HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J. Search for Related Content PubMed Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J. Agricola Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J.

SEED PHYSIOLOGY, PRODUCTION & TECHNOLOGY

Birdsfoot Trefoil Seed Production

III. Seed Shatter and Optimal Harvest Time

^a Instituto Nacional de Investigaciones Forestales y Agropecuarias, CE Pabellon, Carretera Ags.-Zac. km. 32.5, Apdo. Postal 20, Pabellon de Arteaga, Ags. C.P. 20660, Mexico
^b National Forage Seed Production Research Center, USDA-ARS, 3450 SW Campus Way, Corvallis, OR 97331 USA

steinerj{at}ucs.orst.edu

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Seed shattering is a major problem in birdsfoot trefoil (Lotus corniculatus L.) seed production and limited information is available describing the effects of agronomic practices on shatter losses. The objectives of this research were to: (i) quantify the effects of soil-water availability on seed shatter and (ii) determine optimal harvest time on the basis of a heat unit method to minimize birdsfoot trefoil seed losses under western Oregon climatic conditions. Six treatments varying in water depletion percentage and replenishment amount were applied in 1994 and 1995 and two treatments in 1996 on a Woodburn silt loam soil (fine-silty, mixed, mesic Aquultic Argixeroll) near Corvallis, OR. The total amount of shattered seeds was correlated with total harvested seed yield (r = 0.93). Crop water stress index (CWSI) was inversely related to the percentage of seeds shattered (r = -0.76). Increasing amounts of applied water increased the potential of seed yield shattered (r = 0.65). Seed shatter losses fluctuated during the late-reproductive period, but were not influenced by irrigation or fluctuating climatic conditions. A total of 109 heat units (approximately 11 d), which were determined on the basis of a 10°C base temperature, were accumulated from the time of initial pod dehiscence until rapid seed shattering. The average seed yield losses due to shattering was 3 to 5.3 kg ha^- d^-1. The non-irrigated control treatment generally produced more seeds than irrigated treatments. It is, thus, best not to irrigate birdsfoot trefoil grown for seed in western Oregon because increasing amounts of irrigation water increased seed shattering.

Abbreviations: CWSI, crop water stress index • FAWU, fraction of available water used • FC, field capacity • HU, heat units • LS, low-stress treatment • C, non-irrigated control treatment • RH, relative humidity • T, average daily temperature • VPD, vapor pressure deficit

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
BIRDSFOOT TREFOIL is a perennial, non-bloating legume used for forage and hay production in the northeastern and midwestern USA as well as eastern Canada. Fairey (1994) identified some of the most important factors limiting yield in birdsfoot trefoil seed production. Seed shattering is a major problem in birdsfoot trefoil seed production (Seaney and Henson, 1970; Li and Hill; 1989).

As maturity advances in this indeterminate species, the seed yield potential increases because the total number of pods produced increases. The percentage of pods that dehisce and the amount of seed shattering increases as later-developing pods mature (MacDonald, 1946; Anderson, 1955). The anatomy, development, and maturation of birdsfoot trefoil pods have been extensively studied. The anatomical configuration of pod tissues is related to the dehiscence mechanism. Dehiscence occurs along the ventral and dorsal sutures of the carpel margins and along the median vein of the pod (Esau, 1960). Pod dehiscence is caused by different rates of moisture loss from these tissues.

Relative humidity has been reported to be the most critical factor influencing pod dehiscence and seed shattering (Anderson, 1955; Metcalfe et al., 1957). Mature pods shattered freely when relative humidity was below 40% (Anderson, 1955). At low temperatures and high relative humidity, the rate of shattering was less than at high temperatures and lower relative humidity (Metcalfe et al., 1957; McGraw and Beuselinck, 1983). Pod moisture content was influenced by ambient relative humidity, which was a critical factor that determines when pods will dehisce. The critical pod moisture concentration for shattering was between 101 and 104 g kg^-1 (Metcalfe et al., 1957). Under sunny conditions, pod temperature can be 5°C higher than the ambient air temperature, resulting in a change in the relative humidity at the surface of the pod (Metcalfe et al., 1957). Gershon (1961) found no correlation between relative humidity and birdsfoot trefoil pod dehiscence for plants grown under greenhouse conditions, but relative humidity and pod dehiscence were correlated when grown in the field. This was likely due to differences between the humidity in the field and that in the greenhouse (Grant, 1996).

