In order to seek at least a partial explanation of the differential occurrence of Cryptolestes turcicus (Grouv.) and C. ferruginous (Steph.), the interaction between different combinations of the larvae of the two species was studied when each newly hatched larva was provided with 12·5 mg. wheatfeed. It was found that the numbers of C. turcicus surviving to the adult stage in the presence of C. ferrugineus was a linear function of its own initial numbers, whereas the survival of C. ferrugineus seemed to be some function of the size of the environment. This conclusion, taken together with other data, suggests that if the two species were to gain access to cereals together, dry conditions, coarse food, low larval density and the availability of refuges for pupation would favour high populations of C. ferrugineus, whereas moist conditions, fine food, moderate to high larval density and the absence of refuges for pupation would favour high populations of C. turcicus.

A short history of the study of the six tsetse species of the group of Glossina fusca (Wlk.) that occur in Nigeria and the former British Southern Cameroons (now West Cameroun, part of the Federal Republic of Cameroun) is given. Extensive collections of these species were not made until after the scheme for the establishment of the West African Institute for Trypanosomiasis Research was initiated in 1947. More intensive studies, stimulated by the discovery of a technique of searching for the flies in their resting sites, were carried out in southern Nigeria, at the W.A.I.T.R. Field Station and elsewhere, from 1953 onwards.

The situation, climate and vegetation zones of Nigeria are described. Species of the fusca group occur only in the southern part of Nigeria and in West Cameroun and depend on some type of forest vegetation, ranging from relatively dry forest islands and riverine forest in savannah to dense, humid, wet rain forest, for their habitat. Rainfall and relative humidity, in general, decrease with increasing distance from the coast and are factors of great importance in determining the distribution of the fusca group.

Locality records for each species of the fusca group are listed according to the Province and nearest town or village to the place where they were collected, and the results are shown on maps. G. medicorum Aust. has only been recorded from the relatively dry northern part of the rain forest and forest islands or riverine forest in savannah in Western Nigeria. G. nashi Potts occurs in dense, wet rain forest in West Cameroun. G. tabaniformis Westw. occurs, especially, in forest reserves in Ondo and Benin Provinces of Western Nigeria and in wet rain forest in West Cameroun and adjoining forest country in Eastern Nigeria. G. haningtoni Newst. & Evans is common in wet rain forest in West Cameroun and there are also records 175 miles to the west in Delta Province of Western Nigeria. G. fusca occurs in a wider range of climatic conditions and habitat types than the other species; it has been recorded in vegetation varying from forest islands in savannah to wet rain forest. G. nigrofusca Newst. is a rare but widespread species in Nigeria and West Cameroun where it occurs typically in wet rain forest.

The dry and wet limits (indicated by mean annual rainfall) of the distribution of each species in Nigeria and West Cameroun are given, and the importance of climate, especially humidity, in determining these distributions is discussed. The climatic requirements of the six species are confirmed by their distribution outside the boundaries of Nigeria and West Cameroun. G. medicorum extends mainly to the west of Nigeria although there are also two records of the species from Gabon, widely separated from the main area of distribution of the species in West Africa. G. haningtoni and G. nashi, species typical of Central Africa (as here defined), extend only to the east and south of Nigeria (south in the sense of south of the latitude of Nigeria in territories in Central Africa). The remaining three species, those which occur in both reasonably dry and reasonably wet forest conditions in Nigeria and West Cameroun, extend both west and east and south of Nigeria and West Cameroun.

Some type of thicket or forest vegetation is the essential habitat for the survival of all six species of the fusca group. The main areas inhabited by the fusca group now are forest reserves in Western Nigeria and the relatively undisturbed forests of West Cameroun. The species are absent from much of southern Nigeria, and it seems certain that the increase of the human population over relatively recent times, and the effects that this increased density of population has had on the flora and fauna, has been the cause of the present-day discontinuous distribution of the fusca group in southern Nigeria. Man has an adverse effect on fusca-group tsetse populations through clearing of forest vegetation and the hunting and driving away of the game animals on which the flies depend for food. An inverse relationship between human population density and the number of records of the fusca group over southern Nigeria and West Cameroun is demonstrated.

