632 EXPERIMENT STATION EECOED. 



grass showed indications of becoming established if persistently tried on plowed 

 ground, although their introduction was not considered practical. 



Measuring hay in ricks or stacks, H. B. McClure, W. J. Spillman, and 

 J. W. Feoley (U. S. Dcpt. AffT., Bur. Plant Indus. Circ. 131, pp. 19-24, flffs. S).— 

 Ttiis paper gives methods and formulas for measuring the volumes of hayricks 

 and stacks, the heights of stacks and the number of cubic feet in a ton of hay. 

 These methods were determined as the result of a study of the subject covering 

 a period of several years and the actual measurements in 92 cases. 



The improvement of small grains at Macdonald College, L, S. Kxinck 

 {Proc. Amer. Soc. Agron., // (1912), pp. 126-129). — The author here describes 

 the centgener system employed in the improvement of cereals by selection. 



As the result of 5 years' work with this system, from 500,000 single plants 

 only the best from 116,000 planted in the spring of 1907 have thus far been 

 tested side by side with the parent sorts. From 18 of the most productive va- 

 rieties, representing 4 classes of wheat, 3 of barley, and 2 of oats, 37 strains 

 have been isolated, which on an average of the last 2 years have yielded from 

 1 to 5 bu. per acre more than the parent sorts grown side by side with them. 



" It is especially worthy of note that these average increases for 2 years were 

 not obtained over the original, unselected parent stocks, but were obtained over 

 original stocks, which have, for the past 5 years, been subjected annually to 

 the most careful hand selection of heads in the field and of grain in the labora- 

 tory." 



Methods of selection and choice of varieties in Russia, A. Stebut (Jux. 

 Russ. Selsk. Khoz. Oaz., 1912, Nos. 9, pp. 7-10; 10, pp. 1-10; 11, pp. 5-7; abs. 

 in Intemat. Inst. Agr. [Rorne^, Bui. Bur. Agr. Intel, and Plant Diseases, S 

 {1912), No. 6, pp. 1301, 1302).— In this article the author gives 10 suggestions 

 that should be observed in selection and improvement experiments with cereals 

 by Russian investigators. 



Crossbreeding experiments with oats and wheat, H. Nilsson-Ehle (Lunds 

 Univ. Arsskr., n. ser.. Sect. 2, 7 {1911), No. 6, pp. 3^6). — Earlier work on 

 these experiments has already been noted (E. S. R.. 26, p. 831). The author 

 discusses the views of various investigators in regard to the two or more factors 

 involved in modification and variation in crosses or hybrids, and presents addi- 

 tional evidence from his own researches to throw more light on the question. 



The appearance of more than a single color factor in still other kinds and lines 

 than heretofore observed was noted. It was shown that each of the white 

 kernel individuals resulting from crossing 2 red kernel sorts possessed a red 

 color factor. It was also absolutely proved that the truly outward characters 

 of a pure line are constant, but not its gametic characters. 



In a study of the length of internode in the wheat spike, compact spike 

 Swedish varieties were crossed with longer spiked varieties. It was deter- 

 mined that the inherited gradations of a length character occurs through the 

 various interaction of numerous Mendelizing factors, and that these factors 

 are of 2 classes, one class working together as lengthening factors, and the 

 other in the opposite direction as shortening factors. It was found that through 

 the influence of an epistatic factor an apparently noncontinuous variation may 

 be changed into the most delicate kind of continuous variation. A continuous 

 inherited variation can occur therefore in different ways through the combination 

 of Mendelizing factors. The formation of gradations which exceed the limits 

 of the parents in crossing quantitative characters can be explained as resulting 

 from new grouping of Mendelizing factors. 



In 1904 the author observed a wide variation in the rust-resistant ability of 

 certain similar crosses of wheat, and that this character seemed to segregate in 



