)82 



SCIENCE. 



[Vol. II., No. 42: 



average, and the rainfall showed a deficiency of near- 

 ly an inch. The bureau is gradually increasing the 

 number of stations, and malies special efforts to have 

 its observers supplied with standard instruments. In 

 addition to its regular stations, it invites the co-ox^er- 

 ation of voluntary observers, and will furnish relia- 

 ble instruments at reduced prices. The rainfall chart 

 published by this service is deserving of being intro- 

 duced into other similar reports. ' 



Tennessee. — The continued drought has damaged 

 the crops, especially in the eastern portion; but in 

 the middle portion the crops are in fair condition. 

 Frost visited some localities, the temperatures in the 

 state ranging from 32° to 95°. The prevailing wind 

 was north; the average rainfall, 2.06 inches; the aver- 

 age number of clear days, 14. 



LETTERS TO THE EDITOR. 

 Teaching language to brutes. 

 Is it not quite conceivable that some of the lower 

 animals might be taught to use human language 

 rationally? No doubt the reasons for a first hasty 

 answer in the affirmative would be that the animals 

 seem so intelligent as sometimes even to reason, and 

 that they have, in fact, often had human words put 

 into their mouths, and that they seem sometimes to 

 have a language among themselves. Yet, after all, 

 cannot their intelligence, and even wisdom, and oc- 

 casional apparent reasoning, be satisfactorily ex- 

 plained, without attributing to them true reasoning, 

 as the result of hundreds or thousands of generations 

 of experience and transmitted memory, by which cer- 

 tain objects or actions become associated with a feeling 

 of pleasure or pain that induces pursuit or avoidance? 

 How few, indeed, are the cases that cannot readily 

 be so explained, where an animal appears at first' 

 sight to exercise a reasoning-power! and bow ex- 

 tremely simple the effort seems then to be! 



True reasoning can always be reduced to the syllo- 

 gistic form, in effect a statement that what is true 

 of a class is true of something in that class. In order, 

 then, to reason, properly speaking, it is nccessai'y to 

 use a general term (a word or sign with the meaning 

 of a common noun ) to indicate the class ; and, as we do 

 not know of any evidence that brutes have such words 

 or signs, we have no proof that they can reason. In 

 like manner, the lack of evidence that they can rea- 

 son goes far towards showing that they have no lan- 

 guage that includes such general terms, though it 

 may be true that they sometimes understand words 

 in a singular (not general) sense, and have similar 

 expressions for their own feelings. 



The question, then, is whether brutes may not be 

 taught the intelligent use of general words or common 

 nouns, which would enable them to reason. As the 

 step does not seem so very enormous from the unde- 

 niable intelligence of some brutes to the lowest form 

 of generalization, it is perhaps worth while to con- 

 sider how they might possibly be taught to take the 

 step, in hope, that, having once taken it, tliey might 

 be led farther with still greater ease. Since the idea of 

 plurality appears to lie at the very bottom of the idea 

 of class, number would perhaps be the first and 

 simplest step in generalizing, — number, that is, the 

 regarding things merely as individuals or units. It 

 is a step beyond, to regard things as alike in more 

 complex respects. If that is so, the first effort might 

 be made to teach how to count, and, of course, at the 



beginning only to count up to two. If that can be^ 

 accomplished, still further counting can unqueslion- 

 ably be taught, and no doubt by degrees a much 

 greater amount of generalization and reasoning itself. 

 Does it seem impossible that a brute may learn to 

 associate invariably the word ' one ' with a single ob- 

 ject, and ' two ' with a pair of objects, no matter of 

 what kind? At first the two objects should always 

 be two like ones; but by degrees a difference in them 

 might be allowed. The teaching of common names 

 might next be taken up; or it might be l)egun along 

 with the counting, but without the confusing addi- 

 tion of any plural termination. Kven if the mere 

 counting up to two could not be taught successfully 

 to any single individual brute, yet the end might 

 nevertheless be attained, perhaps, in several genera- 

 tions. 



The question tlien comes. With what animal would 

 it be best to begin such experiments, — whether with 

 monkeys, or elephants, or birds, or ants? Of course, 

 articulation is not essential ; for a language of signs 

 might be devised suitable to the animal, — a language 

 corresponding to the deaf-and-dumb one of signs, or 

 to one using the Morse alphabet, or something like it. 

 Elephants are very intelligent, but so very long lived 

 that it would take ages to observe the effect of training 

 through many successive generations. Perhaps the 

 convenience of excellent articulation and rapid propa- 

 gation, both combined with apparently good intelli- 

 gence, might give the preference, on the whole, to a 

 talking bird, such as the Indian mynah. L. B. 



Nov. 9, 1883. 



Climate in the cure of consumption. 



In your issues of Sept. 28 and Oct. 5, Dr. S. A. 

 Fisk of Denver, Col., compares the climates of the 

 principal health-resorts of the United States with one 

 be happens to represent, i.e., Colorado. At the com- 

 mencement of his paper the writer states that " he 

 has given the data for Augusta, Ga., as the best sub- 

 stitute for Aiken, S.C., at which place there is no 

 signal-station; and, in doing so, he thinks that he is 

 presenting data which will fairly represent the cli- 

 matic condition of Aiken." To those familiar with 

 the two places, this is, indeed, a most astounding 

 revelation; and, with your kind permission, I hope to 

 prove, that, although socially very dear to each other, 

 they have climatically but little in common. Augusta 

 is built upon a marshy flat on the Savannah Eiver, 

 which at times overflows its banks, and submerges a 

 portion of the city; while Aiken is located in what is 

 known as the sand-hill region, five hundred and 

 sixty-five feet above sea-level, which is higher than 

 any other town or village witliin a radius of seventy 

 miles. The soil of the latter place is dry and porous; 

 and to obtain water, wells have to be sunk to a depth 

 of from a hundred to a hundred and twenty feet; 

 and there is no water-course within two miles of the 

 town, and even at that distance there are but brooks 

 or small creeks. The result of this absence of soil- 

 moisture, and of large bodies of water, would of itself 

 tend to diminish the amount of humidity in the at- 

 mosphere; but this is still further diminished by the 

 absence of any hill or mountain to interrupt the free 

 circulation of the wind. Augusta, on the contrary, 

 is situated, as before stated, on a plain lying between 

 a range of bills and the river. AH this would lead 

 one to expect that the climate of Aiken would be ex- 

 tremely dry ; and that this is really the case is proved 

 by carefully conducted observations extending over 

 many years, which show that the average relative 

 humidity, fifty-eight percent, is lower than that of any 

 other station east of the Rocky Mountains, and eleven 



