182 HORTICULTURAL SOCIETY OF NEW YORK. 



flowers were found to have developed seed, making a percentage of fertility 

 of eighty-three and one-third per cent, in one-third of the spikelets of the 

 head. Moreover, out of the total of sixty-five grains borne on this particular 

 head, it appears that thirty-five, or more than fifty per cent., were borne on 

 these seven spikelets. 



The attempt will be made to ascertain whether the use as parents for 

 crossing, of plants coming from grains in the more prolific spikelets will be 

 found to result in the production of offspring whose tendency will be to in- 

 crease flower and seed production. It may be urged that the club tendency in 

 this case was simply the result of great vegetative vigor and is not the evi- 

 dence of a mutation form. On account of the severity of the season and the 

 great lack of rainfall during the period when the heads were being formed on 

 the shoots this seems unlikely. 



To ascertain, however, whether we are coming into possession of mutation 

 forms having increased flower production as a distinguishing character we are 

 proceeding in the following manner: The grains in each one of the eleven 

 heads mentioned have been planted this fall in separate rows. In each row 

 the grains from the individual spikelets are arranged in their order from the 

 base to the apex of the spikelet, and the grains of each spikelet succeed each 

 other in serial order in the row. Each grain that grows, therefore, will be 

 definitely referable to its place on the diagram. Next spring all the heads 

 from each plant that grows will be harvested separately, and it is hoped that 

 a statistical analysis of the progeny of the grains from all the different spike- 

 lets of these eleven heads will reveal at least something which may throw light 

 on the facts we are seeking to discover. As a basis for selection of clubbed 

 forms on a larger scale, in case this seems desirable, the remaining si.xty-oue 

 of the clubbed heads of No. 147 have been planted in nursery plots, in which 

 each head is planted by itself, but without any attempt to arrange serially the 

 grains of the different spikelets. 



As an example of the part which natural selection plays in the work of 

 wheat breeding it may be mentioned that out of some three hundred plots of 

 our native cross and pure bred wheats planted last fall ninety-nine were dis- 

 carded in the spring on account of winter killing, while out of one hundred 

 and eight imported wheats from Southeastern Europe furnished by the United 

 States Department of Agriculture fifty-five were discarded for the same rea- 

 son. Some of these turned out to be spring wheats and "durums," whose sur- 

 vival of the winter was not to be expected. 



Further experiments are being carried on in the experiment station to de- 

 termine the relative value of large and small seeds in cereal growing. Experi- 

 ments conducted at the United States Department of Agriculture and elsewhere 

 indicate a marked difference in plants growing from large as compared with 

 those growing from small seeds. Experiments are being conducted by us to 

 demonstrate this fact in the case of the cereal grains and to use the informa- 

 tion obtained as a basis for urging the general use of seed-grading machines 

 among our farmers. 



As an economic cereal Indian corn ranks as an equal with wheat in Kan- 

 sas in point of acreage. Thus far our experiments in corn breeding have been 

 directed almost entirely toward the increasing of the nitrogen content. Since 



