Apbu. 29, 1904.] 



SCIENCE. 



ro7 



In deep water tlie meteorological current is 

 often much stronger than the tidal; but as it 

 has not the period of the latter, the measure- 

 ment of the rise and fall of the tide and of 

 the velocity of the flow and ebb could hardly 

 be seriously affected by its presence. 



The object of looking into the several 

 sources of error has been to ascertain when 

 the errors are small enough to be neglected 

 rather than to attempt correcting for them 

 in actual measurements. R. A. Harris. 



February, 1904. 



SEX DIFFERENCES IN THE SENSE OF TIME. 



In going over the results of a series of dem- 

 onstration tests of the sense of time given 

 to a mixed class recently, the returns from 

 men and women were separately reduced, the 

 summaries of which presented features of 

 sex differentiation concerning which corrobo- 

 ration or revision on the part of others is 

 sought through this note. 



The test involved periods of time extending 

 from a quarter of a minute to a minute and a 

 half in duration. The intervals were filled in 

 four ways : (1) The instructor read aloud to 

 the class from a psychological work unfamiliar 

 to its members; (2) the members of the class 

 marked as rapidly as possible all the letter m's 

 in a page of printed text; (3) the class waited 

 in idleness for the period to pass by, refraining 

 as far as possible from counting or other 

 means of recording the lapse of time; (4) 

 each person estimated as accurately as pos- 

 sible the period in question, using whatever 

 method he had personally found most service- 

 able for the purpose. As there were fifteen 

 men in the class, the tabulation of returns 

 from the wometi was brought to a close when 

 an equal number of judgments had been en- 

 tered therein. 



Only in the case of the one-minute period 

 was estimation made under all four condi- 

 tions mentioned. The results are presented 

 in the following table, in which the signs 

 plus and minus indicate respectively over- and 

 under-estimation of the duration in terms 

 of seconds, and the figures at the tops of the 

 columns the series of conditions enumerated 



Incidentally, the purpose of the test was to 

 call attention to the diiferences in one's esti- 

 mation of time under conditions (1) and (2), 

 and similarly in the case of (3) and (4). 

 The relation of the members of these two 

 pairs to eaxih other is made apparent in the 

 table. It is also to be noted that with the 

 exception of the first entry in column (4), 

 the only minus quantity in the whole series 

 of tests, the error is throughout one of pro- 

 nounced over-estimation. This tendency is 

 very much stronger in the women than in the 

 men, the first point of contrast in the com- 

 parison of sex differences. Por this period 

 of time the constant errors of the two sexes 

 stand in a ratio of one to four. The clearest 

 indication that this over-estimation of short 

 periods of time on the part of women is a 

 persistent habit, and not due to variable fac- 

 tors in the conditions of experimentation, ap- 

 pears in the fourth column of the table. In 

 the ease of men, keeping tally of the passing 

 seconds results in the elimination of the 

 positive error and the appearance of a slight 

 under-estimation. In the case of women, on 

 the contrary, over-estimation still persists to 

 the amount of two fifths of the period in ques- 

 tion; in other words, their unit of measure- 

 ment is much in defect of the objective period 

 which it is meant to represent. 



The results of the various other tests made 

 in the same series are summed in the follow- 

 ing table, in which the same general rela- ■ 

 tions are presented as in the preceding group. 



There is also to be considered in such a 

 comparison as the present the average varia- 

 tions of the individual judgments of men and 

 women respectively from objective accuracy, 

 of which the formulation of their constant 



