"2 kb “st he oe Ce ee ek ee et a ee Te a ae = Se a Ree h tee Maud: ia) ths | tee cee tt Verte 98 ot Greg. a4 Or Was ae ek wy he, ee Oster re ede ade ere eee Madar ae Doh Mora the a Cote a 2 Thee dot 4d Ree het TW ee ek td Oe ed ee Sete ee dat 60 ON Pare een aw ee ee oO ee | Oh bo od Cae dee, We thet try 8 ee get. 04 99) Fw ed ed Sige cet dows te daedom teed oe a ee A ive Meat t te te on aie) eX ee SN ot wide wasn on Wee HE 4 moe ee a wee | WEA ot i eer ee co a re NE tite \ Cs ares pA det die we eae + MA ted don a ked of deve ne i* 498-8 9 leeee a oe oe * aes rao few ee bead eH nm beg, Publik unter reece’ Hi ed ee tf state testeih Y ae wien , aA REE Me he 1 oth He Minateatintat ats ween irae ARR ee Fh nh Si ies Meee i 1 2 enn id Satta AG 4 Garena hehehehe dyed 40d 59.8 Monet Roa h-hh rw) ib t-8 6 real aba bak Wn Be ae (AMO Mee O Hig dee hae GRitewe is ea awe: Se eee i te ene eer 2 MW Ardeden AO We Wed ge PORE Hod te eae ee UD ae ed toy WEY Hele he had Set ke og 4s Me tcl bie te bole ee Vet dewe aaa e Hoe . # ttt Oe Re a Giekieg ote Bt th ft Ce AON Sethe beste args the 44) La ee +44 He 1 tenet 8 te katy ae TPM Mh eth ets geet th ot rk nt VN oe Pe ednitvotrte et teed a al whens Meta w > Hah 94 ow Wet ek ee! heen. due 4 em hd een wow 4 Ot Oe de ed eee de ip we Wrtetenou 1 'E 4 Ae ea a yie pleat, ano at a yee wy 400 iets only Wet ete tetra ae ‘@ Melek ace $e8) ea “| ErWeaei t-H We Riltet ed > Hehotse way ha A Ne the tthe Went «05 he Butpeme. bet “a 1 +i hot bad th he tadog.. Arh bite ihe L tele (CW Goth gett vem. te ed te td ee “elt, iar ae Hg Rete Morea fe they Oy 98 tr TE eh Yk hrteare getty he ed et @ ie Cro ree ae ee teh a een Sve aerate MyM OO CH bot e Ha i erdls a ng A RR, ot aR dae, bh Oou ut 8 Fe he bee dee * AT ew Veted tet Fart hall he St wae i ht Hemet oe ig. Oe, Cacia ee ‘ Career fe E oh th ely het Hee thet 4 note Bring es Ayla diet duet cet ae rt ad nit (Pett atts ae f nee | Lees tented ¥ tk, sabe hel te “t4 Ge er ie. ish Pt eke rer wa BEN rH Her acer ry oeane 4 isha +: he aed. MAT. alsheen 4 Be Om Carbo th a Ae + : UA ep hk hs ea Wea et ik 4s y. feet 4, ky AUN hf Aad mhed 50 lk emgant a | ; : Fath Ah WA EG Hale) tims eaene beat 6 GA ed Geng PET Ede tga eg yah ot Uh oka a | a) ot et, oe ou ee ta | hs nat Aids daded 2A Poe ea ity o 14-4 te- oye - ey oe ee ee Oe aks is Sh Ont WN te a Te a: Pt do toe ae! a Me dl Med emit tote ped We ty Feit thie PoE Whol aan Ut cult Gate | Ah ORAM Mitek tee egg Mi Vege ie Oe 8= EU GS eee ye ee Hee, 4 ~su ited 4 Bred Gevy Wy Ee Pete a Ade Ee dee Qt OL ECCT ONO hae 4g Le ek MRL Mack ek tee | Leas 1s b-WeAl dee be © neitec y ‘i rgiratas ets d + fhe Beis p A Mp Y RE AES Rey Mea ering ib fait oe aaa Wg tase as ala 5% aaa hata 147% Xt Wirnsvans a de WeHog AW od a Vek Qa ek Gh mah in thiod 8 ” HRSA Ho P W tW ie Wee ae a eetn « Ve . 595 4 rh ha ee eee 4 4 8 A ea ged Thal deck | SSUES DAE A Fateh gla Ay 4 \ shad “Wetted 6 dad Te Fath note , Het tev Ree tea Wi ewe dane 4 Laat Ge © Fs rapes hihi ft -” ae te oat uit. 1 45 Ye 0 @ ages Het ‘ Uae hs SN a mk ch 4 HW eit ee dw a, aah © A Hs thee th: eheieyts eet Wo deme te i WM Eee Rees a a ar ee Tihs Wen ae on eit ‘ WA QIN I reat te air ay te nbs te Pipl Fees Pt Ae ces Cite a on s ASA " Se te Ae f bo Gite G8 dente ee tee try eed LAL el ets be a te vot Potetd Marae ae ee Ae, Aa it Sas Be a4 teeth g ow ‘ ‘ 7 a A * Li rb ‘e ee bevy * . Hehe eolpelied te di teh we digg . “w A et TET Gi -4 dot Woe eee y ehore TORR soa @ ht wot go we MM Mees deg 148 Wh ene WA Ww Me dedi read ‘Wed hoe herd) 41 at a4 +E Oe tele C44 HH Gre SOR WOM, Coe ed aged AE Beek. ae SMe OM ere TTA = as nt Wnt Redo deh Haw Goad alae WD ge ese we a he WN EW US AW Adri ie fs ie Miemen ee we ld S454 4 Goquen yy WA Hrt ded aed ag Tete tert as We tree BS 2 Hoe ke Wh 9 We ke he en tho the AiG Hew Ag Te eee: ¢ « FM ems oq ke oe eee ice ou) ores ¥ Fe he totdent iy i ete bot 3h ete ie NSIS tas ree Gods et. bl | i« 4 saat Le Aarh GW is deartsah (hom gO en 24 toe Wi tad e-ee ey ee hm 4 ‘e SU tee BRR Ne He oo eH ef ee JA bu 7c . aA Lt aT athena nor tas Oe Kaw 4 “4 Whew Gee Cg we 4 eet Gey y ww eet Od hPa Wd idee tog owe a WAN Hee eed Le « ‘ashe. Wr f. Ws Hes eth rian Ly ee ” sla ah heh een es Selena co Reet Me mr tr ee art % vi 4 FO ete ee ey Sab NEE hd hte spa i Paha had> es SL rar ine te * we Hertionk TER He eh No ohe pared te € 544 be tow PP Es) eh HMO het tegen “s Ua ate SS Sh RRR aed eek eke get PROCEEDINGS — OF THE ROYAL SOCIETY OF LONDON. From March 24, 1881, to June 16, 1881. VOL. XXXII. LONDON: HARRISON AND SONS, ST. MARTIN’S LANE, | Printers in Ordinary to Her Wajesty. MDCCCLXXXI. LONDON: HARRISON AND SONS, PRINTERS IN ORDINARY TO HER MAJESTY, ST. MARTIN’S LANE. CONTENTS. VOL YOCOM. No. 212.—March 24, 1881. Page THE Croonran Lecrure.—Observations on the Locomotor System of Echinodermata. By George J. Romanes, M.A., F.R.S., and Professor piesa me me yaC VD D8 ds scsaudendadugsen doSoopsoetait (avedas nstensouseedieasenbese, eneseedeaese 1 The Functional Relations of the Motor Roots of the Brachial and Lumbo- sacral Plexuses. By David Ferrier, M.D., F.R.S., Professor of Forensic Medicine, and Gerald F. Yeo, M.D., F.R.C.S., Professor of sem Oa MUMIPM IT’S COMEG CT 210, J cncc.s.c.noscecetenssteasantncvececeseceescaaecseseseseuessdes ones 12 On the Histology and Physiology of the Pepsin-forming Glands.. By J. N. Langley, M.A., Fellow of ‘Trinity College, Cambridge ......... cee 20 March 31, 1881. On the Coefficients of Expansion of the Di-iodide of Lead, Pbly, and of an Alloy of Iodide of Lead with Iodide of Silver, PbI,Agl. By G. F. Rodwell, F.R.A.S., F.C.S., Science Master in Marlborough College .... 23 Permanent Molecular Torsion of Conducting Wires produced by the Passage of an Electric Current. By Professor D. E. Hughes, F.R.S..... 25 On the Tendinous Intersection of the Digastric. By G. E. Dobson, MO Memb LIME mR hed 2x e he ccae neces neconsedhe hed ieidsdeodlcevdsdeesedhibbalasurohduesnue!? 29 Wote on Protagon. By Henry E. Roscoe, LL.D., FBS. .i.......ccceesceecenee 30 April 7, 1881. On the Minute Structure of the Lung of the Newt with especial reference to its Nervous Apparatus. By William Stirling, M.D., Sc.D., Regius Professor of the Institutes of Medicine (Physiology) in the University Bie AE Oc eH remy es 20296 18 2050S t asaya tesa Sten fasedsen asset onhodssveee tvashieansetoess 37 On an Electrodynamic Balance. By H. Helmholtz, For. Mem. R.S., Professor of Physics in the University of Berlin cece eeceesesceeseeeeseeene 39 LV Page On the Internal Forces of Magnetized and Dielectrically Polarized Bodies. By (Protessor EH. Helmboltz, Mor Mem RISa ei oreo. es vied aoieae April 28, 1881. The Influence of Stress and Strain on the Action of Physical Forces. By Eferbert Tomlinson, BiAgk.f0... 6... aoe. Oo olde soto acne eee ec Lucifer : a Study in Morphology. By W. K. Brooks, Associate in Bio- logy and Director of the Chesapeake Zoological Laboratory of the Johns Hopkins University, Baltimore, Md., US.Al i. WPISTOL PESOS \..5= re) is Me ee an ee 4 =) 2 7 Faled ee 4 = it: ‘ ; i ; : 4 § wo a ‘ " i iad TOR 5 sen is Saas a> é Lb Li #y re } ere eA + Ae als ae. ABI i 274 i rie 7 % P 243k ge S LP aa bd A a a5 E Re. 4 Pe : t a i 7 38 = 1 Nah <— : a ees hub it 7 crA te = 2 * ra o ' oot —s , = . t = I = a - 4 . + i are F ~ = - i Pa o f mer ‘ rs 7 4 ( g; 3: ‘ d LA ee H = aN * i X ms " . > 2 > > - = - s = ‘ RS 4 ey > t s 3 Y x » r ‘| 5 iy ‘ Received since the End of the Session. {, a sy US eh M 15 ye, my / | od “ Preliminary Note on the Photographié irene oe Comet 61881. By WiitAm Huveers, D.C.L., LLD., F.R.S. Received June 27, 1881. On the evening of June 24, I directed the reflector furnished with the spectroscopic and photographic arrangements described in my paper “On the Photographic Spectra of Stars”* to the head of the comet, so that the nucleus should be upon one half of the slit. After one hour’s exposure the open half of the slit was closed, the shutter withdrawn from the other half, and the instrument then directed to Arcturus for fifteen minutes. After development, the plate presented a very distinct spectrum of the comet, together with the spectrum of the star, which I have already peeeeied 3 in the paper referred to above. The spectrum of the comet consists of a pair of bright lines in the ultra-violet region, and a continuous spectrum which can be traced from about F to some distance beyond H. The bright lines, a little distance beyond H, with an approximate wave-length from 3870 to 3890, appear to belong to the spectrum of carbon (in some form, possibly in combination with hydrogen), which I observed in the spectra of the telescopic comets of 1866 and 1868. In the continuous spectrum shown in the photograph, the dark lines of Fraunhofer can be seen. This photographic evidence supports the results of my previous observations in the visible spectra of some telescopic comets. Part of the light from comets is reflected solar light, and another part is light of their own. The spectrum of this light shows the presence in the comet of carbon, possibly in combination with hydrogen. On the next night, June 25, a second photograph was obtained with an exposure of an hour and a half. This photograph, notwithstanding the longer exposure, is fainter, but shows distinctly the two bright lines and the continuous spectrum, which is too faint to allow the Fraunhofer lines to be seen. (Postscript, July 9, 1881.) I have since measured the photographs of the comet’s spectrum, and I find for the two strong bright lines the wave-lengths 3883 and 3870. The less refrangible line is much stronger, and a faint lumi- nosity can be traced from it to a little beyond the second line 3870. There can be, therefore, no doubt that these lines represent the brightest end of the ultra-violet group which appears under certain * «Phil, Trans,” 1880, p. 669. 2 Profs. G. D. Liveimg and J. Dewar. circumstances in the spectra of the compounds of carbon. Professors Liveing and Dewar have found for the strong line at the beginning of this group the wave-length 3882°7, and for the second line 3870°5. I am also able to see upon the continuous solar spectrum, a distinct impression of the group of lines between G and h, which is usually associated with the group described above. My measures for the less refrangible end of this group give a wave-length of 4230, which agrees as well as can be expected with Professors Liveing and Dewar’s measure 4220. In their paper “On the Spectra of the Compounds of Carbon,” “Proc. Roy. Soc.,” vol. 30, p. 494, Professors Liveing and Dewar show that these two groups indicate the presence of cyanogen, and are not to be seen in the absence of nitrogen. If this be the case, the photograph gives undoubted evidence of the presence of nitrogen in the comet, in addition to the carbon and hydrogen shown to be there by the bright groups in the visible part of the spectrum. On this hypothesis we must further suppose a high temperature in the comet unless the cyanogen is present ready formed. I should state that Mr. Lockyer regards the two groups in the pho- tograph, and the groups in the visible spectrum, to be due to the vapour of carbon at different heat-levels (“‘ Proc. Roy. Soc.” vol, 30, p. 461). It is of importance to mention the strong intensity in the photo- graph of the lines 3883 and 3870, as compared with the continuous spectrum, and the faint bright group beginning at 4230. At this part of the spectrum, therefore, the light emitted by the cometary matter exceeded by many times the reflected solar light. I reserve for the present the fheoretical suggestions which arise from the new information which the ohaveoraphe have given us. cee ppr~s—-23 | “Note on the Reversal of the Spectrum of Cyanogen.” By G. D. Liveine, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., E.BS., J ncheonent Professor, University of Cambridge. Received July 4, 1881. In the course of many observations on the reversal of lines of metallic spectra, we have frequently noticed dark shaded bands which appeared to be the reversals of bands ascribed to the oxides or chlorides of sundry metals ; more particularly we have seen them when experimenting with compounds of the alkaline earths, and we have repeatedly obtained a reversal of the green magnesium-hydrogen series ; but, until recently, we have never seen any reversal of the shaded bands of the spectrum of cyanogen, though our attention has been Note on the Reversal of the Spectrum of Cyanogen. 3 constantly directed to this spectrum. Quite lately, however, we have obtained photographs which show the reversal of the violet and ultra-violet bands of this spectrum; and the fact is perhaps of suffi- cient interest, especially in connexion with the question of the occur- rence of these bands amongst the Fraunhofer lines, to warrant the publication of this note. We have not yet succeeded in determining precisely the conditions under which the reversal can be produced at will. The most complete reversals of these bands were obtained by the use of the arc of a Siemens’ machine, inacrucible of magnesia, fed with a considerable quantity of cyanide of titanium. The photo- graphs in this case show a very complete reversal of the five bands near L, and of the two strong bands near N, and a less complete reversal of the six bands, beginning at about wave-length 4215. No other metallic cyanides have given, when introduced into the crucible, any such reversal; nor does a stream of cyanogen led in through a perforated carbon produce the effect. Various other nitrogenous compounds have been tried, but the only one which has given us any- thing like the effect of the titanic cyanide is borate of ammonia. Some photographs taken immediately after the introduction of borate of ammonia show distinctly the reversal of the group of bands near L. In one case when metallic magnesium jhad been put into the crucible, the photograph shows a reversal of only that part of the series which is nearest to the magnesium group, indicating that the reversal is due to the bright background supplied by the expanded magnesium lines. There can be little doubt that the greater stability of titanic cyanide and boron nitride than of other nitrogenous com- pounds, has some influence upon the result; and the difficulty in pro- ducing the reversal at will is in securing an absorbent stratum of suffi- ciently high temperature and at the same time a sufficiently luminous background. The circumstances which secure the former condition almost always produce in the arc a still more intense radiation of just those rays which are absorbed, without that expansion of the lines which shows out the absorption in the case of so many metallic spectra. The photographs are, however, conclusive evidence that it is possible to secure both conditions. | OBITUARY NOTICES OF FELLOWS DECEASED. Micuren Cuasies was born at Epernon, in the department of Eure- et-Loire, on the 15th of November, 1793. He began his education at the Lycée Impériale, and even in child- hood showed a decided taste for geometry, and was in the habit of communicating to the pupils of other schools the problems given him by his teachers, obtaining their problems in exchange. In the year 1812 he began his studies at the Ecole Polytechnique. In 1814 he assisted with his fellow-students in the defence of Paris, and in that same year passed the engineering examination. After considerable hesitation he decided to accept the appointment as officer of engineers to which he was now entitled ; but just as he was about to enter the service he was induced to surrender the appointment in favour of a friend, who was next on the list to the successful candi- dates, and to whom it was a great object. He spent a short time with his mother at Chartres, and then resumed his studies at the Ecole Polytechnique. When, in 1814, all the students of the Ecole Polytechnique were abruptly dismissed, Chasles gave hospitality in his home at Chartres to _ his brilliant schoolfellow at the Lycée Impériale, Gaétan Giordini, who through his influence had been induced to study geometry, and who had obtained the first prize at the Concours Général over the head of Chasles himself, and had afterwards obtained the first place in the Ecole Polytechnique. On finally quitting the Ecole Polytechnique Chasles spent about ten years in retirement at Chartres, devoting himself to geometry. In 1837 he published the first edition of his great work, ‘“Apergu His- torique sur l’Origine et le Developpement des Methodes en Géométrie, ” which was characterised by De Morgan as a work of great importance in the historical point of view, and described in the following words by M. Bertrand: “ L’admirable ‘ Apereu Historique,’ qui sous ce titre plus que modeste, restera l’ceuvre la plus savante, la plus profonde, et la plus originale qu’ait jamais inspirée Vhistoire de la science.” In 1841 Chasles became Professeur de Machines et de Géodésie at the Ecole Polytechnique, which appointment he held for ten years, when he resigned it in consequence of some radical alterations which were being made, and of which he entirely disapproved. In 1846 he was appointed to a new chair of Modern Geometry at b il the Faculté des Sciences, which had been established in consequence of the strong recommendation of Poinsot, and for twenty-five years Chasles devoted himself assiduously to his duties at the Sorbonne. The ‘‘Traité de Géométrie Supérieure,” an elaborate and masterly treatise which embodied the substance of a course of lectures given to the Faculté des Sciences, appeared in 1852. This book became scarce, but about two months before his death, Chasles had the satisfaction of seeing a second edition of it published, accompanied by his excellent ‘‘Discours d’ Inauguration.” The ‘‘ Traité de Géométrie Supérieure ” was followed in 1865 by the first volume of his ‘‘Traité des Sections Coniques,” being a sequel to the former. No other volume of this- work ever appeared, though much desired by mathematicians. In 1863 Chasles published his book ‘‘ Les Trois Livres de Porismes d’Euclide, rétablis pour la premiére fois, d’aprés la Notice et les Lemmes de Pappus, et conformément au sentiment de R. Simson, sur la Forme des Enoncés de ces Propositions.” The publication of this work led to a short controversy with M. P. Breton (‘‘Question des | Porismes—notices sur les débats de priorité auxquels a donnée lieu VPouvrage de M. Chasles sur les porismes d’Huclide,” Paris, 1865; and a second part, Paris, 1866). M. Chasles comments on these in his “* Rapport.” In 1851 Chasles was elected a member of the French Academy, of which he had been a corresponding member since 1839. In 1854 he became a foreign member of the Royal Society. and in 1865 he received the Copley medal, which was given to him in acknowledg- ment of his historical and original Researches in Pure Geometry. He was the first, and for some years the only, foreign member of the London Mathematical Society. He was also a member of the Cam- bridge Philosophical Society. Chasles continued his labours in the cause of science without inter- ruption, from the time of his leaving the Lycée Impériale until he was eighty years of age. An interval of sixty-eight years separates the first note of the pupil Chasles, which appeared in the ‘‘ Correspondance sur l’Ecole Polytechnique,” from the last memoir he presented to the Academy. In the “ Catalogue of Scientific Papers” will be found the titles of 177 of his papers, and it is computed that the number published since 1873 would probably bring the total to nearly 270. The subjects range over curves and surfaces of the second and of any degree, geometry, mechanics (and attractions), history, and astronomy. His “ Rapport” perhaps furnishes the best key to his writings, and at pages /2—126, 220—280 will be found an account of his own contri- butions to geometry. Chasles’ life was a happy and simple one. He lived quietly and abstemiously, respected and loved by all the scientific friends whom he il was in the habit of inviting and entertaining with such true kindness and hospitality. His brethren of the Institut, as well as the nume- rous other French and foreign savants, whom he liked so much to collect around him, will not easily forget the cordial and sympathetic reception which they always met with in his hospitable salon, whether in Paris or in his country quarters at Sevres. Chasles was an active member of the Council of the Société. des Amis de France, and it 1s well known how diligently and conscien- tiously he always laboured in finding worthy objects for the charity of this institution, and how generously he frequently supplemented its work by donations from his own purse when the funds of the society did not suffice for its wants. Fifteen years before his death, when the pupils of the Ecole Poly- technique first conceived the idea of founding the Société Amicale, they perceived how important it was, in order to succeed in such an undertaking, that they should place it under the guidance of a man beloved and respected by all, and Chasles was requested to act as president. In spite of his already advanced age, and the necessity he might have urged of repose after his long labours, he accepted the post, and it is well known how zealously he devoted himself to the interests of the society, of which he must always be considered the real founder. His death took place on the 18th of December, 1880. BPH’ Ait. FSR eh ae ease ee he ee a als pa? v = ~~ > ’ 4 we hoee. Wy ‘at ia ST ee ‘ re a LY eps it f° sea pre fA | RES Pees wi atta ib ie 2 : i i, 32 a ? + f Pas in. Os BAGS ww ee Tayi eee Het ‘ P S P= - ee Ss ' =. ; pre wy = ah a re é | eS Oe ! 2 Z , ; : = . t 4 . : ve 1 ‘2 / 7 5 r . aes = q e “ S * 4 et t Es 4 ae ry rn i > =)h¢ a a S “ a f i se i i: ’ Sys \¥ i ‘pee < 4 2 ; 5 . = \ i .- < — \ ‘ E 2 ° 6 ten i ie. é i Ms ; = * - x 4 : c ; y, Ls te * A * ‘ : ~ Bs ras 1 2 " ; : . Van ria 4 - a . ; 5 an 4 E + { x PROCEEDINGS OF Ce eer SO Vy NNO NARA March 24, 1881. THE PRESIDENT in the Chair. The Presents received were laid on the table, and thanks ordered for them. The following Papers were read :— I. THE CRoonIAN LECTURE.—“ Observations on the Locomotor System of Echinodermata.” By GEoRGE J. RoMANES, M.A., F.R.S., and Professor JAMES C. Ewart, M.D. Received March 5, 1881. (Abstract. ) I.—MorPHOLOGY. In Holothuria the polian vesicle opens freely into a wide circular canal a short distance from the termination of the stone canal. From this circular canal five lozenge-shaped sinuses project forwards, and from each of these two large oval sinuses run forward parallel with each other—the ten oval sinuses becoming continuous with the hollow stems of the tentacles. Injection of the polian vesicle shows that it forms one continuous tube system with the circular canal and its sinuses, oval sinuses and tentacles, ampulle and pedicels. Unless the pressure is kept up for a considerable time there is no penetration of the injected fluid into the stone canal, and either the ring, the vesicle, or a sinus gives way before the fluid reaches the madreporic plate. Specimens injected with a gelatine mass show that each canal sinus opens into a cecal tube, which runs forwards internal to the sinuses of the tentacles as far as a wide circum-oral space. This space com- VOL. XXXII. B 2 Mr. G. J. Romanes and Prof. J.C. Ewart. [Mar. 24, municates by well-defined apertures with that portion of the body cavity which lies between the sinuses and the cesophagus, and which is reached through the circular apertures between the sinuses of the circular canal. Hach canal sinus has three other apertures in its walls. It opens by a small round aperture into a radial canal, and the two other apertures occur as minute slits, one at each side of the orifice of the radial canal leading into the adjacent tentacle sinuses. When the tentacle into which the sinus opens is protruded, there is no constriction between the sinus and the tentacle; but when the ten- tacle is retracted, there is a well-marked constriction at the junction of the sinus with the tentacle. The eversion of the perisome and the protrusion of the tentacles are effected chiefly by the shortening of the polian vesicle and the constriction of the longitudinal muscular bands, which run from the inner surface of the body wall between each two adjacent tentacle-sinuses; but the circular fibres of the body wall also assist in the process by contracting immediately behind the group of sinuses, so as to act on them by direct pressure, and also indirectly by forcing the body fluid against them. The amount of the body cavity fluid is constantly changing. At the entrance to the cloacal chamber there is a circular valve which is constantly dilating and contracting, except when the aboral end of the animal is forcibly retracted. When open, this valve allows water to pass into the respiratory tree; when it begins to retract, water escapes from the cloaca. This alternate opening and closing takes place with perfect rhythm, at a rate of about six revolutions per minute. At the end of every seventh or eighth revolution a large stream of clear water is ejected, which sometimes contains sand and the remains of food particles. When the tentacles are being pro- truded, more water is taken in at the cloaca than escapes; on the other hand, retraction of the tentacles is preceded by an escape of a large stream of water. In Echinus, two tubes spring from the under surface of the madre- poric plate. The one is dilated at its origin, so as to include the greater portion of the plate, and ends in the so-called heart; the other is small, deeply pigmented, and runs along a groove in the heart to open into a circular canal at the base of the lantern. From the under aspect of this circular canal the five radial ambulacral vessels take their origin. Immediately within the oral margin of the shell, and alternating with the inner row of pedicels, are the five pairs of ‘“‘tree-like organs.” If a fine glass cannula be forced through the membrane which extends from the apex of each tooth to the oral margin of the interambulacral plates and sides of the alveoli, coloured — fluids may be injected into the space between the membrane and the alveoli of the lantern; the fluid then slowly diffuses upwards into the vesicles around the apices of the teeth. It reaches these ee a ee ee ee ee ine Tee —«A88t.] On the Locomotor System of Echinodermata. 3 vesicles partly by passing directly upwards external to the alveoli, and partly by passing into the cavities of the alveoli and ascending through the circular sinus. In Spatangus the ambulacral circumoral canal has no polian vesicles or sinuses developed in connexon with it. Some of the pedicels have suckers, others are conical and devoid of them, while others again are flattened at their tips, and sometimes split up into segments. If one of the arms of Solaster papposa is divided transversely, and a coloured fluid introduced into the open end of the radial canal, the ampullze and pedicels of the injected arm are at once distended. The fluid next penetrates the circular canal, polian vesicles, ampulle, and pedicels of the other arms; but unless considerable pressure be kept up for some time, none of the solution enters the madreporic canal. If, however, the pressure is maintained for several hours with a column of fluid 2 feet high, the fluid ascends through the stone canal and diffuses slowly through the madreporic plate. When a thin slice is then shaved off the plate, the fluid is observed escaping from a small circumscribed area situated between the centre and the margin of the plate, and corresponding in size and position with the termi- nation of the stone canal on the inner surface. The stone canal, gradually increasing in diameter as it passes inwards from the madreporic plate, runs obliquely over its accompanying sinus, till it finally hooks round this sinus to open into the circular canal. Springing from this canal and opposite to each inter-radial space (with the exception of the space occupied by the stone canal) is a polian vesicle. The size and form of these vesicles are largely de- termined by the amount of fluid in the pedicels. In none of the in- jected specimens was there any evidence of a communication between the ambulacral vessels and the body cavity, or between the ambulacral and the blood (neural) vessels. There was, however, abundant evi- dence of communication between the latter and the exterior. When a cannula was introduced into the outer end of the sinus, a coloured solution could be easily forced through the sinus into the circular blood-vessel, and from this into the radial blood-vessels. But when the cannula was introduced into the proximal end of the sinus, the solution rapidly rushed along the sinus and escaped through the madreporic plate, proving that the blood-vessels of Solaster commu- nicate far more freely with the exterior than do the water-vessels. The ambulacral system of the common star-fish only differs from that of the sun-star in having no polian vesicles. Direction: of meh lechrics a secondary circumstance, arising from the former in consequence of the greater adhesion of the solution than of the mercury to the tube. * The repulsive action being usually at the negative pole in liquids, is a similar fact to the molecular repulsion of highly rarefied gases at the negative electrode in Mr. Crookes’s experiments. WO. XXXIT. H 98 Dr. G. Gore. The mechanical law of action and reaction of equal and opposite forces ° also requires the primary movements of the mercury and solution to be in opposite directions. If this hypothesis is correct, the unequal adhesion of the mercury and solution to the glass is a necessary con- dition of the simultaneous advance of the two liquids in the same direction. If the tube was as free to move as the liquids, it would of course move in an opposite direction tothem. The number of physical actions, however, involved in the case is probably much greater than is represented be the above explanation. Herschel observed that “the peculiar action is only exerted at the common surface of the fluids;”’ and nearly all the evidence supports the conclusion that it is primarily an action of films and not of the mass of mercury or of solution within the capillary tube. Hven when during the passage of a current, the column of liquid is prevented by adhesion from rising or falling in the capillary tube, there is probably a motion occurring in the contiguous portions of the mercury and solution, though not usually visible even by the aid of a microscope. Armstrong’s experiments (“ Phil. Mag.” vol. 23, 1843, pp. 194— 202) also indicate that the movement of the mercury is affected by an action of adhesion between the surface of the liquid and glass within the tube, because in his experiments, when the silk thread, acting ‘as a fulcrum, was removed, the amount of liquid flowing in one direction was equal to that flowing in the other, and there was no manifest transfer of the mass. In all cases with a globule of mercury in a pool of liquid, the outer film of mercury and the film of liquid in contact with it appears to move in the same direction ; the mass of the liquid, and also that of the interior of the mercury, must, therefore, of necessity, move in a direction opposite to that of the films (compare Herschel, “ Phil. Trans.” 1824, p. 165). The opposite movements obtained with dilute sulphuric acid and with solution of potassic cyanide might also be explained on another supposition, viz., that one of these liquids adheres more strongly, and the other less strongly, than mercury to glass; but this does not appear to be a fact. Although the phenomenon of to-and-fro motion appears to be essentially due to a combination of surface actions, it is not confined entirely to capillary spaces, the motion (as is well known) takes place as readily with a large globule of mercury in a pool of liquid as ina capillary tube. (See also “ Movements of Liquid Metals and Electro- lytes in the Voltaic Circuit,” ‘‘ Proc. Roy. Soc.,”’ vol. 10, par. 35.) The to-and-fro movement, in the case of a globule of mercury in a pool of liquid, is dependent upon the circumstance that the solid surface of the containing vessel, upon which the mercury and film of Phenomena of the Capillary Electroscope. a9 solution beneath it lie, constitutes a fulcrum upon which the moving layers of liquid mechanically act. If it were possible to suspend the mercury in the midst of a large mass of a liquid electrolyte of its own specific gravity, the effects would probably be greatly modified, and there would be very little translation of the mass of mercury (com- pare Herschel’s paper, section 10, p. 167). In the experiments by Armstrong, also those of Quincke and Jiir- gensen, of the mechanical transfer of liquids by the passage of electric currents through slender columns of them in tubes, the axial portion of the liquid was observed to move in an opposite direction to the outer layer. This inner moving portion, or that farthest away from the surfaces of adhesion, may be regarded as the return current. It is generally considered that the surface layer of particles of every solid and liquid is in a state of mechanical tension, and con- sequently that every mass of such substance, even that which is only of microscopic magnitude, may be crudely viewed as being bounded by a more or less tightly-fitting envelope. This circumstance also appears to be related to the phenomenon of motion of the mercury. The surface tension of that metal appears to be lessened at the positive mercurial electrode, and increased at the negative one in nearly ail electrolytes. As there is also a surface tension of the liquid as well as that of the mercury, the primary motion is probably a resultant of the two. According to this view there exists two modes by which an electric current may vary the surface tension of a con- ductor in contact with an electrolyte, the one being attended by electrolysis, and the other not. If it were simply a.result of diminu- ‘tion of surface tension of the mercury by the passage of an outgoing current, then it ought not to vary in direction by variation of liquid, strength of liquid, or diameter of tube. As adhesion influences so greatly the action of the instrument, perfect freedom from dust is a most important condition of success, and the most effectual way of securing this is to insert the capillary tube in the pressure-chamber as soon as it is made, and fill it at once with clean mercury. Multitudes of points of adhesion are met with in tubes which have long been open to the atmosphere. The pressure tube should also be completely freed from dust by previously agitating successive portions of mercury init. Dust, not moisture, is the great source of failure. Adhesions of the mercury do not usually interfere unless they happen to be within the range of movement of the meniscus. The influence of friction I have not examined ; it plays, however, an important part in the practical use of the instrument. 7. Relations of the Movements to Heat and Temperature. As heat, applied to the junction of mercury and an electrolyte, H 2 100 DroG. Gore: produces an electric current, it is reasonable to infer that an electric current passed through such a junction in the same direction as the one produced, would tend to lower the temperature. And as it is well known that both the cohesion of liquids and their elevation are diminished by rise of temperature, it is probable that the electric current, by producing a minute change of temperature at the junction, affects to a slight extent the capillary elevation. By comparing, how- ever, the directions of the electric currents obtained by heating mercury in contact with various solutions which exercise no chemical action upon it (see “ Thermo-Hlectric Behaviour of Aqueous Solutions with Mercurial Electrodes,” “‘ Proc. Roy. Soc.,” vol. 29, p. 472), with the directions of movements produced by given directions of electric currents in the same solutions in these experiments, it may be perceived that they do not to any large extent coincide. That the mercury should also conversely suffer slight changes of temperature by electro-capillary action, is in accordance with the discovery made by Pouillet (‘‘ Ann. de Chemie et de Phys.,”’ vol. xx, 1822, pp. 141-162), that heat is evolved by the capillary absorption of liquids by solids; and with the experiments of Jungk (‘“‘ Phil. Mag.,” vol. 2, 1876, p. 454; and ‘‘Pogg. Ann.,” vol. cxxv, p. 292), and further, with those of Melsens, who states that by mixing 4°40 grms. of charcoal and 33 of bromine, the rise of temperature was 35° (‘‘ Phil. Mag.,” vol. 2, 1876, p. 454; also “ Mém. de lAcad. Royal de Belgique,” vol. xxii). The movement is evidently not caused by heat of conduction resistance, because that would produce expansion and advancing motion only. If also the motion is simply due to thermic expansion, then heat of chemical combination at the anode in cases of electrolysis in the capillary ought to cause the mercury to move downwards, but with solutions of potassic cyanide it moves the reverse. The relations of electro-capillary movements to heat require, however, much more investigation. 8. Relations of the Movements to Electric Conditions. The phenomena of the capillary electroscope are not results of electrolysis, nor of disruptive discharge, but of conduction proper, and may occur either with or without electro-chemical change. In nearly all previous investigations of the electric movements of mer- cury, very much stronger electric currents were employed, and the results were complicated by electrolytic phenomena. That electro- lysis, when it does occur, is only a coincident circumstance, is proved by the non-liberation of hydrogen at the meniscus when that surface is made negative in dilute sulphuric acid and_in various other solutions; also, by the mercury at the meniscus not becoming viscid when made the negative pole in a solution of cupric sulphate (see Phenomena of the Capillary Electroscope. 101 Exp. No. 27); and further, by the surfaces of the meniscus not becoming oxidised when made the positive pole in either of the various solutions usually employed; provided in all these cases that ordinary chemical action is absent and that the current is sufficiently weak. The reverse directions of movement produced by the same direction of current in dilute sulphuric acid and in solution of potassic cyanide, probably cannot both be explained by the theory that the movements are due to electro-chemical exidation and deoxidation. The upward movement caused by a downward current can hardly be due to electro- lysis, because that action, by continually destroying the outer film of mercury, tends to diminish its surface tension. Being largely dependent upon electric conduction, the amount of movement of the mercury is affected by every circumstance which alters the conductivity. The part of the circuit which offers the greatest amount of conduction resistance is the slender column of solution between the meniscus and the end of the capillary tube; there the resistance is considerable. The nearer, therefore, the meniscus is to the end of the tube, and the shorter the column of solution (as well as that of the mercury), the more rapid is the move- ment, especially if the tube is slightly larger in diameter towards that end; this circumstance also tends to make the downward movement an accelerated one, even in a tube of uniform diameter, and the upward movement the reverse; and also accounts for the fact that the instrument is more sensitive to a downward current than to an up- ward one, unless the tube becomes narrower downwards at too rapid a rate. The secondary current of an induction coil was not suitable for working the instrument, because it produced electrolysis. On many occasions an electrophorus was used as the source of electricity, and this also produced a similar effect. On charging the electroscope with it, either by induction or by contact, the movements were freely produced, provided the electroscope was not insulated. If it was perfectly insulated the movements did not occur, and if it was imperfectly insulated and then charged by momentary contact, the mercury continued to run out at the end of the capillary tube after removal of the electrophorus, and ceased to flow by discharging the instrument. I made a capillary tube in accordance with the annexed sketch, with a pressure tube A, and a platinum wire electrode Pt; and filled IPTG 3. 102 Dr. G. Gore. it with mercury whilst in a horizontal position; then melted it off at the point B by means of a minute flame; broke it off at C, and fixed it vertically with its end C in dilute sulphuric acid above an electrode. Not a trace of air was visible in any part by the aid of a strong magnifying glass. By charging the upper electrode by induction with an electrophorus, the mercury descended freely; and by charging the lower one it ascended, provided the free end was not insulated. The motion of the mercury, therefore, is evidently not produced by mere electric charge but requires electric flow. Even a residual electric charge of the ebonite base of the reverser (see ‘Proc. Roy. Soc.” vol. 30, p. 32) at the surfaces of contact of that substance with the brass fittings which had been connected with the poles of the nearly exhausted little water-cell was sutf- ficient to work the instrument. Also a voltaic current which raised the meniscus 19°5 millims. would pot produce a visible movement of the needle of a torsion galvanometer having a coil of 100 ohms resistance. These facts illustrate the extreme sensitiveness of the instrument to electric flow of the feeblest tension; and the apparatus might be used for examining the conductivity proper of electrolytes. From the results achieved by Quincke (‘“‘ Pogg. Ann.,” vol. exiii, 1861), and by Jurgensen (Reichert and Du Bois Reymoend’s “ Ar- chiv,” 1860, p. 573; also ‘‘Chemical Physics,’ by W. A. Miller, 4th Edition, pp. 530-533), and the various results obtained by myself, I conclude that the primary mechanical movement in the instrument is due to a more or less charged electric state of the surfaces of the liquids and tube. The electric tension accompanying that state alters the degrees of adhesion of the substances to each other, and produces electric convection, which, in consequence of the unequal adhesion of the solution and mercury to the tube, produces a to-and-fro movement of the mass. The less fundamental results, such as reversal of direction of movement with solutions of potassic cyanide, or with tubes of different diameter, would probably be found to be necessary results of the above causes under the altered conditions. A brief translation of Quincke’s explanation of somewhat similar phenomena may be found in W. A. Miller’s “ Cloner Physics,” 4th Hdition, p- 532. The primary action appears to consist nearly wholly of a direct conversion of electricity into mechanical power; and this accords with the great sensitiveness of the instrument and the comparatively considerable force of the movements. This force has been already applied by Lippmann in the construction of an electro-capillary engine. The converse action, viz., the production of electric currents by mechanically raismg and lowering mercury and an acid solution in capillary tubes, has also been obtained by Lippmann (‘ Phil. Mag.,” x vol. 47, 1874, p. 281). Phenomena of the Capillary Electroscope. 103 9. Influence of Chemical Action. The movements do not appear to depend upon the chemical nature of the solution, because they take place equally well with acid, alkaline, and neutral liquids. Being also purely physical, they are not dependent upon chemical action; such action, when it does occur, appears in every case to interfere with them. Note.—Since the publication of a previous communication ‘‘ On the Capillary Electroscope” (‘‘ Proc. Roy. Soc.,” vol. 30, p. 32), I have been favoured by M. Lippmann with the following remarks respecting that instrument. “1st. The liquid is to be diluted sul- phuric acid, containing something like one-third its weight of sul- phuric acid. Weak acid does not film glass properly; most liquids do not; and then stoppages, or a jumping motion of the mercury, occur, such as you have described. 2nd. The capillary tube is to be cut very short (to about 10 millims.), the motions are in that case ten times more rapid than in a tube of 10 centims., because the friction is reduced in that proportion; besides, possibilities for obstruction are reduced also in the same proportion. 38rd. The instrument is only fit for measuring electromotive forces smaller than one Daniell; by using over-great electrdmotive forces the capillary constant goes over its maximum value, and then the movement of the mercury is reversed as you noticed it to be the case at the end of your communication (see, about this maximum, ‘ Ann. Chimie et Physique,’ 1875, and also 1877). If you will do me the pleasure of visiting M. Jamin’s laboratory in the Sorbonne, you will find there several electrometers in good working order; three of these are being used by different observers for separate researches, with a precision of af = +] sais Daniell. 104 Lord Rayleigh and Dr. A. Schuster. [May 5, May 5, 1881. THE PRESIDENT (followed by THE TREASURER) in the Chair. The Presents received were laid on the table, and thanks ordered for them. In pursuance of the Statutes, the names of Candidates recommended for election into the Society were read from the Chair, as follows :— Ayrton, Prof. William Edward. McLeod, Prof. Herbert, F.I.C., Bates, Henry Walter. HE@zS: Bristowe, John Syer, M.D., | Phillips, John Arthur. F.R.C.P. Preece, William Henry, C.H. Christie, William Henry Mahoney, | Samuelson, Bernhard, M.1I.C.H. M.A., Sec. R.A.S. Stoney, Bindon Blood, M.A., Dickie, Prof. George, A.M., M.D., M.1.C.E. ES: Traquair, Ramsay H., M.D. Kempe, Alfred Bray, B.A. Watson, Rev. Henry William, Macalister, Prof. Alexander, M.D., | M.A. Sec. R.LA. Wright, Charles R. Alder, D.Se. The following Papers were read :— I. “On the Determination of the Ohm in Absolute Measure.” By Lord RAyYLEIcH, F.R.S., and ARTHUR SCHUSTER, Ph.D., F.R.S. Received April 12, 1881. Part I.—By Lord RAYLEIGH. It is generally felt that considerable uncertainty still attaches to the real value of the ohm, or British Association unit of resistance. The ohm was constructed to represent 109 C.G.S. absolute units, but according to Kohlrausch* it is nearly 2 per cent. too great, and according to Rowland} nearly 1 per cent. too small. On the other hand, H. Weber has obtained by more than one method results very nearly in harmony with those of the British Association Committee. Influenced partly by the fact that the original apparatus (though a good deal out of repair) and the standard coils themselves were in * “Phil. Mag.,” xlvii, p. 294, 1874. + “ American Journal of Science and Arts,” 1878. = “ Phil. Mag.,” v, p. 30, 1878. 1881.| Determination of the Ohm in Absolute Measure. 105 the Cavendish Laboratory, I determined last June to repeat the measurement by the method of the Committee, which has been employed by no subsequent experimenter, and sought permission from the Council of the British Association to make the necessary altera- tions in the apparatus. In this way I hoped not merely to obtain an independent result, but also to form an opinion upon the importance of certain criticisms which have been passed upon the work of the Committee. The method, it will be remembered, consists in causing a coil of insulated wire, forming a closed circuit, to revolve about a vertical axis, and in observing the deflection from the magnetic meridian of a magnet suspended at its centre, the deflection being due to the currents developed in the coil under the influence of the earth’s magnetism. The amount of the deflection is independent of the intensity of the earth’s magnetic force, and it varies inversely as the resistance of the circuit. The theory of the experiment is explained very fully in the reports of the Committee,* and in Maxwell’s “ Hlectricity and Magnetism,” section 763. For the sake of distinct- ness, and as affording an opportunity for one or two minor criticisms, a short statement in the original notation will be convenient :— H=horizontal component of earth’s magnetism. y==strength of current in coil at time 7. G=total area inclosed by all the windings of the wire. w=angular velocity of rotation. @=wt=angle between plane of coil and magnetic meridian. M=magnetic moment of suspended magnet. @=angle between the axis of the magnet and the magnetic meridian. K=maenetic force at the centre of the coil due to unit current in the wire. L=coefficient of self-induction of coil. R=resistance of coil in absolute measure. MHz=force of torsion of fibre per unit of angular rotation. _ The equation determining the current is— Ll + Ry=HGw cos at-+ MKw cos (at) Steneeee Ries 6): whence Ww oe Be L?w? {GH(R cos 6+ Lw sin @) +KM(R cos (6—¢)+Lasin (0—¢))} . (2). * Collected in one volume. London, 1873. - Lz 106 Lord Rayleigh and Dr. A. Schuster. [May 5, If L were zero, or if the rotation were extremely slow, the current would (apart from KM) be greatest when the coil is passing through the meridian. In consequence of self-induction, the phase of the current is retarded, and its maximum value is diminished. At the higher speeds used by the Committee, the retardation of phase amounted to 20°. To find the effect of (2) upon the suspended needle, we have to introduce MK and the resolving factor cos (@—®@), and then to take the average. This, on the supposition that the needle remains on the whole balanced at ¢, must be equal to the force of restitution due to the direct action of the earth’s magnetism and to torsion, 7.e., MH sing+MH7¢. Thus— $MKw R24 Lew? In the actual experiment 71s a very small quantity, say 5; and the distinction between 7 ¢ and 7sin dé may be neglected. R2— Rae + Te sec) +L? — Sa 6, T {GH(R cos $+ Lw sin $) + KMR} —MH(sing+7¢)=0. If we omit the small terms depending upon 7 and upon MK/GH, we get on solution and expansion of the radical— R=4GKw cot oi1— aa(ae-1) te n? P— (= (= 1) tant ¢.. + (4). The term in tan‘@ is not given in the report of the Committee, but, as I learn from Mr. Hockin through Dr. Schuster, it was included in theactual reductions. But the next term in tan® ¢, and one arising from a combination of the correction for self-induction with that depending on M, are not altogether insensible, so that probably the direct use of the quadratic is more convenient than the expansion. At the high speeds used by the Committee the correction for self- induction amounted to some 8 per cent., and therefore SHEE be treated as very small. If the axis of rotation be not truly vertical, a correction for level is necessary. In the case of coincidence with the line of dip, no currents, due to the earth’s magnetism, would be developed. If the upper end of the axis deviate from the vertical by asmall angle 8 towards the north, the electromotive forces are increased in the ratio cos (I+): cos I, z.e.. mm the ratio 1+tan 1.6, I being the angle of dip. A devia- tion in the east and west plane will have an effect of the second order only. The magnetic forces due to the currents will not act upon the needle precisely as if the plane of the coil were always vertical, but the difference is of the second order, so that the whole effect of a 1881.] Determination of the Ohm in Absolute Measure. 