ro 



• KNOVS^'LEDGE • 



[Nov. 2'., 1881. 



c.tpird liiiii, nntl shoutrd out his iinmc. lie turned at oncp, 

 cliiiilifd tlip hill, and came into the bungalow, where the 

 Bnnip farce of repentance was gone through. Bully now 

 seemed to have made up his mind that escape was inijios- 

 Bilile ; lie lay down on a mat in tlie verandali, and remained 

 there for a loiij; time, liiit for tiie jiersi.stent cock of the 

 ears, we should have imagined tlio animal really a.sleep. 

 Mr. Cherry eventually went to hi.s otlice-rooni, and 1 re- 

 mained in the verandah, reading the morning paper and 

 occasionally glancing at Bullv. He lay very still, but once 

 or twice 1 detected him opening his eyes and raising his 

 head to look round liim. Each time he caught my eye lie 

 wagged his tail vehemently for a moment or two, and then 

 resorted to his sham sleep. It may liavc been for half-an- 

 hour or thereabouts tliat this state of things continued. I 

 then became interested in an article in the paper, and when 

 I next looked up, Bully was gone. 1 called Mr. Cherry, 

 and the house was searched for Bully. Tlie peon was sent 

 for and interrogated ; he liad not seen the dog. As a last 

 resource, in<iuiry was made of the horse-keepers down at 

 the staliles, D. The reply was " Yes, the dog had passed 

 through the gate, D, some time before." Taking advantage 

 of ni}' occupation, and the absence of his master, Bully had 

 left the house and taken his way to the cantonment by the 

 onlj- path by which he could have escaped unnoticed V)y the 

 peon — that shown by the dotted line. 



It seems to us quite inipossilile to account for the dog's 

 action, as above narrated, without attributing to him tlie 

 exercise of reasoning powers, not merely in selecting the 

 route by whicli he finally escaped, but in the liianceuvres by 

 which he endeavoured to assure those who were watching 

 him that he had given up all hope of escaping. Doubtless, 

 if he had reasoned more perfectly, he would even have 

 allowed his ears to seem asleep, instead of leaving them 

 cocked. But very few of us human beings simulate sleep 

 without some such error, by wliich any observant person 

 would be enabled to detect the trick. Either the muscles 

 of the face are not perfectly relaxed ; or the hands or feet 

 are left in a constrained attitude ; or the position of the 

 body generally is unlike that which a sleeping person would 

 assume ; or else the breathing is unnaturally restrained. 

 And again, though Bully was too demonstrative in his 

 contrition, and afterwards in his attempts to hide the con- 

 sciousness of failure, it was not for want of reasoning 

 power. Few of us know how to act such parts as he tried 

 to play, with perfect correctness ; nor do those who know 

 succeed always in acting such parts as they could wish. 

 Probably Bully was as sensible as the onlookers that 

 he was not quite successful in his acting. It is obvious, 

 however, that he directed his eflbrts as carefully to the end 

 he wished to obtain as a human being of average reasoning 

 powers and skill in counterfiiting sleep, itc, could have 

 done. 



ILLUSIONS. 



By Thomas Foster. 



IN my last jiapcr there were two mistakes, or rather a 

 single mistake, for it amounted really to the inversion 

 of one of the figures. In Fig. .'>, the lines marked AB, CD, 

 should have been put opjiosite the letters EF, Gil, and 

 rict; irrsH. As they actually stand, the description should 

 run as follows: — The lines AB, CI) appear to be curved, so as 

 to be farther apart at the middle than at either end ; whilst 

 tlie lines EF and Gil, in the same figure, appear to be so 

 curved as to be nearer in the middle than at either end. 



Such illusions as are illustrated at pp. 1 1 and iL' vary in 

 etfect according to the position from wliich the illusory draw- 



ing is seen. Thus, if we so place Fig. 1 that the eye looks 

 along, or nearlyjalong, the direction AB, we see at once that 

 CD is in the same straight line with AB. Again, although 

 we cannot look simultaneously in the directions BC and 

 ED in Fig. .'?, it will be found that when we look along 

 BC, so that the illusion ceases so far as the part of the 

 circle near C is concerned, the illusion eea.ses altogether, 

 the part near D no longer appearing to form part of a 

 smaller circle. In Fig. 4, the illusion is not at all efTec- 

 tive when we hold the lines AB, CD, Ac, upright, but Ls 

 very marked when they are looked at aslant. But again, 

 even when we look at thcin aslant, and the illusion is 

 strong, we yet find that running the eye backwards and 

 fonvards along the parallels, the illusion varies all the 

 time ; for the eye cannot fail then to percei\e that the 

 distance between the parallels does not vary as the illusory 

 eflect seems to suggest Thus, the lines CD, EF, which, 

 when the eye is at rest, seem to draw closer at D and F, 

 are found not to do so when tlie eye is carried from CE to 

 DF; and a singular effect of motion is produced, the line.s 

 CD and EF seeming to move apart as the eye i-uns down 

 them towards DF. Tlie other lines seem to move corre- 

 spondingly, and thus we have an illusion of motion, •which 

 will be considered later among illusions of that sort 



In the illusory figures now to be considered, the lines 

 whicli appear curved, though straight, are set in different 

 positions so as to be differently aft'ected, and thus we have 

 illusions affecting the shape of enclosed figures such as 

 shall be considered separately further on. 



If we look at Fig. 8 as it stands, that is with the point 

 A uppermost, we find the lines AB, AE looking strongly 

 curved ; BC and ED rather less curved ; and DC looks 

 almost straight — at least, this is the effect to my eye. (I 

 find EA most curved of all.) Thus the figure, which is 



Fig. 8. 



really a true pentagon with straight sides, looks like a 

 figure having unequally curved sides. Turning the picture 

 round in its own plane, we find that as tlie sides of the 

 pentagon by this turning motion pass to tlie top, they 

 appear more curved, so that tliey appear to change in shape 

 as tliev move, or to have a motion of their ovni. 



The apparent cur^'ature of the really straight lines in 



