Table Of ContentsNext Page

Secondary Dormancy of Oilseed Rape: first Aspects of Heredity

Sabine Gruber and Wilhelm Claupein

University of Hohenheim, Institute for Plant Production and Grassland Research., 70593 Stuttgart, Germany
Email grubersf@uni-hohenheim


Individual plants from four oilseed rape cultivars grown within an assortment of different cultivars in a field were prevented from open-pollination. The comparison of seeds from self-pollination and open-pollination of the same cultivar resulted in somewhat significantly different levels of secondary dormancy induced in the laboratory. Self-pollination enhanced the level of secondary dormancy in highly dormant cultivars and reduced it in low dormant cultivars. In a second experiment, previously dormant and non-dormant seeds were grown to maturity in the greenhouse. All plants produced dormant and non-dormant seeds. The seed progeny of plants deriving from previously dormant seeds had a higher level of dormancy than the progeny of non-dormant seeds and the control, although differences were not significant (P=0.06). All individual plants differed from each other in the dormancy level of their progeny. If the mean dormancy level of a cultivar is consequently composed of various dormancy levels of all individuals in the population, breeding for low dormancy could be performed by simple mass selection from within the cultivar. The risk of affecting secondary dormancy by outcrossing should be considered if gene flow in time by volunteers emerging from the soil seed bank is undesired.

Media summary

First isolation and selection experiments with oilseed rape pointed out chances for breeding low dormant ideotypes to minimise the risk of gene flow.

Key Words

Seed persistence, rapeseed, selection, gene dispersal


Oilseed rape volunteers occur worldwide as a consequence of relatively high harvest losses and the seed’s capability of becoming dormant. Thus the seeds are able to persist in the soil over several months or even years (Roller et al. 2002), and can act as a long-term source for volunteers. In turn, this may contribute toward an undesirable gene flow. It is known that post-harvest conditions have a clear effect on the exhibition of secondary dormancy and seed persistence (Pekrun et al. 1998, Gruber et al. 2004). Additionally, there appears large genotypic variation in the capacity of becoming secondarily dormant as reported by Pekrun et al. (1997), Momoh et al. (2002) and Gruber et al. (2003). Less is known about the inheritance of secondary dormancy in oilseed rape, although breeding oilseed rape genotypes with a low capacity for becoming dormant could minimise the risk of gene flow by volunteers, for instance when transgenic cultivars are grown. The aim of the current study was a first approach to examine heredity of secondary dormancy in oilseed rape by comparing open-pollinated and self-pollinated plants, and by selecting dormant and non-dormant genotypes in two subsequent generations.

Materials and Methods

General Laboratory Method

Seeds were cleaned of impurities and induced into secondary dormancy by a treatment with polyethylene glycol (PEG) 6000 solution at an osmotic potential of -15 bar. 12 100 seeds from each cultivar were imbibed in 8ml PEG solution on 9cm Petri dishes with a double layer of filter paper, and put into boxes lined with black plastic to exclude light. After a 14 day period of dormancy induction at 20C, seeds were transferred to Petri dishes containing 6ml deionised water. During the following 14 days in darkness and at 20C, non-dormant seeds germinated and were removed under a green safety light (500–600 nm). All remaining seeds were classified to be secondarily dormant if they germinated after a following stratification stimulus of alternating light and temperature conditions (3/30C; darkness/light; 12/12 h) over three days.

Experiment 1

The winter oilseed rape cultivars Express, Falcon, Bristol and Mohican were grown from August 2001 to July 2002 in 6 12 m plots as part of a larger set of 31 different oilseed rape cultivars for demonstration purposes. A 10m margin grown with Express enclosed the whole field. Ten plants of each cultivar were isolated in translucent, perforated plastic bags (hole diameter 0.5mm) to prevent plants from open-pollination but allow aeration. The isolated plants were harvested manually, the samples of individual plants mixed and treated in the laboratory as described above, as well as the mechanically harvested, open-pollinated plants of the same plot.