The rate of shattering increases as the rate of water loss from the pods increases, but pods do not shatter as readily when pod drying proceeds slowly (Buckovic, 1952). Although dependent upon environmental conditions, when desiccants and plant growth regulators were used to manage vegetative growth, seed shattering was reduced and seed yield increased (Wiggans et al., 1956). Summations of average daily temperatures have been used to determine when pod dehiscence and shattering will occur (Gataric et al., 1990). These findings suggest that factors other than relative humidity and pod moisture content alone may modify the pod shattering response.

The effects of agronomic practices on seed shattering are not as well defined as the physical factors that trigger pod dehiscence. Grant (1996) suggested that earlier harvest was only partially effective for reducing seed losses due to shattering because fewer mature pods were present in the indeterminate flowering crop. Estimates for proper harvest time have been based on the rate of appearance of pods and pod color (Anderson, 1955; Winch and MacDonald, 1961; Hare and Lucas, 1984; Winch et al., 1985; Pieroni and Laverack, 1994) and the rate of development of reproductive structures (Li and Hill, 1989). Pod color can vary from dark green or dark green-purple to green-white and then to golden-brown (MacDonald, 1946; Anderson, 1955; Winch and MacDonald, 1961). Maximal seed yield was obtained with harvest at the time pod color changed to golden-brown, pod moisture content decreased from 650 to 250 g kg^-1, and initial seed shattering had begun. Optimal harvest time was suggested to be when 70 to 78% of the pods picked at random throughout the field were mature (Winch et al., 1985). A delay in the seed harvest time to allow more developing umbels to mature can result in yield decreases of 50% (Winch and MacDonald, 1961) to 67% (Anderson, 1955).

The objectives of this research were to: (i) quantify the effects of soil-water availability on seed shattering, and (ii) determine optimal harvest time on the basis of a heat unit method to minimize birdsfoot trefoil seed losses due to shattering before the time of seed harvest under climatic conditions of western Oregon. This is the third and final paper that describes the effects of water management on the reproductive development and seed production of birdsfoot trefoil.

	Materials and methods

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
The experiment was conducted in 1994, 1995, and 1996 at the Oregon State University, Crop Science Department, Hyslop Field Laboratory near Corvallis, OR, on a Woodburn silt loam soil (fine-silty, mixed, mesic Aquultic Argixeroll). Details of the agronomic practices used to grow the seed crop are presented in Garcia-Diaz and Steiner (1999, 2000). The plots were arranged in a randomized complete block design with four replications and six treatments in 1994 and 1995. Each plot was 4.5 m wide by 10 m long and had a surface trickle irrigation system to deliver water through 3.5 L h ^-1 in-line, turbulent-flow emitters spaced 0.9 m apart in five plastic drip lines 60 cm apart and perpendicular to the planting rows. A distribution manifold consisting of a mesh filter, ball valve, residential water flowmeter, volumetric controller, and a pressure regulator allowed water to be applied to all four replications of each treatment at the same time. The soil-water status of each plot was monitored weekly until harvest by neutron attenuation (Cuenca, 1988) to determine when to apply the irrigation treatments.

In 1994 and 1995, five supplemental irrigation treatments were applied between the period from the time of herbage clip-back to seed harvest. The treatments were (i) low-stress (LS), in which the soil-water content was maintained close to 100% field capacity (FC) by two or to three water application replacements per week during the period from early-June until 3 wk before seed harvest; (ii) single water replacement to 50% FC when soil-water depletion was 30% (D30–F50); (iii) single water replacement to 100% FC when soil-water depletion was 30% (D30–F100); (iv) single water replacement to 50% FC when soil-water depletion was 60% (D60–F50); and (v) single water replacement to 100% FC when soil-water depletion was 60% (D60–F100). A sixth treatment was a non-irrigated control (C). In 1996, only treatments LS and C were evaluated.

Flower and pod density were estimated weekly in six 0.1-m² random samples per plot from the time of first bud appearance until the end of the season (extending past the time of seed harvest). The pods were classified as (i) immature pods, pod color ranges from dark green to light green; (ii) mature pods, pod color was green-white to dark brown; and (iii) dehisced pods, pods begin to twist and dehisce and the seeds shatter (Winch et al., 1985). Peak flowering was defined as the time when the maximum flower number was observed.

