MOVEMENTS OF JUPITER'S CLOUD-MASSES. 89 



rior activity of the forces at work below them in other words, the in- 

 tense heat of Jupiter's real globe. As to the actual depth of the semi- 

 transparent atmosphere in which these cloud-layers and. cloud-masses 

 float, it would be difficult to express an opinion. We do not know 

 how many cloud-layers there are, how thick any cloud-layer may be, 

 how great may be the depth of the vast rounded masses of cloud 

 whose upper surface (that is, the surface remotest from Jupiter's true 

 surface) we can alone see under favorable conditions. But we can 

 indicate a minimum than which the atmosphere's depth is probably 

 not less ; and, from all the observations which I have examined as 

 bearing on this point, I should be disposed to assign for that mini- 

 mum at least 6,000 miles. I am strongly of opinion that in reality 

 the depth of the Jovian atmosphere is still greater. I cannot doubt 

 that Jupiter has a solid or liquid nucleus, though this nucleus glow- 

 ing, as it must be, with a most intense heat may be greatly ex- 

 panded ; yet I should conceive that, with the enormous attractive 

 power residing in it, containing as it must nearly the whole mass of 

 the planet, its mean density cannot be less than that of the earth. 

 Now, a globe of the mass of Jupiter, but of the same mean density as 

 our earth, would have one-fourth of Jupiter's volume the mean den- 

 sity of Jupiter, as at present judged, being equal to one-fourth that 

 of the earth. The diameter, therefore, of such a globe would be less 

 than the present diameter of Jupiter, in the same ratio that the cube- 

 root of unity is less than the cube-root of 4, or as 1 is less than 1.58Y4. 

 Say, roughly (remembering that the atmosphere of Jupiter must have 

 a considerable mass), the diameter of Jupiter's nucleus would, on the 

 assumptions made, be equal to about five-eighths of his observed di- 

 ameter, or to about 53,000 miles. This is less than his observed diam- 

 eter by about 22,000 miles, so that the radius of his nucleus would be 

 less than his observed radius by about 11,000 miles which, therefore, 

 would be the probable depth of his atmosphere. 



But we have still to consider the velocities with which ronnded 

 masses of cloud travel in the very deep atmosphere of Jupiter. 

 "There is clear evidence," I have pointed out in the article "As- 

 tronomy" of the "Encyclopedia Britannica," "that spots on Jupiter 

 are subject to a proper motion like that which affects the spots on the 

 sun. Schmidt, in No. 1,973 of the Astronomische Nachrichten, gives 

 a number of cases of such proper movements of spots, ranging in 

 velocity from about seven miles to about 200 miles an hour. It may be 

 noted, also, that from a series of observations of one spot, made be- 

 tween March 13 and April 14, 1873, with the great Rosse reflector, a 

 period of 9 h. 55 m. 4 s. was deduced, while observations of another 

 spot in the same interval gave a rotation period of 9 h. 54 m. 55.4 s. 

 The actual difference of velocity would depend in this case on the 

 actual latitudes of the two spots, which were not micrometrically 

 measured. Taking 200,000 miles as about the circumference of a 



