426 



NATURE 



[December 17, 1914 



always take the trouble, and typhoid thus is con- 

 veyed by flies, or by direct iouhnj^ of clothes and 

 hands and food and utensils. Enoug-h has been 

 talked about water-borne typhoid and milk-borne 

 typhoid : we want more talk about man-borne 

 typhoid. 



The protective treatment against typhoid has, 

 on the whole, met with very little opposition, and 

 none of real authority. Certain statements of 

 disastrous results have been inquired into, and 

 have been found not true. Not that no ill-effects 

 have ever followed ; but these are rare. The main 

 fact, that the protective treatment is indeed pro- 

 tective, has been proved past all possibility of the 

 least shade of doubt. 



That old scourg-e of armies and of jails — typhus 

 fe\'er- — ^has long ceased to trouble our country ; 

 probably most doctors now in practice have never 

 seen a case of typhus. It may appear, some day, 

 a most unwelcome visitor, in the western theatre 

 of the war. It is conveyed by body-lice. It is 

 said to be endemic in Silesia. It may find its way 

 westward ; or a trace of it, amid the desolation 

 and wreckage of Belgian villages, may flare up 

 into widespread trouble. Happily, the vigilance, 

 the incessant zeal, of all branches of the Army 

 Medical Service, and of all the many societies 

 working at the Front for the welfare of the Army, 

 will continue to bring forth good fruit. It is not 

 possible, with such splendid organisation, such 

 generous devotion, that any grave outbreak of 

 infection should pass neglected ; and, when the 

 medical and surgical histor}- of the present war 

 comes to be written, it will be a fine record of 

 good work accomplished on a grand scale. 



Stephen Paget. 



THINKING ANIMALS. 1 



ABOUT ten years ago it became known that 

 "Clever Hans," an Arab stallion owned by 

 a Herr von Osten in Berlin, was able to answer 

 arithmetical and other questions, tapping out the 

 reply with his fore-foot. Notoriety led to heated 

 controversy, and the appointment of committees 

 to investigate. The second of these, under Prof. 

 Stumpf, resulted in Pfungst's book, explaining 

 everything in terms of signals consisting in slight 

 movements made unconsciously by some person 

 present knowing the answer. This seemed to 

 have solved the problem finally until the appear- 

 ance of Krall's book in 1912. The author, a 

 wealthy jeweller of Elberfeld and friend of von 

 Osten 's, had after the latter 's death continued to 

 experiment, obtaining results which, he claimed, 

 refuted Pfungst's explanation. This claim found 



1 (i) "Das Pferd des Hei-rn v. Osten (Der kluge Hans)." By O. Pfiingst. 



(Leipzig : J. A. Barth, igo;.) 



(2) " Denkende Tiere." By K. Krall. (Leipiig: W. Engelmann, iqi2.) 

 (^) " Uebsr den dermaligen Stand des Krallismus." By JProf. H. Dexter. 



Reprint from Lotos. Prague, vol. Ixii., 1914. 



(4) "Gibt as denkende Tiere?" By Dr. S. v. Maday. Pp. x + 45i- 

 (Leipzig : W. Engelmarin, 1Q14.) 



(5) " Das Problem der Elberfeltler Pferde und die Telepathie." By Prof. 

 H. V. Buttel-Reepen. Naturwisscnschaftliche IVochensc/tri/t, 1914, 

 No. 13. 



(6) " Meine Er'uhrungen mit den ' denkenden ' Pferden." By Prof. H. 

 V. Buttel-Reepen. 'Naturwisscnschaftliche M'ochrnschrift, IQ14, No. 16. 



(7) " Eine Kritik der Leistungen der ' Elbertelder denkenden Pferde.'" 

 By Prof C. Schroder. Naturwissenscliaftliche Wochenschrift, 1914. 

 Nos. 21, 22. 



NO. 2355, VOL. 94] 



support in a report signed by the zoologists, 

 Kraemer, Sarasin, and Ziegler, asserting that sig- 

 nalling was excluded since correct answers were 

 given even when none of the human participants 

 was visible to the animal. The opinions expressed 

 on Krall's book vary from that of Prof. Dexler — 

 "a shameful blot on German literature," to that 

 of Prof. Ostwald, who foresees that it will "as 

 clearly mark the beginning of a new chapter in 

 the doctrine of man's place in nature as Darwin's 

 chief work did in its day." 



As to the problem itself, a definitive solution 

 could result only from a free and impartial testing 

 of the animals ; as it is one can only indicate 

 probabilities. Intentional deceit is almost certainly 

 too simplicist an explanation, and is in any case 

 inadequate. On the other hand, the probability 

 of obtaining correct answers by chance has been 

 underestimated in view of the number of un- 

 successful attempts and the greater frequency with 

 which certain numbers occur. Very much must 

 be allowed for this and other weaknesses of testi- 

 monv, the demonstration of which has been one 

 of the successes of applied psychology, but which, 

 as every newspaper now shows, are seldom given 

 weight in practice. They particularly affect some 

 at least of the would-be crucial tests. Neverthe- 

 less much remains, of which the following main 

 explanations have been offered. 



The answers are evidence of mathematical in- 

 telligence. This, although a highly developed 

 " number-sense " has been found in persons of low 

 genera] ability, and even in the feeble-minded, 

 conflicts with all that we know from other sources 

 about the animal mind. Detailed scrutiny of 

 Krall's account of his teaching shows that the 

 problem often coulA not have been understood 

 from his exposition. Again, the correcting of a 

 single false figure is done quickly and certainly, 

 as might be expected if signals were being given, 

 since these would be facilitated by concentration 

 of the signaller's attention; if the errors are mis- 

 takes of calculation It is odd. Finally, the in- 

 ability of the animals to prove their understanding 

 by action, compared with their eloquence in the 

 language of taps, is extremely suspicious. 



The answers are due to memory. The horse's 

 memory Is, no doubt, excellent for some things, 

 and the theory has advantages, but also serious 

 difficulties. To associate eight taps with one 

 svmbol and nine with another, the horse must be 

 able to distinguish the two series. But it seems 

 probable that animals cannot distinguish numbers 

 bevond four or five. Rothe trained his dog to 

 come only at the fifth whistle — but this only if 

 the whistles were at regular intervals : his horse 

 would take four lumps of sugar in preference to 

 three, but confused four and five. Again, the 

 horse's eye, while very sensitive to movement, is 

 probably unsuited to the clear perception of com- 

 plex visual forms such as written numbers, and, 

 as a matter of fact, the animals seem to attend 

 to the questioner more than to the blackboard. 

 Finally, the mistakes in cube root, etc., questions 

 strongly suggest the use of tips. 



