- 242 - 



Schlandstedter, and Anderbecker, have not been grown so frequently, 

 but are comparable to the others. The yellow and black varieties 

 never cropped so well at Madryn as many of the white oats. 

 They did better in the early years of the experiment, when the 

 yields were altogether lower and when the land was presumably in 

 poorer condition. Goldfinder proved a useful yellow variety, pro- 

 vided it was sown early enough to ensure ripening in good time." 



R. P. WRIGHT and A. M. M' ALPINE. Report on variety tests of 

 Oats, in Scotland. (West of Scot. Agr. Coll. Ann. Rept. 9 [1909], 

 P a g- 37'77 127-161-167-193, 199-220) E. S. R., vol. XXII, 

 March 1910, n. 4. Washington. 



Experiments were conducted with uniform station-grown seed 

 on twentieth-acre plats on 147 different farms, during the 5-year 

 period ended with the season of 1906. The average yields of the 

 3 typical varieties tested were Banner, 66 bus of dressed grain per 

 acre, Potato, 57 bus, and Sandy, 53 bus. 



In a test of 27 varieties conducted on 33 different farms, 429 iron 

 frames were used, these frames being bottomless and each enclosing 

 a part of the field in which 32 seed oats were sown at uniform 

 distances and under uniform conditions. . " The produce of this 

 miniature acre is intended to show the links of connection between 

 the variety of the oat and the acre yield of the crop. " The cha- 

 racters studied were the mortality of the varieties, their tillering 

 power, proportion of mature straws to barren shoots, straw length, 

 straw bore, ear length, and number of spikelets per ear. 



Studies based on about 200 samples of oats representing 20 va- 

 rieties and grown on 17 different farms, indicated that marks of 

 good seed are proper weight per thousand seeds, and assurance 

 that the seeds have come from prolific ears or heads. A study of 

 maximum and minimum weights per thousand kernels of various 

 varieties showed that among those designated as small oats, the 

 average weight per thousand kernels ranged from 21.8 gm. to 

 23.6 gm. Among the large oats the range was from 25.4 gm. 

 to $*-$ g m ' Even for the same variety of oat, however, higher 

 bushel weight was found to be no guide to higher kernel weight, 

 and manurial applications had little effect on the kernel weight. 

 Data are also presented on the proportion of kernel to husk and 

 its relation to milling power. 



