N. S. Talekar, Asian Vegetable Research and
Development Center
Shanhua, Taina, 74199 Taiwan
A. M. Shelton, Department of Entomology,
Cornell University,
New York State Agricultural Experiment Station, Geneva, New York 14456
With permission from the Annual Review of Entomology, Volume 38, ©1993, by Annual Reviews
Perspective and History of Pest Status
In recent years, the diamondback moth, Plutella xylostella (L.) (Lepidoptera: Yponomeutidae), has become the most destructive insect of cruciferous plants throughout the world and the annual cost for managing it is estimated to be US $1 billion (168). Members of the plant family Cruciferae occur in temperate and tropical climates and represent a diverse, widespread and important plant group which includes cabbage, broccoli, cauliflower, collards, rapeseed, mustard and Chinese cabbage, the most important vegetable crop grown in China (90), the most populous country in the world. Although the diamondback is believed to have originated from the Mediterranean area (64), the origin for some of our most important crucifers (185), the diamondback moth now occurs wherever crucifers are grown and it is believed to be the most universally distributed of all Lepidoptera (107).
Absence of effective natural enemies, especially parasitoids, is believed to be a major cause of the diamondback moth's pest status in most parts of the world (92). Lack of parasitoids in a particular area may have occurred because diamondback moth is better able than its natural enemy complex to become established in newly planted crucifer areas. There are numerous reports on the ability of diamondback moth to migrate long distances (19, 40, 54, 58, 108, 120, 183), but there is no record of migration of any of its parasitoids. Another reason for the lack of effective biological control in an area may be disruption of natural enemies by the use of broad spectrum insecticides. Prior to the introduction of synthetic insecticides in the late 1940s, there were no reports of diamondback as a major pest of crucifers. However, with widespread use of synthetic insecticides on crucifers beginning in the mid-1950s, important natural enemies were eliminated. This, in turn, led to continued use of synthetic insecticides and eventual insecticide resistance and control failures. In 1953, diamondback moth became the first crop pest in the world to develop resistance to DDT (7, 83), and now in many countries diamondback moth has become resistant to every synthetic insecticide used against it in the field (174, 175). In addition, diamondback moth has the distinction of being the first insect to develop resistance in the field to the bacterial insecticide, Bacillus thuringiensis Berliner (62, 86, 151, 164). Insecticide resistance and control failures are now common in tropical climates such as parts of Southeast Asia, Central America, the Caribbean and the southeastern United States. In some of these areas economical production of crucifers has become impossible, a situation similar to the demise of the cotton industry in parts of Central America (106).
The purpose of this paper is to provide the reader with a global overview of the biology and ecology of diamondback moth, its association with its host plants and natural enemies, and past, present and future management strategies. Emphasis has been placed on providing the reader with an overview of the characteristics of diamondback moth and past management practices which resulted in widespread control failures, and to provide insights which will allow better management in the future. This review does not contain all references to diamondback moth, but does provide what we consider to be the most relevant work pertaining to each area we discuss. For a more detailed and broader approach to the literature, the reader is referred to the next section.
Sources of Information
The diamondback moth was the subject of two widely attended international workshops in Taiwan and the proceedings of these meetings (168, 170) are important sources of information. The first volume of an annotated bibliography, published in 1985, compiles most of the literature published until mid-1985 (175). A second volume which includes citations to September 1990 (174) covers those papers published between December 1985 and September 1990. All literature published in Japanese until 1988 has recently been compiled by a group of Japanese entomologists (138). The library of the Asian Vegetable Research and Development Center (AVRDC), Taiwan contains a complete database and collection of reprints for diamondback moth. These resources are available free of charge to scientists interested in diamondback moth research.
Host range and host specificity
The diamondback moth feeds only on members of the family Cruciferae. This diverse family, whose members have various edible plant parts such as roots of radish and turnips, stems of kohlrabi, leaves of cabbage and other leafy brassicas, and seeds of mustard and rape are consumed as fresh, cooked or processed vegetables. Crucifers are grown in tropical and temperate climates and in a variety of cropping systems from backyard gardens to large-scale fully mechanized farming. Crucifers are the most common vegetables in the diet of Asians, as well as important components in the diets of most other cultures. The 1990 FAO production figures (51) indicate that on a worldwide basis cruciferous vegetables were grown on 2.2 X 106 ha with half this production occurring in Asia. When rapeseed is combined to the above figure, it exceeds 17.6 X 106 ha.
