The possible ways in which resistance may arise are, I think, all be included under three heads :

(a) It may exist in the gas itself, to a greater than molecular stance from the point. For this the gas must become possessed something in the nature of structure round about the point , cease to be gas in the ordinary sense of the word-and the ly kind of structure that suggests itself as likely is that of the larised Grotthuss chains, which Prof. J. J. Thomson has used ath such effect in his beautiful theory of striæ.

(b) It may exist at the surface of the metal only, and consist in e tearing away from the point by electrical force of electrified articles which are clinging to it (a special case of which is the rcing of charged particles through a non-conducting layer on the int. (c) Or it may mean the pulling off of something in the nature of film possessing surface-tension.

As a matter of fact (b) and (c) are two opposite extremes of the me phenomenon. In both, particles adhering to the point are ulled off, but the adhesion of (b) is normal to the surface only e., between the particles and the metal-while in (c) it is angential to the surface only-i.e., between the particles themelves. An actual case would lie somewhere between the two, and robably much nearer to (b) than to (c).

Now in the case (a) of Grotthuss chains the field, f, at the point, s measured by the pull upon it, should increase as the sharpness f the point increases; and for two reasons. In the first place, or chains of given length there will be some average strength of ield required to break them, the value of which will always be ess than that measured at the surface of the point on account of he divergence of the lines of force there; and the difference between the measured and the average field will be more marked s the divergence-i.e., the sharpness of the point-increases. Secondly, the chains will be shorter at a sharp point than at a olunt one on account of this same divergence; and this again will necessitate a stronger field to break them, for much the same reason that a short piece of iron is harder to magnetise than a long one.

The same sort of variation of ƒ with sharpness of point will occur in the case of (c), if the mechanical pull of the field on the point-surface be looked on as having to tear off a film stretched over it. But to pull small particles off a point whose radius of curvature is large compared with their diameters will always require the same value of f. Hence (b) differs from (a) and (c) in requiring that ƒ shall be independent of r.

The effect on ƒ of altering r was therefore investigated. Steel needles were ground by a watchmaker to hemispherical points of different diameters and burnished with a hard steeltool. The values of ƒ in air were then determined for each at various pressures; the instrument used being the smaller of the two described in § 8. Curves connecting air-pressure, P, and were then plotted, and from these the numbers given in Table II. were taken. The values of ƒ for positive discharge are given for three pressures-76, 40, and 20 centimetres of mercury. Those for negative discharge are only given as 20 centimetres. The reason being that the latter show a good deal of irregularity at higher pressures, seeming, as will be seen later, to depend largely on the condition of the point-surface, which the positive values do not. The values of ƒ for needle C are inserted with queries, as they are almost certainly too high. The extreme fineness of its point necessitated a very rapid tapering to the sharpest part, so that a good deal of the measured pull must have been due to lines of force starting from its sides. This is borne out by the numbers in the four last columns.

Now, these columns show that below a point radius of millimetre f varies in close proportion with 08, and they consequently negative hypothesis (b). But they also negative (c), for the pull per square centimetre normal to a film of surface-tension, T, and radius of curvature r necessary to break it is proportional to T/r, and if this pull is to be supplied by fit follows that f2 must be proportional to T/r, which gives -0.5 instead of -0.8 as the

power of r with which ƒ varies. Even -0.5 is higher than can be assumed, for it is calculated on the supposition that there is no cohesion between the film and the point, and the existence of cohesion would still further reduce the power of r, as tending to bring the film nearer to class (b).

Resistance to discharge at a point is thus to be found in the surrounding gas, and is, therefore, practically reduced to the breaking down of Grotthuss chains, the length of which are not negligible compared with the radius of curvature of the point. I do not mean by this to exclude surface-resistance as a possible factor in discharge-indeed, what follows shows that it may be very appreciable; but it is not the whole, nor, I believe, an important part, of the resistance at a clean point.