Seed loss due to shattering was determined every Monday, Wednesday, and Friday by collecting shattered seeds from two pans (60 by 40 by 10 cm) that were placed 10 cm above the soil surface between two planting rows and below the crop canopy of each plot. Elevation of the pans allowed air to circulate within the canopy and avoided condensation under the pan. The shattered seeds were collected from the pans with a portable vacuum and then cleaned and weighed. Shattered seed loss at harvest time was the sum of shattered seeds collected from the time of initial shatter to the time of seed harvest. Total shattered seeds was the sum of all seeds collected to the end of the season. The percentage of total seed shatter loss lost by harvest time was calculated by dividing shattered seed loss at harvest time by total shattered seeds and multiplying the results by 100.

Harvest occurred when most pods were light tan to brown and initial seed shattering had begun (Garcia-Diaz and Steiner, 1999). At this stage of pod development, maximal harvested seed yield was obtained. Two subplots, 1 by 4 m, were harvested from both ends of each plot by a gas-powered sickle mower used in the early morning to avoid seed shatter losses. The plant material was collected by hand, bagged, and dried at 32°C for 1 d. Harvested seed yield was the amount of non-shattered seeds at harvest time. Total seed yield was the sum of the harvested seed yield plus the sum of shattered seed losses to harvest time (Garcia-Diaz and Steiner, 2000).

On the basis of preliminary examinations climatic data and literature (Anderson, 1955; Metcalfe et al., 1957; McGraw and Beuselinck, 1983), average daily temperature (T), relative humidity (RH), and vapor pressure deficit (VPD) were chosen as variables to investigate as predictors of pod dehiscence and seed shattering. Daily high and low temperature (T_max and T_min, respectively) and RH were obtained from an automated meteorological station located 400 m from the trial site. The VPD was based on the UN-FAO modified Penman method as used in the calculations for the reference evapotranspiration (ET_r) (Cuenca, 1989). The VPD equation was:

(1)

where E_s = saturation of water pressure (mPa); and E_a = actual water pressure (mPa), and:

(2)

where T_mean (°C) was determined on a daily basis, and:

(3)

and:

(4)

where RH = daily mean relative humidity (%).

Daily degree day heat units (HU) were calculated by Eq. [3] minus a base temperature of 10°C:

(5)

On the basis of the data collected in 1996, the number of accumulated heat units from the time of peak flowering to initial pod dehiscence (HU_SSi) was determined by the equation:

(6)

Equation [6] was then used to estimate time of initial pod dehiscence in 1994 and 1995. To determine whether the estimation of seed harvest time, which was based on accumulated heat units, could be improved, a modification of Eq. [5] comprised of two functional components was tested: (i) accumulation of HU (Eq. [5]) from the time of peak flowering to the time of initial seed shatter (initial seed shatter) plus (ii) the accumulation of HU modified by a VPD threshold from the time of initial seed shatter to harvest time:

(7)

where HU_pfh = accumulated heat units from peak flowering to harvest time, HU_pf = sum of heat units from the time of peak flowering (pf) to the time of initial seed shatter (s_In), and HU_h = sum of heat units from the time of initial seed shatter to the time of seed harvest (h). The efficacy of VPD threshold values to modify the accumulated heat unit model was determined by threshold values less than: 0.40, 0.45, 0.50, 0.55, 0.60, and 0.65 MPa. Few or no days with VPD values below 0.40 MPa were observed at the experimental site during the reproductive period in the 3 yr of the study. The VPD was selected as the threshold indicator on the basis of the hypothesis that under conditions of high T and high RH (low VPD), the process of pod dehiscence does not advance.

Analysis of variance was used to determine differences among irrigation treatments for amount of total shattered seeds. Pearson's correlation coefficient (r) analysis was used to test functional relationship between shattered seed losses with plant water stress index and soil-water status (fraction of available water used, FAWU) (Garcia-Diaz and Steiner, 1999). A technique used to detect sequential time lag effects of temperature on seed development (Steiner and Opoku-Boateng, 1991) and emergence (Steiner and Jacobsen, 1992) was adapted to attempt to predict peak shattering events by means of HU and HU_pfh (Eq. [5] and [7], respectively). Differences between Pearson's r and Spearman's rank correlation coefficient (r_s) were used to determine whether functional relationships between variables and the rank orders among irrigation treatments were similar. Student's t pairwise comparison were used to contrast total shattered seed by irrigation treatments among the 3 yr. Differences reported were significant at P 0.05, unless otherwise was indicated.