In addition to feeding on commonly cultivated crops such as cabbage (Brassica oleracea L. var. capitata), cauliflower (B. oleracea L. var. botrytis), broccoli (B. oleracea L. var. italica), radish (Raphanus sativus L.), turnip (B. rapa L. pekinesis), brussels sprouts (B. oleracea L. var. gemmifera), Chinese cabbage (B. rapa L. cv. gr. pekinensis), kohlrabi (B. oleracea L. var. gongylodes), mustard (B. juncea (L.)), rapeseed (B. napusL), collard (B. oleracea L. acephala), pak choi (B. rapa L. cv. gr. pakchoi), saishin (B. rapa L. cv. gr. saishin), watercress (Nasturtium officinale R. Br.), kale (B. oleracea L. var. alboglabra), diamondback moth also feeds on a large number of cruciferous plants, most of which are considered to be weeds. Diamondback moth maintains itself on these weeds only in the absence of more favored cultivated hosts. The following crucifers have been reported to sustain feeding and reproduction of diamondback moth: Arabis glabra (L.) Bernh., Armoracia lapathifolia Gilib., Barbarea stricta Andrz., Barbarea vulgaris R. Br., Basela alba, Beta vulgaris L. Brassica caulorapha Pasq., Brassica kaber (D.C.) L.C. Wheeler, Brassica napobrassica Mill., Bunias orientalis L., Capsella bursa-pastoris (L.) Medic., Cardamine amara L., Cardamine cordifolia, Cardamine pratensis L., Cheiranthus cheiri L., Conringa orientalis (L.) Dumort, Descurainia sophia (L.) Prantl, Erysimum cheiranthoides L., Galinsoga ciliata, Galinsoga parviflora, Hesperis matronalis L., Iberis amara L., Isatis tinctoria L., Lepidium perfoliatum L., Lepidium virginicum L., Lobularia maritima (L.) Desv., Mathiola incana (L.) R. Br., Norta (Sisymbrium) altissima, Pringlea antiscorbutica, Raphanus raphanistrum L., Rorippa amphibia (L.) Besser, Rorippa islandica (Oeder) Borbas, Sinapsis alba L., Sisymbrium austriacum Jacq., Sisymbrium officinale (L.) Scop., and Thlaspi arvense L. (38, 56, 65, 85, 96, 109, 129). Alternate weed hosts are especially important in maintaining diamondback moth populations in temperate countries in spring before cruciferous crops are planted.
The host range of diamondback moth is limited to crucifers because they contain mustard oils and their glucosides (60, 61, 71, 113, 181, 182). Diamondback moth is stimulated to feed by many glucosinolates, but two of these (3-butenyl-and 2-phenylethyl) are toxic to them at high concentrations (113). The glucosides sinigrin, sinalbin and glucocheirolin act as specific feeding stimulants for diamondback moth and 40 plant species containing one or more of these chemicals serve as hosts. Nonhost plants may contain these stimulants, but also contain feeding inhibitors or toxins (60). Certain chemicals such as sulfur containing glucosinolates or its metabolites, allyl isothiocyanates, are present in crucifers and act as oviposition stimulants (61, 130). Sulfur deficient plants are not attractive to diamondback moth for oviposition (61). At a subcellular level, the oviposition stimulating activity of glucosinolates can be eliminated by treatment with myrosinase or sulphatase enzymes which degrade glucosinolates (130). The oviposition inhibiting property of coumarin present in Melilatus spp. can be overcome by treatment with allyl isothiocyanate but the similar property of an unknown substance in tomato leaf could not be reversed by application of allyl isothiocyanate (61). Allyl isothiocyanate also stimulates egg production in diamondback moth adults (70). Recent studies indicate the presence of unidentified olfactory stimuli that attract diamondback moth to crucifers (121, 125).
General life cycle
Diamondback moth adults become active at dusk and continue so into the night (64). Most adults emerge during the first 8 hours of photophase (124), and mating occurs at dusk of the same day the adults emerge. Female moths start laying eggs soon after mating and the oviposition period lasts four days during which 11 to 188 eggs per female are laid (64). The majority of eggs are laid before midnight with peak oviposition occurring between 1900-2000 hours (11, 124). The ratio of eggs laid on the upper and lower leaf surfaces approximates 3:2, and very few eggs are laid on stems and leaf petioles (10, 64). Eggs are laid preferentially in concavities of leaves rather than on smooth surfaces (61). Lack of light during normal daylight stimulates oviposition but the reverse does not completely inhibit it (11, 163). Plant volatiles, secondary chemicals, temperature, trichomes and waxes on leaf surface all influence oviposition (9, 98, 125, 163, 187). The incubation period, which is influenced mainly by temperature, lasts 5 to 6 days.
Soon after emergence, neonate larvae initiate feeding on foliage. The first instar mines in the spongy mesophyll tissue, whereas older larvae feed from the lower leaf surface and usually consume all tissue except the wax layer on the upper surface, thus creating a "window" in the leaf. There are four larval instars whose duration depends on temperature (18, 77, 98, 140, 141). In 1022 observations during summer in Ontario, Canada, the average duration of the larval instars was 4.0, 4.0, 5.0 and 5.6 days for the first through fourth instars, respectively (64). Faster developmental times are reported in warmer climates and development rates are also influenced by host crop (74).
When the fourth instar has completed its feeding, it constructs an open network cocoon on the leaf surface where it fed and spends a two day period of quiescence marking the prepupal stage. The prepupa sheds its larval skin which remains attached to the caudal end of the pupa. The duration of the pupal period varies from 4 to 15 days depending on temperature (1, 23, 65, 74, 97). Adult moths emerge during scotophase, primarily between 1300 and 1600 hours with a peak at 1400 (124, 139). Adults feed on water drops or dew and are short-lived.
Relationship with Environmental Factors
Diapause Whether diamondback moth diapauses or hibernates in any of its life stages is a controversial topic. In the tropics and subtropics where crucifers are grown throughout the year, all life stages of diamondback moth can be present at any time. In temperate regions where crucifers are not grown year-round, the diamondback's perennial occurrence has led several researchers to believe pupae and/or adults hibernate in host plant debris through the winter (16, 105, 108, 147, 178, 186). However, in none of these studies were insects collected during the coldest months and brought out of hibernation. A single study at Ithaca, New York mentions the presence of motionless diamondback moth adults in crop remnants in the field (64), but it was not clear whether they would have survived the winter. In a study in upstate New York (AMS, unpublished), no diamondback moth pupae survived the winter and, when moth activity was monitored throughout the year using pheromone traps, no moths were caught during the winter months of 1990-1 despite several warm spells lasting for several days. During the same winter in Long Island, New York, however, moths were captured.
