the exact correspondence of numbers of these lines, grouped in a complex manner (so as entirely to eliminate the possibility of a mere chance accordance) with the bright lines seen in the spectra of light from the vapours of those elements. When once Kirchhoff and Bunsen had proved the possibility of exhibiting the same set of lines either as bright lines on a dark ground or as dark lines on a brilliant spectrum, all doubt as to their meaning in the solar spectrum disappeared

at once.

It has been found that in the sun's atmosphere there are present the vapours of iron, copper, zinc, and nickel, besides calcium, magnesium, sodium, and other metals. But the vapours of tin, lead, silver, and gold, do not appear to be present in the solar atmosphere. One of the most remarkable dark lines is due to the presence of hydrogen.

But it has been found possible to extend these researches to the fixed stars. Mr. Huggins and Dr. Miller have done this successfully, and their discoveries afford a means of assigning very sufficient reasons for the colours of the brighter stars. By analogy also we may extend a similar interpretation to the colours of stars not bright enough to give a spectrum which can be satisfactorily examined.

Let us take first the brilliant Sirius. This star belongs to the southern half of the celestial sphere, and although it becomes visible at certain seasons in our latitude, it never rises very high above the horizon. In fact, at its highest,that is, when due south-it is only twenty-two degrees above the horizon, or less than one fourth of the way from the horizon to the point immediately over-head. This peculiarity somewhat interferes with the observation of the star by a method so delicate as that applied by the celebrated physicists we have named. On the other hand the exceeding

brilliancy of Sirius makes some amends for the effects of atmospheric disturbances. By selecting very favourable opportunities, Huggins and Miller were able to analyse the star's spectrum, with the following result:

The atmosphere around Sirius contains sodium, magnesium, hydrogen, and probably iron.

The whole spectrum is covered by a very large number of faint and fine lines, indicating a corresponding variety in the substances vaporised in the star's atmosphere.

The hydrogen lines are abnormally strong as compared with the solar spectrum, all the metallic lines being remarkably faint.

This last circumstance is well worthy of notice, since it is a peculiarity characteristic of white starsso that we begin already to find a hint respecting the source of colour or of the absence of colour in stars.

Take next an orange-red star, the brilliant Betelgeux. The spectrum of this star was very carefully analysed by Messrs. Huggins and Miller. They marked down the places of two or three hundred lines, and measured the position of no less than eighty. They found that sodium, magnesium, calcium, iron, and bismuth are present in the star's atmosphere, but the two strong lines which note the presence of hydrogen are wanting.

Take next the yellow star, Pollux. The observers were not able to obtain very satisfactory measures of this star; but they established the presence of sodium and magnesium in the star's atmosphere; and again the strong lines of hydrogen were found to be missing.

But we are not entitled to assume that red and yellow stars are characterised by the absence of hydrogen from their atmospheres. On the contrary, the noted red star Aldebaran, the spectrum of which was very carefully analysed by Huggins and Miller, was found to exhibit the two lines of hydrogen with perfect distinctness. This star exhibited a richness in the construction of its

atmosphere not presented by any other. The elements proved to be present are sodium, magnesium, calcium, iron, bismuth, tellurium, antimony, and mercury. It must not be supposed, in this or any other case, that other elements might not by a sufficiently laborious scrutiny be proved to exist in the star's atmosphere. The observations required, says Mr. Huggins, are extremely fatiguing to the eye, and necessarily limited to the stronger lines of each spectrum.'

It is clear, however, from the above short list of examples, that a considerable variety exists in the physical constitution of the fixed stars. This of itself affords a suggestive hint respecting the true explanation of the variety of colour which we have described. And the peculiarity that in the white stars the hydrogen lines are singularly strong, while the metallic lines are as singularly weak, is yet more to the point. Sirius was a red star. Was it at that time unlike present red stars? Does it not seem more probable that, if there had existed in those days a Huggins or a Miller, and the instruments used so successfully by these observers had been invented, it would have been found that Sirius did not-when a red star-present peculiarities now observed only in white stars?

