At the time of writing, 1949, field work on mosquitoes has been carried on in Bwamba County, western Uganda, for a period of ten years, with very few intermissions. The topography and vegetation of this area have been described elsewhere (Haddow, 1945a). The main effort has centred round the search for forest vectors of yellow fever, but much work has also been done in native plantations, huts, etc. While genera such asAëdes and Taeniorhynchus have, inevitably, required special attention, records of all mosquitoes encountered have been kept throughout the investigation, whether or not they seemed likely to be involved in the local epidemiology of yellow fever. Certain sections of the work have already been published (Mahaffy & others, 1942; Haddow, 1945a, 1945b, 1945c, 1946 and 1948; Haddow & others, 1947; Haddow & Dick, 1948; Haddow & Mahaffy, 1949), but a great part of it has not as yet been submitted.

Experimental huts fitted with window traps, of the same basic design as those described by Thomson (1948), modified for Malayan conditions, were used in studying the behaviour and mortality of A. maculatus in relation to DDT and BHC. Other species of Anophelines formed a large proportion of the human bait trap catch but rarely entered the huts.

In untreated huts, A. maculatus fed at all hours of the night with a peak between 9 p.m. and midnight. The large majority left before 8 a.m. in search of outdoor resting places, at least 50 per cent. before dawn. It appeared that attraction to light was the strongest, but not the only factor influencing exit from the huts.

Large numbers of Culicines (mostly Culex fatigans) were recovered. In untreated huts about 80 per cent. were blood fed ; 28 per cent. were in the window traps.

DDT (33 per cent.) and BHC (“ Gammexane P530 and P520 ”) water dispersible powders were tested.

An application of 100 mg. DDT and 20 mg. γ BHC (P530) per sq. ft. was unsuccessful, due in part to the small numbers of mosquitos recovered. No conclusions are drawn from the results.

DDT at 200 mg. and BHC (P520) at 40 mg. γ isomer per sq. ft. were both effective for at least 12 to 16 weeks against A. maculatus. With DDT, 63 per cent. initial, and 83 per cent. 24-hour corrected mortalities were recorded over 16 weeks. With BHC, though the initial mortality dropped from 100 per cent. in weeks 1–4 to 19 per cent. in weeks 12–16, the 24-hour mortality, almost 100 per cent. for weeks 1–10, was still 88 per cent. for weeks 12–16.

DDT apparently had little effect on the feeding of A. maculatus ; with BHC there was a reduction in the percentage fed.

DDT was relatively non-lethal to C. fatigans, but had a marked irritant effect driving them into the light trap and the biting rate was reduce to 45 per cent. BHC killed all C. fatigans which entered for 2 weeks but most of its toxic effects were lost by the eighth week. The biting rate was reduced to 27 per cent. for weeks 1 to 8 and the majority of mosquitos were recovered in the light trap.

Ants, cockroaches, spiders, etc., were found to affect the number of mosquitos recovered and special precautions had to be taken against losses.

The results suggested that BHC when freshly applied acts as a repellent to mosquitos.

Mansonia uniformis was shown to be susceptible to both DDT and BHC residual deposits.

The biting habits of Chrysops centurionis Austen and its distribution in Uganda are discussed. This species is everywhere arboreal and mainly crepuscular. The absence of loiasis in man in Uganda is, it is suggested, due to differences between the biting habits of this species and those of the known vectors, C. silacea Austen and C. dimidiata Wulp, which are closely allied to C. centurionis.

The peculiar biting-cycle of another Tabanid, Haematopota nefanda Edw., is described briefly.

Short notes on the biting habits and distribution of some other Uganda Tabanids are given.

The genus Diparopsis Hmps. is confined to Africa. During the present century, an insect always identified hitherto as Diparopsis castanea Hmps. has become a serious pest as a bollworm of cotton in many parts of that continent. Until recently, the genus was thought to consist of only two species, D. castanea Hmps. and D. tephragramma B.-Bkr., but differences in the structure of the frontal process of the adult suggested that the former comprised two or more distinct sub-species or species. The present study was undertaken to determine the status of these different forms. It is based on the material in the collections of the British Museum (Natural History) and on additional material obtained through the courtesy of the Government Entomologist, Anglo-Egyptian Sudan, and the Commonwealth Institute of Entomology.