107 small error of level may be represented by writing G (1+tanI. 3) for G in (8) or (4). The next step is to express GK in terms of the measurements of the coil. In order that there may be a passage for the suspending fibre and its enveloping tube, it is necessary that the coil be double, or if we prefer so to express it, that there be a gap in the middle. If a=mean radius of each coil, n=whole number of windings, b=axial dimension of section of each coil, c=radial dimension of section of each coil, b'=distance of mean plane of each coil from the axis of motion, a=angle subtended at centre by radius of each coil, so that cot a= b'/a, then— (5), Ka" sin? a {lta S2- —15 sin? « cos? «) eo taalage LG) so that ted 2 Gk=2 na sin? ! 1+ S44 pve rae Ok a COS? hee ed sin” \ Cie The correction due to the finiteness of } and ¢ is in practice extremely small, but the factor sin® « must be determined with full accuracy. In order to arrive at the value of MK/GH, which occurs in (3), we observe that the approximate value of K/G is 2sin’a/a?; so that 108 Lord Rayleigh and Dr. A. Schuster. | May 5, MK/GH is equal to tan, where p» is the angle through which the needle of a magnetometer is deflected when the suspended magnet (M) is placed at a distance from it a/sina to the east or west, with the magnetic axis pointing east or west. In practice the difference of readings when M is reversed is taken in order to double the effect, and any convenient distance is used in lieu of a/sin a, allowance being easily made by the law of cubes. The correction for torsion is determined by giving the suspended magnet one (or more) complete turns, and observing the displacement. If this be 6,, reckoned in divisions of the scale, 7.e., in millimetres, and D be the distance from the mirror to the scale reckoned in millimetres, ai AD i (8). The correction for scale reading, necessary in order to pass from 4 tan 2¢ to tan ¢, will be explained under the head of reductions. Corrections depending upon irregularity in the magnetic field, and in the adjustment of the magnet to the centre of the coil are given in the report. They are exceedingly small. The same may be said of errors due to imperfect adjustment of the coil with respect to the axis of rotation. In remounting the apparatus the first point for consideration was the driving gear. The Committee used a Huyghens’ gearing, driven by hand, in conjunction with a governor. This, it appeared to me,. might advantageously be replaced by a water-motor; and Bailey’s ‘“‘ Thirlmere ” engine, which acts by the impulse of a jet of water upon revolving cups, was chosen as suitable for the purpose. As the pres- sure in the public water pipes is not sufficiently uniform, it was at first intended to introduce a reducing valve; but on reflection it seemed simpler to obtain a constant head of water by connecting the engine with a small cistern at the top of the building. This cistern is just big enough to hold the ball-tap by which it is supplied, and gives at the engine a head of about 950 feet. The success of this arrangement depends upon attention to princi- ples, as to which it may be well to say a few words. The work done by many prime movers is within practical limits proportional to the speed. If the work necessary to be done in order to overcome resist- ances, aS n overcoming solid friction, or in pulling up weights, be also proportional to the speed, there is nothing to determine the rate of the engine, and in the absence of an effective governor the motion will be extremely unsteady. In general the resistance function will be of the form— Bv+Cv?+Dv?+ ..., in which the above-mentioned resistances are included under B. The- 1881.] Determination of the Ohm in Absolute Measure. 109 term in C will represent resistances of the nature of viscosity, and that in D a resistance such as is incurred in setting fluids in motion by a fan or otherwise. By these resistances, if present, the speed of working will be determined. In the water impulse engine, however, the work is not proportional to the speed. At zero speed no work is done; neither is any work done at a speed such that the cups retreat with the full velocity of the jet. The speed of maximum efficiency is the half of the last, and the curve representing work as a function of speed is a parabola with vertex directed upwards. If we draw upon the same diagram the curve of work and the curve of resistances, the actual speed will correspond to the point of intersection, and will be well or ill defined according as the angle of intersection is great or small. At the higher speeds of the coil (four to six revolutions per second) so much air is set in motion that the resistance curve is highly convex downwards, and no difficulty is experienced in obtaining a nearly uniform motion. But when the speed of rotation is as slow as once a second, the principal resistance is due to solid friction, and the requisite cur- vature in the diagrams must be obtained in the curve of work. It was necessary in order to obtain a satisfactory performance at low speeds to introduce an additional reducing pulley, so that the engine might run fast, although the coil was running slow. The revolving coil with its frame, and the apparatus for suspending the magnet, were at first arranged as described by the Committee. This description, with drawings, is to be found in the report, and it is reproduced in “ Gordon’s Electricity and Magnetism,” vol. i. The water engine was ready about the middle of June, and towards the end of the month the apparatus was mounted by Mr. Horace Darwin. During July and August preliminary trials were made by Mr. Darwin, Mrs. Sidgwick, and myself, and various troubles were encountered. The only point in which the arrangement adopted by the Com- mittee was intentionally departed from was in the connexion of the magnet and mirror. The magnet is necessarily placed at the centre of the revolving coil, but in their arrangement the mirror is on the top of the frame and is connected to the magnet by a brass wire. In order to save weight, I preferred to have the magnet and mirror close together, not anticipating any difficulty from the periodic and very brief interruption caused by the passage of the coil across the line of sight. A box was, therefore, prepared with a glass front, through which the mirror could be observed, and was attached to the end of a brass tube coming through the hollow axle of the coil. This tube itself was supported on screws resting on the top of the frame. The upper end of the suspension fibre was carried by a tall tripod resting independently on the floor. The first matter for examination was the behaviour of the magnet 110 Lord Rayleigh and Dr. A. Schuster. [May 5, and mirror when the coil was spinning with circuit open. At low speeds the result was fairly satisfactory, but at six or more revolutions per second a violent disturbance set in. This could not be attributed to the direct action of wind, as the case surrounding the suspended parts was nearly air-tight, except at the top. It was noticed by Mr. Darwin that even at low speeds a disturbance was caused at every stroke of the bell. This observation pointed to mechanical tremor, communicated through the frame, as the cause of the difficulty, and the next step was to support the case surrounding the suspended parts independently. ] 0 “4 3) i ” 5 PP] HT 0 0-2 4 Ores ae 0-05 9 40 ” 40 9 In both sets of observations the inverse time-rate is seen to come out true with almost mathematical accuracy. When, however, a relatively small quantity of pancreatic extract was employed, and the time required to reach the achromic point was, in consequence, considerably lengthened, it was found that the advent of the achromic point was postponed beyond the term indi- cated by the rule. If the period occupied in reaching the achromic point fell within the compass of an hour, and the temperature was low, as in the observation just recorded, the inverse time-rate came out true, but when the period of action extended to several hours and the temperature stood higher, the departure from the rule was un- doubted. The annexed table gives the results of an experiment made with a view of testing this point. 150 Dr. W. Roberts. Estimation of the Amylolytic [May 5, Table IT. Experiments showing the postponement of the achromic point when the action is protracted. The quantity of standard mucilage acted on in each experiment was 10 cub. centims. diluted with water up to 100 cub. centims. Temperature 40°C. The “calculated ” time in the third column was obtained by taking the first observation, which was several times repeated, as a standard of comparison. —EE Time in which the achromic point Quantity of pancreatic was reached. extract employed. Found. Calculated. 0:05 cub. centim. 10 minutes. —. 0-005 a 115 Ah 100 minutes. 0-004 a TO} Wh 125 nine 0 002 ¥ NOS doves 250, 0:0005 10380) 4) be 1000 ae It need scarcely be said that when the enzymosis is very slow it is not possible to fix the vanishing point of the colour reaction with the same precision as when the action is more rapid and the change more abrupt. Notwithstanding this source of error, I think the conclusion indicated by the experiments may be relied on. The postponement of the achromic point shown in the table may he explained, as has been suggested, on the assumption that the enzyme liberates its energy at a continually retarded rate. In the case of trypsin, we shall see evidence of a precisely parallel phenomenon. Temperature-—The action of pancreatic diastase on starch mucilage was found to increase in energy (or speed) from zero up to 30°C. From this point to 45° the rate of action continued steady, showing a range or platform of indifferent temperature extending from 30° to 45°. Above 45° the action became iess and less energetic, and finally ceased between 65° and 70°. The following table exhibits the results obtained at various temperatures between 5° and 70°. Table IIL. Showing the effects of temperature on the action of pancreatic diastase ; the amount of the standard mucilage acted on in each experi- ment was 10 cub. centims. diluted with water up to 100 cub. centims. The quantity of pancreatic extract employed in each experiment was 0-1 cub. centim. “N 1881.] and Proteolytic Activity of Pancreatic Extracts. 151 Temperature. Achromic point reached in BOIS | ae eee dees 36 minutes. LOLS: ger Reger Or 43 18 50 LOnig Fey eehaeeesees wees 12 3 ADs erahid- i wretstet Seberebetets salt 8 5 DoOwireman wel sez eres 6 " SO iims: er Taevet ste « aespaye 3 3) 26 Aiea giles Shams 4) 3 A Dee Hh iets Ae ad 5) 5 Ds haya) teas Sek: 7 9 DOGG WT ats eerste eae BG 10 - COM ahs. pe deka eee 40 56 Oo peal jes A203 sc tela s Very slow action. OAD Meath ai. lt Xs No action. These results, thrown into the form of a curve, are shown in the subjoined diagram. The ordinates indicate the diastasic value, or D, as calculated by a method to be presently explained; the abscisse represent the temperatures. Curve illustrating the effect of temperature on the action of pancreatic diastase. Mode of proceeding.—In testing the activity of a sample of pan- creatic extract, it was found on the whole more convenient to operate on a fixed quantity of the standard mucilage, and to vary the quantity of extract added to it, than to proceed contrariwise. The bulk of liquid operated on was thus kept constant. The ordinary proceeding was as follows: 10 cub. centims. of the standard mucilage were mixed in a beaker with 90 cub. centims. of water. The mixture was then warmed to 40° C., or at least to some point well within the range of indifferent temperature extending from 30° to 45°. This was done in order to eliminate the disturbing influence of temperature. The next step was to add the determined quantity of the extract to be tested to the diluted mucilage, and to note the exact time. Then, at short intervals, a drop of the enzymosing liquid was placed on a white slab, or plate, with a drop of the iodine solution. The time and result VOL. XXXII. M 152 Dr. W. Roberts. Estimation of the Amylolytic [May 5, of each testing was noted. When the achromic point was reached the time was marked, and the interval from the commencement of the experiment was computed. If at the end of three minutes the mixture still gave the blue reaction of unaltered starch, a new experi- ment was made, using two, three, or four times the quantity of extract. If, on the other hand, the achromic point was reached in less than two minutes, a new experiment was made, using a smaller quantity of the extract. Two or three experiments generally sufficed to determine the quantity of extract required to bring the achromic point within a period ranging from two to ten minutes. A final control experiment enabled the operator to fix the achromic point somewhere between four and six minutes. The accuracy of the method depends chiefly on the sharpness and precision with which the occurrence of the achromic point can be determined. If it occur earlier than two minutes, the transition is too rapid for exact observation and record. On the other hand, if it occur later than fifteen or twenty minutes the transition is too gradual for precise limitation. The most satisfactory results are obtained when the achromic point falls between four and six minutes. The following example will serve as an illustration of the way in which the experiments were carried out, noted, and expressed :— Table IV. 10 cub. centims. standard starch mucilage+90 cub. centims. water. +0°1 cub. centim. pancreatic extract—at 40° C. Time. Reaction with iodine. 10.30 am. ...... Commencement of experiment. 1.0.31 Wa ee Blue. 10.32 tee oe Violet. HOSS? Ga en ites aa: Brown. PO BA wie ear Yellowish-brown. LOSO2: Gy We ee Pale yellow. VOB Oe oe ireiier ae No reaction—achromic point. 6 minutes. Achromic point reached in 6 minutes. The result of the experiment was expressed in the first instance as follows : 0°1 cub. centim. pancreatic extract + 10 cub. centims. standard mucilage = 6 minutes at 40° C. From this somewhat incongruous expression it is however easy to extract by a simple formula a correct and convenient expression - for the diastasic value of any amylolytic solution. Mode ‘of Calculating and Hzpressing the Diastasic Value. The principle of the method consists, as already stated, in ascertaining 1881.] and Proteolytic Activity of Pancreatic Extracts. 153 the amount of starch mucilage of known strength which can be trans- formed by a unit measure of the diastasic solution to the point at which it ceases to give a colour reaction with iodine, in a unit of time and at a given temperature. In reducing this principle to a definite formula it was necessary to choose arbitrarily a unit of measure and a unit of time. The unit of measure fixed on was | cub. centim., and the unit of time 5 minutes. These selections seemed, on the whole, the best adapted for furnishing a convenient scale. On these bases the formula took the following form: the diastasic value of any solution—or D—is expressed by the number of cubic centimetres of the standard starch mucilage which can be transformed to the achromic point by 1 cub. centim. of the solution to be tested in a period of five minutes at a given temperature. In the process of testing the quantity of the standard mucilage was made constant, namely 10 cub. centims., and the quantity of pancreatic extract and the time were made variable. In order to get the value of D the results must be so transformed as to make the quantity of extract and the time constant, and the quantity of the standard mucilage variable. This is accomplished by increasing or reducing the ‘quantity of pancreatic extract employed to 1 cub. centim., and in- creasing or diminishing the standard mucilage in the same proportion. ‘The product thus obtained is again increased or reduced in the same proportion as is requisite to increase or reduce the time found to five minutes. Taking the example above given, the value of D is obtained by the following formula: Let p signify the quantity of pancreatic extract employed, and m the number of minutes required to reach the achromic point, then :— 10 x Dry) p> mM and in the above example— Dey °=83 at 40° C. The value of D, as already explained, signifies the number of cubie centimetres of the standard starch mucilage which can be changed to the achromic point by 1 cub. centim. of the diastasic solution in five minutes at a given temperature. As the standard mucilage contains I per cent. of dry starch, the value of D divided by 100 gives us the same value in terms of dry starch, and the result of the above experi- ment may be read as follows :— D=83=0°83 grm. of dry starch. This method of diastasimetry is equally applicable to saliva and malt-diastase. It may also be applied to the estimation of the M 2 154 Dr. W. Roberts. Estimation of the Amylolytic [May 5, diastasic agent which is present in urine, and presumably to all dias- tasic solutions. In the case of solid preparations containing diastase like malt or glandular tissue—a solution in known proportions must first be prepared; and from the ascertained activity of such solution the proportionate activity of the solid substance can be easily calcu- lated. I may here mention some of the results which this method has already yielded. Pancreatic Tissue.—The pancreatic tissue of the pig (obtained from animals killed for the market in the fasting state) yielded an extract which, when made on the large scale, possessed a mean diastasic value of 100. This extract is sent out by Mr. Benger, of the firm of Mottershead and Co., Chemists, Manchester, under the name of ‘‘ Liquor Pancreaticus,” and is made in the proportion of one part of pancreatic tissue to four of solvent (water containing 25 per cent. rectified spirit). This value indicates that 1 grm. of the moist pan- creas of the pig is capable of transforming 4 grms. of dry starch to: the point at which it no longer gives a colour reaction with iodine, in five minutes, at a temperature of 40° C. The pancreatic tissue of the ox and sheep yielded an extract (made in the same proportions) which was of far inferior activity. The ox extract hada diatasic value of about 11 and that of the sheep of about 12. These numbers indicate that in point of diastasic activity the pancreas of the pig has ten times the value of the pancreas of the ox and sheep. This extraordinary difference is probably linked with the diversity of their food. The pig is fed largely upon potatoes and meal, which are rich in starch ; the ox and sheep, on the other hand, feed on grass, which is poor in starch. We shall presently find that there is no such difference in regard to tryptic pe in the pancreas of these animals. Human Saliva.—Filtered saliva was found to Here a diastasic value varying from 10 to 17 at 40° C. Its action was influenced by tempera- ture exactly in the same manner as that of pancreatic extract. It increased in energy up to about 30° C., continued steady from this point to about 45°, and then declined, being finally extinguished between 65° and 70°. Mali Diastase—Infasions of malt made in the proportion of one part of crushed malt to four parts of water, exhibited a diastasic value of 4 to 5 at 40° C. But malt diastase did not attain its maximum activity at this temperature. It continued to increase in energy up to about 60° C., when it showed a diastasic value of 10. Above 60° the action diminished in energy, but did not come to a full stop until the tempera- ture approached 80° C. Human Urine.—Several specimens of healthy urine were tested by this method. They showed a diastasic value varying from 0-03 to 0-13 at 40° C. The effect of temperature thereon was not examined. 1881.] and Proteolytic Activity of Pancreatic Extracts. 155 Estimation of the Proteolytic Activity of Pancreatic Kxtracts—Trypsimetry. The writer had found in previous inquiries that when milk is sub- jected to digestion with pancreatic extract, a striking change takes place in it at an early stage of the process. The milk acquires the pro- perty of curdling when boiled. The onset of this reaction occurs at an earlier or at a later period according to the activity of the extract and the quantity of it employed; and it is possible to fix the time of its advent with considerable accuracy—suflicient accuracy to serve as the basis of a method of measuring the proteolytic activity of pancreatic extracts. | The reaction in question depends on the production, as a first step in the pancreatic digestion of casein, of a modified form of that body which I have named metacasein. This substance resembles casein in being curdled by acetic acid in the cold; but it differs from casein in being also curdled by simple boiling. These two reactions together distinguish metacasein from other proteid bodies. The property of curdling when boiled, which may be called the metacasein reaction, continues observable in milk undergoing tryptic digestion until near the termination of the process; it then disappears somewhat abruptly, and the milk, when boiled, remains fluid just as it did at first. We may, therefore, speak of the onset point of the metacasein reaction, and of the vanishing point of the metacasein reaction. These two points mark respectively the initial and the terminal limits of the principal phase in the digestion of milk by pancreatic extract. Before the onset point of the reaction—that is, distinct and un- doubted curdling—is actually reached, its approach is indicated by an appearance of soiling of the sides of the test-tube in which the milk has been boiled. This appearance is due to incipient coagulation, which presently developes into pronounced curdling, and is a useful sign in testing to indicate the coming on of the metacasein reaction. The following typical experiment may serve to give the reader a clear notion of the succession of events—so far as they concern us here—which occur when milk is submitted to digestion with pan- ereatic extract. Table V. 4 cub. centims. pancreatic extract added to 50 cub. centims. milk diluted with water to 100 cub. centims. Temp. 18° C. Time. Reaction on boiling. 2minutes .. No change. Oe 3) .. Slight soiling of the sides of the test-tube. Aiea? 55 .. More soiling. Dy 35 .. Distinct curdling—onset point of the meta- casein reaction. 156 Dr..W. Roberts. Hstimation of the Amylolytic [May 5,. Time. Reaction on boiling. 6 minutes .. More pronounced curdling. LO toS0", .. Pronounced curdling. Ot ae .. Diminished curdling. Or 5; .. Shlght curdling. LOOT, .. No change; vanishing point of the metacasein reaction. The length of time during which the successive steps of the transfor- mation may continue observable depends on the energy of the action ; and this, in its turn, depends on the activity of the preparation and the quantity of it added to the milk; it is also greatly influenced by temperature. By using an excess of an active pancreatic extract, and with a favourable temperature, all the steps of the process may be crowded almost into an instant of time; with converse conditions the action may linger on for many hours. Although milk is a secretion of somewhat variable composition, the oscillations which it exhibits, when it is the product of a dairy, and is not intentionally adulterated, do not materially vitiate it for the purposes of a test fluid such as is here required. The milk delivered at my house presented very little variation. It had a density of 1030 —seldom varying more than a degree from this point—and the results obtained with the milk of different days showed a remarkable uni- formity. Milk from different dairies, and at different seasons of the year, would no doubt present greater irregularities. Milk should,. however, be used fresh, for if it have become slightly acid, as it is apt to do in keeping, the results obtained are untrustworthy. If milk be diluted with water the occurrence of the metacasein reaction is postponed ; and the degree of postponement varies with the degree of dilution. For example, if 50 cub. centims-of pure milk are changed to the onset point of the metacasein reaction in three minutes, the same quantity of milk diluted with an equal volume of water will take six minutes to reach the same point—other conditions being equal. There are, however, several advantages in using diluted milk instead of pure milk as the experimental fluid. The inequalities of the milk are thereby minimised. The “strike” of the reaction is more sharply defined, and the required quantity of pancreatic extract can be included in the water of dilution. This last is an important advantage, because if the extract to be tested is feeble, a considerable quantity of it requires to be added, and this, if pure milk were employed, would seriously alter its degree of dilution, and thereby vitiate the results. In the following experiments, milk dilated with an equal bulk of water was invariably employed ; and if the quantity of pancreatic extract to be added exceeded 3 cub. centims. for every 50 cub. centims. of milk, this was always included in the water of dilution. In principle the method of trypsimetry here proposed consists im 1881.] and Proteolytic Activity of Pancreatic Extracts. 157 ascertaining how many cubic centimetres of milk can be changed to the onset point of the metacasein reaction, in 5 minutes, by 1 cub. centim. of the extract to be tested, at a given temperature. In settling the data on which the method is based, it was necessary, as in the case of diastase, to determine the relations to tryptic action of quantity, time, and temperature. Quantity and Time.—The rule of inverse relations between quantity and time which was found to be valid within a wide range in the case of diastase and starch, is only reliable in the case of trypsin and milk within narrow limits. When the time of action exceeds 8 or 10 minutes the advent of the metacasein reaction is postponed beyond the term indicated by the rule of inverse proportion, and this postpone- ment increases as the time of action is lengthened. The following two sets of observations may be taken as samples of the results obtained by experiment. Table VI. Showing the postponement of the metacasein reaction. The quantity of milk acted on in each experiment was 50 cub. centims. diluted with water up to 100 cub. centims. The “calculated ’’ time in the third column was obtained by taking the first observation in each set as the standard of comparison. Onset point of the metacasein reaction. Quantity of pancreatic extract added. Found. Calculated. ——__—_ ——<——— Set I—Temperature 40° C. 1:0 cub. centim. 3 minutes. 0°8 < 4 i 3} minutes. 0°6 ” 5 ” 5 ” 0-4 ” 9 ” 74 oe) 0:2 i 30 < Te * Set II.—Temperature 16° C. 4:0 cub. centims. 6 minutes. 2-0 * 16 a 12 minutes. 1:0 ” 39 3) 24 ” 0°5 Pm LOSES ee 0-25 ” 280 e 96 ee When the vanishing point of the metacasein reaction was taken as the point of comparison, the results approximated more nearly to the requirements of the rule of inverse proportion, especially at low tem- peratures; but still the evidence pointed in the same direction, and indicated that trypsin, like diastase, exhausts itself in action at a pro- 158 Dr. W. Roberts. Estimation of the Amylolytic [May 5, gressively retarded rate. From the numerous experiments which were performed with a view of elucidating this point, I arrived at the con- clusion that when the onset point of the metacasein reaction fell between 3 and 6 minutes the inverse time-rate gave a reliable basis of calculation, but not beyond these limits. Temperature.—Tryptic enzymosis is exceedingly sensitive to tempera- ture. The action of trypsin on milk increases in energy from zero to 60° C. Above this point there is a rapid fall, and the action is finally arrested between 75° and 80°C. There is not, as with diastase, any range or platform of indifferent temperature. The following table exhibits the degrees of activity from 10° to 80°. In order to obtain the utmost uniformity of results, the quantities of pancreatic extract employed were so adjusted as to bring the incidence of the metacasein reaction within a period ranging from 4 to 6 minutes. Table VII. Showing the effects of temperature on tryptic enzymosis. The quantity of milk employed in each experiment was 50 cub. centims. diluted with water up to 100 cub. centims. In the fourth column the degree of tryptic activity, or T, is calculated by a method to be presently explained. Quantity of pancreatic | Onset point of the | Tryptic value— ae =ature: extract employed. metacasein reaction. or T. 10° C. 6:0 cub. centims. 5 minutes. | 8 15 4-0 D Dy 12 20 3°0 33 4 ~ 21 30 1°0 29 one 45 40 0°6 3 Zines 76 50 0 °4 3 ao 119 60 0-3 9 53s 150 65 0-4 99 4) 36 125 70 0°8 2 ges 78 75 2°0 % 6 3 21 80 4:0 » No action. 0 In the subjoined diagram these results are thrown into the form of a curve. The ordinates indicate the degrees of tryptic activity (or T), and the abscissee indicate the temperatures. In another series of experiments the effect of temperature was gauged by the length of time required to reach the onset or the vanishing point of the metacasein reaction when constant quantities of pancreatic extract were used. The results obtained in this series are tabulated in Table VIII. In the first set the onset point of the re- action was taken as the index of tryptic activity; in the second set — 1881.| and Proteolytic Actwity of Pancreatic Extracts. 159 Curve illustrating the effect of temperature on the tryptic digestion of milk. the vanishing point of the reaction was employed for the same purpose. The results brought out by these experiments correspond pretty closely with those given in Table VII. Table VIII. Showing the effect of temperature by the length of time required to reach the metacasein reaction, when constant quantities of pancreatic extract are used. | I Set. | II Set. | 0'4 cub. centim. pancreatic extract 4 cub. centims. of pancreatic extract with 100 cub. centims. diluted milk. with 100 cub. centims. diluted milk. . f Onset point of the ‘ov navct.., | Vanishing point of the emperature. : 4 Temperature. TR Ne metacasein reaction. metacasein reaction. 2 to 5° C. | 312 minutes. | = | — 10 | 168 a | 10° C. 180 minutes. 15 | 120 RA | — | — 20 | 70 é | 20 | 75 e 30 | Saute | 30 | 26uINt 40 12 5 | 40 12 5 50 6 3 | 50 6 5 60 At aay | 60 rs ame 65 6 of 70 { action suspended, but resumed on cooling. An examination of the table shows how very nearly the results correspond, whether the onset point or the vanishing point of the metacasein reaction be taken as a measure of tryptic activity. This correspondence substantiates the conclusion that the onset point of the reaction furnishes a trustworthy index of the activity of tryptic digestion. The proportionate quantity of pancreatic extract added to the milk in the experiments recorded in Set II of Table VIII was ten 160 Dr. W. Roberts. Estimation of the Amylolytic [May 5, times as great as in those recorded in Set I; and it is seen that, by using these proportions, the vanishing points and the onset points fell out in nearly the same times in both sets of experiments. Mode of Proceeding.—In testing the tryptic activity of a sample of pancreatic extract, the follontne: procedure was adopted :—50 cub. centims. of fresh milk were diluted with 50 cub. centims. of water, less the quantity of extract intended to be added. The diluted milk was then warmed to 40° C., and maintained exactly at that tempera- ture until the close of the experiment. ‘The intended quantity of the pancreatic extract, say 1 cub. centim., was then added, and the time exactly noted. At the end of each minute a portion of the digesting milk was withdrawn, and boiled for a.few seconds in a test-tube, inclining the test-tube to one side after the boiling in order to observe the effect. The result was at once noted down. As soon as distinct curdling occurred on boiling, the experiment was considered finished ; the time was recorded, and the number of minutes which had elapsed from the commencement of the experiment were reckoned. The result came out in the following form :— 1 cub. centim. panc. extract+50 milk=4 minutes at 40° C. If no signs of incipient curdling (soiling of the sides of the test- tube) occurred within 3 minutes, a new experiment was made, using, two, three, or four times as much pancreatic extract. If, on the other hand, distinct curdling occurred in 2 minutes, or less, a fresh experiment was made, using half or quarter the quantity of extract. Three or four such experiments usually sufficed to enable: the operator to fix the onset point of the reaction somewhere between 4 and 6 minutes. Mode of Calculating and Expressing the Tryptic Valwe.—The object of the experiment was to ascertain how many cubic centimetres of milk can be changed to the onset point of the metacasein reaction by | cub. centim. of extract in a period of 5 minutes, at the temperature of 40° C. The tryptic value, or 'T, was calculated-from the first expres- sion of the results of an experiment in exactly the same way as for diastase. If p be made to signify the quantity of pancreatic extract added to the milk, and m the number of minutes which were required to reach the onset point of the metacasein reaction, then the value of T was obtained by the following formula :— oo 1 ip and taking the experiment above given the value of T came out as. follows :— m0 Pega ar Ao? C: ren In judging the practical value of this method of trypsimetry, one must have regard to the inherent difficulty of estimating the activity 1881.] and Proteolytic Actwity of Pancreatic Katracts. 161 of preparations of the proteolytic enzymes. I venture to think that we have in this method a means of estimating the activity of trypsin preparations which is superior in ease and precision to any we possess for the evaluation of pepsin preparations. What may be the limits of error arising from inequalities in the composition of milk I am unable to say, but with the same specimen of milk the limits of error do not certainly exceed 6 to 8 per cent. The tryptic value of pancreatic extract from the pig, made on the large scale, was found to range from 40 to 70 at 40° C. The pan- creatic tissue of the ox and sheep yielded an extract which possessed about the same tryptic activity as that of the pig. Extracts prepared from single glands presented very considerable variations both in regard to their diastasic and their tryptic activity. The following table shows the enzymic values of twelve samples of pancreatic extract prepared with single glands from four pigs, four oxen, and four sheep, killed for the market. Table IX. (All the observations were made at 40° C.) (D stands for diastasic value, and T for tryptic value.) Pig. Ox. Sheep. D=166 D=8 Sie ls ial Nona e, No. of2-5,, D=100 D=10 D=12 Be nas NG peer Nice alae D=100 D=9 D=14 Pah Bo ales on 4] eat ie WS m6 D=100 D=13 No. 4 ‘be No. 8 sige No. Pe as _98 It may be observed that the oscillations in the two enzymic values: bear no mutual relations to one another. The most appropriate standard of temperature for the valuation of tryptic activity, is 40° C., because this corresponds very nearly with the temperature at which trypsin operates in the normal digestion of warm-blooded animals. But itis more convenient to perform the testing at, or near, the ordinary temperature of the room (say, at 20°) inasmuch as in the latter case, it is much less troublesome to main- tain a continuously uniform temperature than at 40°. I have, therefore, taken some pains to ascertain the exact relation between the value of T at 40° and at 20° respectively, and have found that at 40° the value of T is very nearly three and a half times as ereat as at 20°. If, therefore, the testing be performed at 20°, the resulting value of T multiplied by 3°5, will give with sufficient accuracy the value of T at 40°. 162 C. G. Williams and W. H. Waters. [May 12, May 12, 1881. GENERAL STRACHEY, R.E., C.8.1., V.P., in the Chair. The Presents received were laid on the table, and thanks ordered for them. Pursuant to notice Gabriel Auguste Daubrée, Jean Charles Marignac, Carl Nageli, and Carl Weierstrass were balloted for and elected Foreign Members of the Society. The following Papers were read :— I. “ On the Physiological Action of 8 Lutidine.” By C. GREVILLE WILLIAMS, F.R.S., and W. H. Watsmrs, B.A., Demonstrator in the Physiological Laboratory, Cambridge. Received April 23, 1881. ‘In studying the physiological action of f Iutidine the greater number of our experiments were made upon the frog, and more par- ticularly related to the action of this substance upon the heart and central nervous system; our other experiments upon the different organs being for the better interpretation of the action upon these two. Heart. In all cases we first destroyed the brain and spinal cord of the frog by pithing, and with the smallest possible loss of blood. Hence none of the results obtained could be due to the central nervous system. The experiments were made upon the heart to study the effect of the 8 lutidine with regard to its— (1.) Tonicity ; (2.) Time of beat; (3.) Nerve supply. It may be stated at once, that the pure alkaloid applied to the sur- face of the heart causes that organ at once to shrink up, turn pale, and cease to beat. At times it was found possible to restore these beats by washing with normal saline solution (0°75 per cent. sodium chloride), but in no case did they ever reach their former activity. (1.) Tonicity of the Heart— We made use of several methods during our experiments to ascertain the effect of this poison upon the heart’s tonicity. 1881. | On the Physiological Action of B Lutidine. 163 The three chief ones were— (a.) Moist chamber and lever ; (8.) Roy’s apparatus ; (y.) Gaskell’s method of the excised heart. (a.) Moist Chamber and Lever—In this method the frog had just sufficient of its skin and sternal cartilage removed to expose the apex of the heart; a carefully balanced lever (of the first kind) had attached to one end of it, by a horizontal pivot, a light needle, which, by means of a sinall piece of cork, just rested upon the apex of the heart. The other end of the lever carried a fine piece of aluminium to mark the beats upon the smoked surface of a revolving cylinder. Another piece of aluminium recorded the abscissa line from which variations in the height of the curve of ventricular beats could be recorded. A alass shade, having a slit to allow the end of the lever to protrude and work upon the drum, was then placed over the animal, and the air inside kept moist with damp blotting-paper. From the arrangement of the apparatus it is evident that at each ventricular systole the marker would fall, and at each diastole rise; any increase in the tonicity of the heart being indicated by the tracing approaching the abscissa line. A tracing of the heart’s normal beat was in all cases first taken, and then various quantities of various strengths of 6 lutidine were injected into different parts of the body, from a drop or two of 2 per cent. solution in the lymph spaces near the heart, to 0°2 cub. centim. of 2 per cent. in the lower part of the abdominal cavity. In all cases there was a very evident lowering of the lever’s marking point and a decrease in the size of the beats, clearly showing that the heart had experienced a distinct increase in its tonicity. In all cases it became much paler. No. I is a characteristic series of tracings showing the alterations in the ventricular beats. Of these (1) is that made by the normal heart, while 2, 3, and 4, were taken at intervals of one, three, and six minutes after the injection of 0-2 cub. centim. of 2 per cent. 6 lutidine into the abdominal cavity. The increase in tonicity and slowing of the beat are most evident. These tracings should be read from right to left, a a being the abscissa line. (8.) Roy’s Apparatvs.—In this apparatus the heart is tied upon a double canula, one limb of which is in connexion with asupply of fresh diluted sheep’s blood (one part of defibrinated blood to three parts of 0-75 per cent. saline solution), kept at a constant height in a reser- voir ; the other allowing the blood as it is pumped off by the heart to flow away. ‘The surface of the fluid in the reservoir being always at the same height, variations in the heart’s beat could in no way be due to variations in pressure of the fluid. The heart tied to the canula is inserted into a closed chamber filled with oil. By a par- 164 C. G. Williams and W. H. Waters. [May 12, ticular arrangement, a piston, working in a cylindrical prolongation of this chamber, will rise or fall on each systole or diastole of the ventricle. A lever being attached to this piston indicates the con- dition of the ventricle upon a moving surface.: From the construction it is evident that each systole of the ventricle is indicated by a rise in the lever, making evident its character. After the character of the beats of a particular heart had been observed while the normal blood was passing, the blood was then sent through, containing from 0:1 per cent. to 0°4 per cent. of B lutidine, and the character of the tracings studied. These were distinctly modified; the heart still beat, but never relaxed during diastole to its full extent, in fact, showing a most decided increase in its tonicity. The highest portious of the curve, indicating the condition of the ventricle during systole, showed that that condition was much prolonged, that portion of the curve being much rounder and longer. The normal beat could be at once fully restored by repassing the saline blood solution; and a series of experiments, lasting some time, could thus be taken. Ii is a specimen of these tracings, and should be read from right to left. The beats on the extreme right are those of the ventricle before the application of the alkaloid. When weaker solutions are used, similar alterations occur, though naturally not so pronounced. In one experiment, where no beats occurred, the lever rose distinctly on passage of the saline blood containing the f# lutidine, indicating an increase in the tonicity of the ventricle. (y.) Gaskell’s Method.—In these experiments a method due to Dr. Gaskell was used. iS XS = © SN Loa} S eS) ~ Sy 74 On the Physiolo oe Ee i | J hou} aaa eee Te) J le / Nae! Vives wu ee Ao) reas / NE I | fe se 166 C. G. Williams and W. H. Waters. [May 12, ing. The original beats were restored by washing with normal saline- solution. In III are two continuous parts of a tracing taken by this method, and should be read from right to left. From 1 the curve is due to the beats of the ventricle after washing with a very small quantity of normal saline solution. At 2 a very small drop of 0°2 per cent. 8 lutidine was allowed to fall on the ventricle, and the changes in the beats, as above described, at once appear, the normal beats being again restored by washing with normal saline solution as at 3. In the particular experiment from which this tracing was taken the drum was slightly irregular, hence the apparent quickening of the beats after the second washing with the saline solution. Measurement.—The ventricle was just exposed, and during the in- tervals between the observations the opening was covered with damp blotting paper. Even with this rough method it was found that the injection of 2:0 per cent. 6 lutidine into various lymph spaces or into the cutaneous vein, caused a distinct decrease in the length of the yen- tricle, and this was always attended with a slight paleness. From all these cases it cannot but be evident that the introduction of B lutidine into the system, enabling it to reach the heart, causes a distinct increase in the tonicity of its ventricle. The auricles seem to be affected but very little, and even continue to beat after the ventricle has stopped. On Voluntary Muscle—By a method somewhat similar to that of Dr. Gaskell, the effect of the alkaloid was tried upon ordinary striped muscular tissue. A frog was pithed and placed on its belly in a moist chamber. The tendo-achilles laid bare and tied by fine silk to a lever, the knee-joint being fixed. On washing the muscle with 0°75 per cent. saline solution, as was done to the ventricle, a similar fall of the lever occurred ; but on allowing a drop of 0°5 per cent. 8 lutidine (in normal saline) to fall on it, a distinct rise followed. Tracing IV illustrates the above, aa, being the abscissa line; any increase in the tonicity of the muscle will be indicated by an increase in the distance of the lever’s marker from this line. The tracing should be read from right to left. At (1) 0°75 per cent. saline solution was added, and after a short interval the drum allowed to move, a fall of the lever became evident. At (2) and (3) a small drop of 0:5 per cent. of the alkaloid was added, and a distinct rise followed ; the washing again with normal saline solution at (4) resulting in a fall. (2.) Time of Heari’s Beats.—At times we found the results of an experiment undecided, but by far the greater number of them indi- cated a slight slowing after the addition of the alkaloid. It appeared distinctly in the experiments made with strong solutions. (3.) Nerves of the Heart—In all experiments relating to the action 1881.] On the Physiological Action of B Lutidine. 167 of poisons on the heart one naturally turns to trying their effect upon the vagus. Under ordinary conditions this nerve has the effect when slightly stimulated of slowing, when strongly stimulated of stopping the heart’s beat. . The apparatus used was the moist chamber before described, with the addition of a pair of platinum electrodes connected with the secondary coil of a Du Bois Raymond’s induction apparatus. The key for short-circuiting the secondary current carried a marker to make a tracing on the revolving cylinder, indicating the point of time at which the vagus was stimulated and the duration of the stimulus by a rise in the line traced on the drum. To ensure the constant strength of the stimulus, we always allowed the Daniells’ cell to work for about half an hour before using the coil. The frog was pithed, the heart and vagus laid bare with as little exposure of the surrounding parts as possible, the lever and moist chamber arranged, and a tracing taken of the normai ventricular beats. The vagus was next stimulated with the secondary coil at various distances from the primary, till one was found at which the heart was slowed and another at which it was stopped. Experiments were then made with different quantities of solutions of 8 lutidine of various strengths, from ()'2 cub. centim. of 2 per cent. to 0°2 cub. centim. of 5 per cent., which were injected into various parts of the body, the best results being obtained by injecting into the lower part of the abdominal cavity. In all cases the alkaloid deprived the vagus of its power, when stimulated, to arrest the beat of the heart. . As an example, a frog was prepared and the vagus found to cause slowing of the heart’s beat when stimulated with the secondary at 14-5, and stoppage at 14 and 13°5—a long stoppage at the latter. 