We recognise, then, the influence of time upon the spectrum of this celebrated star, as probably tending to render the lines of hydrogen more distinct than of yore, and the lines of the metallic elements less distinct. But what is the meaning of such a change? Suppose a chemist, for example, observing the spectrum of the flame produced by the combustion of a compound body, should notice that the lines of some elements slowly increased in distinctness, while the lines of others grew fainter, how would he interpret such a phenomenon? If we remembered only that the dark lines

are due to the absorptive effect of the vapour they correspond to, on light which is trying, so to speak, to pass through the vapour, we might readily jump at a conclusion, and answer that the extent of absorptive vapour is increasing when the lines are growing more distinct, and vice versa. But we must also consider that these lines are partly the effect of contrast. The limelight held before the sun's disc appears black, though so dazzling when seen alone. It may be, therefore-or rather we may say it certainly is the case-that those parts of the spectral streak which seem dark are in reality luminous; or— which is merely another way of saying the same thing-that the vapours which absorb light from the solar beams, send us light of their own. And SO with stars. Therefore, we have this difficulty to contend against,-that there is no power of determining whether a change in the intensity of a line, or of a set of lines, is due to a variation in the light-giving power of the corresponding vapour, or to a variation in the quantity of vapour whose absorptive effects produce the lines.

But, inasmuch as it resulted from Mr. Huggins' examination of a temporary star which appeared last year, that the increase of light -for it was only the abnormal brilliancy of the star which was really temporary-was due to a sudden outburst of inflamed hydrogen, it seems on the whole more probable that the incandescent vapours of stars burn with variable brilliancy, than that they vary in quantitative distribution.

As regards the constant colours of different stars, we are enabled at any rate to deduce negative results.

For instance, we may dismiss at once the theory started some years ago by a distinguished astronomer. He supposed that the colours of a star are due to the proper motions

of the star, acting so as-in effectto lengthen or shorten the waves of light proceeding from the star to the earth, just as the apparent breadth of sea-waves would be greater or less to a swimmer according as he swam with or against their course. It is quite clear that the effects of a motion rapid enough to produce such a change would be to shift the position of the whole spectrum, and this change, though accompanied by a change of colour, would be readily detected by a reference to the spectral lines.

Another theory-that the orange and red tints indicate a lower degree of temperature, must also be dismissed. For we have seen that the spectra of red stars indicate the presence of the vapour of iron and other metals, and nothing but an exceedingly high temperature could vaporise these.

It seems clear that the difference of tint is due to the different arrangement of the dark lines-in other words, to an absolute difference of physical constitution. "There is a striking difference,' remarks Huggins, 'between the effect on the colour of a star of such closely grouped and very dark lines in the green and blue part of the spectrum of Betelgeux, and of the corresponding part of the spectrum of Sirius, in which the dark lines are faint, and wholly unequal to produce any noticeable subduing of the blue and green rays.'

But we have still to consider the peculiarities presented by the double

stars.

We have seen that amongst the components of these there are observed some which present a distinct blue colour. It has been found possible to analyse some of these with the spectroscope. We have spoken of the charming double star Albireo, the components of which are orange and blue. Both have been analysed,-with this result, that the spectrum of the orange component was remarkable for the

great strength of the lines in the green, blue, and violet, while the spectrum of the blue component is equally remarkable for the great number of groups of fine lines in the orange and yellow.

It would seem, then, that the complementary colours observed in certain double stars, indicate a sort of complementary distribution between the two stars of elements which in our Own sun are associated equably and intimately.

6

And we must note here in passing that it is not absolutely necessary, as some have supposed, that, if there are systems of worlds circulating around such double suns, there should be any remarkable difference in the quality of light distributed to the planets, as compared with that which we receive from the sun. Sir John Herschel has spoken of the charming contrasts and grateful vicissitudes-a red or a green day for instance, alternating with a white one or with darkness, according as one or other, or both of the stars should be above the horizon.' But if the dependent orbs swept in very wide circuits about their double sun, they would receive white light during nearly the whole of each of their days, since it would only be during a brief interval that either sun would be visible alone above the horizon.

Of the deeply coloured stars which are visible with the telescope, none have been found sufficiently brilliant to admit of analysis.

