BIOLOGICAL CONTROL IN FORESTS
----CLICK on desired underlined categories [to search for Subject Matter, depress Ctrl/F ]:
Some unique ecological attributes are present in relatively complex forest environments including a diversity of species, ages, intraspecific genetic composition, spacing and stocking levels (Dahlsten & Mills 1999). Intensively managed forests, even-aged stands, plantations of single and mixed species and seed orchards resemble agriculture, but even these usually exist in a variety of different conditions. It is important to look at some of these ecological attributes in detail as the opportunities for biological control vary depending on the environment and species involved.
In addition to timber production, forests serve as wildlife refuges, recreation, watershed and grazing areas. Where in agriculture the goal of management is to harvest a commodity one or more times a year, in forestry pest management is further complicated by multiple goals and competing interests, including sportsmen, environmentalists, bird watchers, hikers, cattlemen, woodcutters and the Army Corps of Engineers.
Forests tend to be extremely large continuous areas with gradual boundaries, thus quantitative evaluation of controls becomes very difficult and expensive. Control strategy in forests is also affected by the length of time to harvest, which may be 20 to 30 years or more in warm temperate areas to 50 to 100 years in colder areas. Compared to agricultural ecosystems, forests are much more complex ecologically. Forests vary from single species plantations to multistoried stands and plant diversity is greater than in an agricultural field even in the simplest forest stand. Researchers must often deal with stands trees exceeding 70 m in height, a mixture of age and size classes, a mixture of tree species, numerous canopy levels including herbaceous plants and different stocking levels or spacing..
In view of the ecological attributes of forest ecosystems, the choice and evaluation of biological control tactics may vary. The influence on the classical approach to biological control has been analyzed by Pschorn-Walcher (1977). The vast, diverse, relatively less disturbed, long-lived and highly stable in space and time ecosystem confers both advantages and disadvantages for biological control. Diversity confers an advantage for foreign exploration as a large complex of natural enemies is available from which to choose (Pschorn-Walcher 1977). However, this could also make it more difficult for colonization of new species of natural enemies. There would be expected to be a greater chance for the introduced natural enemies to be in competition with related native natural enemies since there is a high probability that relatives would be present in the rich forest fauna. The vastness and diversity create sampling and evaluation problems but less disturbance allows long term evaluations to be more exact.
The collector of natural enemies has an advantage in the relatively uniform forested regions because only minor regional differences are usually exhibited (Pschorn-Walcher 1977). However, any widely distributed pest or a pest introduced in a number of locations in a large forest region would make any colonization program long in term. Pschorn-Walcher (1977) maintains that the great differences between forest and agroecosystems dictate a different approach to biological control in forestry from agriculture. The approach to biological control in agriculture, where there is much less predictability because of continuous disturbance, can be faster using trial and error releases until the best natural enemy is found. With forest insects preintroduction studies are desirable in order to understand the interrelationships of the various natural enemies and finally to select the most likely natural enemies for success. Natural enemy complexes of forest insects can be chosen with a higher degree of predictability for successful introductions and therefore preintroduction studies are justified (Pschorn-Walcher 1977). Studying the parasitoid complex in detail provides information on those species that might be good colonizers, those that would operate at low or high population levels, those that were monophagous or polyphagous, those attacking early or late life stages, those that could adapt to some degree of inbreeding and could then withstand initial low number colonization, or prolonged laboratory rearing, and those that were cleptoparasitoids and then could be selected out .
A variety of approaches in biological control including importation, augmentation and conservation have been used. The major efforts have been in North America (Canada and the United States) and the classical approach of importation has been the most commonly used. Undoubtedly this is because the highest proportion of introduced forest pests occur in North America (Pschorn-Walcher 1977). The majority of insects are lepidopteran and hymenopteran defoliators (sawflies). Since these insects are relatively large hosts it may explain why 9 of the 15 tachinid flies established in biological control attempts were used in forests. It seems that Lepidoptera and Hymenoptera are more commonly pests in the less disturbed, contiguous forest regions. Also forests are not as intensively managed as agricultural ecosystems and it may explain why Homoptera, which are common subjects for biological control in agriculture, are not as common as forest pests.
Importation of Natural Enemies For Introduced Pests.--The most common approach in forestry has been the importation of natural enemies against introduced pests (Turnock et al. 1976, Pschorn-Walcher 1977). This has usually involved colonizing and establishing a relatively small number of natural enemies for control of an exotic pest through direct inoculative releases of newly imported parasitoids. With a few exceptions, parasitoids have been the preferred natural enemies introduced in forestry. Dahlsten & Mills (1999) gave some estimates of the numbers of importations of parasitoids and predators and their success of establishment and control. The data show that 78% of importations involved parasitoids (Hymenoptera or Tachinidae). Only homopteran pests have attracted substantial importations of predators and while the overall rates of establishment of these two groups of natural enemies are equal, the parasitoids have on average been more than twice as successful in achieving some degree of control of the target forest pests.
About 40 species predators were introduced against the balsam woolly aphid, Adelges piceae (Tatz) in an unsuccessful colonization program (Clark et al. 1971), there being no known parasitoids of this species. Attempts to introduce predators against bark beetles have been made on several occasions. Hopkins tried to introduce the clerid, Thanasimus formicarius (L.) from Germany to West Virginia for control of the southern pine beetle, Dendroctonus frontalis Zimm. in 1892-93. Although a complete failure, it was the first attempt to import a natural enemy of a forest insect into the United States (Dowden 1962). Other unsuccessful attempts have been made using Rhizophagus spp. from Britain both in Quebec, Canada in 1933-34 with one species against the Eastern spruce beetle, D. obesus (Menn.) and in New Zealand in 1933 with three species against the European bark beetle, Hylastes ater (Payk) (Clausen 1978). Success was reported in the Soviet Union with Rhizophagus grandis Gyll. against the European spruce beetle, D. micans Kugelann (Kobakhidze 1965, Grégoire et al. 1987). Both predator and host are native and this is a good example of augmentation through periodic inoculation. There are also recent projects in Britain and France with R. grandis against D. micans (Evans & King 1987, Grégoire et al. 1987). Several species of carabid beetles have been imported for control of the gypsy moth, Lymantria dispar (L.), with one species in particular, Calosoma sycophanta (L.) becoming well established (Clausen 1978).
Red wood ants were imported in North America on a few occasions (Finnegan & Smirnoff 1981). Formica lugubris Zett. was imported from Italy in 1971 and 1973 for forests in Quebec (Finnegan 1975), and Formica obscuripes Forel was moved from Manitoba to Quebec in Canada in 1971 and 1972 (Finnegan 1977). The 15 species in the Formica rufa L. complex in North America are not well known but F. obscuripes appeared to have potential and did not occur in the east (Finnegan 1977). The effectiveness of these introductions against defoliators such as the Swaine jack pine sawfly and the spruce budworm is unknown as yet, but the ant populations are still encouraged so that eventually they will be well established in a wide area. These ants have been observed feeding on spruce budworm and other forest insects (McNeil et al. 1978).
One species of vertebrate, the masked shrew, Sorex cinereus Kerr, was colonized in Newfoundland for control of the larch sawfly, Pristiphora erichsonii (Hartig). In a rather unique situation there were no insectivores and few small fossorial animals on the island. These shrews were transported from northern New Brunswick, Canada in 1958 and subsequently released. Shrews also feed on other insects and it is believed that the importation was successful even though there was some public opposition to the operation (Turnock & Muldrew 1971).
Classical biological control using pathogens has not been common in forestry. However, two exceptions are the accidental introduction of a nuclear polyhedrosis virus of the spruce sawfly, Gilpinia hercyniae (Htg.) into eastern Canada (McGugan & Coppel 1962). One nematode, Deladenus siricidicola Bedding was imported for control of the woodwasp, Sirex noctilio F. in Australia (Bedding & Akhurst 1974).
