Studies in the Entomology Division at Ottawa in 1957 and 1958 led to the design and building of a “walk-in” air-conditioned cabinet for rearing insects, air motion and distribution as well as temperature and humidity being precisely controlled. The air is heated, cooled, humidified, and dehumidified in four separate ducts. A quadrate funnel for the air exhaust gives especially even distribution of the air within the cabinet, the air moving at 25 to 30 f.p.m. Temperature and humidity can be held at any desired levels from 65 to 85 ± 0.5°F. and from 45 to an accuracy of ± 1 per cent R.H. provided that ambient conditions are within 65 to 85°F. and 5 to 85 per cent R.H. This paper describes the cabinet and air-conditioning system and the main principles on which the apparatus is based.

During some investigations on grasshoppers it became necessary to know how long they would survive without food or water after hatching. The literature contains only scanty observations. Washburn (1911) reported that grasshoppers “freshly hatched, can live from four to five days without food”. Langford (1930) showed that 30 per cent of Melanoplus femur-rubrum (DeG.) and M. differenrialis (Thos.) (adults?) survived for two days at 95° F. without food and 100 per cent survived for two days at 69° F. At 38° F. 40 per cent survived for 13 days. Telenga (1930) kept first-instar Schistocerca gregaria Forsk. in outdoor cages without food and water where they survived only two days. Bodenheimer (1929) found that the average life of newly hatched S. gregaria in outdoor cages without food at ± 15° C. was 3.6 days, with a maximum of nine days, and that later instars that were starved lived longer at higher humidities. Ludwig (1937) testing the effect of different relative humidities on survival showed that third-, fourth-, and fifth-instar Chortophaga viridifasciata (DeG.) survived starvation for 5.2 to 6.6 days at 25° C, at all re1ative humidities from 5 to 96 per cent.

In 1959 a study was conducted in the Whiteshell Forest Reserve, Manitoba, to determine if insectary-reared larch sawfly larvae developed at the same rate as in the field. Data were collected on the rate of frass production in both situations and used as an index of larval development. This was part of a broader study that will be reported on more fully in a later publication, but the immediate results should be of interest to those concerned with studies on ecological life histories of other insects.

There seems to be only one report of the aphid Myzus certus (Wlk.) transmitting a virus, bean yellow mosaic virus (Swenson, 1957). It is therefore noteworthy that M. certus transmits potato virus Y as readily as does Myzus persicae (Sulz.), until now the most efficient vector known for virus Y. In one experiment that showed this, the numbers of single aphids out of 50 that transmitted virus Y were 27 and 28 respectively. Briefly the experiment was as follows: M. certus was reared on pansy (Viola tricolor) and M. persicae on rape (Brassica napus); the strain of virus Y was the one that has been used at our laboratory for aphid transmission studies during the past ten years; single adult apterae or late instar nymphs each probed once for less than a minute into an infected tobacco plant and then fed overnight on a young healthy tobacco plant (Nicotiana tabacum var. White Burley).

The predacious mite, Melichares dentriticus (Berlese), has been mass-cultured (Burnett, 1960) for the past few years at this laboratory on populations of Tyrophagus putrescentiae (Schrank) (= castellanii (Hirst)) (Acarina: Acaridae) (Robertson, 1959). Individual rearings, however, were required to obtain detailed information on the life history, behaviour, and ecological requirements, prior to the use of this species as an experimental animal in the study of predatorprey interactions. This paper is an account of the various techniques tested and of the observations made throughout these studies.

The literature on Orthoptera (sens. lat.) includes a number of cursory notes describing laboratory drinking in various species, but an extended discussion of this behaviour is not available. The author has spent the past seven years studying the feeding behaviour of Michigan Orthoptera, during which time he has also made observations and conducted experiments on intake of water. Findings with respect to the latter, although of a preliminary nature, are given in the following paper.

On the basis of anatomical and ecological differences, Morgan and Anderson (1957) and Anderson and Morgan (1958) divided the Bryobia praetiosa complex of British Columbia into two species; the brown mite, B. arborea M. & A., 1957, found exclusively on fruit trees, and the clover mite, B. praetiosa Koch, 1836, found chiefly on herbaceous plants and known more commonly as a household pea. Acarologists in Switzerland, The Netherlands, and Russia have separated the complex into a number of species principally on anatomical bases (Pritchard and Keifer, 1958). In a recent exhaustive treatise of the complex in Switzerland, Mathys (1957) proposed five species and four ‘forms’.

