' i ■ PEOCEEDINGS OF THE EOYAL SOCIETY EDINBUEGH. NOVEMBER 1887 to JULY 1888. EDINBUEGH: PEIETEB BY ETEILL AED COMPANY. MDCCCLXXXIX. CONTENTS. PAGE Election of Office-Bearers, . . . . . .1 Chairman’s Opening Address, . . . . .2 The Jubilee Prize, . . . . . . .9 Proposed Additions to the List of Honorary Fellows, . .10 The Edinburgh Equatorial in 1887 ; a Paper with two Appendices. By C. Piazzi Smyth, Esq., Astro nomer-Koyal for Scotland, . 11 On Cauchy’s and Green’s Doctrine of Extraneous Force to explain dynamically Fresnel’s Kinematics of Double Kefraction. By Sir W. Thomson, . . . . . . .21 On the Minimal Tetrakaidekahedron, with Exhibition of Models. By the President, . . . . . . .33 Kesearches on Micro-Organisms, including ideas of a new Method for their Destruction in certain cases of Contagious Diseases. Part II. By Dr A. B. Griffiths, F.E.S. Edin., . . .33 On the Colour of the Skin of Men and Animals in India. By Robert Wallace, Esq., Professor of Agriculture and Rural Economy in the University of Edinburgh, . . . . . .64 On the Height and Volume of the Dry Land, and the Depth and Volume of the Ocean. By Dr John Murray, . . .65 The Pineal Body {Epiphysis cerebri) in the Brains of the Walrus and Seals. By Professor Sir William Turner, M.B., LL.D., F.R.S., . 65 On a Method of graphically recording the Exact Time Relations of Cardiac Sounds and Murmurs. By Byrom Bramwell, M.D., and R. Milne Murray, M.B., . . . . . .66 On Benzyl Phosphines. By Professor E. A. Letts and W. Wheeler, Esq., . . . . . . . .66 A Criticism of the Theory of Subsidence as explaining the Origin of Coral Reefs. By H. B. Guppy, M.B., R.N. Communicated by Dr H. Mill, 84 On the Compressibility of Water and of Different Solutions of Common Salt. By Professor Tait, . ... .84 On a Practical Constant- Volume Air Thermometer. By J. T. Bottomley, Esq., . . . . . . .85 IV Contents. PAGE On the Roots of g2= _ i. By Dr Gustav Plarr. Communicated by Professor Tait, . . . . . . .93 On Vanishing Aggregates of Determinants. By Thomas Muir, LL.D., 96 On a Simplified Proof of MaxwelPs Theorem. By Professor Burn- side. Communicated by Professor Tait, . . . .106 On some Glass Globes with Internal Cavities produced during Cool- ing. By J. T. Bottomley, Esq,, ..... 108 Investigations on the Malpighian Tubes and the “ Hepatic Cells ” of the Araneina; and also on the Diverticula of the Asteridea. By Dr A. B. Griffiths, F.R.S. Edin., and Alexander Johnstone, F.G.S. Bond, and Edin., . . . . . .Ill On the Thomson Effect in Iron. By Professor Tait, . .115 Obituary Notices of former Vice-Presidents of the Society, . .116 Problem in Relationship. By Professor A. Macfarlane, D.Sc., .116 On Transition-Resistance and Polarisation. By W. Peddie, Esq., B.Sc., . . . . . . . .118 The Change in the Thermoelectric Properties of Tin at its Melting Point. By Albert Campbell, Esq., B.A., . . . .125 On the Thermoelectric Properties of Iron. By Professor Tait, . 1 27 On the Constitution of Dielectrics. By W. Peddie, Esq., B.Sc., . 129 On Mr Omond’s Observations of Fog-Bows. By Professor Tait, . 129 Letter from the Astronomer-Royal on the Fall of a Part of the Cliff below Nelson’s Monument, . . . . . .130 On the Causes of Movements in General Prices. By Professor Nicholson, . . . . . . . .130 A New Method for Preserving the Blood in a Fluid State outside the Body. By Professor J ohn Berry Haycraft and E. W. Carlier, M.B., ' 130 The Formula of Morphine. By R. B. Dott, Esq., F.I.C., and Ralph Stockman, M.D., ....... 131 On the Fossil Flora of the Staffordshire Coal Fields. I. The Fossil Plants collected during the Sinking of the Shaft of the Hamstead Colliery, Great Barr, near Birmingham. By Robert Kidston, Esq., F.R.S.E., F.G.S., 135 On a Monochromatic Rainbow. By John Aitken, Esq., . . 135 On Neuropteris plicata, Sternb., and Neuropteris rectmervis, Kidston. By Robert Kidston, F.R.S.E., F.G.S., . . . .137 Reflex Spinal Scratching Movements in some Vertebrates. By Prof. John Berry Haycraft, . . . . . .137 Reply to Professor Boltzmann. By Professor Tait, . . .140 On a Mode of Exhibiting the Action of the Semicircular Canals of the Internal Ear. By Professor Crum Brown, . . .149 On the Temperature and Currents in the Lochs of the West of Scot- land, as affected by Winds. By Dr John Murray, . . 151 Contents. V PAGE ISTote on the Influence of Pressure on the Solubility of Carbonate of Lime in Sea Water containing Free Carbonic Acid. By W. G. Eeid, Esq. Communicated by Dr John Murray, . . .151 On the Distribution of Carbonate of Lime on the Floor, and in the Waters of the Ocean. By Dr John Murray, . . . 158 On the Number of Dust Particles in the Atmosphere. By John Aitken, Esq., ....... 158 Preliminary Note on the Duration of Impact. By Professor Tait, . 159 A Bathymetrical Survey of the Chief Perthshire Lochs, and their relation to the Glaciation of the District. By James Grant Wilson, Esq., H.M. Geological Survey. Communicated by Dr Archibald Geikie, F.R.S., Director-General of the Geological Survey, . 159 Contact-Phenomena of some Scottish Olivine-Diabases. By Ernst Stecher,- Ph.D., Leipzig. Communicated by Dr Archibald Geikie, F.B.S., Director-General of the Geological Survey, . . 160 Experimental Kesearches in Mountain Building. By Henry M. Cadell, Esq. of Grange, B.Sc., F.R.S.E., of H.M. Geological Sur- vey of Scotland, ....... 172 Observations on the Movements of the Entire Detached Animal, and of Detached Ciliated Parts of Bivalve Molluscs, viz,. Gills, Mantle- Lobes, Labial Palps, and Foot. By D. M‘Alpine, Esq., F.C.S. Communicated by W. E. Hoyle, Esq., M.A. (Plates I., II.), . 173 , Keport on the Fishes obtained by Dr John Murray in Deep Water on the North-West Coast of Scotland, between April 1887 and March 1888. By Dr A. Gunther, F.R.S., Keeper of the Zoological Department, British Museum. (Plates III., IV.), . . . 205 Morphological Changes that occur in the Human Blood during Coagulation. By Professor John Berry Hay craft and E. W. Carlier, Esq., M.B., ....... 220 On the Mean Free Path, and the average Number of Collisions per particle per second in a Group of Equal Spheres. By Prof. Tait, 225 Note on the Compressibility of Glass at different Temperatures. By Professor Tait, . . . . . . . 226 Exhibition of Photographs of the Lunar Eclipse of 28th January 1888. By W. Peck, Esq., F.B.A.S., .... 226 Illegitimacy in the Parish of Marnoch. By George Seton, Esq., M.A. Oxon., ....... 227 Notes on the Use of Mercuric Salts in Solution as Antiseptic Surgical Lotions. By G. Sims Woodhead, M.D., F.R.C.P. Edin., 235 The Effect of Differential Mass-Motion on the Permeability of a Gas. By Professor Tait, ....... 249 On a New Diffusiometer and other Apparatus for Liquid Diffusion. Part II. By J. J. Coleman, Esq., F.R.S.E., F.I.C., F.C.S., . 249 Note on the Determination of Diffusivity in Absolute Measure from Mr Coleman’s Experiments. By the President, . . . 256 VI Contents. PAGE On the Soaring of Birds : being an Extract from a Letter of the late Mr William Fronde to Sir W. Thomson, of 5th February 1878, received after Mr Fronde’s death, ..... 256 Preliminary Note on New Determinations of the Electric Resistance of Liquids. By W. Peddie, Esq., B.Sc., .... 259 Notice of the Recent Earthquake in Scotland, with Observations on those since 1882. By Charles A. Stevenson, Esq., B.Sc., Assoc. M.Inst.C.E. (Plate V.), ...... 259 Analysis of the “ Challenger ” Meteorological Observations. By Dr Buchan, ........ 267 On the Conjugated Sulphates of the Copper-Magnesium Group. By Prafulla Chandra Ray, Esq., ..... 267 On the Chemical Composition of the Water composing the Clyde Sea Area. Part II. By Adam Dickie, Esq., . . . 283 On the Principles of Animal Morphology. By Professor Wilhelm His of Leipzig. Letter to Dr John Murray, H.P.R.S.Ed. Com- municated by Professor Sir William Turner, . . . 287 Mathematical Notes. By Professor Tait, .... 298 Analysis of the “ Challenger ” Meteorological Observations. By Dr Buchan, . . ...... 298 Description of the Rocks of the Island of Malta, and Comparison with Deep-Sea Deposits. By Dr John Murray, . . . 298 An Electrical Method of Reversing Deep-Sea Thermometers. By Professor Chrystal, ....... 298 On a Class of Alternants expressible in terms of Simple Alternants. By Thomas Muir, LL.D., ...... 298 Quaternion Note. By Professor Tait, .... 308 Exhibition of Specimens of Eggs of Cod and Haddock. By Dr John Murray, ........ 308 On the Secretion of Lime by Animals. By Robert Irvine, Esq., F.R.S.E., F.C.S., and Dr G. Sims Woodhead, . . . 308 On the Solubility of Carbonate of Lime under different forms in Sea- Water. By Robert Irvine, Esq., F.R.S.E., and George Young, Esq., 316 On a Case of Absence of the Corpus Callosum in the Human Brain. By Dr Alexander Bruce. (Plates VI.-XIII.), . . . 320 Distribution of some Marine Animals on the West Coast of Scotland. By Dr John Murray, . . . . . .341 Remarks on the Larvae of certain Schizopodous Crustacea from the Firth of Clyde. By W. E. Hoyle, Esq., M.A., . . . 341 Exhibition of Photographs of the Nice Observatory. By the As- tronomer-Royal for Scotland, ..... 342 Note on the Hydrodynamical Equations. By Professor Cayley, Hon. F.R.S.E., . . . . . . .342 The History of Volcanic Action during the Tertiary Period in the British Islands. By Dr Archibald Geikie, F.R.S., . . 344 Contents. Vll PAGE Exhibition of Photographs of large Sections of the Lnng. By Dr G. Sims Woodhead, ....... 348 Mean Scottish Meteorology for the last Thirty-Two Years, discussed for Annual Cycles, as well as Supra-annual Solar Influences, on the basis of the Observations of the Scottish Meteorological Society, as furnished to, and published by, the Kegistrar- General of Births, Deaths, &c., in Scotland, after being computed for that Office at the Eoyal Observatory, Edinburgh. By the As.tronomer-Boyal for Scotland, ....... 348 On Sir J ohn Leslie’s Computation of the Eatio of the Diameter of a Circle to its Circumference. By Edward Sang, LL.D., . . 348 On the Four Surfaces of an Aplanatic Objective. By the Hon. Lord McLaren. (Plates XIV., XV.), ..... 355 Quaternion Notes. By Professor Tait, . . . .379 Exhibition of M. Amagat’s Photographs of the Crystallisation of the Tetrachloride of Carbon under Pressure alone. By the Secre- tary, . . . . . . . . .381 On the Development and Life-Histories of the Food and other Fishes. By Professor W. Carmichael MTntosh, F.E.S., and E. E. Prince, Esq., . . . . . . . .381 On certain Theorems mainly connected with Alternants (IL). By Professor Anglin, M.E.I.A., ..... 381 Preliminary Note on a Method by means of which the Alkalinity of the Blood may quantitatively be Determined. By Professor J ohn Berry Haycraft and Dr E. T. Williamson, .... 396 The Electrolytic Decomposition of Proteid Substances. By George N. Stewart, Esq., ....... 399 On the Malpighian Tubules of Lihellula depressa. By Dr A. B. Griffiths, F.E.S. (Edin.), F.C.S. (Bond. & Paris), . . . 401 On a Fungoid Disease in the Boots of Gucumis saliva. By Dr A. B. Griffiths, F.E,S. (Edin.), F.C.S. (Lond. & Paris). (Plate XVI.), .403 On Fossil Fishes from the Pumpherston Oil Shale, with Exhibition of Specimens. By Dr E. H. Traquair, F.E.S., . . . 410 On the Variation of Transition-Eesistance and Polarisation with Electromotive Force and Current Density. By W. Peddie, D.Sc. (Plate XVII.), 411 The Metamorphosis of British Euphausiidse. By George Brook, Esq., and W. E. Hoyle, M.A. (Oxon.), ..... 414 Notes on a Lucifer-like Decapod Larva from the West Coast of Scot- land. By George Brook, Esq., ..... 420 On Invertebrate Blood removed from the Vessels, and entirely sur- rounded by Oil. By Professor John Berry Haycraft and E. W. Carlier, Esq., M.B., . . . . . . . 423 On Laplace’s Theory of the Internal Pressure in Liquids. By Pro- fessor Tait, ........ 426 Vlll Contents. PAGE The Mechanism of the Separation of the Placenta and Membranes during Labour. By Dr Berry Hart, M.D., F.E.C.P.E. (Plates XVIIL, XIX.), 427 The Pathology of Cystic Ovary. By J. W. Martin, M.D. Com- municated by Dr Woodhead, ..... 435 Histological Observations on the Muscle Fibre and Connective Tissue of the Uterus during Pregnancy and the Puerperium. By T. A. Helme, Esq., M.B. Communicated by Dr Woodhead, . 435 The Air in Coal-Mines. By T. G. Nasmyth, M.B., D.Sc., . . 435 Obituary Notice of the late Robert Gray, Esq., Vice-President. By Dr R. H. Traquair, F.R.S., . . . . . 435 On some Relations between Magnetism and Twist in Iron and Nickel. By Cargill G. Knott, D.Sc., ..... 435 On the Fossil Plants in the Ravenhead Collection in the Liverpool Museum. By R. Kidston, Esq., ..... 436 On the Action of Carbonic Acid Water on Olivine. By Alexander Johnstone, Esq., F.G.S., ...... 436 Is the Law of Talbot true for very rapidly Intermittent Light? By George N. Stewart, Esq., . . . . . . 441 On the Specific Gravity of the Water in the Firth of Forth and the Clyde Sea Area. By Hugh Robert Mill, D.Sc., Scottish Marine Station, ........ 465 Arrested Twin Development. By Macdonald Brown, Esq., M.B., . 465 On Numerical Solution of Equations in Variables of the nth Degree. By the Hon. Lord McLaren, ..... 467 Alternants which are the constant Multiples of the Difference- Product of the Variables. By Prof. Anglin, M.A., LL.D., &c., . 468 Chairman’s Closing Remarks, ...... 476 The Theory of Determinants in the Historical Order of its Develop- ment. By Thomas Muir, M.A., LL.D., .... 481 Donations to the Library, ...... 545 Index, ......... 547 PROCEEDINGS OF THE ROYAL SOCIETY OF EDINBURGH. VOL. XV. 1887-88. No. 126. The 105th Session. GENEKAL STATUTOEY MEETING. Monday, 28/^A November 1887. The following Council were elected : — President. Sir william THOMSON, F.R.S. Vice-Presidents. David Milne Home, Esq. of Milne- Graden, LL.D. John Murray, Esq., Ph.D. Professor Sir Douglas Maclagan. The Hon. Lord Maclaren, LL.D. Rev. Professor Flint, D.D. Professor Chrystal, LL.D. General Secretary — Professor Tait. Secretaries to Ordinary Meetings. Professor Sir W. Turner, F. R. S. Professor Crum Brown, F.R.S. Treasurer — Adam Gillies Smith, Esq., C.A. Curator of Library and Museum — Alexander Buchan, Esq., M.A., LL.D. Ordinary Members of Council. Professor Butcher, LL.D. Professor M ‘Kendrick, F.R.S. Thomas Muir, Esq., LL.D. Professor M‘Intosh, F.R.S. Sir Arthur Mitchell, K.C.B., LL. Stair Agnew, Esq., C.B. R. M. Ferguson, Esq., Ph.D. A. Forbes Irvine, Esq. of Drum, LL.D. Dr J. Batty-Tuke, F.R.C.P. Professor Bower. Dr G. Sims Woodhead, F.R.C.P. Robert Cox, Esq. of Gorgie, M.A. By a Resolution of the Society (19th January 1880), the following Hon. Vice-Presidents, having filled the office of President, are also Members of the Council : — His Grace the DUKE of ARGYLL, K.T., D.C.L. The Right Hon. LORD MONCREIFF of Tulliebole, LL.D. VOL. XV. 8/5/88. A 2 Proceedings of Boyal Society of Edinburgh. [dec. 5, Monday, hth December 1887. The Hok Lord M'LAEElSr, Vice-President, in the Chair. 1. Chairman’s Opening Address. In commencing the business of a new Session of the Eoyal Society, it is natural to refer to the work of the Society in the year which has been completed. A Society such as this is, constituted for the promotion of literary and scientific research, works in two ways, and first and chiefly by the reading and publication of papers either extending the boundaries of scientific acquirement, or record- ing the finished results of observation and experiment which are the foundations of theoretical research. To this work only a limited number of our fellows are able to contribute, but it is to be hoped that such of our number as do not contribute to the Society’s publications may at least by their presence, and the interest which they manifest in the subjects of the papers, do something to encourage the more active members of the Society in the work of research. But, secondly, it is part of the proper work of this Society, by its organisation, its influence, and the expenditure of its funds, to aid the work of research in these departments, in which results cannot be achieved by the unaided exertions of individual members. I shall not attempt to enter into particulars regarding the progress of scientific research during the current year. To a proper estimate of its results it would be necessary that a report should be prepared by a combination of men who are themselves engaged in mathema- tical, physical, and biological investigation, and who are acquainted with the work done by specialists in their respective sciences. But it may be interesting to the members to hear something re- garding the business of the Society as an organisation working for public objects, and I shall endeavour to notice briefly the work in which the Council has been recently engaged. The condition of the Society itself, I am happy to say, is entirely satisfactory. Indeed, so far as I know, the only difficulty with which the Council has had to contend is the accumulation of literary material in our Library. Our shelves, as you see, are fiUed with books and 1887.] Chairman's Address. 3 transactions from floor to ceiling, and we have nearly come to the point when we must choose between the sacrifice of a part of our collection, and the alternative of providing additional library accommodation. The Council has been giving its attention to this subject, and I am confident that any practical suggestions which the Council may hereafter be able to bring before us will have the support of the Royal Society. You are aware that our library, I may say like everything that is undertaken by this Society, is distinctly specialised. The object of the Council has been to make it a complete collection of original memoirs on every scientific subject. It includes the transactions of every scientific society of repute throughout the world, and all that are really valuable among the scientific periodicals of our own and foreign countries. Members who are working on special subjects accordingly have at their disposal the original papers in which the discoveries connected with these subjects were communicated to the public, and where the details of investigations are more fully given than in the text-books in which the results of research are put to- gether. This is obviously the most useful kind of reference library for a Society such as ours ; and on the authority of our Librarian, I may state that our library is the most complete collection of the kind in the United Kingdom, more complete even than that of the Royal Society of London. In making this statement, I am not at all afraid that some section of the community will be seized with a desire to appropriate our collection for the purposes of a public library. I am inclined to think that our literary treasures will be found for the most part capable of exerting a remarkable force of repulsion upon any one who approaches them without due preparation. They are eminently useful to students of science, and not particularly interesting to the general reader. Such being its character, it is desirable that our library should be weU cared for, well arranged, and made accessible to members. The Council having found that it was impossible that the duties of an assistant librarian and assistant secretary could be performed single-handed, resolved last summer to engage a second assistant librarian, to take a share of the duties which had hitherto fallen to 4 Proceedings of Royal Society of Edinhurgli. [dec. 5, Mr Gordon, and I think that members have every reason to he satisfied with the manner in which these duties are discharged. One of the events of the year which calls for more than a passing notice, is the institution of the Victoria Jubilee Prize by Dr Gunning, a gentleman distinguished by his enlightened liberality, and who has given away a large part of his fortune to the enrich- ment of the institutions of this city. By the terms of the foundation, the prize is to be awarded triennially, and may be either for work done during the past three years, or for researches to be prosecuted during the ensuing triennial period on a subject approved by the Council. In calling the attention of the Society to the special terms of Dr Gunning’s most useful endowment, I note that this prize fund is, so far as I know, the only fund available for distribution in Scotland in aid of scientific research. There is a research fund, as the Bellows are aAvare, voted annually by Parliament, and placed at the disposal of a committee of the Eoyal Society of London ; but no share of this or any other public fund is placed at the disposal of the scientific profession in Scotland. Ireland has its special grant ; but Scotland has to trust to the liberality, or let me say the sense of justice, of the Royal Society of London in regard to such claims as she may prefer for the work of scientific research in Scotland. We know that when a demand comes from a Scotsman of established reputation in the scientific profession, it is fairly considered by the committee in London, and generally honoured. I need only refer to the sum so liberally voted at the last meeting of the London Committee in aid of Dr Murray’s submarine researches, and, indeed, it is not to be doubted that the eminent men who compose the Research Committee in London are anxious to make a fair distribution of the public money which is placed at their disposal. But we say that it is impossible in the nature of things that a committee in London should have the knowledge that is desirable of the attainments of the younger scientific men of Scotland who may be willing to engage in original work. The work proposed may be most important, and its prosecution may involve the purchase of costly apparatus, such as would be a proper charge on the research fund ; but how is the committee 1887.] Chairman's Address. 5 sitting in London to find out wfietlier the applicant is qualified by training and experience to prosecute the researches which he proposes to undertake h It is plain enough that a committee must to a large extent act on its personal knowledge of the applicants, and that the unknown applicant for a share of the research fund has a tolerable chance of being defeated in that struggle for existence which is illustrated by scientific competition. It is, therefore, very necessary that the distribution of research funds should be in some degree localised. Should any Fellow of this Society be disposed to carry on the work initiated by Dr Gunning, we can assure him that, as to any superfluous money, the possession of which may be useless or injurious to himself, he can- not employ it more usefully than by aiding in the establishment of a research fund, to be distributed by a committee of the scientific body in Scotland. We also invite the public of Scotland, and its representatives in Parliament, to support the Koyal Society of Edinburgh in its claim to have a share of the Government grant allocated to Scot- land for researches conducted within the country. I have now to allude to a cognate subject, which at the present moment possesses a special interest and importance : I refer to the Ben Nevis Meteorological Observatory. I remember to have seen somewhere an engraving after one of the old masters, in which Science is represented as a melancholy female form perched upon a hill and surrounded by circles, instru- ments of alchemy, and other quaint devices. I think that the mediseval artist must have had what we call a forecast of those mountain establishments which have sprung up in different parts of the world, and among which the Observatory at Ben Nevis is the most famed for its admirable situation, and the high character of the work which is there performed. Its hourly observations, taken without intermission during a period of three years by Mr Omond and his assistants, have now been reduced and tabulated. They are in print, and will shortly be issued as an extra number of the Transactions of this Society. It is unnecessary to point out that such a series of observations must be of the highest value for general purposes, such as the deter- mining the constants of atmospheric pressure and temperature in 6 Proceedings of Royal Society of Edinburgh. [dec. 5, relation to height. But this is not all. I believe it will he found, when these observations are discussed, that the variations of atmo- spheric pressure at this elevated station, uninfluenced by local causes, can be reduced to something approaching to a general law, and that the theory of storms or waves of depression and elevation in the atmosphere will receive valuable elucidation from the results of this new departure in meteorological observation. Up to this time the Observatory at Ben Nevis has been supported exclusively by the voluntary contributions of persons interested in scientific research. Out of such contributions the Observatory has been built and equipped, and the salaries (inadequate as they are for such service) have been provided. There is reason to hope that, either by direct grant from the Government, or by the assignment of a part of the sum which is annually voted by Parliament for distribution by the English Meteorological Council, a part at least of our future expenditure may he met. The Fellows will agree with me in congratulating Mr Buchan, and the Scottish Meteorological Society with which he is connected, on his nomination to the vacant seat in the English Meteorological Council. I think that Mr Buchan’s appointment may he regarded as a recognition by very high authority of the value of the Scottish meteorological work ; and it is to he hoped that our representative (if we may so consider him) will be able to give such information to his colleagues as will eventually lead to the establishment at Ben Nevis being made independent of personal contributions. At the same time, it is necessary to add that the establishment is incomplete. A new room is wanted at the Observatory for the seismometers, and a sea-level station at Fort William is necessary for the purpose of proper comparison observations. The Scottish Meteorological Society has undertaken to provide these buildings on condition of being relieved of the future maintenance of the Ob- servatory ; and it is hoped that this expense may be in part met by a grant out of the surplus funds of the Edinburgh Exhibition, if the members of that Association shall see fit to accede to the application that has been addressed to them. I am unwilling to bring these remarks to a close without making reference to a kindred institution, which has even a stronger claim than Ben Nevis on the liberality of the public Exchequer. The 1887.] Chairman’s Address. 7 Edinburgh Astronomical Observatory is a Government edifice, taken over by Government from a private society under an obligation to maintain it in perpetuity. It is understood that, in what relates to original scientific work, the operations of the Observatory have been to a considerable extent suspended in consequence of the want of the necessary instrumental means. The old instruments are worn out, and the new equatorial is incomplete. The English Govern- ment, which in matters relating to science acts spasmodically, and only under pressure, some ten years ago paid a large sum of money for the purchase of a handsome equatorial telescope for the Observa- tory at Edinburgh. During these ten years they have allowed their instrument to remain unproductive, apparently because they would not advance the few hundred pounds that are necessary to adapt the instrument to the special requirements of the work which it is to perform. I shall not enter into further detail in this matter, because the Astronomer-Royal is to read a paper this evening on the Edinburgh Equatorial in 1887. But if the Edinburgh Observatory is to be kept up as a national institution, — and I trust it will remain such in accordance with the agreement between the founders and the Government of the day, — it must be adequately endowed and properly equipped. After the Royal Society has finished the work it has taken in hand in re- lation to Ben Nevis, I hope you will take up the subject of the Edinburgh Observatory, and that you will not let the matter rest until the Observatory has been restored to the position of a first- class observing station. It may be thought by some that it is not desirable to maintain at the public expense two such establishments — the one in London and the other in Edinburgh. Nor would it be desirable, if the two establishments were to be employed in doing the same kind of work. But there is a vast amount of skilled work to be done in Sidereal Astronomy, which can only be done in a public observatory, where, as a matter of duty, assistants are in attendance every night. The researches to be undertaken here need not be of the same description as the work undertaken at Greenwich. In proof of this assertion, I will only mention what is being done at the present time in the Government Observatory at the Cape, by that accomplished astronomer and indefatigable worker. Dr Gill. 8 Proceedings of Royal Society of Edinhurgh. [dec. 5, First, one of the assistants has been employed during the last three years on a photographic survey of the southern heavens : not a survey on the colossal scale recommended by the Paris Conference, hut on a more manageable and perhaps more useful scale — a scale sufficient for the requirements of the practical astronomer, and containing more stars than are contained in the maps of Argelander, which, as you know, are the best existing maps of the northern heavens. Each photographic picture of the Cape series covers an area of four square degrees, and about a half of the work is already done. The places of the stars arebeingVeduced and tabulated by microscopic measurement. Secondly, the Cape Observatory is engaged in a redetermination of the right ascensions of fundamental stars, by means of circle observations taken in the prime vertical ; that is, in the great circle of the heavens which passes from west to east through the zenith. Hitherto right ascensions have been determined mainly by means of time observations, and their places are affected by errors due to the small irregularities of clocks, but still more by errors arising from the impossibility of estimating the one-tenth of a second by eye and ear. Under the Cape series of determinations the right ascensions of all the stars which do not set between south and west will be determined by circle readings with the same degree of precision as polar distances are now determined. Thirdly, the Cape Observatory, as I learn by a communication from its director, has commenced a systematic investigation of the parallax of the southern stars. I should like to say something about the instrument — a masterpiece of mechanical and optical skill — with which these delicate measures are being made. But this would be wandering too far from my subject. What I wanted to say is, that we are not at the present time, either in Greenwich or elsewhere, doing, for the northern half, any one of these three things which three of the chief instruments of the Cape Observatory are employed in doing for the southern half of the celestial sphere. You see there is scope for division of labour in astronomy as in other things. It is really not doubtful that, with a proper equip- ment, our Observatory could be employed on work such as can only be done at a national observatory, under official rules and discipline, which should not in the least compete with the work done at the National Observatory of Greenwich. With the greatest respect for 1887.] Chairman's Address. 9 the views of professional gentlemen near me, I venture to think that the Edinburgh Observatory could be made more useful by being maintained in its present position than by being made over to the University of Edinburgh, according to the Government pro- posals of last session. But whether it is to.be kept up as a national or as an academic institution, let us urge on the Government the duty of making it efficient, by providing the astronomer wuth the necessary instrumental appliances. I have now to open the business of the Session by presenting the Victoria Jubilee Prize to our President, Sir William Thomson. 2. The Jubilee Prize. The Chairman, on presenting the Victoria Jubilee Prize to Sir William Thomson, said : — The Council of this Society, at their meeting on the 15 th of July last, approved of the following Eeport by their Committee : — The Committee appointed to recommend the first award of the Victoria Jubilee Prize, having taken into consideration the terms of the foundation of the Victoria Jubilee Prize, were of opinion that the Council, in making their first award, ought to give the prize for work already done, especially as the Transactions and Proceedings of the Society contain evidence of a large amount of valuable scientific work contributed by gentlemen connected with Scotland during the past three years. Having gone over the list of papers during that period, and con- sidered suggestions regarding the experimental work which had been in progress during the prescribed period, the Committee resolved to recommend to the Council that the prize should be awarded to Professor Sir William Thomson, P.B.S.E., for a re- markable series of papers on Hydrokinetics, especially on Waves and Vortices, which have been for some time, and are still being communicated to the Society. Sir William Thomson has worked at Hydrokinetics almost since he began to publish. Some remarkable papers of his on the Bounding Surface, and on the Vis-Viva of a moving liquid, appeared in the Cambridge and Dublin Mathematical Journal. 10 Proceedings of Boy al Society of Bdinhuryh. [dec. 5, Helmlioltz’s great discovery of the properties of Vortex Motion attracted his special attention in 1867, and led him to the celebrated hypothesis of Vortex Atoms. His great paper on Vortices appeared in our Transactions, and several lesser, hut important, ones in our Proceedings: — such as On Maximum Energy in Vortex Motion. On the Production of a Coreless Vortex. Hext we have his explanation of the apparent attractions and repulsions exerted by bodies vibrating in a fluid. The effect of wind in raising waves. Propagation of ripjdes by surface tension. Motion of solids (with or without perforations) in a perfect liquid, when the motion is irrotational. Stationary waves in running water. Ping-waves produced by a single impulse. Stability of Fluid Motion. Laminar motion in a turbulently moving perfect liquid. These form a collection of most important, and entirely novel, contributions to Hydrokinetics, which will bear comparison with the very best work ever done on the subject. It is understood that a great part of this work has arisen indi- rectly from Sir W. Thomson’s investigations as to the mechanism of the propagation of light, which have been given in outline in the Papyrograph of his Baltimore Lectures. 3. Proposed Additions to the List of Honorary Fellows. The Chairman, in accordance with Law^ XII., read the following list : — Ernest Haeckel, Professor of Zoology and Histology in the Uni- versity of Jena. Pudolph Julius Emmanuel Clausius, Professor of Natural Philo- sophy in the University of Bonn. Demetrius Ivanovich Mendeleeff, Professor of Chemistry in the University of St Petersburg. The following Communications were read : — 1887.] C. Piazzi Smyth on the Edinhurgh Eguatorial. 11 4. The Edinburgh Equatorial in 1887 ; a Paper with two Appendices. By C. Piazzi Smyth, Astronomer-Royal for Scotland. The Equatorial of the Eoyal Observatory, Edinhurgh, is still in October 1887, unfinished, blocked against use, and entirely unusable. This lamentable outcome of so many years, is simply the con- sequence of the necessary funds for finishing and working the instrument having been withheld by Government, after they had been promised by the Board of Visitors, printed again and again in much detail, and thought to have been obtained. The beginning, or first supplying, of the instrument had been carried out by H.M. Office of Works in London, on the strength of a special grant by Parliament ; and the Astronomer knows simply nothing about the money part of that proceeding. He has merely to do with what results it accomplished in its day; — and seeing that these are now freely confessed by the authorities actually concerned to be imperfect, and left standing, as well as locked and blocked, in that condition — to set forth from his point of view how, — assuming the working funds, as well as liberty to act, were to be granted to him, according to the original suggestion of the Board of Visitors, — he would set about the finishing of the instru- ment suitably with both its long ago accomplished beginning, the very peculiar artificial and natural restrictions of the situation, and the high nature of the stellar and spectroscopic observations required, and intended, to be made. Taking the instrument therefore as it is now, or even with such minor improvements as a Government Commission (without the Astronomer upon it) recommended in 1879, and even obtained a grant for their execution by the Office of Works (though they have not been executed yet) — there is no doubt that if any experienced practical astronomer were to endeavour at this time to use the instrument, he would speedily arrive, amongst various other defects, at the following four accusations in chief, viz. : — 1st. The instrument, though not in the slightest degree too large for its intended work, and far smaller than many telescopes else- where, is yet too bulky for its Dome; and that has the largest 12 Proceedings of Royal Society of IJdinbiiryTi. [dec. 5, possible size allowed by the distinguished architectural adviser applied to by Government on the occasion of ordering it. 2nd. The instrument is too heavy and too severe in pressure for its bearings, compatibly with its quick and slow motions, and more especially for its delicate clock-movement. 3rd. The instrument is too awkward and multi-local as to its eye-pieces, handles, cords, finders, ladders, &c. &c. ; and the observer far too much exposed in strained positions to the violence of the wind and intensity of the cold, to be likely to resist their influence long, or make very good observations at any time. 4th. The instrument is too weak in its spectroscope; and the latter too barbarous in its appliances, so far as they have yet been carried out. To meet these evils the Astronomer suggests as follows ; viz. — for Group No. 1, he proposes to shorten both the length of the telescope by its revolving head, and the length of its Declination axis by its outer 1 4 inches of excess far beyond its bearings ; besides stripping off the great outside finder, the small outside finder, the long reading microscopes and a variety of other untoward excrescences ; appropriate sub-arrangements being introduced to render these changes not only compatible with efficiency, but much more efficient, quick and handy. For Group No. 2 he proposes to remove, with the revolving head, both the weights and the counter-weights of the spectroscope, heavy eye-piece plate, and second finder now at the upper long end of the tube ; also the double counterpoise weights thereof at the lower short end of the tube ; and then the triple counterpoise weights of the same at the end of the Declination axis, — thereby getting rid at once of more than 500 lbs. of dead weight, pressing at present with pernicious effect on the lower end of the Polar Axis, which is too small to bear much. For Group No. 3, the Astronomer proposes, by a very simple yet radical change of eye-end arrangement, to have the eye-pieces of telescope, spectroscope and new finder, together with the slow- motion handles in P.A. and Deck, brought to, and arranged round, the end of the Declination Axis as already shortened ; and where they will always be directly accessible to the observer at easy stand- ing height on the floor and never exposed under the open shutter. 1887.] C. Piazzi Smyth on the Edinhurgh Equatorial. 13 Also to exchange the present inefficient, yet cumbersome travel- ling and elevating platform and ladders, for a neat, compact, well seated and. suitably fitted Observer’s travelling hut, — freely tra- versing around, or to and from its generally proper and fully sheltered position at the end of the Declination axis, in any and every position of the telescope for observation arranged on the principles noted on the next page. And for Group No, 4, — he finally proposes to construct a new and grand Spectroscope with two sets of prisms (after the manner of that which he made for himself in 1882, and therewith discovered the exquisite spectral progression of Carbonic oxide, as well as the com- pound triples of pure Oxygen, in gas vacuum tubes) occupying in one plane and chiefly in a diagonal direction therein, all the hitherto unoccupied length between the upper or small-mirror end of the telescope tube, and the outer end of the Declination axis (as shortened) ; thereby balancing in itself, across the Polar axis, the heavy telescope tube, and its very heavy lower, or great-mirror, end ; and allowing an equivalent of dead weight to be taken off the Declination Axis. While he proposes also to utihse the whole length of the telescope, and the axial dark space necessarily running up through it as a Newtonian (or in this case a semi-Newtonian) reflector, — first in the part above the small diagonal mirror, for the objective of a large centrally placed finder to the telescope, always looking fully out of the opened shutter, whenever the telescope itself does (while it sends its cone of rays, by a diagonal mirror of its own, down to the end of the declination axis) ; and then below it, for the collection of rays for an end-on, gas-vacuum and electric lighted tube of his own invention, to form the reference spectrum for stars, in a manner more unexceptionable it is believed and more promising for accuracy than any aiTangement yet in use elsewhere. The observations being always as a rule, — and a rule most essential in the midst of a great and smoky city, growing greater and smokier day by day, — confined to as near the Meridian, and to as high an altitude therein, as possible. And now, as I believe that the above suggestions, worked out already to sufficient extent on paper, meet all the difficulties yet found with even superfluous force, I would try to call attention to 14 Proceedings of Boyal Bociety of Edinburgh. [dec. 5, how remarkably the whole of them, as methods of amelioration for the long considered impossibilities of this Edinburgh Equatorial, which Governments and governing Boards and Central Committees have so totally failed in through more than a dozen years, — flow from, or are bound up with, the one beautifully simple idea, of trans- ferring the eye-piece from the upper end of the telescope, to near the opposite or outer end of the Declination Axis, by altering the angle only, of the small mirror, and not introducing any additional reflection. It is this change which at once renders the Dome quite large enough for the telescope ; which relieves the instrument of the immense amount of dead weight it was found unable to carry ; which gives the observer a sheltered position to observe in, and his assistant plenty of room for working, either on one side or the other ; while it also enables the modern science of spectroscopy to take up its position with power and dignity, in greater space than ever allowed to a star spectroscope before. And where did this simple, yet all powerful and most suitable idea come from 1 It was not with me, at any of the consultations over the instru- ment I assisted at many years ago. I have never heard it hinted at by any one else. Yet here it is now, because it very lately came to me. And came, I know not how, unless as a gracious gift from above, and at a moment of dire extremity, from the Giver of all Good. Wherefore, if in this very advanced Christian age of the world, I were to hesitate for a moment, between misleadingly allowing the public to give the credit to me ; or, on the other hand, attributing it myself frankly and thankfully to God, to whom it is alone due, I should deserve, like another person, well known by name, “ to be eaten of worms and to give up the ghost”; instead of having been thus graciously preserved, through more than one generation of University Professors, up to the present moment;* for further work, may it be, in elucidating the glory of the sidereal creations of the Divine Architect of all things. Appendix I. Memoranda of smaller and local practical matters, considered long ago as necessary to be attended to whenever a practical beginning * October 1887. 1887.] C. Piazzi Smyth on the Eclinlurgh Equatorial. 15 shall be made of what is described in the previous pages, and by whomsoever the work may be carried out. (1) The present very thin flooring of the Dome will have to be propped up, substantially though temporarily only, before any heavy repairs begin. (2) The large timber blocking of Declination axis to be exchanged for a small, compact, iron apparatus, or say merely a steel rod, passing by two small holes through a thick part of the Polar axis, but leaving its ends standing out sufficiently to butt against either one, or other side of the great cradle-frame of said axis ; thereby serving the full purpose of the timber blocking, but without occu- pying one-hundredth part so much space, or fixing the Dome, or interfering with movements of workmen about the instrument. (3) Eevolving Telescope head, with all its weights and attach- ments to be removed ; Telescope tube to be raised 5 inches in its collars (from great mirror end, as assumed below, to eye-end above) ; great mirror to be restored, and residual balance at end of Declina tion axis (after being shortened 14 inches) to be practically ascer- tained as speedily as possible ; all superfluous weights being carried out of the Dome, noting how much. (4) The shutter of the Dome to be altered from present plan of pivoting in the Zenith, to Messrs Cook’s new plan of pivoting on the opposite side of base. The permanent opening of Dome being then taken right up to and 4 feet past the Zenith, with a breadth of nowhere less than 30 inches. A revolving ventilator of much larger diameter than the present one being then attached to the Zenith of the shutter, and three bull’s-eyes illuminators introduced beneath Zenith and horizon. A separate fence of sheet steel about 3 feet high being made to slide horizontally across lower part of Dome’s opening, to keep out the violence of the wind when not observing at very low altitudes ; and the late Mr Grubb’s advice of lining the iron dome, with non-heat-conducting wood, to be no longer delayed. (5) All the air-passages in and around the pier in both Dome- room and under-Dome-room should be caulked with elastic tow, for otherwise both these rooms become a chimney of draughts to all the rest of the Observatory below their level ; and when shutters are opened there, the draughts up into the Dome are very severe. 16 Proceedings of Royal Society of Edinburgh. [dec. 5, (6) The hot-water pipe system, heated from the Computing-room gas-stove, should he exchanged from the under-Dome-room, into the Laboratory, and even have its chief development there ; with revolving cowls in place of simple ones on the roof. Incandescent electric lighting by gas-engine and dynamo to be introduced at all the instruments, because capable of giving light, without oxidizing gases, and without sensible and most pernicious heating. (7) The front projection and elevating part of the so-called Observing chair should be at once removed, together with whatever else may prevent the mere back and wheels of the chair traversing freely both round and past, or from one side to the other of the Declination-axis end, shortened as above, preliminary to the said parts being converted into the new observer’s travelling hut. (8) All axles, bearings, &c., of both Dome, Equatorial, and Clock movement should, without any further loss of time, be taken out of their sockets, well cleaned and re-lubricated ; or mischief may take place amongst them. (9) Both the original proposition of a cylinder chronograph, as well as that for a duplicate speculum for the telescope, should he realised. (10) An engineering opinion to he obtained as to the residual strength of the neck of the Declination axis, considering the holes cut into it by its maker, but not required on the Astronomer’s herein proposed method of working. (11) The toothed spur wheels of Lift to be replaced by endless- screw wheels, as being safer from accidental “ stripping of teeth ” under heavy loads. (12) All the recommendations of the Government Committee of 1879, except such as the Astronomer may agree to dispense with, should be fully and faithfully carried out, before the new works treated of in the first part of this paper are commenced upon. (13) And if by that time Government may have decided to rebuild the Eoyal Observatory, Edinburgh, on a better site, in a more modern manner and supply it with new instruments, as recom- mended by their Commission in 1876, taking however the Equa- torial with them to a larger Dome, — then aU the said new works should still be carried on there, omitting only the shortening of the Declination Axis, which will be better to be kept of its present full 17 1887.] C. Piazzi Smyth on the Edinhurgh Equatorial. length ; eye-pieces, and slow motion handles being equally ex- tended. Appendix II. The Financial Kequirements and Difficulty. Before the Astronomer consented to join in the Board of Visitors’ project, about 1870, of applying to Government for a large Equatorial, he pointed out that such an instrument, even if once set up complete, would require further expenditure year after year to keep it fully efficient. And that the working with it would be so peculiarly onerous and responsible, that the salaries of the officers of the Royal Observatory, Edinburgh, already acknowledged to be at, or below, starvation point, should be raised more nearly to the level of those of other Observatories, or of any ordinary Government offices. He was told in answer that all that was most certainly right, and would be brought about ; while the Board of Visitors — whom Government had appointed years before expressly to advise them on such matters, and how to keep up the Observatory thereby in all future time as “ a proper Royal Observatory,” — did most honour- ably proceed to frame a scheme of modest improvement not only to the Observers’ salaries, but to the available income of the Observa- tory, to be expended by the Astronomer in instrumental repairs, experiments and improvements at his discretion. Under these promising circumstances the Astronomer joined the application of the Board for the large Equatorial. That instrument was accordingly allowed by Government in 1871, was in part set up, under the authority of the Office of Works in London in 1872 ; and in the following year, when the erection was found very incomplete, the scheme of the Board of Visitors for increasing the salaries and available income of the Observatory to a point suffi- cient to finish, maintain, and work the instrument — for a long time not unfavourably entertained by Government, — was suddenly and finally disallowed. The Board of Visitors indeed continued to apply to Government, as represented by the Home Office, until in 1876 the then Home Secretary, Mr, since Sir Richard, now Lord Cross, adopted the following expedient for escaping from the terms of agreement under VOL. XV. 6/6/88 B 18 Proceedings of Boycd Society of EdinlurgJi. [dec. 5, whicli tlie Astronomical Institute of Edinburgh had parted with their Observatory to Government nearly thirty years previously. That is, declining to listen to the long time accredited Board of Visitors, he appointed autocratically a Committee of his own to come down from London, and examine and report on the case. That Committee accordingly arrived in July of 1876, examined at the Observatory, sat and discussed in Queen Street, and then reported for a series of financial improvements of a similar, though altered character to those of the Board of Visitors, because including a rebuilding of the Observatory in a modern manner and on a new site. But the Home Secretary thereupon declined to listen to his own Committee, and neglected all their recommendations, as well as those of the older Board of Visitors. The venerable Mr Duncan MDaren, then Senior M.P. for Edin- burgh, moved thereupon in Parliament to have the Committee’s Keport publicly printed, which was done in 1877. Still however nothing came of it until 1879, when on account of further repre- sentations by the same watchful guardian of Scottish interests, the Home Secretary found it expedient to send another of his Commit- tees to examine and report again. Confining itself however this time to the Equatorial, and without admitting the Astronomer to their Council, this Committee advised certain improvements, obtained a grant for executing them, and handed it over to the OflS.ce of Works, where it is believed either to remain still, or to have lapsed to the Treasury after doing little or nothing at the instrument. This result however is perhaps not very much to be regretted, because the sum was not only absurdly insufficient to go through with all that was, and still more is, required for eflSciency in the mere inorganic instrument, — but the previously admitted starvation of the Observatory in all its oflSces and its various means of doing good after its kind, was left absolutely untouched, and prevails to a degree of intensity, that were it on a larger scale, or nearer London, might, in its crying injustice, excite severe public animadversion, with questions as to the propriety of Home Kule being the only way to obtain justice for Scotland. To compare the case in round numbers with another Koyal Observatory nearer London headquarters, viz., that at Greenwich, 1887.] C. Piazzi Smyth on the Edinhurgh Equatorial. 19 the following contrast comes out, in so far as I may have the cor- rect figures : — (1) Expended on Green- wich. Equatorial and Dome between 1856 and 1879 under the direction of the Astrono- mer-Royal, it is believed . £14,000 (2) Annual budget of Royal Observatory, Green- wich, 7000 (3) Salary of Chief Assist- ant at Royal Observatory, Greenwich, .... 600 (4) New Objective and Dome for Greenwich Equa- torial, 1885-86, perhaps . 5000 But as every one, including myself most heartily, allows that the Greenwich Observatory is one of the best managed, most economical and most efficient Observatories and Government establishments in the world, — the result of the comparison is simply to show that the Edinburgh Observatory is most seriously underpaid. Or to compare in detail the requisite remuneration of one worker in the Edinburgh Observatory, with what is allowed to be the cor- rect thing in any ordinary Government Civil Office, please to com- pare the two following paragraphs which were printed in the public newspapers of this country on the same day. (1) Scotsman bTewspaper, December 6, 1880. “An open competition is to be held in January 1881 simul- taneously in London, Edinburgh, and Dublin for the post of Second Assistant Astronomer at the Eoyal Observatory at Edinburgh. Salary £100 per ann!^ (Tempered however in fact with an extraneous and temporary addition rising by £10 per ann. to £50 per ann., so long as certain extra work is carried on.) (2) Daily Telegraph Newspaper, December 6, 1880. “ An open competition will shortly be held for two Junior Clerk- ships in the Colonial Office, with salaries commencing at £250 per ann. and rising to £600. Five of the Junior Clerks have additional emoluments. The higher clerkships, with salaries from £700 to £1000, are filled by promotion from the junior class.” (!') Expended on Edin- burgh Equatorial and Dome, under the control of the Office of Works, it is believed £3000 (2') Annual budget of Royal Observatory, Edin- burgh, . . . . 1000 (3') Salary of Chief Assist- ant at Royal Observatory, Edinburgh, . . . 200 (4') New Objective and Dome for Edinburgh Equa- torial, certainly ... 0 20 Proceedings of Royal Society of Edinhurgh. [dec. 5, And again on Thursday, Feb. 12, 1885, the country was informed by the Edinburgh Evening Express, that a reorganisation of the Cor- respondence Department of the India Office had just taken place, leaving it thus, — 6 Secretaries at <£1200 per ann. each. 7 Assistant Secretaries, £800 to £1000 each. 11 Senior Clerks, commencing with £600, and rise to £800. 12 Junior Clerks, commence with £200, and rise by £20 per ann. to £600. .Is not this a contrast of most severe kind to occur under the same Government? Especially when one learns further that the Colonial Office Clerks have only day work, and, as may be quite right, in very comfortable, well-warmed rooms of easily accessible buildings kept up at Government expense. While the Edinburgh Observatory Assistants have night, as well as day, work in an inclement little building perched on a hill-top more exposed to storms of wind, rain and snow, and more difficult to get at, or even to leave safely, in the dark than any other Astronomical Obser- vatory or Government Office in any city of the land. Or compare the £100 salary, possibly rising after a time and for a time only, to £\f>0 per ann,, with the £1400^67* ann. of a Clerk in the Treasury, recently defended in the public papers and insisted on as being only just payment for the hard work there, by a Liberal Prime Minister. E"ow these matters, though apparently in my case tinged with personal feelings and sufferings too, yet cannot always, and may not much longer, appertain to me ] while they are otherwise and neces- sarily so intimately connected with the subject of the Edinburgh Equatorial, — if it is ever to be successfully worked and to prove an eventual honour to the country — that they cannot but be entered in any project for the scientific finishing and physical using of the instrument — the largest of its kind that has ever been seen in Scot- land. And then it is also to be remembered that these matters were stipulated for, and promised to the Astronomer before the instru- ment was begun, — see the printed Eeports of the Astronomer ap- proved by the Board of Visitors — and that a Board, appointed nearly a generation before Sir Eichard Cross, having entered the Home Office in London, obtained thereby supreme power over the Eoyal 1887.] C. Piazzi Smyth on the Edinburgh Equatorial. 21 Observatory, Edinburgh; ignored arrangements supposed to have been made for all perpetuity, and rendered impossible, even to the Central Office of Works in London, the completion of this hitherto unfortunate Equatorial tele-spectroscope. A grand beginning, how- ever, of a first class instrument, it must be allowed ; and still safely preserved under a sound wind-and- water-tight Dome for any eventu- alities which the future may be charged to bring along with it. 5. On Cauchy’s and Green’s Doctrine of Extraneous Force to explain dynamically Fresnel’s Kinematics of Double Refraction. By Sir William Thomson. 1. Green’s dynamics of polarisation by reflection, and Stokes’s dynamics of the diffraction of polarised light, and Stokes’s and Rayleigh’s dynamics of the blue sky, all agree in, as seems to me, irrefragably demonstrating Fresnel’s original conclusion, that in plane polarised light the line of vibration is perpendicular to the plane of polarisation; the “plane of polarisation” being defined as the plane through the ray and perpendicular to the reflecting surface, when light is polarised by reflection. 2. E’ow when polarised light is transmitted, through a crystal, and when rays in any one of the principal planes are examined, it is found that — (1) A ray with its plane of polarisation in the principal plane travels with the same speed, whatever be its direction (whence it is called the “ ordinary ray ” for that principal plane) ; and (2) a ray whose plane of polarisation is'[perpendicular to the principal plane, and which is called “the extraordinary ray ” of that plane, is transmitted with velocity differing for different directions, and having its maximum and minimum values in two mutually perpen- dicular directions of the ray. 3. Hence, and by § 1, the velocities of all rays having their vibrations perpendicular to one principal plane are the same ; and the velocities of rays in a principal plane which have their direc- tions of vibration in the same principal plane, differ according to the direction of the ray, and have maximum and minimum values for directions of the ray at right angles to one another. But in the 22 Proceedings of Royal Society of Edinhurgh. [dec. 5, laminar shearing or distortional motion of which the wave-motion of the light consists, the “ plane of the shear ” * (or plane of the distortion,” as it is sometimes called), is the plane through the direction of the ray and the direction of vibration; and therefore it would be the ordinary ray that would have its line of vibration in the principal plane, if the ether’s difference of quality in different directions were merely the aeolotropy of an unstrained elastic solid. t Hence ether in a crystal must have something essentially different from mere intrinsic aeolotropy ; something that can give different velocities of propagation to two rays, of one of which the line of vibration and line of propagation coincide respectively with the line of propagation and line of vibration of the other. 4. The difficulty of imagining what this “ something ” could possibly be, and the utter failure of dynamics to account for double refraction without it, have been generally felt to be the greatest im- perfection of optical theory. It is true that ever since 1839 a suggested explanation has been before the world ; given independently by Cauchy and by Green, in what Stokes has called their “Second Theories of Double Eefraction,” presented on the same day, the 20th of May of that year, to the French Academy of Sciences and the Cambridge Philosophical Society. Stokes, in his Report on Double Eefraction, J has given a perfectly clear account of this explanation. It has been but little noticed otherwise, and somehow it has not been found generally acceptable ; perhaps because of a certain appearance of artificiality and arbitrariness of assumption, which might be supposed to discredit it. But whatever may have been the reason or reasons which have caused it to be neglected as it has been, and though it is undoubtedly faulty, both as given by Cauchy and by Green, it contains what seems to me, in all probability, the true principle of the explanation, and which is, that the ether in a doubly refracting crystal is an elastic solid, unequally pressed or * Thomson and Tail's Natural Philosophy, § 171; (or Elements, § 150). t The elementary dynamics of elastic solids show that on this supposition there might be maximum and minimum velocities of propagation for raj^s in directions at 45° to one another, but that the velocities must essentially le equal for every two directions at 90° to one another in the principal plane, when the line of vibration is in this plane. + British Association Report, 1862. 1887.] Sir W. Thomson on Doctrine of Extraneous Pressure. 23 unequally pulled in different directions by the unmoved ponderable matter. 5. Cauchy’s work on the wave-theory of light is complicated throughout, and to some degree vitiated, by admission of the Navier-Poisson false doctrine* that compressibility is calculable theoretically from rigidity ; a doctrine which Green sets aside, rightly and conveniently, by simply assuming incompressibility. In other respects Cauchy’s and Green’s “ Second Theories of Double Eefraction,” as Stokes calls them, are almost identical. Each supposes ether in the crystal to be an intrinsically aeolotropic elastic solid, having its aeolotropy modified in virtue of internal pressure or pull, equal or unequal in different directions, produced by and balanced by extraneous force. Each is faulty in leaving intrinsic rigidity-moduluses (coefficients) unaffected by the equilibrium-pres- sure; and in introducing three fresh terms, with coefficients (A, B, C in Green’s notation) to represent the whole effect of the equilibrium pressure. This gives for the case of an instrinsically isotropic solid, augmentation of virtual rigidity, and therefore of wave- velocity, by equal pull f in all directions, and diminution by equal positive pressure in all directions ; which is obviously wrong. Thus definitively, pull in all directions outwards perpendicular to the bounding surface, equal per unit of area to three times the intrinsic rigidity-modulus, would give quadrupled virtual rigidity, and therefore doubled wave-velocity ! Positive normal pressure inwards equal to the intrinsic rigidity-modulus would annul the rigidity and the wave- velocity ; that is to say, would make a fluid of the solid. And on the other hand, negative pressure, or outward pull, on an incompressible liquid, would give it virtual rigidity, and render it capable of transmitting laminar waves ! It is obvious that abstract dynamics can show, for pressure or pull equal in all directions, no effect on any physical property of an incompressible solid or fluid. * See Stokes, ‘‘On the Friction of Fluids in Motion and on the Equilibrium and Motion of Elastic Solids,” Camh. Phil. Trans., 1845 ; §§ 19, 20, reprinted in Stokes’s Mathematical and Physical Papers, vol. i. p. 123 ; or Thomson and Tail's Natural Philosophy, §§ 684, 685 ; or Elements, §§ 655, 656. f So little has been done towards interpreting the formulas of either writer that it has not been hitherto noticed tliat positive values of Cauchy’s G, H, I, or of Green’s A, B, C, signify pulls, and negative values signify pressures. 24 Proceedings of Royal Society of Edinburgh. [dec. 5, 6. Again, pull or pressure unequal in different directions, on an isotropic incompressible solid, would, according to Green’s formula (A) in p. 303 of his collected Mathematical Papers, cause the velocity of a laminar wave to depend simply on the wave-front, and to have maximum, minimax, and minimum velocities for wave- fronts perpendicular respectively to the directions of maximum pull, minimax pull, and minimum pull; and would make the wave- surface a simple ellipsoid ! This, which would be precisely the case of foam stretched unequally in different directions, seemed to me a very interesting and important result, until (as shown in § 19 below) I found it to be not true. 7. To understand fully the stress-theory of double refraction, we may help ourselves effectively by working out directly and thoroughly (as is obviously to be done quite easily by abstract dynamics) the problem of § 6, as follows : — Suppose the solid isotropic, when un- strained, to become strained by pressure so applied to its boundary as to produce, throughout the interior, homogeneous strain accord- ing to the following specification : — The coordinates of any point M of the mass which were t], I when there was no strain, become in the strained solid, Uo-,vsIP,tJy (1); Ja, JP, sjy, or the “ Principal Elongations,”* being the same whatever point M of the solid we choose. Because of incompressi- bility we have (2). Eor brevity we shall designate as (a, /?, y) the strained condition thus defined. 8. As a purely kinematic preliminary, let it be required to find he principal strain-ratios when the solid, already strained according o (1) (2), is further strained by a uniform shear, o-, specified as follows in terms of x, y, z, the coordinates of still the same particle, M, of the solid, and other notation as explained below: — * See chap. iv. of “Mathematical System of Elasticity” (W. Thomson), Trans. R. S. Land., 1856, reprinted in vol. iii. of Mathematical and Physical Papers, now on the point of being published ; or Thomson and Tail's Natural Philosophy, §§ 160, 164 ; or Elements, §§ 141, 158. 1887.] Sir W. Thomson on Doctrine of Extraneous Pressure. 25 x = i Ja + apt ] y = r] Jf3 + (rpm (3), 2 C x/y + o-jm J where p = 0~P = Ja + prj J/S + v^ Jy (4), with + -\-7i^ —1 ; \^ + + v'^ — 1 (5) , and l\ + 7nfjL-{-nv = 0 (6); X, fjL, V denoting the direction-cosines of OP, the normal to the shearing planes; and I, w, n the direction-cosines of shearing dis- placement. The principal axes of the resultant strains are the direc- tions of OM in which it is maximum or minimum, subject to the condition + + (7). and its maximum, minimax, and minimum values are the three required strain-ratios. Now we have OM^ = x‘^-\-y^ + z^ = + f/3 + -1- 2cr{li Ja + mrj 4- Jy)p + . . . (8) : and to make this maximum or minimum subject to (7), we have ciaOM^) . c^(iOM2) c?(iOM2) . ... where in virtue of (7), and because OM^ is a homogeneous quadratic function of rj, p = OM2 (10). The determinantal cubic, being -p)- -p)- i w- p) - p) + = 0 , where <^=a(l-l-2o-ZA.-l-(r2A.2) ; J^=/3(1 -f2o-m/i + o-V^) ; <^=y(l + 2t-l-mA.)4-o-2X^] . .(12), gives three real positive values for p, the square roots of which are the required principal strain-ratios. 9. Entering now on the dynamics of our subject, remark that the isotropy (§ 7) implies that the work required of the extraneous pressure, to change the solid from its unstrained condition (1, 1, 1) to the strain (a, /?, y), is independent of the direction of the normal axes of the strain, and depends solely on the magnitudes of a, /3, y. 26 Proceedings of Royal Society of Rdinhurgli, [dec. 5, Hence if E denotes its magnitude per unit of volume ; or the potential energy of unit volume in the condition (a, y) reckoned from zero in the condition (1, 1, 1) ; we have E = i/^(a, fty) ...... (13), where xp denotes a function of which the magnitude is unaltered when the values of a, /?, y are interchanged. Consider a portion of the solid, which, in the unstrained condition, is a cube of unit side, and which in the strained condition (a, y), is a rectangular parallelopiped Ja. Jy. In virtue of isotropy and symmetry, we see that the pull or pressure on each of the six faces of this figure, required to keep the substance in the condition (a, /3, y), is normal to the face. Let the amounts of these forces per unit area, on the three pairs of faces respectively, he A, B, C, each reckoned as positive or negative according as the force is positive pull, or positive pressure. We shall take A + B + C = 0 (14); because normal pull or pressure uniform in ail directions produces no effect, the solid being incompressible. The work done on any infinitesimal change from the configuration (a, /?, y), is A J(fiy)d( Ja) + BV(ya)^(V/3) + C J(ap)d{Jy) , ] or (because a/?y=l) 1-. . (15). =7 J 10. Let 8a, 8^, 8y be any variations of a, J3, y consistent with (2): so that we have (a + 8a) {/3 + 8/3) (y + 8y) = 1 and a/3y = 1 } . . . . (16). How suppose 8a, 8/?, 8y to be so small that we may neglect their cubes and corresponding products, and all higher products. We have ~ ^ + a8j38y + /?8y8a + y8a8/5 = 0 . . (17), a Id y whence whence, and by the symmetrical expressions 1887.] Sir W. Thomson on Doctrine of Extraneous Pressure. 