Silent Witnesses by Nigel McCrery

convince her they were coming to collect her evil soul.
    At last, when all these techniques had failed, he did what he should have done in the first place and searched the murder scene. It didn’t take long before he discovered a bloody mark on a door. Examining it more closely, he realized that it was a fingerprint, and a good one at that. He cut the plank bearing it from the door and took it back to the police station. He then took the prints of Francisca Rojas and compared the two. They matched. He asked Rojas if she had touched her children at all after she had found them dead. She said she hadn’t. If that was the case, he asked, how did her bloody thumbprint get onto the door? He showed it to her. Confronted with this evidence, Rojas finally confessed to murdering her two children with a rock so that she would be free to marry her young lover. She was convicted and sentenced to life imprisonment. The case is generally acknowledged as the first time that a fingerprint was used to solve a murder.
    The Rojas case did for Vucetich what the Ravachol case had done for Bertillon, and he became the most celebrated detective in Argentina. In 1896, Argentina adopted fingerprinting as its main system of identification, and by the first decade of the twentieth century, every major country in South America had followed suit. In England, Galton continued to struggle to devise a satisfactory classification system, but help was about to appear from an unexpected source.
    A civil servant named Edward Richard Henry was the inspector general of police in Nepal. In 1891 he introduced Bertillon’s system there, amending it to involve six measurements rather than eleven in order to make it simpler and faster to use.Even then, he still found the system too complicated, as well as vulnerable to the enthusiasm of the clerks taking the measurements, who often couldn’t see the difference when it came to “only a few centimeters,” and frequently got them wrong.
    While on leave in England, Henry visited Galton. The two men got along well, and when Henry returned to Calcutta, he took with him all Galton’s notes. Henry also saw how difficult the process of categorization was going to be. However, during a train journey in 1896, he suddenly realized, quite out of the blue, how the deltas (those triangular shapes found on the tips of the fingers) could be used to create a proper system of identification. They fell into several clear types. Henry observed that “these deltas may be formed by either the bifurcation of a single ridge or by the abrupt divergence of the two ridges that had hitherto run side by side.” Additionally, the triangular shape conveniently lent itself to geometric measurement. He realized that all he had to do was establish the limits of the triangle, or “the outer and inner terminus.” A line could be drawn between these two termini and the number of papillary lines that this line intersected then counted with a needle. This number was the core of Henry’s classification. The vast majority of fingerprints fall into the simple loop and delta system. There were occasional examples of what he termed “accidentals” (those prints that for one reason or another didn’t match any of the types), but fortunately these could still be absorbed into the general system, meaning that he now had a practical means of categorizing any fingerprint. By 1897, fingerprints had become the sole means of criminal identification in India. And by 1902 they were proving three times more successful at identifying criminals than Bertillon’s system.

    Fingerprints stored in modern police records, illustrating a “whorl” fingerprint. The Henry system described three basic fingerprint patterns—loop, whorl, and arch—which together constitute the majority of fingerprint variations.
    Considering that fingerprinting stood to supplant his own system, it is perhaps