Effect of bee pollination on yield and yield components of field bean

. The effect of bees on the yield of field bean (Vida faba L.) was studied in cage experiments in southern Finland duringtwo successive years. Comparisons were made between field cages with and without honeybees and open-pollinated uncaged controls. Effect of shad- ing was also considered. The yield in open plots, which were freely visited by pollinators, was considerably greater than in plots from which the insects were excluded. Freely visited plants and plants caged with bees produced about twice as many seeds as plants caged without bees, and the number of seeds per pod was also higher. The plants were tallest in cages without bees. In open plots, 44 % of the bee visitors were honeybees, 56 % being bumblebees. The higher number of seeds obtained with honeybee pollination indicates that honeybees are able topollinate the field bean flowers in the conditions prevalent in Finland. It is con-cluded that there is little value in introducing bee colonies in small field bean areas and places where bumblebees are fairly abundant. Honeybees may improve the yield in larger fields, where natural pollinators are often scarce.


Introduction
Field bean has been cultivated in Finland since the 16th century. At the moment its acreage is rather small, being only 70 to 100 hectares. In the recent years there has, however, been increasing interest in cultivating Present address: Vaasankatu 14 B 32, SF-00500 Helsinki, Finland the field bean, and it has been taken into the breeding programme (Hovinen 1988). The protein content of the seeds is relatively high, as is the content of lysine, an amino acid, which makes beans suitable for use in diets based mainly on lysine-deficient cereals (Clarke 1970). The field bean may also be useful as a break crop in farms with inten-45 JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND sive cereal production. It could be more widely grown as an important source of homeproduced feed protein if its yields were more stable. The variations in yield are great (e.g. Varis et ai. 1982), which may partly be caused by weather conditions and susceptibility to pests and diseases; yield variations may also result from problems with pollination (e.g. Lawes 1974).
Considerable yield increases caused by honeybees have been reported for the field bean (e.g. Scriven et al. 1961, Poulsen 1974), a plant where both self-pollination and crosspollination occur. The increases in yield, however, vary considerably in studies performed under different conditions, and according to Free & Williams (1977), the pollination requirements are well worth studying for each type of conditions. The aim of the present study was to clarify the effect of pollination by bees on the podding and yield of field bean in southern Finland.

Materials and methods
Field trials were carried out on the Viikki Experimental Farm of the University of Helsinki (61°12'N)in 1979 and1980 Nylon screen cages with an area of 2 m x 2 m and a height of 1.5 m were placed over the plants just before flowering. Randomized plot design with three replicates was used. The distance between plots was 4 m. The total field bean area was 800 square metres in 1979 and 500 square metres in 1980. In 1980 a field bean area of 3500 square metres was situated at 50 m distance from the experimental field.
In 1980 a further treatment was added in which the plants were covered by a cage with only two walls. The purpose of this treatment was to clarify the role of the possible shading effect of the cages.
For yield data, the total bean yields of the plots were weighed, and the numbers of pods and seeds per pod were counted from 4x5 plants per plot. To obtain the percentage of flowers producing pods, the number of flowers in these plants was counted twice a fortnight during the flowering period in 1979, but because this work was highly time-consuming, in 1980 the counting was based on the scores left by the flowers, and it was done immediately after the harvest. The yield data were analysed by the analysis of variance.
In 1980 the numbers of honeybees and bumblebees were calculated three times per day (between 9 and 10 a.m., 12 and 1 p.m., 3 and 4 p.m.) per 2 square metres in the centre of the field bean area. Only "positive" pollinators which visited the front of the flower for nectar and pollen were counted.

Results
Seed yields. In 1979 the yields from noncaged plots averaged 3600 kg per hectare and in 1980, when the yield ripened too early because of the drought, only 400 kg per hectare.
In both years the highest yields were obtained from the non-caged plots. In spite of the great differences in yield, the tendency between the plots caged with bees and those caged without bees, in favour of the former, was the same in both years (Table 1). In the second year the yield in shading cages was 68 % of that of the non-caged plots. The seed weight of field bean grown in the open field was greatest. Most of the flowers did not produce pods. The Pollinators. 43.6 % of the bee pollinators in the open plots were honeybees, the rest bumblebees. Long-tongued species, B. subterraneus (L.) and B. hortorum (L.) and to some extent also B. lucorum (L.), were the visitors.

Discussion
The yield in open plots freely visited by pollinators was more than twice as high as in plots from which insects were excluded. The level of bumblebee and honeybee activity in these plots was rather high to affect the yield through effective cross-pollination. The yield in plots caged with bees was higher than the yield in plots from which the insects were excluded. The difference was, however, not significant, which indicates that it is difficult to arrange natural conditions for bees in cages and to maintain the optimal amount of bees in them. The results are in accordance with those of Free (1966), who blaimed the detrimental effect of caging, which "is likely to minimize any differences in seed production," and suggested that the beneficial effect of bee pollination would probably be greater than implied by the results. The shading effect of the cages alone does not explain the differences, because the yield in shading cages was more than twice as great as in the cages with bees and several times greater than in the cages from which the pollinators were excluded.
The total area of field bean may also affect the results. In our experiment the total area of field bean was rather small allowing the bumblebees, as more effective pollinators, to achieve adequate pollination in the open plots. In large field bean areas the number of wild bees would most likely be smaller which would emphasize the role of honeybees. Bond and Pope (1974) noticed that crossing decreased as field size increased. In eastern England the critical size above which the number of wild insects would not be sufficient was, however, as high as between 12 and 32 ha.
There are different opinions about the effect of plant density on the crossing frequency of field bean (e.g. Fyfe 1954, Homola 1973).
The number of seeds per plant was the same in the bee cages as in the open plots, and it was twice as high as in plots without pollinators. The greater number of seeds per pod in bee cages and open plots is in accordance with the results of Free (1966) and Poulsen (1974).
Although the honeybee pollination resulted in higher numbers of pods and seeds than did self-pollination, the larger seed size in the latter compensated the difference in yield weight.