Timing of Easter lily (Lilium longiflorum Thunb.) for sales is complex because the date of Easter and the number of leaves formed on plants before flower bud initiation vary from year to year. A process control chart was developed that uses a leaf unfolding rate model of Easter lily to control development rate towards flowering. The technique allows observed and target leaf count to be tracked on a graph and compared visually over time. The optimum leaf unfolding rate and average temperature can be read directly from the chart without the need for mathematical calculation. The approach provides an intuitive method for transferring quantitative models to growers and can be applied to other management problem areas.
P.R. Fisher and R.D. Heins
P.R. Fisher and R.D. Heins
A graphical control chart was developed to monitor leaf count of Easter lily (Lilium longiflorum Thunb.) and make temperature recommendations based on predictions of a leaf unfolding rate (LUR) model. The graph allows observed and target leaf count to be compared visually over time. Timing of the visible bud stage, when flower buds are visible externally on the plant, is important to time flowering for the Easter sales period. The optimum LUR and average daily temperature required to achieve a target visible bud date can be read directly from the chart. The approach provides an intuitive method for transferring quantitative models to growers.
Dustin P. Meador and Paul R. Fisher
The objective was to quantify the effect of water-soluble fertilizers on concentration of free chlorine level in a sodium hypochlorite solution. Research on the disinfestation strength and phytotoxicity risk of chlorine compounds is needed, because control of waterborne pathogens has been based on response to free chlorine, whereas dual injection of fertilizer and chlorine is a common horticultural practice. Free chlorine from sodium hypochlorite was applied at 2.6 mg·L−1 chlorine (Cl) to deionized water only (control) or deionized water with 11 nutrient solutions at 200 mg·L−1 nitrogen (N). Nutrient solutions included reagent-grade ammonium sulfate (NH4)2SO4, ammonium nitrate (NH4NO3), potassium nitrate (KNO3), and urea salts and seven commercial blended N–P–K water-soluble fertilizers that contained both macro- and micronutrients. Commercial fertilizers contained ammonium-N at 0% to 50% of total-N, urea-N at 0% to 14% of total-N, and nitrate-N at 50% to 93% of total-N. Free Cl (mg·L−1), total Cl (mg·L−1), and oxidation-reduction potential (ORP, in mV) were measured 2 min and 60 min after Cl was applied. Combined Cl was calculated as the difference between the total and free Cl measurements. All solutions were maintained at pH 6 and 25 °C. In the control solution, free Cl was 2.6 mg·L−1 after 2 minutes and decreased to 2.2 mg·L−1 after 60 minutes. The ammonium-containing solutions (NH4)2SO4 and NH4NO3 resulted in free Cl below 0.1 mg·L−1 after 2 minutes. Urea reacted more slowly than ammonium salts, whereby free Cl decreased to 2.3 mg·L−1 after 2 minutes and 0.4 mg·L−1 after 60 minutes. In contrast, KNO3 had less impact on free Cl with 2.4 mg·L−1 free Cl available at both 2 minutes and 60 minutes. With all commercial fertilizers tested, free Cl decreased after 2 minutes to below 0.1 mg·L−1. Total Cl remained above 2 mg·L−1 after 60 minutes in all treatments, indicating that the majority of Cl was in a combined form for ammonium and urea salts and commercial fertilizers. The ORP of commercial fertilizer blends and ammonium-containing salts was lower than 600 mV, whereas deionized water, KNO3, and urea treatments had ORP levels above 650 mV. Nutrient solutions containing ammonium or urea required 20 mg·L−1 or more of applied Cl to provide residual free Cl above 2 mg·L−1 at 2 minutes.
A.J. Bishko and P.R. Fisher
Our objective was to systematically quantify the dose response from applications of several basic materials recommended for raising pH in acidic media. A peat (70%)/perlite (30%) medium was mixed with a pre-plant nutrient charge, a wetting agent, and 0, 0.3, 0.6, 0.9, 1.2, or 1.5 kg dolomitic hydrated lime/m3, resulting in a range in initial pH from 3.4 to 6.4. Five rates of flowable dolomitic limestone, five rates of potassium bicarbonate, two rates of potassium hydroxide, a supernatant of calcium hydroxide and a distilled water control were applied as single drenches. The medium was irrigated with distilled water when it dried to 50% container capacity as determined by weight. Media pH and EC of four replicates were tested at 1 day and 1, 2, 3, and 4 weeks after application as a saturated media extract. Flowable limestone and potassium bicarbonate both significantly raised medium pH by up to 2 units compared with the control, depending on concentration. As initial medium pH increased, the effect of the basic chemicals on medium pH decreased. For example, flowable lime applied at 0.5 L·100 L–1 of distilled water increased pH by 2 units at an initial medium pH of 3.4 and by 0.4 units at an initial pH of 6.4. Potassium hydroxide and calcium hydroxide drenches did not significantly raise pH. Potassium bicarbonate was easier to apply than the suspension of flowable limestone, however both chemicals provide practical methods for raising pH of soilless media.
