736 EXPERIMENT STATION EECOED. 



Intensive methods and systematic rotation of crops in tobacco culture, 

 E. H. Mathewson (V. »S'. Dept. Afir. Yearbook 1908. pp. J,()3-Ji20, pis. J).— Tins 

 r.rticle discusses the wastefulness of early methods, present conditions, the 

 adaptation of tobacco to intensive methods, fertilizers for the tobacco crop, 

 effects of crop rotation on the soil, and the effects on the different crojjs grown 

 in succession. The experimental results and observations described are based 

 on material previously noted from other sources (E. S. R., 19, p. 335: 20, 

 p. 833). 



The varietal characters of Indian wheats, A. and G. L. C. Howard {ilcm. 

 Dept. Agr. India, Bot. Her., 2 {1909), No. 7, pp. 11+66). — This is a critical 

 study in which the different characters are discussed with reference to the 

 relation they bear to the length of the growing period and the utilization of 

 the available soil moisture. A classification and a description of the wheats 

 of the Punjab are given and the quality of Indian wheat and natural cross 

 fertilization in India are discussed. A conspectus of the species of wheat is 

 outlined. 



Methods of plant breeding as tested by experiment, K. von IiImkkk 

 (Methoden der rflanzenxuchtung in ccperimentellcr Priifung. Berlin, 1909, 

 pp. VII+321, pi. 1, figs. 7). — In this work the author discusses at length the 

 results secured during a series of years in breeding rape, rye, and wheat. 



The determination of the proportion of first-grade kernels in grain, R. 

 Kegel (Trudiii Bynro Prikl. Bot., 2 U909), Xo. 3, pp. /77-/77).— The author 

 describes and points out the advantages of a ujethod by which the proportion 

 of high-grade kernels in grain may be approximately determined. 



From an average sample, 2 lots of 300 kernels each are taken and weighed. 

 The difference in weight between the 2 lots should not exceed 0.3 gm. From 

 each lot 100 of the heaviest kernels are selected and weighed, the limit of 

 difference in weight between these 2 samples being 0.01 gm. The difference 

 between the lt>0-kernel weight of the selected and the average samples is taken 

 as indicating the degree of uniformity in the grain and incidentally the need 

 of improving either the grain or the results already secured in this line of work. 

 It is shown that a comparison of these 100-kernel weights points out any 

 deficiency of the grain in weight. 



Method of making germination tests, H. Pieper (Jour. Landw., 57 {1909). 

 iVo. 2, pp. 137-lJp). — The present methods of making germination tests are dis- 

 cussed and a new method is proposed and its advantages pointed out. 



The essential feature of the proposed method is to regard kernels failing to 

 germinate as foreign material or impurities. The following concrete examples 

 are given for the purpose of illustration : A sample of orchard grass seed 

 weighing 1 gm. contains 0.1 gm. of foreign seeds and the rest of the sample is 

 composed of l.OSO filled and 120 empty kernels. Of the well-filled seeds. 112 

 germinated in 5 days, 780 in 6, 62 in 7, 4 in 10, 4 in 14, and 2 in 21 days, mak- 

 ing a total of 1,000 germinated seeds or 92.6 per cent of the number of filled 

 kernels. A factor known as energy of germination is determined from the 

 percentage of germination at the end of 7 days, which in this case is 88.15 per 

 cent, 952 of the 1.080 kernels having germinated within this time. The weight 

 of the empty kernels is 0.05 gm. which, added to the weight of the foreign 

 seeds, gives a total of 0.15 gm. or 15 per cent, thus producing a purity of the 

 sample of 85 per cent. 



According to the newly proposed method the filled seeds are not counted, but 

 after inert matter and foreign seeds have been removed the sample is ger- 

 minated and the weight of the kernels not germinating is deducted from the 

 total weight of the sample and calculated to a percentage basis. The average 