	Results and discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Effect of Water Management on Shattering
The total amount of shattered seeds was correlated with the harvested seed yield (harvested seed yield from Garcia-Diaz and Steiner, 2000) (r = 0.93, P 0.001) and influenced by irrigation treatments (Table 1) . However, the relative rankings among irrigation treatments by harvested seed yield and total amount of shattered seeds differed (as shown by the lower Spearman rank correlation coefficient than the Pearson correlation, r_s = 0.85; P 0.001). This suggests that water replacement timing and replenishment amount not only affected harvested seed yield (Garcia-Diaz and Steiner, 2000), but that this effect was different for total amount of shattered seeds. Even though total amount of shattered seeds was correlated with CWSI and FAWU (r = 0.46; P 0.10 and r = 0.51; P 0.06, respectively), harvested seed yield was not (r = 0.33; P 0.26 and r = 0.35; P 0.22, respectively).

Season-Long Shatter Loss Distribution
The percentage of total shattered seeds lost by harvest time per sampling period fluctuated during the course of the reproductive development period (Fig. 1) . The times of peak percentage of total shattered seeds lost by harvest time among the different irrigation treatments generally coincided within years. In 1994, peaks occurred on day of year (DOY) 220, 229, and 238, but the amplitude of the fluctuations varied by water application amount. Plants grown under higher water stress conditions (greater depletion and lower replacement amounts) within similar treatment combination pairs (D30–F50, D60–F50, and C) started to shatter 2 to 4 d earlier and reached peak shatter time earlier than their lower water-stress level compliment (D30–F100, D60–F100, and LS, respectively) (Fig. 1). In 1995, the percentage of total shattered seeds lost by harvest time peaks occurred on DOY 226, 233, and 240, and on DOY 228 and 235 in 1996. Shattering time was generally earlier in the high water-stress treatment levels than with low-stress treatments in 1995 and 1996.

View larger version (30K): [in this window] [in a new window]   Fig. 1 Percentage of the total shattered seeds relative to seeds lost at harvest time of birdsfoot trefoil for six irrigation treatments in 1994 and 1995, and two treatments in 1996. Symbols indicate: D30–F50 see hard copy, D30–F100 see hard copy, D60–F50 see hard copy, D60–F100 see hard copy, LS see hard copy, C see hard copy. The six treatments were: D30–F50, single water replacement to 50% FC when soil-water depletion was 30% of available water; D30–F100, single water replacement to 100% FC when soil-water depletion was 30% available water; D60–F50, single water replacement to 50% FC when soil-water depletion was 60%; D60–F100, single water replacement to 100% FC when soil-water depletion was 60% available water; LS, low-stress, the soil-water content was maintained close to 100% field capacity (FC) during the period from early-June until 3 wk before seed harvest by two or to three water application replacements per week based on neutron attenuation soil-water content measurements; and C, a non-irrigated control. Dashed vertical lines indicate the times of peak shattering events. Vertical bars indicate the average standard error of the mean for the treatments

The number of HUs accumulated between the time from initial flowering to initial pod dehiscence was 381, 383, and 362 for plants grown under LS conditions, and 340, 383, and 362 HUs for plants grown under the non-irrigated treatment conditions, in 1994, 1995, and 1996, respectively. With the exception of Treatment C in 1994 that had earlier seed shattering (DOY 213), all treatments and years had similar initial flowering and seed shatter times. The flowering period over the 3 yr ranged from 34 to 36 d beginning DOY 181 to 183 and ending DOY 216 to 217.

Climatic conditions differed among the 3 yr of study (Fig. 2) . Seasonal fluctuations in T, RH, and VPD coincided with some of the peak shattering events, but no consistent correlations were measured between seed shattering and the individual three climatic factors (data not shown). Multiple regression analyses using the climatic variables to predict seed shatter events also did not produce consistent results among the irrigation treatments and seed production years (data not shown). In addition, utilizing both the accumulated HU (Eq. [5]) and the HU modified by a VPD threshold (Eq. [7]) methods to describe the periods from peak flowering time to the multiple peak shatter events was also not successful (data not shown).

View larger version (31K): [in this window] [in a new window]   Fig. 2 Variation in average temperature, relative humidity, and water vapor pressure deficit in 1994, 1995, and 1996. Dashed vertical lines indicate times of peak shattering events based on the results presented in Fig. 1

Harvest Time and Seed Shatter Losses
Seed harvest time differed among treatments (Garcia-Diaz and Steiner, 1999) and was dependent upon the amount of water applied (r = 0.86; P 0.001). Seed harvest time in 1994, 1995, and 1996 for the non-irrigated control was DOY 216, 220, and 227, respectively. Cumulative seed losses were related to the cumulative number of pods that dehisced, and were also affected by the year of production and irrigation treatment (data not shown). The initial rate of pod shattering was more rapid in the non-irrigated control than in the LS treatment in all 3 yr (Fig. 3) .

