1 905.] Formation of Permanent Pastures. 393 



M'Alpinc refers to the use he has made of Stebler's tables, and 

 he is enthusiastic in their praise. He writes as if Stebler 

 had reduced the preparation of the seeds-mixture to a simple 

 arithmetical problem. It is only necessary for the farmer to 

 decide what proportion of his land he wishes occupied by 

 cocksfoot clover, &c, and then to figure out the quantities. 

 " Like the engineer, plan first and afterwards calculate the 

 number of pounds of seed required to give effect to our plan." 



Before it is examined, perhaps, the method may seem attractive. 

 Here is science once more coming to the aid of the farmer. 

 The formation of pastures need worry him no longer, for his 

 mixture may be " planned and planned to scale." But does 

 the analogy hold ? Can the farmer plan like the engineer ? A 

 brick has a standard size, and the plans and estimates for a wall 

 present no difficulty, but a cocksfoot plant may occupy a 

 fraction of a square inch or many square inches of surface. The 

 expansion of an iron girder, as temperatures vary, is nearly 

 the same year by year, and it may be provided for ; but the 

 white clover plant of January may have expanded five-fold, 

 or it maybe fifty-fold before July, and how is the "planning 

 farmer " to decide whether he shall allot to it 5 per cent, or 

 50 per cent, of his space ? So much depends on soil, manure, 

 weather, and the other plants composing the mixture, that it 

 seems to me to be quite impossible to forecast the space that 

 may be required by any particular plant from the data supplied 

 to us by Stebler. If experiments were to be made in a particular 

 locality, a table that should be useful might be constructed on 

 the lines indicated by Stebler, but Stebler's original figures 

 cannot be employed in Britain with any hope of advantage. 



In experiments carried out by the Cambridge University 

 Department of Agriculture on sowing down land to grass, 

 there have been many illustrations of the slight connection 

 that may exist between seed-rate and crop, and it will be con- 

 venient at this stage to examine some of the figures obtained. 



At Abbotsley, Hunts, in 1900, and at Saxmundham, 

 East Suffolk, in 1903, eight different mixtures, planned by 

 Dr. Somerville, were laid down. In both cases the soil was 

 a poor and heavy clay ; at Abbotsley, Oxford clay, and at 

 Saxmundham, Boulder clay. The first crop at both stations 