Fecundity in the garden chafer, phyllopertha horticola (L.), is extremely variable. Fecundity depends entirely on reserves laid down by the larva.

Previous work in Britain showed that (a) early larval feeding periods in the field produce heavy pupae; (b) in an experiment comparing early with late-hatched larvae, the heavier diapause larvae developed from the early batch; (c) in the field, the earliest larvae to enter diapause were the heaviest.

Further data (hitherto unpublished) showed that larvae reared at 25°C. entered diapause heavier than larvae reared at field temperature, although mortality was high. The above data also indicated that time of entry into diapause was fixed and unaffected by temperature.

A repeat of the field experiment (b) showed the same result and also showed how a population of feeding third-instar larvae can be destroyed by frost while larvae in diapause deeper in the soil are unaffected. Thus a late season produces light larvae and a high risk of death before diapause.

Observations on larvae kept at constant temperatures (12, 16 and 18°C.) showed that the temperature threshold of development is about 12°C. in the first instar but is lower in succeeding stages; the length of the first and the second instars decreases with rise in temperature; high temperature probably produces larger larvae by the end of the second instar, but the effect was small and none at all was detected at the first moult.

The length of the third-instar feeding period is not affected by temperature in the range 12–20°C. Kate of growth (increase in weight) increases with temperature. The higher temperatures (perhaps as a result) produce heavy diapause larvae and pupae in both sexes. Entry into diapause is hindered at 20°C. The weight of a larva at the second moult is positively correlated with its diapause weight.

It is suggested that the temperature effects described can be most simply explained by postulating the existence of differential temperature coefficients for the various growth processes.

Notes are given on the use of a chloral hydrate solution for clearing sandflies (Phlebotomus). Descriptions are given of three new species of sandflies from Uganda and Kenya (Phlebotomus (Sergentomyia) edwini, P. (Sergentomyia) kitonyii and P. (Sergentomyia) teesdalei spp. n.) and of the male of P. (Sergentomyia,) multidens Heisch, Guggisberg & Teesdale. Supplementary descriptive notes are given on Phlebotomus (Phlebotomus) rodhaini Parr., Phlebotomus (Sintonius) graingeri Heisch, Guggisberg & Teesdale, with illustrations of the spermatheca in both species, and the occurrence of morphological variation in the male genitalia of Phlebotomus (Sergentomyia) schwetzi Adl., Thdr. & Parr, in Kenya is discussed in relation to season. Data are given showing natural infection rates with flagellate Protozoa in populations of Phlebotomus (Serigentomyia) garnhami Heisch, Guggisberg & Teesdale in part of the Kitui district of Kenya.

Among larval populations of the mosquito Opifex fuscus Hutton from supralittoral pools on the east coast of South Island, New Zealand, some individuals in the second, third and fourth instars had only simple hairs in their mouth-brushes, others had pectinate bristles. All first-instar larvae had simple hairs, and samples of these from the same pool were reared in the laboratory at 25°C. Of 13 fed on dehydrated blood serum, only two developed pectinate bristles, but of 15 fed on fish food (particle size 0·1–0·6 mm.), all but one did. Of a later sample of 16 reared at room temperature on fish food in the first two instars and then on the blood serum, all developed pectinate bristles in the second instar but had reverted to simple hairs by the fourth. In nature, larvae of O. fuscus observed to be feeding by filtering were predominantly of the simple-hair type and those browsing of the pectinate-bristle type. Environmentally controlled dimorphism may increase the probability that an individual will develop to maturity.

The garden chafer, Phyllopertha horticola (L.), has been recorded in most counties of England and Wales (43 out of 52) and is probably to be found in all. It is more abundant the warmer and wetter the region, i.e., it is more abundant in the west and south and most abundant in the south-west. It flourishes best in permanent pasture on light soil in hilly country, and occurs from sea-level to an altitude of at least 1,160 ft.

Patchiness in local distribution was investigated on the pastures of fell farms in the Lake District.

