46 PROF. O. REYNOLDS ON THE TAILS OF 



case through an angle of 180 in little more than two hours. 

 It seems utterly incredible that in such a case it is one 

 and the same material object which is thus brandished. 

 If there could be conceived such a thing as a negative 

 shadow, a momentary impression made upon the lumini- 

 ferous aether behind the comet, this would represent in 

 some degree the conception such a phenomenon irresistibly 

 calls up. But this is not all. Even such an extraordinary 

 excitement of the sether, conceive it as we will, will afford 

 no account of the projection of lateral streamers, of the 

 effusion of light from the nucleus of the comet towards 

 the sun and its subsequent rejection, of the irregular and 

 capricious mode in which that effusion has been seen to 

 take place, none of the clear indications of alternate eva- 

 poration and condensation going on in the immense regions 

 of space occupied by the tail and coma — none, in short, of 

 innumerable other facts which link themselves with almost 

 equally irresistible cogency to our ordinary notions of 

 matter and force."" 



There can be no doubt that, if these tails are matter 

 moving with the comet, this matter must be endowed with 

 properties such as we not only have no experience of, but 

 of which we can form no conception. This would almost 

 seem a sufficient reason for rejecting the first hypothesis. 

 Moreover, on the second hypothesis there is no difficulty 

 in the immense velocity with which these tails are pro- 

 jected from the head or whirled round when the comet is 

 in perihelio ; for, to take the ' ' negative shadow " as an 

 illustration, here we should have a velocity of projection 

 equal to that of light, and the only effect of the whirling 

 would be a slight lagging in the extremity of the tail, 

 causing curvature similar to that which actually exists ; 

 and whatever the action may be, if its velocity of emission 

 or transmission be sufficiently great, this effect will be the 

 same. But whether this hypothesis is to be rejected be- 



