Plant development mutants : incidence on honey bees behaviour and pollination
1 INRA, Laboratoire de Zoologie, BP 29 35653 Le Rheu cedex France
2 INRA, Station d’Amélioration des Plantes, BP 29 35653 Le Rheu cedex France
The nectar collecting posture of honey bees on the flowers and their flights between mutant and non mutant plants were compared. Observations were done on small areas in three separated experiments. Three mutants were under study : dwarf, apetalous, and cleistogamous oilseed rape lines. On dwarf plants, because the flower morphology is identical to wild types , honey bees behaved in the same manner, i.e. they crawled over the flower and had contacts with the stamens. Their flights between both types were partially reduced only when the difference in height was higher than 80cm and when they differed in their male-fertility (male-fertile versus male-sterile lines). On apetalous flowers, honey bees exhibited a side working posture which reduced their contacts with stamens. Moreover, they were individually constant to one flower type and this behaviour induced a nearly absolute lack of flights between the apetalous and the petalous types. On cleistogamous flowers, about 43% of the nectar foragers were side-working. The others were able to collect nectar by the small hole at the top of the flower. In some cases pollen gatherers opened the flower during their visits. Flights between cleistogamous and conventional flowers were by two times less frequent than within a same type. Thus, incidences of these mutations on the honey bee mediated pollen transfer are different. Dwarf and tall plants could be intercrossed. On the opposite, intercrossing between apetalous and petalous flowers could not be ensured by honey bees only. Cleistogamy could lead to an increased autogamy but a pollen dispersal by visiting honey bees could not be excluded.
Keywords
oilseed rape, dwarf line, apetalous line, cleistogamous line
Introduction
Rapeseed plant development mutants have been selected to provide for various needs. Thus, dwarf and semidwarf genotypes are more resistant to frost, to lodging, they can be sown earlier and used as nitrate traps. Apetalous lines may reduce the contamination by Sclerotinia sclerotiorum, a disease transmitted by contaminated petals (MacLean, 1958), in addition the lack of petals enables a higher incidence of the active radiations to the photosynthetic tissues (Mendham et al, 1991) and the apetalous lines could be more productive and fertile (Fray et al, 1995). Cleistogamous lines have been selected both to increase self-pollination and to reduce gene dispersal. Because honey bees are one of the main insect pollinating rapeseed, it is necessary to know what are the incidences of these plant modifications on their foraging behaviour, especially with regards to their ability to be attracted by these different phenotypes and to their capacity to transfer pollen from a mutant type to a non mutant or vice versa.
Material and method
Three separated experiments were carried on the three different mutants. Similar observations, according to previously described methods (Mesquida et al, 1988; Pierre et al, 1996), were used to determine and compare the attractiveness of the genotypes and the ability of the insect to transfer pollen from one to the other one. So, the density of foraging honeybees per 1000 available flowers along the full flowering period was evaluated along a transect and the calculation of the passing ratio (number of foraging honeybees flying from a given genotype to the neighbouring other genotype divided by the total number of honeybees, and observed on a 4m² area during 5 or 3mn) was made. For comparison, observations were made at the same time within and between the genotypes. Moreover, nectar foraging postures (crawling over the stamens or side-working posture) were registered and nectar production was evaluated.
Statistical analyses were performed by a one-way (genotype) or a two-ways (genotype and date) ANOVA followed by a Newman-Keuls test at a significant level p=0.05.
Dwarf mutant
The dwarf phenotype of Darmor was obtained from a mutation at a single locus named Bzh. The difference between tall and dwarf plants was about 80cm in height. Three open field trials were set up, on small areas, in alternated devices to investigate the effect of dwarfism and its interactions with ogu-INRA cytoplamic male-sterility:
Trial 1, male fertile ‘Bzh Darmor’ versus male-fertile ‘Darmor’ (MF Bzh D/MF D) ; Trial 2, male sterile ‘Bzh Darmor’ versus male-fertile ‘Darmor’ (MS Bzh D/MF D) ; Trial 3, male fertile ‘Bzh Darmor’ versus male-sterile ‘Darmor’ (MF Bzh D/MS D).
