SCIENTIFIC DETECTION OF CRIME — SANNIE 357 



as a negative. We may then superimpose the positive and negative 

 exactly and, if the vahies of the whites and blacks are exactly the 

 same on the positive and the negative, the whites will compensate the 

 blacks and vice versa. If we then make a print through these super- 

 imposed films, all that will show are the canceling marks not belong- 

 ing to the dispatching office. It is thus possible not only to detect 

 these traces, but partially to decipher them, and so discover the office 

 responsible for the first cancelation. 



This method, which theoretically is very simple, is in practice ex- 

 tremely difficult to use, for the perfect compensation of the whites and 

 blacks presents us with a problem of photographic photometry whose 

 difficulties are well known. 



The study of written documents is still more difficult. Any treatise 

 on scientific detection, even of recent date, will show how scanty the 

 means employed by the experts are. They generally consist simply of 

 examination under the microscope and the application of a few chem- 

 ical reagents. More accurate methods exist, but they are much more 

 difficult to use and involve a full-scale physical laboratory. For in- 

 stance, the mineral constituents or, indeed, in some cases, mineral im- 

 purities in ink can easily be identified by spectrography. Spectro- 

 photometric analysis of the pigments in inks enables one ink to be 

 compared with another ; indeed, a method of photographic photometry 

 has been invented that can be used directly by reflection on the docu- 

 ment itself. It is easy to determine the amount of iron in inks once it 

 has been transformed into a colored compound. Variations in the 

 amount of iron in inks containing logwood dyes are particularly sig- 

 nificant, as these info do not normally contain this product. 



It seems probable, however, that the best methods of identifying and 

 comparing the pigments in inks are chromatography on alumina, selec- 

 tive chromatography on paper, and microelectrophoresis on paper, the 

 last of which is still being tested in my laboratory. By these methods, 

 with only one word as a test basis, the pigments in inks can be distin- 

 guished and their impurities revealed. Here, again, however, progress 

 has complicated the expert's work. The inks used in ball-point pens 

 are very different from the earlier ones and cannot be analyzed by the 

 same methods. As a result of their introduction, therefore, the expert 

 finds new problems facing him, and new methods have had to be de- 

 vised to solve them. 



Examination of -firearins. — There have been similar developments 

 with regard to firearms. Scrutiny under the magnifying glass or the 

 microscope is still the main means of identifying bullets and cartridge 

 cases; but we are now able to analyze, by chemical means, the traces 

 of powder and its combustion products in the barrel of a gun or 

 around the hole made by a projectile. Quantitative spectrography or 



