been measured by Berthelot, and found to be 15 atmospheres. As calculated from Berthelot’s theory, the pressure in the wave should be 18 atmospheres ; according to Dixon 57 atmospheres. One reason which led us to dilute the explosive mixture was suggested to us by Professor Osborne Reynolds. If the velocity of the explosion-wave in the gas approximates to the rate at which the distortion-wave in the glass is propagated, the latter might be continually re-inforced, and the tube be broken as the result of a pressure far less than that required to break it under other conditions. The velocity of this wave in glass is nearly 3,000 metres per second. The rate of explosion of equal volumes of cyanogen and oxygen is 2,728 metres per second ; when this mixture is diluted with its own volume of nitrogen, the rate of explosion falls to 2,163 metres per second. In the diluted mixture, therefore, there could be no question of the waves coinciding in rate. The reduction of pressure caused by dilution made the measurement more accurate, as it enabled us to find glass of more nearly equal strength holding and breaking respectively. After several trials a piece of uniform tube was found which broke, and a slightly thicker one which held. Two pieces of the first broke at the following pressures :— I. 950 lbs. on square inch. 2. 925 ;, 9 99 mean 938 ,, a is = 63 atmospheres. Two pieces of the second broke :— I. 1230 lbs. on square inch. 2. I250 9 +) ” mean 1240 ,, is a = 84 atmospheres. The lower limit viz, 63 atmospheres, is rather higher than the pressure calculated by Dixon’s formula. | Instantaneous Pressures in the Explosion-Wave. 179 An indirect way of arriving at the pressures in the explosion-wave is given by Riemann’s* equation for the propagation of abrupt variations in the density and pressure of a gas. Professor Schuster + has given reasons for sup- posing that Riemann’s equation applies for the explosion- wave, and has shown a simple way of calculating the pressures from the known velocity of the explosion-wave and the density of the unburnt gas. According to Rie- mann’s equation the pressure in the explosion-wave of cyanogen and oxygen should be 135 atmospheres, and when diluted with its own volume of nitrogen the pressure should be 71 atmospheres. The calculated and observed pressures may be conveniently compared in the annexed table :— Pressures in the Explosion-wave. CALCULATED. OBSERVED. Gaseous Mixture. Berthelot.| Dixon. | Riemann.||Berthelot. roe | C.Na + Oz a5 it. a7 Ate 1S5 Atl) 125 At}7o—120) C.N,.+ O,+2N, || 18 ,, 5B | AT oll B5cen 103-84] It will be observed that the pressures calculated by Riemann’s equation are about 4 times greater than those deduced from Berthelot’s Theory : and are larger (roughly by 20 per cent) than those calculated from Dixon’s They agree within the limits of error with our observations on the breaking strain of glass tubes. Experiments on the Collision of Two Explosion-waves. The apparatus we employed could readily be adapted to observe the effect of bringing two explosion-waves into collision. Will the result of two waves meeting from * *Gottingen Abhandlungen.’ 8. (1860). + Vide Phil. Trans. Vol. 184, p. 152. (1893). 180 Lnstantaneous Pressures zr the Explosion-Wave. opposite directions be to largely increase the pressure at the point of contact? By analogy one might suppose that such would be the case; but, on the other hand, since the explosion travels much faster than any wave in the unburnt gas the explosion-wave is always, as it were, dashing on a dead wall and piling up pressure, and no further effect, it might be argued, could be produced when it meets and is repulsed by a similar wave. It seemed, however, possible if the wave is propagated partly by the movement of heated yet unburnt molecules in the wave front—that ¢hese mole- cules on coming into collision would cause a measurable increase of temperature and pressure in the wave. We have not been able to measure any such increase by this rough method of trial. The explosion tube, some 3 feet from the firine ‘point, bifurcated )intese ges arms like the letter Y. The two arms were bent round nearly to meet, and the junction was effected by a piece of glass tube inserted in the gap. The centre of the glass tube was exactly equi-distant from the fork by either arm ; consequently the explosion-wave, dividing into two at the fork, traversed the two arms and came into collision in the middle of the glass tube. By a suitable tap, one arm could be closed, and the explosion then traversed the glass tube only in one direction. Experiments made with hydrogen and oxygen, with equal volumes of cyanogen and oxygen, and with the same mixture diluted with nitrogen as before, showed no appreciable difference between the pressures produced in the glass tube when the flame went in one direction only and when the two explosion-waves met end on. Pieces of the tube, which broke in the hydraulic press at 63 atmospheres, broke both ways equally in the explosion apparatus: pieces of the tube, which broke in the hydraulic press at 84 atmospheres, stood the explosion both ways equally. If the collision had caused the pressure to rise by 14 we ought to have detected it. Coast Lines and Magnetic Declination. 