Dahlsten & Mills (1999) noted four cases where mass rearing and release programs were performed in the biological control of forest insects: (1) propagation of 882 million Dahlbominus fuscipennis (Zett.) at Belleville, Canada for Gilpinia hercyniae control (McGugan & Coppel 1962); (2) release of 200 million D. fuscipennis by the Maine Forest Service in the United States against G. hercyniae (Clausen 1978); (3) mass rearing and release of several parasitoids of the gypsy moth, Lymantria dispar, in the eastern United States (Leonard 1974); and (40 the use of a nematode, Deladenus siricidicola against Sirex noctilio in Australia (Bedding & Akhurst 1974).
Importation of Natural Enemies For Native Pests.--As mentioned earlier in other sections, exotic natural enemies may be used effectively against native organisms, even though the procedure is sometimes controversial. This approach was evaluated by Hokkanen & Pimentel (1984) who concluded that it ought to be the preferred approach in biological control. It stems from the idea that through genetic feedback mechanisms host-parasitoid systems evolve toward homeostasis and because of this coevolved equilibrium parasitoids would be limited in their effectiveness as biological control agents (Pimentel 1961, 1963). Generalists would probably be preferable to specialists in the selection of candidate agents. This approach must be done with extreme caution because the Pimentel genetic feedback concept is not wholly acceptable (Huffaker et al. 1971) as it is believed that natural enemies may become better adapted through time in controlling their hosts. To support this are examples of long standing and effective introduced natural enemies such as Rodolia cardinalis and Cryptochetum for control of cottony cushion scale and many others.
Nevertheless, the Hokkanen & Pimentel (1984) analysis concluded that success in biological control was about 75% higher for the new associations. These conclusions were disputed by Goeden & Kok (1986) using biological control examples. They explain that the data used included cacti, which are not representative of target weeds, and that there were inaccuracies with some other examples. Dahlsten & Whitmore (1987) analyzing the 286 examples of successful biological control used by Hokkanen & Pimentel (1984) showed that there was a significant advantage for old associations in terms of complete versus intermediate versus partial success. The use of new associations as the preferred method for biological control is also contradicted by the analyses of Hall & Ehler (1979) and Hall et al. (1980), who found that the establishment rate of natural enemies was significantly higher for introduced pests, the complete success of importations against introduced pests was higher but not statistically significant and the general rate of success for introduced pests higher than for native pests. There appear to be some other misinterpretations in the data of Hokkanen & Pimentel (1984) who used the reference by Clausen (1978) for much of their information. These include the case of the elm leaf beetle, Xanthogaleruca luteola (Müller) and some native Neodiprion sawflies that were controlled by natural enemies in new associations (see Clausen 1956, McGugan & Coppel 1962, DeBach 1964b, Bird 1971, McLeod & Smirnoff 1971, Cunningham & DeGroot 1981, Finnegan & Smirnoff 1981, Laing & Hamai 1976, Clair et al. 1987).
It is encouraging that there are examples of successful introductions of natural enemies for control of both exotic and native pests. Each approach has merit depending on the ecological circumstances. They state that the sawfly examples of efforts against native species are good examples of what can be done. Also the extremely successful project using a parasitoid from a host in a different genus in North America against a native geometric moth, Oxydia trychiata (Guenée), Colombia is a good case. Biological control efforts against native species through the importation of exotic natural enemies or by periodic inoculation of native natural enemies have merit according to Carl (1982). Several ongoing (1996) examples of careful evaluations for Canada are the Douglas-fir tussock moth, Orgyia pseudotsugata (McDunn) (Mills & Schoenberg 1985), the spruce budworm, Choristoneura fumiferana (Clemens) (Mills 1983a) and bark beetles (Mills 1983b, Moeck & Safranyik 1984).
Augmentation of Natural Enemies.--As discussed in an earlier section, the effects of natural enemies can be enhanced by various manipulations of the organisms themselves or by alteration of their environment, such approaches being extremely promising for native pests. Although augmentation and conservation can be distinguished theoretically, it is difficult to distinguish them in practice (Rabb et al. 1976). The two tactics were defined by DeBach (1964c) as to manipulation of natural enemies themselves (augmentation) or their habitat (conservation). Neither approach has been used extensively in forestry, most literature being from agriculture (DeBach & Hagen 1964, van den Bosch & Telford 1964, Rabb et al. 1976, Stern et al. 1976, Ridgway et al. 1977). Augmentation is either by periodic colonization or inoculation, development of adapted strains by artificial selection or inundation (DeBach & Hagen 1964). The tactic may involve either entomopathogens, parasitoids or predators.
Attempts have been made with inoculation of several parasitoids against forest pests in Europe and South America (Turnock et al. 1976). Inoculations were made of Rhizophagus against D. micans in Russia, France and Britain and of the nematode Deladenus against S. notilio in Australia. Red wood ants (Formica spp.) have been moved and relocated in Europe where they are considered to be effective predators on a number of forest pests. Otto (1967) reviewed a number of the programs and concluded that good results were obtained primarily in pin forests against dipterous and lepidopterous larvae. Ants are less effective against sawflies and ineffective against beetles. Effective protection of coniferous forests using ants has been achieved against five lepidopterans and three sawfly pests in Germany, Switzerland, Italy, Russia, Poland and Czechoslovakia (Otto 1967, Turnock et al. 1976). Three to eight species in the Formica rufa complex in Europe are considered to be good biological control agents, with identified species being F. polyctena Forst, F. lugubris, and F. aquilonia Yarrow.
Some examples of parasitoid inundation include Trichogramma minutum Riley against the brown-tail moth, Euproctis chrysorrhea L. in North America (DeBach 1964c, Howard & Fiske 1911); Trichogramma spp. for control of various forest defoliators in Germany and Russia, and Telenomus verticillatus Kieff. against the lasiocampid, Dendrolimus pini (L.) in the Soviet Union (DeBach 1964c). In China inundative releases of Trichogramma spp. are made routinely against various forest defoliators, which is facilitated by a large and economic labor force (McFadden et al. 1981).
Diprion pini (L.) has been successfully controlled in Spain by the collection of sawfly cocoons which were then either placed directly in special emergence cages or exposed to Dahlbominus fuscipennis in the laboratory before return to the field. Parasitoid emergence from these cages contributed about 3 million additional D. fuscipennis and ichneumonids, Exenterus oriolus Htg to the forests, producing about 65% parasitism (DeBach 1964c, Ceballos & Zarko 1952).
Various pheromones and kairomones have been identified for hosts and natural enemies that are considered for implementation in natural enemy release programs to enhance their performance (Haynes & Birch 1985, Borden 1982, 1985, Vinson 1984). Mills (1983b) suggested the use of Dendroctonus aggregation pheromones as a way of selecting useful European bark beetle egg predators for introduction into Canada. Miller et al. (1987) have shown Thanasimus undulatus Say to exhibit cross-attraction in field tests to other bark beetle pheromones and Rhizophagus grandis to be attracted to the frass of three North American Dendroctonus species in the laboratory. Moeck & Safranik (1984) concluded that inundative releases of native clerid beetles against low levels of D. ponderosae offered a good potential.
Bird encouragement programs have been used extensively in Europe by providing nesting boxes in forests for cavity nesting spots (Bruns 1960). In California, Dahlsten & Copper (1979) demonstrated that populations of the mountain chickadee can be increased two to three fold with nesting boxes. It is speculated that bires operate in an inverse density-dependent manner and their importance would be in preventing outbreaks of forest pests rather than in suppressing them.
Since 1980 entomopathogens have begun to play a dominant role in forest biological control. The principal entomopathogens used are the bacterium Bacillus thuringiensis Berl., and baculoviruses. These agents have been tested against a wide variety of forest defoliators in the form of inundative treatments and have the advantage of reduced impact on other groups of natural enemies and non target organisms. Morris et al. (1986) point out that microbial insecticides are likely to receive as wide an application in forestry as in agriculture for several reasons. Forest protection is of much greater concern to the general public due to the more extensive areas covered by forest pests. Forest pest problems also tend to involve only single target species rather than a complex of pests, which requires the development of only a single microbial product. The forest crop is also better able to withstand the slower action of microbial treatments in comparison with agricultural crops. The spruce budworm, Choristoneura fumiferana in North America and the gypsy moth both in Europe and North America have been the main targets of extensive development of Bacillus thuringiensis as a means of inundative biological control. More consistent success has been attained against the spruce budworm and guidelines have been formulated (Morris et al. 1984).
