176 



SCIENCE. 



[Vol. XVIII. No. 451 



complete vocabularies, and using no otber for several years. This 

 tendency to originate language is shown in almost every child by 

 the invention of new words or new uses for words. New terms 

 are often formed by imitating the noise made by the animal or 

 thing named, as "bow-wow" for dog, " choo-choo " for locomo- 

 tive, and sometimes by the repetition of a sound made in perform- 

 ing an act, or an emotional sound made at sight of a new object 

 or act. Any sound thus associated with an object, act, quality, or 

 state of feeling may be used by the child as a word, and, if the 

 parents or playmates accept it as having a certain significance, it 

 becomes fixed as a permanent part of the child's vocabulary. 



Baby talk, or the peculiar pronunciation used by children, and 

 frequently imitated in literature, is a subject of considerable in- 

 terest. How much of what is given as "baby talk" has really 

 been originated by children? How much of the incorrect pro- 

 nunciation of any particular child is due to his inability to pro- 

 nounce correctly, and how much to the foolish habit of mis- 

 pronouncing words when speaking to children practised by so 

 many fond parents? Of still more importance is it to know 

 whether there really is any general law of mispronunciation that 

 may be of practical value to the educator. In only a part of the 

 records sent me was the pronunciation used by the child indicated, 

 hence only the probabilities in regard to the law can be given. In 

 the first place, it must be understood that the ability to pronounce 

 words is entirely independent of the ability to understand their 

 meaning, and either capacity may be developed in advance of the 

 other. However, in the acquirement of new words, difficulty of 

 pronunciation may exercise some influence in preventing the adop- 

 tion of certain words into the vocabulary. Not all children are 

 influenced in this way; some adopt diflicult words but use a sound 

 easily pronounced in place of the one they cannot pronounce, 

 sometimes following a regular system of substitution. The law 

 of mispronunciation proposed by Noble {Education, 1888) seems 

 theoretically quite probable, and some of the facts support it, but 

 not enough have been collected to establish it. He reasons that 

 correct pronunciation depends upon clear perception of the sounds 

 to be uttered and a knowledge of the motions necessary to pro- 

 duce them. The knowledge of the proper movements to be made 

 are partly gained by watching' the motions made by others in 

 speaking. On imitating the sound the errors in movement are 

 detected and corrected by comparing the resulting sound with the 

 sound heard. The sounds then that are most distinctly pronounced 

 and requiring movements that are the most clearly visible will 

 naturally first be learned and be most clearly pronounced. Those 

 made in the front part of the mouth, such as labials and dentals, 

 fulfil both of these conditions, while those made in the back part 

 of the mouth usually fulfil neither of them. This law, if approx- 

 imately true, must yet be modified by the fact that children can 

 usually make every one of the elementary sounds used in language 

 before they begin to talk. The difficulty in pronouncing a word 

 is not to utter the elementary sounds of which it is composed, but 

 to properly combine them. As in learning other complex series 

 of motions, it is not a question of making any one motion, but of 

 propei'ly co-ordinating a series of simple motions. No one has any 

 difficulty in pronouncing such words as "three," and "gray," 

 and " geese " separately, but many do in pronouncing them rapidly 

 one after the other. For a similar reason a child who can pro- 

 nounce perfectly a sound in one word is wholly unable to utter it 

 in another. Besides this, sounds are modified somewhat by the 

 sounds that precede and follow them. Almost every one also 

 slurs some sounds in his pronunciation, and children frequently 

 notice and try to imitate only the most distinctly pronounced 

 sounds. They therefore often mispronounce, not from inability to 

 utter the sounds, but because they have failed to notice some of 

 the less perfectly pronounced ones. Since sounds made at the be- 

 ginning of words are least modified by other sounds, mi'spronuncia- 

 tion can best be studied in the initial sounds of words. The letter 

 with which a word begins usually, but not always, indicates the 

 sound. The following is the order for the letters appearing most 

 frequently as initial letters in children's vocabularies : s, b, o, p, t, 

 w, d, m, h, f, r, 1, g, n. To understand the significance of this, 

 it must be compared with the order of frequency for the difficult 

 letters in the dictionary (s, p, c, a, t, b, r, m, d, f, e, h, 1, g), and 



in "Robinson Crusoe" (s, c, p, a, f, b, r, m, e, t, w, h, 1, i, g). 