An infested fell pasture includes from one to five plant associations which are described as (1) Grass, (2) Sparse Bracken, (3) Bracken Bed, (4) Heath Rush and (5) Common Rush. Typically, all five are present in a pasture. The proportions of ground covered by grass turf (Agrostis-Festuca) in these associations are (1) 100 per cent., (2) 99 per cent., (3) 0 per cent., (4) 80 per cent, and (5) 50 per cent., the grass turf in the rush associations (4 & 5) filling the spaces between clumps of rushes. Soil stages of the garden chafer are found only in the grass turf and so the Bracken Bed association is never infested. In terms of density per unit area of ground, the other four associations exhibit the following relation between their infestations: Grass>Sparse Bracken>Heath Rush>Common Rush. This relation is only partly explained by the similar relation between proportions of ground covered by grass turf (see above). As to other factors that might be partly responsible, no correlation exists between botanical characteristics of grass turf and infestation but there are two negative correlations among physical characteristics which are suggestive: Grass and Sparse Bracken-<Heath Rush and Common Rush with respect to toughness of turf; and Grass<Sparse Bracken<Heath Rush<Common Rush with respect to wetness.

Grassland, an unbroken expanse of grass sward, is, in the Lake District, by far the most important habitat since it carries by far the greatest part of the garden chafer population.

In grassland no evidence was found of an association between density of soil stages and proximity to bracken.

The population of any soil stage, from egg to pupa, is always much greater in drier than in wetter parts of grassland.

Sampling at intervals within a generation indicated that the mortality rate is higher in the wetter parts of grassland. This is not due to any lack of food, or difficulty of utilising it, or to the presence of more enemies (competitors, predators, parasites and pathogens) in the wetter parts, but, as established by laboratory and field experiments, is attributable to a direct effect of moisture. Further experiments suggested that the egg stage is most and the feeding stages are least resistant to excessive moisture.

Field experiments also indicated that the so-called ‘effective bee-liners’, females characterised by a direct and sustained flight from their roosting sites and that lay some eggs far afield, exercise some degree of discrimination in the choice of oviposition sites, the majority tending to choose the drier parts of grassland, the kind of site that does in fact, become more highly infested. Less than a quarter of all females become effective bee-liners and as such lay only about 15 per cent, of all eggs deposited by a population.

There is a dynamic aspect to the phenomenon that population is greater in drier than wetter parts of grassland. Although population is always greater in the drier parts, the magnitude of the population ratio between drier and wetter parts can change considerably from one year to the next as a result of either excessive rain or excessive drought during the larval feeding period. The mechanism of such change is indicated.

Spatial distribution on grassland was assessed statistically by employing Fisher's Coefficient of Dispersion. Eggs, and hence newly hatched larvae, are strongly aggregated. This strong degree of aggregation is gradually lessened, however, by the subsequent wanderings of the larvae in the course of feeding. Nevertheless, even in the final instar, larvae are still appreciably aggregated and hence so also are pupae. By the end of the feeding period, average dispersal from the hatching point is probably not more than 12 inches.

Apart from proper food, which is provided by grassland, the most important living condition for the garden chafer in the soil appears to be proper soil moisture. Although actual limits have not been specified, survival clearly depends on the soil containing neither too much nor too little water for too long.

As regards geographical distribution in England and Wales, the finding that the garden chafer is more abundant the warmer and wetter the climate is explained as follows. Females grow larger, and hence produce more eggs, where the larval feeding period is warmer, and more larvae survive where the climate is wetter, because the greater rainfall means less chance of too little water for an animal that has to live in freely draining soil to avoid too much water.

Fundamentally, the patchy local distribution of the garden chafer is mainly a matter of the patchiness of soil-moisture conditions. The patchiness determined by these conditions is maintained and intensified by the reproductive behaviour of the adults.

Accounts of other species in this and other countries suggest that, for pasture scarabs in general, soil moisture is the chief factor concerned in determining spatial distribution.