Baculoviruses, which include the nuclear polyhedrosis viruses (NPV) and granulosis viruses (GV), have been widely tested in field trials against forest insects (Cunningham 1982). They show a marked degree of specificity for their phytophagous hosts and have no impact on non-target organisms. Natural epizootics of NPV are often responsible for the termination of outbreaks of major forest insect pests, particularly among the Diprionidae and Lymantriidae. The diprionid sawflies provide some of the most striking examples of the use of NPVs in biological control (Cunningham & Entwistle 1981). The virulence of the diprionid NPVs is appreciably greater than that of other host groups (Entwistle 1983) and the gregarious habit of the diprionid larvae promotes the spread of virus through the larval population.
Virus production cannot be achieved on artificial media and for sawflies, in contrast to Lepidoptera which can be reared on artificial diets, foliage fed host larvae are required for mass production of the virus. Host larvae must either be collected from the field for infection in the laboratory (Rollinson et al. 1970) or a heavily infested plantation may be sprayed with virus and the infected larvae harvested as they die (Cunningham & DeGroot 1981). The periodic inundation of the virus can be carried out either by distribution of host cocoons containing infected eonymphs in forest stands or by more conventional aerial or ground spraying machinery. The former methods has potential for Neodiprion swainei (Smirnoff 1962) which has an NPV that spreads rapidly from epicenters, while the latter has been widely used for N. lecontei and N. sertifer (Geoff.) (Cunningham & Entwistle 1981). The NPV of N. sertifer has been successfully used in 12 countries and is the most operationally used of the sawfly NPVs. One factor that contributes to this success is the more synchronous hatching of the larvae of N. sertifer, as a result of overwintering as eggs rather than as eonymphs, which facilitates the timing of spraying to infect the younger more susceptible larval instars.
Conservation of Natural Enemies.--Conservation of natural enemies should be considered a part of all silvicultural systems and treatments. In addition there are measures that can be taken directly to conserve natural enemies. However, studies of existent predators are few (see Legner & Moore 1977 ). There has been much more done in agriculture to conserve natural enemies (van den Bosch & Telford 1964), including strip harvesting and habitat diversification (Stern et al. 1976). Pesticide disturbances should be avoided as much as possible, which includes the forest floor where Syme (1977) has shown that a parasitoid of the European pine shoot moth, Rhyacionia buoliana (Schiff.) requires the flowers of small herbaceous plants for nourishment. The judicious use of chemical insecticides is important for conserving natural enemies. There are undoubtedly many naturally occurring biological controls in forests where often the importance of a natural enemy is not known until their effect on the host is disrupted (Hagen et al. 1971). Secondary outbreaks have been known in forestry, but an extensive outbreak of the spruce spider mite, Oligonychus ununguis (Jacobi) following the application of DDT for western budworm control in Montana and Idaho has been documented (Johnson 1958).
Outbreaks of the pine needle scale, Chionaspis pinifoliae (Fitch) occurred in California on Jeffrey and lodgepole pines near Lake Tahoe when an area was fogged with Malathion to control adult mosquitoes (Luck & Dahlsten 1975). The importance of natural enemies was shown in this study as the collapse of the scale population after spraying was halted, occurred over a three-year period and was shown to be due to a small complex of predators and parasitoids. Other insecticide-induced outbreaks have been reported for the target insects. The elimination of parasitoids and virus diseases of the European spruce sawfly, Gilpinia hercyniae, after three years of spraying with DDT in New Brunswick, Canada, resulted in an outbreak of the sawfly (Neilson et al. 1971). In Texas an increase in an infestation of southern pine beetle, Dendroctonus frontalis, was attributed to the deleterious effects of chemical insecticides on the natural enemies (Williamson & Vité 1971). Swezey & Dahlsten (1983) have documented the effects of lindane on the emergence of natural enemies of the western pine beetle, D. brevicornis (LeConte).
The physical environment in forests may be changed to favor natural enemies. Parasitoids and predators can be benefitted by encouraging specific plants for food, shelter and protection from their natural enemies (Bucklner 1971, Sailer 1971). The effectiveness of natural enemies in Poland in 1958 was increased by applying fertilizers, planting deciduous trees and shrubs and providing nectar plants (Burzynski 1970, Koehler 1970). The presence of wild carrot, Daucus carota L., in pine plantations in Canada increases control of the European pine shoot moth, Rhyacionia buoliana (Syme 1981). Longevity and fecundity of the most effective introduced parasitoid, Orgilus obscurator (Nees), was increased due to its feeding on the nectar of several flowers (Syme 1977).
In efforts to conserve natural enemies of bark beetles, Bedard (1933) recommended examination of infested trees for high degrees of parasitism prior to control in order to conserve parasitoids. The disruption of old infestations of mountain pine beetle in lodgepole pine should be avoided since the braconid Coeloides rufovariegatus (Prov.) is very abundant in old infestations (DeLeon 1935). Wind thrown western white pines should not be disturbed because of the high populations of mountain pine beetle parasitoids (Bedard 1933). Because the parasitoid Coeloides vancouverensis (D.T) is more abundant in small diameter Douglas fir infested with the beetle D. pseudotsugae, such trees ought to be left in place (Ryan & Rudinsky 1962). Clerid predators of the western pine beetle eventually move to the lower portions of the bole of infested trees and thus the lower sections of trees should not be treated with insecticide during control projects (Berryman 1967). Clerids associated with the southern pine beetle emerged later than the bark beetles and it was urged that infested trees not be removed until after clerid emergence (Moore 1972).
There are various world organizations devoted to biological control of forest pests. They indicate that activity has been most prevalent in temperate and Mediterranean regions, but that there are no organizations devoted solely to the biological control of forest insects. References pertaining to organization are Clausen (1956), Beirne (1973), Greathead (1980), Taylor (1981), Embree & Pendrel (1986),
Dahlsten & Mills (1999) provide detailed case histories of biological control projects in forest environments; the following being for the most part from their account:
LARCH CASE BEARER, Coleophora laricella Hübner--Coleophoridae
The larch casebearer is native to central Europe and is relatively innocuous in the alpine area on its normal host, Larix decidua Mill. (Jagsch 1973). A fairly rich complex of parasitoids is thought to maintain the casebearer at lower densities in its endemic region (Ryan et al. 1987). It is a defoliator of Larix species and becomes a pest in Europe and Asia wherever larch is planted. This insect was probably introduced on nursery stock into North America from Europe and was first found at Northampton, Massachusetts in 1896 and in Canada at Ottawa in 1905 (Otvos & Quednau 1981). They spread rapidly on tamarack, Larix laricina (Du Roi) K. Koch, in eastern Canada so that by 1947 it was in Newfoundland, the Maritimes, and Ontario and in the United States, Maine, Michigan and Wisconsin (McGugan & Coppel 1962). It is currently widely distributed in the eastern United States and Canada. In 1957 the casebearer was discovered on western larch, Larix occidentalis Nutt, in Idaho (Denton 1958) and in 1966 in British Columbia (Moinar et al. 1967). It is now widely distributed over the range of western larch including British Columbia, Montana, Idaho, Washington and Oregon (Clausen 1978).
The casebearer has one generation per year. The adults begin appearing in late May and lay eggs on either side of the needles. The larvae hatch and burrow directly down into the needles. In the late summer the larvae emerge from the mined needles and form overwintering cases. They feed for a while and then move to branches and twigs to pass the winter. In the early spring the larvae with their cases move and begin feeding on the young buds and foliage. Pupation occurs within the enlarge case, which is commonly attached to a branch on a leaf whorl. The larval feeding, when extensive, causes a loss of growth that is its greatest impact on larch (Ryan et al. 1987).
