II.3. BRASSICA CROPS (BRASSICA SPP.) – 199
are about 5 mm long, the megaspore in each ovule divides twice, producing four cells,
one of which becomes the embryo sac, while the others abort. The nuclear tissue is
largely displaced by the remaining embryo sac and at flower opening, the ovules mainly
consist of two integuments and the ripe embryo sac.
Pollination, pollen dispersal and viability
Brassica pollen, although heavy and slightly sticky, can still become airborne and
float on the wind due to its minute size (30-40 μm). In addition to wind, pollen can be
transferred by insects, primarily honey bees (Williams, Martin and White, 1987, 1986;
Scheffler, Parkinson and Dale, 1993; Paul, Thompson and Dunwell, 1995; Timmons et al.,
1995; Thompson et al., 1999). Physical contact between flowers of neighbouring plants
also results in pollen dispersal while animals, including humans, passing through
flowering Brassica fields can act as pollen vectors.
Pollen movement can be detected using pollen traps for airborne pollen or by using
bait plants (either male sterile or emasculated) to detect outcrossing, usually through the
use of marker genes such as herbicide resistance. An effective pollen trap, developed in
Germany, combines a sampler that determines pollen deposition rate (Sigma-2 sampler)
and a pollen mass filter apparatus that collects sufficient pollen for polymerase chain
reaction (PCR) analysis (VDI Richtlinien, 2007). Pollen from the Sigma-2 sampler is
analysed as to species and amount under a light microscope and/or by automated imaging
analysis. Strategically located bee hives can also be used to monitor pollen flow whereby
pollen in honey and bee bread samples is concentrated and analysed under a light
microscope or subjected to PCR analysis (VDI Richtlinien, 2006).
Under natural conditions, Ranito-Lehtimäki (1995) reported a gradual decrease in
pollen viability over four to five days. In the laboratory, Mesquida and Renard (1982)
found pollen remained viable between 24 hours to 1 week. However, Chiang (1974)
reported that B. oleracea pollen stored at 4ºC germinated above 20% for the first 10 days,
and even after 6-7 weeks an average 4.5% of the test pollen remained viable.
The greatest pollen outflow from flowering Brassica fields is undoubtedly
wind borne. Studies have shown that the vast majority of the pollen cloud travels less
than 10 m and approximately half the pollen produced by an individual plant falls to the
ground within 3 m (Lavigne et al., 1998). In a two-year study, Bilsborrow et al. (1998)
reported that the pollen concentrations at 10 m was reduced by 48% and 67% compared
to that recorded 2 m from the field border. McCartney and Lacey (1991) found that the
amount of pollen detected at 20 m from the field border was 90% less than that recorded
at the field edge. Over longer distances of 360 m and 400 m, relative to the field margin,
Timmons et al. (1995) and Thompson et al. (1999) reported reductions of 90% and 95%,
respectively. These findings, combined with outcrossing data, established that Brassica
pollen follows a leptokurtic distribution i.e. the presence of pollen shows a steep decline
with distance, but with a long tail containing long-distance events (Figure 3.36;
Thompson et al., 1999; Staniland et al., 2000). These data indicate that at a distance of
50 m from the pollen source, the level of outcrossing is less than 0.5%, even when male
sterile bait plants are used as pollen recipients (Figure 3.36).