The persistence of DDT deposits, derived from dilutions of an emulsion concentrate, Arkotine D.25, on the leaves of riverine vegetation in the Northern Guinea Savannah vegetation zone of Nigeria was studied in the laboratory by a bioassay method using teneral females of Glossina palpalis (R.-D.) as test insects. Irrespective of whether sprayed in the dry or wet season, deposits derived from sprays containing 5.0 per cent. DDT produced by the Warley knapsack sprayer were still reasonably toxic one year after spraying. Deposits derived from sprays having a concentration of DDT varying from 1·25 to 5·0 per cent. DDT were equally toxic when young, but those from the lower concentrations were weathered much more rapidly than those from the 5·0 per cent, sprays. Young deposits produced by Warley and Motoblo sprayers were equally toxic, but deposits from the Motoblo deteriorated more rapidly than those of the Warley.

The toxicity of the DDT deposits on leaves varied regularly with the seasons. The various climatic factors that it is thought influenced the variations in toxicity are discussed. Accumulations of wind-blown dust on the leaves during the late dry season and low evaporation rates at the height of the rains appear to be responsible for reduced toxicity at those times; in the early wet season, the occasional violent storm probably washes off the leaves the dust accumulated during the dry season and makes the insecticidal deposits once more available. Heavy rainfall is important in removing the insecticidal deposits from the vegetation, and leaf decay and refoliation are important in reducing the availability of the deposits. The implications of these variations in toxicity and factors that reduce the availability of the deposits are discussed in relation to the eradication of riverine tsetse.

Great differences in the rates of foliage decay and refoliation occur between different riverine plants, and it is suggested that a thorough study of this feature in riverine plants and of resting sites favoured by tsetse flies might reveal information that would enable tsetse to be eradicated by selective spraying of certain plant species only.

The local blowflies were sampled by bait-trapping at a number of stations in the Northwest Highlands of Scotland and their western seaboard, the Western Isles and eastern and southern England, to supplement the results of an earlier survey of the distribution of Calliphorine blowflies in Great Britain.

Of the species of Lucilia, L. richardsi Collin was as numerous as L. scricata (Mg.) in the south-east, but relatively infrequent further north, and was not found in the Northwest Highlands or Islands. L. ampullacea Villen. and L. silvarum (Mg.), though earlier found in the eastern part of the Northwest Highlands, are rare or absent on the western seaboard and in the Western Isles. L. illustris (Mg.) is relatively infrequent in the Northwest Highlands and Islands, but in these regions was found as far north as Lairg, and in the Outer and Inner Hebrides.

Of the species of Calliphora, neither C. loewi End. nor C. uralensis Villen. were found in any of the eight stations in eastern and southern England. C. uralensis was common in the Western Isles and on the northwest seaboard. C. loewi more common in the Central Highlands and apparently absent from the Islands.

Acrophaga subalpina (Ringdahl) was not taken in the stations in the east and south of England, Cynomyia mortuorum (L.) only in the three more northerly of these, north and west of the Wash, and Phormia terraenovae R.-D. at onlv one station further south, Wye.

Amblyomma lepidum Dön., previously incriminated as a vector of heartwater (caused by Rickettsia ruminantium) among sheep and goats, and possibly camels, in southern Kassala Province, Sudan, is apparently the only Amblyomma species in this area. Systematic collections of all ticks infesting domestic animals were carried out in November 1959, just after the rains, in five ecologically differing areas in the region most severely infected with heartwater. Though rare in desert scrub (north of 16°N. latitude) and only moderately numerous in semidesert grassland (Butana area), incidence of A. lepidum is high in wooded savannah (River Atbara) and in the cultivated Gash Delta, and highest (15·5 per host) in the Acacia seyal savannah of the River Gash. Chief hosts of adults are cattle. Sheep are secondary hosts, and camels, donkeys, and goats are less heavily infested. Two males were taken feeding on humans. Mating (breeding) activity is only slight following the rains (November). Few larvae and nymphs were found during this period; those present were mainly on sheep and goats, with a few on cattle, and these were usually feeding on the hoofs of their hosts.