The biological control program had its beginning in 1928 in western Canada with a request to the Farnham House Laboratory of CIBC for information on the parasitoid complex of the casebearer in Europe (McGugan & Coppel 1962). Importation and field releases of 5 species of parasitoids occurred in eastern Canada between 1931 and 1939 as follows: 1,037 Agathis pumila (Ratz.)--Braconidae, 29,664 Chrysocharis laricinellae (Ratz.)--Eulophidae, 506 Cirrospilus pictus (Nees)--Eulophidae, 3,283 Dicladocerus westwoodii Steph.--Eulophidae, and 97 Diadegma laricinellum (Strobl)--Ichneumonidae (Clausen 1978). All species were subsequently recovered at release sites in Ontario but only two became well established and spread rapidly, A. pumila and C. laricinellae. Between 1942 and 1947 large-scale redistribution releases were made at a number of sites in eastern Canada. The parasitoids were obtained at established colony sites at Millbridge, Ontario (Clausen 1978). By 1948 populations of the casebearer were low on the original release sites. The parasitoids followed the spread of the casebearer to the west assisted by occasional releases (Ryan et al. 1987). This can be cited as an example of a successful biological control program (Webb & Quednau 1971).
A separate, extensive parasitoid importation program was also conducted between 1932 and 1936 in the eastern United States in New England and New York (Clausen 1978). Four of the same parasitoids as released in Canada were used in the U.S. (Clausen 1978) as follows: 8,141 A. pumila, 24,671 C. laricinellae, 231 D. westwoodii, and 3,580 D. laricinellum (Strobl). Although there is little information to go on, the results were apparently the same in the eastern United States with the establishment of A. pumila and C. laricinellae followed by high parasitization rates particularly by A. pumila (Dowden 1962). Releases of the two established parasitoids were also made in 1937, 1950 and 1952 in Michigan and Wisconsin.
In the western United States, the first releases of A. pumila were made in 1960 with 2,360 adult parasitoids that were collected in Rhode Island (Clausen 1978). These were released at 5 locations in Idaho. Recoveries were made at 3 sites in 1962. Between 1964 and 1969 field rearing of A. pumila in whole tree cloth cages permitted the release of this parasitoid at 400 sites in Idaho, Montana, Washington and British Columbia (Ryan et al. 1987). The parasitoid became established and built up at some sites but at other sites it either didn't become established or it didn't build up. In addition, significant defoliation still occurred throughout much of the area by 1970 and the program was rated as a failure (Turnock et al. 1976, Ryan et al. 1987).
Between 1971 and 1983 a new strategy was used as C. laricinellae and five other species of parasitoids from Europe and Japan were released over a period of several years. C. laricinellae became widely established but the other species don't appear to be very important for control of the casebearer though isolated recoveries have been made (Ryan et al. 1987). In an effort to properly evaluate the effect of the parasitoids, the larch casebearer was sampled at sites in Oregon where the casebearer had recently invaded. The populations were followed to the point of severe defoliation from 1972 to 1978 and then parasitoids were released between 1979 and 1985 (Ryan 1983, 1986; Ryan et al. 1987). The first parasitoid to be released was C. laricinellae followed by A. pumila. Parasitoids increased and the casebearer steadily declined and this trend has continued in all plots through 1987 (R. B. Ryan, personal communication). Although the prospects are good for a complete success, Ryan et al. (1987) feel it is too soon to make the claim.
In British Columbia the larch casebearer biological control program was reviewed in 1974 due to the successes in eastern Canada (Otvos & Quednau 1981). Four parasitoids have been released: A. pumila, C. larcinellae, Diadegma laricinellum, and Dicladocerus japonicus Yshm. The story is much the same as with the other release programs--A. pumila and C. laricinellae have become well established and the other two have not been recovered. It is too early to evaluate the effects of the two parasitoids but C. laricinellae is fairly common in British Columbia and may be responsible for the reduction of larch casebearer and less tree mortality (Otvos & Quednau 1981).
The larch casebearer is a successful biological control program in eastern Canada and may shortly be successful in the northwestern United States. It is an example of a classic introduction program with the subsequent redistribution of the parasitoids from areas of establishment to new areas. It is interesting because the two parasitoids complement one another in their action against the casebearer. Agathis is extrinsically superior at low host densities and Chrysocharis is effective at high host densities. Quednau (1970) hypothesized that Agathis can only give partial control on its own and that success is only possible through cooperative interaction with Chrysocharis. Ryan (1985) hypothesizes that Agathis may not be detected in successive samples since parasitized larvae commonly descend to understory vegetation. Samples could be biased toward Chrysocharis due to the absence of Agathis in the foliate that is sampled. There has been no success in establishing other parasitoid species. This program also is an example of one where there was a rigorous attempt to evaluate efficacy of the parasitoids (Ryan 1986, Ryan et al. 1987).
WINTER MOTH, Operophtera brumata (L.)--Geometridae
This polyphagous defoliator of hardwoods is native to most of Europe and parts of Asia, where it is particularly frequent on fruit trees and oak. It was first recognized as an accidental introduction on the south shore of Nova Scotia in 1949 and eventually extended its range to the whole of this region together with small isolated parts of New Brunswick and Price Edward Island by 1958.
In the first few years after its appearance in Nova Scotia, damage was evident in apple orchards, shade trees and oak forests. However, at this time hardwoods were not commercially exploited in the Province and so the winter moth was not considered a serious pest (Embree 1971). Consequently it was possible to initiate a biological control program rather than a program of insecticide eradication. The general research policy in the early 1950's was directed towards population dynamics of forest insect populations and thus the biological control program was initiated in 1954 with a view to population studies of the host and introduced parasitoids.
Prior to the introduction of parasitoids from Europe, the winter moth fluctuated erratically at high population densities. These fluctuations resulted from the coincidence of hatching of the overwintering eggs and bud burst in early spring (Embree 1965). This same key mortality factor was also found to be responsible for changes in population levels of winter moth in Britain (Varley & Gradwell 1968).
Three tachinid and three ichneumonid parasitoids were obtained in sufficient quantity for introduction into Nova Scotia from Europe. The parasitoids were collected and shipped to Canada by staff of the Belleville Laboratory and the CIBC and field releases were made during the period 1954-62. These included releases of over 22,000 individuals of the tachinid Cyzenis albicans (Falk.) and a total of 2,261 individuals of the ichneumonid, Agrypon flaveolatum (Grav.), the only two species that became established. C. albicans is very fecund and oviposits microtype eggs around the edge of damaged foliage where they are ingested by late-instar host larvae. The egg hatch in the midgut of the host and the larvae bore through the gut wall to develop rapidly after the host has pupated. The tachinid pupates and overwinters within the host pupal case in the ground. The biology of A. flaveolatum is similar but its oviposits directly into the host larvae and has larger eggs and much lower fecundity.
Following the establishment of these two parasitoids, parasitism by C. albicans increased rapidly to 50% in 1960 and life table data showed that a considerable increase in prepupal mortality was responsible for the collapse of the winter moth population in the main study site (Embree 1965). Parasitism by A. flaveolatum increased only following the initial decline of the host outbreak and while it may have enhanced the depression of the winter moth density, population models indicate that the efficiency of C. albicans alone is sufficient to account for successful biological control (Hassell 1980). However, a more recent analysis of the life table data from Nova Scotia and Britain (Roland, pers. comm.) indicates that the increased pupal mortality may have arisen only indirectly from the introduction of C. albicans. Increased parasitism by C. albicans is closely followed by an increase in the activity of soil predators, perhaps sustained on overwintering C. albicans puparia through late summer and early spring when prey are generally more scarce. Thus predation rather than parasitism may be more directly responsible for the observed increase in winter moth pupal mortality. Recent unpublished work in British Columbia indicates that staphylinid predators are especially important in regulation and that C. albicans puparia are avoided because they are too large for the predators.
More recently, between 1976 and 1978, winter moth has been noted in Oregon, Washington and British Columbia on various hardwood and fruit trees. Both C. albicans and A. flaveolatum were relocated to these areas between 1979 and 1982 and recoveries were made in many regions (Kimberling et al. 1986). However, it is too early to determine the success of these releases. But in contrast to the earlier program in Nova Scotia, the western program has been conducted at a time when research policy has moved away from population dynamics toward practical application of pest control and thus no detailed monitoring of the winter moth before and after parasitoid release has been made.