 One of the most marked differences is the greater number of 

 words beginning with the dentals 6 and d to be found in the vocabu- 

 laries of children. 



Many very interesting questions were suggested by the study of 

 the records already sent me, but a much larger number of vocabu- 

 laries must be compared before reliable answers can be obtained. 

 I shall be glad to receive such records at any time, or to commu- 

 nicate with any one in regard to methods of carrying on the study 

 of children. Letters directed to Rhodes, Iowa, will always reach 

 me. E. A. Kirkpatrick. 



Rhodes, lo., Sept. 14. 



The Convection Theory of Storms. 



De. Hann of Vienna has published recently an extended discus- 

 sion of this subject, and one which has the extremest significance 

 (Sitzber. d. kais. Akad. d. Wissensch. in Wien, April, 1891). He 

 reiterates his view that in our storms at heights of 10,000 feet 

 there is a fall in temperature, and a corresponding rise in our high 

 areas. These points have been sufficiently answered already 

 {Science, Vol. XVI., p. 136). The remaining discussion merits 

 our attention, as it presents a rather strong attack upon the theo- 

 ries ordinarily accepted. A free translation of the argument is 

 here given. Dr. Hann says : 



" How can we think that such extremely flat disks as the great 

 storms of the higher latitudes are can maintain themselves and 

 advance through a rising of air particles. Our whirls have often 

 more than a hundi-ed times greater extension horizontally than 

 vertically. Doberck gives this ratio as 350: 1. A chimney, as is 

 well known, draws only when its height is many times greater 

 than its interior diameter. But in our whirls the relation is in a 

 most extreme manner opposite. How such an exceedingly flat 

 air-disk, only through an interior force, that is, through a freeing 

 of latent heat by a local interior moisture condensation, can move 

 itself in the atmosphere, appears to me difficult to understand. 

 The whole height of the atmosphere (so far as it can come into 

 consideration for the condensation theory) at the utmost is small 

 as compared to the horizontal diameter of our whirl (above 35,000 

 feet is there no moisture). I do not know that the convection 

 theory has seriously considered this objection. This objection 

 does not hold against the theory that correlates the whirl with, 

 disturbances in the general circulation currents of the atmos- 

 phere. 



" A fact which stands out in sharp contradiction with the plain 

 convection theory of our storms lies in the yearly period of their 

 frequency and intensity. If the convection theory is clearly ap- 

 plicable to most of our storms, how can it be that these storms 

 have their greatest intensity and frequency in the winter, even at 

 a time of the year when the conditions, as well for their origin as 

 for their continuance, are most unfavorable ? 



" In winter the moisture of the air is slight and the thermic 

 equilibrium most stable. Upon the continents the lowest layers 

 are often for a long time the coldest, and the temperature increases 

 above. The heat diminution with height is very small in winter, 

 even less than in a rising air curi'ent due to the distribution of 

 moisture. How can a whirl under such conditions of the convec- 

 tion theory reach to the interior of Siberia, where the temperatures 

 are —33° F. to — 40° F., and there is no moisture. It is an inev- 

 itable consequence of the convection theory that the cyclones of 

 the summer must reach their greatest intensity and frequency, 

 because at this time the moisture of the air is greatest, the heating, 

 of the lowest layers the most active, and the heat diminution with 

 height in consequence the most rapid. 



" In fact, heat thunder-storms and tropical cyclones, the ap- 

 pearances to which rightly the convection theory can find appli- 

 cation, are limited to the warm season. Tropical cyclones reach 

 a maximum of occurrence at a time when the temperature of the 

 sea is highest, or when a generally uniform air pressure and the 

 absence of strong air currents favor largely the development as. 

 well as the advance of such whirls, which, perhaps, have for the 

 great part their driving force in themselves. Also the heat thun- 

 der-storms or thunder-storm whirls of our summers occur most 