27 2S/38y--(^-^-^j 9S s P‘ 8y2 r ho? Y ■ (18). 1/Sy2 Sa2 r a- 11. Now if E + 8E denote the energy per unit hulk of the solid in the condition (a + Sa, ^ + hp, 7 + 8y) ; we have, by Taylor’s theorem, 8E = + H2 + Hg + &c. where H2, &c. denote homogeneous functions of Sa, Sft Sy, of the 1®^ degree, 2"^ degree, &c. Hence omitting cubes, &c., and eliminating the products from H2, and taking Hj from (15), we find 1/A, da' SE = ;f-8a + ?S^ + ^8y + G^ +1^^" t)' (19). where G, H, I denote three coefficients depending on the nature of the function ij/, (13), which expresses the energy. Thus in (19), with (14) taken into account, we have just five coefficients independently disposable. A, B, G, H, I ; which is the right number because, in virtue of a^y= 1, E is a function of just two independent variables. 12. For the case of a=l, /S=l, y=l, we have A = B = C = 0; and G = H = I = G^, suppose ; which give 8E = iGi(Sa2 + 8/32 + Sy2). From this we see that 2Gj is simply the rigidity modulus of the unstrained solid ; because if we make Sy = 0, we have 8a— -8/3 and the strain becomes an infinitesimal distortion in the plane (xy) which may be regarded in two ways as a simple shear, of which the magnitude is 8a * (this being twice the elongation in one of the normal axes). 13. Going back to (10), (11), and (12), let o-beso small that errand higher powers can be neglected. To this degree of approximation, we neglect abc in. (10), and see that its three roots are respectively ^ ^ ^ r-o) * Thomson and Tail's Natural Philosophy, § 175 ; or Elements, § 154. 28 Proceedings of Royal Society of Edinburgh. [dec. 5, provided none of tlie differences constituting tlie denominators is infinitely small. The case of any of these differences infinitely small, or zero, does not, as we shall see in the conclusion, require special treatment, though special treatment would he needed to interpret for any such case each step of the process. 14. Substituting now for a, 6, e in (20) their values by (11) and (12) ; neglecting cr^ and higher powers ; and denoting by Sa, 8ft Sy the excesses of the three roots above a, /3, y respectively, we find .0. <5a = a|2cr^X +^2 + (Z/x + mX)2 2o-m/x + (T2 ]|^ • . . . (21); ^7 — 7 1 2crwy +cr^ |^i/2 - ^ ^ [mv + Ufif' — — J |- and using these in (19), we find ^ E = o-( AZX + 'Qmix + Cnv) +^o-2{ AX2 + B^2^QJ;24.XJ(mv + ?^/x)2+M{wX + ^yp + N(Zyu + ?/^X)2} ^ , + 2(72(0^2X2 + H??1-2^2 ^ I^2y2^ j where L = ?pf^; M = ^AzAl; N = ^ p - y y — a. a — p 15. Now from (5) and (6), we find (mv + np)^==l-l^-\^ + 2{m-rnY-n^y^) .... (24); which, with the symmetrical expressions, reduces (22) to ^E = cr(AZX + B??i/i + Cni;) ^ + i(r2{L + M + N + (A-L)X2 + (B-MV2 + (C-N>2-LZ2-Mm2-Nw2 i ,(25) + 2[(2G+L-M-N)Z2X2 + (2H + M-N-L)m2^2 + (2i + N-L-M)w2i;2]} j (22); 23). 16. To interpret this result statically, imagine the solid to he given in the state of homogeneous strain (a, ft y) throughout, and let a finite plane plate of it, of thickness h, and of very large area Q, he displaced by a shearing motion according to the specification (3), (4), (5), (6) of § 8 ; the hounding planes of the plate being unmoved ; and all the solid exterior to the plate being therefore undisturbed, except by the slight distortion round the edge of the plate produced by the displacement of its substance. The analytical expression of this is <^=f(p) (26), where / denotes any. function of OP such that 1887.] Sir W. Thomson o)i Doctrine of Extraneous Pressure. 29 r dpf{p) = f) (27). J 0 If we denote by W the work required to produce the supposed displacement, we have W = Q^rfj5SE + W (28), 8E being given by (25), with everything constant except o- a function of OP ; and W denoting the work done on the solid out- side the boundary of the plate. In this expression the first line of (25) disappears in virtue of (27) ; and we have w_ w , P — =1{L + M + N + (A-L)X2 + (B-M)/x2 + (C-N)i;2-LZ2-Mwi2-N7i2 Q + 2[(2G + L - M - N)Z2\2 + (2H + M - N - L)m2/x2 + (21 + N - L - ^i?o-2 . . (29) . When every diameter of the plate is infinitely great in comparison with its thickness, W/Q is infinitely small; and the second member of (29) expresses the work per unit of area of the plate, required to produce the supposed shearing motion. 17. Solve now the problem of finding, subject to (5) and (6) of § 8, the values of /, n which make the factor { } of the second member of (29), a maximum or minimum. This is only the problem of finding the two principal diameters of the ellipse in which the ellipsoid [2(2G+L-M-N)A2-L]o;2+[2(2H+M-N-LV2-M]y2+[2(2H-N-L-M)i/2-N]^2::,const. . . (30) is cut by the plane 'Kx + ixy + vz = 0 . . . . . . (31). If the displacement is in either of the two directions (Z, m, n) thus determined, the force required to maintain it is in the direction of the displacement ; and the magnitude of this force per unit bulk of the material of the plate at any point within it is easily proved to be {»•}$ (”). where {M} denotes the maximum or the minimum value of the bracketed factor of (29). 18. Passing now from equilibrium to motion, we see at once that (the density being taken as unity) Y2={M} (33) 30 Proceedings of Eoyal Society of Edinhurgh. [dec. 5, where V denotes the velocity of either of two simple waves, whose wave-front is perpendicular to (A, /x, v). Consider the case of wave- front perpendicular to one of the three principal planes ; {yz) for instance ; we have \ = 0', and, to make { } of (29) a maximum or a minimum, we see by symmetry that we must either have perpendicular to principal plane) 1=1^ m— o,n—o or (vibration in principal plane) l=o, m——v, n=ix Hence, for the two cases, we have respectively YihxdXiou perpendicular to yz . . V2=M + N-f(B-M)fi2 + (C_N’)j;2 . . (35); Vibration m 2/2: V2=L-i-B/x2+Cj/2 + 4(H-l-I-L)|uV . . (36). 19. According to rresnel’s theory (35) must he constant, and the last term of (36) must vanish. These and the corresponding conclusions relatively to the other two principal planes are satisfied if, and require that, A-L = B-M = C-H (37), and H4-I = L; l4-G = M; G-bH = H . . (38). Transposing M and N in the last of equations (37), substituting for them their values by (23), and dividing each member by ^y, we find ~ ^ B — A ^ ^ ^ ^ /3y-a/3 ya-/3y whence (sum of numerators divided by sum of denominators), B-C _ C-A _ A-E ya — a/3 a(3 — /3y (3y — ya . . (40). The first of these equations is equivalent to the first of (37) j and thus we see that the two equations (37) are equivalent to one only ; and (39) is a convenient form Of this one. By it, as put symmetri- cally in (40), and by bringing (14) into account, w^e find, with k taken to denote a coefficient which may be any function of (a, yS, y) : A = A:(S-^y); B = ^(S-ya); C = KS-a/5)l ^ where S = ^{/3y + ya + a/3) f and using this result in (23), we find L=^[a(g+y)-S]; M=^-[d(y+«)-S]; N=k[y{a+g}-B]) or L=k(2S-/3y); M=/c{23-ya); N=/c{2S-ay) i ‘ ‘ ^ ’ By (2) we may put (41) and (42) into forms more convenient for some purposes as follows : — 1887.] Sir W. Thomson on Doctrme of Extraneous Pressure. 31 C-i(S.l) . .i(2S-i);M.i(2S-L)iN.l(2S-l) . where 1/1 1 + 7S + 3\a /I y/ Next, to find G, H, I ; by (38), (44), and (45), we have G + H + I = l(L + M + N) = f/tS = |A:(Ni + -) . . whence by (38) and (44), 20. Using (43) and (47) in (19), we have Sa2 8^2 § 2 .J„2 S 2 ■, (43) , (44) , (45) . (46) , (47) . (48) . •/32 Now we have, by (2) log {a^y) = 0. Hence taking the variation of this as far as terms of the second order. P y which reduces (48) to Remembering that cubes and higher powers are to be neglected, we see that (50) is equivalent to 3E = JA:8(1 + ^ + 1) (51). Hence if we take k constant, we have )• • • • • and it is clear that h must be stationary (that is to say 3/i; = 0) for any particular values of a, /I, y for which (51) holds; and if (51) holds for all values, Ic must be constant for all values of a, /?, y. 32 Proceedings of Eoyal Society of Edinhurgli. [dec. 5, 21. Going back to (29), taking Q great enough to allow W/Q to be neglected, and simplifying by (46), (43), and (44) we find Q = " ii‘ a (d y (53); and the problem (§17) of determining Z, m, w, subject to (5) and (6), to make P'ja + + n^^jy, a maximum or minimum for given values of A, /x, v, yields the equation Z m 71 ■3S-X — •st7 h 0 ; - 7!r'm H = 9 > (^^)> 33-, denoting indeterminate multipliers; whence i2 ^2 73- — 1" "W “1 a (d y r-2 = l^\ 73X — I ■srix = m P rrP n^\] ^ 7/ P 1- / Z2 m- OTj/ = ^^ I )- \ a ^2 ’ 1 - ) (55) , (56) , (57) . P y These formulas are not directly convenient for finding Z, m, %, from A, /X, 7/, given (the ordinary formulas for doing so need not be written here) ; but they give A, /x, v explicitly in terms of Z, m, n, supposed known ; that is to say, they solve the problem of finding the wave-front of the simple laminar wave whose direction of vibra- tion is (Z, m, n). The velocity is given by P „ P 7}P v^ = H - + -^ + — a (d y (58). It is interesting to notice that this depends solely on the direction of the line of vibration ; and that (except in special cases, of partial or complete isotropy) there is just one wave-front for any given line of vibration. These are precisely in every detail the conditions of Fresnel’s Kinematics of Double Eefraction. 22. Going back to (35) and (36), let us see if we can fit them to double refraction with line of vibration in the plane of polarisation. This would require (36) to be the ordinary ray, and therefore re- 1887.] Sir W. Thomson on Doctrine of Extraneous Pressure, 33 tXuires the fulfilment of (38), as did the other supposition : but instead of (37) we now have [in order to make (36) constant] A = B = C (59), and therefore each, in virtue of (14), zero; and therefore by (43), a = /? = y=l; so that we are driven to complete isotropy. Hence our present form (§ 7) of the stress theory of double refraction cannot be fitted to give line of vibration in the plane of polarisation. We have seen (§21) that it does give line of vibration perjpendieidar to the plane of polarisation with exactly FresneVs form of wave-surface, when fitted for the purpose by the simple assumption that the potential energy of the strained solid is expressed by (52) with A: constant ! It is important to remark that k is the rigidity-modulus of the un- strained isotropic solid. 23. From (58) we see that the velocities of the waves correspond- ing to the three cases, Z=l, m = l, 1, respectively, are ^(A:/a), J{klj3), slikjy). Hence the velocity of any wave whose vibrations are in the direction parallel to any one of the three principal elonga- tions, multiplied by this elongation, is equal to the velocity of a wave in the unstrained isotropic solid. 6. Exhibition of Models. The President exhibited Models of the Minimal Tetrakaide- kahedron. His paper on the subject is printed in the London^ Edinburgh, and Dublin Philosophical Magazine, vol. xxiv. 5th series, p. 503, December 1887. 7. Researches on Micro-Organisms, including ideas of a new Method for their destruction in certain cases of Contagious Diseases. Part II. By Dr A. B. Griffiths, F.R.S. (Edin.), E.C.S. (Bond, and Paris), Principal and Lecturer on Chemistry and Biology, School of Science, Lincoln; Science Master in the Lincoln Grammar School, ^’c. In the Proceedings of the Royal Society of Edinburgh, vol. xiv. [No. 123], pp. 97-106, there is a paper of mine under the above title. I wish in the present memoir to communicate to your dis- voL. XV. 7/6/88 c 34 Proceedings of Royal Society of Edinhurgh. [dec. 5, tinguislied Society further details relative to these investigations. The principle of these researches is to find some germicidal agent capable of destroying the microbes of disease, which have been proved to reside in the blood, and are the causes (directly or in- directly) of certain contagious diseases. At the same time, an aqueous solution of such an agent, while destroying the microbes of disease, must have very little or no detrimental action upon the blood. Having found such a substance, the rationale is to inject (hypodermically) a solution of the microbe-destroyer directly into the blood. By so doing, the destruction of the pathogenic organ- isms in situ would be the result. In my last memoir on this subject (Jioc. cit.) aqueous solutions of salicylic acid were shown to materially interfere with the life-his- tories of certain micro-organisms. In the present paper an account will be given of the action of various antiseptic and germicidal agents upon certain microbes and their spores, as well as a practical application of my method in a particular case of advanced phthisis. I. Alkaloids produced by Living Microbes. It appears, as pathological investigations progress, the real cause in many cases of contagious diseases (although not in all) is the for- mation of certain poisonous compounds (ptomaines or alkaloids) by living microbes ; rather than the idea that the mere presence of these microbes in the blood or tissues causes such diseases. It will be remembered that in 1885 Pouchet discovered the ptomain formed by the Comma Tjacillus ; and being very soluble, is easily absorbed into the system. Hence the rapidity of death fol- lowing the first symptoms of the disease. Amongst very recent work on the subject of ptomaines, produced by pathogenic and other microbes, we have the following ; — (a) Dr O. Bocklisch {Berichte der deutschen chemischen Gesellscliaft, vol. XX. p. 1441) found that Vibrio proteus produced in contact with sterilised beef cadaverine (CgH^^Hg) which had been proved by Ladenburg {Berichte der deutschen chemischen Gesellschafty vol. xix. p. 2585) to have all the chemical properties of pentam ethylene- diamine (C5HJ4H2). This alkaloid or ptomaine of Vibrio proteus (Tinkler’s bacillus) is non-poisonous. Bocklisch went a step further, and found that when Vibrio proteus was allowed to live upon 35 1887.] Dr A. B. Griffiths on Micro-Organisms. sterilised beef along with putrefactive germs, besides cadaverine, a very poisonous base methylguanidine is the chief product of their life-histories, (h) Brieger {Berichte, vol. xix. p. 3119) has suc- ceeded in isolating an alkaloid, which he calls tetamine (C13H3QN2O4) from pure cultivations of the bacillus, which causes traumatic tetanus in animals. (c) Although it has not been isolated, M. Pasteur believes that the virus of hydrophobia is a microbe, and that it produces an alkaloid, {d) Dr E. Alvarez {Comjptes Rendus Hehdomadaires des Seances de VAcadeinie des Sciences, vol. cv. [No. 5], 1st August 1887) describes a microbe which he has proved to be the cause of the indigotic fermentation and the production of indigo-blue. This microbe is an encapsuled bacillus (fig. 1), similar in appearance to the bacillus of Rhinoscleroma (Cornil and Alvarez). This bacillus of the indigo fermentation is shown to possess patho- genic properties, and occasions in animals a transient local inflam- mation, or death, with visceral congestion and fibrinous exudations, (e) It has been. shown by Duclaux (in his work on Ferments et Maladies) that when the ptomaine produced by Bacterium cliolerce gallinarum (which possesses narcotic properties) is separated, by fil- tration through a Chamberland filter, from its microbe, it does not produce fowl cholera, but causes a passing sleep, which does not generally end fatally. I have alluded here in passing to recent work on the secretions or products formed during the life-histories of certain microbes, — and 36 Proceedings of Royal Boeiety of Edinburgh. [dec. 5, it appears that the most important problems requiring' the attention of the pathological worker are — (1) To isolate the microbes and their alkaloids in a given contagious disease, and to study their chemical and pathological actions ; (2) To destroy the microbes (if they reside in the blood) in situ by hypodermic injections of some germicide. II. Cellulose the Prohuct formed by certain Micro- Organisms. In the Journal of the Chemical Society [Trans.], 1886, p. 432, Mr Adrian J. Brown, F.C.S., describes an acetic ferment, called by him Bacterium xylinum, which forms cellulose ; the substance of the membranous growth of the so-called “ vinegar-plant,” or the “ Essighautchen ” of Dr Zopf. In my previous paper on this subject (loc. cit.) I alluded to the fact that Dr E. Freund had discovered that Bacillus tuberculosis forms cellulose. My own work on this micro-organism (to be described in this present memoir) entirely confirms Freund’s dis- covery, and somewhat extends his observations. He found cellulose in the organs and blood of tuberculous persons, and I may add that cellulose is also to be found in the sputa of patients suffering from acute general phthisis. This was proved by the reactions used by Freund (see Nature^ vol. xxxiv. p. 581) for the detection of cellu- lose in tuberculosis. Ill, Action op Certain Antiseptics and Disinfectants upon VARIOUS Micro-Organisms. I have already shown that a solution of salicylic acid is a germi- cidal agent of a large number of micro-organisms j and at this point I wish to detail several experiments undertaken to see the action of various reagents upon the life-histories of certain microbes. (a) Sarcina lutea. Several Aitken’s tubes (fig. 2) containing sterilised beef-broth (neutral) were taken and treated as follows : — Tube No. I. was inoculated with the chromogenic saprophyte Sarcina lutea (from a pure cultivation in nutrient agar-agar), and 1887.] Dr A. B. Griffiths on Micro-Organisms. 37 kept at a temperature of 40° C. They grew rapidly, and after four days formed a yellow pellicle upon the surface of the broth. Tubes hTos. II. and III. contained sterilised beef-broth ; and to the broth in each tube was added iodine (in the proportion of 1 millogramme of iodine to 100 c.c. of broth). The tubes were then inoculated with Sarcina lutea from the same source as Tube Ho. I. Ho growths made their appearances after the elapse of twenty-eight days, although the tubes were kept at the most favourable tempera- ture for the development of this micro-organism. After the elapse of twenty-eight days, sterilised platinum needles were dipped into tubes Ho. II. and Ho, III., and the contents of four tubes containing sterilised nutrient agar-agar were inoculated from them. They remained in the incubator at 40° C. for twenty-one days, without any growths making their appearances in the tubes. Other germicidal agents were tried (in a similar manner to the experiments just described) upon Sarcina lutea. Amongst these reagents the following proved fatal to the micro-organism : — 0*5 per cent, solution (sterilised beef broth) of potassium iodate. 3*0 „ „ „ „ ,, salicylic acid. 0-4 „ „ „ „ „ sodium fluosilicate. {h) Micrococcus prodigiosus. I have already shown that salicylic acid is fatal to the growth and multiplication of this organism (see Part I. of this paper). Since the above experiments were performed upon this organism I have tried other experiments. In the preparation of sterilised nutrient agar-agar (according to the well-known methods) the above quantities of potassium iodate, salicylic acidy and sodium lluosili- cate were added before filtration. After preparing a series of tubes containing sterilised nutrient agar-agar, with and also without the germicidal agent, they w^ere all inoculated (the utmost care being observed) from pure cultivations of Micrococcus prodigiosus (fig. 3, A). Tube Ho. I. was inoculated by means of a sterilised platinum wire from the potato cultivation of the micro-organism. After five days’ growth at 34° C. in an incubator, the appearance was similar to the growth in fig. 3, B. (The colour was crimson.) 38 Proceedings of Roycd Society of Edinburgh. [dec. 5, Tube No. II. contained, in addition to the nutrient agar-agar, 3 per cent, of salicylic acid. Tube No. III. contained, in addition to the nourishing medium, 0’5 per cent, of potassium iodate. Tube No. IV. contained, in addition to the sterilised agar-agar, 0*4 per cent, of sodium fluosilicate. B — A growth of Micrococcus prodigiosus in nutrient agar-agar. C. — Appearance of nutrient agar-agar (inoculated) in each tube after the addi- tion of the antiseptics mentioned in this paper, and after twenty-days incubation at 34° C. Fig. 3. — Micrococcus prodigiosus. Tube No. V, contained 1 milligramme of iodine in 100 c.c. of the medium. Tubes Nos. II., III., IV., and Y. did not develop any growths after twenty-five days’ incubation at a temperature of 34° C. After this period had elapsed, sterilised platinum needles were plunged into each tube, and were then transferred to four tubes containing sterilised nutrient agar-agar. No growths made their appearances in any of the tubes after the elapse of three weeks. All the above experiments were performed in duplicate with similar results. 1887.] Dr A. B. Griffiths on Micro-Organisms. 39 (c) Micrococcus tetragonus. This micrococcus (fig. 4) is found in the sputum of patients suffer- ing with phthisis. According to the most reliable sources, Micro- ‘oecus tetragonus is only saprophytic in man, hut pathogenic in animals. Mice inoculated with a small quantity die in a few days Fig. 4. — Micrococcus tetragonus, stained with gentian violet (much enlarged). andthe microbe afterwards is to be found in various organs of the bod'. This microbe grows tolerably well in nutrient agar-agar.* I hve experimented with Micrococcus tetragonus in an exactly sim:ar manner to the experiments with Micrococcus prodigiosus^ and)btained similar results. The agents used completely destroyed thismicro-organism. Te sputum for this purpose was kindly sent to me on 19th July 188, by Dr E. Wood, M.D., L.E.C.P. (Edin. & Bond.), &c., of Broisgrove, Worcestershire, from one of his patients. The bottle senito me was labelled : — “ Thomas Smith [young man), expectora- tion of supposed 'phthisis at base of left lung. Sister died of it '' I fond in the sputum a considerable number of Bacillus tubercu- losii Micrococcus tetragonus, and a large quantity of Ereund’s celliose. [d) Bacillus buhjricus. I will be remembered that in my last memoir [loc. cit.) on this subict, I gave an account of having destroyed Bacillus butyricus by Lsing the germicide salicylic acid in small quantities. This fad has recently been confirmed by M. Pierre Grosfils. M. *Lgar-agar can be obtained from Christy & Co., 25 Lime Street, London, at 8 per lb. 40 Proceedings of Royal Society of Edinlurgh. [dec. 5, Grosfils communicated a paper to the Societe d’Encouragemeiit de Vervier ; and it has recently been published in the Moniteur Indus- trial^ describing a method for preserving butter from the action of Bacillus hutyricus hj addition of 0’0002 per cent, of salicyli/ acid. Butter so treated will keep for an indefinite time, even ik warm countries. / (e) A Neio Micro-Organism. j A new micro-organism, I have recently found upon putrefytg onions kept in a warm, damp, and dark place. The cells are ab^ut 0*005 to 0*007 mm. long, and about 0*0025 mm. in width, 'jhis microbe is capable of forming the zoogloea state. Fig. 5. — Bacterium allium (a new micro- organism) growing on nutrient agar- agar (after Nature, but not the colour). Fig. 6. — Bacterium allium mder high power), stained with gentian violet. The cel are colourless. The green clour- ing matter (which is insuble in water, soluble in alihol) resides in the interstitia]sub- stance. They grow tolerably well in nutrient agar-agar, and prodie a bright green pellicle upon the surface of the nourishing me(uni (fig. 5). This micro-organism, which causes putrefaction in omns, liberates small quantities of sulphuretted hydrogen gas. The sulphuretted hydrogen was proved by the black stain (PbS) prodced upon paper impregnated with a solution of lead acetate ; andjilso the yellow stain (CdS) produced by using cadmium paper (Cdllg). This sulphur gas is also produced to a small extent in the nutent 1887.] Dr A. B. Griffiths on Micro-Organisms. 41 agar-agar during an artificial cultivation of the microbe in that medium. The microbe stains best with gentian violet (fig. 6). I propose to call this microbe Bacterium allium, because it was dis- covered upon Allium ceg^a. Bacterium allium is destroyed by the reagents described under the head of Microeocccus prodigiosus. The colouring matter formed during the life-history of Bacterium allium is soluble in alcohol. Fig. (III.) gives the absorption spectrum of the pigment in alcohol. ABCD E&F n n Fig. 6a. — Absorption Spectrum of an alcoholic solution of the green pigment formed during the life-history of Bacterium allium. It will be noticed that there is an absorption band extending from the extreme violet to the greenish blue part of the spectrum. Also, an absorption band in the green and one in the yellow part of the spectrum. The end of the band in the yellow is exactly in the same position as the D Fraunhofer line in the solar spectrum. It will also be seen from fig. Qa that the spectrum produced by this pigment differs from chlorophyll, although both solutions were of the same intensity of colour, and nearly the same thickness, when placed in front of the slit of the spectroscope. (/) Various Micro-Organisms. The following micro-organisms were destroyed by the germicides already mentioned : — (1) Micrococcus citreus conglomeratus (obtained from the dust of the air). (2) Bacterium urece. (3) Bacterium indicum. (4) Micrococcus violaceus. (5) Sarcina aurantiaca. 42 Proceedings of Royal Society of Edinburgh, [dec. 5, {g) Penicillium glaiicum. Penicillimn glaucum grows well in flour-paste in a warm damp place. It is destroyed by salicylic acid, iodine, potassium iodate, and sodium fluosilicate ; for no growths made their appearance in flour-paste (inoculated with the spores of this fungoid growth) after forty-six days’ incubation. IV. The Vitality op Bacillus tuberculosis and its Spores. Im March of the present year (1887) I received from Mr John Snodgrass,* jun., of Glasgow (who is suffering with acute general phthisis) typical specimens of sputum, which contained a large quantity of old discoloured blood, also lung fibre, debris of various kinds, and numbers of Bacillus tuber cidosis. Fig. 8a is a drawing from a cover-glass preparation. Small quantities of sputum were mixed with calcium sulphate and calcium carbonate, previously sterilised at a temperature of 135° C., and these mixtures were placed in sterilised tubes (fig. 7), and then hermetically sealed. Twelve of these dry tubes, each con- tained about 10 grammes of the mixture. Twelve dry sterilised tubes (fig. 8), not hermetically sealed,, also contained about 10 wool\ plug j ¥ig. 8. grammes of the mixture of sputum, calcium sulphate, and calcium carbonate (these mineral substances constituting the principal in- gredients contained in the dust of the atmosphere). The twenty- four tubes were kept at a dry heat of 32° C., from one to six months. Two hermetically sealed tubes and two of the open tubes were opened after being exposed to a dry heat of 32° C. for one * The translator of Heine’s “ Religion and Philosophy in Germany,” also “Wit, Wisdom, and Pathos, from the Prose of Heinrich Heine.” (Triibner & Co.) 1887.] Dr A. B. Griffiths on Micro-Organisms. 43 month j and four tubes, containing sterilised blood serum, were inoculated from the contents of the tubes. In the two inoculated from the open tubes, growths of Bacillus tuberculosis (proved by staining and microscopical appearance, &c.) made their appearances in sixteen days from the time of inoculation. Growths of Bacillus tuherculosis also made their appearance in the two tubes (after being inoculated from the contents of the sealed tubes) after nineteen days’ incubation. Four more tubes were opened after being Fig. 8a. — Bacillus tuberculosis in Acute General Phthisis ; from sputum of Mr John Snodgrass, jun. Stained by the Weigert- I Ehrlich method, x about 1400. exposed for two months at the temperature already mentioned. Inoculations from two open tubes revealed the vitality of the Bacillus tuberculosis after twenty days’ incubation ; and inoculations from the two sealed tubes proved the vitality of the bacilli after the elapse of twenty- three days’ incubation. The remaining tubes were examined in a similar manner after the elapse of three, four, five, and six months respectively. 44 Proceedings of Royal Society of Edinburgh. [dec. 5, After being exposed to tbe dry heat for three and four months, the vitality of this micro-organism and its spores was not destroyed. But, after being heated for five and six months the vitality of the microbe was completely destroyed; for no growths made their appearance in sterilised blood serum kept at a temperature between 37° and 39° C. for nearly two months. From these experiments it will be seen that Bacillus tuberculosis is capable of being dried up in the dust of the atmosphere for several months without its vitality being impaired. V. Bacillus tuberculosis disseminated by Flies, Paper, &c. It has been shown that farm animals may be inoculated through the bite of flies with Bacillus anthracis (the Bacteridia of Davaine); and Pasteur [Bulletin de V Academic de Medecine, 1880) has shown that the casts of Lumbricus terrestris may contain the germs of splenic fever, at the same time possessing all their original viru- lence. Eecently, MM. Spillman and Haushalter [Gomptes Rendus Hebdomadaires des Seances de V Academic des Sciences , vol. cv. [Illo. 7], 16th August 1887) have discovered that the common house fly in consumptive hospitals is very often seen upon the expectora- tions of the patients. Some of these flies were caught and placed under bell-glasses, and subsequently it was found that their excre- ments contained numbers of Bacillus tuberculosis. Eecently, I have cultivated (using every possible aseptic precaution) from the envelopes containing the letters from Mr Snodgrass (already mentioned), in sterilised solid blood serum growths which had all the macroscopical appearances of Bacilhis tuberculosis. These pure cultivations gave serpent-like twistings in cover-glass impressions, — and under the higher powers of the microscope the characteristic form of Bacillus tuberculosis when stained by the Ehrlich and other methods.* From this investigation we draw the conclusions — (1) that the saliva of consumptive patients used in moistening an envelope may * Although I had been experimenting with Bacillus tiiberculosis for some time, there were no chances of my cultivation plates and tubes becoming con- taminated with Bacillus tuberculosis from sputum, &c., or with foreign microbes. They were inoculated from the envelopes in a room (with closed doors and windows) away from my laboratories ; and further, I had changed my clothes and disinfected my hands. 1887.] Dr A. B. Griffiths on Micro-Organisms. 45 contain the germs of phthisis ; (2) that these germs are capable of travelling a distance of over 200 miles, and then growing, when a suitable medium and temperature (36° - 39° C.) are provided for them. yi. Electrical Experiments on the Bacillus tuherculosis AND ITS Spores. The action of the electric current upon the vitality of various micro-organisms has been very little studied ; therefore, perhaps, the following notes may he of interest. The experiments were performed on pure cultivations of micro- organisms growing in fluid blood serum (slightly alkaline), and other media. (See fig. 9, representing the general arrangements). Fig. 9. — Electrical Experiments on the Vitality oi Bacillus tuherculosis and its spores, &;c.- A = a tube containing growing bacilli in sterilised fluid blood serum slightly alkaline. (1) Bacterium lactis, growing in previously sterilised milk, is killed by an E.M.F. of 2*26 volts. (2) Bacterium acetic growing in previously sterilised alcohol (7 per cent.), is killed by an E.M.F. of 3 '24 volts. (3) Bacillus tuherculosis, growing in previously sterilised fluid blood serum, is killed by an E.M.F. of 2T6 volts. The temperature of the room was 16° C. After allowing the current to pass for 10 minutes in each case, ten tubes containing sterilised fluid blood serum were inoculated from the “ electrified ” 46 Proceedings of Royal Society of Edinhurgli. [dec. 5, (if I may use tliat expression) tubercle-bacilli, and after being kept at a temperature of 38° C. for twenty-five days, no growths made their appearance in any of the tubes. A similar number of tubes containing sterilised sweet milk were inoculated from the “ electrified ” Bacterium lactis, with no results after twenty-five days’ incubation. Seven tubes containing the purest ethyl alcohol and ordinary filtered tap-water* (the mixture contained 6 per cent. ' of alcohol) were inoculated with the ‘‘ electrified ” Bacterium aceti, with negative results. So we have here positive evidence that these micro-organisms were destroyed by the electric current. VII. Is Bacterium aceti the Eeal Cause op the Acetic Feumentation ^ Although Pasteur maintained that Bacterium aceti was the cause of the acetic fermentation ; and Cohn {Biol. d. Pflanzen, vol. ii. p. 173) observed the micro-organism largely’ in sour beers; yet, not until the commencement of 1886, could any one say with certainty that this micro-organism was the real cause of the acetic fermentation. In that year, Mr Adrian J. Brown, F.C.S. {Journal Chemical Society 1886, p. 172), prepared pure cultivations of Bacterium aceti, and found that the well-known reaction — C2H5OH -f O2 = H2O -h CH3.COOH , is produced by the life-history of Bacterium aceti (Mycoderma aceti). I can entirely endorse the correctness of Mr Brown’s observations, for after obtaining pure cultivations of the micro-organism by a com- bination of the fractional and dilution methods (used by Brown), it was found that these cultivations, when used to inoculate steri- lised ethyl alcohol (6 per cent.) gave acetic acid in abundance. VIII. Anal and Hypodermic Injections of Aqueous Solutions OP Salicylic Acid in Cases of Cholera. It will be remembered that early in the present year (1887) the * Tap-water was used in preference to distilled water, on account of the mineral matter it contains — the micro-organisms requiring small quantities of mineral matter. 1887.] Dr A. B. Griffiths on Micro-Organisms. 47 papers were full of accounts of “ human beings dying in heaps ” from cholera in the province of Cordova, in the Argentine Eepuhlic. Ts there no cure for cholera ? Or, in other words, is there no agent that will destroy Koch’s bacillus in the human body % In passing, I may say, through reading the various newspaper abstracts of my memoir {Proc. Roij. Soc. Edin., vol. xiv. pp. 97-106), read before the Society on January 31, 1887, Mr T. F. Agar (Consul-General for the Argentine Eepuhlic in Scotland) kindly wrote for a copy of my memoir. A written copy was forwarded to him, and he has presented it to his Government at Buenos Ayres. I am to have full details of any experiments performed on behalf of the Govern- ment of the Argentine Eepuhlic hearing on my injection method in cases of cholera.'^* It has been suggested some few years ago, that rum or cognac, containing 25 grammes of salicylic acid to the litre,! should he taken when cholera is epidemic. If salicylic acid proves useful as a germicide, or even a pre- servative, from the severer attacks of Koch’s Bacillus komma, would not anal and hypodermic injections of solutions of the acid he the best method of combating this disease ^ By these two kinds of injections, we should meet the growths of the microbe in the intestines, and also those that may have passed into the blood system by absorption. Koch has remarked that acids in general are the greatest hin- drance for the development of the cholera bacillus ; and Dr Klein, F.E.S. {Micro-Organisms and Disease, p. 256), says — “Patho- genic organisms do not thrive in an acid medium.” At any rate, whether the germicidal agent or medicament used be salicylic acid or one more powerful ; I think that such a method as the one described would be the most rational, and evidently would possess a scientific basis, namely, the destruction of microbes in situ. In the case of disinfecting a whole district against the cliolera epidemic, the late Dr Wm. Budd, F.E.S., placed in the sewers of Bristol ferrous sulphate. Dr Budd says : — “ In this way a chemical * Any information received from this source will be embodied in another paper communicated to the Royal Society of Edinburgh. t “Three teaspoonfuls of the mixture to be taken between meals in coffee or tea.” 48 Proceedings of Royal Society of Edinburgh. [dec, 5, bed was prepared for tbe poison, by whose action the population was ensured against barm from any specific germs that by accident or other cause might find their way into the drains or sewers of the town. The sulphate of iron in the drain, thus lying in waiting for the poison, may be likened to the wire gauze of the Davy lamp, always at hand to prevent the explosion of the fatal fire-damp.”* I have shown {Chemical News, vol. xlix. p. 279 ; vol. liii. p. 255 ; vol. Iv. p. 276 ; Journal Chemical Society [Trans.], 1886, p. 119 ; and Cliemiker-Zeitung^ No. 47) that ferrous sulphate destroys parasitic fungi ; and it is probable that on a large scale (for sewers, &c.) it would form a cheap and powerful disenfectant against epidemic diseases in general. IX. Soluble Zymases and their Microbes. What have the soluble zymases (ferments) produced by various pathogenic microbes to do, in connection with contagious diseases ? Arc they the cause of the disease directly or indirectly 1 By their chemical disintegration, do they form the alkaloids (ptomaines) found in disease ? These problems require our earnest attention. Dr Schiavuzzi of Pola (Istria in Austria] {Rendiconti della R. Accademia dei Lincei, December 1886) has confirmed Kleb’s and Tommasi-Crudeli’sf discovery of Bacillus malarix, and that it is the real cause (directly or indirectly) of malarial fever. Schiavuzzi also finds that in the blood of animals infected with the disease, the red corpuscles undergo similar alterations as Marchiafava and Celli {Fortschr. d. Med., vol. iii.) have shown to be characteristic of malarial fever; and he considers these changes in the blood corpuscles to be caused by a “pathological” ferment of a different nature to Bacillus malarise. Most probably a soluble zymase secreted by the microbe itself. Professor Giglioli, in his recent work Fermenti e Microbi, describes the production of soluble fer- ments by micro-organisms. If the soluble zymases produced by living pathogenic microbes * The Cholera Microbe, and how to Meet it, by Sir C, .Cameron, LL.D., M.P., &c., p. 25. t Archiv fur Experimental Pathol., 1879 ; and also Tonimasi-Crudeli’s memoir, “ Der in Erdboden von Seliunte imd Campobello,” Archiv fiir Exp. Pathol., 1880. 1887.] Dr A. B. Griffiths on Micro-Organisms. 49 are the real cause of disease, the hypodermic injection method steps in, for many substances are known to interfere with the action of soluble ferments.* The destruction of the microbes prevents the formation of soluble zymases or alkaloids ; and any given contagious disease (under these circumstances) would he at an end. Nature would then have a chance of restoring to their “ normal standard the lowered vitality which enabled the microbes to get a footing.” X. Salicylic Acid, Natural and Artificial. In the present paper it has been shown that salicylic acid is a good germicide. The natural acid prepared from oil of winter green [Gaultheria procumhens) is a far more powerful germicide than the “artificial” salicylic acid prepared from sodium phenate (CgHgNaO). Hence, it appears from the above observations that the natural variety possesses properties which are not to be found in artificial salicylic acid. This fact supports Pasteur’s idea {Revue Scientifique, January 5, 1884) that organic compounds prepared by synthesis are not altogether identical with the natural compounds. “ Life ” brings into play asymmetrical molecular forces, while in the mineral kingdom and also in our laboratories, only symmetrical molecular forces come into play. Pasteur’s theory is summed up by M. Wyrouboff in these words : — “ Ces theories sont fondees sur une premiere hypoth^se, qui suppose les phenomenes naturels soumis a deux sortes d’actions ; les unes symetriques, les autres dissymetriques ; les premieres president a la mineralit4 et aux syntheses de nos laboratoires, les secondes appartiennent a la vitalite ” {Bulletin de la Societe chimique de Paris, vol. xli. p. :210, March 5, 1884). Pig. 9a represents microscopical slides of pure salicylic acid crystals deposited from alcohol and ether. XL The Treatment of Phthisis by Injection and other Methods. Before I come to my own experiments, I wish to allude to the work of others in the same direction. * Dumas, Comptes Berdus, vol. Ixxv. p. 295 ; Bouchardat, Annates de Chimie et de Physique (3rd series), vol. xiv. p. 61 ; Griffiths, Proc. Roy. Soc. Edin., vol. xiii. [No. 121], p. 527. VOL. XV. 9/6/88 D 50 Proceedings of Royal Society of Edinburgh. [dec. 5, In tlie British Medical Journal for December 18, 1886, there is an article from the pen of Dr J. H. Bennet (of Paris) on Dr Bergeon’s i treatment of pulmonary phthisis by means of anal injections of two ' gases. Bergeon found that sulphuretted hydrogen and carbon I dioxide gases were absorbed by the intestines without any poisonous effects. He uses the natural “Eaux Bonnes” water from the Pyrenees as his source of pure sulphuretted hydrogen, — and by repeated anal injections of these gases has cured the worst cases of pulmonary phthisis and other pulmonary diseases (see Cowfptes Rendus, July 12, 1886, p. 176, and Bulletin de V Academic de Mededne.^ 2nd ser., vol. xvi.) ; and Dr MDaughlin (Physician of the Philadelphia Hospital) recently reports the cure of thirty Salicylic Acid Crystals, crystallised from Salicylic Acid Crystals, crystallised from alcohol. X 340. ether, x about 95. patients in the last stages of consumption, solely by using Bergeon’s method. Bergeon has found that sulphuretted hydrogen, prepared from any other source than Eaux Bonnes ” water or carbon disulphide, will prove a failure. He “ does not propose his method as a microbicide treatment, but merely as one that succeeds,” and that ‘‘the injection of sulphuretted hydrogen is decidedly antiseptic and curative of local lesions.” Therefore, Bergeon’s treatment of phthisis is by anal injection of several litres of the mixed gases into the intestines. There the gases are absorbed into the venous system, and pass out by the 1887.] Dr A. B. Griffiths on Micro-Organisms. 51 lungs. The “ antiseptic ” gases do not pass into the arterial system, as they would by inhalation. Through the kindness of Mr Snodgrass (already mentioned), allowing me to make free use of the important letters he has written me, I am in a position to give critical opinions on the treatment of acute general phthisis by means of — (1) Hypodermic injections of warm solutions of salicylic acid. (2) Bergeon’s anal injections of antiseptic gases. (3) Inhalation of volatilised iodine. (4) The inhalation of, and hypodermic injections of Eucalyptus oil. Sterilised Cotton- wool Plug. Fig. 10. —The destruction of Bacillus tuberculosis by COg and HgS gases {not drawn to scale). A = gas-bag (capa- city about 4 litres) filled with COg gas; B= bottle con- taining Eaux Bonnes water; C = sterilised cotton- wool plug; D = tube containing a pure cultivation of Bacillus tuberculosis in blood serum, inoculated from the sputum of Mr Snodgrass. (The tube D is after Nature.) H2S+CO2, Dr Bergeon gives no details of having studied the action of his gaseous antiseptics on the vitality of Bacillus tuberculosis., therefore it is on this point my next remarks will he directed. Bergeon recommends the anal injection of 4 litres of the mixed gases (see his paper, loc. cit.) ; the time prescribed for injecting this quantity is 20 minutes. I prepared pure carbon dioxide (from the decom- 52 Proceedings of Royal Society of Edinburgh. [dec. 5, position of pure sodium bicarbonate by means of pure dilute sulphuric acid) and filled a 4-litre bag with the gas. This gas was passed through a half-bottle of Eaux Bonnes water (thoroughly impregnated with the H2S gas), and then allowed to pass into a pure cultivation of Bacillus tuberculosis (fig. 10). After all the gases had passed through the cultivation, the tap E (fig. 10) was turned off. Ten tubes containing sterilised blood serum were inoculated from the growths which had been submitted to the Fig. 11. — Action of H2S and CO2 gases directly upon the bacilli in fresh human sputa. A = a drop of human sputum adhering to the cover- glass. action of the gases. The tubes so inoculated were then placed in the incubator at a temperature of 37° C. After forty days’ incuba- tion, no signs of any growths made their appearance in any of the Fig. 12. — Bacilli in Sputum. Case of Miss Green- White. Stained by the Koch-Ehrlich Method. X 750. tubes. These experiments were repeated a second time with similar results. Again, the gaseous antiseptics were allowed to pass for 15 minutes 887.] Dr A. B. Griffiths on Micro-Organisms. 53 into a little glass cell (fig. 11), containing upon the internal surface of the cover-glass (A) a drop of sputum. After allowing the gases to pass through the little cell, the cover- slip was then transferred to sterilised blood serum, and after an incubation of twenty-six days no growths of the Bacillus tuber- culosis (or putrefactive microbes) made their appearance. This experiment was repeated in duplicate with the same results. The sputa used were obtained from Mr Snodgrass and Dr Wood of Bromsgrove. Dr Wood’s tube came to me labelled ’‘^Expec- torated^ May 29, 1887. Girl named Miss Green-White. Incipient phthisis ; night siveats, and harsh breathing under the clavicles.'^ An examination of this specimen of sputum gave numbers of bacilli (fig. 12). From the above experiments, I have reason to conclude that Bergeon’s sulphuretted hydrogen gas is a destroyer of the vitality of Bacillus tuberculosis and its spores. {a) The Case of Mr Snodgrass. A practical Trial of the Bergeon and Griffith^ Methods of treating Phthisis. Mr John Snodgrass, jun., of Glasgow, wrote to me in February of the present year (1887), after reading an abstract of my paper (read before your Society on January 31, 1887) in the Glasgoio Fig. 13. — A = a tin vessel containing water (which is kept near its boiling point) ; B = a small vessel capable of floating in water. This vessel contains tincture of iodine (J oz. of tincture of iodine and ^ oz. of water are used every time); C = tube (of indiarubber) ; D = mouthpiece ; E = spirit- lamp) ; F = a tripod stand. Herald^ and from that day a scientific correspondence has been kept up between us. His case is that of lung disease of thirteen years’ standing, which became distinctly tubercular several years ago. Ever since the discovery of Koch’s bacillus, Mr Snodgrass has tried various devices (of his own) for destroying the microbes in his own 54 Proceedings of Royal Society of Rdinhurgh. [dec. 5, lungs. Amongst these experiments he has used volatilised iodine in the following way: — The apparatus (fig. 13) explains itself, and is very simple. According to Mr Snodgrass, the patient should, if possible, inspire gently by the mouth (from the mouthpiece D) and expire by the nose, taking as full and as deep inspirations as possible. He con- siders that although the iodine may not reach very deeply into the lungs, it will cleanse the throat, larynx, trachea, and the large bronchi. Concerning the value of his device of inhaling volatilised iodine, he says — “ The inhalations of iodine have certainly put the hand hack on the dial in my case for nearly two years.” Concerning the adoption by Mr Snodgrass of Dr Bergeon’s and also my process of hypodermic injection of a solution of a salicylic acid, I have his permission to make free use of his letters, in which he describes the experiments performed and results obtained. Abstracts from these letters I shall give as an appendix to this memoir. By so doing, it will make the paper far more valuable (as they come from a literary man, and a man of sound common sense) than any written description I could give of his trial of the two methods. Mr Snodgrass firmly believes in the value of both methods, and much good has been done by using them. When he wrote to me in February (1887) he was apparently a dying man. He greatly improved by using the methods j in fact, so much so that he was able to leave Glasgow and spend the summer in the Kyles of Bute. During his experiments, I have reported on many occasions the microscopical appearances of specimens of sputa received from him, and it was surprising to note from time to time the decreasing numbers of bacilli present. Although there appears, in the last specimen of sputum received on 28th September 1887, to be an increase in their numbers (due to the fact that Mr Snodgrass has for a short time desisted from using the methods, owing to great physical weakness), there was no increase in the quantity of Freund’s cellulose in the sputum ; showing the inactivity of the bacilli present. In fact, their “ pathological power ” appears to be proportional to the quantity of cellulose found in the sputum. At this point I will refer you to the appendix of this paper, where Mr Snodgrass states in his own words the work that has 1887.] Dr A. B. Griffiths on Micro-Organisms. 55 been done to prove that if consumption is curable at all, it must be done by injection methods of some germicide, either in the gaseous or in the liquid condition. It will he remembered that the concluding words of my first paper on this subject {Proc. Roy. Soc. Edin.. vol. xiv. p. 97) were the following: — “ I have reason to conclude that it may he, with a more extended study of the action of this solution of salicylic acid upon disease ‘ germs ’ and their organisms, we have the most rational mode of treating those contagious diseases whose seat of energy is in the blood.” Perhaps a better germicide than salicylic acid may be discovered, yet it is the principle of the method that is so important. In every case of disease produced by living microbes residing in the blood the most rational and scientific method of treating such diseases would be to destroy the microbes in situ by injections. "When the microbes are destroyed, nature will have a chance of repairing the damage done. It is with this end in view, that forms the basis of the present researches, and of my method for treating contagious diseases. The germicides used in connection with my experiments upon Sarcina lutea, Micrococcus prodigiosus, Micrococcus tetragonus, all destroy Bacillus tuberculosis. In fact, I may say that Mr Wm. Thomson’s* sodium fluosilicate is a very powerful germicide. According to Mr Thomson, F.E.S.E., it is not poisonous, and is inodorous. A saturated aqueous solution contains 0*61 per cent, of the salt. It does not irritate wounds, and it has “ greater anti- septic power for animal tissues than one part of mercuric chloride in 1000 of water : which is a stronger solution than that which can be generally employed for surgical purposes without producing poisonous effects ” (Thomson). (h) The Kolisdier Treatment of Consumption. In passing I wish to record here, that in June 1887, there were accounts given in the newspapers that Dr Kolischer had recently presented a paper to the Vienna Society of Physicians, on a pro- posed method for treating and curing consumption and other * A paper read before the British Association, August 1887, and published in the Chemical News, vol. Ivi. p. 132. 56 Proceedings of Royal Society of Edinhurgh. [dec. 5, tubercular affections of the lungs, or other parts of the body. Dr Kolischer, starting on the assumption that tuberculosis occasion- ally heals naturally, owing to the tubercules being calcined,” hit upon the idea of causing artificial “ calcination ” by means of hypodermic injections of a substance, described as calcium plios- phoricum^ into the limbs of persons affected with tuberculosis. He made a number of experiments with a view to testing his discovery, and in every case the experiments turned out successful. (c) M. BalVs Treatment of Phthisis hy Injections of Euccdyptus Oil. Eecently M. Ball (Membre de TAcademie de Medecine de Paris) read a paper before the Paris Academy of Medicine, stating what he considers to be a cure for consumption, namely, by injections of eucalyptus oil under the skin. I may say here, that Mr Snodgrass has used eucalyptus oil volatilised by heat and inhaled ; and he says : — proved very irritating.^ and I had to desist. I greatly prefer salicylic acid injections to it.^’ (cl) Dr Theodore Williams’ Observations on the Influence of certain Substances on the Growth of Bacillus tuberculosis. In a paper,* kindly sent to me by the author, there are detailed a number of experiments with different reagents on the bacillus of phthisis. Dr Williams found that arsenious and boric acids “ exer- cised no destructive influence on the bacilli ” of consumption. He found that, with solutions of quinine sulphate (2 grains to 10 grains in an ounce of water) in each case the number of bacilli decreased rapidly under its influence, and “ that the bacilli in the sputum after being mixed with theguinine salt could not be cultivated even in beef- broth.” The experiment shows that quinine sulphate is a destroyer of Bacillus tuberculosis. Dr Williams also found that iodine (1 part in 12 of water) reduced the numbers of bacilli, and prevented spore-formation. Mercuric chloride (1 grain to an oz. of water) caused no diminution, but rather an increase of bacilli-spore-forma- ■* ‘ ‘ Observations on the Influence of certain Culture Fluids and Medicinal Reagents in the Growth and Development of the Bacillus tuberculosis fl by C. T. Williams, M.A., M.D., F.R.C.P., Physician to the Hospital for Consump- tion, Brompton, Proc. Pwy. Soc. [Ho. 231], 1884. 1887.] Dr A. B. Griffiths on Micro-Organisms. 57 tion was very marked in the mercury solution. This fact confirms Herroun’s investigation on the value of mercuric chloride as an antiseptic agent. XII. Bacillus tuberculosis a Parasite. From Dr V. Cornil’s researches {Bulletin de V Academic de Aledecine de Paris, 1883) it has been shown that Bacillus tuber- culosis is of a parasitical nature. The microbe is found in the giant cells of the tubercle, also in the colourless blood corpuscles ; and therefore it is to be detected in all organs in which a tubercle can he developed. It passes to the kidneys, and according to Babes {Centralblatt filr d.AIed. Wissensch., 1883, p. 145) has been found in the urine. Recently, M. V. Galtier {Comjptes Rendus, vol. civ. Xo. 19) has shown that whey and cheese from the milk of tuberculous cows often contain the bacilli of phthisis. He has also demonstrated that swine and poultry fed upon dairy produce of this character may contract phthisis. Their flesh may then in turn impart the disease to man. Galtier’s observations appear to explain the hereditary nature of consumption. It is apparently a blood disease of slow growth ; and if milk be a meffium in which the bacillus is capable of living its life-history, we can well understand a phthisical mother suckling her child giving the disease to the child. It will be remembered that Dr Klein has proved the presence of Micrococcus scarlatinae {The Times, 28th May 1887) in the milk of those cows suffering wdth certain diseases of the udders and teats. Therefore, if milk is a nutritive medium for infectious diseases, ' it would he better in every case to nearly boil the milk before using it. Conclusions. 1. It has been proved beyond doubt that microbes are the real cause of certain contagious diseases. 2. In many cases these microbes are capable of being destroyed by various germicides. Therefore, by further investigations, we ought to discover a germicidal remedy for such terrible scourges to humanity as consumption and syphilis. 58 Proceedings of Boyal Society of Edinburgh. [dec. 5, 3. It lias been shown from the researches detailed in this paper that the vitality of Bacillus tuberculosis is considerable, and that it is capable of being dried up in the atmosphere for many weeks without its vitality being impaired. 4. That Bacillus tuberculosis is capable of being disseminated by envelopes coming from phthisical patients. 5. That the electric current destroys the vitality of certain microbes. 6. That a new bacterium is the cause of putrefaction in the onion, liberating as a product of its life-history small quantities of H2S gas. This new microbe I have ventured to call Bacterium allium. 7. That the soluble zymases secreted by living microbes are capable of being destroyed by germicidal agents. Hence, if destroyed, they are incapable of producing chemico-pathological changes in the blood and tissues. 8. The most rational method of treating contagious diseases is by injection of some germicidal agent, either in solution or in the gaseous state. By destroying the microbes, the disease would be at an end. 9. The germicidal agents used for injection purposes must not produce poisonous actions upon the blood and tissues, yet at the same time must be powerful enough to destroy the vitality of the microbes and their spores. 10. It is upon the lines indicated in this memoir that the physi- cian in the future must look for a scientific method of treating those contagious diseases whose microbes reside in the blood. I wish to tender my best thanks to those friends who have ren- dered me assistance during these investigations, but more especially to Mr John Snodgrass and Dr Wood. Appendix. The following are abstracts from Mr Snodgrass’ letters : — (1) Letter of "l^th February 1887. — “ Some time since I obtained remarkable results by a simple process I devised for the inhalation of the volatilised vapour of iodine, with satisfactory results.” (2) Letter of Zrd March 1887. — “The result of inhalation of 1887.] Dr A. B. Griffiths on Micro-Organisms. 59 iodine three days ago has been to cleanse the lung, and to bring away the debris found in the sputum this forenoon. This is, of course, a good result. The iodine inhalation caused headache and considerable depression of heart.” (3) Letter of Wi March 1887. — Since I last wrote to you I have twice injected a 15 minim solution of the acid. The strength was 2 J grains of salicylic acid to a fluid drachm of water ; but the acid was mixed with an equal quantity of borax. My immediate reason for injecting the solution (which would roughly contain fth of a grain of the acid) was a severe attack of rheumatism, of the kind that often accompanies phthisis. The rheumatism disappeared almost entirely I may mention, that before making the injections there was a large deposit of uric acid on the urine. This has quite disappeared, at least to the naked eye. That the salicylic acid passed through the system I am perfectly certain, as I had the usual headache which follows taking it by the mouth, and the taste — that unmistakable taste — was very apparent next morning on the tongue and palate. I made the injections in the calf of the leg, near a large vein I had (it appears) rightly assumed that a cavity was forming in the lung about the time I first wrote to you. About forty-eight hours after inhaling volatilised iodine a considerable quantity of matter came away, with the usual discoloured blood-clot. This debris on examination, contained an abundance of long fibre One most important part of your paper is, that which deals with the action of the gastric juice on medicines. I swallow a great deal of sputum at times — for it is impossible always to eject the whole. Yet I have every reason to suppose that the bacilli thus swallowed in large numbers pass harmlessly through the alimentary tract without getting into the hlood. In fact, it happens with me and Bacillus tubercu- losis, as it happened with M, Bochefontaine and the Comma bacillus. This thoroughly bears out your most important remark that, ‘ there is no doubt the acid properties of the gastric juice .... had acted upon these micro-organisms,’ &c. In cases where con- sumption of the intestines follows upon pulmonary consumption, the inference will be that the gastric juice is either weak or imper- fectly secreted. With reference to the salicylic acid (without borax), what I think of doing is to dissolve a part of the acid in a 60 Proceedings of Royal Society of Ediribiirgh. [dec. 5, small quantity of hot water ; then, if the water takes up the acid in the proportion of 20 to 1, by injecting 15 minims of the solu- tion before it is cold — say at the temperature of blood heat — I shall get into the system about f of a grain of the acid. Now, the medium dose by the mouth being 10 grains, \ of this would be reckoned safe, or at least not dangerous by injection, consequently I am much within the line of safety.” (4) Letter of Wtli March 1887. — “This forenoon I tried the injection of salicylic acid, and after injecting 5 or 6 minims into the tissue of the left thigh, I had to stop, owing to the pain caused by the acid. Judging by the pain that immediately followed the injecting of the drops of fluid, the solution must have been of considerable strength. The only question is, whether the heat of the body is sufficient to dissolve any crystals that remained in the fluid ? I suspect that this is the great fallacy of administering (say) 10 grain doses of salicylic acid by the mouth. Possibly very little of the acid passes into the blood system, the greater part being carried away in the feeces as insoluble. From this, I am sure your method is on the right lines. The micro-organisms must he reached and must he destroy edi’ At this point Mr Snodgrass uses Dr Bergeon’s method along with the salicylic acid injections. (5) Letter of 2d>th March 1887. — “ Dr Bergeon’s instrument (the only one in Scotland) has been seen by my doctor. It is very elaborate ; we think needlessly so. Briefly, the mode of obtaining the mixed gas is to pass the carbonic acid gas through Eaux Bonnes water. Now, it seems that during five or six trials on a patient at the Western Infirmary (Glasgow) no sulphuretted hydro- gen could be detected being emitted through the mouth, showing that the gas did not permeate the lungs. My notion is that too little Eaux Bonnes water was used, that a fresh supply should from time to time have been put into the jar in which the CO2 passed through the water. Your suggestion of the proportion of three volumes of CO2 to 1 of H2S is very valuable.” Mr Snodgrass and his doctor construct a much simpler apparatus for this gaseous injection than that of Bergeon. (6) Letter of March 1887. — “At this point, I may say extensive damage has been done to the throat, for large ulcers are 1887.] Dr A. B. Griffiths on Micro-Organisms. 61 seen on the hack of it, by merely pressing down the tongue. Two litres of gas (CO2 and H2S) were injected, and the sulphur smell was distinctly perceptible five minutes after beginning the injection. Eaux Bonnes water was not used, as it is difficult, but bisulphide of carbon was employed, instead of the mineral water. Some rather disagreeable symptoms followed the injection ; severe headache, colic pains, weakness, and slight pain of the heart. You will be glad to learn that (now for the third time) relief from rather severe rheumatism followed two injections of salicylic acid.” (7) Letter of Wi A;pril 1887. — “After the last injection of 'gas, I again suffered from severe toxic effects, and during the night was much pained with the unabsorbed gas, and with violent and inces- sant purging Many persons would be far too weak for Bergeon’s method, and for them your salicylic acid treatment would be invaluable, for I have not the least doubt that you have discovered an efficient bacillus-destroyer.” (8) Letter of 10th Axjril 1887. — “To-day I have injected 10 minims of salicylic acid solution. It certainly ‘ bites ’ pretty sharply, so it must have been strong enough. I find it of advan- tage to inject it tepid, as it is then more quickly absorbed ; the swelling going down in a few minutes. There is no doubt that I am much better, as far as phthisical disease is concerned. I can now speak without distress, and breathing is much less laboured. Last night I slept seven hours, a thing that has not happened for months, my usual sleep being a broken half hour several times a night, and in all not more than three hours in the twenty -four. I firmly believe that in many cases Dr Bergeon’s system will not be applicable, and in these cases your treatment would be valuable.” (9) Letter of IZth April 1887. — “ Yesterday I again injected the gases. They produced much pain in the left lung, and also in that small portion of the right lung that is impaired. My doctor thinks your observations of immense importance.” (10) Letter of 27th April 1887. — “For the last ten or twelve days I have suffered a great deal of pain. Large quantities of uric acid are being secreted. I inject salicylic acid occasionally, which has the effect of checking the formation of uric acid.”"^ * See “On some Points in the Pathology of Rheumatism, Gout, and Diabetes,” by Dr P. AV. Latham {The Croonian Lectures for 1886.) — (A.B.G. ) 62 Proceedings of Royal Society of Edinhurgh. [dec. 5, In tlie month of June Mr Snodgrass was well enough to travel to the Kyles of Bute from Glasgow. He writes from there as follows : — (11) Letter of Yitli June 1887. — “My lung trouble has of late very greatly improved ; hut the abdominal mischief has been very severe. During the last two days, however, a marked improve- ment has taken place I have the greatest faith in yours and Dr Bergeon’s systems. But, in my case the disease has apparently gone too far, and the condition of the large bowel is such that there is great risk in applying Bergeon’s method.” (12) Letter of \Wi September 1887. — “You will recollect that you were able to make a most favourable report on the sputum sent to you after the salicylic acid treatment and eight or nine injections of gas by the Bergeon mode of treatment. As far as the lung was concerned, a great improvement took place, and for more than two months — certainly during the whole of June and July — I could not have sent you a typical specimen of sputum. Indeed, during that time expectoration almost entirely ceased (as did also the cough), and what there was, was merely mucous phlegm, such as might be present in a slight attack of bronchial inflammation. Otherwise, however, matters were very bad ; the large bowel was severely ulcerated, and adherent in the ileocoecal region to the wall of the abdomen. I continued to inject salicylic acid after ceasing the Bergeon treatment, but this too had to be discontinued on account of the disordered state of the whole system. One remark- able thing, however, has occurred, — ever since the salicylic acid injections I have had no attack of musmlar rheumatism. “ About a month ago, I again began the Bergeon treatment, but very cautiously, and the operations have been continued. There is undoubtedly some improvement, but, as before, uric acid deposits took place^ and I have not ventured to use the treatment on more than two days consecutively. “ I am afraid it must be admitted that (with me at least) there are certain dangerous symptoms caused by the CO2. Amongst these are, obstinate constipation, great difficulty in expelling the unabsorbed residuum of carbonic acid gas in the intestines, and certain rather alarming head symptoms. “ About the 23rd of June last, Dr Coghill, of Yentnor, com- 1887.] Dr A. B. Griffiths on Micro-Organisms. 63 municated an account of his experience of the Bergeon treatment to the British Medical Journal. He there stated that the results, both in his hospital and in his private practice, were not merely remarkable, hut astonishing. In the same number of the British Medical Journal^ however, a physician of one of the London hospital says that not only negative results, or hut very imperfect results, were obtained from experiments in the hospital he repre- sents. But this scarcely surprised me after the failure at the Western Infirmary (Glasgow). Most likely Eaux Bonnes natural mineral water was used. I found it quite inefficient. Again, in the same journal, another communication gives, as a formula from which excellent results had been obtained, the following : — a satu- rated solution (aqueous) of washed sulphuretted hydrogen ; J to 2 oz. being added to 12 oz. of pure water in the bottle through which the stream of carbon acid gas is passed.” It will he gathered from the experiments of Mr Snodgrass and those of my own — 1. That inhalation of iodine vapour has the property of cleansing the lungs, &c., of bacilli, debris, and Freund’s cellulose. 2. That both the salicylic acid injection method, and Dr Ber- geon’s treatment, are capable of preventing the growth and mul- tiplication of Bacillus tuhercidosis. So much so, that Mr Snod- grass was comparatively free from the phthisical complaint for two months ; although the disease is of longstanding^ and there are little hopes of a permanent cure. 3. That Bergeon’s process has a tendency to greatly increase the formation of uric acid in the urine. 4. That salicylic acid injections lessen the abnormal formation of uric acid in the urine. 5. That in severe cases of phthisis, it is difficult for the whole of the gases (in Bergeon’s treatment) to be absorbed by the intestines, — the unabsorbed gases causing an ulcerated state of the intestines. 6. That salicylic acid injections have the power of completely curing muscular rheumatism of the kind which often accompanies phthisis. 