Pyrethrum Flowers: A Quantitative Study Of Their Development

Abstract

Summary. An account is given of the examination of the flowers of pyrethrum plants (C. cinerariafoolium) grown upon a bed in Harpenden. The plants were divided into blocks and randomised, the flowers being harvested from a dozen plants each week over a period of 81/2 weeks, the flower heads ranged from the small bud stage in the first week to the overblown stage in the last week. The categories into which the flowers were divided to indicate maturity are defined and illustrated. The yield in numbers and weight of heads per plant, the diameters of the receptacles and the content of pyrethrin I and II were determined. There was a considerable amount of variation in all these factors in the flowers from different plants. A statistical analysis showed in this experiment: That there was no significant variation in the numbers of the flowers with time, but that position of the plant in the bed had a significant effect. That the time of harvesting had a significant effect upon the content of the pyrethrins whether taken separately or together and whether expressed in percentages, parts per flower head or parts per plant. The effect due to‐ position was only definitely significant in the case of pyrethrin I when expressed in parts per flower head. That there was on the different dates a significant correlation between the contents of pyrethrin I and II expressed in parts per flower head or plant. That there was, for the blocks of material examined, a significant correlation between the contents of pyrethrin I and II, whether expressed in percentages, parts‐per head or parts per plant, but that owing to the heterogeneity of the material the correlation was only significant for individual plants when the pyrethrias were expressed in parts per plant. The data in this experiment indicate that for the material examined, there is a quantitative development of the active principles in the flower heads from the small bud stage up to the time of maturity of the flowers, which more than keeps pace on the whole with the increase in weight of the flowers. Thus the content of pyrethrins, both relatively and absolutely, rises to a maximum at the maturity of the flowers. The mean percentage content of pyrethrins fell after pollination and the fading of the flowers; this corresponds with the rapid increase in weight of the heads on the formation of seed. There would thus appear to be a loss, which might be serious, both in percentage content of active principles and in yield of flowers if harvested before being fully open. There seems to be no useful purpose served in leaving the flowers to the overblown condition. I wish to express my great indebtedness to Dr R. A. Fisher and Dr J. Wishart for carrying out the greater part of the statistica.1 analyses incorporated in this paper and for much helpful advice. I am also indebted to Mr J. T. Martin for checking many of the analyses by an alternative method.