Shape of Capillary Surface in Contact with a Cylinder. 837 



If the optical constants of doubly refracting media are to 

 be observed, the plates are to be mounted for rotation around 

 an axis parallel to the slit and passing symmetrically between 

 the faces. Otherwise the wedge effect is liable to produce 

 excessive distortion in the results. If columns are to be 

 used, plane parallel faces are to be cut at a definite angle to 

 the optic axes, as the columns cannot be rotated. 



The position of minimum thickness of the plate or crystal 

 may be recognized by the aid of two rotations, at right angles 

 to each other, and the reversal of motion of the interference 

 rings. It is thus easy to place the axis of rotation at right 

 angles to the beam. 



The interferometer used in the above experiments was an 

 improvised apparatus, made of 1/4 inch gas-pipe through 

 which water continually circulated. The angle between the 

 component or interfering beams being but 30°, or less, the 

 micrometer was sufficiently close to the observing telescope 

 to admit of easy manipulation. The arms have since been 

 elongated to over 1 metre in length each, and the available 

 free depth below the beam is 15 centimetres. There seems 

 to be no difficulty of increasing these dimensions in any 

 degree. In spite of the lightness of the apparatus, tremor 

 of ellipses does not seriously hamper the observations. 

 Suppose now a glass column 1 metre long with plane parallel 

 end faces is placed in one of the beams. The micrometer 

 displacement which restores the centre of ellipses to the 

 fiducial sodium line will be of the order of 50 centimetres, 

 measurable, so far as the interferences are concerned, to 

 5 X 10~ 5 centimetre. Hence fi — 1 must be measurable with 

 an accuracy of 1 part in a million. 



Brown University, 

 Providence, R. I. 



LXXXIX. On the Shape of the Capillary Surface formed by 

 the External Contact of a Liquid ivith a Cylinder of Large 

 Radius. By Allan Ferguson, B.Sc (LoncL), Assistant 

 Lecturer in Physics in the University College of North Wales, 

 Bangor*. 



THE following analysis is an attempt to find an approxi- 

 mate value for the first integral of the differential 

 equation of the capillary surface formed when a cylinder is 

 dipped vertically into a liquid. The writer's attention has 



* Communicated by Prof. E. Taylor Jones. 



