when the circuit is closed, is most remarkable. In twenty-four hours the electromotive force only loses 0.02 of its original intensity, and then the couple becomes perceptibly constant for several days. Terrestrial couples with a constant current are likewise obtained with iron and lead.

The composite terrestrial couple was formed of a plate of amalgamated zinc with an area of 1 decimetre square, placed in a good vegetable mould at 1 decimetre from the surface, and of a plate of manganized coke, immersed in an earthen cylinder open at both ends, with one end buried in the earth and surrounded with clay to a height of 1 decimetre; then filled with salt water marking 3o, The zinc was 2 metres from the coke.

The resistance of this terrestrial couple is 5.34 greater than that of the nitric acid couple; but this resistance is variable, and depends on the nature of the soil, its hygroscopic state, and the distance between the elements.

FOURTH PART. On Storms.—Amongst the natural sources which furnish electricity continually to the air, we must distinguish the following:

1. The exhalation of oxygen and carbonic acid gas by the leaves of plants, which furnishes positive electricity to the air.

2. The contact of soils and fresh and salt waters; the former acquire an excess of positive, and the latter of negative electricity.

Living vegetables behave in the same way in relation to the earth, that is to say, the parenchymatous tissues are negative, and the earth positive; the two electricities are transmitted to the air by the mediation of vapours.

3. The decomposition of animal and vegetable matters at the surface of the earth and in the water, which gives rise to complex effects. 4. The contact of cold and warm waters circulating on the surface of the seas; the cold waters are positive, and the warm negative. The tension of the electricity, considering the extent of the surfaces, must certainly be exceedingly weak; for the evolution of electricity only takes place by contact where there is a mixture of cold and warm waters.

It has been asked whether, in evaporation in the present and similar cases, that is to say, in a voltaic couple of which the circuit is not closed, each element can furnish electricity to the vapour which escapes from it. We know very well that when one of the elements of a couple is put in communication with the earth, we may give a charge to a condenser by placing one of the plates in communication with the other element; ought the case to be similar in this same couple when vapour escapes from each element? There is nothing to oppose this, seeing that the vapour continually carries off the excess of electricity proper to each element, as might be done by a conducting body which would complete the circuit. It is precisely what takes place when the two plates of a condenser are put in communication with the two elements of a couple.

We see therefore that a great number of facts tend to prove that electricity, sometimes positive, sometimes negative, is evolved in the

air from various natural sources; recompositions of the two electricities must therefore continually take place. The result is an excess of positive electricity when the sky is serene; for this purpose it is necessary that the sources which furnish this electricity should be the most numerous, and these sources are probably vegetation and the contact of soils and waters.

After recalling all that we know of the formation of storm-clouds, I have pointed out that the storm does not generally burst except when two clouds charged with opposite electricity are in presence of each other, or when a strongly electrified cloud comes within the sphere of the earth's activity. In these two cases there is action by influence and a re-establishment of equilibrium. Thus in places where sources of opposite electricity exist at some distance apart, these localities combine the principal conditions for the production of storms.

In the polar regions the rarity of storms is probably connected as much with the absence of evaporation as with the small number of natural sources of electricity; just as under the tropics the frequency of storms is in relation to the abundance of evaporation and that of the sources of electricity.

The diminution of storms in advancing, on the one hand, into the continents, and, on the other hand, in departing from the coast into the open sea, where it very rarely thunders, seems to indicate that the greatest number of sources of opposite electricity exist at but a small distance from the coasts; it is there consequently that they must be sought.

