The puzzle is that a given individual should “hit,” as it were, at approximately the same point in the distribution of nearly every function. VI. CAN AN INTELLECT BE TRAINED AS A UNIT? Studies of the learning process also give light upon the organization of capacities. The question here is as to whether training in one function spreads equally to all other functions. Is it possible to “train the mind” as a whole? Will it raise the proficiency of all performances fifty per cent, if a fifty per cent gain is achieved in Latin composition? Numerous attempts have been made to determine the extent to which skill acquired in one performance increases skill in other performances. The conclusion which emerges from these studies is that _intellect cannot be trained as a unit_. Transfer of training from one function to other functions is far from complete. Apparently, there is spread of improvement from practice in a function only to such other functions as have elements in common with it. If two performances differ in any way, there is something in the second that remains _untrained_ by the practice given to the first. If two performances differ in all respects, the second seems not to derive any benefit at all from training in the first. To a very highly intelligent individual, nearly all situations and performances tend to have some identical elements, no doubt. To a very dull person, relatively few situations or demands present identical elements, for the dull perceive only gross similarities and differences. Thus, spread of improvement is without doubt greatest for the innately gifted, and least for the innately inferior minds. In connection with the present discussion, however, the chief point of interest is that no mind, of whatever degree of innate integrity and sensitivity, can be trained as a unit. Each function has elements special to itself, and some functions are very highly specialized, as regards the amount of transfer of training from them to others, or from others to them. The evidence from learning, therefore, substantiates the evidence from heredity, indicating that intellect is not a unit, but a complex of many capacities, coinciding mysteriously in amount to a very marked extent in an individual. VII. THE HIERARCHY OF ABILITIES