Published online 24 January 2006
Published in Crop Sci 46:243-246 (2006)
© 2006 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
CROP BREEDING, GENETICS & CYTOLOGY
Outcrossing in Annual Canarygrass
M. A. Matus-Cádiz and
P. Hucl*
Dep. of Plant Sciences and Crop Development Centre, University of Saskatchewan, 51 Campus Dr., Saskatoon, Saskatchewan, Canada S7N 5A8
* Corresponding author (hucl{at}sask.usask.ca)
 |
ABSTRACT
|
|---|
Currently recommended minimum isolation distances of 3 and 10 m for pedigreed seed production of annual canarygrass (Phalaris canariensis L.; 2n = 2x = 12) may not be sufficient to maintain the purity of seed lots. The objective of this study was to estimate outcrossing (OC) rates of annual canarygrass over various distances and directions, using the pubescent-hulled trait as a dominant detectable marker. In 2001 and 2002, 5- x 5-m pubescent-hulled pollinator blocks (‘Cantate’) were sown and surrounded by recipient rows of glabrous-hulled annual canarygrass (‘CDC Maria’) at Saskatoon, SK. At maturity, 0.5- x 5-m strips were harvested at 0, 0.5, 1, 2, 3, 6, 9, 12, 15, 18, 21, 24, 27, and 30 m along four transects (N, S, E, W) radiating out from the central pollinator block. Outcrossing from Cantate to recipient plants was identified by the expression of trichomes on the hulls of F1 derived seed. In both years, OC rates adjacent to the pollinator did not exceed 
2.2%, while OC rates declined rapidly to 0% (2001) or trace levels (0.05%, 2002) at 30 m. Elevated OC to the N of the pollinator was associated with prevailing winds in both years. Maximum OC rates of 2.2% support the suggestion that annual canarygrass is predominantly self-fertilizing. In both years, low OC rates of 0.07% were observed at 24 m, suggesting that currently recommended minimum isolation distances may be inadequate to reduce outcrossed derived off-types to acceptable levels in pedigreed annual canarygrass seed.
Abbreviations: OC, Outcrossing
|
INTRODUCTION
|
|---|
ANNUAL CANARYGRASS has become an important alternative crop in western Canada. In Saskatchewan, the area seeded to this crop has increased from 809 ha in 1972 to 340 000 ha in 2004 (Saskatchewan Agriculture and Food, 2004). Currently, its primary market is birdfeed, but the dehulled caryopsis has potential for human consumption (Abdel-Aal et al., 1997). The hull (palea and lemma) surrounding the caryopsis of pubescent-hulled cultivars are covered with silicified trichomes that break off during threshing (Putman et al., 1996). Studies link the trichomes to esophageal cancer in humans and skin cancer in laboratory mice and livestock (O'Neill et al., 1986), indicating that annual canarygrass products destined for human consumption markets must limit trichomes (Matus-Cádiz et al., 2003). The advent of the cultivar CDC Maria, possessing glabrous-hulls (Hucl et al., 2001), and the resulting ‘Canario’ commodity class with a fixed tolerance for 1% pubescent contaminants, has produced a need to validate current isolation requirements for pedigreed seed production. The pubescent-hull trait can be used as a dominant gene marker (Matus-Cádiz et al., 2003) to estimate OC in annual canarygrass.
The Canadian pedigreed seed system produces Select, Foundation, Registered, and Certified seed (Anonymous, 1994). Certified seed, the terminal pedigree class, is recommended for use by the crop producer. In Canada, a maximum impurity tolerance of 0.005% (1 off-type per 20 000 plants) in Select seed, 0.01% (1 off-type per 10 000 plants) in Foundation or Registered seed, and 0.05% (5 off-types per 10 000 plants) in Certified seed is deemed acceptable in annual canarygrass production. The appearance of morphological off-types of the same crop kind indicates that cultivar purity has been compromised by mechanical admixture or OC. In annual canarygrass, minimum isolation requirements of 10 (Select seed) and 3 m (Foundation, Registered, or Certified seed) from the same crop kind are recommended to minimize off-types in pedigreed seed production (Anonymous, 1994).
