The genus Africophilus Guign. was erected by Guignot (1947) to accommodate the species A. inopinatus from the Ivory Coast, West Africa. Unfortunately only the female is known. In subsequent years A. pauliani Legros (1950) and A. nesiotes Guignot (1951) were descriptionbed from Madagascar and A. jansei O-C (1957) from South Africa. The known distribution of A. nesiotes has since been extended outside Madagascar to West, East and Southern Africa. In this paper three new species are descriptionbed, two from Nigeria and one from Malawi.
Four known favoured host plants of Lygus vosseleri were considered as possible trap crops to protect cotton from this pest. In three successive years field trials were done using Cissus adenocaulis as the trap crop. When treated with insecticide to prevent the development of a Lygus population on it, this plant gave considerable protection to the cotton. For various reasons sorghum and Pseudarthria hookeri were considered to be unsuitable. Salvia coccinae was used in a small field trial for protection against nymphal damage. The results obtained did not warrant further investigation.
In previous papers (Robertson 1965 and 1967), details were given of methods of collection and maps were included showing collecting localities. In the present paper records are mostly confined to locality and month of capture, plant names only being included when relevant, as most of the Hymenoptera listed are not restricted in the plants they visit. Mention has been made of the species most commonly collected at the Mercury Vapour light trap in use at Ukiriguru and, as in Robertson (1965) the initials H. J. D. after a record indicates that the specimen was collected by Mr H. J. Disney during his stay at Ilonga Research Centre.
Biological control of the tsetse fly, Glossina spp., by its insect parasites has long been desired. To date, the parasite which has been bred most successfully in large numbers in the laboratory is Syntomosphyrum glossinae (Lamborn, 1925 and Chorley, 1929). However, extensive releases of this parasite by Nash (1933) and Lloyd & Jackson (Swynnerton, 1936) in East Africa proved unsuccessful, largely due to its inability to penetrate the soil to attack the tsetse puparia. Sporadic and unsuccessful attempts have been made at breeding dipterous parasites of the genus Thyridanthrax, which was descriptionbed by Taylor (1932) as ""most unadaptable to artificial conditions"".
The method descriptionbed by Gilbert (1964) for rearing human body lice, Pediculus humanus humanus L., may be summarized briefly as follows: Lice are kept on patches (4 sq cm) of black cotton corduroy cloth. The infested cloth is kept in special containers in cabinets at a constant temperature of 29 ï¿½Oï¿½ 3ï¿½C and a relative humidity of 40ï¿½5 per cent. Rabbits are used as hosts: the whole ventral area of their bodies is closely shaved with an electric clipper.
It has long been customary for arachnologists to kill, fix and preserve specimens by dropping them into 70% ethyl alcohol. This approach yields distorted specimens which, once set, are often unsuitable for detailed taxonomic study. Many scorpions treated by this method bloat as a result of internal decay caused by inadequate preservative penetration (fig. 1b). Perfect study and display specimens (fig. 1c) may be obtained by employing the method descriptionbed below, which can also be adapted for the preparation of other invertebrates.