30G CLASSIFICATION AND DESCRIPTION 



the family or other grGuj) to which ])otanists assigned each 

 plant under consideration, leaving the resemblances and 

 differences thus indicated to be realized more or less vaguely 

 by the student. What was then vague we shall strive now 

 to make more definite, and the student may be assuretl that 

 very much of what he has been learning about economic 

 plants will prove of service in tlie present study. 



83. Early attempts at classifying. Perhaps the reader may 

 ask why it is not sufficient for all purposes of study to classify 

 plants according to their uses, somewhat as we have been do- 

 ing. Such a method of classification was indeed employed by 

 some of the earlier writers upon plants; and this was cjuite 

 natural, since, as we have seen, they were concerned chiefly 

 with plants in their relation to human welfare. But granting 

 that every plant may be of some use (even though not yet 

 discovered) we know that many are useful in more ways 

 than one. Consequently, any classification according to 

 uses would often have to include the same plant in several 

 different groups. Moreover, the great majority of plants 

 are not put to anj' special use, and affect our welfare onl.y in 

 the same general way as do the economic ones apart from their 

 special uses. Hence, any attempt to classify all i)lants ac- 

 cording to use would require us to have besides the economic 

 groups, one general group that would include all plants; 

 and in the subrhvision of this group w(> should be face to face 

 with the original problem. 



One of the earliest attemjits to avoid this diffieulty was a 

 division into herbs, shrubs, and trees. This grouping accord- 

 ing to size and general appearance was a step in the right 

 direction, and for certain purposes is fount! to be a serviceable 

 arrangement even to-day. Yet, aside from the olijection 

 that when ap])lied to all known plants (>ach grou]) includes 

 an enormous numf)er of sorts, there is the further disadvan- 

 tage that such a classification requires onv to plac(^ in differ- 

 ent groups plants which resemble one another more closely 

 than they do any others of the grouji in which they are placed. 

 Thus, for exam])le, certain oaks which are nothing but shrubs 

 would on that account be sei>arat(>(l from all the oth(^r oaks 

 which are trees; the sam(^ is true of willows and of many 



