THE ECLIPSE OF 1870. 51 



But even in Jupiter, notwithstanding the grandeur of 

 his system of satellites, and though total solar eclipses 

 recur at intervals which must be measured by hours 

 rather than by months, as with us, yet such solar 

 eclipses as we see can never take place. For not one 

 of his moons is capable of just hiding the sun's disc and 

 a very narrow border all round while leaving in view 

 beyond that border the coloured prominences, and be- 

 yond the prominences the glory of the corona. If we 

 try ^G conceive the circumstance of an eclipse of the 

 sun by one of Jupiter's nearest moons, we have to 

 imagine a dark disc capable of obliterating a sun more 

 than thirty times larger than that which is actually 

 seen from Jupiter ; and even the farthest of Jupiter's 

 moons covers twice as great a space as the sun. It is 

 easily seen that when a total eclipse is just beginning 

 or just ending, under these circumstances, only a small 

 part of the matter outside the sun can be visible, and 

 nothing resembling that complete ring of such matter, 

 visible to ourselves when the moon obliterates from view 

 the nearly equal solar disc. So also in Saturn, whence 

 the sun must appear as a mere dot of bright light, 

 and in Uranus and Neptune, whence he appears yet 

 smaller, there can be no such eclipses as we inhabi- 

 tants of earth are favoured with. Hence it may not 

 unreasonably be concluded that terrestrial astronomers 

 alone have any knowledge of the coloured solar promi- 

 nences and of the corona. 



It is worth mentioning, also, that, interesting as are 

 the discoveries which have been recently made during 



E 2 



