UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA EXPERIMENTS WITH THE TRANSPLANT ONION CROP IN CALIFORNIA GLEN N. DAVIS and H. A. JONES BULLETIN 682 January, 1944 UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA CONTENTS PAGE Review of literature 3 Spacing tests 4 Time-of -transplanting studies 6 Size-of -seedling studies 10 Pruning studies 13 Time-of -harvest studies 15 Summary 19 Literature cited 20 1 EXPERIMENTS WITH THE TRANSPLANT ONION CROP IN CALIFORNIA 1 GLEN N. DAVIS 2 and H. A. JONES 8 The transplant onion crop of California consists chiefly of the early and the so-called "intermediate" varieties. The early crop includes mainly Crystal Wax, Early Grano, Crystal Grano, and San Joaquin. These are grown pri- marily in the upper San Joaquin, Coachella, and Imperial valleys and in small scattered acreages elsewhere in the state. The intermediate crop is grown chiefly in the lower Sacramento and San Joaquin valleys and to some extent along the coast, south of San Francisco. The varieties are chiefly Stockton Yellow Globe, Red 21, Stockton G36, and Italian Red. Sweet Spanish, a late- maturing onion, is also used to some extent for the transplant crop, especially in Riverside, Los Angeles, and Kern counties. The transplant crop comprises approximately one half of the total annual onion acreage in the state. The culture is somewhat similar for the different varieties and for the dif- ferent districts in the state. The seed of practically the entire transplant crop is sown in open beds in the field, usually in late August or September (accord- ing to the variety and the district) . The seedlings are taken to the field during late fall or winter. Although there is a general conformity to certain practices, there is little specific information about desirable practices in spacing, time of transplanting, size of seedling used, and pruning (see figs. 1-4) . Data reviewed or reported herein from California and several other states have thrown con- siderable light on some of these problems. REVIEW OF LITERATURE In the lower Rio Grande Valley of Texas, Hawthorn (1)* secured heavier yields for four consecutive years from the closer spacing when he compared rows 14 and 16 inches apart, while spacing the plants uniformly in the row. In Montana, Starring (5) compared Sweet Spanish transplants set 3 and 6 inches apart in the row. Although the weight per bulb was heavier for the 6-inch spacing, the heavier per-acre yields were secured for the 3-inch spacing. In Utah, Wilson (6) seeded Sweet Spanish onions directly in the field and then thinned to get the desired spacing of 2 to 10 inches between plants. With each increase in distance between plants came a corresponding increase in the percentage of doubles, and simultaneously a decrease in yield. Thinning beyond 4 inches reduced not only the yield, but also the percentage of large bulbs. In spacing tests with California Early Red in California, Jones {2) found that with each increase in spacing of 3 to 12 inches came a delay in time of maturity, an increase in the size of bulbs, and a decrease in the yield per acre. 1 Received for publication March 26, 1943. 2 Assistant Professor of Truck Crops and Assistant Olericulturist in the Experiment Station. 3 Principal Olericulturist, Division of Fruit and Vegetable Crops and Diseases, Bureau of Plant Industry, Soils and Agricultural Engineering, Agricultural Eesearch Administration, U. S. Department of Agriculture. 4 Italic numbers in parentheses refer to literature cited at the end of this publication. [3] 4 University of California — Experiment Station The results of priming reported by Hawthorn (i) for Yellow Bermuda were somewhat inconsistent from year to year. For a five-year average, how- ever, the unpruned plants gave the highest yield, and those with both tops and roots pruned gave the lowest. Plants with only tops or roots pruned were intermediate in yield. Kraus (3), working in Wyoming with the Early Grano variety, compared unpruned transplants with those having about half the foliage removed. He found no significant difference between pruned and un- pruned plants in the total number of individuals that survived, or in the mean weight per bulb, or in the number of marketable bulbs produced. Hawthorn (1), comparing different-sized transplants of Yellow Bermuda, found that those of medium size, weighing 5 to 7 grams and having a neck TABLE 1 Weight of Onion Bulbs Grown at Different Spacings in the Row Year 1927 1928 1929 1930 1931 1935 1935 1935 1942 1942 Location Davis Davis Davis Davis Davis Liberty Island Liberty Island Liberty Island Shafter Davis Variety California Early Red 21-6 . California Early Red 21-24 California Early Red 21-6 . California Early Red 21-24 Red 21 Early Grano Stockton G36 Red 21 San Joaquin Crystal Grano Mean weight of bulbs (pounds) at the given spacings (inches") 24 0.36 0.26 0.12 0.32 0.39 .41 .65 .41 .45 .36 .47 .38 .17 0.36 0.48 .51 .79 .55 .52 .38 .50 .45 .19 0.44 0.62 .66 .99 .83 .71 .44 .60 .50 .22 0.49 0.72 0.84 1.16 0.96 0.84 0.46 0.60 0.54 12 0.78 0.99 1.28 0.96 Difference required for signifi- cance* 0.07 .08 .10 .08 10 05 08 05 05 06 * Difference required for significance at the 5 per cent point. diameter of 6 to 9 millimeters, produced more U.S. No. 1 onion bulbs than either larger or smaller transplants. If the transplants are very large, a high percentage may form splits, doubles, and seedstems. Jones (2), in California, found that large transplants of California Early Ked produced a higher yield of bulbs and that the average size of the bulbs was larger. SPACING TESTS Methods. — Spacing tests were conducted at Davis from 1927 to 1931 with different lines of California Early Red. Distances between plants in the row were 3, 4, 6, 8, and 12 inches except in 1930, when the 12-inch spacing was omitted. The rows were 18 inches apart. In 1927, they were 48 feet long; in 1 928, 36 feet ; in 1929, 1930, and 1931, 27 feet. Five replicates were planted in 1927, 1929, and 1930; three in 1928 and 1931. In 1935 the spacing tests were located on Liberty Island near Rio Vista, ( 'alifornia. The varieties used were Red 21, Stockton G36, and Early Grano. The plants were spaced 2, 3, 4, 6, and 8 inches in the row. The rows were 15 inches apart and 27 feel long. Four replicates were planted, and each spacing was flanked by guard rows of the same spacing. In 1942, spacing tests were located at Shafter and Davis. The variety San Joaqnin was planted a1 Shafter; Crystal Grano at Davis. Plants were spaced Bul. 682] Experiments with the Transplant Onion Crop 2, 3, 4, and 6 inches apart in the row. The rows were 18 inches apart, 27 feet long. Each spacing was replicated five times. All data treated statistically were reduced by analysis of variance. Results. — The data for 1927 to 1929, though presented earlier by Jones {2) , are included here as parts of tables 1 and 2 to bring together the work on spacing. Results for the first three years showed definitely that the closest spacing matured earliest, and as the spacing increased the plants matured later. Apparently, therefore, environmental factors in addition to photoperiod and temperature may influence the time of maturity. Data on time of maturity were not collected after 1929. TABLE 2 Yield of Onions Grown at Different Spacings in the Row Year Location Variety Yield (100-pound bags per acre) at the given spacings (inches) Difference required for signifi- cance* 2 3 4 6 8 12 218 280 380 278 1927 California Early Red 21-6 California Early Red 21-24 .... California Early Red 21-6 California Early Red 21-24 .... Red 21 195 676 526 175 374 456 429 738 528 515 246 588 502 162 346 409 416 686 485 447 176 488 448 138 299 358 376 573 465 404 172 394 332 107 213 309 357 529 449 366 102 286 256 45 1928 48 1929 102 1930 — t 1931 51 1935 Liberty Island .... Liberty Island .... Liberty Island .... Shafter 78 1935 Stockton G36. .. 