never touches on a metaphysical idea. Fortunately, philosophers are becoming more mathematical, and vice versâ; so the absorption of science and philosophy into poetry seems much less distant than a century ago. G. B. M.

A YEARBOOK OF SCIENCE. Jahrbuch der Naturwissenschaften, 1911-1912. Edited by Dr. Joseph Plassmann. Pp. xvi+ 452. (Freiburg im Breisgau and London: B. Herder, 1912.) Price 7s. 6d.

THE

HE twenty-seventh volume of this useful publication is well up to the level of its predecessors. In spite of the great expansion of all the subjects treated, the size of the work has not been increased. This implies a more and more "intensive treatment, and a careful selection of topics. In physics, the 5000 odd new publications of 1911 have been brought within the compass of fortyeight short notes. The task of selecting one paper out of every hundred must be a formidable one. Dr. Heinrich Konen, to whom it fell, took care to emphasise those which offer a certain amount of novelty or practical utility, such as Lebedef's shortest possible sound-waves (o'2 mm.), which are absorbed by 24 cm. of air; Rubens's longest light-waves (0'116 mm.); Féry's prism with

ment, mining, ceramics, naval construction, freezing plant, gas industry, and firearms.

A calendar of astronomical events and an obituary complete the work, which may be regarded as an almost indispensable work of reference. It should be stated that it is printed in the Gothic type, and not in the Roman type now usual in German scientific publications.

GEOGRAPHICAL WORKS.

(1) The Elements of Geography. By R. D. Salisbury, H. H. Barrows, and W. S. Tower. Pp. ix+616+7 maps. (New York: Henry Holt and Co., n.d.) Price 1.50 dollars. (American Science Series.)

(2) A Geography of the British Empire. By Prof. A. J. Herbertson and R. L. Thompson. Pp. 256+3 maps. (Oxford: Clarendon Press, 1912.) Price 2s. 6d. (The Oxford Geographies.) (3) Forfarshire. By E. S. Valentine. Pp. viii+ 160+2 maps. (Cambridge: University Press, 1912.) Price IS. 6d. (Cambridge County Geographies.)

(4) The Lost Towns of the Yorkshire Coast and other Chapters bearing upon the Geography of the District. By T. Sheppard. Pp. xviii+329. (London: A. Brown and Sons, Ltd., 1912.) Price 7s. 6d. net.

HE American geography under notice

of Rowland gratings.

The chemistry section is rather insufficiently separated from the industrial section, and so it happens that such things as the utilisation of zirconia, and the preparation of illuminating gas free from CO, are dealt with twice over. Dr. Plassmann himself writes the section on astronomy, and devotes considerable space to Martian questions and the mass of the ring of planetoids. Bauschinger's estimate of the latter, amounting to about one-fiftieth of the mass of Mercury, is supported on optical grounds.

Among the subjects dealt with in meteorology we find Wegener's stratification of the atmosphere, wind velocities, sunspots and weather, an aeronautical weather service, and Birkeland's theory of terrestrial magnetism and allied pheno

ment of geography in the University of Chicago. British writers of geographical text-books have yet to follow German and American writers in work of this advanced character. The present volume forms, therefore, an interesting study, possessing many virtues and certain faults. The writers have followed general theoretical lines, avoiding those of the ancient "cosmography" with its principle of description according to countries.

After a short general discussion of the earth as a planet, and of its main features, we find a proper importance awarded to climate and weather, to which seven chapters are devoted out of a total of twenty-one. After these the authors deal with the oceans, then the "materials of the land" (soils, minerals, etc.), and lastly land-forms, with the consideration of the forces which shape them, and their influence on human conditions and on life generally. This is probably the best order that could be followed, though throughout the long section on climate there is some temptation to wish that a few more leading facts concerning the configuration of the surface and the other subjects of the later sections had been transferred to an introductory chapter, so that the student should be, at the outset, more clearly in possession of

the exact meaning of the importance to be attached to the climatic factor.

