to the telephone was determined by extrapolation from measurements on stronger currents with the thermogalvanometer.

The results of experiments on coherers made of oxidised iron wire dipping into mercury, and coherers made of a clean iron point touching an oxidised iron plate, are exhibited as curves connecting-(1) the steadily applied E.M.F. and consequent current through the coherer; (2) the steadily applied E.M.F. and the power given to the telephone, for various rates of delivery of vibration energy to the detector; (3) the power delivered to the detector and the power passed to the telephone, the E.M.F. applied to the coherer being constant. Curves I show that in a self-restoring coherer the current increases more and more rapidly as the E.M.F. is raised, till, in general, a point of inflexion is reached, and then the current increases more slowly. Curves 2 show the rise and fall of sensitiveness to oscillations as the applied E.M.F is increased. Curves 3 show that if W represents the power in watts delivered to the coherer, and w the power passed to the telephone circuit, then w=m(W a) where m and a have values settled by the magnitude of the current through the detector. The quantity m for a good low resistance iron-mercury coherer has been found to be as high as o'06; while a is usually near 10x10-8 watt. These curves show that these coherers are not "voltage-operated" detectors but "integrating" detectors.

The author puts forward the hypothesis that the properties of an oxide coherer may arise solely from the temperature variations caused in the minute mass of oxide at the contact by the electrical oscillations and by the applied E.M.F. He examines the hypothesis mathematically, and shows that most of the phenomena recorded in the curves 1, 2, 3 above, can in this way be accounted for as perfectly as the present state of the measurements permits.

Mr. W. DUDDELL expressed his interest in the paper, and remarked that it was important to know how much energy was needed to work a detector in order to calculate the amount which it was necessary to radiate. He asked the author if the action of the coherer he had described might not depend upon compressions due to electrostatic

attractions.

Prof. C. H. LEES remarked that if the action of the coherer depended upon thermal effects, the sensitiveness would be affected by using materials of different thermal conductivities, whereas if the effects were to be attributed to compressions, change of material would have little effect upon the sensitiveness.

Mr. RAYNER pointed out that if the effect was due to the temperature coefficient of the resistance of the film, the sensitiveness should depend upon the temperature at which the coherer was worked. He asked if Dr. Eccles had conducted experiments at different temperatures, and if he could give any particulars as to how the sensitiveness of the coherer he had used compared with other types.

Dr. RUSSELL suggested that as the thickness of the film of oxide on the iron plate was only about a micron, the❘ electrostatic attraction between the needle and the iron plate would be appreciable even at very low voltages. It was conceivable that, owing to the compression, heat was generated in the film at this point.

Mr. CAMPBELL remarked that in accurate work there would be no trouble in measuring the mutual inductances, but the determination of the damping factors would be difficult.

Dr. ECCLES, in reply to Mr. Duddell, said that the hypothesis that electrostatic attractions between the two conductors separated by the film of oxide caused the change of conductivity produced by electrical oscillations, was negatived by the fact that in such materials as the author had tried the magnitude of the coefficient of decrease of resistance with temperature, and the thermoelectric properties of the two conductors had a great deal to do with the sensitiveness of the detector. In reply to Prof. Lees, no doubt the sensitiveness of a detector would, if the thermal theory were correct, depend upon the slowness with which

the heat was conducted away by the metal near the contact. This consideration suggested one reason why some of the badly conducting oxides formed such good detectors. With regard to Mr. Rayner's observations he had not attempted experiments at other than normal temperatures. The author, replying to Dr. Russell, said the answer to Mr. Duddell was circumvented by his suggestion that the electrostatic attractions caused compressions which, on the whole, produced local heating. But it might be pointed out that if in the secondary circuit the capacity was changed to another value and the inductance altered to keep the period of the circuit the same, the oscillating voltage at the coherer was different, yet no difference worth mentioning was heard in the detector. In answer to Mr. Campbell, no doubt the mutual inductance could be measured accurately given proper experimental facilities. Since, however, another course of experiments was adopted, one that did not need an accurate value of the mutual inductance between primary and secondary, this measurement was not attempted accurately.

A paper entitled "Earth-air Electric Currents" was read by Mr. G. C. SIMPSON.

The paper describes a method for automatically recording the electrical current which passes from the earth into the air during periods of fine weather.