82 1935 1942 Red 21 33 41 1942 162 * Difference required for significance at the 5 per cent point. t No significant difference. Every increase in spacing except one was accompanied by an increase in average weight of bulb (table 1) ; but these increases were not always statis- tically significant. The 3-inch spacing seemed to produce bulbs of the best size. The necks were small, and they cured more readily than plants grown at the wider spacing. Yields per acre were always highest at the 2- and 3-inch spacings. A decrease in yield per acre accompanied each increase in spacing except for Early Grano at the 3-inch spacing. The low yields for Early Grano in 1935 were caused by the high percentage of bolters (table 3). The extremely low yield for San Joaquin at Shafter in 1942 resulted from a rather unfavorable growing season and from lack of sufficient labor to complete the essential cultural operations. However, results from spacing are directly comparable to other plots and in no case did the increase in size of bulb at the wider spacings compensate for the decrease in the number of bulbs harvested. On Liberty Island in 1935, spacing at different distances produced no signi- ficant differences in the percentage of bolters within varieties (table 3) . As is well known, the size of plant is important in determining the percentage of bolters ; but in these studies the increase in size of plant at the wider spacings no doubt occurred too late in the season to be influenced by the environmental conditions that influence bolting. Although the onion varieties differed con- siderably in yield, they could not be compared statistically because they had 6 University of California — Experiment Station been planted separately in several parts of the field. Stockton G36 produced more seeders than Red 21 even though, in nearly all other plantings where these two varieties could be compared, the latter produced the highest percent- age of bolters. TIME-OP-TRANSPLANTING STUDIES Methods. — These studies were conducted at Davis from 1927 to 1934. From 1927 to 1930, different lines of California Early Red were used, whereas in 1931-32, Red 21, Stockton Yellow Globe, and a line of Italian Red were TABLE 3 Percentage of Bolters at Different Spacings, at Liberty Island, 1935* Spacing, inches Early Grano Stockton G36 Red 21 2 59.7 49.0 53.8 39.4 58.0 3.98 6.47 2.24 2.62 3.38 1.00 3 1 62 4 00 6 0.92 8 1 81 * No significant difference appeared in percentage of bolters (within varieties) at different spacings. planted. In 1927 the plots were planted in duplicate; in other years, in tripli- cate. Single-row plots were used. Seeding was done at the usual time in late August or early September ; transplanting, at different times in the late fall and winter. Though only one seeding was made, there were six to eight trans- planting dates. The 1933-34 experiments were handled somewhat differently. Early, intermediate, and late varieties were included. Most of them were from TABLE 4 Effect of Time of Transplanting on Mean Weight per Bulb and Yield per Acre in California Early Red 21-6, at Davis, 1927 Date transplanted Mean weight of trimmed seed- lings, grams Bulbs harvested per plot Mean weight per bulb, pounds Yield per acre (100-pound bags) January 18. January 25. February 1 . February 18 February 22 March 1 . . . . 1.56 1.10 1.26 2.18 2.11 1.58 138 128 144 136 136 136 58 .50 .50 .41* .36* 35* 489 389 436 336t 293* 283* * Highly significant differences when compared with the January 18 planting. t A significant difference when compared with the January 18 planting. seed provided for a study of varietal types (4), and the strains are desig- nated by the numbers originally assigned by the United States Department of Agriculture. Two seedings were made of this material, an early and a late. The early one came on September 6, 1933; the late one on November 6. The early- ^"' the data for yield per acre, the larger seedlings produced the heaviest yield except in the hist two classes, ;is given in the table. The reduction in yield here no doubt results from the large number of bolters in these two classes. The increase in size of hull) did not compensate for the reduction in the number of Bul. 682] Experiments with the Transplant Onion Crop 13 bulbs harvested. Where the seedlings planted were under % inch in diameter the classes showed no significant differences in the percentage of bolters. Bolt- ing was significantly more common in the classes over % inch than in those below that size. Except in the %-inch-or-less class there was no significant dif- ference in the number of bulbs harvested per plot. Many seedlings under % inch in diameter failed to survive transplanting operations. PRUNING STUDIES Methods. — Pruning experiments were conducted in 1930, 1931, 1934, and 1942. In 1930 the variety used was California Early Red 21-24; in 1931, Stockton Yellow Globe ; in 1934, Red 21 ; and in 1942, San Joaquin and Crystal Grano. Four different pruning treatments were given : unpruned, roots TABLE 13 Effect of Seedling Size, in the Variety Babosa, on Percentage of Bolters, Mean Weight per Bulb, and Yield per Acre, at Davis, 1942 Diameter of seedling, inches Bulbs harvested per plot Bolters, per cent Mean weight per bulb, pounds Yield per acre a00-pound bags) 38 54 62 66 55 38 2.6 4 4 22 4 16 0.18 .23 .58 .79 .74 0.83 76.3 Yr^i 137.3 M-% 397 H-H 565.2 444.1 311.0 Difference required for significance. . . 