RECENT TRENDS IN RED CLOVER POLLINATION
tekst: B. Dennis, S. N. Holm
Pszczelnicze Zeszyty Naukowe, Rok XXI, 1977 — B. Dennis, S. N. Holm
strona 149BEE RESEARCH COPIES V.XXI DECEMBER 1977
B. Dennis and S. N. Holm^2
Department of Crop Husbandry and Plant Breeding Royal Veterinary and Agricultural University Copenhagen, Denmark.
ABSTRACT
The pollination of red clover seed crops is often inadequate and seed yields harvested may represent only a small fraction of the yields theoretically possible. The inadequacy of the pollinator population is even more pronounced in autotetraploid red clover which has greatly limited the use of this highly productive form. The full exploitation of the seed setting potential of both diploid and tetraploid forms will require a marked improvement in pollination conditions. While the conservation and propagation of the bumblebee population is necessary on a long-term basis, efforts must be made to improve pollination by honeybees, either by making red clover, particularly the tetraploid form, more attractive, or by inducing honeybees to forage in the crop. Recent research has indicated that this may be possible by utilizing the genetic variation existing in both plant and pollinator. Methods of improving the attractivity of the crop by plant breeding are reviewed, together with the possibility of exploiting heritable difference in pollen source preference as a means of increasing pollinator frequencies in red clover seed crops.
- Paper presented at the International Conference on Pollination of Red Clover held at Pulawy, Poland, 16th—18th August 1976.
- Authors’ address: Research Station Hojbakkegard, DK-2630 Taastrup, Denmark.
INTRODUCTION
The seed yields harvested from red clover crops often represent only a small fraction of the yields theoretically possible (Akerberg & Stapel 1966, Gikić 1973). Dennis & Haas (1967a) concluded that poor pollination was responsible for an average yield reduction of 52% in diploid red clover under Danish conditions. The inadequacy of the pollination was found to be even more pronounced in the tetraploid form, where yield reductions of 69% were observed. This is generally ascribed to the increase in corolla tube length associated with the induction of polyploidy (Julén 1956, Dennis & Haas 1967b, Bond & Fyfe 1968b, Anderson 1973).
Under Northern European conditions, pollination of red clover is dependent to a large extent on the wild bee population, chiefly bumblebees. However, bumblebees are extremely unreliable as pollinators of red clover due to erratic and unpredictable fluctuations in numbers from year to year and place to place (Bingefors & Eskilsson 1962, Hawkins 1961, Hänninen 1962, Valle et al. 1960). Due to their inability to reach the nectar at the base of the corolla tube, honeybees are often unwilling to forage in red clover and prefer other crop or weed species (Fridén et al. 1962, Umaerus & Grazi 1962). It is important to note in this connection that the increasing areas of cruciferous crops are likely to present severe competition to red clover in terms of pollinator availability.
There can be little doubt that the full exploitation of seed yield potential in both diploid and tetraploid red clover, certainly in Scandinavia, will require a marked improvement in pollination conditions. Whilst the conservation and propagation of the wild bumblebee population is necessary on a long-term basis (Fridén 1967), efforts must be made to improve pollination by honeybees, either by making red clover, particularly the tetraploid form, more attractive, or by inducing honeybees to forage in the crop. Honeybees collect both nectar and pollen from red clover. Nectar is collected only from the shortest florets, except when collected from the holes bitten by bumblebees. Pollen is readily accessible for honeybees, but not attractive enough for many colonies. Rapid improvement in pollination is most likely to be obtained partly by improving the attractivity of the plant and partly by breeding honeybees for improved pollen foraging. The present paper will be confined to a discussion of the possibilities of exploiting the genetic variation that exists in both plant and pollinator.
