There are many people who are prepared to regard a book as learned if it is sufficiently scrappy and contains figures arranged in a tabular form. Yet figures when they deal with other than very simple things are almost invariably misleading; and the more so as they have such an appearance of exactness. It is easy for any two people to count the number of men in a room and to agree as to the result; but ask them to say how many tall men, how many with black hair, how many blue-eyed, how many straight-nosed--and you will get a different result each time. The figures will be exact--they cannot be otherwise--but your knowledge will be the reverse. If this is apparent in such a simple matter as the recording of physical characters, how much more apparent it is when an attempt is made to classify and generalise on men. Most books admit that there are not sufficient data on which to base conclusions, and then proceed to suggest conclusions. The whole science of criminology is illustrated by the composite photographs published gravely as contributions; for a composite is a photograph of nobody at all. It is obtained by the superposition of photographs of different persons, and is itself different from any of them. It may represent them all as they ought to be, but it does not represent any of them as he is. It is the criminal in the abstract--who does not exist. It conveys in itself a warning against averages, for it is a pictorial presentment of an average. An average is the mean of different numbers. In dealing with masses of people--feeding them, for instance--by providing a certain average supply for each, all may be satisfied; but whenever the average is applied to individuals it is misapplied, and one finds he has too much, another that he has too little. Measure two men; one is 5 ft. 8 in., the other 5 ft. 4 in.; the average height of both is 5 ft. 6 in., which is the height of neither. So when we have averages of height, weight, etc., given in the case of criminals, we know that we have been told nothing about any of them. The other physical characters of criminals in prison have been noted without any attempt having been made to ascertain whether, and if so when and how, they were acquired, and we are invited to contemplate a number of twisted and bloated faces, many of which could easily be matched among the non-criminals. See these men and women before debauchery has left its mark on them and they are no uglier than some of us who are set over them.