The origin of diamondback moths in an area and their ability to survive in that area during noncropping periods remains an important question. Insecticide resistant diamondback moths which can overwinter may pass genes for resistance to subsequent generations but, if no overwintering occurs, genes for resistance will be lost in that area unless new resistant individuals arrive. Future research on overwintering may give more definitive information but, for the present, we assume that diamondback moth does not overwinter in temperate areas and that immigration occurs by moths moving on wind currents or all stages arriving on contaminated seedlings from tropical or subtropical areas where active breeding occurs throughout the year.
Migration Among the various criteria that make diamondback moth one of the most cosmopolitan pests is its ability to migrate and disperse over long distances. In Britain, where mass migration of diamondback moth has been studied extensively, the yearly occurrence of diamondback moth is attributed to migration by adults from the Baltic and southern Finland, a distance > 3000 km (40, 58, 102, 108, 120, 178). These studies indicate that moths are able to remain in continuous flight for several days and cover distances of 1000 km per day, but how the moths survive at such low temperatures and high attitude is not known. In eastern Canada, annual populations of diamondback moth originate by adult migrations from the United States (67, 152). Similarly in Japan, this insect migrates from southwesterly islands, some of which are tropical, to the temperate climate at Honshu and Hokaido (73). Similar migrations probably occur in other parts of the world such as New Zealand, Australia, South Africa and southern parts of Chile and Argentina.
In recent years in the United States, use of seedlings grown in the southern states contaminated with diamondback moth has proven to be a major source of diamondback moth infestations in northern states (151). High levels of seedlings contaminated with diamondback moth larvae which are resistant to many insecticides has lead to major control failures in several states (151).
Small Scale Farming In developing countries of the tropics and subtropics, production of crucifers is characterized by small farms and intensive use of land, labor and pesticides. Because of the need to produce fresh vegetables for residents of large cities on a daily basis, farms are usually located on the outskirts of such population centers or in cleared areas in the highlands with easy access to cities. Cultivation of fresh crucifers is an important source of income and production of healthy looking, damage free vegetables for the relatively wealthy city dwellers is an important consideration in all cultivation practices, especially plant protection. The mainstay of control is the frequent use of insecticides, often applied with backpack sprayers with few safety features to the applicator. In most developing countries introduction of insecticides, all of which are imported from developed countries, face little, if any, registration hurdles common in the West and Japan. As a result, most insecticides, some of which are not registered in the country of origin, are readily available at a reasonable cost. In some countries pesticides are subsidized, especially for staple foods like rice and export crops like cotton and, because of the absence or poor enforcement of restrictions on pesticide use, insecticides registered for rice and cotton are often applied to cruciferous vegetables. All these factors contribute to the overuse and complete dependence on insecticides to control diamondback moth. In tropical countries where crucifers are grown throughout the year there may be 20 generations per year and the sole reliance on insecticides for control facilitates the rapid buildup of resistance. It is no coincidence that the first report of diamondback moth resistance to an insecticide in 1953 came from one intensive production area in the tropics, Indonesia (7, 83), decades before the appearance of resistance even in the tropical areas of the continental United States (89, 104, 151), Hawaii (165) or Japan (8, 184). To overcome resistance farmers often increase doses of insecticides, use mixtures of several chemicals and spray more often, sometimes once every two days. In most of these areas the insecticide cost amounts to between 30 to 50% of the cost of production, well above the fertilizer cost (91). These high levels of insecticide use have caused diamondback moth to become resistant to practically all insecticides in many areas. Additionally, high insecticide use led to excessive insecticide residues on produce and, since pesticide residue monitoring is absent or not enforced, insecticide contaminated crucifers often pass easily through marketing channels.
Because of the lack of proven alternatives and the continued availability of relatively cheap insecticides, insecticides remain the main control tactic. In a few countries such as Malaysia, Indonesia and the Carribean islands, attempts have been made at alternative control tactics such as introduction of parasitoids (92) but the success of these efforts has been thwarted by the continuing indiscriminate use of synthetic insecticides. An Asian IPM program to combat diamondback moth through the use of natural enemies and judicious use of insecticides is now financed by the Asian Development Bank, but it will be years before benefits of this effort are realized. Similarly, the International Atomic Energy Agency has initiated a cooperative study in Malaysia and Indonesia to assess the possibility of managing diamondback moth with a sterile insect program.
Large Scale Farming In developed countries production of crucifers is characterized by large scale farming practices which include the reduction of labor, the increase of management and capital, and the consolidation of land into larger holdings. Large scale farming of crucifers is common in North America and Europe and is becoming more so in Mexico and Central America. In these areas crop protection decisions tend to be similar over relatively large areas. The primary method of control of diamondback moth in large scale farming involves insecticides applied by air or ground rigs. In North America, Europe and New Zealand researchers have incorporated insecticides into IPM programs which utilize scouting and threshold strategies (15, 21, 75, 76, 146, 149, 179). In the United States management of diamondback moth was not difficult until the mid 1980s, but then control failures were seen in Texas (104, 126), Hawaii (165), Florida (80-82, 89) and throughout North America (151). This rather rapid set of control failures was due to several factors including: the relatively warm growing season during the mid-1980s which lead to an increase in the number of generations produced; the lack of rainfall, a major mortality factor for diamondback moth (64); the recent shift to a shorter or no crucifer free period in southern states; the movement of contaminated transplants within and between states, and; the development of resistance. In several areas of the United States where resistance is high, especially Florida, the frequent use of insecticides and lack of adequate control has made crucifer production often unprofitable. In other parts of the world (e.g. Europe and New Zealand), control still appears to be adequate.