A large plate (17 metres) was placed in the open as tion, this was then connected to an insulated vessel from near to the ground as was consistent with efficient insulawhich water issued through an orifice surrounded by an the insulated vessel carried away, by the well known earth-connected cylinder. The water as it dropped from "collector" action, all the charge which the exposed plate received and the latter remained at zero potential. The charged water drops were collected in a vessel connected to a self-registering electrometer which was earth-connected for an instant at the end of every two minutes. The paper describes the sources of error and the method of deter-. mining the value of the earth-air current, and of the conductivity of the air from the records of the electrometer. The method was used in Simla (India), but owing to the impossibility of obtaining a site on which the normal electrical field of the atmosphere was undisturbed by the surrounding hills and trees, the absolute values obtained were uncertain, but it is very probable that the daily range of the different factors were near approximations to the truth.

Dr. C. CHREE expressed his interest in the paper, and remarked that they had worked at Kew with C. T. R. Wilson's apparatus. It would be interesting to know whether the same results could be obtained from the author's and from Wilson's apparatus. He drew attention to the difficulty of insulating a piece of the earth in England sufficiently well to obtain reliable results.

Dr. RUSSELL said that C. T. R. Wilson had found that the mean earth leakage current in his experiments was about 2'2 × 10--16 ampères per square centimetre. It was therefore exceedingly small.

Linke had shown from the results obtained in balloon ascents that the atmospheric potential gradient diminished regularly up to a height of 6000 metres, and probably much higher. Assuming that air has "conductivity" this shows that the conductivity increases as we ascend. It seems reasonable to suppose that at great altitudes, notwithstanding the intense cold, the highly rarefied air is practi cally a conductor. The earth leakage currents carry charges to these conducting layers, and these charges are continually being returned to the earth by the storms always taking place somewhere or other in the world. The energy expended by the currents was probably due originally to the sun's heat vaporising and raising water to heights in the air. A surprising result obtained by Dr. Simpson was that the sunshine apparently caused a diminution in the conductivity of the air over the surface of the ground. It was customary to suppose that ultra-violet rays improve the conductivity of the atmosphere, and this supposition

VOL. CI., No. 2628.

RECENT CONTRIBUTIONS TO THEORIES REGARDING THE INTERNAL STRUCTURE OF THE EARTH.*

By Sir THOMAS H. HOLLAND, K.T.C.E., D.Sc., F.R.S.

THE intensity and quantity of polemical literature on scientific problems generally varies inversely as the number of direct observations on which the discussions are based; the number and variety of theories in general thus form a a coefficient of ignorance. Beyond the superficial observations direct and indirect made by geologists, not extending below about one two-hundredth of the Earth's radius, we have to trust to the deductions of mathematicians for our ideas regarding the centre of the earth. They have provided us successively with every permutation and combination possible of the three physical states of matter -solid, liquid, and gaseous. Starting, say, two centuries back with the astronomer Halley, geologists were presented wtih a globe whose shell rotated at a rate different to that of its core-in recent times this idea has been re

nitrogen and water vapour; but even now it is possibly not flying off into space. With the growth of the young globe the compression at the centre gave heat enough to melt the accumulated meteoritic matter, and the molten material reached the surface as lava, such volcanic action predominating at the surface till the atmosphere became charged with water, and the familiar processes of weathering started and formed the film of "rust " which geologists know as sedimentary rocks.

With this addition to the variegated array of theories about the physical condition of the earth and its genealogy, the scientific world began to settle down again into serenity, comforted by the feeling that at any rate all agree in the fact that the earth is now a gradually cooling body, with many millions of years still before it. Then came the discovery of radium, and with it the first assurance that geologists were right in claiming a long past to be followed by a longer future than the most optimistic philosopher has dared to predict with our apparently limited store of earth-heat. Now, however, Prof. Joly warns us that if the central parts of the globe contain anything near the quantity of radium found near the surface, we may even be going in the other direction, and that instead of peaceful cooling our descendants may have to face a catastrophic heating; the now inconspicuous little body known as the earth may indeed become famous through the universe as

a new star.