To date, OC rates in annual canarygrass remain unreported and thus minimum isolation distances are based on the assumption that the species is predominantly self-pollinating (
99.5%). The detection of trace levels of OC at 27 m in wheat (Triticum aestivum L.; Hucl and Matus-Cádiz, 2001) has directed our research to assess OC levels in annual canarygrass pedigreed seed. The objective of this study was to estimate the OC rate of annual canarygrass grown at zero to 30 m to the north, east, south, and west of a pubescent-hulled pollinator.
|
MATERIALS AND METHODS
|
|---|
The European annual canarygrass cultivar Cantate, a pubescent-hulled cultivar (Matus and Hucl, 1999), was used as the pollen source to quantify OC rates in 2001 and 2002. The pubescent-hulled pollinator was seeded in a 5- x 5-m block on fallow land at the Kernen Crop Research Farm (Univ. of Saskatchewan) on 25 May 2001 and 30 May 2002 at a low rate (125 seeds/m2) to promote tillering and extend the period of pollen shedding. The soil type was a Dark Brown Chernozem (Typic boroll) Sutherland clay, clay-loam at the Kernen Crop Research Farm. Breeder seed of CDC Maria, a glabrous-hulled cultivar, was seeded perpendicular to the sides of the pollinator block to the north, south, west, and east in 31-m strips at a rate of 250 seeds/m2. Each strip consisted of 16 rows, 31 m long and 0.3 m apart. Seed of Cantate and CDC Maria was obtained from the Crop Development Centre, Saskatoon, SK.
Days to heading, duration of anthesis, and plant height data were collected. Data for the pollinator block were averaged over values collected at five random positions. Data for CDC Maria were averaged over values collected at seven positions (0, 5, 10, 15, 20, 25, and 30 m from the pollinator) in each of four directions (N, E, S, and W). Wind speed (km h–1) and wind direction data were collected within 1 km of the trial site by an Environment Canada weather station (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Meteorological data (± SD) during the estimated pollination period between the pubescent-hulled pollinator source Cantate and glabrous-hulled recipient CDC Maria in 2001 and 2002.
|
|
At maturity, 0.5-m lengths of row were cut within each direction at the following distances from the pollinator block: 0, 0.5, 1, 2, 3, 6, 9, 12, 15, 18, 21, 24, 27, and 30-m. Fifty-six samples of CDC Maria were dried overnight in large driers. In 2001 and 2002, panicles within each of the 56 samples were visually inspected to confirm that each sample contained only panicles with glabrous-hulls. After inspection, each sample was bulk threshed with a rubber-belted deawner. Samples harvested within each direction were threshed in descending order (i.e., 30–0 m), starting with the sample most distant from the pollinator source. A 1- x 1-m quadrant sample of wheat was threshed to clean the deawner between samples from different directions.
A grow-out of seed harvested in 2001 and 2002 was conducted in 2002 and 2003, respectively, to determine OC rates. In 2002 and 2003, seed from each of the 56 samples were grown in strip plots. Strip plots were seeded at a rate of 250 seeds/m2 on 16 May 2002 and 21 May 2003 on fallow land at the Kernen Crop Research Farm. Each strip plot consisted of four rows, 26.6 m long and 0.3 m apart. At maturity, 2000 panicles per strip plot were harvested and dried overnight on air-forced driers. The 2000 panicles within each of the 56 samples of CDC Maria were visually inspected to identify pubescent-hulled panicles.
Ten seeds from each pubescent-hulled panicle were grown out under greenhouse conditions to confirm visual identifications. Fifteen seeds per sample were imbibed within 90-mm Petri-dishes (each containing a Whatman No.1 filter paper) in 2.0 mL of a 0.1% (w/v) KNO3 solution (Matus-Cádiz et al., 2001). Seeds were allowed to imbibe KNO3 for 24 h at 22°C (in darkness) and subsequently prechilled for 3 d at 4°C (in darkness). After prechilling, seeds were incubated for 3 to 14 d at 15°C (in darkness). Ten to 15 pregerminated seeds were planted (1-cm depth) in 15-cm diameter pots. Pots were filled with Terra-Lite Redi-Earth (W.R. Grace and Co. of Canada Ltd. Ajax, ON). Greenhouse conditions were set to 24/18°C (day/night) with 18 h light and a photosynthetically active radiation level of 250 µmol m–2 s–1. Plants were watered every 4 d and fertilized with Type 100 Nutricote controlled release granular fertilizer (14–14–14) (Plant Products Co. Ltd. Brampton, ON) at a rate of 0.8 kg m–2.
Plants were allowed to self pollinate and segregation among F1 derived F2 seed was classified as segregating (3 pubescent-hulled: 1 glabrous-hulled) or non segregating (all pubescent-hulled seeds) for the pubescent-hulled trait. The number of putative pubescent panicles identified and confirmed to be segregating for the pubescent-hulled trait were 257 (year one) and 374 panicles (year two). Outcrossing rates were calculated as follows: OC (%) = (# of pubescent panicles observed in si at dj/2000 panicles of si at dj) x 100, where si is the ith sample and dj is the jth distance. For meteorological data, grand means between years were tested for significance (P = 0.05) by one tailed t tests (Minitab Version 13; Minitab Inc, State College, PA).
|
RESULTS
|
|---|
In 2001, the estimated duration of flowering was 10 d for CDC Maria (5% anthesis on 10 July to 95% on 19 July) and 18 d for Cantate (5% anthesis on 8 July to 95% on 25 July). The estimated overlap in pollination periods between CDC Maria and Cantate was about 10 d (10–19 July) (Table 1). In 2002, the estimated duration of flowering was 18 d for CDC Maria (5% anthesis on 19 July to 95% on 5 August) and Cantate (5% anthesis on 19 July to 95% on 5 August). The estimated overlap in pollination periods between CDC Maria and Cantate was about 18 d (19 July to 5 August). A good level of nicking was expected between recipient and pollinator plants in both years. The estimated periods of overlap in this study may be conservative as the stigma of male-fertile wheat plants are known to be receptive for a period of 4 to 13 d (de Vries, 1971).