§ 5. After a point has been used for discharge for some time its resistance increases greatly, and when discharge occurs it begins with a suddenness and violence which is very suggestive of the bursting of a film. The effect is increased if the point be now reburnished (without regrinding), though this process makes it look as if it had never been used. For instance, Curves II. are those connecting the pressures of the air with f for needle H. The continuous curve AB represents ƒ at the commencement of positive discharge, C D being for negative. The readings for these two curves were taken alternately, two at a time, the pressure being gradually increased. They correspond to an unused clean needle. The cloud of points (0 for positive and x for negative discharge) was obtained on attempting to repeat the curves, and it was noticeable that each repetition increased the values off, though the needle was well polished on dry washleather and rouge each time. Afterwards the needle was reburnished, and the dotted curves EF (+) and G H (-) obtained, showing a still further increase of ƒ and no less irregularity in its

values than before.

The above, therefore, furnishes evidence of the growth of resistance at a point where it is used. That the current does not permanently clear the formation away in getting through is obvious from the readings; but if a reading be taken quickly after another with a point in this condition, the second value of ƒ is often slightly lower than the first; as if the film had not had time to finish forming again after the first discharge. Now if this were so it would follow that the value of f, when the current stops, should be less than when it starts (for a used point only; the two values agree closely for a clean one). To test this, needle H was again taken and the values of ƒ determined for various air-pressures at the point when the current just stopped, the potential of the needle being gradually diminished until a high-resistance (7,000 w) Elliot galvanometer between the needle and the earth stood at zero. The results are given in Curves III. A B and C D for the unused point are repeated from Curves II. for comparison. JK and LM are the new curves. They are marked by circles. The agreement between J K and A B could not be closer, and points unmistakably to a temporary cleaning of the point while current is flowing. LM, for negative discharge, is very curious-coinciding as it does with CD at low pressures, while it leaves it so completely higher up. It was noticeable that just in proportion as the two curves diverged, the stoppage of the current was marked by increasing suddenness. At high pressures it was impossible to get the needle to discharge at all with a small current; either the flow was strong or it stopped altogether with a jerkthe same effect, though almost microscopic, being just perceptible at the cessation of positive discharge also. This being so, it follows that the values of ƒ on the negative curve really correspond to a strong current and not to the point where the current stops. Hence their greater values. The sudden cutting off of the negative current was very striking, and sug. gested rather forcibly the covering over of the point by some sort of film that had been perforated by it. It is true that the positive discharge was hardly affected, but this is only an example of what has been already alluded to-the relatively greater instability of negative than of positive discharge.

(To be continued.)

neither edifying or important, since circulars cóntaining an inaccurate statement concerning us have been printed and distributed, we feel bound to take some notice.

We, in common, we presume, with the rest of the exhibitors of the recent St. Pancras Electrical Exhibition, have received from Mr. Eccleston Gibb, the treasurer, a balance-sheet, accompanied by a circular, concluding thus: "If Messrs. Potter and Joel pay the balance still due from them-viz., £4. 10s. 10d.-this will be distributed at a future date."

The facts are as follows: We refused to pay our account in full, pointing out that, owing to the telephone exhibit having been placed in the ante-room which formed the sole means of access to and egress from our stand for the first week of the exhibition (especially during the evenings), visitors were quite unable to reach our exhibit, the throng in the telephone-room being so dense.

Any of the exhibitors will corroborate this, and it was fully recognised at the time by the committee, the telephones being ultimately removed to a corridor on the ground floor.

Needless to say, the success of our exhibit was seriously marred by this blunder on the part of those responsible for the arrangements.

On June 5 last we wrote to Mr. Eccleston Gibb regretting that a question had arisen with reference to our account and enclosing cheque in settlement, which cheque was accepted, and until the receipt of the circular in question we were under the impression that the matter was settled.

We think that the publication of such a statement as that complained of show very questionable taste on the part of Mr. Gibb, or whoever may be responsible, and, in conclusion, we beg to state that we do not for a moment admit that any balance is due from us, nor have we any intention of making any further payment.-Yours, etc.,

HENRY F. JOEL AND Co.