Throughout the developed countries, there is a movement to reduce pesticide use. Several northern European countries have mandated, or are in the process of mandating, a 50% reduction in the use of synthetic pesticides by the year 2000; other countries will likely follow this lead. To achieve this end and still produce acceptable quality crucifers, crop protection entomologists must come up with alternatives to the sole reliance on synthetic insecticides. In areas of the United States where control problems are most acute, growers are presently testing such tactics as inoculative releases of parasitoids, conservation of natural enemies, mating disruption and cultural controls.
Cultural Control
Because of the failure of insecticides to control diamondback moth, there is increased interest in cultural controls in commercial crucifer production. Some of the classical control measures that have been tried with some success are intercropping, use of sprinkler irrigation, trap cropping, rotation and clean cultivation.
Intercropping Intercropping, the practice of growing more than one crop species together, is a normal cultivation practice in the tropics where farms are small and land is used intensively. However, in these areas intercropping is not presently used for management of diamondback moth, but rather for horticultural and economic reasons. For some crop-insect situations, intercropping has reduced pest populations because the plants act as physical barriers to the movement of pest insects, natural enemies are more abundant, and/or the chemical or visual communication between pest insects and their host plants is disrupted (133, 135, 148). The earliest successes were in Russia where intercropping cabbage with tomato reduced damage to cabbage by several pests, including diamondback moth (190). This practice, however, had only limited success in India (23, 153), the Philippines (103) and Taiwan (11). At the latter location none of the 54 crops tested for their utility in intercropping had any significant impact on the population of diamondback moth on cabbage. Intercropping with Salvia officinalis L., Thymus vulgaris L. and Trifolium repens L. consistently reduced damage to Brussels sprouts from diamondback moth (44, 45), but these crops would not be economically suitable for most small farmers.
Sprinkler Irrigation All but the first instar of diamondback moth are exposed on the leaf surface and influenced by various abiotic factors. Several reports indicate that rainfall is an important mortality factor in diamondback moth (22, 26, 58, 59, 64, 66, 85, 171, 191) and, thus, it is only a serious pest during the dry season. Overhead irrigation has been shown to reduce diamondback moth injury in cabbage (11, 172) and watercress (112, 166). This reduction is believed to be caused by drowning or physically dislodging the insect from the plant surface. This operation at dusk also reduced mating related flight activity (12) and presumably oviposition. Using sprinkler irrigation to control diamondback moth in crops other than watercress, however, is not practical on a commercial farm because of the high cost and probable increase of diseases such as black rot and downy mildew.
Trap Cropping Before the advent of modern organic insecticides, a common practice was to plant strips of a highly preferred but economically less important plant within a commercial crucifer field. The more preferred crops, primarily white mustard (Brassica hirta) or rape (B. juncea (L.) Czern), attracted diamondback moth which spared the commercial crop such as cabbage, Brussels sprouts and others from its attack (56, 84, 131). Now that the same modern insecticides which made past trap cropping practices obsolete are made obsolete by insecticide resistance, trap cropping is becoming more realistic, especially in developing countries. In India when mustard was alternated with every 15 to 20 rows of cabbage, diamondback moth colonized the mustard and spared the main cabbage crop (154). In order to trap most immigrating diamondback moth adults in a field, mustard must be available throughout the cabbage growing period. Effective trap cropping may eliminate all insecticides because diamondback moth larvae are retained in the trap crop and become heavily parasitized. This cultural control practice is now expanding rapidly in India.
Rotation and Clean Cultivation Crop rotation is rarely practiced for control of diamondback moth populations in intensive vegetable growing areas of the tropics and subtropics because of the high prices which crucifers fetch. However, because continuous planting of crucifers allows continuous generations of diamondback moth which leads to frequent use of insecticides and the development of resistance, crop rotation may become a necessity.
Clean cultivation can be an important factor in the management of diamondback moth. Planting seedbeds away from production fields, and plowing down crop residues in seedbeds and production fields is an efficient and easy management practice. Where transplants are grown in the greenhouse, prevention of infestations by immigrating adults can be accomplished through the use of screening. Infestations in greenhouses are usually controlled through the use of frequent insecticide sprays, and this may lead to insecticide resistance (62, 151). Alternative strategies such as plant resistance, use of pheromone disruption, biological control and other tactics should be investigated.
Plant Resistance
Several studies have surveyed existing germplasm for plant resistance to Lepidoptera, including diamondback moth, in crucifers (20, 39, 42, 43, 46, 69, 123, 128, 150). The most notable resistance came from germplasm in the United States Northeastern Plant Introduction Station. Two types of resistance have been identified from material in this collection (48). In two normal bloom cabbage types, resistance is chemically based and elicits antibiosis or nonpreference in the larvae. Polar fractions of ethanol extracts from these types, when incorporated into an artifical diet, caused levels of mortality similar to that observed with intact plants in the field. However, mortality on these types was much lower than on glossy type cabbages derived from a cauliflower accession (PI 234599). Whole leaf ethanol extracts from this type had no activity in the diet and further studies indicated that resistance was due to a change in behavior by neonate larvae (48) caused by differences in the amount and chemical composition of leaf surface waxes (49). Recent work indicates that application of s-ethyldipropylthiocarbamate to normal bloom cabbages changes the leaf surface waxes to become similar to those of the genetic glossy type and thereby become resistant to diamondback moth neonate larvae (47).
