204 BULL SYSTEM TECHNICAL JOURNAL 



edges of the aperture, which lattiT are indicated by parallel straight 

 lines. Owing to the fad that the light valve aperture must have a 

 finite length in the direction of rotation of the cylinder (indicated by 

 the small rectangle in the center of the sketch), there is a certain 

 overlapping of the light pulses on the film. (This is, in fact, neces- 

 sary' for the production of solid i)lacks.) These are indicated dia- 

 grammatically at //. In sketch I are shown, from an actual photo- 

 micrograph, the variations in the image of the light valve as traced 

 out on the moving photographic film. Here the dashed lines repre- 

 sent the limits of the image as formed by one rotation of the receiving 

 cylinder. It will be noted that the images due to the opening of the 

 light \'alve in each direction form a double beaded line. These 

 double lines are juxtaposed, so that the right hand image due to one 

 rotation of the cylinder backs up against the left hand image due to 

 the next rotation, thus forming on the film a series of approximately 

 symmetrical lines of variable width. These are exhibited clearly 

 in the enlarged section of a picture, Fig. 5. It will be understood 

 that for purposes of illustration, the grating used as the test object 

 in the [^receding discussion has been represented as traversing the 

 spot of light at the sending end at such a high speed that the final 

 picture is close to the limit of the resolving power of the system. 

 Thus the photomicrograph shown in I must be viewed from a con- 

 siderable distance in order that its difference in structure from the 

 original object A will disappear. A practical problem in the design 

 of picture transmission ai)paratus is to so choose the speed of rota- 

 tion of the c\linder with reference to the losses in resoKing power 

 incident to transmission that definition is sul)staiitiall\- the same 

 along and across the constituent picture lines. 



There are, in general, two methods by which a transmitted picture 

 ma\' lie recei\'ed. One of these is to form an image of the light \-aIve 

 aperture on the sensitive photographic surface. When this is done, 

 in the manner described in connection with l-'ig. 13 the picture is 

 made up f>f lines of constant density and \ar\iiig width. .\ jiicture 

 of this sort is shown in Fig. 14. A merit of this kind of picture (when 

 recei\'ecl in negative form) is that if the structure is of suitable size 

 (()() to fio lines to the inch) it may be used to print directh- on zinc 

 and thus make a t\pogra|)hic printing plate similar to the earlier 

 forms of half tone, whereby the loss of time usualh' incident to 

 copying a picture for reproduction purposes may be a\-oided. A 

 disad\antage of this form of picture is that it does not lend itself 

 readily to retouching cr to change of size in reproduction. 



Another method of pictin-e reception is to let the light from the 



