442 report — -1878. 



tube placed across the layer, closed at the ends by patches of the heater and cooler, 

 and with sides that reflect molecules by the same law that a polished surface 

 reflects light. By employing this conception Mr. Stoney had shown that for small 

 differences of temperature, and with a heater and cooler that are large, flat, parallel, 

 and at a fixed distance asunder, the polarisation stress will vary nearly as 



CP 



PT 



where G is the flow of heat, P the tension of the gas, and T its temperature from 

 absolute zero. 



By a wholly different method, first suggested by Mr. Fitzgerald in a letter to 

 ' Nature,' which consists in calculating the transverse stress in the same way as 

 Clausius has calculated the stress across the layer, Mr. Stoney had obtained the 

 following symbolical expression for the difference between these stresses, that is, 

 for the polarisation stress, 



ie f + J 



IV 2 (3m 2 -1)*, 



L 



where /* is the cosine of the angle made by the path of a molecule with the 

 normal to the layer, I the proportion of molecules moving at that angle, and V 2 

 the mean of the squares of their velocities. 



By applying the same method of investigation to the general case of gas 

 polarised in any way, Mr. Fitzgerald had obtained the expressions for the most 

 general stresses that can arise, radial and tangential, and had shown them to be 

 identical in form with the corresponding equations given by Maxwell in his electro- 

 magnetic theory of light. 



And, finally, by attending to the conditions indicated by the theory, Mr. Moss 

 had succeeded in keeping spheroidal drops in existence for upwards of an hour and 

 a half, and had ascertained that spheroidal drops may be supported on air 

 devoid of vapour. This last very important confirmation of the theory was estab- 

 lished by experiments on melted paraffin at temperatures at which separate experi- 

 ments had shown that no evaporation could be detected after a prolonged exposure 

 of the paraffin in vacuo. 



5. On the Cause of Travelling Motion of Spheroidal Drops. 

 By G. Johnstone Stoney, M.A., F.B.S. 



In the course of the experiments referred to in the last communication a 

 tendency in spheroidal drops to travel about over the surface of the liquid on which 

 th«y rest had to be checked. This motion, where it was not simply due to con- 

 vection currents in the supporting liquid, was traced to differences of surface 

 tension, and may, by suitable arrangements, be made very swift. Thus, if sphe- 

 roids of spirit of wine or ether be floated on a surface of warm water, the heavy 

 vapour pouring down from the floating drops will, as is known, very much diminish 

 the surface tension of the water wherever it comes in contact with it ; and, as it 

 cannot be made to pour down quite symmetrically all round, the drop finds itself 

 on one side of the position of minimum tension, and is accordingly hurried along 

 one radius of the rapid surface current which travels outwards in all directions 

 from this point. The consequence is a very beautiful spectacle, the drops dashing 

 about with an appearance of great sprightliness. 



G. The Stanhope "Demonstrator" or Logical Machine* By Robert Haklet. 

 * A model was exhibited. See ' Mind,' for April, 1878. 