181 On the Influence of the Configuration and Direction of Coast Lines upon the Rate and Range of the Secular Magnetic Declination. By Henry Wilde, F.R.S. (Recezved April 3rd, 1594.) In a paper which was read before the Royal Society in June, 1890, I showed that the principal phenomena of terrestrial magnetism and the secular changes in its horizontal and vertical components could be explained on the assumption of an electro-dynamic substance (presumably liquid or gaseous) rotating within the crust of the earth in the plane of the ecliptic, that was to say, at an angle of 23°°5, and a little slower than the diurnal rotation. By means of some electro-mechanism, new to experimental science, which I termed a Magnetarium, the period of backward rotation of the electro-dynamic sphere required for the secular variations of the magnetic elements on different parts of the earth’s surface was found to be 960 years, or 22°5 minutes =0'375 annually. From the relations of a magnetic needle on the earth’s surface, and an electric current circulating round the internal electro-dynamic sphere, it will be obvious that the magnetism of such a system would be symmetrically distributed, with similar lines of declination and inclination on meridians and parallels 180° from each other. An examination, however, of the lines of declination over the terrestrial globe, as determined by careful and repeated observations, exhibits wide divergencies from the sym- metrical lines of declination obtained with the. electro- dynamic sphere alone, Thus, there are on the variation chart (Plate X.) four 182 Mr. HENRY WILDE oz well-defined lines of no declination in the northern hemi- sphere, for two similar lines of no declination in the southern hemisphere. The declination also varies very considerably for equal latitudes and longitudes in the northern and southern hemispheres, for the same or for different epochs. This is seen in the large amount of the declination at the Cape of Good Hope, and the small amount on the great land areas of Eastern Europe and Asia, as well as over the Eastern States of North America. The comparatively small area, or oval, of westerly declination in Eastern Asia, surrounded by considerable areas of easterly variation, together with the closed curve of small easterly variation in the equatorial parts of the Pacific, contributes still further to increase the difficulty of the problem of reducing the distribution of the earth’s magnetism to general laws. The unsymmetrical character of the lines of equal variation, and the devious courses of the lines of equal inclination and the magnetic equator, are no less perplexing to magneticians than the irregularities of the declination at different epochs for equal latitudes. In the course of my experiments, it was noticed that the lines of no declination of the internal sphere of the magnet- arilum were generally in advance of those on the charts for a given epoch. Thus the two antartic lines of no declination were more than 40’ east of the similar lines of the electro- dynamic sphere for the epoch 1880. With the object of ascertaining what influence the configuration of the surfaces of the terrestrial globe, as indicated by the general distribution of land and water, had on the magnetic elements, the ocean areas of the outer globe of the magnetarium were covered with thin sheet iron roughly contoured to the coast lines in both hemi- spheres. On turning the internal electro-dynamic sphere 84° W. to correspond with the epoch 1880, a remarkable change in the magnetic elements was manifested. The two Coast Lines and Magnetic Declination. 183 lines of no declination in the southern hemisphere of the outer globe were nearly coincident with those on the chart; and in the northern hemisphere four zero lines appeared similarly coincident ; two of which lines on the North American and European continents were continuations of those in the southern hemisphere. But the most remark- able and unexpected feature of the distribution of the magnetism on the iron-covered globe was the reproduc- tion of the oval area of small westerly declination in Eastern Asia (110 —160° E.), surrounded by large areas of eastern declination. The oval also agreed in detail with that on the chart in having the largest westerly declination, about 8° in the centre, between the lines of no declination. Scarcely less interesting was the unlooked-for reproduc- tion of the oval area of small easterly declination, about 5°, surrounded by a large area of greater eastern declination in the equatorial parts of the Pacific (120°—170° W.), while the unsymmetrical form of the magnetic equator was very similar in its deviations to that of the earth for the