persons of genius to insanity with the similar liability of corresponding normal classes, there is one comparison which it is interesting to make. We may compare the liability of persons of genius to insanity with the similar liability of their wives or husbands. It is noted by the national biographers that in fourteen cases the wives or husband (there is only one case of the latter) became insane. We may be fairly certain that this is a decided underestimate, for while the biographers would hold themselves bound to report the insanity of their subjects, they would not consider themselves equally bound to give similar information concerning the wives, while in other cases it may well be that the record of the fact has been lost. If now, in order to make the comparison reasonably fair, we omit the cases of senile decay, and only admit two-thirds of the doubtful cases of insanity, we find that the proportion of cases of insanity among the persons of genius is 3.3 per cent. Among the conjugal partners, on the other hand (I have not made any allowance for second marriages), it is 2.4. Thus we see that on a roughly fair estimate the difference between the incidence of insanity on British persons of genius and on their wives or husbands is less than one per cent. When we bear in mind that the data on which one of our groups is based are much more complete than those on which the other is based, it is not hazardous to assert that British men of genius have probably not been more liable to insanity than their wives.
At the first glance it might seem that this may be taken to indicate that the liability of genius to insanity is exactly the normal liability. That, however, would be a very rash conclusion. If the wives of men of genius were chosen at random from the general population it would hold good. But there is a well-recognized tendency—observed among all the mentally abnormal classes—for abnormal persons to be sexually attracted to each other. That this tendency prevails largely among persons of eminent intellectual ability many of us may have had occasion to observe. What we see, therefore, is not so much the conjunction of an abnormal and a normal class of persons, but the presence of two abnormal classes.
With regard to the significance of insanity, it must be pointed out that, although there may be an unusual liability to insanity among men of genius, there is no general tendency for genius and insanity, even when occurring in the same individual, to be concomitant. Just as it is rare to find anything truly resembling genius in an asylum, so it is rare to find any true insanity in a man of genius when engaged on his best work. The simulation of it may occur—the divine mania' of the artistic creator, or a very high degree of eccentricity—but not true and definite insanity. There seem to be only two certain (and two or three possible) cases—mostly poets—in which the best work was done during the actual period of insanity. Periods of insanity may alternate with