0-2 cub. centim. of 2 per cent. solution of f lutidine were injected into the abdominal cavity near the right leg ; after three minutes the vagus on stimulation had but very slight effect, and after six minutes no action. From time to time the secondary was moved nearer and nearer the primary, but stimulation had little or no effect, and after twenty-five minutes the action was imperceptible even when the secondary coil was right over the primary. Reflex Action. Our next series of experiments were made to ascertain what effect this alkaloid would have upon reflex action. The frogs used had their brains destroyed, with as small a loss of blood as possible, by dividing the spinal cord at the junction of the first vertebra and passing a wire into the skull. The animals were then left for some time to recover from the effects of the operation. The stimulus used was a very weak solution of hydric sulphate, and the VOL. XXXII. N 168 C. G. Williams and W. H. Waters. [May 12, time elapsing between the placing of the foot in this stimulant and its withdrawal by the animal was measured by a metronome beating 120 to the minute and taken as the indication of the time required for reflex action. The frog was worked upon either resting on its belly with its legs hanging down or suspended by its head. In order that the results might not be thought due to fatigue, owing to the suspen- sion of the animal, in the earlier experiments, another frog similarly pithed, was suspended, and its reflex time measured simultaneously with the one treated with the alkaloid. These frogs suffered no change, therefore after the first experiments their use was discontinued. _ After suspension of the animal its reflex action was tested for half an hour at intervals of five minutes, in order to get the length of its normal reflex time. The alkaloid was then injected, sometimes pure, into the stomach, and at other times dilute into the dorsal lymphatic sac; here the effect could be produced by 0°2 cub. centim. of 1 per cent. solution in ten minutes to a quarter of an hour. The action of the f lutidine was most distinct, the reflex time, after an interval. growing longer and longer at each stimulation, then the leg would be only just removed, and that feebly from the acid; and, finally, after a time varying from a quarter to half hour the foot was incapable of being withdrawn though immersed in the acid for a minute or a minute and a half. Decidedly, then, 6 lutidine destroys reflex action in the frog. The disappearance of the reflex action might be attributed to the following causes :— (1.) To the poison acting upon the heart, perhaps stopping its beat, and thus cutting off the blood supply to the spinal cord. In all cases, how- ever, the frog was dissected and the heart was invariably found beating. (2.) It might be due to its action on the nerves, thus preventing the conduction of impulses to the muscles of the legs, necessary to cause them to contract and withdraw the foot from the acid. (8.) It might be said to be due to an effect like that of urani, preventing the communication of an impulse from the nerve to the muscle, blocking their physiological communication. That it was not due to either of these causes will be seen from the following experi- ment: A frog was very carefully pithed, the sciatic nerve on each side was exposed and a ligature tied round the whole of the left leg— except the nerve. The nerves on each side were each stimulated with such a: feeble current as to cause only a slight twitching of the toes (secondary coil at about 32). Far more 6 lutidine than is necessary to produce the described effect . upon reflex action was then injected into the dorsal lymphatic sae, at times as much as two or more drops of the pure alkaloid being used, Although the experiment lasted in some cases over two hours, the feeble stimulus caused the same amount of movement in both feet as before. 1881. | On the Physiological Action of B Lutidine. 169 (4.) The effect of the poison might also be thought due to its action upon the muscles themselves, causing them to lose their irritability and cease to respond to stimuli reaching them from the nerve. The last experiments would go strongly against this, but to be perfectly sure we performed the following: A frog, whose brain had been de- stroyed, received urari up to a point at which no reflex action occurred, and no contraction of the leg muscles on stimulation of the sciatic nerve. The animal was placed ina moist chamber, and by means of the pendulum myograph tracings were taken of the contractions of the . gastrocnemius, caused by direct stimulation. A number of these. ' tracings of the muscular contraction was taken before application of the poison, and then from time to time # lutidine was injected into the dorsal lymphatic sac, commencing with 0°1 cub. centim. of 10 per cent. solution, and finally using as much as two drops of the pure alkaloid ; but there was no alteration of a decided character in any of the curves, either upon the latent period, contraction, or relaxation. It may be mentioned that after acting on frogs retaining their spinal cord with # lutidine, destruction of the spinal cord with a wire, causes little or no movement of the limbs, as is the case under ordinary conditions. From all these results it cannot but be evident that the disappear- ance of reflex action is due alone to the effect of the poison on the nerve cells of the spinal cord. Strychnine and B Lutidine—— Such being the action of 6 lutidine upon the spinal cord, we were naturally led to ascertain whether it was antagonistic in its action to strychnine. If a frog, whose brain has been destroyed, be treated with a small quantity of strychnine solution, the slightest stimulus is sufficient to throw the animal into strong tetanic convulsions. Various experiments were tried; as examples we will give the following :— (1.) A frog whose reflex action was good (the brain having been destroyed in the usual way) had sufficient 8 Iutidine solution given it, to cause this reflex action to disappear. A drop of 0°5 per cent. strychnine was injected into its dorsal lymphatic sac, a quantity which injected into another frog similarly prepared brought on the strongest convulsions. Still no effect was produced, neither did any change appear when another large drop was injected. (2.) As an example of another method, a frog (whose brain had been destroyed) was thrown into strong ‘convulsions by the use of strychnine, then 0°5 cub. centim. of 10 per cent. f Iutidine were injected into the dorsal lymphatic sac, also a drop of the pure alkaloid ; in fifteen minutes the strychnine tetanus was passing off and in twenty-five minutes had quite gone. The same effect happened with smaller quantities of the alkaloid. N 2 170 Prof. G. G. Stokes. Discussion of the Results [May 12, (3.) Another, better, method was used. Two frogs, whose brains had been destroyed as before, were taken, and into the dorsal lym- phatic sac of each a small drop of 5 per cent. strychnine was injected. A minute afterwards only 0°1 cub. centim. of 10 per cent. 6 lutidine was injected into one frog (A), the other (B) remaining as before. After fifteen minutes (A) gave no signs of strychnine tetanus, while (B) gave distinct signs. After twenty minutes (A) gave only very faint reflex action; but (B), on being touched, went into strong tetanus. After thirty minutes (A) gave no signs of reflex action, while (B) went into strong tetanic convulsions on simply touching the table. These results lasted over an hour; then into (B) 0°1 cub. centim. of 10 per cent. 6 Iutidine was injected, and in ten minutes the effect of the strychnine began to pass off, and in thirty minutes was quite gone, the frog not even giving signs of reflex action. Again, in some experiments strychnine was injected into one frog (prepared as before), and strychnine with 6 lutidine into another. After twenty-four hours the first went into strong tetanic convulsions on touching, but never the latter; that of the former disappearing after the injection of the alkaloid. From our experiments we hope we have made it clear that 8 lutidine causes a distinct increase in the tonicity of both cardiac and voluntary muscular tissue, also a slowing in the rate of the heart’s beat, and that it arrests the inhibitory power of the vagus. That by its action upon the nerve cells of the spinal cord, it, in the first place, lengthens the time of reflex action, and then arrests that function; finally, that it is successfully antagonistic to strychnine in its action upon the spinal cord. In conclusion, we feel much pleasure in acknowledging our grate- ful thanks to Dr. Michael Foster, both for his kind help and happy suggestions, which have been of great assistance to us in our investi- gations. Il. “ Discussion of the Results of some Experiments with Whirled Anemometers.” By Professor G. G. STOKES, Sec. R.S. Received April 26, 1881. In the course of the year 1872, Mr. R. H. Scott, F.R.S., suggested to the Meteorological Committee the desirability of carrying out a series of experiments on anemometers of different patterns. This suggestion was approved by the Committee, and in the course of same year a grant was obtained by Mr. Scott from the Government Grant administered by the Royal Society, for the purpose of defraying the expenses of the investigation. The experiments were not, however, 1881.] of some Experiments with Whirled Anemometers. 171 carried out by Mr. Scott himself, but were entrusted to Mr. Samue! Jeffery, then Superintendent of the Kew Observatory, and Mr. G. M. Whipple, then First Assistant, the present Superintendent. The results have never hitherto been published, and I was not aware of their nature till on making a suggestion that an anemometer of the - Kew standard pattern should be whirled in the open air, with a view of trying that mode of determining its proper factor, Mr. Scott informed me of what had already been done, and wrote to Mr. Whipple, requesting him to place in my hands the results of the most complete of the experiments, namely, those carried on at the Crystal Palace, which I accordingly obtained from him. The progress of the enquiry may be gathered from the following extraet from Mr. Scott’s report in returning the unexpended balance of the grant. “The comparisons of the instruments tested were first instituted in the garden of the Kew Observatory. This locality was found to afford an insufficient exposure. ‘“‘ A piece of ground was then rented and enclosed within the Old Deer Park. The experiments here showed that there was a consider- able difference in the indications of anemometers of different sizes, but it was not possible to obtain a sufficient range of velocities to furnish a satisfactory comparison of the instruments. Hxperiments were finally made with a rotating apparatus, a steam merry-go-round, at the Crystal Palace, which led to some results similar to those obtained by exposure in the Deer Park. “The subject has, however, been taken up so much more thoroughly by Drs. Dohrandt and Thiesen (wide “ Repertorium fiir Meteoro- logie,” vols. iv and v) and by Dr. Robinson in Dublin, that it seems unlikely that the balance would ever be expended by me. I, there- fore, return it with many thanks to the Government Grant Com- mittee. ‘The results obtained by me were hardly of sufficient value to be communicated to the Society.” On examining the records, it seemed to me that they were well ‘deserving of publication, more especially as no other experiments of the same kind have, so far as I know, been executed on an anemo- meter of the Kew standard pattern. In 1860 Mr. Glaisher made experiments with an anemometer whirled round in the open air at the end of a long horizontal pole,* but the anemometer was of the pattern employed at the Royal Observatory, with hemispheres of 8:75 inches diameter and arms of 6°725 inches, measured from the axis to the centre of a cup, and so was considerably smaller than the Kew pattern. The experiments of Dr. Dohrandt and Dr. Robinson * “ Greenwich Magnetical and Meteorological Observations,” 1862, Introduction, p. li. i2 Prof. G. G. Stokes. Discussion of the Results [May 12, were made in a building, which has the advantage of sheltering the anemometer from wind, which is always more or less fitful, but the disadvantage of creating an eddying vorticose movement in the whole mass of air operated on; whereas in the ordinary employment of the anemometer the eddies it forms are carried away by the wind, and the same is the case to a very great extent when an anemometer is whirled in the open air in a gentle breeze. Thus, though Dr. Robinson employed among others an anemometer of the Kew pattern, his experiments and those of Mr. Jeffery are not duplicates of each other, even independently of the fact that the axis of the anemometer was vertical in Mr. Jeffery’s and horizontal in Dr. Robinson’s experi- ments ; so that the greater completeness of the latter does not cause them to supersede the former. In Mr. Jeffery’s experiments the anemometers operated on were mounted a little beyond and above the outer edge of one of the steam merry-go-rounds in the grounds of the Crystal Palace, so as to be as far as practicable out of the way of any vortex which it might create. The distance of the axis of the anemometer from the axis of the ‘““merry”’ being known, and the number of revolutions (7) of the latter during an experiment counted, the total space traversed by the anemo- meter was known. The number (N) of apparent revolutions of the anemometer, that is, the number of revolutions relatively to the merry, was recorded on a dial attached to the anemometer, which was read at the beginning and end of eachexperiment. As the machine would only go round one way, the cups had to be taken off and replaced in a reverse position, in order to reverse the direction of revolution of the anemo- meter. The true number of revolutions of the anemometer was, of course, N-+n, or N—n, according as the rotations of the anemometer and the machine were in the same or opposite directions. The horizontal motion of the air over the whirling machine during any experiment was determined from observations of a dial anemo- meter with 3-inch cups on 8-inch arms, which was fixed on a wooden stand in the same horizontal plane as that in which the cups of the experimental instrument revolved, at a distance estimated at about 30 feet from the outside of the whirling frame. The motion of the centres of the cups was deduced from the readings of the dial of the fixed anemometer at the beginning and end of each experiment, the motion of the air being assumed as usual to be three times that of the cups. The experiments were naturally made on fairly calm days, still the effect of the wind, though small, is not insensible. In default of | further information, we must take its velocity as equal to the mean velocity during the experiment. Let V be the velocity of the anemometer (7.e., of its axis), W that of the wind, 0 the angle between the direction of motion of the 1881.] of some Experiments with. Whirled Anemometers. Aus anemometer and that of the wind. Then.the velocity of the anemo- meter relatively to the wind will be— A, VipeONEWRCOSIOLEINVC So. Wiss uy (a) The mean effect of the wind in a revolution of the merry will be different according as we suppose the moment of inertia of the anemometer very small or very great. If we suppose it very small, the anemometer may be supposed to be moving at any moment at the rate due to the relative velocity at that moment, and therefore the mean velocity of rotation of the cups in one revolution of the merry will be that corresponding to the mean relative velocity of the anemometer and the air. If, as is practically the case, W be small as compared with V, we may expand (a) in a rapidly converging series according to ascending powers of W. All the odd powers will disappear in taking the mean, and if we neglect the fourth and higher powers we shall have for the mean Ww? \V ae Ww so that W?--4V is the small correction to be added to the measured velocity of the anemometer in order to correct for the wind. On the other hand, if the moment of inertia of the anemometer be taken as very great, the rate of rotation of the cups during a revolu- tion of the merry will be sensibly constant. If V’ be the velocity of the anemometer relatively to the air, v the velocity of the centre of one of the cups, and if we suppose the rotation of the anemometer resisted by a force of which the moment is F, then, according to Dr. Robinson’s researches, we have approximately K=AV”—2BvV’— Cv’. In the present case friction is not taken into account, and instead of F we must take the moment of the effective moving force. Further- more, it appears from the experiments of Dr. Robinson, in Dublin, that the observations were almost as well satisfied by taking the first two terms only of the above expression for I' as by taking ali three, and this simplification may be employed with abundantly sufficient accuracy in makifig the small correction for the wind. We have, therefore— F=AV”?—2BoV’, where V’ is given by (a). In order that the anemometer may be neither accelerated nor retarded from one revolution of the merry to another, the mean effective force must be nil; and taking the means of both sides of the above equation, observing that, in conse- quence of the supposed largeness of the moment of inertia, v is ve! Prof. G. G. Stokes. Discussion of the Results ee 12 sensibly constant during one revolution of the merry, we, have on employing the approximate value of the mean of V’ or (a) already used— But if U be the constant velocity of air relatively to the anemometer which would make the cups turn round at the same rate, we have similarly— 0O=AU?2—2BvU. Hliminating Buv/A, between these two equations we get— ns Ul Vs =e Wet (V+T-)=v+ OF and as the fourth and higher powers of W have been neglected all along, we get from the last— UAaV+e0 a re so that, on this supposition, the mean correction for the wind is 3W?/4V, or three times the correction of the former supposition. The mean value of the radical (a) is given by an elliptic function ; but even in an extreme case among the experiments, when the ratio of the velocity of the wind to that of the anemometer is as great as 3 to 5, the error of the approximate expression V-+W2?/4V amounts only to about 0-01 mile an hour, which may be quite disregarded. The error in employing (d) for the determination of U instead of (c) is of about similar amount. Three anemometers were tried, namely, one of the old Kew standard pattern, one by Adie, and Kraft’s portable anemometer. Their dimen- sions will be found at the heads of the respective tables below. With each anemometer the experiments were made in three groups, with high, moderate, and low velocities respectively, averaging about 28 miles an hour for the high, 14 for the moderate, and 7 for the low. Each group again was divided into two subordinate groups, according as the cups were direct, in which case the directions of rotation of the merry and of the anemometer were opposite, or reversed, in which case the directions of the two rotations were the same. The data furnished by each experiment were: the time occupied by the experiment, the number of revolutions of the merry, the number of apparent revolutions of the anemometer, given by the difference of readings of the dial at the beginning and end of the experiment, and the space S passed over by the wind, deduced from the difference of readings of the fixed anemometer at the beginning and end of the experiment. 1881.| of some Experiments with Vhirled Anemometers. 175 The object of the experiment was, of course, to compare the mean velocity of the centres of the cups with the mean velocity of the air relatively to the anemometer. It would have saved some numerical calculation to have compared merely the spaces passed through during the experiment; but it seemed better to exhibit the velocities in miles per hour, so as to make the experiments more readily comparable with one another, and with those of other experimentalists. In the reductions I employed 4-figure logarithms, so that the last decimal in V in the tables cannot quite be trusted, but it is retained to match the correction for W, which it seemed desirable to exhibit to 0'U1 mile. On reducing the experiments with the low velocities, I found the results extremely irregular. J was subsequently informed by Mr. Whipple, that the machine could not be regulated at these low velocities, for which it was never intended, and that it sometimes went round fast, sometimes very slowly. He considered that the experiments in this group were of little, if any, value, and that they ought to be rejected. They were besides barely half as numerous as those of the moderate group. I have accordingly thought it best to omit them altogether. In the following tables the first column gives the group, H standing for high velocities, M for moderate; the subordinate group, — stand- ing for rotation of the anemometer opposite to that of the machine, + for rotations in the same direction, and lastly the reference number of the experiment in each subordinate group. ‘T gives the duration of the experiment in minutes; » the number of. revolutions of the machine; N the number of apparent revolutions of the anemometer ; S the space passed over by the natural wind, in miles. These form the data. From them are calculated: V, the velocity of the anemometer, in miles per hour; W, the velocity of the wind; W?/2V the mean of the two corrections to be added to V on account of the wind, according as we adopt one or other of the extreme hypotheses as to the moment of inertia of the anemometer, namely, that it is very small or very large. The actual correction will be half the number in this column on the first supposition, and once and a-half on the second. Vj, V, denote the velocity of the anemometer, or, in other words, of the artificial wind, corrected for the natural wind on these two suppositions respectively, so that the last two columns give 100 times the ratio of the registered velocity to the true velocity, or the registered as a percentage of the true, the registered velocity meaning that deduced from the velocity of the cups on employing the usual factor 3. The dials of the first two anemometers read only to 10 revolutions, which is the reason why all the numbers N end with a 0. 176 Prof. G. G. Stokes. Discussion of the Results [May 12, The Old Kew Standard. Diameter of Arms between Centres of Cups 48 inches; Diameter of Cups 9 inches. Fixed to Machine at 22°3 feet from the Axis of — Revolution. - Ge W: | 3000 | 3000 BHUNS cht eee eee eee ae Wo oy ane Mee H— : i 15 303 | 1690 OFF NN 3iral7 2°80 | 0:13 | 126°9 | 126°3 2 18 301 | 1690 0°5 | 26°63 1°67 4\.0°05 | R241 ess 3 16 301 ; 1580 0°6 | 29°96 2°2 0:08 | 114°2 | 1138 °9 4, i9/ 300 | 1710 2S ial 83°88 | 0°27 | 125-9" Mea e Ty 5 17 300 | 1720 Lot | 28d 3°88 | 0°27 | 126 -85ilZacs 6 22 400 | 2210 1:3 | 28:96 3°27 | 0:18 | 121-4 | 120°6 if 23 400 | 2220 Ik dP Ae o@) 2°87 | 0-15 1 122 aes 8 19 300 | 1670 O°7 | 25°14 2°21 4\ 0-10 1) 122-6222 9 19 300 | 1640 O°8 | 25:14 2°53 | 0°13) | TAG-R elses 10 i7/ 301 | 1670 0°8 | 28:20 2°71 |) O43) 122 ieee 11 19 300 | 1670 0°8 | 25°14 2-53. |- O13) 12265 k22 7-0 Mean 26°75 2.-78-| 0°15 | 1225602 H+ 1 i'7/ 302 980 0:0 | 28°29 0:00 | 0:00 | 114:°2 | 114-2 2 17 3860 960 1:0 | 28°11 3-53. 10-22 | 192 6a eles 3 Les 300 | 1000 1°4 | 30°82 5-429) O-A8 lila oles) A 223 300 | 1080 M-8 | 2il-23 4°80 | 0°54 | 122°3 | 119 °2 5 19 300 | 1020 0-7 5°14 2-21 | 0-10 | das alee 6 16 300 | 1030 0-9 | 29°86 38°37 | 0:19 | 119:0 | 118-2 7 18 3800 | 1050 0°8 | 26°54 2°67 | 0°13 | 120°8 | 120°2 8 18 301 | 1060 0-6 | 26°63 2°20 | 0:09 | 121-4 | 121°0 9 18 3800 | 1000 0°7 | 26°54 2543 4 (ON SUM 7es lee ies Mean 27 (02 2°96 | 0-21 | 118°4 | 117°5 M-— 1 30 300 | 1650 0°8 | 15°92 1°60 | 0:08 |; 120°8 | 120°2 2 381 300 | 1670 1°6 | 15°41 3°10°) O33 12-7 hos 3 84 300 | 1570 1:9 | 14°05 3:01 | 0:22 | 112°6 | 110°1 4. 36 3800 | 1540 1h O/B CXS 2°83 | 0°30 | 110°0 | 107°6 5 36 3800 | 1540 1°3 | 13°26 2°17 | 0°18 | 110°5 | 109-0 Mean... 14°38 2°54 | 0°24 | 11b5"1 | 113-2 M+ 1 28 301 880 0:0 | 17°63 0:00 | 0:00 | 102°6 | 102°5 2 38 300 940 | PA 557/ 3°15 | 0°38 | 109°5 | 106°4 3 38 300 890 16 ) 12-57 2°52 | 0°25 | 105-7 | 103-7 A 36 300 990 0:8 | 13°26 1°33 | 0:07 | 115°4 | 114 a 5 35 3800 990 1:0 | 13°66 1:71 | 0:07 | 115-3) ae Mean... ct eae i) ut 13 :94 1°74 | 0°15 | 109°7 | 108°5 1881.] of some Experiments with Whirled Anemometers. 177 Adie’s Anemometer. Diameter of Arms between Centres of Cups 13:4 inches; Diameter of Cups 2°5 inches. Fixed to Machine at 20°7 feet from the Axis of Revolution. Group W? | 3000 | 3000 and No. Tr MN : Me My 2V Vani Ve H— 1 17 | 300 | 3860] 1:0/ 26-16] 3:°53|0-24| 95-2 | 94-4 2 15: | 300 | 3650] 1-4] 28-61] 5:45 10-52 | 89:5 | 87-9 3 22: | 300 | 3940 | 1-°8|19-70| 4:80|0-58| 96:7) 24-0 4 19 | 300 | 3760] 0-7 | 23-33| 2-21]0-10| 93:1] 92-7 5 16 | 300| 3780! 0:9 | 27-71! 3:3710-20| 93:5] 92°8 6 is | 300 | 3890| 0-9 | 24-63! 2-67|0-14| 96:4| 96-0 7 18 | 301 | 3980| 0-7 | 24-72 | 2:33 |0-11| 98-6] 98-2 8 is | 300 | 3940] 0:8 | 24°63 | 2-67|0-14]| 97°8| 97°38 Mean aa 24:04 | 3°38 | 0-25 | 95:1| 94-2 H+ 1 17 | 300 | 3240] 1-1 | 26-:08| 3°89} 0-29] 94:°9| 93-9 2 17 | 300 | 3330] 1-1 | 26:08] 3:89|0-29| 97:3] 96°3 3 19 | 300 | 3760 | 0-7 | 23:33 | 2-21 | 0-10 | 109-2 | 108-7 4 16 | 300 | 3780| 0-9 | 27-71 | 3-37 | 0-20 | 109-6 | 108°8 5 19 | 300 | 3060] 0-8 | 23°33| 2°53 |0-14| 90°3| 89°8 6 17 | 301 | 3120| 0-8 | 26-17 | 2°8210-15 | 91-6 | 91-1 7 19 | 300] 3160! 0-8! 23:33] 2°53) 0-14! 93-0] 92°5 Mean is .. | 25°15 | 3-03! 0-19| 98-0) 97:3 M— 1 38 | 300 | 3620| 2-0] 11-67| 3:°16|0-43| 87-7] 84-9 2 38 | 300 | 3500] 1-6/11°67| 2:°5310-43| 84-7 | 81-7 3 36 | 300/ 3910] 0-8 | 12-26} 1:33|0-07| 97°53 | 96:9 4 35 | 300 | 3430] 1-0 | 12-67| 1-71|0-07| 841] 83-7 Cee. |. |... | 42-07 | 2-18 | 0-25 | 88-5. |) 86-8 M+ 1 31 | 300 | 3250| 1:6| 14:30] 3:10] 0°34| 94:3] 93-7 2 34 | 300] 2920! 1:7/|13:04| 3:00| 0-35 | 83-7] 85-5 3 34 | 300] 2940 |, 1:9| 13-04| 3:06| 0-36! 86-1] 83-1 4 36 | 300| 2760/ 1-7| 12°31 | 2°83 | 0:33 | 81:41 79-2 5 36 | 300 | 2780| 1:3| 12°31 | 217|0-19| 65-5 | 64-7 Mean 13°09 | 2°83 | 0-31 | 82:6] 81-0 178 Kraft’s Portable Anemometer. Prof. G. G. Stokes. Discussion of the Results [May 12, Diameter of Arms between Centres of Cups 8°3 inches; Diameter of Fixed to Machine at 19°10 feet from the Axis of Cups 3°3 inches. | —________ ee Revolution. Group and No. Tq. ieee 1 19 2 174 3 15 4, 18 5 16 6 17 yy 17 8 22 9 23 10 19 11 19 12 17 13 19 Mean H+ i 17/ 2 154 3 225 A, 19 5 16 6 18 i 18 8 18 Mean... M— 1 30 2 31 3 34 A 34 +) 56 6 36 Mean... M+ | 1 | 38 2 38 3 36 A BY) Mean... ga SCOCOPHEHOOOOO DHOODIWEH OHH ANATIA Seo oor its OD AOTAOKRO V. W. 21°53 | 1°89 23 °60. | 2°69 2722 280 22°79 | 1°67 25 65 | 2:25 24°06 | 3°88 24°06 | 3°88 24°79 | 3:27 Ha ST |p OX 7he\ Pall O33 |) 2 oD 21°53 | 2°58 24°14. | 2-82 21°53 | 2°53 P33) ay | Bo7/1L 24°06 | 3°53 26°39 | 5°42 18°18 | 4:80 PAL S533.) D2 OI PASS SIS ( |G 7) A) TION ROG) 22°79 | 2-00 DPR) || OD 87) 63 21S) 03 03 36 "36 27 1-60 3°10 2°53 3°39 2°83 74 ALE 0-09 0°36 0-27 0°47 0°35 | 0°21 2°60 | 0°29 3°16 2°53 1°33 Weg 0°46 0°30 0-08 0°12 22°99 | 3°33 | 0-27 | 100°8 2°18 | 0°24 300v | 38000 Val ee 95°6 | 95°4 96°2 | 95-4 102 °8 | 102 °4 1042 | 104-0 104°7 | 104°3 105 °2 | 103 °6 106°9 | 105°5 101 °4 | 100-4 101 °5 | 100°9 102 °3 | 101-7 Shei ay | 8) °/ Se) O | Ge) 0 100°9 | 100°1 101°5 | 100°8 102 °2 | 101°0 Seg | Bm 102°5 | 98-9 97°4 | 96°8 1006 | 99°8 100°6 | 99°8 100 °9 | 100°5 102 °2.| 101 -4 99° 4 93°3 | 92°5 99 72) |) 1 96R6 86°8 | 84°8 88°8 | 85°4 84°0 | 81°4 82°3 | 80°9 89:1 | 86°9 84°8 | 81°4 809 Siw 95°3 | 94°7 90°3 89 °3 87°8 | 86:0 The mean results for the high and moderate velocities, contained in 1881.] of some Hxperiments with Whirled Anemometers. 179 the preceding tables, are collected in the potlowang table, in which are also inserted the mean errors. High Velocities. Moderate Velocities. iB Directions of | Mom. inert. | Mom. inert. | Mom. inert. | Mom. inert. q Rotation. small. large. small. large. Be | 5 < eens meses es: Coals es nunca) rine elk 0s Coa unpes | Opposite.... | 122-6 | 2:4 | 121-9 | 2-3 | 115-1 | 4-9 | 113-2 | 5-2 e hich Elica e2cO il for eZ eSa O97 a) Aco) | LOS oii orl i Mean .....| 120°5 so |) LL o/ arene Fe 4 qe) EROS : Opposites...) 9d a | 2:3) | 94:2) | 2-3 | 88:5 | 4:5 | 86:8 | 5-0 zs PANTIE o.0, 3 ose 98°0 | 6°5 | 97°3 | 6:5 82:6 | 7:3 81°0 | 7°3 ‘ Mean 96 °5 95-7 85 °5 83 9 : Opposite’... . 101°5 | 2°6 | 100°8 | 2°5 SIF "4-8 86°9 | 5:1 = liera pears |) LOO;8) | We2eiy 99:4) lS.) 8758.1 5-0 | 86-0) | 630 H =| = = — MS Mean ..... 101 ‘1 100 °1 88 °4 86 4 The mean errors exhibited in the above table show no great difference according as we suppose the moment of inertia of the anemometer small or large in correcting for the wind. There appears to be a slight indication, beyond what may be merely casual, that the errors are a little greater on the latter supposition than on the former, which is what we should rather expect; for an anemometer would get pretty well under way in a fraction of a revolution of the whirling instrument. However, the difference is so small that 1t will suffice to take the mean of the two as the mean error belonging to the par- ticular anemometer, class of velocity, and character of rotation under consideration. From the mean errors we may calculate nearly enough, by the usual formule, the probable errors of the various mean percentages for rotations opposite and alike. The probable errors of these mean percentages come out as follows :— Kew, 1:0 for high velocities; 2:7 for moderate velocities. Adie, 15 99 pe) 2°0 99 99 Kraft, 0-9 Mi in les 5 at These probable errors are so small that it appears that for the high and even for the moderate velocities the experiments are extremely trustworthy, except in so far as they may be affected by systematic sources of error. If we compare the registered percentages of the true velocity of 180 Prof. G. G. Stokes. Discussion of the Results [May 12, the air relatively to the anemometer according as the rotations are in opposite directions or in the same direction, we see that in five out of the six cases they are slightly greater when the rotations are opposite. The sole exception is in the group ‘‘ Adie, high velocities,’ which is made up of the groups “ Adie H—” and “ Adie H+.” On referring to the principal table for the Adie, we see that Experiments 3 and 4 in group H+ give percentages usually high, depending on the high values of N. These raise the mean for the group, and make the mean error far greater than those of the other five groups for high velocities. There appears little doubt, therefore, that the excess of percentages obtained for rotations opposite is real, and not merely easual. It is, however, so small as to give us much confidence in the correctness of the mean result, unless there were causes to vitiate it which apply to both directions of rotation alike. It may be noticed that the difference is greatest for the Kew, in which the ratio of r to Ris greatest, r denoting the radius of the arm of the anemometer, and R the distance of its axis from the axis of revolution of the machine, and appears to be least (when allowance is made for the two anomalous experiments in the group ‘* Adie H+’) for the Kraft, for which 7/R is least. In the Kraft, mdeed, the differences are roughly equal to the probable errors of the means. In these whirling experiments 7/R is always taken small, and we might expect the correction to be made on account of the finiteness of R to be expressible in a rapidly converging series according to powers of r/R, say— 2 3 We ay * o(4) sed ye lee TM ha fees We may, in imagination, pass from the case of rotations opposite to that of rotations alike, by supposing R taken larger and larger in successive experiments, altering the angular velocity of revolution so as to preserve the same linear velocity for the anemometer, and suppos- ing the increase continued until R changes sign in passing through infinity, and is ultimately reduced in magnitude to what it was at first. The ideal case of R=oo is what we aim at, in order to repre- sent the motion of a fixed anemometer acted on by perfectly uniform wind by that of an anemometer uniformly impelled in a rectilinear direction in perfectly still air. We may judge of the magnitude of the leading term in the above correction, provided it be of an odd order, by that of the difference of the results for the two directions of rotation. Unless, therefore, we had reason to believe that A’ were 0, or at least very small compared with B’, we should infer that the whole correction for the finiteness of R is very small, and that it is practically eliminated by taking the mean of the results for rotations opposite and rotations alike. 1881.] of some Experiments with Whirled Anemometers. 181 We may accept, therefore, the mean results as not only pretty well freed from casual irregularites which would disappear in the mean of an infinite number of experiments, but also, most probably, from the imperfection of the representation of a rectilinear motion of the anemometer by motion in a circle of the magnitude actually employed in the experiments. Before discussing further the conclusions to be drawn from the results obtained, it will be well to consider the possible influence of systematic sources of error. 1. Friction—No measure was taken of the amount of friction, nor were any special appliances used to reduce it; the anemometers were mounted in the merry just as they are used in actual registration. Friction arising from the weight is guarded against as far as may be in the ordinary mounting, and what remains of it would act alike in the ordinary use of the instrument and in the experiments, and as far as this goes, therefore, the experiments would faithfully represent the instrument as it is in actual use. But the bearings of an anemometer have also to sustain the lateral pressure of the wind, which in a high wind is very considerable; and the construction of the bearing has to be attended to in order that this may not produce too much friction. So far the whirled instrument isin the same condition as the fixed. But besides the friction arising from the pressure of the artificial wind, a pressure which acts in a direction tangential to the circular path of the whirled anemometer, there is the pressure arising from the centrifugal force. The highest velocity in the experiments was about 30 miles an hour, and at this rate the centrifugal force would be about three times the weight of the anemometer. This pressure would consider- ably exceed the former, at right angles to which it acts, and the two would compound into one equal to the square root of the sum of their squares. The resulting friction would exceed a good deal that arising from the pressure of the wind in a fixed anemometer with the same velocity of wind (natural or artificial), and would sensiblv reduce the velocity registered, and accordingly raise the coefficient which Dr. Robinson denotes by m, the ratio, namely, of the velocity of the wind to the velocity of the centres of the cups. It may be noticed that the percentages collected in the table on p. 179, are very distinctly lower for the moderate velocities than for the high velocities. Such an effect would be produced by friction; but how far the result would be modified if the extra friction due to the centrifugal force were got rid of, and the whirled anemometer thus assimilated to a fixed anemometer, I have not the means of judging, nor again how far the percentages would be still further raised if friction were got rid of altogether. ' Perhaps the best way of diminishing friction in the support of an anemometer is that devised and employed by Dr. Robinson, in which 182 Prof. G. G. Stokes. Duscusston of the Results [May 12, the anemometer is supported near the top on a set of spheres of gun- metal contained in a box with a horizontal bottom and vertical side which supports and confines them. For vertical support, this seems to leave nothing to be desired, but when a strong lateral pressure has to be supported as well as the weight of the instrument, it seems to me that a slight modification of the mode of support of the balls might be adopted with advantage. When a ball presses on the bottom and vertical side of its box, and is at the same time pressed down by the horizontal disk attached to the shaft of the anemometer which rests on the balls, it revolves so that the instantaneous axis is the line joining the points of contact with the fixed box. But if the lateral force of the wind presses the shaft against the ball, the ball cannot simply roll as the anemometer turns round, but there is a slight amount of rubbing. This, however, may be obviated by giving the surfaces where the ball is in contact other than a vertical or horizontal direction. Let AB be a portion of the cylindrical shaft of an anemometer ; CD, the axis of the shaft; HEFGHI, a section of the fixed box or cup containing the balls; LMN, a section of a conical surface fixed to the shaft, by which the anemometer rests on its balls; FIKM, a section of one of the balls; F, I, the points of contact of the ball with the box; M, the point of contact with the supporting cone; K, the point of contact or all but contact of the ball with the shaft. The ball is sup- posed to be of such size that when the anemometer simply rests on the balls by its own weight, being turned perhaps by a gentle wind, there are contacts at the points M, F, I, while at K the ball and shaft are separated by a space which may be deemed infinitesimal. Lateral pressure from a stronger wind will now bring the shaft into contact with the ball at the point K also, so that the box on the one hand and the shaft with its appendage on the other, will bear on the ball at four points. The surface of the box as well as that on the cone LN being supposed to be one of revolution round OD, those four points will be 1881.] of some Experiments with Whirled Anemometers. 183 situated in a plane through CD, which will pass of course through the centre of the ball. If the ball rolls without rubbing at any one of the four points F, I, K, Mas the anemometer turns round, its instantaneous axis must be the line joining the points of contact, F, I, with the fixed box. But as at M and K likewise there is nothing but rolling, the instantaneous motion of the ball may be thought of as one in which it moves as if it were rigidly connected with the shaft and its appendage, combined with a rotation over LNAB supposed fixed. For the two latter motions the instantaneous axes are CD, MK, respectively. Let MK produced cut CD in O. Then since the instantaneous motion is com- pounded of rotations round two axes passing through O, the instan- taneous axis must pass through O. But this axis is FI. Therefore, FI must pass through O. Hence the two lines FI, MK, must intersect the axis of the shaft in the same point, which is the con- dition to be satisfied in order that the ball may roll without rubbing, even though impelled laterally by a force sufficient to cause the side of the shaft to bear on it. The size of the balls and the inclinations of the surfaces admit of considerable latitude subject to the above condition. The arrangement might suitably be chosen something like that in the figure. It seems to me that a ring of balls con- structed on the above principle would form a very effective upper support for an anemometer whirled with its axis vertical. Possibly the balls might get crowded together on the outer side by the effect of centrifugal force. This objection, should it be practically found to be an objection, would not of course apply to the proposed system of mounting in the case ef a fixed anemometer. Below, the shaft would only require to be protected from lateral motion, which could be done either by friction wheels or by a ring of balls constructed in the usual manner, as there would be only three points of contact. 2. Influence on the Anemometer of its own Wake.—By this I do not mean the influence which one cup experiences from the wake of its predecessor, for this occurs in the whirling in almost exactly the same way as in the normal use of the instrument, but the motion of the air which remains at any point of the course of the anemometer in consequence of the disturbance of the air by the anemometer when it was in that neighbourhood in the next preceding and the still earlier _ revolutions of the whirling instrument. It seems to me that in the open air where the air impelled by the cups is free to move into the expanse of the atmosphere, instead of being confined by the walls of a building, this must be but small, more especially as the wake would tend to be carried away by what little wind there might be at the time. On making some enquiries from Mr. Whipple as to a possible vorticose movement created in the air through which the anemometer passed, he wrote as follows :—“ I feel VOL. XXXII. 0 184 Prof. G. G. Stokes. Discussion of the Results [May 12). confident that under the circumstances the tangential motion of the air at the level of the cups was so small as not to need consideration in the discussion of the results. As in one or two points of its revolution the anemometer passed close by some small trees in full leaf, we should have observed any eddies or artificial wind had it existed, but T am sure we did not.” 3. Influence of the Variation of the Wind ; first, as regards Variations which arenot Rapid.—During the 20 or 30 minutes that an experiment iasted, there would of course be numerous fluctuations in the velocity of the wind, the mean result of which is alone recorded. The period of the changes (by which expression it is not intended to assert that they were in any sense regularly periodic), might be a good deal greater than that of the merry, or might be comparatively short. In the high velocities, at any rate, in which one revolution took only three or four seconds, the supposition that the period of the changes was large compared with one revolution is probably a good deal nearer the truth than the supposition that it is small. On the former supposition, the correction for the wind during two or three revolutions of the merry would be given by the formule already employed, taking for W its value at the time. Consequently, the total correction will be given by the formule already used, if we substitute the mean of W® for the square of mean W. ‘The former is necessarily greater than the latter; but how much, we cannot tell without knowing the actual variations. Weshould probably make an outside estimate of the effect of the variations, if we supposed the velocity of the wind twice the mean velocity during half the duration of the experiment, and nothing at all during the remainder. On this supposition, the mean of W? would be twice the square of mean W, and the correction for the wind would be doubled. At the high velocities of revolution, the whole correction for the wind is so very small, that the uncertainty arising from variation as above explained is of little importance, and even for the moderate velocities it is not serious. A. Influence of Rapid Variations of the Wind.—Variations of which the period is a good deal less than that of the revolutions of the whirling instrument act in a very different manner. The smallness of the corrections for the wind hitherto employed depends on the cir- cumstance that with uniform wind, or even with variable wind, when the period of variation is a good deal greater than that of revolution of the merry, the terms depending on the first power of W, which letter is here used to denote the momentary velocity of the wind, disappear in the mean of a revolution. This is not the case when a particular velocity of wind belongs only to a particular part of the circle de- scribed by the anemometer in one revolution. In this case there will in general be an outstanding effect depending on the first power of W, 1881.] of some Experiments with Whirled Anemometers. 185 which will be considerably larger than that depending on W?. Thus suppose the velocity of whirling to be 30 miles an hour, and the average velocity of the wind 3 miles an hour; the correction for the wind supposed uniform, or if variable, then with not very rapid variations, will be comparable with 1 per cent. of the whole; whereas, with rapid variations, the effect in any one revolution may be com- parable with 10 per cent. There is, however, thisimportant difference between the two : that whereas the correction depending on the square leaves a positive residue, however many experiments be made, the correction depending on the first power tends ultimately to disappear, unless there be some cause tending to make the average velocity of the wind different for one azimuth of the whirling instrument from what it is for another. This leads to the consideration of the following con- ceivable source of error. 5. Influence of Partial Shelter of the Whirling Instrument.—On visiting the merry-go-round at the Crystal Palace, I found it mostly surrounded by trees coming pretty near it, but in one direction it was approached by a broad open walk. ‘The consequence is, that the anemometer may have been unequally sheltered in different parts of its circular course, and the circumstances of partial shelter may have varied according to the direction of the wind. This would be liable to leave an uncompensated effect depending on the first power of W. I do not think it probable that any large error was thus introduced, but it seemed necessary to point out that an error of the kind may have existed. The effect in question would be eliminated in the long run if the whirling instrument were capable of reversion, and the experiments were made alternately with the revolution in one direction, and the reverse. lor then, at any particular point of the course at which the anemometer was more exposed to wind than on the average, the wind would tend to increase the velocity of rotation of the anemometer for one direction of revolution of the whirling instrument just as much, ultimately, as to diminish it for the other. Mere reversion of the cups has no tendency to eliminate the error arising from unequal ex- posure in different parts of the course. And even when the whirling instrument is capable of reversion, it is only very slowly that the error arising from partial shelter is eliminated compared with that of irregularities in the wind; of those irregularities, that is to say, which depend on the first power of W. For these irregularities go through their changes a very great number of times in the course of an ex- periment lasting perbaps half an hour; whereas, the effect of partial shelter acts the same way all through one experiment. It is very desirable therefore, that in any whirling experiments carried on in the open air, the condition of the whirling instrument as to exposure or shelter should be the same all round. 0 2 186 Prof. G. G. Stokes. Discussion of the Results [May 12, The trees, though taller than the merry when I visited the place last year, were but young, and must have been a good deal lower at the time that the experiments were made. Mr. Whipple does not think that any serious error is to be apprehended from exposure of the anemometer during one part of its course and shelter during another. # From a discussion of the foregoing experiments, it seems to me that the following conclusions may be drawn :— 1. That, at least for high winds, the method of obtaining the factor for an anemometer, which consists in whirling the instrument in the open air is capable, with proper precautions, of yielding very good results. 2. That the factor varies materially with the pattern of the anemo- meter. Among those tried, the anemometers with the larger cups registered the most wind, or in other words required the lowest factors to give a correct result. 3. That with the large Kew pattern, which is the one adopted by the Meteorological Office, the register gives about 120 per cent. of the truth, requiring a factor of about 2°5, instead of 3. Even 2°5 is pro- bably a little too high, as friction would be introduced by the centri- fugal force, beyond what occurs in the normal use of the instrument. 4, That the factor is probably higher for moderate than for high velocities; but whether this is solely due to friction, the experiments do not allow us to decide. Qualitatively considered, these results agree well with those of other experimentalists. As the factor depends so much on the pattern of the anemometer, it is not easy to find other results with which to compare the actual numbers obtained, except in the case of the Kew standard. The results obtained by Dr. Robinson, by rotating an anemometer of this pattern without friction purposely applied, are given at pp. 797 and 799 of the “‘ Phil. Trans.” for 1878. The mean of a few taken with velocities of about 27 miles an hour in still air gave a factor 2°36, instead of 2°50, as deduced from Mr. Jeffery’s experiments. As special antifriction appliances were used by Dr. Robinson, the friction in Mr. Jeffery’s experiments was probably a little higher. If such were the case, the factor ought to come out a little higher than in Dr. Robinson’s experiments, which is just what it does. As the circum- stances of the experiments were widely different with respect to the vorticose motion of the air produced by the action of the anemo- meter in it, we may [ think conclude that no very serious error is to be apprehended on this account. In a later paper (“ Phil. Trans.” for 1880, p. 1055), Dr. Robinson has determined the factor for an anemometer (among others) of the Kew pattern by a totally different method, and has obtained values considerably larger than those given by the former method. Thus the limiting value of the factor m corresponding to very high 1881.] of some Experiments with Whirled Anemometers. 187 velocities, is given at p. 1063 as 2°826, whereas the limiting value ob- tained by the former method was only 2'286. Dr. Robinson has expressed a preference for the later results. 1 confess I have always been disposed to place greater reliance on the results of the Dublin experiments, which were carried out by a far more direct method, in which I cannot see any flaw likely to account for so great a difference. It would be interesting to try the second method ina more favour- able locality. I take this opportunity of putting out some considerations respect- ing the general formula of the anemometer, which may perhaps not be devoid of interest. The problem of the anemometer may be stated to be as follows :— Let a uniform wind with velocity V act on a cup anemometer of given nattern, causing the cups to revolve with a velocity v, referred to the centre of the cups, the motion of the cups being retarded by a force of friction F'; it is required to determine v as a function of V and F, F haying any value from 0, corresponding to the ideal case of a friction- less anemometer, to some limit F',, which is just sufficient to keep the cups from turning. I will refer to my appendix to the former of Dr. Robinson’s papers (‘‘Phil. Trans.” for 1878, p. 818), for the reasons tor concluding that F is equal to V* multiplied by a function of V/v. Let V/Ju=e, B/V2=», then if we regard £ and y as rectangular co-ordinates, we have to determine the form of the curve, lying within the positive quadrant £Oy, which is defined by those co-ordinates. We may regard the problem as included in the more general pro- (188 Experiments with Whirled Anemometers. [May 12, blem of determining v as a function of V and F, where v is positive, but F may be of any magnitude and sign, and therefore, V also.