1879.] oil Generic Images. 167 



this discourse to point out. Exceptional occurrences leave an 

 impression on the brain of i\ir greater strength, and conversely 

 habitual occurrences leave one of far less strength, than their numbers 

 warrant. The physiological oifect of prolonged action, or of reitera- 

 tion, is by no means in direct proportion to the length of the one or 

 to the freiiuency of the other. The magnitude of the " subjective " effect 

 never bears a simple, direct proportion to the magnitude of the " ob- 

 jective " cause. The relation between them, in a very wide circle of 

 physiological phenomena, is expressed by the law of Weber or Fechner, 

 which it is sufficient for our present purposes to state in its original 

 form, because it is exceediugly simple, and is at the same time suffi- 

 ciently correct for all except extreme cases, in which certain alien 

 considerations begin to exert a sensible influence. According to this 

 law (sensation = log. stimulus) the more the senses are stimulated, 

 the more is their discriminative power blunted. If a room is lighted 

 by only a single candle, and a second one is brought in, the eye feels 

 a certain increase of light. Now, if 1000 candles had originally been 

 in the room, it would require the addition, not of one caudle, but of 

 another 1000 candles, to produce the sense of a similar increase. In 

 order that the magnitude of any sensation shouUl increase by a series of 

 equal steps, the magnitude of the stimulus that causes it must increase 

 by successive multiples. The one follows an arithmetic progression, 

 the other a geometric one. 



A few simple experiments will illustrate this. Five perfectly 

 black cards are taken, each of the size of half a sheet of note paper ; 

 also a sheet of perfectly white note-paper. The latter is torn in two, 

 and one half is laid upon card No. 5, which it exactly covers. The 

 remaining half is carefully folded down its middle, and torn in two, 

 and one portion is laid on card No. 4, of which it exactly covers one 

 half. The same process is continued, so that card 3 is covered to the 

 extent of one quarter of its surface, 4 to one-eighth, and 5 to one- 

 sixteenth, and there is a remnant of one-sixteenth, which may be thrown 

 away. To avoid fractions, let>ais count the quantity of white on the black 

 card No. 1 as one, then that on Nos. 2, 3, 4, 5 will be as two, four, 

 eight, and sixteen respectively, the latter standing for pure white. 

 The next step is to cut the portions of jmper into shreds, and to 

 scatter them uniformly over their resjpective cards. In the specimens 

 now upon the table tliis has been already done, and the shreds are 

 pasted down. The effect, when they are looked at from a little 

 distance with the eye not focussed too sharply uj^on them, is that 

 of a series of greys, which appear to be separated by equal intervals 

 of tint from one another, although we know that the differences 

 in the amount of white material is by no means uniform. The eye 

 judges card No. 3, which contains four portions of white, to be of 

 a medium tint between Nos. 1 and 5 ; but, as No. 1 contains one 

 portion, and No. 5 contains sixteen portions, the medium quantity 



of white is really eight and a half (because — - — = 8 J), and this 



