The vapor, as it emanates from the nucleus, is repelled by the sun with a force often two or three times as great as the ordinary solar attraction. The most probable explanation of this phenomenon is, that it is a case of electrical repulsion, the sun and the particles of the cometary mist being similarly electrified. With reference to this electrical theory of the formation of comets' tails, Professor Peirce makes the following observation: "In its approach to the sun, the surface of the nucleus is rapidly heated: it is melted and vaporized, and subjected to frequent explosions. The vapor rises in its atmosphere with a well-defined upper surface, which is known to observers as an _envelop_.... The electrification of the cometary mist is analogous to that of our own thunder-clouds. Any portion of the coma which has received the opposite kind of electricity to the sun and to the repelled tail will be attracted. This gives a simple explanation of the negative tails which have been sometimes seen directed towards the sun. In cases of violent explosion, the whole nucleus might be broken to pieces, and the coma dashed around so as to give varieties of tail, and even multiple tails. There seems, indeed, to be no observed phenomenon of the tail or the coma which is not consistent with a reasonable modification of the theory." Professor Peirce regarded comets simply as the largest of the meteoroids. They appear to shine partly by reflected sunlight, and partly by their own proper light, which seems to be that of vapor rendered luminous by an electric discharge of slight intensity. [Illustration: Fig. 357.] 317. _Collision of a Comet and the Earth._--It sometimes happens that the orbit of a comet intersects that of the earth, as is shown in Fig. 357, which shows a portion of the orbit of Biela's comet, with the positions of the comet and of the earth in 1832. Of course, were a comet and the earth both to reach the intersection of their orbits at the same time, a collision of the two bodies would be inevitable. With reference to the probable effect of such a collision, Professor Newcomb remarks,--