Breeding honeybees for red clover pollen collection. Breeding work in honeybees has previously been concentrated on characters such as honey production, temperament, egg-laying ability, etc., of commercial interest to the beekeeper. In 1965 Nye & Mackensen in the US
strona 151demonstrated heritable differences in lucerne pollen collection between honeybee colonies. Subsequent breeding work has shown that quite considerable increases in the frequency of honeybee pollen collectors in lucerne seed fields could be obtained by selection (Nye & Mackensen 1968, 1970, Cale 1970). Very little information is available on the effect of inproved pollen collection in terms of increased seed yields, although in preliminary trials Johnson (1970) reported increases varying from none to 27%. The high and low APC (Alfalfa Pollen Collectors) lines developed by Nye & Mackensen in Utah were tested under Danish conditions in 1970 and 1971 (table 1). In the first year the high line collected almost entirely white clover pollen, whereas in 1971 a greater pollen collection was observed in red clover and other plants. On August 19th in the latter year about 80% of all bees in a lucerne seed field collected pollen, and samples taken from pollen traps showed this activity to be confined to the colonies of the high APC lines. These results demonstrate that a preference for lucerne pollen collection can be achieved in areas with other legumes such as red clover. At the same time it can be seen from Table 1 that the differences between the high and low APC lines are expressed in their general pollen-collecting habits, and Mackensen and Tucker (1973) also concluded that the variation in choice of pollen source is expressed as a general preference rather than a specific preference for pollen from a certain crop.
In view of the scale of success achieved in lucerne, similar genetic differences in pollen collection might be expected to exist in red clover.
Table 1 Pollen collection (%) by colonies of local, high and low APC honeybee lines in Denmark
| Lucerne | Red clover | White clover | Other plants | |
|---|---|---|---|---|
| Vemmetofte 1970 | ||||
| Local bees | 0,2 | 8 | 84 | 8 |
| High line | 0,1 | 2 | 90 | 9 |
| Low line | 0,1 | 26 | 55 | 19 |
| Højbakkegård 1971 | ||||
| Local bees | 0,1 | 29 | 33 | 39 |
| High line | 3,1 | 41 | 25 | 32 |
| Low line | 0,6 | 24 | 3 | 74 |
| Højbakkegård 18th Aug. 1971 | ||||
| Local bees | 0 | 40 | — | 60 |
| High line | 47 | 19 | — | 34 |
| Low line | 0 | 4 | — | 96 |
In Denmark Poulsen (1973) and Holm (1975) have demonstrated significant differences in pollen collection between neighbouring honeybee colonies for other plant species including red clover, white clover and field beans. 35 colonies were tested and in cruciferae the percentage of pollen collected varied form 2 do 20, in red clover from 0 to 53%, and in white clover from 2 to 64%. The majority of the colonies collected between 5 and 10% of their pollen from cruciferae, between 5 and 35% from red clover, while in white clover the colonies were more evenly distributed and showed wider variation.
A prerequisite for selection for improved pollen collection is a knowledge of the variation in pollen preference, and the extent to which this variation is due to genetic or environmental effects. In Sweden Schwan (1976) has tested a number of colonies over a five-year period and has found considerable intercolony variation (table 2); in 1971, for example, the percentage of pollen collected from red clover ranged from 2.3 to 59.7. Variation of a similar magnitude (11—43%) was also reported by Holm (1975) in a Danish test comprising 10 honeybee colonies. Schwan (1975) was able to compare a number of colonies in which no queen change had occurred in two years, and found a reasonable high degree of correlation between years; on average over 25% of the variation in one year could be described by colony differences in the other year. Thus this variation was not random, but probably of a heritable nature.
In 1975 work started in Denmark and Sweden to establish whether differences in pollen collection between honeybee colonies are genetically controlled, and if so, to improve pollen collection by means of a suitable selection programme (table 3). From an initial population of unrelated colonies, the four best (A, B, C and D) have been selected and lines produced from each colony by sister/brother matings. Twelve inbred colonies of each line have been tested for pollen collection from red clover. Due to the common drone source, the intra-line variation will
Table 2. Frequency (%) of red clover pollen collectors in honeybee colonies 1969—73. (After Schwan 1976).
| Year | Number of colonies | Average | Range |
|---|---|---|---|
| 1969 1) | 21 | 27,6 | 8,7—48,6 |
| 1970 1) | 21 | 28,6 | 12,4—43,0 |
| 1971 1) | 19 | 33,6 | 2,3—59,7 |
| 1972 1) | 17 | 81,7 | 64,5—90,8 |
| 1973 1) | 10 | 45,8 | 17,9—71,4 |
| 1973 2) | 10 | 27,2 | 10,3—47,3 |
- diploid red clover. 2) tetraploid red clover.
