486 Scientific Proceedings, Royal Dublin Society. 
facilitate the understanding of the simple apparatus to be described 
in this paper if the nature of this chart is first briefly referred to 
(fig. 1). 
From left to right, over the rectangular area of the chart, are 
reproduced the successive colours, according to Newton’s scale, as 
derived from interference produced by increasing retardations in a 
crystalline plate of differing elasticities. The chart is thus, in the 
original, traversed by vertical bands of colour, which begin at 
black, and, passing through grays, white-grays, white, yellowish- 
white, yellow, orange-yellow, ete., etc., reproduce the first four 
orders of Newton’s scales—beyond which the tints become so 
diluted with white as to be useless. The scale, according to which 
these colours are distributed horizontally, is toa unit of 1x 10° mms., 
or millionths of millimetres retardation. 
The amount of retardation being made up of the product of 
two quantities—the thickness of the plate and its birefringence— 
we can assign to its place on this scale any suitably oriented section 
of known thickness and birefringence; 7.e., we can determine the 
interference colour to which it will give rise. Let the entire vertical 
height of the chart represent to scale a mineral section of the thick- 
ness 0°06 mms. We can now assign to any substance of known 
birefringence a place along the upper boundary of the chart, 
scaling from the left (according to the scale of millionths of milli- 
metres used to distribute the colours) the retardation proper to a 
plate of the substance possessing the thickness of 0:06 mms. Thus 
ul, for example, the substance be quartz, which has a birefringence 
of 0:009, we place it on this upper line, at a distance measured to 
the right of 54 x 10° mms. ; this being the product of the thick- 
ness and birefringence. The colour at this point will be that pro- 
duced by a quartz plate of the thickness 0:06 mms. But as the 
retardation is proportional to the thickness, we can, by joining this 
point on the upper boundary to the lower left-hand corner (the 
origin) of the diagram, obtain at once an indication of the tint 
which any thickness of quartz less than 0°06 mms. will occasion. 
We simply ascend this uniformly inclined line proper to quartz 
till we have risen in the diagram to a height corresponding to the 
thickness of our section, and at this point find the tint to be 
expected. Accordingly the diagram is divided, as shown by hori- 
zontal lines drawn at heights representing to scale thicknesses of 