A tendency observed in other American textbooks is also to be noticed in this-that of introducing details which can scarcely be considered to have any relation with geography, even following the widest connotation of that term. The very close details of output in the economic chapters provide an illustration in point, valuable as they are, no doubt, in themselves. This book is very fully illustrated; many both of the diagrams and of the views are on too small a scale to fulfil their purposes properly.

(2) The "Geography of the British Empire," by Prof. Herbertson and Mr. Thompson, is arranged on a simple descriptive plan, and illustrated with a very large number of sketch-maps mostly showing very clearly the special points which they are intended to show, though not all are free from the charge of over-reduction. The book is of an elementary character, and little or no endeavour is made to deal with the inter-relation of the various parts of the Empire, though these are treated individually with a due sense of proportion. This proportionate treatment, within the compass of one volume, is in itself a valuable educational achievement, indicating what should be the first object of geographical teaching in British schools. It may be regretted, perhaps, that the coloured physical maps are confined to the representation of the British Isles.

(3) The Cambridge County Geographies have unquestionably improved since the inception of the series, and Mr. Valentine's volume on Forfarshire maintains the standard. In general reputation for scenic and kindred interests the eastern counties of Scotland have suffered in contrast with the western, yet Forfarshire is an area possessing many natural beauties, both on the coast and inland: its archæological interests are considerable, and its economic importance is high. All these aspects are clearly illustrated, both textually and by means of photographs, though the statistics freely quoted in the economic connection will not long maintain their value. The descriptions of ancient remains and buildings (for which Forfar is scarcely surpassed by any other Scottish county) are specially good.

(4) Mr. Sheppard provides a complete physical and historical setting for his study of the villages of Holderness which have been destroyed by the encroachment of the sea on the land. He cites authorities very fully, and has investigated old maps with great care; there is a chapter on these, with a number of reproductions, some of which have been reduced so far that not only the

minutiæ, but also the more salient features, are lost; in such cases the reproduction of the pertinent section of the map on a larger scale would have been preferable. There are many appropriate photographs and reproductions of old prints. In one respect the title of the book does less than justice to its scope, for the last six chapters are descriptive of the East Riding generally, and will serve as a useful guide to that district.

The books by Mr. Valentine and Mr. Sheppard both contain, as it happens, an explanation of the word "shire"; the two writers curiously disagree on the point.

OUR BOOKSHELF.

Photography of To-day. By H. Chapman Jones. Pp. 342+ plates. (London: Seeley, Service, and Co., Ltd., 1913.) Price 5s. net. THERE are a variety of text-books of photography in the market, one of which is by Mr. Chapman Jones. On turning to the work under review, to our great relief it is found to be of a totally different character from the ordinary variety. It contains no formulæ for developers or for anything else, but is what it professes to be "a popular account of the origin, progress and latest discoveries in the photographer's art, told in nontechnical language "-and is illustrated with excellent illustrations of pictorial art, and with some passable diagrams. The author commences with light and its effects, then continues with lenses, and follows on with a short history of photography told in a bright and readable manner.

The history of photography before the use of gelatine is cut rather short, but perhaps it is well, as those who read the work will, as a rule, be those who use a Kodak-the "press the button and we do the rest "kind of people. To such photographers the chapters on the gelatine process will be read with pleasure, and will at all events enable them to talk rationally about their hobby, which is seldom the case at present, with few exceptions, and it may be that by reading it they may wish to "press the button " and do the rest themselves. The printing processes are fully described, as are instantaneous photography and telephotography. Truth and error in photography have a chapter devoted to them. There is a saying as to "lying photography down easily in this respect. like a photograph." Mr. Chapman Jones lets

The author has produced a book which it is a pleasure to read, and with some small omissions has carried out its intention admirably. Allusion has already been made to the illustrations, which are all distinctly good. It would have been interesting if he had told us the method adopted of reproducing the picture of the frontispiece, "A Rainbow from an Autochrome," in more detail than he does. We can recommend the book to all, more especially to those who are not expert photographers.

The Botany of Iceland. Edited by Dr. L. Kolderup Rosenvinge and Dr. Eug. Warming. Part i. "The Marine Algal Vegetation." By Dr. Helgi Jónsson. Pp. vi+186. (Copenhagen: J. Frimodt; London: John Wheldon and Co., 1912.)