Rhipicephalus s. sanguineus (Latr.), the most numerous and widely distributed tick here, infests chiefly sheep and goats, is common on cattle, and also occurs on camels and donkeys. Mating activity of this species is high in November. Hyalomma dromedarii Koch is equally widely distributed and follows R. s. sanguineus in numbers, but is most frequent on camels, common on cattle, and rare on donkeys and sheep; goats were uninfested. Mating is moderately active at this season. The incidence of H. impeltatum Schulze & Schlottke, which occurs in each zone, equalled that of A. lepidum. Cattle are the chief hosts and camels are secondary hosts; donkeys, sheep, and goats are less frequently infested. Mating activity is moderate in November.

Other more or less common tick species on these domestic animals are Boophilus decoloratus (Koch), R. e. evertsi Neum., R. simus Koch, H. anatolicum excavatum Koch, H. truncatum Koch and H. marginatum rufipes Koch, evidence for the status of which as a subspecies of H. marginatum Koch is to be presented in a subsequent paper. Several other rare species are also recorded.

Investigations have been made on the influence of environmental conditions on the contact toxicity to adult mosquitos (Anopheles stephensi List.) of two kinds of insecticidal deposits of importance in sprayed houses, superficial deposits from wettable powders and insecticides sorbed on dried soils.

Increasing the relative humidity from 43 to 80 per cent, during the contact period had no effect on the toxicity of wettable-powder deposits on plywood of dieldrin, Sevin and O-methyl O-(2,4,5-trichlorophenyl) ethylphosphoramidothioate. On the other hand, the availability of DDT, dieldrin and 3-isopropylphenyl N-methylearbamate sorbed on dried soils increased considerably as atmospheric humidity increased. The logarithm of the median lethal exposure time was linearly related to atmospheric humidity, and decreased by a factor of 2·8, 2·2 and 2·1 for the carbamate, dieldrin and DDT, respectively, for each 10 per cent, increase in relative humidity. There was also a linear relationship between the logarithm of the water uptake of a given soil and humidity.

Temperature affected the action of insecticides during both the contact period and the post-treatment period. Median lethal doses of dieldrin, Sevin and the phosphoramidothioate decreased, and that of DDT increased, as the post-treatment temperature increased from 20 to 30°C. These effects of the post-treatment temperature were also found when treatment was by exposure to residues. When these effects were eliminated it was found that the contact toxicity of superficial deposits on plywood and of insecticides sorbed on dried soils increased as the temperature during the exposure period increased. A simple relationship between median lethal time and exposure temperature was indicated.

The known external characters, and some new ones, for separating the sexes of pupae and adults of stored-products Coleoptera are described and in many cases figured.

A description is given of the life-cycle in Britain of Theocolax formiciformis Westw. (Pteromalidae), a parasite of Anobium punctatum (Deg.) (Anobiidae). Adults emerge from infested wood in large numbers from April to June. Eggs are laid through the wood surface and use was made of this habit as a means of rearing this parasite under observation in the laboratory. Anobium larvae were placed individually in channels gouged in the surface of 3-inch squares of plywood. They were covered with tracing paper and the parasites were caged on this by means of a glass ring covered with a glass plate. Development from egg to adult at 22 and 25°C., and 75 per cent, relative humidity, averaged about five and six weeks, respectively, compared with 12 weeks outdoors in summer. The number of progeny per female was highest (approximately five) at 22°C., at which the ratio of males to females was 1:3. Only about one per cent, of the adults reared were winged. The use of Theocolax as a biological-control agent is not considered practical.

A method of rearing Gryllus pennsylvanicus Burm. is described, designed to yield large numbers of first-instar nymphs of uniform age for toxicological investigations Nymphs are reared to the fourth instar in glass vessels containing a little soil, which prevents cannibalism, and then transferred to glass-topped stock cages having sides and floors of aluminium screening and each accommodating 250 individuals. A mixture of dog-food pellets, oatmeal and milk powder, with fresh oats or clover, is provided throughout as food, and petri dishes of moist fine sand for oviposition. After separation from the sand by sieving, the eggs are incubated in moist soil at 24°C. for one month, subjected to 4°C. for three months to eliminate the obligate diapause, and then returned to 24°C., when 70 per cent, hatch after 15–18 days. The diapause enables eggs to be stored at 4°C. until required. At 24°C., the nymphal stage lasted about 81 days, the adult males and females, which are produced in equal numbers, survived about 62 and 56 days, respectively, oviposition extended over 11 weeks and the average number of eggs laid per female was 970·6.