Table 3 Proposed programme for breeding honeybees with preference for tetraploid red clover as pollen source in Scandinavia
| Year | Procedure | Material | |||
|---|---|---|---|---|---|
| 1 | Test | Initial population | |||
| 1 | Selected | A | B | C | D |
| Number of queens or colonies | |||||
| 1 | Inbreeding * | 15 | 15 | 15 | 15 |
| 2 | Test | 12 | 12 | 12 | 12 |
| 2 | Selected | A1A2A3 | B1B2 | C1 | D1 |
| 3 (B1 × A2) | |||||
| 2 | Crossing | 20 (C1 × A1) | 20 (D1 × A1) | ||
| 3 (B2 × A3) | |||||
| 3 | Test | 15 | 15 | ||
| 3 | Selected | 1 | 1 | ||
| 3 | Crossing | 20 ((C1 × A1) × (B1 × A2)) | |||
| 20 ((D1 × A1) × (B2 × A3)) | |||||
| 4 | Test | 15 | 15 |
- Sister × brother
express the variation among the heritable characters of the queens. Three, two or one colony will be selected according to the average of the line (A highest, D lowest). The lines will then be intercrossed in two steps. First queens from lines B, C and D will be mated to drones from line A. The best queens from C and D will be selected and their daughters (CxA, DxA) mated to drones from line B (BxA). In this way the final colonies will be founded by queens which will give worker offspring of the three-way hybrid type, i.e. (CxA)xB. Preliminary results (Holm & Schwan, unpublished) have shown an increase in mean red clover pollen collection from 16% to almost 48%. These results justify a certain degree of optimism. If successful, this project will not alone lead to the exploitation of the highly productive tetraploid red clover, but also inevitably to a stronger integration of honeybees in agricultural production and hence economy.
Breeding red clover for attractivity to honeybees. Information on genotypic differences in attractivity in red clover is scarce, but in lucerne differences in the attractiveness of clones and varieties have been reported by several authors (Pedersen & Nye 1962, Boren et al. 1962, Loper & Waller 1970). Kauffeld et al. (1969) found such differences in attractiveness to be stable in different locations, seasons and years. In red clover Bond & Fyfe (1968a) found that honeybees visiting F1 plants from a partial diallel cross were able to recognize plants having one parent in common.
Preferential honeybee visitation in red clover can be due to corolla
strona 154tube morphology, nectar volume and sugar concentration, flower colour, aroma, and pollen availability or any combination of these factors. Selection for shorter corollas an a means of improving attractivity has been attempted by several authors (Starling et al. 1950, Kress & Fuchs 1959), and Bond & Fyfe (1968a, b) found this character to be highly heritable in both diploid and tetraploid red clover. The short corolla tube character has, however, been found to be associated with lack of vigour and reduced agronomic performance. Dennis (1975) obtained several lines of evidence to suggest that there is probably no absolute relationship between vigour and corolla length, or between seed yields and corolla length. Path coefficient analyses showed negative direct and indirect effects of corolla length on seed yields, but these effects were extremely small compared to other components of seed yield. These results indicate that other floral attributes, possibly nectar production or flower aroma, are probably also of importance as determinants of plant attractivity. In lucerne Pedersen (1953), Pedersen & Bohart (1953), Wykes (1952), Kauffeld & Sorensen (1971) and others have shown nectar production and sugar concentration to be important attractiveness factors. In red clover Bond & Fyfe (1968a, b) reported genotypic differences and high degrees of heritability for nectar height and nectar depth in both diploid and tetraploid forms. Some of the variation in nectar height was independent of corolla length, and they concluded that it might be possible to improve by selection both the height of nectar and its accessibility.