This program is often considered a good example of biological control in which, in contrast to earlier multiple introduction programs, selective introduction were made. These led to the establishment of a high host density specialist (Cyzenis), with high fecundity to bring about the collapse of an outbreak, and a low host density specialist (Agrypon), that has good searching ability to maintain the collapsed population at a low level of abundance. However, the main reason for the release of a smaller number of parasitoid species was the relatively meager size of collections in Europe, where winter moth abundance was not high at the time. Thus the only conscious selection process was of parasitoid species obtained in sufficient quantity for meaningful release (Mesnil 1967), although once the two established parasitoids were becoming effective in the early 1960's a decision was made to curtail releases of other species (Embree 1966). The end results was the successful establishment of two particularly suitable parasitoids and the program provides one of the best examples of the detailed evaluation of a biological control project. Also as was pointed out in earlier sections, the development of a detailed model in England prior to the importations tended to show very little regulatory impact by Cyzenis, which might have precluded its importation into North America.
COLOMBIAN DEFOLIATOR, Oxydia trychiata (Guenée)--Geometridae
A successful example of the use of an exotic parasitoid to control a native forest pest was the importation of the egg parasitoid, Telenomus alsophilae Viereck, from North America to Colombia in South America against a geometrid defoliator (Bustillo & Drooz 1977, Drooz et al. 1977). There are a number of interesting facets to the program since the normal geometrid host of the parasitoid in North America, the fall cankerworm, Alsophila pometaria (Harris), is in a different subfamily and genus than the target pest, Oxydia trychiata, in South America. The Colombian geometrid, O. trychiata, has a wide distribution extending from Costa Rica to most of the countries in South America. The moth has 3 generations per year and apparently is capable of normal development on introduced tree species (citrus, coffee, pine and cypress). There has been an attempt to establish exotic conifer species in Colombia for the production of pulp and paper. This previously unimportant insect became a pest in these pine and cypress plantations (Drooz et al. 1977).
The egg parasitoid, T. alsophilae (Scelionidae) has several biological attributes that are well worth noting since they may have influenced this unique cross genus introduction. First, its normal host, the fall cankerworm, feeds on several broad leaved trees but its host in South America feeds on conifers. This indicates that host plant odors or other differences between conifers and broad leaved trees are unimportant in host egg finding. There may have been a clue to this because the fall cankerworm feeds on several genera of deciduous hardwoods. The parasitoid is apparently easily to handle as changes in photoperiod and lack of cold in the winter did not interfere with development (Drooz et al. 1977). The climate at the origin of the parasitoid in Virginia (30° N. Lat., el. 370 m, mean winter temperature 2°C and mean summer temperature 24°C) compared to that of the release site in Colombia (6° N. Lat., 2340 m, temperature range 6° - 26°C with annual mean of 16°C) shows a shift from a temperate to a tropical climate although the extremes are about the same. The rainfall patterns in the two regions also differ. The ecological plasticity of this parasitoid is thus demonstrated, and in addition it is long-lived (>6 months) (Drooz et al. 1977).
The parasitoid may be easily reared, which is important to a biological control project (Drooz et al. 1977), and eggs of another species of geometrid, Abbottana clemataria (J. E. Smith) are used because it could be propagated on artificial diet. Around 18,000 parasitoids were sent to and released in a pine plantation in Colombia between October and December in 1975 (Bustillo & Drooz 1977, Drooz et al. 1977). Parasitization rates on O. trychiata eggs were very high and by the time the parasitoid had undergone three generations in April of 1976 few adults could be found at normal emergence time. Only 13 egg masses of O. trychiata could be found and these were 99% parasitized. By May the outbreak was controlled when larvae could not be found in the area (Drooz et al. 1977). It is speculated that the parasitoid maintains itself on any of the four species of Oxydia or other geometrids in Colombia.
EUROPEAN PINE SHOOT MOTH, Rhyacionia buoliana (Schiff.)--Tortricidae
This species occurs throughout Europe and parts of Asia where it is a major pest of pine plantations. It was first discovered in North America at New York in 1914 and was later also found on imported nursery stock in Canada in 1925. While its distribution extended throughout the northeastern United States and eastern Provinces of Canada, as well as in British Columbia and the northwestern United States, it was considered an important pest only in the red pine plantations in the northeastern United States and southern Ontario.
In 1927, the Commonwealth Institute of Biological Control was engaged to collect parasitoids in Great Britain for introduction into Canada and this led to the release of eight species during the period 1928-43 and an additional five species from material collected in continental Europe during 1954-58 (McGugan & Coppel 1962). Two additional species were released during 1968-74, one from Germany and one from Argentina (Syme 1981). A similar program of parasitoid introductions was carried out in the New England states from 1931-37 (Dowden 1962). This program is another example of the multiple introduction approach where emphasis is placed on the need to provide rapid results without detailed preintroduction studies. Of the 15 species of parasitoids released in New England and in southern Ontario, only three larval parasitoids, the braconid Orgilus obscurator (Nees), and the ichneumonids Eulimneria rufifemur (Thoms.) and Temelucha interruptor (Grav.), became firmly established. However, it was not until the early 1960's that T. interruptor was disclaimed as a cleptoparasitoid detrimental to the potential impact of O. obscurator (Arthur et al. 1964).
Orgilus obscurator is a specific larval parasitoid with a high fecundity and an efficient host finding ability that permits it to avoid both superparasitism and very low host density situations (Syme 1977). In contrast, T. interruptor is a more general parasitoid of Microlepidoptera and while it also has a high fecundity it is inefficient at host finding and oviposits most successfully in host larvae previously attacked by O. obscurator. Both parasitoids attack young host larvae and only develop further when the host larvae approach maturity. However, the first instar larva of T. interruptor is competitively superior to that of O. obscurator, which is killed at an early stage to ensure the successful development of the cleptoparasitoid (Schroeder 1974).
Although the biological control program against pine shoot moth in North America is considered to be unsuccessful, there are isolated reports of high levels of parasitism by O. obscurator followed by the collapse of shoot moth populations at Dorcas Bay in Ontario (Syme 1971) and near Quebec City (Béique 1960). The occurrence of wild carrot, Daucus carota (L.) at Dorcas Bay where parasitism by O. obscurator reached 92% prompted further investigations on the influence of this nectar and pollen source on parasitism in Ontario. Syme (1977) demonstrated the beneficial influence of flowers on the longevity and fecundity of O. obscurator and was able to show increased rates of parasitism and elimination of pine shoot moth populations when the parasitoid was released into plantations where D. carota was plentiful (Syme 1981).
GYPSY MOTH, Lymantria dispar (L.)--Lymantriideae
This insect is native to the Palearctic region where it is a pest of broadleaf forests in eastern and southern Europe. It was brought to North America and accidentally released in Massachusetts in 1868. Since then it has become a serious pest of hardwoods throughout the northeastern United States and has a continually expanding range which currently extends into Ontario, Quebec and southward into Virginia with isolated infestations in Minnesota, Oregon and occasionally California.
A biological control project was organized by the U. S. Department of Agriculture, Bureau of Entomology in 1905 and extensive foreign exploration for parasitoids and predators was carried out in Europe, Japan, North Africa and Asia at various intervals since that time (Doane & McManus 1981). This was the first major classical biological control project against a forest insect. The gypsy moth project has revealed that (1) insect disease was recognized as an important biological control factor, (2) the sequence theory of natural enemies was introduced by W. F. Fiske, (3) a number of future important contributors to biological control were trained on the project (H. S. Smith, W. R. Thompson and W. D. Tothill), (4) sleeve cages were invented as well as other equipment and techniques that are still in use today and (5) L. O. Howard and W. F. Fiske were the first to clearly distinguish between those causes of mortality that act in relation to the density of the population and those that do not. L. O. Howard also stimulated the Canadian interest in biological control in the early 1900's by making available facilities and scientific assistance from the Melrose Highlands Parasite Laboratory of the U. S. Bureau of Entomology.
