(b) Investigation of forgeries, erasures of vari ous kinds, and overwriting.
In work of the first group photography is used only to make a permanent record of the conditions revealed by examination of the documents by means of lenses of high magni fying power, and the photographic work is usually directed by the expert entrusted with the examinations. The originals to be compared must be copied under identical conditions with extreme sharpness, and enlarged to the same scale, so as to show the hesitancies of a forger in imitating a handwriting, or the individual defects of the type-letters which may differ tiate typewriters of the same make.' The photographer plays a more active role in work of the second class, which often allows of the detection of differences which it is im possible to observe visually. The document is photographed with an enlargement of six to ten times linear, if possible with a lens of short focal length.
Erasures are brought out by using a lighting which just grazes the surface, or by means of transmitted light. 2 The slight traces of yellow remaining after chemical erasure of writing can be accentu ated by photographing with an ordinary (non colour-sensitive) emulsion, or, better still, by wet collodion, the document being illuminated by a light rich in violet and ultra-violet rays, such as an ordinary mercury arc.
Differences between two inks of the same appear ance can sometimes be ren dered visible by placing the document for about 8 days in contact with a print out paper, which is then exposed uniformly to light, the text (or a mechanically or chemically erased text) appearing lighter or darker than the ground.
Another more general method for differentiating between inks of the same appearance consists in photographing the document several times, each negative being made with a light-filter of different colour ; differences of composition are then often revealed by differences in the depth of the lines.
Finally, photo-micrographs of low magnifica tion (X 2o approximately), either single views or stereoscopic views, sometimes permit of ascertaining in which order two lines that cross have been. drawn, and thus of detecting addi tions made after the writing of the main text.
742. Copying a Plane or Cylindrical Surface in Successive Strips. It has been suggested that when copying at one sitting originals of which one dimension exceeds that of the largest useful dimension of the camera, the copy should be made by moving both the original and the sensitive surface at uniform speeds correctly proportioned (C. A. Bruere, 1923).
Let us suppose (Fig. 188) that the lens 0 projects on to a sensitive surface p the sharp image of the plane P, both perpendicular to the optical axis, and that these two planes move relatively to each other at such speeds that when the point B has reached the point A at present on the optical axis, the point b of the sensitive surface which was receiving its image will itself have replaced, on the optical axis, the point a (image of A). Under these conditions (ratio of speeds equal to the scale of reproduction) each point of the image will remain in invariable position on the sensitive surface during the whole of its displacement in the field of the lens.
The sensitive surface may be wound round a cylinder c turning at the same peripheric speed, on the one condition that an opaque shield E pierced with a slit F parallel to the axis of the cylinder must be used to cover the portions of the cylinder other than those practically merged in the tangential plane /.
Under the same conditions and with the same proviso, the flat original P could be replaced by the cylinder C with a peripheric speed equal to the speed of the plane P. This arrangement has been especially used in photographing metal conduits subjected to erosion tests in various soils (R. Davis, 1925).
The interior walls of a cylinder (rifle barrel) have been photographed by displacing, following the axis of the said cylinder, a periscopic device forming a same-size image of a fraction of the surface on a photographic film moved, in a suitable direction, at a speed equal to that of the periscope. If, for instance, the image includes I/6th of the circumference, it will suffice to repeat the operation 6 times (each time after rotating the prism of the periscope ifith of a complete revolution) in order to obtain the entire image (I. C. Gardner and F. A. Case, 1.926).