A peculiarity has been remarked by a distinguished modern observer which is worthy of careful attention. Many of the regularly variable stars, when passing into their phase of minimum brightness, exhibit a ruddy tinge which is very conspicuous in instruments of adequate power. It does not seem easy to explain this as due to any change in the vaporous constitution of a variable star-since it seems difficult to show why such changes

should occur at regular intervals. It would appear to be more probable that, in general, these changes are due, either to the rotation of the star itself, and the presentation, in a cyclic order, of the different parts of an unequally illuminated globe, or to the revolution round the star of an extensive vaporous mass, whose interposition cuts off from us at regular intervals a portion of the star's light.

It is remarkable that a large number of the known variable stars are red or orange. There is one notable exception, however, for Algol -the celebrated variable in Medusa's head-is a white star.

It is probable that a careful examination of the stars with any efficient colour-tester' would lead to the discovery of many cases of variation in colour. Admiral Smyth adopted a chromatic scale of colour -but a test of this sort is not very satisfactory. Opaque colours generally vary with time, so that it is impossible to say that two observers, even if they have used the same strip of coloured discs, have really made observations fairly

comparable inter se. And it is further to be noted that there are many persons who find a difficulty and uncertainty in the comparison of stars, or brilliants, with opaque colour-scales. An ingenious astronomer has suggested the use of chemical solutions, which can always be reproduced with certainty; and he has described a method for forming an artificial star in the field of view of a telescope, and for gradually varying the colour of the star until it should coincide with that of a fixed star whose colour we may desire to determine. The great objection to the plan is its complexity. Coloured glasses, through which a small white disc within the telescope might be illuminated (just as the wires are illuminated in the ordinary transit telescope), would serve the same purpose much more simply. The inquiry is an exceedingly interesting one, and Sir John Herschel has expressed the opinion that there is no field of labour open to the amateur telescopist which affords a better promise of original discoveries than the search for such variations as we have described.

A

VERY hot controversy has broken out between two eminent men in France. They belong to the class which the financial world is fond of calling authorities, and they have been advanced to one of the highest literary dignities in Europe, on account of their reputation in the science in which they now come forward as contending champions of contradictory opinions. They are Members of the Institute of France, because they are held to be great in currency: and yet the principles which they proclaim on this subject are diametrically opposed to each other. The conflicts of such men are likely to be full of instruction for those who desire to discover the truth on a question of such great practical importance as currency, if indeed the confusion, which universally prevails in respect of it, does not deter them from engaging in a study which seems as yet barren of well ascertained results. The point on which the quarrel turns is one which compels the combatants, on both sides, to dive down to first principles, and to make a direct application of them to the issue in debate: and thus those who follow the argument have the great advantage of knowing clearly what those principles are, and still more, of testing their value, by observing how they stand under fire, and how they succeed in meeting the requirements of the problem to be solved.

The matter in dispute is Free Banking, a question as vexatiously debated in England as in France. This expression, however, does not cover so great an extent of ground as it seems to imply. Neither the limitation of the liberty to set

La Banque d'Angleterre et les Banques l'Institut.

up a bank at pleasure, nor restrictions on bankers to carry on their business in any way that they may choose, except in one single respect, are at stake in this struggle. The term Free Banking refers to one single function, exercised by some banks, few in number, but great in importance; and it is here intentionally chosen by one of the parties in the struggle for the purpose of investing the view he espouses with the dignity and authority which belong to the expression, Free Trade. Free Banking, as discussed by M. Wolowski and M. Michel Chevalier, means banking subject to no restriction of any kind, and consequently empowered at pleasure to issue bank notes to the public. Ought the issuing of bank notes to be confined to a single central bank of issue, or ought it to be open to every bank, without limitation, as a right intrinsically inherent in the business of banking? On this question the quarrel hangs. Besides the warmth natural to scientific controversialists, a practical interest imparts additional acrimony to the discussion. The privilege of the Bank of France is involved in the answer given to the question; and a large profit, eagerly coveted by one side, and as eagerly retained by the other, gives increased vivacity and intensity to the argument. The nature of the question brings forward, clearly and categorically, the leading principles usually proclaimed in currency; and it is for this reason, beyond every other, that is, because an opportunity is gained for seeing what these principles become, when handled for winning a battle of pecuniary as well as scientific value, by men ac

d'Écosse. Par L. Wolowski, Membre de