It is not at all certain that there is any well marked boundary line between genius and some forms of imbecility. Many quite irresponsible idiots have marvelous verbal memories, and can repeat parrot-like, page after page of books of which they have no comprehension. Dr. Barr tells of cases of prodigies, musical, mathematical and mechanical, who except in their specialty were almost totally deficient mentally.[63] Many of the world's most brilliant musicians, mathematicians and even military leaders have been men of one-sided mental development, whose ability in other lines was so slight that they were little better than imbeciles, and it is not at all surprising that their children are sometimes truly idiotic. [Footnote 63: Barr, op. cit., p. 301 _et seq._] The best writers of the present day no longer recognize consanguinity as a cause _per se_, of idiocy. The heredity of neuroses, however, is so strongly established that few would dispute the proposition that where the morbidity is inherited through both parents it appears more frequently and in a more marked degree than where one parent is entirely free from taint. This is what occurs when a consanguineous marriage takes place between descendants of a neurotic family. The percentage of idiotic children would then be somewhat higher from consanguineous marriages than from the average marriage purely through the action of the laws of heredity. Dr. Barr finds 49 out of 4050 cases of idiocy or 1.21 per cent, in which there was a family history of consanguinity. This is little higher than the average frequency of first cousin marriage, and an analysis of 41 of these cases does not show one case that can be attributed to consanguinity alone. To quote: "Two were the result of incestuous connection--one of brother and sister, the other of father and daughter, and in the others there was an undoubted history, of grave neuroses."[64] "Beach and Shuttleworth find in the consideration of their 100 cases (out of 2,380 idiots), giving 4.2 per cent (of consanguineous parentage) that the bad effects are due rather to the intensification of bad heredity common to both parents."[65] [Footnote 64: Barr, op. cit., p. 94.] [Footnote 65: Ibid., p. 109.] Dr. Arthur Mitchell examined all idiots in nine counties of Scotland and found that 42 out of 519 or 8.1 per cent of whom the parentage was known, were children of first cousins.[66] Dr. Down found 46 out of 852 or 5.4 per cent to be children of first cousins.[67] Dr. Grabham of the Earlswood Idiot Asylum in Surrey, England, stated that 53 out of 1388 patients were the offspring of first cousins. The facts, he adds, were obtained from the parents and are "therefore tolerably trustworthy."[68] Other investigations give percentages as follows: Kerlin, 7; Rogers, 3.6; Brown, 3.5 and C.T. Wilbur, 0.3.[69] [Footnote 66: Darwin, see _Jour. Stat. Soc._, p. 173.] [Footnote 67: Huth, _Marriage of Near Kin_, pp. 210-211.] [Footnote 68: Darwin, op. cit., p. 166.]