View larger version (33K): [in this window] [in a new window]   Fig. 3 Shattered seeds (see hard copy), shattered pods (%) (see hard copy), and seed losses before (solid vertical bars) and after (hatched vertical bars) harvest time as a percentage of actual harvested seeds from low stressed (LS) and non-irrigated (C) birdsfoot trefoil irrigation treatments in 1994, 1995, and 1996. The treatments were as described in Fig. 1. Harvest time was when the color of most pods were light-tan to brown and seed shattering had begun. Vertical bars indicate the average standard error of the mean for each seed shatter component

The percentage of dehisced pods at harvest time in the non-irrigated control was 38, 36, and 35% in 1994, 1995, and 1996, respectively (Fig. 3). For the LS treatment, harvest dates in 1994, 1995, and 1996 were DOY 238, 241, and 238, with the percentage of dehisced pods at harvest time being 36, 20, and 7%, respectively.

On the basis of the calculated number of 238 accumulated HU from the time of peak flowering until the initial pod dehiscence, a total of 109 HU were needed from the time of initial pod dehiscence until rapid shattering occurred. This was approximately 11 d with the average daily HU equaling 9.5 HU d^-1. Delayed harvest time has been shown to reduce birdsfoot trefoil seed yields (MacDonald, 1946; Anderson, 1955). Harvested seed yield was maximal when the rate of pod maturation was greater than the rate of accumulating dehiscent pods. Seed shatter losses due to pod dehiscence could not be reduced by earlier harvest because the number of harvested pods with mature seeds would be reduced. When harvested after the time of maximal harvested seed yield, seed yield will always decline because of the rate of dehiscing pods with the accompanying shattered seed losses will be greater than the rate of newly maturing pods that are ready to harvest.

The average rates of pod dehiscence in the non-irrigated control and LS treatments were 6 and 2% per day, respectively. These results are similar to previously published results for birdsfoot trefoil of 5 to 71% after 12 d (Anderson, 1955). In big trefoil (L. uliginosus Schk.), pods shattered at a rate of 10% per day and resulted in seed yield losses ranging from 7 to 88% (Hare and Lucas, 1984). Comparing the LS and non-irrigated control treatments in our experiment, the average seed yield losses per day to shattering were 3.0 and 5.3 kg ha^-1, respectively. Given the findings that (i) increasing amounts of applied irrigation water increased the percentage of the potential seed yield that would shatter by harvest time and (ii) the non-irrigated control treatment generally produced more seeds than the other treatments (Garcia-Diaz and Steiner, 1999 and 2000), it is best to not irrigate birdsfoot trefoil grown for seed under the humid temperate marine climatic conditions found in western Oregon. Further research is needed to determine the relationships between climatic conditions at harvest time and the mechanisms that cause pod dehiscence so that optimal harvest time can be predicted and seed losses due to shattering minimized.Cuenca Nicholson 1982

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Oregon Agric. Exp. Stn. Technical Paper No. 11,410.

Received for publication October 5, 1998.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 

