So far as exact evidence on the colour of the hair goes, it points chiefly to a relative deficiency of red-haired persons among criminals. This may perhaps be accounted for. There seems to be a lessened power of resistance to disease among persons of brilliant pigmentation. The extensive anthropological statistics of the American War showed a very marked inferiority on the part of fair persons. These statistics have been criticised by De Candolle, who believes, however, that even with deductions they may probably still be accepted. Our evidence as to the proportion of bright-haired people in lunatic asylums seems to point in this direction. These red-haired people, with their “sanguine” temperament of body, are peculiarly susceptible to zymotic disease; they take scarlet fever, for instance, very easily, and suffer from it severely. Among the manifold risks of a criminal life the brightly pigmented person, with his sensitive vascular system, seems to be soon eliminated. § 4. _Criminal Physiognomy._ The science of physiognomy is still in a vague and rudimentary condition, although the art has long been practised with more or less success. There are, for instance, a large number of proverbs in which some of the most recent results reached by the criminal anthropologists of to-day were long ages back crystallised by the popular intelligence. Such are the Roman saying, “Little beard and little colour; there is nothing worse under heaven;” the French, “God preserve me from the beardless man;” the Tuscan, “Salute from afar the beardless man and the bearded woman;” the Venetian, “Trust not the woman with a man’s voice.” Many of the old physiognomists, especially the two greatest, Dalla Porta and Lavater, tell us how they immediately recognised criminals, although they sometimes ludicrously failed; and Lavater once mistook the portrait of an executed assassin for Herder’s. A criminal anthropologist of to-day, Professor Enrico Ferri, declares that out of several hundred soldiers whom he examined, he found one, and one only, whom his face declared to be a murderer; he was told that this man had, in fact, been found guilty of murder. Garofalo, the Neapolitan jurist, observes that he is scarcely deceived twice out of ten times. Nor is this acuteness of perception by any means confined to skilled observers. It is very commonly found among women. Many persons, on first meeting an individual, are conscious of an unfavourable impression which they succeed in out-living, but which is subsequently justified. Sometimes the revealing glance is found, perhaps with a shock of horror, in a face already familiar. It is a mistake to attempt to stifle such instinctive impressions as irrational. They are part of the organised experiences of the race, and, subject to intellectual control, they are legitimate guides to conduct.