PROCEEDINGS OF SECTION A. 187 
tions of the standards in use at the Melboure Observatory, 
both metrical and English inch values. On carefully examining 
these I found a slight discrepancy between the inch scale, as 
copied directly from the standard, and the same values obtained 
by computation and ruling from the metrical standard. As 
I had no means of determining which of the two scales was 
most likely to be correct, I adopted the metrical scale as it 
stood as my standard for metrical values, and the inch values, 
as copied from the standard inch scale, as a standard for frac- 
tional values of an inch. At a later date I submitted several 
micrometer rulings to Mr. E. M. Nelson, a recognised authority 
upon ali matters connected with measurements of this cha- 
racter, with the result that it was found that the ratio of inch 
to m.m. was, in the case of my inch rulings, 25.3821, instead 
of 25.39997, but as the metrical values proved to be correct, 
in comparison with the best standards, I have since adopted 
this scale as a basis for both systems. It may be of interest to 
knew how I determine that lines stated to be ruled, say, at the 
rate of 90,000 per inch, are really of that value. For this it is 
only necessary to adjust the relationship of the wedge to the 
screw once for all, so that forty revolutions of the latter give 
a movement = .02 in., in which case one revolution will equal 
.0005 in. As the error in forty revolutions can easily be 
brought within 1/50,000th of an inch, the error in 1/40th of 
this is a negligible quantity. The screw head being divided 
into 360 dee. reading by a vernier to 1/10th of a degree, 8 
deg. cf movement of ‘the screw head advance, the plate being 
ruled the 90,000th part of an inch, and so, proportionately, 
for other values up to 120,000 lines per inch, the finest I have 
ruled, which have so far been resolved. . In passing, I may state 
that the finest lines it has been possible to resolve or separate, 
by means of the most perfect microscopical apphances hitherto 
constructed by the best makers, have not exceeded 120,000 per 
inch. 
I have as yet said nothing concerning the glass most: suitable 
for ruling upon. Ordinarily the outer. crust. or surface of the 
glass as it leaves the makers’ hands is much too hard and brittle 
for the purpose, and speedily ruins the hardest diamonds. This 
is especially so in the case of thin unannealed microscopical 
cover glass, which it is essential to use for many purposes. 
Hence it occurred to me that it might be possible to so modify 
and alter the surface of this glass by a process of annealing that 
better results would be obtained. After some few trials I 
found that by enclosing carefully cleaned cover glass in a metal 
‘capsule, and slowly heating to a certain point, short of actual 
softening, and allowi ing the cooling process to extend over as 
long a period as possible, the glass proved to be both softer and 
tougher, and at the same time far less liable to any alteration 