Body above pale yellow, abdomen with faint brown mid-dorsal and dorso-lateral spots; a vinous stripe in front of wing-base; body beneath and legs whitish buff. Wings above yellow, paler distally on fore wing and costally on hind wing. Fore wing with faint indications of an arcuate, blackish-fuscous, broken antemedial band; orbicular represented bv a black dot in cell; reniform obsolete; postmedial line strongly zig-zagged, inierrupted at veins, preceded by black dashes in cells M2, to Cu2; a subterminal row of brownish-fuscous spots between veins; fringe yellow; hind wing with interrupted, zig-zagged postmedial line, preceded by black dashes, as on fore wing; subterminal spots stronger than on fore wing; fringe yellow. Wings beneath whitish buff; costa of fore wing weakly infuscated; a fuscous dot-like orbicular and lunate reniform; no other markings. Expanse 67 mm.

Extensive investigations on the biology and control of biting flies were conducted in northern Canada from 1947 to 1953 (Brown, et al., 1951; Haufe, 1952; Hocking et al., 1950; Twinn et al., 1948, 1950; Twinn 1950). The difficulty of assessing populations of mosquitoes both in chemical control and in ecological investigations limited the interpretation of some of the work that was undertaken. One of the primary objectives of ecological and behaviour studies was to determine the weather conditions that are favourable for attraction of mosquitoes to man. Reliable estimates of the unattracted population must, for this purpose, be obtained independently of the observer. Large catches are also important in estimating abundance when the time for individual catches is reduced to one hour or less. For these reasons a satisfactory mechanically-operated device for sampling mosquitoes in flight became essential.

In his treatment of Chrysops wiedemanni Kröber, Philip ( 1959, pp. 203-4) stated, “Syntypes in Hamburg were casualties in the war, but one intact syntype female remaining in the Canadian National Collection is available as lectotype.” Later, Philip (in litt. Jan. 19, 1960) explained, “my wording was ambiguous about the lectotype of Chrysops wiedemanni. I indicated its availability in contrast to the loss of all other syntypes in Hamburg but this was intended to officially designate your specimen as lectotype and it should be so labelled”.

Freeze-drying equipment at this station capable of handling bulk quantities of fresh plant material was designed as a single unit and was therefore unsatisfactory for use with plant materials having insecticide residues because decontamination was difficult and quantitative recovery of the condensate after each run was almost impossible. The condensate may contain a considerable portion of the insecticide and contamination could influence assay results. To overcome these problems a simple apparatus, modified after the Cole-Palmer bulk freeze-dryer (Cole-Palmer Instrument and Equipment Company, Chicago, Illinois), was constructed from readily available materials with a minimum of machine-shop labor.

The following five undescribed species of mites were collected on citrus trees in the provinces of Assam and Mysore in India by Dr. C. A. Fleschner of the Citrus Experiment Station, Riverside, California. These mites are of the suborder Mesostigmata and are probably predacious. They may be introduced to California in attempts to control phytophagous mites on citrus and, therefore, may be recorded from this state in the future. I am grateful to Dr. Fleschner for the opportunity to examine this material.

This species keys to eleodis Vier. in Muesebeck's revision (U.S. Dept. Agric., Misc. Pub. No. 241, p. 16, 1936) and is very similar to that species and M. aethiops Nees. The female of M. eleodis differs in having a longer ovipositor sheath (about 1.2 times as long as abdomen bevond petiole), shorter petiole (only 1.9 to 2.2 times as long as its apical width), less strongly excavated and longer propodeum (only about 1.5 times as wide as long). The female of M. aethiops differs in having a pair of small dorsal pits near the middle of the petiole (both sexes), shorter ovipositor sheath (about 0.75 times as long as abdomen beyond petiole), slightly shorter petiole (about 2.2 times as long as apical width), and completely black petiole.

Coenosia tigrina (F.), an important anthomyiid predator of the onion maggot, Hylemya antiqua (Meig.), in Quebec, was found in all stages of development in field populations of the latter. The species was named by Fabricius (1775). A description of the adult was given by Perron et al., 1956, but no description of the immature stages is available. This is a report on the external morphology of the immature stages of this species, the determination of the overwintering sites of the larvae, and the effect of predation by adults on adult H. antiqua in field cages.

