Geological Society. 73 



each tension \\\\e\\ the two disks are at any distances asunder 

 exceeding 20 millims. 80 far as we have been able to ascertain, 

 it seems to arise partly from the sudden expansion of the gas 

 throughout the extent of the Crookes's layer as soon as the light 

 is turned on (an expansion which acts on the swinging disk as if 

 a feeble explosion had taken place in front of it), and partly from 

 a Crookes's force acting between the upperside of the tube and the 

 swinging disk (caused by the inside of the tube becoming sensibly 

 heated by the convection-current that commences as soon as the 

 Crookes's layer is established). We do not think that any appreci- 

 able part of it is due to the direct action of the convection- 

 current. 



The rest of the diagram represents the forces which we found 

 when the disks were at distances of 10 and 5 millims. asunder. 

 The forces which presented themselves at these distances are 

 to be attributed mainly to a true Crookes's reaction between the 

 disks ; and they seem to warrant the conclusion that Crookes's 

 reaction was manifested at a distance of at least 10 millims. in a 

 hydrogen vacuum, when the outstanding tension was as much as 

 5 millims. of mercury. 



At distances of from 20 to 80 millims. the very feeble force 

 acting on the glass disk in our apparatus seemed to vary abouD 

 inversely as the tension. As already mentioned, it appeared to 

 be nearly independent of the distance when the distance exceeded 

 20 millims. 



At distances of 5, 10, and 20 millims. the force on the swinging 

 disk made some approach to varying at each tei.sion inversely as 

 the distance. But, so far as may be judged from our measures of 

 such exceedingly feeble forces, there is a sensible deviation from 

 this law at most of the tensions. 



Moreover the diagram, taken as a whole, seems to suggest, in 

 conformity with the dynamical theory, that the law changes vrith 

 variations of density ; for if the law were the same at all the 

 observed tensions, the converging lines in the diagram should 

 converge to points in the axis of abscissas, whereas they converge 

 towards points lower dowTi. 



We will postpone the further discussion of the observations 

 already made with this apparatus until we can supplement them. 

 by others. 



GEOLOGICAL SOCIETY. 

 [Continued from vol. iii. p. 541.] 

 May 9th, 1877.— Prof. P. Martin Duncan, M.B., F.E.S., 

 President, in the Chair. 

 The following papers were read : — 



1. " On the Agassizian genera Amhhjpteriis^ Palr^oniscKS, Gyro- 

 lepis, and Pyqopterus.''^ By Ramsay H. Traquair, Esq., M.D., 

 r.R.S.E., F.G.S. 



2. " On the Circinate Vernation. Fructification, and Varieties of 

 Sphenopteris offinis, and on Str(pJi7j7oj)teris ? PeacJiii, Etheridge and 