14t 19* 11. 6t 16 Oi 12f 0.16| 41. 9t 57. 8t * Did not enter into the analysis. t Difference required for significance at the 5 per cent point. % Difference required for significance at the 1 per cent point. pruned, tops pruned, roots and tops pruned. In 1930, 1931, and 1942 the tops were trimmed 5 inches above the stem plate ; in 1934, 6 inches above it. Roots were trimmed to a length of 1 inch. This method of trimming both roots and tops conforms closely with commercial practice. All plants were set in shallow furrows made with a hand plow so that the long roots of the unpruned plants could be covered. Transplants were set 4 inches apart in rows 18 inches apart and 27 feet long. In 1930 there were seven replicates ; in 1931, three ; in 1934, fourteen ; and in 1942, five. In 1930 the seedlings were transplanted on Febru- ary 8 ; in 1931, on February 3 ; the 1934 crop, on November 21, 1933; and the 1942 tests, on December 19, 1941. Results. — The pruning of onion seedlings to facilitate transplanting is a common practice. To what extent it influences yield has not been definitely determined. According to the results of other investigators, pruning is of no benefit except in making transplanting easier. At Davis, California, in 1930, the unpruned seedlings of California Early Red 21-24 produced a larger bulb and a greater yield per acre than any of the pruning treatments (table 14). Cutting back both roots and tops of an individual plant reduced the weight of 14 University of California — Experiment Station the mature bulb. Pruning either the roots or the tops gave the plant a less severe check than pruning them both. Root pruning seemed to produce the least injury. In 1931 (table 15) the differences in weight of bulb between the various treatments were not significant. The treatments arranged themselves, how- TABLE 14 Effect of Pruning Onion Transplants on Stand, Mean Weight per Bulb, and Yield per Acre, in thei Variety California Early Red 21-24, at Davis, 1930 Treatment Unpruned Roots pruned Tops pruned Roots and top3 pruned Difference required for significance Bulbs harvested per plot 78 No significant difference Mean weight per bulb, pounds 0.69 0.62 0.62 0.57 0.03* 0.06f Yield per acre (100-pound bags) 589 534 527 47S 33* 45f * Difference required for significance at the 5 per cent point, f Difference required for significance at the 1 per cent point. ever, in about the same order (relative to bulb size) as in 1930. Unpruned^ seedlings and those with only the roots pruned produced the heaviest bulbs. In yield per acre the transplants with both tops and roots trimmed yielded sig- nificantly less than the unpruned or those with only the roots pruned. In 1934 the unpruned lots were significantly heavier, in both weight per TABLE 15 Effect of Pruning Onion Transplants on Stand, Mean Weight per Bulb, and Yield per Acre, in the Variety Stockton Yellow Globe, at Davis, 1931 Treatment Bulbs harvested per plot Mean weight per bulb, pounds Yield per acre (100-pound bags) 78 SO 78 78 0.51 50 49 44 427 428 412 Roots and tops pruned 367 Difference required for significance. . . No significant difference No significant difference , nfi* * Difference required for significance at the 5 per cent point. bulb and yield per acre, than those of other treatments; and, as before, the seedlings with both tops and roots pruned made the poorest showing (table 16). In 1942 pruning studies were made at Davis with two varieties, San Joaquin and Crystal Grano (table 17). For San Joaquin there was no significant dif- ference in the mean weight of bulbs from the different treatments nor in the yield per acre, though there was a significant difference in the number of bulbs harvested ]>er plot. With Crystal Grano tliere was no significant difference Bul. 682] Experiments with the Transplant Onion Crop 15 between the number harvested nor the mean weight. The yield per acre from transplants with both tops and roots pruned was significantly lower, however, than the yield from any other treatment. TIME-OF-HARVEST STUDIES Methods. — To determine how time of harvest affects yield, two blocks were selected in a field of California Early Red onions at Davis. Block 1 was divided TABLE 16 Effect of Pruning Onion Transplants on Stand, Mean Weight per Bulb, and Yield per Acre, in the Variety Eed 21, at Davis, 1933-34 Treatment Bulbs harvested per plot Mean weight per bulb, pounds Yield per acre (100-pound bags) 71 67 G7 67 48 44 44 43 361 314 314 307 Difference required for significance . . No significant difference 02* 03t 31* 41* Difference required for significance at the 5 per cent point. i required for significance at the 1 per f Difference requi cent point, into three sub-blocks, each having nine rows or plots. The first harvest was made on June 3, and succeeding harvests came at intervals of 4 or 5 days until July 12. Immediately after each harvest the plants were dried in shallow trays. Weighing was done 10 days after harvest. Block 2 was divided into five sub-blocks, each having six rows or plots of onions. The first harvest was made on June 17, and the last on July 12. TABLE 17 Effect of Pruning Onion Transplants on Stand, Mean Weight per Bulb, and Yield per Acre, in the Varieties San Joaquin and Crystal Grano, at Davis, 1942 Treatment San Joaquin Bulbs harvested per plot Mean weight per bulb, pounds Yield per acre (100-pound bags) Crystal Grano Bulbs harvested per plot Mean weight per bulb, pounds Yield per acre (100-pound bags) Unpruned Roots pruned Tops pruned Roots and tops pruned 0.47 45 48 48 320 273 273 262 0.34 0.34 0.35 0.31 147 151 127 105 Results. — The object of the time-of -harvest experiments was to determine about how much the yield was reduced when the plants were pulled at differ- ent times before maturity. As is shown by data from block 1 (table 18), there was a gradual increase in weight of bulb from June 3 to July 12, and a con- siderable increase in yield per acre even after most of the tops were down. The weight of the bulb was affected somewhat by the climatic conditions during 16 University of California — Experiment Station TABLE 18 Effect of Harvest Date on Mean Weight per Bulb and Yield per Acre in California Early Eed Onions, Block 1, at Davis, 1927 Date harvested Bulbs harvested per plot Tops down, per cent Mean weight per bulb, pounds — ■ 1 Yield per acre (100-pound bags) 133 133 133 130 133 125 130 131 136 12.3 26.5 45 6 76.9 87.3 94.6 100.0 100.0 100.0 37 .42 .42 .47 .48 .45 .51 .51 0.52 296 June 8 335 343 June 17 354 June 22 388 June 27 342 July 2 401 July 7 404 July 12 430 Difference required for significance. . . No signifi- cant difference 9.3* 12. 8f 0.04* 0.06t 34* 47t * Difference required for significance at the 5 per cent point. t Difference required for significance at the 1 per cent point. TABLE 19 Effect of Harvest Date on Mean Weight per Bulb and Yield per Acre in California Eary Red Onions, Block 2, at Davis, 1927 Date harvested Bulbs harvested per plot Tops down per cent Mean weight per bulb, pounds Yield per acre (100-pound bags) June 17 June 22 June 27 July 2 July 7 July 12 134 127 131 140 137 140 28.0 36.1 78.1 79.3 94 9 92.8 33 .35 .38 .38 39 0.39 270 268 303 325 324 335 Difference required for significance . No signifi- cant difference 8.2* 11. 2f No signifi- cant difference No signifi- cant difference * Difference required for significance at the 5 per cent point. t Difference required for significance at the 1 per cent point. the 10-day curing period before weighing;. Lots exposed to strong desiccating winds lost more weight than lots not so exposed. Probably a better method would have been to remove the tops and roots and weigh the bulb at time of harvest. The data from block 2 (table 19) show the same trend. There was, however, no significant difference between harvest dates in mean weight of bulb or in yield per acre. Bul. 682] Experiments with the Transplant Onion Crop 17 Fig. 1. — Plants transplanted to the field at different dates ; dug and photographed on March 2, 1928, to show the extent of root and top development. From left to right, these groups of plants were set out, respectively, January 13, 20, and 27 and Febru- ary 6, 10, and 17, 1928. Fig. 2. — Stockton Yellow Globe seedlings for transplanting, grown on raised beds in peat land in the Delta region of California. Seed was drilled thickly in rows 3 to 4 inches apart. (From Cir. 357.) 