382 Improvement of Crops by Selection, [august, 



superior types by the old method would necessitate, first, 

 selection in the field and then selection in the plot, i.e., if 

 1,000 ears had been planted in 1,000 plots or rows, the best 

 must be selected from these by weighing and measuring the 

 produce of each plot, and from the residue the best again 

 selected. This is incredibly laborious, and entails the growth 

 and cultivation over many years of many hundreds or 

 thousands of individuals which must ultimately be discarded. 

 If a superior oat plant, for example, could be isolated at the 

 first selection, and at once put into competition with the 

 existing strains of the type which it is intended to displace, 

 much intermediate work could be saved. This is exactly 

 what has become possible. It has been demonstrated that 

 certain characters of apparently negligible importance are 

 actually trustworthy indicators of the productive powers of 

 an individual and of its quality. An example of this law of 

 correlation or association of characters from each of the 

 common cereals will suffice. Oats may be found with 

 panicles of different types, and many of the individuals are- 

 pure. A plant which has the main branches of the whorls 

 longer than the main stem is prolific ; further, in this prolific 

 group is a sub-race in which the larger grains are above the 

 middle of the panicle, and this sub-race is the best; and in 

 this sub-race is a strain in which the colour of the grain is 

 an additional indication of productiveness. With this 

 knowledge, the original selection of types from the field is 

 easy. In wheat the number of kernels in each spikelet, the 

 density of the ear, the shape and venation of the scales, and 

 even the hairiness of the scales, are correlated with practical 

 qualities. A concrete case taken from the record of barley 

 improvement will best illustrate this shorter method of selec- 

 tion. 



A barley was wanted in Sweden with the brewing quality 

 of Chevalier, but without its fatal weakness of straw. To 

 breed a strong-strawed Chevalier had proved impossible, but 

 it had been found that the hairs on the base of the scales 

 of barley were correlated with the composition of the kernel, 

 and that short, stiff, woolly hairs were marks of a good brew- 

 ing barley. A stiff-strawed but coarse barley called Imperial 

 was carefully searched for ears showing this character of hairs, 