Weekly observations on two populations of Culicoides impunctatus Goet. were made in 1948 at Bannachra and Wester Bannachra, near Loch Lomond, throughout the whole of the C. impunctatus season by means of “ sticky ” traps. Each site had an area of 5 acres of which half was composed of woodland and half of moorland. Twenty-two catching stations, each consisting of three traps placed 2 ft., 6 ft., and 10 ft. above the ground were erected on each site, equal numbers being placed in the woodland and moorland. There were therefore 132 traps to be changed weekly and a total of 67,033 Culicoides were trapped of which 65,576 (97·9 per cent.) were C. impunctatus. In 1949 observations were repeated only in Bannachra woodland.

The results have been analysed in the light of two alternatives (1) that C. impunctatus is a single biological entity and (2) that C. impunctatus is composed of two biological races.

When the seasonal effect was eliminated the stations showed very real differences between their weekly catches. Some stations recorded a large catch every week whilst others always recorded low catches. The total seasonal catch at each station was plotted on plans of the two sites and it then became apparent that there were two centres of high C. impunctatus density at Bannachra, one in the woodland and the other on the moorland, whilst at Wester Bannachra there were two foci in the woodland and a very diffuse zone of high midge density on the moorland. It has been shown that the centre of high midge density in Bannachra woodland was associated with a concentrated breeding place which supplied all the adults for the woodland (2½ acres) and moreover this centre remained unaltered in 1949. The importance of these observations on the control of C. impunctatus is discussed.

After the breeding place in Bannachra woods was discovered it became clear that the majority of adult C. impunctatus moved in an eastward direction from the breeding place and very few went westwards. This was attributed to the influence of the prevailing SW. wind. Knowing the centre of breeding in Bannachra woods it was possible to calculate the regression of the midge density on the distance from the breeding site. It was found that there was a linear relationship between the logarithm of the catch and the distance from the breeding site. A useful statistic is the reciprocal of the regression which may be designated as the density coefficient. It is the distance at which the density has decreased to l/10th of its initial value. The density coefficients for male and female C. impunctatus are 62·5 yards and 64·9 yards respectively. An attempt has been made to separate the effects of “ dilution ” and “ mortality ” on the density of C. impunctatus.

The mortality of male and female C. impunctatus is shown to be normally distributed with respect to the logarithm of the distance flown. These distributions are male l·867±0·197 (=73·6 yds. ; 46·8–115·9 yds.) and female 1 ·885±0·197 (=76·7 yds. ; 48·8–120·8 yds.).

From the density/distance formula the average distances flown by male and female C. impunctatus have been calculated. They are 79·0 yards (males) and 81·4 yards (females).

The analysis of the vertical distribution of C, impunctatus revealed that each station had its own definite pattern. The stations sited amidst the birch saplings showed very real differences between themselves which were not related to the ground flora. It is suggested that these are related to the height of the leaf canopy. Evidence was also produced which pointed to the fact that in the oak wood where the base of the leaf canopy was about 20 ft. from the ground, C. impunctatus had a self-imposed “ ceiling ” of little more than 10 ft.

The ground flora was found to have an appreciable effect upon the vertical distribution of C. impunctatus on the moorland. Thus where the ground flora provided only very shallow shelter, i.e., about 6 ins. in depth, equal numbers of adults were captured at each height, but when the stations were sited amongst Juncus sp. or in small sheltered clearings in the herb layer then more adults were caught on the 2-ft. trap. However, the shelter provided by bracken Pteridium aquilinum appeared to be less attractive and there were fewer adults caught on the 2-ft. traps at stations sited in dense bracken than at other moorland stations.

In the woodland the females were slightly more abundant on the 2-ft. trap and the males more numerous on the 10-ft. trap but the differences in distribution, although significant, were small. On the moorland the males were more abundant at 2 ft. and the females at 6 ft.

Although the general pattern was very similar there were highly significant differences between the horizontal and vertical distributions of the two races of C. impunctatus.

The development of Ptinus fur (L.) and P. sexpunctatus Panzer at 70 per cent. R.H. and several temperatures has been studied.

The optimum temperature for rapid development of P. fur is about 23°C., at which temperature it completes its development in a mean period of 132·1 days on fishmeal. For P. sexpunctatus the optimum temperature for rapid development is slightly below 30°C., development being completed in approximately 111 days on wheatfeed at this temperature.

Eggs of P. sexpunctatus, but not those of P. fur, hatch at 30°C., which is near the maximum temperature for development of P. fur.

Larvae of P. fur normally moult three times at 23°C. on fishmeal, but some have an extra moult.