J.R. Fisher and S.G.P. Nameth
Cucumber mosaic virus (CMV) was isolated from the perennial ornamental mint, Ajuga reptans L. `Royalty', using melon aphids (Aphis gossypii Glover). The isolate and its associated satellite RNA (satRNA) were biologically and chemically characterized. The satRNA was cloned and sequenced and is 338 nucleotides long and does not induce lethal necrosis on `Rutgers' tomato (Lycopersicon esculentum Mill.) or severe chlorosis on tobacco (Nicotiana L. spp.). The virus is ≈28 to 30 nm in diameter and reacts to CMV serological subgroup I antibodies. The virus is able to infect `Black Beauty' squash (Cucurbita pepo L.), cucumber (Cucumis sativus L.), and `Howden' pumpkin (Cucurbita pepo) but is not able to infect green bean (Phaseolus vulgaris L.) or cowpea [Vigna unguiculata (L.) Walp. ssp. unguiculata]. The virus is able to efficiently replicate its satRNA in tobacco and `Black Beauty' squash but replication is less efficient in cucumber, based on accumulation of double-stranded satRNA.
J.H. Lieth, P.R. Fisher, and R.D. Heins
A growth function was developed for describing the progression of shoot elongation over time. While existing functions, such as the logistic function or Richards function, can be fitted to most sigmoid data, we observed situations where distinct lag, linear, and saturation phases were observed but not well represented by these traditional functions. A function was developed that explicitly models three phases of growth as a curvilinear (exponential) phase, followed by a linear phase, and terminating in a saturation phase. This function was found to be as flexible as the Richards function and can be used for virtually any sigmoid data. The model behavior was an improvement over the Richards function in cases where distinct transitions between the three growth phases are evident. The model also lends itself well to simulation of growth using the differential equation approximation for the function.
P.R. Fisher, J.H. Lieth, and R.D. Heins
Stem elongation of commercially produced flowering poinsettia (Euphorbia pulcherrima L.) is often sigmoid. However, sigmoid mathematical functions traditionally used for representing plant growth fail to adequately describe poinsettia stem elongation when a shoot has a long vegetative growth period. A model was developed that explicitly described three phases of poinsettia stem elongation: 1) the initial lag phase, where stem length increases approximately exponentially; 2) a period when elongation is linear; and 3) a plateau phase, where elongation rate declines to zero and stem length reaches an asymptotic maximum length. The timing of the plateau phase was linked to flower initiation date. Fit of the resulting model to data from single stem `Freedom' poinsettia grown with different periods between transplant and flower initiation had an R2 of 0.99. Model parameters had clear biological meaning, and the poinsettia model has horticultural application for simulation and graphical tracking of crop height.
P.R. Fisher, J.H. Lieth, and R.D. Heins
A model was developed to quantify the response of Easter lily (`Nellie White') flower bud elongation to average air temperature. Plants were grown in greenhouses set at 15, 18, 21, 24, or 27C after they had reached the visible bud stage. An exponential model fit the data with an R 2 of 0.996. The number of days until open flowering could be predicted using the model because buds consistently opened when they were 16 cm long. The model was validated against data sets of plants grown under constant and varying greenhouse temperatures at three locations, and it was more accurate and mathematically simpler than a previous bud elongation model. Bud length can be used by lily growers to predict the average temperature required to achieve a target flowering date, or the flowering date at a given average temperature. The model can be implemented in a computer decision-support system or in a tool termed a bud development meter.
P.R. Fisher, J.H. Lieth, and R.D. Heins
The objective was to predict the distribution (mean and variance) of flower opening for an Easter lily (Lilium longiflorum Thunb.) population based on the variability in an earlier phenological stage and the expected average temperature from that state until flowering. The thermal time from the visible bud stage until anthesis was calculated using published data. `Nellie White' grade 8/9 Easter lilies were grown in five research and commercial greenhouse locations during 1995, 1996, and 1997 under a variety of temperature and bulb-cooling regimes. Distributions of visible bud and anthesis were normally distributed for a population growing in a greenhouse with spatially homogenous temperatures. The variance at anthesis was positively correlated with variance at visible bud. The mean and variance at visible bud could therefore be used to predict the distribution of the occurrence of anthesis in the crop. The relationship between bud elongation, harvest, and temperature was also incorporated into the model. After visible bud, flower bud length measurements from a random sample of plants could be used to predict the harvest distribution. A computer decision-support system was developed to package the model for grower use.
P.R. Fisher, R.D. Heins, and J.H. Lieth
Stem elongation response to a single foliar application of the growth retardant chlormequat chloride [(2-chloroethyl) trimethylammonium chloride] for poinsettia (Euphorbia pulcherrima Klotz.) was quantified. Growth retardant applications did not affect final leaf count or timing of visible bud, first bract color, or anthesis. There was a statistically significant effect of growth retardant concentration on stem elongation, with a range from 289 ± 15 mm (mean 95% confidence intervals) for the control plants to 236 ± 17 mm at 4000 ppm. The growth-retarding effect during the first day after the application was not significantly different between 500 and 4000 ppm, and concentration primarily affected the duration of growth-retarding activity. A dose response function was incorporated into a three-phase mathematical function of stem elongation of single-stem poinsettia to predict elongation of treated and untreated plants. The model was calibrated using a data set from plants receiving 0, 500, 1000, 1500, 2000, 3000, and 4000 ppm, with a resulting R 2 of 0.99. Validation of the dose response model against an independent data set resulted in an r 2 of 0.99, and predicted final stem length was within 12 mm of observed final length.