The glossy leaf trait derived from PI 234599 is inherited as a simple recessive gene. Cabbage lines derived from this parent have been successfully tested in Honduras under extreme pressure and provided > 95% control (42). Other genes for glossiness were examined for insect resistance (155, 156) and the results indicated high levels of resistance and that the cause for resistance was leaf surface waxes (49). Especially important are recent results which indicated high levels of resistance can be obtained from glossy dominant genes (155). Currently two major seed companies are developing glossy lines for resistance to diamondback moth (AMS, unpublished).
Sex pheromone
Evidence for sex pheromone emission in female adults of diamondback moth was initially demonstrated in Taiwan (31). The pheromone consists of three components: (Z)-11-hexadecenal (Z-11-16: Ald), (Z)–11-hexadecenyl acetate (Z-11-16:OAC) and (Z)-11-hexadecenyl alcohol (Z-11-16:OH) (5, 29, 35) and is now available commercially. The exact proportion of the three components in a pheromone blend that will attract the maximum number of male moths is influenced mainly by air temperature, possible strain differences in female reponse that might exist at widely spaced locations, but not humidity (99). Extensive studies have been conducted to determine the optimal proportion and loading of the pheromone components, effect of environmental factors, effective distance, longevity, etc. (27-29, 32-36, 88, 94, 99-101) in order to use the pheromone more effectively in the field. The pheromone has been used for monitoring diamondback moth populations in the field (13, 88) and, during the past three years, Japanese scientists have succeeded in achieving mating disruption in cabbage fields using high concentrations of the pheromone (114 - 116). A 1:1 mixture of (Z)-11-16:Ald and (Z)–11–16:OAC, known as KONAGA-CON, is now commercialized in Japan and collaborative multilocation studies in that country have shown promising results (115), but its use is still not cost effective. Similar pheromone disruption studies have been initiated by McLaughlin and his colleagues in Florida.
Biological Control
All stages of diamondback moth are attacked by numerous parasitoids and predators with parasitoids being the most widely studied. Additionally, adults are often attacked by polyphagous predators such as birds and spiders. Although over 90 parasitoid species are reported to attack diamondback moth (57), only about 60 of them appear to be important. Among them six species attack diamondback moth eggs, 38 attack larvae and 13 attack pupae (92). Egg parasitoids belonging to the polyphagous genera of Trichogramma and Trichogrammatodic contribute little to natural control and require frequent mass releases. Larval parasitoids are the most predominant and effective. Many of the effective larval parasitoids are host specific and belong to two major genera, Diadegma and Cotesia (=Apanteles); a few Diadromus spp., most of which are pupal parasitoids, also exert significant control. The majority of these species came from Europe where diamondback moth is believed to have originated. In Moldavia in Romania, 25 species of parasitoids occur and provide 80-90% parasitism of diamondback moth (111).
Southeast Asia, Pacific Islands, Central America, the Caribbean and most of the sub-Saharan Africa are constantly plagued by diamondback moth because these areas lack effective larval parasitoids. This is in contrast to countries in continental Europe and North America which are endowed with many Diadegma, Cotesia and Diadromus spp.
Parasitoid Introductions Introduction of exotic parasitoids to control pest insects and weeds has been practiced for decades (41, 78, 160). This approach has considerable promise in the control of diamondback moth; however, it has been practiced only sporadically over the past 50 years. Widespread and often indiscriminate use of insecticides has frustrated recent efforts and delayed the establishment of parasitoids and their beneficial effects.
One of the earliest parasitoid introductions was made in New Zealand. When no significant parasitism of diamondback moth by three native larval parasitoid spp. was found (110, 134), Diadegma semiclausum Hellen and Diadromus collaris (Gravenhorst) were introduced to New Zealand from England (68, 180). These introductions continue to suppress diamondback moth populations and the challenge today is to incorporate this natural control into a commercial IPM program (15).
A somewhat similar situation existed in Australia where, prior to the introduction of effective exotic parasitoids, diamondback moth caused serious damage (192). Among the introduced parasitoids, D. semiclausum became established throughout Australia and on the island State of Tasmania. Diadromus collaris was established principally in Queensland, New South Wales, Victoria and Tasmania and C. plutellae Kurdjunov in Australian Capital Territory, New South Wales and Queensland. These introductions resulted in heavy parasitism of diamondback moth (72 to 94%) and marked reduction in damage to crucifers (57, 63, 192).
Efforts to control diamondback moth in Indonesia by introduction of parasitoids was initiated in 1928 (50). However, it was only in the early 1950s that the exotic parasitoid D. semiclausum was actually introduced from New Zealand into the crucifer growing areas in the highlands of Java and became established (189). However, because of overuse of insecticides, it was not until mid 1980s that the beneficial effects of this parasitoid in the control of diamondback moth in the field were realized (143). With the substitution of B. thuringiensis in the early 1980s the parasitoids proliferated (143). This parasitoid has now been introduced from Java to the highlands of other islands in Indonesia.
