In this method of limiting choice there is much that is attractive. I feel that attraction strongly. Under the exceptional conditions which exist at the Johns Hopkins University, a group system has done excellent work. Like all the rest of the world, I honor that work and admire its wise directors. But group systems seem to me to possess features too objectionable to permit them to become the prevalent type of the future, and I do not see how these features can be removed without abandoning what is distinctive, and changing the whole plan into the elective system, pure and simple. The objectionable features connect themselves with the size of the unit of choice, with difficulties in the construction of the groups, and with the attempt to enforce specialization. But these are enigmatic phrases; let me explain them. Obviously, for the young, foresight is a hard matter. While disciplining them in the intricate art of looking ahead, I should think it wise to furnish frequently a means of repairing errors. Penalties for bad choices should not be too severe. Now plainly the larger the unit of choice, the graver the consequences of erroneous judgment. The group system takes a large unit, a body of studies; the simple elective system, a small unit, the single study. Errors of choice are consequently less reparable under the group system than under pure election. To meet this difficulty the college course at Baltimore has been reduced from four years to three; but even so, a student who selects a group for which he finds himself unfit cannot bring himself into proper adjustment without the loss of a year. If he does not discover his unfitness until the second year has begun, he loses two years. Under the elective system, the largest possible penalty for a single mistake is the loss of a single study, one quarter of a year's work. This necessary difference in ease of reparability appears to me to mark an inferiority in group systems, considered as methods of educating choice. To the public it may seem otherwise. I am often astonished to find people approving irreparable choices and condemning reparable ones. That youths between nineteen and twenty-three should select studies for themselves shocks many people who look kindly enough on marriages contracted during those years. Boys still unbearded have a large share in deciding whether they will go to college, to a scientific school, to a store, to sea, or to a cattle-ranch. Their lives are staked on the wisdom of the step taken. Yet the American mode of meeting these family problems seems to our community, on the whole, safer than the English way of regulating them by tradition and dictation. The choice with heavy stakes of the boy who does not go to college is frequently set off favorably against the choices with light stakes of the boy who goes. Perhaps a similarly lenient judgment will in the long run be passed on the great stakes involved in group systems. I doubt it. I think it will