Clergymen, however, who also constitute an educated professional class, have a suicide rate which is only half that of the population as a whole, and this is undoubtedly due to the restraining influence of religion, which is much stronger in clergymen than in laymen. The relation of suicide to religion raises a number of curious and interesting questions, but, unfortunately, the religious factor is so involved with other factors in the complicated problem of self-destruction that it is almost impossible to isolate it so as to study it alone. For example, the suicide rate of Protestant Christians in the northern part of Ireland is twice that of Roman Catholics in the southern part; but here education comes in as a complication: the Protestants are generally better educated than the Catholics, and their higher suicide rate may be due to their education and not to the form of their religion. In Europe generally, the tendency to suicide is much greater among both Protestants and Catholics than among Jews; but here education, race, and economic condition all come in as complicating factors, so that it is impossible to credit the Jewish faith alone with the lower rate. In view, however, of the fact that the suicide rate of the Protestant cantons in Switzerland is nearly four times that of the Catholic cantons, it seems probable that Catholicism, as a form of religious belief, does restrain the suicidal impulse. The efficient cause may be the Catholic practice of confessing to priests, which probably gives much encouragement and consolation to unhappy but devout believers, and thus induces many of them to struggle on in spite of misfortune and depression. The Salvation Army, in attempting to lessen self-destruction by opening "anti-suicide bureaus" in large cities, and by inviting persons who are contemplating suicide to visit these bureaus and talk over their troubles, is virtually introducing a system of confession which, so far as this particular evil is concerned, resembles that of the Catholic Church. In view, however, of the conflicting nature of the evidence, and the extreme difficulty of disentangling religious factors from other important factors, I doubt the possibility of drawing any trustworthy conclusions with regard to the dependence of suicide upon religious belief. It may be said, as a matter of record, that the tendency to self-destruction is greatest among Protestant Christians, next largest among Roman Catholics and Orthodox Greeks, and lowest among Mohammedans and Jews; but the differences are not certainly due to religion. The dependence of suicide upon nationality and race presents a number of problems of great interest, but of extraordinary difficulty and complexity. I can state a few of these problems, but I cannot solve any of them.