Four methods have been tested of dividing cultures consisting of wheat grains containing larvae and pupae of Sitophilus granarius (L.) into samples weighing just under 10 g. and suitable for use in fumigation experiments. Two mechanical methods, each dividing a culture into two equal parts, the division being successively repeated to yield 32 samples, were shown to be satisfactory in every trial. One apparatus divided the culture on the chess-board principle, the other by means of radial fins equally spaced round the circumference of a conical divider. The standard method used at the Pest Infestation Laboratory also proved satisfactory; it consists of scooping up samples by hand radially, at random, from a culture dispersed symmetrically by pouring it over the bottom of a spherical flask, but it was evident that personal bias could be introduced by its use. The fourth method, consisting of ‘ random scooping ’, is widely used but yielded samples having considerable heterogeneity from all cultures containing larvae older than the late third instar and having a density exceeding one insect in 30 grains of wheat. This method was the only one that divided standard cultures into samples showing heterogeneity in their age structure. When cultures with an age range of only four days were divided, all four methods yielded samples for which the emergence curves were heterogeneous, apparently as a result of heating induced in samples as small as 9 g. as a result of the metabolism of the larger larvae.

Results are presented of 15 experiments made in north coastal districts of New South Wales from 1957 to 1961 with insecticides against the weevils Amnemus quadrituberculatus (Boh.) and A. superciliaris (Pasc.), the larvae of which have been reported as pests attacking the roots of clover (Trifolium repents, T. pratense and T. subterraneum) there. The adults live above ground, and single surface applications of insecticides were made to pastures in the form of dusts, granules or dilutions of emulsifiable concentrates in January, when maximum numbers of weevils were present and before significant oviposition had taken place. Insecticidal control was assessed by soil sampling for larvae in July–August, during the following winter. Pasture cuts were made in an attempt to assess the effect of treatments on the yield of clover.

Dieldrin at rates of 2 and 4 lb. per acre gave near eradication of Amnemus for one year, and satisfactory control was obtained with 1 lb. per acre. Aldrin, endosulfan (Thiodan) and BHC were less effective, and limited tests with heptachlor suggested that it was inferior to dieldrin. The scanty results available suggest that dieldrin is effective against Amnemus for only one year, but it is doubtful if annual applications are warranted. Owing to the patchy distribution of clover in the pastures, only one experiment showed an increased yield of clover associated with the reduction in numbers of larvae of Amnemus brought about by the application of 1 lb. dieldrin per acre. Similarly, poor persistence of clover on all plots prevented assessment of any residual effect on yield in the second year.

No method was found by which accurate estimates could be made of the density and distribution of adults of Amnemus in pastures to which insecticides were to be applied. Treatments, therefore, were replicated more frequently than was needed statistically in order to ensure that measurable levels of larval infestation were present at sufficient sites.

Small plots (500–700 sq. ft.) proved of little value, and this is attributed not only to contamination of control plots but possibly also to the death of weevils that moved from untreated into treated plots. Large plots (13,000–21,000 sq. ft.) gave the most consistent results in assessment of insecticides.

Control of Amnemus did not increase the productivity or persistence of clover, and it is concluded that any benefits from control are unlikely to be significant unless other factors, such as competition from grasses, drought, floods, nematodes and diseases of clover, which affected the density and persistence of clover in these experiments, are not limiting.

Brief notes are given on five species of boring insects that attack date palms in Iraq and between them cause considerable damage, based on observations in 1957–62. They comprise the Pyralid Arenipses sabella Hmps., the larva of which mines the spathes and fruit stalks, preventing fruit maturation; the Cerambycid Pseudophilus testaceus Gah., the larva of which mines the frond bases and stem;the Dynastid Oryctes elegans Prell, of which the adult makes a surface mine in the frond mid-rib or fruit stalk and the larva develops, usually, inside the stem of dead or moribund palms; the Bostrychid Phonapate frontalis (Fhs.), which mines in the frond mid-rib; and the termite Microcerotermes diversus Silv., which makes covered galleries up the outside of the stem and mines within the stems and frond bases of weakly palms.

