440 Major L. Danviu. On the Method of 



nodal point Ni ; ef would be the same in the two cases, but gh would 

 be less in the second ca,se, and the curvature would therefore exceed 

 the tolerated limit : with two lenses giving an equal total distortion 

 at the margin, one should be passed and the other rejected. The 

 total distortion at any one point will not therefore give a measure by 

 which the lens should be judged, the greatest rate of change in the 

 distortion more nearly representing what is required to be known ; 

 and, as the rate of change is certain to be greatest at the margin, the 

 Kew certificate supplies the information required. 



The tourniquet has already been mentioned as an apparatus which 

 has been specially recommended for the purpose of testing photo- 

 graphic lenses ; by means of this invention, Commandant Moessard 

 obtains an excellent means of detecting distortion, but hardly of 

 measuring it in a way to indicate the curvature produced in an 

 image. It will be remembered that the lens can be revolved about 

 an axis which passes through the nodal point NI, whilst the eyepiece 

 remains stationary; the effect of this movement can be seen in 

 fig. 17 by imagining the lens to be stationary, whilst the object and 

 the eyepiece revolve about the nodal point, the arc abcdF being the 

 path traversed by the eyepiece. Let a be the image of the object 

 after the lens has been revolved through an angle /3, and e the posi- 

 tion where the image would be seen on the photographic plate ; for 

 there is no reason to believe that the line ea will coincide exactly 

 with the line eS/s ; if there were no distortion, b would be the image 

 as seen in the tourniquet, and the distance moved by the image from b 

 to a is what is measured by that apparatus. It will be noted that the 

 image a will be much out of focus if the lens has a fairly flat field ; and 

 that, after re-focussing, a' will represent the image, and b' the point 

 from which the measurement is taken ; this re-focussing will tend to 

 reduce any error which may be due to ae not being coincident with eS/s, 

 but such a movement in the middle of an operation is rather objection- 

 able on mechanical grounds. Putting this objection aside, it will be 

 seen that we do not get a ready means of finding the curvature pro- 

 duced in the image as seen in plan in fig. 18 ; for, if c'd' is the length 

 measured by the tourniquet when the lens is revolved through an 

 angle 0, then the sagitta of the curve is equal to 



a'V sec /3 cos a. c'd' sec 0. 



15. Achromatism. After Focussing in the Centre of the Field in White 

 Light, the Movement necessary to bring the Plate into Focus in 



Blue Light (dominant wave-length 4420), = 0' inch. Ditto 



in Red Light (dominant wave-length 6250), = 0' inch. 



The photographer may be said always to adjust his focus in day- 

 light, and if the actinic rays are not brought to the same focus as the 



