1 64 THE TEACHING BOTANIST 



cracks and folds are simply due to individual differences in the 

 mode of drying, etc., and that shape and size are variable, 

 though within limits. In his treatment of Exercise 3 he should 

 be led to distinguish clearly resemblances and differences, 

 and to describe them separately. 



(3) On Representation.- -The general principles of this 

 part of the work are discussed in Chapter IV. Observation 

 should be fully made before recording is begun. As to draw- 

 ing, the students should first be allowed to do the best they 

 can unaided, judging for themselves how many and what kind 

 of drawings are necessary to show completely the seed and its 

 parts. They should be made to complete a subject the best 

 they can before it is examined or criticised by the teacher ; 

 this is to inculcate self-reliance. After they have done their 

 very best, their work should at once be examined, and wherever 

 it shows marked deficiencies they should be encouraged to 

 look and try again. After they have finally done all they 

 can, the teacher should, step by step, carefully explaining 

 the logic of each point, show them the best way he knows for 

 representing the objects, with which they may compare their 

 own efforts ; this may well be done for them all together on a 

 blackboard ; they are, after their own trials, in a position to 

 profit by all of the advice thus given. A good representation 

 of the Lima Bean as an example for beginners is shown in Fig. 

 12, though the faint radiating lines might have been added. 

 But while representation is made thus important, the teacher 

 must not go so far as to make a fetich of it ; for after all it is 

 but a means to an end. At first only clear diagrams should be 

 insisted upon - - shading, etc., may better come later. It is, 

 moreover, very important not to insist upon too many things 

 at once, as this tends but to confusion ; earlier exercises may 

 well be left somewhat incomplete for this reason. The descrip- 



