THO UGHI -READING. 



327 



a power of reading his master's thoughts akin to that 

 which, on the thought-reading hypothesis, is possessed 

 by some of ourselves. 



To return to Professor Barrett's inquiries : 

 " We endeavoured/' he says, ' ' to gather such indi- 

 cations as we' could of the way in which the impression 

 flashed on the mind of the child. The first question 

 concerns the respective parts in the phenomena played 

 by mental eye and mental ear. Among the experi- 

 ments which we counted as failures were very many 

 where the number or card selected was guessed, as it 

 were, piecemeal. For instance, the number 35 was 

 selected, and the guesses were 45 and 43. So 57 was 

 attempted as 47 and 45. So with cards : the seven of 

 diamonds being chosen, the guesses were six of 

 diamonds and seven of hearts j the three of spades 

 being chosen, the guesses were queen of spades and 

 three of diamonds. These cases seem somewhat in 

 favour of mental eye, the similarity in sound between 

 three and thirty in 43 and 35, or between five and 

 fifty in 45 and 57, not being extremely strong; while 

 the picture of the three or the five is identical in either 

 pair. A stronger argument on the same side is the 

 frequent guessing of king for knave, and vice versa. 

 On the other hand, names of approximate sound (also 

 reckoned as failures) were often given instead of the 

 true one ; ' Chester ' for Leicester, ( Biggis/ for 

 Billings. Frogmore was guessed first as { Freemore/ 

 Snelgrove was given as { Singrore/ the last part of the 

 name was soon given as ' grover/ and the attempt was 