An identical situation existed in the Cameron Highlands of Malaysia, the major vegetable production area for Malaysia and Singapore, where crucifers are grown year round and diamondback moth is a serious pest. Prior to 1977 only one parasitoid, Tetrastichus ayyari Rohwer was present in this area but it did not adequately suppress diamondback moth populations (91, 117). Malaysian entomologists in 1977-78 introduced one larval parasitoid, D. semiclausum, and two pupal parasitoids, Tetrastichus sokolowaski Kurdjumov and D. collaris, in this area (119). Although these parasitoids were recovered from diamondback moth in this area one and six years after the release, parasitism was only 6% for both D. semiclausum and D. collaris. Tetrastichus sokolowski was recovered in 1978 but was not found in 1984 (37). Parasitism by C. plutellae, which entered this area accidently (91), amounted to 11% in 1978 and 20% in 1984. Because insecticides were still effective, farmers continued to use them intensively which kept the parasitoid population very low. Toward the end of the 1980s, however, diamondback moth developed resistance to practically all synthetic insecticides and officials in Singapore, the major market for vegetables grown in Cameron Highlands, rejected cabbage because of high insecticide residues. These two events forced farmers to start using B. thuringiensis (118) and resulted in an increase in the parasitoid population and reduced diamondback moth damage. Surveys in 1989 showed that D. semiclausum and D. collaris, respectively, have become the dominant parasitoids with C. plutellae contributing little control. The combined parasitism has drastically reduced the need for insecticide applications and cabbage production is increasing (162).
In 1970, C. plutellae obtained from India was released in the Caribbean islands of Grenada, St. Vincent, St. Lucia, Dominica, Antigua, Montserrat, St. Kitts and Nevil, Belize, Trinidad, Barbados and Jamaica. In some of these release sites, the parasitoid has been recovered but it appears to have little effect on diamondback moth suppression. Attempts to introduce A. vestalis Haliday and D. collaris were not successful (17, 194). Re-introduction of C. plutellae into Jamaica in early 1989 resulted in its establishment with parasitism between the date of introduction (5.4%) increasing to 88.7% by March of 1990. This parasitism resulted in reduced plant damage from 75% before introduction to 38% in March 1990 (3). On the Cape Verde Islands off Africa, locally occurring predators and parasitoids were not able to exert adequate pressure to control diamondback moth (93). Introduction of C. vestalis and T. sokolowskii and the use of B. thuringiensis resulted in the establishment of these parasitoids and control of diamondback moth (188).
In Taiwan diamondback moth has been a serious problem since the mid-1960s (24). Cotesia plutellae, reported to parasitize diamondback moth since at least 1972 (30), was not able to give adequate control so D. semiclausum was imported from Indonesia and released in the lowland crucifer growing areas in 1985 (9); however it failed to become established. When broad spectrum insecticides were replaced by B. thuringiensis, C. plutellae became established and provided adequate control, but D. semiclausum still did not become established (167). However, when D. semiclausum was introduced in the highlands, within one season it parasitized > 70% and towards the end of that season diamondback moth could not be found in the field (12). Diadegma semiclausum now occurs throughout the highland areas of Central Taiwan and provides substantial savings in diamondback moth control (169, 176). Differential establishment of these parasitoids in highland and lowland areas appears to be due to their different temperature requirements. Laboratory studies indicate a temperature range of 20 to 30oC is optimum for parasitization of diamondback moth by C. plutellae and 15 to 25oC for D. semiclausum (173). At temperatures approaching 30oC parasitism by D. semiclausum drops sharply. In tropical and subtropical areas only temperatures in the highlands are suitable for the establishment of D. semiclausum whereas temperatures in the lowlands are suitable for C. plutellae (169). These studies help explain the successful establishment of D. semiclausum in the highlands of Indonesia, Malaysia, and Taiwan and C. plutellae in the lowlands of the latter two countries.
In the highlands areas of northern Philippines a single release of D. semiclausum in 1989 at the beginning of the season resulted in 64% parasitism of diamondback moth at harvest (127). The ideal temperature range of 12 to 28oC most of year in this area is expected to help in the spread of D. semiclausum in the remaining crucifer areas in this region.
Insecticides
Chemical Use Patterns Because diamondback moth larvae feed on cruciferous vegetables which usually have high cosmetic standards, effective control is necessary and historically the mainstay of control has been the use of synthetic insecticides. In a reveiw of publications on diamondback moth (175), nearly a third of the papers focused on insecticidal control and the majority of these involved screening compounds.
General use patterns of insecticides vary widely over geographic locations and decades. The driving forces behind these changing patterns are the development of new, more effective insecticides and the loss of older insecticides because of resistance. The most dramatic patterns have occurred in Southeast Asia where diamondback moth is abundant. The best example of the rapid change in use patterns is illustrated by Rushtapakornchai & Vattanatangum (136) who complied a list of screening results in Thailand from 1965 to 1984. A dominant product like mevinphos provided excellent control in 1965, fair control in 1974 and poor control in 1984. In 1976, permethrin was introduced and provided excellent control in the central region, but provided only fair control two years later. In the early 1980s insect growth regulators (IGRs) were introduced and IGRs like triflumuron provided good control in 1982 but poor control by 1984. Other biologicals like B. thuringiensis were introduced in the early 1970s and provided fair to good (but never excellent) control when they were first introduced. Because of the lack of excellent control when used alone, the primary use of B. thuringiensis has been in IPM programs with thresholds and conservation of natural enemies.
Reports from Thailand on the introduction and eventual failure of many insecticides (136, 195) are not unique. Similar patterns have also been documented in other parts of the world like Taiwan (159), Japan (62), Malaysia (161), the United States (80, 89, 104, 126, 151, 165), and Central America (6). Because of the magnitude of the diamondback moth problem and the worldwide importance of cruciferous vegetables, new potential insecticides such as genetically improved strains of B. thuringiensis, neem (145), macrocyclic lactones (2, 52, 161), baculoviruses (79) and fungi (132) are being explored. However, as with all previously used products, their long-term effectivenes is questionable because of resistance.
Insecticide Resistance Factors which influence the development of resistance in diamondback moth include high fecundity and reproductive potential, rapid turnover of generations, a long growing season, extensive acreage of crucifers and frequent insecticide applications (104, 159, 193).