DANISH botanists are to be congratulated on the vigorous manner in which they attack the botany of the various dependencies of their kingdom. In the "Botany of the Faeröes" (1901-1908) the results of a systematic investigation of the flora and vegetation of those islands were presented, and with the completion of that work a similar survey of the botany of Iceland has been commenced.

The first part of the Iceland series, namely, the marine algæ, by Helgi Jónsson, has now appeared. It begins with the systematic list, which is concisely dealt with. An interesting account of the phytogeographic components of the flora follows, together with a comparison of the floristic features. of neighbouring areas. The remaining pages are occupied with a detailed description of the algal communities, and notes on the biology of the A new species. method of classification is employed; three main vertical "zones are recognised, and the communities of the littoral zone are subdivided according to their illumination requirements. It is open to question whether these divisions will meet with general approval, but all will agree that Dr. Jónsson has furnished a most valuable contribution to algological literature.

A. D. C.

A Medical and Surgical Help for Shipmasters and Officers in the Merchant Navy; including First Aid to the Injured. By W. Johnson Smith. Revised by Dr. Arnold Chaplin. Fourth edition, revised. Pp. xviii+355. (London: Charles Griffin and Co., Ltd., 1912.) Price 5s. net. DR. CHAPLIN has re-written the portions of the work dealing with the causation of diseases, so as to incorporate the recent advances in our knowledge, especially of tropical diseases. The new scales of drugs and medical and surgical appliances, issued by the Board of Trade in January, 1912, have been included, and in other ways the volume has been brought into line with present-day requirements.

A Handbook of Wireless Telegraphy: its Theory and Practice. By Dr. J. Erskine-Murray. Pp. xvi + 442. Fourth edition. (London: Crosby Lockwood and Son, 1913.) Price 10s. 6d. net. A REVIEW of the third edition of Dr. ErskineMurray's book will be found in the issue of NATURE for August 24, 1911 (vol. lxxxvii., p. 240). The additions to the present edition include a new chapter on the telegraphic efficiency of a wireless system; a theory of abnormal ranges, by night and by day, deduced directly from telegraphic observations, now included in the chapter on transmission; and new sections in other chapters on the Poulsen, Goldschmidt, and new Telefunken systems.

LETTERS TO THE EDITOR.

[The Editor does not hold himself responsible for opinions expressed by his correspondents. Neither can he undertake to return, or to correspond with the writers of, rejected manuscripts intended for this or any other part of NATURE. No notice is taken of anonymous communications.]

On the Appearance of Helium and Neon in Vacuum Tubes.

Ar the last meeting of the Chemical Society, Sir William Ramsay, Prof. Collie, and Mr. Patterson described some experiments which they regard as proving the transmutation of other elements into helium and neon. I have been making experiments of a somewhat similar character for some time, and though the investigation is not yet finished, the results I have obtained up to the present time seem to me in favour of a different explanation from that put forward at the Chemical Society. I described some of these experiments in a lecture at the Royal Institution on January 17, but as the separate copies of that lecture have not yet been issued, I will give here an account of some of the experiments which seem to me to have the most direct bearing on the phenomenon in question.

of a

I used the method of positive rays to detect the gases; this method is more sensitive than spectrum analysis, and furnishes much more definite information. I may say that the primary object of my experiments was to investigate the origin and properties new gas of atomic weight 3, which I shall call X,, which I discovered by the positive-ray method. This gas, as well as one with an atomic weight 20 (neon?), has appeared sporadically on the photographs taken in the course of the last two years; the discharge in the tube being the ordinary discharge produced by an induction coil through a large bulb furnished with aluminium terminals, and containing gas at a very low pressure. There seems to be no obvious connection between the appearance of either of these lines and the nature of the gas used to fill the tube; the 3 line has appeared when the bulb was filled with hydrogen, with nitrogen, with air, with helium, or with mixtures of hydrogen and oxygen in various proportions; the 20 line when the bulb cor.. tained hydrogen, nitrogen, air, hydrochloric acid gas, mixtures of hydrogen and oxygen.