* Negative values of F mean, of course, that the cups instead of being retarded by friction, are acted on by an impelling force making them go faster than in a frictionless anemometer, and values greater than F’, imply a force sufficient to send them round with the concave sides foremost. Suppose now F' to be so large, positive or negative, as to make v so great that V may be neglected in comparison with it, then we may think of the cups as whirled round in quiescent air in the positive or usual direction when F is negative, in the negative direction when F is greater than F,. When Fis sufficiently large the resistance may be taken to vary as v*. For equal velocities v it is much greater when the concave side goes foremost, than when the rotation is the other way. For air impinging perpendicularly on a hemispherical cup, Dr. Robinson found that the resistance was as nearly as possibly four times as great when the concave side was directed to the wind as when the convex side was turned in that direction. When the air is at rest and the cups are whirled round, some little difference may be made by the wake of each cup affecting the one that follows. Still we cannot be very far wrong by supposing the same proportion, 4. to 1, to hold good in this case. When F is large enough and negative, F may be taken to vary as v*, say to be equal to —Lv”. Similarly, when F is large enough and positive, F may be taken equal to L'v?, where in accordance with the experiment referred to, L’ must be about equal to 4L. Hence we must have nearly— n= — Lé, when € is positive and very large; 7 negative ,, ee Hence if we draw the semi-parabola OAB corresponding to the equa- tion 7=4Lé in the quadrant 7O—é, and the semi-parabola OCD with a latus lectum four times as great in the quadrant €O0—y7, our curve at a great distance from the origin must neariy follow the parabola OAB in the quadrant 7O0—€, and the parabola OCD in the quadrant £O—y, and between the two it will have some flowing form such as PNMK. There must be a point of inflexion somewhere between P and K, not improbably within the positive quadrant €Oy. In the neighbourhood of this point the curve NM would hardly differ from a straight line. Perhaps this may be the reason why Dr. Robinson’s experiments in the paper published in the ‘‘ Phil. Trans.” for 1878 were so nearly represented by a straight line. . * Of course v must be supposed not to be so large as to be comparable with the velocity of sound, since then the resistance to a body impelled through air, or haying air impinging on it, no longer varies as the square of the velocity. + ‘Transactions of the Royal Irish Academy,”’ vol. xxi, p. 168. 1881. ] Investigations on the Spectrum of Magnesium. 18) Ill. “Investigations on the Spectrum of Magnesium.” By G. D. Livetne, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., F.R.S., Jacksonian Professor, University of Cambridge. Received April 28, 1881. [Prate 1.] Since our last communication on this subject (“ Proc. Roy. Soc.,” vol. 30, p. 93) several authors—Ciamician, Cornu and Fievez—have published observations on the spectrum of magnesium, to some of which allusion is made in the sequel, but these observations by no means exhaust the subject. Our own observations, carried on for a considerable time, have extended to new regions and a variety of cir- cumstances, and the summary of them which we now present to the Society will, we hope, help to bring out the connexion between some of the variations in the spectrum of this element and the conditions under which it is observed, and throw additional hght on the ques- tion of the emissive power for radiations of short wave-length of substances at the relatively low temperature of flame to which we alluded in our paper on the spectrum of water (‘* Proc. Roy. Soc.,” vol. 30, p. 580). _ We begin with an account of these observations. Spectrum of the Flame of Burning Magnesiune. When magnesium wire or ribbon is burnt in air, we see the three lines of the 6 group, the blue line about wave-length 4570, first noticed by us in the spark spectrum (‘“‘ Proc. Roy. Soc.,” vol. 27, p. 350) ; and photographs show, besides, the well-known triplet in the ultra- violet between the solar lines K and L sharply defined, and the line for which Cornu has found the wave-length 2850 very much expanded and strongly reversed. These lines are all common to the flame, are, and spark spectra; and the last of them (2850) seems to be by far the strongest line both in the flame and arc, and is one of the strongest in the spark. But, in addition to these lines, the photographs of the flame show a very strong, somewhat diffuse, triplet, generally re- sembling the other magnesium triplets in the relative position of its components, close to the solar line M; and a group of bands below it extending beyond the triplet near L. These bands have, for the most part, each one sharply defined edge, but fade away on the other side; but the diffuse edges are not all turned towards the same side of the spectrum. The positions of the sharp edges of these bands, and of the strong triplet near M, are shown in the figure, No.1. Itisremarkable that the triplets near P and S are absent from the flame spectrum, and that the strong triplet near M is not represented at all either in the are or spark. The hydrogen-magnesium series of lines, begin- ning at a wave-length about 5210, are also seen sometimes, as already - 190 Profs. Liveing and Dewar. — [ Mar. 12, described by us (“‘ Proc. Roy. Soce.,’’ vol. 30, p. 96), in the spectrum of the flame; but we have never observed that the appearance of these lines, or of the strong line with which they begin, is connected with the non-appearance of b,. Indeed, we can almost always see all three lines of the b group in the flame, though as 0, is the least strong of the three, it is likely to be most easily overpowered by the continuous spectrum of the flame. The new observations recorded below leave, we think, no room for doubt that the series of lines beginning at wave-length 5210 are due to a combination of hydrogen with mag- nesium, and are not dependent solely on the temperature. The wave-lengths of the strong triplet near M are about 3720, 3724, 3730, and of the defined edges of the bands about 3750, 3756, 3765, 3772, 3777, 3782, 3790, 3799, 3806, 3810, 3815, 3824, 3841, 3845, 3848, 3855, 3858, 3860, 3865. Burning magnesium in oxygen instead of atmospheric air does not bring out any additional lines; on the contrary, the continuous spec- trum from the magnesia overpowers the line spectrum, and makes it more difficult of observation. Magnesia heated in the oxyhydrogen jet does not appear to give the lines seen in the flame. We have left out of the figure and from the enumeration of lines the well-known bands of the oxide. Spectrum of the Are. By examining the arc of a battery of 40 Grove’s cells, or that of a Siemens’ machine, taken in a crucible of lime, under the disper- sion of the spectrum of the fourth order given by a Rutherford grating of 17,296 lines to the inch, we are able to separate the iron and magnesium lines which form the very close pair 6, of the solar spectrum. Hither of the two lines can be rendered the more prominent of the pair at will, by introducing iron or magnesium into the crucible. The less refrangible line of the pair is thus seen to be due to iron, the more refrangible to magnesium. Comparison of the solar line and the spark between magnesium points confirms this conclusion, that the magnesium line is the more refrangible of the two. In the ultra-violet part of the spectrum photographs show several new lines. First, a pair of lines above U at wave-lengths about 2942, 2938°5.* These lines are a little below a pair of lines given by the spark for which Cornu has found the wave-lengths 29349, 2926-7. The latter pair are not seen at all in photographs of the arc, nor the former in those of the spark. The strong line, wave-length about * [Many of our photographs show besides these two lines a third line wave-length about 2937°5, but we have not been able to determine certainly that it is due to: magnesium, If so this group probably belongs to the series of triplets.—June 2. | 1881. | Investigations on the Spectrum of Magnesium. 191 2850, is always seen, very frequently reversed. Of the quadruple sroup in the spark to which Cornu has assigned the wave-lengths 2801°3, 27971, 27945, and 2789°9, the first and third are strongly developed in the arc, the other two not at all. Next follows a set of five nearly equidistant lines, well-defined and strong, but much less strong than the two previously mentioned, wave-lengths about 2782°2, 2780°7, 2779°5, 2778:°2, 2776°9. The middle line is a little stronger than the others. The same lines come out in the spark. ‘Beyond these follow a series of pairs and triplets; probably they are triplets in every case, but the third, most refrangible, line of the triplets is the weakest, and has not in every case been noticed as yet. These succeed one another at decreasing intervals with diminishing strength, and are alternately sharp and diffuse, the diffuse triplets being the strongest. The positions are shown in fig. 2. The series resembles in general character the sodium and the potassium series described by us in a former communication, and we cannot resist the inference that they must be harmonically related, though they do not follow a simple harmonic law. The most refrangible line in the figure at wave-length 2605 represents a faint diffuse band which is not resolvable into lines; it belongs, no doubt, to the diffuse members of the series, and, to complete the series, there should be another sharp group between it and the line at wave-length 2630. This belonging to the weaker members of the series is too weak to be seen. The approximate wave-lengths found by us for these lines are as follows :—2767°5, 2764°5, 2736, 2732°5,. 2731, 2698, 2695, 2693°5, 2672°5, 2670, 2668°5, 2649, 2646, 2635, 2630, 2605. It is worthy of remark that the line at wave-length 5710, described by us in a previous communication (‘‘ Proc. Roy. Soc.,” vol. 30, p- 98), is very nearly the octave of the strong line at 2850. Moreover the measures we have taken of the wave-length of this last line, with a Rutherford grating of 17,296 lines to the inch, indicate a wave- length 2852 nearly, which is still closer to the half of 5710. In Cornu’s map of the solar spectrum a line is ascribed to mag- nesium with the wave-length 3278. Although a line at this place appears in many of our photographs of the arc, we have not been able to identify it as a line due to magnesium. It does not show any increased strength when magnesium is introduced into the are. When metallic magnesium is dropped into a crucible of magnesia or lime through which the arc is passing, the electric current seems some- times to be conducted chiefly or entirely by the vaporised metal, so that the lines of other metals almost or wholly disappear; but the line at wave-length 3278 does not in such cases appear, though the other magnesium lines are very strongly developed. The line at wave-length 2850 is often, under such circumstances, enormously expanded and reversed, those at wave-lengths 2801, 2794, and the 192 Profs. Liveing and Dewar. [May 12, alternate diffuse triplets, including those near L and near 8, much expanded and reversed, and the group of five lines (2776—2782) sometimes reversed. | ~ When the arc of a Siemens machine is taken in a magnesia crucible, the strong line of the flame spectrum, wave-length 4570, is well seen sharply defined; it comes out strongly and a little expanded on drop- ping in a fragment of magnesium. When a gentle stream of hydrogen is led in through a hollow pole, this line is frequently reversed as .a sharp black line on a continuous background. From comparing the position of this line with those of the titanium lines in its neighbour- hood, produced by putting some titanic oxide into the crucible, we have little doubt that it is identical with the solar line 4570°9 of Angstrom. When the arc is taken in a crucible into which the air has access, it may be assumed that the atmosphere about the arc is a mixture of nitrogen and carbonic oxide. When a stream of hydrogen is passed, either through a perforated pole or by a separate opening, into the erucible, the general effect is to shorten the length to which the arc can be drawn out, increase the relative intensity of the continuous spectrum, and diminish the intensity of the metallic lines. Thus, with a very gentle stream of hydrogen in a magnesia crucible, most of the metallic lines, except the strongest and those of magnesium, dis- appear. Those lines which remain are sometimes reversed ; those at wave-length 2850 and the triplet near Iu being always so. Witha stronger stream the lines of magnesium also disappear, the 0 triplet being the last in that neighbourhood to go, and }, and by remaining after b, has disappeared. ; Chlorine seems to have an opposite effect to hydrogen, generally intensifying the metallic lines, at least those of the less volatile metals, but it does not sensibly affect the spectrum of magnesium. Nitrous oxide produces no marked effect; coal-gas acts much as hydrogen. Spectrum of the Spark. In the spark of an induction coil taken between magnesium points in air we get all the lines seen in the arc except two lines at wave- lengths 4350 and 4166, two lines above U, and the series of triplets more refrangible than the quintuple group about wave-length 2780. The blue line wave-length about 4570 is seen in the spark without a jar when the magnesium electrodes are close together, and the rheotome made to work slowly, but requires for its detection a spec- troscope in which the loss of light is small. On the other hand, some additional lines are seen. Of these, the strong line at wave-length 4481 and the weak line at 4586 are well known. Another faint line in the blue at wave-length 4808* has been * This line we first noticed in a former communication (“ Proc. Roy. Soc.,” 1881.] Investigations on the Spectrum of Magnesium. 193 observed by us in the spark, and two diffuse pairs between H and the triplet near L. Two ultra-violet lines at wave-lengths 2934°9, 2926-7 (Cornu) are near, but not identical with, two lines of the arc above mentioned ; and two more lines at wave-lengths 279771, 27899 (Corna) make a quadruple group with the very strong pair which are con- Spicuous in the arc in this region. The spectrum of the spark ends, so far as we have observed, with the quintuple group (2782—2776) already described in the arc. The lines of this spectrum are given in fig. 3. When a Leyden jar is used with the coil, some of the lines are reversed. ‘T'his is notably the case with the triplet near L, the line at wave-length 2850, and those at 2801 and 2794. Cornu (‘* Compt. Rend.,” 1871) noticed the reversal of the less refrangible two lines of the triplet near L under these circumstances. This effect is very much increased by increasing the pressure of the gas in which the spark is taken. For the purpose of observing the influence of increased _ pressures, we have used a Cailletet pump and glass tubes similar to those employed in the liquefaction of gases by means of such a pump, but with an expansion of the upper part in which were magnesium electrodes attached to platinum wires sealed into the glass. The tube having been filled with gas at the atmospheric pressure, was sealed at its upper end, while the lower end dipped into mercury contained in the iron bottle of the Cailletet pump, and the gas was afterwards com- pressed by driving more or less mercury into the tube. The gases used were hydrogen, nitrogen, and carbonic oxide; and the image of the spark was thrown on to the slit of the spectroscope by a lens. In hydrogen, when no Leyden jar was used, the brightness of the yellow and of the blue lines of magnesium, except at first that at wave-length 4570, diminished as the pressure increased; while, on the other hand, the b group was decidedly stronger at the higher pressure. The pressure was carried up to 20 atmospheres, and then the magnesium lines in the blue and below almost or entirely disappeared, leaving only the b group very bright, and the magnesium-hydrogen bands which are described below; even the hydrogen lines F and C were not visible. When a jar was used, the magnesium lines expanded as the pressure was increased; all three lines of the 6 group were expanded and reversed at a pressure of 5 atmospheres ; the yellow line, wave- length 5528, was also expanded but not reversed ; and the line at 4481 became a broad, very diffuse band, but the line at wave-length 4570 was but very little expanded. The expansion both of the b group and of the yellow line seemed to be greater on the less refrangible than on the more refrangible side of each line, so that the black line in those vol. 27, p. 353), but the wave-length is there given, through an error in taking out the ordinate of the curve of interpolation, as 4797 instead of 4807. Another measure has given the wave-length 4808. 194 | Profs. Livemg and Dewar. [May 12, which were reversed was not in the middle. When the jar was used the pressure could not be carried beyond 10 or 124 atmospheres, as the resistance became then so great that the spark would not pass across the small distance of about 1 millim. between the electrodes. At a pressure of 25 atmospheres, with a jar, the ultra-violet magnesium triplet near L was very well reversed, and the two pairs of lines on its less refrangible side (shown in fig. 3) were expanded into two diffuse bands. In nitrogen and in carbonic oxide the general effects of increased pressure on the magnesium lines (not the magnesium-hydrogen bands) seemed to be much the same as in hydrogen. Without a jar the blue and yellow lines were enfeebled, and at the higher pressures dis- appeared, while the 6 group was very brilliant but not much ex- panded. With the jar all the lines were expanded, and all three lines of the b group strongly reversed. The bands of the oxide (wave- length 4930—5000) were not seen at all in hydrogen or nitrogen; they were seen at first in carbonic oxide, but not after the sparking had been continued for some time. The disappearance of certain lines at increased pressure is in har- mony with the observations of Cazin (‘‘ Phil. Mag.,” 1877, vol. iv, 154), who noticed that the banded spectrum of nitrogen, and also the lines, grew fainter as the pressure was increased, and finally disappeared. When a Leyden jar is employed there is a very great increase in the amount of matter volatilised by the spark from the electrodes, as is shown by the very rapid blackening of the sides of the tube with the deposited metal, and this increase in the amount of metallic vapour may reasonably be supposed to affect the character of the discharge, and conduce to the widening of the lines and the reversal of some of them. Without a jar the amount of matter carried off the electrode also doubtless increases with the pressure and consequent resistance, and may be the cause of the weakening, as Cazin suggests, of the lines of the gas in which the discharge is passed. It is to be noted, moreover, that the disappearance of the hydrogen lines depends, in some degree, on the nearness of the electrodes. The lines C and F which were, as above stated, sometimes invisible in the spark when the electrodes were near, became visible, under circumstances otherwise similar, when the magnesium points had become worn away by the discharge. M. Ch. Fievez (“ Bull. de PAcadémie Royale de Belgique,” 1880, p- 91) has investigated the variations in the appearance of the spark spectrum of magnesium under certain different conditions. Using a Rutherford grating of 17,296 lines to the inch, he has noticed certain lines about the 6 group which increase in number with the order of the spectrum observed. He has also noticed dark lines in the solar spectrum corresponding to these lines of magnesium when the two spectra were superposed (fig. 5). We have noticed similar lines in the — 1881.] Investigations on the Spectrum of Magnesium. 195 spectrum of magnesium given by a Rutherford grating, but attribute them to a different cause. The Rutherford gratings have a periodic in- equality in the ruling, due to an imperfection of the screw of the ruling machine, in consequence of which the image of every bright line is accompanied by a series of fainter images at nearly equal distances on either side of it, diminishing rapidly in brightness as they recede from the principal line. These ghosts are so much fainter than the principal lines that they are not noticed in the case of any but bright lines, and except in the case of very bright lines only two, one on each side, are seen to accompany each principal line. Solar spectrum by Rutherford grating 4th order after M. Fievez. Magnesium. Fie. 7. Magnesium b Spectrum group with ati wen ghosts pro- Bacal by order. Rutherford grating. Spectrum of 3rd order. Fig. 6. The positions of these ghosts have been investigated by Mr. Peirce in the ‘Mathematical Journal” of the Johns Hopkins University, Baltimore, who has found theoretically, and confirmed it by actual 196 Profs. Liveing and Dewar. — [May 12, observation, that the distance between successive images of the same line is directly proportional to the dispersion and inversely as the order of the spectrum. Our own observations of the positions of the ghosts of the 6 group of magnesium lines in spectra of different orders agree closely with Mr. Peirce’s theory, and two different Rutherford gratings both give us the same results. The annexed dia- gram (figs. 6 and 7) gives the relative positions of the first pair of ghosts of each of the lines of the 6 group in the spectra of the third and fourth orders, when the angle between collimator and telescope is 45°. If this is compared with M. Ch. Fievez’s map, it will be seen that he has probably been deceived by these ghosts, both in the solar spectrum and in that of the spark; but as he does not state the angle between his collimator and telescope, no exact comparison can be made. These ghosts are sometimes very embarrassing when many lines are in the field of view, but they may be detected by comparing the spectra of different orders, as the ghosts have different relative positions in the spectra of different orders. In the spectrum of the third order the first ghost of 6, on the more refrangible side falls on b,, and that of b, on its less refrangible side falls on bg. The Magnesium-hydrogen Spectrum. In the “ Proc. Roy. Soc.,”’ vol. 27, p. 494, and vol. 30, p. 93, we have recorded a series of experiments which led us to attribute to magnesium together with hydrogen a peculiar spectrum. This spectrum we have on no occasion been able to detect in the absence of hydrogen. Ob- servations on the spark discharge in nitrogen, in carbonic oxide, and in hydrogen, at reduced pressures, confirmed the results given in the first-mentioned paper, when the discharge was taken in the gases at atmospheric pressure. It was further shown that this peculiar spec- trum could be reversed during the voltaic discharge in a lime crucible, provided magnesium and hydrogen were both present, but not in the absence of hydrogen. Likewise the flame of burning magnesium was found to emit this spectrum when the combustion occurred in ‘an atmosphere containing either free or combined hydrogen. In summing up our results the following opinion was expressed :— “The experiments above described, with nitrogen and carbonic oxide at reduced pressures, are almost 1f not quite conclusive against the supposition that the line at 5210 is due merely to the lower tem- perature of the spark in hydrogen. From De La Rue and Miller’s observations it would appear that nitrogen ata pressure of 400 millims. should produce much the same effect on the spark as hydrogen at 760 millims. Now the pressures of the nitrogen and carbonic oxide were reduced far below this without any trace of the line in question being visible. Moreover, the magnesium line at 4481, which is not 1881. ] Investigations on the Spectrum of Magnesium. US 2 seen in the arc, and may be reasonably ascribed to the higher tem- perature of the spark, may be seen in the spark at the same time as the line at 5210 when hydrogen is present. Nevertheless, tem- perature does seem to affect the result in some degree, for when a large Leyden jar is used, and the gas is at the atmospheric pressure, the line almost disappears from the spark, to reappear when the pressure is reduced; but by no variation of temperature have we been able to see the line when hydrogen was carefully excluded. ‘A line of the same wave-length has been seen by Young in the chromosphere once. Its absence from the Fraunhofer lines leads to the inference that the temperature of the sun is too high (unless at special times and places) for its production. If it be not due to a compound of magnesium with hydrogen, at any rate it occurs with special facility in the presence of hydrogen, and ought to occur in the sun if the temperature were not too high. “We have been careful to ascribe this line and its attendant series to a mixture of magnesium and hydrogen rather than to a chemical compound, because this expresses the facts, and we have. not yet obtained any independent evidence of the existence of any chemical compound of those elements.” Fig. 4 shows more completely than we have given it before the general character of this spectrum, which consists of two sets of flatings and a pair of fainter bands, the flutings closely resembling in character the hydrocarbon flutings, each fluting consisting of a multitude of fine lines closely set on the less refrangible side and becoming wider apart and weaker towards. the more refrangible side, but extending under favourabie circumstances much further than is shown in the figure. The set in the green is the stronger, and it was to this that our former observations were confined. It has two flutings, one beginning at about wave-length 5210 and the other close to b, on its more refrangible side. The other set consists of three principal flutings, of which the first begins at about wave- length 5618, the next at about wave-length 5566, and the third begins with three strong lines at about the wave-lengths 5513, 5512, 5511. Both sets are very well seen when a magnesium wire is burnt in the edge of a hydrogen flame, and in the arc in a crucible of magnesia when a gentle current of hydrogen is led into it. The less refrangible edges of the bands are at wave-lengths about 4849 and 4803. As Mr. Lockyer, in a paper entitled “A New Method of Spectrum Observation” (‘‘ Proc. Roy. Soc.,” vol. 30, p. 22) has brought for- ward this spectrum as illustrative and confirmatory of his views regarding the possibility of elemental dissociation at different heat- levels, we have been induced to review our former work. The view taken by Mr. Lockyer may be expressed in his own words. “The flame spectrum of magnesium perhaps presents us best with 198 Profs. Liveing and Dewar. — | May 12, the beautiful effects produced by the passage from the lower to the higher heat-level, and shows the important bearing on solar physics of the results obtained by this new method of work. ‘‘In the flame the two least refrangible of the components of b are seen associated with a line less refrangible so as to form a triplet. A series of flutings and a line in the blue are also seen. ‘“‘On passing the spark all these but the two components of 6 are abolished. We get the wide triplet replaced by a narrow one of the same form, the two lines of b heing common to both. ‘“‘ May we consider the existence of these molecular states as forming a true basis for Dalton’s law of multiple proportions? If so, then the metals in different chemical combinations will exist in different molecular groupings, and we shall be able, by spectrum observations, to determine the particular heat-level to which the molecular complexity of the solid metal, induced by chemical affinity, corresponds. “* Hxamples.—None of the lines of magnesium special to the flame spectrum are visible in the spectrum of the chloride either when a flame or a spark is employed.” In order to ascertain if this spectrum could be produced at a high temperature in the presence of hydrogen, which element we have already shown to be essential to its production at the atmospheric and at reduced pressures, the series of experiments already mentioned in describing the spark spectrum were made with hydrogen at pressures increasing up to twenty atmospheres. On the supposition that this spectrum originates from the formation of some chemical compound, probably formed within certain limits of temperature when vapour of magnesium is in presence of hydrogen, the stability of the body ought to depend largely on the pressure, of the gaseous medium. Like Graham’s hydrogenium, this body might be formed at a temperature at which 1 would under ordinary cireum- stances be decomposed, provided the pressure of the hydrogen were correspondingly increased. In fact, it has been shown by Troost that the hydrides of palladium, sodium, and potassium all follow strictly the laws of chemical dissociation enunciated by Deville; and increased pressure by rendering the compound more stable, provided the secondary effect of such pressure in causing a higher temperature in the electric discharge were not overpowering, ought to conduce to a more continuous and brillant spectrum of the compound. Conversely, if such a more continuous and brilliant spectrum be found to result, in spite of the higher temperature, from increased pressure, it can only be explained by the stability of the substance being increased with the pressure. Now what are the facts? When the spark of an induction coil, without a Leyden jar, is passed between magnesium electrodes in hydrogen at atmospheric pressure, the flutings in the green are, as before described, always seen, but they are much stronger at the poles 1881.] Investigations on the Spectrum of Magnesium. 199 and do not always extend quite across the field. As the pressure is increased, however, they increase in brilliance and soon extend per- sistently from pole to pole, and go on increasing in intensity, until, at fifteen and twenty atmospheres, they are fully equal in brilliance to the b group, notwithstanding the increased brightness these have acquired by the higher temperature, due to the increased pressure. The second set of flutings, those in the yellowish-green, come out as the pressure is increased, and, in fact, at twenty atmospheres only the b group and the flutings are noticeable ; if the yellow magnesium line be visible at all it is quite lost in the brilliance of the yellow flutings. The tail of fine lines of these flutings extend at the high pressure quite up to the green, and those of the green flutings quite up to the blue. On again letting down the pressure the like phe- nomena occur in the reverse order, but the brilliance of the flutings does not diminish so rapidly as it had increased. If, now, when the pressure has again reached that of the atmosphere, a large Leyden jar be interposed in the circuit, on passing the spark the flutings are still seen quite bright, and they continue to be seen with gradually diminishing intensity until the sparks have been continued for a con- siderable time. It appears that the compound, which kad been formed in large quantity by the spark without jar at the higher pressures, is only gradually decomposed, and not re-formed, by the high temperature of the spark with jar. This experiment, which was several times repeated, is conclusive against the supposition that the flutings are merely due to a lower temperature. When the pressure was increased at the same time that the jar was employed, the flutings did not immediately disappear, but the expansion of the magnesium lines and the increase of the continuous spectrum seemed to over- power them. When nitrogen was substituted for hydrogen, the strongest lines of the green flutings were seen when the spark without jar was first passed at atmospheric pressure, probably from hydrogen occluded, as it usually is, in the magnesium electrodes. As the pressure was increased they speedily disappeared entirely and were not again seen either at high or low pressures. With carbonic oxide the same thing occurred as with nitrogen; but in this gas the flutings due to the oxide of magnesium (wave-length 4930 to 5000) were, for a time, very well seen. Ciamician (“Sitzungsber. Akad. Wissensch.,” Wien, 1880, p. 437) has described a spectrum of magnesium of the first order (in Plicker’s nomenclature) obtained by taking sparks from an induc- tion coil, without a jar, between magnesium electrodes in an atmo- sphere of hydrogen. He gives a figure to a scale of this spectrum, but it is not to a scale of wave-lengths, so that exact comparison of his observations with ours is difficult. The least refrangible set of VOL. XXXII. P 200 Profs. Liveing and Dewar. [May 12 flutings in his figure corresponds very well with that we have described in the yellowish-green. The next set, in the green, in his figure does not, however, correspond exactly with ours; it begins nearer to ) than we have observed and consists of four flutings, whereas we observe but two in this set. It looks as if, in his figure, the magnesium-hydrogen spectrum were superposed upon the hydro- carbon spectrum in this region. Further, he gives a third more refrangible set of flutings which we have only observed as two blue bands, not fluted. This third set of flutings, as drawn in his figure, appears to be somewhat more refrangible than the set due to the oxide, and occupies partly the place of the blue hydrocarbon series, but a passage in the text, in which he says that the mag- nesium spectrum of the second order might, without measurement, easily be taken to be identical with that of carbon, almost negatives the supposition that this set of flutings is the blue hydrocarbon set and mistaken for a magnesium spectrum of the first order. ‘To what- ever it may be due, we have not seen anything closely resembling it under the circumstances described by him, though our observations on the spark spectrum of magnesium in hydrogen have now been repeated with all the variations of circumstance which we could devise. Mr. Lockyer states (loc. cit.) that none of the lines of magnesium, special to the flame spectrum, are visible in the spectrum of the chloride, either when a flame or aspark is employed. But we find that when the spark is taken between platinum points from a solution of the chloride of magnesium, in a tube such as those used by Delachanal and Mermet, the line at wave-length 5210 can frequently be seen in it when the tube is filled with air, and that if the tube be filled with hydrogen the green flutings of magnesium-hydrogen are persistent and strong. Repeated observations have confirmed our previous statements as to the facility with which the magnesium-hydrogen spectrum can be produced in the are by the help of a current of the gas. In a magnesia crucible, by regulating the current of hydrogen, the flutings can be easily obtained either bright or reversed. Comparison of the Spectra. When we compare the spectra of magnesium in the flame, are, and spark, we observe that the most persistent line is that at wave-length 2850, which is also the strongest in the flame and arc, and one of the strongest in the spark. The intensity of the radiation of magnesium _ at this wave-length is witnessed by the fact that this line is always reversed in the flame as well as in the arc when metallic magnesium is introduced into it, and in the spark between magnesium electrodes when a Leyden jar is used. It is equally remarkable for its power of 1881. ] Investigations on the Spectrum of Magnesium. 201 expansion. In the flame it is a broad band, and equally so in the are when magnesium is freshly introduced, but fines down to a narrow line as the metal evaporates. Almost equal in persistence are the series of triplets. Only the least refrangible pair of these triplets is seen in the flame, another pair is seen in the spark, but the complete series is only seen in the arc. We regard the triplets as a series of harmonics, and to account for the whole series being seen only in the arc we must look to some other cause than the temperature. This will probably be found in the greater mass of the incandescent matter contained in the crucible in which the arc was observed. The blue line of the flame at wave-length 4570 is well seen in the are, and is easily reversed, but is always a sharp line, increased in brightness but not sensibly expanded by putting magnesium into the crucible. In the spark, at atmospheric pressure, it is only seen close to the pole or crossing the field in occasional flashes; but seems to come out more decidedly at rather higher pressures, at least in hydrogen. The series of bands near UL, well developed in the flame, but not seen at all in the arc or spark, look very much like the spectrum of a compound, but we have not been able to trace them to any particular combination. Sparks in air, nitrogen, and hydrogen have alike failed to produce them. The very strong, rather diffuse triplet at M, with which they end, so closely resembles in general character the other magnesium triplets, that it may well be connected with that constitu- tion of the magnesian particle which gives rise to the triple sets of vibrations in other cases, but, if so, its presence in the flame alone is not easily explained. The occurrence of this triplet in the ultra-violet, and of the re- markable series of bands associated with it, as well as the extra- ordinary intensity of the still more refrangible line at wave-length 2850, which is strongly reversed in the spectrum of the flame, corroborates what the discovery of the ultra-violet spectrum of water had revealed, that substances at the temperatures of flames while giving in the less refrangible part of the spectrum more or less continuous radiation, may still give, in the regions of shorter wave-length, highly discontinuous Spectra, such as have formerly been deemed characteristic of the highest temperatures. This subject we will not discuss further at present, but simply remark what we have stated formerly, that ‘it opens up questions as to the enormous power for radiation of short. wave-length of gaseous bodies at the comparatively low temperature of flame with regard to which we are accumulating facts.” In the are and spark, but not in the flame, we have next a very striking group of two very strong lines at wave-lengths about 2801 and 2794, and a quintuple group of strong but sharp lines above Pp 2 202 Investigations on the Spectrum of Magnesium. [May 12, them. The former are usually reversed in the spark with jar, and all are reversed in the arc when much magnesium is present. There are also several single lines in the visible part of the spectrum common to the arc and spark. All-of these may be lines developed by the high temperature of the arc and spark. An indigo and a violet line in the arc have not been traced in the spark, but their non-appearance may be due to the same cause as that above suggested for the non- appearance of the higher triplets, the smallness of the incandescent mass in the spark. A pair of lines in the are near U appear to be represented in the spark by an equally strong, or stronger, pair near but not identical in position. The possibility of such a shift, affecting these two lines only in the whole spectrum and affecting them unequally, must in the present state of our knowledge ‘be very much a matter of speculation. Perhaps sufficient attention has not hitherto been directed to the probability of vibrations being set up directly by the electric discharge independently of the secondary aetion of elevation of temperature. Some of the observations above described, and many others well known, indicate a selective action by which an electric discharge lights up certain kinds of matter in its path to the exclusion of others ; and it is possible that in the case of vibrations which are not those most easily assumed by the particles of magnesium, the character of the impulse may slightly affect the period of vibration. The fact that, so far as observations go, the shift in the case of this pair of magnesium lines is definite and constant, militates against the supposi- tion suggested. On the other hand, the ghost-like pairs of lines observed in the spark below the triplet near L, suggest the idea that some of the particles have their tones flattened by some such cause. The strong pair at wave-length 2801, 2794, are accompanied in the spark, but not in the arc, by a much feebler, slightly more refrangible pair, but these have not the diffuse ghost-like character of those just alluded to. These lines are phenomena of the high potential discharge in which particles are torn off the electrodes with great violence, and may well be thrown into a state of vibration which they will not assume by mere elevation of temperature. There are two lines in the spark besides the well-known line at wave-length 4481 which have not been observed in the arc, but they are feeble and would be insignificant if it were not the fact that they, as well as the line at wave-length 4481, all short lines seen generally only about the poles, appear.to be present in the solar spectrum. In thesun | we seem to have all the lines-common to the flame, arc, and spark (unless the line given in Angstrém’s map at 4570°9 be not identical as we believe it to be with the magnesium line), and possibly, judging by Rutherford’s photograph, the strong triplet of the flame at M; but one line common 46 | | Ca HeROCr LOVE 0G) VOU. Sele Vale 41 Ce eral We heer eae . — oe eee beri | LOUeren ean 3000 29 28 R17 26 2500 i | | aa ieee Geeben dimer t as | a West,Newmon & C? se. = Liveing & Dewar. Proc. Roy. Soc. Vol. 82 Pl. 1. SPECTRUM, ; 4 53 52 51 5000 49 48 47 46 45 44 4 42 ii pauila ee ee revel irene rare ere tee Menno uel walle sec thot Fig.l. Flame - L Ss setae a ee ote ae tobi mellacie aeeudace Geen all Bonu beeeeemade aeudladema ueeulasuadaeuc ai Geena MenmblMinoen Gen! moumeece buena Fig. 2. ; hieee ie Cocoddl doteuenbebuubellauelbenielmaiil Fis Fee Ee sie eeeel euebumed lenin / Hobo Fig.3. [ | sparle boob a mul at ave Wooded 41 4000 39 38 37 36 35 x 34 Bis) 32 31 3000 29 28 27 26 2500 got reread ere re ere oe ere renee Toren Te eee seinen cree eee neliceeeudiee| seme = tical eel eters rere enel lh cr ener vse vet bine enlaere Creer resister elcid beatae feretilen oreteonet i ie Fig. 2. | ; Ly ieee Once feted Li Lil speed ec enue i lL iil irerae Stet | E LI arene ee ATI i ee ee ii L | Lit i (ite era eae os spare | | | Paeee| eee! West, Newman & C? #c. i . Keites Che a de a OL ea BRE piety nF nfias Wy “cided Beh h 12 J : Lea sth Peles x REF > . ay) ( < oF * RE Ab - f ; gee we ria cS - é Veo i aa 5 | b A ‘| : = ni Mi . 1 Ab Tonh | 7 a > i Peer | F f i ‘ . <4 i } ‘ < {: ; j Bey : a “0 3 i ‘ 5 Z s 2 y . 3 > f mt ray ~ 4 é et ted ll ee Sein ihe re Se . * ' ee t ae ‘ 2 Val ' Cs ae i on ‘ . is We f pet] i : ; i , can h ‘ pantie ri im Si aici = eh: cal . “ 7 5 * b t i » te V , 1 art ony i x 4 bs ceatylt i ro. ~. ir ‘ “ Roan ; i fe y : i it mw a 5, « - i pf f Py x 1881.] On the Spectra of Sun-spots. 203 to the are and spark at wave-length 4703 does not appear in Angstrém’s map. It is hard to account for its absence, as it is present in Kirch- hoff’s map and in Rutherford’s photograph. We have noticed, however, that when the spark is taken in hydrogen, the line at wave-length 4570 appears stronger than that at wave-length 4.703, while the reverse is the case when the atmosphere is nitrogen. It is possible then that the atmosphere may, besides the resistance it offers to the discharge, in some degree affect the vibrations of the metallic particles.. When we have made all the simplifications that we can by elimi- nating, as we hope we have done satisfactorily, the hydrogen-magnesium flutings, and by supposing the whole series of triplets to be harmoni- cally related, and possibly some of the single lines also to be similarly related, we have still the fact that the chemical atoms of magnesium are either themselves capable of taking up a great variety of vibra- tions, or are capable by mutual action on each other, or on particles of matter of other kind, of giving rise to a great variety of vibrations of the luminiferous ether; and to trace satisfactorily the precise con- nexion between the occurrence of the various vibrations and the circumstances under which they occur, will require yet an extended series of observations. IV. “ Note on the Reduction of the Observations of the Spectra of 100 Sun-spots observed at Kensington.” By J. NORMAN LockYEr, F'.R.S. Received May 12, 1881. [PLATE 2. | In anticipation of a more detailed communication, I beg to lay before the Royal Society some of the results of the reduction of the six most widened lines between F and 6 seen in the spectra of 100 sun-spots, observed at Kensington between November 12th, 1879, and September 29th, 1880, limiting my remarks solely to the spectrum of iron. | In the accompanying map, the Fraunhofer lines agreeing in position with the iron lines given by Angstrom and Thalén are entered in the horizon headed “Sun,” in the next are plotted the lines assigned to iron by Angstrom, who used the electric arc in his experiments. In the next horizon are entered the iron lines given by Thalén, who employed the induction coil in his experiments. In these three horizons the lengths of the lines represent their intensities. The individual observations of the sun-spots having been plotted out on another map, the number of times each line was seen was ascer- tained, and is entered in the next horizon under “ Frequencies in Sun- Spots.” 204 Mr. J. N. Lockyer. | [May 12, In the next horizon are entered the frequencies with which iron lines were seen in observations of a hundred flames by Tacchini. In both these horizons the lengths of the lines represent frequency. This map shows that within the region F to b, there is ne line com- mon to spots and flames, if the lines 6, b,, which are so frequently seen affected in both spots and flames, be neglected. If, therefore, we were unacquainted with the spectrum of iron, we should be justified in say- ing that the spectrum of the prominences was due to one substance, and that of the spots to another.* In a paper presented to the Royal Society in December, 1878, I drew attention to the fact that many of the lines most frequently seen in the flames by Young and Tacchini were assigned by Angstrém to the spectra of two so-called elements. In the map I have entered his results in the horizon arc A. I have since that time had prepared a table, showing the lines having coincident readings in two or more metals, according to Thalén, and in my paper of March 5th, 1879, “ Discussion of Young’s List of Chromospheric Lines,” showed that some of these lines could not be due to impurity. The lines given coincident by Thalén in his tables, are entered in the horizon spark T. | I have confirmed in most cases Thalén’s and Angstroém’s work, and have proved that these lines could not be due to an impurity of the one in the other, as the longest lines of each were absent from the other. I used the arc, quantity coil, and intensity coil, which are respectively indicated by are L, spark LQ, and LI, with high dis- persion. : Besides confirming Thalén’s and Angstrom’s work, I have been able to add a few more basic lines to the list. I have already pointed out that the fact of different rates of motion being indicated by different iron lines in the same field of view at the same time, afforded important evidence that we were not dealing with iron itself, but with primitive forms of matter contained in iron, which are capable of withstanding the high temperature of the sun, after the iron, observed as such, has been broken up as suggested by Brodie.+ An appeal to the principles of continuity and evolution now enables us to add another argument of equal weight. If on the cooling of a * Tt will be observed that the line at 50175 given among the flame observations in the map is not recorded as an iron line by either Angstrom or Thalén. The way, however, in which it sympathised with the line at 4923 in the flames induced me to look for it in iron with the intensity coil. I at once found it, and it is as sympa- thetic with the line at 4923 in the spark as it is in the flames themselves. Though ~ omitted by Thalén (most likely accidentally, thougn it may well be that he did noé use sufficient tension to bring it out strongly), it is recorded as an iron line by other observers. + Or rather, of course, before “iron” had been formed by condensation. Proce. Roy. Soc. Vol. 02 .PU.Z. 6 8 | OPARK THALEN 20 | FREQUENCIES i50| SUN_SPOTS | 30| KENSINGTON FREQUENCIES FLAMES PALERMO Are West, Newman & C? sce. d ’ 0 4 cay 7 , aK 1 2 i fey ve Ne Z \ ‘ “i } Pa sa GP eRaR ERE EOUGREOAUUGULLUERL | as Oa 4 7 = F TO TTT So | Hii (apenas HH CH re] ese [cme] 2] to SSL SUN —|ARC ANGSTROM aaa ee THALEN : : ° E 20 | FREQUENCIES 30 poset 3 ! 