124 



of its life-history. However great such differences might "be 

 they would obviously not justify our classing different forms 

 of one and the same organism as distinct species. When 

 defining species, therefore, care must be taken not only to 

 observe what characters are likely to vary in the immediate 

 offspring of one and the same individual plant, but also to 

 note how the characters of the same plant vary at different 

 periods of its life-history. An imperfect knowledge of life- 

 histories has led to mistakes being made in the classification 

 of some fungi, different stages in the life-history of one and 

 the same individual having been sometimes described as entirely 

 distinct species. The words " always able to transmit" in 

 definition 1, while indicating constancy under varying condi- 

 tions of existence, also imply that, while plants capable 

 of both sexual and asexual reproduction cannot constitute a 

 species if they only transmit their essential characters truly by 

 asexual reproduction, organisms only capable of asexual repro- 

 duction are not thereby precluded from forming a true species. 



The characters which are used to define species must 

 not only be constant, but, also, in order to facilitate identi- 

 fication, they should, as far as possible, be such as can be 

 easily recognised, and further, to be of use in written descrip- 

 tions, they must be such as can be easily described in words. 



Among the higher flowering plants, with which we are 

 chiefly concerned, characters of the floral structure are as a 

 rule most constant and important and, hence, special atten- 

 tion is paid to them in most Floras, but these, owing to their 

 minuteness, are often difficult to recognise, while flowers are 

 also only available at certain seasons ; just as, however, the 

 savage, who pays no attention to such minute characters, 

 is still able to recognise the different trees in the forest, so can 

 also the expert forester, by utilising characters referring to 

 the buds, leaf-scars, kind of bark on young and old stems, 

 the method of branching, the colour of the foliage at different 

 seasons, and others, which, as a rule, are not included in 

 botanical books, owing to their not being easily described in 

 words, to their not being easily recognisable in herbarium 

 specimens, or to their not being sufficiently constant over 

 large areas and in different localities. For this reason, also, 

 a botanical classification based on a single character is not 

 necessarily artificial, for such a character may be correlated 

 with several others, the latter having been omitted from the 

 scheme of classification owing to their being difficult to de- 

 scribe or to distinguish. ^ 



