ANNUAL REPORT, 1931 197 



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



Fertilizer Ratios. Except for a few changes the fertihzer and lime experiments 

 with onions were conducted as they were last year. One ton of lime per acre was 

 applied to the south half of all plots, making possible a comparison between 

 three 1-ton and one 2-ton applications over a period of seven years. Nine of the 

 former cover-crop check plots, used last year for 6-8-8 plots, were changed over 

 to 2-12-8 and 6-12-8 plots. (Figures refer to percentage of NH3, P.Os, and KjO.) 

 The 4-12-8 organic gave appreciably higher yields this year, followed in order by 

 the 4-12-4, 6-8-4, 4-7-4 (2-1-0 first top-dressing), 4-8-4 (nitrate top-dressing), 

 and 8-16-8. The 2-12-8, 4-8-8, 6-8-8, and the 4-12-8 inorganic were all inferior to 

 their adjacent checks. The three 1-ton applications of lime gave higher yields 

 than one 2-ton application in almost every treatment. One ton of lime, applied 

 to the plots of the old lime series which had had heavy applications in 1925, had 

 a depressing effect on yield. Extra applications of superphosphate alone or with 

 lime increased the yield, but of the two the lime seemed to be the more effective. 



Size of Sets. Following the line of work started last year, a comparison was 

 made of commercially graded sets ranging from M to K inch with specially 

 graded sets ranging from K to % inch, to determine whether the range in size 

 of set could be broadened to include more sets without lowering their value for 

 planting. It was found that the commercially graded sets produced a larger 

 percentage of split bulbs and seed stalks as well as a greater variation in size of 

 bulb. After all bulbs which were small, divided, or had seed stalks were deducted 

 from both grades, it was found that the specially graded sets produced slightly 

 more bags per acre than the commercially graded sets. 



Groimng Onion Sets. An experiment conducted last year showed quite conclu- 

 sively that a medium-sized set was far more desirable for planting than either 

 small or large sets. This made it desirable to know how to grow sets which would 

 give a maximum number of the medium-sized sets. Accordingly an experiment 

 was planned having two fertilizer levels, two widths of rows, and three rates of 

 seeding. Best results were produced by .50 pounds of seed and 500 pounds of 

 fertilizer per acre and 2-inch rows. 



Onion Breeding. (M. E. Snellj. This project has been developing grad- 

 ually and expanding along the lines adopted at the beginning. Of the 80 strains 

 and varieties tried, 36 have been discarded as inferior. The past season 992 bulbs 

 which were selections from 142 of the better groups selected from the 1344 lots 

 of bulbs grown in 1930, were planted for seed production. These bulbs repre- 

 sented 31 of the remaining 44 strains. Of the 142 groups planted for seed pro- 

 duction, 13 were used in some crosses which were started this year. Vegetable 

 parchment paper bags were very successfully used for covering the self fertilized 

 heads in place of the cheesecloth coverings formerly used. The paper bags should 

 be much more effective in preventing cross fertilization and seemed to give 

 flowers and seeds better protection from unfavorable weather. The past season 

 284 lots of bulbs representing 14 of the original strains were produced from seed, 

 165 of them from self-fertilized seed and 118 from mass-fertilized seed. It has 

 been observed that within groups there is a marked tendency toward uniformity 

 of characters in the growing seedlings, the grown bulb, and the second-year 

 plants. Some of the more outstanding characterisitcs are: color of leaves, coarse- 

 ness of leaves, and early dormancy in the seedlings; firmness, shape, skin texture, 

 and keeping quality in the bulbs; length of seed stalk, earliness of bloom, produc- 

 tion of viable seed, and number of seed heads per bulb in the second-year growth. 