In its potential outbreak areas in Northern Ehodesia and Tanganyika, sexually immature populations of adults of Nomadacris septemfasciata (Serv.) tend to concentrate in the islands of grass that remain unburnt after the annual fires that occur during the dry season. The locusts use the tall grasses as roosting sites and descend in the daytime to feed on adjacent short grasses, where these occur as constituents of an internal mosaic within the islands, or, in the case of islands composed wholly of tall grasses, on the fresh grass-shoots on the burnt-over ground surrounding them.

Two methods are described whereby emulsified solutions of dieldrin were applied in the Central Bukwa plain, Tanganyika, in 1960 as a stomach poison at dosages of 2·9 and 4·5 oz. active ingredient per acre from aircraft to the whole of individual islands showing the internal mosaic pattern (4 trials) or at 2·9, 4·5 and 8·2 oz. per acre to a swath 20 or 40 yd. wide around the perimeter of tall-grass islands (8 and 4 trials, respectively). Results were assessed by counting the numbers of locusts flushed by a vehicle driven along strips parallel to one another and 0·1 mile apart, and checked by counts of dead locusts. Mortality reached at least 90 per cent, within four days, at either dosage, following the first method, and 21–89 per cent, within 14 days following the second, being significantly greater at the higher dosages but not with the wider swath.

These results are compared with those obtained in preceding years by the standard method of applying DNC from aircraft at 20 per cent, in oil as a contact insecticide; in the latter case, the cost of insecticide was more per acre sprayed, but less per 1,000 locusts killed because the density (5 locusts per sq. yd. of grass island) in the trials was greater than in those using dieldrin (1 per 3 sq. yd.). Even so, when dieldrin was used, many more locusts were killed per sortie flown, and the total cost (of flying plus insecticide) per island was much less, and per 1,000 locusts killed very much less, than in the standard method, which nevertheless has to be used against mobile swarms because DNC is quick acting.

The dispersion of groups of adult grain weevils (Sitophilus granarius (L.)) was investigated under controlled conditions by observing the numbers occurring on the surface, or at various depths below it, in small bulks of grain. The effect of population density, and temperature and moisture content of the grain, on numbers at the surface was studied by using groups of weevils in 500 g. wheat, and on dispersion in the vertical plane by confining weevils, isolated or in groups of twenty, within a thin layer of 200 g. wheat contained between vertical glass plates, through which the positions of individuals could be recorded.

In grain at 25°C. and 14 per cent. moisture content, the proportion of the population at the surface was about 17 per cent. for densities of 2–100 insects in 500 g., and was only temporarily changed by adding or removing weevils at the surface; the proportion was higher at nine per cent. moisture content or at 15 or 35°C.

In grain at 25°C. and 14 per cent. moisture content, the typical behaviour pattern of individuals consisted of random vertical and lateral movements, so that there was a constant turnover within a group, but at nine per cent. moisture content and 25°C. they tended predominantly to move upwards. At 15 and 35°C., isolated individuals were stationary, but upward movement was elicited by increase in density; relatively static behaviour at 35°C. is thought to be associated with the need for water conservation. The possible relevance of these results to the detection and control of pest populations and their spread in storage conditions is discussed.

In a long series of experiments with Lasioderma serricorne (F.) carried out at 30°C. and 60 per cent. R.H. it was established that shortage of food prolongs development and reduces survival of the immature stages and also reduces the weight of the resulting adults. Competitive effects between larvae over and above those caused by food shortage alone were not detected. Larvae ate eggs and pupae only in the complete absence of any other food, and they did not appear to eat one another. There was no trend in the performance of the immature stages attributable to the age of the female parent at the time of oviposition. Variation between the times taken by different individuals to attain a subsequent stage was positively skewed. Adult females weigh more than males, this difference in weight first occurring during the third (penultimate) larval instar. There were no significant differences in the developmental periods of the sexes.