64 Froceedings of Royal Society of EdinhurgJi. [dec. 5, 8. On the Colour of the Skin of Men and Animals in India. By Robert Wallace, Professor of Agriculture and Rural Economy in the University of Edinburgh. Indian cattle have, with few exceptions, jet hlack skins. The hair is frequently white, grey, hrown, or hlack, but only in rare cases are the skins white. The animals having white skins are weakly, if not unhealthy, are liable to blister with the sun, and contract after a time a form of leprosy. Black Skin an Advantage. — This opens up and widens the scope of a most interesting question on the relation of colour to climate, which — I have it on the very highest authority, that of Professor Huxley — is by no means at present understood. The field of investigation as regards India is a large one ; it embraces the human races, and the breeds of cattle, sheep, pigs, buffaloes, and horses. The skins of all, as a rule, are black or dark coloured; the few white exceptions I have noticed particularly in buffaloes and cattle, and in one case of a goat, are as stated delicate. The white or grey hair so prevalent in cattle extends to the Arab horse, and would appear to be, when associated with the black skin, peculiarly well adapted to resist the extreme heat of a tropical sun. It has always been a marvel that the white skin, which on account of its colour does not absorb heat so quickly as a black skin, should not prevail in the human species within the tropics ; and it becomes even more wonderful now, when it begins to dawn upon us, that the skins of the lower animals follow the same great law of nature, whatever that law may be. Black Skin Theory Explained. — It would seem at first sight that the black skin should rather be a disadvantage than otherwise ; but in the reality it is not so. The black colour of the skin causes it to absorb more heat than a white skin, but while it is doing so, at the same time and for the same reason, it is giving off more heat — its absorbing power and also its radiating power being greater. Therefore, when the sun’s rays impinge upon the skin, the heat is rapidly absorbed; but, as the rate of absorption of heat is greater than the rate of radiation, unless the temperature of the skin were lowered by some other influence, the whole surface of the body would become extremely hot. 1887.] Prof. Wallace on the Colour of Skins. 65 To complete the explanation, we must here take into consideration what is known of black-skinned men. Any one who has been in India can see that natives, although they drink water freely, do not appear to perspire so copiously as Europeans, but this is simply because more of the perspiration comes from them in the form of vapour, and less is seen to stand like dewdrops on the surface of the skin. In the evaporation of the moisture exuding from the skin, we have a demand for heat far greater than an ordinary observer might imagine ; and by it can be disposed of all the surplus heat which the black skin absorbs over and above what it gives off by radiation. It is a fact which few realise, that the amount of water is small indeed which, by being evaporated, could transform into its latent condition all the heat derived from the warming influences of the sun in the hottest climates. PRIVATE BUSINESS. Mr D. S. Sinclair, Dr A. D. Leith Napier, and Mr Alexander Galt were balloted for, and declared duly elected Fellows of the Society. Monday, IWi December 1887. Sir DOUGLAS MACLAGAN, M.D., Vice-President, in the Chair. The following Communications were read : — 1. On the Height and Volume of the Dry Land, and the Depth and Volume of the Ocean. By John Murray, Esq., Ph.D. (Published in the Scottish Geographical Maga- zine for January 1888. 2. The Pineal Body {Epiphysis cerebri) in the Brains of the Walrus and Seals. By Prof Sir Wm. Turner, M.B., LL.D., F.K.S. In this paper the author described the pineal body in the walrus and in Phoca vitulina and Macrorhinus leoninus, in which animals, but more especially in the walrus, it is of larger size than is usual in mammalia. In one walrus it measured 30 mm. (IT 8 inch) in VOL. XV. 12/6/88 E 66 Proceedings of Eoyal Society of Edinburgh. [dec. 19, length, and 18 mm. (0’7 inch) in its greatest transverse diameter; in another it was 29 mm. long, 13 mm. broad, and 13 mm. in vertical diameter. It rested on the superior vermiform process of the cerebellum, and was visible between the two diverging hemispheres of the cerebrum when the brain was looked at from above. [The paper is printed in extenso in the Journal of Anatomy and Phy siology, January 1888, and as a part of the “Eeport on the Seals ” collected by H.M.S. “ Challenger,” Part LXYIIL, 1888.] 3. On a Method of graphically recording the exact Time Relations of Cardiac Sounds and Murmurs. By Byrom Bramwell, Esq., M.D., and R. Milne Murray, Esq., M.B. (Printed in full in Brit. Med. Journal, Jan. 7, 1888.) 4. On Benzyl Phosphines. By Professor E. A. Letts and W. Wheeler, Esq. The phosphines have, comparatively speaking, been little studied, and most of our information concerning them is due to the investiga- tions of only two or three observers. There are consequently many points in their history which require examination, and the object we had in making the present research was to extend our knowledge of them as a group. We experimented in the benzyl series for several reasons, among which we may mention the following : — Monohenzyl phosphine is a liquid at ordinary temperatures, whereas the corresponding methyl and ethyl derivatives are gase'ous, hence it is more easily worked with than the latter. Then, again, benzyl derivatives have, as a class, considerable chemical activity ; and lastly, one of us in conjunction with another chemist had already studied somewhat exhaustively the quaternary phosphorised compounds which this radical forms.* Hofmann f was the first to obtain mono- and di-henzyl phosphine, by heating a mixture of chloride of benzyl, phosphonium iodide, and oxide of zinc in sealed tubes. He apparently submitted them to a somewhat cursory examination, and only determined their leading * Letts and Collie, Trans. Boy. Soc. Edin. t Hofmann, Ber. d. d. Chem. Ges. 1887.] Prof. Letts and W. Wheeler on Benzyl Pliospliines. 67 properties. He mentions that bye-products are formed along with them which he did not further investigate. We have repeated Hofmann’s experiments, and have submitted both the mono- and di-benzyl phosphine to a very careful examina- tion. We have also investigated the bye-products which are formed and have determined as far as possible their composition. Preparation of Mono- and Di-Benzyl Phosphine. — Hofmann recommends digestion during six hours at 160° C. of a mixture of 4 parts of oxide of zinc, 16 of iodide of phosphonium, and 12 of chloride of benzyl. Experiments conducted in this way with commercial chloride of benzyl from Kahlbaum gave in the tubes a viscous semicrystalline mass. To obtain a good result, thorough mixing of the materials in the sealed tubes by shaking before heating seemed to be necessary. On opening the tubes much phosphuretted hydrogen escaped, but on heating for a longer period or to a higher temperature, the escaping gas seemed to consist of hydrochloric acid only. It was soon found that at the temperature of 160° C., a great deal of hydrochloric acid is formed, and but little of the primary phosphine. The best results were obtained by a six hours’ digestion of the mixture at temperature of 120° C. Experiments tried at 100° to 110° C. showed that but little of the primary phosphine is formed. With the quantities Hofmann recommends and a digestion for six hours at 120°, the tubes when cold contain a viscous semi- transparent mass, sometimes of a brown colour, sometimes red and opaque from the separation of free phosphorus. Above this a small quantity of a liquid usually floats, which at times is mobile, but at others thick and slightly fluorescent. A few crystals of undecomposed iodide of phosphonium are also frequently present. The liquid floating on the viscous mass (consisting of benzyl chloride, toluol, &c.) we usually poured off, whilst the viscous mass itself we removed by inverting the tubes and blowing a current of steam through them — the operation being so conducted that no air was admitted, whilst the viscous mass (which liquefies when warmed) was allowed to run into a distilling flask — without coming in con- tact with the air. The crude primary phosphine (liberated by the action of water on the product of the reaction) was then distilled off in a current of 68 Proceedings of Eoyal Society of Edinhiirgh, [dec. 19, steam, a stream of carbonic acid passing through the apparatus to prevent oxidation. Under favourable circumstances, from 60 to 70 grms. of the impure primary phosphine were obtained from 360 grms. of benzyl chloride. The residue in the distilling flask usually consisted of a slightly brown viscous mass, which solidified on cooling. It contained the secondary phosphine and several bye-products. The water in the distilling flask along with it also contained phosphorised bodies. After various experiments, we found that the best method for isolating the secondary phosphine is as follows : — The mass is boiled several times with water until the latter ceases to extract zinc salts. It is then boiled with strong potash solution, which removes a further quantity of zinc salts, and after washing with water it is extracted with boiling spirit, in which most of it dissolves ; leaving, however, a small quantity of a black viscous syrup. The alcoholic solution on evaporation yields crystals of crude di- benzyl phosphine, which may be purified by recrystallisation from spirit. The mother liquors on further evaporation yield, in addition to more of the crude dibenzyl phosphine, a viscous substance which contains phosphorised bodies. In addition to mono- and di-benzyl phosphine we obtained the following bye-products : — (A) A crystalline substance precipitated on addition of hydro- chloric acid to the potash solution used to extract the viscous mass containing dibenzyl phosphine. (B) A viscous substance remaining in the alcoholic mother liquors after the di-benzyl phosphine had crystallised out. (C) A solid substance separating from the aqueous solution obtained by treating the contents of the sealed tubes with water. (D) A crystalline zinc salt also contained in the aqueous solution, from which (C) bad been separated. In this paper we shall first discuss the properties of mono- and di- benzyl phosphine, and afterwards the nature of the bye-products. Monobenzyl Phosphine. — Hofmann purified the crude phosphine obtained by distilling the contents of the sealed tubes with water by fractional distillation only. He states that, after two distillations in 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines, 69 a stream of hydrogen, the phosphine is obtained of the constant boil- ing point, 180°. Our own experiments, repeated again and again, and with the greatest care, have satisfied us that the pure phosphine cannot be obtained thus readily. In our attempts to obtain it by simple fractional distillation, we operated altogether upon 50 to 60 grms. of the crude substance. The phenomena observed were much the same in each case. The crude products began to boil at about 100°. The thermometer then rose rapidly to 160°. From 160° to 170° most distilled, whilst from 170° to 190° very little passed over. The residue in the retort usually decomposed suddenly above this temperature, with separation of red phosphorus. All the fractions contained the primary phosphine, for they all had its powerful and characteristic odour, and when mixed with fuming hydriodic acid they all gave the crystalline hydriodate. On repeatedly re- distilling them no substance of constant boiling point could be obtained. Considering that the boiling point of chloride of benzyl is 177°, and that much of that body is contained in the crude phos- phine, it is not surprising that mere fractionation fails to separate it from a body boiling only a few degrees higher. Eventually we decided to separate the phosphine from the crude product by obtaining its crystallised hydriodate, but ownng to the bulky nature of that compound and to its insolubility, we experienced considerable difficulty in effecting this. After several experiments, we found that either of the two following methods may be employed : — (1) The crude phosphine is placed in a retort and a stream of perfectly pure hydriodic acid gas (dried by passing over phosphoric anhydride) is conducted by a long tube into the body of the retort. As soon as saturation seems to be complete the retort is placed in an oil bath, and heated to a temperature of 160° to 180°, a very slow current of hydriodic acid passing all the time. The hydriodate then sublimes in beautiful colourless scales, and when most has thus volatilised into the neck of the retort, the latter is allowed to cool — the hydriodate shaken out, and well-washed with pure benzol. (2) The crude phosphine is mixed with about twenty times its volume of pure dry benzol, and the mixture satured with dry hydriodic acid. It grows w^arm, and eventually almost solid from 70 Proceedings of Boyal Society of EdinhitTgli. [dec. 19, the separated hydriodate. The mass is then thrown on to a linen filter and thoroughly squeezed, then dried between blotting paper, broken up, and washed with benzol so long as the latter dissolves anything. The benzol is then removed by squeezing, drying on filter paper, and exposure of the pounded mass in vacuo. The hydriodate prepared by either of these methods is snow white, and tolerably permanent in air ; but if not carefully prepared, it rapidly becomes brown. Its purity was established by a deter- mination of iodine. The whole of the crystallised hydriodate (about 60 grms.) was placed in a separating funnel, and caustic potash added until the latter was nearly full ; the mixture was then shaken, when the hydriodate rapidly decomposed, and the phosphine separated as an oily layer which floated. It was then decanted and submitted to fractional distillation in a stream of hydrogen. The thermometer rose rapidly to 178°, then slowly to 190°. It was fairly constant from 180° to 182° when most distilled. Only a little passed from 182° to 190°. Fraction 178° to 190° was redistilled. The ther- mometer rose at once to 177°, then rather more slowly to 178°. From 178° to 185° most distilled. The exact boiling point could not be fixed, but most of the liquid distilled from 180° to 183°. These experiments, conducted with the greatest care, and repeated two or three times, appear to indicate that monobenzyl phosphine suffers a slight decomposition at its boiling point, which lies some- where about 180° to 183° C. (uncorrected). Properties. — Monobenzyl phosphine is a colourless, highly re- fracting liquid, possessing a very characteristic and penetrating odour. Its smell remains for days on the hands after operating with it, and in one case the smell was observed months after an instrument had been handled by one of us for some time — the fingers having been previously in contact with a trace of the phos- phine. Exposed to the air, it at once fumes powerfully, and grows very hot. Its vapour, indeed, often inflames on unstopping a bottle containing it. Mixed with fuming hydriodic or hydrobromic acid, it gives a bulky crystalline precipitate of the haloid salt ; and it also, though with more difficulty, gives a hydrochlorate. Hydriodate. — This salt is easily formed either by subliming the 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines. 71 phosphine in dry hydriodic acid gas, — by saturating a solution of the phosphine in benzol with dry hydriodic acid, or by dissolving the phosphine in warm fuming hydriodic acid. By the first method it is obtained in snow-white scaly crystals, like benzoic acid ; by the second, as a seemingly amorphous, bulky precipitate ; whilst by the third, it is also obtained in the crystalline state. A specimen prepared by the first method was analysed — •6090 gave *5630 Agl = -30431 = 49-97 %. calculated per C^H^PH2.H1 - 1 = 50-39 %. The hydriodate, when pure and dry, is permanent in dry air ; but a trace of impurity causes it to become brown. It is rapidly decom- posed by water, and instantly by caustic potash solution, the phos- phine being set free. It is very insoluble in benzol, slightly soluble in ether, and spar- ingly soluble in warm fuming hydriodic acid. Hyclrohromate. — Hofmann could not obtain this compound, but we found that it could be prepared with the greatest ease, either by saturating a solution of the phosphine in benzol with gaseous hydro- bromic acid, or by dissolving the phosphine in the fuming (aqueous) acid. By the latter method it is obtained in scaly crystals, very similar to the hydriodate. Its analysis gave a small deficiency of bromine — probably due to slight deliquescence, or to a trace of impurity — Calculated for Obtained. BzPH2.HBr. Bromine, . 37-9 39-0. The salt is insoluble in benzol, and only very slightly soluble in warm fuming hydrobromic acid. It decomposes rapidly in contact with water, and instantly with caustic potash. Hydroclilorate. — Hofmann did not succeed in obtaining this salt, but it may be produced by similar methods to those which we em- ployed for obtaining the two last-named compounds. On passing gaseous hydrochloric acid into the pure phosphine dissolved in benzol, no effect is produced until the solution is quite saturated. Then white crystalline scales begin to form. 72 Proceedings of Boyal Society of Edinburgh, [dec. 19, On shaking the phosphine with a saturated aqueous solution of hydrochloric acid, a white crystalline precipitate is produced, which dissolves on shaking or on gently warming. Owing to our small stock of the phosphine, we were unable to obtain sufficient of this compound for analysis, but there can be little doubt as to its composition. Action of Bisidphide of Carbon on Monobenzyl Phosphine. — We thought it possible that monobenzyl phosphine might react with bisulphide of carbon, so as to give^a phosphorised sulphur urea, and the following experiments were accordingly tried : — Two grms. of the pure phosphine were sealed up with 2 grms. of bisulphide of carbon, and heated at 120° C. for two days. On examining the tube after heating, the contents were found to consist of a viscous, colour- less substance and a number of colourless needle-shaped crystals. When opened, a considerable quantity of sulphuretted hydrogen escaped. The contents of the tube were treated with bisulphide of carbon, which dissolved the viscous substance, but left the crystals. The latter were repeatedly washed with the bisulphide, then dried, and submitted to a combustion. =0-0118 H =5-42% Calculated Obtained. C7H7PH2.S. C7H7PS. Carbon, . 54‘3 53-85 54-54 Hydrogen, . 5-4 5-79 4-5 Owing to the very small quantity of product at our disposal (about 0-25 grm.), we were unable to examine it further ; hence its composition must remain doubtful, though its formula is pro- bably either one or other of the two given above. The bisulphide of carbon washings from the crystals were warmed to get rid of the bisulphide. A slightly yellow gummy mass remained, which was insoluble in water, alcohol, and ether, and from which no definite product could be obtained by the action of various reagents. It was, however, noticed that boiling glacial acetic acid dissolved it to a certain extent, and it was therefore treated with a considerable quantity of this solvent, in the hope that, if it consisted of two or more products, a separation might be effected. 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines. 73 The hot acetic solution deposited on cooling oily droplets, which eventually formed a viscous mass, exactly like the original substance. The acetic solution decanted from this was evaporated to dryness ou the water-bath, until the whole of the acetic acid had volatilised. There remained a gummy mass not different in appearance from the original body. An analysis was made of this portion (^.e., that which remained dissolved in the cold acetic solution), and also of the portion which the acid had not dissolved. The results show that both portions have the same composition — but no probable formula could be calculated. (A) Portion dissolved hy Acetic Acid. 0*4355 gave { 0*828 002 =0*22582 0 = 51*8% 0*1982 H2O = 0*02202 H = 5*12 %. (B) Portion not dissolved hy Acetic Acid. 0*3643 gave 0*654 002 =0*17836 0 = 51*5% 0*1592 H02 = 0*1772 H = 5*1%. Obtained. a' ^ Oarbon, 51*8 51*5 Hydrogen, 5*02 5*1 Calculated for (C7H7)PH2.CS2. (C^H7)PCS. (C7H7PH)2CS 48*0 57*9 62*1 3*5 4*2 5*5 Oxidation of Monohenzyl Phosphine hy Air. — As before mentioned, the primary phosphine attracts oxygen with great energy from the air. The temperature rises considerably; dense white vapours are produced, and these occasionally take fire spontaneously. The sub- stance which eventually results is a viscous liquid which refuses to crystallise, and could not be obtained in a fit state for analysis. It dissolves somewhat sparingly in water, and behaves as an acid, its solution reddening litmus paper and neutralising alkalies. Its lead salt is easily prepared by adding acetate of lead to its aqueous solution, when a white flocculent precipitate is produced. This is by no means completely insoluble, so that its bulk dimin- ishes considerably on washing. The following results were obtained on submitting this compound to analysis : — 0*1060 gave 0*0925 PbS04 = 0*06319 Pb = 59*6% (C7H^)POPb requires . . . . 60*0 %. These numbers seem to show that the lead salt is a derivative 74 Proceedings of Boyal Society of Edinburgh. [dec. 19, of the oxide of inonobenz5d phosphine PH2(C^H^)0 — in which both atoms of hydrogen are replaced by the metal. The composition of the oxide was also established to a certain ex- tent by the increase in weight which a sample of the pure phosphine experienced on spontaneous oxidation, — a rough experiment giving an increase of 14T% instead of 12 ‘9%, the calculated amount. The authors regret that the small quantity of phosphine at their disposal prevented them from making further or more exact experi- ments. No oxide of a primary phosphine containing a single atom of oxygen has as yet been obtained (excluding the substance under discussion). Action of Bromine on Oxide of Monobenzyl Phosphine. — When bromine is added to the syrupy oxide, it is rapidly decolorised, and the mixture grows hot, whilst hydrobromic acid is evolved, and the pungent odour of bromide of benzyl is noticed. If suffi- cient bromine has been added, a crop of crystals forms after some time, which dissolve both in water and ether, and may be obtained colourless by recrystallisation. These crystals, on analysis, gave the following results: — 0-221 gave 0*166 AqBr” = 0*07065 Br = 31*9 % 0*384 Bromine, . Carbon, Hydrogen, The formula calculated from the analytical results is, as will be noticed, somewhat complex, and at a first glance may appear to be improbable. The following equations show, however, that such a body could conceivably be formed from the phosphine oxide — 0*6245 CO2 =0*18735 C = 48*9% 0*1535 H2O =0*01705 H = 4*9 % Calculated for Obtained. (C7H7)3P 2Br2H30 31*9 31*1 48*9 49*0 4*9 4*7 So>r '^P— 0— H (2) (C^H.K H ^P+3Br2-f3H20-C7H7Br-f5HBr+H3P04 H C7H7 (3) (C7H7 ^p_o_p^C7H7_^ _ C^H^^P— 0-P Br Br 1887.] Prof. Letts and W. Wheeler on Benzyl Phosyliines. 75 Owing to the small quantity of the primary phosphine at our command, it was not possible to obtain sufficient of the brominated body for further experiments. We also tried the action of alcoholic caustic potash and chloro- form on the phosphine, to see whether, under these conditions, a phosphorised nitrile could be obtained but with negative results, the phosphine remaining unacted upon. Our experiments with the primary phosphine are necessarily incomplete, owing to the very small quantity at our disposal, and the difficulty experienced in obtaining it in sufficient quantity. Bibenzyl Phosphine. — This body is readily purified from the crude product described on p. 68 by two or three recrystallisations from boiling alcohol. It crystallises from that liquid in colourless needles, which are quite unchanged by exposure to air. Dibenzyl phosphine is only sparingly soluble in cold alcohol, but it dissolves pretty readily in boiling alcohol. It is readily soluble in chloro- form, and is also soluble both in iodide of ethyl and in bisulphide of carbon. It is insoluble both in ether and in water. Glacial acetic acid is its best solvent. It melts at 205° (uncorrected), and sublimes at a higher temperature, with considerable decomposition. The following are the results of its analysis : — Carbon and Hydrogen — {: T795 gave 1171 H20 = -013011 H= 7-2% 5158 C02= -14067 C = 78-4% Phosphorus — -3645 gave -1963 Mg2P207= -05473 P= 15-0 % Calculated for Obtained. (C7H)72PH Carbon, .... 78-4 78-5 Hydrogen, ... 7-2 7-0 Phosphorus, . . .15-0 14-5 Hofmann could not obtain any salts of dibenzyl phosphine, and considered it to be devoid of alkaline properties. We have found. 76 Proceedings of Royal Society of Edinhurgh. [dec. 19, however, that although a very inert body yet under certain con- ditions, it does combine with the hydracids to form compounds, which, however, are unstable. Among other compounds it forms a very characteristic salt with bromine, and it also combines with chloride of platinum. Platinum Salt. — On mixing alcoholic solutions of chloride of platinum and dihenzyl phosphine a light yellow crystalline powder is produced, hut its composition varies considerably according to the conditions under which it is prepared, as the following analyses show : — (A) (B) (C) Carbon, . 59-5 56-6 58-5 Hydrogen, . . 5-8 5-3 5-9 Platinum, . . 12-8 13*1 13-1 (A) Prepared by mixing cold alcoholic solutions of chloride of platinum and dihenzyl phosphine, and washing the product with alcohol until the washings were colourless. (B) Prepared by mixing very dilute boiling solutions of the two bodies, and repeatedly boiling with alcohol. (The alcohol dissolved a colourless body.) This product was very crystalline, and of a full yellow colour. (C) Prepared as (B), but washed with cold alcohol. It is probable that the products are loose compounds of chloride of platinum and dibenzyl phosphine. The nearest formula for A is 5{(C7H5-)2HP},PlCt^, which requires Carbon, 59*5 Hydrogen, . 5*3 Platinum, 14-0 Action of Hydrol)romic Acid on Dibenzyl Phosphine. — An aqueous solution of hydrobromic acid is without effect on dibenzyl phosphine, but if the latter is dissolved in glacial acetic acid, and the mixture then saturated with hydrobromic acid gas, a crystalline precipitate falls, which usually redissolves as the solution grows warm, and subsequently, when the latter is saturated and has grown cool again, is deposited in small colourless crystals, which, when examined with the microscope, are found to consist of small but perfect plates. 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines. 77 Two separate specimens — A and B — of the compound were pre- pared, each being washed with glacial acetic acid, and finally dried in vacuo over sulphuric acid. B was prepared more carefully, and washed more thoroughly than A. Determinations of bromine gave the following results : — A. B. Calculated for (1) (2) (Bz^HP^HBr 2(Bz2HP),HBr Bromine, 19'5 20-5 16-3 27*4 15-9 The compound is very unstable. It is decomposed by boiling its solution in acetic acid with water, by dissolving it in alcohol, and by an alcoholic solution of potash — dibenzyl phosphine resulting. In the two first cases the decomposition is gradual, in the third it is immediate. Dibenzyl Phosphine and Hydriodic Acid, — On passing gaseous hydriodic acid into a solution of dibenzyl phosphine in glacial acetic acid, a crystalline precipitate is produced, which dissolves as the solution grows warm, and subsequently is again deposited in thin rect- angular plates. The compound, after careful washing with glacial acetic acid, and drying in vacuo, yielded the following numbers : — Iodine, Obtained. Calculated. 2(Bz2HP),HI 22-8 The compound, like the hydrobromate, is unstable, and is decom- posed in a precisely similar manner. Action of Bromine on Dihenzyl Phosphine. — On mixing solutions of the phosphine and bromine in glacial acetic acid, heat is developed, and a light orange crystalline precipitate is thrown down. The compound was prepared several times under varying conditions. For analysis it was thoroughly washed with glacial acetic acid, then dried in vacuo. The following results were obtained : — A. B. C. D. E. F. Carbon, . . . . 56-5 56-5 56*4 Hydrogen, . ... 5-6 5-0 5-3 Bromine, . 26-0 26-4 27-2 26*7 26-6 A, B, C, D, E, and F were separate preparations. At first we naturally anticipated that an addition product had 78 Proceedings of Royal Society of Edinhurgh. [dec. 19, been obtained, as no free bydrobromic acid was observed, but tbis supposition seemed to be negatived by tbe fact that tbe above analyses indicate that for every molecule of tbe pbospbine only one atom of bromine bad been added. Tbis fact appeared to indicate that a substitution product bad really been formed, but as we shall sbow presently tbe reactions of tbe brominated body do not support sucb a view of its constitution. Tbe subjoined numbers sbow that only a very slight difference would exist in tbe composition of these two bodies, and that the analytical results obtained by us would agree almost equally well with either. Obtained Calculated for Mean Eesults. Bz2HP.Br BzgBrP. Carbon, . 56-5 57T 57-3 Hydrogen, . . 5-3 5T 4-8 Bromine, . . 26-5 27-2 27-3 We therefore turned to tbe^ reactions of tbe brominated body for some clue to its constitution. Tbe compound, when warmed with an alcoholic solution of potash or soda, was at once decolorised and a colourless body resulted, which crystallised from alcohol in colourless needles. A specimen of this crystalline body, after several crystallisations from alcohol, gave the following numbers : — Carbon, . . . . 76’5 Hydrogen, . . .6*7 When boiled with spirit it was also decolorised, a volatile body smelling like bromide of benzyl being disengaged, whilst colourless crystals separated from the solution. These after recrystallisation gave an analysis : — Carbon, . . . .76*8 Hydrogen, . . . 6’6 These two determinations proved that the brominated body decom- posed when treated either with an alkali or with alcohol, and we suspected, from the'appearance and properties of the crystalline body formed, that dibenzyl phosphine had been reproduced. The above numbers, however, do not agree with those required for dibenzyl phosphine, and we thought it possible that the brominated body had not been submitted for a sufficient length of time to the action of the decomposing agent. 1887.] Prof. Letts and W. Wheeler on Benzyl Phos'pliines. 79 Accordingly we prepared a new specimen of the brominated body with great care (specimen E). A sample of it was evaporated three times to dryness with alcoholic potash, then carefully recrystallised from alcohol. In appearance the product thus obtained exactly resembled the original dibenzyl phosphine. Its melting point (205°) was the same, and when treated with bromine it gave the charac- teristic yellow compound. On analysis numbers were obtained agreeing with those required for dibenzyl phosphine. Calculated for Obtained. Bzg H P Carbon, . . .78-0 78-5 Hydrogen, ... 7*1 7'0 We are of opinion that these experiments prove that the brominated body is a product of addition, and that the alkali simply removes the bromine ; but as a compound of one atom bromine with a molecule of dibenzyl phosphine is not in harmony with the modern views of atomicity, the following structural formula may be written for the brominated compound ; — Bz — P — Br h/ Bz H H yP— Br / — indicating that the formula for dibenzyl phosphine itself should be doubled, and this would certainly account for its remarkable inert- ness, in which respect it differs from all secondary phosphines hitherto obtained. The compounds which it forms with the hydracids also favour this view of its constitution, for it will have been noticed that both the hydriodate and hydrobromate contain a single molecule of the hydracid for the double molecule of the phosphine. Action of Chlorine on Dihenzyl Phosphine. — Chlorine was passed into a solution of the phosphine in glacial acetic acid, when a white body was first precipitated, but this dissolved up partly, and even- tually a light yeUow crystalline substance separated. Bz^HPCl 14-2 Chlorine, . 12'0 80 Proceedings, of Royal Society of Edinhurgli. [dec. 19, As the quantity of dibenzyl phosphine at our disposal was small, ' we could not repeat the experiment so as to obtain a new quantity of the chlorinated body for analysis, but it can scarcely be doubted that it is of the same nature as the brominated body, and that its formula is in all probability (C7H7)4H2P2Cl2. Examination of Bye Products. — (A) Crystalline Substance 'precipi- tated hy Hydrochloric Acid from the Potash Solution used to extract the viscous mass containing the crude Dihenzyl Phosphine. — This substance was precipitated in crystalline flocks on addition of hydrochloric acid to the potash solution. It was very sparingly soluble in cold water — rather more so in hot water, and crystallised from a boiling aqueous solution in indistinct leaflets. That the substance had acid properties was proved by the readiness with which it dissolved in potash solution and in caustic baryta. A slight residue was, however, left in both cases, and indeed an impurity appeared to be present extremely difflcidt to get rid of, as the following analyses show : — Obtained, K Calculated for (C7H7)20HP0 68-3 Carbon, a) 59-69 (2) 66-3 (3) 67.06 Hydrogen, . 5-8 6-3 6-5 6-1 Phosphorus, 16-2 12-5 12-6 (1) Crude product washed with water. (2) Precipitated from a solution of the crude product in baryta water by hydrochloric acid, and carefully washed (melting point 183° C.). (3) Several times precipitated from baryta solution by hydro- chloric acid, then recrystallised from alcohol and water (melting point 185’5). Barium Salt. — Obtained by dissolving the crude product in baryta water and subsequent precipitation of the excess of baryta by a stream of carbonic anhydride. The salt crystallised from the highly concentrated solution in radiating tuffs of crystals. Obtained. Calculated for f > ■[(C7Hjr)2P02|Ba Barium, .22-2 21 *8 21 -8 Calculated for {C,H,)2P02}^Ba,llH20 23-7 Water. . 24-0 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines. 81 Zinc Salt. — Obtained as a white amorphous precipitate on adding acetate of zinc to a solution of the barium salt : — Calculated for Obtained. {{C^HjisPOj I^Zn Zinc. . . . 12-1 ■ 11-7 Silver Salt. — Obtained by adding a strong aqueous solution of nitrate of silver to a solution of the acid in alcohol, when the salt Calculated for {(C^H, bPO^ }Ag. 30-6 Although the analyses of the acid itself are not very satisfactory, the composition of its salts shows pretty conclusively that it is di- benzyl phosphinic acid. This is also proved by its artificial pro- duction from dibenzyl phosphine, as we shall presently explain. Dibenzyl phosphinic acid has the following properties ; — It is very sparingly soluble in water, but readily dissolves in hot alcohol. Drom a mixture of the two it crystallises in thin scales with mother-o’-pearl lustre. Its melting point is 183°-186° C. Its salts with the alkalies and alkaline earths are readily soluble, whilst those which it forms with lead, zinc, and silver are very sparingly soluble. Production of Dibenzyl Phosphinic Acid from Dihenzyl Phos- phine.— When dibenzyl phosphine is heated with caustic potash or soda it fuses and floats on the surface of the melted alkali. No violent action occurs, but on cooling the mixture and treating it with water, the greater portion dissolves, and acids then precipitate a flocky crystalline substance, which is dibenzyl phosphinic acid, as the following data prove : — Melting point after two recrystallisations from a mixture of alcohol and water, 186° *5 C. Obtained. Calculated. Carbon, . . . 67T 68‘3 Hydrogen, . . 6 '5 6T Lead Salt— Lead, . . . 30*5 29 '7 Barium Salt — (dried at 110° C.) Barium, . . 22-0 21*8 VOL. XV. F separated in thin colourless needles : — Obtained. Silver, . . 30T 82 Proceedings of Royal Society of Edinburgh. [dec. 19, We think it probable that the occurrence of the acid, as a bye- product in the preparation of the secondary phosphine, is due to the digestion of the latter with strong potash solution. B. Solid Substance separating from the Aqueous Solution ob- tained by treating the contents of the tubes with Water. — This separated out spontaneously from the aqueous solution after it had been concentrated, and consisted of a white powder stained brown by iodine. On boiling with dilute spirit the crystals dissolved, and on cool- ing the solution deposited colourless very thin plates, which, when dried, had the lustre of mother-o’-pearl. An analysis was not made, as the melting point, reactions, and appearance of the salts which the substance yielded proved suffi- ciently that it was dibenzyl phosphinic acid. It may here be mentioned that this body is also produced when chloride of benzyl and phosphonium iodide are heated alone in sealed tubes, the product of action being afterwards treated with water. C. Viscous Substance remaining in the Alcoholic Mother Liquors after Dibenzyl Phosphine had crystallised out. — This was squeezed out through a linen filter from the crude dibenzyl phosphine, and was obtained in considerable quantity. Various experiments were tried with it, but without getting any satisfactory results. At last it was found that boiling water dissolved a portion, and that in concentrat- ing an oily liquid separated out, which partly solidified on cooling. By draining the crystals thus obtained on filter paper, and re- peated recrystallising them from hot water, a product of constant melting point, viz., 104° to 105°, was obtained. On analysis the following numbers were obtained : — I. II. III. Carbon, . .71-5 71 *3 71*7 Hydrogen, . . 6'86 6'96 6*78 The body had the following properties : — (1) When heated most of it appeared to distil unchanged, but at the same time an odour similar to that of the primary phosphine was noticed. (2) When its aqueous solution was mixed with caustic potash. It was precipitated unchanged. 1887.] Prof. Letts and W. Wheeler on Benzyl Phosphines. 83 (3) Chloride of platinum gave no sparingly soluble or crystalline compound. (4) Bromine vapour liquefied it, and an additional quantity of bromine gave an oil which partly solidified. (5) Aqueous solution of mercuric chloride gave an immediate flocculent white precipitate. (6) Iodide of zinc gave an uncrystallisable oil. (7) Iodide of cadmium gave with a dilute solution a white crystalline body soluble in boiling water, which crystallised as the solution cooled in minute square plates. On analysing this compound, the following numbers were ob- tained : — Iodine, . . . . 31-5 % Cadmium, . . . 12*3 % If it be assumed that the percentage of iodine was correctly determined, the calculated percentage of cadmium amounts to 13 '9, and the percentage of iodide of cadmium to 45*4. On the assumption that the compound contains a single molecule of the latter, and two molecules of the phosphorised body, the molecular weight of the latter amounts to 220° C. As the quantity of phosphorised body at our disposal was exceedingly small, we were not able to make any determinations of phosphorus. Arguing, however, from the percentages of carbon and hydrogen, and from the molecular weight deduced as above, we find as the most probable formula for the body C^gH^gPO, as the following numbers show : — Obtained. Calculated for C13H15PO . Molecular weight, . 220 216 % Carbon, . . . 7L5 71 ’5 % Hydrogen, . . 6*9 6*9 Its reactions resemble those of a tertiary phosphine oxide j for instance, its ready solubility in water, the fact that it distils almost unchanged, that its solution is precipitated by potash, and that it unites with iodide of cadmium. We hesitate, however, to express any positive opinion with regard to its formula or constitution, as we have not suflicient data for a complete argument. 84 Proceedings of Royal Society of EdinhurgJi. [dec. 19, The body is possibly a tertiary phosphine oxide containing both aromatic and fatty radicals. We have repeated some of the experiments described in this paper, and hope to be in a position to publish the results shortly. 5. A Criticism of the Theory of Subsidence as explaining the Origin of Coral Reefs. By H. B. Guppy, Esq,, M.B., RK Communicated hy Dr H. Mill. Published in the Scottish Geograjphical Magazine. 6. On the Compressibility of Water and of Different Solutions of Common Salt. By Prof. Tait. [Ahstract.) Within the limits of the experiments, which were for t from 0° C. to 15° C., and forp from 1 to 3 tons-weight per square inch, it was found that the average compressibility of water per atmosphere may be fairly represented by 0-28 (36+p)(150 + ^)’ a formula which, while very convenient for application in the hydrostatic equations, extends to the whole range of temperature and pressure ordinarily occurring in nature. Some speculations, connected with Laplace’s theory of Capillary Action and with the Kinetic Theory of Gases, are given as to the meaning of the 36 tons-weight per square inch which occurs in the formula. Similar experiments made on solutions of common salt, of various strengths up to saturation, give analogous formulae. As a rough indication of the results, it may be stated that at 1° C. the average compressibility per atmosphere for the first 150 atmospheres is somewhere about 0-002 40-l-s where s is the mass of salt dissolved in 100 of water. 1887.] J. T. Bottomley on a Tractical Air Thermometer. 85 Friday, January 6, 1888. Sm WILLIAM THOMSON, F.R.S., President, in the Chair. The following Communications were read : — 1. On a Practical Constant- Volume Air Thermometer. By J. T. Bottomley, Esq. In the fourth Memoire of his celebrated Relation des Experiences, published in 1847, Begnaulb gives cogent reasons for preferring the air thermometer before any other as the instrument by means of which temperatures may be defined, and high temperatures deter- mined. The thermodynamic researches of Sir William Thomson have furnished an absolute thermodynamic definition of temperatures ] and the experimental researches of Dr Joule and Sir William Thomson have established the practical agreement of Eegnault’s air ther- mometers with the thermodynamic scale of temperatures. Lastly, the air thermometer is at present the only instrument, with the excep- tion of a mercurial thermometer which has been compared with an air thermometer, by means of which temperatures higher than, say, 150° C. or 200° C. can be determined within 3° C. or 4° In experimenting on the resistance of platinum and carbon fila- ments at high temperatures, in connection with a research on ther- mal radiation with which I have been engaged, I have used air thermometers of various forms ; and I have recently been using a constant-volume air thermometer, which I first described to Pro- fessor Gray of University College, Bangor, just two years ago (January 1886), and'partially constructed for him at that time. It is this instrument, greatly improved as to practical details, which I now desire to bring before the Koyal Society. The best known constant-volume air thermometer is that of Jolly of Vienna. It is a convenient instrument, and is fairly accurate for moderate temperatures ; but for high temperatures a correction, which it is necessary to apply on account of expulsion of air from the heated part of the thermometer, becomes serious, at any rate * Mr H. L. Calleiidar has proposed to use the resistance of platinum for thermometric purposes ; hut in this case also the final standard of reference is the air thermometer. 86 Proceedings of Eoyal Society of Edinburgh. [jan. 6, with the dimensions commonly given to the instrument. It has also some other defects, among which may he mentioned difficulties as to the capillary surfaces of the mercury, want of flexibility or adaptability for various positions, and the proximity of the mano- metric column to the heated regions. The modiflcations which I have made in the construction of the air thermometer have a threefold object, one part of which is to improve on the accuracy of the instrument, and reduce to the minimum that is practicable the correction above referred to for the air expelled by heat from the thermometer bulb or air reservoir. A second object is to increase the range of the instrument by giving it a form in which the hard Bohemian glass can be used in the con- struction of the part to be heated. The third object is to make that part of the thermometer which is to be heated, and which, in the use of the instrument, must be put in position with other pieces of experimental apparatus, of such a form as to be easily handled. 1888.] J. T. Bottomley on a Practical Air Thermometer. 87 For all these objects I find it most convenient to construct separ- ately the manometric columns, and the air reservoir with its volume indicator; connecting these two parts of the instrument only by flexible tubing. This arrangement necessitates an apparatus for regulating the pressure under which the air in the thermometer is maintained. The complete instrument is shown in fig. 1. A is the air reservoir and volume indicator, B is the manometric gauge, and C is the pressure apparatus. The air reservoir and volume indicator I shall call for brevity the volume gauge. It is made in two forms (figs. 2 and 3) — the form shown in fig. 2 for the lower, and the other for the higher temperatures. The bulb a, which is generally either globular or cyhndrical, is connected by a very fine capillary tube e with a some- what wider tube d. At &, V there are two cylindrical bulbs of the same size. The tubes dd and d'd' are of precisely the same diameter, being cut from the same length of uniform glass tubing. The 88 Proceedings of Royal Society of Edinhurgli, [jan. 6, diameter of this tube is about 1 mm. It is such tubing as is used for the fall tubes in a Sprengel pump, t and t' are two stop-cocks ; t being a tbree-way stop-cock, connecting together the volume gauge, the manometric gauge, and the pressure pump ; and t' is a stop-cock used for adjusting the quantity of liquid in the volume gauge. The object of the two cylindrical reservoirs h and h' in the volume gauge is to give space into which the air in the bulb a may expand during heating, or in which a supply of the air may be kept during the cooling of the thermometer. The tube d is very small in capa- city in comparison with the bulb ; and were it not for these reser- voirs, a very small change in temperature would cause the air to be driven out round the bend of the U, or the liquid in the bend to be drawn over into the bulb, unless the observer were incessantly on the watch to prevent this occurring by regulating the pressure. The U of the volume gauge is filled so full of liquid that the equilibrium reading is taken at the points pp of the tubes d and d' ; and both in the selection of the tubes c and d, and in the glass- 1888.] J. T. Bottomley on a Practical Air Thermometer. 89 blowing at the junction, as well as in tlie adjusting of the quantity of liquid in the bend, the endeavour is made to keep the volume of the air-space between the bulb and the point p as small as possible, consideration being given to the capillarity of the tube d. Either mercury or sulphuric acid may be used in the volume gauge. I prefer sulphuric acid on account of its smaller density. The great- ness of the density of mercury, and the uncertainty of its capillary action, make its use very liable to produce serious errors in reading. But, on the other hand, in the case of sulphuric acid, the wetting of the tubes, which constitutes its advantageous quality so far as capil- larity is concerned and gives regularity of capillary action which mercury never possesses, renders watchfulness necessary to keep the acid well clear of the fine tube c. If once the acid is allowed to enter that tube, it tends to make its way along it towards the bulb. The manometric tube is simply a U tube capable of giving a difference of levels of from 100 to 150 centimetres of mercury, and wide enough to make capillarity very small and difference of capillarities in the two tubes negligible. With a tube giving a difference of levels of 150 centimetres, a temperature of about 550° C. may be reached, starting with air at normal density at common temperatures. The difference of levels may be read by means of a kathetometer, or, what is preferable, the tubes themselves may be graduated to millimetres. The tubes which I use are graduated from a zero line which is at the middle of the long branch of the U (see fig. 1). The longer tube is numbered upward, and the shorter downwards from the zero line, and the mercury is filled in so as to stand at the zero in both branches when there is no difference of pressure, and thus the sum of the readings of the two tubes is equal to the difference of pressures when any difference of pressure exists. The pressure apparatus consists of a simple pressure syringe which forces air into a small air-bag of india-rubber fortified with canvas. The air-bag is placed between two boards, which are con- nected by a leather hinge, and pressed together by means of a nut which works on a wooden screw. The air-bag is also connected by means of a T tube with the three-way stop-cock t' ; and, by means of this stop-cock, presses both on the liquid in the volume gauge, and on the shorter column of the manometer. The india-rubber tubes 90 Proceedings of Royal Society of Edinburgh. [jan. 6, used for these connections require to be strengthened with canvas to resist the pressure. The form of volume gauge shown in fig. 3 is designed for use at very high temperatures. It is made in two parts, which are con- nected together at the cup e (shown enlarged, fig. 4). The bulb and tubes c, 6^, and h, are made of hard Bohemian glass; the remaining part of the gauge is of German glass or English flint glass. The stopper of the cup e is made to fit the throat of the cup closely, and just below the throat an enlargement / is blown out, through which the elongated point of the stopper passes. The stopper Is fastened air-tight into the cup with German “ Siegel wachs ” ; and the object of the enlargement is to furnish a cushion of air which prevents the liquid of the volume gauge from coming in contact with the cement. The making of this joint is a little troublesome, and it requires to be protected against radiation from the hot source. There are various stoppers and joints well known, which prevent leakage inwards from without; but it is much more difficult to find an efficacious stopper which will act against pressure from within outwards. The thermometer bulb is filled with per- fectly pure dry air, and it is desirable to have the bulb filled with such a quantity of air that the pressure is approximately that of a normal atmosphere when the temperature is freezing. For, if the quantity of air be con- siderably greater than corresponds with this condition, there is a loss of range in the instrument ; whereas, if there be but a small quantity of air, there is a tendency for the liquid of the volume gauge to be drawn over into the bulb when the temperature of the room comes down (as in winter it may) to about the freezing point, unless the instru- ment be left with the three-way stop-cock closed and the air under diminished pressure. For special circumstances the quantity of air may be made to suit the conditions; for, as Kegnault has 1888.] J. T. Bottomley on a Practical Air Thermometer. 91 shown, the results obtained from the instrument are but very slightly affected by the initial pressure of the air, and this with very wide limits ; and by commencing at common temperatures with air of small density, very low pressure, the upper limit of the range may be extended without increasing the length of the mano- metric tubes. , The filling I accomplish in the following way : — The proper quantity of liquid is first introduced into the volume gauge, and the stop-cock t helps in introducing the liquid and in adjusting the quantity. For this and the subsequent operations I use a really good Bunsen water aspirator, with a Woulfe’s two-necked bottle interposed between the aspirator and the work, and a good length of small bore non-collapsible india-rubber tubing. With the india- rubber tubing the apparatus to be exhausted can be turned into any required position while the exhaustion is being carried on, and air-bubbles can be got rid of with ease. When the volume gauge has been supplied with liquid, I connect the three-way stop-cock t' to the aspirator, and draw the whole of the liquid up into the bulb h' and the tube leading up to the stop-cock itself. The size of the bulbs and of the tubes is, as has been explained, such that when this has been done the bulb h is empty as well as the tubes on the left-hand side of the gauge almost down to the bend. The three-way stop-cock is then closed, and the aspirator disconnected. I now, with the help of a temporary three-way stop-cock, connect together the tail piece of the bulb shown in fig. 3, the aspirator, and a train of drying and purifying tubes (sulphuric acid and caustic potash). The arrangement is such that, on turning the tap of the three-way stop-cock into position No. 1, the aspirator draws the air out of the bulb ; while, on turning it into position No. 2, air flows into the bulb passing through the drying tubes. The bulb is emptied and refilled many times ; and during this process the bulb and all the tubes are heated with a Bunsen flame very nearly to the melting point of the glass.* When it is perfectly certain that there is nothing but pure dry air in the bulb and tubes, * By this process every trace of moisture and condensed air is driven up from the walls of the tube ; and, the bulb being filled with perfectly dry air, it seems certain, from the experiments of Bunsen and from experiments which 92 Proceedings of Royal Soeiety of Edinlurgh. [jan. 6, these are allowed to cool with free passage to the atmosphere through the drying tubes. The bulb is then surrounded with broken ice, and the three-way stop-cock t' is opened. The liquid of the volume gauge now finds its level ; and, noting the barometer roughly (merely to know approximately the pressure), I seal the tail- piece at the extremity. The bulb now contains about the quantity of air required, and it is now only necessary to remove the tail-piece. For this purpose the ice is taken away, and the liquid of the gauge is once more drawn back to a considerable extent, thus making a partial vacuum to avoid blowing out of the air during sealing. The blowpipe flame can then be applied, and the sealing finished off as in fig. 2. Finally, the manometer and pressure apparatus are con- nected to the volume gauge, and the constant of the instrument is obtained by determining the pressure required, including the barometric pressure, to bring the liquid of the volume-gauge into the marked position, first at the temperature of melting ice, and then at the temperature of steam at normal pressure. When read- ing the standard barometer, I also, in accordance with a most convenient suggestion by Professor Quincke, read at the same time my standard aneroid j and this for most purposes, with occasional comparison with the standard, is amply sufficient to give the baro- metric variations. As in the case of the mercurial thermometer, so also in the air thermometer there is sure to be a secular con- traction of the bulb ; and, with the large bulbs used for the air thermometer it is quite possible that the redetermination of the constant of the air thermometer from time to time may be necessary. Convenient formulm for calculating temperatures from the indica- tions of the air-thermometer are easily obtained. Such formulae were given by Jolly [Jubelhand von Poggendorf s Annalen), who also made fresh determinations of the expansion of air and other gases. Some of these formulae are quoted in the Leitfaden der Prahtiselien Physik of Kohlrausch ; but curiously enough there is nothing said in the description of the air-thermometer by Kohl- I have myself carried out, that there is no subsequent perceptible condensation of air at the surface of the glass, such as has sometimes been supposed to vitiate the readings of the air thermometer. Air only condenses on the surface of the glass when there is moisture present — at any rate in such quantity as would he perceptible in a case like the present. 1888.] J. T. Bottomley on a Practical Air Thermometer. 93 rausch, as to determination of the boiling point, the “ ice point ” merely being determined. An experimental determination of each point is, however, absolutely essential. Addition, Juneh, 1888. Shortly after the reading of the foregoing paper, I commenced to use the coal-gas oxygen blowpipe — employing Fletcher’s oxygen blowpipe and oxygen supplied in steel cylinders by the Scotch and Irish Oxygen Company (Brin Process). For convenience these cylinders, with the automatic apparatus supplied by the company for reducing the pressure of the gas, leave nothing to be desired ; and the use of the oxygen blowpipe makes easy and simple many operations which were formerly all but impossible. In particular, the working of Bohemian tubing becomes, without the slightest exaggeration, as easy as that of common flint or soft German glass ; and in addition it is a perfectly simple matter to make a junction between flint glass and Bohemian glass tubing (Bohemian glass does not join well with soft German tubing). Another great advantage in the use of oxygen with the Bohemian glass is, that the glass does not become porcelainised when worked with this flame, as it does when worked with the ordinary flame. With this new power to assist I have now abandoned completely the form of gauge shown in fig. 3, and instead 1 am using a gauge in which the main part is made of flint glass (stop-cocks of Bohe- mian glass cannot, so far as I know, be procured), but in which the air bulb a and capillary tube e are made of Bohemian glass, and the two glasses joined together a little below the bend at the top of the tube dd. I have not yet been able to obtain from any of the first- class makers of Bohemian tube a supply of fine capillary tubes, but this I make for myself by fusing up a piece of thick wide Bohemian tubing and drawing it down. 2. On the Roots of gy Gustav Plarr, Docteur es-Sciences Math. Communicated by Prof. Tait. The imaginaries of Algebra have done good service during the process of discovery of the principles of quaternions. Now that those principles have been founded on the basis of operations on real lines, we must no more admit ^ - 1 as the equivalent of a 94 Proceedings of Boyal Society of Pdinhw^gh. [jan. 6, unit-vector. Vectors, and unit-vectors, by their very definition, represent real lines, each of given direction and length, and a real quantity cannot be represented by an imaginary one. We define the symbol Up to be the operator by which we con- struct the length Tp, so as to give to that length the direction belonging to p. In expressing the vector p by the symbolical product p = Up X Tp , and in attributing reality to p, we are constrained to attribute reality to Up also. We may put the expression of p under the form p = (TJpxOx(^P), where I is supposed to represent the unit line : thus (Up x Z), or simply Up, represents a vector of unit length : hence the denomina- tion of unit-vector given to Up. Tor the proposed equation, we remark first that the symbolic square of the unit-vector 2 is not, properly speaking, a square in the algebraic sense, and that stands for the symbol SS which may be looked upon as a collocation arrived at in the process of the symbolic multiplication of two vectors, which contain the element S. We must therefore refrain from thinking that the element £ can be extracted from the symbol ££, by simply apply- ing to it an algebraical extraction of the square root. There exists a more extensive class of cases, in which we must avoid the introduction of the symbol - 1 in the place of unit- vectors, and for this end we shall propose the adoption of a certain principle which has its analogue in the principle of the inconverti- bility of the order of the factors in a vector product. We have, namely, for the versor q — cos w -f 2 sin , the result Vl = cos w-\-Z sin 10 , Tb where the fraction — is supposed to be reduced to its lowest terms, m n , w = —ti\ — N , m m and 95 1888.] Dr G. Plarr on the Boots of - 1. the integer N being susceptible of taking the m values 0_, 1, 2, . . . . , m - 1 . When the angle u takes the value of a right angle, say then we get W = l7T, (2) = s , so that g becomes the equivalent of what is called a quadrantal versor. In this case we have n S’"" — cos + £ sin , n 27t,^ ^ ^ m m N again taking the m values 0, 1, 2, , (m- 1) . n The second member of this expression of g”^ represents m versors differing from each other, but they all are real quantities (real as opposed to imaginary of the form A + B 1), their axis being the unit-vector g. The first member, that is the expression of the — ^ power of g, m may be considered principally under two forms, and for stating them we will suppose the case : n %i m 2m' -f- 1 2^^' The power ^ ^ of g may be calculated : I. Either under the form II. or under the form 1 \2n' 2m'+l 1 . ' / ’ ^g2n' • In this second case, applying = ( - 1)**', the result will be This is the expression of the roots of the scalar equation: 96 Proceedmgs of Royal Society of Edinburgh. [jan. 6, These roots are evidently, all of them, scalars, of the form A + B^-1, adniitting J-l to be a scalar, in consequence of what is admitted in the theory of biquaternions. We cannot, therefore, accept the expression II. for the repre- sentation of the real versors^ depending on S, which form the second members of the expression of the ^ power of g, and we are consequently led to admit the form I. as being the true expression of that power of £. Generalising this result, we adopt the principle in virtue of /I \ w _1 which must be represented by \g”"j , and not by (g”)”"’ We may apply this principle even to the case ?7^ = 2, 7^ = 2, when we get (2-)2=S, and we have to reject (g^)^ Should - converge towards an incommensurable value t (like that of a surd, &c.), we may still apply the principle by forbidding the separation of t into 2 x when g^ would not be admitted equal to (g^)^; but we might write (g^) =^\ As to the point of view under which the symbolic square (IJp)^ of a unit-vector Up may be looked upon, I would refer the reader to my paper printed in the volume xxvii. part ii. pages 175 to 202, of the Trans. Roy. Soc. Edin., and particularly to the view taken there of treating ii,ij, ih:, ji, &c., &c., each as a symbol by itself, to be determined by the condition that the products of two vectors should remain unaltered, whatever the system of the directions may be by which the components of the vector factors are reckoned. 3. On Vanishing Aggregates of Determinants. By Thomas Muir, LL.D. 1. In a paper‘d communicated to the Berlin Academy on 27th July 1882, Kronecker pointed out that certain sets of minors of * Kronecker L., “ Die Subdeterminanten symraetrischer Systeme,” Sitzungsb. , d. Tc. Akad. d. Wiss., 1882, pp. 821-824. 1888.] Dr T. Muir on Aggregates of Determinants. 97 any axisymmetric determinant are connected by a linear relation, or, as I have tried to put it more definitely in the present title, that certain aggregates of minors are equal to zero. This discovery has attracted considerable attention in Germany, as the list of papers herewith given suffices to show.* The object of the present communication is, in the first place, to point out how much the subject gains in simplicity and clearness, if we consider such identities altogether apart from axisymmetric determinants ; and, in the second place, to direct attention to a new class of identities which have a similar special application, 2. Let us take then any general determinant of the order, but for shortness’ sake let it be written of the 5*^^ order, viz., I 1 • We have clearly at the outset the vanishing aggregate a^ «2 ^3 ^4 ^5 ^2 ^3 ^4 ^5 (Xi «3 «4 5i l>2 ^3 62 &3 &5 Cl Cg Cg C^ C5 - C2 C3 C4 Cg -b ^1 ^3 ^4 ^5 c?i c?2 d^ ^2 ^3 ^4 ^5 c?i cfg d^ d^ Cl 62 H ^4 ^5 ^2 a^ a^ a^ «1 ^2 a^ ^1 ^2 ^3 ^4 &2 ^4 ^5 &1 &2 ^3 ^5 62 63 \ Cl C2 C^ Cg + Cl C2 C3 Cg - Cl C2 C3 C4 di d^ d^ d^ di d^ d^ d^ c?i dc^ c^3 d^ ^3 ^4 If we complete the first column of each of the last five determin- ants by inserting the elements A, B, C, D, the result is still a vanishing aggregate : and if the last row of the second determinant be completed by inserting the elements a, y8, y, S, and if, at the same time, one of these elements be inserted in the place (5,2) of the remaining determinants, no alteration is even then made in the value of the aggregate : that is to say, we have the identity * Runge, C. “ Die linearen Relationen zwischen den verscliiedenen Subdeter- minanten symmetrischer Systeme,” Crelles Joiirn., xciii. pp, 319-327. Mebmke, R. “ Bemerkung iiber die Subdeterminanten symmetrischer Systeme,” Math. Annalen, xxvi. pp. 209, 210. Schendel, L. “ Der Kronecker’sche Subdetermintensatz,” /. Math, u. Phys., xxxii. pp. 119, 120. VOL. XV. 15/6/88 G 98 Froceedings of Boyal Society of Edinburgh, [jan. 6, «1 ^2 «3 0^4 % A ^2 ^3 «4 A «i «3 % h h h h B h h \ h B h h Y ^2 ^3 ^4 ^5 - C ^2 ^4 ^5 + C «i ^3 ^4 ^5 d^ d^ d. D dc) d^ d. d. D d. ^3 d^ d^ ^2 % ^4 ^5 Y a P y 8 ^2 a A % «4 A «3 ^5 A «2 ^3 «4 B h h h h B h h ^3 h B \ ^2 ^3 ^4 C ^2 ^4 ^5 + C ^2 C3 “ C Y ^2 ^3 ^4 D d. C?2 d^ d, D d. d^ d. ^5 D di d^ ^3 d. h 64 y *^5 S Continuing this last process, viz., completing the 5*’" row of the third determinant, by inserting the elements X, Y, Z, and at the same time inserting one of these new elements in the place (5,3) of the remaining determinants, and so on, we finally arrive at the identity «2 «3 ^4 U5 A «2 «3 ^*4 6^5 A U3 «4 U5 h ^-3 ^4 B \ ^3 &4 B h h &4 h ^1 ^2 % ^4 ^5 - C ^2 ^3 ^4 ^5 + C ^3 ^4 ^5 d^ <^2 <^3 d^ C?5 D d. (^3 C?4 d. D di f?3 d. ^5 Y ^2 ^3 64 ^5 a y S a X Y Z A ^1 «2 «4 «5 A ^2 % A % «3 «4 B h ^4 h B ^3 B h h ^>3 \ C C-2 ^4 ^5 + C ^2 ^3 ^5 - C Cl ^2 C3 «4 D d^ C?4 D d. 6^2 ^3 d^ D C?1 ^2 cfg ^3 /? X 9. ^4 y Y 0} % S z involving 5^ + 2 *4 + 3 + 2+ 1 elements, or in general +1) (3/»-2). 3. The connection of this with Kronecker’s theorem is easily made apparent. Take a determinant with umbral elements, and for the sake of variety, let it now be of the 4^^ order, viz. — • 1.1 1,2 1,3 1,4 2.1 2,2 2,3 2,4 3.1 3,2 3,3 3,4 4.1 4,2 4,3 4,4 , ' and let the new elements inserted during the formation of the identity be 1888.] Dr T. Muir on Aggregates of Determinants. 99 1,4 2,4 3,4 5.6 5,7 5,8 6.7 6,8 7.8 so that the identity is 15 16 17 18 14 16 17 18 14 15 17 18 25 26 27 28 24 26 27 28 24 25 27 28 35 36 37 38 34 36 37 38 + 34 35 37 38 45 46 47 48 45 56 57 58 46 56 67 68 14 15 16 18 14 15 16 17 24 25 26 28 24 25 26 27 34 35 36 38 + 34 35 36 37 47 57 67 78 48 58 68 78 A glance at this suffices to show that the second determinant here would be a minor of any determinant of the 8^ order, in which the elements 54 and 45 were identical: that the third determinaot would likewise be a minor of the same determinant, if 46 = 64 and 56 = 65; also the fourth determinant, if 47 = 74, 57 = 75, 67 = 76; and the fifth determinant, if 48 = 84, 58 = 85, 68 = 86, 78 = 87 . Now in the axisymmetric determinant 11, 22, 33,...., 88 all these conditions hold. Consequently the above identity is an identity connecting five of the minors of I 11, 22, 33,...., 88 and this is Kronecker’s theorem. 4. It has been said that the number of elements occurring in the identity is ^(n+ 1) (on - 2), ?i being the order of the determinants involved. When, therefore, the identity is given in connection with an axisymmetric determinant of the ^n^^ order, which, as we know, has 1) distinct elements, it is suggested to inquire which elements of the latter do not occur. Their number evidently is n(f.n + 1) - J(^^ + 1 ) if>n - 2) i.e. + -^n + 1 100 Proceedings of Pioyal Society of Edinburgh. [jan. 6, Now among these the 2n elements of the axis of symmetry (or main diagonal) of the determinant must be included, as from the law of formation of the determinants of the identity their presence in the determinants is impossible. Consequently the number of elements outside the main diagonal of an axisymmetric determinant of the order, which are not involved in Kronecker’s identity, is - |9^ + 1 , Le. \(gi - 2) (n-V) , i.e. 1 + 2 + 3 + . . . . + (gi — 2) . For example, when 2?^ = 6, the number not involved is 1 only: thus, in the case of the axisymmetric determinant of the 6**" order the typical identity is Oj^ ^5 ^6 d^ d^ d^ d^ 65 or D, say. ^4 0^5 Ug Ug <2g ^6 Ug ^4 ^5 ^6 - ^3 ^5 ^6 hg b^ 5g - h ^4 h C4 Cg Cg C4 d^ dg Cg d^ 6q % ^6 ^6 which involves all the elements outside the main diagonal, except a^. 5. Denoting by 1 2 3 4 5 6 that minor of D whose elements belong to the 1®*, 2”*^ and 3’’*^ rows, and 4^^, 5*^ and columns of D, we may write this identity in the more convenient form 1 2 3 4 5 6 1 2 4 3 5 6 1 2 5 3 4 6 1 2 3 4 It is then easily seen that the exclusion of the main diagonal elements is accomplished by making the numbers of the rows (e.g. 1,2,5), different from the numbers of the columns (e.g. 3,4,6) : and that the exclusion of the one other element U2, which occurs in both the 1®‘ and 2'^'^ columns of D, is accomplished by always including the numbers 1 and 2 among the numbers of the rows. 1888.] Dr T. Muir on Aggregates of Determinants. 101 and thereby preventing them occurring among the numbers of the columns. Further, we readily conclude that as there are fifteen elements outside the main diagonal of D, there must be connected with D fifteen identities like the above, the four determinants of each identity being easily got as soon as the element to be omitted has been decided upon. The number of different determinants involved in the fifteen identities is the number of ways in which the numbers 1, 2, 3, 4, 5, 6, can he separated into two sets, and therefore, is ^Cg,3 ^.e. 10. Of these fifteen identities connecting ten determin- ants, it will be found that only five are independent. 