300 



SCIEJSrCE. 



[Vol. IX., No. 216 



The mental span increases quite constantly with 

 the age. Boys of 11 years could grasp 6.5 numerals 

 and 5.5 letters ; of 12 years, 6,8 numerals and 5.7 

 letters ; of 13 years, 8.8 numerals and 6.9 letters. 

 The follotting table shows the result of a more 

 extended set of observations on the girls of the 

 North London collegiate school : — 



lum showed a somewhat higher mental span, ac- 

 companied by a higher capacity generally. 



In the same journal, Dr. J. M. Cattell records 

 some ' Experiments on the association of ideas. 

 His object is to measure the time needed for the 

 characteristic processes of ordinary thinking. The 

 experiments were made on himself and a German 



Age 



Number of subjects 



Average number of numerals 

 Average number of letters ... 



19 



14 

 8.6 

 7.9 



While the limit for numerals was, as a rule, 

 higher than that for letters, cases when the re- 

 verse was true were not infrequent. In one set 

 of 88 schoolboys, 14 could repeat more letters 

 than numerals, while 33 of the remainder had the 

 same limit for both. No definite conclusions can 

 be drawn as to the relative spans of the two sexes, 

 as the boys and girls came from different classes 

 of society. It may be worth noting, that, at the 

 age of 13, the boys could repeat 8.8 numerals to 

 the girls' 8.3, but only 6.9 letters to the latter's 7.3. 



A very clear result was, that the span bore a 

 definite relation to the rank in the class. Thus, 

 the 10 boys who stood highest among 30 twelve- 

 year-olds had an average span for numerals of 

 9.1 ; while themiddie 10 had only 8.3, and the low- 

 est 10, 7.9 ; and the same holds for the girls. The 

 first half of a class almost invariably shows a 

 higher span, both for letters and for numerals, 

 than the second half. 



Mr. Francis Galton and Professor Bain applied 

 a similar method of observation to the memory- 

 powers of idiots. While most idiots can hardly 

 add two figures together, some have a decided 

 knack for remembering figures, dates, and so on. 

 Nine of the best girl-idiots at an asylum (none of 

 whom could add 3 to 5) had an average span for 

 numerals of only 4. Two girls who could not re- 

 peat more than two figures without mistake were 

 tested with three figures. In 23 trials the last 

 figure was rightly repeated 17 times, the second 

 10 times, and the first 7 times, showing that the 

 last- uttered sound is most readily repeated. 



Idiots with peculiar memories were also tested. 

 One could repeat pages of Maynall's ' History ' 

 with considerable exactness ; another had a re- 

 markable intimacy with the calendar. But they 

 all failed on the numeral test, being hardly able 

 to repeat three figures. Their memories seemed 

 deeply rutted in one groove ; not strong, but very 

 limited. 

 • The experiments on the idiots of another asy- 



friend, Dr. Berger. A few of his results are 

 these. To give the name of the picture of an ob- 

 ject in a foreign language (English for a German, 

 and German for an American) required .649 and 

 .694 of a second respectively, which is .172 and 

 .149 of a second longer than to name objects in 

 one's own language. 



Experiments on the time necessary for translat- 

 ing words showed that it took longer to translate 

 from the foreign to the vernacular than the re- 

 verse, and also that the time itself might indicate 

 one's familiarity with the two languages. 



Given a city to name the country in which it is 

 situated required about .400 of a second. Given 

 a month to name the following month required 

 .367 of a second, while to name the preceding 

 month took as long as .798 of a second, showing 

 how much more readily the mind moved forwards 

 than backwards. Similarly, it is easier to proceed 

 from the part to the whole than from the whole 

 to the part. Given a month to name the appro- 

 priate season requires .363 of a second ; given a 

 season to name a month in it, .498 of a second. 



When the association is less restricted, — as, for 

 instance, to name a subject for an intransitive 

 verb (swim-Jish), or an object for a transitive one 

 (write - letter), — the time is longer. The former 

 operation took .646 of a second, and the latter .517, 

 the mind moving logically towards the object. 



The time necessary to judge the length of aline 

 suddenly revealed was very long (nearly one sec- 

 ond), showing that the judging process forms 

 slowly. 



It is, however, to be remembered that in all the 

 above processes individual variations are ex- 

 tremely large. While such experiments are rather 

 suggestive and personally interesting, they can 

 hardly be said to have the scientific character or 

 importance belonging to the measurement of more 

 elementary processes. There is little guaranty 

 that the process in different minds is sufficiently 

 alike to make an average significant. 