earth in which the heavier substances conformed to the

equatorial belt. Late in the seventies, Dr. Ritter started the idea of a gaseous core surrounded by a solid crust, and this was modified in 1900 by the Swedish philosopher Svante Arrhenius, whose globe with a solid crust, liquid substratum, and gaseous core is now a favourite among many geologists. Though the proportions of the layers differ, Arrhenius's globe and that of Dr. Wilde are in general agreement, and of the two the latter more nearly satisfies the simple deduction from the remarkable phenomena regarding the transmission of distant earthquake shocks noticed only four years ago by Mr. R. D. Oldham. But the variety of ideas does not end with theories on the present constitution of the globe. Poisson required the process of solidification to begin from the centre and to progress outwards, while other mathematicians had been happy with the Leibnitzian consistentior status as the first external slaggy crust. Since the days of Laplace all naturalists have been made to swallow the idea of a solar system formed by the cooling and condensation of a spheroidal gaseous nebula; and all except those geologists who have vainly searched for traces of the primeval crust have been happy in this belief. Now some American astronomers and geologists have combined to form the solar system by the aggregation of innumerable small bodies, "planetesimals," which gathered gradually into knots; thus, the earth grew gradually by accretion from quite a small body, and even now, although the process has nearly stopped, it receives much meteoric matter from outside. With this theory there must have been a time when the gravity of the earth was too small to hold an atmosphere of any but the heaviest gases, such as carbonic acid; later, it was heavy enough to retain oxygen, then

Abstract of the Wilde Lecture delivered before the Manchester Literary and Philosophical Society, March 22nd, 1910.

others, in spite of the armour-plating of physicists, have recently been riddled by the a-rays of Rutherford. Still some survive, battered and scarred, but nevertheless seaworthy; and with these well tried craft it is proposed to examine some recently acquired information.

So far as the centre of the earth is concerned the nearest approach to actual observation is due to recent developments of seismology. It has been shown that when a large earthquake occurs, two groups of vibrations are recorded on distant seismographs-those due to the vibrations that pass through the earth, and those that travel The vibrations that stick around by the superficial crust. to the crust travel at a fairly uniform rate of 3 kilometres, or 1.85 miles a second; those that pass through the earth have a greater speed, and the deeper they penetrate the earth the faster they travel. But Oldham has noticed that if that they penetrate to within about 1600 miles of the the second group of vibrations enter the earth so deeply centre, they become retarded by a core which is different in its physical characters to the rest of the earth.

It is possible that this is the gaseous core postulated by Ritter, Wilde, and Arrhenius. Oldham aptly remarks that the study of seismic waves thus promises to give us data of the kind that we now get from the spectroscope regarding the constitution of distant bodies that cannot be sampled physically.

With regard to problems nearer the crust, among the most promising results recently obtained are those which show the local variations in gravity. The remarkable work recently done in India, more especially by Colonel Burrard, has a direct bearing on geological problems regarding the state of affairs immediately below the surface; for the geodetic results can be correlated with the simple geoIn the first place, it has now logical history of the area. of gravity, but not so great as was assumed by Archdeacon been shown that under the Himalayas there is a deficiency

Pratt when he framed his famous theory regarding the irregularities of the earth's surface being coincident with the variations in the mass of solid matter below.

The deficiency of gravity under the Outer Himalayas is about equivalent to 3500 feet thick of solid rock; hence at stations 8000 feet above sea-level there must be a mass of rock some 4500 feet high maintained by the rigidity of the

earth's crust. But at the foot of the mountain range almost the same deficiency of gravity occurs, and it is towards this lighter area that the rocks are being thrust from the north. Further to the south, on the alluvial plains of the Ganges, there is found to be a band of high gravity parallel to the Himalayan range, and this presumably indicates the existence some thirty miles below the surface of a band or ridge of heavy basic rock. This heavy ridge, of which no other signs appear at the surface, deflects the plumb-line from its normal vertical position, and counteracts the attraction of the Himalayan mass at stations between it and the mountains. It was because the earlier observations (made between this unsuspected heavy band and the Himalayas) showed the mountains to have a small attraction that Pratt assumed the higher hills to be com pensated by deficient or light material below.

Thus

The geological history of the area gives a clue to the nature of this mysterious subterranean ridge. For geological ages before the Himalayas began to rise there was a great continent stretching away to the south from India to Africa, and an ocean then stretched across Europe and Central Asia, with its southern shore near the present line of Himalayan snowy peaks. For ages material was carried away by the rivers from this continent, and was deposited near the southern shore of the great Eurasian ocean. the continent gradually became unloaded, while the ocean bed near by was correspondingly weighted down; there was a gradual sagging down of the crust towards the north, with a stretching of the rocks on the continent. The evidence of stretching is now preserved as great fissures running roughly east and west, and of all ages varying from Lower Palæozoic times on to the end of the Mesozoic era, when apparently the strain had reached its limit, and the molten basaltic magma lying below welled out and flooded the country for over 200,000 square miles. Besides the masses of lava which flooded the surface, the subterranean fissures were injected with the basic magma, many of which have since been laid bare by erosion, while greater masses consolidated below are still concealed. It is probably these that lie below ground, between the lava flows and the foot of the Himalayas, which now disturb the plumb-line by their high gravity.