Well defined diapause as mature larvae in cocoons occurs at low temperatures in some individuals of P. fur ; at 23°C. this lasts about 220 days after normal larvae have pupated ; the period at 20°C. is about 280 days. Diapause at this stage also occurs in P. sexpunctatus. In experiments, the development periods of the diapausing and non-diapausing larvae overlapped to form a continuous series.

Some P. sexpunctatus pass through a period of quiescence as mature adults in cocoons ; at 23°C., this lasted 43·5 days on fishmeal and 103·0 days on wheatfeed (mean periods). In P. fur the pre-emergence period is 30 to 60 days, considerably longer than that of other warehouse Ptinids studied.

Of the foods, wheatfeed allows the quickest development for both species, but fishmeal produces the heaviest beetles.

The adult beetles live for several months. They require food and drinking water. Egg output is small, the mean number of eggs laid per female at 25°C. being 38·6 for P. fur and 21·3 for P. sexpunctatus.

P. sexpunctatus was able to overwinter in an unheated outbuilding and could possibly become a pest on some foods.

P. fur is unlikely to occur in large numbers, owing to its long development period, small egg output and tendency to larval diapause.

Field trials in British Somaliland and in the Anglo-Egyptian Sudan showed that dry bran bait can be used for complete destruction of Desert Locust hoppers (Schistocerca gregaha). Wet bait was never found to be completely effective when dry bait was ineffective.

Failures of dry baiting were found which were attributable to several causes : hoppers not feeding just before moulting ; ground too hot for hoppers to remain on it ; too low or too patchy a deposit of bait ; adult locusts probably not making contact with bait. None of these failures could be attributed to the bait being dry.

The main constituent of standard bait, wheat bran, should not be replaced by barley husks or peanut flour, but dom flour could be used and dura and maize flours could be tried instead of bran.

The use of dry bait makes it convenient and practical for locust scouts who use camels, vehicles, etc. to destroy small bands of hoppers as soon as they find them.

Dry baiting against marching bands and ring baiting can be done very conveniently and speedily by means of the Shendi Funnel, designed to be tied to the back of a vehicle.

Area baiting can be done rapidly by blowing dry bait up from a dusting machine, so that the wind spreads it. An area-dosage of 25 lb./acre is recommended, and a rate of ½-acre/minute should be readily achieved.

The amount of labour required is reduced to about 25% by the use of dry bait instead of wet. Dry baiting by machine uses only 10 per cent. of the labour required for dry-baiting by hand, but the cost would not fall as much as this, because higher wages would have to be paid for the greater skill necessary.

Adults can be killed by baiting, but it is much more difficult than with hoppers to ensure that the locusts make contact with the bait.

The insecticides already in use against the Desert Locust, arsenic and benzene hexachloride, can completely destroy hoppers when applied in bait. An insecticide is required with a long persistence like that of arsenic and a low mammalian toxicity like that of BHC.

In 1932, Clark investigated the exotic and indigenous insect relationships with Pinus radiata in New Zealand, recording six species in the former group and 19 in the latter. Three previously unrecorded insect associations are now added to this list.

Laboratory experiments showed that piperonyl butoxide (PB) is a powerful synergist of pyrethrins in Shell oil P31.

The toxicity of PB alone to different species of insects was determined. Tribolium castaneum was resistant to direct sprays of PB in P31, but appreciably susceptible to films : the toxic action of the films on this species was unusually delayed. Ptinus tedus and Calandragranaria were resistant to films, but C. oryzae somewhat susceptible.

In view of claims that PB makes pyrethrum films more persistent, an experiment specifically designed to reveal such an effect was carried out. This failed to show that PB makes the films any more persistent.

A formula was obtained giving the approximate pyrethrin equivalent of different pyrethrin-PB mixtures applied to T. castaneum as films on filter paper. It was shown, however, that the pyrethrin equivalent of a given pyrethrin-PB mixture could differ according to the method of application and according to the species of test insect.

An experiment is described in which three light traps were tested for relative efficiency by having three locations in a wood and moving the traps each day so that over a period of three nights each trap had been once in each position. The actual number of moths caught were converted to a logarithmic scale and the mean log. (i.e., log. of the geometric mean catch) was used for comparison. By analysis of variance it is possible to show what portion of the differences between the catches is due to the difference between the locations, to the difference between the nights, and the difference between the traps.

A brief historical account is given of the spread of the Eucalyptus Snout Beetle, Gonipterus scutellatus Gyll., fron Australia, its country of origin, and of the utilisation of its natural parasite Anaphoidea nitens Gir. for its control on the African mainland.