5) uncsaons | : ; : ele 2 le : - \__| g9| KENSINGTON ae eh = = Li s - ae ee 90 100 Baers ete , ! ag | ie | | | _|FREQUENCIES _——— a aa Peeabe l -r, | eee | ¢0] FLAMES aaa Taga 7a as : er ease II | a is eee ——| 40} PALERMO | | | i ie " : oe Wost, Neveman & C? 20 ae re mie) es Fe es ee 4 ; 1881.] | On the Spectra of Sun-spots. 205 mass of vapour ultimately to form a star, each reduction of tempera- ture increases the number and complexity of the chemical forms by rendering new combinations possible, these new combinations will con- tain the earlier ones in different proportions. If, for instance, the members of the iron groups are not elementary, they will contain earlier forms, as the salts of calcium contain calcium, which once existed as calcium in the atmosphere of the earth before the salts were produced as the result of a subsequent condensation brought about by cooling. The discussion in the accompanying map shows that this is really the case, and that it is the lines which are common to two or more substances which in the main produce the spectra of the lower, and therefore hotter, region of the solar atmosphere. This natural result at once explains the strange variations from the ordinary spectra which have puzzled observers ever since the new method was intro- duced. More than this, these are precisely those lines which have their intensities strengthened when we pass from the arc to the coil. It is very instructive to note the gradual simplification of the iron spectrum by increased temperature as we pass from the are through the spots to the flames, and how with this increasing simplicity we find . the basic character of the lines increasing, only one basic line on the map having been yet observed among those lines seen in the arc alone. Solar spectrum. mm Arc. Quantity coil, with jar. as 57) BON Ars Intensity coil, with jar. ” », nO jar. eae Spots. fees Prom. Tacchini. ye wounss @a Reversed in penumbra of spot, August 5, 1872. Young. ; = Motion. The accompanying figure shows what happens with regard to three adjacent iron lines under different solar and terrestrial conditions, and I give it because it indicates that one of the most important inquiries to be taken up in the eclipses of next year and 1883, will be the 206 Mr. W. Crookes. [May 19, determination of the intensities of the spot and flame lines, and the heights in the solar atmosphere at which they are visible at the moment of disappearance and reappearance of the sun. The spectrum of the lower parts of the corona itself should also be observed, with a proper amount of dispersion with instruments affording a great anantity of light. I stated some time ago to the Royal Society that the iron lines visible in the are but not in the spark, in all probability owe their existence among the Fraunhofer lines to an absorption going on in the cooler levels, above the flame and spot regions. A look at the map will show how important it will be to adjust a spectroscope on the lines shown in the figure, and contrast their behaviour with the other twelve adjacent arc lines which are prominent neither in spark nor spot, nor flame. It is to be feared that any observer who attempts to do more than such a restricted piece of work as this will doubt his own results. The time has arrived when minute observations must take the place of mere general ones. May 19, 1881. THE PRESIDENT in the Chair. The Presents received were laid on the table and thanks ordered for them. The Right Hon. William Ewart Gladstone was admitted into the Society. The following Papers were read :-— J. “On Discontinuous Phosphorescent Spectra in High Vacua.” By WILLIAM CROOKES, F'.R.S. Received March 31, 1881. In a paper which I had the honour of presenting to the Royal Society in March, 1879,* I drew attention to the fact that many sub- stances, when in high vacua and submitted to the molecular discharge by means of an induction coil, emitted phosphorescent light; and I especially mentioned the phosphorescent sulphides, the diamond, the ruby, and various other forms of alumina, crystalline and amorphous. Pure alumina chemically prepared has very strong phosphorescence. Sulphate of alumina is dissolved in water, and to it is added an excess * “Phil. “Prans;;’ vol: 170: 1881.] Discontinuous Phosphorescent Spectra in High Vacua. 207 of solution of ammonia. The precipitated hydrate of alumina is filtered, washed, ignited, and tested in the molecular stream. It phosphoresces of the same crimson colour, and gives the same spectrum as the ruby. Alumina in the form of ruby glows with a full rich red colour, and when examined in the spectroscope the emitted light is seen to be dis- continuous. There is a faint continuous spectrum ending in the red somewhere near the line B; then a black space, and next an intensely brilliant and sharp red line, to which nearly the whole of the intensity of the coloured glow is due. The wave-length of this red line, which appears characteristic of this form of alumina, is, as near as I can measure, \ 689°5 m.m.m. This line coincides with the one described by H. Becquerel as being the most brilliant of the lines in the spectrum of the light of alumina in its various forms, when glowing in the phosphoroscope. This coincidence is of considerable inierest, as it shows a relation between the action of molecular impact and of sunhght in producing luminosity. The phosphorescence induced in a crystal of ruby by the molecular discharge is not superficial, but the light comes from the interior of the crystal, and is profoundly modified according as its direction of vibration corresponds or makes an angle with the axis of the crystal, being quenched in certain directions by a Nicol prism. Sunlight falling on the ruby crystal produces the same optical phe- nomena. The light is internally emitted, and on analysis by a prism, is seen to consist essentially of the one brilliant crimson line, \ 689°5. This fact may account for the extraordinary brilliancy of the ruby, which makes it so highly prized as a gem. The sun not merely renders the red-coloured stone visible, as it would a piece of coral, but it excites the crystal to phosphorescence, and causes it to glow with a luminous internal light, the energy of which is not diffused over a broad portion of the spectrum, but is chiefly concentrated into one wave-length. The crimson glow of alumina remains visible some time after the current ceases to pass. When the residual glow has ceased it can be revived by heating slightly with a spirit-lamp. After long experimenting with chemically pure alumina precipitated from the sulphate as above described, a curious phenemenon takes place. When sealed up in the vacuum two years ago it was snow white ; but, after being frequently submitted to the molecular discharge for the purpose of exhibiting its brilliant phosphorescence, it gradually assumes a pink tinge, and on examination in sunlight a trace of the alumina line can be detected. The repeated molecular excitation is slowly causing the amorphous powder to assume a crystalline form. Under some circumstances, alumina glows with a green colour. Ammonia in large excess was added toa dilute solution of alum. The 208 Mr. W. Crookes. [May 19, strong ammoniacal solution filtered from the precipitated alumina was now boiled. The alumina which the excess of ammonia had dis- solved was thereby precipitated. This was filtered off, ignited, and tested in the molecular discharge. It gave no red light whatever, but phosphoresced of a pale green, and on examination with a prism the light showed no lines, but only a concentration of ight in the ereen. Two earthen crucibles were tightly packed, the one with sul- phate of alumina, the other with acetate of alumina. They were then exposed, side by side, to the most intense heat of a wind furnace —a heat little short of the melting-point of platinum.* The resulting aluminas were then tested in the molecular stream. The alumina from the sulphate gave the crimson glow and the spectrum line. The alumina from the acetate gave no red glow or line, but a pale green phosphorescence. In my examination of rubies, many pounds of which have passed through my apparatus, | have been fortunate enough to meet with one solitary crystal, not to the eye different from others, which emits a green light when tested in the molecular stream. All others act as I may call normaily. Thespectrum of this green-glowing crystal shows, however, a trace of the red line, and on keeping the discharge acting on it for a few minutes the green phosphorescence grows fainter and a red tinge is developed, the spectrum line in the red becoming more distinct. Besides the ruby, other native forms of crystallised alumina phos- phoresce. Thus corundum glows with a pink colour. The sapphire’ appears to be made up of the red-glow and green-glow alumina. Some fine crystals of sapphire shine with alternate bands of red and green, arranged in layers perpendicular to the axis. Unfortunately it is im- possible to prepare a tube for exhibition containing this variety of sapphire, as it is constantly evolving gas from the numerous fissures and cavities which abound in this mineral. The red glow of alumina is chiefly characteristic of this earth in a free state. Few of its compounds, except spinel (aluminate of mag- nesium), either natural or artificial, show it in any marked degree. All the artificially crystallised aluminas give a strong red glow and spectrum line. An artificially crystallised aluminium and barium fluoride phosphoresces with a biue colour, but shows the red alumina line in the spectrum. Spinel glows red, and gives the red line almost as strong as the ruby. The mineral spodumene (an aluminium and lithium silicate) phos- phoresces very brilliantly with a rich golden-yellow colour, but shows no spectrum line, only a strong concentration of light in the orange * This operation was kindly performed for me by Messrs. Johnson and Matthey. 1881.] Discontinuous Phosphorescent Spectra in High Vacua. 209 and yellow. A phosphorescing crystal of spodumene has all the internal light cut off with a Nicol prism, when the long axes of the Nicol and the crystal are parallel. It became of interest to see if the other earths would show phos- phorescent properties similar to those of alumina, and especially if any of them would give a discontinuous spectrum; considerable interest attaching to a solid body whose molecules vibrate in a few directions only, giving rise to spectrum lines or bands on a dark background. Glucina prepared with great care is found to phosphoresce with a bright blue colour, but no lines can be detected in the spectrum, only a concentration of light in the blue. The rare mineral phenakite (silicate of glucinum), sometimes used as a gem, phosphoresces blue like pure glucina. This mineral shows a residual glow after the current is turned off. Thorina has very little, if any, phosphorescence. This earth is, however, remarkable for its very strong attraction for the residual gas in the vacuum tube. On putting thorina in a tube furnished with well-insulated poles, whose ends are about a millimetre apart in the centre, and heating strongly during exhaustion, the earth, on cooling, absorbs the residual gas with such avidity that the tube becomes non- conducting, the spark preferring to pass several inches in air rather than strike across the space of a millimetre separating the two poles. It is probable that this strong attraction for gas is connected with the great density of the earth thorina (sp. gr. = 9°4). Zirconia gives a very brilliant phosphorescence, approaching in in- tensity that of sulphide of calcium. The colour is pale bluish-green, becoming whiter as the intensity of the discharge increases: no lines are seen in its spectrum. Lanthana precipitated as hydrate and ignited shows no phos- phorescence. After it has been heated for some time before the blow- pipe it phosphoresces of a rich brown. Didymia, from the ignition of the hydrate, has sandal any phos- phorescence ; what little there is appears to give a continuous spec- trum with a broad black band in the yellow-green. On examining the light reflected from this earth when illuminated by day or artificial light, the same black band is seen, and with a narrow slit and sunlight the band is resolved into a series of fine lines, occupying the position of the broadest group of absorption lines in the trans- mission spectrum of didymium salts. Yttria shows a dull greenish light, giving a continuous spectrum. Erbia phosphoresces with a yellowish colour, aud gives a continuous spectrum, with the two sharp black bands so characteristic of this earth cutting through the green at \ 520 and 523. These lines are easily seen in the light reflected from erbia when illuminated by day- light. Itis well known that solid erbia heated in a flame glows with 210 i Mr. W. Crookes. : OE Miary 8 a green hght, and gives a spectrum which chiefly consists of two bright green lines in the same place as the dark lines seen by reflected hight. A curious phenomenon is presented by erbia when the spark passes over it at a high exhaustion. The particles of earth which have acci- dentally covered the poles are shot off with great velocity, forming brightly luminous lines, and, striking on the sides of the tube, rebound, remaining red-hot for an appreciable time after they have lost their velocity. They form a very good visible iliustration of radiant matter. Titanic acid phosphoresces dark brown, with gold spots in places. Stannic acid gives no phosphorescence. Chromic, ferric, and ceric oxides do not appreciably phosphoresce. * Magnesia phosphoresces with a pink opalescent colour, and shows no spectrum lines. Baryta (anhydrous) scarcely phosphoresces at all. Hydrated baryta, on the contrary, shines with a bright orange-yellow light, but _shows no discontinuity of spectrum; only a concentration in the yellow-orange. Strontia (hydrated) phosphoresces with a beautiful deep blue colour, and when examined in the spectroscope the emitted light shows a greatly increased intensity at the blue and violet end, with- out any lines or bands. Lime phosphoresces of a bright orange-yellow colour, changing to opal-blue in patches where the molecular discharge raises the tem- perature. In the focus of a concave pole the lime becomes red and white-hot, giving out much light. This earth commences to phos- phoresce more than 5 millims. below the vacuum, and continues to grow brighter as long as the electricity is able to pass through the tube. On stopping the discharge there is a decided residual glow. No lines are seen in the spectrum of the light. Calcium carbonate (calcite) shows a strong phosphorescence, which begins to appear at a comparatively low exhaustion (5 millims.). The interior of the crystal shines of a bright straw colour, and the ordinary and extraordinary rays are juminous with oppositely polarised light. Calcite shows the residual glow longer than any substance I have as yet experimented with. After the current has been turned off the crystals shine in the dark with a yellow light for more than a minute. | Calcium phosphate generally gives an orange-yellow phospho- rescence and a continuous spectrum. Sometimes, however, a yellow-. green band is seen superposed on the spectrum. Potash phosphoresces faintly of a blue colour. The spectrum shows a concentration at the blue end, but the light is too faint to enable lines, if any, to be detected. 1881.] Discontinuous Phosphorescent Spectra in High Vacua. 211 Soda phosphoresces faintly yellow, and gives the yellow line in the spectrum. Lithium carbonate gives a faint red phosphorescence. Hxamined in the spectroscope, the red, orange, and blue lithium lines are seen. I have already said that the diamond phosphoresces with great ‘brilliancy. In this respect perfectly clear and colourless stones “‘ of the first water’ are not the most striking, and they generally glow of a blue colour. Diamonds, which in sunlight have a shght fluores- cence disappearing when yellow glass is interposed, generally phos- phoresce stronger than others, and the emitted light is of a pale yellowish-green colour. Most diamonds which emit a very strong yellowish light in the molecular discharge give a continuous spectrum, having bright lines across it in the green and blue. A faint green line is seen at about X 537; at \ 513 a bright greenish-blue line is seen, and a bright blue line at ) 503. A darkish space separating the last two lines. Diamends which phosphoresce red generally show the yellow sodium line superposed on a continuous spectrum. There is great difference in the degree of exhaustion at which various substances begin to phosphoresce. Some refuse to glow until the exhaustion is so great that the vacuum is nearly non-conducting, whilst others commence to become luminous when the gauge is 5 or 10 millims. below the barometric level. The majority of bodies, how- ever, do not phosphoresce till they are well within the negative dark space. During the analysis ef some minerals containing the rarer earths experimented on, certain anomalies have been met with, whieh seem to indicate the possible presence of other unknown elements awaiting detection. On several oecasions an earthy precipitate has come down where, chemically speaking, no such body was expected; or, by frac- tional preeipitation and solution, from a supposed simple earth some- thing has separated which, in its chemical characters, was not quite identical with the larger portion; or, the chemical characteristics of an earth have agreed fairly well with those assigned to it in books, but it deviated in some physical pecuharity. Jt has been my practice to submit all these anomalous bodies to molecular bombardment, and I have had the satisfaction of discovering a class of earthy bodies which, whilst they phosphoresce strongly, also give spectra of remark- able beauty. The spectrum seen most frequently is given by a pale yellowish coloured earth. It consists of a red, orange, citron, and green band, nearly equidistant, the citron being broader than the others and very bright. Then comes a faint blue, and lastly two very strong blue- violet bands. These bands, when seen at thei: best, are on a per- fectly black background ; but the parent earth gives a continuous 212 Phosphorescent Spectra in High Vacua. [May 19, spectrum, and it is only occasionally, and as it were by accident, that I have so entirely separated it from the anomalous earth as to see the bands in their full purity. Another earthy body gives a spectrum similar to that just described, but wanting the red, and having a double orange and double citron band. A third gives a similar spectrum, but with a yellow line interposed between the double orange and the double citron, and having two narrow green lines. At present I do not wish to say more than that I have strong indications that one, or perhaps several new elements are here giving signs of their existence. The quantities I have to work upon are very small, and when each step in the chemical operation has to be checked by an appeal to the vacuum tube and to the induction coil the progress is tediously slow. In the thallium research it only occupied a few minutes to take a portion of a precipitate on a platinum loop, introduce it into a spirit-flame, and look in the spectro- scope for the green line. In that way the chemical behaviour of the new element with reagents could be ascertained with rapidity, and a scheme could be promptly devised for its separation from accompany- ing impurities. Here, however, the case is different: to perform a spectrum test, the body under examination must be put in a tube and exhausted to a very high point before the spectroscope can be brought to bear on it. Instead of two minutes, half a day 1s occupied in each operation, and the tentative gropings in the dark, unavoidable in such researches, must be extended over a long period of time. The chemist must also be on his guard against certain pitfalls which catch the unwary. I allude to the profound modification which the presence of fluorine, phosphorus, boron, &., causes in the chemical reactions of many elements, and to the interfering action of a large quantity of one body on the chemical properties of another which may be present in small quantities. The fact of giving a discontinuous phosphorescent spectrum is in itself quite insufficient to establish the existence of a new body. At present it can only be employed as a useful test to supplement chemical research. When, however, I find that the same spectrum- forming earthy body can always be obtained by submitting the mineral to a certain chemical treatment; when the chemical actions which have separated this anomalous earth are such that only a limited number of elements can possibity be present; when I find it impossible to produce a substance giving a similar discontinuous spectrum by mixing together any or all of the bodies which alone could survive the aforesaid chemical treatment; when all these facts are taken into consideration, and when due weight is given to the very characteristic spectrum reaction, I cannot help concluding that the most probable explanation is that these anomalies are caused by the presence of an unknown body whose chemical reactions are not 1881. ] Prof. D. E. Hughes. Molecular Magnetism. 213 sufficiently marked to have enabled chemists to differentiate it from associated elements. III. “ Molecular Magnetism.” By Professor D. EK. HuGHEs, F.R.S. Received May 10, 1881. During the course of some late researches, which I had the honour to communicate to the Royal Society, March 7th,* and experimentally illustrate on the reading of my second paper, March 31,¢ many experimental facts occurred, all pointing to the conclusion that ordinary molar magnetism is entirely due to the symmetrical arrange- ment of the polarised molecules, and that these molecules can be rotated by torsion, so as to decrease the longitudinal magnetism, or increase it, if the effect of the elastic torsion is to rotate the molecules into the required longitudinal symmetrical arrangement. And observing that molecular magnetism could induce an electric current upon its own molar constituents, or that an electric current by its passage through an iron wire would produce molecular magnetism, I have continued these researches in the hope of elucidating, as far as possible, the pheno- menon of the transformation of electricity and magnetism by the changes produced in the molecular structure of its conducting wire. For this purpose I have employed three separate methods of investi- gation, each requiring a slightly modified form of apparatus. The first relates to the influence of an elastic torsion upon a magnetic or conducting wire; the second, to the influence upon the molecular struc- ture of an iron wire of electricity or magnetism; the third, to the evident movement of the molecules themselves as given out in sonorous vibrations. The general details of the apparatus employed having been given in my paper of March 7th, I will only briefly indicate any modification of the method employed. 1. Influence of an Elastic Torsion upon a Magnetic or an Electric Conducting Wire. In my paper of March 7th on ‘‘ Molecular Electro-Magnetic Induc- tion,” I showed that induced currents of electricity would be induced in an iron wire placed on the axis of a coil through which intermittent currents were passing, and that these currents were produced only when the wire was under the influence of a torsion not passing its limit of elasticity. It became evident that if the intermittent mag- netism induced by the coil produced under torsion intermittent currents * “ Molecular Electro-Magnetic Induction,” “ Proc. Roy. Soc.,” vol. 31, p. 525. 7 “ Permanent Molecular Torsion of Conducting Wires produced by the Passage of an Electric Current,” Zbid., vol, 32, p. 25. 214 Prof. D. E. Hughes. [May 19, © of electricity, an intermittent torsion under the influence of a con- stant current of electricity or a constant magnetic field would pro- duce similar currents. This was found to be the case, and as some new phenomena presented themselves indicating clearly the molecular nature of the actions, I will describe a few of them directly relating to the subject of this paper. The apparatus used was similar to that described in my paper of March 7th. An iron wire of 20 centims. was placed in the centre or axis of a coil of silk-eovered copper wire, the exterior diameter of the coil being 55 centims., and that of the interior vacant circular space 34 centims. The iron wire is fastened te a support at one end, the other passing through a guide, to keep it parallel to the axis but free, so that any required torsion may be given to the wire by means of a connecting arm or index. A sensitive telephone is in direct communication with the coil, or a galvanometer may be used, as the currents obtained by a ,slow elastic torsion are slow and strong enough to be seen on a very ordinary galvanometer. I prefer, however, the telephone, because it has the inestimable advantage in these experiments of giving the exact time of the commencement or end of an electric current. It has, however, the disadvantage of not indicating the force or direction of the current; but by means of the sonometer the true value and direc- tion of any current is at once given. Again, the telephone is useless for currents of slow intermittence; but, by joining to it the microphonic rheostat described in my paper of March 7th, a slowly intermittent or permanent current is broken up into rapidly intermittent currents, and then we are able to perceive feeble constant currents. For this reasona microphonic rheostat is joined to the telephone and coil. The current from a battery of two bichromate cells is sent constantly through the wire if we wish to observe the influence of the torsion of the wire upon the electric current, or a constant field of magnetic energy is given to the wire by either a separate coil or a permanent magnet. The currents obtained in the coil are induced from the change in the mole- cular magnetism of the wire, but we may equally obtain these currents in the wire itself without any coil by joining the telephone and rheotome direct to the wire; in the latter case, it is preferable to join the wire to the primary of a small induction coil, and the tele- phone and rheotome to the secondary, as then the rheotome does not interrupt the constant electric current passing through the wire. As the results are identical, I prefer to place the telephone on the coil first named, as the tones are louder and entirely free from errors of experimentation. If we place a copper wire in the axis of the coil we produce no effect by torsion, either when under the influence of a constant mag- netic field or a current passing through it, nor do we perceive any effects if we place an iron wire (2 millims. in diameter), entirely free 1881. ] Molecular Magnetism. 215 from magnetism and through which an electric current has never passed. I mention this negative experiment in order to prove that all the effects I shall mention are obtained only through the magnetism of the wire. If now I pass an electric current for an instant through this same wire, its molecules are instantly polarised. I have never yet been able to restore the wire to its original condition, and the magnetisation induced by the passage of a current is far more powerful and more persistent in soft iron than tempered steel. This may be due, however, to the fact that in tempered or softened steel we find traces only of a current during to the rotation by torsion of its molecules, some two to three degrees of sonometer, whilst iron gives constantly a current of 70 scnometric degrees.* In order to obtain these currents, we must give aslight torsion of 5° or 10° to and fro through the zero point. We then have a current during the motion of the index to the right, and a contrary current in moving the index to the left. Ifwe use a galvanometer, we must time these movements with the oscillations of the needle; but with the telephone it gives out continuous sounds for either movement, the interruptions being only those caused by the rheotome. The direction of the current has no influence on the result; either positive to the free arm or index or negative gives equal sounds, but at the moment of reversal of the current a peculiar loud click is heard, due to the rapid change or rotation of the polarisation of its molecules, and this peculiarly loud momentary click is heard equally as well in steel as in iron, proving that it is equally polarised by the current, but that its molecular rigidity prevents rotation by torsion. We can imitate in some degree the rigidity of steel by giving the iron wire several per- manent twists. The current due to elastic torsion is then reduced from 70° to 40°, in consequence of the mechanical strain of the twists remaining a constant; and a weakening of the current is also remarked if with a fresh wire we pass in torsion its limit of elasticity. If a new soft iron wire of 2 millims. (giving no traces of a current by torsion) has passed through it a momentary current of electricity, and the wire is then observed free from the current itself, it will be found to be almost as strongly polarised as when the current was con- stantly on, giving by torsion a constant of 50 sonometric degrees. If, instead of passing a current through this new wire, I magnetise it strongly by a permanent magnet or coil, the longitudinal magnetism gives also 70° of current for the first torsion, but weakens rapidly, so that in a few contrary torsions only traces of a current remain, and we find also its longitudinal magnetism almost entirely dissipated. Thus there is this remarkable difference, that, whilst it is almost impossible to free the wire from the influence produced by a current, the longitudinal magnetism yields at once to a few torsions. We may, however, trans- * 0°8 of a Daniell battery. VOL. XXXII. Q 216 Prof. D. E. Hughes. [May 19, form the ring or transversal magnetism into longitudinal magnetism by strongly magnetising the wire after a current has passed through it. This has the effect of rotating the whole of the molecules, and they are all now symmetrical with longitudinal magnetism; then, by a few torsions, the wire is almost as free as a new wire. I have found this method more efficacious than heating the wire red hot, or any other method yet tried. If I desire a constant current from longitudinal magnetism, I place at one of the extremities of the wire a large permanent magnet, whose sustaining power is 5 kilogrammes, and this keeps the wire constantly charged, resembling in some respects the effects of a constant current. The molecular magnetism or the current obtained by torsion is not so powerful from this, my strongest magnet, as that produced by the simple passage of a current, being only 50 sonometric degrees in place of 70° for that due to the passage of a current. The mere twisting of a longitudinal magnet, without regard to the rotation of its molecules has no effect, as is proved by giving torsion to a steel wire strongly magnetised, when only traces of a current will be seen, perhaps one or two degrees, and by using a constant source of magnetism or electricity, when no mea- surable effect will be obtained. Hvidently we have as much twisted the magnetised steel as the soft iron. In the steel we have a powerful magnet, in the soft iron a very feeble one; still the molecular rotation in iron produces powerful currents to the almost absolute zero of tempered steel.* If we magnetise the wire whilst the current is passing, and keep the wire constantly charged with both magnetism and electricity, the currents are at once diminished from 70° to 30°. We have here two distinct magnetic polarisations at right angles to each other, and no matter what pole of the magnet, or of the current, the effect is greatly diminished ; the rotation of the two polarities would now require a far greater arc than previously. The importance of this experiment can- not as yet be appreciated until we learn the great molecular change which has really occurred, and which we observe here by simply diminished effects. If we heat the wire with a spirit flame, we find the sounds increase rapidly from 70 to 90, being the maximum slightly below red heat. I have already remarked in my previous paper this increased molecular activity due to heat, and its effects will be more clearly demonstrated when we deal with the sounds produced by intermittent currents. Another method, by means of which I have again received proofs of the rotation of the polarised molecules, is to pass an intermittent current through a soft 0°5 millim. iron wire, listening to the results by the telephone joined direct and alone to the coil, as described in * T purposely avoid using the terms “ magnetic fluid”’ and “ coercitive force.” 1881. ] Molecular Magnetism. 217 my paper of March 7. If the wire is then entirely free from strain, we have silence, but a torsion of 20° produces some 50 sonometric degrees of electric force. If, now (the wire being at zero strain), I bring one pole of the permanent magnet I have already described near the side of the wire, the sounds increase from zero up to Ole being at their maximum when this magnet is 5 centims. distant; but if we continue to approach the magnet the sounds gradually weaken almost to zero. The explanation of this fact can be found when we know that the greatest inductive effect on the wire would be when a magnet is at an angle of 45° with the wire. And, also, considering each molecule as a separate independent magnet, we find that at a given distance for a given magnet the force of rotation is equal to that of 45°; by approaching the magnet we increase the rota- tion but diminish the angular polarity in relation to the wire, hence the decrease of force by the near approach of the magnet. And to prove that the function of the elastic torsion is simply to rotate the polarised molecules similarly to the magnet, we place the wire under an elastic torsion of 20°, and approach gradually the magnet as before. One pole now will be found to increase the sounds or \‘s angular polarity, the other will diminish them, until at 5 centims. dis- tance, as before, we have perfect silence ; the torsion exists as before, but the molecules are no longer at the same angle. On removing the magnet we find that instead of the usual 50 of current we obtain barely 5 or 10: have we then destroyed the polarity of the molecules, or do they find a certain resistance to their free rotation to their usual place ? To solve this question we have only to shake the wire, or give it a slight mechanical vibration, and then instantly the molecules rotate more freely, and we at once find our original current of 50=. wall forbear mentioning many other experimental proofs of my views by this method, as there are many to relate by different methods in the following sections. 2. Influence wpon the Molecular Structure of an Tron or Steel Were of Hlectricity or Magnetism. Being desirous to modify the apparatus already described, so that it should only give indications of a current if it were of a spiral nature, the wire was kept rigidly at its zero of strain or torsion, and the coil was made so that it could revolve on an axis perpendicular to the wire; by this means, if the wire was free from strain, the centre or axis of the coil would coincide with that of the wire. Thus, with a straight copper wire, we should have a complete zero, but ey this wire formed a right or left-handed helix, the coil would require moving through a given degree (on an arbitrary scale) corresponding to the diameter and closeness of the spirals in the helix; the degrees through which the coil moved, were calibrated in reference to known Q 2 218 Prof. D. E. Hughes. [May £9, copper helices. 50° equalled a copper wire 1 millim. diameter, formed into a helix of 1 centim. diameter, whose spiral turns were separated 1 centim. apart. In order to obtain a perfect zero, and wide readings, with small angular movement of the coil, it is necessary that the return wire should be of copper, 2 millims. diameter, offering comparatively little resistance, and that it should be perfectly parallel with the steel or iron wire. In order that it may react upon the exterior of the coil, it is fastened to the board, so that it 1s near (1 centim.) the exterior of the coil, and parallel to the iron wire, at a distance of 4 centims. If we consider this return wire alone, we find, as in the sonometer, that if the wire is perpendicular to the exterior wires of the coil, we have a zero or silence, but moved through any degree, we have a current proportionable to that degree; by this means, we have an independent constant acting on the coil, constantly aiding the coil in finding its true zero, and allowing of very wide readings, with a comparatively small angular movement of the coil. The rheotome is joined to a battery of two bichromate cells, and by means of a reversing switch, an intermittent current of either direc- tion can be sent through the wire. The telephone is joined direct and alone to the coil; thus no currents react upon the coil when perpen- dicular to the iron, and its return wire, if not of a spiral nature. Placing an iron wire 0°5 diameter, and passing a current through it, I found a change had taken place similar to those indicated in my paper of March 17th; but it was so difficult to keep the wire free from magnetism and slight molecular strains, that I preferred and used only in the following experiments tempered steel wire (knitting needles I found most useful). All the effects are greatly augmented by the use of iron wire, but its molecular elasticity is so great that we cannot preserve the same zero of reading for a few seconds together, whilst with steel, 0°5 millim. diameter, the effects remained constant until we removed the cause. I have not as yet been able to obtain a steel wire entirely free from magnetism, and as magnetism in steel has a remarkable power over the direction of the.spiral currents, I will first consider those in which I found only traces. On passing the intermittent current through these, the sounds were excessively feeble for either polarity of current, but, at each reversal, a single loud click could be heard, showing the instant reversal of the molecular polarity. The degree of coil indi- cating the twist or spirality of the current was 5° on each side of its true zero. The wire was now carefully magnetised, giving 10° on each side for different currents. The positive entering at north pole indica- ting 10° right-handed spiral, the negative entering the same pole, a left- handed spiral, we here see in another form, a fact well known and de- monstrated by De la Rive by a different method, that an electric current 1881. | Molecular Magnetism. 219 travels in spirals around a longitudinal magnet, and that the direction of this spiral is entirely due to which pole of an electric current enters the north or south pole. I propose soon, however, to show that under certain conditions these effects are entirely reversed. If through this magnetised wire I pass a constant current of two bichromate cells, and at the same time an intermittent one, the spiral is increased to 15°, but the direction of the intermittent current entirely depends on that of the constant current. Thus, if the positive of the constant current enters the north pole, the intermittent positive slightly increases the spiral to 17°, and the negative to 12°, both being right-handed; the two zeros of the constant battery are, how- ever, as we might expect from the preceding experiment, on opposite sides of and at equal distances from the true zero; but if we maenetise the wire whilst a constant current is passing through it, a very great-molecular disturbance takes place; loud sounds are heard in the telephone, and it requires for each current a movement of the coil of 40°, or a total for the two currents of 80°. This, however, is not the only change that has taken place, as we now find that both constant currents haves a right-handed spiral; the positive under which it was magnetised, a right-handed spiral of 95°; the negative, a right-handed spiral of 15°, and the true central or zero: point of the true currents indicates a permanent spiral of 55°. This wire was magnetised in the usual way, by drawing the north pole of my magnet from the centre to one extremity, the south from the centre to the other, and this was repeated until its maxi- mum effects were obtained. In this state I found, sliding the coil at different portions, that the spiral currents were equal, and in the same direction throughout. It now occurred to me to try the effect of using a single pole of the magnet; this was done whilst a constant current was passing through the wire, commencing at the extremity where the positive joined, drawing the north pole through the length of the wire, from positive towards the negative; the effect was most remarkable, as the steel wire now gave out as loud tones as a piece of iron, and the degree on the coil showed 200°. The constant and intermittent currents now showed for either polarity a remarkably strong right- handed twist; the positive 200 right, and the negative 150 right- handed spirals. The molecular strain on the wire from the reaction of the electric current upon the molecular magnetism was so great, that no perfect zero would be obtained at any point, afact already observed when a wire was under an intense strain, producing tertiary currents that superposed themselves upon the secondary. In order to compare these spiral currents with those obtained from a known helix, I found that taking a copper wire of similar diameter (0'5 millim.), and winding it closely upon the steel wire ten turns to each centimetre, 220 Prof. D. E. Hughes. [May 19, having a total of 200 turns, with an exterior diameter of 1:5 millims., withdrawing the steel wire, leaving this closely wound helix free, it gave some 190°, instead of the 200 of the steel wire alone; thus the spiral currents fully equalled a closely wound copper wire helix of 200 turns in a similar length. If it were possible to twist a magnetised wire several turns to the right, and that its line of magnetism should coincide with that of the © twist, then on passing a positive or negative current, there would be an apparent augmented or diminished spirality of the current, but both would have a right-handed twist. The result would be identical with the phenomenon described, although the cause is different. The explanation of this phenomenon can be probably found in the fact that the constant spirality now observed is that of the electric current under which it was magnetised, for whilst magnetising it we had a powerful source of magnetism constantly reacting upon the electric current, and the constant spirality now observed is the result or remains of a violent molecular reaction at the instant of magnetisa- tion, and the remaining evident path or spiral is that of the electric current. On testing this wire as to its longitudinal magnetic force, I found that it was less than that of a wire simply magnetised in the usual way; thus the effects are internal, affecting the passage of the electric current, giving, however, no external indications (except apparent weakness) of the enormous disturbance which has taken place. If, instead of drawing the north pole of the magnet as above, from positive towards negative, I draw it from negative to positive, all the effects are repeated, except that we have now, as we should expect, a left- handed spiral. But if I draw the magnet from the extremities of the wire to the centre, then at this centre I find an absolute zero of twist, but on each side a contrary twist, the wire then having a left and right-handed twist, the positive travelling towards the centre in a right-handed twist gradually ceasing in zero; this is as we might expect, but if done under the influence of a constant current, no matter what pole of the battery enters afterwards the north pole of the magnet, it will have during its first half a right-handed, and its second a left-handed spiral. It became important to know if a wire which had been magnetised under the influence of a current could be re- stored to something like its original condition. Hlectric currents had no effect. Heat, which would not destroy its temper, had no effect. Mechanical vibrations and torsions failed to disturb the molecular arrangement; but magnetising it strongly by a magnet, when no current was passing, at once broaght the wire to its usual apparently rigid state, and the constant or intermittent currents now indicated only 18° of spiral currents against a previous 200°, and the sounds were, as usual from steel, excessively weak. I have since used this method with invariable success, when I wished to repeat 1881.] Molecular Magnetism. 221 the experiments upon the same wire. If these experiments are repeated upon an iron wire, the effects are far greater in the first instance, so great that they were thrown out of the range of my measurements ; it was only after a few seconds of successive reversals that the zero of the wire was brought within range, and although these rapidly decreased, exactly similar effects were observed as in the steel. And as with all moderate ranges, I could bring the iron at once to a complete zero by torsion, and as torsion alone would produce this com- plete zero, I believe we have here effects from causes identical with those related in the first chapter. Having noticed in my previous papers the increased molecular activity caused by the approach of a powerful permanent magnet, and believing that the permanent spirality above mentioned was due to this alone, and not to an increased polarity, I magnetised strongly an iron wire giving as usual a reversed spiral for different currents of but 10°. I now heated the wire by a spirit flame to a dull red heat, whilst the current was passing through it, and on cooling I found a similar but stronger permanent torsion of 250°; both currents, as in the previous experiments, having a right-handed spiral. Thus a current of electricity passing through a wire nearly red hot determines a molecular arrangement, or path, which on cooling forces currents of either direction to follow the path which had been deter- mined under the influence of heat. 3. Molecular Sounds. The passage of an intermittent current though iron or other wire gives rise to sounds of a very peculiar and characteristic nature. Page, in 1837, first noticed these sounds on the magnetisation of wires in a coil. De la Rive published a chapter in his “ Treatise on Electricity ” (1853) on this subject, and he proved that not only were sounds produced by the magnetisation of an iron wire in an inducing coil, but that sounds were equally obtained by the passage direct of the current through the wire. Gassiot, 1844, and Du Moncel, 1878-81, have both maintained the molecular character of these sounds. Reis made use of them in his, the first electric telephone invented, and these sounds, since the apparition of Bell’s telephone, have been often brought forward as embodying a new form of telephone. These sounds, however, for a feeble source of electricity, are far too weak for any applied purposes, but they are most useful and interesting where we wish to observe the molecular action which takes place in a conducting wire. I have thus made use of these sounds as an inde- pendent method of research, and by their means verify any point left doubtful by other methods, some of which I have already de- scribed. © | The apparatus was the same as that described in the last section, 222 Prof. D. E. Hughes. 