FILE: <bc-57.htm>                                                                                                                                                 GENERAL INDEX    [Navigate to   MAIN MENU ]

 

 

 

NUTRITION OF ARTHROPOD NATURAL ENEMIES

                                                                                                             (Email Contacts)

 

 

 -----Please CLICK on desired underlined categories [to search for Subject Matter, depress Ctrl/F ]:

 

Overview

Parasitoid Culture In Vitro

History

Adult Entomophage Nutrition

Utilization of Food

Continuous Culture on Artificial Media

Nutritional Requirements in Development

Contemporary Applications

Predator Culture In Vitro

References

                  ----------------------------------------------------------------------------------------------------------------------------------------------------------------

 

Overview

The nutrition of entomophagous arthropods was originally discussed in detail by Doutt (1964) and Hagen (1964). Slansky (1982, 1986) and Thompson & Hagen (1999) illustrate the complex interactions of behavioral, physiological and nutritional factors in arthropod nutrition. Nutrition is thus the action or processes of transforming substances found in foods into body materials and energy to do all the things attributed to life. Nutritional requirements are dependent on the synthetic abilities of the organism, which is controlled genetically. House (1977) stated that "... through nutrition we have a direct and essential connection between an environmental factor, foodstuff and the vital processes of the insect organism." Most nutrition research with insects has been aimed at improving rearing and not developing a basic understanding of their nutrition. Research has emphasized feeding and the development of artificial diets, which are concerned with dietetics (Beck 1972). Although critical to insect rearing, such research has given only a little understanding of insect nutrition per se.

Qualitative nutritional requirements of all insects are very similar in spite of a great diversity of feeding habits (Beck 1972, Dadd 1973, Hagen 1986b). Although a knowledge of dietetics and nutrition has advanced, practical application of principles to insect rearing to support biological control is lacking. Rearing each insect species is a unique challenge as there is meager knowledge of nutrition principals that might provide a broad and sound basis for approaching insect husbandry (house 1977). With entomophaga, foodstuff is in a continuous state of qualitative and quantitative change, and very little is known of the quantitative nutritional requirements for various life stages and physiological functions of these insects. The requirements for many nutrients are often dependent on the presence and concentration of others and correct nutrient balance may be critical for successful nutrition. Those parasitoids and predators for which artificial diets have been developed may serve as models for in vitro investigation on quantitative requirements for specific nutrients.

Thompson (1976a, 1982) used a defined artificial medium to examine the quantitative requirements for supporting larval growth of Exeristes roborator. Parasitoid development is intimately associated with host physiology. Changes in the host's physiology following parasitism are adaptive for the parasitoid, which insures successful development (Vinson & Iwantsch 1980, Thompson 1986). Parasitoids overcome potential nutrient constraints by altering their host's behavior and physiology (Slansky 1986). Changes in composition of the host's internal milieu may have significant nutritional consequences for a parasitoid (Grenier 1986, Thompson 1989). Endocrine interactions seem critical to successful parasitoid development. Synchrony in development between many larval endoparasitoids and their hosts occurs (Beckage 1985), and this suggests that the host's hormones and endocrine physiology influence parasitoid development (Lawrence 1986a.). The physiological basis of developmental synchrony is not well understood and knowledge is restricted to investigation of the relationship of Biosteres longicaudatus with its host Anastrepha suspensa (Lawrence 1982, 1986b). Some studies have tested the effects of hormones on the development of parasitoids in vitro with little success. The potential of using insect hormone supplements in artificial media to achieve successful growth and development of parasitoids in vitro deserves research emphasis.

The importance of ecological considerations in the nutrition of insects was discussed by Slansky (1982). It was emphasized that behavior and regulatory physiology of insects are in a state of continuous flux in response to food supply, and that nutrition can be fully understood only by considering the insects "nutritional ecology." With entomophaga both the ecology of the entomophage as well as that of the host or prey needs to be known.

Dietary and nutritional requirements are genetically based and genetic manipulation holds promise as a way to modify the nutrition of entomophages. Chabora (1970) suggested that nutritional content varies between strains of insects when he demonstrated that the yields of two parasitoids, Nasonia vitripennis (Walker) and M. raptor Girault & Sanders were significantly increased when they were reared on a hybrid of two strains of the host, Musca domestica L. The selection of desired traits for insect rearing was discussed by Collins (1984). The potential for genetic improvement of entomophages was outlined by Rousch (1979) and Hoy (1979, 1986). Most genetic selection has been directed to increase field effectiveness of entomophages, such as improving sex ratio, host finding ability, host preference, pesticide resistance and improved climatic tolerance. However, genetic improvement must also guarantee the preservation of vigor and vitality of the entomophage. Because these are intimately associated with nutrition, genetic programs may involve selection for nutritionally related traits.

Advances in recombinant DNA technology indicates a possibility for genetic manipulation of the nutrition of entomophages (Thompson 1989). The incorporation of foreign or in vitro altered genes for the expression of desirable traits by an organism, is rapidly advancing (Beckendorf & Hoy 1985), but is still not suited for practical application as of 1991.

History of Parasitoid Nutrition.--Salt (1941) probably was the first to emphasize the complexity of parasitoid nutrition in studies that demonstrated that the host influences growth and survival of the developing parasitoid as well as sex ratio, fecundity, longevity and vigor of the adult wasp (Clausen 1939, Salt 1941). Such complexities were demonstrated in work by Arthur & Wylie (1959), Wylie (1967), Nozato (1969) Sandlan (1979a) and others (Vinson & Iwantsch 1980). It has long been known that there is a relationship between host biomass and size of solitary parasitoids, larger parasitoids developing from larger hosts. This relationship exists for parasitoids which attack every host developmental stage, but applies more generally to parasitoids of host eggs and pupae where host size is fixed (Sandlan 1982). The relationship applies when a parasitoid is reared on different host species of variable size as well as when reared on different sized individuals of a single host species (Salt 1940, Jowyk & Smilowitz 1978, Mellini & Campadelli 1981, Sandlan 1982, Mellini & Beccari 1984). It does not seem to hold with ectophagous parasitoids, however (Legner 1969 ). The size of adult Trichogramma pretiosum Riley reared on the eggs of five hosts showed a direct correlation between parasitoid size and the volume of the host egg from which it emerged (Bai et al. 1989). A correlation also exists between total parasitoid biomass and/or numbers with host size in the case of gregarious larval parasitoids (Wylie 1965, Bouletreau 1971, Thurston & Fox 1972). The means by which gregarious organisms moderate their development relative to host size has been shown (Beckage & Riddiford 1983).

The relationship between size of host and parasitoid is closely associated with food quality and quantity (Arthur & Wylie 1959, Sandlan 1982). Salt (1940) found that adult Trichogramma evanescens Westwood display behavioral dimorphism related to host size. Large females obtained from large hosts failed to oviposit on small hosts, whereas small females accepted hosts of all sizes. Male wing development was influence by host size, and this was also found in Gelis corruptor (Foerster) by Salt (1952). Adult female Coccygomimus (= Pimpla) turionellae (L.) did not show morphological and behavioral polymorphism, but larger females found it difficult to oviposit in small hosts. On the other hand small females were more efficient in attacking small hosts. Fecundity was influenced by longevity with the greatest longevity reported for larger individuals reared from large hosts (Sandlan 1982).

