TEACHING AND LEARNING. 15 



taken by the students in the best obtainable book is much 

 greater than with a cheaper one, not to mention the advantage 

 of its much greater chance of preservation. In these books the 

 students record results, using all care and taking an artistic 

 pride in their completeness and neatness. Each record should 

 be made complete in itself and a model of scientific exposition. 

 It should include : 



(i) A statement of the exact object of the experiment. 



^2) A description of the practice and theory of the method 

 used, with illustrations of the appliances. The latter 

 may best be annotated blue-prints made by the 

 students themselves from a negative of one of the 

 experiments actually set up. It is not profitable to 

 spend time upon drawing apparatus, though much 

 drawing of diagrams, etc., must be done. In the 

 latter free use of colors should be made, for which 

 colored pencils are sufficient. 



(3) The exact results obtained, expressed in tables of 



figures and in curves (or polygons). Whenever 

 possible it is well to construct a curve averaged from 

 the total results of the class, which for comparison 

 is to be plotted on the same sheet with the individual's 

 curve. Indeed this method offers the simplest means 

 for obtaining an approximate idea (the more accurate 

 the larger the number of students whose records are 

 combined in the average) of the probable error in the 

 individual's results, whether this be due to the " per- 

 sonal equation ' of the observer, individual variation 

 on the part of the plant, or uncontrollable variations 

 in external conditions. 



(4) The conclusions as to the result of the experiment, 



together with remarks as to its general bearings. 



In their proper places in the sequence of topics should come 

 the accounts of the subjects studied non-practical ly and the 

 essays. These essays are of particular value. They are by 