Soon after this catastrophe occurred, the enormous masses of sediment which had accumulated in the great Eurasian ocean having reached a depth sufficient with their store of radium to become softened, were buckled up to form the great plateau of Tibet. By the action of great thrusts from the north these are now being pushed over on to the plains of India, folding over like breakers on a sea-shore.

The lecturer then discussed the relation between the central granitic core of the Himalayan range and the basaltic magmas lying below. He concluded that the solid crust could not be more than about twenty-five miles thick, and that the variations in the distribution of gravity are due to great masses of heavy basic lava, which reach higher levels under conditions of strain like those developed on the old continent that existed before the Himalayas began to rise. The bearing of the facts on Dutton's theory of isostasy was discussed, as well as the evidences regarding the supposed permanence of the oceanic depressions and Prof. Love's recent mathematical examination of the figure of the earth.

Laundry Exhibition.-On April 2nd a Laundry Exhibition was opened at the Agricultural Hall, London, remaining open till Saturday, the 9th. In addition to the largest display of laundry machinery and appliances ever got together in this country, there was a competition for finished work, in which about fifty of the leading laundries of the country entered. It included a competition for finished work, and also a display of linen, &c., from the London County Council Technical Institutes and Schools.

A NEW METHOD OF

ESTIMATING CHLORATES IN THE PRESENCE
OF NITRATES AND CHLORIDEs.
(SUPPLEMENT).

By TARAK NATH DAS, B.Sc., F.C.S.,
Central Hindoo College, Benares City, India.

THE method has been suggested by the author in the CHEMICAL NEWS of January 28, 1910 (vol. ci., p. 38), of boiling a weighed quantity of KCIO3 with a known volume of SnCl2 solution and Na(OH) in excess. The introduction of iodine solution into the flask as there shown in figure (a) is inadmissible owing to the action of iodine on NaOH :2NaOH + I2 = NaI + NaIO + H20.

Apparatus (b) should be used with the side tube ending in a beaker as in (a). Caustic soda solution coloured with phenolphthalein is just neutralised with HCl from the dropping funnel after the boiling operation is finished. Aniline hydrochloride indicator must be omitted as it cannot act in an alkaline solution. Presence of free acid is likely to interfere owing to the liberation of nitric acid (from nitrates present), which will oxidise SnCl2, or, latterly, iodine. The former should therefore be made alkaline with NaHCO3 and rochelle salt (Clowes and Coleman, "Quantitative Analysis"), as no reducing action of these on chlorate is known under the conditions. Further experiments are in progress.

I have to express my thanks to Mr. G. S. Arundale, M.A., F.R.H.S., Principal and successor of my most esteemed Professor, Dr. Arthur Richardson, for allowing me to go on with original investigation in the Chemical Laboratory.

THE HALLEY COMET.

By C. J T. HANSSEN, C.E.

THE observations of the Halley comet which have been published show that the comet's orbit does cross the earth's orbit, but from the dates given it is evident that on April 20th the comet will be in its perihelium, and on May 18th just in the straight line between the earth and the sun, at a considerable and safe distance from the earth. A collision with the comet is not to be feared, as the earth at the end of April will have passed that point of its orbit where the comet will cross several weeks later, but on May 18th we shall, according to Prof. H. H. Turner, of Oxford University, pass right through the comet's tail, and the question is, therefore, what does the tail consist of? and can the tail do us any harm?

A comet, according to the best observations, consists of a solid core, surrounded by a cluster of stones or metallic balls of all sizes from a fraction of a pound to many tons in weight, and at such distances apart from each other that even small stars can be clearly seen through the whole nucleus of the comet; and the tail, which can be many millions of miles in length, is a swarm of similar stones, but at greater distances from each other than in the head of the comet. The comet and its tail is, in fact, a modified form of the showers of meteors which frequently pass through our atmosphere as star shots or fire balls, and of which many fall to the ground, as related in the Book of Joshua, x., ver. 11, and in other ancient and modern historical records. These meteoric globes have, of course, the low temperature of universal space, say, 2° to 3° Absolute, and where they on their way pass through upper regions of planetary atmospheres, they will condense and freeze portions of the gases they meet, and be covered with a crust of frozen hydrogen, oxygen, carbon, and other gases, which causes them in bright sunlight to shine like a very clear white cloud (Comet 1843, Astronomische Nachrichten, No. 491, p. 175). All astronomers agree