But psychological experiments have produced much more striking examples. Claparède arranged for a man, wearing the mask of a clown, to enter his lecture room while a lecture was in progress. The students were afterwards asked to pick out this mask from a series of ten, and out of twenty-three who attempted the task five only were successful--and even these successes were probably due, largely or wholly, to chance. The appreciation of distances, measured by the eye, is also very likely to be erroneous, the rule being that large distances are under-estimated and small ones over-estimated. A similar rule holds good of estimations of intervals of time. Errors of this kind are not pure errors of perception; they are due chiefly to lack of experience. A carpenter or builder would usually make a much more accurate estimate of the dimensions of, say, the side of a house than would the ordinary person; and astronomers who work with transit instruments have, as a class, very accurate perceptions of small intervals of time. It is chiefly lack of experience, also, which is responsible for the absurdly different estimates different observers will make of the number of people in a crowd. Dr. Locard states that, on questioning the policemen employed to keep order during a procession as to the number of people they estimated to be taking part in the procession, he obtained the figures five thousand, ten thousand, twenty-four thousand. The actual number, he states, was three thousand. And during another procession two middle-aged, intelligent, educated Paris journalists gave as their estimates for the number of people engaged, the one thirty thousand and the other three hundred thousand.