With all the precautions then available, it was found that mistakes in identification involved unjust suffering. A man named Coyle was sentenced for larceny in 1889, a Millbank warder swearing to his previous conviction, ten years before, as one Hart. The jury having examined Hart’s photograph gave a hostile verdict, the distinctive marks of the two men were found to be different, and Coyle moreover showed that he had been doing a short term when Hart was in prison. As is wisely stated in p. 23 of the Report: “The true test of the efficiency of a system of identification is not the number of identifications made, but the number of mis-identifications, or of failure to identify.” A woman lacking her left breast was identified with another who had suffered in the same way, and who had been previously convicted. It became clear that the women were different, and poor Eliza ---- had her punishment accordingly reduced from seven years’ penal servitude to six months’ imprisonment. A case in 1908 was that of two men charged with burglary, both of whom were short of a fore-finger, and were about the same age and of similar appearance. A man named Blake was found under circumstances that suggested an attempt at burglary, and was identified by a constable and several others, including a prison warder, as a convict called Steed, under supervision. It was found, however, that Blake had clearly been at liberty when Steed was in prison, and the former was promptly acquitted. One Callan was convicted as an incorrigible rogue, but had been identified wrongly with another man, he himself at the time of the alleged offence certainly having been in St. George’s Workhouse. He was, however, afterwards rightly convicted for a similar offence. It would appear from these and numerous other cases not referred to in this Report, that those mistakes affect only the criminal class. Probably there is a little too much readiness to identify a known rogue with the offender wanted, and those unfortunate victims often of disease and early training deserve fair and just dealing. Alas, however, the really innocent have sometimes suffered dreadfully from judicial blunders. The famous Beck case is too recent and tragic to require recall. But, besides occasional false identifications of innocent persons, the old system, now happily superseded, was admittedly very ineffective in detecting old offenders passing under different disguises and with false names. The time spent on each identification of old offenders was very great, an average of eight hours being required for one identification. A few minutes is now found to be sufficient.