DISTRIBUTION OF TIME OF CORNELL STUDENTS. 551 



Let us next turn to the variations and extremes as related to the 

 final average. The variations here expressed are necessarily some- 

 what high and should not be confounded with mean variations of in- 

 dividual cases, because they represent merely the variations of averages 

 of groups, each one of which stands for a particular set of conditions 

 and thus tends to exaggerate some one or more items at the expense 

 of the rest. TJnder these conditions we should expect just what the 

 table shows — that the hours for sleep and meals, which are little 

 affected by course or class, are the most constant, the variation being 

 but 3 per cent, for sleep and 8 per cent, for meals. In the case of uni- 

 versity work, we may note that there is considerable variation in the 

 distribution of the type of work, e. g., 67 per cent, for laboratories, 139 

 per cent, for shop and field, but that the final outcome is fairly con- 

 stant, being slightly under 10 per cent, variation. 



When we glance at the individual extremes many of the figures 

 are surprising. Thus, while we know that many students give no time 

 to laboratory or shop work or to self-support, we should scarcely antici- 

 pate that there are students who give no time (at least for this week) 

 to outside stud}', and others who have no lectures or recitations. It 

 is equally unexpected to find that some students give more time by an 

 hour or more a day to a single phase of work, like laboratory or shop 

 work, than the average student gives to all phases of university work 

 combined. The highest lecture time is that, of course, of a student 

 who is attending as an auditor several courses in which he is not 

 registered. With regard to amusement, the chief interest lies in the 

 fact that four students, two in medicine, a freshman in civil engineer- 

 ing and a junior in electrical engineering — report no time for amuse- 

 ment, while periods aggregating only an hour a week, i. e., about ten 

 minutes a day, are reported by several others. Similarly, we find six 

 students who gave no time at all to any form of physical exercise, not- 

 withstanding the temptations of fine skating. 



But it is with regard to meals that the extremes are most interest- 

 ing. We have already noted that the average time is considerably 

 below the time allotted. One student, a freshman in engineering, 

 reports three and a half hours daily for meals; a sophomore in civil 

 engineering reports 3.22 hours, and four students report three 

 hours each. In other words, only six (less than seven tenths of one 

 per cent.) give the time to meals recommended by President Eliot. 

 But it is equally true that but few students give even the two hours 

 recommended by President Schurman. Indeed, no group or class of 

 students, women excluded, average even an hour and a half daily for 

 meals, while the women students average but 1.55 hours. 



This is the feature of the president's speech which, next to his 

 advocacy of eleven hours for work, appears to have aroused the most 