Apetalous mutant
The apetalous mutant was selected from an interspecific hybrid between B. oleracea and B. napus backcrossed with B. napus. It was compared to a petalous doubled haploid line from the cross ‘Prota’ x ‘Brutor’. Density and passing ratio observations were made on small plots (1.5 x 7.5m) of each type which were contiguous.
The foraging behaviour was fully observed under net cages (3 x 3 x 2 m high) by video recording and confirmed in the field.
Cleistogamous mutant
The closed flowers of the cleistogamous line were due to a single mutation (Clg1) and this line was compared to two non mutant genotypes : male-fertile ‘Gaspard’ (Trial 1) and male-sterile ‘Darmor’ (Trial 2). The design and methods were the same as in the experiment on apetalous flowers.
Results
Dwarf mutant
Number of foraging honeybees/1000 available flowers ( ± Standard error)
Trial |
Dwarf genotypes |
Tall genotypes |
Significance |
T1 |
MF Bzh D = 6.05 ± 0.9 |
MF D = 5.52 ± 0.8 |
NS p=0.4651 |
T2 |
MS Bzh D = 3.16 ± 0.4 |
MF D = 4.51 ± 0.6 |
S p=0.0308 |
T3 |
MF Bzh D = 4.16 ± 0.9 |
MS D = 3.92 ± 0.4 |
NS p=0.4104 |
The dwarfism did not induce a reduction of the attractiveness to honeybees (Trial 1) excepted when it was associated to male-sterility (Trial 2). In addition, a comparison of the dwarf genotype, made between the three trials, indicated that the male-fertile dwarf Darmor was significantly (p=0.0002) less attractive when it was contiguous to male-sterile Darmor (Trial 3) than when the trial was composed of only male-fertile plants (Trial 1). This can be explained by the highly significant difference in nectar production (µl/flower) observed between the male-sterile lines (MS Bzh D = 1.01 ; MS D= 1.38) and the male-fertile lines (MF Bzh D = 2.06 ; MF D = 2.38). That indicates that the mean value of nectar production of a trial, considered as a whole, played a role on attractiveness.
Passing ratio of foraging honeybees within and between genotypes ( ± Standard error)
Trial |
Within dwarf genotypes |
Within tall genotypes |
Between |
Significance |
T1 |
MF Bzh D = 1.33 (a) |
MF D = 1.42 (a) |
0.94 (a) |
NS p=0.2158 |
T2 |
MS Bzh D = 1.55 (a) |
MF D = 1.42 (a) |
0.59 (b) |
S p=0.0008 |
T3 |
MF Bzh D = 1.33 (b) |
MS D = 2.44 (a) |
0.53 (c) |
S p=0.0001 |
On average, the passing ratio was not significantly modified by the difference in height alone (Trial 1) even if it was reduced on 3 dates among the 5 dates of observation. The passing ratio was about 3 times lower between plants differing both in height and male-sterility (Trial 2 and 3) than within the area of plants of a same type used as references.
Because the shape of the flowers of the dwarf plants were unchanged, the foraging posture were identical to those commonly observed on rapeseed, i.e. most of the honeybees crawled over the stamens. Consequently, honeybees were able to load pollen on their body and to deposit it on the stigma from flower to flower.
Apetalous mutants
Number of foraging honeybees/1000 available flowers ( ± Standard error)
Apetalous genotype |
Petalous genotype |
Significance | |
Date 1 |
1.6 ± 0.1 |
1.8 ± 0.1 |
NS p=0.6541 |
Date 2 |
3.2 ± 0.4 |
2.2 ± 0.2 |
S p=0.0385 |
The flowers without petals were at least as attractive as petalous flowers. Even more, it appeared that sometimes the density of honeybees was higher on the apetalous line. There was no difference in nectar production between both types.
Passing ratio of foraging honeybees within and between genotypes ( ± Standard error)
Within petalous |
Between |
Significance |
2.5 ± 0.32 |
0.25 ± 0.04 |
S p=0.0000 |
The honeybees seldom flew from a type to the other. They frequently exhibited an adapted posture on the apetalous flower by inserting their tongue between the base of the sepals. This side-working reduced their contact with reproductive organs of the plant. The proportion of inserting postures were about 55% on the apetalous flowers and 6% on petalous flowers.