The success of parasitoids in parasitization activity is directly related to nutritional factors. Smith (1957) found differences in larval mortality and adult size, sex ratio and reproductive rate of several species when reared on Aonidiella aurantii (Maskell) and Comperiella bifasciata Howard maintained on different food plants. Habrolepis rouxi Compere displayed limited mortality on A. aurantii when feeding on citrus, but 100% mortality when feeding on sago palm. Pimentel (1966) and Altahtawy et al. (1976) showed differences in parasitoid fecundity and longevity depending on host food source. Thurston & Fox (1972) reported that nicotine influenced the emergence of Cotesia (= Apanteles) congregata (Say) when reared on Manduca sexta (L.) feeding on tobacco. Hyposoter exiguae (Viereck) was harmed by tomatine in Heliothis zea (Boddie) feeding on tomato (Campbell & Duffey 1979).

Aphelinus asychis Walker required a longer larval developmental time and showed a decreased adult longevity when reared on Myzus persicae (Sulzer) fed on defined diets deficient in sucrose or iron (Zohdy 1976). The effects seemed related to decreased host size rather than a difference in nutritional quality of the host, however. The survival of Aphaereta pallipes (Say) was affected by the balance of amino acids and glucose in the artificial diet used for rearing its host, Agria housei (= affinis) Shewell (House & Barlow 1961). Differences in larval development and adult size, fecundity and sex ratio were observed in Tetrastichus israeli (Mani & Kurian) when reared on several host species, which was correlated to the total level of essential amino acids in host tissues (Nadarajan & Jayaraj 1975). Even though parasitoids reared from some host species with high levels of essential amino acids were larger and longer lived, the results were variable, as were the specific amino acid compositions of the different hosts. In general it may be assumed that parasitoid fecundity, reproductive size, sex ratio and longevity are correlated with host size and nutritional factors (Charnov et al. 1981, Charnov 1982, Luck et al. 1982, Mackauer 1986, Strand 1986). The importance of rearing Chelonus sp. nr. curvimaculatus on the natural host for vigor retention was demonstrated by Legner & Thompson (1977), as discussed in previous sections.

In contrast to parasitoids, few studies have been done on the effects of various natural foods on the biological character of predators. Smith (1965) reported that 10 coccinellid species fed dried, powdered aphids, grew larger and faster when feeding on Acyrthosiphon pisum (Harris) and Rhopalosiphum maidis (Fitch) than on Aphis fabae Scopoli. Coccinella septempunctata L. gained more weight when feeding on Lipaphis erysimi (Kaltenbach) than on two other aphid species, and it was demonstrated that L. erysimi had higher protein levels (Atwal & Sethi 1963).

Utilization of Food

Parasitoids have been thought to show high efficiencies in food utilization. Larvae consume food of high nutritional content and are mostly inactive within the host which offers a limited food supply, which points to selection for high food efficiency (Fisher 1971, 1981; Slansky & Scriber 1985, Wiegert & Petersen 1983). Parasitoids examined for food utilization include Coccygomimum (= Pimpla) instigator (F.), Pteromalus puparum (L.) (Chlodny 1968), Gelis macrurus (Thompson), Hidryta frater (Cresson) (= sordidus) (Edgar 1971), Brachymeria intermedia (Nees) and C. turionellae (Greenblatt et al. 1982), Diadromus pulchellus Wesmael (Rojas-Rousse & Kalmes 1978) and Trypatgilum (= Trypoxylon) politum (Say) (Cross et al. 1978), Phanerotoma flavitestacea Fischer (Hawlitzky & Mainguet 1976), Venturia (= Nemeritis) canescens (Gravenhorst) (Fisher 1968), Cidaphus alarius Gravenhorst and Phygadeuon dumetorum Gravenhorst (Varley 1961), and Cotesia glomerata (L.) (Slansky 1978). In these species, the mean net conversion efficiency (= proportion of assimilated food converted to body mass (Petrusewicz 1967, Calow 1977, Hagen et al. 1984) varied broadly (11-62%), with a mean of 37% that was < than for many groups of insect herbivores and detritivores. Cameron & Redfern (1974) of two studied parasitoids, Eurytoma tibialis Boheman and Habrocytus elevatus (Walker), were at the high end of this range. Net conversion efficiencies may not be very high because selection might have been for rapid rather than efficient growth (Slansky 1986). Possibly the well known inverse relationship between growth efficiency and assimilation (Welch 1968) may also be important. In contrast to net conversion efficiency, the above parasitoids had relatively high percentages of assimilation (= percentage of ingested food that is assimilated) ranging from 55-94%, with mean of 67%, compared with means of 40-50% for most herbivores and detritivores.

Howell & Fisher (1977) reported the highest nutritional efficiencies for a parasitoid in the ichneumonid V. canescens. Larvae had a 65% net conversion efficiency and 95% assimilation when maintained on the host Anagasta (= Ephestia) kuehniella (Zeller); net conversion efficiency to the adult was 20%.

The proportion of food/host available that is consumed by the parasitoid and converted to parasitoid biomass was calculated by Slansky (1986) and Howell & Fisher (1977). Calculated exploitation indices varied among species from 3-80%, and V. canescens larvae consumed 90% of its host's biomass and converted 55%, but there was no clear correlation between host size and parasitoid size nor biomass conversion.

Food utilization by predators has also been thought to be highly efficient, for reasons similar to that for parasitoids. This is especially true when predators spend much time waiting for their food (Lawton 1971), thus avoiding metabolic expenditure. Studies on food utilization of 11 predators was reviewed by Slansky & Scriber (1985). All had similar net conversion efficiencies (4-64%, mean = 34%), but higher assimilation efficiencies (37-98%, mean = 86%) than those of parasitoids. Cohen (1984, 1989) in studies of food utilization by Geocoris punctipes (Say) when reared from 1st instar nymphs to adults on eggs of Heliothis virescens (F.), found an assimilation efficiency of ca. 95%, gross conversion efficiency of 53% and net conversion efficiency of 55%.

Nutritional Requirements in Development

Qualitative nutritional requirements of insects, determined by use of defined and deficient artificial diets, were presented by several authors (Dadd 1973, 1977, 1985; Friend & Dadd 1982, Hagen et al. 1984). All insects have similar requirements for ca. 30 chemicals that include protein and/or 10 essential amino acids (arginine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan and valine), the B-vitamin complex (biotin, folic acid, nicotinic acid, panthothenic acid, pyridoxine, riboflavin and thiamin), as well as other water soluble growth factors, including choline and inositol, some fat soluble vitamins, cholesterol or a structurally similar phytosterol, a polyunsaturated fatty acid, minerals and an energy source usually provided by simple or complex carbohydrates and/or lipids.

