38 MISSOURI AGRICULTURAL REPORT. 



able ears to plant a block from 10 to 15 hills square. The general 

 character and yield of each plot is accurately recorded as in the row 

 method. The idea in this method is to prevent cross pollination, since 

 the mixing" of the seed before planting insures quite thorough crossing 

 between the progeny of the two ears planted. The method requires 

 that the land be extremely uniform in character and it does not permit 

 of such accurate data being kept as to individual ears. The method 

 evidently requires considerably more time in developing a particularly 

 desirable strain than the row method. 



The breeding of corn for a change in its chemical composition is 

 being conducted at the Illinois station with considerable success, and 

 this work has been receiving no little comment, especially at the hands 

 of writers for popular magazines. The details of this work are quite 

 complicated and only the general method will be mentioned here. It 

 should be stated that it has been found possible to very greatly change 

 the proportions of protein, starch and oil in the grain of a particular 

 variety of corn by a careful selection, accompanied by chemical analysis 

 to indicate the varying proportions of these constituents. Strains have 

 been developed some of which yield over twice as much protein or 

 almost three times the amount of oil as do others, while the starch can 

 be changed accordingly. The breeding plots for this work are arranged 

 after the order of tiie row method before mentioned, the selection being 

 made largely on the basis of high protein, high oil or high starch, as the 

 case may be, with a fair consideration of yield per acre. It has been 

 found possible in this work to rely partially on a simply physical ex- 

 amination of the grain which lessens considerably the cost of work by 

 decreasing the number of chemical analyses necessary. It has been 

 shown for instance, that the germ contains the greater proportion of 

 the oil and by selecting for large germs a fairly accurate estimate can 

 be made of the proportion of oil in the grain. The protein is very 

 largely contained in the clear or horny starch of the grain so that by 

 selecting ears whose grains show a large proportion of this horny starch 

 there is every probability of getting the high protein cars. It should 

 be noted that while there is some variation in the composition of in- 

 dividual grains on the ear they run fairly uniform on the whole so that 

 the composition is an ear characteristic rather than a characteristic of 

 individual grains and consequently by determining the protein or oil in 

 a few grains from an ear a very accurate indication is given of the total 

 protein or oil content. The means of developing varieties of corn thus 

 varying in composition comes from the fact that for special uses as for 

 feeding, for starch manufacture or for production of corn oil, the corn 