Early importations of natural enemies occurred between 1905-14 and again between 1922-33. While some collections were made in Japan, attention focused on Europe where temporary field laboratories were placed wherever gypsy moth outbreaks were sufficient to permit the rearing of parasitoids from a large number of hosts. Frequent shipments of parasitoids and predators were made to the gypsy moth laboratory at Melrose Highlands, Massachusetts and this resulted in the liberation of >690,000 living insects of more than 45 species during this period (Dowden 1962). The enormous importation and multiple release program enabled two larval/pupal predators, two egg parasitoids, six larval parasitoids and one pupal parasitoid to become established in the New England states. The two egg parasitoids were also subject to either large scale rearing releases in the case of Ooencyrtus kuwanae (How.), or to large-scale relocation releases in the case of Anastatus disparis Ruschka. Most of the establishments occurred rapidly after the initial field releases but the tachinids Parasetigena silvestris (R.-D.) and Exorista larvarum (L.) were not recovered until 1937 and 1940 respectively and the chalcidid Brachymeria intermedia (Nees) was only recovered in 1965.
Biological control by established parasitoids and predators in New England was limited and large scale aerial applications of DDT were used until the early 1960's. Since 1960 renewed interest in the search for additional natural enemies has extended explorations in Europe, Japan, Morocco, India, Iran and Korea (Doane & McManus 1981). Since 1963 the USDA Agricultural Research Service Beneficial Insects Research Laboratory has continued to receive gypsy moth natural enemies in their quarantine facilities and have been able to distribute more than 200,000 individuals of about 60 species to other State and Federal facilities for culture, study and field release. From 1966 until 1971, the Gypsy Moth Methods Improvement Laboratory at Otis Air Force Base in Massachusetts was charged with the development of rearing procedures for the imported natural enemies. From 1963-71 in conjunction with the New Jersey Department of Agriculture about 7 million parasitoids of 17 species were reared and released in the forests of New Jersey and Pennsylvania. Then from 1971-77 a Gypsy Moth Parasite Distribution Program was established in which the New Jersey Dept. of Agriculture and the University of Maryland reared and released an additional two million parasitoids of 18 species throughout the New England states. Since the late 1970's more new parasitoids and a predator from Japan and Korea and from the Indian gypsy moth, Lymantria obfuscata Walk., have been imported (Coulson et al. 1986). More than 100,000 individuals of nine new species or strains have been released in the field in Delaware, Massachusetts and Pennsylvania.
Although much knowledge of the biology and rearing methods of the imported parasitoids was gained during this massive program of importation, propagation and release, it has resulted in the addition of only a single pupal parasitoid, Coccygomimus disparis (Vier.) to the complex of 10 species established during the initial importation program. This has prompted Tallamy (1983) to compare the establishment of gypsy moth parasitoids with island biogeography theory, suggesting that a dynamic equilibrium now exists between further introductions and the extinction of established parasitoids. In the last 30 years two of the parasitoids that were initially established, Anastatus disparis and Exorista larvarum have become very rare, while two pupal parasitoids Brachymeria intermedia and C. disparis have become established. However, the main reasons for the failure to establish additional parasitoids in recent years are the parasitoids' requirements for suitable alternative overwintering hosts for their second generation each year and the fact that several of the parasitoid species released during the 1960's were not closely associated with gypsy moth as a principal host in their areas of origin.
The failure of the established natural enemies to control expanding outbreaks of the gypsy moth encouraged attempts during the 1970's to augment the impact of previously established species. Through inundative releases of Cotesia melanoscelus (Ratz.), Weseloh & Anderson (1975) were able to show significantly increased rates of parasitism but this had little influence on foliage protection or egg mass counts for the following generation. On the other and several other inundative releases of this and other species failed to provide any evidence of increased parasitism in comparison to control plots (Doane & McManus 1981). The combined release of parasitoids and pathogens has been used as a method of augmentation. Wollam & Yendol (1976) were able to show a synergistic effect of the release of C. melanoscelus in plots treated with a double application of low concentration Bacillus thuringiensis over plots treated with each of these natural enemies alone. The resultant reduction in defoliation and subsequent egg mass densities has more recently been attributed to the retarding effect of B. thuringiensis on host larval growth which exposes the younger larvae to parasitism for a longer period of time (Weseloh et al. 1983). A similar effect of C. melanoscelus in conjunction with viral treatments is unlikely to occur since this parasitoid avoids oviposition in moribund host larvae (Versoi & Yendol 1982).
Augmentation through use of microbial pathogens has been of considerable importance against gypsy moth with significant advances in recent years. Early trials with B. thuringiensis in the 1960's were not effective in providing foliage protection; but the discovery of improved strains (Dubois 1985b) and successive improvements in formulation and application technology during the late 1970's and early 1980's led to greater success. The results of aerial applications during the 1970's remained highly variable but a recommendation of double application of low concentrations was developed and used operationally for the first time on a large scale in 1980. This also met with limited success but further experimental work in the early 1980's (Dubois 1985a) indicated that the use of higher concentrations and acrylamide stickers could provide not only good foliage protection but also could reduce subsequent egg mass densities significantly with a single application. This development reduced the cost of B. thuringiensis applications and has been used operationally with success on 40-70% of the 1.3-1.5 million ha. of hardwood forest treated since 1983.
Many field trials have been conducted with virus sprays against gypsy moth both in North America and Europe (Cunningham 1982). An NPV virus strain (Hamden standard) isolated from a natural epizootic in Connecticut in 1967 forms the basis for the commercially produced "Gypchek" that was registered for use against gypsy moth in North America in 1978. However, early trials of the baculovirus produced erratic results and while continued improvements in formulation and application have produced more positive results, it has never been accepted for operational use (Podgwaite 1985). Reasons for this are the relatively low virulence of the virus, its rapid degradation on foliage in the field and the more recent successes with the use of B. thuringiensis.
The gypsy moth program has been spectacular in both the scale and the continued enthusiasm with which it has been conducted, but that the results have been disappointing and serve as a good example of the failure of classical biological control in situations where the introduced pest is also severe in its region of origin. Therefore the search for natural enemies in areas where gypsy moth is not a pest, in non-outbreak populations or from related non-pest Lymantria species may prove to be a better strategy.
EUROPEAN SPRUCE SAWFLY, Gilpinia hercyniae (Hartig)--Diprionidae
A spruce (Picea spp.) feeding insect native to most of Europe, the European spruce sawfly was first noted as an accidental introduction in Canada in 1922. By 1930 a severe outbreak was causing concern in the Gaspe Peninsula and by 1936 the sawfly threatened to devastate the spruce forests of eastern Canada by extending its range across all eastern Provinces and adjacent United States and causing severe damage over an area of more than 10,000 sq. miles (McGugan & Coppel 1962).
One of the most extensive projects undertaken in classical biological control was begun against European spruce sawfly in 1933. Gilpinia hercyniae was not at first distinguished from G. polytomum (Htg.) and the Farnham House Laboratory in England (now known as CIBC) was engaged to make large-scale parasitoid collections from the latter species in Europe. Initial studies revealed that apart from the egg parasitoids, all other parasitoids develop so as to overwinter in the host cocoon. This simplified parasitoid collections in Europe to those stages of development. A team of about 30 persons collected >1/2 million cocoons of G. polytomum in Europe for shipment to Canada during 1932-40. Additionally more >1/2 million eggs and 31 million cocoons of other spruce and pine feeding sawflies were shipped to supplement the numbers of the less host specific parasitoid species available for field release (Morris et al. 1973, Finlayson & Finlayson 1958). There were 96 species of primary and secondary parasitoids obtained from these cocoon collections at the Belleville Laboratory in Canada and a multiple introduction program involving two egg parasitoids and 25 larval and cocoon parasitoids was initiated in 1933-51. The importation of a wide variety of parasitoids from diverse hosts permitted the inclusion of several sawfly pests as additional targets for some of the releases (McGugan & Coppel 1962).
The addition of an elaborate controlled environment quarantine building was made at Belleville in 1936 allowed the mass rearing of several of the imported European parasitoids. Dahlbominus fuscipennis, a gregarious ectoparasitoid of prepupae, readily attacked cocoons in the laboratory and was selected along with several other species for a massive program of mass rearing for release. The mass-rearing peaked in 1940 when a total of 221.5 million D. fuscipennis was released and by the end of the program in 1951 a total of 890 million directly imported or laboratory reared parasitoids had been liberated (McGugan & Coppel 1962).