6. Further, ''each identity consists of 4 x 6, i.e. 24, terras separable into 12 pairs, the terms of each pair being equal in magnitude, and opposite in sign. Now it is a curious fact, that the twelve different terms are exactly the twelve terms of the Pfaffian got from D, by deleting the elements which are not found in the identity. For example, the twelve different terms of the typical identity given above, are exactly the twelve terms of the Pfaffian, I 0 a^ &3 b^ C4 Cg Cj. ^6 i so that, in fact, the identity may be put in the form (X3 a^ «g ^3 ^4 ^6 «3 «4 a 5 &4 ^5 b^ - h h h + h ^4 h - h ^4 5 C4 Cg dg ^5 ^5 ^6 e, 6 . «3 a, 1 ^5 ^6 • 1 . % «4 h h : h h 1 \ . h h h ^4 ^4 ^5 ^6 d^ d^ cZg f?g % 1 5 Apparently this is equally true for all the higher orders. Thus, in the case of the axisymmetric determinant, 111, 22,33,....,88 L,=, 102 Proceedings of Royal Society of Edinlurgh. [jan. 6, we have 15 16 17 18 25 26 27 28 35 36 37 38 - 14 16 17 18 24 26 27 28 34 36 37 38 -h 14 15 17 18 24 25 27 28 34 35 37 38 45 46 47 48 45 56 57 58 46 56 67 68 14 15 16 18 14 15 16 17 - 24 25 26 28 -f 24 25 26 27 34 35 36 38 34 35 36 37 47 57 67 78 48 58 68 78 14 15 16 17 18 _ 1 . . 14 15 16 17 18 24 25 26 27 28 ' . 24 25 26 27 28 34 35 36 37 38 34 35 36 37 38 45 46 47 48 45 46 47 48 56 57 58 56 57 58 67 68 67 68 78 78 the zero-elements of the Pfaffian occupying as before the places of those elements of the axisymmetric determinant, which do not occur in the identity, and which, in this case, are three in number (v. § 4). 7. The next theorem in regard to vanishing aggregates of deter- minants lends itself readily to formal enunciation. It is as follows : — If any tivo determinants A and B of the n*^ order he taken, and from them two sets of determinants he formed, viz., first, a set of n determinants, each of which is in one row identical with A, and in the remaining rows loith B, and secondly, a set of n determinants, each of inhich is in one column identical with A, and in the remain- ing columns with B, then the sum of the first set of determinants is equal to the sum of the second set. Let the two determinants A and B be ^2 ^3 ^6 ^3 h h \ Cl Cg Cg } C4 Cg Cg then the first set of determinants derived from them is 1888.] Dr T. Muir on Aggregates of Determinants. 103 «3 «4 a^ ttg ^4 ^5 ^6 h h h -t ^1 ^2 ^3 + ^4 ^^5 ^6 C4 C5 Cq O4 Cg Cg ^2 ^3 and the second set O5 ttg Ct^ ^2 ^6 a^ «3 h h h &4 &2 ^6 + &4 h^ Z>3 Cl Cg Cq C4 Cg Cq O4 Cq Cg Expressing each determinant of the first set in terms of the elements of a row and their complementary minors, viz., the first determinant in terms of the elements of the first row, the second determinant in terms of the elements of the second row, and so on, we obtain the nxn terms - «2l^4«6l + “3IM5I. - h,}a^c^\, + c^\aj}^\ - + CjlVsI. But the sum of all the first terms of the expansions is expressible as a determinant of the third order, so also is the sum of all the second terms, and so on ; the result being I I + I I \ 5 as was to be proved. 8. This new identity, it will be seen, depends for its existence on the possibility of a double application of a certain expansion- theorem ; and as this theorem is but the simplest case of Laplace’s expansion-theorem, we are prepared to find that the dependent identity likewise is widely generalisable, so as, in fact, to be co- extensive with the theorem of Laplace. The generalisation is as follows : — If any two determinants A and Bo/ the n^’^ order he taken, and from them two sets of determinants he formed, viz., first, a set of n(n-l) . . . (n-r-f 1)/1.2 . . . . r determinants, each of which is in r roivs identical with A, and in the remaining rows with B, and secondly, a set of the same number of determinants, each of which is in r columns identical with A, and in the remaining columns ivith B, then the sum of the first set of determinants is equal to the sum of the second set. 104 Proceedings of Royal Society of Edinburgh. [jan. 6, Thus taking for A and B the determinants I af^cgd,^ I , I af^d^e^ | , and forming from them the set of six determinants ^2 ^3 «4 ^1 «2 CO «4 ^2 «3 a^ h h h ^8 h h h, h % ^6 ^8 ^1 ^2 ^3 ^4 «5 ^6 ^8 ^5 ^6 ds j d^ d& dr^ C?8 j d. C?2 <^3 d. «6 a^j «8 «5 Uq a>j «8 h h h h ^4 h b, ^8 ^1 ^2 % ^4 ^5 ^6 ^8 ^2 ^3 ^4 C?6 dfj ds j d. (^2 C?3 d^ j d. ^2 ^3 d. by repeatedly choosing two rows from A and two from B, we then expand each of the six in terms of minors of the second order and their complementary minors, the minors being formed in every case from the rows originally taken from A. There will be six terms in each of the six expansions ; so that if we write the second expansion under the first, the third under the second, and so on, we shall have a collection of thirty-six terms which may be viewed as arranged in six columns as well as in six rows. Expressing the sum of the terms in each column as a determinant, we find for the total \af^c^d^\ -1- + Vhh^id^\ -t- \af>.2^^d^ -p \af>2^'^d^ + Wf^cyi^ , in accordance with the theorem. 9. If A and B be the alternants the identity becomes {aP + + a^d^ -P -l- h^d^ + c^df\a%'^cH^\ = -p \a%^+hH^+^\ -P -p ! + \a^h'^rPH^\ + , or = a(0,l,s -P 2,s -P 3) -p a(0,s -P 1, 2,s + 3) -P a(0,s + l,s -h 2, 3) -p a(5,l,2,s -p 2) -p a(s,l,S -P 2,3) -p a(s,S -P 1,2,3), which we know on other grounds to be true. 1888.] Dr T. Muir on Aggregates of Determinants, 105 10. Keturning now to § 7, and taking the originating determinants A and B in the form 41 42 43 46 45 44 51 52 53 56 55 54 61 62 63 ) 66 65 64 that is to say, with their elements in the umbral notation_, we obtain the identity in the form 41 45 44 46 42 44 46 45 43 51 55 54 + 56 52 54 + 56 55 53 61 65 64 66 62 64 66 65 63 41 42 43 46 45 44 46 45 44 56 55 54 + 51 52 53 + 56 55 54 66 65 64 66 65 64 61 62 63 hTow the first three determinants here are minors of any deter- minant of the 6*^ order, and the second three would he minors of a determinant of the 6*^ order if its elements were such that 41, 42, 43, 51, 52, 53, 61, 62, 63 = 36, 35, 34, 26, 25, 24, 16, 15, 14, — in other words, were such that r,s=7-r,7-s. But this is exactly the definition of a centro-symmetric determinant. Conse- quently the above identity is an identity connecting six of the minors of the centro-symmetric determinant 1 11, 22, 33, . . . , 66 Irs-7-r,7-«) and thus we have in regard to such determinants a theorem quite co-ordinate with Kronecker’s regarding axisymmetric determinants. In the notation of § 5 it would stand as follows: — ^ I 11, 22, . . . , 66 I he centro-symmetric, i.e. if its elements he such that in every case r,s = 7 - r,7 -s, then 4 5 6 + 4 5 6 6 + 4 5 1 4 2 4 5 1 4 2 6 4 5 6 + 4 5 6 + 6 6 6 I 6 I 106 Proceedings of Royal Society of EdinhurgJi. [jan. 6, 4. On a Simplified Proof of Maxwell’s Theorem. By Pro- fessor Burnside. Communicated hy Professor Tait. In the course of verifying some of the mathematical work in the third instalment of Professor Tait’s paper on the Kinetic Theory, the following simplification of his proof of Maxwell’s theorem occurred to me. The number of pairs of particles, one from a set (P, h)^ the other from a set (Q, ^), for which the velocities parallel to the axes lie between X and x-\-dx ^ y and y + dy^ z and z-vdzm the one case, and between a?' and x + dx, y' and y + dy\ z' and z' 4- dz' in the other, oc e~^^^'^'^y^+^‘^^~^^^"^'^y"^^^'‘^'>dxdydzdx dy dz' Write Pa? + Qa?' = (P + Q)a x-x' =d and similar substitutions for y,f and z,z. Then lix^ + hx"^ = {li + h)o? + Q2/z+P2^ ,2 , oQ^^-P^ (P + Q)^"" "pW aa = aa? + ba!^ -1- 2caa! say . d^az e-«(a24-^2+Y2)_&(a'2+^'2+y2)_2c(aa'+i3^'-l-yy') X dad(^dyda! d(^' dy Hence if a? + +y^ = r? a'2 +/5'2 +y'2 aa! + f^f^' -4- yy = VV^ COS 6 , (1-) the numbers of pairs of particles one from each system for which the velocity of the centre of inertia lies between v and v + dv, the relative velocity between and + dv^^ , and the angle included by the directions of v and between 6 and 6+dQ Qc Q-av^-hvo^-cvvQco?,d vHf^dvdv^ sin 0 dO . The energy exchanged between a P and a Q at impact is X product of components of v and in the line of centres at impact. 1888.] Professor Burnside on MaxweWs Theorem. 107 If y is the angle between the line of centres and the direction of Vq , and the angle between the plane parallel to these directions and the plane parallel to the directions of v and , the above quantity is cos and since in the process of averaging the terms involving ^ will obviously vanish, they may be omitted from the first. Hence, the probability of a collision being proportional to the relative velocity, the average energy exchanged between a P and a Q at impact 2PQ - / /) P + y COS y + Sin y 2PQ ^P + Q in r ^7 -f*«2-&Do2-2ci5yoCOS^-o n • n • _ o dv / di'Q / d9 / dye sm 6 , Vq sin y cos y . vVq cos-y 0 y 0 y 0 y 0 7T y<*°° f'^i -ay2-6t>o2_2cOToCOs0 o o • /, dv / dv^ / dd / dye sm 6 . Vq sm y cos y . 0 y 0 y 0 -'0 0 ^ 0 ^ 0 Performing the integrations with respect to 0 and y , this becomes PQ /•oo r" T ^-av'^-lvA 0 y 0 L -av'^-hvM 3 cosh 2cWn~ ^^sinh 2c ^0 ] dvdvQ P+ Q y^OO /-•GO y ^ vVq^ smh 2cvV(f . dvdvQ [No element of the integral in the numerator can be negative hence it can only vanish if 1 cosh 2cvvn 2cVVn sinh 2cvVn = 0 always, i.e. if c = 0 .] The integrals involved all depend on ^00 /^oo 0 Q which when ah — is positive, as it is in this case, can at once be shown to be equal to i{ah - c^) * Finally then, the average energy exchanged at an impact 108 Proceedings of Roycd Society of Edinburgh. [jan. 6, 1 1 4 dc^ 4c dc ah - pTq Id 1 ' 2 dc ab- 1 2PQc -(P + Q)(a6-c2). 5. On some Glass Globes with Internal Cavities produced during Cooling. By J. T. Bottomley. the Royal Society a number of flint-glass globes having remarkable internal cavities produced during cooling. Along with these globes there is exhibited for comparison one globe having no perceptible internal cavity. The making of these globes was shown to me by Mr John Griffin, manager of the St Eollox Flint Glass Works, Glasgow. The globe having no internal cavity is marked A. A set of globes, four in number, are marked B^, B2, Bg, B^, and the sixth globe is marked C. The globes which have cavities were formed in the following way : — A pot of the finest glass having been selected, very free from any appearance of scum or bubbles, a ball of glass was gathered in the usual way at the end of the glass-blowers’ iron or long blowing tube, and it was worked with the help of a wooden mould into the form of a glass ball projecting out, by a short neck of glass, from the end of the iron. When the red hot ball had been thoroughly examined to see that there was no flaw of any kind in the mass, the operator quickly brought it to the open window, and held it in the draught which was blowing sharply into the glass-house. The ball was at this time very bright red hot — glowing, in fact — and it was quietly turned round and round in this cold position, so as to cool equally all round. At first it looked perfectly uniform throughout, but soon there {Abstract.) The object of this communication is to exhibit and describe to 1888.] Mr J. T, Bottomley on Glass Globes with Cavities. 109 became visible in the middle of the mass a few very minute bright points — little specs it seemed — and these quickly grew, and were then perceived to be hollow spaces in the midst of the glass, and finally they assumed the appearance of the large bubbles now to be seen. The conclusion of the operation was the cutting across of the neck of glass which supported the globe, and allowing the globe to drop off, and the passing of it through the leer or annealing furnace in the ordinary way. The place where the neck was broken ofi was finally ground flat by the polishers. The explanation of the cavities is obvious. The sudden cooling of the outside of the glass globe caused the outer layer to become rigid, while the interior mass was still extremely hot and molten. But when the cooling reached the interior, that portion underwent great contraction of volume, and as the outer skin, which had enormous strength, on account of its shape, refused to be drawn in, the interior was forced to part somewhere, and these cavities were formed. To follow a little further this interesting phenomenon, my friend Mr Griffin was good enough to make for me a globe without cavities, by allowing the whole mass to cool more uniformly, and letting the skin fall in towards the centre along with the interior mass. Thus he produced the globe marked A. The process, which, of course, is the ordinary one for producing large glass paper-weights, and articles of that kind where flaws or bubbles are considered as blemishes, consisted in very frequently putting the cooling globe into the mouth of the glass-pot, and thus warming up the outer skin sufficiently to keep it from becoming suddenly rigid. Thus the cooling was gradually carried on with frequent partial reheating of the surface skin. On examining the globes marked B^, &c., it will be seen that the distribution of the smaller cavities is very curious in appearance. It will easily be noticed that these little cavities are distributed over concentric spherical surfaces (to speak roughly). The cause of this was not difficult to trace. The glass worker, in gathering a ball at the end of his iron, is in the habit of taking from the glass-pot a small quantity of glass to begin with. He then draws out the point of the iron into the air 110 Proceedings of Royal Society of Edinhurgh. [jan. 6, for an instant, allowing the glass to cool very slightly j and imme- diately thrusts it hack for another charge which covers the first. This is repeated several times, till a sufficient quantity of glass has been gathered, when he puts the whole mass hack into the mouth of the pot, and turning it round and round heats it up preparatory to blowing. Now, on considering the matter, it seemed probable that these little cavities would form at the surfaces thus produced and exposed successively to the air. Any dust which might fall on the surface would give a starting place for a cavity ; or a place where the con- tracting glass would part under the negative pressure produced in the way I have described above. Accordingly, I asked Mr Griffin if he could manage to gather a ball of glass of considerable size, without bringing it out from beneath the cover of the glass-pot, and he very kindly made the attempt and soon succeeded, and produced the ball marked C, which, when cooled in the manner already described, exhibited three or four beautiful large cavities, but none of the small cavities which are possessed by the others. I do not propose to enter at all into a discussion of what I may be allowed to call the lessons to be learned from a study of these phenomena, though there are several points which are well worthy of consideration. I wish only to refer to two matters which may be thought of in this connection. The first of these is one which was pointed out to me by Mr W. H. Barlow, F.E.S., the engineer of the Tay Bridge, and a Fellow of this Society. Mr Barlow is a member of the Ordnance Committee, and he was greatly interested in examining these globes, and in thinking of the possibilities of similar flaws being produced during the cooling of large castings, such as those used in the construction of big guns. The second question to which I would direct attention is, perhaps, one which is already in the minds of all who have looked at these globes. It is the case of the cooling of a body like our earth. It seems certain that if the interior of the earth is of material which shrinks in cooling, cavities such as these would of necessity be formed by the shrinkage of the interior parts after an outside shell has become rigid. 1888.] Dr A. B. Griffiths on the Malpighian Tubes. Ill 6. Investigations on the Malpighian Tubes and the “ Hepatic Cells ” of the Araneina ; and also on the Diverticula of the Asteridea. By Dr A. B. Griffiths, F.R.S. (Edin.), F.C S. (Bond. & Paris), Princi]pal and Lec- turer on Chemistry arid Biology^ School of Science^ Lincoln; and Alexander Johnstone, F.G.S. (Bond. & Edin.), Assistant Professor of Geology and Mineralogy^ University of Edin- burgh., <^'C. The present memoir is a continuation of those already published by one of us, on the physiology of the Invertehrata. 1. Malpighian Tubes o/Tegenaria domestica. The intestines of Tegenaria domestica form a tuhe-like body, which dilates into a short rectum, and into this rectum the Mal- pighian tubes open (fig. 1, A and B). The secretion obtained from a large number of these tubes yields uric acid. The secretion is neutral to test papers. The secretion was examined chemically by two separate methods : — (a) The clear liquid from the Malpighian tubes was treated with a hot dilute solution of sodium hydrate. On the addition of hydro- chloric acid a slight flaky precipitate was obtained after standing six hours. These flakes were seen (under the microscope) to con- sist of rhombic plates (figs, ^a) and other crystalline forms. When these crystals are treated with nitric acid and then gently heated with ammonia, the beautiful four-sided prisms of reddish purple murexide [CgH4(N’H4)N'50g] are obtained (fig. 25). The crystals (another portion) produced by sodium hydrate solution were dis- solved in a drop or two of sodium carbonate solution, and then poured upon a piece of filter paper moistened with a solution of silver nitrate, when a dark brown stain of metallic silver was pro- duced : thus showing, according to the test of Schiff, the presence of uric acid. ib) Another method used was as follows (for testing the secretion of the Malpighian tubes of Tegenaria) : — The secretion was boiled in distilled water, and evaporated care- fully to dryness. The residue obtained was treated with absolute 112 Proceedings of Boyal Society of Edinhurgh. [jan. 6, ethyl alcohol and filtered. Boiling water was poured upon the resi- due, and an excess of pure acetic acid added to the aqueous filtrate. Intestinel Rectum Fig. 1. — Tegenaria domestica. A dissection showing the position of the Malpighian tubes and the so-called “liver.” A (greatly enlarged) ; B shows the same parts, hut isolated from the abdomen. After standing eight hours, crystals of uric acid made their appearance, and were easily recognised by the chemico-microscopical tests already mentioned. The waste nitrogenous products of Tegenaria domestica are con- verted by the Malpighian tubes into uric acid ; but the uric acid is not in that condition (^.e., the acid condition), for it was found in combination with sodium. Sodium is easily found in the secretions 1888.] Dr A. B. Grififiths on tlu Malpighian Tubes. 113 of the tubes ; therefore the secretion contains sodium urate. This fact also points to the inference that sodium is a constituent of the blood of Tegenaria. Ko urea, guanin, or calcium phosphate could he detected in the secretion. This investigation proves the true renal function of the Malpighian tubes of the Araneina, 2. The Hepatic Cells of Tegenaria domestica. The “ liver ” ducts (hepatic cells) are to he found anteriorly to the rectum (fig. la and b), and pour their secretions into the ali- Tig. 2, — a and h. Crystals of Uric Acid and Murexide. a, the uric acid crystals covered with a brown pigment ; &, Murexide crystals. mentary canal (intestine). They appear under the microscope to consist of cellular tissue. The secretion of the “ liver ” of Tegenaria domestica^ when perfectly fresh, gives an acid reaction to litmus paper. The following reactions were obtained from the secretions of a very large number of animals : — 1. The secretion forms an emulsion with oils yielding suhse- quently fatty acids and glycerol. 2. The secretion decomposes stearin, with the formation of stearic acid and glycerol— C57HnoOr, + . 20/6/88 VOL, XV. H ri4 Proceedings of Royal Society of Edinhurgli. [jan. 6, 3. The secretion acts upon starch paste with the formation of dextrose. The presence of dextrose was proved by the formation of brownish red cuprous oxide from Fehling’s solution. 4. The secretion dissolves coagulated albumin (hard white of an egg). 5. Tannic acid gives a white precipitate from the secretion. 6. When a few drops of the secretion of the organ were examined chemico-microscopically, the following reactions were observed : — On running in between the slide and the cover-slip a solution of iodine in potassium iodide, a brown deposit is obtained ; and on running in concentrated nitric acid upon another slide containing the secretion, yellow xanthoproteic acid was formed. These reac- tions show the presence of albumin in the secretion of the organ in question. 7. The presence of albumin in the secretion was further confirmed by the excellent tests of Dr E. Palm [Zeitsclirift fur Analytische Ghemie, vol. xxvi. part 1). 8. The soluble zymase (ferment) secreted by cellular tubes was extracted according to the Wittich-Kistiakowsky method (Pfliiger’s ArcMv filr Physioloyie, vol. ix. pp. 438-459). The isolated fer- ment converts fibrin (from the muscles of a young mouse) into leucin and tyrosin. 9. The albumins in the secretions are not converted into tauro- cholic acid or glycocholic acid ; for not the slightest traces of these biliary acids could be detected by the Pettenkofer and other tests. 10. The secretion contains approximately 4 per cent, of solids, of which we could detect sodium. The slight residue (solids) effervesced on the addition of a dilute acid. 11. No glycogen was found in the organ or its secretion. From these investigations the so-called “ liver ” of the Araneina is similar in physiological functions to the pancreas of the Vertehrata. 3. The Diverticida of the Asteridea. The saccular diverticula of Uraster ruhens (fig. 3) have also been examined by similar reactions to, those applied to the reactions of the ‘‘ hepatic cells” of ih^Araneina^ and with the same results. Therefore, we conclude that the diverticula of the Asteridea are pancreatic in function. 1888.] Dr A. B, Griffiths on the Malpighian Tubes. 11'5 It appears from the investigations already published by one of us on the so-called ‘‘ livers ” of the Invertebmta, that the pancreas was Fig. 3. — Uraster rubens. Dissection from dorsal or aboral aspect, showing the diverticula (2 lobes) and the pyloric sac. the chief digestive organs in the early forms of animal life ; for, as far as these investigations have progressed, there seems to be no organ in the Invertebmta to answer to the liver of higher forms. 7. On the Thomson Effect in Iron. By Prof. Tait. IIG Procee-dings of Royal Society of Edinburgh, [jan. 16, Monday, l^th January 1888. Professoe CHRYSTAL, LL.D., Vice-President, in the Chair. The following Communications were read : — 1. Obituary Notices of former Vice-Presidents of the Society. 2. Problem in Relationship. By Professor A. Macfarlane, D.Sc. The following problem was sent me by Mr Kirkman, F.R.S.; its solution illustrates the Calculus of Relationship, Proc. Roy. Soc. Edinh., vol. xi. pp. 5 and 162: — Two hoys, Smith and Jones, of the same age, are each the nephew of the other ; how many legal solutions 1 The statements are that Smith is the nephew of Jones, and Jones the nephew of Smith. Let c denote child, then ^ denotes parent, and the equations are S = cc- J and J = cc-S . c e Hence S = cc-cc-S . c e Let m denote male and / female : then all the varieties of relationship are obtained by introducing m or / before the second, third, fifth, and sixth symbols ; for the sex of the first and fourth symbols is given to be m. Thus there are sixteen varieties in all. Of these mminm, mffm, fmmf, and ffff are chronologically impossible, because they make a man his own grandson, or a woman her own granddaughter. For example, S = cmcf- cfcfni-Si , c c reduces to S = CMcfcm-^ , 1888.] Prof. A. Macfarlane on ProUem in Bdationshi'p, 117 i,e. mis = , f e., tlie father of S is his own grandson. Again, there are eight which are legally impossible, because a person may not marry his or her grandparent, namely, mmmf, mmfm, mfmm, fmmm^ fffm, ffmf, fmff, mfff. For example, S = cmcm}-cmcf^-^ , c c S = cmcmc/is, m-S = mcmc/-S, i.e., the father of S marries his own grandmother. There remain the four cases mmff, mfmf, ffmni, fmfm\ each of which is legally and physiologically possible. For there is nothing to prevent two persons, each twenty years old say, from marrying each the appropriate parent of the other, each of whom may he forty years old, and the Smith and Jones of the problem may he the result of these contemporaneous marriages. For example, take the first of these, S = cw^c77^ic/q/■is G G . *. mi^TTzis^ = 77zi^/c^/is^ ; z.e., the father of Smith’s father marries the daughter of Smith’s mother. Thus there are four solutions — (1) Two men marry each the mother of the other. (2) Two women marry each the father of the other. (3) A man and a woman marry the mother and father of one another, which comprehends the two cases of Smith being the son of the old woman, and Smith being the son of the young woman. Note, — Here Smith and Jones are taken as arbitrary names, equivalent for example to Tom and Hugh. If the convention about surnames is taken into account — that a child’s surname is identical with that of his father — then only the first and second solutions are possible. 118 Proceedings of Boy at Society of Edinburgh, [jan. 3. On Transition-Resistance and Polarisation. By W. Peddie, B.Sc. In a paper communicated to this Society last Session, I gave an empirical formula representing the relation between current-strength and time when platinum plates are used for the passage of a current through acidulated water, the electromotive force being that of a single Daniell cell. As is well known, we can look upon the electrodes as condensers of very great capacity. If E be the electromotive force of the battery used, while e is the reverse electromotive force of polarisation, and r, x, and c are the values of the resistance, the current-strength, and the capacity respectively, the equations of conduction through the condenser are as follows: — E - e = e de R being the resistance of the dielectric. If, in the 'case we are considering, no decomposition of the liquid occurs, we may suppose R to be infinite. Hence the second equation becomes and we get for the law connecting current-strength and time, the relation where x^ is the value of x when ^ = 0. But the curve represented by this equation does not represent the actual variation of x with time. Hence, either r or c must vary, or both must vary simul- taneously. In a paper also communicated last Session, I showed that there is a very considerable transition-resistance at the surface of platinum, and that this resistance goes on slowly increasing as the time that has elapsed after heating the platinum to redness increases, but the law of increase is unknown. I have made the assumption that the rate of increase is proportional to the excess of the final value of the resistance over its value at the given time. If we use R to 1888.] Mr W. Peddle on Transition- Resistance, 119 represent the total increase of transition-resistance in time, t this gives E = Ko(l-s-‘% (A) where Eq is the final value of E. Hence, if we delete the suffix, and write for the value of the total resistance in the circuit when ^ = 0, we get for the value of the total resistance at any time r = ro + E(l-s-'^). So, if we assume the capacity to be constant, the equations of. conduction give approximately with this value of r , = .... (B) where _E ,_1 “ ^ c(K + r,)’ e being the capacity of the condenser. The approximation is obtained by assuming that the reciprocal of the quantity ch(r^ E), is very small compared with unity. Since C and E are very large quantities, while h is not very small, this assumption may safely be made. This equation contains four unknown constants, and their numerical value cannot be determined by elimination, for the degree of the resulting equation is too large. So I have obtained them by giving probable values to a and a?Q, and then calculating h and k. In this way, by giving different values to a and curves were obtained between which the observed curve lay, and so by varying a and satisfactory values were finally obtained. The curve to which I liave applied the equation is that drawn through the group of points marked a in the plate which accompanies my paper already referred to on transition-resistance. The value of the constants were a =10, aTQ = 420, & = 0‘3924, /t = 0‘0404. The calculated and observed values of x for different values of t are given below. t 1 2 4 8 12 16 x{ohserved), 93 60 40-4 28-4 23 20-2 x{calculated)^ 95 60-2 40-8 28-8 23-7 20 The coincidence is obviously extremely close. I also applied the equation to the curve marked h in the plate alluded to, assuming the same values for h and k, which is at least approximately correct. The values of a and Xq are 5*3 and 230. The results are 120 Proceedings of Royal Society of Edinburgh, [jan. 16, t 1 2 4 8 12 x{ohserved\ 83 53*5 38-6 27*3 22-3 xicalculated), 81-9 55*1 37*9 26-5 22*9 the coincidence again being close, though not so good as before. The value of the constant b shows that in at most twenty minutes (the unit in terms of which the constants are calculated being one minute), the quantity practically reaches its final value ; that is, the resistance has practically reached its final value. This is not in accordance with the idea on which the above investigation was based. But the close agreement of calculation and experiment renders it probable that the resistance may increase more rapidly when polarisation is going on, than when the plates are merely having oxygen slowly deposited on them. I have remarked in my paper on transition-resistance, that certain experi- ments I made seemed to indicate this. To obtain definite information on the point, I have used a dead beat galvanometer, so that the reading of the deflection could be taken at an interval of five seconds after starting the current. In all other points the arrangement of the apparatus was the same as in the previous experiments ; that is, the platinum electrodes 60 square centimetres in area, which di]3ped into a dilute solution of sulphuric acid, were connected with the terminals of the galvanometer. A single tray-Daniell cell was placed in circuit when required. When the cell was joined in, readings were taken at intervals of five seconds for a short time. Then the plates were polarised until the reverse electromotive force was as nearly as possible equal to that of a Daniell cell, after wliich the battery was thrown out of circuit, and the readings during discharge of the plates were taken as before. Experiment la. Time in seconds, 5 10 15 20 25 30 Deflection during charge, 8 5-7 4-6 4 3-5 3*2 Deflection during discharge, ... 3-64 2-94 ... 2-64 2-34 Experiment 2a. Time in seconds. 5 10 15 20 Deflection during charge, 9-03 7*03 6-03 5-43 Deflection during discharge. 4-7 3-8 3-3 3-2 121 1888.] Mr W. Peddie on Transition-Resistance. Experirrwnit 3a. Time in seconds, 5 10 15 20 Deflection during charge, 9*38 6 ’3 8 5 ’38 4 '68 These three experiments were performed in the order indicated on the same day. The next experiment was made on the following day. Experiment 4 a. Time in seconds. 10 15 20 25 Deflection during charge. 8-6 6-5 5-5 4*8 Deflection during discharge, 3-41 2-81 2-51 2*21 In another experiment care was taken to ensure that the electro- motive force of the battery had the same value when the plates were to be charged as it had when thrown out of the circuit previous to discharge of the plates. The following table gives the result : — Experiment 5 a. Time in seconds, 5 10 15 20 25 Deflection during discharge. 3-98 2-98 2-38 2-08 1-93 Deflection during charge. 7-35 5*85 4-95 4-45 4-05 These experiments show that the magnitude of the transition- resistance is about doubled when the plates are polarised fully by a Daniell cell. They were all made when the plates had been unheated for some weeks. Corresponding experiments were made after the plates had been heated, and little difference could be detected in the values of the initial deflections during charge and discharge. This shows that the difference formerly observed was due to the transition-resistance. The only cause of uncertainty lies in the fact that, in the time which elapsed between taking out the battery and connecting the plates, the electromotive force of polarisation might have largely diminished. This interval was only a small fraction of a second. But the polarisation does not so diminish, for the initial value of the depolarising current was found to be the same even if the discharge was commenced one minute after breaking the circuit. When the plates were heated a resistance of about 600 ohms had to be placed in the circuit to reduce the deflections to their value before heating. In the formula (B), the constant a, which represents the ratio of 122 Proceedings of Boyal Society of Edinhurgh. [jan. 16, the final to the initial resistance, has the value 10. But we have just seen from experiment that the true value of that ratio is 2. Hence the close correspondence between the curve represented by the equation (B), and the observed curve merely shows that the other causes of variation in the current strength are of such a nature that their effects can be represented as due to variation in the value of the transition-resistance, according to the law (A) : so that in equation (A), the constant Eq does not give the true final value of transition-resistance. One cause of variation of the strength of the current of which I have taken no account, consists in the decomposition of the liquid by smaller electromotive forces than are necessary for visible decom- position, the products of decomposition being dissolved in the liquid. Also no account has been taken of any variation of capacity which may occur. How Varley has shown that the capacity of such an arrangement as we are dealing with, increases as the electromotive force increases. He used one of the ordinary methods of determining the capacity of a condenser. It is easy to show the same effect by the methods I have been using. If the circuit including an electrolyte and a battery be broken, for a short time, during the process of charging the plates, and be again completed, the current is found to be stronger. This shows that the capacity of the plates has increased in the interval, thus diminishing the reverse electromotive force of polarisation. In the following experiment readings were taken at intervals of five seconds, and a break was occasionally made in the circuit. The interval of break was half a minute. Experiment \h. Deflection. 6'41, 5*31, 4-71, 4*21, 3’81, 4‘11, 3’41, 3-11, 2-91, 2-71, 2-59, 3T1, 2-71, 2*47, 2*33, 2*24, 2*15, 2-51, 2-26, 2T2, 2-03, H96, 1*91, 2T6, 1-91, 1*84, 1-79, 1-75, H86, 1*76, H71, 1*67, 1-73, 1-66, 1-58, 1-55, 1*52, 1-47, 1*45, H42, H61, 1-51, 1-46, 1-44, 1-40, 1-54, 1*46, H41, H37, 1-36. The same phenomena appear during discharge of the plates, the increase of current-strength after the break being due, in this case, to the decrease of capacity causing increase of the electro- 1888.] Mr W. Peddle on Tmnsition-Eesistance. 123 motive force of polarisation. The results are given below. The last interval of break was 1 minute. All the other intervals were half a minute. Experiment 2h, Deflection. — 3*60, 3’10, 2-84, 2*60, 2*45, 2*34, 2*20, 2T2, 2-04, 2-00, 1*92, 1-87, 1*82, P78, P74, P70, P70, 1*60, 1-56, 1-52, 1-49, P47, 1*44, 1*42, P40, P38, P36, 1-37, 1-34, 1-32, 1-30, 1*28, P26, P25, 1*24, P23, 1-22, 1-21, 1-23, 1-20, 1-19, M7, M6, M5, M4, M3, M2, Ml, 1-10, M3, Ml, MO, P09, 1*08, 1-07, 1-06, 1-06, 1-06, 1-05, 1-04, P07, 1*05, P04, 1-03, 1-02, 1-02, 1-01, 1-01, 1-01, 1-00, 1-02, POl, 1-00, 0-99, 0-99, 0-98, 0-98, 0*97, 0*96, POO, 0-99, 0-98, 0*97. In order to explain the increase of capacity, Varley {Proc. Roy.. Soc., 1871), supposed that there was a film of gas separating the plate from the liquid, and that this film was compressed by the tendency of the oppositely electrified surfaces to approach. Since the electric attraction is inversely as the square of the distance, he supposed that the molecular forces keeping the particles of the gas apart were inversely as the cube of the distance. I do not think that his supposition of compression is necessary; but, as he does not enter into detail regarding his method of experimenting, a conclusion cannot be easily arrived at. However that may be, the phenomena exhibited in experiments 2a and 26, merely show that electric absorption is occurring in the one case, and that residual discharge occurs in the other. The insulating portion of the circuit consists of two parts, — the gaseous film which conducts appreciably, and the space between the gas and the plate, across which no conduction can occur so long as decomposition does not take place. Thus Clerk-Maxwell’s theory of a composite dielectric applies to it, and shows that electric absorption and residual discharge must occur. The oppositely charged layers of electricity are not separated merely by the molecular distance between the gas and the plate until conduction has taken place through the gaseous film. This fact has important bearings on all determinations of the capacity of electrodes. I give below the results of an experiment on the discharge of 124 Proceedings of Boyal Society of Edinhurgli. [jan. 16, the plates, no break being made in the circuit. The first reading was taken ten seconds after commencing the discharge. Deflection at intervals of five seconds. — 2*03, 1*43, 1*23, ITO, 1*02, 0*93, 0*86, 0*81, 0*76, 0*71, 0*68, 0*64, 0*62, 0*59, 0*56, 0*54, 0*52, 0*51, 0*49, 0*48, 0*46, 0*45, 0*44. At intervals of ten seconds. — 0*42, 0*40, 0*38, 0*37, 0*35, 0*34. At intervals of one quarter of a minute. — 0*32, 0*30, 0*29, 0*28, 0*27, 0*25, 0*24, 0*24. At intervals of half a minute. — 0*22, 0*21, 0*20, 0*19. At intervals of one minute. — 0*16, 0*15, 0*13, 0*12, 0*12, 0*11. If we take the values of x for ^ = 10 and ^ = 20, and calculate from them the value of h in the equation /yi /y* C ;r(C2H30)N20g- It was prepared by the action of acetic anhydride on morphine, “when a considerably smaller quantity” of the anyhdride is taken than is required to form diacetylmorphine. The product “resembles /^-diacetylmorphine in every particular, save that it yields different numbers on analysis.” The numbers given by Wright agree only fairly well with theory. “ That the substance is truly a monoacetyl morphine, and not a mixture of morphine and diacetyl derivatives, is shown by the fact that the base itself is soluble in ether j whereas morphine is practically not soluble in that menstruum. Moreover, a mixture of ^-diacetyl- 134 Proceedmgs of Royal Society of Edinburgh, [jan. 30, morpliine hydrochloride and morphine hydrochloride in equal quantities dissolved in a little water, allows almost the whole of the latter salt to crystallise out, and does not dry up to a varnish over sulphuric acid, hut to a crystalline mass wetted by a syrup, which finally dries up to a glaze over the crystals.” Wright obtained from the chloroplatinate 19’45 per cent. Pt ; the mono- acetylmorphine compound requiring 19*29 per cent, of metal. We have repeated Wright’s experiments, with the results under- noted. 15 grams of morphine were dried at 120° C., and thoroughly mixed with 1*25 gram acetic anhydride. The mixture was then warmed on the water-bath for half an hour, treated with water and with sodium carbonate in slight excess, and the whole shaken up with ether. On separating and evaporating the ether a non- crystalline residue remained, to which dilute hydrochloric acid was added in quantity just sufficient to render faintly acid. The strong solution showed no signs of crystallisation even after the lapse of two days. To a portion of the solution platinic chloride was added, and the washed precipitate dried at 120° C. On ignition 0*124 gram gave 0*023 gram Pt = a yield of 18*54 per cent. ^34^36(^2^3^)2^2^6*P^^2^^6 givcs 18*48 per cent. Pt. C34H3^(C2H30)]Sr20g.PtH2Clg gives 19*53 per cent. Pt. In this experiment there is no evidence that any monoacetyl- morphine was formed, although the conditions were most favourable for its formation. The solution must have contained either diacetyl- morphine or a mixture of tetracetylmorphine and morphine. Wright’s assumption, that the supposed monoacetylmorphine could not have contained morphine, because the latter is insoluble in ether, is not well founded. Under certain conditions, probably when it is freshly precipitated and partially amorphous, morphine is soluble in ether. The evidence from the non-crystalline condition of the product is far from conclusive, as crystallisation is influenced by a variety of circumstances not well understood; and it must be remembered that tetracetylmorphine hydrochloride is a very soluble and not readily crystallisable salt, which may form a basic compound with morphine. (3) Ethylmorphine. — Since Wright contributed his papers on morphine derivatives, it has been shown by Grimaux,^ that part of * Comptes Bend. , xcii. 1888.] Mr Dott and Dr Stockman on Morphine. 135 the hydrogen in the morphine molecule may be replaced by alcohol radicals. We endeavoured to prepare a monoethyl derivative [C34H3h(C2H5)N206], by beating together equivalent quantities of morphine, soda, and ethyl iodide, in alcoholic solution. The alcohol having been evaporated, the residue was exhausted with chloroform, and the chloroform extract converted into hydrochloride. The resulting crystalline mass was pressed in calico, and a chloroplatinate prepared from the crystals. Dried at ISO"", 0’44 gram gave on ignition 0'083 grani = 18-86 per cent. Pt; 0-2965 gram gave 0-056 gram Pt= 18-88 per cent. Pt. The expressed mother-waters from the crystals yielded a platinum salt, which gave on ignition — 0’3239 gram = 0*061 gram Pt= 18*83 per cent. Mean of three determinations = 18*85 per cent. Pt. C34H36(C2H5)2N06.PtH2Cl6 = 18*97 per cent. Pt. C34H37(C2Hg)N20g. PtH2Clg = 19*50 per cent. Pt. Whence it is manifest that under the conditions described, only diethylmorphine is formed (using the nomenclature adopted by Wright). In fine, there does not appear to be any evidence to justify the adoption of a higher formula for morphine than the empirical (Cj^Hj^glSrOg), which is the formula still in general use. It follows that Wright’s *‘ diacetyl morphine ” should be named acetijlmorpliine^ and his “ tetracetylmorphine ” diacetylmorphine. 4. On the Fossil Flora of the Staffordshire Coal Fields. I. The Fossil Plants collected during the Sinking of the Shaft of the Hamstead Colliery, Great Barr, near Birmingham. By Robert Kidston, F.R.S.E., F.G.S. 5. On a Monochromatic Rainbow. By John Aitken. A monochromatic rainbow looks like a contradiction in terms. As a rainbow of this kind was, however, seen lately, its occurrence seems worth putting on record. On the afternoon of Christmas day I went for a walk in the direction of the high ground to the south of Falkirk. Shortly after starting I observed in the east 136 Proceedings of Boy at Society of Edinburgh, [jan. 30, what appeared to be a peculiar pillar-like cloud, lit up with the light of the setting sun. What specially attracted my attention was that the streak of illumination was vertical, and not the usual horizontal band-form we are accustomed to. I looked in the direction of the sun to see if I could trace any peculiar opening in the clouds through which the light passed, but failed to do so. I continued observing for some time the peculiar appearance, when at last the pillar-like illumination became more elevated, and by the time the sun was just setting and I had arrived on the high ground, it had reached to a considerable height, and I at last began to suspect that what I had been looking at was not a cloud at all, but the “ tooth ” of a rainbow. Soon all doubt was put at rest by the red pillar extending, curving over and forming a perfect arch across the north-east sky. The rainbow when fully developed was the most extraordinary one I ever saw. There was no colour in it but red ; it consisted simply of a red arch, and even the red had a sameness about it ; all the other colours were absent. Perhaps this is stating it too strongly, as after careful observation I succeeded in detecting at one or two points traces of yellow; but of green, blue, or violet there was not a vestige, and in their place there was a dark band extending inwards to about the breadth usually occupied by these colours. This band, though distinctly darker than the sky, to the inside of it was not greatly so. Outside the rainbow there was part of a secondary bow, and inside, at certain places, there were indications of a supernumerary bow, as short detached red patches were visible at different points on the inner edge of the dark band. Por some time before the bow developed itself I had been watch- ing the Ochil Hills, which lay to the north of me. These hills at the time were covered with snow, and the setting sun was shining brightly on them. On many occasions I have seen snow-clad hills, in this and other countries, lit up with the light of the setting sun, and glowing with rosy light, but never have I seen such a depth of colour as on this occasion. It was not a rosy red, but a deep furnacy red. How, why was the colour on the hills of so deep a red on this evening? The monochromatic rainbow gives its own explanation; it also tells us why the hills glowed 1888.] Mr John Aitken on a Monochromatic Bainhow. 137 with so rich a colour. The rainbow is simply nature’s spectrum analysis of the sun’s light, and it showed that on that occasion the sun’s light was shorn of all the rays of short wave lengths on its passage through the atmosphere, and that only the red rays reached the surface of the earth. But it will he said, that every object on the surface of the earth on that afternoon ought to have appeared red, and nothing hut red, if nothing but red rays reached the earth from the sun. ISTow this was by no means the case. Everything looked not very different from their usual ; they appeared simply tinted with red. The reason for this evidently was, that while we received only red light direct from the sun, there was a great deal of green, blue, and violet light reflected from the sky overhead, and these com- bining with the red caused the light to he but little different from the usual. The reason why the Ochils glowed with so deep a red was owing to their being overhung by a dense curtain of clouds, which screened off the light of the sky. Their illumination was thus principally that of the direct red light of the sun. 6. On Neuropteris plicata, Sternb., and Neuropteris rectinervis, Kidston. By Robert Kidston, F.E.S.E. 7. Reflex Spinal Scratching Movements in some Verte- brates. By Prof. John Berry Haycraft. Many, who have kept dogs, are aware that if the skin covering the side of the body be scratched, a dog will move the leg of that side as if itself to scratch the part touched. This fact is known to the physiologist, and a, so-called, scratching centre, to which the sensory impulses are carried, and from which motor impulses to the muscles pass, has been shown to exist in the spinal cord. I would venture to lay before the Society one or two additional facts in this connection which have been observed by me. First, as to the sensory area from which this reflex may be initiated. The only sensitive part of the skin in most cases is that covering the lower ribs in about the middle of their course. If, however, the dog be sensitive to irritation, if it has suffered from 138 Proceedings of Royal Society of Edinburgh. [jan. 30, vermin, or from any irritative condition, the area is much greater. Practically it includes those parts of the skin to which the hind foot can be approximated. It commences posteriorly at the part which the hind leg can reach, generally 2 or 3 inches in front of the flank, though this will vary according to the size of the animal. It extends forwards to the shoulder, including the whole side of the animal, and even reaches up the side of the neck, and on to the root of the ear. The most sensitive portion is that part which in most dogs alone gives the reflex. The scratching area is very sharply defined. If in a sensitive animal the skin on the hack within half an inch of the middle line be scratched the leg of that side will move. If, however, the skin at a corresponding part of the opposite side be touched, the animal will scratch at once with the other leg. The same observations apply to the scratching areas when they extend ventrally to the middle line. The skin of the flank, of the muzzle, of the fore leg are outside this area, and outside the reach of the hind leg. They are scratched by the teeth or fore leg. As in the case of the pithed frog, if one side of the animal be scratched, and if the leg of that side be forcibly held, scratching movements of the opposite leg may often be observed. These movements may be observed in young puppies, and can readily be called forth in animals which are sound asleep. I have been unable to get these movements from cats, although the cat tribe is probably related to the dog tribe by common ancestry. In the rabbit, too, I have been unable to observe them. If, however, a guinea-pig be killed by a blow on the back of the neck, and if the skin at the side of the belly be gently tickled, the animal will bring the leg of that side rapidly to the part, and scratch it violently for some time. I have noticed, too, that after an ox has been killed by a blow of the pole-axe, the hind leg will be brought to the side of the body if that part be rubbed. The move- ment is similar to that made during life to get rid of flies. We see then that these reflex scratchings are sometimes present, sometimes absent in animals nearly related. This variation depends, no doubt, on the habits, but more especially upon the build of the animal itself. The cat possesses great mobility of the head and neck. It can lick its sides, and can reach most parts of its body with its fore 1888.] Prof. Haycraft on Reflex Scratching Movements. 139 claws. The dog cannot do this, and the hind leg is used instead. The rabbit has a mobile and flexible body, which it cleans in a sitting posture with its mouth and fore paws. The guinea-pig and the ox are shorter, with thick -set necks, and the hind leg is called into requisition. One cannot attain spinal scratching reflexes from the human sub- ject, and probably not from the apes. The theory of build, and bodily mobility, will not entirely explain these cases, however, for another consideration appears. This we shall now consider. Co-ordinated reflex movements may be divided into two classes. In the first class we have movements of limbs, the aim of which is to bring them into relationship with other parts of the body. Such are the complex movements of the pithed frog, and the scratching movements we are discussing. In these cases all that is essential for the acquisition of the power of bringing one part of the body into connection with another part is tactile sensibility of the skin. The other classes of co-ordinated reflexes are those which change the position of the body in respect to its surroundings, e.p., walking, swimming, &c. In this case sensation of sight, hearing, &c. are required in addition. The nerves of tactile sensibility for the trunk and limbs pass to the cord in which they make connection with motor-fibres passing to those parts of the body. The nerves of hearing, sight, &c., pass to the brain. It follows from this that the first class of movements may take place in a pithed animal; the latter, never. Now, one constantly finds the remarks that spinal, co-ordinated movements present in lower animals, e.g., the frog, have their centres in higher regions of the nervous system in higher animals. This and similar remarks indicate, I think, a grave misconception. The second class of co-ordinated movements are never purely spinal either in the frog or in any other vertebrate. It is probable in those cases of the first class, in which the co-operation of the brain with the spinal cord is necessar^q that this is not the result merely of higher development, but depends upon other causes, some of which I have touched upon. The scratching movements, quite as complex as any of the move- ments of the pithed frog, require for their performance the cord 140 Proceedings of Royal Society of Edinhurgh, [jan. 30, alone, both in the case of the guinea-pig and dog. The rabbit and the cat certainly do not possess more highly developed brains, yet no such scratching movements can be elicited. The difference does not depend then upon a question of development either upwards or downwards, but rather upon a variation of habit or build. From increased mobility of the body, or from altered habits, the cat and rabbit may have come to use their eyes and head, whereby the brain is called into action, in place of the leg used by some ancestral type. Or, again, it is possible that the dog and guinea-pig may have acquired the use of the leg for scratching from altered habits, or from loss of mobility. JSTor is it difficult to explain the condition of things seen in the human subject. A child is born without such working connections between the sensory surfaces of its body and the corresponding groups of muscles as would lead to the approximation of a limb to any particular part of the skin. This comes only by laborious ex- perimentation on the part of the child. It sees its foot, and directs its hand to it. It feels the touch, and is conscious of the move- ment it has made. By continual practice it can touch most parts of its body. This is learnt only by experience, which has involved the use of sight, and therefore depends largely on the action of the brain. On this account, if the spinal cord be divided, we should not expect a man, upon having the calf of one leg tickled, to be able to Scratch it with the foot of the other leg, because during his extra- uterine development the brain was a necessary factor in producing such movements. Such is, indeed, the case, for although spasmodic jerks of muscles may be called forth by stimuli applied to the skin, an absence of purposive movements is noticed as soon as the cord is severed. 8. Reply to Professor Boltzmann. By Prof. Tait. Hearing, again by accident, that Professor Boltzmann has in the Vienna Sitzungshericlite published a new attack on my papers about the Kinetic Theory, I at once ordered a copy, which has at length arrived. As my papers appeared in our Transactions, I think my answer to this fresh attack should be communicated in the first 141 1888.] Prof. Tail’s Beply to Prof. Boltzmann, place to this Society. The time I can spare for such work at this period of the year is very scant, and Prof. Boltzmann has raised a multitude of questions. I will take them in order. But I must commence by saying, with reference to Prof. Boltzmann’s peculiar remarks on my behaviour as a critic, that, while leaving them to the judgment of readers, I shall have to bring before the same readers several instances in which Prof. Boltzmann has completely misstated the contents or the objects of my papers. This is not a new departure. In his first attack on me he said that I had nowhere stated that my investigations were confined to hard spherical particles ; whereas I had been particularly explicit on that very point. But fresh cases of a similar character abound in this new attack. First. There runs through this paper an undercurrent, at least, of accusation against me for putting forward my results as new, and thus ignoring the work of others. I had no such intention, and I do not think anything I have said can bear such a construction. My knowledge of the later history of the subject is no doubt now considerably greater than it was about two years ago when, at Sir W. Thomson’s request, I undertook an examination of Clerk- Maxwell’s first proof of his own Theorem. But it is still of a very fragmentary character. I had, years ago, read papers by Maxwell and by Clausius ; and had glanced at the treatises of 0. E, Meyer and Watson. I had also made a collection of various papers by Prof. Boltzmann. But I found that, without much expenditure of time and labour, it would be impossible to master the contents of the three last-named works, mainly because the methods em- ployed seemed to me altogether unnecessarily intricate. [I have already stated the impression produced on me by such of Prof, Boltzmann’s papers as I have tried to read, and I need not recur to it.] I therefore set to work for myself, having certain definite asserted results in view, but little knowledge of the processes which their discoverers or propounders had used. After obtaining a demonstration of Clerk-Maxwell’s Theorem, I was led to pursue my investigations into other matters, such as the rate of restoration of the special state, the size of molecules, &c, I brought before the Society such of these investigations as I had more fully developed; and I hope to communicate others, One object which I tried to 142 Proceedings of Boy al Society of Bdinhurgli. [jan. 30, keep constantly in view was to make my papers at least easily intel- ligible. Intelligibility is not too common a characteristic of papers or treatises on this subject. But if I have succeeded in putting some parts of the Foundations of the Kinetic Theory (for to these alone do my papers profess to extend) in a form which renders them easily apprehended, I shall have done a real service to students of Physical Science. The other object at which I aimed was, of course, the verification of Maxwell’s Theorem; and of the extension of it (to all degrees of freedom of complex molecules) which was made by Prof. Boltzmann. Sir William Thomson and myself were, in fact, called to the question by the discrepancies between the observed behaviour of gases and the behaviour which Prof. Boltzmann’s Theorem would have led us to expect. To test this excessively general theorem, I determined to examine certain special cases, and (that these might be, however imperfectly, repre- sented by systems of free particles) it was necessary to assume want of freedom for collision, though confessedly as one step only. I could not, of course, in this way put limits on the excursions or the admissible speeds for different degrees of freedom. Second. While examining, and seeking to improve, the proof which Clerk-Maxwell originally gave of his Theorem, I found it impossible to begin without the assumption of a certain regularity of distribution of masses and velocities ; and of course I sought how to justify such an assumption. I was thus led to believe that collisions, not merely of particles of the two kinds with one another but among those of each kind, are absolutely necessary for this justification. Then I saw that, in complex molecules, perfect freedom of collisions of all kinds of “ degrees of freedom ” could not possibly be secured, and that this might, in part at least, account for the discrepance between Prof. Boltzmann’s Theorem and the observed behaviour of gases. I saw also that, for the truth even of jMaxwell’s Theorem, it was necessary that neither of the two gases should be in an overwhelming majority. Thus these two things, which Prof. Boltzmann now speaks of as “physically less important,” are from my point of view vital to the general truth of his Theorem. Prof. Boltzmann commences his recent paper by citing a “general equation” from the Phil. Mag. of April 1887; and of it he says: — “Bei Ableitung dieser Gleichung habe ich dort im Ubrigen 1888.] Prof. Tait’s Reply to Prof. Boltzmann. 143 genau dieselben Voranssetziingen zu Grunde gelegt, welclie aucli Herr Tait maclite, nur dass icli liber die relative Grosse der Durcli- messer X und A der Molecule beider Gase, sowie liber den Grossen- werth des Verlialtnisses : H2 niclit die mindeste Annahme gemaebt babe.” Tliis is so far from being tbe case, that it was precisely bis assump- tions, and not bis proof, wbicb I disputed. My remark was; — “I tbink it will be allowed that Prof. Boltzmann’s assumptions, wbicb (it is easy to see) practically beg tbe whole question, are tbemselves inadmissible, except as consequences of the mutual im- pacts of the particles in each of the two systems separately. Of course, with bis assumptions, Prof. Boltzmann obtains tbe desired result : — baving in them virtually begged tbe question. He now blames me for not having said a word in refutation of bis proof, for I bad professed my willingness to allow its accuracy without even reading it. There was no discourtesy in that remark : — nothing but a cheerful admission that, in tbe hands of Prof. Boltzmann, such premises could not fail to give tbe result sought. My comments were in fact necessarily confined to the assumptions. For, as I could not admit them, the proof founded on them had no interest for me. Professor Boltzmann assumed that two sets of particles, even if they have no internal collisions, will by their mutual collisions arrive at a state of uniform distribution in space, and of average behaviour alike in all directions. This may possibly be true, but it is certainly very far from being axiomatic, and thus demands strict proof before it can be lawfully used as a basis for further argument. In quoting my remarks on this point Prof. Boltzmann very signi- ficantly puts an “&c.” in place of tbe following words: — “it is specially to be noted that this is a question of effective diameters only and not of masses : — so that those particles wbicb are virtually free from the self-regulating power of mutual collisions, and therefore form a disturbing element, may be much more massive than tbe others.” It was of this preliminary matter, of course, that I spoke when I wrote the following sentence, which seems to have annoyed Prof. Boltzmann : — “ I have not yet seen any attempt to prove that two sets of particles, which have no internal collisions, will by their mutual collisions tend to the state assumed by Prof. Boltzmann,” 144 Proceedings of Royal Society of Edinburgh, [jan. 30, I think it probable that Prof, Boltzmann has not fully appre- hended the meaning of the word ‘‘assumed” in this sentence. Otherwise I cannot understand why he is annoyed because I took his proof for granted. In taking leave, for the time, of this special question, I need scarcely do more, and I cannot do less, than reaffirm the assertion just quoted : — while adding the remark that this is very far from being my sole objection to Prof. Boltzmann’s very general Theorem. In fact Professors Burnside * and J. J. Thomson f have quite recently advanced other serious objections. Prof, Boltzmann’s Theorem, in a word, is not yet demonstrated. Third. As to the questions of viscosity and heat-conduction ; my investigations were expressly made on the assumption that change of permeability, due to motion, was negligible. When I found that I had obtained in a very simple way certain char- acteristic results of Clerk-Maxwell and of Clausius respectively, I was satisfied with the approximation I had made. Prof. Boltz- mann does not allude to the fact that my investigation was distinctly stated to be an approximate one only, and that the additional consideration he now adduces had been before me and had been rejected (rightly or wrongly) for reasons given. I said — “ Strictly speaking, the exponent should have had an additional term See the remarks in § 39 below.” And, in the § 39 thus pointedly referred to, one of the remarks in question is — “ We neglect, however, as insensible the difference between th-e absorption due to slowly moving layers and that due to the same when stationary.” And, in fact, the result which I gave for the viscosity (and which Professor Boltzmann, without doubt justly, claims as his own) is correct under the conditions by which I restricted my investigation. The introduction of the consideration of change of permeability due to the shearing motion involves an alteration of about eleven or twelve per cent, only in this avowedly approximate result. Of this I have assured myself by a rough calculation, and I will work it out more fully when I have leisure. It seems that I missed this in looking over Meyer’s book, and, according to Prof. Boltzmann, ^ Trans, M.S.K, 1887, t Phil Trans., 1887. 1888.] Prof. Tait’s Reply to Prof. Boltzmann. 145 all investigators except Meyer have fallen into the same trap. .Meanwhile the calculation with which Professor Boltzmann has furnished me gives an excellent example of his style, for it is altogether unnecessarily tedious. And it seems to contain two gigantic errors which, however, compensate one another. For his integrand contains the factor Here / is a signless quantity, and the limits show that x is always positive and p always negative. As written, therefore, the integral is infinite, though in the result it is made to come out finite. The object of the paragraphs 1 and 2, which immediately follow, is unintelligible to me. The former seems to suggest the use of an unsound method, the latter has no discoverable bearing on anything that I have written. Prof. Boltzmann has also afforded an idea of the value which he himself attaches to the terrific array of symbols in the 95 pages of his 1881 paper (to which he refers me) by now allowing that he is not prepared to assert that any one of three determinations of the coefficient of viscosity which he quotes (mine, or rather his own, being among them) is to be preferred to the others ! Fourth. Prof. Boltzmann refers to my remarks on Mr Burbury’s assertion that a single particle, with which they can collide, would reduce to the special state a group of non-colliding particles. Prof. Boltzmann signified his belief in the truth of this proposition ' and in answer I showed that (were it true) seons would be required for the process, even if that were limited to a single cubic inch of gas. He now calls this an “ entirely new question ” and will not “ prolong the controversy by its discussion.” I do no see that, so far at least as the “ controversy ” is concerned, it is any newer than the rest. It is contained in the first instalment of his attack. Why then should he now desert it ? But Prof. Boltzmann, in thus leaving the subject, takes a step well calculated to prolong the discussion, for he represents me as speaking of the instantaneous reversal of the motions of all the particles, whereas my argument was specially based on the revei'sal of the motion of the single stranger alone, a contingency which might possibly occur by collision even with a particle of the gas, certainly by collision with the containing vessel. There is a common proverb, “ All roads lead to Home.” It seems it ought now to be amended by the addition, “whether you go backwards or forwards along them.” VOL. XV. 26/6/88 K 146 Proceedings of Royal Society of Pdinhurgh. [jan. 30, Fifth, As to my proof (so designated) of the Maxwell Law of distribution of velocities : — I have already explained that this part of my paper was a mere introductory sketch, intended to make into a connected whole a series of detached investigations, and therefore contained no detailed and formal proofs whatever. Maxwell’s result as to the error-law distribution of velocities, being universally accepted, was thus discussed in the briefest manner possible. I said also that a detailed proof can be given on the lines of § 21 of my paper. Prof. Boltzmann* at first accused me of reasoning in a circulus vitiosus, and went the extreme length of asserting that the inde- pendence of velocities in different directions can do no more than prove the density (in the velocity space diagram) to be dependent on the radius vector only. Now, when I have taken the trouble to point out briefly and without detail what I meant by the state- ments he misunderstood, he says I have admitted that my proof is defective ! For my own part, I see no strong reason wholly to reject even the first proof given by Maxwell ; and it must be observed that although its author said (in 1866) that it depends on an assumption which “ may appear precarious,” this did not necessarily imply that it appeared to himself to be precarious. The question really at issue was raised in a very clear form by Prof. Newcomb, who was the earliest to take exception to my first sketch of a proof. He remarked that it seemed to him to possess too much of a geometrical character {i.e. to prove a physical statement by mere space-reasoning), while Maxwell’s seemed to involve an unauthorized application of the Theory of Probabilities. In consequence of this objection I examined the question from a great many points of view, but I still think my original statement correct. What I said was “ But the argument above shows further, that this density must be expressible in the form f{r)f{y)f{z) whatever rectangular axes be chosen passing through the origin.” In my second paper I said (in explanation of this to Prof. Boltz- mann), that the behaviour parallel to y and 2 (though not the * This addition to Prof. Boltzmann’s first attack on me seems to have appeared in the Phil. Mag. alone. It is not in either of the German copies in my possession (for one of which I am indebted to the author), nor do I find it in the Sitzungsbericlite of the Vienna Academy 147 Prof. Tail’s Reply to Prof. Boltzmann. number) of particles whose velocity components are from xiox-\- dx, must obviously be independent of x, so that the density of “ ends ” in the velocity space diagram is of the form f{x). F('y,i2). The word I have underlined may be very easily justified. No collisions count, except those in which the line of centres is practically perpendicular to x (for the others each dismiss a particle from the minority ; and its place is instantly supplied by another, which behaves exactly as the first did), and therefore the component of the relative speed involved in the collisions which we require to consider depends wholly on y and motions. Also, for the same reason, the frequency of collisions of various kinds (so far as x is concerned) does not come into question. Thus the y and 2 speeds, not only in one X layer but in all, are entirely independent of x ; though the number of particles in the layer depends on x alone. Prof. Boltz- mann’s remark about my quotation from De Morgan will now be seen to be somewhat irrelevant so far as I am concerned, though he may (perhaps justly) apply it to some of his own v/ork. Sixth. As to the Mean Path, though I still hold my own definition to be the correct one, I would for the present merely say that Professor Boltzmann entirely avoids the statement I made to the effect that those who adopt Maxwell’s definition, which is not the ordinary definition of a “mean,” must face the question “Why not define the mean path as the product of the average speed into the average time of describing a free path T The matter is, however, of so little moment, that a very great authority, whom I consulted as to the correct definition of the Mean Free Path, told me that the preferable one was that which lent itself most readily to integration. Seventh. In his remarks upon the effect of external potential, Prof. Boltzmann does not defend his proof to which I objected, but gives a new and fearfully elaborate one. And he quotes, as a remark of mine on this entirely different proof, the phrase “ this remarkable procedure ” which I had applied to his objectionable old one ! He also treats in a disparaging manner the assumption on which my very short investigation is based; viz. “When a system of colliding particles has reached its final state, ive may assume that {on the average) for every particle ivhich enters, and undergoes collision in, a thin layer, another goes out from the other 148 Proceedings of Royal Society of Edinhurgh. [jan. 30, side of the layer precisely as the first would have done had it escaped collision^ Of course it would be easy to make a 20 page proof of this by tbe help of an imposing array of multiple integrals. But this would be the sort of thing which I have called “ playing with symbols,” i.e. using them instead of thought^ while their proper function is to assist thought. A mathematical demonstration does not necessarily imply the use of symbols, any more than that of diagrams : — and, when we find an author continually using symbols to establish what is obvious without them, we very naturally question the validity of his symbohcal processes when they are employed for their legitimate purpose. I still think the assumption above a legitimate and indeed almost an obvious one ; but it is strange that an objection of this kind should come from a writer like Prof. Boltzmann, who (see head Second above) has made, and still defends, a fundamental assumption (of the class to which he applies the term “ unbewiesene Voraussetzung ”) which most clamantly demands proof. Finally, as Prof. Boltzmann objects alike to Greek, and to English, quotations, although they have Plato and De Morgan for their authors, what does he say to the Latin one “ Quis tulerit Gracchos de seditione qiierentes'^ % PRIVATE BUSINESS. Eudolph Julius Emmanuel Clausius, Professor of Natural Philo- sophy in the University of Bonn; Ernest Haeckel, Professor of Zoology and Histology in the University of Jena; Demetrius Ivanovich Mendeleff, Professor of Chemistry in the University ofK|j St Petersburg, who had been proposed as Foreign Honorary Fellows, If I and had been named from the Chair in terms of Law XIL, at theS; Meeting of 5th December 1887, were balloted for, and declared'" duly elected Foreign Honorary Fellows of the Society. 