31, Wilson-street, E.C., August 26, 1891.

ELECTRICAL TRADES SECTION.—FRANKFORT

EXHIBITION.

SIR, The sub-committee recommends that gentlemen joining the party in connection with the Electrical Trades Section of the London Chamber of Commerce should travel via Dover and Ostend, leaving Holborn Viaduct Station by the train at 10 o'clock a.m. on Saturday, the 5th September. Each member of the party should provide himself with the necessary travelling ticket, and it is recommended that such tickets be obtained through Messrs. Cook and Son, Ludgate-circus, E.C., the fare being, vid Dover and Ostend, £5. 16s. first-class return, available 30 days. The ticket should be taken on or before the 2nd September, and in the event of this being done, Messrs. Cook and Son will engage reserved railway carriages between London and Frankfort, and members are requested to specify that they belong to the Electrical Trades Section party. Members are further notified that ladies may be included in the party. With regard to hotel accommodation the sub-committee could not well engage definitely sufficient rooms beforehand for the entire party, but the committee suggests to members to adhere to the following suggestions, in which case they will no doubt be suited for rooms. They can either buy coupons for rooms and dinners, through Messrs. Cook and Sons, for the Swan Hotel, Frankfort, and if such coupons are taken at the same time as the ticket, Messrs. Cook will take steps to reserve sufficient and good accommodation. Or members of the party can secure rooms at the Britannia Hotel, or Hotel Romscher Kaiser, if they write to the proprietors about a week beforehand, mentioning that they belong the Electrical Trades Section party.

The committee is in communication with the president of the Electro-Technical Society of Frankfort with respect to tickets for the various events in connection with the exhibition and congress, and gentlemen will be advised, if necessary, as to what special arrangements can be made in this respect. Yours, etc.,

KENRIC B. MURRAY, Secretary.

THE BRITISH ASSOCIATION AT CARDIFF.

PRESIDENTIAL ADDRESS BY WILLIAM HUGGINS, Esq., D.C.L. (OXON.), LL D. (CANTAB., EDIN., ET DUBLIN.), PH.D. (LUGD. BAT.), F.R.S., F.R.A.S., HON. F.R.S. E., ETC., CORRESPONDANT DE L'INSTITUT DE FRANCE.

(Continued from page 186.)

Prof. Rowland looks to the solar lines which are unaccounted for as a means of enabling him to discover such new terrestrial elements as still lurk in rare minerals and earths, by confronting their spectra directly with that of the sun. He has already resolved yttrium spectroscopically into three components, and actually into two. The comparison of the results of this independent analytical method with the remarkable but different conclusions to which M. Lecoq de Boisbaudran and Mr. Crookes have been led respectively, from spectroscopic observation of these bodies when glowing under molecular bombardment in a vacuum tube, will be awaited with much interest. It is worthy of remark that as our knowledge of the spectrum of hydrogen in its complete form came to us from the stars, it is now from the sun that chemistry is probably about to be enriched by the discovery of new elements.

In a discussion in the Bakerian lecture for 1885 of what we knew

up to that time of the sun's corona, I was led to the conclusion that the corona is essentially a phenomenon similar in the cause of its formation to the tails of comets-namely, that it consists for the most part probably of matter going from the sun under the action of a force, possibly electrical, which varies as the surface, and can therefore in the case of highly-attenuated matter easily the coronal particles may return to the sun, those which form the master the force of gravity even near the sun. Though many of long rays or streamers do not return; they separate and soon become too diffused to be any longer visible, and may well go to furnish the matter of the zodiacal light, which otherwise has not received a satisfactory explanation. And further, if such a force exist at the sun, the changes of terrestrial magnetism may be due to direct electric action, as the earth moves through lines of inductive force.

