No abstract

Many observers have recorded that populations of Aphids and Coccids, visited by ants for the honeydew they excrete, multiply more rapidly than those not attended by ants (Nixon, 1951;Flanders, 1951). Several explanations have been advanced: ants give different degrees of protection against the natural enemies of the Homoptera; the colonies are '' healthier '' because the ants remove exuviae and the sticky honeydew; the ants sometimes transport the Aphids or Coccids to favourable feeding sites where their nutrition is improved.Some Homoptera are unable to survive without attendant ants because they become immobilised and are killed in the accumulated honeydew in which fungal parasites of the insects may develop, and because of attacks by insect enemies (Way, 1954). Like many other Homoptera, Aphis fabae Scop, usually colonises its various host-plants very successfully without being attended by ants. This Aphid ejects its honeydew forcibly and easily and this, therefore, does not accumulate and kill the insects.A. fabae on beans is occasionally visited and attended by the ant, Lasius niger (L.), and such colonies are said to benefit from the consequent protection from insect enemies, and possibly from a direct effect which the ant has in stimulating the Aphid's excretion rate. Herzig (1937) believed that there is, therefore, an increased absorption of plant sap and consequently a higher rate of reproduction. El-Ziady & Kennedy (1956) concluded from experiments that, whether predators were present or not, the ant, L. niger, increased the rate of multiplication of A. fabae and delayed the production of winged forms, and they suggested that the ant might raise the "plane of nutrition" of the Aphids. They considered that the protection of the Aphids against attacks by insect enemies was perhaps of secondary importance.The object of the work described below was to assess the effect of the ant, L. niger, on the reproduction rate of A. fabae on broad beans (Vicia faba) in the absence of the Aphid's insect enemies under experimental conditions; and to reconsider the manner in which this effect is brought about. Methods.The experiments were carried out in a garden where natural nests of the ant, L. niger, were common under paving stones. Single bean plants in 5-inch flower pots were infested with A. fabae from a stock culture and were placed in large insect-proof cages in the garden close to the ants' nests. The cages (3 ft. x 2 ft. x 2 ft.) were of wood with the door and rear panel fitted with glass, the sides and top being of fine muslin. Some of the cages,, all of which were raised above ground on up-turned flower pots, were made ant-proof by smearing the insides and outsides of the supporting pots with a ring of tree-banding grease.Ants were established in the other cages in the following way before experiments started. Large numbers of ants were collected from under the paving stones with an aspirator attached to an electrically-driven vacuum pump, the

Small populations of Aphis fabae were wholly or partially protected by the ant Lasius niger which drove most predators away. Predators eliminated whole colonies of unprotected aphids or persistently restricted their numbers. Protection of the aphid against predators is probably more important than hitherto thought, and may be more important than the other ways in which ants affect the multiplication rate of A. fabae.

Changes in numbers of adult Coccinellids on nettles infested with Microlophium evansi (Theo.) and on three experimental bean plots infested with Aphis fabae Scop. are described for the spring and summer of 1952 at Rothamsted. Of the three common species, Adalia bipunctata (L.), Coccinella septempunctata L. and Propylea quatuordecimpunctata (L.), the first was always the most abundant.The overwintered Coccinellids produced two broods: one on nettles in the spring and one on beans in early summer.On bean plots, changes in numbers of the adult insects occurred in three phases: (1) overwintered ladybirds, dispersing from nettles and other places, accumulated on the beans in the early stages of the infestations by Aphis fabae; (2) ladybirds, which had developed from eggs laid by the overwintered insects on nettles, emigrated from nettles and accumulated on the beans when A. fabae populations were at their height; (3) ladybirds, developed from eggs laid on the beans also by the overwintered insects, reached the adult stage when the bean aphid infestations had finished. The rise and fall of the A. fabae populations are discussed.One of the bean plots had always a high population of Coceinellids, which was attributed to its situation between two of the nettle sites. The other two bean plots were remote from nettles and one was sheltered by trees and buildings; their coccinellid populations were much lower.

Despite the economic importance of aphid feeding, there are few satisfactory estimates of the rate of sap ingestion by aphids and apparently none has been made of the total amount of sap removed from a plant or of the rate of feeding of any aphid throughout its whole lifetime. Previous estimates of feeding rates have been made for 230 C. J. BANKS AND E. D. M. MACAULAY short periods only, although Mittler & Sylvester (1961) established the sap intake of two aphid species during their whole larval lives.We now describe our method and present data on the rates of feeding, growth and reproduction of Aphis fabae Scop. throughout its life on field beans (Vicia faba L.)under standardized experimental conditions. T H E E X P E R I M E N T A L C O N D I T I O N SMethod of estimating the feeding rate Of several methods used previously (reviewed by Auclair, 1963), only one is suitable for estimating the sap intake of an individual aphid continuously from birth to death. This is by adding the weights of honeydew excreted and of water lost by evaporation to the weight of plant material converted into aphid substance (by growth and reproduction) during a period of time; gases exchanged during respiration are neglected. Auclair (1959) used this method for estimating the food intake of larval Acyrthosiphon pisum (Harr.) on peas for 71 hr. and Mittler & Sylvester (1961) similarly compared the feeding rates of A. pisum and Therioaphis maculata (Buckton) on alfalfa during their larval lives. We also use this method.Direct weighing measures increases in body weight, the weight of larvae born and losses by evaporation. Excretion rate, the product of the number and size of the excreted droplets of honeydew, is strongly affected by wind and by changes in temperature and humidity (Mittler, 1958). Therefore, in the work now described, the