Shuel (1955) concluded that nectar quantity and sugar concentration were heritable traits in red clover. He suggested, however, that the physiology of nectar secretion was influenced to such a large degree by factors such as light intensity, temperature, humidity, soil moisture and photoperiod that breeding for sugar concentration would not be practical. Walker et al. (1974) observed significant differences among lucerne clones for both quantity of nectar per floret and per cent total sugars in the nectar and concluded that simultaneous selection for increased sugar concentration and increased nectar quantity should be possible. Hawkins (1971) obtained a significant increase in nectar height from 1,5 mm to 2,5 mm after five generations of selection in English singlecut (late-flowering) red clover. This increase in nectar hight was not associated with a reduction in corolla length, but probably in nectar production as such. This would agree with the results of Jabloński (1972) who found a significant positive correlation between height of nectar in the tube and the sugar yield per flower.
Loper & Waller (1970) observed clonal preference by honeybees foraging in lucerne and noted that flower aroma might be the cause. By means of a bioassay procedure, Loper et al. (1974) were able to identify lucerne clones that were olfactorily distinct to the honeybee,
strona 155and concluded that this method might be of value in selecting parental clones in breeding work.
It would appear that several components of attractivity show genetic variation and may respond well to selection. Kauffeld et al. (1969) drew an interesting parallel to phytophagous insects which show and stable degrees of preference for oviposition or feeding, and in which selection for insect resistance, i.e. for non-attractiveness in this case, has been successful. In the case of red clover attractivity to honeybees, we still need considerable information on the components of attractivity, their interrelationships and the extent of genetic variation.
CONCLUSION
Sufficient genetic variation appears to exist in both the honeybee and in red clover to warrant intensive selection programmes aimed at improved pollen foraging activity and plant attractivity. The complex nature of the pollination problem will necessitate a close collaboration between entomologists and plant breeders, not only in terms of the mutual exchange of information at conferences such as this, but also on an even closer basis with the development of joint breeding projects.
LITERATURE
Akerberg, E. & C. Stapel 1966: A survey of pollination and seed growing of red clover in Europe. — Bee World 47 (1) Suppl.: 15—42.
Anderson, L. B. 1973: Relative performance of the late-flowering tetraploid red clover 'Grasslands 4706′, five diploid red clovers, and white clover. — N.Z. J. Exp. Agr. 1: 233—237.
Bingefors, S. & L. Eskilsson 1962: Pollination problems in tetraploid red clover in Central Sweden. — Z. Pflanzenzüchtg. 48: 205—214.
Bond, D. A. & J. L. Fyfe 1968a: Corolla tube length and nectar height of F1 red clover plants (Trifolium pratense) and their seed yield following honey-bee pollination. — J. agric. Sci., Camb. 70: 5—10.
Bond, D. A. & J. L. Fyfe 1968b: Variation between tetraploid red clover plants in corolla tube length and height of nectar. — J. agric. Sci., Camb. 71: 113—116.
Boren, R. B., R. L. Parker & E. L. Sorensen 1962: Foraging behavior of honeybees on selected alfalfa clones. — Crop Sci. 2: 185—188.
Cale, G. H. 1970: Breeding bees for alfalfa pollination. — The Indispensable Pollinators, Rep. of Ninth Pollination Conf., Hot Springs, Arkansas, 1970: 69—72.
Dennis, B. A. 1975: Seed-set and related problems in autotetraploid Trifolium pratense L. — Ph. D. thesis, University of Reading, 158 + X pp.
Dennis, B. A. & H. Haas 1967a: Pollination and seed-setting in diploid and tetraploid red clover (Trifolium pratense L.) under Danish conditions. I. Seed-
strona 156-setting in relation to the number and type of pollinating insects. — Arsskr. Kgl. Vet. & Landbohojsk. 1967: 93—117.
