8 INTRODUCTORY STATIMEN^T. 



kernel, size of head, season of maturity, etc. Doctor Lyon's exten- 

 sive researches will thus be found very valuable in enabling us to 

 understand more clearlj^ these complex relations and in pointing out 

 the main factors to be considered in breeding wheats to increase the 

 gliadin and glutenin content, and still obtain increased yield and 

 better bread-making qualities. 



The gross selection of wheat seed heretofore has largel3" been based 

 on the separation of large and heavy kernels. Doctor Lyon's re- 

 searches have demonstrated that the smaller and lighter kernels 

 contain the largest percentage of nitrogen, and that while the yield 

 from kernels of this kind at first gives a smaller yield of grain, the 

 total yield per acre of nitrogen is nevertheless greater. By con- 

 tinuous selection of the smaller and lighter kernels for several gen- 

 erations he shows that the grain 3'ield gradually increases and finally 

 approaches or equals the 3'ield derived from the select large and 

 heavy kernels. This gives us a new view of the process of wheat" 

 selection necessar}^ to increase the nitrogen yield per acre. 



The very numerous chemical anal3"ses made by Doctor L^^on give 

 an indication of the great variation of the percentage of proteid 

 nitrogen present in different plants. In the analyses of samples in 

 1902 the plants varied from 2.02 per cent to 4 per cent, while in the 

 analj'ses of the next jesir a variation from 1.20 per cent to 5.85 per 

 cent was found. The existence of this wide variation affords abund- 

 ant opportunity for improvement hj selection. 



Evidence is also given which shows conclusivel}^ that the average 

 composition of a spike of wheat may be judged from the analyses 

 of a row of its spikelets. A satisfactory method of conducting selec- 

 tions has thus been devised. 



The results also show. that early-maturing plants give much the 

 largest average yield, which is a most important point in guiding 

 selection to increase the j-^ield. The percentage of proteid nitrogen 

 is rather less in the early plants, but the total nitrogen per plant is 

 probably greater. 



The quality of the gluten largely determines the bread-making 

 value of a variety of wheat, and it is thus important to keep the 

 ratio of the two elements constituting the gluten — the gliadin and 

 glutenin — the same. Doctor Lyon has shown that as the gluten 

 content is increased b^^ selection the ratio of gliadin to glutenin 

 remains about the same, so that the value of the wheat for bread- 

 making purposes is not impaired. 



The extensive data presented in this bulletin bearing on important 

 matters relating to the improvement of wheat b^^ breeding will 

 enable wheat breeders to plan and conduct their operations with a 

 degree of certainty which would otherwise not be possible. 



Herbert J. Webber, 

 Physiologist in Charge of Lahoratory of Plant Breeding. 



Washington, D. C, MarcJi 30, 1905. 



