Further, the statements of the number of all the nouns, adjectives, verbs, and adverbs used by a child of two years are of interest, although they present several errors: e. g., supper makes its appearance twice in the case of the same child under s, and enough figures as an adjective. For the three girls, in their twenty-fourth month, the results were:

Parts of Speech.First child.Second child.Third child.
Nouns285230113
Verbs1079030
Adjectives343713
Adverbs29176
Other Parts of Speech282511
Total483399173

A fourth child, brother of the first and second, made use (according to the lists kindly communicated to me by the author), in his twenty-fourth month, of 227 nouns—some proper names among them—105 verbs, 22 adjectives, 10 adverbs, and 33 words of the remaining classes (all these figures being taken from the notes of the child's mother).

From these four vocabularies of the twenty-fourth month it plainly results that the stock of words and the kinds of words depend primarily on the words most used in the neighborhood of the child, and the objects most frequently perceived; they can not, therefore, be alike in different children. The daughters of the astronomer, before their third year, name correctly a portrait of Galileo, and one of Struve. A local "tone," or peculiarity of this sort, attaches to every individual child, a general one to the children of a race. I may add that the third child (in England) seems to have been less accurately observed than the others (in Madison, Wisconsin). Great patience and attention are required to observe and note down every word used by a child in a month.

Without mentioning the name of Holden, but referring to his investigations, which, in spite of the defects mentioned, are of the very highest merit, M. W. Humphreys, Professor of Greek in Vanderbilt University, Nashville, has published a similar treatise, based on observations of his own ("A Contribution to Infantile Linguistic," in the "Transactions of the American Philological Association," 1880, xi, pp. 6-17). He collected, with the help of a dictionary, all the words that a little girl of just two years "had full command of," whether correctly pronounced or not, and whether they appeared exactly in the twenty-fourth month or earlier. He simply required to be convinced that every one of the words was understood and had been spontaneously used, and could still be used. He did not include proper names, or words (amounting to hundreds) from nursery-rhymes, or numerals, or names of the days of the week, because he was not sure that the child had a definite idea associated with them. The vocabulary thus numbered 1,121 words: 592 nouns, 283 verbs, 114 adjectives, 56 adverbs, 35 pronouns, 28 prepositions, 5 conjunctions, and 8 interjections. In this table irregular verb-and noun-forms are not counted as separate words, except in the case of defective verbs, as am, was, been. The author presents the 1,121 words according to their classification as parts of speech, and according to initial letters, not according to initial sounds, although he himself declares this an erroneous proceeding, as I did in discussing Holden's paper. The only reason for it was convenience.

In the adoption of a word by the child, difficulty of utterance had some influence in the first year; when the little girl was two years old, this had ceased to have any effect whatever. She had by that time adopted certain substitutes for letters that she could not pronounce, and words containing these letters were employed by her as freely as if the substitutes had been the correct sounds. In regard to the meaning, and the frequency of use dependent upon it, it is to be observed that the simplest ideas are most frequently expressed. When two words are synonymous, one of them will be used exclusively by a child, because of the rarer employment of the other by persons speaking in the child's presence. Here, too, the local "tone" that has been mentioned made itself felt; thus, the little girl used the word "crinoid" every day, to designate sections of fossil crinoid stems which abounded in neighboring gravel walks.

As to parts of speech, nouns were most readily seized; then, in order, verbs, adjectives, adverbs, pronouns. Prepositions and conjunctions the child began to employ early, but acquired them slowly. Natural interjections—wah, for instance—she used to some extent from the beginning; conventional ones came rather late.

The following observations by Humphreys are very remarkable, and are, in part, up to this time unique:

When about four months old the child began a curious and amusing mimicry of conversation, in which she so closely imitated the ordinary cadences that persons in an adjacent room would mistake it for actual conversation. The articulation, however, was indistinct, and the vowel-sounds obscure, and no attempt at separate words, whether real or imaginary, was made until she was six months old, when she articulated most syllables distinctly, without any apparent effort.

When she was eight months old it was discovered that she knew by name every person in the house, as well as most of the objects in her room, and the parts of the body, especially of the face. She also understood simple sentences, such as, "Where is the fire?" "Where is the baby in the glass?" to which she would reply by pointing. In the following months she named many things correctly, thus using words as words in the proper sense. The pronunciation of some final consonants was indistinct, but all initial consonants were distinctly pronounced, except th, t, d, n, l. These the child learned in the eleventh month. At this period she could imitate with accuracy any sound given her, and had a special preference for ng (ngang, ngeng), beginning a mimicry of language again, this time using real or imaginary words, without reference to signification. But an obscurity of vowel-sounds had begun again. After the first year her facility of utterance seemed to have been lost, so that she watched the mouths of others closely when they were talking, and labored painfully after the sounds. Finally, she dropped her mimicry of language, and, at first very slowly, acquired words with the ordinary infant pronunciation, showing a preference for labials (p, b, m) and linguals (t, d, n, not l). Presently she substituted easy sounds for difficult ones. In the period from eighteen months to two years of age, the following defects of articulation appeared regularly: v was pronounced like b, th (this) like d, th (thin) like t, z like d, s like t, r like w, j like d, ch like t, sh like t; further: