18 THE QUANTITATIVE METHOD IN BIOLOGY 



idea that plants and animals are more variable under cultiva- 

 tion and domestication than in the state of nature is adopted 

 by almost every biologist. 



It may be asked, however, whether there is not a certain 

 exaggeration in that belief. 



When a plant species, taken from the state of nature, is 

 cultivated in a garden, we ordinarily grow a large number of 

 specimens side by side. We pay much attention to our plants, 

 looking at them again and again, and comparing them with one 

 another. In this way we discover numerous and often un- 

 expected individual differences. In the state of nature, on the 

 contrary, the specimens of a given species are ordinarily more 

 isolated from each other and therefore we have less opportunity 

 to make comparisons. It happens very rarely that we observe 

 carefully more than half-a-dozen specimens : this brings about 

 the impression that there is less variation than among the 

 cultivated plants. 



If we compare a large number of specimens growing wild we 

 are often compelled to collect them from a rather large area. 

 A botanist who has collected two hundred specimens of a 

 given species from an area of a hundred square miles, who has 

 studied attentively the variation of the collected material and 

 compared this with the variability of an equal number of culti- 

 vated specimens of the same species, belongs to a rarissimu 

 species. 



If we only take the necessary trouble we shall see that varia- 

 tion (it may be complexity or plasticity) in the state of nature 

 is greater than we fancied. In July and August, 1916, I com- 

 pared several hundreds of specimens of Dactylis glomerata 

 collected between Wlthington and Alderley Edge (near Man- 

 chester), along the roads, on meadows, on waste ground and on 

 the heath. Although shady and distinctly wet places were 

 purposely excluded, the plants were very variable with regard 

 to the length of the intemodes, the dimensions of the leaves, 

 the number and the length of the branches of the panicle and 

 the density of the clusters of spikelets. When the extreme 

 forms were placed side by side it seemed at first sight as if they 

 belonged to different species. Phleum pratense was observed in 

 the same way in the same district : in this species the variation 

 w^as on the whole less marked, although notable in the length 

 of the spike-like panicle (from 8 to 166 mm.).^ I have made 

 similar observations in Flanders, observing the variation of 

 several properties of Centaurea nigra, Calluna vulgaris, Stellaria 

 holosfea, Stellaria media, Stellaria uliginosa. Primula elatior, 



^ Holcus mollis is exceedingly variable (plastic) in the Manchester district. 

 The herbarium of the University of Manchester contains an unlabelled 

 specimen of Ph. pratense the inflorescence of which has a length of 194 mm. 