Dennis, B. A. & H. Haas 1967b: Pollination and seed-setting in diploid and tetraploid red clover (Trifolium pratense L.) under Danish conditions. II. Studies of floret morphology in relation to the working speed of honey — and bumble-bees. — Arsskr. Kgl. Vet. & Landbohojsk. 1967: 118—133.
Fridén, F. 1967: Humlorna och jordbruket. — Medd. Sverig. FröodlFörb. 8: 65—82.
Fridên, F., L. Eskilsson & S. Bingefors 1962: Bumblebees and red clover pollination in Central Sweden. — Proc. lst Int. Symp. on Pollination, Copenhagen 1960. Medd. Sverig. FröodlFörb. 7: 17—26.
Gikić, M. 1973: (The relationship between the calculated and actual seed yields of cultivars and ecotypes of red clover (Trifolium pratense var. sativum). — Poljoprivredna Znanstvena Smotra 30: 405—421. (Herbage Abstracts 45: 2149).
Hänninen, P. 1962: Bumblebee species on red clover in Central Finland. — Finn. agr. Res. Board, publ. 197, 21 pp.
Hawkins, R. P. 1961: Observations on the pollination of red clover by bees. 1. The yield in relation to the number and kinds of pollinators. — Ann. appl. Biol. 49: 55—65.
Hawkins, R. P. 1971: Selection for height of nectar in the corolla tube of English single-cut red clover. — J. agric. Sci., Camb. 77: 347—350.
Holm, S. N. 1975: Variation among honeybee colonies in pollen collection from different agricultural crops. — Proc. III Int. Symp. on Pollination 1974, Suppl. Bulletin Technique Apicole 2: 219—228.
Jabloński, B. 1972: (The height of the nectar column in a red clover flower and the quantity of nectar and of sugars contained in it). — Pszczelnicze Zeszyty Naukowe 16: 125—131. (Plant Breeding Abstracts 44: 4498).
Johnson, I. J. 1970: Effectiveness of Hy-queens as alfalfa pollinators. — The Indispensable Pollinators, Rep. of Ninth Pollination Conf., Hot Springs, Arkansas, 1970: 73—74.
Julén, G. 1956: Practical aspects on tetraploid red clover. — 7th Int. Grassld Congr., Palmerston North 1956: 471—478.
Kauffeld, N. M. & E. L. Sorensen 1971: Interrelations of honeybee preference of alfalfa clones and flower color, aroma, nectar volume, and sugar concentration. — Res. Publ. Kansas State Univ. Agric. Appl. Sci. 163: 14 pp.
Kauffeld, N. M., E. L. Sorensen & R. H. Painter 1969: Stability of attractiveness of alfalfa clones to honeybees under varying locations, seasons, and years. — Crop Sci. 9: 225—228.
Kress, H. & A. Fuchs 1959: Über den Züchtungswert von Rotkleeformen mit kurzer Blütenröhre. — Z. Pflanzenzüchtg. 41: 167—194.
Loper, G. M. & G. Waller 1970: Alfalfa flower aroma and flower selection by honey bees. — Crop Sci. 10: 66—68.
Loper, G. M., G. D. Waller & R. L. Berdel 1974: Olfactory screening of alfalfa clones for uniform honey bee selection. — Crop Sci. 14: 120—122.
Mackensen, O. & S. C. Tucker 1973: Preference for some other pollens shown by lines of honeybees selected for high and low alfalfa pollen collection. — J. apic. Res. 12: 187—190.
Nye, W. P. & O. Mackensen 1965: Preliminary report on selection and breeding of honeybees for alfalfa pollen collection. — J. apic. Res. 4 (1): 43—48.
Nye, W. P. & O. Mackensen 1968: Selective breeding of honeybees for alfalfa pollen: Fifth generation and backcrosses. — J. apic. Res 7 (1): 21—27.
Nye, W. P. & O. Mackensen 1970: Selective breeding of honeybees for alfalfa
strona 157pollen collection: with tests in high and low alfalfa pollen collection regions. — J. apic. Res. 9: 61—64.
