Introduction
Broccoli is not often transplanted in the Willamette Valley, but transplanting offers advantages in multiple cropping and establishing an early stand. The cost of using plug-grown transplants can be reduced by using the smallest plug capable of producing a quality transplant, thus reducing greenhouse bench space, number of trays, and amount of rooting medium needed to produce the crop. In addition, costs might be reduced further by growing multiple seedlings per plug and reducing the number of plugs needed to transplant a given area.
The purpose of these trials was to examine the feasibility of using very small plugs in combination with multiple seedlings per plug to produce broccoli transplants and to follow their performance in the field.
Methods
On 31 March, 1989, 'Gem' broccoli was seeded into either 5.1 or 31 cm3 plugs (Landmark Plastic Corp. 288- and 128-cell trays, respectively) containing a peat-vermiculite medium. Either 2, 3, or 5 seeds were placed in each cell and the seedlings were thinned to a final population of 1, 2, or 3 per cell two weeks later. The seedlings were watered daily as needed and fertilized weekly with a 10N-13P-16.7K soluble fertilizer at 100 ppm N.
The plugs were transplanted on 1 May to a Willamette silt loam, to which had been applied 500 pounds per acre of a 10N-8.7P-16.7K fertilizer, 2.0 pounds B, 0.75 pounds trifluralin, and 1.3 pounds chlorpyrifos per acre. Propachlor herbicide was applied at 4.0 pounds per acre immediately after planting. An additional 75 pounds of nitrogen per acre was applied as ammonium nitrate on 29 May and again on 13 June.
Treatments consisted of a factorial combination of the two cell sizes and three populations per cell. Spacing between plugs in the row was 18 inches for all plots. Plots consisted of 15 feet of a four-row bed with 16 inches between rows. Plots were harvested on 28 June, and again on 7 July. Only main shoots were harvested from the center two rows of each plot.
Methods in 1990 were similar except as follows. Seeding was on 27 March and the plugs were again transplanted on 1 May. Preplant fertilizer application included 200 pounds of triple superphosphate, 50 pounds of ammonium nitrate, and 2.0 pounds of B per acre. An additional 150 pounds of N per acre was applied as ammoniun nitrate on 17 May.
Spacing between plugs in the row was 9, 18, or 27 inches for one, two, and three seedlings per plug, respectively. Plots were harvested on 3 July, and again on 10 July.
Results
The interaction of plug size and number of seedlings per plug did not significantly affect any yield component in 1989 and only main effects are given in Table 1. The larger plug size increased mean head weight and bead size, indicating that the relative crowding in the smaller plugs either delayed maturity or resulted in smaller plants. The highest population of seedlings per cell (3) also delayed maturity, reflected in the smaller number of heads cut at the first harvest, the reduced head size, and the tendency toward smaller bead size.
For the total of two harvests, plug size did not significantly affect the number of heads harvested or total yield. However, the mean head weight and bead size were slightly smaller with the smaller plug size. With greater number of seedlings per plug, the total number of heads harvested increased, but the fraction of plants resulting in marketable heads (number of heads harvested divided by the target plant population) decreased markedly with increasing number of plants per plug. Mean head weight decreased dramatically with increasing number of seedlings per plug, indicating either severe competition among plants originating from the same plug or delayed maturity. Total yield did not vary significantly with seedlings per plug as the effect of increasing numbers of plants was almost exactly offset by the decline in mean head weight.
Target plant population per acre ranged from 21,780 for 1 seedling per plug to 65,340 for 3 seedlings per plug. At the higher populations, it is apparent that the competition among seedlings in a plug severely reduced head weight. Thus, no yield advantage was gained by using multiple seedlings and no cost benefits would have been realized. Some of this crowding effect could be offset by using a larger plug size, but only at the cost of increased bench space, trays, and medium needed to produce the plugs.
Table 1. Main effects of plug size and number of seedlings per plug on broccoli yield, 1989, NWREC, Oregon First harvest Total No. of Yield Mean head Bead No. of % of Yield Mean head Bead headsz T/A wt. (g) sizey heads maximumx T/A wt. (g) size ______________________________________________________________________________ Plug size 31 cm3 10.5 2.7 175 2.7 21.3 48.5 4.7 168 2.8 5.1 cm3 9.3 2.2 152 1.8 22.3 50.7 4.3 147 2.4 NSw NS ** ** NS NS NS * * No./plug 1 11.1 3.7 233 2.5 12.9 58.5 4.1 223 2.7 2 12.3 2.6 147 2.4 23.2 52.8 4.7 140 2.8 3 6.4 1.1 111 1.9 29.4 44.5 4.6 110 2.4 ** ** ** NS ** ** NS ** NS zNumber per plot, 44 plugs/plot. yFive point scale with 1 = very tight and fine, 5 = loose, open, near anthesis. x100% x the total number of heads cut divided by the number of seedlings originally planted. wNS,*,**: no significant differences, differences significant at the 5% and 1% levels, respectively.
