254 



NA TURE 



[July i i, 1901 



In such a case the friction on the surface of the body, caused 

 by its motion through the resisting medium must be greater 

 on the side next the centre of the nebula than on the side 

 next its boundary. This difference of resistance must obviously 

 result in imparting to the impinging body a rotatory movement. 



Of course a tremendous translatory velocity would be required 

 to produce any sensible motion of rotation in the impinging body 

 itself. But by following Prof. Seeliger's reasoning it becomes 

 easy to understand how even a comparatively small translatory 

 motion suffices to originate enormous gyratory movements in the 

 strata of the atmosphere surrounding the body. Obviously, the 

 immediate consequence of a collision between body and nebula 

 will be a superficial heating of the former and the resulting 

 formation of an incandescent atmosphere around it. Now Prof. 

 Seeliger has pointed out that the attraction of the body on the 

 nebular mass through which it travels must greatly enhance 

 the relative velocity of those particles which pass near 

 the surface. In his opinion, " no extravagant assumption is 

 required to obtain very great velocities for these particles, 

 velocities such as have been proved to exist in the case of 

 Nova Aurigx. " Hence, even when the initial translatory motion 

 is small, the attractive force of the body would cause enormous 

 difi'erences of velocity between the impinging particles of the 

 cosmical cloud and the atmosphere of the intruding body. And 



The assumption made so far, that the rotation of the particles 

 takes place in circles concentric with the circumference of the 

 body, must, however, be modified. The fan-like action of the 

 body's atmosphere will draw in towards the poles of rotation quan- 



■^u 



it is easy to perceive that, in case of a preponderance of impacts 

 on one side over those on others, there must result a gyratory 

 movement of the atmosphere, the velocity of which will, in 

 course of time, become much of the same order as that of the 

 impinging nebular matter. 



The assumption of a cosmical cloud, the density of which increases 

 towards the centre of gravity, leads, therefore, to the necessary 

 conclusion that the incandescent gaseous matter near the body 

 must assume a vorticose motion of probably very high velocity. 

 This motion has its maximum near the surface of the body, 

 whence it will grow less with increasing distance from the 

 centre. 



.According to the fundamental laws of gyration the rotatory 

 motion must vanish at a certain distance, beyond which it will 

 assume the opposite direction. Let Fig. 2 represent a section 

 through the centre of the vortex in a plane perpendicular to its 

 axis of rotation. Let aa be the surface of the body, bb the 

 locus of the stationary sphere separating the two oppositely- 

 gyrating systems, cc the outermost boundary of the whole 

 system of gyration. Then we have between A and B a rotatory 

 motion of high velocity in one direction, decreasing in amount 

 from .\ towards B, and a rotatory motion in the opposite direc- 

 tion between B and c of less average velocity than the former. 

 The space from A to c is filled with incandescent nebulous 

 matter, the maximum incandescence being at A, whence it 

 decreases towards c. The space beyond c, on the other hand, 

 is filled with nebulous matter of low temperature and no rotatory 

 motion. 



The whole vortex travels, of course, along with the central 

 body in a certain direction. This obviously imparts to the light 

 emitted by every particle of the whole system exactly the zame 

 displacement, and hence the motion of translation may be left 

 out of consideration in questions dealing with relative velocities. 

 NO. 1654, VOL. 64] 



litiesof nebulous matter which, yield ing to centrifugal force, will 

 flow towards the equator, and be thence projected outwards again. 

 The nature of this action may best be seen by considering a section 

 of the system in a plane passing through the centre of the vortex 

 and in the direction of the axis of rotation. The figure so 

 obtained (Fig. 3) is precisely the same as that arrived at by 

 Dr. Siemens in his ingenious theory of the conservation of solar 

 energy. (Nature, ^larch 9, 1SS2. ) In fact, the conditions 

 postulated by Dr. Siemens in his theory — viz., that the sun is 

 surrounded by matter in a rarefied form, filliilg interplanetary 

 and even interstellar space — are precisely the conditions under 

 which the phenomenon of a new star is here supposed to occur. 



We have, then, to expect an indraught of cool nebulous matter 

 at the poles of the intruding body, and an outflow in all direc- 

 tions of hot nebulous matter at its equator. 



In spite of the apparent complexity of the different motions 

 involved in the gyration here described, it is comparatively easy 

 to indicate the influence they must have on the appearance of 

 the lines of a substance present in the nebular matter. Let us 

 first consider the influence of the /<j«^««/zVi/ components of the 

 gyratory motion. 



Reverting to Fig. 2, and assuming AD to be the direction of the 

 line of sight, it is clear that in the space a.adt) we have to deal 

 with an incandescent nucleus w whose light is intercepted by 

 incandescent matter at»a lower temperature between A and c, 

 and by dark nebulous matter of still lower temperature between 

 c and D. The resultant eftect would be exactly that which 

 Sir William Abney has described in M. A' xxxvii. p. 27S. The 

 displacements of the line in opposite directions from the normal 

 caused by the approach and recession of the limbs of the rotating 

 body and its atmosphere would broaden the absorption band, 

 which would therefore appear dark in the centre and would 

 gradually shade off towards the edges. The intensity curve of 



the band produced by this part of the gaseous envelope, still 

 provided there be no radial motion of matter in the vortex, 

 would thus be similar to that exhibited in curve A in Fig. 4. 

 Taking next the segment bab'b" on the left hand side of the 



