Heliothis armigera (Hb.) has been shown to develop continuously in tropical areas but to become more dependent with increasing latitude on a pupal diapause to survive the winter. The changes that occur in pupal period under ambient conditions in Botswana were examined. Pupae formed in May were able to survive the winter period and produce adults in September. Laboratory work using artificial environments showed that this extended pupal period is not simply a response to low pupal temperatures but is induced by the low temperatures and short daylengths experienced by the larva. Pupae from larvae reared at 27° C and a 14-h daylength (non-diapause pupae) initiated adult development even at 13° C. Pupae from larvae reared at 18° C and a 9-h daylength (diapause pupae) had similar pupal periods to non-diapause pupae at 33 and 27° C. However, at 18 and 13° C, diapause pupae had much longer developmental periods than non-diapause pupae. More diapause pupae were produced at 18° C from larvae reared in a 12-h daylength than from those in 15- or 9-h daylengths. The variation and importance of this temperature-labile facultative diapause over the extensive geographical range of H. armigera is discussed.

Night-time observations on adults of Heliothis armigera (Hb.), and data from light-trap and hand net catches in irrigated and dryland crops, showed that both sexes flew and fed and females oviposited between 20,00 and 23.00 h. From then until 02.00 h both sexes were inactive, but. from 02.00 to 04.00 h the males flew above the crop in a ' purposeful' manner, while the females were stationary and releasing pheromone. During this period of high male and low female activity, copulation in cages, assembly of males to females in cages and copulating pairs on crop plants were all observed. Inseminated female H. armigera were collected in crops which were at a stage suitable for oviposition, while traps away from crops or near mature crops collected mainly virgin females. It is suggested that susceptible crops attract H. armigera adults and that, once within the crop, inseminated females are ' trapped' by suitable physiological cues from the plants. Identification of such cues could assist in the breeding of crop varieties resistant to H. armigera attack.

Observations of oviposition behaviour by Chilo partellus revealed that antennae, ovipositor tip and possibly tarsi were all involved in the choice of oviposition site. Among a very large number of typical mechanoreceptor hairs on the tip of the ovipositor valves, two pairs of stouter, blunt‐tipped hairs were discovered. These hairs alone took up silver nitrate stain and transmission electron microscope sections revealed that the hair shaft had a hollow lumen within which ran a group of four or five dendrites. It is concluded that these sensilla are contact chemoreceptors and they are positioned in such a way that, with the placing of each egg, they would contact the oviposition substrate. Similar hairs were found on the ovipositor of Spodoptera littoralis. It is suggested that the hairs may prevent oviposition on surfaces directly chemically harmful to the eggs.

Six trials with an unpurified suspension of Heliothis armigera nuclear polyhedrosis virus (NPV) against Heliothis armigera (Hb.) larvae on sorghum (five trials) and on cotton (one trial) during 1971-73 are described. Commercial preparations of H. zea NPV and of Bacillus thwingiensis were included in two of the trials. H. armigera NPV was as successful as a standard insecticide in preventing losses of sorghum in one trial. Lowering the rate of NPV reduced the control of H. armigera larvae. There was no difference in control achieved by spraying in the morning or in the evening. Spraying every second or every third row (relying on natural spread of the< virus) reduced control. Low concentrations of molasses (0-6%) made control by 100 larval equivalents (L.E.) per hectare as good as that by 200 L.E. per hectare without molasses. Neither the commercial H. zea NPV nor the Bacillus preparation was as effective as the local virus. The local virus was reasonably effective against H. armigera on cotton at very high rates and with molasses. It was not as damaging as carbaryl to populations of egg parasites and predators, but the predators did not control damaging populations of aphid and Jassid on the virus-sprayed plots.

No abstract