1888.] Prof. Cmm Brown on Canals of Internal Ear. 149 Monday, February 6, 1888. Sir william THOMSON, President, in the Chair. The following Communications were read : — 1. On a Mode of Exhibiting the Action of the Semi- circular Canals of the Internal Bar. By Professor Crum Brown. The problem is to contrive an apparatus in which, by means of inertia, acceleration of angular velocity about any axis in either sense may he observed and approximately measured. In the internal ear we have such an apparatus, if we consider the two ears with their six canals as forming one organ, each canal sensitive to acceleration in one sense about the axis at right angles to the average plane of the canal. The model shown illustrates the principle. It consists of a rectangular frame turning on a vertical axis parallel to the short sides of the rectangle. On each side of this axis there is a heavy wheel with its axle vertical. Each wheel 150 Proceedings of Royal Society of Edinhurgh. [fee. 6, is fitted with a stop which prevents its rotating, relatively to the frame, beyond a certain position (which we may call its normal position) in one sense. One wheel, W, cannot rotate beyond its normal position in the positive sense, the other, W', cannot rotate beyond its normal position in the negative sense. W can rotate negatively, hut in doing so stretches a spring, and the spring is made strong enough to prevent any hut a very slight angular movement, with the greatest acceleration to which the instrument can he exposed. Similarly W' can rotate positively, hut in doing so stretches a spring ; in fact, W and W' are mirror images of each other, the two springs S and S' being as nearly as possible equal. If, now, the frame receives an acceleration of positive rotation, the two wheels tend to rotate negatively, relatively to the frame ; hut W' cannot do so, it is forced by its stop to rotate with the frame. But W does rotate relatively to the frame and stretches its spring. The extent to which the spring is stretched is approximately a measure of the acceleration. If we could keep the acceleration con- stant, the wheel would remain at the same angular distance from its normal position with its spring stretched. But if we make the acceleration zero, i.e., make the rotation uniform, the spring brings hack the wheel to its normal position. What is true of W and W' with positive acceleration, is of course true of W' and W with negative acceleration. While the frame is rotating, we cannot easily see whether the wheels are in their normal position or not, or how far they have rotated from them. It is necessary, therefore, to contrive some way of indicating this. In the model shown this is done by leading gas through the lower part of the axis of the frame, and by two pipes, one to each wheel. On the axle of each wheel, where the gas-pipe passes it, there is a stop-cock. In the normal position this stop-cock is nearly closed, so as to allow only a little gas to pass ; as the wheel rotates away from its normal position the stop- cock opens. From the stop-cocks the gas-pipes pass round to the upper part of the axis of the frame, and pass out through it through a joint to two fixed gas jets. Acceleration in the positive sense opens the stop-cock of W, and the corresponding gas jet flares up. Acceleration in the negative sense opens the stop-cock of W', and its jet flares up. When the rotation of the frame is uniform, whatever its rate may he, the wheels remain in their 1888.] Prof. Crum Brown on Canals of Internal Ear. 151 normal position, and the gas jets at their minimum. When this rate changes, we have acceleration in the one or other sense, and this is indicated by the flaring up of the corresponding gas jet. The model illustrates merely one-third of the complete apparatus, as it shows the results of acceleration about one axis only, and in both senses about that axis. Perhaps, in making such a model, it would be better to work the stop-cocks by means of cranks from the wheels, and so diminish the friction of the axles on their bearings. However, in the model shown, this friction is so small that very moderate acceleration is well indicated. The apparatus was made, from Professor Crum Brown’s instruc- tions, by Mr Alexander Prazer, 7 Lothian Street, Edinburgh. 2. On the Temperature and Currents in the Lochs of the West of Scotland, as affected by Winds, By John Murray, Esq. 3. Note on the Influence of Pressure on the Solubility of Carbonate of Lime in Sea Water containing Free Carbonic Acid. By W. G-. Reid. Communicated hij John Mukrat, Esq. Analysis of the dredgings brought to the surface during the Y oyage of H.M.S. “Challenger,” has shown, that in deeper water as the depth increased, the quantity of carbonate of lime shells decreased,* and as the pressure is in direct proportion to the depth under water, it was surmised that some connection existed between the pressure and the disappearance of lime shells. To ascertain if there was any truth in this surmise, Mr Murray suggested the following experiments. The results are unfortunately incomplete j neverthe- less, Mr Murray thinks it advisable to publish them. During this investigation, I had the honour of working with Mr H. N. Dickson, who, with his hands full of more important work, * Murray, “On Coral Keefs,” Proc. Roy. Proc. Eclin., 1880, p, 509 ; and Narrative of the Cruise of the Challenger, p. 923. 152 Proceedings of Royal Society of Edinhurgli. [feb. 6, superintended the physical part of these experiments with cha- racteristic patience and kindness. This subject does not seem to have been previously investigated. Th. Schloesing and others have demonstrated that the solubility of carbonate of lime, &c., in water containing carbonic acid, increases as the pressure of carbonic acid increases, and that according to a definite law ; but nothing is said about the effect on the solubility, when, the quantity of carbonic acid in solution remaining the same, the pressure is in- creased. The latter was the object aimed at in these experiments, and as they had special reference to the conditions existing in the ocean, sea water was taken and charged with a definite quantity of carbonic acid, that the effect might be exaggerated, and therefore more easily studied. Tor the experiments done under pressure the following modus operandi was adopted: — A (see sketch) is a Bohemian glass funnel, having a glass cover ground to fit; a flat india-rubber band of the same circumference as the cover is put between it and the funnel. A rubber capsule is now stretched over the top, and for greater security the neck of the funnel is passed through a slit in a strong rubber band, which is then stretched over the capsule. The funnel contains a filter paper, and a cambric bag, within which is a weighed quantity of shells. To the funnel the bulb B is attached, and this is connected by the glass tube C to a vessel containing mercury D. The whole is tied to a suitable support, and is ready for immersion in the water, contained in the pressure apparatus, i.e., the celebrated “gun” belonging to the “ Challenger ” Commission. The bulb B, which has been accu- rately measured, contains sea water charged in the following manner Avith a definite quantity of carbon dioxide. B is first filled with sea 1888.] Mr W. G. Reid on Carlonate of Lime in Sea Water. 153 water, noting the temperature, and the pieces of india-rubber tubing at each end securely clamped. The tube C is attached and secured with wire, then filled with mercury, and another bulb, also accurately measured, and filled with carbon dioxide at a known temperature and pressure, is attached in the same way as B, to the other end of the tube. The clamps are now unscrewed, and the carbon dioxide is allowed to come in contact with the water, and as it is absorbed the vacancy so caused is filled up by allowing mercury to be sucked in. When the carbon dioxide is all dissolved, the mercury is allowed to flow to the smaller bulb (always that which contained the carbon dioxide), which may be detached after clamping the rubber tube at the bottom of tube C. Where pressure is applied, the mercury is forced up into the bulb, and the water into the funnel, the air in which at the pressure employed (4 tons per square inch, or nearly 600 atmospheres) contracts to a very small bubble, and thus allows the water to come into contact with the shells. In these experi- ments the pressure was kept at its maximum for 30 minutes, then released, and again applied and kept up for 60 minutes. When the pressure is let off, the air expands and drives the water out of the funnel ; the effect therefore of the break in the application of the pressure is to cause a slight agitation, and to bring a fresh portion of the liquid into contact with the shells. After each experi- ment the alkalinity of the water was carefully determined, and the original alkalinity deducted therefrom: the quantity of carbonate of lime dissolved was thus ascertained. Each of the experiments made under pressure w^as repeated, at the ordinary pressure, under as nearly as possible the same conditions. The funnel taken in this case was larger, and a scratch on it indicated the capacity of the funnel used for the corre- sponding pressure experiment. A glass cover protected the con- tents of the funnel from dust. The lower end of the bulb B was attached to a tube which passed through an india-rubber stopper in one neck of a small Woolff’s bottle, through the other neck a funnel tube about 20 inches long was passed. Both tubes dipped under the surface of mercury contained by the Woolff’s bottle, and by pouring mercury into the funnel tube, the water was forced into the funnel until it reached the scratch aforementioned. Table showing Influence of Pressure on the Solubility of Carbonate of Lime in Sea Water containing Carbonic Acid. 154 Proceedings of Royal Society of Edinburgh. [fee. 6. ^'00 "rajg jad ^ paAiossfa ^00^0 ^ OT'^SOCDOCMlCt^COlC’tOOii— IOiCCVOrH(M^CiOOt'~^001>.(NO'iD r30i-HOOCO'SHC-^DOCOi-HCiOOt^Or-(00<^i— l(MC^C.COt^aD(M(iikO'^OOCMiL|cf lO ^ CM CM CO 'SI rH 1— I CD ^ I-H t-H r-UO pH CM rH pH •aJXTl ^ aad jCxju’n'BJnY g Sutiinsa'a ^ ipOOi^Oc:»?prHipOi'^T7HO'^05^(rocX3g5(:pOiHHf^j;Ht^O'COI>. i-nt^'HOicbcOlOWCiUO)— ICOOt)?DrHOC75^(X)a54Ht^t^lOO:)COCOlO 01i-HOa5t>-l>-CJ5CX)OiCC)iX>fH05t'-l>.?OOCOOiI>-QOlOOOlOUOOCO pH rH r-l pH Tempera- ture. (0 pcpopcMOM^c»pq.iOCOOOOO'H>0'SilOkO pppppppppppppppppppppppppoppo ^pH 1 — li — ipHpHi— li — li — li— li— Hi — li — IpHpHi — 1 i — li — li — ll — ll — 1 i — ii — li — li — IrH Yol. (h) ,Y^pppp-^pi^ppp'^^qocbrLioocMCM oOCOCOi— ll— I'H'HpHCOOOCOpHCMOOOCOh)H'SHCOoppppiopppppppHHpY'‘P‘P ‘^3jZ];ZlS2PlPlS^dS?HgHSjHSgH^H?Hl^gHgH'^S?H?Hl^jHjH Carbonate of Lime used. Quantity. (d) . CM t^t^O O O O O O O O O »0 O O O >n O O t^C35i-UOOO ^CTlCOCSiOCMCMCMCOr-'CMOOOCMt'-CMCXDOiOiOCM OOt^t^lOiCO gooo<:oooo<:o?oO'coocoioocoocono<:ocx)o : : '.cocococooi ^ppioppuopppppppppppppp • • ■pppp'? ^iL| PPP^PPPPi-Hi-Hi-H,— |rHi-HpHiL|P' 05 COCOCOCOCM Kind. (c) Glob. I. Do. Do. Do. Do. Do. Do. Glob. II. Do. Do. Do. Coral sand I. Do. Do. Do. Do. Coral sand II. Do. Pteropods. Crystal. bo. Do Do. Do. Do. Do. Do. I Do. ground. Capacity of Funnel. ih) . p p p p ^ p p p p rH p p tH 1^ pH p pH tH 17c glOVOHHlOkO'^THlO'SHlOHtHiHlOlHldpHpHpHpHiHpHpHr-HpHiHlHiHiH «'CDC30aj?OCX)CO00i31<:OOSSDOiO5:55aiO5O:iO^OiO^C5iO5C35Oi No. of Experiment. (a) I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XV. XVI. XVII. XA'IIT. XIX. XX. XXL XXII. XXIII. XXIV. XXV. XXVI. XXVII. XXVIII. 1888.] Mr W. G. Reid on Carhonate of Lime in Sea Water. 155 The accompanying table gives the results of the experiments. The columns of the table are explained by the headings, and Mr Murray adds the following notes with reference to the contractions used in column (c). Globigerina Ooze. — Collected on the 21st March 1876, in the South Atlantic. Lat. 21° 15' S. ; long. 14° 2' W.; depth, 1990 fathoms. Specimen /., consists of the larger shells in this deposit, such as the shells of the Pelagic Globigermidse, such as Orhulina universa, GloUgerina liustigernia, Splieroidina, Pullenia, Pul- vinulina. In addition to these there were the shells of a few bottom living Poraminifera, as Biloculina, and fragments of Echinoderms, Lamellibranchs, and otoliths of fish. The average size of the shells and particles in this specimen is about *6 of a millimetre. Specimen //., consists of the smallest shells in the same deposit, being almost wholly made up of young shells of the above mentioned Globigermidse. The average diameter of the grains of this fine sand are less than T of a millimetre. Coral Sand. — Collected off the Great Barrier Reef of Australia on the 31st August 1874. Lat. 11° 35' 25" S. ; long. 144° 2' E.; depth, 135 fathoms. Sample /., consists chiefly of the coarser fragments of these deposits, and is made up of particles of broken Pteropods, Gas- teropods, Lamellibranchs, Echinoderms, Polyzoa, SerpuliB tubes, and numerous Poraminifera, The average size of the fragments were from 2 to 3 millimetres in diameters. Sample II. This was a sample from the same deposit, and made up of the same kind of fragments as Sample L, but these were con- siderably smaller in size, Pteropods. — These consisted of the shells of Cavolinia clio^ Cuvierina^ Limacince, and shells of Atlanta. These were com- plete, or nearly complete shells, and apparently free from sand and mud, and were picked out from the coral sand above men- tioned. The last column in the table (column^?) contains the results stated, so as to render all the experiments comparable. Taking these figures, we have the following average results : — 156 Proceedings of Royal Society of Edinhurgh. [feb, 6, Glohigerina Ooze, I. At 4 tons pressure, . At 2 „ 5, At atmospheric pressure, Amount of CaC03 dissolved per grm. , COg taken. Difference from Extremes. •1121 ±*03 •1019 •0553 ± -0009 GloJ)igerina Ooze, II. At 4 tons pressure, . At atmospheric pressure. Coral Sand, I. At 4 tons pressure, . At atmospheric pressure. •0846 •0252 •1155 •0419 ±•009 ±•0017 ±•015 ± -0006 Pteropods. At 4 tons pressure, . . •lOlS Crystal of Iceland Spar (XXIV.- XXVII.). At atmospheric pressure, • ^0050 ± -0018 Crystal, ground to coarse powder. At atmospheric pressure, . ^03 2 2 The disparity between the various results obtained in the pressure experiments I am unable to account for satisfactorily. Xevertheless, the amount of carbonate of lime dissolved at a pressure of 4 tons per square inch, is so much greater than the amount dissolved at the ordinary pressure, that I think it justifies the conclusion that the effect of pressure is to increase the rate of solution; or, in other words, that the chemical activity of a solution of carbonic acid is increased by pressure. It is to he noted, that although these results may indicate that the solution of carbonate of lime in carbonic acid water is more rapid under high pressures, it by no means follows that the solu- bility is greater than at the ordinary pressure {ceteris paribus). Schloesing and other investigators have shown, that in order to get 1888.] Mr W. G. Reid on Carbonate of Lime in Sea Water. 157 tlie maximum amount of carbonate of lime dissolved, the carbonic acid solution had to be left in contact, and agitated with the carbonate for five or six days. With the apparatus at our command we could not accomplish this, and had to rest contented with the results given. In the experiments XX. to XXIII. , a crystal of Iceland spar was taken. The results show a gradual falling olf in the quantity dis- solved. The reason for this I cannot explain, but that it is not due to the properties of Iceland spar is shown by the experiments XXIV. to XXYII. Tor these another crystal was taken, and after each experiment it was washed, dried, and weighed carefully. The amount of carbonate dissolved by 117*5 c.c. of sea water (the total quantity taken for each experiment) was as under. A is the amount obtained by titration (alkalinity), and W the loss as observed by weighing. Considering the smallness of the quantity to be measured, and the opportunities for observational error, the results agree fairly well with each other. XXIV. XXV. XXVI. XXVII. A *0009 grms. *0006 „ *0004 „ *0006 „ W *0008 grms. *0008 „ •0010 „ *0007 „ For the last experiment (XXVII.) the crystal, used in the preceding four, was ground to a powder, the grains of whicli varied from about 1 mm. square down to impalpability. This was done to try the effect of increasing the surface exposed. As was expected, the amount dissolved was much greater (six times). My thanks are due to Mr T. Lindsay for kind assistance in some of these experiments. 158 Proceedings of Royal Society of Edinhurgli. [fee. 6. 4. On the Distribution of Carbonate of Lime on the Floor, and in the Waters of the Ocean. By John Murray, Esq. {With Lantern Illustrations.) 5. On the Number of Dust Particles in the Atmosphere. By John Aitken, Esq. PRIVATE BUSINESS. Mr James Mactear, Mr John M Arthur, Mr Charles A. Fawsitt, Mr George Brook, Professor W. H. Perkin, Mr H. N. Dickson, Mr David Prain, Mr George Muirhead, and Mr Cathcart W. Methven were balloted for, and declared duly elected Fellows of the Society. PEOCEEDINGS OF THE ROYAL SOCIETY OF EDINBURGH. VOL. XV. 1887-88. No. 127. Monday, 20th February 1888. Sir william THOMSON, President, in the Chair. The following Communications were read : — 1. Preliminary Note on the Duration of Impact. By Professor Tait. The results already obtained were got by means of a roughly made apparatus designed for the purpose of testing the method used, so that only a single instance, to show their general character, need now be given. When a wooden block of 10 lbs. mass fell through a height of 18^ inches on a rounded lump of gutta-percha, the time of impact was found to be somewhere about O’OOl sec., and the co- efficient of restitution was 0‘26. As the principle of the method has been found satisfactory in practice, new apparatus is in course of construction, which will enable me to use a fall amounting to 10 feet at least. It is proposed to make a series of experiments on different substances, Avith great varieties of mass and of speed in the impinging body. 2. A Bathymetrical Survey of the Chief Perthshire Lochs, and their relation to the Grlaciation of the District. By James S. Grant Wilson, Esq., H.M. Geological Survey. Communicated by Dr Archibald Geikte, F.R.S., Director-General of the Geological Survey. VOL. XV. 1/10/88 L 160 Proceedings of Boy at Society of Edinlv^rgh, [fee. 20, 3. Contact-Phenomena of some Scottish Olivine-Diabases. ^ By Ernst Stecher, Ph.D.,' Leipzig. Communicated hy * Archibald Geikie, F.R.S., Director-General of the Geological ^ Survey."^ In the second and third numbers of the ninth volume of R I TscliermaP s Miner alogisclie und petrographische Mittlieilungen^ I published the results of an examination of some “ dolerites ” from the Carboniferous basin of the Firth of Forth, the material for which | I obtained through the kindness of Prof. Archibald Geikie. It is true most of the “ dolerites ” are already well known through the valuable writings of Professor A. Geikie t and Mr Allport, J but a special study of the modifications of the rock in the different parts of any one volcanic mass has not yet been the subject of any paper. Having elicited many facts of petrological interest, I purpose giving- in this communication a short account of my investigations. At the outset, I may state that the specimens which I have examined were obtained from the following localities : — Salisbury Crags, Hound Point, Stewartfield (near Broxburn), head of Aherdour pier, Hawk-Crag (near Aherdour), Colinswell, the coast west of St Monan’s Church, Dodhead Quarry and Kilmundy Quarry (near Burntisland), Simnyhank Quarry (near Inverkeithing), and FTew- halls (near Queensferry). These rocks have all been grouped hy Mr Allport under the common name of dolerites. Professor Geikie, however, assigned them partly to the dolerites, partly to the diabases. My conviction, arrived at hy the study of the rocks to he described. * My friend Professor Zirkel, having told me that he was often at a loss for a subject of original investigation to prescribe as a thesis for the doctorate at Leipzig, and having asked me for assistance in the matter, I suggested the contact-metarnorphism and connected phenomena of the eruptive rocks in the basin of the Firth of Forth, which had never been fully investigated. He accepted the suggestion, and I forw-arded to him a series of specimens collected as material for the purpose of working out the subject. The thesis was eventu- ally taken up by Dr Stecher, and the results of his investigations are con- densed in the present paper. — A. G. d Arch. Geikie, “ On the Carboniferous Volcanic Rocks of the Basin of the Firth of Forth,” 1879, Text- Book of Geology, p. 535, &c. d S. Allport, “On the Microscopic Structure and Composition of British Carboniferous Dolerites,” Quart. Jour. Geol. Soc., 1874, vol. xxx. p. 529, &c. A i 1888.] Dr Stecher on Contact-Phenomena. 161 is, that they belong solely to one group, and this for the following reasons : — 1. Orthoclase does not occur in any notable quantity. The “ divergent herring-bone lineation from the plane of twinning,” that Professor Geikie* * * § believes to be of orthoclase, was observed but rarely, and where found (Hillwood Bath) proved to be the weU-known micropegmatitic structure. The species of felspar in which the quartz is included could not be satisfactorily determined. Excepting these few undecided cases, the felspars represent plagio- clase. Although in the heart of the more extensive intrusive sheets, this plagioclase always occurs as untwinned crystals, it is never- theless a triclinic felspar. It exhibits zonal structure (the so-called “ isomorphe Schichtung”), and the extinction angles, measured in the most widely separated zones of one and the same individual, vary up to 40°. These peculiarities have been shown by Tbrnebohmt and HopfnerJ to be characteristic of plagioclase felspar. 2. As will be showm in a subsequent part of the present paper, the quartz grains that occur in these rocks are due to secondary influences ; they cannot, therefore, be considered as characteristic. However, many of Geikie’s “ dolerites ” {e.g., Hound Point) contain in certain zones a considerable quantity of quartz, occurring as an authigenic constituent. 3. With respect to the olivine, what I have just stated with reference to the quartz is also true here. All the differences, even the most extreme, can be reduced to the influence of the associated rocks on the intrusive sheets which have invaded them. 4. Moreover, we must bear in mind the peculiarities of structure. But these were in no case found to prevail throughout the whole of a volcanic mass. Neither could I find in the specimens examined by me the distinctions pointed out by Professor Eosenbusch, § namely, that Geikie^s “ diabase ” has a “ normal hypidiomorphic,” and the “ dolerite ” a “ typiscli diabasiscli-hdrnigC or ophitic structure. * On the Carl. Vole. Rocks, &c., p. 488. t Tornebohm, On Sveriges wichtigare diahas och gabhro arter. + Hopfner, “ Ueber das Gestein vom Monte Tajumbina in Peru,” Neucs Jahrh.f. Min., 1881, vol. ii. p, 164. § Rosenbusch, M.ikroskopische Physiographie, Aufl. 2, Band. ii. p. 193. 162 Proceedings of Boyal Society of Edinhurgh. [fee. 20, To sum up these results, we may say that all the rocks in question belong to a single type, which we may call olivine-didbases. The material at my disposal, having been selected in the field, enabled me to make a special study of the endogenous contact- phenomena. The rocks are comparatively fresh ; and they represent specimens from the peripheral portion (Salband) as well as from the central and from intermediate parts of the volcanic body. Thus I was able to point out that the volcanic mass has been modified, both in regard to substance and to form, and not only throughout the whole mass, but also in its different parts, and that this modifi- cation is to be ascribed to the influence of the associated rocks. I. Modification in Substance. We are often astonished at the great quantity of fragments enclosed in volcanic rocks, and it has been stated in several treatises that fragments of the associated rocks have been melted down by the igneous magma. Nobody, however, has succeeded in proving that a given modification of the volcanic magma has been produced by partial or complete assimilation of fragments of other rocks. The investigation of this question could be most favourably under- taken in the intrusive masses of the Salisbury Crags, Hound Point, and Stewartfield. These bodies, up to 100 feet in thickness, present in their central portions a rock which has been formed by a magma that cooled down with extreme slowness. At the immediate junction with the associated rocks, however, the magma consoli- dated rapidly. In this way the volcanic mass avoided all the influences that would have arisen if the magma had melted down extraneous material. We should, therefore, expect the portion of the rock that consolidated most rapidly to represent that rock- type to which the magma originally belonged. All these phenomena are very clearly illustrated by the above-mentioned specimens. At the immediate junction the diabase exhibits a great quantity of well-formed crystals of olivine. At a small distance from the con- tact these crystals of olivine are more or less corroded. Towards the heart of the mass the olivine is either absent, or it occurs sparingly in isolated rounded grains. Olivine, which, owing to its basic composition, is one of the first minerals to crystallise 1888.] Dr Stecher on Contact- Phenomena. 163 out, underwent re-solution in the magma ; the latter being rendered acid by the assimilation of acid material. If this material were added in sufficient quantity, and the magma were maintained during an enormous period of time at a uniform temperature, all the olivine would be dissolved; while in a rapidly consolidating mass such a destruction of the olivine is impossible. In all probability, there is a close analogy between these facts and the data given by Liebe and Zimmermann,* Rohrbach,f &c., who regard the olivine as a so-called endogenous contact-product. Such an interpretation, unsatisfactorily established, is only apt, I think, to lead us to another problem. It seems to me more advisable to apply the above explanation to those olivines which have been proved to be endogenous contact-products. We may thus obtain further support for the opinion just expressed. The supposition that allothigenic material has been corroded and eaten into by the igneous magma is supported by the following observations : — The Salband of the olivine-diabases of Hound Point and Salisbury Crags envelops numerous fragments of the associated rocks. The fragments vary in size from minute microscopic grains up to great masses, some of which have been figured by Professor Geikie in the sections he gives {loc. cit.). At a short distance from the contact the diabase invades the enclosed quartz grains in a sinuous manner, and there occur aggregations of quartz grains which have the appearance of being mechanically welded together. Professor Geikie { has remarked that large fragments of sedimentary rocks must have been melted down by the igneous magma. But the slow-cooling in the central portion of the volcanic masses favoured the neutralisation of the acid substance which had been added to the original basic mass; nearer the contact such chemical changes could not occur. In this latter case, a surplus of silica remained, which separated and crystallised as automorphic quartz (Hound Point, 12 feet above base). The specimens from these parts of the massive rock cease to exhibit anything that might suggest the earlier existence of olivine. The volcanic * Liebe and Zimmermann, “ Die jiingeren Eruptivegebilde im S. W. Ostthiiringens,” Jahrh. d. preuss. Landeranstalt, 1885, p. 178. t Rohrbach, “ Ueber die Eruptivgesteine im Gebiete d. ma’hr. Kreideform,” Tschermak' s Mineral, u. petr. Mitth., 1885, vii. pp. 27 and 54. X Text-Book, pp. 535, 536. 164 Proceedings of Royal Soeiety of Edinhurgli. [feb. 20, magma must liave been predisposed to consolidate as olivine-diabase with a great quantity of olivine. In the central portion of the large bodies, however, the igneous magma, as I explained above, acquired a greater acidity. Thus olivine could not separate, or the olivine crystals, already formed in the fluid magma at an earlier period, were corroded or wholly dissolved, so far at least as they were not preserved by a sudden consolidation of the whole mass. Such a sudden consolidation took place near the Salhands, and here the diabases are often exceedingly rich in well-defined oli- ! vine crystals. These suggestions may be further supported by the following details. The plagioclase-crystals in the heart of ' the more extensive diabase masses — as, for instance, from Auchen- , ; sterry, Hillwood Bath, Avoabridge near Linlithgow, St Margaret’s , ! Hope, Hound Point — show well-developed zonal structure, caused , by the acidity gradually increasing towards the centre of the ' “ crystal. Thus the extinction angles often differ in the different . j | zones of one crystal up to 40°. Although we must bear in mind , ; that the more basic minerals, as a rule, are the first to separate, and that the remaining magma thus gradually acquires a greater _ f acidity, I suppose this extreme variation in the acidity of the ; • zones of one felspar individual to signify that the acidity of the j . .. magma must have been absolutely increased by means of a real j addition of silica. In fact, I was able to prove by analysis that the ' diabase becomes more basic the nearer it approaches to the central j | portion. Further, I think it worth noticing that the immediate ' . i vicinity of the corroded olivine crystals scattered through the dia- ? ; base exhibits a specially fine-grained structure. This observation , f' j points, I think, to the conclusion that there is a chemical difference ^ ^ between the substance immediately surrounding the olivine and I that of the whole rock. i While the fresh diabases represent intrusive sheets invading the I Carboniferous sedimentary rocks of the basin of the Firth of Forth, \ there are also the so-called “white traps,” which, however, only i reach a thickness of 3 to 4 feet. At first sight, they seem to exhibit j' no similarity to diabase, t Colinswell, near Burntisland, however, f there occurs an intrusive sheet of 30 feet in thickness, which con- I' sists in its interior of normal diabase, while towards the outside the « | rock passes into “ white trap.” There appears to be evidence that 1888.] Dr Stecher on Contact- Phenomena. 165 the “white trap” is but the result of the decomposition of diabase rock. The original identity of these two rock-types can also be very clearly demonstrated by examining the rocks under the micro- scope. The “ white traps ” all exhibit a microporphyritic structure and an ophitic ground-mass : the scattered crystals, often occurring in abundance, may be recognised by their outlines as having once represented olivine. These porphyritic crystals seldom show rounded forms; in such a case we are unable to say whether we are dealing with olivine or with augite (Newhalls). To sum up the description of these “ white traps,” it is chiefly to be noticed that nearly all of them occur in sheets of small thickness, and that, for the greater part, they are rich in disseminated olivine individuals. II. Modification in Form. (a) Modification of the Structure of the Rocks. — In the interior of the more extensive massive sheets, — as, for instance, of the Salisbury Crags, Hound Point, and Stewartfield, — the rock always exhibits a granitic or doleritic structure, while towards the Salband it assumes that of a microdiabase-porphyry, of which the ground-mass is ophitic. Passing through gradually finer-grained varieties, it is finally found as a true glass, in such places where, under the influ- ence of the associated rocks, the liquid magma was able to cool down with extreme rapidity. In spite of the great age of the diabases we are describing — Professor A. Geikie has pointed out that they are of Carboniferous age — a thin vitreous layer still exists at the outmost Salband of some rocks ; it is even to be detected at the Salband of two white traps, viz., from the shore west of St Monan’s Church and Kilmundy Quarry, where it forms a “ Gangbreccia.^' It is interesting to note that the glass is insoluble in hydrochloric acid. Finally, we may refer to the white traps, which, forming sheets of only small thickness, never possess the microdiabase- porphyritic structure. ifi) Modification of the Form and of the Structure of the Component Minerals. — The titaniferous iron-ore occurs in the rock-specimens from the interior of the larger massive sheets, as irregular patches, and, gradually diminishing in size as the distance from the centre increases, finally assumes the form of minute globulites. This tran- 166 Proceedings of Royal Society of Edinburgh, [fee. 20^ sition represents a slow gradation ; and I was thus able to show that network-skeletons always occur as the intermediate form between the irregular patches and the globulites. On the one hand, these skeletons increase regularly in quantity, while that of the ilmenite patches diminishes ; and, on the other hand, the former gradually diminish in favour of the globulitic grains. The apatite, which occurs in some specimens in large crystals visible to the naked eye, becomes a constituent of the ground-mass, where the rock assumes a microporphyritic structure. The olivine, however, continues to be separated out from the magma in large porphyritic crystals. According to these observations, we must conclude that the olivine belongs to an earlier period of consolida- tion than the apatite. In considering the modifications of the structure of the indivi- dual minerals, we have first to call attention to the plagioclase. In the interior of the larger diabase masses the plagioclase crystals seldom exhibit polysynthetic twinning; they are in most cases simple twins, or even totally untwinned crystals. In the same manner the lath-shaped crystals of the coarse-grained ophitic types are seldom composed of poly synthetic twin-lamellse. ITear the junction with the associated rocks, however, the great majority of the felspar crystals are finely twinned, often on two types. Besides the single twin-lamellse, which exhibit sharply-defined and entirely uncorroded forms, they sometimes appear to have been separated with intercalated interspaces between each other. The first impression given is, that the lamellae have yielded to a strain which operated on the crystal from without, and endeavoured to pull it asunder. Many lamellae really appear to have been drawn out of the crystal. As regards the augite, I have pointed out an analogous pheno- menon. Where the augites are sufficiently fresh to exhibit dis- tinct optical properties, they may all be recognised as untwinned crystals, if the specimen has been taken from the central part of a large massive bed. But the greater the proximity to the outside of the body the more does the quantity of untwinned augites diminish, while that of twinned individuals increases, till at last all the augites are twinned. I have expressed these data in the following table : — I s I i i 1888.] Dr Stecher on Contact- Phenomena. 167 Distance from the Junction. Salisbury Crags. Found Point. Stewartfield. Pier, Aberdour. Heart. All nntwin- ned. All untwin- ned. 4-5 feet. Untwinned : twinned = 16 ; 3, &c. 2 feet. About half are twinned. j 1 From the vicinity of an enveloped fragment. J The slice is per- pendicular to the junction. Mostly- twinned. 4 inches. Seldom twin- ned. 1 inch. Twinned : un- twinned =4:3. 1 inch. Mostly twinned. From the im- mediate junc- tion. Nearly all twinned. So far as I am aware, such a relationship has not hitherto been noticed; nevertheless, I suppose that a similar dependence might also be established at other localities as well as in other rocks. I have already visited a great many localities in Saxony, Bohemia, and Thuringia, where augitic volcanic rocks occur in junction with the associated rocks ; yet in the few cases where I found the volcanic rock fresh up to the contact, I never succeeded in detecting this interesting phenomenon. Bor in none of these cases did the volcanic mass exhibit any modification of its structure towards the junction. There appears to be evidence that the associated rocks, where ex- amined, have exerted no cooling influence on the igneous magma, a condition which would be necessary for the production of such variations in crystalline structure. As will be seen in the above table, the proportion of twinned individuals of augite not only increases towards the junction with the contiguous rocks, but also depends on the distance from the junction with included fragments, so far is these have a considerable size. In the latter case this twinning of the crystals is found within a smaller distance. 16B Proceedings of Poijal Society of Pdinhurgh. [fee. 20, Finally, I have to call attention to the quartz. This mineral exhibits most peculiar properties. In some specimens of the diabases (viz., Hound Point, 8 feet above the contact with a fragment of sandstone, which is enclosed about 4 feet above the base of the bed, and 15 inches from the junction with an included fragment of sand- stone) there appear (most clearly seen during the preparation of the slices) quartz grains, surrounded by sharply contoured hexagons. An examination with the naked eye, however, suffices to show that these hexagons are sometimes filled up witli calcite substance, and that in most cases their central portion only represents a round grain of quartz, the hexagonal outline being given by calcite which surrounds the quartz grain. I have only once been able to find a hexagon that was filled throughout with quartz substance represent- ing a true quartz crystal. It seems possible that all these hexagonal outlines were in former times due to automorphic quartz crystals, which no longer occupy all the space they may have filled before. Being unable to offer a sufficiently clear explanation of these, I have contented myself with describing them. In the sliced specimen the sharp outlines of the hexagons have disappeared. With regard to the quartz substance itself, we find that between crossed nicols it exhibits a peculiar structure. It might fairly be supposed that, between crossed nicols, a section of a simple crystal should exhibit the same colour in all parts. We are therefore astonished to see the rounded quartz grains which I have described, variously coloured along their radii, or even consisting of several grains, each of which exhibits a homo- geneous optical orientation. The former phenomenon might be taken for the spheroidal structure of chalcedony, unless the quartz sub- stance enclosed Sorby’s “stone-cavities,” and gave no interference figure between crossed nicols. The latter phenomenon, however, might be interpreted as the combination of a right-handed quartz with a left-handed one, unless the single grains within one hexagon exhibited, between crossed nicols, more than two different colours. It remains to answer the possible objection that several quartz crystals could have grown together with parallel crystallographic axes. This interpretation is refuted by tlie fact that all the hexagons are always sharply contoured and regularly formed. Therefore we cannot but assume that the quartz exhibits an anomalous pheno- menon of polarisation, hitherto unnoticed, at least as far as my 1888.] Dr Stecher on Contact-Fhenomena. 169 experience goes. Bearing in mind the statements of Mligge^* that “ ahnlich wie stark erhitzte nnd der Abkiihlung ausgesetzte und deshalb doppelt brechende Objectglaschen sofort ganz oder nahezii isotrop werden, wenn das Glas zerspringt,” the anomalous double refracting glass of natural pitchstone returns to its normal isotropic state as soon as the glass develops perlitic fissures, I shall try to interpret the structure of the quartz grains we have been describing as the effect of a strain with the tendency to split each quartz crystal into diverse grains, differing in their optical orientation. This ex- planation is fully borne out by the following arguments. In the outer portions of the quartz, where it might seem to have been in contact with the magma, and therefore to have been exposed to a specially powerful strain, the optic anomaly exi3lained above is sometimes repeated on a minute scale. The supposition that the quartz crystals had been exposed to a strain will be afterwards discussed. Another fact observed, as proving the pyrogenic origin of the quartz, and otherwise illustrating the effect of a straining force, should be men- tioned. The very few crystalline quartz grains that do not exhibit between crossed nicols anomalous phenomena, represent, as a rule, the smallest individuals ; these, besides, are more or less free from enclosures, while the crystals that give anomalous phenomena are rich in gas- vesicles and “stone-cavities” with fixed vesicles. These “stone-cavities” often exhibit the dihexagonal form of the including quartz, the crystallographic axes being parallel to those of the latter ; and thus undoubtedly prove their glassy nature. All the inclosures are sometimes arranged in rhombohedral planes, parallel to which there occur in one specimen microscopical fissures that might suggest either cleavage-planes or gliding-planes. Eeturning to the conclusions we arrived at respecting the quartz, and relying upon the state- ments of Mligge,f who has pointed out a distinct relation between the planes of structure and those of twinning, there are already strong reasons to suppose that in the rocks we are considering the twinning of the augites and felspars is nothing but the effect of strain. As further results from observations, such a tractive power must * Neues Jahrbuch fur Mineralogie, B.B. iv. p. 590. See also Quart. Jour. Geol. Soc., xl. p. 343, where Rutley describes analogous phenomena in some obsidians. t A^eues Jalirhuch fur Mineralogie, 1883, i. p. 54. 170 Proceedings of Boy al Society of Edinburgh, [feb. 20, have been greater the nearer we approach to the outside of the volcanic mass or to the junction with a large enveloped fragment. Even in the contact with such a fragment we have been able to show that, proceeding from the enclosed fragment towards the central portion of the volcanic body, a constant diminution of the twinned augite crystals takes place in favour of the untwinned individuals. In explaining the cause of these interesting facts, there is ample ground for assuming that all these analogous phenomena are to be considered as consequent upon sudden cooling. Just as glass rapidly cooled down proves to be brittle, and exhibits anom- alous optic properties, so the volcanic rock, rapidly cooled down near the junction with cold associated rocks, will likewise be subject to a molecular tension. This tension, I believe, manifests itself by producing twinning both in augite and felspar, as well as by conferring anomalous properties on the quartz. I am unable to offer a suitable physical explanation of these phenomena, and can only offer the following suggestion : — The porphyritic crystals scattered through the microporphyritic diabase must have separated a long time before the consolidation of the ground-mass. Therefore the strain manifested in the twinning of the porphyritic crystals must have had its origin in those portions of the magma which, after the consolidation of the whole mass, represent the ground-mass. The latter had alone been directly influenced by the rapid cooling. There first resulted a contraction from the cooling of the ground- mass, and then this contraction extending to the porphyritic crystals, seized on the crystal at some point of its outer planes, with the tendency to draw it asunder. To sum up these results, we may say that the conditions, which we must postulate for the strain, correspond with the effects we have observed in different minerals. The conclusions as to the effect of this tractive power, that we were able to draw from either mineral, represent, therefore, three views. These fairly coinciding one with another, point to one hypothesis, which thus assumes a high degree of probability. The results we have arrived at agree wuth the facts that Prof. Judd* has noticed, in his paper “On the Tertiary and older Peridotites of Scotland.” There, however. Prof. Judd drew * Qua7't. Jour. Geol. Soc., 1885, xli. pp. 354-418. 1 F I t r 1888.] Dr Stecher on Contact- Phenomena. 171 other conclusions. It is true he has also stated that the twin- ning of the plagioclase is nothing but a phenomenon produced after solidification. But he suggested that the crystals had been twinned by a pressure which he interprets as resulting from the weight of the rock masses superposed upon the volcanic sheets. Assuming such a pressure, I should be unable to explain why the above-described phenomena should occur in so striking a manner only round an enveloped fragment. Besides these chief results, I may state some other interesting data. Dr Sorby* has stated, concerning the rock of the Salisbury Crags, that, at the j unction of the diabase and sandstone, the fluid-cavities in the quartz grains of the sedimentary rock have been emptied by contact with the igneous magma. In contradiction to this, I succeeded in finding fluid-cavities containing vibrating bubbles in the quartz grains, both near the immediate junction and in frag- ments totally enveloped by the volcanic mass. Nevertheless, the statements of Dr Sorby may be correct, since the specimen he examined may have been taken from a point of the volcanic mass where the magma altered the associated rocks to a greater degree. The rocks associated with the olivine-diabases are often black Carboniferous shales. Where these shales have come in contact with the volcanic rocks, the carbonaceous substance has been driven back for a minute distance (at the utmost 2V inch). Thus the shale exhibits at the junction a small white border, followed by a zone which is darker from having absorbed a great part of what has been driven back frchn the immediate junction. The olivine-diabase of the Salisbury Crags proves to be rich in microscopic crystals of analcime, which, between crossed nicols, exhibit the well-known anomalous phenomena. I tried the following experiment : — A slice was prepared without application of heat. In this slice the anal- cime shows the same beautiful polarisation of light, as when prepared by the usual method. This result disproves the statements of Eohrbach.t Finally, I will cite the result of an analysis of the “ white trap ” from Newhalls, near Queensferry : — * Address, Quart. Jour., xxx. t Tsclitrmak' s Mineral, u. petrogr. Mitth., 1885, vol. vii. p. 32. 172 Proceedings of Boy al Society of Edinburgh, [fee. 20, SiO^ T1O2 CO2 PA AI2O3 FeO MgO CaO K2O Na20 H26 36*8 per cent. 2-6 >} n-9 0-75 jj )) 22-95 „ 4-08 2-85 9-73 1-1 0*5 7.7 3) 53 53 33 53 33 100*96 per cent. 1 *44 hygrosc. water. According to this result, the “ white trap” may, in its main part, he regarded as consisting of 56 per cent, kaolin, and 26*6 „ carbonates. Ample details of all these investigations are given in my paper mentioned at the beginning of this communication. By permission of the Society, Dr John Murray exhibited some phosphorescent animals brought from the deep water of Loch Long and Loch Goil. 4. Experimental Researches in Mountain Building. By Henry M. Cadell, Esq. of Grange, B.Sc., F.R.S.E., of H.M. Geological Survey of Scotland. Ppoc.Roy. Soc.EciinR Vol. XV, Plate I. laid open^dorndeair'alSu/dcce^ 7/ jdadilctly idiz^s di.M'i/ip rdadae posdiz>7W/Fdt^s^ ffdlr,Jfaride fee J Tu^Ai draiee Tie/ilnn^r-Pae/uxZIiiAo ZeetGiep -7h^U\ Fr&e-ZCar^uiy I 7^Mcm.eie.-7ci>e . dtptAe lijii/f?. /■ft^riPK lower' ' Pip. ip' UpfreenieP ee/ieu' T^ucSiwLxr mrzj'auo'ofi \ / G^reil 7/J. 7^e7b./}i/ier (7el7 ^eisTer^cor- Por.^oeP -Z<^ Ouler Chitri/i Au^ irtner/fxce-. C/^rreue. Fia. W/.-Fp/f. ddis, JAzier Sc PuPsr ' deiceeJuuF. ^ZttPr'U.>r- . J’t^. V7/7 JleFt dild/nnerdOulc^^'^ c/tea/mzf jTTzzer dilLi'<^arad/u/ ppTUr_ae IromyOuiP^. 7ree- lorcZep oi J??j7.erLam-ellay 'li:er'ijyp J'irUj ezG Tos^Cerua- die/ 7Si.7lu/7ii Fnaer- Fdl ac- divuddc So P'sC (TiPefZzzeriy pPJioect/ioTiZ dnterior C Pun//7-ajn oP7?jip7le Cp7npiG7'llamen& Poseerior Pace^' Tr , /PPreximal ertZe Pock-lOcri. [S JteOeOeZ' 7^ VJzc/lPdds, Finer side op rridlmierFcdicdFadi adadied. i i ecp^n/ru .eri'nyd7 die ramce -dip.X/. -dhoi. Ve/cO ed Viecr. dhee^. d.dirUeredlnizncd remrneci dae/i.slidl ce jfi/sidu/rtrpF FBod^^'^ccps JVccturaZ-dosedrPrey oZJSepc/k^ yiiAcZd^ . dPiUdim oFJjcdddFa/prr. f Jd'^daz'.J IDeo-^- '2. ' . ay -LeFA- ( {inner Ixd/j ^PoS/W'Wf Ji/pAt 7u/M.ded Le/t-liarided.^ a d.efc Inner -rei/ersedc Ij duiaAldnner-ree^erred. du/hr - hetnded ^ leaded hpld- lamdedc. d> duyTiI I rener . Fed - haeidedy ’pdi Proc. Roy. Soc. EdinT, Vol. XV, Plate II Jti^h &,-7l.cr^j3y^- ri£>rrn - Ze/?y - /zcTr^txleZ'. Jiu77it -h.ci-nZeermost. Jiryh/- AcfriZrZ: J>u/e/rum. I 1888.] Mr D. Alpine on Bivalve Molluscs. 173 Monday, Wi March 1888. JOHN MUEKAY, Esq., Ph.D., Vice-President, in the Chair. The following Communications were read : — 1. Observations on the Movements of the Entire Detached Animal, and of Detached Ciliated Parts of Bivalve Molluscs, viz., Gills, Mantle-Lobes, Labial Palps, and Foot. By D. M ‘Alpine, P.C.S. Communicated hy W. E. Hoyle. (Plates I, II.) (Abstract.) In the introduction it is pointed out that, although ciliary motion lias been observed and noted in the mussel, independent locomotion communicated to the gill itself lias not yet been studied, and it is pointed out that a sea-mussel when detached is capable of roaming about — can float and can sink. Dr Aug. A. Gould and Dr Lockwood (American Naturalist, voh iv. p. 331) have observed Mytilus edidis to climb the side of a glass jar by the aid of its foot and byssus to the extent of three inches in a single night. Besides the entire animal, the mantle- lobes, gills, labial palps, and foot all exhibit locomotive activity, and all of these parts are endowed with cilia, not merely motile, as in the organ of Bojanus, but cilia which, when they are attached to the mass, cause it to move from point to point. Interest attaching to Investigation. — Before entering into dateils, it may be remarked that there is a threefold interest attaching to an investigation of this sort. There is, first of all, the peculiarity of detached portions of an animal, comparatively high in the scale, retaining to a certain extent independent vitality, moving about and often rotating, as we shall see, in a certain definite manner and direction. Such an appearance is always interesting, whether it be the detached portion of a hydra or of an earthworm, the wriggling tail of a lizard or the amputated leg of a spider, the writhing por- tion of an eel or the palpitating locomotive detached heart of a frog. Then there is a further interest when it is known that this move- ment is due, in whole or in part, to the action of ciliu. Indeed, there 174 Proceedings of Poijal Soeiety of Edinburgh, [march 5, is quite a parallel case in the ciliated epithelium of our own bodies, say of the lining membrane of the nose or of the windpipe, for, not only does this ciliary motion continue when the epithelium is scraped olf and mounted in water, and not only may the motion persist for days, hut detached cells or even groups of cells of the epithelium may swim about freely on their own account. There is this important difference, however, in the sea-mussel, that it is not merely subordinate parts which are capable of independent loco- motion, hut large and conspicuous organs of the body. And lastly, it will he interesting to determine the function of the parts when attached to the body, in so far as that function depends on movement, judging from their behaviour when free, and to see if such movements can throw any light upon their actions when in organic connection with other parts. Origin of the Present Inquiry. — While examining the sea-mussel in the ordinary course for medical students at the Biological Labora- tory, Ormond College, University of Melbourne, the gills came in for their share of study, not only to determine their minute structure, but to observe the beautiful play of cilia. Eecognising the import- ance of seeing ciliary motion as far as possible under natural condi- tions, each student was directed to detach a portion of the gill, and examine it uncovered under a low power of the microscope. It was soon discovered that the rapid rhythmic motion of the countless cilia continually drove the portion under observation from the field of view, and at first I was inclined to attribute such an unlooked- for motion to the flotation of the thin membranous portion of the gill in the liquid necessary for mounting. It soon became evident, however, on introducing the eye-piece micrometer, and noting the rate of motion over a measured distance, that the movement was pretty regular and in the same constant direction. The cilia seemed to propel the mass like so many oars all acting together, and with measured stroke. The naked eye was next called into requisition to observe the movement, in this instance truly making that which should have been first, last. Suffice it to say, that the clue was promptly followed up, and unexpected movements and modes of motion revealed themselves in other parts as well, even in the entire animal. The course of the investigation, in fact, is a striking instance of 1888.] Mr D. M‘ Alpine on Bivalve Molluscs. 175- the progressive development of ideas. At first the motion was supposed to be that mechanically caused by a thin film of substance in an agitated liquid, the agitation being due to the act of mount- ing the specimen. jN'ext the movement was seen to be not tem- porary but persistent, and probably due to ciliary motion, but supposed to be imperceptible to the naked eye, and so compara- tively slow as only to be measured under the microscope. Then the microscopic slide was kept moist, and the moving mass seen with the naked eye to glide perceptibly along, and the climax in this direc- tion was reached, when the entire detached gill was transferred to a moistened plate where there was plenty of room to move, where it could either be completely immersed or simply kept moist, and where it travelled, not only horizontally, but vertically, and even when the plate was turned upside down. During the movement, when held fully in the light, the flicker of the lashing cilia could be plainly seen, and this is also beautifully visible when the gill is looked at in the opened shell. Last in the order of discovery was the movement of the entire animal, but it may appropriately take the first place in the order of description. Previous Observations. — Sharpey, under heading “ Cilia ” in Todd and Bowman’s Cydo'pcBdia of Anat. and Phys., 1835, and in Quain’s Anatomy, gives a resume of what has been done up to that date, and a full list of references. He there points out that in the detached arms of the plumed polyp of Trembley we have independent life and locomotion in the opposite direction to the currents set up by the cilia, and he notes that pieces of gills of the common sea-mussel, detached portions of the gills of the salamander, the external gills of the tadpole, the gills of the larvae of the newt or water-salamander, all when cut off “ moved through the water with the cut extremity forwards in a direction contrary to the currents.” Most of the ex- periments were made with the sea-mussel (Mytilus), in which the ciliary motion and locomotion are much more marked than in the fresh-water mussel and the oyster (Ostrea), with which, however, confirmatory experiments were made. Preliminaries. — It will be necessary to be agreed as to the posi- tion from which the moving are to be viewed, since it is impossible to have them detached and observed in motion in their natural position. If the valves of the shell are separated in the VOL. XV. 1/10/88 M 176 Proceedings of Royal Society of Edinburgh, [march 5, usual way by inserting a knife at the ventral surface, and passing it round the posterior end until the posterior adductor muscle is cut through, then if the two valves are spread out, with their pointed ends directed forward, the right and left valves will lie just reversed from our own right and left. This is the position from which our observation will be made, and fig. 1 shows the four parts so arranged. In describing the rotation of the labial palps, it will be found very convenient to use the terms right-handed and left-handed, as is done in connection with the rotation of the plane of polarisa- tion. So when rotation occurs in the direction of the hands of a watch placed face upwards, i.e., from left to right, it will be called right-handed, and when in the opposite direction left- handed. The mussel, as seen in fig. 1, may be regarded as the watch, only the labial palps represent the hands as they are when half round, so this must be remembered in settling the direction of rotation. The labial palp will accordingly be spoken of as right-handed or left-handed according to its rotation ; and the direction of watch- hands, as seen by the observer, will be taken as the right-handed standard. The attachments and connections of the various parts will be briefly noted, in order to understand how they perform their functions, but minute structure will not be regarded, except in so far as it explains the movements. In most cases fresh mussels were taken, because the object of the inquiry was to find out how certain detached portions behaved under circumstances as near as possible to that under which they naturally existed, and fresh mussels were considered indis|)ensable. I even tried some experiments at the sea- side, taking the mussels direct from the sea, but equally good results were obtained after bringing them home. With regard to the temperature at which the observations were made, it ranged from 13° to 19° C. The specimens were examined principally in August, September, and October, and no attempt was made to increase the speed by artificial heat, as might have been done, since the effect of such variations will be considered under the appropriate heading. The exact temperature is usually given in the tables, which is the temperature of the room. An increase from 19° to 28° C. is known 1888.] Mr D. M‘ Alpine on Bivalve Molluscs. 177 to render the movement of cilia six times as fast, and so increase of temperature will alter the rate of motion considerably. It may be useful to mention, once for all, that for ordinary pur- poses of observation T found nothing better than an upturned plate whereon to lay the object. The rim of the bottom just served to keep the fluid in, and enough was usually obtained from a shell, while the potter’s marks upon it served to indicate the smallest amount of movement. Each dot, line, figure, letter, or other device was a handy guide, and in the plates I used there happened to be a rhomboid figure, exactly 1 inch in length, which just suited my purpose. When plenty of room was “required, as, for instance, to allow the gill free scope in its movements and rotations, the largest dish pro- curable was used, and there, too, the leaves and lines were excellent guides. As I used plates of the same size and pattern, it was easy, when found necessary, to trace the course of any of the parts with pen and ink on an empty plate, — say the track of a rotating gill or palp, or of an advancing foot, and transfer it direct to paper. The gills are very delicate structures and readily break up, but if sea-water is placed in the valve of the shell, so as to keep them afloat while being detached with a fine pair of scissors, they may be transferred to the plate entire. The parts are laid out with their inner surface uppermost, as they lie in fig. 1, and this is the position of movement recorded, unless specially stated otherwise. Section A. — Entire Animal removed from Shell. — A small mussel, when carefully removed from its shell and immersed in a vessel of sea-water, exhibited a rotatory movement with slight for- ward movement, the posterior end sweeping round whilst the anterior extremity formed a sort of movable pivot. The two ends moved round in opposite directions, the anterior being the point of least motion. This motion was continuous, and always in the initial direction. The movement of the foot was quite independent and irregular, in one case the direction being right-handed, in another left. After several trials, some light was thrown upon this change of direction. It is difficult to lay out specimens quite evenly, so that the body-axis will be exactly central and the lateral members equally disposed on each side. I observed that in sped- 178 Proceedings of Boy at Society of Edinhurgh. [march 5, mens rotating left-handed (as in the latter case) the hulk of the body-axis was towards the right side (diag. 1 6), consequently left mantle-lobe and gills were most outspread, so that the balance of power was on the left side, and drove the animal to the right or relatively weaker side, and conversely. It may be presumed that if the animals could be laid out flat they would move straight forward. Duration. — The movement lasted for nearly 21 hours, and I was fortunate in observing it at the very last. When moving very slowly, it is difficult to fix the precise time when it ceases to move or completes a round ; but, by recording the times for slight move- ments, it is possible to fix a time after which it did move a little. Taking the last recorded time after which it moved slightly, the duration of movement was 20 hours 51 minutes, or say 21 hours. The palps and foot were still sensitive although not moving. Direction and Rate. — Another specimen of the same size as the first was laid out, measuring when removed from the shell If long and 1 J inch broad. It was first placed at the bottom of a bottle of sea-water, where it rotated left-handed, the body-axis being towards the right side. The rate of motion was very slow, only half a round being performed in 44 minutes. Throughout this and all the succeeding movements the foot never appeared, only the byssus projecting slightly. It was next transferred to a plate, where it was laid out as before ; it rotated left-handed, doing the first quarter round in 12 minutes and the second in 1 hour. The same was again laid out with the body-axis towards the left, when it rotated right-handed, the first quarter round being performed in 14 minutes and the second in 23. It was finally placed with the body-axis towards the right side ; it now rotated as at first, left-handed. The first quarter round was done in 18 minutes, the second in 28 minutes, and the next two quarters took 3 hours 5 minutes, thus completing the round in 3 hours 51 minutes. It was now 1 inch higher up than, and exactly above, its original position, i.e.y it had moved forward 1 inch in the course of one rotation, without ultimately shifting its position otherwise. The next quarter round took 2 hours 2 minutes, and one might think that its powers were beginning to fail, but, as will be seen immediately, the falling olf was probably due to its getting some- what out of gear. 1888.] Mr D. M' Alpine on Bivalve Molluscs. 179 The same specimen was again carefully spread out, with the body-axis in the centre, to see if rotatory might be converted into translatory movement, but it commenced rotating at once right- handed, and the axis was soon visibly inclined to the left side. Although but 8 minutes elapsed from the last-mentioned quarter round, it was remarkable how the speed was suddenly increased when the parts were properly outspread. The last quarter round took 2 hours, and, starting anew, 8 minutes ; afterwards a quarter round was performed in 5 minutes, and the next in 3, the complete round being done in 23 minutes. Next day a quarter round was observed, the rotation still being right-handed, and it took 2 hours 27 minutes, this being the last recorded. As regards translatory movement, it was only by accident, as it were, that it could be measured for any distance, since rotation might begin at once or be delayed for a little, and I did not use any guides to compel it to move straight forward. In one case inch was covered in 7 hours, and in another \ inch in 3 minutes, and I inch in 6 minutes, showing how variable the forward movement was when considered apart from rotation. Duration. — From the detachment of this specimen up to the last quarter was 37|- hours, and thus it had continued to move longer than the first specimen ; but as it only rested finally after doing about half a quarter round more, and as it moved very slightly after being noted 13 hours afterwards, we can say that it retained the power of motion for at least 50 J hours. The movements of detached parts will now be considered. Labial Palps — General Description. — The labial palps are two pair of triangular bodies of a deep flesh colour, one pair on each side of the mouth, attached dorsally by their broad base, and free at their pointed ends. The outer or anterior margin is more or less convex, while the inner or posterior margin is a little concave or almost straight. The inner and outer pairs will be considered separately. The two apposed faces of the outer and inner palps have a ridge running lengthways down their centre, with close-set transverse stripes proceeding from it towards the outer convex margin, while the other face of each is comparatively smooth. The outer is the stouter of the two, capable of more prolonged exertion, and of expelling larger masses from the body (fig. 2, a, h). 180 Proceedings of Boy al Society of Edinburgh, [march 5, The right and left palps are usually of the same size, hut occasion- ally one is half the size of the other and I have met with a left inner palp only Jth the size of its fellow. Since the palps may sometimes change their contour when de- tached, so that it is not easy to tell which is the outer and inner margin, this may always be determined by noting that, on both inner and outer palps, the outer margin is transversely striped. Inner Palps — Descrijption. — Each inner palp lies inside the inner gill of either side, to which it is attached by its anterior basal corner, which also forms the angle of the mouth. This connection with the gill is very slight, and yet sufficient to allow the palp to come away with the gill when detached (fig. 5), The two inner palps are connected with each other at their base by a thin mem- branous portion forming a lower lip to the mouth, just as the two outer are similarly connected to form an upper lip (fig. 2 c). If a palp is detached as near its base as possible, having the form shown in fig. 2 a, and laid on a plate with the liquid from the shell, then its movements may be easily observed. Nature of Movement. — The movement is one of regular rotation, the palp revolving about one end in a steady manner and in a definite direction. There may be forward or backward or lateral movements combined with this, but when once the palp has fairly become accustomed to its free condition of existence, rotation is its characteristic movement. This rotatory motion is probably due to the fact that the basal (cut) end is destitute of cilia, and so there is a tendency to turn round that spot as on a pivot. The palp, how- ever, can also rotate upon its tip, and we can hardly account for its being made the pivot, on purely mechanical grounds. Muscular contraction also takes place in these palps, assisting the cilia or modifying the direction of motion of the palp. The usual movement of rotation is as follows : — The broad base of the triangular palp forms the pivot round which it turns, while the exceedingly sensitive tip is directed in the opposite direction to that of the motion, just as the rudder is constantly directed in turning a boat in motion. The nature of the movements, when the palp is completely immersed, will be considered later. Direction of Movement. — The right and left inner palps, detached 1888.] Mr p. M‘ Alpine on Bivalve Molluscs. 