The experiments I made had for their object the discovery of the circumstances which favour the production of X,, and to test whether it was triatomic hydrogen produced by the discharge, as this is the alternative to its being a new element. I have found that the conditions which lead to a considerable production of X, generally give rise to the appearance of helium and neon. Indeed, in the great majority of cases in which I have observed the appearance of traces of helium and neon these gases have been accompanied by larger quantities of X,; this gas seems to have escaped the notice of the readers of the paper at the Chemical Society. I may mention, too, that along with neon of atomic weight 20 there is a line in these circumstances corresponding to an atomic weight 10 or thereabouts. Though this is probably due to neon with two charges of electricity, it is generally brighter in comparison with the neon line than is usual for the lines corresponding to doubly and singly charged atoms, so that it is not impossible, though perhaps unlikely, that it may be due to a new gas. The positive rays for the analysis of the gases were produced in a vessel containing gases at a low presI shall call this the testing vessel; the vessel in which the various processes for generating X,

sure.

were tried (the experimenting chamber) was sealed on to the testing vessel, but separated from it by a tap. Thus the pressure in the experimenting chamber was not restricted to being the same as that in the testing vessel, but might have the value which seemed most appropriate for any particular type of experiment. After these experiments were over, the tap was turned and some of the gases from the experimenting chamber let into the testing vessel; a photograph was then taken, and by comparing it with one taken before turning the tap the new gases present in the experiment chamber could be detected. The processes by which I have hitherto got the most plentiful supply of X, are:

3

(1) By bombarding with kathode rays metals and other bodies.

(2) By the discharge from a Wehnelt kathode through a gas at a low pressure.

(3) By an arc discharge in a gas at a comparatively high pressure.

By far the larger number of the experiments were made by bombarding metals, but I will begin by describing an experiment with the arc, as it raises the question of the origin of these lines in a very direct way. An arc between iron wires passed through hydrogen at about 3 cm. pressure (in this case all the kathode rays would be absorbed quite close to the electrode) for an hour or so, and the gases liberated in the experimenting chamber tested; X,, helium, and neon were found. The experiment, using the same wires for terminals, was repeated the next day; the three gases were again found. On the next day, still using the same wires, the arc was passed through oxygen; the X, line was still there, though much fainter than before; the helium and neon could not be detected with certainty. The next day, using the same terminals, the arc was again passed through oxygen; not one of the lines could be detected. This looks as if these substances were produced by the arc passing through hydrogen. It was found, however, that, still keeping to the same terminals, on pumping the oxygen carefully out and filling up again with hydrogen, the arc through the hydrogen now did not give even a trace of these lines. On replacing the old iron wires by new ones, and sending the arc through the hydrogen, the lines reappeared. This experiment seems to me to point very clearly to the conclusion that these gases were in the terminals to begin with, were removed from them by the longcontinued sparking, and were not produced de novo by the arc.

In the experiments when the discharge was produced in a tube with a Wehnelt kathode, the potential difference between the terminals was only 220 volts, so that the kathode rays in the tube had only a fraction of the energy they had when the discharge was produced by an induction coil; X, and helium appeared when the discharge passed through this tube. I did not detect any neon.

The method which gave X, and also the other gases, in the greatest abundance, was to bombard metals, or indeed almost any substance, with kathode rays. The tube used for this purpose had a curved kathode, which focussed the rays on a table on which the substance to be bombarded was placed. The substance, round the spot struck by the rays, was generally raised to a bright red heat by the bombardment; the bombardment was as a rule continued for five or six hours at a time. I have got the X, line, as a rule, accompanied at first by the helium line, and somewhat less frequently by the neon line, when these following substances (which include nearly all I have tried) were bombarded iron, nickel, oxide of nickel, zinc, copper, various samples of lead, platinum, two

meteorites, and a specimen of black mica given me by Sir James Dewar, which was remarkable for the amount of neon it gave off.

