Contamination in seed crops

A NUMBER of results are now available from systematic experiments on natural cross-pollination between varieties of crops grown for seed, e.g. on radish by Crane and Mather (i); on radish and turnip, using various planting arrangements by Bateman (z 947a) ; and on the wind-pollinated crops, beet and maize, by Bateman (1947b). These data are in a consistent form which makes comparison and generalisation possible. The conclusion that may be drawn from these experiments is that whatever the absolute level of contamination or the range of distance involved in any experiment, the shape of the curve relating contamination to isolation distance is the same. It is such that the rate of decrease of contamination per unit increase of isolation distance itself decreases with that increase. At first, increases in isolation distance rapidly reduce contamination, but at greater distances the contamination, though small, becomes persistent and in one instance (Bateman, i 947a) there was no detectable decrease in contamination when the isolation distance was increased from 50 yards to 200 yards. The consistency of these results justifies the present attempt to derive a general mathematical expression which, by the substitution of appropriate values for the constants, could express the variation of contamination with distance under any conditions and for any crop. In the following sections (the first of which contains new observations on the pollinating methods of bees) the argument is roughly divisible into three parts :—