18 University of California — Experiment Station Fig. 3. — California Early Red and Stockton Yellow Globe seedlings for transplanting; grown in raised beds (two rows to the bed) near Sacramento. I i %**$ Hl ■:■ A - . IP v^^f^c^w .;0- J %£' : *l"n:.r. ::::;>: ;<»-, "-Mi ■? % : " , : /' .■:■/''." V Slip::,. ■ :, ■ ; ■ ■ ■ . ■ ■ ■ ■■■"■■■-"-■ :'";■■■ ;: ", ■■■■'. Y\g. 4. — California Early Red seedlings for transplanting; grown on raised beds (four rows to the bed) near Sacramento. The seedlings in the two outside rows of each bed were the largest and were pulled first. Bul. 682] Experiments with the Transplant Onion Crop 19 SUMMARY In spacing tests during seven years, as the spacing distance between plants in the row was increased beyond 2 or 3 inches there was an increase in the size of bulb and a decrease in yield per acre. For the varieties tested, the 3-inch spacing appeared best. The plants matured well, and there was no crowding in the row. The different spacing had no influence on the percentage of bolters for the one year (1935) in which records were taken. Time-of -transplanting studies throughout six years show definitely that early planting produces larger bulbs and heavier yields per acre. Plants set early form a vigorous root system during the winter and therefore develop more rapidly with the advent of warm weather. Early transplanting of bolting- varieties, however, greatly increases the percentage of bolters. In every comparison between large and small transplants made at Davis during five years, the differences in favor of large transplants for size of bulb were either significant or highly significant, and the differences in yield per acre were all highly significant. In a one-year test at Milpitas the large seed- lings produced the largest bulbs ; but not all differences were significant, and no significant differences occurred in yield per acre. A gradual increase in the percentage of bolters (0 to 100 per cent) occurred in the variety Babosa as the size of transplant was increased from under % inch to % or 1 i ncn in diameter. In pruning tests for four years, a considerable reduction in yield often occurred when transplants had both roots and tops trimmed, as is usually done in commercial practice. In two of the four years the differences in yield between unpruned transplants and those with roots pruned, tops pruned, or both roots and tops pruned were highly significant. In two of the years there was very little difference between the unpruned transplants and those with either tops or roots pruned. In all four years, except with the variety San Joaquin grown at Shaf ter in 1942, the difference between the unpruned trans- plants and those with both tops and roots pruned was either significant or highly significant. According to time-of -harvest studies, onions continue to increase in size and in yield per acre for some time after most of the tops break over. 20 University of California — Experiment Station LITERATURE CITED 1. Hawthorn, L. R. 1938. Cultural experiments with Yellow Bermuda onions under irrigation. Texas Agr. Exp. Sta. Bui. 561:1-30. 2. Jones, H. A. 1929. Spacing, time of planting, and size of seedling studies with California Early Bed onions. Amer. Soc. Hort. Sci. Proc. 26:114-18. 3. Kraus, J. E. 1942. Effects of partial defoliation at transplanting time on subsequent growth and yield of lettuce, cauliflower, celery, peppers, and onions. U. S. Dept. Agr. Tech. Bui. 829:1-35. 4. Magruder, Roy, et ah 1041. Description of types of principal American varieties of onions. U. S. Dept. Agr. Misc. Pub. 435:1-85. 5. Starring, C. C. 1935. Comparison of some methods of growing onions. Montana Agr. Exp. Sta. Bui. 305:1-11. 6. Wilson, A. L. 1934. Influence of spacing on the formation of "double onions." Utah Acad. Sci. Arts and Letters 11:177-84. 12m-2, '44(8277)