The most abundant supply of X, came from platinum, and I will describe an experiment with this metal. A piece of platinum foil was bombarded on four days, and the gases produced each day examined. At the end of the first day's bombardment it was found that the line due to X, was very strong, those due to helium and neon weaker, but still quite conspicuous. The gases produced the first day were well washed out of the tube, and the foil bombarded for a second day. The gases formed proved to be much the same as on the first day; there was no appreciable diminution. The examination of the result of the third day's bombardment showed that the X, line had diminished considerably, the lines due to helium and neon perceptibly. When the gases produced on the fourth day's bombardment were examined it was found that the X, and helium had diminished to such an extent that the lines were barely visible. I could not see the neon line at all. In this case the helium was not eliminated until the fourth day. In general I have found that the helium disappeared long before the X, gas. Thus a piece of old lead I bombarded gave off appreciable quantities of helium from the first day's bombardment, very little on the second day, and none that I could detect on the third or subsequent days. The X,, on the other hand, came off in considerable quantities up to the end of the experi ment, which lasted for six days. I attribute the superior elimination of X, in the case of the platinum foil to the fact that during the whole time the bom bardment was concentrated on a patch only about 2 mm. in diameter, while the lead melted under the bombardment, so that fresh portions were continually being exposed to the rays. A piece of Kahlbaum's chemically pure lead gave appreciable amounts of X, and helium, though not nearly so much as the old lead. I tried some lead which had just been precipitated, but could not detect either X, or helium.

In the course of the experiments with old lead I let hydrogen into the experimenting chamber to see if it would increase the amount of X, but could not detect any effect. On one occasion I let in oxygen when nickel was bombarded, also without any appre ciable effect. I think these experiments are in favour of the view that these gases are present in the metal independently of the bombardment, and are liberated by the action of the kathode rays. They are surprisingly firmly held by the metal, and cannot, so far as my experience goes, be got rid of by heating. I kept a piece of lead in a quartz tube boiling in a vacuum for three or four hours, until all but a quarter of the lead had boiled away, and examined the gases given off during this process; neither X, nor helium could be detected. I then took the quarter that remained and bombarded it, and got appreciable amounts of X, and helium. On a second bombardment the X, was visible but the helium had disappeared. As an instance of the way these gases can stick to metals even when in solution or chemical combination, I may mention that though, as I have said, platinum foil after long exposure to kathode rays is freed from these gases, yet I got appreciable quantities of X, and helium, though no neon from platinum sponge freshly prepared from platinic chloride.

The reason helium is obtained by heating the glass of old Röntgen-ray bulbs is, I think, that after liberation by the kathode rays, the helium either adheres to the surface or is absorbed in a much looser way than before it was liberated. The question as to how these gases get into the metals is a most interesting one; are they absorbed in the process of manufacture?

note

In this connection it is interesting to that X, does not appear to occur to any appreciable extent in the atmosphere. Sometimes when suffering from the difficulty of clearing out these gases I have been goaded into speculating whether they do not represent the partially abortive attempts of ordinary metals to imitate the behaviour of radio-active substance; but whereas in these substances the a particles and the like are emitted with such velocity that they get clear away from the atom, in ordinary metals they have not sufficient energy to get clear, but cling to the outer parts of the atom, and have to be helped by the kathode rays to escape.

I would like to direct attention to the analogy between the effects just described and an everyday experience with discharge tubes-I mean the difficulty of getting these tubes free from hydrogen when the test is made by a sensitive method like that of the positive rays. Though you may heat the glass of the tube to melting point, may dry the gases by liquid air or cooled charcoal, and free the gases you let into the tube as carefully as you will from hydrogen, you will still get the hydrogen lines by the positive-ray method, even when the bulb has been running several hours a day for nearly a year. The only exception is when oxygen is kept continuously running through the tube, and this, I think, is due, not to lack of liberation of hydrogen, but to the oxygen combining with the small quantity of hydrogen liberated, just as it combines with the mercury vapour and causes the disappearance of the mercury lines. I think this production of hydrogen in the tube is quite analogous to the production of X,, of helium, and of neon. I have been greatly assisted in the experiments I have described by Mr. F. W. Aston, Trinity College, and Mr. E. Everett. J. J. THOMSON.