• Anderson S.R. Development of pods and seed of birdsfoot trefoil, Lotus corniculatus L., as related to maturity and to seed yields. Agron. J. 1955;47:483-487.[Free Full Text]
• Buckovic, R.G. 1952. Some of the morphologic and agronomic factors associated with pod dehiscence in Lotus corniculatus.M.S. thesis. Oregon State College, Corvallis.
• Cuenca R.H. Hydrologic balance model using neutron probe data. J. Irrig. Drain. Engin. 1988;114:644-663.
• Cuenca R.H. Irrigation system design: an engineering approach. Englewood Cliffs, NJ: Prentice Hall, 1989.
• Cuenca R.H., Nicholson M.T. Application of Penman equation wind function. J. Irrig. Drain., Div., Am. Soc. Civil Engin. 1982;108:13-24.
• Esau K. Anatomy of seed plants, 2nd ed New York: John Wiley, 1960.
• Fahn A., Zohary M. On the pericarpial structure of the legumen, its evolution and relation to dehiscence. Phytomorphology 1955;5:99-111.
• Fairey, D.T. 1994. Seed production in birdsfoot trefoil, Lotus spp.: A review of some limiting factors. p. 81–85. In P.R. Beuselinck and C.A. Roberts (ed.), First Int. Lotus Symp., St. Louis, MO. 22–24 March 1994. Univ. Exten., Univ. of Missouri, Columbia, MO.
• Garcia-Diaz C.A., Steiner J.J. Birdsfoot trefoil seed production: I. Crop-water requirements and response to irrigation. Crop Sci. 1999;39:775-783.[Abstract/Free Full Text]
• Garcia-Diaz C.A., Steiner J.J. Birdsfoot trefoil seed production: II. Plant water status on reproduction and seed yield. Crop Sci. 2000;40:449-456 (this issue).[Abstract/Free Full Text]
• Gataric, D., S. Alibegovic-Grbic, N. Gabocik, V. Krajcovic, and V.M. Zimkova. 1990. The influence of weather conditions and some agronomy on plant development and yield components in seed production of birdsfoot trefoil. p. 305–308. In N. Gaborcik, et al. (ed.) Soil-grassland-animal relationships. Proc. Gen. Meeting Eur. Grassland Fed., 13th. Banska Bystrica, Czechoslovakia.
• Gershon, D. 1961. Breeding for resistance to pod dehiscence in birdsfoot trefoil (Lotus corniculatus L.) and some studies of the anatomy of pods, cytology and genetics of several Lotus species and their interspecific hybrids. Ph.D. Thesis (Diss. Abstr. AAG6104877) Cornell Univ., Ithaca, NY.
• Grant W.F. Seed pod shattering in the genus Lotus (Fabaceae): A synthesis of diverse evidence. Can. J. Plant Sci. 1996;76:447-456.
• Hare M.D., Lucas R.J. "Grasslands Maku" lotus seed production. Development of Maku lotus seed and the determination of time of harvest for maximum seed yields. J. Appl. Seed Prod. 1984;2:58-64.
• Li Q., Hill M.L. A study of flower development and seed yield components in birdsfoot trefoil (Lotus corniculatus L.). J. Appl. Seed Prod. 1989;7:65-69.
• MacDonald, H.A. 1946. Birdsfoot trefoil (Lotus corniculatus L.). Its characteristics and potentialities as a forage legume. Cornell Agric. Exp. Stn. Memoir 261.
• McGraw R.L., Beuselinck P.R. Growth and seed yield characteristics of birdsfoot trefoil. Agron. J. 1983;75:443-446.[Abstract/Free Full Text]
• Metcalfe D.S., Johnson I.J., Shaw R.H. The relation between pod dehiscence, relative humidity and moisture equilibrium in birdsfoot trefoil, Lotus corniculatus L. Agron. J. 1957;49:130-134.[Abstract/Free Full Text]
• Pieroni S.J., Laverack G.K. Determination of harvest date in Lotus corniculatus by pod color. J. Appl. Seed Prod. 1994;12:62-65.
• Seaney, R.R., and P.R. Henson. 1970. Birdsfoot trefoil. p. 1–30. In N.C. Brady (ed.) Adv. Agron. Acad. Press Inc., New York.
• Steiner J.J., Jacobsen T.J. Diurnal soil temperature effects on cotton seedling emergence and rate of development. Crop Sci. 1992;32:238-244.[Abstract/Free Full Text]
• Steiner J.J., Opoku-Boateng K. Natural season-long and diurnal temperature effects on lettuce seed production and quality. J. Am. Soc. Hort. Sci. 1991;116:396-400.[Abstract/Free Full Text]
• Wiggans S.C., Metcalfe D.S., Thompson H.E. The use of desiccant sprays in harvesting birdsfoot trefoil for seed. Agron. J. 1956;48:281-284.[Abstract/Free Full Text]
• Winch J.E., MacDonald H.A. Flower, pod and seed development relative to the timing of the seed harvest of Viking birdsfoot trefoil, Lotus corniculatus. Can. J. Plant Sci. 1961;41:523-531.
• Winch, J.E., S.E. Robinson, and C.R. Ellis. 1985. Birdsfoot trefoil seed production. Min. Agric. and Food. Order No. 85-111. Ontario, Canada.

This article has been cited by other articles:

				Y. Wang, C. Mu, Y. Hou, and X. Li Optimum Harvest Time of Vicia cracca in Relation to High Seed Quality during Pod Development Crop Sci., March 19, 2008; 48(2): 709 - 715. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J. Search for Related Content PubMed Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J. Agricola Articles by Garcia-Diaz, C.A. Articles by Steiner, J.J.