These conclusions appear to be in accordance broadly with the subsequent eclipses. Prof. Schuster takes an essentially similar lines along which thought has been directed by the results of view, and suggests that there may be a direct electric connection between the sun and the planets. He asks further whether the sun may not act like a magnet in consequence of its revolution about its axis. Prof. Bigelow has recently treated the coronal forms by the theory of spherical harmonics, on the supposition that we see phenomena similar to those of free electricity, the rays being lines of force, and the coronal matter discharged from the sun, or at least arranged or controlled by these forces. At the extremities of the streams, for some reasons, the repulsive power may be lost, and gravitation set in, bringing the matter back to the sun. The matter which does leave the sun is persistently transported to the equatorial plane of the corona. In fact, the zodiacal light may be the accumulation at great distances from the sun along this equator of such like material. Photographs on a larger scale will be desirable for the full development of the conclusions which may follow from this study of the curved forms of the coronal structure. Prof. Schaeberle, however, considers that the coronal phenomena may be satisfactorily accounted for on the supposition that the corona is formed of streams of matter ejected mainly from the spot zones with great initial velocities, but smaller than 382 miles a second. Further, that the different types of the corona are due to the effects of perspective on the streams from the earth's place at the time relatively to the plane of the solar equator.

Of the physical and the chemical nature of the coronal matter we know very little. Schuster concludes, from an examination of the eclipses of 1882, 1883, and 1886, that the continuous spectrum of the corona of the corona has the maximum of actinic intensity displaced considerably towards the red when compared with the spectrum of the sun, which shows that it can only be due in small part to solar light scattered by small particles. The lines of calcium and of hydrogen do not appear to form part of the normal spectrum of the corona. The green coronal line has no known representative in terrestrial substances, nor has Schuster been able to recognise any of our elements in the other lines of the corona.

The spectra of the stars are almost infinitely diversified, yet they can be arranged with some exceptions in a series in which the adjacent spectra, especially in the photographic region, are scarcely distinguishable, passing from the bluish-white stars, like Sirius, through stars more or less solar in character, to stars with banded spectra, which divide themselves into two apparently independent groups, according as the stronger edge of the bands is towards the red or the blue. In such an arrangement the sun's place is towards

the middle of the series.

At present a difference of opinion exists as to the direction in the series in which evolution is proceeding, whether by further condensation white stars pass into the orange and red stages, or whether these more coloured stars are younger and will become white by increasing age. The latter view was suggested by John. stone Stoney in 1867.

About 10 years ago Ritter, in a series of papers, discussed the behaviour of gaseous masses during condensation, and the probable resulting constitution of the heavenly bodies. According to him a star passes through the orange and red stages twice, first during a comparatively short period of increasing temperature which culminates in the white stage, and a second time during a more prolonged stage of gradual cooling. He suggested that the two groups

of banded stars may correspond to these different periods: the young stars being those in which the stronger edge of the dark band is towards the blue, the other banded stars, which are relatively less luminous and few in number, being those which are approaching extinction through age.

Recently, a similar evolutional order has been suggested, which is based upon the hypothesis that the nebulæ and stars consist of colliding meteoric stones in different stages of condensation. More recently the view has been put forward that the diversified spectra of the stars do not represent the stages of an evolutional progress, but are due for the most part to differences of original constitution. The few minutes which can be given to this part of the address are insufficient for a discussion of these different views. I purpose, therefore, to state briefly, and with reserve as the subject is obscure, some of the considerations from the characters of their spectra which appeared to me to be in favour of the evolutional order in which I arranged the stars from their photographic spectra in 1879. This order is essentially the same as Vogel had previously proposed in his classification of the stars in 1874, in which the white stars, which are most numerous, represent the early adult and most persistent stage of stellar life, the solar condition that of full maturity and of commencing age; while in the orange and red stars with banded spectra we see the setting in and advance of old age. But this statement must be taken broadly, and not as asserting that all stars, however different in mass and possibly to some small extent in original constitution, exhibit one invariable succession of spectra.

