248 REPORTS FROM THE SECTIONS. 
as before, but each thousandth of an inch that the lenses were sep- 
arated made the definition worse. In this case the curve of the 
second surface has a smaller radius than the third, so that the lenses 
put together without foil touch in the centre, and the maker sent 
them out with pieces of foil just thick enough to support the ~ 
edges, and at the same time let the lenses touch in the centre, and 
a beautiful batch of Newton’s rings marked the point of contact. 
hese lenses, as usual, were supported only on three points, 
but in this case I found that with full aperture and high powers 
there was a decided tendency to a triangular figure in the rays 
about bright stars ; I therefore removed the three points in the 
brass cell on which the lenses rested, and allowed them to rest 
directly on a plain brass ring touching all round, and I put three 
more pieces of foil between those placed b by the maker, so that the 
lenses were supported on six points inside, and on the brass rings 
Gees ; this nearly cured the triangular figures, and I determined 
to put a piece of paper in the form of a ring so as to support the 
lenses all round. Now several difficulties arose — this ring of 
paper was to be 114 inches in diameter, and -+},th of an inch 
broad, and more es aed ap exactly 0 010 ‘tick thick ; the 
m, was not a convenient thing to place, and in some parts it 
extended farther from the edge thas was desirable. I removed it 
and tried to replace it with foil exactly 0-010 inch thick, and here 
a fact became evident that has had a great deal to do withmy writing 
of these remarks. e thickness "Of foil which I wanted was 
0-010 inch ; the nearest I could command was 0-0105 inch, and this 
was ee in ‘and the lenses did not touch in the centre. e next 
oil measured 0-009 inch, and I tried this ; now the lenses 
not aaty touched in the centre, tet one would spin freely on the 
other as if it did not touch Ke foil at all; I was obliged there- 
fore to use 0:0105 inch, and twelve pieces oe this were , carefully 
placed round the edge of the lower — and then the upper one put 
on it—no oe appeared, and I put my finger on the centre and 
o see if I could produce them, but without effect. 
You will thas that the difference in thicknesses of tin foil on 
for the lower lens was only supported in ‘the centre, bend un 
lenses as now mounted cannot be more than 0-0005 inch, and yet — 
two lenses, when pressed in their centres, will not 