Diamondback moth has a long history of eventually becoming resistant to every insecticide used extensively against it. In 1953, Ankersmit (7) noted the development of resistance to DDT in Lembang, Indonesia. Subsequently, diamondback moth has become resistant to most of the other major classes of insecticides used in Indonesia. In the Philippines the development of resistance by diamondback moth was first reported in 1974 by Barroga (14) when she confirmed field failures to EPN and mevinphos. In Malaysia diamondback moth has become resistant to all groups of conventional insecticides (161). Additional reports from Taiwan (25, 159), Japan (62, 177), Australia (4), and North America (104, 151, 165) have documented resistance to a variety of insecticides. Once resistance occurs, several authors have reported success with tank mixes (81, 89), but the long term usefulness of this strategy is questionable.
As a first step in managing resistance, Magaro and Edelson (104) developed a technique using disposable cups in which larvae were exposed to discriminating doses of several insecticides at the LC90 level. Such techniques can be used to identify populations resistant to specific insecticides and enable growers to use alternatives, if they are available. Understanding the genetics, field dynamics, mechanisms and stability of resistance is necessary if resistance management is to succeed, but too often these studies are done once resistance has developed. Studies on the mechanism(s) of resistance to carbamate, organophosphate, pyrethroid, abamectin and benzoylphenyl urea (BPU) insecticides against diamondback moth are summarized by Sun (158).
Insect growth regulators and pathogens offer promise as alternatives to broad spectrum insecticides which often disrupt the control exerted by natural enemies. Products like BPU which interfere with chitin synthesis provide an alternative to the more common classes of insecticides and may help in resistance management. However, unofficial reports from Thailand indicate that diamondback moth has already developed significant resistance to several BPUs only 2-3 yr after their introduction (122). Additional studies indicated that insects collected from Thailand in 1988 showed resistance to several IGRs including chlorofluazuron, diflubenzuron, hexaflumron and several experimental IGRs. Resistance was considered due to a recessive monofactorial and autosomal gene. Rearing populations for 40 generations did not result in loss of resistance (87).
Bacillus thuringiesis offers tremendous hope for diamondback moth control because of its specificity and the fact that no serious control failures in the field have been documented despite the use of B. thuringiensis for > 20 years (55). However, Kirsch and Schmutterer (86) found low efficacy of B. thuringiensis for controlling diamondback moth in the Philippines and speculated that it may have been due to resistance. Tabashnik et al (164) reported results on populations of diamondback moth collected from commercial fields of watercress, cabbage and broccoli in Hawaii. In laboratory bioassays, diamondback moth collected from watercress fields that had been heavily treated with B. thuringiensis had LC50 and LC95 values 25 to 33 times greater than those of two susceptible laboratory colonies. In 1990 Shelton & Wyman (151) collected 11 populations of diamondback moth from Brassica plants in six states and Indonesia and tested for their responses to two natural formulations of B. thuringiensis and a genetically engineered form of the bacterial preparation. High levels of resistance (>200-fold) were found in populations which originated from Florida. Additional field studies conducted in 1992 (A. M. Shelton, unpublished) have documented control failures in several locations in Florida to commercial formulations of B. thuringiensis subspecies kurstaki and laboratory assays of these same populations have documented LC50 values > 1000-fold. In these same locations, field tests with a formulation of B. thuringiensis subspecies aizawai provided adequate control and laboratory assays indicate < 10-fold variation in LC50 values. Populations which had high LC50 values originated from fields in which B. thuringiensis had been used extensively. Hama (62) found high levels of resistance to B. thuringiensis in glasshouses in Osaka where watercress had been grown throughout the year. These studies demonstrate that with frequent foliar applications of available B. thuringiensis products, resistance and control failures to the HD-1 isolate of the kurstaki serotype of B. thuringiensis will occur in the field.
Resistance to B. thuringiensis is thought to be due to lack of binding of the crystal to the brush-border membrane, either because of strongly reduced binding affinity or the complete absence of the receptor molecule (53). Recent studies (C. W. Hoy, unpublished) indicate that larvae do not avoid droplets containing B. thuringiensis on treated foliage but do consume less foliage after they ingest the droplets. Tabashnik et al (personal communication) conducted a genetic analysis which indicates that resistance to B. thuringiensis was autosomal, recessive and controlled primarily controlled primarily by one or a few loci. The current B. thuringiensis products registered in the United States for diamondback moth control contain only the HD-1 isolate of the kurstaki serotype, and that isolate contains only the Cry IA and Cry II toxins (72). A recent report from Malaysia (161) indicated that a diamondback moth population which had a resistance ratio of 112 to the HD-1 isolate of the kurstaki serotype had a resistance ratio of only 3.3 to a product which contains the aizawai serotype which has additional toxins. Although published reports indicate a lack of cross resistance between some serotypes of B. thuringiensis and thereby offer some hope for managing B. thuringiensis resistance, limiting selection pressure to any of the toxins will be necessary if B. thuringiensis is to remain a durable insecticide complex. We should be warned by the work of Jansson and Lecrone (82) who reported that genetically improved strains of B. thuringiensis were effective in managing diamondback moth during the first two years of a study in Florida, but efficacy declined markedly in the third year.