Experiment 2

Seeds of the winter oilseed rape cultivar Zenith (open-pollinator) were harvested in 2002 and then separated into dormant and non-dormant seeds by the laboratory method. From each fraction 15 seedlings were grown in the greenhouse in a row-column design, together with a control of 15 plants from the original cultivar. After isolation of the plants in translucent, perforated plastic bags (hole diameter 0.5mm) during flowering, seeds of all individuals were sampled separately and tested for their capacity for secondary dormancy in the laboratory.

Statistical analysis

The data were transformed by arcsine transformation (experiment 1) or log-transformation (experiment 2) in order to stabilise the variance and normalise the data. For the analysis of variance, the SAS mixed model procedure MIXED (SAS Institute Inc.) was used. Significances between cultivars were determined by Fisher’s LSD (P = 0.05).

Results and Discussion

Experiment 1

The level of secondary dormancy in open-pollinated plants of the cultivars tested in the study and those in the directly adjacent plots varied from 2.5 to 69.3% dormant seeds/viable seeds (Table 1).

Table 1. Secondary dormancy (dormant seeds/viable seeds, %) of the open-pollinated, tested cultivars and cultivars in the adjacent plots



Dormant /viable seeds (%)


Dormant /viable seeds (%)


Dormant /viable seeds (%)


Dormant /viable seeds (%)




























Express and Falcon exhibited low dormancy levels between 1 and 4% dormant seeds/viable seeds, Bristol resulted in 35–36% and Mohican in 51–78% in the self- and open-pollinated fractions (Figure 1).

Figure 1. Secondary dormancy (dormant/viable seeds, %) of four oilseed rape cultivars after self-pollination and open-pollination; bars: standard error of mean, significances according Fisher’s LSD ( P=0.05)

Outcrossing can be assumed at least from one adjacent plot into the other, but also pollination with a more or less uniform pollen mixture from all 31 cultivars with a mean dormancy level of 33% dormant seeds/viable seeds may have occurred.

When Bristol and Mohican, cultivars with a comparatively high capacity for seed dormancy, have been prevented from pollination with genotypes characterised by a lower dormancy potential, more seeds could be induced into secondary dormancy compared to open-pollinated cultivars. In contrast, the low dormant cultivars resulted in higher dormancy levels when open-pollinated and a possible fertilisation by high dormant genotypes took place. The differences were significant between the self-pollinated and the open-pollinated fraction of Mohican (P < 0.01) and Falcon (P < 0.001).

Since outcrossing affected dormancy of the seeds produced, the capacity for becoming secondarily dormant may be regarded as under genetic control. However, the potential still exists that the isolation or harvest method may have had an effect on seed dormancy. The results corroborate the hypothesis that the level of secondary dormancy in oilseed rape can be partly influenced by genotype. Besides maternal effects during maturation influencing dormancy (Fenner 1991), environmental factors after harvest including humidity, light, temperature or gaseous conditions (Lpez-Granados and Lutman 1998; Momoh et al. 2002), a third crucial factor for secondary dormancy may be the nuclear genotype.

Experiment 2

Figure 2. Secondary dormancy (untransformed, dormant seeds/viable seeds, %) in the progeny of dormant and non-dormant seeds and the genuine cultivar as control. Bars represent standard deviations.

Plants grown from the previously dormant and non-dormant seeds of cultivar Zenith and from the original cultivar as control produced dormant seeds as well as non-dormant seeds, respectively (Figure 2).

