CLOUD OBSERVATIONS IN VICTORIA. 261 
fixed rules, which, once mastered, would require no further skill 
or judgment in the operators 'at the instrument, but only a 
methodical and conscientious attention to the rules. This was a 
necessary condition which I had to take into account in deciding 
upon my course. 
IT will now describe the main principle on which the Kew 
photographic method is based. 
The absolute height and velocity of a cloud, provided it be 
suitably situated, can be determined from two photographs of it 
taken simultaneously with two cameras placed at a distance of 
from a few hundred yards to 1 or 2 miles from each other, the 
cameras being in all respects equal and rigidly mounted so as to 
point accurately to their respective zenith—namely, having their 
collimation axes truly vertical. 
If two threads, at right-angle to each other, permanently fixed 
in a plane parallel to and very near the sensitive film, in such a 
way that the line passing through their intersection and the 
optical centre of the objective, is maintained in an invariable 
position relatively to the body of the camera, and falls perpen- 
dicularly to the face of the plate, then, whenever a picture is 
taken, these threads will be shadowgraphed on the plate, and the 
image of their intersection may be adopted as the centre of 
the plate. So that in all plates exposed in a truly horizontal 
position, a point in the zenith would form its photographic image 
at this fiducial centre. Such are the conditions required by the 
Kew method. 
Having photographed a cloud under these conditions we obtain 
on each plate a picture of the cloud and an image of the respective 
zenith of the two stations. Let us now superimpose the two 
pictures, so as to make them coincide exactly. 
To insure accuracy in this operation it would be necessary in 
practice to take a contact print on glass of one picture, and super- 
impose this positive on the negative of the other picture. 
Then it is clear that the line joining the centres of the two 
plates represents the photographic image of a line joining the 
two zenithal points of the stations at the height of the cloud, and 
the ratio of its length to the actual distance between the stations 
would be the same as the ratio of the focal length of the objective 
to the height of the cloud, and this height can thus be determined 
from the data obtainable on the photographs. The process of 
superposition is, however, somewhat unsatisfactory, and can be 
obviated by orientating the cameras so that one of the fiducial 
lines in each is made to coincide with the direction of the line 
joining the stations. 
It is then sufficient to measure on each negative the co-ordi- 
nates of any corresponding point in the image of the cloud, taking 
the centre of the plate as the origin. 