In the spectra of the white stars the dark metallic lines are relatively inconspicuous, and occasionally absent, at the same time that the dark lines of hydrogen are usually strong, and more or less broad, upon a continuous spectrum, which is remarkable for its brilliancy at the blue end. In some of these stars the hydrogen and some other lines are bright, and sometimes variable.

As the greater or less prominence of the hydrogen lines, dark or bright, is characteristic of the white stars as a class, and diminishes gradually with the incoming and increase in strength of the other lines, we are probably justified in regarding it as due to some conditions which occur naturally during the progress of stellar life, and not to a peculiarity of original constitution.

Con

To produce a strong absorption spectrum a substance must be at the particular temperature at which it is notably absorptive; and, further, this temperature must be sufficiently below that of the region behind from which the light comes for the gas to appear, so far as its special rays are concerned, as darkness upon it. sidering the high temperature to which hydrogen must be raised before it can show its characteristic emission and absorption, we shall probably be right in attributing the relative feebleness or absence of the other lines, not to the paucity of the metallic vapours, but rather to their being so hot relatively to the substances behind them as to show feebly, if at all, by reversion. Such a state of things would more probably be found, it seems to me, in conditions anterior to the solar stage. A considerable cooling of the sun would probably give rise to banded spectra due to compounds, or to more complex molecules, which might form near the condensing points of the vapours.

The sun and stars are generally regarded as consisting of glowing vapours surrounded by a photosphere where condensation is taking place, the temperature of the photospheric layer from which the greater part of the radiation comes being constantly renewed from the hotter matter within.

At the surface the convection currents would be strong, producing a considerable commotion, by which the different gases would be mixed and not allowed to retain the inequality of proportions at different levels due to their vapour densities.

Now the conditions of the radiating photosphere and those of the gases above it, on which the character of the spectrum of a star depends, will be determined, not only by temperature, but also by the force of gravity in these regions; this force will be fixed by the star's mass and its stage of condensation, and will become greater as the star continues to condense.

In the case of the sun the force of gravity has already become so great at the surface that the decrease of the density of the gases must be extremely rapid, passing in the space of a few miles from atmospheric pressure to a density infinitesimally small; consequently the temperature gradient at the surface, if determined solely by expansion, must be extremely rapid. The gases here, however, are exposed to the fierce radiation of the sun, and unless wholly transparent would take up heat, especially if any solid or liquid particles were present from condensation or convection

Currents.

From these causes, within a very small extent of space at the surface of the sun, all bodies with which we are acquainted should fall to a condition in which the extremely tenuous gas could no longer give a visible spectrum. The insignificance of the angle subtended by this space as seen from the earth should cause the boundary of the solar atmosphere to appear defined. If the boundary which we see be that of the sun proper, the matter above it will have to be regarded as in an essentially dynamical condition-an assemblage, so to speak, of gaseous projectiles for the most part falling back upon the sun after a greater or less range of flight. But in any case it is within a space of relatively small extent in the sun, and probably in the other solar stars, that the reversion which is manifested by dark lines is to be regarded as taking place.

Passing backward in the star's life, we should find a gradual weakening of gravity at the surface, a reduction of the temperature gradient so far as it was determined by expansion, and convection currents of less violence producing less interference with the proportional quantities of gases due to their vapour densities, while the effects of eruptions would be more extensive,

At last we might come to a state of things in which, if the star were hot enough, only hydrogen might be sufficiently cool relatively to the radiation behind to produce a strong absorption. The lower vapours would be protected, and might continue to be relatively too hot for their lines to appear very dark upon the continuous spectrum; besides, their lines might be possibly to some extent effaced by the coming in under such conditions in the vapours themselves of a continuous spectrum.

In such a star the light radiated towards the upper part of the atmosphere may have come from portions lower down of the atmosphere itself, or at least from parts not greatly hotter. There may be no such great difference of temperature of the low and less low portions of the star's atmosphere as to make the darkening effect of absorption of the protected metallic vapours to prevail over the illuminating effect of their emission.

