SECTIONAL TRANSACTIONS.— M. 443 



and fix characters separately. If other breeders have fixed other characters, 

 he will then be able the more easily to incorporate them in his own herd. 



The dual purpose type (beef and milk) has a certain justification. By 

 suitable crossing immediate adjustments can be made to suit changing 

 conditions both of production and market requirements. Thus a variety 

 of breeds is an advantage to a country, but basal knowledge concerning the 

 inheritance of milk yield is desired in order that in the future such adjust- 

 ments may be made more accurately and more rapidly. 



Discussion of preceding papers (11.40). 



Afternoon. 

 Excursion to Lancashire County Institute of Agriculture, Hutton. 



Saturday, September 12. 



Excursion to farms in the Lake District. 



Monday, September 14. 



Presidential Address by Prof. J. Hendrick on Soil science in the twentieth 

 century (10.0). 



Followed by Discussion on Soil problems. 



Discussion on Soil problems (11.0). 



Prof. G. W. Robinson. — Problems in soil classification. 



Soils may be classified in a variety of ways. The first problem is to 

 decide what is to be classified, i.e. what is the soil-individual. There is 

 now general agreement in regarding the soil profile as the individual ; but 

 some difference of opinion may exist as to the definition of the lower bound- 

 ary of the profile. 



Ideally, a system of classification should illustrate the genetic relationships 

 of the material classified. The elaboration of a system of classification of 

 soils is rendered difficult by the fact that it is the exception rather than the 

 rule to encounter soils in which contemporary pedogenic factors have 

 resulted in climax development. Complications are introduced by human 

 interference and its consequences. Apart from objective difficulties, a 

 comprehensive system of classification can be devised only when it has been 

 possible to consider most of the existing varieties of soil. 



In reviewing soil data, it is important to distinguish the relevant from 

 the irrelevant. Published descriptions of profiles often give minute details 

 which, although they may serve to recall field appearances to the actual 

 observers, are not clearly significant for the purposes of objective definition. 



Profile descriptions, in order to serve as material for studies in classifica- 

 tion, should always include quantitative data expressing the proportion and 

 character of the reactive colloidal material in each recognizable horizon. 



The principal factors determining the course of profile development are : 

 (1) the moisture regime ; (2) the temperature ; (3) the parent material ; 

 and (4) the vegetation. A provisional system of classification is outlined. 