7 [May 19, except that no telephone was used. By means of a switch key the in- termittent electric current was either connected with the coil inducing longitudinal magnetism in the wire, or could be thrown instantly through the wire itself, thus rapid observations could be made of any difference of tone or force by these two methods ; a reversing key also allowed, when desired, a constant current of either polarity to pass through the wire under observation. Of all metals that I have yet tried, iron gave by far the loudest tones, though by means of the microphone I have been able to hear them in all metals; but iron requires no microphone to make its sounds audible, for I demonstrated at the reading of my paper, March 31st, that these sounds with two bichromate cells were clearly audible at a distance. similar but reverse action takes place as regards torsion ; for if the wire is magnetised by the coil we obtain an almost complete zero of sound by simply moving the torsion index 45° on either side, and as this was the degree which gave silence in the previous experiments for the same wire, it was no doubt due to the same rotation of its polarised molecules. If we now pass a constant current through the wire whilst the intermittent one is upon the coil, we hear augmented sounds, not in pitch but loudness; and if we give torsion of 45° to one side we have silence, or nearly so, whilst, to the other side, it gives increased tones which become silence by reversing the battery. If, whilst the wire by torsion has been brought to zero, we decrease or increase the mechanical longitudinal strain, then at once the polarised molecules are rotated, giving loud sounds; and we further remark that when the wire is loosened, and we again tighten it, we gradually approach a zero, and on increasing the strain the sounds return; thus we can rotate the molecules by a compound strain of torsion and longitudinal strain. If we wish to notice the influence of a constant current passing through the wire under the influence of the intermittent current in the coil, we find, if the wire is free from torsion, that, on passing the current, the tones are diminished or increased according to the direction of the current. The tones then have an entirely distinctive character, for whilst preserving the same musical pitch as_ before, the tones are peculiar, metallic, and clear, similar to those given out when a glass is struck, whilst the tones due to longitudinal magnetism are dull and wanting in metallic timbre. If we now turn the index of torsion upon one side, we have a zero of sound with or withont the current; but turning in the opposite direction gives increased tones whilst current is passing through the wire, but zero when not. Here again a peculiarity of timbre can be noticed, as although we have loud tones due onlv to the action of the current through the wire, the timbre is no longer metallic, but similar to that previously given out by the influence of the coil; evidently then the metallic ring could only be due to the angular polarisation of the molecules, and when these were rotated by torsion the tones were equally changed by its action upon the wire. I have already shown that a permanent magnet brought near the wire can rotate its polarisation, and it equally can produce sound or silence (while the wire is at its zero of torsion, and a constant current is sent through the wire as in the last experiment): we find that either pole of the natural magnet has equal effect in slightly diminish- ing the sound by an equal but opposite rotation from:the line of its maximum effects; but if the wire is brought nearly to zero by torsion, then on approaching one pole of the natural magnet we produce a complete silence, but the opposite pole at once rotates the molecules 224 Prof. D. E. Hughes. Molecular Magnetism. [May 19, in such a manner as to produce the maximum loudness, and on taking away the magnet we have comparatve silence as before. Heating the wire to nearly red heat by a spirit lamp increases the tones of longitudinal magnetism induced by the coil some 25 per cent., but it effects a much more marked increase in the tones produced by the direct passage of the current, which are increased by more than 100 per cent.; and if we pass the intermittent current through the coil and constant through the wire, we find no direct rotation of the molecules by heat. Although an apparent rotation takes place if, by the required torsion, we first place the wire at its zero, for then on the application of heat faint sounds are heard, which become again almost silent on cooling, this is simply due to the diminution by heat of the effect of the elastic torsion. Tempered steel gave exceedingly faint tones, requiring the use of the microphone; but on magnetising with a constant current, inducing spiral magnetism, the sounds became audible, some 15° sonometer against 175° for iron; thus the molecular rigidity of steel as observed by previous methods was fully verified. I have mentioned only a few of the numerous experiments I have made by the three methods described, all of which, however, bear directly upon the molecular arrangement of electric conducting bodies. I have selected a few bearing directly upon the subject I have chosen for this paper. I have, I believe, demonstrated by actual experiments which are easy to repeat, that— 1. An electric current polarises its conductor, and that its molecular magnetism can be reconverted into an electric current by simple torsion of its wire. 2. That it is by the rotation of its molecular polaniy alone that an electric current is generated by torsion. 3. That the aa of an electric current through an iron or steel wire is that of a spiral. 4, That the direction of this spiral depends on the polarity of the current, or that of its magnetism. 5. That a natural magnet can be produced, having its molecular arrangement of a spiral form, and consequently reversed electric currents would both have a similar spiral in passing through it. 6. That we can rotate the polarised molecules by torsion or a com- pound strain of longitudinal and transversal. 7. That the rotation or movements of the molecules give out clear audible sounds. 8. That these sounds can be increased or decreased to zero by means - that alone have produced rotation. 9. That by three independent methods the same effects are produced, and that they are not due to a simple change or weakening of polarity, 1881.] Identity of the Spectral Lines of Different Elements. 225 as when rotation has been incomplete a mere mechanical vibration has at once restored the maximum effect. 10. That heat, magnetism, constant electric currents, mechanical strains and vibrations, have all some effect on the result. III. “ On the Identity of Spectral Lines of Different Elements.” By G. D. Livetne, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., F.R.S., Jacksonian Professor, University of Cambridge. Received May 12, 1881. THE question of the identity of spectral lines exhibited by different elements is one of great interest, because it is very improbable that any single molecule should be capable of taking up all the immense variety of vibrations indicated by the complex spectrum of iron or that of titanium, and it might therefore be expected that such substances consist of heterogeneous molecules, and that some molecules of the same kind as occur in these metals should occur in more than one of the supposed elements. Further, the supposed identity of certain lines in the spectra of more than one element has been made by Mr. Lockyer the ground of an argument in support of a theory as to the dissociation of chemical elements into still simpler constituents, and in reference to this he wrote (“‘ Proc. Roy. Soc.” vol. 30, p. 31), ‘‘ the ‘basic’ lines recorded by Thalén will require special study with a view to determine whether their existence in different spectra can be explained or not on the supposition that they represent the vibrations of forms, which, at an early stage of the planet’s history, entered into combination with other forms, differing in proximate origin, to produce different ‘elements.’ ” Young, on examining with a spectroscope of high dispersion the 70 lines given in Angstrém’s map as common to two or more sub- stances, has found that 56 are double or treble, 7 more doubtful, and only 7 appear definitely single (“‘ American Journal of Science,” vol. xx, 119, p. 353), and he remarks, “The complete investigation of the matter requires that the bright line spectra of the metals in question should be confronted with each other and with the solar spectrum under enormous dispersive power, in order that we may determine which of the components of each double line belongs to one and which to the other element.” It is this confronting of the bright line spectra of some of the térrestrial elements which we have attempted, and of which we now give an account. For the dispersion we have used a reflecting grating similar to that used by Young, with 17,296 lines to the inch, and a ruled surface of about 34 square inches; telescope and 226 Profs. Liveing and Dewar. [May 19, collimator, each with an aperture of 13 inch and focal length 18 inches, the lenses being of quartz, cut perpendicularly to the axis and unachromatised, giving a very good definition with monochromatic light. The chromatic aberration is in this case an advantage, for when the telescope is in focus for lines in the spectrum of any given order, the overlapping parts of spectra of different orders are out of focus, and their brightness consequently more or less enfeebled. We have sometimes used green or blue glasses to enhance this result. The telescope and the collimator were generally fixed at about 45°, the collimator being more nearly normal to the grating than the telescope, and the grating moved to bring in successive parts of the spectra. For the parts of the spectra less refrangible than the Fraunhofer line K the spectrum of the third order was employed, for the more refrangi- ble rays that of the fourth order. The source of light was the electric are taken in a crucible of magnesia or hme; and for the examination of any supposed coincidence first one metal was introduced into the crucible and the line to be observed placed on the pointer of the eye- piece, the second metal was next introduced, and then in most cases, as detailed below, two lines were seen where only one was visible before, and the pointer indicated which of the two belonged to the metal first introduced. In some cases where both metals were already in the crucible, we had to reinforce the spectrum of one of the metals by the introduction of more of that metal, which generally brought out the spectrum of that metal more markedly than the other, and enabled us to distinguish the lines with a high degree of probability. Thus the crucibles of magnesia, or the carbons, always contain sufficient lithium to show the orange line and the calcium line hereto- fore supposed coincident with it Gare length 6101°9), but we observed these lines quite distinct and separated by a distance, estimated by the eye in comparison with the distance of neighbouring titanium lines, at about one division of Angstrém’s scale. On dropping a minute piece of lithium carbonate into the crucible the less refrangible line was seen to expand, and for a short time to be reversed, the other line remaining narrow and quite unaltered. When the lithium had evaporated, and both lines were again narrow, a small piece of Iceland spar was dropped into the crucible, which immediately caused the expansion, and on one occasion the reversal, of the more refrangible line, while now the less refrangible line was unaffected. In this way we satisfied ourselves that the calcium line is the more refrangible of the two, and is probably represented by the line at wave- length 6101-9 in Angstrém’s normal solar spectrum, while the lithium line appears to be unrepresented. In the case of iron, which gives such a multitude of lines, it was a@ priort probable that some lines would be coincident, or nearly so, with lines of other elements; and in fact we find that in five-sixths of 1881.] Ldentity of the Spectral Lines of Different Elements. 227 the supposed coincidences lines of iron are involved. We have, there- fore, chiefly directed our attention to iron lines. Taking iron and titanium, we find that we can resolve the lines at the wave-lengths following :— 6064-7* is a wide double, the iron line less refrangible than the titanium by something like a division on Angstrém’s scale. Young did not resolve this line, probably because he was looking for a much closer double line. 5714-09. A very close double, the iren the less refrangible, but not in Thalén’s list. 5661°65. A double titanium line, but there appears to be a single iron line between the titanium lines. 5489:05. p de sa grandeur naturelle. Du reste on verra tout-a-l’heure qu'il n’est besoin de mesurer ni a ni J. ‘‘ Supposons la tige supérieure de du bifilaire déviée de 90°. Alors, le poids P étant de 1,000 grms., la force directrice du bifilaire pour 2=90°, sera a peu pres 53,.1,000 grs.=5 grms. On équilibre cette force directrice horizontale par une torsion de 180° que lon donne au fil métallique /7 attaché en haut a la piece tournante R et en bas au point 7 milieu de ab. a tringle supérieure dc du bifilaire tient aussi a une autre piece rotative faisant partie de R et qui permet de tourner cd de 90°. Ces deux piéces de R tournent l’une dans l’autre, a Vordinaire. ‘La lunette inférieure, N, a des fils micrométriques qui vont se mirer en 7m sur un petit miroir vertical fixé au poids P. En ramenant Vimage de ces fils a la coincidence avec les fils eux-mémes, comme dans le pointé nadiral, on raméne la direction ab de la partie inférieure du bifilaire 4 sa position primitive. De méme, des miroirs attach4s aux parties tournantes en R, et réglés sur la lunette Q, assurent des rotations connues pour le haut du bifilaire de et pour le fil de torsion fz. Voici maintenant l’opération complete :— “T’appareil étant calé par ses niveaux, et bien stable, j’ameéne le bifilaire dans la position ot image des fils de la lunette N coincide exactement avec ses fils. A ce moment un petit miroir vertical fixé latéralement a la piece quien R donne le mouvement de rotation au bifilaire, ce petit miroir, dis-je, renvoie la partie inférieure des fils de la lunette Q sur eux-mémes ; et de plus un autre petit miroir pareil, fixé au-dessus du précédent a la piece qui donne la rotation au fil métallique fz, renvoie dans la partie supérieure du champ de la lunette ( Vimage de la moitié supérieure des fils de cette lunette sur ses fils eux-mémes. “Tout étant ainsi réglé :— ‘“‘ Je fais tourner de 90° la piece qui guide la tige supérieure de du On Gravimeters. 513 bifilaire, ce que l’on obtient au moyen d’un second petit miroir vertical fixé latéralement sur cette piece et 4 90° azimuthalement du premier. Ce miroir est amené 4 remplacer le premier miroir de la méme piéce pour le renvoi des rayons dans la lunette Q, et la rotation de de est alors de 90°. ‘a piece tournante a laquelle est fixée ’extrémité supérieure f du fil de torsion a aussi un second petit miroir vertical qui fait azimuthale- ment avec le premier miroir de cette piece un augle égal a trois fois 90°, et en tordant par son moyen le fil fz de trois fois 90° en sens con- traire du mouvement donné a de du bifilaire, la torsion de fz restera en définitive une torsion de 180° quand sa direction ab du bifilaire aura repris sa position primitive. ‘¢ Soit P le poids fixe du bifilaire (y compris a, b, &c.). On ramenera ab &sa position primitive au moyen d’un petit poids additionnel p. Alors la pesanteur locale, agissant sur une masse P+yp, par l’inter- médiaire du bifilaire équilibre la torsion de 180° du fil métallique fz. Si dans une autre latitude il faut P+p’ pour équilibrer la méme tor- sion, les intensités de la pesanteur dans les deux localités seront entre elles comme P+p est 4 P+ ’. * * * * * * * * ** Pour zod00 (qui est réclamé dans l'article des ‘ Outlines’) voyons Veffet produit par un poids de 0'l gr., ajouté a un poids de 1,000 grs. Ce poids de 1,000 grs. équivaut dans le bifilaire a 5 grs., et un poids de 0-1 gr. équivaudra a Saga Belvo = 00005 or. ‘“‘Mais 5 germs. sont équilibrés par une torsion de 90° [180° | du fil fv. Done 1 gr. correspondrait 4 une torsion de 36°, ou bien de 129,600”. Alors 0:001 répond a4 une torsion de 129”°6, et 355 gr. correspond a la moitié de ce nombre; mais comme le mouvement du miroir double le déplacement de limage, il restera un déplacement de 129’°6 (un peu plus de 2’) pour 5355 de variation dans la pesanteur. Si la lunette N a un peu plus de 30centims., le déplacement de l'image des fils sera de + millim. ** Nota.—ll faudra prendre la position de ab ramené a sa position primitive en faisant osciller ab pour éviter les adhérences de départ. “ Voyez, jugez, et surtout conseillez ; votre avis sera suivi. Je finis par ces mots des ‘ Outlines’: ‘The great advantages . . . render the attempt well worth making.’ “ J’ai Vhonneur, &c., oul BANG Bal WS Bh ‘““A Sir JouHn HERSCHEL. “P.S.—Il faut que je me hate de publier ce que j’ai trouvé, car a 69 ans ce serait vraiment étre trop exigeant envers la bonne providence que de lui demander encore indéfiniment du temps pour remplir la 514 Major J. Herschel. tache qu’il lui a plu de m’assigner. Vous avez été plus sage, pour le grand bien de la science. Moi, je puis malheureusement me dire, ““< Oras vives! hodié jam vivere, Posthume, serum est. Ille sapit quisquis, Posthume, vixit heri..—Manrrrat.” I have also seen a copy of the reply to this letter. It is needless to say that it was full of approval, and only made some suggestions, which were at once adopted or accepted. There was no mention of any other form of gravimeter having been proposed in England. I need not transcribe from the subsequent letters; until, on the 15th February, a year later, Babinet again writes :— ‘“‘M. Le Verrier me presse de terminer notre balance gravimétrique. Je ne yeux pas commencer imprudemment avant d’avoir votre ulti- matum d’approbation.”’ Here follows a réswmé of the principles involved in the design, as to which I need only remark that the elasticity of the suspending wires is not taken into account, nor is the necessary effect of this elasticity upon the balance in any way referred to. “T’appareil sera établi et essayé al’ Observatoire Impérial de Paris, puis essayé dans les latitudes voisines. Je serai heureux si cette mesure de la gravité et les déplacements nécessaires qu’elles entrainera me fournissent l’occasion de vous voir en Angleterre.” The letter then passes on to other matters—a new plan for measur- ing aberration by means of gratings—and only returns to the gravi- meter at the close. ‘‘ J’aurais encore attendua vous parler de ceci sans la circonstance de notre balance gravimétrique. En cas de succes de cette balance pourriez-vous, vous ou quelqu’un des votres, vous charger des stations du Royaume Uni? II faudrait au moins une mesure de gravité pour chaque degré carré de surface. “‘ Nota.—Sitdt apres votre réponse on se mettra a lceuvre pour Vappareil gravimétrique. ** Recevez, &c., ‘“ BA BINEI? The paper in the ‘Comptes Rendus”’ (lvi, pp. 244—248) was read on February 9, 1863, or a week earlier than the date of this last letter, which therefore appears to have resulted from the approval with which the paper was received at the reading. The published account adds little to what the letters telli—on the contrary—which is partly my reason for transcribing them. It is noticeable that there is from first to last no mention of any prior attempt in the same field. On Gravimeters. 515 I have endeavoured to ascertain, by enquiry from a gentleman who was associated with Babinet in physical researches of this nature, whether the instrument was ever constructed; but I do not gather from the reply that such was the case. It would seem as if there was too much uncertainty as to the constancy of the force relied upon in torsion. The original paper communicated by Broun to the Royal Society of Edinburgh follows immediately upon two others by the same author, in one of which the theory of the bifilar magnetometer is considered, and mention is there made of the elasticity of the suspending wires. Their effect is considered to be of little importance. The effect of temperature in modifying the elasticity is also alluded to. We cannot therefore suppose Mr. Broun to have disregarded these considerations altogether. Itis perhaps the more remarkable that he has left no indication of being alive to the really imteresting balance which occurs in his instrument, and which must equally have occurred, and can hardly have escaped his notice, in the former one. In his letter, above referred to, he speaks of his gravimeter as if the instrument ultimately perfected was substautially the same as the one described in 1861. But, as I have had occasion to point out, it differs in one respect which I cannot but consider important —the difference between a spiral balance spring and a twisted wire. Itis to be regretted that we are not in possession of any information as to the reasons which induced him to discard the one in favour of the other. In any case, the change is distinctly a part of the history of the invention. The next step was the construction of the instrument as mentioned in his letter; and then its exhibition and description by himself in the Catalogue of the Loan Collection of Scientific Instru- ments at the South Kensington Museum, where, he tells us sugges- tively, it was shown alongside of Mr. Siemens’ bathometer. We may now turn our attention to this last. My knowledge of the bathometer designed by Dr. C. W. Siemens is confined to the account given of it in the “ Philosophical Trans- actions’ for 1876, and I should not presume to examine it were it not necessary to do so for the purpose of this review. Dr. Siemens, it is true, regards it, in the introductory passages of that paper, as having a different sphere of action; but, on a careful consideration of the principles—so far as I understand them—of its construction, it appears to me to belong decidedly to the class of instrament, whatever we may call them, typified by the one already mentioned as described in the “Outlines of Astronomy.” The special purposes for which it has been designated, and the actual uses to which it has been put, do not of themselves preclude it from being set to the more genera] pur- poses and uses of a gravimeter. Reference is made in the cited paper to another form of instrument 516 Major J. Herschel. constructed by the author in (or about) the year 1859, which, how- ever, was abandoned, although it enabled him ‘‘to predict approxi- mately the depth that would be found on the use of the sounding line.” In this instrument the specific gravity of mercury would seem to have been measured by determining the column necessary to balance the pressure of a fixed quantity of confined air. The rise of the surface under a change of gravity was magnified by a lighter supernatant liquid being forced up a narrow tube. The principle, in this case, is analogous to the former typical one only if we regard the compressed air as acting the part of a spring. It is obviously dependent on the possibility of guarding against, or exactly allowing for, the change of elasticity with temperature—in this, too, being analogous to the typical form. It was sought to meet this very serious difficulty by keeping the air at the temperature of melting ice. There is another method—perhaps not less dificult to practise successfully; that, namely, of selecting three (or preferably four) more accessible temperatures, and so experimenting at these that in the long run the actual observations shall supply a constantly accu- mulating body of evidence as to the requisite factor, as well as two or more actual data in each case from which to infer what is required. But it might turn out after all that the uncertainty must remain greater than the variation to be measured. The possibility of such a result militates strongly against methods involving large corrections. But, perhaps, least of all in cases such as this where the whole instrument may be immersed without injury. It then becomes chiefly a thermometric question. In the bathometer, on the contrary, which superseded the one just mentioned, ‘‘changes of temperature are entirely eliminated from the result,” at least in theory. In practice such a thing would perhaps be hardly possible ; but in any case residual effects may be regarded as subjects for ultimate consideration. I will now endeavour to con- dense a description of this instrument, so as to assign it a place in the present category. There are two spiral steel springs which, in the drawing of the smaller of two instruments, are about 16 inches long and rather more than one in diameter. The lower ends are attached to a cross-bar which is to bear the intended weight. This weight is that of a column of mercury resting upon a certain surface to be presently described. There is an arrangement for maintaining the height of the column constant. JI am not sure that I understand how this works; but supposing it effective, the weight upon the cross-bar must depend upon the area pressed by the mercury; and if this also is constant we have in effect the same principle as in the gravimeter depicted in the ‘‘ Outlines.” The mercury is contained in a vertical steel tube having cup-shaped ~ On Grarvimeters. 517 enlargements above and below. The lower end is closed by a flexible corrugated steel plate, which has a solid central disk, by which the pressure on the plate is transferred to the cross-bar. When this plate is horizontal, the mercury presses equally on every part; but it is clear that the whole of the pressure is not borne by the cross-bar. In fact, it would only by trial be possible to ascertain what weight is borne by the springs. Any change in the position of the disk must alter the curvature of the plate, and with it the proportion of the weight borne by the springs and by the rim of the cup—irrespective of any change in the height of the column. The diameter of the steel tube would seem to be adjusted so as to cause a proper change in the level of the upper surface, when the bulk of the mercury is affected by change of temperature. I am not quite sure that, in presence of these several causes of variation, the principle of con- stancy of weight on the springs can be said to be retained. Alterations of position of the disk are read by means of a micro- meter below the cross-bar. It seems clear that the desired indications must be obtained from the readings of this contact micrometer, and some curiosity may be experienced as to-the manner of interpreting its readings. On this point the author says:—‘“ It would be difficult to determine the actual scale of the instrument a4 priori; and I there- fore adopted the easier and safer method of relying for its final adjustment upon the result of actual working.” If it were possible to ascertain, with some approach to certainty, the weight borne by the springs in some particular position of the diaphragm and cross-bar (to be called the zero position) the instru- ment might be used, on the principle of the simple gravimeter, by ascertaining what weight upon the cross-bar would be necessary, under a diminished gravity, to bring it again to this zero position—tempera- ture being the same. This, however, is not what is intended in this form of instrument. The subject of compensation for temperature receives considerable - attention. ‘This, indeed, is very necessary, when we remember the sensitiveness of springs to heat. The author gives, as the result of experiment upon these springs, a factor of variation which appears to be zon for each degree Centigrade. The importance of temperature in such case will be better understood if we consider that at this rate it would need but 21° C. to effect as great a change in the elastic force, as transference from equator to pole would effect in gravity. Change of temperature has here, it would seem, an effect twelve times as great as it has in the case of the pendulum. Certainly there is room for compensation; and Dr. Siemens obtains it, to some extent, by apportioning the diameter of the vertical tube to the cups. It is not necessary, for the present purpose, to consider closely how this exercises a compensating effect. It is estimated, by experiment, that 518 Major J. Herschel. the factor is reduced to gogg555- It seems highly probable, how- ever, that the residual effect would no longer be a simple function of the change of temperature, and might be quite incalculable. It is also necessary to consider that the masses of metal concerned would probably, except under controlling conditions, cause consider- able uncertainty in this effect. Enough has been said to show that the bathometer, although a gravimeter in principle, cannot be regarded as likely to prove of ser- vice in measuring small changes of gravity under different climates. Its eticacy as an instrument for measuring sea-depths need not be considered here. It may have peculiar properties in that connexion which I have not understood; for indeed I may confess a certain hesitation in believing that any instrument can do that, by deter- mining the change of attraction, except under conditions favouring an empirically deduced scale. There are rumours of other designs for measuring small changes of gravity, of quite recent invention, of which I can only say at present that it is very much to be hoped that they wil] serve their intended purpose, and that we shall soon be in possession of experimental evidence to that effect. II. On the Torsion Gravimeter, constructed on the Design of the late J. Allan Broun, by Dr. C. 8. Miiller, of Stuttgart. Before entering upon any description of this instrument it may be well to explain that the title of this paper is intended to take cogni- zance of the fact that the designer did not experiment with it or improve upon it after it left the maker’s hands, and that consequently some portion of the merit, whatever it may prove to be, is due to the appreciation of the design by the latter. Conversely, it is due to the designer to recognise that failure of any kind may to some extent be attributable to misapprehension on the part of the maker. At the same time, it must be said that the workmanship is of a high order, and betokens a more than ordinary attention as well as great skill and delicacy. Should the instrument justify its existence by its ultimate utility, it cannot be denied that it will owe its success, in a high degree, to the intelligence of the constructor as well as to the genius of the inventor. There are two ways of describing a new form of instrument. We may either approach it from outside, and learn its functions by con- sidering its parts, or we may study it with a prior knowledge of its intention. The latter method is, perhaps, the best for one to adopt who wishes to describe an instrument of his own designing; for he cannot fail to indicate truly what design he had. But for one whose only knowledge of the functions of an instrument are inferred from what he sees, it might be dangerous to presume a full knowledge of On Gravimeters. 519 the design. I shall, therefore, describe the instrument as I see it, with direct reference to what I conceive to be the intention.* The gravimeter stands, when set up for use, on a tripod; but it might equally be stood upon a table with a bay cut out of the side two or three inches deep. In either case it would stand on three ad- justing foot-screws carrying a thick brass fowndation plate. Below this plate projects what I will call the well and a cathetometer. Above it stands the chamber, consisting of two brass plate sides and glass plate front and back, besides various clamping and manipulating appliances. On the top of the chamber are two cross levels, and from the centre rises a hollow shaft. Measuring from the surface of the foundation plate, the depth of the well is about 84 inches (26 centims.), and the height of the chamber 53 inches (16 centims.). The shaft stands about 184 inches (47 centims.) above the plate, or 13 inches above the roof of the chamber. The latter is about 24 inches wide and 22 inches deep (say 6 centims. each way). This general description will suffice to give an idea of the framework. More minute descriptions will follow as they become necessary. : Within the chamber are suspended by wires, one below the other, two weights, which for distinction I will call the major weight and the minor weight. The major weight consists of a brass cubical block, from which rise two lateral rectangular pillars, crossed at top by a somewhat slighter bur. The block is perforated vertically by a large cylindrical hollow. This major weight is suspended by two parallel wires which traverse the shaft; being just visible at its lower end, where they enter two small screws on the cross-bar. There are means of adjusting both their length and distance apart, above and below. The principle here is that of the bifilar torsion balance: if the block be turned through any angle less than 180°, its tendency to return is (mainly) due to the weight of the mass suspended. This, however, is not that of the major weight only; for from the under side of the cross-bar hangs a single wire, which, descending through the hollow in the block, is attached to and supports the minor weight; hence the mass suspended by the double wire is the sum of the masses of the two weights. The minor weight consists of a light frame, which I will call the head, and a long thin glass rod or plunger, which descends into the well. Suppose the major weight turned through an angle 6. When the disturbance has subsided and oscillations have been quelled, it will, of course, be found that the minor weight has turned through a like * Since this report was written, two plates have been drawn to illustrate the description which follows. Necessarily there is no direct reference to them in the text. 520 Major J. Herschel. angle. The double wires are in a state of torsion, but the single wire hangs freely, helping by its tension to increase the tension on the double wires caused by the major weight, but not itself in a state of torsion. Now, suppose the minor weight turned round : it will imme- diately begin to exert through the resistance to torsion, or elasticity, of the single wire, a force which will tend to relieve or oppose, accord- ing to the direction of the new application, the external force which keeps the major weight detorted. Suppose the new force applied in the same direction, so as to relieve the former. As the torsion of the single thread increases—with the increase of the angle through which the lower weight is turned—a point is at length reached when it exactly relieves the whole of the external force applied to the upper. Suppose this to occur when it has turned through an angle @, #.e., through an angle 6+4¢, from the initial position. Then it is clear that the force which turned the upper end of the single wire through an angle 0, has been found equal to that which turns its lower end through an angle ¢, relatively, and 0+4¢, absolutely. .We have now to consider what these forces are. Before doing so, it would be advisable to recognise the means pro- vided for observing these angles. Jn some way, which there are no means of exactly discovering, the designer or constructor has ascertained that by a certain apportion- ment of lengths, weights, and thicknesses, the proportion of 0 to (which of course is a variable one) can be made 1:3 when 0=90°. The result of this is, that when 6+¢@ amounts to one complete revolu- tion, @ alone is one quarter of a revolution. [In this position the re- sistance of a bifilar suspension is a maximum. I do not know that it has any strong advantage except what may turn on that. It is only necessary to allude to the fact, to take occasion to add that it is cer- tainly of no importance, either practically or theoretically, whether 0 is exactly or only approximately, equal to 90°. We shall see even- tually that another consideration (perhaps not considered by the in- ventor) entirely overrides the one mentioned. This by way of parenthesis. | The head of the minor weight carries a small flat mirror, which faces in the position of rest, as also after one revolution, a horizontal colli- mator. The block of the major weight carries three similar mirrors, of which one is parallel to the former (or may be made so) when in a state of ease, and the others become so (under proper conditions) when the major weight has been turned to the right or left through one quarter of a revoiution. These mirrors face the upper half of the - object-glass of the collimator. A fiducial mark in the focus of the latter is seen by reflection from the mirrors when they are perpen- dicular, or nearly so, to the line of sight. It will be necessary to return to this in describing the intended observation. On Gravimeters. 521 To understand thoroughly the conditions of equilibrium we must now study in detail the counteracting forces. We have seen that the weight supported by the double wire is the sum of the two weights described as the major and the minor weights. One-half of this sum is, of course, borne by eaeh of the wires, which are further twisted in- dividually through a quarter of a circle. Let R be the length of these wires, and 2r their distance apart. Consider one only: the upper end being fixed, the lower will describe a curve which will be the inter- section of a sphere by a vertical cylinder, the ordinate from which to a plane drawn horizontally through the lowest point may be shown to be equal to | r : re ‘ z: rversin 0} 1437... versin 0 — de. \ As r: Ris a small fraction, less than ;4,, it will be a question whether the second and further terms may not be neglected ; for the present we may be content with the first, as an approximation only is wanted here. This ordinate is the height through which the weights rise as they are turned round. Let this be ealled h. A little consideration Ligeti =". a constant, shows that in this case the r versin 0 restriction to the first term makes the path of the lower end of the wire a circle lying in an inclined plane,* and the force tending to cause rotation (irrespective of the torsion of the parallel wires) is that of a body having the joint weight of the two, whose path of descent is to bealong this circle. The gradient along this path (which is the dh (Re 180°, and a maximum at 90°. Let A+B=P be the weights, which are augmented for adjustment by a small weight, p. Then the force, depending on gravity, which tends to turn the system, resolved hori- of the meaning of tangent of the inclination) is =" sin 6, which is zero at 0° and 0 R zontally, is (P+) tan Z inclination =(P+ P)-e sin 0. Neglecting for the present the torsion of the parallel wires, we see that this force is resisted by the torsion of the single wire, as to which we must have regard to two propositions :— (1.) Torsion is independent of tension ; (2.) Torsion varies directly as the angle of twist. We may, therefore, express the torsion by p,¢ (at the distance unity), remarking that @ is mechanically increased until it balances the force tending to turn the system as above described. Now, since (P+p) = sine attains a maximum when 0=90°, if at * 'This demands an elliptical cylinder. VOL. XXXII. Ze 522 Major J. Herschel. this point it were balanced by p,#, any increase of torsion caused by increase of ¢, however small, would be answered by no corresponding augmentation of the opposing force. The upper system would obey, @ would increase, but the resistance would diminish instead of increas- ing. In short, 90° would represent a position of unstable equilibrium. This state of things is prevented by the torsion of the parallel wires, which act in aid of the weight and defer the condition of unstable equilibrium to a point some degrees beyond 90°. The perception of this may or may not have been present to the designer; but it may obviously be made use of advantageously as follows :—Let 2,0 denote the torsion of the two parallel wires when turned through an angle 0. Then the forces which balance are (P+ i .sin 0 + 2p,0, and p,p; and the utility of the instrument, depending in the first place on the abso- lute constancy of p,%, depends also on the possibility of exactly equali- sing to this the other force by varying either 9 or P+p. It is in this respect exactly analogous to a balance, the desired relation of 6 to d being equivalent to the demand for horizontality in the latter. The sensibility of such an equipoise will be measured by the smallness of the change in P+» requisite to produce, or to correspond to, a given small change of @. Let (P+ ) z . sin 6+ 2p,0=F, then ae . sin 0, =P +p) = . COS 9+ pz, R r and op=— cosec . 0 ((P TP) . COS 0+2p2) 50. i When 0=0°, dp is infinite; 7.e, no change of weight will affect @. Similarly when 6=180°. But when 06=90°, Sp pene! At this r point no relation would subsist between ép and 60 but for pz; 1.e., if the equilibrium were independent of the torsion of the parallel wires. As it is, the equilibrium is a compound one, and is disturbed by a change of weight, the effect of which is inversely proportional to the torsion of the parallel wires, in apparent contradiction to the first of the above pro- positions regarding torsion. If the torsion of these wires be doubled, the addition of the same small weight 6p will cause a change of @ to one-half the amount. The sensibility of the instrument is therefore proportional (within limits) to the fineness of these wires; but this must not be carried so far as to bring the position of unstable equili- brium too near to 90°, or else the stability of the balance will become too delicate and observation too difficult. On Gravimeters. 5s To determine the position of unstable equilibrium, consider the general statical equation (P+ p) = . sin 0+ 2p.0=p). When the left-hand side is a maximum its differential with respect to @ equals zero, therefore (P+p) = cos 0) +2p,=0, whence cos y= ae wo. “ 1g ie 2 Now, as in the intended position of rest, 0=_, = we have (BaP) Cri 42) 5 = 20a 5 PED Combining this with the former, it appears that sec 0) = = : A_1) a 2 py 2 The double wires and the single wire are no doubt of the same material, but the former are thinner. Their lengths are as 4:1. Suppose their thicknesses as 2:3. Then p,: pg:: 4 3*: 2?::9:1, which gives secO,—=— 1275 5 1964, and .*. 0)=92° 55’. That is to say, the position of unstable equilibrium lies less than 3° beyond the intended position of rest. [This is closer than I should have thought, but it is borne out by the fact that I have not been able (with the existing adjustments) to reach the intended position of rest both ways, t.e., turning both to the right hand and to the left hand, before reaching the point of unstable equilibrium. This is easily ac- counted for by supposing the primary position to be a balanced one, a slight deviation from parallelism of the mirrors being induced by a slight initial torsion of the parallel wires.] To return :— The angle through which the lower end of the single torsion wire is turned, by the stop acting on the minor weight, is@+@. As this angle approaches 360°, 6, which is the angle through which the upper end and the major weight are turned, approaches, or should approach, 90°. As this stage is approached a very small increase of 6+¢ in- duces a larger and larger increase of 9, according to the proximity to the value 0. And the above expression for sec 9) shows that this depends on the ratio of p,: pz. By varying the strength of the double 7 Te 524 Major J. Herschel. wires the position of unstable equilibrium may be placed anywhere from 6)=90° to 6;=180°; and the nearer it is to 90°, 7.e., the weaker they are, the more sensitive will the instrument be at the position 6=90°. This suggests, I think, that the double wires should be coarse and inelastic rather than the reverse. The ratio r: R is as nearly as I can judge by measurement 1 : 138, and (P+) would be, at the equator, 3,725 grs.; so that (Pt?) = 2/ grs. very nearly. Now, 2 ip 1 a-r=a(P+p)— .— 1 _ ; Be eee ee Pl 2 py =18(1+ ae ee This is the force, exerted at a distance r=‘09 inch, which is requisite to turn the single wire, of length 3 inches, through the angle z. If P2 be taken as above, equal to py7=192 grs., at 09 inch; and val pow 22 grs. nearly. We can now form some estimate of the effect of a small change of e R ° weight. We haveseen that ép=2p,. —.60. Substituting numerical r values, as above, we find ép=138 x 42 x S22 oe nearly. de T The collimator contains a scale, exactly 1th of an inch long, divided into 30 parts. The focal distance being 8°5 inches, the angular value of the scale is tan 12, or 1° 41'; and of each part, 3° 22", or sgo5 7 approximately. Putting one of these parts equal to 260 (doubling on account of the reflection) we find ép=-323,='092 gr., the change of weight which will deflect the mirrors relatively one division. It follows that the 30 divisions of the scale correspond to a change of weight of about 2°76 ers. But we must remember that this relation depends on an assumption, viz., that p::py:: 9:1. To provide for a more accurate estimate being 2 18 27k ee hereafter obtainable, let far bs Chen Pi epee Pot = ana ép= 7452 ~. as and in the case of the scale-division pea és ; gTs., —flt 7 — or very nearly ao hi 7 for the whole scale. Let us now cousider what may be expected as the actual conse- quence of a change of gravity. The most practical evidence of this is the effect on the rate of a pendulum. A pendulum which would beat seconds at the equator would gain 225 seconds, or beats rather, at the On Gravimeters. 525 poles. The corresponding increase of gravity is in double this pro- portion, t.e., 450 on 86,400, or 1 on 192, nearly. Weights, as mea- sured by a constant force such as the torsion of a wire is supposed to be, are affected in like proportion but inversely. Now the weight of the mass suspended by the double wires is 3,725 grs.; and the change of weight, if one may so express it, from pole to equator, will be 3223, or 19°4 gers. We have seen that the range of the scale in the collimator will measure a deflection due to a change of weight of 2°76 grs., or thereabouts. It follows that the extreme change of gravity which can be looked for would cause a deflection about seven times as great as what the scale will measure. This is provided for by the auxiliary weights, of which there are five, weighing 4°4 grs. each, and by the contrivance of the glycerine well, which will now be described. Before quitting this part of the subject, however, I should point out that one division of the scale (992 seems to correspond to a change of pendulum rate of Ou x 225=1:07 second per diem. The buoyancy of the glycerine in the well depends on the volume of liquid displaced by the plunger, and therefore on the diameter of the latter, which can only be got at awkwardly, in one place. To ascer- tain this diameter I cut a slit in a piece of zinc plate, rather wider at the mouth than at the inner end, and used it as a gauge, marking the place where the narrowing width fitted the plunger. A piece of copper wire, ‘05 in diameter, was found to fit it at the same spot; perhaps ‘048 would be more correct. The specific gravity of glycerine is 1°26; hence a cubic inch weighs 349°4 ors. With these data we find that one inch rise of glycerine will buoy 0°628 gr. The cathetometer scale being divided metrically, this corresponds to 0°0247 gr. per millimetre. The micrometer reads to hundredths of a millimetre. The riders weigh, as nearly as I can determine, 4°40 gers. each. Hence one rider has the same effect as 178 millims. of rise; but the scale only runs up to 150 millims., so that additional riders will be requisite. I found by trial that 100 millims. rise of the glycerine buoyed about 2grs. It was not a careful observation, but agrees fairly with the above calculation; so that it seems pretty certain that the riders are too heavy. The total weight being 3,725 grs., the variation of gravity from pole to equator (being as 193: 192 nearly) will require an augmentation of 19°4 grs. This corresponds to 225 seconds, therefore 1 second corre- sponds to 0:0867 gr., or 35 millims. of glycerine, if the above is correct. 526 Major J. Herschel. We must look for accuracy which may be represented by +0:1 second, or the instrument is not worth testing severely. This corre- sponds to 0°35 millim. of glycerine, or 35 divisions of the micrometer. The foregoing description and investigation represent what I had learnt about the gravimeter by inspection, measurement, calculation, and reflection, before meeting with any instruction from the designer. In his letter to General Walker, Mr. Broun mentions having seen and corrected the proof of the description which was to appear in a new edition of the Catalogue of Instruments exhibited in the Special Loan Collection. Not expecting that a catalogue description would be otherwise than brief, and confined to principal features such as I could not fail to perceive unaided, I did not wait until I could procure it, to study the instrument. Some delay also occurred before I received, through the courtesy of the Secretary of the Science and Art Department, the extract in question. It seems right to mention this, in explanation of the independence which will be remarked in what I have said. Whereas, in what follows, I acknowledge the said description as an additional source of information. Mr. Broun’s description is accordingly inserted here; and I shall then add some comments upon points which seem to need further elucidation or notice. “421d. Gravimeter. An instrument for the measurement of the variations of the earth’s attractive force, invented by J. A. Broun, F.R.S., and constructed from his drawings by Dr. C. Miller, of Stuttgart. J. Allan Broun, F.R.S. “The instrument consists of a weight suspended by two gold wires; a single wire fixed to the top of the weight and passing through its centre carries a cylindrical lever; when the lever is turned through 360° at the normal (say southern) station, the torsion of the single wire thus produced carries the weight round through an angle of 90°. The forces then in equilibrium are, the torsion force of the single wire and the attraction of the earth on the weight, which, as the two wires are no longer vertical, has been slightly raised and seeks to attain its lowest point. ‘On proceeding from a southern to a more northerly station the earth’s attraction increases; the amount of this increase may be measured in two ways :— ‘1st. The lever will require to be turned through more than 360° in order to carry the weight to the height due to turning © it through 90°. (Had the station been more southerly the lever would be turned through less than 360°.) The difference of the angle from 360° measures the increase (or diminution) of weight. On Gravimeters. 