80 TRANSACTIONS OF THE [Sess. lxxx\i 



formation of nitrates and ammonia from farmyard manure 

 is not a simple chemical process, as Liebig taught, but 

 that it is due to bacterial action. The chain of evidence 

 has been forged link by link, so that now we recognise 

 that the soil contains innumerable minute organisms 

 (bacteria, etc.) all actively changing the soil, either im- 

 proving its fertility or diminishing it. The soil is like the 

 plate of nutrient jelly of the bacteriologist, the important 

 part is the food for the living organisms. The mineral 

 particles, sand and clay, are mainly a skeleton to carry the 

 jelly or colloid, and to control the supply of air and water. 

 The word colloid was first used by Graham in 1861 in his 

 studies of liquid diffusion. Colloids have great powers of 

 absorption, hence excess of colloid, as in a peaty soil, leads 

 to a poor crop {e.g. heather), whereas the proper colloidal 

 state assists fertility. The utility of soil micro-organisms 

 was demonstrated in one direction by the experimental 

 work on the root-nodules of Leguminosae, chiefly that of 

 Hellriegel and Willfarth described in 1885. Another 

 step was the recognition of nitrification, the formation of 

 nitrates by soil bacteria. This was first suggested by 

 Warington at Rothamsted in 1879, and since then his 

 views have been confirmed that the nitrogen of organic 

 matter is made available for green plants by the activity 

 of soil micro-organisms. The pioneer work on mycorhiza 

 (fungus-roots) on trees by MuUer in Denmark, about 1878, 

 opened another aspect of plant nutrition. This has been 

 extended, so that now a number of species are known to 

 supplement their food supply with the assistance of fungi 

 in the .soil. 



Apart from symbiotic organisms, there is evidence that 

 the plant bears a relationship to the soil that cannot be 

 expressed in terms of abundant or deficient nitrogen, 

 phosphate, etc. When plants are grown under close 

 observation, great differences are known to exist ; one soil 

 is fertile, another is infertile. This problem lias not yet 

 been solved, and it is too early to say yet which of the 

 methods now being tested, if any, will form the test of 

 .soil-fertility. 



Plant Breeding. — Rather more than fifty years ago 

 Darwin's "Origin of Species" brought together numerous 