Pedersen, M. W. 1953: Seed production in alfalfa as related to nectar production and honeybee visitation. — Bot. Gaz. 115: 129—138.
Pedersen, M. W. & G. E. Bohart 1953: Factors responsible for the attractiveness of various clones of alfalfa to pollen-collecting bumblebees. — Agron. J. 45: 548—551.
Pedersen, M. W. & W. P. Nye 1962: Alfalfa seed production studies. II. Additional factors associated with seed yields. — Utah State Univ. Agr. Exp. Sta. Bull. 436: 9—15.
Poulsen, M. H. 1973: Diurnal and Inter-colony Variation in Pollen Collection from Different Seed Crops by Honeybees (Apis mellifera L.). — Ent. scand. 4: 100—104.
Schwan, B. 1965: Foraedling af bier til bestovning of rodklover. — Nordisk Bitidsskrift 26: 32—35.
Shuel, R. W. 1955: Nectar secretion. — Amer. Bee J. 95: 229—234.
Starling, T. M., C. P. Wilsie & N. W. Gilbert 1950: Corolla tube length studies in red clover. — Agron. J. 42: 1—8.
Umaerus, M. & F. Grazi 1962: Observations on bumblebees and red clover pollination in southern Sweden in 1957—1960. — Proc. lst Int. Pollination Symposium, Copenhagen 1960. Medd. Sverig. FröodlFörbund 7: 86—94.
Valle, O., M. Salminen & E. Huokuna 1960: Pollination and seed-setting in tetraploid red clover in Finland. — Acta agral. Fenn. 97: 1—64.
Walker, A. K., D. K. Barnes & B. Furgula 1974: Genetic and environmental effects on quantity and quality of alfalfa nectar. — Crop Sci. 14: 235—238.
Wykes, G. R. 1952: Preferences of honey bees for solution of various sugars which occur in nectar. — J. Exp. Biol. 29: 511—518.
NAJNOWSZE TENDENCJE W BADANIACH ZAPYLANIA KONICZYNY CZERWONEJ
B. Dennis i S. N. Holm
Uniwersytet w Taatsrup, Dania
STRESZCZENIE
Zapylenie czerwonej koniczyny nasiennej często nie daje spodziewanych wyników, czego przyczyną jest struktura kwiatu tego gatunku. Zbiory prezentują często tylko część z tego, co możnaby uzyskać. Brak zapylaczy dla tetraploidalnej koniczyny ogranicza jej uprawę. Pełne wykorzystanie diploidalnej i tetraploidalnej koniczyny wymaga zabezpieczenia określonych warunków. Trzmiele nie bardzo wchodzą w rachubę ze względu na duże wahania ich populacji i na zmniejszanie się liczbowe gatunku. Nie zaniedbując metod zachowania i adaptacji różnych gatunków trzmieli, co powinno być działaniem długofalowym — należy więcej uwagi poświęcić zapylaniu przez pszczołę miodną. Szczególnie zainteresować się trzeba rasami oblatującymi koniczynę i to tetraploidalną. Najnowsze badania wykazują, że od strony genetycznej jest to możliwe. Atrakcyjność koniczyny dla pszczół może być zwiększona przez taką hodowlę, która zapewni zbieraczkom pyłku bogatsze jego źródło (dostępność) i to zwiększy oblot kwiatów tego gatunku.
Errata — błędy druku znalezione podczas digitalizacji
- s. 151: „inproved” → improved
- s. 152: „varied form 2 do 20″ → varied from 2 to 20
- s. 152: „reasonable high degree” → reasonably high degree
- s. 154: „corollas an a means” → corollas as a means
- s. 154: „nectar hight” → nectar height
- s. 156, literatura: „Fridên” → Fridén
- s. 156–157, literatura: „Proc. lst” → Proc. 1st
- s. 157, literatura: „bestovning of rodklover” → bestovning af rodklover
- s. 157, streszczenie: „Taatsrup” → Taastrup