527 “2nd. By removing a small portion of the weight, equal to that due to the increased attraction of the earth, the weight can be turned through exactly 90° by rotating the lever through 360°, as at the normal station. (On proceeding south weight has to be added.) “The following are the instrumental arrangements in order to make these observations :— “The weight has on each of three sides, at its base, a vertical mirror (silvered, not quicksilvered) ; the middle mirror makes an angle of exactly 90° with the other two. The lever also carries a vertical mirror, which, when there is no torsion in the suspension wire is immediately below and in the same vertical plane with the middle mirror of the weight. A telescope, having a glass scale at the focus of the eye-piece, is adjusted so that images of the scale can be seen (one higher than the other) reflected from the middle mirror of the weight and the lever mirror. When both of these mirrors are exactly in the same plane, the middle division on the scale seen directly with the eye-piece, coincides with the same division in the two reflected images. ‘““By a wheel and pinion (with endless screw and clamp for delicate movement) placed below the instrument, a polished agate point can be made to act on a similar agate point fixed to the lever, so as to turn the latter through any angle. When turned through 360° the middle scale division again agrees with the image from the lever mirror. If the image reflected from one of the side mirrors of the weight does not agree also, the lever is turned through a greater (or lesser) angle than 360°, till this agreement is obtained ; the difference of the angle through which the lever has been turned from 360° is obtained from the scale reading, as seen on the lever mirror. “The following apparatus is employed for very small increases or diminutions of the weight. Suspended to and vertically below the lever is a carefully calibrated glass wire (1 millim. diameter), which enters a glass tube fixed below the instrument. At the lower end of this tube is a cistern containing a liquid (distilled water, or as at present, chemically pure glycerine). This liquid can be forced into the glass tube by a screw and piston (as in some barometer cisterns). The liquid is then raised till such a diminution of weight is produced by the immersion of the glass wire as to bring the mirror of the weight through exactly 90°, when the lever is turned through 360°. The length of glass wire immersed is read, by a micrometer microscope and scale, to a thousandth of a millimeter. ‘Though finely polished agate points have been employed for 528 Major J. Herschel. turning the lever so as to diminish the friction, there is an additional apparatus to ensure that vertical friction has no effect on the observation at last. The lever contains a magnet; and two bar magnets, with rack-work adjustments for height, are placed one on each side of the instrument, so that by a pinion and rack movement they can be approached to the lever magnet till their force is exactly equal to the torsion force of the single wire, and the agate points are no longer in contact. “The instrument is made to serve for latitudes differing about 10° or 15°, but an auxiliary apparatus carries five platinum rings, which can be lowered upon the weight, so as to make the in- strument serve from the equator to the poles, and to any height in the atmosphere. “There are special appliances for portability, by one of which the weight is fixed; another fixes the lever; so that strain is removed from the suspension wires, and the suspended parts can- not be shaken from their places. Levels, a thermometer, and other details fit the instrument for the most accurate observations. The suspension wires are fixed at their cnds in a special manner, so that the fixed points cannot vary. All the suspended apparatus is electro-gilt.” (1.) We learn from this that the suspending wires are of gold, and that suspended parts are electro-gilt. I observe that there is a tendency to spottiness, resembling mould, on some parts of the gilded surface. (2.) The part which I have called the ‘minor weight” is here designated as a ‘‘ cylindrical lever.”” The latter term is but remotely descriptive. ; (3.) The torsion of the double wires zs not alluded to in describing the forces which balance each other. This confirms my doubt whether the very important part played by this torsion was recog- mised. It igs an essential feature, without which the position of maximum gravity-action could not be chosen; and in making the adjustments it is impossible to disregard it. (4.) here is no necessity for the two side mirrors on the major weight to be inclined at ‘‘ exactly 90° ” to the middle one. Indeed, it is scarcely possible to tell exactly what their inclination is. It is about 90° ; and that is all that can be said or desired. (5.) Of the two terms ‘“telescope’’ and ‘collimator,’ the latter describes more correctly the function of the appliance by which the angular positions of the mirrors are observed. The scale is in the focus of the object-glass of this collimator, rather than in that of its eye-piece, though, of course, the latter is also true. The divisions and value of this scale are not mentioned. If my estimate is right, which ~ On Gravimeters. 529 makes one division correspond to 1°07 per diem, it is clear that this is a weak point, especially as the collimator is optically indifferent. (6.) This ‘‘ telescope” or collimator is said to be ‘‘adjusted ”’ so as to perform certain functions. Unfortunately it has no means of adjustment except in a vertical plane. There is no horizontal motion; nor has the object-glass a power of focal adjustment by rack and pinion. It is very difficult to get a good sight of the reflected scale. (7.) The only dimension stated in the whole description is the diameter of the “plunger” (1 millim.). Neither are the weights of the suspended masses stated. A precise knowledge of the joint weight is necessary for the calculation, even of differential results. By precise I mean to within 4 or 5 grs. (8.) The plunger is described as a “calibrated glass wire.” I imagine this to mean that the glass red was specially made, and tested at every point. I do not see any advantage in its being of glass. (9.) “The length of glass wire immersed ”’—that is to say, the length of liquid displaceed—“is read . . . . toa thousandth of a millimeter.” The graduation of the micrometer head enables hundredths of a millimeter to be read. The thousandths are by estimation. But this probably far exceeds the power of observation of the surface of a liquid, such as glycerine, in a glass tube. (10.) “‘The finely polished agate points” appear to be of ruby- coloured glass—if ready fusion be any test. (11.) The magnetic holder. I imagine this is better removed. It would be nearly impossible to guarantee its action being entirely horizontal. And the presence of a magnet as part of a mass, the weight of which is under examination, is inadmissible. Should these comments give the impression that I wish to cavil at the description, I must reply that it is necessary for the present pur- pose. The question before us is not whether the instrument is ingenious ; but whether it can be used for the intended purpose, in preference to other existing instruments whese use and powers are well known. It may be said that that question can best be answered by trying i. Unfortunately this is not the case. It will already have become apparent that it is one of a class of instruments in which the observa- tion is nothing but the last of a series of elaborate and difficult adjustments—adjustments which require patience and skill and no little time, all of which would be thrown away if the ultimate obser- vation should prove abortive. I will now proceed to indicate more exactly what these adjustments seem to be. The principle on which the instrument is designed involves, as a primary consideration, the angular rotation and ultimate angular position of two bodies. The angles 9 and @ of the theory above 530 Major J. Herschel. explained, although unimportant in absolute magnitude, are all- important relatively. The prime defect of the instrument is to be seen in the insufficient optical means of noting and recording this relation. This might be remedied without altering the instrument in any way, by changing the collimator; so I will not dwell further on that, but pass on to the means of controlling and adjusting this angular position. Inasmuch as torsion of wires is in question, it is obvious that the way in which their ends are held and turned is a detail of extreme importance. Mr. Broun has alluded to this. He says they ‘‘are fixed at their ends in a special manner, so that the fixed points cannot vary.” It is unfortunate that one cannot learn cer- tainly by inspection, what the attachment is. It is probably by the pinch of a split screw, for the wires appear to pass through the axes of the holding screws. Whatever the method is, it ought to be un- impeachable. But net only should the holding be secure, it should also be easily manageable. It is of little avail to attach a fine wire to a delicate screw which can only be manipulated with caution by a steady hand, for the adjustments depend mainly on these ends of wires being turned accurately through very small angles. I regard it as a capital error of construction that the grasp. of the ends of these suspension wires is made as small instead of as large as possible. This is an opinion based on wearisome experience no less than on common sense, for I have spent many hours in endeavouring to obtain the adjustment in question, with no other result than this experience and the discovery of the cause of repeated failure—as I will now explain. The first adjustment required would seem to be to make the minor weight hang so that its mirror shall be parallel to the middle mirror of the major weight. To secure this the holding of the single sus- pending wire must be turned, either above or below. It is difficult, if not impossible, to get at the lower holding. But the upper one offers no difficulty except what is to be expected from the smallness of the parts. The necessary adjustment was at length made, approxi- mately. The second adjustment consists In so managing the torsion of the double wires, that when the minor weight (with its mirror) is turned through 360° either way, the major weight shall present its. right or its left hand mirror equally short of or beyond the ultimate position. The observation in that case will consist in ascertaining what alteration of weight will bring about exact conjunction, in either case. The difficulty, which I have already commented on, of giving any precise amount of rotation to the holding screws, is in this matter also so great as to make exact adjustment quite fortuitous. This will explain why I made the same adjustment several times wm succession (on each occasion with sorely tried temper and patience) before On Gravimeters. 531 becoming aware that there was something wrong. I traced this at last distinctly to a want of permanence of the first adjustment. I had spent from first to last not less than ten hours on these adjustments alone, which may give some idea of their uncertainty. It was now necessary to ascertain what was the source of the instability. It must be clearly understood that there is not any part of the whole design which is of greater importance than the attain- ment of permanence in this part. If, after disturbance, the lower weight does not return to the same position, relatively to the upper, with absolute exactness, I see no chance of obtaining anything which can be called a result. It is literally a sine quad non. The instability might be due to one of two causes—I see no third alternative. Hither the holding was insecure; or the wire was strained beyond what its elasticity would bear. I tried various plans to test this. At first it seemed clear that the holding was in fault. Then I fancied that the wire was strained. I mention this vacillation purposely, because my final conclusion, which condemns the holding after all, though more hopeful, might be wrong, and as it challenges the “special manner” noticed in Mr. Broun’s description, it will be best not to be too certain. The following test, however, seems con- clusive. I prepared two needles of deal, about 2 inches long and as thin and light as seemed necessary. These were split at one end, and thrust upon the taut wire, so as to stand out from it horizontally. One was placed close to the upper holding, but free from contact; the other about 0°3 inch lower. The upper weight was clamped, and the lower then turned through two entire revolutions. [1 did not scruple to overdo it, having ascertained from collateral experiments with other wire that a wire will bear being turned twenty or thirty times to every inch, without any other ill effect than a permanent twist. | The result was that the upper index turned through a few (8 or 10) degrees, and the lower through 70° or 80°—the latter being sensibly in due proportion to its distance: as to the former, I could not say exactly where the holding point might be. Now, the test would be in the positions to which they would return. If the wire was strained, the lower index would not return to conjunction with the upper: if the holding had failed, the two would not return to the starting point. The event proved the latter alternative. The lower index returned to conjunction with the upper; but both failed to return to the original position. I repeated the experiment, giving the weight only a single turn. The result was the same, in less degree: the holding had again failed, z.e., had allowed the wire to turn in its socket still further. The screws were all firm, on trial. The above test is so easy, and useful, that a description of it needs no apology. Still, itis rather by way of proof that I give it; for a 532 ; Major J. Herschel. bare assertion of belief in the insecurity of the holding might fairly be challenged. The above result could in no way be passed by. It was necessary to run any needful risk to ascertain further whether the evil was accidental and capable of being remedied without some decisive altera- tion. I therefore loosened the fastening and drew out the wire. The only result of this was to show that the wire was held solely by the pinch of a split screw. This is the ‘“‘ special manner” alluded to in the description, and I have no hesitation in saying that it is bad, because it relies on the almost microscopic accuracy of cutting of the parts. Assuming that no play is required in respect of length, there can be no objection to doubling back the wire before clamping it—and - that in a large screw, split boldly for half an inch or more. No slip could possibly happen in that case, even with very slight clamping, This is the hold I have used (in the experiments alluded to), the clamp being a cleft in wood, pinched ad libitum. Under the circumstances I replaced the wire, and screwed the hold- ing nut nearly to the utmost the screw would bear; und then repeated the test—with the same result. The hold by friction (of brass upon gold) is, at any rate in this particular case, insecure. Nor do I see any means of remedy short of partial reconstruction, which should, of course, extend also to the other five holding points. Should any doubt remain as to the slipping I add the following. Wishing to see whether it was due to the split being too coarse, I decided on extricating the holder and examining it under a micro- scope ; as also to see if the wire showed any abrasion or destruction. To get the holder out involves a great deal of dismemberment; but the choice lay between this and leaving the question in some doubt. I succeeded in doing it—it is needless to describe how. The first result was the possibility of recognising clearly the nature of the holder, of which I give here an enlarged view. It is a screw about 7 inch long, with capstan head, and three capstan nuts; of which two hold it on the beam, and the third pinches the split point. The split is quite flat- sided. The shank is hollow, as far as to the end of the slit, where there is a cross tunnel. By way of trial I pinched a piece of silver wire in the way the suspension wire is pinched, and twisted it. It was thinner than the latter. I think the pinch held it, for it soon broke. This was against the theory of slipping. Fortunately, I had saved a fragment of the gold wire (about } inch) which had pro- jected at the free end and had been broken off by the insertion of the pin (I suppose). This piece I inserted and pinched firmly, as firmly as I dared without spoiling the screw. The projecting end I bent into a crook, which I grasped with tweezers. I then turned the screw. Would the wire strain, or break, or slip? That was the question. The event justified previous experience. It slipped freely. On Gravimeters. 533 T could not pinch it tight enough to hold. Subsequent examination showed no sign upon the gold wire. This proves conclusively that, under these circumstances, the pinch of a round wire is an insecure hold against torsion.* I cannot but regret this result, for the sake of the exquisite work- manship; but there is no help for it, the wires must be held in some other way before anything can be done to test the imstrument as a gravimeter. I find, for instance, that a slight flattening of the wire meets the slipping difficulty perfectly ; but a flattened wire strains (i.e., takes up a twisted or strained condition) very readily, and it is doubtful whether any alteration of the cylindrical form, even for a very small part of the length, can be permitted without risk. My examination having reached this stage, I found it necessary to decide whether to stop here or run some risk of injuring the apparent perfection of the instrument, in the endeavour to obtain the first requisite, a firm hold of the ends of the torsion wire. I defer, for the * T have since tried a steel wire pinched between lead sheets in avice. No pinch seems sufficient to prevent the slipping, if the wire is straight, but a slight crook suffices. 534 | Major J. Herschel. present, all reflections in order to proceed with my narrative. I decided to flatten the wire slightly at the place where it passes the © split. Whether I did so insufficiently or not at the right place it is impossible now to say. The result was slipping, as before. I then determined to have recourse to the loop—by which I mean giving to the wire at the place where it is pinched a sharp bend back upon itself. To do this the whole instrument had again to be dismembered. I cannot give an adequate idea of the anxiety attending a step of this kind, in the case of an unfamiliar instrument of delicate construction. My anxiety would have been greater perhaps, but hardly my care, had I known, what I now learnt, that the gold wires were nearly as brittle as untempered steel. In dismounting the major weight (without which the lower attachment of the single wire could not be reached) one of the double suspending wires snapped off short at the fastening. I did not recognise the cause until on attempting to double back the single wire to give it the crook it also snapped short. These mishaps were experienced without much cost, for the new plan demanded but a fifth of an inch of wire; which the other end, in each case, could easily spare. But before trying to bend it again I took the precaution to anneal the end. At length, after a deal of trouble, I had succeeded in fastening in this way both ends of the single wire, and of the broken one of the pair. I ought to have done the same with the other as well, but courage was wanting to go further in this direction than accident had rendered necessary. The torsion on so long and thin a wire (the pair are much thinner than the single one) turned through only one quarter of a revolution, would probably be so slight as not to exceed the holding power of the existing attachment. I should say here, that before putting the parts together again, © I weighed them, and measured the wire-lengths. The weighing is elsewhere recounted. The lengths are 12°6 and 3:08 inches respec- tively. LIalso took the opportunity to measure exactly the diameter of the glass plunger, having reason to doubt the correctness of the measurement assigned in Mr. Broun’s description. The result justified my suspicion. The end of all this is now at hand. In due time the imstrument was once more in a condition to recommence the adjustments. Warned by previous experience I wasted no time over perfecting the first; but, noting the actual position of the lower mirror when at rest, I turned it through two revolutions and allowed it to return. I[é failed to reach its former place. I tightened the holding nuts to the utmost which the met would bear; with no better result than to reduce the slipping, but not to prevent it. The index test seemed to exonerate the upper holding, but there was either slipping or straining—and that to a variable extent—on every trial, whichever way the lower weight was turned. On Gravimeters. 535 At this point, I decided on abandoning the investigation, until I should receive further instructions. Considering that the instrument has been entrusted to me to experiment with rather than upon, I have already dared more perhaps than I ought to have—certainly more than most persons would have felt justified in doing. J presume, per- haps, in thinking that a rather long experience of instruments of precision will be accepted as my excuse for having gone so far; in the earnest endeavour to ascertain whether an instrument of such exceed- ing beauty (as to workmanship) would prove as valuable as it looked, aud as, I must say, the principle of its construction leads one to expect. In pursuing this endeavour I have done some slight injury to it—which can be easily remedied if necessary. This I admit: but per contra I have ascertained a good deal without which it would be useless, besides gaining some experience which should avail in per- fecting it, and making it (or another similar in principle) useful for its intended purpose. Finally, if any further apology is necessary, I will add this—that in no case could the instrument have been actually and efficiently employed for that purpose without material alterations. I cannot too often repeat, that both as regards the design and its execution, the instrument deserves high praise. Nevertheless it is a failure. I have felt this all along, and I ought not to conclude this paper without pointing out what I conceive to be its chief defects. Of the defect which has brought this trial to a premature conclusion —the insufficient hold of the ends of the wires—it is only necessary to say that some plan should be discovered of putting this beyond question. This is the first consideration. I do not regard this as an error either of design or of construction such as can be complained of. But I do regard as such that which gives the instrument, as actually constructed, its beauty, viz., the minuteness and delicacy of its parts. There is a wealth of adjustment which is not necessary, and their details are all on far too small a scale. I suppose that the major weight consists of not less than 100 parts. Of these probably 80 could be set apart whose total weight would not reach 200 ers. out of the 3,100 which the whole weighs. It is clear that, supposing the total to be restricted to that, a more generous distribution to the smaller parts might have been made, without any disadvantage; but on the contrary, a great gain in handiness. JI fear it is useless to add that a large proportion of these smaller parts are of the kind which instrument makers delight to show their wonderful skill in producing —as nature does flowers. Of course, one cannot tell how far these multifarious adjustments may not owe their presence to a conscientious endeavour on the part of the maker to give effect to his instructions : but that must not prevent my saying that they are, to a large extent, redundant, unnecessary, if not useless. I regard it as a mistake that the whole instrument is on so small a 536 Major J. Herschel. scale. We should not be far wrong in estimating the power of such an instrument, not 7m proportion to its size, but in an even higher ratio. Even in scale-balances there is an advantage in size; though there it is the absolute weight put into the scale which is measured, whereas in the gravimeter we have a balance whose delicacy is measured by the relutive minuteness of the weight added. The torsion gravimeter relies on the perfect obedience of the bifilar suspension, due to the absence of friction. Until it can be practically shown that this obedi- ence is, in practice, not perfect, it is an abuse of the leading principle to refrain from drawing upon the resource it offers. This is what is done when high constructive art and skill are exerted to keep down the weight and size of parts instead of the contrary. The minor weight is turned as one turns the hand of a clock with the finger. I cannot imagine why this one-sided action is preferred— for I suppose it is preferred—-to the obvious two-fingered action, which comes into play in so many common practices where one wishes to avoid displacing the central axis of motion. The effect is to give a wobbling swing to the whole hanging system, increasing the risk of jar and strain, besides displacing the centre, and thereby altering the normal direction of the pull of the lower weight as well as the verti- cality of the mirrors. [It is true that magnets are provided to relieve this, but I hardly suppose any one desiring to make accurate observa- tions would allow them to remain as preferable to the two-fingered stop.] I have already remarked on the so-called “ finely-polished agate points,’ which I have been obliged to replace by steel ones, because being very thin and of glass they soon got broken off. Ido not recognise any objection to their being of metal. A tripod stand is furnished with the instrument, and I always used it; but it is very unsuitable, for the following reason. A portable tripod almost necessarily requires a large splay; this involves risk— even in the hands of a surveyor habituated to three-legged stands— and risk of a kind which, in my opinion, is fatal in the case of such an instrument as this. This is why I notice the stand. I doubt if anything like the necessary permanence of condition could be looked for in a fine wire which, when supporting such a weight as 600 grs., had to sustain a jolt, or such a shock as would be caused by a slight kick to one of the legs of the stand. I have no proof of this. It is one of the tests I intended to inflict on the instrument. In the absence of anything but a strong doubt, I can do no better than set it down here for future trial. The last point which I shall dwell upon is one which has already been noticed, the effect of torsion of the double wires in deferring the position of unstable equilibrium beyond that of maximum gravity resistance. As already pointed out, the possibility of choosing 90° as the place of rest turns on the alliance of this torsion with the force of — ~ On Gravimeters. 537 gravity in opposing the torsion of the single wire. If Mr. Broun was aware of this, it is strange that he gives no hint of it in mentioning the forces in equilibrium ; anyhow, the third force is there, and might, I think, be taken advantage of. I found by trial that the torsion of the single wire might be increased so cautiously as to cause the major weight to stand nearly stationary at the position of unstable equili- brium. If the angle of position of the lower weight were read off (by vernier or otherwise) when this happened, both right hand and left hand, I imagine a delicate measure of the variation of gravity would be obtained, without the need of any appliances for varying the weight, or of a collimator. It would be foreign to the purpose of this paper to pursue this design further here; it is enough to have indi- cated it in connexion with the principle involved in Broun’s design. I will only add that the two principles here indicated, viz., the balancing of gravity by torsion, and the determination of the condition of equili- brium when unstable, are both involved in a simple bifilar suspension ; and I see reason to think that this form of gravimeter, from its ex- treme simplicity and great adaptability, is worth consideration.* It now only remains to offer some apology for the length of this report, and for its discursiveness. It will have been quite apparent to any one who has had the patience to read it, that it has been written from day to day, as the examination proceeded; and now that the latter has to be closed without anything of the nature of an ‘‘ observa- tion ” having been possible, it may be that there is a certain advan- tage in letting the facts, so brought forward, tell their own tale. Much, no doubt, could well be spared, but at the risk of laming the narrative. . . . . . . At the same time, I may say that Tam quite willing . renew the attack, and do what can be done to reach a more promising conclusion, and to make the instrument efficient ; provided I am so instructed. In that case, however, the sanction of the Royal Society, to whom it belongs, will be also * Tf (P+ 1) sin@=2o¢ be the general statical equation of a simple bifilar system, the equilibrium is unstable when sec?= —— . “, in which position p @p=—tan %. From this it is at once apparent, since ¢ must be positive, that 9, must lie between 90° and 180°. It must also be greater than 90°, or ¢) would be in- finite ; but beyond this I do not see any theoretical restriction. I think it would be advisable to make ¢, as large—several revolutions—as the wires will bear without injury. All that is necessary is that they be provided with means of turning their upper ends, until the weight reaches its position of unstable equilibrium (which should happen when @ is little more than 90°), and with means of recording the angle through which they have been turned so as to reach this condition. The ap- plicability of such an arrangement, as a sensitive gravimeter, will depend entirely on the adaptation ; which, I believe, would be found quite feasible. The equations here given are sufficient to determine suitable proportions. . VOL. XXXII. 2 Q 538 Major J. Herschel. necessary ; aS alterations may be necessary which one would otherwise — have no right to make. I append a separate account of the weighings of the major and minor weights. Supposing the instrument ultimately brought into use, it would become necessary to know the total weight of the suspended parts with some accuracy. This need should be kept in view, if alterations are made. Account of the Weighing. I have weighed the suspended parts on three occasions, the circumstances differing in each case somewhat. In the first two weighings, the cross piece at the top of the shaft having been removed, the whole of the swinging parts were suspended from one arm of a balance—with exception of three mirrors absent for repair at the time. This suspension involved certain additions and sub- tractions by way of allowance for parts not present or redundant. These were either determined or estimated for. I have since been able to correct all the estimations, and the first weighing makes the sum of the two parts A and B equal to 3,721+69+14—35—18—22 =3,/29 ors. In the second weighing—being uncertain of the accuracy of the weights used—I prepared two lead blocks representing approxi- mately the masses A and B; and, by more directly counterpoising the redundant parts, and reducing to a minimum all corrections, I ob- tained a mass A equal to 3,109 grs. and a mass B equal to 615 ers., the sum of which or 3,724 grs. balanced, or would have balanced (for I could not eliminate two small pieces which had to be allowed for) the whole of the suspended parts. The separation into two parts provided for the removal of the lesser when the minor weight was supported. I was quite aware that this was a doubtful partition, owing to the impossibility of exactly supporting the lower weight without imparting a thrust through the single wire. But there was no way of obviating this without a separation of the two paris of the mechanism. The total weight was free from suspicion. When later events led me to take the whole apparatus to pieces, I took care before building it up again to weigh the two principal parts separately—as well as to take some measurements which seemed important. The result showed that this thrust had been much stronger than I supposed, having transferred 9 or 10 gers. from the lower to the upper estimate. I now found, directly and without any allowances or reductions—the large mass being in one scale, and the major weight as zt would hang in the other, that the former required paring down. The parings weighed 9 grs. Conversely the smaller mass required an addition of 10 grs. to balance the minor weight. Having made this transfer, I now have the two leaden masses, repre- On Gravimeters. Hao senting (each within 1 gr., I think) the major and minor weights, as they actually hang: they weigh (according to my scale weights) 3,100 and 624 grs. respectively. The sum agrees with the second weighing to a grain—the exact agreement being unintentional. These weighings do not include the platinum wire riders. I may add here an account of the measurement of the glass plunger—described by Mr. Broun as “ calibrated”’ to 1 millim. My rough (?) measurement had indicated, as above said, ‘048 inch as its diameter. The cathetometer microscope seemed to show it about 120 divisions, or 1°2 millims.=-049 inch, but I did not trust this. I hung a small weight by a fibre of raw silk and wound up 100 turns of it on the glass rod. The length absorbed was 14°9 inches. I esti- mate the diameter of the silk, which was coarse, at ‘001; but it is very uneven. This would give VEY Ail =e inch. Not satis- TT fied, I repeated this with some fine silver wire, annealing it first. This gave similarly O1S4* .0036=-0454 inch. The diameter of the TT wire was got by measuring the length covered by the close coil of 100 turns. The true diameter of the glass rod is rather larger than this last, as the outer part of the wire would stretch more than the inner would compress. It may be taken as ‘046 inch at the place chosen, which was unfortunately near the top. As 1 millim. is ‘0394. inch, Mr. Broun’s statement on this point must be rejected. I cannot imagine on what ground this plunger has been made of glass. Surely it cannot be contended that a metal wire would be of uneven diameter, in a way that a glass one would not be? On the other hand the risk of injury is considerable, and such a rod would be impossible to replace in foreign parts. DESCRIPTION OF PLATES 8 AND 9. As has been mentioned in the note to p. 519 the plates were prepared subsequent to the submission of the Report, which they are intended to illustrate. Plate 8 is a general oblique view of the gravimeter on its tripod stand. The chief parts seen are—the shaft through which the parallel wires descend :—the chamber through the glass front of which is seen obscurely what is represented very faithfully in Plate 9 :—the collimator, or observing telescope, to which a somewhat undue pro- minence is given in the drawing, by the effect of foreshortening :—the ftadle and foot-screws :—the tripod stand :—the well and glycerine reservoir, with the cathe- tometer on the left. There is hardly any part of this plate which requires more special explanation than will be found in the foregoing pages, if we except perhaps the arrangement of nuts and screws on the outside of the chamber ; and these will be readily understood by consulting the other plate, where it is seen that they serve to govern the positions of the two stages inside by which the hanging part (the major weight of the Report) is held so as to maintain a fixed position, with slackened supporting wires, when Owe, 540 Mr. G. F. Rodwell. packed. There is a different arrangement for holding the minor weight, which the engraver has failed in representing so well as the rest. Plate 9 shows also a variety of milled screw-heads, each of which has of course a purpose ; but as neither of them, except one by which the lower weight is turned round, has any part in the observations suchas they are described above, but only in ultimate manipulations which it is unnecessary to dwell upon here, they may be regarded as ornaments. It will be noticed that the collimator is removed, and one of the side supports is supposed broken off, to discover the minor weight; which last, {with its mirror and magnet bar, is seen turned through 30° or 40° into an oblique position. At thesame level and outside the chamber are seen arms, one of which resembles a cross. These are the guides for two magnets which have been removed. Their intended purpose is mentioned by Broun in his description. The pillar alongside the shaft is a case for a thermometer, the bulb of which is within the chamber. . “On the Coefficients of Expansion of the Diiodide of Lead, PbI,, and of an Alloy of Iodide of Lead with Iodide of Silver, Pbl,.Agl.” By GF. Ropwenn, F.R.AS, EC Ss Science Master in Marlborough College. Communicated by Professor A. W. WILLIAMSON, For. Sec. R.S. Received March 10. Read March 31, 1881. In former communications which I have had the honour of sub- mitting to the Royal Society, I have given determinations of the coefficients of expansion by heat of the chloride and bromide of silver and the iodide of mercury between 0° C. and the fusing point; also determinations of the coefficients of expansion and contraction of the jodide of silver, and of certain chlorobromiodides of silver. (“‘ Proc. Roy. Soe.,”’ vol. 25, pp. 280-303, and vol. 28, p. 284.) The iodide of lead, and an alloy of iodide of lead with iodide of silver, were thought to be very suitable substances for a continuation of these experiments. The following pages describe the results obtained. The experimental method was precisely similar to that before described, but the expansion apparatus was rendered more delicate by several notable changes suggested during the course of the former experiments. It is unnecessary to describe this apparatus again (for description vide “ Proc. Roy. Soc.,” vol. 25, p. 281-2), but it may be remembered that a homogeneous rod of the substance under examina- tion is connected with a series of levers which multiply 5,382 times, while the value of the movements is estimated by a micrometer screw ” reading to =535,5 of an inch. The following alterations were made mainly with a view of reducing the resistance by diminishing fric- tion, and thus adding to the sensibility of the apparatus :— 1. The wooden base N (fig. 1) was replaced by a massive stone Herschel. ehOO MvOVA OC VO) sO debe ee sa = Saas ain ny ro iy a =x — oo mM i\ a ii): Tw VAN Mi NN ru MN WH. Wesley del. W.Rhziu Vowmnon & CO imp Herschel. Proc. Roy. DOC Vol or tayo) lll all 7h AL, tual sc On the Coefficients of Expansion, §c. 541 block, to which the box containing the levers and the upright carrying the micrometer head were firmly bolted. ——— = Wy _———o "oT V Eel HTT eh a eae This block with figs. 1, 2, 3 is a reprint from “ Proceedings,” vol. 25, p. 280-282, where all are described. 2. The levers 8, U (fig. 2) were reduced in weight. 3. The spring Y (fig. 2) was removed, as it was found that the recoil of the helical spring Z was quite sufficient to bring the index back to zero, and the presence of Y served only to increase the resist- ance and general strain. 4. The vertical axis X (figs. 1 and 3) was no longer allowed to work in the upper confining plate of the framework b, but it was caused to turn lightly on a bent arm above, while below it rested upon a slightly hollowed ruby. 0. The steel chain W communicating motion from the lever U to the vertical axis X, which carries the index, was shortened and caused to wind upon the barrel in such a manner that when unwound to the BAD | MiG F. Rodwell: extent of half a single coil, it moved the index through its entire range, viz., from 0° to 180° of arc. By this means any possibility of the chain doubling upon itself was obviated. 6. But perhaps the most important changes were the removal of stuffing boxes I and the rods H from the trough F, and the substitu- tion of levers working over the rim of the trough; and the suspension of the rod of substance under examination in a cradle between the Jevers. This was effected in the following manner :— "tS MAT SIZE. Section through the trough longitudinally, showing the mode of suspension of the bar, and the position of the levers. A horizontal bar LL (fig. 4) was supported by rods N, N, strengthened by cross bars (not shown) Jet into the stone base of the instrument ; it carried Y-shaped brass levers B, B, moving about axes at C, CO (figs. 4 and 5), attached at the points A, A. F is the trough in which hot ceresine is used for heating the bar under examination, H, supported by the cradle G. Two rods I, I, which slide in holes K, K, and are capable of being held at any height by screws, support the cradle G. D is the rod (figs. 1 and 2) which bears upon the lever S, and M the point of the micrometer screw. The apparatus was standardised at frequent intervals by the use of a rod of fine homogeneous silver. Ceresine boiling at 430° C. was used to heat the rods in F, and it was heated to any desired tempera- ture by means of a Bunsen burner placed beneath, and near the ceutre of the trough. On the Coefficients of Hxpansion, §c. 943 End section of the trough, showing one of the levers, and the bar which carries it. Iodide of Lead.—Pure iodide of lead was cast into rods one-third of an inch in diameter and 6 inches long. The ends were made plane by a fine steel saw, and they were furnished with copper caps. Great difficulty was experienced in casting the rods, owing to the brittleness of the iodide. Slightly greased tubes of very thin German glass were used as the moulds, and as the rods would rarely slip out of the tubes the glass had usually to be chipped away along the whole length of the rod by the point of a knife. The iodide underwent the same changes of colour as were observed in the iodide of silver; that is to say, it fused to a bromine-red liquid, which, when solidified, became red-brown, and, while cooling, brick-red, reddish-yellow, and, when completely cool, orange-yellow. Harsh noises, like those produced by bending tin, were heard during the cooling of the mass, and the frac- ture was highly crystalline. Differences of opinion appear to exist as to the effect of fusing iodide of lead in the air. In the same volume of a standard work I find two exactly contrary opinions: for it is stated, on the one hand, that the iodide if fused in contact with air gives off a part of its iodine, becoming oxyiodide of lead; while elsewhere the iodide is classed among those which may be fused in an open vessel without change. _ In order to set this matter at rest, 56°1690 grms. of iodide were fused in a covered porcelain crucible. The fusion was continued DA4 Mr. G. F. Rodwell. for eight minutes, during which the cover was three times momentarily removed. Violet fumes of iodine escaped on each occasion, but on weighing the loss was found to amount to only "11036 per cent. Again, the mass was kept fused for four minutes, and the crucible cover was twice removed, but the loss had only increased to "1584 per cent. After a third fusion the total loss only amounted to ‘1718 per cent. Hence it is manifest that iodide of lead may safely be fused out of contact with air, with scarcely appreciable loss. When, however, the crucible cover was permanently removed, the iodide rapidly decom- posed. When the iodide was heated in a current of carbonic anhydride, it sublimed unchanged in crystals; while if it was heated in a eurrent of dry oxygen it rapidly decomposed, fine crystals of iodine collecting in the fore part of the tube. The specific gravity of iodide of lead, in common with the iodides of copper, silver, and potassium, is less than the mean specific gravity of its constituents. Karsten found it to be 6°0282, Boullay 6:11, and my own determinations gave 6°12. The calculated specific gravity is 6°629. The fusing point as determined by Mr. Carnelley is 383° C. The coefficient of cubical expansion for 1° C. was found to be ‘00007614 for temperatures between 0° C. and 205° C. It increased to 00008317 between 205° C. and 253° C. Between 253° C. and 265° C. the mass expanded rapidly, with a co- efficient nearly eight times greater than the previous, viz. :— "0006378. After the subsidence of this rapid expansion it no longer retained the original coefficient, but assumed one of more than double the amount, VIZ. :-— 000180. At temperatures some distance from the melting point the rod began | to bend, and it became necessary to assume that this last coefficient continues to the melting point. The expansion in passing from the solid to the liquid condition was determined by the method described in my previous paper. On the Coefficients of Hxpansion, &c. 545 It will be observed that the iodide of lead, as in the case of the iodide of mercury (“ Proc. Roy. Soc.,” vol. 28, p. 284), has three coefficients of expansion, viz. :—(a@) a coefficient somewhat less than that of chloride of silver up to 253° C.; (0) a coefficient during 12° C., nearly eight times greater than the preceding; and (c) finally a coefficient somewhat more than twice as great as that between 0° and 253° C., at temperatures above 265° C. Undoubtedly the iodide of lead, as in the case of the iodide of mercury, undergoes a molecular change while rapidly expanding between 253° and 265° C., and before assuming the higher coefficient. This is supported by the fact that the highly brittle and crystalline rod showed itself capable of bending after having undergone the rapid expansion. It will be remembered that the iodide of silver, which is very crystalline and brittle below 145° C., becomes amorphous and plastic above that temperature. The familiar example of sulphur will also recur to the mind. If we suppose a mass of iodide of lead to be heated from 0° C. to the melting point (383° C.) the following will be the volumes at the respective temperatures. Volume at 0° C. =1-:000000. = 205° =1°015608. te Dio =1:019595. of 265° = 1'027248. % 383° (solid) =1:048488. 383° (liquid) =1:078080. The curve is shown in Table A. The specific gravity of the iodide in the molten condition is 5°6247. The fact that a substance may possess two or three different co- efficients of expansion has apparently only been observed hitherto in the case of such substances as fusible alloy, because in determining the coefficients of solid bodies temperatures exceeding 100° C. have rarely been employed. Paraftfine or ceresine used as a. heating medium will allow the determination of coefficients to a temperature of 300° C., and, undoubtedly, many bodies when thus examined would be found to present anomalies similar to those remarked in the case of the iodides of lead and mercury. The Lead-Silver Iodide Alloy. Bearing in mind the peculiar nature of the coefficients of certain alloys of iodide of silver with the chloride and bromide (‘‘ Proc. Roy. Soc.” vol. 25, p. 292), it was thonght to be advisable to determine the coefficients of an alloy of iodide of lead with iodide of silver. These bodies were accordingly fused together in the proportion of one molecule of each, viz., PbI,.AgI. This contains in 100 parts e and Volume of loy, Pbiz.Agi. 000+ G00: 010+ G1 0-1 L 020-1 Sc0:l O¢0-1 Ge0-l [| \Ov0-1 iat a SIWNIOA S70: 0c0-1 Fee ane eee ationship between the Temperatur f Silver, and of the Lead-Silver Iodide Al o 2 O Se) pS Fy o é a Iodide o ? Iodide of Lead +9, 08S 09S OFS OES 00G OB O9b Ob Or OO OBES 09 OS OEE OOS O82 092 Ob2 O22 002 OBI OSI Ob! OdI OOl O08 09 Ob 0@ O an SSYUNLVYHSdW3al Table A.—Table showing the Rel On the Coefficients of Expansion, Sc. SAT Todide of lead...... 66°206 ieadtas est BS Las 29-7449 Iodide of silver .... 33°794 TVET he eece ee ies: 1575642 LOGINS IRE: 546909 100:000 100-0000 The substances were fused together in a porcelain crucible, and cast in thin glass tubes 9 inches long by one-third of an inch in diameter. The molten mass underwent the same changes of colour in cooling as either one of its constituents, and ultimately became a dull orange-coloured compact mass. Although composed of two sub- stances which are highly crystalline and brittle, the alloy was found to be hard and tenacious. Although the constituents are coarsely crystalline in structure, the alloy is finely granular. During the cooling of the mass it expanded with sufficient force to break the glass tube. Harsh noises were emitted during cooling, and the whole mass was sometimes jerked from its position; while, if held in the hand, it was felt to be agitated by strong tremors. Mr. T. Carnelley has determined for me the melting point of the alloy, which he finds to be 350° C. The specific gravity is 5°923. By repeated digestion with large volumes of boiling water the alloy is decomposed, the iodide of lead being dissolved, while the iodide of silver remains as a dull green powder. On examination in the expansion apparatus the alloy was found to undergo slow expansion to a temperature of 118° C., then, for 6° C., it simply absorbed heat without either contracting or expand- ing. At 124° C. contraction commenced, and continued at unequal rates till a temperature of 139° C. was attained. Then, again, the mass underwent neither contraction nor expansion during heating through 5° C., and then it commenced to expand somewhat rapidly. The most rapid contraction on heating took place between 130° and 133° C. Thus, in all, for the temperatures during which the mass con- tracted, the index moved through fifteen revolutions of 180° to 0° of are, and these were related to the temperatures in the following manner :— 1 revolution of index took place during heating from 124—128° C. 2 2 x o 128—130° C. | , . ¥ 130—13T?-e, 6 i : 131—133° ©. 2 - e f 133-—139° C. The details of these contractions are shown in Table B. The heating, especially at these temperatures, was excessively slow, and so moderated that a complete observation of the behaviour of the ‘Substance in the expansion apparatus lasted from three to four hours. Above 144° C. the alloy expanded with a coefficient about three times greater than that which it possessed between 0° and 118° C. , and of the Lead- ] Es | OS66: SAWNIOA. irre a GLE: 2 be 2 o Le) “4 ne) ° = H o ss =| M ~ |0000:1 Ge00:l Mr. G. F. Rodwell. 