approaching a real solar observing neighbourhood. The thermometer under such shade as the carriage roof with open window ventilation afforded, rose from 71° to 87° Fah. The geraniums and juicy grasses of Cintra disappeared from the road side. No more now of its "sunken glens, whose sunless shrubs must weep," nor its hoary cork trees waving pendant mosses from every bough; but gaunt aloes stood on arid slopes, and scorched up Australian gum trees with shadeless dangling leaves, looked at us over stony walls; the homeless dogs and equally public cats were sleeping, or lying exhausted, under the eaves of roofs, or wherever a handful of shadow was to be found; and when we reached Lisbon, the fair city seemed positively white-hot in the sunshine, though hazy with eternal smoke.

But at the Royal Observatory, away to the west, we hoped presently to have the sun of Lisbon without the smoke thereof.

P. S. THE ATMOSPHERE OF THE PLANET JUPITER.

BY E. NEISON.

(Continued from page 230.)

The mathematical formulæ by which these results have been obtained have been given in the Quarterly Journal of Science for October, 1875, and no doubt can exist as to their accuracy. The results may be given as follows: If d, t, be the density and temperature of the atmosphere at the uppermost cloud-bearing strata of the atmosphere which is visible to us, then at any distance s in miles beneath this layer, the density and temperature will be d and t respectively. And the value of d may be calculated from the formula

In the last equation ẞ is a constant depending on the temperature to at the surface of the planet, and the specific gravity D of the gas constituting the atmosphere, in the last air being taken as unity. Therefore

B-1 = D.

1 +0.003665 to
0'500

The two factors f and r are constants, on whose values depend the rapidity and degree in which the temperature of the atmosphere decreases as the height above the heated surface of the planet increases. The value of these two factors depends primarily

on the temperature of the surface of Jupiter. Thus the value of f will be

when is the temperature of the uppermost layers of the atmosphere, that is to say, the temperature of space. It is evident that f must also lie within zero and unity. It cannot be zero, for this would make the temperature of the atmosphere uniform throughout ( really depends on f, for it must vanish with f). The surface of the planet Jupiter, being supposed to be intensely heated, and the temperature of space being known to be much under zero centigrade, it can easily be shown that the value of ƒ must be greater than two-thirds and less than unity. By assuming different values for t, and to any different value may be arrived at, and its results tried.

The rate at which the temperature falls depends mainly on the value of r, and its value can be found from the equation—

where & being the density of the absorption at the surface of the planet, the factor u, is found from the equation

u1 = log. (-
&

*

This last equation is practically of little use, so that remains to all intents an arbitrary constant. If its value be assumed to be unity, its influence disappears. If it is taken greater than unity, it corresponds to a quicker decrease of temperature, and reduces the depth of atmosphere which can possibly exist for a given temperature at the surface of the planet. If supposed less than unity, it corresponds to a slower decrease of temperature. It will be found that it cannot be assumed to be much less than if the temperature of the planet be supposed to be high, without making the temperature of the upper cloud-bearing strata greater than is consistent with the fact that they are supposed to be the uppermost cloud-bearing strata. Because if it be supposed that the upper cloud-bearing strata of the atmosphere reached a considerable temperature, the clouds would be vaporised, and ascending condense still higher in a cooler region. The new strata would be then the region to which the considerations would apply. The value of r must never, therefore, be made so

These formulæ are not the most convenient which could have been

contrived for this special purpose. They were put in this form for another end, and it has not been thought necessary to transform them.

small that the temperature of the upper cloud-bearing strata is made too great. This means really that this temperature should not differ from o° centigrade.

The following considerations will show how these formulæ can be applied to the end in view. When & becomes 1,000 times the density of that of our terrestrial atmosphere, it would to all intents cease to behave as a gas, and become to all intents a liquid. No increase of temperature would of course affect this. It can, therefore, be supposed that at the depth where d is equal to 1,000 times the density of our own atmosphere-that is when it would be denser than water-the surface of the planet (liquid or solid) would be reached. Next, considering d, it is probable that no cloud layers would permanently exist if the density of the atmosphere was less than one-tenth of that at the surface of the earth. This is the case on the earth, and would be probably the case on Jupiter. Take, however, an extreme supposition and suppose that clouds would exist on Jupiter in strata of its atmosphere only one-hundredth of the standard density of the terrestrial atmosphere. Putting then Too and 8= 1000, and the first d = equation gives

= 1000 e

-26

ulog. (100,000) = 11515

(The logarithms are throughout hyperbolic or natural logarithms.) Substitute this value in the equation to s and—

By giving different values to D fand r in these formulæ, different permissible atmospheric conditions may be found for the planet Jupiter. It must be remembered that ƒ cannot exceed unity, nor r be so small that (1+0.003665 t) (1 −ƒ + ƒ[100000]') differs sensibly from unity.

