1898] NOTES AND COMMENTS 83 



analogous in structure to an onion. This theory is now exploded. 

 Dr Liversidge's researches prove that the nuggets have a well-marked 

 internal crystalline structure (reproduced in an exquisite series of 

 [photographs) ; and this structure is also observable in gold ingots 

 which have been fused. The nuggets are shown to commonly 

 •enclose foreign substances, such as quartz and clay ; and Dr Liver- 

 sidge is inclined to think that the gold was originally deposited from 

 solution in veins or pockets in the solid rock. During denudation 

 ithe nuggets would thus find their way into the gravels and alluvium 

 when already completely formed. 



Fatigue in Reading 



The Psychological Review contains in its number of September 1896 

 ;an article, left too long unnoticed, " On the Conditions of Fatigue in 

 Eeading," which is of some practical importance, as well as of 

 theoretical interest. The authors, Messrs Harold Griffing and 

 Shepherd J. Franz of Columbia, following the idea suggested by 

 Prof. Cattell in a well-known paper, and supplementing his work, 

 ■show how facility of reading is affected by size and quality of type, 

 'by ' leading,' by the intensity and quality of the illumination, and 

 by the quality of the paper. The result to which they come is 

 'that " the size of the type is the all-important condition of visual 

 fatigue. No type less than Vo mm. in height (eleventh point) should 

 ever be used, the fatigue increasing rapidly even before the size 



■ becomes as small as this." The intensity of illumination is " of 

 little consequence within the limits of daylight in well-lighted rooms. 

 Very few intensities less than 3 to 10 candle-metres (a candle- 

 metre being the light of a standard candle at a perpendicular 



■ distance of one metre) are sources of even greater fatigue than small 

 type, and 100 candle-metres may be considered a type limit." The 

 •experiments on the relative legibility of different kinds of type were 

 ^carried out by different methods, the results of which agree fairly 



well — by determining the times of reading certain passages, by 

 .finding the percentage of words which could be seen in certain 

 phrases when cards containing them are exposed for a given time, 

 by determining (through the falling chronometer) the time of ex- 

 posure necessary for reading certain words in different type, or the 

 amount of illumination necessary to see letters of different sizes. 

 The experiments are of a careful character. But it would have 

 fbeen well in determining the effect of these various conditions to 

 try a parallel series with nonsense- words instead of real ones, so as 

 to eliminate the element of familiarity, and the various accidental 

 elements arising from special association, though this has been done 

 in part by taking a considerable number of observers. 