0900+) SZ00°1 ete ¢2dlo:l 0G10-\ Qbl Lvl 9bl Stl bbl Svl Sbl Ib! Ovl GSI SSI ZEl OSI Gel FEI Sei |ct isi OF! 62l Sal Zal 921 Gal pel Sal eal lal ol SAYNLVYAdWSaL H o e — os M Cy ° o =) | We) (e) ke Gy (o) at) io) by by r2 q ° os Bt) 5) S eH ialeal a jo) ie) ®o ro mle Cy (o) 7) =| ins] + Q jaa) o — ae) S H On the Coefficients of Expansion, &e. 549 Coefficients of Cubical Hepansion and Contraction of the Alloy for 1° C. Between 0° and 118° C.= +:0000306 , 124 ,, 128° C.= —-0003240 , 128 ,, 130° C.= —-0012990 1800 y, 6 181°C = 0017330 , J31 ,, 133° C.= —-0039000 , 188 ,, 139° C.= —-0004329 he 144 ,, 350° C.= +:0001150 Plus has been placed before the coefficients of expansion on heating ; minus before the coefficients of contraction. ‘The expansion in pass- ing from the solid to the liquid condition was determined as before. The coefficient between 144° C. and the fusing point increased rapidly with the temperature. If we take the volume at 0° C. as unity, we have the following volumes corresponding to the temperatures given :— Volume at 0° C. =1:000000 : 118 1:003610 i 124: 1:003610 A 128 1:002314 3 130 -999716 Me SIL "994517 2 133 986717 i 139 °984120 i. 144 -984120 As 150 984810 it 300 1:006500 : 350 (solid) 1:013790 : 350 (liquid) 1-024370 In regard to this alloy the following points may be noted :— 1. It possesses a similar density at three different temperatures. Thus, it is obvious that the density is the same at 0° C., at just below 180° C., and at 282° C. 2. Although the alloy contains only 33°794 per cent. of iodide of silver, it contracts as considerably during heating as the iodide itself. 3. While the iodide of silver commences its contraction at 142° C., and finishes it at 145°°5, the alloy commences to contract 18° C. lower (viz., at 124° C.) and finishes 6°°5 C. lower (viz., at 139° F.). 4, The chlorobromiodides of silver also began to contract on heat- ing (an effect which, of course, we must attribute solely to the pre- sence of iodide of silver) at 124° C., but they finished at 133° C. 5. The harsh sounds emitted by the alloy during cooling, and the tremors simultaneously propagated through the mass, prove that violent molecular agitation is going on at such time as the iodide of 550 Mr. G. F. Rodwell. silver is passing from the amorphous plastic condition to the brittle crystalline condition, within the mass of the iodide of lead. 6. The fusing point of the alloy is 177° C. lower than that of the iodide of silver, which constitutes one-third of its weight, and 33° C. lower than that of the iodide of lead, which constitutes two-thirds of its weight. 7. If the lowering of the fusing point (also markedly apparent in the case of the chlorobromiodides of silver) is due to the fact that similar particles of matter attract each other more powerfully than dissimilar, and hence, when the particles of two bodies are mutually diffused, the attraction becomes less, and the molecular motion is consequently more readily assimilated; the same cause may serve to explain the commencement of the phase of contraction on heating the alloy at a temperature 18° C. lower than the substance to which it owes this property. 8. It is interesting to compare one of the chlorobromiodides of silver with the lead-silver iodide alloy. For this purpose we will take the chlorobromiodide which contains the nearest approach to the same quantity of iodide of silver as the alloy. The second of the chloro- bromiodides before described (‘‘ Proc. Roy. Soc.,” vol. 25, p. 295) contains 41°484 per cent. of iodide of silver, and 58°5160 per cent. of the chloride and bromide of silver, which latter, from the heat point of view, may be regarded as the same substance, because their co- efficients of expansion are practically the same. It may be noted (vide below) that while the expansion of the bromide (which is slightly greater than that of the chloride) scarcely exceeds that of the iodide of lead, and while, moreover, the chlorobromiodide contains 8 per cent. more iodide of silver than the lead-silver iodide alloy, the amount of contraction by heat of the latter is more than twenty times greater than that of the former, although we must believe this effect to be solely due to the iodide of silver in each case. Comparison of the Coefficients of the Iodide of Lead and the Bromide of Silver, used in conjunction with Iodide of Silver in the forma- tion of the two Alloys given below. Todide of lead. Bromide of silver. Meltunie,qeing 2... 2c. Bert Osa aes o 427° C. NWolume at cOmC on wi: L000000 Fae 1:000000 Ae DOD oe ae il LOLS 60S (0 fee ae 1:021945 Bee es, TOLO5O4 ee 1027369 & PAG EON ie O2M2AS Sees 1028725 ea (cold) | IOS tccn 1:042531 a 383 (liquid) .. 1:078080 on AQT. (Sold) ioe negates ol ae iaieee 1:047855 LG OMe ns 1107225 551 » OC PANsion ficients of Ex On the Coe 0 L6G OS16c0- T OSSE66- F60L86. BG8E86- OOLG86- OCT 866- SO0S866- 809866. C9L866- 000000: T ‘1SV “LOATIS JO OPIpPOT ‘0 L&E 98P 460: T 9P99FO- T 662680: T Z€9900: T LE0Z10-: T 000000: T OSV tas Vy ISy “IQATIS JO OpIpot Fo "yuao zed PRP. Tp SuUTUTLIUOD ‘TOATIS JO 9pIpoTULoAqo.LoT YO ‘O 0&6 OLEVCO.- T O6LET0- T 090200: T OLS8F86- OZTFS6: OZTESG. LIL986- LISh66. 9TL666. PLEZO0- T 8019800. T SOT9E00- T 000000. T 1°V 14d ‘IOATIS JO OpIpOr Fo “quae tod F6Z. EE SuUIeZOD ‘KOT[V OPIPOT AOATIs-pvoT ‘LOATIG FO 9plpoy Fo pur “IOATIG JO OPIporwmtoarqoaopyH wv Fo osoy} YIM AO[[W optpoy TeATis-pvor] oy} Jo spuoroooy oy} Jo wostrweduto‘ SOU DO OD80-0 00 0000.00 GOOCH A TOL Chay AT quiod woreoytprpos ye prnbrt ° ° ° ° ° 008 OST ¢. PPL PVL aia I 6&1 eéL T&T O&T 861 Pel SIL ‘0 0 66 ress ss qurod surpout 4B prjos Jo ouINTO A 66 66 66 “c 66 66 6é (a3 (73 “ (74 qe OUINTO A ore ee oe uorytsoduog 552 On the Coefficients of Expansion, Se. The accompanying tables show the curves of contraction and ex- pansion of the lead-silver iodide alloy, and of its constituents. In Table B the scale has been enlarged in order to show the details of the contraction of the alloy on being heated from 124° C. to 139° C. (Addendum. Received April 8, 1881.) Other alloys of iodide of lead with iodide of silver have since been made, having the following composition :— (1.) 2AgI.PbI,, containing 50°517 per cent. of iodide of silver. (2) BAglL2Pbl, -: <, 9 43:360 i : G)) ) 4ersepl, =.) 40497 . 2 (4) SAcL4PblL, ,, 38:950 : i (5) lOAhOPpI, — _,” = 364190 : These all possessed the same general appearance as the alloy AgI.PbI, described above, which contains 33°794 per cent. of iodide of silver. But with the exception of No. 5 they were all so brittle that they could not be cast into rods suitable for use with the expan- sion apparatus. During cooling large rifts appeared in the rod at right angles to its length, at the time when the iodide of silver com- menced to expand. In the case of No. 1 the rod was violently broken during its cooling by the expanding iodide of silver; even when slowly annealed in hot paraffine. It may be noted that no such effect was produced in the case of the chlorobromiodides of silver, having the composition respectively: Ag,I,.AgBr.AgCl; Ag,I,AcBr.AgCl ; Ag I,.AgBr.AgCl; and containing in each case a larger percentage of iodide of silver (viz., 58°6404; 68°0171; and -73°9285) than the silver-lead iodide alloy No. 1. The chlorobromiodides, although, of course, their brittleness increased with the percentage of iodide of silver, formed less brittle rods than the iodide of silver, and than the first of the silver-lead iodide alloys, although the latter contains 23 per cent. less iodide of silver than the chlorobromiodide Ag I,.AgBr.AgCl. INDEX to VOL, XXXII. ABNEY (W. de W.), note on the spec- trum of sodium, 443. Absorption of gas by the intestines and action of carminatives uponit (Brunton and Cash), 456. of gases by solids (Hannay), 407. spectra of cobalt salts (Russell), 258. Acipenser ruthenus and A. Sturio, struc- ture and development of the skull in (Parker), 142. Ammonia and its salts, on the action of, upon muscle and nerve (Brunton and Cash), 384. Amylolytic and proteolytic activity of pancreatic extracts, on the estimation of the (Roberts), 145. Anemometers, discussion of the results of some experiments with whirled (Stokes), 170. Annual meeting for election of Fellows, June 3, 1881, 332. Atkinson (R. W.) on the diastase of Koji, 299. Ayrton (W. E.) elected, 332 ; admitted, 339. Bates (H. W.) elected, 332; admitted, 333. Blood corpuscles, the relation of the white, to the coagulation of the blood (Wool- dridge), 413. Brachial and lumbo-sacral plexuses, func- tional relations of the motor roots of the (Ferrier and Yeo), 12. Bristowe (J. 8.) elected, 332; admitted, 333. Brooks (W. K.), Lucifer: a study in morphology, 46. Broun (J. A.), torsion gravimeter de- signed by the late (Herschel), 507. Brunton (T. L.) and T. Cash on the absorption of gas by the intestines and the action of carminatives upon it, 456. VOL. XXXII. Brunton (T. L.) and T. Cash on the action of alkali and acid on muscle: frog and rabbit, 456. on the action of ammonia and its salts, and of hydrocyanic acid upon muscle and nerve, 384. on the effect of electrical stimulation of the frog’s heart, and its modification by cold, heat, and the action of drugs, 383. Candidates, list of, 104. Capillary electroscope, phenomena of the (Gore), 85. Carbonic acid, note on the spectrum of (Wesendonck), 380. Cash (T.) and G. F. Yeo, the effects of certain modifying influences on the latent period of muscle contraction, 456. and T. L. Brunton on the absorp- tion of gas by the intestines and the action of carminatives upon it, 456. on the action of alkali and acid on muscle: frog and rabbit, 456. Chasles (Michel), obituary notice of, 1. Christie (W. H. M.) elected, 332; ad- mitted, 333. Coagulation of the blood, the relatiou of the white hlood corpuscles to the (Wooldridge), 413. Coal-dust, on the influence of, in colliery explosions, No. III (Galloway), 454. Cobalt salts, on the absorption spectra of (Russell), 258. Coefficients of expansion of the di-iodide of lead, PbI,, and of an alloy of iodide of lead with iodide of silver, PbI,AgI (Rodwell), 23, 540. Crookes (W.) on discontinuous phos- phorescent spectra in high vacua, 205. Croonian lecture.—Observations on the Zeer D4 locomotor system of Echinodermata (Romanes and Ewart), 1. Darwin (G. H.) on the stresses caused in the interior of the earth by the weight of continents and mountains, 432. Daubrée (G. A.) elected, 162. Definite integrals, on certain, No. 9 (Russell), 450. on some, which satisfy Riccati’s equation and its transformations (Glaisher), 4.44. Determination of the ohm in absolute measure (Rayleigh and Schuster), 104. Dewar (J.) and G. D. Liveing, inves- tigations on the spectrum of magne- sium, 189. on the identity of spectral lines of aifferent elements, 225. on the reversal of the lines of metallic vapours, No. VIII, 402. Diastase of 467i (Atkinson), 299. Dickie (G.) elected, 332. Diffusion of liquids, influence of voltaic currents on the (Gore), 56. Digastric, on the tendinous intersection of the (Dobson), 29. Di-iodide of lead, PbI,, coefficients of expansion of, and of an alloy of iodide of lead with iodide of silver, PbI,AgI (Rodwell) , 23, 540. Discontinuous phosphorescent spectra in high vacua (Crookes), 206. Dobson (G. H.) on the tendinous inter- section of the digastric, 29. Dodgson (W.) and B. Stewart, note on a comparison of the diurnal ranges of magnetic declination at Toronto and Kew, 406. Echinodermata, locomotor system of (Romanes and Ewart), 1. Election of Fellows, annual meeting for, 332. Election of foreign members, 162. Electric current, permanent molecular torsion of conducting wires produced by the passage of an (Hughes), 25. Electrical stimulation of the frog’s heart, on the effect of, and its modification by cold, heat, and the action of drugs (Brunton and Cash), 383. Electricity, on the refraction of (Tribe), 435. Electrodynamic balance, on an (Helm- holtz), 39. Ellis (A. J.), postscript to the chronolo- gical summary of methods of comput- ing logarithms in my paper on the potential radix, 377. INDEX. Ewart (J. C.) on a new form of febrile disease associated with the presence of an organism distributed with milk from the Oldmill Reformatory School, Aberdeen, 456, 492. and G. J. Romanes, observations on the locomotor system of Hehino- dermata (Croonian lecture), 1. Ewing (J. A.), effects of stress on the thermoelectric quality of metals. Part I, 399. Febrile disease, on a new form of, asso- ciated with the presence of an organism distributed with milk from the Old- mill Reformatory School, Aberdeen (Ewart), 456. Female organs and placentation of the fda (Procyon lotor) (Watson), 272. Ferrier (D.) and G. F. Yeo, the functional relations of the motor roots of the brachial and lumbo-sacral plexuses, 12. Forbes (G.) and J. Young, experimental determination of the velocity of white and of coloured light, 24:7. Foreign members elected, 162. Formule for sn 8x, en 8u, dn 8u, in terms of sn wu (Glaisher), 444. Functional relations of the motor roots of the brachial and Ilumbo-sacral plexuses (Ferrier and Yeo), 12. Galloway (W.) on the influence of coal- dust in colliery explosions: No. III, 454, Gladstone (W. H.) admitted, 206. Glaisher (EH. H.), formule for sn 8u, cn 8u, dn 8w, in terms of sn uw, 444, (J. W. LL.) on Riccati’s equation and its transformations, and on some definite integrals which satisfy them, 444), Gore (G.), influence of voltaic currents on the diffusion of liquids, 56. -—— phenomena of the capillary elec- troscope, 85. Gravimeters, on, with special reference to a torsion gravimeter designed by the late J. A. Broun (Herschel), 507. Hannay (J. B.) on the absorption of gases by solids, 407. on the states of matter, 408. Heart, on the effect of electrical stimula- tion of the frog’s, and its modification by cold, heat, and the action of drugs (Brunton and Cash), 383. Helmholtz (H.) on an electrodynamic balance, 39. on the internal forces of magnetized INDEX. and dielectrically polarized bodies, 40. Herschel (Major J.) on gravimeters, with special reference to a torsion eravimeter designed by the late J. A. Broun, 507. Histology and physiology of the pepsin- forming glands (Langley), 20. Hughes (D. E.), molecular magnetism, 213. permanent molecular torsion of conducting wires produced by the passage of an electric current, 25. Hydrocyanic acid, on the action of, upon muscle and nerve (Brunton and Cash), 384. Identity of spectral lines of different elements (Liveing and Dewar), 225. Internal forces of magnetized and dielec- - trically polarized bodies, on the (Helm- holtz), 40. Intestines, on the absorption of gas by the, and the action of carminatives upon it (Brunton and Cash), 456. Kempe (A. B.) elected, 332. King crab, note on the existence of stig- mata in the (Lankester), 391. K6ji, on the diastase of (Atkinson), 299. Langley (J. N.) on the histology and physiology of the pepsin-forming glands, 20. Lankester (H. Ray), note on the exis- tence in the king crab (Limulus poly- phemus) of stigmata corresponding to the respiratory stigmata of the pulmo- nate Arachnida, and on the morpho- logical agreements between Limulus and Scorpio, 391. Light, experimental determination of the velocity of white and of coloured (Young and Forbes), 247. Limulus polyphemus, note on the exis- tence of stigmata in (Lankester), 391. Liquids, influence of voltaic currents on the diffusion of (Gore), 56. Liveing (G. D.) and J. Dewar, investiga- tions on the spectrum of magnesium, 189. on the identity of spectral lines of different elements, 225. on the reversal of the lines of metallic vapours. No. VIII. (Iron, titanium, chromium, and aluminium), 402. Lockyer (J.-N.), note on the reduction of the observations of the spectra of 100 sun-spots observed at Kensington, 208. Locomotor system of EH-:hinodermata (Romanes and Ewart), 1. D905 Logarithms, postscript to the chrono- logical summary of methods of com- puting, in my paper on the potential radix (Ellis), 377. Lucifer, a study in morphology (Brooks), 46. Lung of the newt, on the minute structure of the, with special refe- rence to its nervous apparatus (Stir- ling), 37. Macalister (A.) elected, 332. Macewen (W.), observations concerning transplantation of bone. Illustrated by a case of inter-human osseous transplantation, whereby over two- thirds of the shaft of a humerus was restored, 232. McLeod (H.) elected, 332; admitted, 3098. Magnesium, investigations on the spec- trum of (Liveing and Dewar), 189. Magnetic declination, note on a com- parison of the diurnal ranges of, at Toronto and Kew (Stewart and Dodg- son), 406. Magnetism, molecular (Hughes), 213. Magnetized and dielectrically polarized bodies, on the internal forces of (Helmholtz), 40. Mannheim (A.) sur la surface de l’onde, et théorémes relatifs aux lignes de courbure des surfaces du second ordre, 44:7. Marignac (J. C.) elected, 162. Matter, on the states of (Hannay), 408. Metallic vapours, on the reversal of the lines of, No. VIII (Liveing and Dewar), 402. Metals, effects of stress on the thermo- electric quality of (Ewing), 399. Miik, on a new form of febrile disease associated with the presence of an organism distributed with (Ewart), 456. Molecular magnetism (Hughes), 2138. torsion, permanent, of conducting wires produced by the passage of an electric current (Hughes), 25. volume of solids (Wilson), 455. Morphologicalagreements between Limu- lus and Scorpio (Lankester), 391. Motor roots of the brachial and lumbo- sacral plexuses, functional relations of (Ferrier and Yeo), 12. Moulton (J. F.) and W. Spottiswoode on stratified discharges. VI. Shadows of striae, 385. on stratified discharges. VII. Multiple radiations from the negative terminal, 388. Muscle, on the action of alkali and acid on, 456. D096 Muscle and nerve, on the action of am- ,; monia and its salts, and of hydro- cyanic acid upon (Brunton and Cash), 384. contraction, the effects of certain modifying influences on the latent period of (Yeo and Cash), 456. Nageli (C.) elected, 162. Nervous apparatus, on the minute structure of the lung of the newt, with especial reference to its, 37. Newt, on the minute structure of the lung of the (Stirling), 37. Obituary notice of Michel Chasles, 1. Osseous transplantation, a case of in- terhuman (Macewen), 232. Pancreatic extracts, on the estimation of the amylolytic and proteolytic activity of (Roberts), 145. Parker (W. K.) on the structure and development of the skull in sturgeons (Acipenser ruthenus and A. sturio), 142. Pavy (F. W.), a new line of research bearing on the physiology of sugar in the animal system, 418. Pendent drops (Worthington), 362. Pepsin-forming glands, histology and physiology of the (Langley), 20. Permanent moiecular torsion of con- ducting wires produced by the passage of an electric current (Hughes), 25. Phillips (J. A.) elected, 332; admitted, 333. Physiological action of £8 (Williams and Waters), 162. Placentation of the racoon (Procyon lotor), (Watson), 272. Poisons, on the differences in the physio- logical effects produced by the, of certain species of Indian venomous snakes, (Wall), 333. Potential radix, postscript to the chro- nological summary of methods of com- puting logarithms in my paper on the (Ellis), 377. Preece (W. H.) elected, 332; admitted, 333. Presents, lists of, 48, 249, 498. Procyon lotor, on the female organs and placentation of, 272. Protagon, note on (Roscoe), 35. lutidine Racoon, female organs and placentation of the, 272. Rayleigh (Lord) and A. Schuster, on the determination of the ohm in absolute measure, 104. Refraction of electricity (Tribe), 435. INDEX. Reversal of the lines of metallic vapours, No. VIII (Liveing and Dewar), 402. Riccati’s equation and its transforma- tions (Glaisher), 444. Roberts (W.) on the estimation of the amylolytic and proteolytic activity of pancreatic extracts, 145. Rodwell (G. F.) on the coefficients of expansion of the di-iodide of lead, PblI., and of an alloy of iodide of lead with iodide of silver, PbI,AgI, 23, 540. Romanes (G. J.) and J. C. Ewart, ob- servations on the locomotor system of Echinodermata. (Croonian lecture) 1. Roscoe (H. E.), note on protagon, 35. Russell (J. S.), the wave of translation and the work it does as the carrier wave of sound, 382. Russell (W. H. L.) on certain definite integrals. No. 9, 450. Russell (W. J.) on the absorption spectra of cobalt salts, 258. Samuelson (B.) elected, 332. Schuster (A.) and Lord Rayleigh, de- termination of the ohm in absolute measure, 104. Skullin sturgeons, structure and develop - ment of the (Parker), 142. Snakes, on the differences in the physi- ological effects produced by the poisons of certain species of Indian venomous, 339. Sodium, note on the (Abney), 443. Solids, molecular volume of (Wilson), 455,457. Sound, the wave of translation and the work it does as the carrier wave of (Russell), 382. Spectra, on discontinuous phosphores- cent, in high vacua (Crookes), 206. Spectra of 10Q sun-spots observed at Kensington, note on the reduction of the observations of the (Lockyer), 203. Spectra, on the absorption, of cobalt salts (Russell), 258. Spectral lines of different elements, on the identity of (Liveing and Dewar), 2205. Spectrum of carbonic acid, note on the (Wesendonck), 380. of magnesium, inyestigations on / the (Liveing and Dewar), 189. of sodium, note on the (Abney), 443. ; Spottiswoode (W.) and J. F. Moulton on stratified discharges. VI. Shadows of striz, 385. on stratified discharges. VII. spectrum of INDEX. Multiple radiations from the negative terminal, 388. Stewart (B.) and W. Dodgson, note on a comparison of the diurnal ranges of magnetic declination at Toronto aud Kew, 406. Stigmata, note on the existence in the king crab of, corresponding to the re- spiratory stigmata of the pulmonate Arachnida (Lankester), 391. Stirlmg (W.) on the minute structure of the lung of the newt with especial reference to its nervous apparatus, 37. Stokes (G. G.), discussion of the results of scme experiments with whirled anemometers, 170. Stoney (B. B.) elected, 332. Stratified discharges. VI. Shadows of strie (Spottiswoode and Moulton), 385. — VII. Multiple radiations from the negative terminal, 388. Stress, effects of, on the thermoelectric quality of metals. Part I (Ewing), 399. and strain, the influence of, on the action of physical forces (Tomlinson), 41. Stresses caused in the interior of the earth by the weight of continents and mountains (Darwin), 432. Sturgeons, structure and development of the skull in (Parker), 142. Sugar, a new line of research bearing on the physiology of, in the animal system (Pavy), 418. Sun-spots, note on the reduction of the observations of the spectra of 100, observed at Kensington (Lockyer), 203. Surface de l’onde, sur la, et théorémes relatifs aux lenes de courbure des sur- faces dusecond ordre (Mannheim), 447. Tendinous intersection of the digastric, on the (Dobson), 29. Thermoelectric quality of metals, effects of stress on the (Ewing), 399. Tomlinson (H.), the influence of stress and strain on the actien of physical forces, 41. Torsion gravimeter, on gravimeters, with special reference to a, designed by the late J. A. Broun (Herschel), 507. Transplantation of bone, observations concerning (Macewen), 232. qr On ~T Traquair (R. H.) elected, 332. Tribe (A.) on the refraction of elec- tricity, 435. Velocity of white and of coloured light, experimental determination of the (Young and Forbes), 247. Voltaic currents, influence of, on the diffusion of liquids (Gore), 56. Wall (A. J.) on the differences in the physiological effects produced by thie poisons of certain species of Indian venomous snakes, 333. Waters (W. H.) and C. G. Williams, on the physiological action of 6 lutidine, 162. Watson (H. W.) elected, 332; admitted, 333. Watson (M.) on the female organs and placentation of the racoon (Procyon lotor), 272. Wave of translation, the, and the work it does as the carrier wave of sound (Russell), 382. Weierstrass (C.) elected, 162. Wesendonck (C.) note on the spectrum of carbonic acid, 380. Whirled anemometers, discussion of the results of some experiments with (Stokes), 170. Williams (C. G.) and W. H. Waters on the physiological action of 6 lutidine, 162. Wilson (E.), the molecular volume of solids, 455. Wooldridge (L. C.), the relation of the white blood corpuscles to the coagula- tion of the blood, 413. Worthington (A. M.) on pendent drops, 362. Wright (C. R. A.) elected, 332; ad- mitted, 333. Yeo (G. F.) and Dr. Cash, the effects of certain modifying influences on the latent period of muscle contraction, 456. and D. Ferrier, the functional relations of the motor roots of the brachial and lumbo-sacral plexuses, 12. Young (J.) and G. Forbes, experimental determination of the velocity of white and of coloured light, 247. END OF THE THIRTY-SECOND YOLUME. 7VOl, XXXT. ip! : Pe ae ate ae ey : mate DA mrep eye! trek ny J ; ; T4157 aa i HoA "" 4 tes e i ES rl AS ‘ as i HARRISON AND SONS, PRINTERS 1 n ‘st. MARTIN’ 3 2 » ¢ } PROCEEDINGS OF ae “ig ad ce) Vek Ge Gs Wits GO at Dead Le ee. XXXH. Nos 212: | Ste | ee ~H oe CONTENTS. wy “OY March 24, 1881. Stes I. Tue Croon1an Lecrurz.—Observations on the Locomotor System of Echinodermata. By Grorce J. Romanss, M.A., F.R.S., and Professor James C. Ewart, M.D. : : ; ; : II. The Functional Relations of the Motor Roots of the Brachial and Lumbo- Sacral Plexuses. By Davip Frrrizr, M.D., F.R.S., Professor of Forensic Medicine, and Gpratp F. Yro, M.D., F.R.C. S, , Professor of Physiology in King’s College . , : . - III. On the Histology and Physiology of the Pepsin-forming Glands. By J. N. Laneuuy, M.A., Fellow of Trinity College, Cambridge - — March 31, 1881. I. On the Coefficients of Expansion of the Di-iodide of Lead, PbI,, and of an Alloy of Iodide of Lead with Iodide of Silver, PbI,AgI. By G. F. Ropwett, F.R.A.S., F.C.S., Science Master in Marlborough College II. Permanent Molecular Torsion of Conducting Wires produced by the Passage of an Electric Current. By Professor D. E. Hucuzs, F.R.S. . III. On the Tendinous Intersection of the Gee By G. E. Dozson, M.A., M.B. : : ; : : ; ; IV. Note on Protagon. By Henry E. Roscoz, LL.D., F.R.S. For continuation of Contents see 4th page of Wrapper. Price Three Shillings. PAGE 12 20 23 25 29 30 PHILOSOPHICAL TRANSACTIONS. Contents oF Part ITI, 1880. XXI. A Memoir on the Single and Double Theta-Functions. By A. CAYLEY, F.R.S., Sadlerian Professor of Pure Mathematics in the University of Cambridge. XXII. Revision of the Atomic Weight of Aluminum. By J. W. Manet, F.RB.S., Professor of Chemistry in the University of Virginia. XXIII. Description of some Remains of the Gigantic Land-Lizard (Megalania prisca, OWEN), from Australia.—Part II. By Professor Owen, C.B., E.R.S., Xe. XXIV: On the Ova of the Echidna Hystrix. By Professor Owen, O.B., F.R.S., &e. XXYV. On the Determination of the Constants of the Cup Anemometer by Experi- ments with a Whirling Machine.—Part II. By T. R. Rosryson, D.D., E.R.S., &e. XXVI. On the Dynamo-electric Current, and on Certain Means to Improve its Steadiness. By C. Witur1Am Siemens, D.C.L., F.R.S. Index to Part ITI. Contents OF Part I, 1881. On the Structure and Development of the Skull in the Batrachia—Part III. By W. K. Parxer, F.R.S. PHILOSOPHICAL TRANSACTIONS. Part III, 1880, price £1 1s. Part I, 1881, price £2 10s. Extra volume (vol. 168) containing the Reports of the Naturalists attached to the Transit of Venus Expeditions. Price £3. Sold by Harrison and Sons. Separat ate copies of Papers in the Philosophical Transactions, commencing with 1B may be had of Triibner and Co., 57, Ludgate Hill. Phenomena of the Capillary Electroscope. 103 9. Influence of Chemical Action. The movements do not appear to depend upon the chemical nature of the solution, because they take place equally well with acid, alkaline, and neutral liquids. Being also purely physical, they are not dependent upon chemical action; such action, when it does occur, appears in every case to interfere with them. Note.—Since the publication of a previous communication “ On the Capillary Hlectroscope” (‘‘ Proc. Roy. Soc.,’” vol. 30, p. 32), I have been favoured by M. Lippmann with the following remarks respecting that instrument. ‘‘1st. The liquid is to be diluted sul- phuric acid, containing something like one-third its weight of sul- phuric acid. Weak acid does not film glass properly; most liquids do not; and then stoppages, or a jumping motion of the mercury, occur, such as you have described. 2nd. The capillary tube is to be eut very short (to about 10 millims.) the motions are in that case ten times more rapid than in a tube of 10 centims., because the friction is reduced in that proportion; besides, possibilities for obstruction are reduced also in the same proportion. 38rd. The instrument is only fit for measuring electromotive forces smaller than one Daniell; by using over-great electromotive forces the capillary constant goes over its maximum value, and then the movement of the mercury is reversed, as you noticed it to be the case at the end of your communication (see, about this maximum, ‘ Ann. Chimie et Physique,’ 1875, and also 1877). If you will do me the pleasure of visiting M. Jamin’s laboratory in the Sorbonne, you will find there several electrometers in good working order; three of these are being used by different observers for separate researches, with a precision of Daniell.” pape 10000 te wi CONTENTS (continued). April 7, 1881. PAGE I. On the Minute Structure of the Lung of the Newt with especial reference to its Nervous Apparatus. By Wiutiiam Sriruine, M.D., Se.D., Regius Professor of the Institutes of Medicine a in the University of Aberdeen. : : TI. On an Hlectrodynamic Balance. By H. Hetmuourz, For. Mem. B.S., Professor of Physics in the University of Berlin III. On the Internal Forces of Magnetized and Dielectrically Polarized Bodies. By Professor H. Hetmuoutz, For. Mem. B.S. April 28, 1881. I. The Influence of Stress and Strain on the Action of Physical Forces. By HERBERT TOMLINSON, B.A. : : ; . II. Lucifer: a Study in Morphology. By W. K. Brooxs, Associate in Biology and Director of the Chesapeake Zoological Laboratory of the Johns Hopkins University, Baltimore, Md., U.S.A. < ; 3 List of Presents A : i t 2 : 3 : 5 : Infivence of Voltaic Currents on the Diffusion of Liquids. By G. Gor, LL.D., F.B.S. : : : : : : ; Phenomena of the Capillary Electroscove. By G. Gorz, LL.D., F.R.S. 37 39 40 AL 46 48 56 85 CATALOGUE OF SCIENTIFIC PAPERS, COMPILED BY THE ROYAL SOCIETY. Published by Her Majesty’s Stationery Office. 8 vols., 4to. 1800—1873. Per vol.: 20s., cloth; 28s., half-morocco. On Sale by Murray, Albemarle Street, and Triibner and Co., Ludgate Hill. HARRISON AND SONS, 45 & 46, ST. MARTIN’S LANE, W.C., AND ALL BOOKSELLERS. PROCEEDINGS OF THE ROYAL SOCIETY. VOL. XXXII. . List of Candidates FE. 1G JOE IIT. IV. CONTENTS. May 5, 1881. On the Determination of the Ohm in Absolute Measure. By Lord RayieEieH, F.R.S., and ARTHUR ScuustTER, Ph.D., F.R.S. : On the Structure and Development of the Skull in Sturgeons eae ruthenus and A. stwrio). By. W. K. Parker, F.RS. . ; 5 On the Estimation of the Amylolytic and Proteolytic Activity of Pan- creatic Extracts. 'By Witi1am Roserts, M.D., F.R.S., Physician to the Manchester Royal ee and Professor of Clinical Medicine in Owens College : 5 : ; : : : , May 12, 1881. . On the Physiological Action of 6 Lutidine. By C.:GREVILLE WILLIAMS, F.R.S., and W. H. Waters, B.A., Demonstrator in the Physiological Laboratory, Cambridge : : A ; ; 3 A . Discussion of the Results of some Experiments with Whirled Anemo- meters. By Professor G. G. Stoxss, Sec. B.S. Investigations on the Spectrum of Magnesium. By G. D. Liverne, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., F.R.S., Jack- sonian Professor, University of Cambridge (Plate 1) ; Note on the Reduction of the Observations of the Spectra of 100 Sun- spots observed at Kensington. By J. Norman Lockyer, F.R.S. (Plate 2) : ‘ 4 : j For continuation of Contents see 4th page of Wrapper: Price Light Shillings. No. 2138. 145 162 170 189 203 PHILOSOPHICAL TRANSACTIONS. LLLP LLIN LLL LLLP DDD DD IPT Contents oF Part III, 1880. XXI. A Memoir on the Single and Double Theta-Functions. By A. CAYLEy, F.R.S., Sadlerian Professor of Pure Mathematics in the University of Cambridge. XXII. Revision of the Atomic Weight of Aluminum. By J. W. Mattet, F.R.S., Professor of Chemistry in the University of Virginia. XXIII. Description of some Remains of the Gigantic Land-Lizard (Megalania prisca, OWEN), from Australia.—Part II. By Professor Owzn, C.B., E.RB.S., &e. XXIV. On the Ova of the Hchidna Hystrix. By Professor Owen, C.B., F.RBS., &e. XXYV. On the Determination of the Constants of the Cup Anemometer by Experi- ments with a Whirling Machine.—Part Il. By T. R. Ropinson, D.D., E.R.S., &e. XXVI. On the Dynamo-electric Current, and on Certain Means to Improve its Steadiness. By C. WittiAm Siemens, D.C.L., F.R.S. Index to Part IIT. Contents OF Part IJ, 1881. On the Structure and Development of the Skull in the Batrachia—Part III. | By W. K. Parker, E.R.S. PHILOSOPHICAL TRANSACTIONS. Part III, 1880, price £1 1s. Part I, 1881, price £2 10s. Extra volume (vol. 168) containing the Reports of the Naturalists attached to the Transit of Venus Expeditions. Price £3. Sold by Harrison and Sons. Separate copies of Papers in the Philosophical Transactions, commencing with 1875, may be had of Triibner and Co., 47, Ludgate Hill. On the Diastase of Koji. Doll water from 077. Referring always to the material dried at 100° C., the total albuminoids in kéji amount to 9°84 per cent., whilst 8°34 per cent. was found in solution. In the case of rice, the amount of albuminoids dissolved by water is very small; in one specimen of cleaned rice examined, it was 1°38 per cent., the total albuminoids amounting to 6°47 per cent. It is evident, therefore, that one change effected by the growth of the fungus is to increase the proportion of soluble nitrogen. Ifthe total amount of albuminoids in kéji be compared with the amount in rice, it will be seen that it is greater in the former than in the latter, evidence of the removal of the carbohydrates, alluded to in Section I. The broken grains which result from the cleaning of the common rice are also converted into /éji. As the bran is mixed with the broken grains, the whole contains a larger proportion of albuminoids than the original rice. The following analyses were made of the broken grains, and of the /0ji formed from them. In both cases the results are given upon the material dried at 100° C. Table XIX.—Composition of K6j7 prepared from Broken Grains. Soluble solid Soluble Insoluble | Dextrose. iE tele Beis matter. | albuminoids. | albuminoids. Per cent. Per cent. Per cent. Per cent. Broken rice...... 5-04 a WAG 7 OBB JKOi0e ee eee 25 -90 12 -80 7-08 rule erg LE 8s Sa: f tes The amount of solid matter dissolved from the 071 is just about five times as much as that dissolved from’the rice, and the soluble albuminoids have increased in nearly the same proportion. Beyond the recognition of this alteration in the solubility of the albuminoids, and the related increase in the amount of solid matter dissolved, I have been able to come to no definite conclusion. I hope, however, to have the opportunity of examining more-in detail the nature of the individual albuminoids present in koji, but thus far, that is a part. of the problem I have not touched. It is, indeed, generally believed, that the converting effect of diastase is owing to the existence of certain albuminoids in solution, and the results obtained in this research go to show that the active properties of kdji are accompanied by the presence of soluble albuminoids, but to go beyond this, and to show how, by the growth of the fungus, this change is. effected, is a problem which, however interesting, lies beyond my power. Having drawn attention to the fact that a particular fungus has VOL. XXXII. 2A 332 On the Diastase of Koji. this power of rendering the rice grain diastatic, an effect which has hitherto been attributed only to the germination of the embryo, the question whether the effect 1s a general one or not, must be left to professed vegetable physiologists. In conclusion, the pleasant task remains to me of expressing my obligations, and offering my thanks to Hiroyuki Kato, Hsq., President of the University of T6kid, who has rendered my task a comparatively easy one, by the assistance he has given me in various ways. To Mr. Jihei Kameyama also, the proprietor of the /6ji manufactory in Yashima, T6ki6, I am deeply indebted for the willingness with which I have been allowed to make experiments, and to collect information in his works. I wish also to thank my assistant, Mr. Nakazawa, for the interest he has taken in the research, and for much. assistance which I have received from him. Now published. Price 4s. CATALOGUE OF THE SCIENTIFIC BOOKS IN THE LIBRARY OF THE ROYAL SOCIETY. Hirst SECTION :-—Containing Transactions, Journals, Observations and Reports, Surveys, Museums. Fellows of the Royal Scciety desiring to have direct information, by Post card, of the Papers to be read at the Ordinary Meetings of the Society, may obtain it by sending their names to Messrs, Harrison and Sons, Printers, 45, St. Martin’s Lane, W.C. CONTENTS (continued). May 19, 1881. PAGE I. On Discontinuous Phosphorescent Spectra in High Vacua. By WinLiAmM Crooxss, F.R.S. ; : : } ‘ P ; ‘ : . 206 II. Molecular Magnetism. By Professor D. EH. Hueuss, F.R.S. . ; . 218 III. On the Identity of Spectral-Lines of Different Elements. By G. D. Liveine, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., F.R.S., Jacksonian Professor, University of Cambridge. : : . 220 IV. Observations concerning Transplantation of Bone. Illustrated by a Case of Inter-human Osseous Transplantation, whereby over two-thirds of the Shaft of a Humerus was restored. By W. Macewen, M.D.. . 282 V. Experimental Determination of the Velocity of White and of Coloured Light. By Dr. J. Youne, F.R.S., and Professor G. ForBEs : . 247 List of Presents : : : : : 5 2 : : . 249 On the Absorption Spectra of Cobalt Salts. By Witrt1am J. RussELt, Ph.D., F.R.S., Treas. C.S., Lecturer on Chemistry at the Medical School, St. Bartholomew’s Hospital . : : : . : : . 258 On the Female Organs and Placentation of the Racoon (Procyon lotor). By M. Watson, M.D., Professor of ee: Owens es Manchester (Plates 3-6) : , ‘ ; : ene On the Diastase of Ké7i. By R. W. Arxinson, B.Se. Ge Professor of Analytical and Applied Chemistry in the University of Tékié, Japan . 299 Obituary Notice :— MiIcHEL CHASLES . Wagers . ‘ i : ? : 5 : i Received since the end of the Session. Preliminary Note on the Photographie Spectrum of Comet (June, 1881). By Witiiam Hueeins, D.C.L., LL.D., F.R.S. Note on the Reversal of the Spectrum of Cyanogen. By G. D. Livzrne, M.A., F.R.S., Professor of Chemistry, and J. Dewar, M.A., F.RB.S., Jacksonian Professor, University of Cambridge. CATALOGUE OF SCIENTIFIC PAPERS, COMPILED BY THE ROYAL SOCIETY. Published by Her Majesty’s Stationery Office. 8 vols., 4to. 1800—1873. Per vol.: 20s., cloth ; 28s., half-morocco. On Sale by Murray, Albemarle Street, and Tribner and Co., Ludgate Hill. HARRISON AND SONS, 45 & 46, ST. MARTIN’S LANE, W.C., AND ALL BOOKSELLERS. ~ VOL. XXXII. Annual Meeting for Election of Fellows 4 VI. VIL. Vill. PROCEEDINGS OF Pee ROYAL SOC LE TY. J CONTENTS./ ues yh Ss June 3, 1881.\ © Ve © ah . June 16, 1881. . On the Differences in the Physiological Effects produced by the Poisons of certain species of Indian Venomous Snakes. By A. J. Watt, M.D. (Lond.), Surgeon H.M. Indian Army . On Pendent Drops. By A. M. Worrsinerton, M.A. . Postscript to the Chronological Summary of Methods of Computing Logarithms in my Paper on the Potential Radix. By ALEXANDER J. Huuis, B.A., F.R.S., F.S.A. . . Note on the Spectrum of Carbonic Acid. By CHarLes WESENDONCK (Berlin) . The Wave of Translation and the Work it does as the Carrier Wave of Sound. By Joun Scoorr Russext, F.R.S. On the Effect of Electrical Stimulation of the Frog’s Heart, and its Modification by Cold, Heat, and the Action of Drugs. By 1. LAvDER Brunton, M.D., F.R.S., and THropoRE CasH, M.D. . : On the Action of Ammonia and its Salts, and of Hydrocyanic Acid upon Muscle and Nerve. By T. Lauper Brunton, M.D., F.R.S., and THEODORE CasH, M.D. : p : ; : : ; : On Stratified Discharges. VI. Shadows of Strie. By Wut11aM SPoTTiswooDE, President R.S., and J. FuetcHer Mouton, F.R.S. IX. On Stratified Discharges. VII. Multiple Radiations from the Negative Terminal. By Witztiam Sporriswoopz, P.R.S., and J. FrrrcHer Movtton, F.R.S. For continuation of Contents see 3rd and 4th pages of Wrapper. Price Five Shillings. No. 214. 333 362 377 380 382 383 388 PHILOSOPHICAL TRANSACTIONS. ContENtS oF Part III, 1880. XXI. A Memoir on the Single and Double Theta-Functions. By A. CAYLEY, F.R.S., Sadlerian Professor of Pure Mathematics in the University of Cambridge. XXII. Revision of the Atomic Weight of Aluminum. By J. W. Mazer, F.RS., Professor of Chemistry in the University of Virginia. XXIII. Description of some Remains of the Gigantic Land-Lizard (Megalania prisca, OWEN), from Australia.—Part II. By Professor Owen, C.B., E.RS., &e. XXIV. On the Ova of the Echidna Hystrix. By Professor Ow=n, C.B., F.B.S., &e. XXV. On the Determination of the Constants of the Cup Anemometer by Experi- ments with a Whirling Machine.—Part II. By T. R. Rosinson, D.D., E.R.S., &e. XXVI. On the Dynamo-electric Current, and on Certain Means to Improve its Steadiness. By C. Witiiam Siemens, D.C.L., F.R.S. Index to Part ITT. ConTENTS OF Part I, 1881. On the Structure and Development of the Skull in the Batrachia—Part III. By W. K. PaRgKER, F.R.S. PHILOSOPHICAL TRANSACTIONS. Part III, 1880, price £1 1s. Part I, 1881, price £2 10s. Extra volume (vol. 168) containing the Reports of the Naturalists attached to the Transit of Venus Expeditions. Price £3. Sold by Harrison and Sons. Separate copies of Papers in the Philosophical Transactions, commencing with 1875, may be had of Triibner and Co., 57, Ludgate Hill. CONTENTS (continued). PAGE X. Note on the Existence in the King Crab (Limulus polyphemus) of Stigmata corresponding to the Respiratory Stigmata of the Pulmo- nate Arachnida, and on the Morphological Agreements between Limulus and Scorpio. By E. Ray Lanxestur,’M.A., E.RBS., Jodrell Professor of Zoology in University College, London . . 3891 XI. Effects of Stress onthe Thermoelectric Quality of Metals. Part I. By J. A. Ewine, B.Sc., F.R.S.E., Professor of Mechanical Hngi- neering in the University of Tokio, Japan . : : 3 . 399 XII. On the Reversal of the Lines of Metallic Vapours. No. VIII. (Iron, Titanium, Chromium, and Aluminium.) By G. D. iiverne, M.A., F.R.S., Professor of .Chemistry, and J. Dewar, M.A., FE.RB.S., Jacksonian Professor, University of Cambridge . : : . 402 XIII. Note on a Comparison of the Diurnal Ranges of Magnetic Declina- tion at Toronto and Kew. By Batrour Stewart, LL.D., F.BS., and WiLLIam Dopa@son : : : ; : . : . 406 XIV. On the Absorption of Gases os Solids. By J. B. Hannay, F.R.S.E., HOS... ; 3 ; : : . 407 XY. On the States of Matter. By J. B. Hannay, F.R.S.E., F.C.S. . 408 XVI. The Relation of the White Blood Corpuscles to the Coagulation of the Blood. By L. C. WooupRip@ez, B.Sc. Lond. ie ase Anstalt, Leipzig . : : : . ; 413 XVIL A New Line of Research bearing on the Physiology of Sugar in the Animal System. By F.W. Pavy, M.D.,F.RS. . ; . 418 XVIII. On the Stresses caused in the Interior of the Earth by the Weight of Continents and Mountains. By G. H. Darwin, F.R.S. . . 432 XIX. On the Refraction of Electricity. By Atrrep Trips, F.LC., Lecturer on Chemistry in Dulwich College. : : 5 » 435 XX. Note on the Spectrum of Sodium. By ee W.de W. ee R.E., F.R.S. : : : 443 XXI. Formule for sn 8u, cn 8u, dn 8u,in terms of snw. By Ernest H. GLAISHER, B.A., Trinity College, Cambridge 4 : ; . 444 XXII. On Riccati’s Equation and its Transformations, and on some Definite Integrals which satisfy them. By J. W. L. GuAIsHER, M.A., F.R.S., Fellow of Trinity College, Cambridge . : . 444, XXIII. Sur la Surface de l’Onde, et Théorémes relatifs aux Lignes de Courbure des Surfaces du Second Ordre. Par A. MANNHEIM . AAT XXIV. On certain Definite Integrals. No. 9. By W. H. L. RusseEtz, E.RS. . A é : , ; , 3 : é . 450 XXY. On the Influence of Coal-dust im Colliery Explosions. No. III. By W. GaLLowAY . : : - 454: CONTENTS (continued). PAGE XXVI. The Molecular Volume of Solids. By E. W11tson : ; . 455 XXVII. The Effects of certain Modifying Influences on the latent Period of - Muscle Contraction. By Dr.G. F. YEoand Dr. CasH . . 456 XXVIII. On the Absorption of Gas by the Intestines and the Action of Car- minatives upon it. By T. LaupER ee M.D., E.RS., and THEODORE CasH, M.D. . : : . ° . 456 XXTIX. On the Action of Alkali and Acid on Muscle: Frog and Rabbit. By T. Lauper Brunton, M.D., F.R.S., and THEoporE CasH, M.D. . ; : : é : d : ° - ° . 456 XXX. Ona New Form of Febrile Disease associated with the presence of an Organism distributed with Milk from the Oldmill Reforma- tory School, Aberdeen. By J. C. Ewart, M.D., Professor of Natural History in the University of Aberdeen. ; : . 456 Now published. Price 4s. CATALOGUE OF THE SCIENTIFIO BOOKS IN THE LIBRARY OF THH ROYAL SOCIETY. First SECTION :-—Containing Transactions, Journals, Observations and Reports, Surveys, Museums. Fellows of the Royal Society desiring to have direct information, by Post card, of the Papers to be read at the Ordinary Meetings of the Society, may obtain it by sending their names to Messrs. Harrison and Sons, Printers, 45, St. Martin’s Lane, W.C. CATALOGUE OF SCIENTIFIC PAPERS, COMPILED BY THE ROYAL SOCIETY. Published by Her Majesty’s Stationery Office. 8 vols., 4to. 1800—1873. Per vol.: 20s., cloth ; 28s., half-morocco. On Sale by Murray, Albemarle Street, and Triibner and Co., Ludgate Hill. HARRISON AND SONS, 45 & 46, ST. MARTIN’S LANE, W.C., AND ALL BOOKSELLERS. PROCEEDINGS OF Pita wOYAL°SOCLE TY, WOE, XX X11. No. 215. a Bie? PAGE The Molecular Volume of Solids. By Epwarp WItson, 457 On a New Form of Febrile Disease associated with the presence of an Organism distributed with Milk from the Oldmill Reformatory School, Aberdeen. By J. C. Ewart, M.D., Professor of Natural History in the University of Aberdeen . : : : : : : : ~ 492 List of Presents . : : : ‘ : : ; : . 498 On Gravimeters ; with special reference to a Torsion Gravimeter, designed by the late J. Allan Broun, F.R.S. By Major J. Herscuet, R.E., F.RS., Deputy Superintendent of the Survey of India. (Plates 8, 9) : + 007 On the Coefficients of Expansion of the Di-iodide of Lead, PbI,, and of an Alloy of Iodide of Lead with Iodide of Silver, PbI,AgI. By G. F. Rop- WELL, F.R.A.S., F.C.S., Science Master in Marlborough College. . 540 Podex : 3 : : : : ; 2 5 : . 053 Title and Contents. Price Four Shillings. PHILOSOPHICAL TRANSACTIONS. — OOOO Contents OF Part I, 1881. On the Structure and Development of the Skull in the Batrachia—Part ILI. By W. K. Parxer, F.R.S. ConTENTS OF Part II, 1881. IT. The Cochlea of the Ornithorhynchus platypus compared with that of ordinary Mammals and of Birds. By Urspan PritcHarp, M.D., F.R.C.S. III. On the Organization of the Fossil Plants of the Coal-Measures.—Part XI. By W. C. Wiuutamson, F.R.S. ; IV. On the Induction of Hlectrie Currents in Infinite Plates and Spherical Shells. By C. Niven, M.A. | Y. Electrostatic Capacity of Glass, II, and of Liquids. By J. Hopxrrysoy, M.A., D.Sc., F.R.S. VI. On the Forty-eight Coordinates of a Cubic Curve in Space. By WILLIAM SPOTTISWOODE, Pres. R.S. VII. On the Viscosity of Gases at High Exhaustions. By WilL1AmM CROOKES, F.R.S. Note on the Reduction of Mr. Crookes’s Experiments on the Decrement of the Arc of Vibration of a Mica Plate oscillating within a Bulb containing more or less Rarefied Gas. By Professor G. G. Stokes, Sec.R.8. VIII. On the Electrical Resistance of Thin Liquid Films, with a revision of Newton’s Table of Colours. By A. W. Rernoxp, M.A., and W. Rucker, M.A. IX. On the Tidal Friction of a Planet attended by several Satellites, and on the Evolution of the Solar System. By G. H. Darwin, E.R.S. X. On the Thermal Conductivity of Water. By J. T. BorromieEy. XI. Description of some Remains of the Gigantic Land-Lizard (Megalania prisca, Owen), from Australia.—Part III. By Professor Owen, C.B., F.R.S., Xe. Price £1 10s. PHILOSOPHICAL TRANSACTIONS. Part III, 1880, price £1 1s. Part I, 1881, price £2 10s. Part II, price £1 10s. Extra volume (vol. 168) containing the Reports of the Naturalists attached to the Transit of Venus Expeditions. Price £3. Sold by Harrison and Sons. Separate copies of Papers in the Philosophical Transactions, commencing with 1875, may be had of Triibner and Co., 57, Ludgate Hill. GOVERNMENT GRANT OF £1000. A Merrine of the Government-Grant Committee will be held in February, 1882. It is requested that applications to be considered at that Meeting be forwarded to the Secretaries of the Royal Society, Burlington House, before the 31st December, 1881. Fellows of the Royal Society desiring to have direct information, by Post card, of the Papers to be read at the Ordinary Meetings of the Society, may obtain it by sending their names to Messrs. Harrison and Sons, Printers, 45, St. Martin’s Lane, W.C. Now published. Price 4s. CATALOGUE OF THE SCIENTIFIC BOOKS IN THE LIBRARY OF THE ROYAL SOCIETY. Fizst SEcTIon :—Containing Transactions, Journals, Observations and Reports, Surveys, Museums. CATALOGUE OF SCIENTIFIC PAPERS, COMPILED BY THE ROYAL SOCIETY. Published by Her Majesty’s Stationery Office. 8 vols., 4to. 1800—1873. Pervol.: 20s., cloth ; 28s., half-morocco. On Sale by Murray, Albemarle Street, and Triibner and Co., Ludgate Hill. HARRISON AND SONS, 45 & 46, ST. MARTIN’S LANE, W.C., AND ALL BOOKSELLERS. A fr ln ' ae > A Le ao) om hs { inp