1891.] Cloud Photography. 475 



where 6 and p are the drift and parallax as measured on the photo- 

 graphs, and t the interval in seconds between the pictures being taken. 



The method of reduction of the photographs first adopted and em- 

 ployed during the early part of the past summer was as follows : 

 Prints were made on albuminised paper of the set of four pictures, 

 two taken at each end of the base with an interval of time between 

 them, and they were mounted on stout cards in order to avoid the 

 usual curling up of the paper. When necessary, new fiducial lines 

 were then drawn in the proper direction through the points that had 

 been ascertained to represent the corrected position of the lines of 

 reference as before described, and these lines were extended to the 

 margins of the cards. 



If possible, five or six cloud-points were then selected in each print, 

 capable of satisfactory identification. A sheet of paper was next 

 procured, larger than the pictures, and lines intersecting at right 

 angles were drawn across it. Punctures were then made, by means 

 of a needle, through all the selected cloud-points in the four pictures, 

 which were successively placed over the reference sheet (termed here- 

 after the receiver), so that the fiducial lines upon the pictures coin- 

 cided with the lines drawn upon the receiver, thereby ensuring the 

 points of intersection being directly superimposed, and, by means of 

 a needle passed through the pricked holes, the marked cloud-points 

 were transferred to the receiver. 



This having been done in turn for all the four pictures of the set, 

 the points thus pricked off were joined by inked lines, those obtained 

 from the pair of pictures taken simultaneously being drawn in black 

 ink, and those from the other pair in red, by which a series of 

 parallelograms was formed, equal in number to the number of points 

 selected for treatment. 



The black lines or sides of these parallelograms then represented 

 the parallax of the several cloud-points, being proportional in length 

 to the tangent of the angle subtended by the base line at the altitude 

 of the cloud, whilst the red lines forming the other two sides of the 

 quadrilaterals represented on the same scale the drift of the cloud 

 daring the interval which elapsed between the taking of the two sets 



pictures. 



The measurement of these black and red lines provided the means 

 already explained of determining the height of the clouds and the rate 

 of their motion, the direction being given by the inclination of the 



o lines, of which the black one represented the base. 



In dealing with the direction of the drift when thus obtained from 

 positive prints, it has to be remembered that by the printing the 

 right and left of the pictures are transposed, so that the east is on the 

 ! left and the west on the right in a picture the top of which is directed 

 to the north. 



of 



t 