In 1990, the trial was repeated with single versus multiple seedlings per plug, but with spacing in the row varied to maintain the same target plant population per acre, assuming survival of all seedlings.
Again in 1990, the interaction of plug size and number of seedlings per plug did not usually significantly affect yield and only main effects are given for the first harvest in Table 2. The few significant interactions are found in Table 3. At the first harvest, the larger plug size increased the number of heads cut and the mean head weight and width, indicating that the relative crowding in the smaller plugs either delayed maturity or resulted in smaller plants.
Increasing the number of seedlings per cell also delayed maturity, reflected in the smaller number of heads cut at the first harvest, the reduced head weight, and the tendency toward smaller bead size and reduced head width. There was a significant interaction of cell size and number per cell affecting mean head weight and width (Table 3). The combination of small cell size and three seedlings per cell resulted in greatly reduced head weight and width.
For the total of two harvests (Table 4), the larger plug size significantly increased total yield, but not the number of heads harvested or mean head weight. However, both number of heads harvested and head weight tended to be greater with the larger cells. Head width was significantly greater with the larger cells. Mean bead size and incidence of hollow stem (data not shown) were unaffected by cell size. There were no significant interactions affecting any component of yield for the sum of the two harvests.
With greater number of seedlings per plug, the total number of heads harvested decreased. This is in contrast to 1989 when the number of heads harvested increased with greater number of seedlings per cell. Plugs were set at a uniform 18-inch spacing in 1989, resulting in a greater target plant population at greater number of seedlings per plug. In 1990 the yield per plot also decreased with increasing number of seedlings per cell, due in part to the decrease in number of heads harvested, but also due to decreased mean head weight. Mean head width and bead size were not significantly affected by number of seedlings per plug.
The reduction in number of heads harvested per plot at greater number of seedlings per plot indicates that fewer seedlings survived to produce a head or that fewer heads matured in time to be cut in two harvests. The approximate halving of yield in 1990, as the number of seedlings per cell went from one to three, indicates that multiple seedlings per cell causes too much crowding for rapid, uniform seedling growth and survival. These results argue against the practice of using multiple-seedling plugs for transplanting broccoli.
Table 2. Main effects of plug size and number of seedlings per plug on broccoli yield, first harvest, 1990, NWREC, Oregon No. of Yield Mean head Mean head Bead headsz (T/A) wt. (g) width (cm) sizey ___________________________________________________________________ Plug size 31 cm3 8.6 1.5 151 8.5 2.6 5.1 cm3 4.9 0.8 126 7.0 2.5 **x ** * ** NS No./plug 1 12.8 2.2 233 8.4 2.8 2 6.3 1.0 147 8.0 2.7 3 1.3 0.2 111 6.8 2.0 **lin **lin **quad NS NS zNumber per plot. yFive point scale with 1 = very tight and fine, 5 = loose, open, near anthesis. xNS,*,**: no significant differences, differences significant at the 5% and 1% levels, respectively. Lin, quad: linear and quadratic regression, respectively. Table 3. Interaction of plug size and number of seedlings per plug on broccoli mean head weight and width, first harvest, 1990, NWREC, Oregon Plug size No./plug Mean head wt. (g) Mean head width (cm) 31 cm3 1 162 8.7 2 144 8.2 3 145 8.6 5.1 cm3 1 154 8.0 2 149 7.9 3 75 5.1 LSD (0.05) 25 1.3 Table 4. Main effects of plug size and number of seedlings per plug on broccoli yield, total of two harvests, 1990, NWREC, Oregon No. of Yield Mean head Mean head Bead heads (T/A) wt. (g) width (cm) size Plug size 31 cm3 23 3.7 148 9.2 2.9 5.1 cm3 20 3.1 141 8.6 2.9 NS ** NS * NS No./plug 1 25 4.5 159 9.1 3.1 2 22 3.5 144 8.8 2.9 3 17 2.4 130 8.8 2.8 **lin **lin **lin NS NS