Nutritional requirements of entomophagous insects are similar, and similar to those of nonentomophagous species. House (1977) referred to this common characteristic of insect nutrition as the "rule of sameness" (House 1966a, 1974). The rule has been confirmed by recent studies with parasitic and predaceous insects. In assessing the need for nutrients, it is important to consider that most studies were done by rearing a single generation on a synthetic or semi-synthetic diet. Some investigations overlooked the potential contribution of nutrients stored within the egg. Stored nutrients may support limited development and, in the case of trace nutrients, supply a sufficient quantity to ensure development of at least one generation. Studies with Itoplectis conquisitor (Say) (Yazgan 1972) and Exeristes roborator (F.) (Thompson 1981a) demonstrated partial larval development on diets lacking various essential amino acids and B-complex vitamins. Other studies have demonstrated that entomophagous insects have no unusual qualitative nutritional requirements. A requirement for asparagine by Eucelatoria bryani Sabrosky (Nettles 1986a) and the absence of a requirement for a polyunsaturated fatty acid by A. housei (House & Barlow 1960) were consistent with findings for nonparasitic Diptera (Dadd 1977).

The quantitative balance of different nutrients is a critical and dominating factor determining dietary acceptability and suitability (House 1969, 1974). The predominant foods of both parasitic and predaceous insects are of animal origin and, thus, are generally high in protein content and low in carbohydrate and fat (House 1977). Thompson (1986a) found a high requirement for protein and/or amino acids in parasitoids. Exeristes roborator at the 6% amino acid level completed larval development without glucose and/or fatty acids (Thompson 1976a). However, glucose markedly improved survival when the amino acid level was reduced to 3% and at 1% amino acid, no development occurred with the carbohydrate. Similar effects of amino acid level on larval development were reported by Yazgan (1972) for I. conquisitor. Adult eclosion was reduced by dietary amino acid levels of <6% and by deletion of glucose. Fatty acids were only marginally beneficial in enhancing growth and development rates of both species. A polyunsaturated fatty acid, however, was required in small amounts. Adult I. conquisitor (Yazgan 1972) and E. roborator (Thompson 1981a) displayed crumpled wings and/or bent ovipositors without a polyunsaturated fatty acid in the larval diet. Linolenic acid alleviated these deformities in I. conquisitor, and linoleic and linolenic acids were provided together in the case of E. roborator .

Thompson (1983a) described the effect of nutritional balance on larval growth of Brachymeria lasus (Walker). Media containing 0-10% glucose with 2% amino acids, and 1-8% amino acids with or without 2% glucose were tested. All media contained 15% albumin and 2.5% lipids. Weight gain increased on diets containing 2% glucose when the amino acid level was increased from 1-4%, but was reduced at the higher amino acid levels. Similar effects of varying the amino acid level were obtained with diets lacking glucose, but the overall weight gain was less than observed with the diets containing glucose. On diets containing 2% amino acids, weight gain increased dramatically when glucose was increased from 0.5-4%, but decreased at higher glucose levels. Growth rates on the above diets were generally in the range of 15-200 mg/g/day. The maximal rate, 260 mg/g/day, was obtained on a medium containing 2% glucose and 2% amino acids. The effects of nutrient balance were closely related to the osmolality of the artificial medium (Thompson 1983b).

House (1966b) demonstrated similar quantitative requirements to those of hymenopterous parasitoids in the dipteran Agria housei Shewell. Maximal growth and survival were achieved when all nutrients were increased proportionately over the levels in a basal diet that contained 2.25% amino acids, 0.05% salts, 1.16% lipids and 2.25% other ingredients, including glucose, ribonucleic acid, vitamins and agar. When amino acid level alone was increased, survival was reduced. On a diet containing nutrient levels equivalent to pork liver (= 20% amino acids, 4% glucose, 3.5% lipids, 2% salts and 0.75% ribonucleic acid), survival was >80%. House (1967, 1970) showed that the relative balance of amino acids and glucose was critical in determining growth or development and that A. housei larvae selected diets for feeding on the basis of nutrient balance.

Quantitative nutritional studies with parasitoids have generally evaluated the effects of nutritional balance by univariate or monofactorial analysis. Grenier et al. (1986) thought that such an approach had severe limitations because it ignored potential interactions between nutrients, including "...additivity, competitivity, antagonism or synergy." Thus, interpretation of effects of nutrient variation aimed at medium optimization was difficult, and it was suggested that nutritional studies be designed and analyzed in a multidimensional manner that accounted for interactions between all nutrients and biological criteria.

Canonical correlation analysis, which constructs maximum correlations between all linear combinations of variables within sets, such as between growth and development, and dietary parameters, were reported with Lixophaga diatraeae (Townsend) by Bonnot (1986, 1988). Because biologically meaningless correlations may be generated, accurate interpretation requires knowledge of biological correspondence between variables. Bonnot varied the concentrations of 30 medium components and determined the effects on 9 developmental criteria. Nine linear correlations were obtained and three had correlation coefficients of >0.95.

There is little information about the effects of developmental nutrition on the behavior of parasitoid larvae apart from measurements related to growth and development rate. However, Veerman et al (1985) reported that a photoperiodic response by C. glomerata was influenced by the carotenoid content of its host's diet. Vitamin A was essential for photoperiodic induction of diapause and it was suggested that this vitamin or a derivative may function as a photoreceptor pigment.

Optimal nutritional balance can be influenced by environmental factors, as was shown by House (1966b) with A. housei that the effects of dietary glucose level on larval survival and development could be modulated by temperature. The nutritive value of a basal medium (House 1966a) was increased by increasing the temperature from 20 to 25 and 30°C at glucose levels between 0-1.5%. At higher glucose levels larval survival and development were reduced with increasing temperature. Two media of different composition were formulated whose superiority for larval growth and development of A. housei was reversed at two different temperatures (15 & 30°C) (House (1972). Such nutritional effects might have ecological significance in affecting insect host range (House 1966b). It was thought that in establishing host range, an insect might be affected differently if the nutrient composition of its food were uniform but the temperature varied within the range rather than if the temperature were uniform but the composition of food was variable. On the other hand, the insect might not be affected if variation in food composition was accompanied by compensatory changes in temperature. Therefore, an insect species that attacks a particular foodstuff in a region with a specific temperature might, if introduced into another area with a different temperature, adapt to a different food source whose nutrient composition is favored at the new temperature .

Non-nutritional factors are intimately and intrinsically involved in food acceptance and ingestion. These include physical properties such as form, texture, etc., but also non-nutritive chemicals that elicit specific behavioral and/or physiological responses essential for finding and accepting foodstuff and in some cases for initiating behaviors associated with the feeding process itself (Bernays & Simpson 1982, Bernays 1985). Although such factors have been best shown in phytophagous insects, they also play a role in the biology of entomophaga and will likely be of importance in the development of continuous in vitro culture.

