On the Formation of the Tails of Comets. 127 



and apparent length, when the supply of emitted matter is in 

 greater quantities than before in a given time. Whether we sup- 

 pose, with Herschel and others, that the materials of the tail are 

 furnished by the heat of the sun, or, with Olbers and Bessel, that 

 the particles of the tail flow directly from the nucleus, under the 

 operation of a repulsion consequent upon some polarizing action 

 of the sun, it must increase in apparent length and brightness as 

 the comet approaches the sun. But it is supposable, on either 

 theory, that the maximum action may not be till a certain time 

 after the perihelion passage. The tail may also undergo varia- 

 tions of brightness and of length, as seen from the earth, by rea- 

 son of changes in the obliquity under which it is viewed. Thus, 

 the great brilliancy of the recent comet on the 28th of February, 

 and its great length in the latter part of March, are partially attri- 

 butable to the small angle of inclination of the line of sight to the 

 line of the tail. 



As to multiple tails they may originate in separate nebulosities, 

 or may be collections of matter upon which the nucleus has dif- 

 ferent degrees of repulsive action. The fact often noticed, that 

 multiple tails spring up suddenly and generally soon disappear, 

 (of which we had an exemplification in the comet of 1843, as 

 seen about the 3d of March,) may be alluded to here as affording 

 striking evidence of the truth of the theory under consideration ; 

 for, it will be seen that their disappearance is a simple conse- 

 quence of the waste of the local stock of materials which in the 

 act of escaping had formed the supernumerary tail. 



There is but little space left for the explanation of phenomena. 

 Let us first see how the theoretical agrees with the actual situa- 

 tion of the tail. Let PCA be a portion of a comet's orbit, the sun 

 being at S : and suppose a particle to be expelled in the direction 

 SAD, when the head is at A, and another particle to be driven 

 off in the direction SBE, when the head is at B. Each particle 

 will retain the orbitual motion which obtained at the time of its 

 departure, as it moves away from the sun ; and thus, when the 

 comet has reached the point C, instead of being at any points 

 D and E on the lines SAD and SBE, will be respectively at cer- 

 tain points a and b farther forward. The line Cab, which when 

 the comet is at C, is the locus of all the particles that have been 

 emitted during the interval of time in which the comet has been 

 moving over the arc AC, is the tail. It is easy to see that this 