In 2001, prevailing winds were from the SSE and S during 4 d (averaging 17 km h–1), from the NE, ENE, and E during 4 d (averaging 13 km h–1), and from the SW for the remaining 2 d of pollination (averaging 15 km h–1) (Table 1). Mean temperature was slightly higher in 2001 than 2002. In 2002, prevailing winds were from the W, WSW, and SSW during 8 d (averaging 17 km h–1), the S and SE for 3 d (averaging 18 km h–1), the E and ENE for 3 d (averaging 11 km h–1), and the WNW, NW, and N for the remaining 4 d of pollination (averaging 21 km h–1) (Table 1).
Outcrossing rates were higher in 2002 than 2001 (Table 2). In 2001, percentage OC declined rapidly as distance increased and leveled off at 0.05% by 9 m to the N of the pollinator, 0.15% by 3 m to the E, 0.10% by 3 m to the S, and 0.05% by 24 m to the W. Approximately 95.0, 99.7, and 100% of cumulative OC occurred within 3, 9, and 24 m of the pollinator block, respectively. Approximately 72% of cumulative OC occurred to the N and W of the pollinator. Prevailing wind direction was associated with elevated OC rates to the N and W of the pollinator during the entire pollination period of 10 d (Tables 1 and 2). In 2002, percentage OC declined rapidly as distance increased and decreased to 0.10% by 9 m to the N and E of the pollinator and 0.15% by 21 m to the S and W. Approximately 95.0, 98.9, and 100% of cumulative OC occurred within 18, 27, and 30 m of the pollinator block, respectively. Approximately 65% of cumulative OC occurred to the N and S of the pollinator. Prevailing wind direction (SW, S, SE, NE, NW) during 12 out of the 18 d of pollination was associated with elevated OC rates to the N and S of the pollinator (Tables 1 and 2).
View this table:
[in this window]
[in a new window]
|
Table 2. Outcrossing rates (%) observed for CDC Maria grown zero to 30m to the north, east, south, and west from pubescent-hulled pollinator Cantate at Saskatoon, SK in 2001 and 2002.
|
|
|
DISCUSSION
|
|---|
Pollen dispersal during flowering varies with pollinator field size (de Vries, 1974). Hybrid wheat production studies have studied optimum sizes of pollinator blocks for cross-pollination. These studies generally use small pollinator plots and, thus, have limited application in estimating the amount of OC taking place between commercial-scale fields (Matus-Cádiz et al., 2004). Caution must be applied to extrapolating the OC rates reported in the present study to larger-scale field studies using commercial-scale pollinator sources as our study may underestimate the level of OC occurring between commercial-scale fields (e.g., pubescent-hulled vs. glabrous-hulled annual canarygrass fields). In addition, trace OC rates of 0.05% may occur at distances exceeding 30 m when using commercial-scale pollinator sources. Further research would be needed to determine the level of OC occurring between neighboring commercial-scale annual canarygrass fields.
Pollen dispersal during flowering is affected by plant height (de Vries, 1972). Averaged over years, CDC Maria (63 cm, SE ± 1) was only 5 cm taller in plant stature compared with Cantate (58 cm ± 1). This indicates that the height of the pollinator relative to target plants was probably not a major factor in promoting the downward dispersal of pollen onto recipient plants.
Pollen dispersal during flowering varies with environmental conditions such as prevailing winds, wind speed, temperature, humidity, and precipitation (de Vries, 1971, 1972, 1974). Outcrossing was generally dependent on wind direction during the flowering period of the crop (Tables 1 and 2). Wind direction and OC rates showed a weaker association in 2002 compared with 2001, indicating that OC rates should not be based on experiments oriented in only one direction from the pollinator. The pollination period in 2001 was considerably shorter and slightly warmer relative to that observed in 2002. Higher OC rates in 2002 likely reflect that the pollination period in 2002 was nearly twice as long as that observed in 2001. Elevated OC rates are also promoted by cooler conditions (de Vries, 1972), suggesting that the elevated OC rates in 2002 relative to 2001 were at least, in part, promoted by cooler temperatures. de Vries (1972) reported that the highest concentration of wheat pollen appeared to be dispersed at a temperature of 16 to 20°C and relative humidity of 70 to 75%. Wheat pollen grains have been reported to be viable for 15 to 20 min or up to 30 min under optimal conditions (de Vries, 1971). In the present study, the weather conditions in 2002 fall within the optimum temperature and humidity range reported by de Vries (1972).