Integrated Pest Management
For the past 30 years farmers have depended exclusively on insecticides to control diamondback moth, but resistance to presently available insecticides and lack of new insecticides has stimulated research on alternate control measures. In some cases these alternatives are essentially the same ones that were discarded in favor of synthetic insecticides. Since parasitoids play such an important role in checking diamondback moth populations, introduction and conservation of parasitoids will be basic to any sustainable IPM. To implement IPM a coordinated effort among growers must occur since practices of one grower will influence that of his neighbor. This applies to the development of insecticide resistance or the introduction and conservation of natural enemies. Such coordination will be most needed in small scale agriculture where farms are often < 0.1 ha and where many farms in an area are owned by different growers. An example of a successful coordinated effort was the establishment of D. semiclausum in the highlands of Indonesia, Malaysia, Taiwan and the Philippines and the conservation of the parasitoid with B. thuringiensis (119, 127, 143, 169). A similar successful program based on introduction and conservation of natural enemies was developed in Missiouri (Biever, personal communication). An IPM project funded by the Asian Development Bank will soon cover all countries in Southeast Asia where, if not already present, D. semiclausum will be introduced in the highlands and C. plutellae in the lowlands. In the lowland areas of the tropics and subtropics where temperatures are high, C. plutellae is the only larval parasitoid that can survive. Although this parasitoid has been established in a number of crucifer growing areas in the tropics and subtropics (171, 176), except for Cape Verde Islands this parasitoid alone is not effective in controlling diamondback moth and supplemental measures are required.
In areas where other pests besides diamondback moth are important, their management must also be taken into consideration. For example, Crocidolomia binotalis (Zeller) is a major pest of crucifers in the highlands of Indonesia and presently marketed strains of B. thuringiensis are not effective against it. Growers who have used synthetic insecticides routinely against it have caused occasional flareups of diamondback moth due to insecticide induced mortality of D. semiclausum (144). Throughout much of North America, cabbage is also attacked by imported cabbageworm, Artogeia rapae (L.), cabbage looper, Trichoplusia ni (Hubner), onion thrips, Thrips tabaci Lindeman, and cabbage aphids, Brevicoryne brassicae (L.). Presently, only commercial control of onion thrips can be accomplished using host plant resistance (150, 157) and B. thuringiensis is not effective against aphids and only marginally effective against cabbage looper. Other control tactics which are compatible with diamondback moth control must be developed for these other pests.
Because of the magnitude of control failures of diamondback moth, as well as the pressure to reduce insecticide input in small and large scale agriculture, both systems must be open to alternatives to broad spectrum insecticides. Traditionally, such ideas as trap cropping, adult trapping and pheromone disruption were considered more amenable to small scale agriculture, but this is no longer true. Researchers in India have demonstrated the benefits of using a mustard (B. juncea) trap crop to attract diamondback moth away from the principal crops (154), thus reducing the need for insecticides to a maximum of two sprays compared with 10 or more per season for conventional control methods. A team of Thai and Japanese scientists has demonstrated the utility of yellow sticky traps to capture diamondback moth adults thereby reducing their oviposition and subsequent damage by larvae (137). In fields with such traps three sprays of B. thuringiensis achieved better control and twice as much yield as five sprays of B. thuringiensis mixed with mevinphos in a check field. It may be possible to combine mustard trap cropping and yellow sticky traps to reduce the need for insecticides even more. In Japanese field tests of mating disruption by pheromones, populations of diamondback moth have been reduced by 95% compared to control fields (116). Preliminay tests in Florida have not been as successful but pheromone disruption may still serve as an important component along with parasitoids and B. thuringiensis.
The concept of sampling populations and treating when thresholds are exceeded is fundamental to IPM and has been promoted in developed countries (15, 21, 75, 76, 146, 149, 179) and in many developing countries of the tropics (23, 24, 95, 137, 142). Its adoption, however, is hindered because it requires regular scouting by trained personnel who may not be available. In developing countries adoption of IPM is also hindered because many farmers cannot differentiate pests from beneficials, some farmers have difficulty in counting because of their illiteracy, and resistance to multiple insecticides makes most insecticide applications useless. Thus, in the tropics and subtropics community wide management will most likely rely primarily on the release and establishment of as many parasitoids as possible and the use of cultural practices.
In the last 40 years the diamondback moth has become one of the most difficult insects in the world to control because of its intrinsic biology and ecology and its large host range which includes many crops which have high cosmetic standards. Central to control failures is the development of resistance by diamondback moth to every insecticide which has been widely used against it, including B. thuringiensis. Because of insecticide resistance, concern for insecticide residues on the crop and in the environment, and deleterious effects of synthetic insecticides on natural enemies, alternatives to the regular use of synthetic insecticides are sorely needed.
Parasitoids, especially D. semiclausum and C. plutellae have been tremendously successful in controlling diamondback moth populations in the highlands and lowlands, respectively, in Southeast Asia and provide a model for the basics of a successful IPM program. Importation of these or functionally similar biological control agents can serve as the basis of a management program, but this will require a switch to insecticides which are compatible with natural enemies. Use of B. thuringiensis in this context has proven successful in several parts of the world, but the isolated cases of resistance to B. thuringiensis provide warning for the future. How stable this resistance is and what is the potential for cross-resistance between toxins of various strains, isolates and serotypes of B. thuringiensis are questions which must be addressed if B. thuringiensis is to remain a viable tool for diamondback moth management.
Past experiences with diamondback moth management have reinforced the belief that single component strategies will fail. New technologies such as host plant resistance, development of new pathogens and insecticides and mating disruption with pheromones must become available to complement our traditional strategies of biological control, trap crops, host free periods and the like. Because of the importance of crucifers in the human diet and local and world economies, entomologists will continue to be challenged to develop rational and sustainable management systems for diamondback moth.