FIFTH PART. On the Crystalline Compounds formed in slow actions with or without the concurrence of electrical forces.-The examination of the natural sources of electricity which must intervene in the phænomena of decomposition and recomposition of mineral substances, has led me again to the study of the slow actions which take place with or without the apparent concourse of electricity, a question which possesses a high degree of interest for geologists. In the first place, I have indicated the improvements which I have introduced in the processes which I have hitherto made use of to obtain various metallic oxides, and double or simple basic salts in a crystalline form; I have then shown how the following products may be obtained crystallized in the space of several years, upon mineral substances placed in suitable conditions:

1. The double carbonate of lead and soda upon galena in small acicular crystals, which will be found some day in lead mines into which water charged with bicarbonate of soda filters;

2. Carbonate and sulphate of lead upon galena in crystals similar to the native ones;

3. Arseniate of lime upon sulphate of the same base, exactly similar to what is met with in some veins;

4. Bibasic phosphate of lime crystallized, with phosphate of copper upon limestone;

5. Double combinations, which will be found described in my memoir;

6. Copper pyrites, which was made to undergo such a decompo.. sition that its surface became adorned with the fine colours presented by iridescent copper pyrites.

When I commenced, thirty years ago, occupying myself with the reproduction of mineral substances by the concurrence of the elec trical forces with those of the affinities, it was objected against me that nature did not make use of voltaic couples. This objection did not appear to me to be of a serious nature. I showed that I sometimes arrived at the same end without the intervention of a couple by placing insoluble mineral substances in contact with various solutions for several years. From this has resulted this new principle in chemistry, that the insolubility of bodies is not absolute, and that, in course of time, when they are in contact with certain solutions, double decompositions take place, which give origin to crystallized compounds similar to those found in metalliferous strata. -Comptes Rendus, Dec. 15, 1856, p. 1101.

TWINKLING OF THE STARS.

The stars twinkled much at Brighton on the evening of the 16th of March. Sirius and Aldebaran were in sight; and on passing the axis of the eyes across them so as to produce a changing place for their image on the retina, it was seen that the difference due to twinkling was so great as at certain moments to cause the apparent extinction of the stars. A mirror was therefore held in the hands so as to send a reflected image of the selected star to the eye, and then the mirror was moved in such a manner as to throw the image of the star into a line or a circle, which could easily be done. Upon examining the brightness of the star image at the moment when it was reflected from different parts of the moving mirror, the light was seen to differ enormously, and very frequently, indeed, to be replaced entirely by darkness. In fact, the successive phases of the star, which, being seen in one direction, gave a continual but wavering light, were now separated into their luminous and their absolutely dark conditions; and thus the effect of twinklings made far more manifest than by the ordinary mode of observation. The apparent extinctions were not for long together; but they often cut up a circular path of light of about 10 degrees angular space into six, seven, or eight parts, separated by short dark intervals without sensible light. M. F.

NOTE ON THE DENSITY AND THE MASS OF COMETS.

BY M. BABINET.

All astronomers are agreed that the mass and density of comets are very small, and that their attraction cannot produce any sensible effect upon the movements of the planetary bodies. We shall see that from the effects observed, combined with the laws of optics, we may deduce the conclusion that the direct shock of one of these

bodies could not cause the penetration of the infinitely rarefied matter of which they are composed, even into our atmosphere.

It is a well-ascertained fact that stars of the tenth and eleventh magnitude, and even lower ones, have been seen through the central part of comets, without any sensible loss of brilliancy. Amongst the observers who have frequently proved this optical fact, we find the names of Herschel, Piazzi, Bessel, and Struve. In most cases, says Mr. Hind, there is not the least perceptible diminution in the brilliancy of the star. As to the pretended phases of the cometary nuclei, the direction of the line of the horns was incompatible with the hypothesis of the illumination of an opake nucleus, and the modern representations of the appearance of comets readily explain the error of those who have assumed opake nuclei. I shall take as an example the well-known comet of Encke, which is sometimes visible to the naked eye, and generally presents a rounded mass. In 1828 it formed a regular globe of about 500,000 kilometres in diameter, with no distinct nucleus; and M. Struve saw a star of the eleventh magnitude through its central part, without noticing any diminution of brilliancy. In an observation of M. Valz, on the other hand, a star of the seventh magnitude almost entirely effaced the brightness of a brilliant comet. Let us start from these observed facts.

