THE DETERMINATION OF THE SIZE OF PLOT SUITABLE FOR COTTON SPRAYING EXPERIMENTS IN THE SUDAN GEZIRA

Abstract

An insecticidal experiment in the season 1955‐56 investigating interplot effect—defined as the interaction of one plot with the insect population on an adjacent plot—is described. This effect is considered to operate in three possible ways: (a) spray drift; (b) insect (pest) movement; (c) insect (parasite and predator) movement.In the experiment, care was taken to minimize (a). The plot sizes were 10‐ or 20‐feddan fields within areas of 90 and 180 feddans, respectively.There was strong evidence that yields of fields on the boundaries between sprayed and unsprayed fields were affected by the treatments on the other side of the boundary. That is, sprayed fields yielded less when adjacent to unsprayed fields and unsprayed fields yielded more when adjacent to sprayed.When cotton fields of 10 feddans in areas of about 90 feddans were unsprayed and separated by 150 or 450 m. of unsprayed cotton from sprayed fields, Empoasca libyca de Berg. and bollworm infestations were not measurably changed, but the Hercothrips spp. infestations were significantly reduced and the Bemisia tabaci Genn. infestation slightly increased compared with those found on 10‐feddan fields within 90 feddans of completely unsprayed cotton, the changes in infestation corresponding to the distance from the sprayed cotton. Yields from such fields tended to be higher than those from within completely unsprayed cotton, although these differences were not significant.When fields of 20 feddans in areas of 180 feddans were unsprayed and separated by 150 or 450 m. of unsprayed cotton from sprayed fields of similar size, there were no apparent differences when their insect infestations or yields were compared with those from 20‐feddan fields within 180 feddans of completely unsprayed cotton.These results are adduced as evidence that interplot effect might have affected infestations and yields in 10‐feddan fields of unsprayed cotton at distances of 150 or 450 m. from sprayed cotton. No such interplot effect was apparent when 20‐feddan fields were used. It is considered that spray drift was unlikely to have caused these results which must, therefore, have been due to insect (pest, parasite and predator) movement.It is concluded that in experiments in the Sudan Gezira, where treatments affect the insect complex on cotton, it is dangerous to assume that a 10‐feddan field in a 90‐feddan area is independent of differently treated fields in that area. On the other hand, independence can probably be assumed in 20‐feddan fields in 180‐feddan areas of cotton, provided that the field is separated from its differently treated neighbour by at least 150 m.