150 



in a similar manner. The obliquity of the axis of rotation is known 

 •with considerable accuracy in the cases of Mars and Jupiter ; and 

 also in that of Saturn, if it coincide with the axis of rotation of his 

 ring. Venus presents great difficulties to the observer, but it ap- 

 pears now to be pretty satisfactorily determined, that the period of 

 rotation about her own axis is nearly the same as that of the Earth, 

 and tbat the obliquity of her axis is large, amounting to as much as 

 about 75°. This must produce an extraordinary difference between 

 tbe changes of annual temperature in that planet and those which we 

 experience. The author has endeavoured, in this paper, to estimate 

 numerically the effect of this anomalous obliquity. Practical astro- 

 nomers have entertained the opinion that Venus likewise has an 

 atmosphere. Of Mercury we know too little by direct observation 

 to form any opinion on those points founded on observed facts, and 

 the same remark will apply to the remoter planets beyond Saturn ; 

 but most astronomers probably feel much the same conviction that 

 Mercury, Uranus, and Neptune have atmospheres of greater or less 

 extent, as that they revolve round their own axes with greater or 

 less angular velocity. 



It is not the author's object, however, to adjust the balance of pro- 

 babilities for particular hypotheses in favour of planetary atmospheres 

 or against them; but assuming their existence, to estimate their effects 

 on the planetary temperatures. And in like manner he points out the 

 influence which must be exercised by a greater or less conductivity, and 

 specific heat in the superficial matter of a planet, without professing to 

 discuss the probability of such properties being materially different in 

 the different planets. The Earth's atmosphere is known to be almost 

 completely diathermanous for heat radiating directly from the Sun ; 

 and it is assumed to be equally so for the heat which proceeds directly 

 from the fixed stars, and to which the general temperature of space 

 is due. This radiating heat therefore has little or no effect in heat- 

 ing the atmosphere during its transmission to the Earth's surface ; 

 but after falling upon and heating terrestrial objects, it loses the 

 power of radiating completely through the atmosphere, and is trans- 

 mitted back into space through the atmosphere by conduction, con- 

 vection, and partial radiation to limited distances. But for any of 

 these modes of transmission, it is essential that the temperature of 

 the atmosphere should be greater in its lower than in its upper por- 

 tions, and in a degree greater as the quantity of heat to be trans- 

 mitted is greater. The temperature (r 2 ) of the upper portion must 

 be determined by the condition, that, in a given time, a quantity of 

 heat must radiate from it into surrounding space equal to that which 

 falls upon it from external sources in the same time, and is trans- 

 mitted back after reaching the surface of the earth or objects near 

 to it. Consequently r 2 must be independent of the height of the 

 Earth's atmosphere. At lower points the temperature will increase 

 till we reach the surface of the Earth ; and if we denote the tempe- 

 rature there by r x , it is manifest that r, will be greater, the greater 

 the height of the Earth's atmosphere. 



It must here be particularly observed, that t 2 is the proper tern- 



