350 Transactions. 



differs from the coastal Nuggets form in some minor details, but 

 only in such as might be expected from ordinary fluctuating 

 variability, so that the two plants according to recognised 

 floristic rules must be considered identical. 



But are such floristic methods sufficient in this case ? Mr. 

 Matthews thus writes, and shows very clearly the position he 

 takes with regard to the coastal and alpine forms : "In regard 

 to Celmisia lindsayi and the Bonpland plant (sent yesterdav), 

 there is perhaps no clear actual botanical distinction, but when 

 seen growing side by side they are dissimilar in many respects 

 that could not be reduced to writing." This statement of Mat- 

 thews opens up a very wide and most important question, but 

 here a few general remarks must suffice. Professor L. H. Baile 

 thus writes : " Many of us feel that the present methods of no- 

 menclature and description will be outgrown, for these methods 

 are made for the herbarium and museum rather than for the 

 field. It is a most suggestive commentary that the botanist 

 may know the species when it is glued on an herbarium sheet, 

 but may not know it when it is growing. The nurseryman or 

 gardener may know it when growing, but not when it is in an 

 herbarium. This is not merely because the botanist is un- 

 familiar with the field or the gardener with the herbarium. 

 These men have a different fundamental conception of what a 

 species is; they use different 'marks' — one morphological, the 

 other largely physiological. I believe that the gardener is 

 nearer the truth."* The fact is, such cases as the one under 

 consideration, and dozens of others much more striking which 

 could be selected from the New Zealand flora, cannot be settled 

 by a mer.^ morphological examination. The truth does not 

 rest on the dictum or perhaps whim of one man, or indeed of 

 a number of men, but upon the observation of a simple fact — 

 the power of the particular form in question to reproduce itself 

 " true," or the contrary, from seed. If the progeny resembles 

 the parent in those characters which distinguish this latter 

 from all its allies, then we have to do with a distinct entity — an 

 elementary species, as De Vriesf has termed it — and such must 

 receive a name ; such elementary species are realities, whereas 

 collective Linnaean species are merely ideas. The final court of 

 appeal as to " specific value " is no longer the herbarium or 

 study of the systematist, but the seed-bed of the experimental 

 garden. 



* " The Mutation Theory of Organic Evolution " : Six addresses 

 given before the American Society of Naturalists at Philadelphia, "28 th De- 

 cember, 1903. "Systematic Work and Evolution," L. H. Bailey (Repr. 

 Science, n. s., vd. xxi. Xo. 536, 7th April, L904, p, 12). 



t H. de Vries, ''Die Mutationstneorie," hand i, chapter v, §21,. 

 "Species, Subspecies, and Varieties," pp. 115-20. 



