NOTES ON FLYING MACHINES. 45 
or riding a bicycle. In larger machines this will have to be done 
by making the area of the tail variable for ascending or descending, 
and tilting one corner up or down for turning to either side. 
The ascending and descending may also be managed by making 
the framing between the universal joint and the crank shaft vari- 
able in length. By increasing this distance, more than half the are 
oni which the wings move is brought below the body plane, so 
the machine was previously balanced for horizontal flight 
the iesbatinis will make it ascend. 
Likewise the turning may be effected by making the upper end 
of one connecting-rod fast to a sleeve sliding on the midrib. The 
position of the sleeve determines the arc of that wing measured 
from the body plane, but the two wings will still make equal ares 
with the horizon as the body will be rocked from side to side. 
Regarding the proportion of weight to area, the models are 
comparable more to butterflies than to the majority of ying 
area of the waves thrown; or, that a beetle and a mot 
weight expend the same amount of strength in flying the same 
distance, and the result is produced by throwing the same number 
of cubic inches of air in each case 
It takes 14 lb. to stretch one ‘of these red elastic bands 10 
inches ; that is, when they are looped so that there are four folds 
of the india: rubber. ere is one movable sheave introduced, so 
that the barrel is turned by about 7 lb. when the machine is 
wound up tight.* 
The pitch of the equivalent plane when the connecting-rod 
equals 24 cranks is about 75 inches when arranged as in that 
model with the long tail, which is one of the oldest and least 
efficient I have, the proportion now used is the crank equals one- 
third the connecting-rod. These results are arrived at by a system 
of selection ; those machines that fly at all have been kept till 
some slight improvement, suggested by er opt or judgment, 
has been embodied in the next ones, the older machines are then 
broken up or ateneed. It is found that this renee method is 
more fruitful than attempting to make any elaborate calculations, 
although differently constituted minds may find the latter way the 
best. 
* Nore.—A marked cr tp diel has since been made by making the 
barrel flat instead of cylindri now a i rotary 
motion, and there isa — absence of that jerk that was so icuous as 
centres; the motive power is also much 
