different words used by 2 (b.), 479 or 64 per cent, were used in the first five hours or half of the day.
Such a record does not by any means include all the words which it would be possible for a child to use in one day, could it come into contact with its entire little world of experience by playing all its plays, looking at all its books, going on all its occasional visits, seeing all its acquaintances, living through all the days of the week and seasons of the year—that would involve almost its entire vocabulary up to that date. These two children had used respectively, up to two and a half years, 1,432 and 1,308 different words, almost all of which constituted their still usable vocabulary. So that on these days they used only about half of the words they might have used. When a child's world of named things is smaller it will naturally use a larger proportion of its vocabulary on any one day. For instance, the eighty different words vised by 3 (g.) at the beginning of the twentieth month were 96 per cent. of the vocabulary up to that day, but by the next month this proportion had dropped to 60 per cent. and in subsequent months varied from 54 per cent. to 43 per cent.
These children we believe to be but slightly, if at all, above the average child in the use of language. As an example of what a reputedly talkative child can do, we took the words used on his second birthday by Carl Andrist (whose father was instructor in French in the University of Minnesota); he used 803 different words.
Most of the records of children's vocabularies hitherto published have been gross underestimates, chiefly, we think, from two causes. In the first place, the estimate has almost always been made on the first or only child; for the enthusiasm for child study or psychology usually breaks out with the advent of the wonderful first child and lessens as this novelty wears away and heavier domestic cares with the second child discourage scientific ardor. But the later children have an advantage in learning much from contact with the older child. Thus our first child used only 400 words at two years, and 769 at two and a half years, that is, about half as many as the second or third child. Of three other published records of two children in the same family, two cases show a much slighter increase while one shows a decrease.
The second cause of the underestimation of a child's vocabulary is the use of unreliable methods. Even the almost constant presence of the mother with the child, and her daily noting of the new words it uses from the beginning of speech to the two or two and a half year limit, we found insufficient. For when in the case of the second child we also followed the child about and made a complete record, we found that the former method had failed to note many of the commonest words. We followed the third child through eight complete days, on the first of months 20, 21, 22, 23, 24, 25, 28 and 31; and, on checking