The discovery of G. scutellatus in Mauritius is described. Its entry was effected between 1935–40, and by 1944 it was threatening Eucalyptus plantations over the whole island.

A. nitens was introduced in 1946 from East Africa. Establishment was extremely rapid and two years later G. scutellatus was effectively controlled.

Notes are given on the bionomics of both host and parasite under Mauritian conditions.

Ecological data are briefly discussed.

Tests of the acceptability of crushed dura (Sorghum millet) of various particle size composition suggested only small differences between any of them and wheaten bran. There seemed to be a slight advantage to dura passing a sieve of 20 meshes to the inch and stopped by 30, and a slight disadvantage to dura stopped by a 20-mesh sieve. There is certainly no reason to use wheaten bran in preference to crushed dura.

Maize bran containing 2·5 per cent. sodium arsenite was less acceptable than the other materials, but crushed maize was not. Nevertheless, this arsenic bait and also coarse dura with BHC produced a mortality apparently no lower than the most acceptable baits containing 0·5 per cent. BHC.

It is suggested that experiments with milling and separating machinery may produce a first-class dura bait and at the same time reserve the flour for human consumption.

The Citrus Leaf-miner (Phyllocnistis citrella) has been recorded from plants belonging to several widely differing botanical families.

It is shown that the true food-plants of this pest are the various species of Citrus and Aegle marmelos.

Elephant lemon (Citrus medica), which has comparatively succulent leaves, is the most susceptible species of Citrus, whilst lime (C. aurantiifolia), with its comparatively thick and coarse leaves, is the least susceptible. Aegle marmelos is not of any importance as the attack in this case is negligible.

C. medica should be excluded from citrus orchards as it is of least economic importance, and is highly susceptible to attack. C. limonia, which is usually grown as a hedge around gardens, provides an undisturbed breeding ground for the pest and should as far as possible not be planted around citrus plantations.

All varieties of sweet corn, in the seedling stage, are susceptible to frit fly attacks in England.

Different open-pollinated, hybrid and inbred varieties differ in susceptibility to damage by frit, indicating a possible genetical basis for resistance. Some breeders have produced less susceptible types than others.

Susceptibility is not associated with adaptability or maturity (i.e., comparative earliness when whole rows were ready for picking) of varieties but there is indication that seed origin is of some importance.

Sowing sweet corn at different dates influences the severity of attack. Peaks vary according to variety and year but generally sowings late in May are preferable.

Distribution of attack on a crop is at random, although there is a suggestion that plants in stringers from a central mass are mere damaged in the direction of the prevailing wind but there is no regular decrease in incidence with distance from the central mass.

Methods of tackling the frit fly problem include growing sweet corn in frit free areas, timing sowing dates, using prophylactic measures and breeding more resistant strains on genetical principles.

All stages of Tineola bisselliella were subjected to temperatures between 30°C. and 43°C. for periods varying between a few hours and a number of days.

Although humidity was not an important factor, it was controlled at 70 per cent., and various other humidities were used at some of the lower temperatures.

Complete mortality occurred at relatively low temperatures, 41°C. for four hours being lethal to all stages of the insect.

The eggs appear to be the most resistant stage. Some are able to hatch at 35°C. and some to survive exposure of four hours at 40°C.

Development through the entire life-cycle is possible at 33°C.

Characters are given whereby the gall midges attacking St. John's Wort (Hypericum), a noxious weed, may be distinguished.

Four species, Dasyneura hyperici (Bremi), D. serotina (Winnertz), Macrolabis marteli Kieffer and Zeuxidiplosis giardi (Kieffer), inhabit leaf galls in Europe.

One species, Dasyneura toweri Felt, inhabits the flowers in the U.S.A. Three species, Contarinia hyperici, sp. n., Clinodiplosis hyperici, sp. n. and Lestodiplosis (Coprodiplosis) sp., live in the unopened flowers in Europe.

Two species, Lasioptera virginica Felt and Cecidomyia ? triadenii Beutenmüller, are recorded as attacking the stems in the U.S.A. The latter is a name given merely to a gall.

Four species, Geocrypta braueri (Handlirsch), G. hypericina (Tavares) (perhaps a synonym of G. braueri), Trisopsis hyperici Tavares and Lestodiplosis (Coprodiplosis) hyperici (Tavares), live in underground galls in Europe.

Descriptions of Contarinia hyperici, sp. n. and Clinodiplosis hyperici, sp. n. are given.

Field and laboratory studies were made of the mosquitos of the Wuhan area, central China, between June 1947 and March 1948.

