Feb, 16, 1883.] 



KNOWLEDGE • 



101 



remaining unevaporated half of the particle, and as the 

 evaporation continues, the velocity of the unevaporated 

 residue will increase, until it may become much j^reater 

 than the velocity with which the gaseous molecules are 

 driven off from the heated body. 



It is evident that evaporation will take place most 

 rapidly where the sun's rays strike vertically ; and if we 

 suppose evaporation to take place so that the particles all 

 fly off in directions which are inclined less than 45° to the 

 line passing through the sun's centre, the resultant effect 

 of all the recoils would give the unevaporated half of tlic 



particle a velocity greater than in a direction away 



from the sun. 



The velocity with which an evaporating molecule leaves 

 the evaporating body must at least be equal to the velocity 

 of the free molecule in the gaseous state immediately after 

 evaporation. We know that the average velocity of 



v/Z 



•273 



miles 



bydrogen molecules at a temperature r is 1 OG' 



per second ; where r is the number of degrees centigrade 

 measured from the absolute zero of temperature, and we 

 consequently do not need to assume a temperature higher 

 than the temperature of an English summer day to give a 

 velocity of I'l miles per second for the molecvdes of 

 hydrogen gas, which would correspond to a velocity of 

 nearly a hundred thousand miles per day. 



Hitherto we have only considered the motion due to 

 evaporation of a part of the mass of the moving particle, 

 bat in the case of a gaseous mixture, where one element 

 is precipitated by cold, while the rest remain in the gaseous 

 form, we may have condensation going on on the shaded 

 side of the precipitated particles and evaporation on the 

 illuminated side, while the condensed matter continually 

 flows round to the illuminated side, or the precipitated 

 particles may be turned by the forces acting on their illu- 

 minated surfaces. 



Prof. Arthur Wright, of Yale, has showTi that when 

 stony meteorites are heated to a temperature of only 

 100° C. a mixture of several gasses, carbonic dioxide, 

 carbonic oxide, hydrogen, &c., is given off. In such a 

 mixture at ordinary atmospheric pressure the carbonic acid 

 would be condensed into the liquid form, at a temperature 

 of about - 80°C., and in vacuo, the condensation would take 

 place at a still lower temperature. We can, therefore, 

 conceive of the action described going on at a very low 

 temperature, and can understand the formation of comets' 

 tails at distances from the sun considerably greater than 

 the earth's distance. 



An evaporating particle near to the comet's nucleus would 

 he acted upon by four forces, viz., gravitation towards the 

 nucleus and heat repulsion from the nucleus, gravitation 

 towards the sun and heat repulsion from the sun. If the 

 repulsive action is in each case greater than the attractive 

 action, the particle will be driven towards a hyperbolic 

 surface, which has the comet's nucleus in one focus and the 

 sun's centre in the other focus. 



This is evident, for on such a hyperbolic surface, the 

 tangent TP bisects the angle SPX. . If the forces in the 

 direction SP and NP are equal, the resultant will bisect 

 the angle between them, and the particle will be driven 



along in the hyperbolic surface ; while if the force in the 

 direction NPis greater than the force in the direction SP, 

 thi- particle will be driven outwards, till it reaches a con- 

 focal hyperbolic slieet, where the forces acting from the 

 sun and nucleus are equal. 



Particles of different sizes would be acted upon by 

 gravitating forces corresponding to their masses — that is, 

 'jy attractive forces varying with the cube of the diameter 

 of the particle ; while the repulsive forces would vary 

 with the surfaces exposed to radiation, or as the square of 

 the diameter of the particles. Consequently particles of 

 difierent sizes, as well as substances evaporating at dif- 

 ferent temperatures, would be driven towards difierent 

 confocal hyperbolic sheets, and as seen in projection, these 

 would appear as hyperViolic envelopes one within the other. 

 Such ditl'erenccs in the repulsive forces would evidently 

 also account for differences of curvature in the plane of 

 the orbit of diHVrent parts of the tail. Fan-shaped, inter- 

 secting arcs in front of the nucleus have frequently been 

 observed, which may possibly be accounted for by sup- 

 posing that hyperbolic envelopes about different nucleii 

 are occasionally seen in projection one through the other. 

 Probably all the phenomena of cometary structure cannot 

 be accounted for on the above supposition, but I wish to 

 point out that it is not necessary to call in a hypothetical 

 electrical repulsive force in order to account for the thief 

 features of cometary structure. Possibly, as Bredichin, 

 Norton, and Zollner have suggested, electricity may be 

 generated as physical changes take place during the heat- 

 ing of the nucleus : and though I cannot accept their 

 electrical theories with regard to the formation of comets' 

 tails, I would suggest that it is possible that the bright 

 line spectrum in the neighbourhood of the nucleus may be 

 due to a rapid succession of electrical discharges, while the 

 general temperature of the gas remains far below the tem- 

 perature of incandescence. 



If the earth had no atmosphere, a particle driven up- 

 wards from the earth's surface with a velocity of a mile a 

 second would, under the action of terrestrial gravity, only 

 rise to a height of 82 miles, and then fall again towards 

 the earth ; but a particle driven away from a nucleus 

 weighing only a few pounds, or even a few million tons, 

 with a velocity of a mile a second, would be driven off into 

 space. It is evident that with temperatures such as we 

 have been supposing, there would not be an atmosphere 

 about a flight of meteoric stones similar to the earth's 

 atmosphere, but the molecules would stream away with 

 approximately parallel motions, and precipitated particles 

 in the neighbourhood of the heated side of the nucleus 

 would be bombarded with molecules of evaporating gases 

 moving in very long free paths. This condition Mr. Crookes 

 has termed " the fourth state of matter,' and he has in 

 a series of beautiful experiments shown that such a bom- 

 bardment may be caused in a vacuum tube about a negative 

 pole when a discharge from an induction coil is passing 

 through the exhausted tube, and that bodies struck by 

 particles of such radiant matter shine with a phos- 

 phorescent light which I conceive may be the cause of the 

 intense brightness of the suspended matter in the neigh- 

 bourhood of the nucleus, while the gaseous molecules 

 during their long free paths would give out their own 

 distinguishing wave lengths. Thus, without electrical dis- 

 charges tlie whole phenomenon of the continuous and 

 bright line spectrum in the neighbourhood of the nucleus 

 may be accounted for. 



The few oliservations I have made with regard to the 

 polarization of the light derived from the tails of comets 

 show that there is usually but little polarization in the 

 neighbourhood of the nucleus, but in the further parts of 