Predator Culture In Vitro

The artificial rearing of predators has stressed maintenance of the adult stage for maximizing egg production rather than complete in vitro culture. Predator larvae are the preferred biological control agent, and eggs and larvae produced by adults are placed directly in the field. However, some effort has been aimed at complete artificial culture of predators.. Among the first reared artificially from egg to adult was the coccinellid Coleomegilla maculata maculata (DeGeer) by Szumkowski (1952). Adults fed on raw liver or meat being kept for months on these food in the absence of prey. However, survival of larvae was poor on meat products alone and only 38% reached the adult stage. Supplementing vitamins resulted in ca. 86% of the larvae reaching adults. Oviposition and egg viability were increased by addition of vitamin E to the adult diet. The culture methods were refined and a diet of fresh yeast and glucose supported larval development (Szumkowski 1961a,b). Smith (1965, 1966) reared several coccinellid species including C maculata lengi on dried aphids supplemented with pollen. Success also was achieved on a diet of 40% brewer's yeast, 55% sucrose, inorganic salts, cholesterol, RNA, wheat germ oil and vitamins. Adults were fed the same diet supplemented with powdered liver. Attallah & Newsom (1966) reared 8 generations of this coccinellid on a defined diet of casein, sucrose, wheat germ, soybean hydrolysate, glycogen, butter fat, corn oil, a liver factor, dextrose, cotton leaf extract (with carotenoids and steroids), brewer's yeast, ascorbate, inorganic salts, vitamins and agar. Adults reared in vitro were fecund and mating was stimulated by addition of vitamin E to the diet. The medium failed to support growth of Coccinella novemnotata Herbst, Cycloneda spp., Hippodamia convergens Guérin and Olla v-nigrum (= abdominalis) (Mulsant). The last species was successfully cultured in vitro by Bashir (1973). Optimum egg production required inclusion of vitamin E in the larval diet, which was in contrast to the results of Szumkowski (1952) where supplementation of the adult diet alone was insufficient for maximum egg production.

Several coccinellid species were reported to be successfully cultured in vitro by Smirnoff (1958). These included Psyllobora (= Thea) virgintiduopunctata (L.), C. septempunctata, Oenopia (= Harmonia) doublieri (Mulsant), O. (= Harmonia) conglobata (L.), Rhizobius lophantae (Blaisdell), R. litura (F.), Rodolia cardinalis (Mulsant), Exochomus anchorifer Allard, E. quadripustulatus (L.), E. nigromaculatus Erhorn, Scymnus suturalis Thunberg, S. pallidivestis Mulsant, S. kiesenwetteri Mulsant, Stethorus punctillum Weise, Chilocorus bipustulatus (L.), Clitostethus arcuatus Rossi, Pharoscymnus numidicus Pie, P. ovoideus Sicard and Mycetaea tafilaletica Smirnoff (Endomychidae). The diet contained sucrose, honey, alfalfa flour, yeast, royal jelly and agar supplemented with dried pulverized prey. Larval rearing in a few species was improved by adding beef jelly. All species developed more rapidly and lived longer on the artificial diet compared with insects reared under natural conditions, and the adults were very healthy. Harmonia axyridis (Pallas), C. septempunctata and Chilocorus kuwanae Silvestri were reared on Smirnoff's (1958) diet and other artificial media by Tanaka & Maeta (1965). Successful culture of all species was obtained but adults failed to lay eggs. Chumakova (1962) reared Crytolaemus montrouzieri Mulsant on similar crude diets supplemented with dried prey.

Okada et al (1971a, 1972) and Matsuka et al. (1972) successfully reared H. axyridis on diets containing powdered larvae and pupae of drone honeybees (Apis mellifera L.). Sixteen generations of H. axyridis and three generations of Menochilus sexmaculatus (F.) were cultured in vitro. Okada & Matsuka (1973) and Matsuka et al. (1982) later improved the rearing method for maintaining adult Rodolia cardinalis. Chilocorus rubidus Hope, Scymnus hilaris Motschulsky, S. otohime Kamiya, Vibidia duodecimguttata Poda and S. hilaris adults were maintained on the diet but did not lay eggs (Matsuka et al. 1972). Niijima et al. (1986) described the use of drone honeybee powder for rearing several coccinellids including A. bipunctata, Anatis halonis Lewis, Coccinella explanata Miyatake, C. septempunctata, Coccinula crotchi (Lewis), Eocaria muiri, H. axyridis, Harmonia yedoensis Takizawa, H. convergens, Hippodamia tredecimpunctata L., Lemnia beplagiata (Swartz), M. sexmaculatus, Propylea japonica, S. hilaris and S. otohime. Variable results were obtained, but 11, 16 and 25 successive generations of E. muiri, H. axyridis and M. sexmaculatus respectively were cultured from the egg to adult stage. Larval development, adult longevity and fecundity were satisfactory.

The fractionation of honeybee powder was described by Matsuka & Okada (1975) who found that the active factor stimulating predator growth was unstable but nonproteinaceous. Expanded attempts to analyze bee powder was described by Niijima et al. (1977). Niijimi et al. (1986) then formulated several chemically defined diets for rearing H. axyridis. Larvae developed from the 1-3rd instar on a diet containing 18 amino acids, sucrose, cholesterol, 10 vitamins and 6 minerals.

Kariluoto et al. (1976) described rearing of A. bipunctata. About 60 variations of seven artificial diets were tested. These contained varying amounts of wheat germ, brewer's yeast, casein, cotton-leaf extract, egg yolk, sucrose, liver fractions, honey, glycogen, soybean hydrolysate, butter fat, corn oil, amino acids, dextrose, ascorbate, choline, inorganic salts, vitamin E and antibiotics. The best diets yielded 60-80% of larvae that became adults, but development time was slowed and adult weight lowered. Kariluoto (1978) modified the medium, and Kariluoto (1980) obtained fecund adults of A. bipunctata, C. septempunctata and others reared in vitro.

In vitro culture attempts with Chrysopa species did not succeed until Hagen & Tassan (1965) got a complete culture of Chrysoperla carnea (Stephens) on an encapsulated liquid medium (in paraffin droplets). The diet consisted of enzymatic yeast, protein hydrolysate, ascorbate, fructose, choline and casein hydrolysate. Adults were fecund but development time from the egg stage was ca. 2X that of insects reared on aphids. Vanderzant (1969) then successful cultured C. carnea for 7 generations on pieces of cellulose sponge soaked in enzymatic casein and soy hydrolysates, fructose, inorganic salts, lecithin, cholesterol, choline, ascorbate, vitamins and inositol. Development on this diet was slow, but 50-65% of larvae reached the adult stage compared with 85% when reared on natural insect eggs. Hassan & Hagen (1978) reported obtaining three generations of C. carnea on an artificial diet of honey, yeast flakes, sucrose, casein, yeast enzymatic hydrolysates and egg yolk. Developmental time and pupal weights were similar to those of insects on eggs of Sitotroga cerealella (Olivier). Chrysoperla sinica (Tjeder) was cultured for 10 generations on a diet of egg, brewer's yeast, sucrose, honey and ascorbate (Ye et al. 1979). Adults were fed powdered liver, honey and brewer's yeast. Cai et al. (1983) reared this species on an encapsulated medium of soybean and beef hydrolysates, egg yolk, sucrose, honey, brewer's yeast, ascorbate and linoleic acid, with similar success reported by Zhou & Zhang (1983).