Cleistogamous mutants
Number of foraging honeybees/1000 available flowers ( ± Standard error)
Trial |
Wild type |
Clg mutant |
Significance |
T1 |
MF = 1.66 ± 0.30 |
0.97 ± 0.20 |
NS p=0.0557 |
T2 |
MS = 1.43 ± 0.20 |
0.96 ± 0.18 |
NS p=0.0829 |
The density of honeybees on cleistogamous flowers was not significantly lower than on opened flowers. Nevertheless, one can notice that the probability are not very far from 0.05. Compared to these wild types, the nectar production was much higher in cleistogamous flowers.
Passing ratio of foraging honeybees within and between genotypes ( ± Standard error)
Trial |
Within wild type |
Between Clg1 mutant |
Significance |
T1 |
MF = 0.84 ± 0.06 |
0.34 ± 0.04 |
S p=0.0000 |
T2 |
MS = 0.97 ± 0.09 |
0.37 ± 0.04 |
S p=0.0000 |
The flights between Clg1 mutant and non mutants were reduced in the same proportion (about 40%) whatever the neighbouring genotype was male-fertile or male-sterile. Under cages, about 43% of the nectar foraging honeybees were side-working, the remaining nectar gatherers collected nectar by the little hole at the top of the flower and had some contacts with the stamens in the front part of their body. In addition, pollen gatherers were also able to enlarge this hole to reach the stamens.
Conclusion
The movements of honeybees between dwarf and tall plants are reduced only when plants also differ in male fertility. Altough this interaction, « male » and « female » parents differing in height can be intercrossed by honeybees. The main factor sufficiently ensuring flights between both parents is their similarity in the nectar production as it has yet been shown in other experiments with tall plants (Renard and Mesquida, 1987). Moreover, it has been demonstrated that semi dwarf genotype seeds can be produced using alternated designs with a male fertile or a male-sterile dwarf parent (submitted paper). In the case of the apetalous mutant, although honeybees are well attracted by this phenotype, these pollinators exhibit an individually specialised behaviour inducing a high reduction of their capacity to transfer pollen between petalous and apetalous flowers and reciprocally. Cleistogamy also induces an adapted behaviour but in a less clear-cut manner and the flower constancy of honeybees to these mutants is lower. Therefore, pollen dispersal by honey bees foraging on cleistogamous flowers to conventional rapeseed cannot be excluded.
References
Fray M. J., Evans E. J., Kelly A., Arthur A. E., Lydiate D. J., 1995.Evaluation of the effect of apetalous flowers and upright pods on seed yield using doubled haploid lines. Proc. 9th International Rapeseed Congress 1995, Cambridge, U.K., 497-499.
Maclean D. M. 1958. Role of dead flower parts in infection of certain crucifers by Sclerotinia sclerotiorum (Lib ;) D. Plant Disease Report 42, 663-666.
Mendham N. J., Rao M. S. S., Buzza G. C., 1991. The apetalous flower character as a component of a high yielding ideotype. Proc. 8th International Rapeseed Congress 1991, Saskatoon, Canada, 596-600.
Mesquida J., Marrileau R., Pham-Delègue M. H., Renard M., 1988. A study of rapeseed (Brassica napus L. var. oleifera Metzger) flower nectar secretions. Apidologie 19, 307-318.
Pierre J., Pierre J.S., Marilleau R., Pham-Delègue M.H., Tanguy X., Renard M., 1996. Inlfuence of the apetalous character in rape (Brassica napus) on the foraging behaviour of the honeybees (Apis mellifera). Plant Breeding 115, 484-487.
Renard M., Mesquida J., 1987. Significance of nectar secretion for honey bee foraging Apis mellifera and consequences on pollination in oilseed rape Brassica napus seed production. Proc. 7th International Rapeseed Congress 1987, Poznan, Poland, 222-227.
