388 REPORT—1883. 
by Professor Osborne Reynolds in teaching his class. We have here a 
central roller a, pressed on by three intermediate rollers B, By B3, which 
are all held in what I have called a nest ring D pressing the whole together. 
The pressure is caused and maintained by coning the rollers as shown 
in section. When the two halves of pulleys B are pinched together, 
they are wedged against p by A, and the necessary adhesion obtained, 
allowing D to drive A or A to drive p.! Mr. Foster has informed me 
that he considered his invention to be this mode of tightening, for 
that he had met with examples of similar nests in which an attempt 
had been made to get the pressure simply by initial fitting. Mr. Foster’s 
mode of tightening is ingenious, and a modification of it will probably be 
found very useful; the surfaces of the rollers do not rell true on each 
other, and although the friction from this cause is less than in the old V 
friction gear, it is considerably greater than we shall have in true rolling 
nest gear. The possibility of tightening by cones and by double-cones 
occurred to me independently; and at about the same time Mr. 
Williamson, a draughtsman then employed by Messrs. Ayrton and Perry, 
conceived the idea of a nest which was almost identical with Mr. Foster’s. 
My own favourite idea when I took out my first provisional specification 
in the spring of this year was to tighten the rollers by the means shown 
in fig. 4. A is placed excentrically to D; there are three rollers, B, By B3, of 
which B, is smaller than the two others; all the rollers have simple cylin- 
drical surfaces which would develop as planes; the tightening is effected 
by forcing one of the rollers, as B2, from a wider into a narrower part of 
the space between A and p. When the excentricity is not great a very 
moderate force on the spindle of the tightening roller will maintain a great 
pressure between the rollers. Mechwart, I have since found, employs 
a similar adjustment to vary the space between his laminating rollers, 
but, as I have already pointed out, he did not construct gear to be sub- 
stituted for toothed wheels in any case, whereas the nests which I am de- 
scribing are gear in the sense that they can be used instead of spur-wheels 
and pinions. Nests tightened by this excentric method are exhibited at 
Southport as part of a large winch made by Messrs. Stothert and Pitt. 
Tt has also been employed in a telpherage locomotive designed in the 
office of the Telpherage Co., and made by Messrs. Crompton. In both 
cases the gear works extremely well, and the rolling friction has proved to 
be smaller than I anticipated. I am not yet able to give accurate informa- 
tion as to this friction, nor as to the wear, but I can say that the friction 
is less than that of spur-wheels when these are transmitting considerable 
force, although more than that of spur-wheels when these are running 
quite slack. The efficiency of the winch when lifting 1] tons is about 
80 per cent. It contains two large nests, or has, in other words, double 
purchase, and the speed of the chain lifting the weight is about zisth the 
speed of the handles. The design of the winch is in several respects 
defective, the rollers B, B) B; are alloverhung and imperfectly supported ; 
they do not therefore lie with the axes quite parallel. This leads to two bad 
results. 1. There is a screwing action tending to move the rollers length- 
ways against the collars or flanges by which they are retained. 2. It tends 
to reduce the line of contact between the convex rollers to a point where 
the cylinders cross, and at this point wear occurs. It will be quite easy 
1 Since reading this paper I have found a patent by Mr, Tibbitts containing the 
same arrangements. 
