380 Dr. N. Ernest Dorsey on the Surface- Tension of 



regions I could detect not the slightest shift ; so I think the 

 readings are entirely free from any error due to inaccuracies 

 in the ways. Any error in the other adjustments also causes 

 a shift in the positions of the crosses ; but this is uniform 

 throughout the length of the ways, and so can easily be dis- 

 tinguished from the non-uniform displacements due to faults 

 in the ways. 



Extraneous I) 1st urban ces. 



Having disposed of these difficulties we come to another 

 that gave great trouble. This is the disturbance of the liquid 

 by extraneous vibrations. After trying many plans I finally 

 placed the box containing the tray of water and the fork F 2 

 upon an iron slab (H) weighing about 150 pounds, and sup- 

 ported by long steel springs hung from a joist whose ends 

 rested in the walls of the building but which was free from 

 the floor above. It was hung so that when weighted for 

 work it was about three or four centimetres above the top of 

 a brick pier, and the intermediate space between it and the 

 pier was loosely filled with cotton batting. If the cotton is 

 packed neither too tightly nor too loosely, this arrangement 

 cuts out most extraneous disturbances. The dividing-engine 

 was supported on this pier. The fork which generated the 

 ripples, and the vessel of water were each supported on pieces 

 of rubber tubing, so that the water might not be disturbed 

 by any slight vibration of the stai:d of the fork. 



Influence of Glass Plate. 



Another point of prime importance is to have the glass 

 plate which generates the waves well wetted by the liquid. 

 If it is not, the crests will be so irregular and so distorted 

 that no correct setting can be made. A microscope-slide was 

 first used, but it was very difficult to get its surface in such a 

 condition that water would wet it readily ; and one lot of 

 slides I was entirely unable to use. 1 tried to clean them 

 with caustic potash, sulphuric acid, nitric acid, chromic acid ; 

 I even boiled them in chromic acid and still water w r ould not 

 readily wet them, and never under any condition did I 

 succeed in wetting them uniformly. I finally used a piece 

 of German thin plate mirror from which the silvering was 

 removed. This was easily cleaned, and there was no trouble 

 in keeping it in such a condition that water and salt- solutions 

 would wet it readily and uniformly. 



"When I began this work I always took one reading within 

 two centimetres of the fork, in order that I might run the 

 fork with as small an amplitude as possible, so as to avoid 

 reflected waves. The other reading was taken some five or 

 ten waves further from the fork. 



