THE EDUCATIONAL HUMANE SOCIETY 



123 



and the letter or number is called out 

 aloud, and at the same time the dog is 

 shown which one it is. After several 

 trials, he will select this one and dis- 

 regard the others, when it is called for. 

 This once learned, the next letter is 

 taught in like manner, until a large 

 number are recognized by the dog, and 

 he is able to pick out any of them at 

 will. Plants are also to be selected in 

 a similar manner, from a row placed 

 on the table, and so forth. 



It must be admitted, however, that 

 most feats of this character, as per- 

 formed in public, are the result of some 

 trick, rather than any marvelously 

 elaborate training on the part of the 

 dog, which would be necessarv if these 

 feats were genuine — granting them to 

 be possible at all. As a matter of fact, 

 most of these apparently marvelous 

 feats are based on a very few cues, 

 given to the dog at the appropriate 

 time, to which he has been taught to 

 respond in a simple manner. A few 

 examples will make this clear. 



Many of these feats are performed 

 by means of a cue word, in just the 

 same kind of way as "mind-readers" 

 entertain and puzzle their audience. 

 As soon as this word is given, it may 

 be in the course of a sentence, the doe 

 knows that he is to perform a certain 

 action. It is not necessary for him to 

 understand the whole of the sentence; 

 only one word in it. As soon as that 

 word is caught, the action is performed. 

 Each action corresponds to a certain 

 cue word. Again, there is the method 

 of training by the use of the eyes. The 

 dog watches his master's eyes, and 

 when his master glances in any direc- 

 tion — at a card, for example — the dog 

 can follow his glance, and pick out the 

 card in turn. Or the dog may be told 

 to bark a certain number, in which case 

 the dog watches his master's face 

 closely, and simply barks until the 

 eyes, or some movement, tell him to 

 stop. He does not have to know that 

 he barks nine times. All he has to 

 know is that he must go on barking 

 until he is told to stop by his master's 

 signal ; and the trainer is the one who 

 does all the counting. 



There are certain stage tricks which 

 depend very largely upon the dog's 

 memory, however — such as picking up 

 a numbered card, and the like. The 

 cards are arranged in a row, and the 



trainer stands in front of the row in 

 which the card rests. A string is at- 

 tached to the dog's neck. First, the 

 dog is trained to go to the row of cards 

 nearest the trainer; then, if he is in- 

 clined to pick up one too near, a slight 

 pull on the string" is given, pulling the 

 dog up to the required number. The 

 trainer stands at a certain distance 

 from the table in these tricks ; if close 

 to the table, the dog knows it means 

 card one ; if farther away, card two, 

 and if still further, card three. By 

 care in training, the dog can be taught 

 to pick out any required card, without 

 in any way knowing the number writ- 

 ten upon it. When the dog has been 

 taught to pick up any card by means 

 of this code, the trainer may appear 

 to make it far more complicated by caus- 

 ing the dog to add, subtract, multiply, 

 divide, etc. All that is necessary, of 

 course, is that the performer himself 

 should do the sum, mentally note the 

 position of the card giving' the answer, 

 and indicate this card to the dog by 

 means of some hidden code. 



In the same way, horses can be made 

 to stamp out any desired number, tell 

 the date of a coin, etc., by simply going 

 on pawing the ground until the trainer 

 gives them the signal to stop by means 

 of some secret sign, unnoticed by the 

 audience. — Scientific American. 



Notes on the Armadillo of Texas. . . 



BY DR. R. W. SHUFELDT, C. M. Z. S., WASH- 

 INGTON, D. C. 



There occur in the American fauna 

 quite a number of different kinds of 

 armadillos, their habitats ranging all 

 the way from southern Texas, down 

 through Mexico, the Central American 

 states and South America. They are 

 all timid, inoffensive animals with very 

 interesting life histories, and they are 

 related to the extinct glyptodons, 

 which, like the living forms, were pro- 

 tected by an osseous coat of mail com- 

 posed of little confluent scutes formed 

 in the integuments. Most of the glyp- 

 todons were large animals, their armor 

 often protecting the entire body, but 

 sometimes only the back where it is 

 known as the carapace. In our own 

 existing armadillos, however, this coat 

 of mail never covers the belly, while 

 above it is divided into three definite 

 sections, an anterior, a posterior and a 

 middle part. This middle part is gen- 



