and then they assigned them to a place in the scale of values according to the results of this analysis. They did not think it imperative to observe whether or not the organism would easily and economically assimilate any particular food, or whether the factor of appetency should be taken account of. They did not inquire whether there were elements in cheese, say, which the organism might resent, so that instead of this being a good article of food it might be nearer a poison.
We seem to-day to be abandoning the practise of relying wholly upon the analytic method of determining food values. We are now attaching chief importance to observing how the organism reacts upon any article of food when it is taken into the system. This method is likely to modify greatly our conception of food values. It is of special significance for our purpose that it seems already to have been shown that an article of food which may be of great service to the organism at one period of its development may be relatively valueless or positively detrimental at another period. It has been shown, for example, that while a calf may thrive on milk during the first months of its life, still if it be kept on a milk diet too long it will begin to decline, and it will literally starve unless other foods are added to the dietary. But the chemical methods of determining the values of foods make out milk to be a valuable food without regard to age or individual differences.
The principle under consideration can be illustrated further by referring to the methods employed in an earlier day in the study of the parts of speech in children's language. Men like Hale and others wrote down the words an eighteen-months old child, say, used in his daily expressions. Then they went to work and classified these words according to their grammatical properties, so that they found that 60 per cent. of the words a child used were nouns, 20 per cent. were adjectives, and so on. Then they inferred that the child's thought relates mainly to objects as contrasted with actions, and qualities, and spacial relations, since nouns predominate in his vocabulary, and they denote things. But to-day we appreciate that the outward form of a word does not furnish clear evidence of the way it functions in the child's expression. He may use the word "cat," say, with verbal, adjectival, and exclamatory as well as pure nominal function. That is to say, the fact that "cat" is grammatically a noun does not show that the child employs it as such. Indeed, it is certain that at the outset he does not use it with strict nominal function. The only effective way to determine the parts of speech in a child's vocabulary, viewing the matter from the standpoint of the function of words in expressing thought, is to observe the child as he reacts upon his environments when he uses particular words, so that we may notice what his attitudes are when he employs them.
This principle is mentioned here simply to impress it as of special importance in application to the study of educational values. In order
