836 EXPERIMENT STATION RECOBD. [Vol. 86 



weekly intervals, at the rates of 2, 4, 6, and 8 pk. per acre, respectively, from 

 September 16 to October 20. In 1913 and 1914 the experiments were repeated 

 with the additional dates of September 8, October 28, and November 4, Tlie 

 yields of grain and straw for each date and each rate of seeding for each of 

 the three years are given in tabular form, together with the averages for the 

 3-year period. The results are also shown graphically. 



According to these results the 2-pk. rate of seeding produced as large yields 

 as the 4-, 6-, or 8-pk. rates when sown before the last week in September, but 

 after that date the heavier seeding produced larger yields on the average. This 

 condition progresses until the 8-pk. rate, sown between the second week of 

 October and November 1, produces nearly twice as large yields as the 2-pk. 

 rate. The high yield of the 2-pk. rate can be explained by the heavy stooling 

 of Turkey winter wheat when sown early in the season, while as the season 

 advances this stooling decreases. 



Field investigations have shown that it is unsafe to sow winter wheat any- 

 where east of the western third of Kansas earlier than September 25 in 

 northern Kansas, and October 5 in southern Kansas, owing to the Hessian fly. 

 It is evident, however, that such factors as rate and date of seeding must be 

 kept in mind when efforts are being made to increase crop yields. 



Wheat, yields per acre and prices, by States, 50 years 1866—1915 (17. S. 

 Dej)t. Agr. Bui. ol-i {1917), pp. 16). — Statistics are given for wheat similar to 

 those for corn noted on page 832. 



Table for converting weights of mechanical separations into percentages 

 of the sample analyzed, E. G. Boekner {U. S. Dept. Agr. Bui. 516 (1916), pp. 

 21, figs. 2). — The regulations under the United States Grain Standards Act 

 provide that an original sample of shelled corn must not be less than 2 qt. in 

 quantity and that about IJ pints must be placed in an air-tight container for 

 moistui'e determinations. The remainder is to be placed in a cloth bag for the 

 determination of color, damaged corn (not including heat damage), heat-dam- 

 aged corn, and foreign material and cracked corn. The maximum and mini- 

 mum percentages allowed for each of these factors is specified for each grade 

 and these percentages are to be determined by weight. To save time the 

 2.5-pint samples are very frequently divided into smaller samples of approxi- 

 mately 250 gm. and, after the mechanical separations and weighings have been 

 made, mathematical calculations are necessary to convert these weights into 

 percentages. A table given in this bulletin enables the analyst to compute these 

 percentages without any calculations for percentage equivalents, ranging from 

 to 40 for samples weighing from 240 to 260 gm., inclusive. These limits 

 cover the entire range of corn of other colors, damage, heat damage, and for- 

 eign material and cracked corn specified in the numbered Federal Corn 

 Standards. 



Directions are given for the use of the table. It is deemed essential for 

 accuracy that considerable care be exercised in obtaining the 250-gm. sample 

 from the 2.5 pints of the original sample and, to insure representative sampling, 

 it is recommended that some such device as that described in Bulletin 287 

 (E. S. R., 33, p. 836) be used. 



Report of the department of seed analysis, J. P. Helyab (New Jersey Stas. 

 Rpt. 1915, pp. 143-150). — The results of seed analyses for 1915 are reported and 

 briefly discussed. Twenty per cent of the alfalfa seed examined showed the 

 presence of Turkestan alfalfa, which, in a large number of cases, had been 

 used as a substitute for American-grown seed. 



An alien weed (Jour. Dept. Agr. So. Aust., 19 (1916), No. 10, pp. 898-900, 

 fig. 1). — This is a brief note on the appearance of Xanthium canadense in 

 South Australia, with directions for its eradication. 





