566 Notes. 



be liable to defects of the same kind and order of magnitude as those it was 

 required to detect. This led the authors to construct an apparatus based 

 upon interference methods, by which the above difficulty is done away with, 

 as the measurements of the displacements are made directly in terms of a 

 known wave-length of light. 



The interference rings produced are viewed through a Microscope, in the 

 eye-piece of which are three cross-wires, one central and at right angles to 

 the other two. 



The definition and separation of the rings are sufficiently good to enable 

 one to estimate Jg of the distance between two rings fairly correctly. 



Since a shift of one ring to the next, past any given point, is equivalent 

 to one wave-length alteration in thickness of the path which is being 

 measured, a great degree of accuracy is attained. 

 A A full description of the instrument and its mode of application is given. 



The Paeallel Plate Micrometer.* 

 By J. II. Pvynting, F.B.S. 



If a parallel plate of glass is interposed between the objective of a Micro- 

 scope and the object, the image is seen in its true direction when the plate is 

 perpendicular to the axis. When the plate is tilted the image is shifted 

 sidewise, and by an amount which, for angles less than 10 degrees, is very 

 nearly proportional to the tangent of the angle of tilt, and for such angles 

 when a low-power objective is used, the definition is not appreciably im- 

 paired by the tilt. 



To use the plate as a micrometer, it may be fixed to one end of an axis 

 which turns in bearings and is perpendicular to the axis of the Microscope. 

 A pointer attached to the revolving axis moves over a straight scale, and the 

 number of divisions of the scale from the centre is proportional to the 

 tangent of the angle of tilt, and therefore nearly proportional to the shift of 

 the image. Suppose that it is required to measure the diameter of a small 

 particle. The plate is tilted so that one side, a, of the particle is on the 

 crosswire in the eyepiece, and the position of the pointer on the scale is read. 

 The plate is then tilted so that the other side, b, of the particle is on the 

 crosswire, and the position of the pointer is again read. The difference of 

 the two readings gives the diameter in scale divisions. The value of a scale 

 division may be determined by using as object a finely divided scale. 



The micrometer may be entirely detached from the Microscope, so that in 

 manipulation there is no risk of disturbing the Microscope. There is no 

 backlash. 



For powers higher than 1^ or 1 in. there is insufficient space for the plate 

 between the object and the objective, and the tilting affects the definition. 

 The plate may then be interposed in the tube between the objective and the 

 eyepiece, and in this position it is, of course, much more sensitive, and the 

 definition remains cood. 



In a measuring bench or comparator in the Physical Laboratory of the 

 Birmingham University, we use two Microscopes with 2-in. objectives. The 

 plates are 6 mm. thick, the pointers are 25 in. long, and move on millimetre 

 scales with about 100 divisions for a shift of 1 millimetre. 



The parallel plate micrometer was described by Clausen as far back as 

 1841. f It was re-invented by Porro in 1812.J Porro used both the form 

 described above and a double image form for the Telescope. In the double 

 image form there are two plates, each occupying half the field, placed in 



* Printed in extenso by permission of the author. 

 t Ast. Naoh., xviii. (1841) col. !)5-%. t C. R., xli. (1S55) p. 1058. 