epoch 1880. Further experiments with the iron-covered globe showed that the land areas, besides retarding the translatory movement of the lines of thedeclination, generally diminished the amplitude of the declination itself, and to a greater amount as the broad features of the continental coast lines extended more or less in a direction parallel to the earth’s equator. On the other hand, continental coast lines extending more or less parallel to the earth’s axis and terminating in capes or headlands, diminish the horizontal force, and, consequently, increase the rate and range of the declination ; as instanced :—(1) In the large amount of the secular change along the South African coasts, where at the Cape of Good Hope the declination is 30°W. (2) On the South American | coasts about Cape Horn, 20°E. (3) Onthe Greenland coasts at Cape Farewell, 50°W. (4) The coasts of Southern India 184 Mr. HENRY WILDE oz at Cape Comorin, where the declination was 16°W. in the year 1601, and is now I°E. vy The observations of deep-sea temperatures made during recent years have brought out the important fact that, at great depths, the temperature of the ocean beds is little above the freezing point of water. Prestwich and others have inferred that this low temperature of ocean depths is competent to produce a greater thickness of the earth’s crust under the oceans than under the land. The large amount of iron which enters into the composition of the earth’s crust is well known from the analysis of volcanic ejections from all parts of the globe, while at extreme depths this element exists in the metallic state, as at Ovefak, off the coast of Greenland, where it is found diffused in the basaltic rocks and in separate masses. We have, therefore, through the low temperature and increased thickness of the ferruginous ocean beds, the precise con- ditions required for producing the differences in the magnetic elements which have been shewn on the mapped globe when the ocean areas were covered with iron. Now that the great influence which the land areas exercise in retarding the translatory motion of the lines of the declination has been shewn, which is distinct from the magnetism of local geological formations, the same influence in determining the form and position of the declination lines on the terrestrial surface becomes very apparent on the charts. An important negative feature of this influence is the symmetry and simplicity of the declina- tion lines in the southern hemisphere, where the ocean completely encircles the globe in latitude 60°, as compared with the devious lines of the declination on the great land areas in the same latitude of the northern hemisphere. The dominant influence of longitudinal coast lines is well seen in the bend of the zero and other declination lines towards the north pole of the earth’s axis in their westerly Coast Lines and Magnetic Declination. 185 march over Europe ;—the effect of the intersection of the land areas by the basin of the Mediterranean and other great inland depressions, extending more or less parallel to the equator over 60° of longitude. The polarising effect of the arctic coast line appears in the small amount of the declination at St. Petersburg, which has not varied more than 8° during the last 150 years. The polarisation of the coast lines is again seen on the chart at the shoulder of the South Amerian continent at Pernambuco, where the lines bend upwards towards the polar axis, and resume their westerly direction in the Carribean Sea and in the basin of the North Atlantic. Strong polarising effects, to diminish and retard the declination, are also produced by the longitudinal coast lines of the Gulf of Mexico, the West Indian Islands, the north and south coasts of Australia, the great coast lines within the Antarctic circle, the Malayan Archipelago and the southern coasts of India and China. So great is the polarity of the West India Islands, that the secular change of the declination at Jamaica and Cuba has not amounted to more than 3° during the last 200 years. From a comparison of the zero lines of declination of the internal electro-dynamic sphere in relation to the earth’s axis and to the zero lines on the terrestrial surface, it will be seen that the appellation of magnetic poles and zeros of declination applies with strictness only to the poles of the earth’s axis, and to the poles and meridians of the internal electro-dynamic sphere, as the zeros of declination on the earth’s surface, for the present epoch, are generally the resultants of the changing electro-dynamic and permanent magnetic forces acting through and upon the outer crust of the terrestrial globe. An interesting instance in confirmation of my views and experiments on the polarising action of longitudinal coast lines which I desire to bring before the Society, on account of its importance to practical navigation, was brought to 186 Coast Lines and Magnetic Declination. light during the Admiralty enquiry into the causes that led to the disastrous loss of H.M.S. “Serpent” off Cape Villano, on the coast of Spain, November, 1890.