Since the interposition of a comet illuminated by the sun does not sensibly weaken the light of a star in front of which it forms a luminous curtain, it follows that the brilliancy of the comet is not a sixtieth part of that of the star, for otherwise the interposition of a light equal to a sixtieth part of that of the star would have been sensible. We may therefore assume, that at the utmost the brilliancy of the comet equalled a sixtieth part of the light of the star. Thus, by this hypothesis, if the comet were rendered sixty times more luminous, it would have a lustre equal to that of the star; and if it had been rendered sixty times sixty times, that is to say three thousand six hundred times, more luminous than it was, it would then have been sixty times more luminous than the star, and in its turn would have made the latter disappear by the superiority of its lustre.

The conclusion from this is therefore that it would have been necessary to illuminate the cometary substance more than three thousand six hundred times more than it was then illuminated by the sun, to enable it to cause the disappearance of a star of the eleventh magnitude.

We may assume that the light of the moon causes the disappearance of all the stars below the fourth magnitude; thus the atmosphere illuminated by the full moon acquires sufficient luminosity to render stars of the fifth, and all lower magnitudes invisible.

Between the fifth magnitude and the eleventh, there are six orders of magnitude, and according to the fractional relations of these different orders, we may admit that a star which is a single degree of magnitude above another, is two and a half times more luminous than the latter. In the publications of the Observatory of Oxford, we may see a good compilation of that excellent astronomer, Mr,

Johnson, upon this subject, and very recently a work by Mr. Pogson on the valuations of magnitudes has appeared. From this we learn that a star of the fifth magnitude is about two hundred and fifty times more brilliant than a star of the eleventh magnitude. Thus the illumination of the atmosphere by the moon is much more intense than the illumination of the cometary substance by the sun itself, since it would be necessary to render the comet three thousand six hundred times more luminous to enable it to extinguish a star of the eleventh magnitude, whilst the luminosity of the atmosphere illuminated only by the moon is sufficient to render invisible stars which are two hundred and fifty times more brilliant.

The disproportion becomes still more striking when we consider, that, according to the measurements of Wollaston, to which Sir John Herschel says he sees no objections to be made, the illumination of the full moon is a little less than the eight hundred thousandth part of the full illumination of the sun.

To complete the data of our definite calculation, we shall call to mind, that, according to the density of the air in the lower strata of the atmosphere and its total weight as indicated by the barometric column, the whole stratum of air which constitutes the atmosphere is equivalent to a stratum of about 8 kilometres in thickness, and possessing the density of the air at the surface of the earth.

We have alseady found that it would be necessary to render the comet 3600 times more luminous for it to extinguish the lustre of a star of the eleventh magnitude. To render a star of the fifth magnitude invisible, it would require to be made 3600 x 250 times more brilliant than it is. In other words, if the atmosphere were 3600 × 250 times less compact than it is, it would be equivalent to the comet.

As 3600 x 250 make 900,000, the nine hundred thousandth part of the atmosphere would suffice to produce the same effect of illumination as the comet; but as the latter is in the full light of the sun, whilst the atmosphere is only illuminated by the moon when it extinguishes stars of the fifth magnitude, this circumstance gives the atmosphere a further advantage in the proportion of 800,000 to 1; which under ordinary circumstances gives the atmosphere a superiority equal to 900,000 x 800,000, or 720 billions.

But this is not all: the thickness of the cometary substance being 500,000 kilometres, whilst that of the atmosphere is only 8 kilometres, we must increase the above relation in the proportion of 500,000 to 8, which brings it to forty-five millions of billions, thus

45,000,000,000,000,000.

Thus, according to these data, the density of the substance of a comet could not be calculated at so high a quantity as that of the atmosphere diminished by the enormous divisor, forty-five millions of billions. The shock of a substance so rarefied would be nothing at all, and not the least particle of it could penetrate even into the most rarefied parts of our atmosphere.

According to experiments of my own, gases lose their property of elasticity long before they are reduced to such a low density. I do