181 and placed as shown in fig. 1, turn inwards, the left turning to the left, while the right turns to the right (diag. 2). Both, as already remarked, may move forward or backward or to the side, but through it all there is this revolving motion ; and when not roaming about, hut rotating steadily, the base is comparatively stationary. If, however, there are obstacles in the way, such as dirt-particles in the water, or solid bodies of any kind, then the sensitive tip, ever seemingly on the alert, soon backs out and clears away from it, even although it should involve a change of course. Thus I have seen a palp, when placed in a dirty liquid, turn reversely for a short distance, until it had shaken itself clear of adhering rubbish, and then go forward in its regular course as if nothing had happened. If either palp is reversed, then it might be anticipated that the direction of movement would also be reversed. The right reversed, just behaved^like the left already described (diag. 3 h), and the left reversed ought to have behaved like the right as already given, but it did not. The unexpected happened here, for the tip formed the pivot, and merely shifted a very little to the right — about \ inch — in twelve revolutions (diag. 3 a). The tip was curved inward upon the body of the palp, making the tip end truncated like the basal end. This mode of rotation was evidently exceptional, and so another specimen was tried. It rotated on its base, but, contrary to expectation, the rotation was right-handed (diag. 3 a'), and thus the very reverse of the right, inner side uppermost. Twelve revolutions were recorded for comparison, and, with the exception of momentary reversions, there was no change in the direction during this time. About two hours afterwards it was observed rotating in the same direction, completing its round in 9 minutes, but in an hour and a half afterwards it was observed rotating left-handed (diag. 3 a"). The rate was now more rapid, being \ round per minute. Almost every possible mode of rotation was here shown, on tip and base, right-handed and left-handed. This variability of the rotation of the palp, when detached, is a sign of its vita- lity— that it is not a mere rigid body blindly obeying impulsive forces. A right reversed was also tried again, and it too behaved 182 Proceedings of Royal Society of Edinhuo^gli. [march 5, differently. It turned at once to the right and the rotation was thus left-handed (diag. 3 &'), or opposite to its previous direction. Twelve rotations were recorded, and indeed a thirteenth, all keeping steadily in this direction. It was observed twice after- wards like the left, and there was no change of the left-handed direction, but instead of moving quicker than at first, it moved exceedingly slow. Owing to these discrepancies, it became necessary to find out, if possible, which was the primary and normal direction for each palp reversed. Accordingly several were tried, and their first direction noted, with the following result : — Left inner reversed, right-handed (on base) ; and right inner reversed, left-handed. Rate of Movement. — Numerous continuous observations were made over extended periods of time. It generally happened that the rate was slow at first, then gradually quickened, attained its maximum speed, and finally declined. The greatest speed attained was found to be a complete revolution in If minutes. It is hardly possible to take the maximum and minimum speed, and determine the mean, for now and again the palp will stop, and after a short interval resume, so that there is not always continuous movement throughout. But at that stage, when there is a regular constant rotatory movement, without the complication of to-and-fro movements, a fair average rate may be struck for that period, to be called the 'partial average. Two averages will thus sometimes be given — a general average, in- cluding the rotations from the very commencement; and a partial average, only extending over a limited and selected number of rotations. For determining rate of movement, a right and left inner palp were selected from the same mussel, and placed together in the same plate, under the self-same conditions. Although placed under similar conditions, the two did not behave alike, as the results will show. Left. — For fifteen recorded rotations, the slowest was 17 minutes, the quickest minutes, and the average 6 minutes. The first revolution took 11 minutes, and the last (recorded) 17 minutes. A partial average, including from the 4th to the 12th round, when the rate was comparatively regular, gave 3 minutes per round. After the 15th round, the movements became very irregular. 1888.] Mr D. M' Alpine on Bivalve Molluscs. 183 and the palp turned irregularly from right to left, or left to right, without completing a round. All its varied movements were followed, which need not be given, but after about two hours it completed a regular round to the right in 3 minutes. After oscil- lating for a quarter of an hour, it completed a round to the left in 5 minutes, when it was accidentally stopped by a hair. This change of direction and return to the original is rather interesting to follow, since it shows that there are more than mere mechanical arrangements concerned in the movement, but as there is a very striking case of change of direction with the outer palps, the subject will be more fully referred to then. The left reversed performed twelve revolutions, at an average rate of 8 minutes. After the first round, which took 16 minutes, the rate was either 7 or 8 (diag. 3 a). In a second series of observations with another palp, twelve revolutions were performed at an average rate of 6J minutes. The first round took lOJ minutes, and afterwards they varied from to 5 minutes. Right. — For twenty-six recorded revolutions, the slowest was 60 minutes, the quickest If minutes, and the average 8J minutes. It commenced with a revolution in 5 minutes, about the middle (14th) attained to the quickest in If minutes, and ended with the slowest in 60 minutes. A partial average for the more steady rounds, from the 6th to 19 th inclusive, gave minutes per round. The record was closed for the right after completing twenty-six rounds, when it became perfectly still, as if exhausted. It was still sensitive, however, as it quivered on being touched with a pin, and next morning it had shifted its position. The right reversed moved very slowly, although it rotated in the usual manner by making the base the pivot. The first round occupied an hour, but deducting time stuck, it only took 28 minutes, the second round 22, and the third 20 minutes (diag. 3 h). This was abnormally slow, but in a second specimen tried, twelve revolutions gave an average rate of 5^ minutes per round. The first round took 7J minutes, the last 8|- minutes, and the intermediate rounds from 4 to 5|- minutes. It was left rotating at the rate of 7 minutes per round, as shown by the 13th. In the outer palps the movement generally resembles that of 184 Proceedings of Royal Society of Edinburgh, [march 5, the inner palps, but, as a rule, is in the opposite direction. It sometimes changes, especially after a short halt, and is accom- panied by a corresponding change in the movement of the cilia, as described in the gill (Purkinje and Yalentiu, Physiologic). The motion is longer continued and more rapid than in the inner palps. Left. — The left was observed for twenty rounds moving to the right with great regularity. The average was minutes to the round, the slowest being 9|- minutes, and the quickest 6 minutes. It commenced at the rate of 6 minutes per round, and with a steady pace, varying from 6 to 9 minutes, the 20th round was performed in 7|- minutes. The movement still continued when I ceased recording. Right. — The right was observed continuously for 50 rounds, and for given periods of time the rate was pretty constant : the general average was 5 minutes to the round ; the slow^est record was at the commencement, with 25 minutes to the round; and the quickest was 2 minutes. The partial average for the twenty best continuous rounds, from the 13th to 32nd inclusive, was 3 minutes; and the middle round of the whole (25th) was 2 minutes. The palp was going at the rate of 4 minutes to the round when I left off record- ing, and the 51st round took 5| minutes. Both left and right continued to move for some time afterwards, as I observed them for 25 minutes, before leaving them for the niglit, rotating as usual. General Remarks. — The left outer moved a little to the left at first, without turning round. Then it began to turn very, very slowly, but there was no elevation of the tip, only expansion and contraction of it. After an hour and a half had elapsed without the round being completed, I pricked it a little, and it drew itself up. Shortly after it began to elevate its tip and turn back, reach- ing its original position 1 hour 46 minutes from the start. It immediately began to turn to the right or outward^ and in 6 minutes had completed its first round. I note this partly to explain the phenomenally rapid first round, and partly to show that parts may not behave normally for a little after they have been detached, and until they have become adapted to a free, instead of a fixed, condition of existence. 1888.] Mr D. Alpine on Bivalve Molluscs. 185 The right outer palp behaves in a similar manner, and the general results were confirmed by other series of observations. The labial palps immersed in water have the power of moving in the direction of the cut end, and in one case a left labial palp moved 9 inches in 22 minutes ; 36 inches were covered in 82 minutes, one 9-incb section taking only 16 minutes. The labial palp has also the power of turning itself from side to side, and could creep along on one margin. Duration of Independent Movement of Palps when kept moistened ivith Sea-Water. — At the end of seven days the palps of the sea- mussel reacted to stimulation ; those of the fresh-water mussel (Unio) reacted at the end of eight days. They rotated in both directions ; the outer palps rotating outwards, the inner normally inwards. It should be noted that the ridged and transversely striped exterior surface of the inner palp corresponds to the interior surface of the outer palp, the smooth outer surfaces of the one again correspond- ing to the inner surface of the others. Gills — Description. — The gills are attached, on either side to the body-wall, and posteriorly by their pointed ends to the dark brown tentacular margin of the mantle-lobes. There are a pair of gills on each side of the body, the outer next the mantle and the inner next the body, and both are about equal in size. No deficiencies have been met with here, as in the labial palps. Each gill further con- sists of two lamellae, and in the case of the inner gill, as seen from the inner surface, the inner lamella, at the margin next to the body, is a thick unattached border, while the other is fixed. There is a small space between the two lamellae, interrupted by delicate bands stretching obliquely across, and appearing on the surface of the gill as dark brown streaks. Hence each pair of gills in transverse section has the appearance of a W, the two outer free legs repre- senting the outer and inner lamellae of the outer and inner gill, while the two united legs represent the other two lamellae. The inner gills, drawn and experimented with, were detached just along the thickened free border of the inner lamella (fig. 3). The gills will be named as in the following scheme ; — 186 Proceedings of Royal Society of Edinburgh, [march 5, The left and right gills were first experimented upon as a whole, i.e., taking inner and outer of same side together ; next, inner and outer were observed separately ; and lastly, small portions were taken. As both gills, inner and outer, move and rotate in the same direction, it might be thought that they would keep together when once started; but, as the inner ultimately separated itself entirely from the outer, it shows that the former possesses greater power. Proportion by Weight capable of Movement. — Owing to the pre- sence of cilia on the parts, about body by weight was capable of independent movement. For structure of the gills, see K. Holman Peck, Quart. Jour. Micr. Science., 1877. For power of imbibition, see Dr Th. W. Engelmann, Jour, of Anat. and Pkys., vol. iii. 1868-69. The author finds in connection with this that the gill loses three- fourths of its weight on drying in the atmosphere. Left Gills. — The left gills were detached together and placed in liquid, with the inner surface of the inner gill uppermost. They soon began to glide along in the direction of the cut surface. The anterior end moved quietest and more quickly than the other, and covered 1 inch in 10 minutes, while the posterior end had only moved f inch. Thus the anterior end moved more than twice as quickly as the posterior, at first, the result being that the gills turned completely round upon their posterior ends, which moved forwards at the same time about f inch from their original position. The perpendicular gills thus became horizontal, and the quarter of a round was completed in a few minutes less than 2 hours. It is the movement of rotation, rather than that of translation, which is here attended to (fig. 7, diag. 5). Right Gills. — The right gills were also detached together, but in such a way that the one was free to move upon the other. In this instance the posterior ends remained practically at rest, while the anterior ends went round. A quarter round was completed in 1 hour 14 minutes, when the inner gill began to leave the outer (fig. 8). The second quarter was performed in 35 minutes, when the inner had separated itself from the outer (diag. 6). How the two gills separated was rather interesting. The inner 1888.] Mr D. Alpine on Bivalve Molluscs. 187 began to separate at the posterior end, and gradually wrought itself off, until at last there was just a very narrow connection with the anterior end. The anterior end of the outer gill was now dragged round by the inner, so that, when they separated, the inner was perpendicular with its anterior end forward, while the outer was sloping towards it, with its anterior end backwards. The inner was now parallel with its original position, and away from it just the breadth of itself, or | inch. The outer soon became perpendicular too, and knocking against the other, which was now fixed at the edge of the plate, the course of both ended. Such an observation as the above shows that there is sufficient motive power in the inner gill to move away from the outer when placed upon it, also that the inner is the dominant gill (fig. 8). In both the inner and the outer gills the anterior end moves more quickly, the entire gill rotating on its posterior extremity. Portions op Gills. — If a small piece of inner or outer gill is taken and placed in liquid, it usually begins to move at once in the direction of the inner cut surface, presuming that the outer free margin forms the other boundary. It does not move straight forward, but usually to one side, and finally reaches the end of the plate with its cut edge, where it remains. If set adrift again, it moves about as before until it reaches the edge. It is also noted that the motive power in any piece of a given size is greater in the inner than in the outer gill, but that a strip of the margin does not move constantly, but commences to wriggle like a worm. It may bring the two ends together, then become straightened out ; but it is continually changing its position. Nature of Movement. — The movement is composite, consisting of translation and simultaneous rotation. It consists of a forward gliding movement in the direction of the cut surface, necessarily combined with a rotatory one. The rotation of the entire gill is almost always round the posterior end. Direction of Rotation of Portions of the Gills. — The result stated in the most general way is, that the gill in whole or in half turns in the same direction (upon posterior end), that extreme anterior and posterior pieces also turn similarly (upon posterior end), that the upper and lower corresponding pieces sometimes vary in their direction, and that lower pieces, unless at the extreme ends, in- 188 Proceedings of Boyal Society of EdinhurgJi. [march 5, variably turn in one direction (upon anterior end). Such is the normal way in which the gill, or portions of it, turn, but sometimes the entire gill will turn upon its anterior end in succession, and I have met with the anterior half also turning upon its anterior end. In some cases, however, the exact reverse is the condition, the rota- tion taking place on the anterior end. The three currents of the gill Surface. A Alntra-lamellar. drive it according to the directions of the arrows and the intra- ^^■“^Marginal. lamellar currents, by their action or inaction, could turn the scale either in favour of the anterior or posterior end turning round. Minor variations were noted as follows : — Pieces of sea-weed laid upon the surface and carried to the margin were then thrown off and carried posteriorly by the counter current, necessarily created by the forward marginal current. At the anterior end the currents were normal, but at the posterior end the surface current drove particles forward for a short distance instead of towards the margin. Rate of Movement. — The rate of movement \yas determined both in the entire gill and in small portions of it ; also when moving horizontally, vertically, and upside down ; as well as when immersed and merely kept moist. The rate of rotation has already been mentioned in connection with the different gills, but as forward movement in the direction of the cut surface is most characteristic of the gill, that is what we have now to consider. In some cases the gill would move several inches in succession without departing from the horizontal, but usually there was a movement of rotation which interfered with this horizontality. So I adopted the plan of starting the gill level for each inch, and fixing upon a point about the middle of the outer margin, from which to reckon the forward distance travelled. Entire Gill. — As the result of numerous determinations at different times, I have found that 2 minutes to the inch is a good average rate of speed, both for the outer and inner gills, when travelling horizontally. Sometimes the gill is sluggish at first, and especially in large specimens the rate may be slower, but the ordi- nary rate is as above. Thus in the first specimen of the left inner 1888.] Mr D. M'Alpine on Bivalve Molluscs. 189 gill, on the first trial, 26 minutes elapsed from the time of detacli- ment until it had travelled an inch, but this was hardly a fair test of its speed, as the three succeeding periods of 8, and minutes respectively showed. The most favourable vertical rate was 7 minutes to the inch, and when upside down 2 minutes to the inch. A left inner gill placed in sea-water travelled in a horizontal direction at the rate of about 1 inch in 3 J minutes. When merely moistened it travelled 1 inch in 6J minutes, whether it was on the upper or the under surface of the dish. Small pieces of the inner gill progressed at an average rate of about 1 inch in 2J minutes, especially if the tendency to turning on the posterior end was counteracted by placing glass slides on the plate on each side of the portion of the moving gill. This movement lasts at least for 48 hours. The movements of the detached gill will be due to various currents differently directed. On its inner and outer surface the currents are outwards, on the opposed faces of its lamella the currents are inwards, and on the margin of the gill the direction is forward. Explanation of Direction of Movement. — The gill is the centre of various ciliary currents which will drive it in various directions when detached. We are so accustomed to and familiar with one fixed direction of the ciliary current, that w^e are hardly prepared for its diversity in the mussel. In the nose and in the wdndpipe the cilia habitually work outwards, but in the gill of the mussel they work backwards and forwards, upwards and downwards, and in a sloping direction. Three principal currents may be noted, and these are not invariable in their direction. 1. There is the outer surface current from the attached to the free margin, which considered alone would drive the gill forward, and from the very contour of the gill the posterior end w^ould be less rapidly propelled than the anterior end. 2. There is the marginal current towards the anterior end, and this alone would tend to drive round the posterior end and cause rotation on the anterior. But this is to a certain extent counter- balanced by the slower motion of the posterior end, as indicated above, although the main cause lies in the third ciliary current. 3. There is the intra-lamellar current towards the inner or attached 190 Proceedings of Royal Society of Edinburgh, [march 5, margin and the posterior end, which will at least counteract the marginal current, so that the turning round of the posterior end will not take place. When the posterior end turns round, as it occasionally does, we might infer that for some reason or other the intra-lam ellar current was not acting. Thus, combined with the translatory there will he a rotatory movement, and the anterior end will move quickest, making the posterior end relatively the pivot. The current produced moves at the rate of 2 inches per minute. The problem of motion here is not quite so simple as it may appear to some. To account for the currents is comparatively easy. If the gill in its natural position be conceived of as a boat moored to the shore (the body), and the cilia as so many rowers with their oars, then the current is simply due to* the successive sweeping of the oars through the water, while the boat to which they belong is stationary, and when the gill is detached it is simply a case of the boat let loose from- its moorings and free to move. We assume here, and we are justified in the assumption, that the ordinary motile functions of the gill continue after detachment as before. Further, it is easy to account for the two kinds of movement — translatory and rotatory — depending as they do on the mode of distribution, the number and the power of the cilia. If the boat were regularly manned, and the rowers equally dis- tributed and of equal strength, then the movement would be straight forward or translatory, but if unequally distributed, there would be a rotatio-n upon the Aveak side — a turning round of the boat. So far the matter is simple, but when we consider that the cilia, although they perform mechanical work, do not do their work mechanically (and in this respect they resemble oarsmen), that they can move without causing visible motion, and that they change their direction, then the problem becomes rather complicated. How- ever the normal direction of the currents is as enumerated above, and the normal direction of movement ought to be capable of mathe- matical expression. The movement of the entire mantle-lobe is rotatory, as cilia are found on one side only, and the posterior end remaining nearly stationary acts as a pivot, the direction of rotation always being towards the attached margin or inwards. 1888.] Mr D. M‘ Alpine on Bivalve Molluscs. 191 The brown margin of the mantle-lobe and the whitish muscular margin both move about briskly, especially small pieces taken from the latter. The movement continues for several days, and in one instance it continued for a period of eight days, the rate of move- ment being about 2 inches per minute. Expla7iation of Duection of Movement. — The strong marginal backward current will tend to drive the mantle-lobe round at its anterior end, and the surface current from the attached or cut margin will drive it inward, and the result of these combined currents will be to cause the gill to rotate in a right-handed direc- tion. The margin of the mantle is a heavy muscular mass, particu- larly towards its posterior end, forming a constant drag, otherwise there might be considerable forward movement. Hence, I take it, the almost purely rotatory movement in striking contrast to the composite movement of the gill. The Foot is richly ciliated, and there is a slight notch at the free end, usually making the top slightly bifid, from which passes a thin white line (the byssal grove) to the posterior extremity. This groove may be converted into a closed canal by the meeting of the muscular sides. Movement — a, of entire free Portion. — If the free portion of the foot is detached and laid in water sufficient to cover it, a movement will take place in the direction of the tip or away from the cut surface. In all the parts hitherto tried the translatory movement Avas in the direction of the cut surface. The movement is usually in a direct straight line away from the cut surface, but the foot made a complete rotation in 6 hours 47 minutes, and during this rotation the tip would be occasionally raised and swept round, or it might oscillate to and fro. The forward motion is at an average rate of I inch per hour, the highest speed attained being 1 inch in 24 minutes, the lowest 1 inch in 3 hours 4 minutes. This is when moving on the ventral surface, but when moving on the dorsal surface the speed is only about one-third as great. The direction of rotation differs in the different specimens. Sometimes there is no movement at all, the specimen simply curling itself up and refusing to move. h, of Tip. — If a small portion of the tip is snipped off it moves freely to and fro, and is very sensitive to contact of any kind. VOL. XV. 2/10/88 N 192 Proceedings of Eoyal Society of Edinlurgh. [march 5, It moves about, seemingly without any definiteness in its move- ments. To give some idea of its rate of movement, it may be mentioned that in one trial, at the end of 43 minutes, it was 1 inch in a straight line from its starting point, although in its irregular movements it had actually traversed a much greater distance. The duration of movement in four specimens was from one to three days. The rate of movement of the current produced was about 1 inch in 3 minutes. Explanation of Direction of Movement. — The foot detached is constant in its direction, always moving forward in the direction of the tip, never backward in the direction of the base of cut surface. When examined under the microscope the cilia are plainly seen, and the direction of the ciliary current, as might be anticipated from the movement of the mass, is towards the base. At the tip there is a strong inward surface current, and on each side the marginal current passes backwards ; but if the cilia working at the posterior end of the grooved ventral surface are carefully examined, they are found to create a strong current forward. This forward current is too confined, however, to interfere with the general movement, unless perhaps to retard it a little; but its probable use will be afterwards pointed out in dealing with function. The direction of the currents on the attached foot may be very easily demonstrated by means of small pieces of sea-weed placed upon it. A tiny piece of sea-weed placed on the tip (ventral surface) travelled backwards to the attachment of the byssus, a measured distance of 1 inch, in 3 minutes. Afterwards it- crept much more slowly along the projecting portion of the abdomen. The rate, of course, will depend on the bulk of the body carried, for a larger piece of sea-weed placed on the tip travelled \ inch only in about 1 hour 20 minutes, audit was carried towards the margin. Any- thing placed upon the ventral surface of the foot is sooner or later driven backward, and either thrust off at the side or carried towards the base. The reason for the direct straight forward course of the foot is now evident. The surface and marginal currents are towards the base, hence The direction of motion is towards the tip. 1888.] Mr D. M‘Alpine on Bivalve Molluscs. 193 Eeferring generally to the molluscan foot, Professor Lankester* writes — “ The most permanent and distinctive molluscan organ is the foot. It may be compared, and is probably genetically identical, with the muscular ventral surface of the Planarians and with the suckers of Trematoda, hut is more extensively developed than are those corresponding structures;” and it is interesting to find that the movement of the “ genetically identical ” foot when detached is similar to that of the Planarians. Their movement is due to cilia, and the description of it in the land Planarians by Professor Semper f would apply equally to the detached foot of Mytilus. He says — “ They move by means of fine microscopically small hairs the cilia or flagella which are attached to the skin, and which by their peculiar motions can carry the animal forward when it is surrounded by a sufficient quantity of trickling water or of mucilage. On a perfectly dry surface, therefore, they cannot creep about for any length of time ; the rapidly drying skin would soon yield up all the moisture which the cilia on the under side require for their motions.” Darwin J also says — “ Hone of these [land] species have the quick and vivacious movements of the marine species ; they progress by a regular wave-like movement of the foot, like that of a gasteropod, using the anterior extremity, which is raised from the ground, as a feeler.” Leaving out of account the rapidity of movement, the looping of the detached foot of Mytilus, its forward movement, and its capability of turning round, are all strikingly suggestive when compared with a Planarian. Bearings of the observed Movements and the Directions of the Ciliary Currents on the uses of the Parts. — The functions with which we will particularly concern ourselves are those of respiration and nutrition. 1. Gills. — The gills, inner and outer, create currents from the posterior towards the anterior end, along their free margin, as well as from the attached to the outer edge. The result is, not only that a constantly renewed stream of water bathes the gills, and thus serves respiratory purposes, but solid matter is likewise sifted from the water and carried along the free margin. This often accumu- * Ency. Brit, 9th ed., art. “Mollusca.” + Animal Life, p. 186. I Ann. Mag. Nat Hist., vol. xiv., 1844. 194 Proceedings of Royal Society of Edinhurgh. [march 5, lates as a long streak, held together by some viscid elastic sub- stance, and is gradually driven towards the anterior end as new material is added behind. The gills not only serve to convey solid particles inward towards the anterior end, hut likewise outward towards the posterior end. This is done by means of the currents on the opposed faces of the lamellae. Dr Sharpey discovered,* on putting moistened charcoal powder on the surface of the gills, that the finer part of the powder soon disappeared, passing into the interlamellar space. There it was rapidly carried backwards and out at the excretory orifice with the general current. The coarser particles remained on the outside, and were carried towards the free margin, thence forward towards the mouth. My own observation confirms this, as small pieces of sea-weed are driven backward and inward in a sloping direction, as shown by the dotted arrows in fig. 1, and then they are readily passed back- ward and outward along the surface of the body. The meaning of this arrangement is evident. The newly intro- duced water bathing the gills passes through their interstices, and thus the tissues are completely aerated, and the refuse water is passed out, either along the body-surface or the inner surface of the mantle-lohe. But there is not that distinction here of the “inhalent” and “ exhalent ” currents which some maintain, since the pallial chamber is not subdivided into a ventral or branchial, and a dorsal or anal chamber. It may be mentioned here, however, that I have more than once seen, when the valves gaped a little, the free margins of the inner gills close together, and these would enclose a space posteriorly, shut off from the general cavity. That the gills serve a respiratory purpose is pretty certain, but that they also directly subserve nutrition is doubtful, since the finer and lighter particles best adapted for that purpose pass through the lattice-work of the gills and out of the body, while it is only the coarser particles which are driven forward. 2. The function of the laUal palps is purely respiratory. In the inner, the current passing towards the outer margin and away from * Art. “Cilia,” Todd and Bowman’s Cydop(xdia. 1888.] Mr D. Alpine on Bivalve Molluscs. 195 the mouth; in the outer, the currents running towards the inner margin, so that particles thrust off at the anterior end of the outer gill are driven towards the inner margin of the palp, and so con- veyed aioay from the mouth. The highly vascular nature of the palps suggests or indicates that they have a respiratory function. 3. Mantle- Lohes. — There can be little doubt that the inner sur- face serves a respiratory purpose, by reason of the ciliary currents towards the margin. The muscular margin, too, has its own proper work to do, fitted as it is by its superior strength and eminent con- tractility, to expel intruders and rejected products from the body. In Mytilus the direction of the ciliary current is outward on the body of the mantle and posterior along the inner edge of the muscular margin. 4. In the foot in the byssal groove the ciliary current is towards the tip, and since this groove may be converted into a closed canal, it would serve remarkably well to convey the generative products out of the body and beyond reach of its own currents, that is, in front of the animal (see Leuwenhoeck, Select Works^ 1800, vol. i. p. 84). Or it may be that the ventral byssal groove, with its tipward ciliary current, is used for directing the byssal filaments to their appropriate spot before fixing. It may like- wise keep off other sedentary animals from settling down upon the shell, at least round about the anterior end. But it also removes unnecessary material from the body, as will be shown more fully afterwards, and thus the tongue-shaped foot, which might serve as accessory to nutrition, and behave like a tongue to the headless mollusc, co-operates with the palps in removing matters from the neighbourhood of the mouth. From experiments made on the direction of the currents, it is found that small pieces of coloured sea-weed travel along the free margin of the inner gill towards the mouth at an average rate of 2 inches per minute (Bronn’s Thier- reichs, vol. hi. p. 415), but that both pairs of palps arrest these pieces at the mouth. Particles on the margin of inner gill are taken up by the inner palp. It can sweep along the margin with its tip, but it is towards its attached end that particles are picked up, then passed for a short distance backward along its outer margin, and finally thrown off. The same applies to the outer gill, only here it is the outer palp 196 Proceedings of Royal Society of Edinhirgh. [march 5, wliicli takes charge of the particles. The way in which the outer palp does its share of the work is rather ingenious. It is curved round the anterior end of the gills on its own side, its outer margin parallel with theirs, and its outer convex surface sloping away to the mantle-lobe. Any material brought along the free margin of the gill is quickly transferred along the convex outer surface of the palp, thence along the lower (inner) margin towards the tip, and finally handed over to the muscular margin to he conveyed out of the body. The outer palp sometimes appears to carry away material brought by the inner gill, and these particles sooner or later are passed to the muscular margins of the mantle-lobes, whence they are rapidly conveyed outside. The mantle-lobes and the foot also drive out foreign particles from the posterior end of the body by means of their cilia. Huxley, in his Anat. of Invert. Animals^ says: — “ The same agency [^.e., the ciliary currents created by the gills] bring the nutritive matters suspended in the water within reach of the labial palpi, hy tvhich they are guided to the mouth.’’’ Lankester, in his article “ Mollusca,” already referred to, says: — “ The food of the Anodon, as of other Lamellibranchs, consists of microscopic animal and vegetable organisms, which are brought to the mouth by the stream, which sets into the subpallial chamber at the lower sephonal notch. Probably the straining of water from solid particles is effected by the lattice-work of the ctenidia or gill- plates.” And Keferstein* says : — “ Organic particles in the water-stream are driven along the free margin of the gills and forwards between the mouth-lobes, where the position and movements of the latter convey them into the mouth.” But he immediately afterwards adds, “ the mouth, in an unknown manner., appropriates the solid particles.” Claus, in his Zoology (English translation), vol. ii. p. 17, says : — “ Food materials pass with the water to the labial palps and so to the mouth;” and at p. 21— “The mouth leads into a short oesophagus, into which the cilia of the labial palps drive small nutritive particles received into the mouth-cavity with the water.” How, unfortunately for these statements, the cilia of the palps. Bronn’s Thierreichs, vol. iii. p. 415. 1888.] Mr D. Alpine on Bivalve Molluscs. 197 both of the outer and inner, do not drive particles towards the mouth but away from it, and the normal direction of rotation indi- cates the same. The only one I have found to question the asser- tion is Herr J. Thiele,* and it is a significant fact that he has examined the palps of no less than eighteen families of Lamelli- branchs. In dealing with the physiology of the palps, the first point he attempted to settle was the possibility of their having any function in relation to the ingestion of food. He cannot but recognise the outward direction of the currents, and at the same time he cannot shake his mind free from prejudice in favour of the palps, and so he attempts a compromise. He comes to the conclu- sion that the use of the marginal currents appear to be to drive away the water from which the food has been obtained, but he omits to notice the very point at issue, how the food is obtained — how the nutritive particles in the water can travel one way and the water another. According to one set of observers or describers, the gills strain off the solid particles from the water, but here it is the palps which remove the water from the solid particles of food. In this connection, we must not omit to notice the observations of Alder and Hancock,! entirely agreeing with our own, as far as they go. They observed the currents in a Pholas, laid out similarly to the Mytilus, by means of indigo particles. They found an out- ward surface current and a forward marginal current on the gills. The stream of indigo particles, constantly being reinforced, increased in volume towards the anterior extremity of the gill, and they were not loosely driven along but bound together in cords or threads by some tenacious fluid. This stream of particles continued to be formed and to move forward for hours. The rest will be given in their own words : — “Thus considerable quantities of indigo were accumulated in the vicinity of the mouth and oral tentacles. These accumulations were composed of ravelled threads, spun as it were by the branchial apparatus, from the scattered, nearly invisible particles of indigo, in the surrounding medium. This examina- tion of the gill in its living state throws some light upon the sustentation of the Laniellibranchiate molluscs; for it would appear evident that all the minute particles of matter sus- * Zeitschr.f. Wiss. Zool., xliv,, 1886. t A7in. Mag. Nat. Hist., vol. viii., 18.51. 198 Proceedings of Roycd Society of Edinhurgh. [march 5, pended in the water are collected and carried to the month without any apparent selection. The labial tentacles may possibly have the power of rejecting distasteful matters; but it is difficult to conceive how this can be, if the particles, as in the present case, always form a continuous cord, which would have to be severed before any part could be disengaged.’’ The indigo, however, must have entered the mouth somehow, for five or six individuals placed in a small vessel strongly coloured with indigo, removed it visibly, and it was afterwards found cramming the alimentary canal when opened, and was passed out, but little altered in appearance, along with fsecal matters. This uniting of the loose particles into a coherent mass is a convincing argument against the selective function of the palps, and it must not be forgotten that if the animal were in its natural position, these ravelled threads of matter would not collect in heaps about the mouth (as shown in drawings by these authors), but would have dropped or been pushed away as each succeeding mass came forward. According to Keferstein,* feeding is a passive operation, inde- pendent of the will of the animal, which accepts whatever the ciliary stream brings to the mouth, and this is the view of the pre- ceding observers. But I have followed minute particles of sea-weed, even to the very margin of the mouth, and they were not mechani- cally taken in, but systematically removed. Hence the animal can abstain at least from feeding when it wishes, and consequently can open or close its mouth for or against the admission of food. Material lying immediately in front of the mouth was likewise not drawn in. In order to come to some definite conclusion about this food- question, I first of all determined in a general way what the mussel feeds upon. This is easily done by inserting a small pipette through the wide mouth and gullet into the stomach, whence the contents can be drawn for examination under the microscope. The contents are usually of a pale, dirty white or yellowish colour, con- sisting of minute organisms and debris of various kinds. Diatoms are always present in great variety and abundance, and sometimes the colouring matter is seen partially removed. JSTo Desmids were found, although carefully looked for. Any amount * Bronn’s Theirreiclis, vol. iii. p. 417. 1888.] Mr D. M'Alpine on Bivalve Molluscs. 199 of egg-like bodies, large and small, yellow and colourless, and sometimes with the contents wholly or partially removed. In- numerable minute rounded bodies moving slowly about, and giving motion to neighbouring masses, only definitely seen under a high power of the microscope. In some specimens numerous Spirillum-like bodies were also observed. In addition to these there was a miscellaneous collection of various things — stray pieces of filamentous algse, small pieces of limbs of minute crustaceans, empty egg-shells, and inorganic matter. The most noticeable bodies present were Diatoms, yellow egg-like bodies, and ova of various kinds and sizes or spores. After deter- Qiining generally the nature of the food, I examined specially a male and a female,— the sexes being separate in Mytilus. The object was to see if either of the sexual elements occurred in the food. After examining spermatozoa from the deep salmon-coloured mantle of the male, I took some of its food from the stomach, and found spermatozoa there in great abundance. After viewing ova from the pale salmon-coloured mantle of the female, I likewise examined its food, and found occasional ova of its own present there. The presence of the animal’s own sexual elements in the food goes to show that what enters the mouth is not so carefully tested and tried as some would make us believe. A little consideration will show that there are grave difficulties in the way of getting food into the mouth by the direct interven- tion of the palps, although apparently so natural. Firstly, the direction of their currents is against it. Secondly, the particles conveyed by the gills are bound together by viscid material, and this arrangement is, against selection by the palps, rather in favour of rejection. Thirdly, the viscid secretion of the palps themselves likewise binds together particles for convenience of rejection. This viscid matter is also elastic. I have seen particles smeared with this viscid material carried the whole length of the body, along the muscular margin of the mantle, without breaking the thread, so that where it should have left the body it was pulled back again. This elasticity will enable the aggregated matter collected at the anterior end of the gills to form more readily into halls. Such balls of refuse matter are often met with, and sometimes they are shot 200 Proceedings of Royal Society of Edinhnrgk. [march 5, out from among the gills and palps with sufficient force to send them to the muscular margin. Fourthly, the finer particles, which might he supposed to serve as food, are sifted out by the gills, and only the coarser particles are carried forward. Fifthly, the mouth-lips can discriminate suitable or unsuitable food-particles, and so the mouth-lobes or palps are not required for that purpose. The lips may be seen to pass to one side or another matters presented to them, and they can also pass them in as well as out. Sixthly, sensitive tips of palp readily reject, but do not retain solid particles. Seventhly, the palps may control to a certain extent the opening or closing of the lips, and may even allow matter to pass, but there is no evidence to show that they carry supplies to the mouth. It seems to me that in all this speculation about the mode of feeding one very obvious fact has been overlooked, viz., the counter - currents necessarily accompanying the original currents. A glance at fig. 1 will show what a network of currents exists in Mytilus, and they are only partly represented there. There are four inward currents towards the mouth-end, along the free margins of the gills, two on either side of the body, going at a rate of at least 1 inch per minute, and sometimes double that speed. And there are two outward currents, one on each mantle-lobe margin, streaming posteriorly at a rate of 2 inches per minute, and some- times even quicker. One effect of these currents will be to create counter-currents, and we have already noticed the counter-current of the right inner gill carrying particles posteriorly. The counter marginal current of each mantle-lobe will carry particles towards the anterior end. There the converging and diverging currents of the gills and mantle-lobes will create an eddy in the confined triangular space in front of the mouth, and food- particles will be brought within reach of the mouth. In this neighbourhood, too, the constantly renew^ed water will be as constantly bringing fresh supplies, and a portion of the micro- scopic food must inevitably enter such a wide opening as the mouth, which, moreover, has its own ciliary current. By the mere act of opening or closing its mouth, it can feed or abstain from feeding. 1888.] Mr D. M'Alpine on Bivalve Molluscs. 201 In the one case, the cilia send matters to one side, in the other they send food down the gullet ; and it may be that the relaxation or stretching of the labial palps has to do with the opening or closing of the mouth. It may be objected to this view, that there is no selecting of suitable food before the mouth is reached, but after passing through the efficient strainer of the tentacular margin of the mantle-lobe, only relatively small particles will be carried forward, and that is the main selection exercised, the lips doing the rest before allowing it to enter. The indigo-particles were swallowed, although they were not food, but they were afterwards ejected along with the faecal matter. Indeed, the mussel is not particularly fastidious, nor are its wants difficult to supply, as it even makes use of its own generative products, as does the oyster. The main purpose of the labial palps will be to clear away intro- duced matter that might otherwise clog up the body — one palp to intercept the strained matter of each gill ; and in a sedentary animal, with streams of tidal sea-water periodically flowing in, this is quite as necessary an operation as the aeration of the tissues or the actual supply of food ; in fact, it renders these operations easy of accom- plishment and certain in effect. If asked. How does food get into the mouth of Mytilus I would reply that the counter-currents of the muscular margin of the mantle-lobe carry forward food-particles in their stream, and the ingoing in conjunction with the outgoing currents create such an eddy at the anterior end as to bring within reach of the mouth abundance of food. Once there, the lips can reject whatever is unsuitable, if not presented in too large quantity, or pass along whatever is suitable. Whether this view be accepted or not, I trast that its being brought forward may lead to more attention being paid to the food and mode of feeding of Lamellibranchs. The ciliary action is sufficient to move the mantle-lobe weighing about 5 grains in a horizontal direction at the rate of one mile in 88 days, or in a vertical direction one-fifth of a mile in the same time. Comparing these figures with those obtained by Wyman and Bowditch {Bost. Med. and Stirg. Jour.^ August 10, 1876), it may be calculated that the force of the ciliary motion in the sea-mussel is to that of the frog’s palate, &c., is as 18 to 5, or nearly four times as great. 202 Proceedings of Royal Society of Edinburgh, [march 5, "Valentin (Purkinje and Valentin, Physiologic) arrived at the same results. The observations and conclusions of this paper, referring solely to Mytilus ed.ulis, may he summarised as follows : — 1st. Entire animal removed from the shell moves ; the movement is rotatory; it rotates at an average rate of 15 minutes per round, and it retained power of movement for 21 hours in one case, and 50 J hours in another. 2nd. Detached mantle-lobes, gills, labial palps, and foot, either entire or in parts, also move. 3rd. Mantle-lobes rotate in the direction of attached margin, posterior end acting as pivot ; the rate of rotation is 4J hours per round ; the rate of currents on muscular margin is fully 2 inches per minute. 4th. Gills have a composite movement, consisting of translation and simultaneous rotation ; they rotate in the direction of detached margin, usually on posterior end as a pivot; they travel in the direction of the cut surface, either horizontally, vertically, or upside down ; the horizontal rate of movement is 2 minutes per inch, or 44 days per mile ; the vertical rate is 7 minutes per inch ; and the upside down rate is 2J minutes per inch ; rate of current on free margin is J minute per inch or 2 inches per minute ; similar to that on muscular margin of mantle-lobe. 5th. Labial palps have various movements, but principally rotatory; the inner rotate inward normally, and the outer rotate out- ward ; the inner and outer of same side rotate in opposite directions ; the rate of rotation of inner is 2J minutes per round, and of outer 3 minutes; rotation may occur either on base or tip; graceful gliding translatory movement on surface or margin at the rate of 1 inch in 2J minutes; also turning over from one surface to the other in succession; duration of movement for 7 days. 6th. Foot has both a translatory and rotatory movement, usually the former ; rate of rotation, nearly 7 hours per round ; average rate of translation, 1 inch per hour, or 7 years per mile ; duration of movement for 3 days at least ; direction of translatory movement always away from the cut surface. 7th. Labial palps act as guards and not guides to the mouth, since the direction of their ciliary currents is always away from the mouth on the margin continuous with the lips. 1888.] Mr D. Alpine on Bivalve Molluscs. 203 8th. Food consists principally of diatoms of various kinds, ova and spores (including its own reproductive elements). 9th. Particles strained off by the gills probably do not enter the mouth, but ultimately leave the body. The counter currents of the muscular margin of mantle-lobe carry particles within reach of the mouth, and the converging and diverging currents in the confined space at the anterior end assist in this action. 10th. Motive power exerted by the gills when detached is equal to lifting its own weight 1 inch in 7 minutes or y inch in 1 minute. The ciliated cells of the frog’s epithelium are calculated similarly to do J- inch. But in the case of the gill there is the clinging force to overcome in addition to the force of gravity, and approximately estimating this force and allowing for it, the gill raises its own weight at least f inch in 1 minute, or does between three and four times more work than the frog’s ciliated epithelium. The movements are not entirely due to the action of cilia, muscular contraction playing a most important part in altering the shape and dimensions of the part, and in giving it outlines which enable it to get rid of obstacles, or to make a more judicious use of its motive power. The ciliary and other activity of all these parts is stimulated by direct mechanical irritation, ample proof of this being obtained by numerous experiments. It appears, too, that there is just as much reason to recognise volition in the detached parts as in the ciliated infusoria, from the fact that the direction of the moving pieces of gill is so frequently changed as they pass from point to point on a moistened plate. In the common sea-mussel there is a latent power of independent movement in the entire animal as well as in the detached parts which has hitherto escaped notice. [Table Table slwiuing Nature, Direction, Rate, and Duration of Movement in the 4 Parts. 204 Proceedings of Royal Society of Edinhiryh. [march 5, Duration. 7 days 48 hours at least 3 days at least 50J hours at least © ra cn '2 ^ .3 rd ^ 'T3 2 a p S S o ffl !> ^ O Cd Ph 2 HlusHN-ln IX> oo lO a> a a o3 c3 ra to ra &J0 P2 0^ &c a c3 ra -U C3 ^ CO CP ^ P-l CO ^ &^.a ^ CO rt CO © O H gj '=Sf « a c5 a ra A r^ .r2 b2o o a -1-0 o3 |h w § ^ ^ ^ t>o B O g a -ij :::::.. M r2 o3 ©a § o 2 aP5 -gH H ■ P^ a c3 !>o ° o a §P5 s a M I I hJ Ph © © II: I © 02 Si la ■ m o bD S a a a a 2 = a S >-H0QOi- off Skate Island, was obtained in 100 fathoms, on 4th November. An adult male, 19 inches across the disk, was caught in Kil- brennan Sound, in 20 fathoms. March. Also Collett found this species to be an inhabitant of deep water on the coast of Norway, viz., in from 80 to 250 fathoms. Raja intermedia (Parn.). A female, with a disk 19 inches wide, was obtained between Sanda Island and Ailsa Craig, 24 fathoms. March 6. Raja circularis. Sandy Ray. An adult male, 14 inches across the disk, and a very young female, 3f inches broad, were obtained in the Sound of Sanda, at a depth of 20 fathoms, on March 10. The former seemed to have been engaged in the work of propagation when caught ; its stomach contained prawns and sand-eels. The young female specimen is extremely similar to the adult male, but is armed on the tail with a median line of spines (beside the lateral ones), which is absent in the male. Another adult male from the Sound of Sanda, 49 fathoms, caught on March 17, had spawned. Contents of stomach, sand-eels. Scyllium canicula. Lesser Spotted Dog-Pish. A very young specimen, 8 inches long, was obtained in the Sound of Sanda, at a depth of 20 fathoms, on March 10. Proc. Roy. Soc. Edin A. GAD US ESMARKII. B. GOBIUS JEFFREYSII .U. Proc.Ko^/.Soc. Edin. Vol.XV. PI. IV. B, A, Peter Smit del etlitk. A.TRIGLOPS MURRAYI. B. COITUS LILLJEBORGII . C. RHOMBUS NORVEGICUS. Mmiem. Bros . imp . 1888.] Dr Gunther on Deep- Water Fishes. 207 Notidanus griseus. Grey Shark. I take this opportunity of reporting the regular occurrence of this shark in British waters. The specimens hitherto known from the British seas were few in number, and taken on the south coast, and therefore this species was considered to be an accidental visitor from the Mediterranean and neighbouring parts of the Atlantic, where it is not uncommon. However, it appears to be well known to the fishermen who frequent the banks between the Orkneys and Shet- land ; and from this locahty Mr William Cowan obtained two adult specimens last summer, one of which is now in the British Museum. Acanthias vulgaris. Spiny Dog. One adult female, 26 inches long, with large ova in the oviduct, and a young specimen, 10|- inches long, were obtained in Kilbrennan Sound, 26 fathoms. December. One immature male in Upper Loch Fyne, 38 fathoms. January. Pristiurus melanostomus. Black-Mouthed Dog. Two adult males in Upper Loch Fyne, 37 fathoms. January. Coitus lilljehorgii. Norway Bullhead. (PI. IV. fig. B.) Coitus lilljehorgii^ Collett, Norg. Fish, p. 25, pi. i, figs. 4-5 ; Llitken, Vid. Medd. Foren., Kjobenk, 1876, p. 376; Lilljeb., . Sver. Fish, i. p. 158. D. 8|11-12. A. (6) 8-9. P. 15-16. V. 3. C. 13-14. Lateral line protected by bony plates ; the back above the lateral line and the upper side of the head with numerous minute ten- tacles j interorbital space deeply concave, narrow, only half as wide as the orbit ; upper part of the snout with two short spines ; parietal region with two subparallel ridges, made uneven by two pairs of projecting tubercles ; the ridges include a quadrangular space, which is half as long again as it is broad. The hind margin of the praeoperculum is armed with four spines, two of which are at the angle, the third is in the middle of the hind margin of the bone, and the fourth at its lower extremity ; these spines are very short and obtuse, with the exception of the upper, which is sharply VOL. XV. 3/10/88 O 208 Proceedings of Boy al Society of Edinhurgh, [march 5, pointed and not longer tlian the eye. The ventral fin nearly reaches the long and prominent anal papilla ; dorsal fins separated from each other by a short interspace. Reddish-olive, crossed hy five broad blackish bands, of which the first crosses the parietal region, the second corresponding to the anterior dorsal fin, the third and fourth to the posterior, and the fifth occupying the end of the tail •, fins marbled with blackish ; lower fins whitish. This species is allied to Coitus huhalis^ but distinguished by the lesser development of the armature of the head ; the spines, which in C. huhalis project as sharp points beyond the skin, being covered in the present species by the skin ; also the parietal area has a different shape. Our specimen has only six anal rays, whilst the Scandinavian authors ascribe eight or nine to this species. On account of this discrepancy, I felt some doubt as to the correctness of my determination, which, however, was confirmed by Dr Liitken, who kindly acceded to my request to compare the Scotch specimen with those in the Copenhagen Museum.* This is a new addition to the fish fauna of Great Britain. The specimen is 2| inches long, and the species generally seems to be much inferior in size to the other British species of the genus. It was caught off Ardrossan, at a depth of between 15 and 30 fathoms, in the month of July. Previously this species had been found on various parts of the coast of Norway and near the Faroe Islands. Coitus scorpius. Large Bullhead. Two immature specimens, between Cloch Lighthouse and Whit- ing Bay, from 15 to 30 fathoms. July. One immature specimen, between French and Kilbrennan Sound, 10 to 14 fathoms. March. Coitus huhalis. Long-Spined Bullhead. A very young specimen; in the Mull of Cantyre, from 60 fathoms. February. An immature specimen ; Sound of Sanda, 20 fathoms. March. * Whilst this paper was passing through the press, I received a second specimen, If inches long, from the Sound of Sanda, 20 fathoms, which has eight anal rays. 1888.] Dr Gunther on Bmjp- Water Fishes. 209 Triglops murropi^ sp. n. (PI. IV. fig. A.) D. 10|19. A. 19. P.17-18. V. 3. C. 17. Head hut little compressed, its length being contained thrice or thrice and one-third in the total length, without caudal. Eye one- third, and in large specimens somewhat less than one-third, of the length of the head. Interorhital space very narrow and but slightly concave. Maxillary not extending to below the middle of the eye ; prseoperculum with four very small and obtuse promi- nences on the hind margin. The head and the back are covered with shagreen-like skin, but below the lateral line the integuments form oblique folds as in Triglops pingelii; also the series of larger tubercles, which in that species runs close to the base of the dorsal fins, is present in our species. Thorax, in front of the ventrals, covered with transverse folds of the skin. Structure and proportions of the fins as in T. pingelii. Anal papilla long. Whitish, clouded with darker, dorsal and caudal fins with black spots. Although this species is closely allied to T. pingelii, it may be readily distinguished, not merely by the less number of fin rays, but also by the different form of the bead, size of the eye, and more compressed tail, which in T. pingelii is singularly depressed. Several specimens, from to 4 inches long, were obtained in the Mull of Cantyre, at a depth of 64 fathoms, in the months of Feb- ruary and March, and 4 miles south-east of the island of Sanda, in 35 fathoms, in the middle of March. Agonus catapJiradus. Poggy ; Lyrie. Five specimens from the Mull of Cantyre, 49 and 64 fathoms. February. Two specimens from Kilbrennan Sound, 10 to 20 fathoms, Miirch. Trigla gurnardus. Gurnard. Two half-grown specimens from Lamlash Bay, in 6 to 8 fathoms. April. Three half-grown specimens, and one 2 inches long, from Kil- brennan Sound, in 26 fathoms. December. 210 Proceedings of Boy al Society of Edinburgh, [march 5, Four young specimens, 3 inclies long, from Kilbrennan Sound, in 24 fathoms. March. Six adult, half-grown and young specimens, between San da Island and Ailsa Craig, in 24 and 30 fathoms. March 6 and 17. The adult specimens are almost ready to spawn, and have the stomachs empty, whilst the immature individuals had freely fed on crustaceans. Gobius minutus, Polewig. One specimen, in 20 fathoms, between Cumbrae and Skelmorlie Buoy, in April. Many specimens, in 26 fathoms, Kilbrennan Sound, in December. One specimen, in 43 fathoms, Cloch Lighthouse, in August. Four specimens, in 45 fathoms. Loch Goil, in March. Four specimens, in 30 to 40 fathoms, between Cumbrae and Wemyss Point, in February. Gobius Jeffrey sii. Jeffreys’ Goby. (PI. III. fig. B.) Three specimens were obtained in April, viz., in Lamlash Bay (6 to 18 fathoms), off Whiting Bay (20 fathoms), and off Cumbrae Light (56 fathoms). Five specimens in August, off the Cloch Lighthouse, in 43 fathoms. Two adult specimens, male and female, obtained in Kilbrennan Sound, at an uncertain depth (10-45 fathoms), on March 22. These last specimens are of special interest, being a pair which were engaged in spawning at the time of capture, and the breeding dress of the male not having been previously observed. The female is nearly 2 inches long, and has the ordinary coloration as shown in the original figure {Ann. and Mag. Nat. Hist.^ 1867, xx. pi. v.). The male, of which I give a figure here, is only 1 J inch long, and has a uniformly coloured body, without any spots. The first dorsal fin is very high, much higher than the body, the five anterior spines being nearly equally prolonged, so as to produce a horizontal upper outline of the fin ; only their points project beyond the membrane. The pectoral, anterior dorsal, ventral, and anal fins are blackish, and the caudal fin uniform white. The second dorsal is very beautifully ornamented, the lower half being of a blackish colour. 1888.] Y)v OuniYier on Deep-Water Fishes. 211 and the upper of a deep black ; a series of round white spots, one corresponding to each interradial space, occupying the boundary line between the black and grey divisions of the fin. Gallionymus lyra. Common Dragonet. Two adult specimens from Ardrossan (10 to 15 fathoms) and from the Firth of Clyde (20 fathoms). April. Several specimens, from 30 to 40 fathoms, south-east of Island of Sanda. Contents of stomach, tubicolous worms. Adult male and female from Kilbrennan Sound (26 fathoms). December. Gallionymus maculatus. Spotted Dragonet. The occurrence of this dragonet within the limits of the British fauna has been made known by me as far back as the year 1867, in Ann. and Mag. Nat. Hist., vol. xx. p. 289, “where also an adult male specimen is figured on pi. v. fig. A. The three specimens then known to me, and placed by me in the British Museum, came from the neighbourhood of the Hebrides, from a depth of from 80 to 90 fathoms. Mr Murray has now rediscovered this beautiful species in Kilbrennan Sound, where it seems to be rather abundant, at a depth of 26 fathoms. The largest male specimen measures 4J inches. Other specimens were obtained in the Sound of Sanda, 24 to 28 fathoms. Ligoaris ligjaris. Sucker. Many specimens, from 49 and 64 fathoms, in the Mull of Cantyre. February and March. Three specimens, from 30 to 40 fathoms, between Cumbrae and Wemyss Point. February. Sticliaeus lampetraeformis. Discovered by Mr Sim off Aberdeen. This species proves to occur also on the West Coast of Scotland, three adult specimens having been found between Cumbrae and Skelmorlie Light, in 20 fathoms, in April, and at Cumbrae Lighthouse, in 60 fathoms, in February. Gentronotus gunellus. Butterfish. One specimen, at the Cloch Lighthouse, in 43 fathoms. August. One specimen, in Kilbrennan Sound, in 20 fathoms. March. 212 Proceedings of Royal Society of Edinhitrgli. [march 5, Lophius xnscatorius. Sea Devil. A specimen, 16 inches long, was obtained between Cumbrae and Wemyss Point, in 30 to 40 fathoms. Pebruary. Crenilabfus melops. Corkwing. One specimen from Lamlasb Bay, in 6 to 18 fathoms. April. Ctenolahriis rupestris. Goldsinny. One specimen was obtained in Lamlasb Bay, in 6 to 18 fathoms, April; and another between Cumbrae and Skelmorlie Buoy, in 20 fathoms, in the same month. Gadus morrhua. Cod. Two young, 9 and 12 inches long, from Kilbrennan Sound, 26 fathoms. December. An adult specimen from the same locality, and captured at the same time, was in very poor condition. One young, 5 inches long, from the same locality, 15 fathoms. March. Gadus esmarkii, Nilss. Norway Pout. (PI. III. fig. A.) This species, as far as I know, has not previously been recorded from British Seas. The specimens in which I recognised it were obtained by means of the trawl in Kilbrennan Sound, at a depth of between 26 and 46 fathoms, together with a host of other small Cadi, especially Gadus minutus^ with which it may be readily confounded. The species does not seem to be unfrequent in that locality; the specimens measured from to 7 inches. The Norway Pout has been recognised as a distinct species for many years on the coasts of Scandinavia, where it occurs locally in deep water during the winter months. Dr Ltitken records its occurrence near the Faroe Islands. The characteristics by which it can be distinguished from its British congeners are — the lower jaw, which projects beyond the upper ; the dentition, the teeth of the outer series in the upper jaw being a little larger than the inner ones ; the length of the snout, which is almost equal to the length of the diameter of the eye ; the large size of the eye, which is a little less than one-third of the length of the head; the slender 1888.] J)i 0\mih.QY on Deep-Water Fishes. 213 barbel, which is about half as long as the eye ; and finally the fin formula, it being D. 15-16 123-25|22-25. A. 27-29 123-25. However, I have to mention that I somewhat hesitated before finally identifying the British specimens with the species described by the Scandinavian ichthyologists. On comparing them with specimens which the British Museum has received through Hr Collett’s kindness from Christianiafiord, I find that they are some- what stouter, which is not wholly accounted for by the fact that they are approaching the spawning season, the ovaries being much ex- panded by a very large number of minute ripening ova. Secondly, the eye of Norwegian specimens is conspicuously larger than in the British specimens ; thus in a British specimen, the head of which measures 40 mill., the eye is 12 mill, long, whilst in a Norwegian specimen, with a head of 42 mill., the eye measures 13 J mill. Finally, I have to mention that the stomach' of these fishes con- tained nothing but a large quantity of mud. Many of them suffered from a singular affection of the eye, nearly the whole eyeball, and also a greater or lesser part of the iris, being covered with cysts containing a cheesy matter. Young specimens of this fish were also found in tolerable abund- ance, off the Island of Sanda (35 fathoms, March), in Lower Loch Fyne (80 fathoms, January), in the Sound of Mull (70 fathoms, Sep- tember), in the Mull of Canty re (65 fathoms, March), in Upper Loch Nevis (50 fathoms, September), in Loch Sunart (45 to 50 fathoms, September), and in Loch Aber (70 to 80 fathoms, September). Gadus minutus. Power Cod. Seems to be generally distributed on the West Coast, and was obtained — In the Sound of Sanda, . in 22 faths.. March 10. Off the Island of Sanda, . 30 March 17. At Ardrossan, 10 to 15 55 in April Off C umbrae Light, . 