It is only by a vibratory motion corresponding to a very high temperature that the bright lines of the first spectrum of hydrogen can be brought out, and by the equivalence of absorbing and emitting power that the corresponding spectrum of absorption should be produced; yet for a strong absorption to show itself, the hydrogen must be cool relatively to the source of radiation behind it, whether this be condensed particles or gas. Such conditions, it seems to me, should occur in the earlier rather than in the more advanced stages of condensation.

The subject is obscure, and we may go wrong in our mode of conceiving of the probable progress of events, but there can be no doubt that in one remarkable instance the white-star spectrum is associated with an early stage of condensation.

Sirius is one of the most conspicuous examples of one type of this class of stars. Photometric observations combined with its ascertained parallax show that this star emits from 40 to 60 times the light of the sun, even to the eye, which is insensible to ultraviolet light, in which Sirius is very rich, while we learn from the motion of its companion that its mass is not much more than double that of our sun. It follows that unless we attribute to this star an improbably great emissive power, it must be of immense size, and in a much more diffuse and therefore an earlier condition than our sun; though probably at a later stage than those white stars in which the hydrogen lines are bright.

A direct determination of the relative temperature of the photospheres of the stars might possibly be obtained in some cases from the relative position of maximum radiation of their continuous spectra. Langley has shown that through the whole range of temperature on which we can experiment, and presumably at temperatures beyond, the maximum of radiation power in solid bodies gradually shifts upwards in the spectrum from the infra-red through the red and orange, and that in the sun it has reached the blue.

The defined character as a rule of the stellar lines of absorption suggests that the vapours producing them do not at the same time exert any strong power of general absorption. Consequently we should probably not go far wrong, when the photosphere consists of liquid or solid particles, if we could compare select parts of the continuous spectrum between the stronger lines or where they are fewest. It is obvious that if extended portions of different stellar spectra were compared, their true relation would be obscured by the line-absorption.

The increase of temperature, as shown by the rise in the spectrum of the maximum of radiation, may not always be accompanied by a corresponding greater brightness of a star as estimated by the eye, which is an extremely imperfect photometric instrument. Not only is the eye blind to large regions of radiation, but even for the small range of light that we can see the visual effect varies enormously with its colour. According to Prof. Langley, the same amount of energy which just enables us to perceive light in the crimson at A would in the green produce a visual effect 100,000 times greater. In the violet the proportional effect would be 1,600, in the blue 62,000, in the yellow 28,000, in the orange 14,000, and in the red 1,200. Captain Abney's recent experiments make the sensitiveness of the eye for the green near F to be 750 times greater than for red about C. It is for this reason, at least in part, that I suggested in 1864, and have since shown by direct observation that the spectrum of the nebula in Andromeda, and presumably of similar nebulæ, is in appearance only wanting in the red.

The stage at which the maximum radiation is in the green, corresponding to the eye's greatest sensitiveness, would be that in which it could be most favourably measured by eye photometry. As the maximum rose into the violet and beyond, the star would increase in visual brightness, but not in proportion to the increase of energy radiated by it.

The brightness of a star would be affected by the nature of the substance by which the light was chiefly emitted. In the laboratory solid carbon exhibits the highest emissive power. A stellar stage in which radiation comes, to a large extent, from a photosphere of the solid particles of this substance would be favourable for great brilliancy. Though the stars are built up of matter essentially similar to that of the sun, it does not follow that the proportion of the different elements is everywhere the same. It may be that the substances condensed in the photospheres of different stars may differ in their emissive powers, but probably not to a great extent.

All the heavenly bodies are seen by us through the tinted medium of our atmosphere. According to Langley, the solar stage of stars is not really yellow, but even as gauged by our imperfect eyes, would appear bluish-white if we could free ourselves from the deceptive influences of our surroundings.

From these considerations it follows that we can scarcely infer the evolutional stages of the stars from a simple comparison of their eye-magnitudes. We should expect the white stars to be, as a class, less dense than the stars in the solar stage. As great mass might bring in the solar type of spectrum at a relatively