256 
NATURE 
[May 9, 1912 
elevated perch, such as a bough of a tree or a rock, 
nearly always finish their flight in an upward direc- 
tion; but neither this nor wing flapping is at present 
open to flying machines on account of the mechanical 
difficulties of construction. 
Alteration of the trim of the wings, however, pre- 
sents no great constructional difficulty, but when the 
angle between the wings and the path is large the 
effect of accidental variations of pressure due to eddy 
formation is more serious, and the instability is 
greater than when the angle in question, is the 
gliding angle; and here, therefore, automatic correc- 
tion would be very important. If this could be used 
successfully, a machine the flying speed of which was 
40 miles an hour and which had a gliding angle of 
1/7 could, as may be found from the resistance 
diagrams, reduce its velocity by alteration of the trim 
of the wings to 25 miles per hour before the weight 
ceased to be air-borne. Further, since for the whole 
time the resistance would average about one-fourth 
of the whole weight, the time taken in effecting the 
reduction of speed would be four times that required 
for gravity to generate the difference between 40 and 
25, being 15 miles per hour. During this time—2-7 
seconds—the average speed would be 32 miles per 
hour, and the machine would cover about 120 feet. 
These rough figures can be easily corrected from the 
curves giving lift and resistance for any particular 
machine, but there can be no doubt that it would be 
a substantial gain if the high speeds, which are 
becoming more and more common, could be quickly 
and safely reduced before reaching the ground. 
It is quite possible to imagine a flying machine 
made with lifting screws which would rise vertically 
from the ground and remain poised and stationary in 
the air; but no success has hitherto attended any 
attempts in this direction, partly because the inventors 
have not realised the very large blade area necessary 
for: reasonable economy of power. One way of 
realising the stationary condition would be to con- 
nect two flying machines travelling at the same speed 
in opposite directions with a length of rope and letting 
them circle round one another. No ‘banking ”’ 
would take place, as the centrifugal force of each 
would be taken by the pull of the rope. If the latter 
were shortened as far as possible, the pair would, in 
effect, form a single machine with a lifting screw. 
The experiment would be dangerous, and is not 
recommended: for trial, but is mentioned rather as 
indicating the size of the screw blades which the 
hovering type of machine would require. 
In taking a general view of the present condition 
of the art of flying, it must be admitted that much 
remains to be done before it ceases to be a fine: 
weather sport, and I think the right course to pursue 
would be to try to evolve a type of machine which js 
fairly safe even in turbulent winds, and can 
arise and alight on the smallest possible area. When 
the essential’ features of the design which secures 
these results are recognised, the machines may be 
specialised for war or other purposes, and additional 
improvements may be introduced for convenience, 
comfort, or speed. 
The opinion seems to be gaining ground that fly- 
ing machines are more likely to be usefully developed 
than dirigible balloons, and in this opinion I fully 
concur, more especially as regards the larger 
dirigibles, which I have always considered too frail 
and too liable to accident to be of much real service. 
All aircraft, whether heavier or lighter than air, 
will for some time to come be designed for the pur- 
poses of sport or war rather than for commerce, and 
although for war-machines cost tales a second place, 
it must be remembered that a dirigible costs rather 
more than a torpedo-boat, whilst a flying machine 
NO. 2219, VOL. 89] 
costs rather less than a torpedo. Further than this, 
there are very few services to be performed by a 
dirigible which could not be carried out as well, or 
better, by a flying machine, the only, and rather 
dearly purchased, advantages attaching to the balloon 
being its power of rising quickly and of leaving the 
ground without the necessity of taking a run; and I 
think the best policy for us would be, while recog- 
nising the occasional usefulness of dirigibles of 
moderate size (and building a sufficient number for 
experiment), to devote our attention chiefly to the 
elaboration of the most efficient means of destroying 
them. 
From the purely scientific point of view it cannot 
be said that the ascents of any large balloon have 
added much to our knowledge. 
The small balloons, however, recently used for 
carrying self-recording instruments have ascended to 
heights (60,000 feet or more) at which personal 
observation is impossible, and have brought bacl: 
valuable information which could scarcely have been 
attained in any other way; and although the records, 
as a rule, only deal with pressure and temperature, 
there is no reason why solar radiation should not alse 
be measured by suitable apparatus. Such measures 
would give a better knowledge of the temperature of 
the sun than could be got by direct observation, even 
on the highest mountains. 
In conclusion, and speaking generally, I may say 
that it seems desirable to encourage experiment on 
the widest scale, even if much of the work is not on 
strictly scientific lines; bearing in mind ‘that great 
improvements may result from the working out of 
ideas which, as originally conceived, were unsound 
or even absurd, and that this is the more likely to be 
the case in such a subject as flight, for which, as I 
have endeavoured to point out, a considerable part is 
not yet subject to accurate theoretical treatment. 
APPENDIX. 
The relative densities of different gases at the same 
altitude may be conveniently expressed in terms of 
heights of homogeneous atmosphere of each. 
The height of the homogeneous atmosphere for a 
gas is defined as the height of a column of the gas 
of uniform density (equal to that which it has at sea- 
level) the weight of which produces the atmospheric 
pressure at its base. Thus the height of the homo- 
geneous atmosphere H, for air is in feet the number 
of cubic feet which weigh 2100 Ib. nearly, and since 
1 cubic foot of air weighs o-o80 Ib., H,=26,000 feet 
nearly. ct 
For hydrogen H,=H, x the ratio of the densities 
of the two gases (namely, 16), so that H,=416,000 
feet nearly. 
If the distribution of temperature in the atmosphere 
is isothermal, the actual height (h) above sea-level at 
which the pressure is p is h=H log Po Thus when 
h=H the pressure is p,/e, and the pressure does not 
vanish until an infinite height is reached. 
If, on the other hand, the temperature decreases 
according to the adiabatic law (that is, if the tempera- 
ture of the air at height h and pressure p is what it 
would be if with surface temperature to start with it 
was lifted without loss or gain of heat to the given 
height), 
at (s-(2) so A (-(2)) 
In this case, therefore, there is a definite upper limit 
to the atmosphere, for when p=o, h=H in (rather 
more than 17 miles for air and for 
275 miles 
hydrogen). 
ge 