The dormancy levels of the progeny of the individual plants varied within each selected fraction and within the control. Therefore, the dormancy level that characterises a cultivar, at least over two years (Gruber et al. 2003), is not represented uniformly in all plants of the population, but is composed by various dormancy levels of all plants. An explanation might be the involvement of several genes in the onset of secondary dormancy in oilseed rape, as reported for other species (Foley and Fennimore 1998). The mean dormancy level of seeds from all plants in each fraction was 7% in the progeny of dormant seeds and 4% in the progeny of non-dormant seeds. The control resulted in a mean dormancy level of 2%. According to previous dormancy tests (Gruber, unpublished), secondary dormancy in the cultivar Zenith grown in the field was much higher (about 60%) in the harvest years 2001 and 2002. But this may be due to the unusually high temperatures in the greenhouse in summer of 2003. The phenomenon of generally low dormancy levels of seeds harvested in the exceptionally hot summer of 2003 was also observed in 18 oilseed rape cultivars grown in the field (Gruber, unpublished).

Differences between the dormancy levels of the progenies and the control were not significant in the F-test (P = 0.06). Nevertheless, 14 of 15 previously dormant seeds grew to produce plants with dormant seeds, whereas only 8 plants derived from non-dormant seeds and 9 plants of the control produced any dormant seeds at all. This indicates that selection could be successful in breeding of low dormant genotypes.

It can be assumed that other open-pollinator cultivars also consist of dormant individuals, if these cultivars are not completely homozygous. In hybrids, segregation in the F2 generation may produce seeds giving rise to volunteers, even when parents appear to lack seed dormancy.


As secondary dormancy in the seeds of a cultivar can be influenced by outcrossing from cultivars in adjacent fields, it may be possible that parts of a crop located close to adjacent crop may exhibit higher seed dormancy level. The consequence would be a higher risk of gene flow via the soil seed bank and emerging volunteers. Furthermore, the results suggest that the physiological onset or release of secondary dormancy in oilseed rape is, partly, located in nuclear tissue containing both maternal and paternal genes i.e. embryo. A number of genes is probably involved in the onset of secondary dormancy. Even if there is no evident success in selection after the first generation, further experiments with a higher number of cultivars, with selection over more generations and including specific matings can provide information about the number and function of genes involved in secondary dormancy of oilseed rape. As high and low dormant plants can be found in the same cultivar, selection for low dormancy may be readily performed.


Fenner M (1991). The effects of the parent environment on seed germinability. Seed Science Research 1, 75–84.

Foley ME, Fennimore SA (1998). Genetic basis for seed dormancy. Seed Science Research 8, 173–182.

Gruber S, Pekrun, C and Claupein W (2004). Population dynamics of volunteer oilseed rape (Brassica napus L.) affected by tillage. European Journal of Agronomy, 20, 351–361.

Gruber S, Pekrun, C and Claupein W (2003). Seed persistence of genetically modified and conventionally bred oilseed rape in laboratory and burial experiments. Proceedings of the 11th International Rapeseed Congress, Copenhagen, Denmark (Groupe Consultatif International de Recherche sur le Colza), 876–878.

Lpez-Granados F and Lutman PJW (1998). Effect of environmental conditions on the dormancy and germination of volunteer oilseed rape seed (Brassica napus). Weed Science 46, 419–423.

Momoh EJJ, Zhou WJ and Kristiansson B (2002). Variation in the development of secondary dormancy in oilseed rape genotypes under conditions of stress. Weed Research 42, 446–455.

Pekrun C, Potter TC and Lutman PJW (1997). Genotypic variation in the development of secondary dormancy in oilseed rape and its impact on the persistence of volunteer rape. Proceedings of the 1997 Brighton Crop Protection Conference – Weeds, Brighton, UK (British Crop Protection Council), 243–248.

Pekrun C, Hewitt, JDJ and Lutman PJW (1998). Cultural control of volunteer oilseed rape (Brassica napus). Journal of Agricultural Science, Cambridge, 130, 155–163.

Roller A, Beismann H and Albrecht H (2002). Persistence of genetically modified, herbicide-tolerant oilseed rape–first observations under practically relevant conditions in South Germany. Journal of Plant Diseases and Protection, Special Issue XVIII, 255–260.

Top Of PageNext Page