120 THE INDIAN ECLIPSE, 1898. 



generally called the " polar rays." But in truth a careful 

 examination of such photographs as those of Mr. Thwaites and 

 Mr. Fred Bacon, already referred to, shows clearly that similar 

 " plumes " may be seen also in every region of the limb. There 

 can, we think, be little mistake as to the nature of these. On 

 two if not three occasions chance has afforded a very palpable 

 hint as to their nature. In 1882 a small comet was seen near 

 the sun during the eclipse of that year, and it so resembled the 

 coronal " plumes " that, though it was quite detached from 

 the sun, and showed a distinct cometary nucleus, more than 

 one distinguished astronomer was inclined to believe it simply 

 a detached fragment of the corona. Again, in the eclipse of 

 1893, a comet was actually seen projected upon the corona, 

 from which it would have been almost impossible to distinguish 

 it but for the fortunate circumstance that the interval in time 

 between the taking of the photographs in Chili, Brazil, and 

 West Africa was sufficient for the motion of the comet to render 

 its character unmistakable. Mr. Eanyard also claimed, though 

 on less convincing evidence, that a comet formed part of the 

 coronal forms in 1871. Whether this last was a real instance 

 or not, we may look upon the eclipse comets of 1882 and 

 1893 as we might upon detached leaves fluttering down from 

 a tree, affording us by their separation an easier demonstra- 

 tion as to the character and nature of those left on the parent 

 stem. 



We know perfectly the sort of changes through which a 

 comet passes as it approaches the sun : how thin shells of 

 material rise from the nucleus towards the sun, and are then 

 violently repelled. Professor Bredichin has investigated the 

 forms of a great number of cornets' tails, and if his results be 

 accepted we may take it that cometary tails may be grouped in 

 three principal classes, in proportion to the strength of the 

 repulsion exercised upon their material by the sun a re- 

 pulsion which is the greater, the lighter the material acted 

 upon. Professor Bredichin therefore considers that the three 

 types of tail correspond to hydrogen, to the hydro-carbons, and 

 to the heavy metals. 



What we know of the orbits of comets, and of the meteor 

 streams, leads us to conclude that a vast number of these must 

 have their perihelia within the coronal region. There can, 

 therefore, never be any lack of small bodies in the immediate 

 neighbourhood of the sun to act as cometary nuclei, and to 

 give rise to cometary tails, presenting themselves to us under 

 the form of these beautiful coronal plumes. In this way we 

 can understand the curious coronal rifts, so hard to explain on 

 any other ground. 



If the great rays stood alone they might naturally be explained 

 in the same way. A tail, long, narrow, and straight, is neces- 



