ARABLE WEEDS. ASSOCIATION WITH SOILS 139 



Bath, it was sub-dominant ; in a large proportion of instances 

 it only occurs in traces, isolated individuals here and there 

 being the sole representatives. 



The true charlock (Brassica sinapis] needs to be 

 carefully distinguished from some half dozen other weeds 

 which pass under the same name. Charlock does not 

 exhibit such a marked preference for heavy soil as the fore- 

 going weeds, but a consideration of its dominance or scarcity 

 on the various soils show that it has a very strong preference 

 for the heavier land. On clay, heavy and medium loam and 

 chalk it is very often dominant, but on the light loams and 

 sand it comparatively seldom reaches such a position. On the 

 other hand, it is much more often scarce on the lighter soil, 

 indicating that the habitat is less congenial. The preference 

 for heavy land in this case is probably real, as the plant 

 has no underground part which aids in reproduction, but is 

 entirely dependent upon seed for its propagation. Charlock 

 is very generally considered to be a chalk weed, but investiga- 

 tion shows that in many cases on the chalk the dominant 

 " charlock " is really white mustard (Brassica alba]. 



(/3) Equally Plentiful on Heavy and Medium Soils.-. The 

 weeds described above all prefer really heavy land (heavy 

 loams and clay) to any other type of soil. A few weeds, 

 however, are as much at home on the medium loams, but are 

 much less prevalent on the lighter soils and chalk. 



The large-flowered field speedwell ( Veronica tournefortii) is 

 really an alien, but since its introduction it has spread so 

 rapidly that it is now one of the most common of arable 

 weeds and has to a large extent usurped the place of the 

 ordinary field speedwell ( Veronica agrestis). This fact has not 

 been fully recognised, and as in many points the two plants 

 very much resemble one another V. tournefortii is still 

 easily overlooked and confused with V. agrestis unless special 

 care be taken. In fact, although it was reported from as 

 many as 29 per cent, of the fields visited in the Rothamsted 

 investigation it was certainly present more frequently, as for 

 two years, in Bedfordshire and the West Country, we ourselves 

 failed to make the distinction and put all records under Veronica 

 agrestis. Long, 1 too, fails to distinguish between the two species 

 in his list of weeds, though Buckman ' 2 in 1855 realised the 

 difference. 



p. 364. 



1 Long, H. C. (1910), " Weeds of the Farm and Garden," p. 399. 



a Buckman (1855), " On Agricultural Weeds," Jour. Roy. Ag. Soc., XVI, 