It must be pointed out that to suppose D to be materially less than unity, is equivalent to assuming the atmosphere of Jupiter to be composed of some element or elements entirely unknown. For the only element which possesses a gaseous specific gravity materially less than that of atmospheric air is hydrogen, and it is obviously out of the question to suppose the Jovian atmosphere to consist of incandescent hydrogen. The spectroscope, moreover, has shown that it is impossible for the principal constituent of the atmosphere of Jupiter to be a gas much lighter than atmospheric air. Thus Dr. Vogel and others have shown that the

Jovian spectrum contains dark lines, similar to the ordinary telluric lines, and known to be due principally to aqueous vapour. The light received from Jupiter being solar light reflected from the upper cloud-bearing strata, this shows that aqueous vapour must exist at least in the cloud-bearing strata, and probably above it. In fact the presence of aqueous vapour in the atmosphere of Jupiter shows that probably, nay, almost certainly, that the elouds in the higher regions of the atmosphere of Jupiter are mainly due to aqueous particles, exactly in the same manner as the terrestrial clouds. The proof of this conclusion, like that on many other points connected with the physical condition of the planet, is necessarily primarily a chemical one, and rests on the known influence of aqueous vapour on other chemical compounds. Next the upper portions of the atmosphere of Jupiter must be considered to be in statical equilibrium, that is to say, although, doubtless, subjected to numerous minor disturbances, such as air currents, and to be generally free from any very violent disturbances, like those which occur in the solar atmosphere for example. Under these conditions it is a known physical law; a law which has also been deduced by Clerk Maxwell from the dynamical theory of gases, that in a mixture of different gases, each one so disposes itself that it would be in equilibrium were there no other present. Consequently aqueous vapour would not be found permanently in the atmosphere of Jupiter at any greater height above the surface than it would occupy were it alone to constitute the atmosphere. Suppose then the Jovian atmosphere to contain some hypothetical gas much lighter than aqueous vapour. Then the last conclusions show that this hypothetical constituent of the atmosphere of Jupiter may be neglected in finding the greatest depth beneath the upper cloudbearing strata that the atmosphere possesses. For it has been shown that the upper cloud-bearing strata of Jupiter's atmosphere permanently contains large quantities of aqueous vapour (were the amount small no absorption lines would be shown by the spectroscope), and this aqueous vapour would not ascend permanently to higher altitudes than it would occupy were it alone to form the atmosphere, consequently the height above the surface of these visible uppermost clouds cannot be greater than what would be their height were the hypothetical gas non-existent, and the atmosphere composed of aqueous vapour only. This conclusion proves that the presence of a hypothetical new gas, much lighter than air, would not disturb the conclusions which have been arrived at with regard to the maximum depth beneath the visible cloud-bearing strata to which the atmosphere may extend. At a subsequent period it will be shown that besides

being immaterial it is impossible that the atmosphere of Jupiter can contain such a gas.

*

It may, therefore, be taken as established that the planet Jupiter cannot possess an atmosphere which extends to a greater depth than 150 to 200 miles beneath the visible uppermost cloud-bearing strata. It will be found, in fact, that it is only by violating conditions which must be observed, that is to say, by assuming the atmosphere of Jupiter to be in a condition which it is known it either does not or even cannot possess, that any greater depth can be deduced. The formula which have been given will prove this. It is probable, however, that the atmosphere of Jupiter does not extend 100 miles beneath the uppermost layer of clouds; yet this depth, small as it looks, is fully ten to fifteen times greater than the corresponding depth. in the case of the earth.

Returning to the interpretation which has been put on the phenomenon observed by Mr. Todd, at Adelaide, it remains to see what depth of atmosphere to Jupiter these observation are claimed as proving. Both Mr. Todd and Mr. Ringwood saw the I. satellite of Jupiter projected on the disc of Jupiter, as if at internal contact, so that the whole disc of the satellite was seen within the limb of the planet. That is, the apparent centre of the satellite was rather more than half-a-second of arc within the apparent edge of the planet, the diameter of the first satellite being taken as 108". The image of the satellite not being distorted, necessarily the preceding edge of the satellite must have been rather over one second of an arc distant from the apparent edge of the planet.

Mr. Todd's interpretation of this phenomenon seen by himself, and supposed to be due to the action of an atmosphere, is very simple. His words are very definite. The satellite appeared "as if seen through the edge of the planet, as if the latter were surrounded by a transparent atmosphere laden with clouds.” The view taken, therefore, was that the phenomenon was exactly what might be expected were the planet a circular disc with a semi-transparent border of rather over one second of an arc in thickness. As the satellite passed behind the transparent border, the satellite was seen through it, and only disappeared when it passed behind the central opaque portion. The phenomenon

*The above applies solely to a gas much lighter than air; therefore the fact that the spectroscope shows lines of unknown origin does not apply, for this substance may be of similar density to air. It has not, however, been shown that the new lines may not owe their origin to the fact that a greater mass of atmosphere is traversed by the solar rays on the planet Jupiter than on the earth.