The hemipteran predator, Geocoris punctipes may be reared on several diets (Dunbar & Bacon 1972). Media were nevertheless supplemented with insects. Cohen (1981) reported in vitro culture of G. punctipes from 1st stage nymph to adult on encapsulated semidefined diets. Six media containing casein hydrolysates, yeast, sucrose, cholesterol, corn oil, lecithin, agar, inorganic salts, phenylalanine and a vitamin mixture were formulated and encapsulated in different forms. The latter included mixtures of polybutene 32, dental impression wax, Vaseline, epoline C-16, candelilla wax, Sunoco, and Paraplast. Best results were with vitamin enriched medium encapsulated in a mixture of 5% polybutene 32 and 95% dental impression wax. Development of G. punctipes in vitro was better than when reared on Spodoptera exigua (Hübner). The percent of nymphs that reached adults and survival of the in vitro reared predators were significantly greater on the artificial diet. Cohen (1983) then described modifications of media content, preparation and encapsulation and could rear two generations of G. punctipes. Geocoris pallens Stal, H. convergens, H. axyridis and Nabis spp. also successfully fed on the encapsulated medium. In all cases superior results were obtained on medium encapsulated with 30% polybutene 32 and 70% dental wax. A diet composed of equal parts of fresh ground beef and beef liver supplemented with sucrose for continuous rearing of G. punctipes was produced (Cohen 1985). The ingredients were blended into a paste and small aliquots wrapped in stretched Parafilm presented to developing nymphs for feeding. Twelve generations were successfully cultured, and artificially reared predators displayed greater fecundity and adult weight than individuals reared on insect eggs and coddled larvae (Cohen & Urias 1986). Nevertheless, development was slower on the artificial diet.

Parasitoid Cultures In Vitro

In vitro culture offers a simple alternative for mass culture (Mellini 1978, Greany et al. 1984), and also enables dietary and nutritional manipulations for fundamental studies of nutrition and biochemistry. Some benefits of in vitro culture were given by Greany et al. (1984). However, the physiological and metabolic adaptations exhibited by insect parasitoids in relation to their parasitic way of life are of critical importance for successful in vitro culture (Mellini 1975a, Thompson 1981a, Grenier et al. 1986, Campadelli & Dindo 1987). Parasitoid/host relationships are often incorrectly thought to lack the complex physiological interactions typical of the host associations of other Metazoa (Thompson 1985, 1986a, Dindo 1987). The immature stages of many parasitoids are truly parasitic and such parasitoid/host relationships are characterized by extensive physiological and biochemical interaction (Beckage 1985, Thompson 1985, 1986a; Lawrence 1986a). Such interactions are often intimately associated with nutrition and successful development of the parasitoid in the host (Beckage & Riddiford 1983, Thompson 1983a, 1986a). The potential importance of the host endocrine system and of hormonal interaction in in vitro culture was discussed by Mellini (1975b, 1978, 1983) and Grenier et al. 1986). Greany (1986) discussed physiological interaction with reference to the culture of hymenopterous larval endoparasitoids. The extent that parasitoid/host physiological interactions need to be considered in the successful development of in vitro culture must still be determined but will undoubtedly vary with the parasitoid species.

Diptera.--A variety of natural foodstuffs, including fish and liver products, were utilized in early rearing attempts with parasitoids. House & Traer (1948) reared the sarcophagid A. housei for many generations on a diet of salmon and liver. Contrasted to 38% pupation among larvae reared on the host, Choritoneura fumiferana (Clemens), 88% pupated when reared on the artificial medium. A related species, Sarcophaga aldrici Parker was reared on the same medium and on liver alone (Arthur & Coppel 1953) and subsequently Coppel et al. (1959) maintained A. housei in the laboratory on fresh pork liver. About 1,000 A. housei larvae were reared on 1/2 lb. of sliced liver and were not affected by putrefaction of the tissue. Smith (1958) maintained Kellymyia kellyi (Aldrich) for 40 generations on pork liver and was also able to rear larvae on a mixture of powdered milk, powdered egg and brewer's yeast moistened with water to form a thick paste.

House (1954) developed the first chemically defined medium for rearing a parasitoid, using A. housei. The diet contained 19 amino acids, ribonucleic acid, dextrose, inorganic salts (U.S.P. XII), B vitamins, choline and inositol. It was prepared aseptically and gelled with agar. About 84% of the larvae reached the 3rd instar, 60% of those pupated and 32% of the pupae emerged as adults. The medium was later refined and many of the developmental nutritional requirements of A. housei were determined (House 1977). Vitamin E was necessary for reproduction (House 1966c).

Other dipterous parasitoids have been more difficult to culture outside the host. Many of these species have specialized physiological adaptations associated with parasitism that are lacking in sarcophagids. Tachinids, for example, have relatively high respiratory rates (Ziser & Nettles 1979, Bonnot et al. 1984) and during or immediately following the first stadium form a direct connection to the host's tracheal system (Kellen 1944, Fisher 1971). First instar larvae of the parasitoid E. bryani attach to the host's tracheal system 12 hrs after hatching, and respiratory considerations were critical for the development of in vitro cultures (Nettles et al. 1980). During initial studies, first instar larvae dissected from the host were placed directly in a liquid artificial diet. They were then transferred to diets gelled with agar, thereby exposing larvae directly to atmospheric oxygen. Improvements in the methods allowed development without transfer. Powdered artificial diet containing 1.5% agar was preconditioned by maintaining it at 5% RH for 24 hrs. The diet was then poured into petri dishes and held at 90% RH. Young larvae dissected from the host 18-24 hrs after larviposition fed on the liquid diet covering the surface of the gelled medium, and this was consistent with the normal feeding habit of first instar larvae that feed on and develop in the host's hemolymph. As the liquid was slowly absorbed by the agar gel, the surface of the gelled medium dried and larvae were exposed to the atmosphere. The artificial medium for rearing E. bryani was composed of mixtures of organic acids, amino acids, nucleic acid bases, B and fat soluble vitamins, phospholipids and derivatives as well as ATP, lactalbumin hydrolysate, bactopeptone, yeastolate, albumin, cholesterol, triolein, glucose and trehalose. When thus reared, larvae developed at an equivalent rate as when reared on the host, H. virescens, and 13% developed into adults with a sex ratio of ca. 66% females. Adults were fecund but produced fewer progeny than host reared insects. The medium was later refined and simplified and some of the basic developmental nutritional requirements of E. bryani were determined (Nettles 1986a). The nutritive values of adding albumin or soy flower to the medium was tested, which greatly increased adult yields and fecundity (Nettles 1986b).

Other tachinid parasitoids have been successfully reared on artificial media. Larval development of Phryxe caudata Rondani to the 3rd instar was obtained with a liquid artificial diet (Grenier et al. 1974). However, in contrast to the results of Nettles et al. (1980) with E. bryani, development of P. caudata was not improved by rearing larvae on gelled diets (Grenier et al. 1975). It was suggested that this may have resulted from the slower development rate and respiratory requirements of the latter when reared in vitro (Nettles et al. 1980). Bonnot (1986) discussed the importance of respiratory requirements in the in vitro culture of P. caudata. The first tachinid that was successfully cultured in vitro on artificial media from the first instar larvae to the adult was Lixophaga diatraeae (Townsend) (Grenier et al. 1978). This medium contained organic acids, amino acids, B and fat soluble vitamins, gelatin, enzymatic hydrolysates of casein, soy protein, lactalbumin, ovalbumin, ATP, cholesterol, lecithin and gelled with agarose. Adults were fecund and their progeny developed normally on Galleria mellonella. One critical factor for successful development of both P. caudata and L. diatraeae was osmolality, which could not exceed 450 mOs/Kg (Grenier et al. 1986).

Grenier (1979) investigated the embryonic development of P. caudata and L. diatraeae on artificial media. Newly fertilized eggs were removed from adult females and placed on an agarose-gelled medium similar to that for the larvae. Larval yield was equal to that observed in vivo and was much greater than when reared on a liquid diet. Again, respiratory requirements seemed critical for success.