56 55 55 In the Firth of Clyde, 20 55 55 In the Mull of Cantyre, . 65 55 March 21. In the Sound of Mull, 70 55 September 5. In Lamlash Bay, 6 to 18 55 in April. In Upper Loch Fyne, 37 55 in January. 214 Proceedings of Royal Society of Edinburgh, [march 5, The specimens, obtained on March 10 and 17, were ready to spawn, and had fed on Nyctiplianes, sand-eels, and Aphrodite. Gadus ceglefinus. Haddock. Three yonng specimens, to inches long, were obtained off Ardrossan, in 10 to 15 fathoms, in April, and off Cumbrae, in 90 fathoms, in August. . Three half-grown specimens in Kilbrennan Sound (26 fathoms), in December. One young (4 inches) specimen, between Cumbrae and Wemyss Point (30 to 40 fathoms), in February. Three immature specimens, Sound of Sanda, 22 fathoms, in March, Gadus raerlangus. Whiting, An adult female and several young specimens were caught off the Island of Sanda, in 23 fathoms, on March 17, They had fed chiefly on young Gadus esmarkii; and the ova of the female were far advanced towards maturity, Numerous half-grown and young specimens from Kilbrennan Sound, 26 and 46 fathoms (from December to March). All the other specimens obtained were young, 3 to 6 inches ) * Off Ardrossan, in 10 to 15 faths., in April. Off Cumbrae Light, . 56 5? 55 Between Cumbrae and Skelmorlie Buoy, , 20 JJ 55 In the Sound of Bute, 90 5J in July. In Upper Loch Kevis, 50 55 September 3, Between Cumbrae and Wemyss Point, in 30 to 40 55 February. Merluccius merluccius. Hake. 'wo specimens, 12 to 13 inches long. were obtained in April. at a depth of 10 to 20 fathoms, in the Firth of Clyde. Three others, 8 to 11 inches long, from Kilbrennan Sound, 26 fathoms. December and March. One, 11 inches long, between Cumbrae and Wemyss Point, 30 to 40 fathoms, February. 1888.] Dr Gunther on Deep- Water Fishes. 215 Two, 15 and 22 inches long, between Island of Sanda and Ailsa Craig, 24 fathoms. March 6. Molva molva. Ling. A young specimen, 11 inches long, from 30 to 40 fathoms, between Cumhrae and Wemyss Point. February. Onus cwihrius. Four-Bearded Eockling. Very common and generally distributed, as will be seen from the following list : — 4 spec, off Cumbrae, in 70 faths., in August. 3 ,, Cumbrae Light, 56 ?) April. Many spec, between Cum- brae and Skelmorlie Buoy, 20 April. 4 spec, in the Sound of Bute, .... 90 5? July 1 spec, in Lamlash Bay, . 6 to 18 April. Many spec, in Loch Fyne, 100 5? Nov. 4. 1 spec, in Upper Loch Fyne, 37 55 January. 2 spec, in Kilbrennan Sound, ... 46 2 spec, between Cumbrae and Wemyss Point, . 30 to 40 Dec. & Mar. February. Onus maculatus. Greater Three-Bearded Eockling. Of the two species of three-bearded Eockling which are found on the British coasts, this is the larger species, attaining to a length of 18 inches. It is the species which I have described in the Cata- logue of Fishes as Motella maculata ; whether or not it is identical with Eisso’s Onos maculata I am unable to decide from insufficient materials of the Mediterranean forms, but it is the. Motella vulgaris of Ltitken. I cannot consider the large size of its front teeth to be a sign of age, having compared specimens of Onus tricirratus of the same or even larger size of body, which lack the large teeth altogether. The pectoral fins of the British specimens of Onus maculatus have twenty-two rays, and the ventrals eight. A specimen, inches long, was obtained in Loch Fyne, in 40 216 Proceedings of Royal Society of Edinburgh, [march 5, fathoms ; February. A young one, 4 inches long, in the Mull of Cantyre, in 65 fathoms ; March. Ammodytes lanceolatus. Greater Sand-Eel. Young specimens (4-5 inches long) were obtained in abundance in 22 fathoms, in the Sound of Sanda, in the latter half of March. Hippoglossoides platessoides. Eough Dab. The Eough Dab is the most common of the flat-fishes of the West Coast j at any rate, many more specimens entered the trawl than of any other species of this family. They were from 2 to 9 inches long, and obtained at these depths — 2 spec, in the Firth of Clyde, in 20 faths.. in April. 2 ,, between Cumbrae and Skelmorlie Buoy, 20 April. 1 spec, in Loch Fyne, off Skate Island, . 100 Nov. 4. ] spec, in Lamlash Bay, 6 to 18 April. 4 spec, in Loch Sunart, 45 to 50 )J Sept. 5. 3 ,, in the Sound of Mull, 70 J) Sept. 5. 1 ,, in Upper Loch Nevis, 50 Sept. 3. 2 „ in Loch Duich, 60 J) Aug. 31. 1 „ in Loch Horm, 70 }} Aug. 29. Many adult and young speci- mens (from 2 in. in length), Kilbrennan Sound, 46 and 26 33 December. Many specimens (7 in. long), in Kilbrennan Sound, . 20 33 February. Two females in the Sound of Sanda, . 30 35 March 17. Many adult and young speci- mens caught between Cum- brae and Wemyss Point, 30 and 40 33 February. In some of the specimens caught in February the ovaries showed conspicuous signs of enlargement, whilst the testicles were in a collapsed condition. But those obtained towards the end of the month were ready to spawn, and those caught on March 17 had 1888.] Dr Gtinther on Dee^-Water FUlies. 217 finished spawning. This species continues to feed during the time of propagation, the stomach of many of the fishes being crammed full with large Annelids, Crustaceans, and Ophiurids. Two speci- mens had swallowed large pieces of a green Medusa, which discoloured not only the walls of the stomach, but also the ab- dominal muscles. Rliomhus megastoma. Sail Tluke. An adult female, 18 inches long, was caught in Kilbrennan Sound, 40 fathoms, on March 22. The ova were ripe for shedding. Stomach empty. Rhombus norvegicus. l^orway Top-Knot. (PI. IV. fig. C.) This species has been known to Scandinavian ichthyologists for the last fifty years, as a rare fish on the coasts of Sweden and Norway. Its first discoverer, Fries, considered it to be the Pleuronedes cardina of Cuvier, and in this he was followed by Krbyer, Sundevall, and Nilsson, until I pointed out that the species must be distinct, giving to it the name of Rhombus norvegicus. More recently this species was found by Collett to be more abundant on the northern shores of Norway. With regard to its occurrence on the coasts of Great Britain, Couch is the first not only to have noticed, but also to have correctly determined it. He had received early in the year 1863 a specimen from the Bristol Channel, apparently 5 inches long. The figure which he gives of it belongs to the more accurate of his work, and is perfectly recognisable. I myself became acquainted with the existence of this fish in the British seas in 1868 from a small specimen 2 inches long, which had been dredged in the sea off Shetland, at a depth of about 90 fathoms. The third specimen, now obtained, was dredged in Lamlash Bay, at a depth of from 6 to 18 fathoms; it is inches long, and in excellent condition. A fourth smaller specimen was recently caught in February off Cloch Lighthouse, in 43 fathoms ; and a fifth, 3 J inches long, on March 22, in Kilbrennan Sound, in 45 fathoms. The following is a description of the larger example : — D. 80. A. 66. L. lat. 50. The greatest depth of the body is contained twice and two-thirds 218 Proceedings of Royal Society of Edinburgh, [march 5, in the total length (without caudal), the length of the head three and a half times. The scales are of moderate size, regularly arranged and ciliated on the blind side of the body as well as on the coloured : they cover nearly the whole head, -even the maxillary, leaving only the foremost part of the snout uncovered. The eyes are two-sevenths of the length of the head, and separated from each other by a high sharp, <>^-shaped ridge. Each fin-ray is accompanied by a series of minute rough scales. Lateral line with a sub-semicircular curve above the pectoral fin. Lower jaw prominent; the length of the maxillary is two-fifths of that of the head, the bone extending beyond the front margin of the eye, but not reaching to below its middle. Teeth on the head of the vomer extremely small. Lower eye a little in advance of the upper. The dorsal fin commences in front of the eye, and is rather low ; the hindmost very small rays are inserted on the blind side of the body. Ventral fins separated from the anal. The pectoral fin of the coloured side is rather small, but larger than that of the blind side ; it consists of eight rays, of which the fourth is the longest, extending beyond the bent portion of the lateral line. Brownish, marbled with darker ; a large blotch at the commencement of the straight portion of the lateral line, and a transverse band on the tail behind the dorsal and anal fins, are the most conspicuous markings. The rays of the vertical fins are irregularly annulated with blackish-brown. Rhombus punctatus. Bloch’s Top-Knot. One specimen from Cumbrae Lighthouse, 60 fathoms. February. Phrynorhombus unimacidatus. Top-Knot. One specimen was obtained off Ardrossan, in 10 fathoms, in April. Arnoglossus laterna. Scald Fish. One specimen from Kilbrennan Sound, 20 fathoms. March. Pleuronectes platessa. Plaice. Six adult specimens from Kilbrennan Sound, 26 fathoms, in December. One adult male from the same locality, caught on March 7, was ready to spawn. Stomach empty. 1888.] Dr Gunther on Deep-Water Fishes. 219 Three adult females and three young, from 30 to 40 fathoms, between Cumbrae and Wemyss Point. February. Ova nearly mature ; stomach empty. Many specimens were obtained in the Sound of Sanda, in 24 to 30 fathoms, in the first half of March, others in Kilbrennan Sound, in 20 fathoms. Their food consisted of shells, worms, and Ophiures. Mature specimens were approaching the time for spawning. This species is known to descend to a depth of 200 fathoms, on the Norwegian coast, and is reported from the North-West Atlantic to extend to a greater depth than any other flat fish, viz., to more than 700 fathoms. Also the specimens dredged by Mr Murray come from considerable depths, viz. : — From Loch Horm, . 70 faths., August 29. „ Kilbrennan Sound, 46 to 70 „ Dec. & Mar. 22. „ ,, . 20 ,, March. In specimens obtained between Cumbrae and Wemyss Point, 30 to 40 fathoms, February, the ovaries show considerable enlargement; those caught on March 22 had flnished spawning. The stomach contained pieces of a green Medusa, of Grangon dllmani and Nephrops, and a large number of Nereis; also an Opliiura. Through the kindness of Mr Cowan, I have also received a singular specimen of the Craig Fluke from the Orkneys, 10 inches long, which is very deficient in the pigmentation of the integuments, and before it was immersed in spirits, was so transparent that the Angers of the hand could be seen through its body when it was held against the light. Adult females are ornamented on the right side of the tail with a straight blackish band, which in extent and shape corresponds to the intermuscular cavity containing the ovaries. Pleuronectes Umanda. Dab. Pleuroneetes cynoglossus. Craig Fluke. „ „ Canon, „ „ Fyne, . From Lower Loch Fyne, . 60 ,, September 2. 100 ,, November 4. 80 „ January. Pleuronectes microcephalus. Smear Dab. Two examples were obtained, one from the Firth of Clyde, 10 220 Proceedings of Royal Society of Edinburgh, [march 5, fathoms, in April ; and another from the mouth of Loch Fyne, from 40 to 60 fathoms. Many adult female and immature specimens were caught off the Island of Sanda, in 30 to 35 fathoms, on March 17. The ova were not yet fully developed, and the fishes had fed freely on Solens and Annelids. To extract the former from their shells and holes, the fish must exercise great rapidity and energy of motion. Solea solea. Sole. Two females and one male specimen were caught between Sanda and Pladda Islands, in 26 fathoms, on March 8. They had spawned some time previously to their capture. Solea variegata. Thick-Back. Two immature specimens from the Mull of Canty re, 65 fathoms. March 21. Argentina sphyrcena. This deep-sea fish does not seem to he at all uncommon on the West Coast. By sinking a small-meshed trammel to a depth of 30 and more fathoms, Mr Murray succeeded in getting a number of specimens, which probably would have been still larger but for the attacks on the captured specimens by Crustaceans and Starfishes. The specimens obtained in February had not yet spawned ; there were three obtained in 32 fathoms, between Little Cumbrae and Briguird Point, on February 7 ; and five obtained in 37 fathoms, in Loch Striven, on February 13. 3. Morphological Changes that occur in the Human Blood during Coagulation. By Professor John Berry Haycraft and Mr E. W. Carlier, M.B. {Abstract,') A grant was made by the British Medical Association, on the recommendation of the Scientific Grants Committee of the Associa- tion, towards the expenses of a research, a part of which appears in this communication. Sir Joseph Lister showed that the coagulation of blood is induced 1888.] Messrs Haycraf t & Carlier on Morphological Changes, 221 by solid matter. Dr Freund and Professor Hay craft* have advanced further proof of the correctness of this view. The exact action of a chemically inert solid, such as glass, in producing coagulation remains an undetermined point. This we have resolved to investigate, and have introduced to the Society at a recent meeting a method by means of which experiments may be made with human blood. When a drop of blood is allowed to flow into oil, and another drop, for purposes of comparison, is received on to a glass slide, the former will not clot ; the latter will clot in from five to ten minutes. If, on examination of the two drops, any difference in the blood-cells is visible, it must be due to absence or presence of solid matter. This leads us to consider the action of inert solids on blood-corpuscles. Action of Solid Matter on White Blood-Corpuscles. There are at least two sorts of white blood-corpuscles in circu- lating blood. Both these kinds when in the circulation are rounded in shape, exhibiting no amoeboid movements except in diapedesis. If human blood be received on a slide at a temperature below 65° F., the white corpuscles remain rounded; if the temperature be elevated to about 68° F., they soon exhibit movement ; if a tempera- ture of 74° F. be attained, they become very active. Experiments. Temperature of room and oil, 70° F. A drop of blood was received from a well-greased finger into a tube full of pure castor oil, in which it could be kept free from solid matter; another drop was received on a clean slide, where it coagulated in about ten minutes. This was examined before coagulation had begun, and the white corpuscles were seen to be actively amoeboid, their movement continuing even after the field of the microscope was thickly covered with fibrin threads. At the end of thirty minutes the blood was removed from the oil, placed upon a clean slide, and examined. The white corpuscles were all globular, but after two or three minutes they began to show amoeboid move- ments. * “An Account of some Experiments, which show that Fibrin Ferment is absent from circulating Blood Plasma,” Proc. Roy. Soc., July 1887, and Jour. Anat. and Phys., vol. xxii. 222 Proceedings of Royal Society of Edinlurgh. [march 5, The red corpuscles were in most cases crenated. The white cor- puscles continued to move as long as examined (some thirty minutes) , though fibrin threads had formed long ere that. Some white cor- j)uscles, however, did not exhibit amoeboid movement, but appeared abnormally transparent. This experiment was repeated several times with similar results. Many experiments were also performed at various temperatures, leading us to believe that a temperature of 65° F. was too low to allow of amoeboid movements in the white blood-corpuscles though the blood clotted, showing that metabolic changes were occurring in the blood. At a temperature above 74° F. the changes occurred in the blood so rapidly as to prevent the examination of the fluid before their commencement. We believe that these experiments demonstrate conclusively that glass and other chemically inert solids act as stimuli to the white corpuscles, as indicated by the fact that they exhibit amoeboid movements if the temperature permits. The stimulus is of the nature of a purely mechanical stimulus. As a result of its action, metabolic changes occur in the cells, associated at certain temperatures with changes of form. The white blood-corpuscles, devoid of an envelope, are exposed to the full stimulating effect of mechanical irritation, exhibiting changes in shape if temperature permits. The white corpuscles were also observed to tend to stick to the glass. Do White Blood-Corpuscles tend to h, euh down during Coagidation ? Schmidt * and others maintain that coagulation of blood is the direct result of death of the corpuscles, especially the white ones. We are certain, on the other hand, that some at least of both varieties of white blood-corpuscles are always found alive after coagulation. We have drawings of moving cells in blood which had clotted two days previously. We believe that very few, if any, corpuscles break down during coagulation. If a drop of blood be examined at intervals, noting the position of the white corpuscles, we find that some exhibit amoeboid movements, * Muller’s ArcMv, 1861, pp. 545-587 and 675-721. 1888.] Messrs Haycraft & Carlier on Morphological Changes. 223 and that some become abnormally transparent, as if breaking down ; but they are readily stained by dyes, and these changes occur only about half an hour after coagulation. It has been advanced as a proof that white blood-corpuscles break down on shedding of blood, that the number of corpuscles in defibrinated blood is less than in undefibrinated blood. Some observers have stated that if blood be examined as soon as shed, blood-corpuscles may be seen to break down; this we have failed to observe. The blood was examined in a protecting covering of vaseline, made by smearing a slide and cover glass with a thin layer, and placing a drop of blood from a greased finger between them. Vaseline has been proved to prevent coagulation of blood for a time, so that by this method ample time was given us to examine the blood before coagulation occurred. With this method about fifty specimens were examined, and at intervals the white corpuscles drawn and counted in all cases. The coarsely granular corpuscles were seen to undergo some change, by which their granules accumulated in the centre of the cells, and so their outlines became indistinct, but in no case were any of the cells seen to disappear as long as observed (some thirty minutes), that is long after coagulation of the blood. The finely granular corpuscles behaved in the same manner, with the exception that their granules did not run to the centres of the cells. From these experiments we draw the following general conclusions as regards blood shed from the body ; — If weather be warm, amoeboid movement begins after from one to ten minutes, depending on the temperature. The movement in some cases lasts for hours. In other cases the cells change in from a quarter of an hour to two or three hours, becoming pale, indistinct, granular masses, with their nuclei still visible, and still capable of being easily stained. If the weather be cold, amoeboid movement is not discernible, but the other changes go on as above. Conclusion. — Solid matter mechanically stimulates the white corpuscles of the blood, leading to amoeboid movements if the blood be not cooled. In any case some metabolic change, associated with formation of fibrin, occurs in the white corpuscles, whereby they are led to contribute to the production of fibrin. The stimulus in the VOL. XV. 5/10/88 p 224 Proceedings of Royal Society of Edinburgh, [march 5, case of exceptional cells may be so strong or so continued as to lead to an apparent or real breaking down, which occurs, however, only after, and sometimes long after, coagulation is complete. When removed from the body, of course all the cells eventually die. Inert Solids and their Action on Blood Plates. This subject was suggested by Professor Greenfield, and the work was done by his kind permission in his wards at the Eoyal Infirmary. We had also the advantage of the assistance of his demonstrator, Dr Gibson, who has made the blood plates a special object of study. In all these experiments the blood of patients suffering from chronic diseases was examined, as in these cases blood plates are more numerous than in healthy individuals. The method used was in all cases the one mentioned at the commencement of this communication. In all cases the blood was allowed to remain for about thirty minutes in the oil, after which it was placed in a slide in osmic acid, and examined with yV inch water immersion lens. The blood plates were found floating free in the fluid. They had not changed in any way. In drops of blood mounted fresh from the body, the blood plates ran together and changed their shape. Conclusion. — The action of an inert solid on blood plates is much the same as its action on white blood -corpuscles. It causes them to become sticky and to run together, lose contour, and change their shape. The life history of blood plates has certainly not been made out. They have been described as special and peculiar elements of the blood, but their origin and ultimate destiny have never been explained. They appear to be simply bits of protoplasm like white blood-corpuscles, both from their appearance and their under- going similar changes on irritation. These changes which we have described are the morphological changes which occur in the blood during coagulation. These experiments do not in any way determine the part played by the white corpuscles and so-called blood plates in the chemistry of coagulation. 1888.] Prof. Tait on the Mean Free Path. 225 4. On the Mean Free Path, and the average Number of Collisions per particle per second in a Group of Equal Spheres. By Prof. Tait. There is general agreement as to the value of the quantity which expresses the fraction of the whole group whose members have speeds from v to v + dv; and as to the corresponding Mean Path The “mean,” in this case, is found accord.ing to the definition given by De Morgan: — “ The arithmetical mean, or average, is always to be understood when the word mean is mentioned, unless the contrary be specified.” Thus, using the word in its proper sense, the mean speed is the mean time of describing a free path is '^(n^p^/v); and thus also the Mean Free Path is %{n^p^). A quite different thing is the Mean of the Free Paths described by one particle in T seconds, or by T particles in one second (which, in a perfectly communistic group of S.lO^o per cubic inch, is of course the same). Yet the term Mean Free Path is usually applied to this quantity. The matter is of no consequence whatever in investigations con- nected with the Kinetic Theory of Gases, because in these the distribution of speeds has to be taken account of, and thus (about which all are agreed) alone comes in. The value of this wrongly named mean is easily found. During nJT: a particle has speed from v to v + dv, and its mean path is p^. Let be the number of such paths, i.e. the number of collisions it had after describing such a path. Then, if be the space it described under these conditions, we have = nJY.v = Q^p^ . Taking means (properly so called) we get S/T = :S(SJ/T = ^(?^^?;), the mean speed of a particle during T seconds. This is the same as the mean speed of all, at any instant. C/T = 2(C„)/T = the mean number of collisions per particle per second. 226 Froceedings of Eoyal Society of Edinhurgh. [march 5, Hence the Mean of the Free Paths per particle in T seconds is and is thus seen to be, not a true mean but, the ratio of two proper means. It can be put, no doubt, in the form which seems to agree with De Morgan’s definition. But it must be remembered that C is itself only a mean; and thus that CfG is not strictly analogous to n^. As soon as we admit a departure from the ordinary sense of a term, we pave the way for other departures, some of which may be at least plausible though many may be grotesque. It is for this reason that I said {Trans. K.S.E., 1886, xxxiii. 75) “those who adopt this divergence from the ordinary usage must, I think, face the question: — Why not deviate in a different direction, and define the mean path as the product of the average speed into the average time of describing a free path?” 5. Note on the Compressibility of Glass at different Temperatures. By Prof. Tait. 6. Exhibition of Photographs. The Astronomer-Eoyal for Scotland exhibited Photographs of the Moon during the recent Eclipse by Mr W. Peck. Mr James Durham, Mr William James Bell of Scatwell, the Bev. John Stevenson, minister of Glamis, Mr Henry Brougham Guppy, Professor Thomas Hudson Beare, Mr Andrew H. Turnbull, Mr John Alfred Jones, and Professor John Eerguson were balloted for, and declared duly elected Eellows of the Society. PEIVATE BUSINESS. 1888.] Mr G. Seton on Illegitimacy in Pa7nsh of Marnoch. 227 Monday, March 1888. Sir william THOMSON, President, in the Chair. The following Communications were read : — • 1. Illegitimacy in the Parish of Marnoch. By George Seton, Esq., M.A. Oxon. An examination of the Birth Eegisters pertaining to the parish of Marnoch, in Banffshire, for the ten years ended 1886, exhibits the following startling results : * — Year. Total Births. Illegitimate Births. 1 Paternity not acknowledged. Paternity acknowledged. Paternity found. Occupations of Mothers. Occupations of acknowd. Fathers. Domestic Servants. Farm Servants. Agricultl. Lab’rers. Miscel- laneous. Farm Servants. Miscel- laneous. 1877, . 103 27 26 1 1 19 4 3 1 0 1 1878, . 123 26 19 7 1 18 5 3 0 5 2 1879, . 108 39 34 5 1 31 4 3 1 1 4 1880, . 102 24 18 6 2 19 1 1 3 3 3 1881, . 108 25 17 8 2 22 0 2 1 4 4 1882, . 110 26 24 2 0 21 0 3 2 1 1 1883, . 102 33 24 9 0 26 0 2 5 3 6 1884, . 106 24 18 6 2 21 0 1 2 4 2 1885, . 109 26 23 3 1 18 1 2 5 0 3 1886, . 111 27 22 5 0 22 0 3 2 3 2 Ten Years, 1082 277 225 52 10 217 15 23 22 24 28 Ann. Aver. 108 27-7 22-5 5-2 1 21-7 1-5 2 ’3 2-2 2-4 2-8 The most striking facts displayed in the preceding table are — In the average annual number of 108 births, the illegitimates amount to nearly 28 (27*7), or upwards of one-fourth of the whole; while in the year 1879 these births were considerably above one- third. 2nd. During the ten years in question, the paternity was acknow- ledged at registration in 52 cases, and found by decree of Court in 10 others — together 62 — ^.e., considerably less than one in four. Transcripts included and twins counted two. 228 Proceedings of Royal Society of Edinburgh, [march 19, 2>rd, Of the 277 mothers, no fewer than 217 were domestic servants while of the remaining 60, 23 were agricultural labourers^ 15 farm servants, and 22 dressmakers and other occupations. 4:th. While 24 of the fathers who acknowledged the paternity at registration were farm servants, \ the occupations of the remaining 28 appear from the Eegisters to have been as follows : — Farmers, . . . . 5 Labourers, ..... 5 Innkeepers, . . . . 2 Granite-workers, .... 2 Eoad engine-drivers, . . . 2 Other occupations (1 eacli), . . 12 28 With regard to the 277 mothers, 59 gave birth to 153 children — or considerably more than one half of the whole — being an average of 2*6 births to each. Of these — 2 produced 5 children each. It „ (1 case of twins). ,, (3 cases of twins). surmised, moreover, that some of the mothers, towards the beginning and the end of the decennial period in question, gave birth to other children, which could, of course, be ascertained by a farther search in the Eegisters. The same surname occurs repeatedly, no doubt indicating, in some instances, that sisters, and occasionally mothers and daughters, appear in the same category. Thus, in 5 cases, the same surname occurs twice, and in 2 cases, three times. The same locality also frequently turns up as the place of birth, * This term includes servants in farm kitchens, never employed at out-work, and also those whose chief work is in-door, hut who are sometimes employed at out-work in time of turnip-hoeing and harvest. + There are no bothies in the parish. The male farm servants get their food in the farm kitchen and sleep in an outhouse. It is believed that they wander about a good deal at night, being under little supervision. 6 11 40 59 It may be reasonably 1888.] Mr G. Seton on Illegitimacy in Parish of Marnoch. 229 the village of Aberchirdir furnishing 179 of the 277 illegitimate births. The parish of Marnoch lies inland, along the north bank of the Deveron, on the north-east side of the county, and derives its present name from Saint Marnoch. The extent of the parish is about 15,000 acres, or about six miles by five. At the last census the population amounted to 3220 — 1507 males and 1723 females — being a slight decrease (64) as compared with that of the census of 1871. At the same period the population of the village of Aber- chirdir amounted to 1358, of whom 562 were males and 796 females — showing a preponderance of the latter to the extent of 234. Accordingly, it would appear that while the village of Aberchirdir, with a population of 1358, furnished 179 of the illegitimate births in question, the remainder of the parish, with a population of 1872, furnished only 98. • In other words, while the proportion of these births to the population, in the purely rural portion of the parish, was about 5 per cent. (5 ’2), the proportion in the village of Aberchirdir was about 13 (13-2). The proportion of illegitimate births to the population in the whole of Scotland is only 0*27 (or about one to every 400 persons), from which it would appear that the rural portion of the parish of Marnoch, in the matter of illegitimacy, is about 20 times worse than Scotland generally, and the village of Aberchirdir about 50 times w^orse ! The inhabitants of the former are chiefly farmers and crofters (agriculture and cattle-rearing being the principal industries) ; and of the latter, the followers of the various trades found in any ordinary country village, besides a good many farm servants and agricultural labourers. In both, the households frequently consist of from 7 to 10 members. According to the Neiu Statistical Account of Scotland, published in 1845, “the parishioners of Marnoch are an industrious, quiet, well-behaved people, and possessing a high degree of intelligence. Many of them are much given to reading ; and it may be mentioned that, in the course of two weeks, 60 copies of Dr Dewar’s Body of Divinity'^ were sold in the parish.” Besides two or three unendowed schools, “ the parochial school is taught in the most efficient manner.” At the same date there were regular feeing * The body of Humanity does not appear to have been much influenced by Dr Dewar. 230 Proceedings of Royal Society of Edinburgh, [march 19, markets in the village (Aberchirclir) at Whitsunday and Martinmas ; and an annual market for horses and cattle, called “ Marnoch Fair,” took place on the second Tuesday of March. There were six public-houses, of which five v'ere in the village. At present there are no fewer than six places of worship in the parish, of which five are situated in the village of Aberchirdir, Of these, two pertain to the Church of Scotland, and the four others to the Free, U.P., Baptist, and K.C. Churches respectively. Besides an Episcopal school, there are four public schools and a large paro- chial library ; and the markets appear to be the same as in 1845. A recent writer on the illegitimacy of Banffshire, in the columns of the Scotsmaii (March 3, 1887), speaks of Marnoch as the historical parish, in the Presbytery of Strathbogie, that may almost be said to have given birth to the Free Church j” and suggests that, along with those of five other representative parishes, the registers might divulge some interesting results in connection with the occupations of the parents, which “ could not fail to have a most instructive and beneficial effect.” He also alludes to the unfortunate neglect on the part of so many registrars to make any comments on the figures embraced in their quarterly returns, in the compartment headed “ Eemarks by Eegistrar.” “ Had registrars,” he says, “ carried out their instructions, we should long before this have had a mass of evidence on this question of illegitimacy, that might have enabled measures to be taken for its amelioration. Take, for example, the case of Marnoch. For the long space of twenty-nine years the registrar of this interesting parish has pre- served an unbroken silence on this subject, even when the cases were mounting up to 40 per cent. The experiences of this official must be of a valuable character, if he can only be induced to break the silence.” After alluding to the failure of philanthropists and the local clergy to take any action in the matter, he further says : — “ If it does not fall within the province of a Eoyal Commis- sion, surely the General Assembly would show its wisdom by sending a deputation or commission to investigate and devise remedies for the evil. Better send a deputation to Banffshire than to Beyrout or Bombay.” * * The same intelligent writer has lately contributed a series of valuable papers on “Banffshire Illegitimacy” to the columns of the Banffshire Journal. 1888.] Mr G. Seton on Illegitimacy in Parish of Marnoch. 231 It may, perhaps, seem somewhat invidious to have selected a single parish, in order to illustrate the prevalence of the unfortunate social blemish by which the whole of Scotland is more or less characterised ; but there appears to be ample justification for the present paper, when it is borne in mind that, with the occasional exception of Wigtownshire,* the county of Banff has, during the past thirty years, steadily held the highest place in the illegitimacy tables, and that, while one of its parishes (Seafield) exhibits a percentage of little more than 7, and two others (Cullen and Bath veil) each about 10*5, Marnoch, during the same period, reached the enormous proportion of upwards of 24 per cent. The diagram on the wall (which has been most carefully prepared by Per Cent. Mr Hubert H. Gray of the Kegistrar General’s Department), forcibly exhibits the position of Marnoch {red line), in the matter of illegitimacy during the thirty years ended 1884, as compared with * During the decennial period in question, the parish of Penninghame, in Wigtownshire, with a population, in 1381, of 3777 (1755 males and 2022 females), had 1083 births, of which 146 were illegitimate, thus presenting a very favourable contrast to IVIarnoch. These figures, however, do not embrace t ranscripts, which would probably bring up the illegitimate births in Penning- hanie to about 166, against 277 in Marnoch, with 557 fewer inhabitants, and a larger proportion of houses to the population. (Penninghame — 610 houses to 822 families. Marnoch — 678 houses to 712 families.) 232 Proceedings of Eoyal Society of Edinhurgli. [march 19, Banffshire generally {blue line), Scotland {black line), and the district of Seafield {dotted red line). The registration district of Seafield consists of parts of the parishes of Cullen and Bathven, and is mainly seaboard; while Marnoch, on the other hand (as already stated), is a purely inland parish, 8 or 9 miles from the coast. Possibly some social reformers may lay too much stress upon the vice which leads to the results in question ; but, on the other hand, there seems to be a growing tendency in the public mind to regard these results as normal, if not inevitable, or at least to treat the subject with indifference. Habit, we all know, is a ‘‘second nature”; and if bad habits are not discouraged, they are very apt to become indurated. Places, as well as persons, may reach that unaccountable condition which leads them to be rather proud of pre-eminence, even in bad qualities. It is recorded of an elderly English couple, that they used to boast of being “ acknowledged ” to be the ugliest pair in the kingdom. Let us hope that Marnoch has not arrived at that point of degradation which would induce it to survey its moral condition with pride, or even with complacency. Postscript. — Since writing the above, I have been informed by an intelligent correspondent, tong resident in the parish of Marnoch, that “ the evil has become so inveterate that it is not looked upon as a disgrace. It does not seem to be any bar to a female servant’s getting married or obtaining a situation that she has been the mother of an illegitimate child. In all other respects, they seem to be well-behaved. As a rule, they are honest and trustworthy servants, and when married are as faithful wives as those who have a better previous record. Indeed, misconduct on the part of farm servants, male and female, after marriage, is rare.” This view is confirmed by an anecdote which I lately heard from a northern ex-sheriS-substitute. The wife of a Banffshire minister was, some years ago, applied to by a lady regarding the character of a servant girl. After vouching for her honesty, tidiness, activity, and other good qualities, the denizen of the manse calmly added, in a self-satisfied tone — “ Ay, and she’s had her bit bairnie, too ! ” The contributor to the Scotsman and Banffshire Journal has 1888.] Mr G. Seton on Illegitimacy in Parish of Marnoch. 233 kindly favoured me with the perusal of a thoughtful essay, by a clergyman in the locality, on “ The Immorality prevalent in Banff- shire and the north-eastern counties of Scotland,” in which the writer arrives at the painful conclusion that both religion and education have failed to check the evil in question. Among its causes, he specifies the following : — 1. Drink. 2. Seduction under promise of marriage, rarely fulfilled, 3. Insufficient supervision of servants. 4. Inadequate house accommodation. 5. Concubinage. 6. Tolerance of public opinion. 7. Protection of the male offender, to some extent, by the existing law. In the village of Aberchirder there are, at present, seven un- married couples cohabiting, some of whom have large families ; while between thirty and forty single women live in houses by themselves, some of whom have had as many as five or six illegi- timate children. It need scarcely be added that the circumstances of the latter give every encouragement to what are euphemistically termed “ midnight courtships,” The author of the essay considers that the comparative immunity of the fishing population from the stain of illegitimacy partly arises from the fact of their children being kept longer at home, and consequently being better looked after than those of the agricultural class. As the most serious results of the evil he enumerates — 1. Pauperism. 2. Lunacy. 3. Abnormal infantile mortality. 4. Child murder. 5. Low tone and coarse language, even among females. (This last is, of course, a cause as well as a result.) Religion and education having both failed to diminish the social blemish, he falls back upon legislation as the sole remedy, making the following specific suggestions : — 1st. All cases of seduction should be placed in the hands of the 234 Proceedings of Royal Society of Edinburgh, [march 19, public prosecutor, as ofiences against society as well as against an individual. 2nd. The oath of the mother should always be accepted as proof of guilt, unless an alibi can be established. 3rd. The employment of females as out-door workers, without due supervision, ought to be made penal. 4th. The Local Authority should be entrusted with the duty of seeing that female servants are provided with proper sleeping apartments. 5th. Obscene language and indecent exposure should be more strictly dealt with than at present. The only one of these suggestions to which, I think, any reason- able objection can be offered is the second. All Scottish lawyers are familiar with the difficulties which used to attend the question of semiplena probatio and the oath in supplement. The relative rules of law have been superseded, but not abolished, by the Evidence Act of 1853, which allows parties to a suit to be examined as witnesses; and the woman, being now entitled to give testimony, is not permitted to emit her oath in supplement, on the ground of a semiplena probatio having been established.* It has also been suggested that it should be made compulsory for the mother of an illegitimate child, at the registration of its birth, to report the supposed father, with a view to the insertion of his name in the Eegister. In a good many cases the truth 'would, no doubt, be recorded; but, in some instances, I have reason to fear that false statements would be made, in order to screen the real offender, while in others the mother might, in good faith, report the wrong man. As the Eegistration Act at present stands, the registrar is debarred from inserting the name of the reputed father, unless he attends along with the mother, and they make a joint request to that effect ; and it is highly improbable that the Legisla- ture would consent to remove that important condition. See Fraser’s Laio of Parent and Child, p. 139. 1888.] Dr Woodhead on Mercitric Salts as AntiseiMcs. 235 2. Notes on the Use of Mercuric Salts in Solution as Antiseptic Surgical Lotions. By G. Sims Woodhead, M.D., RRC.P. Edin. Koch, in his article on Disinfection and Antiseptics in the Mlttheilungen aus dem K. Gesundheitsamte, vol. i., 1881, p. 264 (abstracted by Dr Whitelegge in the publications of the New Sydenham Society, vol. for 1886, “Microparasites in Disease”), gives the result of an elaborate series of experiments with mercuric diloride, and comes to the conclusion that a single application of a very dilute solution (1 to 1000, or even 1 to 5000) is sufficient to destroy the most resistant organism in a few minutes. He further states that, “ with longer exposure, it only begins to he unreliable when diluted beyond 1 to 20,000.” In consequence of Koch’s experiments, corrosive sublimate solu- tion was, very rightly, introduced as one of the most valuable, if not the most valuable, of all antiseptic lotions in surgical and obstetrical practice. A most important fact is, however, too frequently ignored — a fact on which Koch laid some stress, but one in connection with which his argument has not been followed, or has certainly not been sufficiently attended to. He points out that, if liquids contain- ing albumen or sulphuretted hydrogen, or other substances forming insoluble compounds with mercury salts, are to be disinfected, enough of the mercuric chloride must be added to leave at least 1 to 5000 in solution, after all precipitation has ceased. In the disinfection of such liquids by means of mercurial salts, it should be remembered that precipitates will form, which, if the process is frequently repeated, may accumulate, and become dangerous on account of the amount of mercury which they contain. The following expresses roughly Koch’s results with bichloride of mercury, used as a germicide : — In solution of I to 1000, it kills anthrax spores in ten minutes, and even those spores found in garden earth which are much more resistant than anthrax spores. 1 to 5000 also kills, if allowed to act for a considerable time. 236 Proceedings of Roy cd Society of Edinhm^gh. [march 19, 1 to 10,000 acting for ten minutes on spores of Anthrax hacilli did not weaken them sufficiently to render them innocuous to mice. 1 to 20,000 in ten minutes kills spores (so that they will not germinate on nutrient media). 1 to 50,000 acting for sixty minutes, has no effect on spores. 1 to 300,000 prevents the growth of Bacillus anthracis. 1 to one million restrains growth of Bacillus anthracis. These results do not seem to coincide throughout, hut it must he remembered that in some cases he used the cultivation medium tests, whilst in others he used inoculation of animals as his test. Klein {Fifteenth Annual Report of the Local Government Board, Supplement, 1886, p. 155 et seq.) refers to many of the points taken up hy Koch, agreeing with him generally as to the efficiency of the corrosive sublimate solution, but maintaining that Koch has over- rated its antiseptic properties. In connection with this paper it may be observed, that Klein points out the necessity for using distilled water, “since the dis- infectant may have entered into combination with proteids, salts, or other substances, whereby its action on given infective material may have been seriously interfered with.” He seems, however, to think that this is specially true in the case of albumen, and he insists that an albuminate of mercury is then formed. These albuminates may be dissolved by adding excess of albumen, but it may be pointed out that the mercury, in such cases, does not necessarily again enter into combination with chlorine to form a chloride. In spite of the great advantages claimed for the use of this salt, it soon becomes evident that the above-mentioned formation of albuminates must greatly detract from its potency as an antiseptic. In October of last year (1887), Dr Laplace of New Orleans con- tributed a paper to the Berlin W ochenschrift on the use of acid solutions of corrosive sublimate. The work on which his paper is founded was carried out in Koch’s Laboratory in Berlin, and Dr Laplace, in his contribution to the subject, opened up many interesting questions in connection with the antiseptic properties of the mercuric salts. Any one who has watched a surgical operation, or who has put 1888.] Dr Woodhead on ATercuric Salts as Antiseptics. 237 specimens away in corrosive sublimate solution, has bad ample evid- ence of the fact, that a quantity of the albuminate of mercury is rapidly formed as a kind of coaguluin. So much is this the case, that a large organ put into a saturated sublimate solution will soon give up so much albumen to combine with the mercury, that there is little or none of the antiseptic material left. This recurs again and again, and the organ cannot be satisfactorily preserved unless the fluid is changed frequently, and for a considerable period. After going carefully into the subject for some little time past, I have come to the conclusion that several of the series of experiments on the antiseptic properties of bichloride of mercury lose a great part of the value claimed for them, for the simple reason that, al- though the above conditions have been known, they have been ignored, and it has not been borne in mind that the salt has not been left in its original condition. Whilst working at this salt, and noting that Koch had used with such good results the mercuric sulphate and nitrate, the other mer- curic salts for which antiseptic properties had been claimed were naturally suggested, and I commenced to carry on a series of parallel experiments with the biniodide of mercury (red iodide of mercury), dissolved in iodide of potassium, and bichloride of mercury. Pro- fessor Crum Brown suggested those 6f the cyanide of mercury for a similar purpose, but the experiments with that salt are yet incom- plete, though I hope to be able to give the results very shortly. The following are, briefly, the results of my experiments with the two salts used along with materials added to keep them in solu- tion : — (In every case ox blood fresh from the abbatoir was used, as the albuminous solution. The bichloride and biniodide were made up to a strength of 1 to 1000.) The modus operandi was as follows : — 50 cc. of the antiseptic fluid was placed in a narrow glass jar; to this was added the salt or acid, and then 5 c.c. of blood. To test for the mercuric salt remaining after the mixture of blood and mercuric salt had been mixed, the mass was filtered, and stannous chloride was added to the filtrate. If any mercuric salt was left in solution, the characteristic reactions were obtained, a white pre- cipitate, gradually turning black, in the case of the bichloride, and yellow, turning black on boiling, with the biniodide of mercury. On adding 5 c.c. blood to 50 c.c. of corrosive sublimate (1 to 238 Proceedings of Royal Society of Edinburgh, [march 19, 1000), nearly every trace of the corrosive sublimate is thrown down with the albumen in the form of the albuminate of mercury. The precipitate is very coarse, and much paler than blood. On adding stannous chloride to the filtrate, there is the merest trace of a darker colour ; so slight is this, that at first it is scarcely perceptible, and there is no definite precipitate in the test tube for several hours. Prom this it is evident that Laplace’s statement that the mercury from 5 c.c. of the sublimate solution (1 to 1000) is precipitated by 5 c.c. of blood, is very much under-estimated, as we see that 5 c.c. of blood is quite sufficient to precipitate the salt in 50 c.c. of the mercuric solution. Along with the serum albu- men the hsemoglobin is also carried down, as the filtrate is almost clear, and contains scarcely sufficient hsemoglobin to give the two characteristic spectroscopic absorption bands. (When a larger quantity of blood, say 50 c.c., is added and the mixture filtered, a considerable amount of hsemoglobin in solution comes through in the filtrate. This hsemoglobin solution, if put into a clean bottle, may be kept for a considerable time, apparently unaltered. Tested with stannous chloride, no trace of the corrosive sublimate can be detected in the filtrate.) If, instead of filtering, the mixture be allowed to stand in the tall glass jar for twenty-four hours, there is found in the lower part a pale brick-red deposit, occupying five- sixths of the whole column, the remaining one-sixth being occupied by a clear supernatant fluid, in which there is not a trace of hsemo- globin, and scarcely a trace of the sublimate. If only 2^ c.c. of blood be added to 50 c.c. of the sublimate solution, a large proportion of the mercury is still precipitated as albuminate, and can be separated by filtrati'' but on the addition of the stannous chloride there is a precipitate sufficient to render the mixture dirty brown in colour. On allowing this to subside, however, the precipitate is certainly not more than one-sixth as bulky as that obtained from a similar quantity of the sublimate solution unmixed with blood. With 4 c.c. of blood there is a very slight precipitate, which becomes dark much more slowly than in the above cases. As Lister and others have pointed out, the albuminate is redis- solved in excess of albumen ; after adding such excess, it was found that mercuric salts of some form or other again passed 1888.] Dr Woodhead on Mercuric Salts as Antiseptics. 239 through the filter paper, and could be detected in the filtrate by means of the stannous chloride. It has for long been known that common salt is a solvent of the albuminate of mercury, and that if it be added to the sublimate solution, the albuminate is dissolved so rapidly that practically it is never formed. That acids acted in a similar manner was also known; and not only these, but that the halogen salts, of which iodide of potassium, bromide of potassium, chloride of sodium, amongst others, may be taken as examples, have a similar solvent action. If to the mercuric chloride and blood solution 2 c.c. of a saturated solution of common salt be added, the precipitate is much finer, and the blood, instead of turning pale, exhibits a slight opacity, and is only slightly lighter in colour. On filtering, haemo- globin comes through with the filtrate; and on addition of the stannous chloride, some of the mercuric chloride is found in solution, though evidently not the whole of it, some of it being thrown down as albuminate in the fine precipitate. On examining the mixture after twenty-four hours, there is found a well-marked very fine precipitate, occupying the lower half of the glass with a clear super- natant fluid, containing haemoglobin and a diminished quantity of corrosive solution. Two c.c. of common salt is therefore not suffi- cient to dissolve the whole of the albuminate formed. If 7 c.c. of the saturated salt solution be added to the sublimate solution on the addition of the blood, there is a slight opacity, which lasts for a very short time. On examination after twenty-four hours, the fluid is quite clear, is unaltered in colour, and there is scarcely a trace of precipitate. The characteristic haemoglobin absorption-bands are well marked. On the addition of stannous chloride to the filtrate, there is rapidly formed a dense black precipitate, equal in bulk to that obtained from the pure sublimate solution. On evaporating and weighing out the salt from the saturated com- mon salt solution, it was calculated that J per cent, to 1 per cent, salt solution should be used instead of distilled water, as the fluid in which to dissolve the corrosive sublimate in making up the 1 to 1000 antiseptic lotion. Seven c.c. of sodium phosphate (saturated solution) added to 50 c.c. sublimate solution, gives no marked precipitate on the addi- VOL. XV. 5/10/88 Q 240 Proceedings of Boycd Society of Edinburgh, [march 19, tion of 5 c.c. of blood. There is at first a slight cloudiness, the result of the formation of a fine precipitate, which may afterwards be seen adhering to the sides of the glass. After twenty-four hours a deposit of a light brick-red colour constitutes about four-fifths of the whole column. The supernatant fluid is clear, and contains no haemoglobin. Merely a trace of corrosive sublimate is to be detected in this clear fluid. The haemoglobin is carried down with the precipitate. Two c.c. of 25 per cent, solution of potassium hydrate added to the sublimate gives a slight yellow coloration. On the addition of 5 c.c. of blood, the fluid becomes turbid and light in colour, but it rapidly clears up, at the same time becoming dark in colour, almost like dark-brown vinegar, the whole or the greater part of the preci- pitate disappearing. Even a few drops of the potassium hydrate is quite sufficient to dissolve the whole of the coagulated albuminate of mercury. After twenty-four hours the haemoglobin is reduced. The characteristic absorption bands are absent ; there is simply a darkening at both the red and violet ends, and a thin dark line in the yellow band. If 1 c.c. of tartaric acid be used, there is at first no precipitate, and very little alteration in colour. Later, a very filmy precipitate makes its appearance, the fluid becomes quite black and slightly viscid, especially near the bottom. The haemoglobin is completely reduced. It is difficult to determine the presence of the mercuric salt in the filtrate, on account of the dark colour of the fluid ; but if the stannous chloride be added, and the fluid be allowed to stand for a few hours, there is a distinct dark-brown precipitate, showing that a large portion of the mercuric salt is left in solution. At the end of three weeks this reaction is not nearly so definite, and it appears that partial oxidation of the mercury has taken place, giving rise to the formation of calomel. If caustic potash or soda be now added in excess to the mix- ture of bichloride, blood, and tartaric acid, there is no precipitate thrown down, whatever quantity of the alkali be used. The fluid becomes slightly viscid, and no haemoglobin can be detected. It is evident from this experiment that the albuminate, once dis- solved in tartaric acid, cannot again be thrown down, even in the presence of a large excess of an alkali. It should here be pointed 1888.] Dr Woodhead on Mercuric Scdts as Antiseptics. 241 out that this holds good only where weak solutions (1 to 1000 of the bichloride in this instance) are used. If 1 c.c. citric acid he used in place of the above, there is first a slight opacity which soon disappears, the fluid becoming slightly darker. After twenty-four hours the precipitate occupies one-fifth of the column, and the supernatant fluid is very dark, almost the colour of porter. The whole of the hfemoglobin is reduced. With 3 drops of hydrochloric acid the fluid is turned muddy for a short time, and then it rapidly turns dark brown (again almost the colour of porter). At the end of twenty-four hours there is a precipitate which occupies two-fifths of the column. There are no haemoglobin absorption bands, but much closing in at the two ends of the spectrum. If only a single drop of hydrochloric acid be used, the fluid is darkened for a few seconds only, it then suddenly clears up almost completely, ilext day there is only a slight precipitate; but the whole of the haemoglobin is reduced. If before blood is mixed with the bichloride solution a couple of drops of liq. ammoniae be added, there is a white band at the point of contact, insoluble in excess of ammonia. A dense precipitate is ultimately formed, the so-called white precipitate or infusible white precipitate. According to Fownes, this is mercurammonium chloride (NHIHg"). Cl, which is insoluble in excess of ammonia, but is soluble in hot water or with great excess of cold water, when, however, it becomes converted into a higher mercuric ammonium chloride. This white precipitate, harmless as a germicide, remains in the fluid, and it is oiilj’- when acted upon by the heat of the body or by an excess of cold water that it can again become active as a mercurial salt, and then, naturally, only in very weak solution, in which form it is quite possible it may retain feeble antiseptic properties. Lauder Brunton states {Pharmacology, 3ded., p. 695) that “ it is used in combination as an ointment to destroy parasitic fungi, but more especially to kill pediculi in the hair on the body. It is also useful in impetigo con- tagiosa, lichen, pityriasis, subacute eczema, and other skin diseases.” It is important to bear in mind this action of ammonia on the bichloride, for it is found that, if we add blood to the mixture of ammonia and bichloride solution, there is no precipitate of albu- minate of mercury, or if it is formed, it is very rapidly dissolved. 242 Proceedings of Royal Society of Edinhurgli, [march 19, It appears, however, as though in this case the whole of the mercury were converted into the white precipitate which is almost insoluble, and is an antiseptic of little potency. The blood is here simply held in suspension or solution in the aseptic, but not antiseptic fluid. (This is a point of some importance, as an aseptic fluid is of compara- tively little value in the treatment of a septic wound.) Haemoglobin absorption bands are very distinctly marked in fluid taken from any part of the glass vessel. If before adding the ammonia to the bichloride solution an excess of tartaric acid be introduced, there is no white precipitate formed, and none will make its appearance, however much ammonia be added. Weak potash or soda solutions may he added with similar results, no precipitate being thrown down even after the fluid becomes strongly alkaline. Similarly the white precipitate is dis- solved by a slight excess of tartaric acid, and the whole of the mercury is again thrown into solution, evidently retaining most of its antiseptic properties. In this respect mercury behaves with tartaric acid somewhat as do some of the metals of the copper and iron group, e.g., chromium, aluminium, iron, copper, zinc, lead, and molybdenum, — all of which, in combination with tartaric acid, are extremely soluble, and once so combined it is a matter of extreme diffi- culty to reprecipitate them (^.e., get them in an insoluble form). In the case of some it is necessary to fuse them before this can be done.* Citric acid acts in the same way, and it is in connection with this peculiar solvent action (amongst others) that chemists ascribe to those vegetable acids the character of an alcohol as well as of an acid. This action upon mercury has been apparently overlooked by most observers, for I can find no mention of it in any of the ordinary text-books on chemistry, and several authorities to whom I have spoken on the subject were unaware of it. Laplace, in the record of his experiments, makes no mention of the fact that tartaric acid will keep the bichloride of mercury in solution for a time, on the addition of this excess of an alkali. In making use of the second mercuric salt, biniodide of mercury, it must first be rendered more soluble. This is done by dissolving I * Since the above was read I have learned that some of these experiments have been repeated, and that Dr Dott has determined that in about a fortnight the tartaric acid converts the corrosive sublimate into the almost inert calomel. This is a point of great practical importance. 1888.] Dr Woodhead on Mercuric Salts as Antiseptics. 243 gramme of the salt with a slight excess of iodide of potassium in 1000 c.c. of distilled water. We have thus formed a 1 to 1000 solution, not of hiniodide of mercury, but of a solution of the double iodide of potassium and mercury, just as above we have a double chloride of ammonium and mercury (?). It is this property of forming double salts with the more basic or positive metallic iodides, as those of the alkali metals and alkaline earths, that renders the biniodide available in a soluble form. A hot solution of the salt with iodide of potassium deposits crystals of potassio-mercuric iodide — 2(KIjHgl2)3H20 (Fownes). These crystals are decomposed by water, and there is then a separation of about half the mercuric iodide, the solution containing the salt 2KI.Hgl2, which remains as a saline mass on evaporation. This combination of iodine, mercury, and potassium, used as above stated, instead of the bichloride of mercury, with the same series of reagents, gives the following results : — On adding 5 c.c. of blood to 50 c.c. of the antiseptic fluid there is not the slightest degree of opacity in the fluid. It remains beautifully clear and unaltered, even as to colour. At the end of ten days there is still no change, and the hmmoglobin bands are well marked in the spectrum. When stannous chloride is added to the filtrate a dense yellow precipitate is formed, which on being heated turns black, and there is as bulky a precipitate as if an equal quantity of the I to 1000 solution of pure biniodide of mercury solution had been used. It is evident, therefore, that in this case the whole of the mercuric salt remains in solution. This is a most important point, as it is at once seen that mercury in the form of an iodide of potassium and mercury does not combine with an albumen, but remaining in solution it (unlike the corrosive sublimate, which combines with the albumen of the blood under these conditions, and loses its antiseptic power) retains its germicidal properties. As we have seen, choride of sodium acts with the bichloride of mercury as does the iodide of potassium with biniodide of mercury. The only difference therefore is really that of the different solubility of the salts in the pure form. The mercury combines more readily with either iodine or chlorine than with albumen, and if these be kept in excess no albuminate of mercury can be formed. If 1 c.c. of a saturated solution of common salt be added to the 244 Proceedings of Royal Soeiety of Rdinburgli, [march 19, biniodide solution before the blood is put in, there is still no pre- cipitate formed, all the reactions are just those above noted, and the same series of reactions are still obtained on the addition of 7 c.c. of the salt solution to the mixture of mercurial salt and blood. If 3 c.c. of a saturated solution of sodium phosphate be added, the reactions are still much the same, and haemoglobin may still be detected in considerable quantities by means of the spectroscope, the red colour of the blood also being retained. 3 c.c. of a 25 per cent, solution of caustic potash added to the biniodide solution gives no precipitate, and when the blood is added the fluid first becomes slightly milky or muddy, but after this there is evidently rapid reduction of the haemoglobin, for the fluid is rapidly cleared up and assumes first a dark red colour, and latterly becomes almost black. The haemoglobin absorption bands have, at the same time, disappeared from the spectrum. If 7 drops of tartaric acid be added to the biniodide and blood solution, the mixture immediately assumes a darker colour, which gradually deepens until within a few minutes it becomes as black as porter. The haemoglobin is completely reduced. At the end of twenty-four hours there is a distinct dirty brown gelatinous deposit at the bottom of the glass jar, occupying about one-fifth of the whole column. When only 1 drop of tartaric acid is added, the darkening is not quite so distinct, but the precipitate is even more marked, both in volume and density. Citric acid used instead of tartaric acid gives much the same reactions, and the precipitate occupies about one-fourth of the whole column. Three drops of hydrochloric acid added to the biniodide solution produces a yellow tinge, and on the addition of blood the changes are much the same as with tartaric and citric acids. The hsemo- globin is reduced, the fluid becomes black, and has a peculiar opacity characteristic of the presence of all of these acids in the mixture of blood and antiseptic. There is a deposit at the bottom of the jar, which occuj^ies about one-fourth of the whole column. The fluid looks like muddy-brown vinegar. If a couple of drops of ammonia be added, the effect is much the same as if a similar quantity of caustic potash had been used ; the 1888.] Dr Woodhead on Mercuric Salts as Antiseptics. 245 liquid becomes darker in colour, but tbere is no precipitate of any kind. From these experiments it is seen that the best results are un- doubtedly obtained with the biniodide of mercury in solution of iodide of potassium alone. There is no coagulation of the albumen of the blood, no reduction of the hsemoglobin ; in fact, there appears to be no change of any kind, and the fluid at the end of six weeks is just as clear and as little changed as it was on the first day. The immediate results are much the same if common salt, phosphate of soda, &c., be used; but, as we shall find afterwards in the case of the latter salt, the permanent results are not so satisfactory. The stronger alkalies and the acids all reduce the haemoglobin, and appear to diminish the germicidal activity of the potassio mercuric iodide. Compare those results with what was observed where these various reagents were used with the corrosive sublimate solution. In that case, common salt and the vegetable and hydrochloric acids Avere all available as solvents of the precipitate of albuminate of mercury formed AAdien the bichloride was mixed with blood. In order to see Avhat further changes would take place in these fluids if left exposed to the air, they were placed in a warm room in which were numerous spores of organisms floating in the atmosphere. Along with them was exposed a glass jar in which 5 c.c. of blood was mixed with 50 c.c. of distilled water. At the end of six weeks they were examined, with the following results : — At the surface of each there Avas, of course, a thin film of dust, in which could be found a feAV spores of fungi, &c., and in some cases a few bacteria or even micrococci, but except in those to be mentioned, there was no growth or proliferation of micro-organisms of any kind. In the mixture of bichloride of mercury and blood, which as we saw was eventually an aseptic but not an antiseptic fluid, there were a few micrococci and bacilli. On the surface Penicillium (jlaucum Avas growing luxuriantly, but there Avas no odour of any kind, even when the fluid was stirred. When sodium phosphate had been added, micro-organisms were very abundant, and they appeared to be proliferating very rapidly. Penicillium glaucum was found growing on the surface. Here again, however, there was no putrefactive odour. Of the biniodide series, that in which the biniodide and iodide 246 Proceedings of Royal Society of Edinburgh, [march 19, of potassium alone had been used was in the best condition. It appeared to be absolutely free from any change, in marked contrast to the bichloride specimen above mentioned. When sodium phosphate had been used, we had the most marked changes in the whole series. Although no penicillium was growing, the fluid appeared to be swarming with micro-organisms, espe- cially near the surface, and the odour coming from the fluid was very pronounced. On microscopic examination, micrococci and Bacterium termo were found in great numbers, and in a state of great activity. Penicillium and micro-organisms were also found in the mixture of biniodide, blood, and hydrochloric acid, and in this instance, too, the smell was somewhat offensive. The control experimental jar with the distilled water and blood was found to be swarming with micro-organisms, and the smell was very strong indeed. A few of the more important deductions to be drawn from the above observations, may be here briefly summarised. All observers refer to the fact that in experiments with antiseptics made on micro-organisms cultivated in fluid media, the results are never so satisfactory as when the organisms are cultivated on solid media, such as gelatine or agar agar. No doubt, the greater vitality of organisms so cultivated may have something to do with these results, but it must also be borne in mind that phosphate of soda is almost invariably one of the ingredients of a cultivating fluid, and we have seen that wherever it is present some micro-organisms flourish in spite of the presence of either bichloride or biniodide of mercury. The conflicting results of various observers are only to be explained by taking such factors into consideration. Klein’s and Law’s observations, though interpreted by them in different ways, may in all probability be brought under the same heading ; and it appears to be possible that Klein’s lower results, as compared with those of Koch, in the use of bichloride of mercury as an antiseptic, may have some such explanation as the following: — The peptones, like other albuminoids, are coagulable by bichloride of mercury ; hence a large proportion of the salt may be rendered completely inactive. It may be pointed out that, in Klein’s experi- ments, a single drop of the fluid in which the micro-organisms had been cultivated was drawn into a pipette, and then 100 drops (or these proportions) of the sublimate solution. It will be evident 1888.] Dr Woodhead on Mercuric Scclts ccs Antiseptics. 