ARABLE WEEDS. ASSOCIATION WITH SOILS 125 



large broad leaves nearly 2 inches long. It is an accepted 

 fact that many of these variations constitute true sub-species, 

 but the classification and differentiation are still so hazy and 

 uncertain that it has not yet been practicable to make use of 

 these " splits " of species in studying the weed flora. It is, 

 however, very probable that when matters are cleared up more 

 it will be found fhat some of the main splits are of very great 

 importance and possibly certain splits will prove to be as 

 characteristic of different types of soil as such species as spurry 

 are of sand. Meanwhile we must be content to accept these 

 weeds as occurring on all soils, remembering that we may be 

 dealing with a variety of types instead of with a single one. 



The account of the weeds of general occurrence may 

 conveniently be broken up into sections according to the 

 classes of soils that are more or less particularly favoured. 

 The division is made with due reserve, and is purely artificial 

 in nature, because its truth and convincingness will vary ac- 

 cording to the districts, soils, and cultural conditions with 

 which the reader is acquainted. Weeds that are classified 

 into one section on a general analysis may belong to another 

 section for a particular area, so it must be borne in mind that 

 the classification is the result of observations made on many 

 hundreds of fields scattered over the country, and so may give 

 a fairly true general picture. 



(#) Plants that are Equally Distributed on all Types of Soil. 

 A certain number of weeds appear to be truly ubiquitous, 

 occurring on any and every soil without preference. Naturally 

 enough some of the very commonest farm pests come under 

 this heading, as the faculty of being able to live happily under 

 any circumstances favours the spread and persistence of these 

 plants to an exceptional degree. Some of them are very 

 widely spread, as is shown by the following table which gives 

 the number of times some of the commonest weeds were 

 noted during the investigation, together with their percentage 

 of frequency x on the various classes of soils : 



1 Percentage of frequency = number of times seen per 100 fields. 