Environmental conditions from year to year are considered to be random factors or factors that cannot be controlled by the researcher. Consequently, the weather conditions in some years will promote high OC rates and conditions in other years will depress OC rates, indicating that OC rates should be based on experiments conducted over multiple years. This suggests that recommended isolation distances used in pedigreed seed production should be based on data from environments with near-optimal conditions for elevated OC. For example, our data suggests that in years with depressed OC rates (e.g., 2001), a minimum isolation distance of 27 m would likely suffice to minimize OC-derived off-types in the subsequent generation of pedigreed annual canarygrass. In contrast, in years where weather conditions promote OC (e.g., 2002), a minimum isolation distance of >30 m would be needed. Generally, our results indicate that current isolation distances in Canada may be inadequate to mitigate outcrossed derived off-types to acceptable levels in pedigreed annual canarygrass seed production.
|
ACKNOWLEDGMENTS
|
|---|
Appreciation is expressed to L. Ehman, G. Trowell, W. Schatz, M. Grieman, M. Cardenas, N. Kwan, and H. Braitenbach for their technical assistance throughout the research. This research was funded by a grant from the CSGA.
Received for publication April 15, 2005.
|
REFERENCES
|
|---|
- Abdel-Aal, E.S.M., P. Hucl, and F.W. Sosulski. 1997. Structural and compositional characteristics of canarygrass (Phalaris canariensis L.). J. Agric. Food Chem. 45:3049–3055.[CrossRef]
- Anonymous. 1994. Regulations and procedures for pedigreed seed crop production. Circular 6–94. Canadian Seed Growers' Association. Ottawa, ON. Available at http://www.seedgrowers.ca/regulationsandforms/circular.asp#1 (verified 20 September 2005).
- de Vries, A.P. 1971. Flowering biology of wheat particularly in view of hybrid seed production: A review. Euphytica 20:152–170.[CrossRef][ISI]
- de Vries, A.P. 1972. Some aspects of cross pollination in wheat (Triticum aestivum L). 1. Pollen concentration in the field as influenced by cultivar, diurnal pattern, weather conditions, and level as compared to the height of the pollen donor. Euphytica 21:185–203.[CrossRef]
- de Vries, A.P. 1974. Some aspects of cross pollination in wheat (Triticum aestivum L). 4. Seed set on male sterile plants as influenced by distance from pollen source, pollinator: Male sterile ratio, and width of the male sterile strip. Euphytica 23:601–622.[CrossRef][ISI]
- Hucl, P., M.A. Matus-Cádiz, A. Vandenberg, F.W. Sosulski, E.S.M. Abdel-Aal, G.R. Hughes, and A.E. Slinkard. 2001. CDC Maria annual canarygrass. Can. J. Plant Sci. 81(1):115–116.
- Hucl, P., and M.A. Matus-Cádiz. 2001. Isolation distances for minimizing outcrossing in spring wheat. Crop Sci. 41:1348–1351.[Abstract/Free Full Text]
- Matus, M., and P. Hucl. 1999. Isozyme variation within and among accessions of annual Phalaris species in North American germplasm collections. Crop Sci. 39:1222–1228.[Abstract/Free Full Text]
- Matus-Cádiz, M., P. Hucl, and G. Munasinghe. 2001. Seed dormancy & germination in three annual canarygrass (Phalaris canariensis) cultivars relative to spring wheat (Triticum aestivum). Seed Sci. Technol. 29:523–531.
- Matus-Cádiz, M.A., P. Hucl, and A. Vandenberg. 2003. Inheritance of hull pubescence and seed color in annual canarygrass. Can. J. Plant Sci. 83:471–474.
- Matus-Cádiz, M.A., P. Hucl, M.J. Horak, and K. Blomquist. 2004. Gene flow in wheat at the field scale. Crop Sci. 44:718–727.[Abstract/Free Full Text]
- O'Neill, C.H., P. Jordon, T.S. Bhatt, and R.H. Newman. 1986. Silica and oesophageal cancer. Ciba Found. Symp. 121:214–230.[Medline]
- Putman, D.H., P.R. Miller, and P. Hucl. 1996. Potential for production and utilization of annual canarygrass. Cereal Foods World 41:75–83.
- Saskatchewan Agriculture and Food. 2004. Specialty crop report. Saskatchewan Agriculture and Food. Regina, SK. Available at http://www.agr.gov.sk.ca/DOCS/crops/special_crops/production_information/specialtycroprpt2004.pdf (verified 20 September 2005).
TOP
ABSTRACT
INTRODUCTION