* Among other reports read before the Court was one from the captain of the Spanish screw steamer “ Beneta,”’ who stated that, during the 15 years he had been trading along the north coast of Spain, he did not remember to have observed any deviation of the compass on account of the attraction of the iron ore mountains, but he always noticed that when steering on a southerly course the error of the compass was N.E.; that was to say, that the local variation of the compass was eight or ten degrees N.W. instead of 18 or 20 degrees as shown on the Admiralty chart of the declination. Now, a glance at the chart will show that the north coast of Spain extends parallel to the earth’s equator for a distance of nearly eight degrees, or 400 miles, and has, consequently, a maximum polarising influence on the compass to diminish the amount of the secular change of the declination, as observed by the captain of the “ Beneta” and as set forth in my papers. It would therefore appear that some of the declination lines, as represented on the charts, do not partake of that symmetrical character that is generally accorded to them and that caution will be required in the use of variation charts off the greatly extended coast lines of deep seas where the rate and range of the secular declination are large in amount. * The London Zzmes, December 17th, 1890. — LK Ly, V7 {= ga 1 WAL A yes * oe Ko ae a : Bai) Dy ay ay \ . a ff NSS Af of y al j Ne RY —Influence of Coast Lines on Magnetic Declination. “ow o 0 m0 PY Goole aces ad ] “i } EX PLAT AT | SS Wy) CURVES OF EQUAL MAGNETIC VARIATION FOR THE EPOCH. 1880. 4th Series, Vor. VIII. ieee vi CY A a ¥ - 3 ; JD e } ' " whe “tyr ~~ — ~ Ph ganas tage cape tipioeks cry eegieree- +> i oteamihe z > . ~h b'R anaes * ~ ag x ae er a } ~~ ay F32 eARME ADEE RATA TS CEL Sef FE IE pa D.C.L., LE.D., F: RS; Savilian Prof. of Geom. in the Univ. of Oxford, Cor. Mem. Inst. Fr. (Acad. Sci.), &c. Mew College, Oxford. Tait, Peter Guthrie, M.A., F.R.S.E., &c., Professor of Natural Philosophy, Edinburgh. 38, George Square, Edinburgh. Thorpe, T. E., Ph.D., F.R.S. Laboratory, Somerset House, London, W.C. Trécul, A., Membre de Il’Institut. Parzs. Turner, Prof. Sir Wm., F.R.S. Edinburgh. ‘iylor; Edward: Burmett, F.k.S., D.C.L. (Oxon), LED: (St. And. and McGill Colls.), Keeper of University Museum. Oxford. Radcliffe Observatory, Vines, Sidney, Professor of Botany, F.R.S. Headington fill, Oxford. Waage, Professor P. Christiania, Norway. Walker, General Francis A., Professor of Political Economy. 237, Beacon Street, Boston, U.S.A. Warburg, Professor E. 8, Goethestrasse, Freiburg, Baden. Ward, Professor H. M., F.R.S. Cooper's Hill, Englefield Green, Surrey. Weismann, Professor August. retburg, Baden. Wiedemann G., Prof. of Physics, For. Mem. R.S. 35, Thalstrasse, Letpsic. Williamson, Alexander William, Ph.D., LL.D., F.R.S., Corr. Mem. Inst. Fr. (Acad. Sci.). Aigh Pitfold, Shotter- mill, Haslemere. Williamson, Wa -C., LL.D., Clapham Common, London. Young, Prof. C. A. Princeton College, N. J., U. S.A. F.R.S.- 43) 2772s Koad, Zirkel, Ferdinand, Professor of Mineralogy. Unversity of Letpsic. Date of Election. 1870. March 8. 1866, Jan. 23. 1861, April 2. 1849, April 17. 1850, April 30. 1882, Nov. 14. 1859, Jan. 25. 1857, Jan. 27. Corresponding Members. Corresponding Members. Cockle, The Hon. Sir James, M.A., F.R.S., F.R.A.S., F.C.P.S. 12, St. Stephen’s Road, Bayswater, London. De Caligny, Anatole, Marquis, Corres. Mem. Acadd. Se. Turin and Caen. Socc. Agr. Lyons, Sci. Cherbourg, Liége, &c. Durand-Fardel, Max, M.D., Chev. of the Legion of Honour, &c. 36, Rue de Lille, Paris. Girardin, J., Off. Legion of Honour, Corr. Mem. Instit. France, &c. Lille. Harley, Rev. Robert, M.A., F.R.S. Savile Park, Halifax, Yorks. Herford, Rev. Brooke, 91, Fitzjohn’s Avenue, Hampstead, London, N. W. Le Jolis, Auguste-Francois, Ph.D. Archiviste perpétuel and late President of the Soc. Nat. Sc., Cherbourg, &c. Cherbourg. Lowe, Edward Joseph, F.R.S., F.R.A.S., F.G.S., Mem. Brit. Met. Soc., &c. Shirenewton Hall, near Chepstow. Date of Election. 1870, Dec. 13. 1861, Jan. 1837, Aug. 1881, Nov. I. 1887, Nov. 16. 1865, Nov. 15. 1888, Nov. 13. 1888, Feb. 7, 1894, Jan. 9. 1868, Dec. 15. 1861, Jan. 1875, Nov. 16. 1889, Oct. 15. 1894, Mar. 6. 1855, April 17. 1861, April 2. 1844, Jan. 22. 1889, April 16. 1860, Jan. 23. 1886, April 6. 1846, Jan. 27. 1889, Jan. 8. 1880, Oct. 15. 1872, Nov. 12. 1893, April 18. 1854, April 18. 1884, Nov. 4. 1853, Jan. 25. 1893, Jan. 10. 1859, Jan. 25. 1876, April 18. Ordinary | Members. 233 Ordinary Members. Angell, John, F.C.S., F.I.C. 6, Beaconsfield, Derby Road, Fallowfield, Manchester. Anson, Rev. George Henry Greville, M.A. Sirch Rectory, Rusholme. Ashton, Thomas. 36, Charlotte Street. Ashton, Thomas Gair, M.A. 36, Charlotte Street. Ashworth, J. Jackson. 39, String Gardens, City. Bailey, Charles, F.L.S.