The other mistake is odder still, though it may also be attributed to Rector. Robert Hyslop's second wife was named Margaret, familiarly called Maggie. Now, although it was impossible to misunderstand when Robert Hyslop was talking of his wife, this name Maggie never came correctly. Professor Hyslop waited a long time without rectifying the mistake; he waited for the communicator to perceive it and correct it himself, but this spontaneous correction was not made. At last he wanted the matter cleared up, and Dr Hodgson explained that the name of Professor Hyslop's step-mother had not been given. Rector, failing to understand, gave up his place to George Pelham, who began by administering a tolerably sharp scolding to the sitters. "Well, why do you not come out and say, Give me my step-mother's name, and not confuse him about anything except what you really want? By Jove! I remember how you confused me, and I don't want any more of it. I am going to find out, and if your step-mother has a name you shall have it." George Pelham went out of the "machine" and returned shortly, saying, "I do not see any reason for anxiety about _Margaret_." Margaret was really the name asked for, but one would have expected to obtain it in its more habitual form, Maggie. However, it is easy to understand that Robert Hyslop should not have given the familiar name of his wife to a stranger like George Pelham. While Professor Hyslop was preparing his report, a number of his friends who knew of his researches asked him what proportion of truth and error he had met with in these manifestations. This frequently-repeated question suggested to him the idea of making tables in which this proportion should be made clear at a glance. This kind of statistics would be important for the class of persons who think themselves stronger-minded than the rest, and who tell you, "I only believe in the eloquence of figures." Such people do not realise that battalions of figures are like battalions of men, not always so strong as is supposed.