February 8.

The Water-surface "Halo."

THE "halo" which a happy memory of eighty years enables the Rev. O. Fisher to recall in NATURE of February 6 was probably one to which the explanation offered by Dr. Franklin Parsons does not apply.

There is a very striking phenomenon of separate rays or shafts of light converging on the shadow of the observer's head when this shadow is thrown on water. The phenomenon requires for its production certain conditions:-(1) A bright sun, high in a clear sky. For this reason in these latitudes the appearance is best seen about midday in summer. In winter it is scarcely noticeable. (2) The water must not be quite clear; on the other hand it must not be very turbid. (3) The surface must not be smooth, but may be fairly briskly agitated, but again not too briskly. (4) The water should be deep.

If any one of these conditions is absent the phenomenon is not seen, or is only imperfectly seen, as I was able to satisfy myself about twenty-five years ago by observations made, day after day, on the lake of Ullswater, where a stream discharged the muddy water of a mine far into the lake, and thus provided one of the necessary factors of variation. The necessity of these conditions, when once discovered, makes the explanation easy. The irregular convexities of the ruffled surface acting as condensing lenses separate the light penetrating the water into converging shafts. Along certain lengths of each or many of these shafts a sufficient condensation of light takes place to render them visible by means of the additional illumination of the slight turbidity. Thus the water is filled with luminous parallel shafts of varying lengths, which,

seen in perspective, have their vanishing point in the shadow of the observer's head. I remember that it was long before I realised that the rays were below and not on the surface. When the observer's head is not many feet above the water the rays may be traced to great distances-50 or 60 degrees-from the shadow of the head.

The phenomenon, though often very brilliant, is often unnoticed, even by good observers-I think because it requires a certain comprehensive glance, no doubt in the first instance accidental, to recognise that the widely separated broken radiations belong to a single convergent system. But when this system has once been realised it becomes hauntingly present, and one glimpses portions of it at every glance at the water, even though the shadow of the head is cut off from the surface. A. M. WORTHINGTON. Exmouth, February 9.

An X-Ray Fringe System.

By allowing a diverging pencil of Röntgen radiation to fall at nearly grazing incidence on one of the sets of cleavage planes of a crystal of rock-salt, and observing the intensity of the reflected pencil by a photographic plate, we find a series of well-marked and equal-spaced maxima in positions corresponding to equal increments of cos 0, where is the angle of incidence of radiation on the cleavage planes. In the directly transmitted beam there is no indication of variation of intensity with angle of incidence We thus have what appears to be a series of X-ray spectra of different orders, due to agreement in phase of waves from successive layers of molecules. Calculating on this assumption we get a wave-length of the order of magnitude in agreement with that calculated from the velocity of ejection of electrons by a substance exposed to this particular radiation-that is, assuming the results of the experiments of A. L. Hughes and others on ultra-violet light are equally applicable to Röntgen radiation. While only few experiments have yet been made on which to base any interpretation, this is in agreement with what we have already observed. Of the experimental results there is no doubt, and we cannot at present suggest any probable explanation except the very obvious one of interference. Further experiments are in progress. C. G. BARKLA.

King's College, London. February 11.

G. H. MARTYN.

Atmospheric Potential.

He

IN NATURE of December 12, 1912 (p. 411), Dr. George C. Simpson directs attention to several outstanding problems in atmospheric electricity. says, inter alia: "Everywhere it has been found that the air is a conductor, and that the potential gradient is practically the same." It is not the object here to consider these statements, however questionable.

The potential gradient of the atmosphere is the difference of electric potential between two points in the same vertical one metre apart; which, for the first few kilometres above the earth's surface, is about 100 volts.

Now one problem which Dr. Simpson does not mention is the absence of current from the upper regions of the atmosphere to the lower corresponding to this difference of potential between them. It is a fundamental law of electricity that an electric current will flow in a conductor from a high potential to a lower

one.

A conductor projecting vertically from the earth's