247 that, under these conditions, all the albumen in the cultivation medium would be coagulated immediately, and would so remain, for there is an excess of the mercury salt, not of the albumen. In such a case it is quite possible that there is actually a coating or pellicle of albuminate of mercury formed at an early stage around the spores or micro-organisms which, protecting them against the action of the added sublimate solution, is only dissolved when the organisms with their pellicles are again intro- duced into a nutrient fluid in which, of course, there is sufficient albumen to form an excess, and so to dissolve the pellicle and set the organism free to flourish in its new surroundings. If the inoculation be made into gelatine, the solid mass does not yield the excess of albumen so readily, and the slight quantity of mercury remaining in position around the organism prevents its development. In the blood of an animal there are, of course, the same conditions that are present in a fluid medium, as regards the presence of albumen and its action as a solvent of the albuminate. In the case of the biniodide of mercury, we could have no such fallacy creeping in, as no insoluble compound is formed, and the whole of the mercuric salt acts directly on the micro-organism. As the result of the experiments of numerous workers, it may be concluded that all the mercuric salts which can be kept stable, and in solution.^ have powerful antiseptic or germicidal properties, varying (a) according to the quantity of mercury they contain, and (&) according to the acid or halogen with which they combine (a somewhat important factor). If the bichloride be used, it readily unites with albumen. When used as a lotion in surgical cases, this must often prove a serious drawback in cases of profuse h^emorrhage, one part of blood being sufficient to neutralise the antiseptic power of ten parts of the 1 to 1000 sublimate solution, twenty parts of the 1 to 2000 solution, and so on in proportion. We are then using an aseptic, but not an antiseptic solution. In certain cases this is doubtless an advantage, but in many cases it is a source of danger. The biniodide does not unite with the albumen, hence the whole of its antiseptic power is available at the time, and so continues as long as it remains in contact with the tissues. As we have seen, the whole of the mercury in the bichloride solution may be rendered available by adding common salt or 248 Proceedings of Royal Society of Edinhurgli. [march 19, tartaric acid (these are probably the best solvents), the latter only in perfectly fresh solutions ; but, in so rendering the bichloride of mercury soluble, we greatly increase the risks of poisoning by absorption into the system, so that in increasing the antiseptic powers of the lotion we also increase its poisonous properties. In obstetrical practice, and in wounds extending over a large area, such a factor cannot be left out of account. Here the biniodide possesses advantages over the bichloride which cannot be lightly esteemed. It is considerably less poisonous. Cash has pointed o\i\, {Local Government Board Rejport, Supplement^ 1885, p. 186) that guinea pigs are much more tolerant of it than of the bichloride ; and pharmacologists are agreed that the poisonous dose is probably at least double that of the bichloride. The bichloride dissolved in tartaric acid is probably more dangerous, because of the extreme difficulty of rendering it again insoluble. If taken internally, none of the ordinary antidotes for the bichloride could be got to take effect. I venture to suggest the following as reasons why surgeons and obstetricians are gradually coming to see that they have in the biniodide of mercury a safer and more reliable antiseptic than the bichloride : — 1. It is not so poisonous, hence the risks of poisoning by absorp- tion are not so great. 2. It does not form an albuminate, consequently the whole of the salt is available as an antiseptic. 3. It may be used with either acids (especially vegetable acids) or alkalies, neither of which appear to interfere immediately with its antiseptic properties. 4. It is not necessary that the solution should be made with distilled water; all that is necessary is a slight excess of iodide of potassium. 5. The mercury from this solution is not deposited on the surface of the skin or on instruments, or the deposit is exceedingly slight, so slight indeed that it will not injure the most delicate instrument. 6. The exact strength of the solution is always known, as its properties remain constant. As a preservative fluid for pathological and other specimens, it is 1888.] Dr Woodhead o% Mercuric Salts as Antiseptics. 249 much more reliable than the bichloride of mercury. It may be used from the commencement, being changed from time to time as required, either alone or with Muller’s fluid, or it may be used to continue the preservation after the organ or piece of tissue has been placed in perchloride of mercury solution for the purpose of bringing about coagulation of the albuminoid materials. When used for this purpose, great care should be taken to change the fluid frequently, for a few days, if putrefaction has commenced in the slightest degree, whether the bichloride or the biniodide be used, otherwise the sulphuretted hydrogen developed is quite sufficient to render a great part if not the whole of the mercuric salt inert, as it is converted into the sulphide. After the putrefactive process has been checked, it is not necessary to change the fluid. In conclusion, I must thank my friends Drs Gibson, Stockman, and Edington for several valuable hints and corrections given during the time that I have been carrying out my experiments. 3. The Effect of Differential Mass-Motion on the Permeability of a Gas. By Professor Tait. This will appear in the Transactions in Part III. of Prof. Tait’s paper ‘‘On the Foundations of the Kinetic Theory of Gases.” 4. On a New Diffusiometer and other Apparatus for Liquid Diffusion. Part II. By J. J. Coleman, E.R.S.E., E.I.C., E.C.S. I had the honour of communicating to the Royal Society of Edinburgh, upon the 15th July last, a paper describing a new diffusiometer. This instrument wns devised to make visible to the eye the diffusion of acids or alkalies into a supernatant column of water. In ease of acids the column of wnter was made of a yellow colour by minute measured quantities of an alkali and methyl orange, and in case of alkalies the water was made acid and of a red colour by minute but measured quantities of an acid and methyl orange. The coloured water was placed in a tube resem- 250 Proceedings of Boy cd Society of Edinburgh, [march 19, bling a barometer tube, which was, after being inverted, immersed in the reservoir of the substance being diffused. The progress of the diffusion was registered by a sharp change of colour and line of division, as easily read as that of oil standing over water, at a point where the ascending particles became sufficiently numerous to neutralise the acid or alkaline condition, so that the instrument afforded an ocular demonstration of the upward march of an ascending column, which, from observations conducted for thirty or forty days, was found to follow with extreme precision the square root of the time of diffusion. It is obvious that the use of this instrument is limited to acid or alkaline liquids, unless it can be made to work with some other in- dicator than methyl orange. The neutral salts, such as chlorides and sulphates, being desirable subjects for investigation, it occurred to me that soluble salts of silver could be used as indicators for chlorides, and soluble salts of barium as indicators for sulphates, provided jelly be dissolved in the water, so as to entangle the precipitates produced. This expectation was very satisfactorily realised by filling the diffusion tube with jelly containing 5 per cent, of isinglass, and TU(To weight of the indicator, viz., a salt of silver or barium. Under such circumstances, the ascending column converts the transparent jelly into what looks like a rod of ivory surmounted with transparent jelly. Although the progress upward is not so mathematically exact as in the case of the employment of methyl orange, it certainly follows with wonderful precision a speed corresponding with the square root of the time of diffusion. Compared, however, with diffusion into pure water, there is a retardation, owing to the presence of the jelly. Gelatine combines to some extent with salts of silver, but there seems to be less objection to the employment of gelose or Japanese isinglass, 1 per cent, of which is as effectual as 5 per cent, of ordinary gelatine, besides which gelose does not prevent the precipitation of silver salts by chromates, which occurs with gelatine ; and this is a matter of interest, as red insoluble chromate of silver suspended in jelly is converted into white chloride by diffusing into it any soluble chloride. I have annexed a table herewith of experiments which [lasting Diffusion of Hydrochloric Acid, HCl; Sulphuric Acid, H2SO4 ; and various Chlorides into Jelly. 1888.] Mr Coleman on a New Diffusiometer. 251 CO 00 ^ 1--0 rH CO IG CO HI CO HI 05 05 CO CO CO CO ca ca CM ca 42 .-H CO rH 00 CO ca 00 CO CO 05 00 1 CO CO 0 1 l^pO CO CO 05 05 ca CM -rfi Ol 259 259 CO CO 1 CO CO C<1 CN 6a iM 33 lO 0 1 CC l'~ 1^00 1 • • >c >c CO CO 1 • • 01 01 30 (M CO UO -rfi 1 : : CO CO ! ■ ■ 00 ca CO CO - 05 CM i>.oo 1 >0 00 CO ^ 0 C20 0 0 1 Ct 05 10 lO CO ^ CX> Oi 00 CO IG oa (M (N • m mho • ® :g -2 rG • 0 rC) a; ^ . • tn g tfH 0 . c5- 0 0, rr:' (H '“''S rl. ^ 0 > 0 02 > 02 Q 0 ft 0 _L 0 atl i 2 0 2 g (M 0 0 0 ^ *50 . 05 S , 02 &0 02 Cte S 111 ' &a-2 -2 ^ ,g o> g, “ 0 0 5h • O) Ph -l-:> 02 -M -45) G2 -g 0 s > « 0 ling 154 mgs. p ined nitrate sih . to precipitate c.c. NaCl as ab 0 rO G 01 G 0 ft 02 G G 0 bO 02 0 02 a> g 1 1 i' 0 .0 fth-. ^2.0 'S;^2 a 1 . g |o Jh .'o2 cS "S ^ C3 Ol "Jf .04 m • c3 4^ P 2 .§ ^ a 0 fl ^ '0 -M ■ 0 <1^ " G s-< ^ <15 .^ • o « o ^ C '4H^ G -e G2 02 G G G G C 'S3 3 & G X 02 .G ra H -3 G « HOG 3 02 02 02^ & 3 -go «4H ft ® ®!=! 02* G 02 X bo g 0 ca rH <4H 0 0 S) 2 CO 00 '0 6 G 02 ft X bo g C20 05 «4H 0 '|-S &D?S 5C-2.2 S iffusion into 1 tained c.c. to 0.3, G S'34 .2-g.S-S G 3 ^ 5)m 0^ 02 02 G 3 O TO H Gh G 0 1-5 G.2,^0 •|o2 G "h ^ G _o w 5G G _o "m G _o ' X 5G ft ft < ft < ft ft ft ft oa‘ co’ Hj4 in 1 co‘ ihI 00 05’ Note. The lines of figures maikecl Theory are calculated in ratio of square root of time, starting from results of first five days. 252 Proceedings of Boy al Society of Edinlurgh, [march 19, thirty or forty days], demonstrate (1) the retardation which takes place in the ascent of HCl into jelly coloured yellow with methyl orange; (2) which demonstrate that the results by the silver nitrate methods corresponds closely with those obtained by the diffusion of same acid into jelly coloured with methyl orange. The same table also contains results of diffusing HgSO^ into jelly impregnated with barium nitrate, and of diffusing KCl, NaCl, SiCl and MgCl2 into jelly impregnated with silver nitrate. In point of fact, when jelly is used the methyl orange method and the silver nitrate and barium nitrate methods are about on a par, so that the two latter become very useful in making experiments of a comparative nature with neutral salts, either for investigation or lecture illustration. The principal results I have obtained, together with some of those obtained in the paper read July 15, 1887, are given in the diagram of curves herewith, in w^hich the vertical lines represent time, and the horizontal lines height of diffused columns in milli- metres. The observations were taken daily, and the regularity of the curves came out wonderfully in accordance with theory. [Diagrams Height diffused in Centimetres. 1888.] Mr Coleman on a Neio Diffusiometer. 253 Abstract of Results obtained in Present and Former Paper \_\^tli July 1887] arranged in Curves. u. Diffusion of water containing 216 mgs. per c.c. of hydrochloric acid into alkaline methyl orange, requiring mgs. of HCl per c.c. to neutralise. b. Ditto, with 5 per cent, gelatine, or 1 per cent. Japanese isinglass, added. c. Diffusion of water containing 150 mgs. of HCl into jelly which con- tained nitrate of silver, requiring mgs. per c.c. of HCl to pre- cipitate. d. KCl into jelly plus nitrate silver, ratio yi-jj (see Table). e. NaCl „ „ „ /. LiCl „ „ „ (J MgCl „ „ „ 5 10 15 20 25 80 Hydrochloric acid into water. Ditto into jelly. Potassic chloride 10 15 20 Days of 24 hours. into jelly. Height diffused in Centimetres. 254 Proceedings of Royal Society of Edinburgh, [march 19, Abstract of Results obtained in Present and Former Paper \\^tli July 1887] arranged in Curves — continued. a. Dijffusion of water containing 330 mgs. per c.c. of KHO into acid methyl orange, requiring nigs. KHO per c.c. to neutralise. h. Diffusion of water containing 185 mgs. KHO per c.c. into acid methyl orange, requiring mgs. of KHO per c.c. to neutralise. c. Diffusion of water containing 100 mgs. KHO per c.c. into acid methyl orange, requiring mgs. KHO per c.c. to neutralise. d. Diffusion of water containing 285 mgs. of KaHO per c.c. into acid methyl orange, requiring flf mgs. NaHO per c.c. to neutralise. 10 20 25 30 Soda. Potash do. do. into water. 15 20 Days of 24 hours. 0 5 10 25 30 Height diifused in Centimetres. 1888.] Mr Coleman on a New Dijfusiometer. 255 Abstract of Results obtained in Present and Former Paper \\bth July 1887] arranged in Curves — continued. a. Diffusion of water containing 377 mgs. of nitric acid per"! c.c. into alkaline methyl orange, requiring mgs. per c.c. to neutralise it. h. Diffusion of water containing 289 mgs. of sulphuric acid per c.c. into alkaline methyl orange, requiring mgs. per c.c. to neutralise. c. Diffusion of water containing 215 mgs. of SO3 per c.c. into jelly im- pregnated with nitrate barium, requiring mgs. of SO3I per c.c. to precipitate. YOL. XV. lO/il/88 R 256 Proceedings of Royal Soeiety of Edinburgh, [march 19, 5. Note on the Determination of Diffnsivity in Absolute Measure from Mr Coleman’s Experiments. By the President. 6. On the Soaring of Birds : being an Extract from a Letter of the late Wilham Froude to Sir W. Thomson, of 5th February 1878, received after Mr Froude’s death. So much for sails. Now I want to make some suggestions, or suggest some queries, as to the shimming flight of birds, in reference to which a good deal of fresh observation has been possible during the voyage. You perhaps recollect that when the British Association was at Glasgow, you asked me to put into ‘writing, briefly, as a paper for your section, some remarks on this subject which I had made to you in conversation, but that, owing to my hasty departure to attend the trial of H.M.S. “ Shah,” I omitted to do this. I had better briefly recite the above particulars here in order to make more clear the bearing of the new observations we (I and Tower) have made. The view was that when a bird skims or soars on quiescent wings, without descending and without loss of speed, the action must depend on the circumstance that the bird had fallen in with, or selected a region where the air was ascending with a sufflcient speed. In still air the bird, if at a sufficient height, could continue to travel with a steady speed, using his extended wings as a sort of descending inclined plane, the propelling force depending on the angle of the plane and on the equivalent of “ slip,” that is to say, on the excess of the angle of actual descent compared with the angle of the inclined plane. The steady speed would be attained when the weight of the bird and the sines of the angle of the plane = the bird’s air resistance., including skin friction of wings, in fact one might say = simply the skin friction of the whole area, for the bird’s lines are fine enough to justify this statement, since there is no wave-making to be done, and indeed experiment shows that the 1888.] Mr William Froude on the Soaring of Birds. 257 statement is true for “ fish-formed ” bodies moving wholly and deeply immersed in water. Of course the bird’s angle of actual descent is greater than that of the quasi-inclined plane, owing to the equivalent of “ slip ” in the wings. Under these simultaneously acting and correlated conditions there is of course — or probably — some total angle of descent which enables the bird to minimise his rate of approach to the earth in still air. If when there is a wind the configuration of the ground or any other circumstances can produce a local ascent of air more rapid than the bird’s minimum rate of descent when soaring in still air, he may continue to soar indefinitely by keeping in the region where the air is thus ascending. Now in most cases where one sees birds “soaring,” it is easy to see that they have plainly selected such a region, and for a long time I felt confident that the only two even apparent exceptions I had encountered were such as to prove not to invalidate the rule. One of these exceptions was that once, when the sea in Torbay was in a state of glassy calm, I noticed a large gull thus soaring at some distance from the shore, — watching it with a pair of binoculars, so that I was sure of the quiescence of the wings. But here the riddle was at once solved by the observation of what I had not at first noticed, — the dark trace of the front line of a fresh sea breeze advancing all across the bay. Such an advance with a definitely marked front, encountering an extended body of quiescent air, involved of course an ascent of air in the region of the encounter, and this was where the bird was soaring. The other exception was that when at sea I had often noticed birds thus soaring near the ship. The solution was that, so far as I had then noticed, the birds always selected a region to leeward of the ship, where the eddies created by the rush of air past her hull, &c., might readily have created local ascending currents. The new exceptions we have seen since we have approached the Cape entirely sets these two solutions at defiance. The first exception we noticed was in the flight of some albatrosses. We were sailing, and steaming (at low speed, being short of coal), nearly due east in the latitude of the Cape, with the wind light and variable abaft the beam, and with a well-marked S.W. swell of about 8” to 9” period, and varying from 3 or 4 feet to 8 or 9 feet 258 Proceedings of Boy at Society of Edinhurgh. [march 19, from hollow to crest. The speed of such waves would he from 24 to 27 knots. Under these conditions the birds seemed to soar almost ad libitum both in direction and in speed. ISTow starting aloft with scarcely, if any, apparent loss of speed. Now skimming along close to the water, with the tip of one or other wing almost touching the surface for long distances, indeed now and then actually touching it. The birds were so large that the action could be clearly noted by the naked eye even at considerable distances ; but we also watched them telescopically and assured ourselves of the correctness of our observations. The action was the more remarkable owdng to the lightness of the wind, which sometimes barely moved our sails, as we travelled only 5 knots before it, by help of the screw. After long consideration the only explanation of at all a rational kind which presented itself was the following, which indeed pre- sents the action of a vera causa, and one which was very often certainly in accordance with the birds’ visible movements, though it was often also impossible either to assert or to deny the accord- ance ; and anyhow the question arises. Is the vera causa sufficient % I will try to trace its measure. When a wave is say of 10 feet in height and say 10" period (a case near enough to ours to form the basis of a quantitative illustra- tion) the length of the wave from crest to crest is just 500 feet, the half of which space, or 250, the wave of course traverses in 5", and assuming the wave to be travelling in a calm, it must happen approximately that during the lapse of this 5" the air which at the commencement of the interval lay in the lowest part of the trough has been lifted to the level of the crest, or must have risen 10 feet, so that its mean speed of ascent has been 2 feet per second (10 feet in 5 seconds). And since (as is well known) the maximum speed of an harmonic motion is ^ times, or nearly IJ times its mean speed, it follows that all along the side of the wave at its mid-height the air must approximately be ascending at the rate of 3 feet per second, and if the bird were so to steer its course and regulate its speed as to conserve this position he would have the advantage of a virtual upward air current having that speed. if fe:-' Eroc.Roy Soc.Eain" Vol XV. PI v; Korl'liUnsf 1888.] Mr Peddie on the Electric Resistance of Liquids. 259 7. Preliminary Note on New Determinations of the Elec- tric Resistance of Liquids. By W. Peddie, B.Sc. {Abstract.) In this paper Mr Peddie described some preliminary experi- ments made on the resistance of dilute solutions of sulphuric acid (ordinary commercial). The solution is enclosed in a glass tube, which passes through a metal vessel containing water. Each end of the tube dips into a separate vessel containing some of the acid solution, and a platinum electrode is placed in each vessel. This apparatus is joined in circuit with a Helmholtz tangent galvano- meter and a Brush dynamo. The current from the dynamo is then passed through the arrangement, and the water surrounding the tube is stirred constantly until its temperature ceases to rise. The temperature of the water, the temperature of the air, and the current strength are then noted, and the resistance is obtained by the application of Joule’s Law, the rate of loss of heat being known. This method avoids the difficulties of polarisation and transi- tion resistance. The results already obtained agree roughly with Kohlrausch’s determinations. A full series of experiments will be made, and the results will be communicated to the Society. 8. Notice of the Recent Earthquake in Scotland, with Observations on those since 1882. By Charles A. Stevenson, B.Sc., Assoc. M. Inst. C.E. (Plate V.) In 1880 a rather sharp shock of earthquake was felt in Scotland, an account of which I communicated to this Society. Since then seven shocks have been felt in Scotland, and a notice of them, especially of the last, which occurred this year, will, I hope, be of interest to the Society. On 8th April 1882, the lightkeepers at Phladda Island reported that “at 7.37 p.m. a sudden shock of an earthquake passed the island from west to east, approaching with wave-like motion, dying away with a noise like distant thunder, lasting about three seconds. It was felt in the neighbouring islands.” 260 Proceedings of Royal Society of Edinburgli. [march 19, It may be mentioned that Phladda lies nearly on the line of the great fracture which runs through Scotland from Inverness in a south-westerly direction. This was the third earthquake felt at this station within five years. On 18th June 1885 a slight shock was experienced at Ballachulish, and also in Glencoe. On 26th September 1885, at 10 p.m., the lightkeepers at North Unst, which also lies nearly on the line of fracture, reported that “ we felt the tower shake very suddenly; the men in bed, as well as the man on watch, felt it the same. We can’t account for it, unless it was a slight shock of an earthquake ; no heavy sea, and the wind light from north. Barometer, 29*75 ; thermometer, 46° T.” On 18th December 1887, between five and six in the evening, a very slight shock of earthquake was felt in the Loch Broom district of Eoss-shire. On 12th January 1888 the meteorological observer at Glenquoich, Inverness-shire, observed a mild shock of earthquake. On Tuesday, 31st January of this year, an earthquake was reported in the newspapers as being felt in the midland counties of England. Eartlujuakes of 2nd February 1888. At 3.30 A.M., 2nd February, a slight shock was felt at Comrie, in Scotland, and about the same hour a slight tremor was felt at Loch Broom. About 5 a.m. on the same day a rather sharp shock of earthquake was experienced along the line of the Great Glen, making itself felt over a large portion of Scotland. The rupture seems to have taken place at Loch Ness, and the shock was propagated in all directions, but with diminishing severity. The following account, kindly communicated to me by Mr Paterson, Engineer to the Highland Eailway Company, is of special interest and value, coming as it does from an accurate observer ; — “ As regards the actual time the earthquake occurred, one of my assistants fixed it at 5 a.m. exactly, and others at 5.2 a.m. There were the usual six of us, including my daughter, aged 13, and two servant girls under my roof that night. Four of us were awakened by the shock ; my daughter and one of the servant girls were not. On awaking I felt a tremor, considerably exceeding the vibration we are accustomed to from shunting operations at the station (we are 1888.] on the Recent Eartliq^iiahe in Scotland. 261 on the Crown Terrace, and the engine-shed and station yard are immediately below us), and a creaking of the house timbers, similar to that of a vessel in the trough of the sea, and then a distinct upheaval for from one to two seconds. On collecting myself, I sprung out of bed, turned on our lowered gas, and found that the time by my watch — which was exactly to Greenwich time at 9 a.m. on the previous morning — was 5.2. I have, therefore, concluded that the shock occurred at 5.1 a.m. I can come no nearer to it than that.” The Inverness High Church clock is reported to have been stopped by the shock, but this is very doubtful. The following reports from lightkeepers who were on watch at the time the shock occurred, and consequently under favourable cir- cumstances for observing such phenomena, will be of interest : — Tarhetness. — “At 15 minutes to 5 a.m. (sun time), while I was on watch in the lightroom, all' of a sudden the tower shook very much — so much that the shades and lamp glasses rattled a good deal. The sensation I felt at the time appeared to me something like being on a railway bridge and heavy waggons passing over it. The vibra- tion of the tower continued, as far as I could judge, for three or four seconds. I could not say I heard any noise before or during the shock. It was blowing fresh at the time, and when such is the case our dome and lantern make a good deal of noise.” Barom., 29*69. Chanonry. — The lightkeeper writes — “ The only sensation I felt was a quick motion of the tower, and the chair I was sitting on shook very much. This occurred at 4,30 a.m. on the 2nd, and lasted for four seconds. I heard no noise, neither was it felt in the dwelling- houses.” Barom., 29*85. Gormn. — The lightkeeper says — “ I was awakened by a strange rumbling sound, and simultaneously came a rude shaking of the dwelling-house to the very foundations, as if caused by a very weighty machine passing. The keeper who was on watch in the lightroom at the time, said he heard a sound as if a flock of wild birds passed the lantern, and almost immediately the whole tower shook so that the spare lamp glasses in the tray and some of the red panes, &c. rattled, but nothing was broken or displaced.” Barom., 29*93. Oronsay. — The lightkeeper says that “at 4.41 a.m. (by our time), 262 Proceedings of Royal Society of Edinhu7''gh. [march 19, we felt a shock of earthquake, which lasted about ten seconds, accompanied by a dull, heavy rumbling sound. I came off watch at 4 A.M., and was in bed at the time of the shock, but not asleep, and I felt the bed, as it were, bodily lifted from the ground, and some of my children were awakened out of their sleep by the sound. The lightkeeper in the tower felt himself going backwards and forwards on the chair on which he was sitting.” Barom., 29*86. Ardnamurchan. — The lightkeeper writes as follows : — “ I was on watch in the lightroom when the earthquake occurred. I first heard a noise as if some heavy weight had fallen, but no shock, and in about one minute afterwards a similar sound, but much louder, with a distinct upheaval but no oscillatory movement. It occurred at 4.40 A.M. (dial time, 5.5 Greenwich). The whole time it lasted did not occupy more than three or four seconds. The lightkeeper, who had retired to rest at four o’clock, had not been asleep. He and his wife heard a low rumbling noise, as if of distant thunder, but no perceptible movement. The noise was not so loud as to awaken any of the others at the station who were asleep.” Barom., 29*73. Lismore. — The lightkeeper on watch “ distinctly felt and heard the noise of the earthquake. It began at 4.45 a.m., and continued one and a half minute. The noise was pretty loud, and awakened all the inmates at the station.” Barom., 29*91. This earthquake took place in the month of February, thus adding one more to the already long list of British earthquakes happening during the winter months — from November to the beginning of February. The diagram of British earthquakes, fig. 1, for the last eight years is sufficient to show this tendency. The shock occurred during cold, wet, and stormy weather, the average rainfall for January from five stations on the line of the Great Glen being 4*27 inches, the previous summer and spring having been unusually dry. The barometer was falling very uniformly over the whole of Scotland at the rate of \ inch in twelve hours. There was no steep barometric gradient. In the vicinity of the Great Glen the average height of the barometer for eleven stations was 29*7. The thermometer was falling over Scotland, the average at 9 p.m. on the previous evening being 32°, and at 9 a.m. on the 2nd 39° F. The moon at the time was in perigee ; it was nearly on the meridian, and the earthquake occurred shortly (five days) after full moon. 1888.] Mr Stevenson on the Recent Earthquahe in Scotland. 263 The area over which it was felt is shown on the accompanying map, Plate V., and measures about 15,000 square miles. At Inverness, Ardnamurchan, and Oronsay the shock seems to have been distinctly of a vertical character, and at many places an undulatory motion was experienced, the direction in most cases being stated to have been roughly at right angles to the line of the Great Glen. Two or more waves in quick succession are reported from Kyleakin, Grantown, Port Augustus, Dufftown, Strathpeffer, Perth, Oban, and Falkirk ; but at Dalwhinnie a shock is reported one and a half minute after the first. The duration of the shock at Inver- ness seems to have been at least two seconds, the average of all observations at a greater distance from the centre of disturbance being about four seconds. The earthquake was in all probability due to a rupture of the crust of the earth or slip of the strata in the Great Glen, probably of some length, at or near Loch Ness at about 5.1, Greenwich time, and seems to have spread out at a speed of about fourteen miles per minute (which is the average in seven directions) ; although this is 264 Proceedings of Boyal Society of Edinburgh, [march 19, only approximate, owing to the want of accurate observations at a considerable distance from the source. It reached Inverness at 5.1, Fort William and Glenluichart at 5.3, Oronsay 5.4, Cults, Broad- ford, Ardnamurchan at 5.5, and Banff at 5.9. The case of the two shocks one and a half minute apart at Dal- whinnie would lead to the supposition that there had been an overlapping of two earthquake waves here, and the times south of this would appear to point to this having been the case, namely, Ballinluig 5.1 to 5.2, Edinburgh 5.2. It is probable, therefore, that there was a subsidiary rupture, perhaps near Comrie, not caused by the passage of the wave from Loch Ness, but nearly simultaneously with it. The main shock, which was originated in the metamorphosed Lower Silurian rocks, made itself felt over Scotland, regardless, apparently, of configuration or geological formation, except, per- haps, that it was not propagated so far to the N.W. over the Laurentian and Cambrian formations of the north-west of Scot- land. One marked difference between this earthquake and that of 1880 was, that in 1880 the rumbling noise was confined to a very limited area near the source, whereas, in this case, all places, no matter how' far distant, heard the sound. Tarbetness and Chanonry lighthouses and Falkirk are the only three places at which “no noise” was reported. Chanonry is on a gravel spit, and Tarbetness is founded on a rock, with 10 feet of gravel over it. Mr Omond informs me that the earthquake w^as not felt on Ben Nevis. In conclusion, I have to thank Mr Murdoch Paterson, C.E. ; Mr C. Livingstone, Fort William ; Mrs Fowler ; Miss Sherriff j Dr Buchan ; Eev. Mr Hall, Comrie ; Mr W. Anderson Smith, and many others, who gave valuable information. The information received from various places is given in the accompanying table : — [Table 1888.] Mr Stevenson on the Recent Earthquake in Scotland, 26 o Stations. Character of Sound. Character of Disturbauce. Duration of Disturbance. Direction in which Wave appeared to travel. Loch Broom. Inverness. Rumbling. Shaking, tre- mor, and creaking of house tim- bers, and then distinct upheaval for from 1 to 2 seconds. 1 to 2 secs. N.W. to S.E. Dollar. Tarbetness. Rumbling. No noise. Shaking. 3 or 4 secs. Chanonry. No noise. Quick shaking. 4 secs. Corran. Oronsay. Rumbling (pre- ceding). Rumbling (ac- Rumbling and then shaking. Up and down. 10 secs. Ardnamurchan. companying). Rumbling. and back- wards and forwards. Distinct up- 3 or 4 secs. Edinburgh. Golspie. Sound of Mull. Rumbling. Shaking up ? down ? Rumbling (ac- heaval, but no oscillatory movement. Shaking. 12tol5secs. Tobermory. Cromarty. Kyleakin. companying) Rumbling (pre- ceding). Rumbling noise Trembling. Shaking, two 4 secs. W. to E. Dalmally. Macduff. (accompany- ing). Rumbling noise. Rumbling going waves up and down. Vibration. 1 min. Torridon. west. Noise. 1 to 2 secs. From W. or Glenmoristou. Rumbling (pre- Oscillation. 5 secs. S.W. Corrymony. ceding). Rumbling (fol- 10 secs. E. to W. Glen Urquhart. Tremor. 4 or 5 secs. S.W.to N.E. Strath Glass. lowing). Rumbling (pre- Shake. 4 or 5 secs. N.W. to S.E. Eskadale, Strath ceding). Rumbling (pre- Tremor. Over ^ a Glass. Strath Errick. Dornoch. ceded and fol- lowed tremor). Noise. Rumbling (ac- Trembling. Tremor and min. 3 or 4 secs. S.W. to N.E. 1 companying and followed). jolting ; two irregular up- ward and one side move- ment. 266 Proceedings of Royal Society of Edinhurgh. [march 19. Table — continued. Stations. Character of Sound. Character of Disturbance. Duration of Disturbance. Direction in which 'Wave appeared to travel. Clashmore. Noise. 4 to 6 secs. Loch. Buie. Rumbling. Easterly. Lismore. Noise. mins. Dalnaspidal. Spean Bridge. Rumbling. Oban. Rumbling before Wave motion, From N.E. 3 or 4 waves. Fort-William. Noise. Shaking (4 or 5 undulations). 1 min. N.E. to S.W. Grantown. Rumbling (pre- Tremor. 1 min. W. to E. ceded). Dingwall. Comrie and Criefl’. Rumbling. Oscillation. 10 secs. Banff. Aberlour. Noise. Shake. Dufftown. 3 vibrations. Duthil and Dulnain Rumbling. Shake. From W. Invergarry, Glen Rumbling. Shake. 4 secs. W. to E. Garry, and Glen Quoich. Strath Nairn. Rumbling (fol- Shake. lowing). Tain. Rumbling (pre- ceding). Undulation. 8 secs. W. to E. Plockton (Ross- Shaking. shire). Fort- Augustus. Rumbling. 3 distinct waves 2 or 3 secs. with tremors. Keiss. Vibrations. Lochluichart. Noise. Shake. S.W. to N.E. Nairn. Noise (preceded 2 to 4 secs. N.W. and accom- panied). 20 secs. S.W. to N.E. Garve. Noise. Tremor. Perth. Breadalbane, Aber- Rumbling (pre- 5 or 6 waves. 6 secs. W. to E. feldy, Grantully. ceding). Beauly. Strathpeffer. Rumbling. 3 distinct shocks 15 secs. W. to E. Glen Nevis. N.E. to S.W. (?) Cults (Aberdeen). • 3 to 4 secs. N. to S. Dollar. Rumbling (pre- ceding). Falkirk. No noise. 4 shocks. Inveraray. Rumbling noise. Vibration. 1888.] Prafulla Chandra Ray on Copper- Magnesium Groups. 267 Monday, 2nd April 1888. The Rev. Pkofessor FLINT, D.D., Vice-President, in the Chair. The following Communications were read : — 1. Analysis of the “Challenger” Meteorological Obser- vations. By Dr Buchan. 2. On the Conjugated Sulphates of the Copper-Magnesium Group. By Prafulla Chandra Ray, Esq. Historical and Introductory. From time to time memoirs have appeared by various chemists, pointing out that there is a tendency among the sulphates of the magnesium group to combine with one another in definite molecular proportions. This tendency has been brought into connection with the fact that these sulphates have almost identical atomic volumes. So early as the year 1840 Kopp drew attention to the fact that all the vitriolic sulphates have almost the same “ atomic ” volume {Ueher Atomvolum, Isomorpliismus und speeifiselien Gewiclit, Ann. xxxvi. p. 1, 1840). Playfair and Joule {flhem. Soc. Jour., 121, 1848), Schiff, and, recently, Thorpe have confirmed and extended Kopp’s classical work. Schauffele found (“ Ueber die mehrbasischen schwefelsauren Salze der Magnesiareihe,” Jour, fur PraTct. Chem., Iv. 371, 1852) that when one sulphate of the magnesium group is dissolved to the point of saturation in a previously saturated solution of another, the crystals which are obtained contain the component sulphates in definite proportions. In 1854 Eammelsberg published an elaborate paper on this subject. From the result of his researches he concluded that two sulphates of the copper-magnesium group often crystallise together in very simple ratios when they are dissolved together in equivalent propor- tions, the solution allowed to evaporate spontaneously, and the crystals collected fractionally as they are formed (“Ueber das Verhaltniss in welchem isomorphe Korper zusammen krystallisiren und den Einfluss desselben auf die Form der Krystalle,” Pogg. Ann., xci. 321). 268 Proceedings of Royal Society of Edinhurgli. [apeil 2, About the same time Vobl described a very large number of “ double-double” sulphates of the general formula (M"S04 . M'2S04 . 6H2O) -f (M”S04 • . 6H2O) , obtained by mixing the component sulphates in equivalent propor- tions, and allowing the solutions to evaporate spontaneously. Von Hauer, who followed a similar line of investigation to Schauffele, drew special attention to a series of definite compounds amongst the vitriols (‘‘Uber eine Eeihe von Verbindungen der Vitriole in bestimmten Aequivalent verbal tnissen,” Pogg. Ann., cxxv. 635, 1865). J. M. Thomson Assoc. Rep., 1877) endeavoured to prepare a nickel-cobalt- potassium sulphate of the type described by Vohl, but his efforts were unsuccessful. Still more recently, Aston and Pickering (Chem. Soc. Jour. Trans., 1886) have made another unsuccessful attempt to obtain salts of Vohl’s type. They, however, arrive at the conclusion that not only are salts of the type described by Vohl not formed when the con- stituent sulphates are mixed together according to his directions, but further that there is not the slightest tendency among the so- called vitriolic sulphates to form definite molecular compounds with one another. These latter investigations certainly seemed to throw very grave doubts on the trustworthiness of Vohl’s early research, but seemed to me insufficient to justify the sweeping statement of Aston and Picker- ing. As an attempt to settle the question, I took up this investigation. General RemarTcs on Methods of Preparation and Analysis. In all the preparations which I am about to describe, the con- stituent sulphates, after being weighed out, were dissolved together with the calculated quantity of an alkaline sulphate in cold water by trituration in a mortar, and a little free sulphuric acid added. In every case the amount of water was more than what was exactly necessary to effect solution j in other words, the solution was never allowed to be saturated, although it was always nearly so. In order to obtain results strictly comparable with One another, the con- stituent sulphates were usually mixed together in equivalent or very nearly equivalent proportions. The solutions thus obtained were allowed to evaporate spontaneously in flat-bottomed crys- 1888.] Prafulla Chandra Ray on Co;pper- Magnesium Group. 269 tallising dishes. Crystallisation did not in most cases go on continuously as the liquid evaporated, hut commenced somewhat abruptly, and having once commenced proceeded pretty rapidly. After a time crystallisation ceased altogether or nearly so, and at this stage the “ crop ” was collected. After a comparatively long pause, crystallisation recommenced and a second “crop” separated out. The second crop differed generally in composition from the first. Xot unfrequently, however, the end of the first crystallisation or crop overlapped the beginning of the second, so that a heterogeneous deposit of crystals of indefinite composition occurred. The amount of such heterogeneous deposits was usually small compared with that of the crops proper. The deposit or crust which adhered to the walls of the crystallising dish was rejected, as I found them to be of a different composition from the main mass of the crystals. For each case the combined weights of the sulphates taken was never less than 35 grammes, and the crops collected amounted usually to from 3 to 5 grammes. The crystals thus obtained were pulverised and air-dried. Detailed Remarks on each Preparation. I. Cobalt-Mckel-Potassium Sulphate. — One crop of this pre- paration was collected in accordance with the general plan. The crystals belonged to the rhombic system, and were of a pale blue colour. Examined by means of a Haidinger’s prism, they were found to be dichroic, as remarked by Thomson. The cobalt was separated as “Fischer’s salt,” which was then dissolved in hydrochloric acid, and the excess of acid got rid of by evaporatiou. From the solution the cobalt was precipitated by pure caustic soda. The hydrated oxide was dried and ignited in a current of hydrogen, according to Rose’s method. The reduced metal was thoroughly washed with hot water, to remove alkalies which adhere so tenaciously to the oxides of cobalt and nickel, and was then again heated in a current of hydrogen and weighed. The nickel in the filtrate was precipitated by pure caustic soda, and estimated as the protoxide. II. Zinc-Manganese-Ammonium Sulphate. — In this preparation the component sulphates were mixed together as above. The first crop consisted of limpid and transparent rhombic prisms. In 270 Proceedings of Royal Society of Edinhiirgh. [april 2, the analysis, the solution of these crystals was made neutral by the addition of a few drops of sodium carbonate solution, a small quantity of sodium acetate added, and the zinc then precipitated by sulphuretted hydrogen gas. The precipitate was dissolved in hydrochloric acid, and the zinc reprecipitated by sodium carbonate. Before being thrown on the filter, the precipitate was boiled re- peatedly with water, the water being decanted after each boiling. In some cases the ignited oxide had to he extracted with boihng water, it not being quite free from sodium carbonate. The manganese in the filtrate was precipitated as carbonate, and similarly treated. III. Copper-Iron-Ammonium Sulphate. — Only one crop was collected for analysis. IV. Copper - Iron - Ammonium Sulphate. — 11 grammes (*04 equivalent = 11*12), ferrous sulphate were mixed with 10 grammes ('04 equivalent = 9*98) cupric sulphate. Only one crop was col- lected for analysis. V. Copper-Iron-Ammonium Sulphate. — This is a duplicate of IV. Only one crop was collected for analysis. VI. Copper - Cobalt - Potassium Sulphate. — 25 grammes ( 1 equivalent = 24*95) copper sulphate were mixed with 28 grammes (*1 equivalent = 28*06) cobalt sulphate in presence of 35 grammes (*2 equivalents = 34*84) potassium sulphate. Three successive crops — Via, VI^, and Vly — were collected. The crystals of all the crops were rose-coloured. In the analytical separation of these samples some difficulties were experienced. When the copper was precipitated by means of sulphuretted hydrogen a re-precipitation was found to be absolutely necessary. Precipitation as cuprous sulphocyauate, by means of potassium sulphocyauate in presence of sulphurous acid, was found to work very well. To obtain perfectly accurate results, however, it is necessary to dissolve the precipitate in sulphuric acid, and to reprecipitate as sulphide. The precipitated sulphide was in all cases treated according to Rose’s method. The copper was also in some cases separated electrolytically. The first and second crops were kindly analysed for me by Messrs J. P. Macfarlan and Hugh Marshall, and the third crop by Mr Alexander Drysdale. VII. Copper - Magnesium - Potassium Sulphate. — This prepara- 1888.] Prafulla Chandra Ray on Copper- Magnesium Croup. 271 tion was made by mixing in equivalent proportions previously pre- pared double sulphates of copper-potassium (CuSO^. K2SO4. 6H2O) and magnesium-potassium (MgSO^. K2SO4. 6H2O). Ror the preparation of these double sulphates I am indebted to Mr Andrew King. Only one crop was collected. YIII. Copper-Kickel-Potassium Sulphate. — Unknown quantities of the double salts previously prepared, as in the preceding case, were dissolved in water. One crop only was collected. IX. Copper-Kickel-Potassium Sulphate. — Sulphate of copper and sulphate of nickel were weighed out in equivalent proportions, and dissolved in water along with the calculated quantity of potassium sulphate. X. Copper-Cadmium-Ammonium Sulphate. — The component salts for this preparation were mixed in equivalent proportions. One crop only was collected. The copper was separated as in IX. The cadmium was estimated as sulphide. XI. Copper- Cadmium- Ammonium Sulphate. — For this pre- paration the sulphates of copper and cadmium were mixed in the equivalent ratios 1:4. 10 grammes (‘04 equivalent = 9 ‘98) copper sulphate and 41 grammes (T6 equivalent = 40*96) cadmium sulphate were taken and dissolved together, with the calculated quantity of ammonium sulphate (27 grammes). The copper and cadmium were estimated as in X. XII. Nickel-Iron-Potassium Sulphate. — The component sul- phates were mixed in equivalent proportions. In the analysis the iron was separated as basic acetate. XIII. Xickel-Iron-Potassium Sulphate. — The sulphates of nickel and iron were taken in the equivalent ratio 1:4. 11 grammes (*04 equivalent = 11*22) nickel sulphate and 44*5 grammes (*16 equivalent = 44*48) iron sulphate were dissolved together, with the calculated quantity of potassium sulphate. XIV. Iron-Cadmium-Ammonium Sulphate. — The component sulphates were mixed in equivalent proportions. Only one crop was collected. XV. Iron - Manganese - Ammonium Sulphate. — 14 grammes (*05 equivalent = 13*9) ferrous sulphate were mixed with 12 grammes (*05 equivalent = 12*05) manganese sulphate and the calculated quantity of ammonium sulphate. One crop was collected. VOL. XV. 10/11/88 s 272 Proceedings of Royal Society of Edinburgh. [april 2, XVI. Zinc - Iron - Ammonium Sulphate. — 12 grammes (‘04 equivalent = HAS) zinc sulphate were mixed with 11 grammes (*04 equivalent = 11-12) ferrous sulphate, together with 11 grammes (*08 equivalent = 10-56) ammonium sulphate. Four crops — XYIa, XVI/?, XV ly, and XVI8 — were collected. In the analyses of all these four crops, and also in those of XVII. and XVIII., the following method for the separation of the iron and zinc was rigidly adhered to, a fact on which I lay special stress. In order to convert the iron into the ferric state, the substance was evaporated to dryness with nitric acid and dissolved in water with the addition of a few drops of hydrochloric acid. The iron, after addition of sodium carbonate to neutralisation and then of sodium acetate, was, by boiling, thrown down as basic acetate. This latter was dissolved in hydrochloric acid, and the iron reprecipitated by ammonia. The zinc in the filtrate from the basic acetate was estimated as in II. As a check, the iron was occasionally estimated by tritration with standardised permanganate solution. XVII. Zinc - Iron - Ammonium Sulphate. — The component sulphates in this case were mixed in the same quantities as in XVI. One crop only was collected. XVIII. Zinc-Iron-Ammonium Sulphate. — 28 grammes {'1 equivalent = 27-8) ferrous sulphate and 29 grammes (-1 equivalent = 28-7) zinc sulphate were dissolved in water with the calculated quantity of ammonium sulphate. Eight crops were collected. Each crop was, weighed. The weights in grammes were: — a = 3-5, y8 = 2-9, 7 = 3-0, 8= 2-1 €= 1-2, C = '9, 0 = 2-6. XIX. Magnesium-Iron- Ammonium Sulphate. — 12 grammes (*05 equivalent = 12-3) magnesium sulphate were mixed with 14 grammes (-05 equivalent = 13 -9) ferrous sulphate and the calculated quantity of ammonium sulphate. One crop was collected. XX. Magnesium-Iron-Ammonium Sulphate. — The components were mixed in very nearly the same quantities as in XIX. Seven crops were collected, the first three of which were weighed — a=Tl; y8 = 3; y = 2-3. In this case and also in XIX., the iron was estimated as described in XV. 1888.] Prafulla Chandra Ray Copier-Magnesium Group. 273 Discussion of Results. The following table contains all the necessary data. Of course, my analyses do not prove that in every case the com]3osition of the crystals is really represented by the formulae which I have assigned to them. This no mere analysis can do, hut there is an unmistak- able and remarkable agreement between theory and experiment, the differences between which in no case exceed the range of unavoid- able experimental error. The conditions under which I worked sometimes necessitated my using rather small quantities for analysis, but as a rule in the case of the metal present in larger quantity, the weight of the substance estimated was about 1 decigram. In the majority of cases, I have not contented myself with the determination of one metal only, but I have estimated both. It stands to reason that the estimation of that metal which is the pre- dominant one should have the principal say in determining the con- stitution of the salt. For instance, in No. X. the ratio of metals Cu : Cd is 12 : 1, and therefore the estimation of the copper is more to be relied upon than the estimation of the cadmium. From Column II. it will be seen that in the case of cadmium the in- fluence of a possible error of weighing is about eight times greater than in that of the copper. As pointed out by Professor Pickering, the estimations of sulphuric acid, of the alkalies, and of the water of crystallisation, are of little moment in the determination of the con- stitution of these salts, and I have made these estimations in the case of one salt only, viz. 11a, zinc-manganese-ammonium sulphate. I. II. III. Mean Theory. Zn : Mn Zn : Mn 9-2 2-1 Zn =13-33 13-20 13-69 13-41 13-36 10-93 Mn = 2-56 2-50 2-53 2-50 4-61 (NH^)^ 8-88 8-92 8-90 9-01 9-04 (SOJ =48-10 47-72 47-91 48-08 48-27 H,0 = (by difference) 27-25 27-05 27-15 100 00 100-00 100-00 From this it will be seen how little the estimations of the con- stituents other than the two metals Zn and Mn are to be relied upon in determining the constitution of the salt. The differences Table of Eesults of Analyses. 274 Proceedings of Boyal Society of Edinhurgh. [april 2, X! , , { — \ o w" tS CO CO In In o o m m K-i to X X ' — ' " — ' T)i o o 02 VO o O O VO CO o •OTJ'B.l juai.Tidttia jsaiduiig i-H jcM 05 |(M MI |vO CM |cO ? •sjiigtaAV^ituojv aAijoadsa.i am iiq papiAip •A uumpo UI saSi?juaa.ia. O VO CO O MI p O I— 1 CM <05 i-H CO O <05 O MI O VO I— 1 I— i 'AI uumpQ UI saSi?j -uaa.iaj jo uoapi VO CO o 05 rH CO P CO CM iM CO p p cb CO O rH p OO VO <05 •A I 'III Pu^ 'II uiojj pajujnopa jps ui pjait JO aJ^ujuaa.iaj O O O oooX CO CO CO p MI 05 13-34 Zn 13-20 Zti 13-69 Zn 2-56 Mn 2-50 Mn <0 O [++IQ <05 CM CO p p p CO C50 S3 pi S CS3 Cs3 ^ S O CO O CO O 05 CO CO CM O 05 VO CO rH CM o o o o o CO 05 o5 f+ f+ CO o : j— 1 t-H CO CO P p <“> o M CO OO^co 05 05" ;0 [+ +i O O MI O <50 CO <05 !>. CO rH O O tH s •ua5[t?j aauBjsqus jo saunuB.io ut jqStOAL CM OO CO CM QO CO CO OO CO p pi O O O O VO M MI VO <05 MI CO MI 1>- VO p p p p O 0-9678 0-4200 0-9678 t-H •sdo.iQ JO uoijuuSisao; s a 5*65 I - -06 ( -0337 I - *0003 1888.] Prafulla Chandra Ray Copper-Magnesium Group. 275 CO O K o 6 CO • O rH O O O o o o o o p p p p p + + + ’l T— ll^i— COOOOi-Hr-1 COMHCO 10^1— ICOOO CD^'rH OOi-HOO OOO O rH CM CO CO . o o ; o o CO 00 p P o o O! O! O O O Ol -rH !=*:::! 2d p Ci Ol ITJ 00 PmO OO OO OOO OO^ O^ O^^ OOOOO 05 . CO 05 00 05 CD rH (N . CO kO o 05 OO ■ oo M O O (N oo CO O CO o CO CO oo oo I— 1 CN Ml M O CO O CO O O kO kO (N O (51 CO kO Ml O CO oo CD Ml Ml CO CD CO OO (N CO CO o oo CO rH 1-1 p P P IM p P p P p P p p p p p p rH O O O O OO lM (N l-H O O (50 00 CO CD CO CO O (N CO O (N O oo O oo 05 M 05 l-H kO iH kO kO 05 CO (N 05 CO (N CO 00 CO 05 05 05 O M . (N M ?0 CO O OO CO CO OO CO : p M p p P p M p M M ; M p p p p p M p p M o O O iM o iM OOO O O O . O OO 00 CO 00 CO o o o o VO lo I-I O C5 O r-l T— I CO ^*1 CO CO 05 O r-l O O O CO VO CO CD (M i-H CD OO CO CO CO CO -^05 O O O O rH O 1— ' i>> J>- 05 . 1—1 VO CO 01 CM - CM CD CO O O O O O rH + I + + I I I 1 O I-H 05 VO

0 00 05 CO O O O VO (M VO 05 CO CO O 7-H CO CO OOo ODD pJH p!H Cs3 CD CM (M O O 'cH CO CO (05 OOO fO OO PH ; VO 05 CM CO o o ^ O 0> O Ph Ph N] CM CD OO CO rH CO CO CO CD OOO D a Ph !S3 o o 00 'CH Htl 05 O 1 — i CO o Ph ; VO OO CM O Oo p^ N VO o ^ CM CD o o O 01 Ph 1S3 OO o CM hH CO o O ,-H VO rH t-H CO O rH

' X > a CQ. Table of Results of Analyses — continued. I- n. III. IV.* V. VI. VII. VIII. IX. X. XL XII. i. It s li pi i Percentage of | calculated from ] n. and HL || bE • 1 1 1 d ¥ 1 Calcolated Percen- from Theory in accordance with 1 Column VII. il flsg 1 i XL 07580 07580 0-6098 -1068 CrtoS -0771 CclS -0615 CdS 11-24 Cu 7-91 Cd 7-84 Cd 11-24 j 7-88 •1774 ■0704 10-94 7-74 + ■30 + -14 { ::: 1 5Cu . 2Cd . 7[SOj , (NHJjSOj . 6H„0] XII. 0-4568 0-4175 0-4956 0-4175 oo iiii 1-06 Fo 1-11 Fe 12 85 Ni 13-01 Ni 1 1-09 } 12.93 ■0195 •2203 i ::: { ::: 1 This sample has nearly the ratio > Fe : Ni«=l : 11 (see discussion \ below), XIII. 0-6072 0-5733 0-6072 0-5733 -0210 FeoOo -0203 Fe'A •0690 Ni' ■0650 Ni 2-4-2 Fe 2-48 Fe 11-36 Ni 11-34 Ni ■0437 •1934 1 2-33 11-00 + •12 + ■35 !::: 1 2Fe . 9Ni . ll[SO, . KjSOi , 6H.O] XIV. o-Ssl 0-6197 0-4033 •1003 F00O3 •0962 Fo;03 •0435 CdS •0293 CdS 11-83 Fo 11-31 Fe 5-46 Cd 5-65 Cd |ll.32 •2020 T 11-11 5-55 + •21 j -0983 I -0288 + -0020 - -0007 + -0005 1 4Fe . Oil . 6[SOj ; (NH.)aSOj . 6HjO] r XV. 0-6171 0-6015 0-5467 0-6171 •0808 Fe„Os •0786 Fea'Oa •0394 MdjO^ •0445 MD3O4 9 -17 Fe 9-13 Fe 519 Jin 5-20 Mn }9.15 j 5-20 •1633 •0945 5-11 + -05 + 09 j *078? 1 -0388 + -0006 + -0003 + -0006 + -0008 1 7 Fe. 4Mn. ll[SOj.(NH,)jSO,. 6I4O]! XVI. n 0-8622 0-8622 •0355 FegOa •1392 ZnO 2-88 Fe 12-97 Zn 12-97 •0514 •1984 1 2-80 ^ -]2 •0345 •1405 + •0010 -•0013 j Fe . 4Zn . 6[SO, . (NHj)jSOj . 6H3O] 0-6205 0-6185 0-6205 •0306 FcaOa •0302 FegOa •0942 ZuO 3 ‘45 Fe 3-42 Fe 12-19 Zn ^ 12-19 •0614 -1365 1 3 3-51 12-28 -■07 -•09 i -0311 •0310 •0949 - -0005 - -0008 - -0007 1 Fe . SZn . 4[SOj . . 6H.O] 0-5230 0-7530 •0277 Fe^Og •0395 Fe'03 3-71 Fe 3-67 Fe ... Zn 1 3-69 ■0659 I ::: 1 A mixture of the preceding salt and J 2Fe.5Zn.7[S04.(NH,)oS04.6Hs0] 0-4887 0-6683 0-4887 •0332 FesOg •0316 Fe •0668 ZnO 4-76 Fe 4-73 Fe 10-98 Zn } 4-75 10-98 •0848 ■1680 1 2 4-69 10-94 \ -0327 j -0313 •0665 + -0005 + *0003 + •0003 1 Fe . 2Zn . 3[SOj . (NHj).SO, . 6H.O] XVII. 1 -2068 1-2068 •0480 FejOa •1940 ZnO 2-78 Fe 12-91 Zn 2-78 •0496 •1975 L 2-80 - -02 0483 •1966 - -0003 - -0026 j Fe . 4Zn . 5[SOj . (N-HJ^SOj . 6HjO] XVIII. a 0-5866 -0285 FegOg 3-40 Fe ... Zn 3-40 ■0607 ~S 3-51 12 28 - -11 •0294 - -0009 & 0-4200 0-4259 •0204 FcjOg •0645 ZnO 3-40 Fe 12-17 Zn 3-40 12-17 •0607 ■1862 L 12-28 -•11 - -11 •0210 •0651 - -0006 - -0006 Fe . SZn . 4[SOj . (NHJ^SO. . 6H.O] 0-6720 0-6720 •0328 Fe-.Og •1040 ZuO 3-42 Fe 12-43 Zn 3-42 12-43 •0610 •1901 1 3-51 12-28 + -IE •0337 •1027 - -0009 + -0013 Proceedings of Boyal Society of Ediiiburgh. [apbh2, j ,jgg] Prafulla Chandra Ray o» 277 Table op Results of Analyses — continued. 278 Proceedings of Royal Society of Edinhurgh, [april 2 ><1 I — I go Co tr» w IS CQ • .5 CO o> ^ Stn •5o O ^ OJ ’ ^ !=! ■B w (O o 00 d T|1 w o o &0 X} ’X enuiui ‘in !>. CO r-t »0 O O i-( i-l O O O O O . O O o o o : o o + + + I CO i-H Oi o o o o o o o p o 1 + + IIIA puT? 'll rao.ij pg^Binoi^JD -idpaaj JO JuSioAi CO lO 05 ^ {BJ 99UBJSqUS JO S9iuraB.iD ui jq§i9A\. •sdo.iQ 1 JO U0lJBuSlS9a o o 05 P) P^ IS3 O OO O lO OO o o OO XO 00 CM O CO CO o o OO (M a; 0^ XO (M CM o o OOO O »0 CO X'~ CO CO -rfi OOO p p b b OO CM !>. OO (05 XO X>. . XO XO • XO lO W I-H a XO (35 o CO O CO o CM CO r* O O 05 05 XO 1888.] Prafulla Chandra Ray Copper-Magnesium Grou'p. 279 o (M CO O W o VO^ c Table or Results of Analyses — continued. I. 11. III. IV. V. VI. VII VIII. IX X. XI. XII. 1 II 1“ 3 1 ■punoj mSpAv jjil sl If ife Percentages in Column V. Atom^Hveifihts. || Calculated Percen- tages of Metal Column VII. 1 1 1 Ill liJg i Fomnilaj of Salts. -0300 Fe.YOa 4-n Fe 4-11 •0734 4-01 + ■10 ■0293 + -0007 0-5114 -0758 ZnO 11-91 Zn 11-91 •1822 11-70 + •21 ■0745 + ■0013 0-5788 ■0320 Fe.03 3-87 Fo 3'87 •0691 2 4-01 - -14 ■0331 - -0011 ... Zu y , c 0-5592 ■0325 Fe.,03 4-07 Fe 1 4-01 ■0716 401 •00 / -0320 + -0005 ■ 2Fe . 5Zn . 7[SO, . (NH^ . )SO, . 6H.O] 0-5457 •0308 FejOg 3 -95 Fe \ -0313 - -0005 ... Zn 5 0-4575 -0274 FeoOg 4-19 Fe 4-19 •0748 2 5 4-01 + •18 ■0262 + ’0012 0-3997 •0248 FeoOg •0251 Fe'oOa 4-34 Fe 4-34 Fe ... Zn } 4-34 ! ::: 1 A mixture of the preceding salt and J Fe.2Zn.3[SO,.(NH,)jSOj.6HjO] XIX. 0-9020 •0770 FesOg 5-98 Fe ( -0778 - -0003 \ 0-5066 0-4940 •0435 FesOg •0438 FegOg 6-01 Fo 6-13 Fe 1 6 04 •1078 6-00 + ■04 \ -0434 { -0424 + •0001 + -0009 1 2Fe . 3Mg . 5[SOj . (NHj).,S04 . 6HaO] , 4-03 Mg •1679 8-92 + •11 j ^ ^ XX. a 0-4526 •0392 Fe.Og 6-06 Fe ... Mg 6-06 •1082 2 3 ^92 •0388 + -0004 0-4150 •0364 Fe^Oa 6-14 Fo ... Mg 6-14 •1096 1 6-00 3-92 + •14 ■0356 + -0008 1 2Fe . 8Mg . 6[SO. . (NH.).SOj . 6H.O] y 0-4955 ■0436 Fe„03 616 Fe ... Mg ■1100 y ^92 + •16 •0425 + ■0011 J 5 0-2056 ■0188 Fe.Og 6-40 Fe ... Mg 6-40 •1142 1 6-42 3-72 -•02 •0188 ■0000 j 3Fe . 4Mg . 7[SO, . (NH,),SOj . 6H.O] f 0-4165 0-5123 ■0392 Fe^s ■0490 Fe^Oa 6-59 Fe ... Mg 6-70 Fe ... Mg 6-59 6-70 •1176 ■1196 I 6-65 3-62 6-65 3-62 -■06 + •05 •0396 ■0487 -•0004 + -0003 1 4Fe . 6Mg . 9[SO, . (NHJ.SO, . 6H.O] 0-3370 0-3120 ■0332 Fe..Oa •0308 FCoOg 6-90 Fe 6-91 Fe ... Mg 1 6-91 ■1234 1 1 A mixture of the precediug salt and r Fe.Mg.2[S0j.(NH.).S0..6H.0] Proceedings of Eoljal Society of Edinhiirgh, [april 2 | ;^388.] Prafulla Chandra Ray o?i Coj7^er-il/«^7icsi«m G^ro!t2^. 279 280 Proceedings of Royal Society of Edinhurgh. [april 2, in tlie percentages of these two metals, calculated from theory for the two salts of ratios, 9 : 2 and 2 : 1 is very marked, and the results of the analysis agree very well with the theoretical percen- tages of the formulae assigned. It is in accordance with the behaviour of isomorphous mixtures, that when two salts are in solution together that which is less soluble has a tendency to crystallise out at first in excess of the other. Thus it is that in the case of ISTo. I. we have a salt formed having the ratio of the two metals Co : M = 1 : 2. So also in the case of copper-cobalt-potas- siuin sulphate, the first two crops have the ratio Cu : Co = 5 : 4. The withdrawal of a proportionally large amount of copper-potas- sium sulphate having altered the ratio of the metals in the mother- liquor, the influence of the greater solubility of the cobalt-potassium sulphate was counteracted by its presence in a relatively larger quantity; and the next crop of crystals consisted therefore of a salt of Volil’s type, in which the ratio of the metals Cu: Co = l : 1. Preparations X. and XII. show striking instances of the influence of solubility. The ratios in both are nearly 1:12, the greater number referring in each case to the less soluble salt. The results of the analysis of these samples I do not take for more than an indication of how the respective sulphates tend to behave when taken in equivalent proportions. To study the behaviour of these sulphates further, preparations XI. and XIII. were undertaken. In each of these two cases the more soluble salt was taken in four times the equivalent proportion of the less soluble one. Under these circumstances, salts were obtained having the formulae assigned to them in the table. The difficulties in the way of discriminating between a homo- geneous crystallised salt and a mixture consisting of two or more salts in variable proportions are undoubtedly great, Rammelsberg says — ‘‘ The analysis of a certain number of crystals is no guarantee for the constitution of individual crystals . . , . The result in general must be regarded as an expression for the mean of one entire crop. .... Unfortunately, I could never analyse any of the crystals singly on account of their diminutive size ” (loc. cit.^ page 329). Von Hauer attempted to get over this difficulty by analysing individual crystals which he was fortunate enough to obtain sufficiently large. 1888.] Prafulla Chandra Ray on Copper- Magnesium Croup. 281 There is generally a considerable pause between the deposition of two successive crops, during which no crystallisation takes place, as I have already pointed out. Dr Gibson suggested to me that I should collect a given crop fractionally, and ascertain whether each fraction had the same com- position, and thereby obtain proof positive of the homogeneity of the whole crop. This suggestion I have attempted to carry out in preparations XVIII. and XX. The fractions deposited during each period of crystallisation were carefully collected, and in a number of cases weighed. These weights will be found under the details of the individual preparations in the early part of the paper. In XVIII. the numbers obtained by the analy&is of the first three fractions collected (a, and y) agree with those calculated from the formula Ee. 3Zn. 4[S04. (XHJ2SO4. 6H2O]. The analyses of the next four fractions (8, e, 4 ’?) gave results agreeing in each case with the formula 2Fe. 5Zn. 7[S04. (XH4)2S04. 6H2O]. Thus I have actually succeeded in collecting two crops of crystals having definite though different molecular compositions, in three and four fractions respectively. This behaviour has an evident and important bearing upon the question raised by Aston and Pickering, and cannot be explained it seems to me in accordance with their views. During the deposition of the first three fractions a considerable change in the composition of the mother-liquor necessarily occurred, and yet these three fractions proved on analysis to be all of the same composition. The numbers obtained by analysis agree closely, moreover, with those calculated for a simple compound of the two constituent double sulphates. After the deposition of the third frac- tion of this first crop, some time elapsed before any further crystal- lisation was observed. The results of the analysis of the four next fractions in which the second crop was collected, proved these four fractions to be all of practically identical composition, unmistakably different from that of the first three fractions, but agreeing closely with the calculated composition of the not very complex com- pound assumed above. The change in composition between the two “crops” was not in any sense continuous, as it should have been according to Aston and Pickering, but was distinctly abrupt, thus the percentages of iron found were 3’40, 3 -40, 3*42, 4*11, 3*87 (?), 4*01, 4*19. 282 Proceedings of Eoyal Society of Edinhurgh. [apb.il 2, In preparations XX. I have similarly been able to prove the same thing. The first three fractions have all the composition, 2Fe . 3Mg . 5[S04 . (XH4)2S04 . GHgO]. The next salt which forms is 3Fe. 4Mg. 7[S04. (XH4)2S04. GHgO], and this came down as a whole in the fourth fraction (8), but with greater care it also might have been fractionated. The next salt which formed was 4Fe . 5Mg . 9[S04 . (XH4)2S04 . 6H2O], and this I was able to collect in two fractions (e, ^). The next fraction (rj) appears to consist of a mixture of the two salts — 4Fe. 5Mg . 9[S04. (XH4)2S04 . 6H2O] and Fe . Mg. 2[S04. (XH4)2S04 . 6H2O]. The percentage of iron is greater than what is required for the former, hut it is less than what is required for the latter. This is a case of overlapping, and we have further instances of the same thing in XVIy and XYIII^. In preparation XVI. we have no fraction which agrees with the formula, 2Fe . 5Zn . 7[S04. (XH4)2S04 . 6H2O] ; hut as in preparation XVIII. we have undoubted proof of its existence, it must have been present in XVIy, mixed with the salt Fe . 3Zn. 4[S04 . (XH4)2S04.6H20]. In preparation III. as compared with preparations IV. and V., and in preparations XVI. and XVII. as compared with preparation XVIII., there are apparent anomalies. Why has the salt No. III., the ratio Fe : Cu = 4 : 5, and the salts for Nos. IV. and V. the ratio Fe : Cu = 2 : 3 1 And again, why have the first salts deposited from XVI. and XVII. the ratio Fe : Zn = 1 : 4, and the salt from XVIII. the ratio Fe : Zn = 1 : 3 ^ The answer to these queries probably lies in the fact that in the solutions, as originally made, the ratio of the metals must have varied slightly in the respective cases, as the salts were weighed out only very roughly. Further, the temperature may have been different in the two cases. I am far from denying that slight changes in the various conditions under which the crystals are formed may have a determining influence upon the com- position of the compounds separating out from solution, but main- tain that by proceeding cautiously, and especially, as above described, by taking advantage of spontaneous evaporation, I have succeeded in proving the existence of definite compounds of the class dis- believed in by Aston and Pickering. The reason why these investigators failed to obtain definite compounds seems to me to lie in their modus operandi, wPich necessarily brought about the 1888.] Prafulla Chandra Kay on Copper- Magnesium Group. 283 formation of heterogeneous deposits. The conclusions which they have drawn from the analysis of such preparations are therefore unwarranted. It seems evident that the results of former investi- gations, as well as those obtained by me, prove the existence of “double-double” sulphates of definite composition. The chemical affinity which determines the formation of such compounds is not indeed very powerful, but its influence is quite unmistakable if proper care be taken to avoid conditions under which it is necessarily obscured. Professor Crum Brown pointed out to me that there is room for subsequent research in the determination of the limits of the composition of the solution between which given definite com- pounds separate out. I accordingly made some experiments in this direction, but must leave any discussion of this interesting subject to a future date. The whole question seems to me worthy of the attention of those who can find time to carry out the numerous and often troublesome analyses necessary for the solution of the various problems which present themselves. In conclusion, I beg to offer my grateful thanks to Dr John Gibson for his help and advice, and for the many important sugges- tions which he has made throughout the course of the investigation. I am also deeply indebted to Mr T. F. Barbour for the ready and constant aid I have received from him while drawing up this paper ; indeed, it is not too much to say that without Mr Barbour’s un- grudging labour and sacrifice of time, the table of the Eesults of Analyses could not have been presented in the very complete form in which it now appears. I also gladly avail myself of this op- portunity to express my cordial thanks to Professor Crum Brown for counsel, criticism, and encouragement during the progress of my work. 2. On the Chemical Composition of the Water composing the Clyde Sea Area. Part II. By Adam Dickie. This is merely the concluding portion of the tables attached to a paper which was formerly communicated to the Society by Dr Murray, and which is printed in your last Part of Proceedings. I need only state that the same methods described in that paper Table of Eesults. 284 Proceedings of Royal Soeiety of Edinhurgli. [april 2, were used in these analyses, of which the accompanying tables give the results : — * O OO 05 CM j:^ Oi iH CO tH CM VO c:> (M 05 rfi CO l>. CO CO CO Ci CO rH CO . irj . •H Hi hH rH hH -HH rH rH rH rH I— I rH rH rH rH rH |H rH rH rH 1 — I rH ^ — 1 rH O P, CM ■a 00 CO OO rH CO CO CM a a <1 OO ■jH ip (M p p p p p p p OT »o O 05 05 <50 <“> CO C5 o oT rs a 0 a rQ ra CQ o o *o zn ■% VO '13 Do. ’3 a cS a o ■rs S ..a o c3 a -M 3 o 5a .a fi a a cS .a o O O CO o CO 0 q-l ^ CO CM rH VO CO 00 (M CO VO 0 <35 ° a 05 <35 OO OO <35 <35 OO OO CO o 3 -H H H hH hH a rH a '^02 i—H rH I— i rH rH rH rH rH rH rH a rH Table of Results — continued. 1888.] Mr A. Dickie on the Composition of Clyde Water. 285 O r-H oo CO CO 05 CD lO cq Cl VO O lo O (M oo VO ■dH rH CO oo 'di CM rA ^ rA 'Hi 'HH hH H4 VO -HH VO VO 60 t-H 1-H 1— i rH 1-H 1-H rH 1—i rH rH pH r— ( rH 1 — 1 .M o « (M CD cq CD cq cq Hfl OO CO CO vp cp vp *P p cp 'H p • ^ oo lO 05 i>- 05 CD 05 oo • CD CO VO OO '=^1 'dH 'Cfl 'di 'di Hhi HtH VO rt1 HfH u o CO CM oo 05 i—H 05 oo q-- Ht1 7 — q cq OO oo CO OO oo o VO VO VO rd o CM (M VO VO CO VO CM (M VO 'd cq CO o 1 — I 05 1-H O o 1-H o rH 1-H O 1 — ! o p p o a Cfi ^ cq 1— ( CM (?q (N OJ cq cq cq cq cq cq cq cq oT 3 05 05 CO VO CD CO CO 1 — 1 CD oo cq 9 1—4 CO oo CO CD oo 05 oq 05 CO CO <05 q^ O Ip vp — * t-H W W oo 00 VO VO 05 05 CO CO rH tH VO VO 00 00 t-H rH 1 — 1 1 — 1 rH rH rH ^ — 1 w x^ q:^ q>. q^ q^ q^ qr^ q:^ oo oo oo oo OO oo oo oo 00 oo oo oo 00 oo oo oo oo oo oo oo oo oo oo CO oo oo oo oo 1—1 rH rH 7— H 1-H 1-H 1 — 1 rH rH pH rH pH I-H Pd o5 05 05 00 oo 05 05 C55 05 05 05 05 05 o o c3 ft cq CM (N (M oq H XO lO 'dv VO VO XO VO VO VO VO VO VO VO CO m c/2 CO CJl m a a s a 3 a o o o CO P CO o o o rP rP rP CO CO Vh 1— 1 ^ s rH ."P Cr-t O .a «a cq ft 0) ft T— 1 a"' o o CO a &B sS o o ’Jh o d Z/1 6 *o o d nq 6 o n:^ 6 i-q -ps rP Pj o 'o o 'o w o 'o w a C3 CJ o o CJ aJ OJ d o T3 o o hq rP o o i-q rP P O D o3 o Do. o C3 a 6 6 'o fi D a o d CIS rP d d o ft M Q o Ph Q o +5 a C Q Ph U) Q P !-i O P Co cd o3 p o rP rCj o Q o m Q H 02 'S’2 VO VO VO VO VO VO VO VO VO VO ^ CJ rH 1-H rH t-H pH i-H iH rH t-H t-H Table of Eesults — continued. 286 Proceedings of Boyal Soeiety of Edinburgh. [april 2, •p9nnu.i9:}ap ;ou s;u9iiodraoo J9ii:^o o SI, Q Q !zi PQ be o O £>0 o Q to o p-< cS O to o fl o niJ a ci ito 02 to o o • P3 . o B ^ d S S •+J o