fi.—_ GEOGRAPHY. 93 
for it. How is it to be obtained ? Will stereoscopic survey do anything 
for us ? 
I feel certain that the answer is Yes; but much less certain thatthe 
photographs should, at least in the first instance, be taken from the air, 
which seems to be contemplated by the advanced school, and taken for 
granted by the newspapers. Air photography made a brilliant success 
jn the War, when the cost was not too severely scrutinised. It did its 
best work in France, but demanded a pretty close plane table survey of 
well-marked points, to give it a rigid skeleton. In the East it also did 
well, but perhaps on the ground that any sort of patchwork mosaic was 
a good deal better than nothing at all. In peace we have to approach 
the problem with the fear of the Treasury in our hearts, and with more 
respect for that sort of precision which lets one go on in an orderly way 
for ever, without leaving accumulations of errors ‘ to be absorbed in the 
_ desert,’ as they say in the Sudan. 
Now, photographs taken with axis vertical cover a surprisingly small 
ground, even from extreme heights. With a lens of 6 in. focal length, 
about the minimum, you must go to 25,000 ft. to get a result on the scale 
1/50,000, and then you can photograph only about three miles square on 
each plate. Flying at ninety miles an hour you must take plates every 
few seconds to avoid getting too much stereoscopic relief. It looks as if 
vertical photographs combined stereoscopically will fail in mountainous 
country. 
I turn to obliques. The photographs taken in the air are taken from 
unknown points, and the first thing to do is to determine the position of 
the camera at the instant of exposure. This requires at least three 
recognisable fixed points on each plate ; and the first adjective is as essential 
as the second. The geometry of the method is none too strong, anyhow, 
and we could not expect to find the resulting place of the aeroplane with 
anything like the accuracy of a ground station. This leads me to think 
that stereographic survey from ground stations will be found to play an 
indispensable part in the future survey of mountainous country. 
Suppose, for example, that political difficulties did not exist, and that 
we were able to survey the country south of Mount Everest. I think I 
would rather start out with a series of camera stations along the Singalela 
ridge, and fix all the visible crests stereographically with horizontal axes 
and vertical plates. A large part of the ground would be dead ground ; 
_ but quite a good deal could be putin. Would this not solve the question 
of providing a fixed framework for the obliques from the air, perhaps 
‘combining each with a plate from a ground station, rather than in pairs 
of obliques? It will at any rate be worth the trial; and therefore I am 
anxious that we should not fail to exploit the relatively easy and in- 
expensive ground stations, while we are perfecting the vastly more difficult 
process of the oblique air photograph. And that is the reason why I 
would urge a start with the best of existing apparatus: though which that 
is I am not at present prepared to say, for there are several models in the 
field, including a new one by the ingenious Mr. Wild, already mentioned. 
I suppose there never was a time when it was more difficult than now 
- to forecast the future in surveying. We have seen already that Geodesy 
is in a state of flux; we are not even allowed to believe that the pole or 
the continents stand fast in their right places. The methods that were 