Only 5 species of parasitoids out of 27 released became established over more than several generations, although four additional species were recovered during the years shortly after release. Three of the five species, D. fuscipennis, Exenterus amictorius (Panz.) and E. confusus Kerr, were widely established only during the outbreak and have since not been recovered from G. hercyniae. Although E. amictorius had little impact, the other two species achieved variable but appreciable levels of parasitism and have been credited with the decline of the outbreak in at least some areas. Two other parasitoids, Exenterus vellicatus Cush. and Drino bohemica Mesn., never became important until the collapse of the outbreak but have replaced the three species present during the outbreak to maintain host population at low, non-damaging densities.
The epizootic of European spruce sawfly began to decline in 1939-40, which coincided in the southern part of the range with the occurrence of a nuclear polyhedrosis virus, Borrelinavirus hercyniae. This virus is thought to have been accidentally imported and released in Canada along with the parasitoid. It spread rapidly to produce virus epizootics throughout most of the outbreak range and by 1943 host population densities had declined to very light infestations. Unlike other diprionid sawflies, G. hercyniae larvae are not gregarious and the rapid spread and subsequent impact of the virus was attributed to its virulence (Bird & Elgee 1957). More recent studies in the Great Britain, where G. hercyniae was accidentally introduced from the European continent in 1968, indicate that birds play an important role in virus transmission (Entwistle 1976). The importance of D. bohemica, E. vellicatus and the NPV virus in maintaining the spruce sawfly at low population densities in Canada has been inadvertently demonstrated through chemical spray treatments aimed against spruce budworm. Both in the early 1960's and again in the 1970's sawfly population levels increased immediately following the cessation of a 2-3 year spray treatment, due to the detrimental effects of the spray on natural enemies, but declined after several generations as a result of increased parasitism and the reappearance of the virus (Neilson et al. 1971, Magasi & Syme 1961).
There are several interesting features of this successful biological control program. First the success of the accidental introduction of the virus provides to date the most outstanding example of the use of a pathogen in classical biological control. Its ability to control the sawfly population in the absence of parasitoids has been demonstrated (Bird & Burk 1961, Entwistle 1976) and in Canada it has persisted in the forest environment since the initial introduction despite the low host densities (Magasi & Syme 1981). The multiple introduction programs of parasitoids resulted in the establishment of the two more effective and specific species, despite the release of a wide range of potential competitors. However, the continuous and large scale release of poorly adapted parasitoids, which were later recovered only from other sawfly hosts, was successful in inducing significant levels of mortality prior to the introduction of the virus.
LARCH SAWFLY, Pristiphora erichsonii (Hartig)--Tenthredinidae
A comparatively rare insect in Europe, the larch sawfly was first generally recognized as established in larch forests throughout the eastern Provinces of Canada in 1884. Several short lived but severe infestations were observed in 1906-16 in which hugh quantities of tamarack (Larix laricina) were destroyed (McGugan & Coppel 1962). Ever since the sawfly has been found throughout the range of larch in North America but remains more important on tamarack than on western larches. It is unknown whether the sawfly was a recent introduction in the late 19th Century or of much older origin in North America (Ives & Muldrew 1981). But the lack of native parasitoids prompted a classical biological control program in 1910-13, 1934 and 1961-64.
Collections were made in Great Britain during the early phase of introductions (McGugan & Coppel 1962). They were shipped to Canada for quarantine, screening and direct release. This led to the establishment of the specific ichneumonid larval parasitoid Mesoleius tenthredinis Morley, which in Manitoba was found in 20% of sawfly cocoons in 1960 and had parasitized over 80% of the population by 1927 (Criddle 1928). Subsequently a tachinid Zenillia nox (Hall), was collected in Japan in 1934 by the U. S. Dept. of Agriculture and released both in New Brunswick and British Columbia but failed to establish. The success of parasitism by M. tenthredinis prompted an extensive relocation program to distribute this parasitoid throughout Canadian larch forests. Rapid establishment was reported with subsequent reductions in sawfly populations and reduced timber losses.
This appeared to be another example of the success of classical biological control in Canada, but in the late 1930's larch sawfly defoliation again became prevalent in Manitoba. Parasitism by M. tenthredinis appeared to have dropped to low levels, so 75,000 parasitoids were transferred from British Columbia across central Canada. While the parasitoids' range increased, levels of parasitism remained low due to the encapsulation of parasitoid eggs by host larvae (Muldrew 1953). The appearance of a resistant European strain of the sawfly, capable of encapsulating M. tenthredinis eggs, appears to have resulted from the parasitoid introduction program in 1913, when imported larch sawfly cocoons were placed directly in the field. The resistant strain has since spread across Canada and into neighboring states of the United States, becoming predominant in most regions (Wong 1974).
Renewed efforts were made in 1957 to obtain more parasitoids from Europe and Japan, and long term study plots were chosen in Manitoba to better evaluate the dynamics of the larch sawfly populations and the impact of introductions. These studies (Ives 1976) indicated that mortality in the cocoon and adult stages determined population trends and that high water tables and predation by small mammals were largely responsible for the erratic population abundance. The native tachinid, Bessa harveyi (Tns.), considered the most important parasitoid in the renewed outbreaks, had little impact.
The CIBC collected 11 parasitoids in Europe and Japan and shipped them to Canada between 1959-65. Five of the more abundant species were selected for release and >200 adult were liberated. A separate introduction of the masked shrew, Sorex cinereus Kerr from New Brunswick to the island of Newfoundland was made in 1958 in order to fill the vacant niche for an insectivore and to increase cocoon predation. The shrew as successfully established as well as two of the parasitoids. One of these parasitoids, the ichneumonid Olesicampe benefactor Hinz., attacks young sawfly larvae, the second, a Bavarian strain of M. tenthredinis, was shown to be only weakly encapsulated by the resistant sawfly strain and was able to pass its characteristics on to the progeny of mixed (Britain X Bavarian) crosses (Turnock & Muldrew 1971). Parasitism by M. tenthredinis initially increased following the release of the Bavarian strain but O. benefactor became the dominant parasitoid influencing cocoon survival. Parasitism by the latter at the release point in Manitoba attained levels of ca. 90% between 1967-72 (Ives & Muldrew 1981) and was the dominant factor for the collapse of the sawfly epizootic (ives 1976). Olesicampe benefactor was relocated from Manitoba to most other Provinces in Canada (Turnock & Muldrew 1971) as well as to Maine (Embree & Underwood 1972), Minnesota (Kulman et al. 1974) and Pennsylvania (Drooz et al. 1985).
Effects of the masked shrew on larch sawfly cocoon survival in Newfoundland have never been adequately estimated. Predation of cocoons is thought to have increased, but outbreaks have continued through the 1970's (Ives & Muldrew 1981). Therefore, O. benefactor seems to offer the greatest potential for controlling larch sawfly in Canada. However in 1966 a hyperparasitoid, Mesochorus globulator Thunb. began to attack this parasitoid in Manitoba. The polyphagous hyperparasitoid is common in Europe and may also have been accidentally introduced during the initial 1910-13 introductions. It has spread throughout the region and into Wisconsin, although it hasn't been reported from Pennsylvania (Drooz et al. 1985). While hyperparasitism attained very high levels (80-90%) in Manitoba during 1970's, sawfly populations continue to remain low in abundance, and thus O. benefactor despite the occurrence of the hyperparasitoid may achieve control.
The larch sawfly program gives further evidence of the value of the more specific and well adapted parasitoids in classical biological control. As in the case of the European spruce sawfly, while a wide range of parasitoids was released, only the more specific species became established. However, while in the absence of hyperparasitism O. benefactor may have been an ideal control agent, its competitive superiority over the Bavarian strain of M. tenthredinis may have prevented the latter from establishing and spreading more widely. This and the known occurrence of various geographic strains of M. tenthredinis differing in ability to avoid encapsulation by the host, emphasizes the value of detailed studies of parasitoid biologies prior to introduction. Also, the accidental introduction of both a parasitoid resistant strain of the host and probably also a hyperparasitoid indicates the critical need for quarantine handling of imported material to avoid unnecessary liberations.
