ANNUAL REPORT, 1934 13 



best. Also the relationship of nitrogen, ammonia, and nitrate content of each 

 plant to the relative abundance of either phosphorus, potash, and nitrogen in the 

 medium is being determined. 



Magnesium Requirement of Crops. (A. B. Beaumont and M. E. Snell.) Barley, 

 Sudan grass, millet, rutabagas, turnips, mangels, spinach, peppers, sweet corn, 

 and field corn were grown in this experiment in 1934. This experiment in its 

 present form has now run 6 years. Including tobacco, grown in the previous 

 experiment, and alfalfa and red clover, grown in pots in soil taken from the ex- 

 perimental field, 20 varieties of field and garden crops have been studied. These 

 crops have differed to a marked degree in their reaction to magnesium deficiency, 

 in yield as well as in physiological symptoms. Buckwheat and spinach were most 

 affected; turnip, mangels, corn, and tobacco considerably; the small grains, 

 grasses, clovers, alfalfa, and potatoes were only slightly affected; and others 

 hardly or not at all. Plants sensitive to magnesium deficiency developed charac- 

 teristic physiological symptoms which have diagnostic value. Chlorosis of the 

 older leaves developed in the intervascular tissue. In leaves of plants with parallel 

 veins this produced a striped appearance; while in leaves having netted venation, 

 a mottled chlorotic pattern was produced. Results of this experiment for 6 years 

 have been summarized for publication. 



Onion Set Production. (M. E. Snell.) During the past summer two new phases 

 of growing onion sets of desirable size for planting have been studied: first, the 

 possibility of controlling the size of set by time of pulling; and second, the effect 

 of later planting on development of desirable sets. It was found that within broad 

 limits it was possible to control the size of sets by time of pulling. By this method 

 one can seed quite heavily (50 or more pounds per acre) and fertilize heavily 

 enough to get good strong growth (1000 pounds per acre on fair onion land), 

 and by spreading the seed well over a broad row get very large yields of sets, with 

 a high percentage of them of any desired size. The past season's results would 

 indicate that good sets may ordinarily be produced by planting as late as May 20. 

 This would sometimes be an advantage, as planting early while the ground is 

 still cold and wet often gives very poor germination. 



The sets grown during 1933 were stored for about six months to check the 

 previous year's findings on keeping quality. The storage loss was somewhat less 

 than for the 1932 sets. In respect to size of set, keeping quality compared fa- 

 vorably with the 1932 results. Different fertilizer treatments did not seem to 

 make any material difference in keeping quality of sets. 



Sets from each fertilizer treatment (with commercially grown sets as checks) 

 were again planted to study influences on yield of first-grade bulbs and produc- 

 tion of seed stalks and multiple bulbs. Marked difference between the two sizes 

 planted was found in development of seed stalks and multiple bulbs, but the net 

 yield of first-grade bulbs was nearly the same in all cases. None of the fertilizer 

 treatments showed marked differences, however, in any respect. 



Onion Breeding. (M. E. Snell.) The past summer was very favorable for 

 breeding work. Seed production was particularly good with most of the selfed 

 bulbs, but not so good with the crosses which were attempted. Many good groups 

 of bulbs were grown from seed sown in the field, as well as a number of very good 

 lots of sets for planting. For the first time seed from some of the inbred strains 

 and crosses was planted in the greenhouse and later transplanted to the field, 

 with very good results. This method permits a much greater number of bulbs 