Hymenoptera.--Simmonds (1944) made the first attempt to rear hymenopterous parasitoids in vitro. Three species of ichneumonid ectoparasitoids were maintained as larvae for extended periods on raw beef and gelatin. Although some growth was observed, none could complete their development. Bronskill & House (1957) did succeed in rearing C. turionellae on a slurry of pork liver in 0.8% saline. An autoclaved homogenate of the liver was dispensed into sterile test tubes and surface sterilized eggs were dissected from host pupae and transferred to this medium. Mature larvae were placed in gelatin capsules for pupation and 7% of the eggs developed to adults. When reared naturally on G. mellonella, 50% parasitoid adults were obtained. Culture of the ichneumonid I. conquisitor on a diet similar to that developed by House (1977) for A. housei was obtained by Yazgan & House (1970). The first holidic diet for rearing a hymenopterous parasitoid in vitro was reported by Yazgan (1972) for I. conquisitor. The diet was a mixture of amino acids, fatty acids, fat soluble vitamins, B vitamins and lipogenic growth factors, and glucose, RNA and gelled with agar. It was ground into a viscous slurry. Parasitoid eggs dissected from the host were placed directly on this medium, and development from egg to fecund adult was obtained with a development time twice that observed on the natural host, G. mellonella. Exeristes roborator was reared on a diet with a similar nutrient composition (Thompson 1975), but unlike I. conquisitor, larvae of this parasitoid would not tolerate direct contact with gelled media. Direct exposure to atmospheric oxygen was important for successful in vitro culture of E. roborator and success was achieved by retaining suspensions of the liquid diet in lipipholic Sephadex LH-20 gel filtration medium. Mortality, size and development time of the parasitoid reared in vitro were similar to those of individuals reared on Pectinophora gossypiella (Saunders). Many of the developmental nutritional requirements of I. conquisitor and E. roborator were determined by Yazgan (1972) and Thompson (1976a,b).

Thompson (1980, 1981d) described the various chemically defined diets for rearing various chalcids of the genus Brachymeria. Complete development of B. lasus from egg to adult at rates approximating those observed in G. mellonella were obtained on diets containing heat-denatured albumin, amino acids, glucose, B vitamins, inorganic salts, lipogenic growth factors and Intralipid. The latter, a phospholipid emulsion of soybean oil, was necessary for complete development. Larvae were reared from eggs dissected from host pupae immediately following oviposition and parasitoids were cultured individually in the wells of micro tissue culture plates. Development of larvae was ca. 2X as long on the synthetic medium compared to the insect host, and ca. 80% reached the active adult stage. Interestingly, the yellow coloration of the femur did not develop if vitamin A was lacking.

A critical factor in formulating the artificial media for B. lasus was osmotic pressure (Thompson 1983b). The effect of both carbohydrate and amino acid levels was similar and appeared closely related to osmolality. Optimum osmotic pressure in the artificial diets ranged from 550-700 mOs/Kg which was much greater than the 350-450 mOs/Kg of host hemolymph and tissues.

Complete development of the pteromalid Pachycrepoideus vindemiae (Rondani) was not obtained on an artificial medium similar to that used successfully for in vitro culture of Brachymeria (Thompson 1981c). When the amino acid component was replaced with a mixture of the corresponding polymerized amino acids and the osmolality was reduced to ca. 390 mOs/Kg, development from egg to adult was obtained (Thompson et al. 1983c).

These studies demonstrate that the importance of osmotic pressure varies with the parasitoid species. Parasitoids such as I. conquisitor and E. roborator are very tolerant of osmotic pressures. Artificial diets that supported in vitro culture of these species had osmolalities of ca. 2,000 mOs/Kg. On the other hand, the tachinids, P. caudata and L. diatraeae (Grenier et al. 1986), and the pteromalid P. vindemiae did not develop at osmolalities of >450 mOs/Kg.

Pteromalus puparum was cultured in vitro by Bouletreau (1968, 1972). Complete development on host hemolymph in hanging drop slide mounts was obtained. Similar results were reported by Hoffman et al. (1973). Hoffman & Ignoffo (1974) had limited success with an artificial medium containing yeast hydrolysate, fetal bovine serum and Grace's tissue culture medium.. Tetrastichus schoenobii Ferriere was reared on modified Gardiner's tissue culture medium supplemented with egg yolk, milk and hemolymph from Anteraea pernyi Guérin (Ding et al. 1980a). About 60% of the parasitoids completed development to the adult stage with no deformities nor abnormal fecundities. Greany (1980, 1981) described studies on the in vitro embryonic development of the braconid Cotesia (= Apanteles) marginiventris (Cresson), maintained in Grace's tissue culture medium supplemented with fetal bovine serum, bovine serum albumin and whole egg ultrafiltrate. Insects were reared from the embryonic germ band stage to mature first instar larvae on this diet cocultured with host fat body tissue. Greany (1986) obtained similar results with Microplitis croceipes. Emphasis was placed on the importance of protein nutrition for success and protein secretion by the fat body was considered a factor to explain the need for this tissue for successful embryonic development.

Vinson & Iwantsch (1980) found that teratocytes (cells derived from the embryonic membrane of the parasitoid egg) are released into the host hemocoel at the time of egg hatching. It was suggested that the teratocytes may play a role in parasitoid nutrition. Sluss (1968) demonstrated that the teratocytes of Perilitus coccinellae (Shrank) increased in volume several times in the coccinellid host and where then subsequently eaten by the developing parasitoid larvae. Greany (1980) found that teratocytes present in artificial culture medium for C. marginiventris caused dissociation of cocultured fat body and suggested that the teratocytes might facilitate larval growth. Rotundo et al (1988) obtained complete larval development of the braconid Lysiphelebus fabarum (Marshall) on a similar artificial diet that was lacking in fat body and teratocytes.

Strand et al. (1988) demonstrated a role for teratocytes in the successful in vitro culture of the egg parasitoid Telenomus heliothidis Ashmead. Embryonic development of T. heliothidis was obtained in Hinks TNH-FH medium containing 30% w/v M. sexta hemolymph. Mature embryos were transferred to a medium containing 40% M. sexta hemolymph, chicken egg yolk, trehalose and milk. Development to the adult stage required one day more than on the host H. virescens and 42% of the larvae became adults. The sex ratio was ca. 50% females. The presence of teratocytes had no effect on larval development to the third instar. However, when teratocytes were removed from the medium during larval development, pupation was greatly reduced and the development time of parasitoids that completed development increased. The authors concluded that the teratocytes aided larval feeding by dispersing the particulate material in the medium and solubilizing nutrients. It was suggested by Strand et al. (1986) that teratocytes of T. heliothidis aided in decomposition and necrosis of host tissue partially due to release of lytic enzymes. Therefore their function in vitro might be the same that occurs during the normal development of the parasitoid in the host, Heliothis virescens.