Since 1947 a blueberry thrips, Frankliniella vaccinii Morgan, has been reported as a common pest of the low-bush blueberry in New Brunswick. In a previous note Wood (1956) described the injury caused by the thrips under the assumption that only one species was involved. During the summer of 1958, however, several specimens of thrips ollected in Charlotte County, N.B. were sent, by request, to L. J. Stannard of the State Natural History Survey Division, Urbana, Illinois, who discovered two species, F. vaccinii and Taeniothrips vaccinophilus Hood, were included. This information was quite surprising and prompted further examination of specimens collected in this area prior to 1958. These specimens were examined by W. R. Richards, Entomology Research Institute, Ottawa, who reported that both species were present. Specimens collected between 1947 and 1951, however, were all F. vaccinii, those collected in 1951 were about 75 per cent F. vaccinii, and those collected since 1951 were mostly T. vaccinophilus. These records show that T. vaccinophilus, which appeared in or about 1951, has rapidly replaced F. vaccinii as the predominant species infesting blueberry in this area.

Four steps are necessary in constructing and utilizing a mathematical model for control of a specific insect pest. First, a large volume of field data must be collected according to a plan which is sound from both statistical and biological points of view (Morris, 1955). Secondly, a suitable procedure for developing component parts of the whole model and fitting these together must be developed (Watt, ms.). Thirdly, after exhaustive study of available data and testing of equations against these data we must determine the appropriate form of equation to express the action of each type of phenomenon governing dynamics of the pest population. This paper and other in the series (e.g., Watt, 1959) are designed to obtain such formulations. Which phenomena are worthy of smdv for a given pest species will become clear through mathematical analysis of life-table data of the type collected by Morris and Miller (1954). Finally, parameter values are obtained from the data and the equations are manipulated io determine optimum control practices, using methods outlined by Watt (ms.).

Many attempts to establish introduced parasites in Canada failed though the climatic conditions at their release points were apparently similar to those of their native habitats (Thorpe, 1930; Baird et al., 1947). It is known that many adults of hymenopterous parasites feed on flowers, aphid honeydew, host body fluids, and other substances (Jackson, 19 37; Thorpe and Caudle: 1938; Schneider, 1939; Györfi, 1945, 1951; Zoebelein, 1956). This brings up the possibility that nourishment from one or more of the above sources is necessary for surviial and reproduction, and that some of the failures may have been because of the absence or scarcity of suitable adult food (Clausen. 1956). At present data on the food preferences of adult hymenopterous parasites are scarce. Except for a few field and laboratory observations, there are no experiments that demonstrate distinct preferences for one food over another. The results of the experiments described in the present paper demonstrate that adults of the species studied have preferences for some flowers and food substances tested.

This is the second of a series of papers (Miller, 1959) describing the interaction of primary parasites and the spruce budworm, Choristoneura fumiferana (Clem.), based on data collected during an outbreak of the budworm in northern New Brunswick during the period 1947–1958. The first paper showed that the interaction between the spruce budworm and Apanteles fumiferanae Vier. is adequately described by the general mathematicai model developed by Watt (1959). The data on the parasite Glypta fumiferanae (Vier.) to be presented in this paper are also analysed by means of Watt's model and consequently the method is essentially the same. There is, however, one important difference. In the case of A. fumiferanae, the estimated number of adult parasites was only an index based on the potential number emerging from the previous host generation. The observed density of G. fumiferanae is a more realistic estimate. It is based on the actual number of cocoons found on the foliage during the adult emergence period.

Several species of resin midges (or pine-pitch midges) have been recorded from species of Pinus in North America. These are all at present referred to the genus Retinodiplosis Kieffer 1912; for reasons given below this genus is here treated as a subjective synonym of Cecidomyia Meigen 1803. At least two species of Cecidomyia have been recorded from jack pine, Pinus banksiana Lamb., in Ontario, Manitoba, Saskatchewan and Michigan (Bradlev (1946), Barker and Wong (1948), anonymous (1946), Prentice and Hildahl (1957, 1958) and Mc-Daniel (1938)); they were sometimes determined as Retinodiplosis sp., sometimes tentatively or definitely as R. resinicola (O.S.). Study of specimens of Cecidomyia from P. banksiana collected during the past three years by officers of the Forest Biology Regional Research Laboratories, Canada Department of Agriculture, has shown that three species of the genus occur on this host. All three are described as new in this paper. C. resinicola (O.S.) 1871, originally described from Virginia pine, Pinus virginiana Mill., was not present in this material nor among Michigan specimens from P. banksiana in the U.S.N.M.; it almost certainly does not occur on P. banksiana.