The following species were collected : Anopheles hyrcanus var. sinensis, Mansonia uniformis, Armigeres obturbans, Aëdes albopictus, Culex fuscanus, Culex vagans, Culex fatigans, Culex pipiens and Culex tritaeniorhynchus. A brief account is given of the topography and other environmental features of Wuhan.

The seasonal incidence and feeding habits of the adults, and the breeding sites favoured by the various species are recorded.

Monthly mean figures for rainfall, temperature and humidity are examined in relation to incidence of malaria and abundance of adults and larvae of Anopheles hyrcanus var. sinensis. The peak of malaria incidence coincides with that of Anopheles population which occurs two to three months after the peak of rainfall at an average relative humidity of 66 per cent.

A chemical analysis of waters from different types of breeding sites was made. Both the total organic nitrogen, and the amount of dissolved oxygen were considerably greater in the breeding site of Armigeres obturbans than in those of other genera. A sample of water from a site in which no mosquito breeding was observed contained very much smaller amounts of organic substances and minerals than samples from known breeding sites.

Other ecological factors affecting the early stages in their breeding sites are briefly referred to.

The habits and habitations of the natives in the Bwamba lowlands, Uganda, are briefly described.

A series of catches designed to assess the relative importance of the various species of biting insect and of various routes of transference in the transmission of yellow fever virus from the monkey population of the forest to the nearby human population is reported.

The microclimate in a native hut is discussed.

The species of biting insect encountered are listed, the results of the catches given, and the relative importance of the different species and routes assessed by means of indices.

It is concluded that yellow fever virus is most likely to be transferred to the native population by A. (S.) simpsoni in plantations, the infection in this insect being derived from forest monkeys which raid the plantations for food.

During a survey for whitegrub parasites, six Thynnid species were found in parts of western Patagonia and studied in varying degrees of detail.

Definite host records are established for Elaphroptera scoliaeformis (Hal.), E. (Paralycus) sp. near nigripennis (Smith) and Pseudelaphroptera brevipilosa Durán, which are solitary external parasites of the mature larvae of Aulacopalpus pilicollis (Fairm.) (Rutelinae), Macrosoma glacialis (F.) (Melolonthinae) and Adioristus acuminatus Hust. (Curculionidae), respectively.

No adult feeding was observed during the mating flight of E. scoliaeformis and E. sp. near nigripennis, but pairs of P. brevipilosa and Ornepetes semicincta Turner fed extensively on flower heads of Umbelliferae. Some evidence is presented suggesting that at the conclusion of the mating flight the winged males tend to return the apterous females to the sites where they picked them up.

Hosts are permanently paralysed and malaxated before eggs are laid on them. The eggs of the two Elaphroptera species are placed parallel to the longitudinal axis of the host, on the mid-ventral line of abdominal segments two to seven. In the case of P. brevipilosa, the egg is lightly cemented by its caudal end at an angle of about 45 degrees to the host surface, the attachment being near the spiracular line of the host and between the third thoracic and sixth abdominal segments.

There are five larval instars. The larvae do not normally move over the host surface. In the first four instars feeding is by suction, but the fifth-instar larva tears a hole in the host in the region of the earlier feeding punctures. The film of liquid with which the larva is bathed is produced from its posterior extremity and is not exuded from the feeding puncture. Cocoons of two types may be recognised, corresponding to the sexes, the female type being the shorter and of more rigid construction.

E. scoliaeformis, E. sp. near nigripennis, and P. brevipilosa have a single generation a year. The period from egg deposition to cocoon formation is generally covered in three weeks, and at least nine months are spent in the cocoon as a mature larva or prepupa. Adults probably live up to two months.

The only natural enemies observed were a Bombyliid of the genus Acrophthalmyda and a Carabid predator, Cnemacanthus sp., attacking the cocoon stages of E. sp. near nigripennis. The adult of this Thynnid is also attacked by a mimicking Asilid fly, Lycomya germainii Big.

The function of the mating flight and the economic value of members of the Thynnidae as controlling agents for Scarabaeid pests are briefly discussed.

Pyrgomorphella arachidis, sp. n.

♂ (Type). Of medium size, slender, finely rugulose.

Antenna of 17 segments, shorter than head and pronotum together, thick, slightly flattened and scarcely broadened at the base. Fastigium of vertex long, parabolic, ratio of its length to the width between eyes, 1·8. On the lateral sides of head behind the eyes there is a series of callosities forming a longitudinal stripe, which continues on to the lateral lobe of pronotum.