Culture of Trichogramma pretiosum in vitro was first attained by Hoffman et al. (1975) following unsuccessful attempts by Rajendram (1978) with T. californicum Nagaraja & Nagarkatti. Trichogramma pretiosum completed development on filter paper discs soaked in sterile H. zea hemolymph. In vitro culture to the adult stage required ca. 25% more time than observed on the host, Trichoplusia ni (Hübner). Even though most adults did not fully expand their wings, they mated and laid eggs without difficulty. Progeny from eggs of in vitro cultured parasitoids had a sex ratio of 1.2:1 males/females when reared on host eggs. Hoffman et al. (1975) reported development to the prepupal stage on a semisynthetic artificial diet similar to that described by Hoffman & Ignoffo (1974) for P puparum, but supplemented with wheat germ oil. Strand & Vinson (1985) obtained complete in vitro culture of T. pretiosum on an artificial medium similar to that used by Thompson (1981d) for B. lasus but supplemented with ca. 40% M. sexta hemolymph, the latter being required to induce pupation. Survival to the adult stage was 70s% and the sex ratio ca. 1:2 males/females. Xie et al. (1986a) also reported that host hemolymphs was required for pupation of T. pretiosum and that factors in the host egg influenced adult emergence. Irie et al. (1987) reported that the requirement of host hemolymph for the complete in vitro development was due to the presence of specific factors that could be extracted in 76% ethanol. Purification of the pupation factor by chromatographic methods showed the presence of two active carbohydrate containing factors.

Trichogramma dendrolimi Matsumura was cultured in vitro in hanging drop mounts of hemolymph from A. pernyi Guan et al (1978). Liu et al (1979) reported success in hanging drop mounts containing media with A. pernyi or Attacus cynthia (Drury) hemolymph and chicken egg yolk, bovine milk, organic acids and procine serum. The extent of development of Trichogramma japonicum Ashmead, T. australicum Girault and T. evanescens was not reported, however. Wu et al. (1980, 1982) and Wu & Qin (1982a) obtained successful culture of T. dendrolimi to the adult stage on media without host hemolymph but containing chicken egg yolk, chicken embryo fluid, bovine milk, and amino acid mixture and peptone. However, only 16% of the eggs completed development, and most adults were females of poor vitality. The results did suggest that in contrast to T. pretiosum, the in vitro culture of T. dendrolimi does not require host factors (Xie et al. 1986a, Irie et al. 1987). Liu & Wu (1982) reported on in vitro culture of T. dendrolimi, Trichogramma confusum Viggiani and T. pretiosum on a medium of yeast hydrolysate, fetal calf serum, Grace's tissue culture medium, chicken embryo extract, bovine milk and chicken egg yolk. However, adults were less viable than normal and displayed abnormal wing development. The cooperative Research Group of Hubei Province, China (CRGHP 1979) has carried out extensive studies on the complete in vitro culture of T. dendrolimi in artificial media encapsulated in artificial eggs into which the adult females oviposited. Gao et al (1982) reported rearing 35 continuous generations of this species in hanging drop mounts of the artificial medium.

Some studies have tried to determine the effects of hormone supplementation on parasitoid development in vitro, with generally negative results. The tachinid Gonia cinerascens Rondani depends on its host's endocrine system for growth and development, but was not induced to mold from the 1-2nd instar by addition of 20-hydroxy (b) ecdysone to an artificial medium of host tissue homogenate and Grace's tissue culture medium. Development from the 2nd instar to adult was reported on artificial medium in the absence of hormones, indicating that some hormones may be necessary for the 1-2nd instar molt in vitro. The 20-Hydroxy ecdysone failed to stimulate development of B. intermedia in vitro (Thompson 1980); however, Greany (1980, 1981) reported that this hormone inhibited egg hatching in C. marginiventris and ecdysone, 20-hydroxy ecdysone and the juvenile hormone analog hydroprene had no effect on larval growth or development. The deleterious effect of this hormone could be overcome by simultaneous application of hydroprene.

Nenon (1972a,b) demonstrated that hormones greatly increased in vitro survival of developing embryos and larvae of the encyrtid Ageniaspis fuscicollis (Dalman). The parasitoid was maintained on a diet of chicken embryo extract, beef peptone and equine serum. Ecdysteroid or juvenile hormone added in the medium had little effect, but when included together, resulted in nearly 100% survival to the 2nd instar. Further study of the effects of host hormones in vitro culture systems must require careful and detailed experimental design. Hormones act in a complex and often synergistic way, and the timing of their application as well as the method of exposure may prove critical to assessing their potential. There is no doubt as to the importance of hormonal interaction to the successful development of parasitoids in vivo, particularly with regard to synchronizing parasitoid development to the host's life cycle.

Adult Entomophage Nutrition

Adults of many entomophaga must feed, and although adult parasitoids and predators are usually fed in the laboratory, the significance of such feeding in nature had been largely ignored by early workers. Bierne (1962) considered that many biological control attempts failed as a result. Leius (1967a) gave one of the first field demonstrations of the importance of adult feeding when he reported a relationship between the natural abundance and variety of wild flowers in apple orchards and the incidence of parasitism of Malacosoma americanum (F.) and Laspeyresia (= Carpocapsa) pomonella (L.) by the parasitoids, I. conquisitor, Apophua (= Glypta) simplicipes (Cresson), Scambus hispae Harris, Telonomus sp., Ooencyrtus clisiocampae (Ashmead), and Eupelmus spongipartus Foerster. Eighteen times as many M. americanum pupae, four times as many M. americanum eggs and five times as many L. pomonella eggs were parasitized in orchards with an undergrowth of wild flowers when compared with other orchards lacking such flora.

The early literature describing how adult parasitoids feed from flowers and other plant parts was reviewed by Leius (1960). Generally insects fed on floral and extrafloral nectars as well as pollens. Although knowledge of the specific nutritional requirements of adult entomophagous insects is limited, much data are available on the chemical and nutritional requirements of adult entomophaga is limited, much is available on the chemical and nutritional composition of these plant products. Floral nectars contain up to 75% by weight of simple sugars, mainly sucrose, fructose and glucose (Baker & Baker 1983), but considerable qualitative and quantitative differences exist between plant species. Free amino acids are also abundant in nectars although most nectars do not contain all 10 essential amino acids. Small amounts of proteins, lipids, dextrins and vitamins that are nutritionally beneficial are also found. The composition of extrafloral nectars is also complex (Baker et al. 1978). Pollens have a complex composition of small molecular nutrients and many pollens have high levels of free amino acids (Barbier 1970, Stanley & Linskens 1974). By comparison, pollens generally have higher levels of protein, lipid and polysaccharides. Pollens and nectars together can provide a complete diet for successful growth, development and reproduction. The predator Coleomegilla maculata lengi Timberlake can complete larval development on pollen alone (Smith 1961); therefore, when prey are scarce, plant products may play a critical role in maintaining predators (Hodek 1973). Hagen (1986a) discussed the complex ecological and evolutionary interactions between plant flowers, nectars and pollens and several insect groups.

Leius (1960) examined the plant feeding habits of I. conquisitor, Scambus buolianae (Hartig) and Orgilus obscurator (Nees). The attractiveness of the flowers of wild mustard, white sweetclover, wild parsnip, silky milkweed and annual sowthistle were tested. Except for annual sowthistle, I. conquisitor was attracted to and fed from all flowers tested, but was most attracted to wild parsnip. Similar results were shown with S. buolianae. Orgilus obscurator was attracted to and fed on wild parsnip only, but further tests revealed that this parasitoid also fed on other umbelliferous plant flowers, including those of wild carrot and water hemlock. The nutritive value of various pollens for fecundity and longevity of S. buolianae was reported by Leius (1963). Itoplectis conquisitor and S. buolianae accepted various semi-natural foods also, including honey, sucrose solution with or without plant pollens and raisins. Plant feeding behavior of O. obscurator examined by Syme (1975) showed a broad range of food plants, including species from five families. Adult parasitoids may emerge prior to the availability of the insect host, and Syme (1977) suggested that a variety of plant species be provided as food to ensure sufficient longevity of the adult female.

