ENCLOSURE TO PART III. ) this latter being the altitude of the main area for cotton in Nyasaland. The climates of these two stations differed greatly ss regards heat intensity and he expected to find some difference in the correlated results. So far he had oniy had experience during one season and that at one station; the work was really only just being started. The observations taken were the growth of the plant from the ground, measured weekly; its flowering, measured daily; losses attributed to pests, measured daily; boil weight, etc. He was endeavouring to plot the results in curves. So far he had seen no natural shedding. Losses were due in the main to boll worm; in Nyasaland other forms of shedding were entirely obscured by that due to boil worm. His boll worm records were at present incomplete, but he hoped to carry out further observations during next year. He considered on present evidence there was a seasonal distribution of boll worm species in Nyasaland, although previous observations had been against this view. Tobacco was a host of American boll worm in Nyasaland. Natives found boll worm attack to be negligible between August and October and they were altering their planting date accordingly. A complication in Nyasaland was the attack of the American boll worm, following that of the Sudan boll worm. Isolated records were of very little use unless combined with agricultural records. It appeared to him necessary that correlation should start with the layout of experimental stations and that plant development, agricultural, entomological, and meteorological records should all be got together on the same basis. For the present season he was working out his records on seven day means. THE CHAIRMAN (Mr. Milligan) remarked on the difficulty of getting information of real value regarding the rainfall; totals and averages were of course always available, but the more important details were wanting. Mr. PARNELL considered that the records of maximum and minimum temperatures were not of such value as might be expected, owing to the rapidity with which temperatures changed. = What would be really valuable would be a thermograph record from which could be seen the amount of time during which the temperature was above or below a certain point. At this point the class of instruments required for records was discussed and the various members of the Section expressed their views on the types they themselves were using. Mr. PEREIRA said that at the present stage there were not many well equipped stations in Portuguese East Africa. Mr. CAMERON considered that wind records correlated with rainfall would be valuable, anyhow in South Rhodesia, although he agreed that wind records in many countries would probably not be of much value. As regards the question of records in the field, the Section desired to emphasise the importance of correlating plant records and meteorological data and considered it desirable to draw up a statement showing the minimum requirements in instruments for individual stations. Messrs. Ducker and Nye were asked to draw up a list of : — (a) .The minimum requirements for a cotton station. (b) Additional useful apparatus. 14%