ORIGIN OF SPECIES 91 



five-lettered and six-lettered groups extremely cumbrous, as 

 Linnaeus found his orders Pentandria and Hexandria. 



De Candolle made an immense advance when, in 1809, he 

 pointed out that plants have certain natural affinities, and 

 that therefore they should be classed according to the sum 

 of these affinities. It is only in this way that a "natural 

 classification ' ' can be drawn up ; but De Candolle be- 

 queathed to his successors an almost endless task. How is 

 the sum of natural affinities to be measured ? How is the 

 product of so many variants to be estimated ? Root, stem, 

 branching, thorns, leaves, stipules, bracts, inflorescence, 

 calyx, corolla, stamens, carpels, placentation, fruits, verna- 

 tion, aestivation ; it is impossible to give a numerical value 

 to each of these variable organs or characters. Who is to 

 decide whether and to what extent marked similarity in one 

 character shall outweigh dissimilarity in many others. 

 Imagine two examiners differing as to whether A or B shall 

 have a prize ? (We are about to spoil a well-known story. ) 

 "I have given B more marks than A," says one, "and if 

 we add your marks to mine B still comes out first, and yet 

 you persist that A is the cleverer boy. On what do you 

 base your conviction?" "On my general impression." 



"Mr. , if your examiners had trusted to their general 



impression, you would never have been in a position to 

 examine for this prize." A general impression has no value 

 in an examination unless it be the sum of a number of 

 particular impressions, each accurately expressed in marks. 

 There is no conceivable plan by which the value of variable 

 characters can be marked for purposes of classification. 



With the publication of the "Origin of Species" (1859) 

 the problems of classification acquired an entirely new and, 

 for the first time, a really natural aspect. So much clearer 

 and more comprehensive was Charles Darwin's theorem of 

 Natural Selection than any of the statements of Erasmus 

 Darwin, Lamarck, St. Hilaire and others who had recog- 

 nised that the fact of the variability of species indicates a 