EUROPEAN WOOD WASP, Sirex noctilio F.--Siricidae
Biological control attempts against the woodwasp are one of the very few large projects directed against a wood boring insect. Woodwasps usually are considered secondary pests and attack dead or dying trees. Sirex noctillio occurs in Canada and throughout Europe but is most common in the Mediterranean area. It is somewhat specific to Pinus species (Spradbery & Kirk 1978), and is unique among the siricids in Europe in that it is able to kill standing green trees. Under the right circumstances, as occurred in New Zealand and Australia, this insect was able to cause serious losses to Monterey pine, Pinus radiata D. Don., plantations. The pest was first discovered on the North Island of New Zealand about 1900 but it was not until 1927 that it was abundant enough in exotic pine plantations for control to begin (Taylor 1981). High mortality occurred in P. radiata plantations between 1940-49 in New Zealand, and S. noctilio reached Australia in southern Tasmania in 1952 and Victoria in 1961 (Taylor 1976).
There is a special relationship of S. noctilio to a symbiotic fungus, Amylostereum areolatum (Fr.) Boidin, that serves as a kairomone for the parasitoids of the woodwasp. Also the parasitic nematode, Deladenus siricidicola Bedding, is wholly dependent in nature on the woodwasp and the fungus (Bedding 1972). Adults of S. noctilio emerge from midsummer to late fall and mate in the upper foliage of trees. Female wood wasps oviposit by drilling holes through the bark into the sapwood of trees that are usually predisposed or damaged. At the time of oviposition the symbiotic fungus is introduced (Taylor 1981). Adults live only a few days in nature. The eggs hatch when the fungi have invaded the surrounding area and this occurs when some drying has taken place to favor the fungi. First and second instar larvae feed exclusively on fungus and third and fourth instars begin to tunnel into the wood. The larvae turn back toward the bark to about 5 cm from the bark surface to enter the prepupal stage. Pupation may not occur until the second or third year after hatching, depending on the weather. After pupation adults emerge in about three weeks, and each generation emerges over a period of two to three years with the proportion of individuals emerging in the first, second and third year varying by site (Taylor 1981).
Biological control was initiated in New Zealand in 1927 (Taylor 1981). During 1929-32 the ichneumonid, Rhyssa persuasoria L. was introduced but the control was not satisfactory (Turnock et al. 1976). Then Ibalia leucospoides (Hochenw.) (Ibalidae) was colonized in 1954-58, which resulted in improved control (Zondag 1959). The two parasitoids were then colonized in Tasmania. A large-scale biological control effort did not begin until 1961 following the discovery of S. noctilio in Victoria, Australia. A National Sirex Fund was established, which consisted of a consortium of federal, state and private agencies, and a committee was formed to coordinate research and control in Victoria (Taylor 1981). A world wide search for natural enemies was begun by the Division of Entomology, CSIRO in 1962. The search for parasitoids in the northern hemisphere was completed by 1973, and during the 11-year period 21 species of parasitoids were sent to Tasmania for culture (Taylor 1976). The plan was to obtain all the available parasitoids of siricids in conifers and as many strains as possible from different climatic zones with emphasis on the Mediterranean area. This included collections of siricids in conifers other than Pinus and from genera and species other than Sirex noctilio. Ten different parasitic species were released in Tasmania and Victoria, six having become established and one additional species, the ichneumonid Rhyssa hoferi Roher, probably established (Taylor 1981). Of the seven species two are holarctic (R. persuasoria and I. leucospoides), two are palearctic (I. rufipes drewseni Borries and the ichneumonid Odontocolon geniculatus Kreichbaumer) and three are nearctic [the stephanid Schlettererius cinctipes Cresson and the ichneumonids Megarhyssa nortoni (Cresson) and R. hoferi].
These species tend to be complementary, although there might be some competition within the guild attacking larger larvae. The Ibalia species attack first or second instar siricid larvae and the two species have different emergence times so that they do not compete directly. The ichneumonids attack the more developed larvae of their host and there may be differential preference based on tree diameter (Taylor 1981). Schlettererius cinctipes emerges after the peak emergence of the ichneumonids, while the other two are also complementary as O. geniculatus is small, emerges in springtime and attacks late hatching larvae that are still closer to the bark surface. Rhyssa hoferi is adapted to drier areas and could do well in drier climates (Taylor 1981).
A parasitic nematode, Deladenus siricidicola, was found in New Zealand in 1962 (Zondag 1969). It causes female wood wasps to lay infertile eggs. Additional nematodes wee sought during 1965-75 without success (Bedding & Akhurst 1974). Different strains of the nematode have also been released throughout wood wasp infested areas in Tasmania and Victoria and it is well established throughout. This nematode also affects the reproduction of some of the female parasitoids (Bedding 1967), which apparently does not adversely affect biological control. The nematode is credited with reductions of wood wasp populations to very low levels in certain areas.
The Sirex noctilio biological control program is significant for several reasons. A large group of organizations cooperated in a well funded, extensive worldwide search for parasitoids as well as a research program that examined many aspects of the host tree/Sirex/fungus/parasitoid relationships (Taylor 1981). As with Gilpinia hercyniae there was a fortuitous introduction (the nematode). Sirex noctilio was introduced from the northern to the southern hemisphere and attacked an exotic host plant Pinus radiata (native to California). The search for parasitoids in the north was made from S. noctilio and its host trees to siricids in other genera and species in Pinus as well as other conifers. The project was well planned with attention given to colonizing strains of parasitoids suited to different climatic zones and developmental stages of the host.
It is believed that this biological control project will eventually be completely successful (Turnock et al. 1976). It has been thought that the combination of parasitoids and nematodes along with sound forest management should hold S. noctilio down to the level where losses are not serious (Taylor 1976).
GREATER EUROPEAN SPRUCE BEETLE, Dendroctonus micans (Kugelmann)--Scolytidae
This bark beetle, probably native to coniferous forests of eastern Siberia, is one of only two Dendroctonus species occurring in the palearctic region. Dendroctonus micans is primarily a pest of spruce, Picea spp., but will occasionally attack Pinus sylvestris L. The beetle has been expanding its range for many years and is still spreading. About 200,000 ha are currently suffering from epizootics and recently invaded areas include Great Britain, France, The Georgian S.S.R. and Turkey (Grégoire et al. 1987, Evans 1985). In the inner parts of its range where the beetle has been established for a long time populations remain at low densities and it is not a pest.
This bark beetle differs from the more aggressive North American Dendroctonus species in that it attacks its host tree in low numbers, killing the bark in patches. Successive attacks over a period of five to eight years may be necessary to kill a tree except during beetle outbreaks (Grégoire 1985). The beetle shows kin-mating, gregarious larvae and apparently lacks associated pathogenic fungi that are characteristic of many Scolytidae. Dendroctonus micans has very few natural enemies which may be due to its unique biology that seems to protect the beetles from competitors and generalist natural enemies by the defenses of its living host (Everaerts et al. 1988).
One specific predator Rhizophagus grandis Gyllenhal is very abundant in areas where the bark beetle has been present for long periods of time. This rhizophagid beetle is believed to be responsible for maintaining the low, stable D. micans population in these areas (Kobakhidzi 1965, Grégoire 1976, Moeck & Safranyik 1984).
A massive biological control project was initiated against D. micans in Georgia S.S.R. in 1963 (Kobakhidze 1965). The spruce beetle had extended its range into Georgia following World War II in timber imported from the north. A predator relocation program was planned as the predator did not follow its host. Rhizophagus grandis was released in large numbers as larvae and adults on spruce trees infested by D. micans (Kokakhidze et al. 1968). Effective control apparently has been achieved (Grégoire et al. 1987).
First observed in the Massif Central of France in the early 1970's, D. micans was targeted for biological control in a program funded by the European Economic Community in 1983. Its main thrust was to establish the predator, R. grandis (Grégoire et al. 1987). A similar program was initiated in 1983 in Great Britain (Evans 1985, Evans & King 1987). Evaluations are still in progress, but knowledge that the predator is attracted to the frass of three North American Dendroctonus species (Miller et al. 1987) suggests its possible use against species other than D. micans.