Lingren & Lukefar (1977) demonstrated that adult Campoletis sonorensis (Cameron), a parasitoid feeding on the extrafloral nectar of cotton, lives longer when exposed to extrafloral nectaried cotton than nectariless cotton. Parasitism of hosts was higher on the nectaried form. Adejei-Maafo & Wilson (1983) showed that 15 categories of entomophaga, including the predators Deraeocoris signatus (Distant), Geocoris lubra (Kirkaldy), Nabis capsiformis Germar, Chrysopa spp., Laius bellalus Guérin, Coccinella repanda (Thunberg) and Verania frenata Erichson, were present at densities of 2-3 times higher on nectaried versus nonnectaried cotton. Although semiochemicals contribute to attraction for plants in these insects, the nutrition provided by nectars and pollens seems to be important. Hemptinne & Desprets (1986) reported that following hibernation Adalia bipunctata (L.) fed on pollens as an alternate food which allows the predators to lay eggs as soon as prey become available.

As was discussed in an earlier section, in addition to feeding plants and plant products many parasitoids are host-feeders. Adult female Hymenoptera puncture or damage host larvae or pupae and feed on the hemolymph and/or internal tissues. Kidd & Jervis (1989) estimated that as much as 1/3rd of the world's parasitoid fauna (>100,000 species) host feed. Some parasitoids may kill more host individuals by host feeding including ovipositor probing followed by host rejection, than by parasitization (Johnston 1915, DeBach 1943, 1954). Legner (1979) emphasized that consideration of a parasitoid's host destructive capacity was important to correctly evaluate the impact of periodic inundative field releases on pest populations, and Greathead (1986) and Yamamura & Yano (1988) suggested that host feeding behavior was important for assessing the potential of a biological control agent. Kidd & Jervis (1989) recently discussed the significance of host-feeding on parasitoid-host population dynamics.

Bartlett (1964) in examining host-feeding in the encyrtid, Microterys flavus Howard, was among the first to correlated host-feeding behavior with nutrition. He hypothesized that host feeding developed coincidentally with depletion of eggs and suggested that host mutilation was a reflection of "frustrated" host feeding when the host failed to bleed readily. Host feeding by M. flavus was usually displayed following egg-laying, and oviposition resumed after host feeding. Reviewing this predatory habit for adults from 20 families of Hymenoptera, Bartlett concluded that the behavior was indicative of the necessity for dietary supplementation of some ubiquitous substances required by many diverse species. He reported that a food supplement of enzymatic yeast and soy hydrolysate with honey satisfied the nutrient requirements for sustaining reproductive activity in M. flavus, and suggested that a protein nutrient source may be necessary.

The difference between proovigenic and synovigenic Hymenoptera was discussed earlier, categories proposed by S. E. Flanders (1950). Females of proovigenic parasitoids complete oogenesis prior to or shortly after emergence and lay eggs over a relatively short period of time principally on larval stages of their host. Host feeding is important for ensuring that the female lives long enough to deposit all eggs. In contrast, females of synovigenic species eclose with a minor fraction of their total egg complement as mature eggs. Synovigenic parasitoids attack primarily host eggs and pupae, are longer lived than proovigenic species and produce eggs throughout their adult lives. To sustain oogenesis the females of many synovigenic species require additional nutrients. Based on the egg type, Dowell (1978) described two types of synovigenic parasitoids: (1) those producing large anhydropic or yolk-rich eggs that contain sufficient nutrient for completion of embryonic development prior to oviposition. Parasitoids that produce anhydropic eggs obtain nutrition for sustaining egg production by host-feeding; (2) those producing hydropic or yolk-deficient eggs. Embryonic development in hydropic eggs occurs in the host following oviposition, in which case the adult does not require additional nutrient to support egg development and has no requirement to host feed. Legner & Gerling (1967) showed the importance of early host feeding and oviposition to pteromalids of the first type, as was previously discussed. Leius (1962, 1967b) demonstrated the importance of feeding habits to fecundity of S. buolianae. Egg production was reduced to 1/3rd and longevity to 2/3rds, when females were permitted to host-feed intermittently or were deprived after 15 days of age. No eggs were laid if females were deprived for 20 days. The effects of feeding host body fluids, in conjunction with honey, pollen and raisins on fecundity and longevity of S. buolianae and I. conquisitor were examined by Leius (1961a,b). Maximum fecundity and longevity of both species were obtained when host fluids and seminatural foods were provided together. Host feeding was nevertheless essential, and S. buolianae did not lay eggs when deprived of host hemolymph or tissues.

The feeding behavior of 140 hymenopterous parasitoids was also reviewed by Jervis & Kidd (1986), who concluded that host feeding was important for egg fecundity or egg production, while non-host foods were important for maintenance and longevity. Four types of host feeding distinguished were (1) concurrent feeding where the female used the same host individual for feeding and oviposition, (2) nonconcurrent if the female used different host individuals for feeding and oviposition, (3) the feeding habit may be nondestructive or destructive (the host may survive or may die), and (4) destructive feeding which generally resulted in a host that was unsuitable for oviposition. Parasitoids were found to differ in their lifetime and diurnal patterns of feeding, and it was concluded by Jervis & Kidd (1986) that concurrent/nondestructive feeding was most likely when hosts were readily available and that destructive feeding was advantageous when host density was low.

Jervis & Kidd (1986) also gave several models to assess how the energetic demands and constraints on a parasitoid affect its host-feeding strategy. One model predicted the feeding strategy for maximizing egg production of a single synovigenic female (see Thompson & Hagen 1999, for formulae).

Host feeding also occurs among dipterous parasitoids but is not as common as in Hymenoptera (Clausen 1940). Host feeding by tachinid parasitoids may affect longevity and fecundity (Shahjahan 1968)(. Nettles (1987b) demonstrated that fecundity was prolonged by feeding E. bryani host hemolymph compared with feeding a sucrose solution. The effect of host feeding on fecundity could not be simulated by substituting a solution of free amino acids or bovine serum albumin.

The excretion of various Homoptera, such as honeydew, may serve as a food for many adult entomophaga. Neuropterans of the genus Chrysoperla and other genera with nonpredaceous adults feed actively on honeydew as well as on nectar and pollen (Principi & Canard 1984). Although honeydew does not contain all the essential amino acids, yeast symbiotes residing in the gut can provide the missing amino acids in some nonpredaceous species (Hagen & Tassan 1972). Neuropteran predatory adults also feed on honeydew, but reproductive activity ensues only after prey are eaten (Hagen 1986a). Hagen (1962) found that honeydew alone will not stimulate egg production in coccinellid predators. Dipterous and hymenopterous parasitoids also have been found to feed on honeydew (Clausen 1940, Zoebelein 1956). The importance of honeydew as a supplementary food was suggested by Clausen et al. (1933) in work with Tiphia