10 MASS. EXPERIMENT STATION BULLETIN 293 



(Ired poiiiKis of the 0-12-8 did not give a high yield of either total sets or sets of the 

 desired size; but gave a very high percentage of sets of the desired size, especially 

 with 30 and 40 pounds of seed per acre. In general, 30 pounds of seed per acre 

 were not enough for best results, and the three-inch rows seemed to give better 

 opportunity for development than the narrower rows. Several of the plots 

 yielded over 12,000 pounds per acre of sets of the 1/2- to 3/4-inch size. 



Onion Breeding. (M. E. Snell.) This project was continued during the past 

 summer as previously, except for one change in technique. Previously two types 

 of inbreeding had been used. In one, designated as "mass fertilized," all bulbs 

 of each selection were planted close together and just before flowering were sur- 

 rounded by a netting-covered frame which permitted fertilization among bulbs of 

 the same strain. In the other, known as "self fertilized," each flower head was 

 given a separate covering. Since the two methods seemed to give very similar 

 results, and the former was the more expensive, it has been discontinued. 



The seed from the inbred strains, which were crossed a year ago was planted, 

 and a few more cross-matings were made this year. A good crop of bulbs was 

 produced and is being housed in the new storage which was recently completed. 

 This storage should greatly facilitate the work by permitting bulbs to be carried 

 over for breeding purposes, which was not possible with the former facilities. 



Field Experiments with Tobacco. (A. B. Beaumont and M. E. Snell.) 

 Cropping systems. Results of eight years (1924-1931) of experiments on the 

 elTect of cropping systems on the yield and quality of tobacco have been sum- 

 marized for publication in bulletin form. Cover crops were in general beneficial. 

 Of those studied, red top {Agrostis alba) was the most, and timothy {Phleum 

 pratense) the least beneficial. The highest yield was produced where tobacco 

 was grown continuously with fertilizer and a cover crop of red top, but the best 

 quality of tobacco was produced where manure was used as a supplement to the 

 fertilizer. Tobacco grown in rotation with other crops was low in yield and poor 

 in quality. 



In the season of 1932 some changes were made in the cropping systems, in- 

 cluding the substitution of oats for rye as a cover crop. The concentration of the 

 nitrogen in the fertilizer used was raised from 4.1 per cent (used 1924-1931) to 

 5.0 per cent. Red top cover again produced the highest yield (1907 pounds) 

 of all cover crops used, but was slightly surpassed by the manure supplement 

 with no cover. In comparison with no-cover (check) plots, oats, timothy, and 

 rye cover crops depressed the yield in the order given. The yield from the rotation 

 plots was of the same low order as in previous years. The addition of sodium 

 nitrate and calcium nitrate to certain rows of tobacco in the rotation plots did not 

 increase yield but did raise the nitrate content of the soil to a high point. The 

 addition of dried peat in a similar way increased the yield of tobacco. 



Form of nitrogen. The old experiment on single forms of nitrogen was modified 

 in 1932. One half the fertilizer nitrogen was supplied from cottonseed meal and 

 the other half from each of the tollov\ing carriers: sodium nitrate, ammonium 

 sulfate, urea, and calcium cyanamid. The standard fertilizer mixture containing 

 nitrogen from cottonseed meal, sodium nitrate, and calurea was used as a check. 

 In case of tobacco following tobacco, calcium cyanamid and the standard mixture 

 produced the highest yields, followed closely by ammonium sulfate, urea, and 

 sodium nitrate; but after timothy cover or sod, the order was standard mixture 

 (highest), ammonium sulfate, urea, sodium nitrate, and calcium cyanamid. 



A new experiment on different proportions of organic and inorganic nitrogen 



