Since Location II seems to have a richer insect fauna than Location
I, the higher percentage of contaminated plants found at Location I this
year was quite unexpected. Since no insecticide was sprayed on Location
II in 1984, this perhaps should have allowed a further increase in the
percentages of contaminated plants and seeds. It did not. Yet, because
the field was not sprayed, seed production was greatly reduced as a re-
sult of a high incidence of disease, particularly viral disease. This
explains the difference in the average seed production per plant between
the two fields (46.6 seeds per plant on Location I vs. 28.2 seeds per
plant on Location II).
Cross fertilization rates were also determined using WL 145 (long,
late, non-fasciated, green cotyledons). Table 3 shows the results for
Location I and Table 4 for Location II.
Usually 20 seeds were sown for each control group in both fields,
so the loss of plants was especially high in this line. The difference
between the average seed production per plant (45.8 seeds per plant on
Location I and 38.9 on Location II) was not as great as in the case of
WL 6040. The percentage of contaminated plants and seeds was greater in
Location II than Location I but the high values on Location II were
mainly due to the surprisingly high rates of cross fertilizations of
control groups 9 and 10 and for which there is no satisfactory expla-
nation. Having watched (and filmed) the different kinds of bumble bees
and a solitary bee species for many hours, I noted that these bees do
not seem to have the refined information systems that the honey bee has
for communicating the most productive parts of a field. Moreover,
bumble bee behavior is often rather erratic: leaving a red pea flower,
turning to different plant species of the meadow nearby, returning to
the pea field, flying from white to red flowers and vice versa, some-
times crossing the entire field and beginning its activity anew at the
most distant part of the field, etc.
Besides the two lines mentioned, WL 118 (medium height, as early as
DGV NF 15-1, non-fasciated, green cotyledons) also was investigated.
The control group, consisting of 36 plants, had two contaminated plants
(=5.6%) and produced 1682 seeds, two of which had yellow cotyledons
(=0.12%).
Moreover, a few recombinants were selected, resulting from cross
fertilizations between WL 6040 and probably DGV. They were in most res-
pects similar to DGV (height, non-fasciation, flowering and ripening
times, etc.) except, of course, for the green cotyledons (Table 5).
