126 Trmisactiom. — Miscellaneous. 



increases the tube expands. At the most suitable time, while the rate of 

 rotation is increasing, let the revolving tube be bent round till the ends 

 touch ; join the two ends fast. We have now a rapidly revolving tubular 

 ring,* the length suitably chosen and the velocity of rotation having also been 

 suitably determined, the insides of the tube may be considered to touch, or 

 rather nearly to touch. The outsides of this tubular ring are going one 

 way, while the insides of the tubular ring, almost touching, are moving in 

 the opposite direction. If two masses of matter touch this tubular ring on 

 opposite sides they will both be impelled in the same direction. If two 

 masses of matter touch a revolving sphere on opposite sides they will be im- 

 pelled in ojyposite directions. 



The aether may be conceived to be made up of these tubular rings, which 

 may be called " ^ethereal corpuscles."! The axis of revolution is a circular 

 axis. One of the fundamental laws of mechanics is that no revolving body, 

 or system of bodies, can, by the mutual interaction of its parts, either ac- 

 celerate or diminish its rate of rotation. The ethereal corpuscle would 

 therefore continue to revolve with an undiminished velocity so long as it 

 came in contact with nothing. These corpuscles must be conceived as so 

 small that they will freely enter the pores of any solid. 



Mass for mass they may be considered as much stronger than steel, 

 that is as we know the strength of steel by subjecting it to a strain. In 

 particular it would be found advisable though not necessary to ascribe an 

 almost perfect elasticity to the matter of the shell of this tubal ring. If 

 this is not done, the corpuscles must be considered as of every size. It 

 would be best to proceed with the enquiry on the supposition that these 

 corpuscles are all of the same mass. If we alter the direction of our 

 enquiry this would not be necessary, though less satisfactory. 



These corpuscles constituting the sther, of course do touch each 

 other, but, their direction of rotation being the same, there would be no 

 conflict unless any of the corpuscles move slower than the contiguous ones. 

 In case of conflict, all contiguous corpuscles would adjust their directions of 

 rotation to the larger and therefore predominant mass close to them. We 

 must follow out this notion of adjustment : — If a mass of matter were sud- 

 denly placed in the aether — say, far from any other mass of matter, — and 

 kept immoveable in its place, the corpuscles would proceed to adjust them- 

 selves to it. For convenience of conception, a mass of matter may be con- 



* In Chambers' Encyclopsedia, article " Vortex," and in Tait's " Advances in Physi- 

 cal Science," (last lecture), will be found wood-cuts of smoke-rings. They will illustrate the 

 action, and to some extent the form, of this tubular ring. 



t See note to Sir John Herscliel's lecture on " Weather," where he speaks of " aethe- 

 real molecules." 



