2 Stromeyer, Parallax Determinations by Photography. 



case the normal distance of any object measured along 

 the axis of the photographic objective is exactly equal to 

 the product of the base line into the focal length divided 

 bv the micrometric displacement. Thus, in the case of 

 landscape views, if we shift one of the slides relatively to 

 the other so as to be set for a certain distance, say i,ooo 

 yards, we can draw a curve through those regions which 

 register correctly, and which appear of uniform shade. 

 This curve would be a section line corresponding to a 

 distance of i,ooo yards. This process can be repeated 

 for every lOO yards or other intervals. 



If the cameras are placed in front of each other, the 

 planes of the negatives still being parallel, the contours 

 obtained are intersections of hyperboloids of rotation 

 with the undulating land, and it is difficult, but not 

 impossible, to reduce these results. 



If the two cameras are placed at an angle, the reduc- 

 tion of the resulting photographs has to be made by the 

 present rather tedious methods, or a negative and a 

 transparency have to be illuminated by two lanterns 

 tilted to the right angle. In this case there will be much 

 inconvenience due to the photographs not being in focus 

 over their whole surfaces. 



The instrument now shown has been made of suffi- 

 ciently small dimensions to be used in one lantern. It con- 

 sists of two light frames with clamping arrangements, the 

 negative being clamped to one frame and the transparency 

 to the other. One of the frames carries two vertical 

 adjusting screws and one horizontal one, with which the 

 other frame can be tilted, lifted, or moved horizontally. 



The accuracy of even this very simple instrument is 

 such that the turning of the micrometer screw through 

 one-hundredth revolution is detected on the screen ; this 

 represents an angle of only 5" on an 8 in. focus, and far 



