Abstract

Outbreaks of diamondback moth, Plutella xylostella (L.), are being documented more frequently in many regions of the world and the causes for the outbreaks are often associated with insecticide resistance and lack of natural enemies, the latter often due to use of broad-spectrum insecticides. Growers are likely to think of insect management on an individual field basis (i.e. their fields) without sufficient regard to the larger context in which the insect operates on a regional basis, and this may limit their ability to manage the pest. Like any insect, P. xylostella populations fluctuate on an intra-regional and inter-regional basis and outbreaks in either case may be due to many factors. In this paper two causes of specific outbreaks of P. xylostella, movement of contaminated plants and an unusual climate within a region, are discussed.

Keywords

transplant, contamination, dispersal, insecticide resistance, permethrin, methomyl, seedlings

Introduction

Outbreaks of diamondback moth, Plutella xylostella, frequently occur in various parts of the world (Talekar & Shelton 1993) and result in severe losses (Javier 1992, Shelton 2001). Outbreaks are often the result of a population of P. xylostella having developed resistance to the available insecticides or lack of sufficient control by natural enemies. In the latter case, populations of natural enemies, which would have provided some level of control, may have been reduced due to the use of the broad spectrum insecticides used against P. xylostella and against which P. xylostella may have developed resistance. However, outbreaks may also be caused by movement of P. xylostella or climatic conditions which favour host development or reduce the occurrence of fungal diseases.

Growers are likely to think of insect management on an individual field basis (i.e. their fields) without sufficient regard to the larger regional context in which the insect operates, and this may limit their ability to manage the pest. Intra-regional and inter-regional approaches are needed for more sustainable management practices and these are being developed through approaches such as investigations of movement studies of P. xylostella and its natural enemies and the ‘window strategy’ for resistance management.

In this paper I will discuss two examples of how an improved understanding of inter-regional movement of contaminated plants and an unusual climate within a region can affect outbreaks of P. xylostella. Full reports of both papers have been published, as noted in the section headings.

Movement of contaminated plants and resistant insects (Shelton et al. 1996)

In New York State, ca. 5,000 ha of cabbage, Brassica oleracea capitata (L.) are grown annually and beginning in the late 1980s we began to see problems in controlling P. xylostella with our commonly-used insecticides. Because of the short growing season in New York and the absence of P. xylostella overwintering in an adjacent area (Ontario, Canada) (Harcourt 1986), high levels of resistance would not be expected to have developed in New York. Thus, we suspected insects either flew into our area or were being transported on plants which were grown in another location and shipped to New York for transplanting. It is estimated that nearly 70% of the cabbage grown in New York is grown from transplants and the majority of these are produced in other states.

From 1989 through 1992, cabbage transplants were obtained from growers or brokers in New York who received shipments of transplants from Florida, Georgia and Maryland, the states which supply the majority of transplants to New York. We also obtained locally grown transplants as a check. We sampled cabbage transplants as they arrived in crates or boxes at the growers’ fields or at the brokers. From 1989 to 1992 we sampled a total of 49 different shipments of transplants. As soon as a shipment was received, a sample of transplants was inspected visually for P. xylostella larvae. In addition, transplants were inspected for cabbage looper, Trichoplusia ni (Hübner), imported cabbageworm, Pieris rapae (L.) and cabbage webworm, Hellula rogatalis (Hulst). Plutella xylostella larvae collected during the first inspection were counted and transferred to rape seedlings, Brassica napus, and reared (Shelton et al. 1991) for insecticide assays. The inspected transplants were then placed in soil in large pots and kept for two additional weeks in a greenhouse at 20–25°C, at which time a second inspection was performed.

In addition to examining the plants for insect contamination, we also tested some colonies for their susceptibility to commonly used insecticides. In 1989, we were able to establish four colonies of P. xylostella: two from Florida and one each from Maryland and Georgia. In 1990 we established one colony of P. xylostella from Florida and one from Georgia. All colonies were established from the P. xylostella taken from the transplants and then we used the F₂ generations to test for susceptibility to permethrin and methomyl using a leaf dip bioassay (Shelton et al. 1993a).

Samples collected from 1989 to 1992 documented that P. xylostella, was introduced into New York in early spring on cabbage transplants grown in the southern United States (Table 1). During 1989, transplant shipments from five transplant companies in Florida, Georgia and Maryland had average seasonal infestations ranging from 1.3 to 3.5 P. xylostella per 100 transplants. During June, when the majority of transplants arrived in New York, P. xylostella infestations were as high as 12.8 insects per 100 transplants on an individual shipment. Infestations by T. ni, P. rapae, and H. rogatalis on an individual shipment were as high as 19.7 insects per 100 transplants. Compared with a standard susceptible field population, the P. xylostella which were collected from transplants demonstrated moderate to high (> than 100-fold in one case) levels of resistance to permethrin or methomyl (Table 2). In 1990, average seasonal infestations per transplant company varied from 0.3 to 12.0 P. xylostella per 100 plants, but an individual shipment from Florida had 30.4 P. xylostella per 100 transplants. A population of P. xylostella collected in 1990 from Florida transplants had >200-fold resistance to methomyl. Despite intensive treatments, a New York grower who used the transplants with high contamination of resistant P. xylostella was unable to achieve acceptable control in his field. Samples collected from 1989 to 1992 from a transplant grower in Maryland indicate that better management in the field can reduce contamination levels to <0.5%.

As a result of this work, the Cornell Cabbage IPM Program now recommends that growers inspect transplants before putting them in the field and consider rejecting the load if >5% of the plants are contaminated with P. xylostella. While New York growers depend on having transplants as part of their production practices, transplant growers in the southern states should realise that their management practices may influence the level of P. xylostella control that can be obtained by the growers in other regions who receive their transplants. Since this study was performed and presented to New York cabbage growers, they have become acutely aware of the risk involved in receiving contaminated transplants, especially from regions in which insecticide resistance in P. xylostella is known to occur. As a result, some New York growers have constructed their own greenhouses to ensure their plants can be available at the proper time and be free from contamination of resistant P. xylostella.

Table 1. Insects (P. xylostella and others) found on cabbage transplants shipped to New York by companies from other states, 1989–1990^a

		April		May		June		Season average
Company	n	P.x	otherb	P.x	other	P.x	other	P.x	other
1989
Georgia A	2957	0	0	2.7	0.1	c	c	1.3	0
Georgia B	1059	1.3	0.1	c	c	c	c	1.3	0.1
Georgia C	1892	4.3	0.2	c	c	2.7	0.4	3.4	0.3
Maryland A	8755	c	c	0.2	0.5	7.3	5.6	3.5	2.8
Florida A	5702	0	0	0.3	0.1	8.2	0.1	2.5	0
New York A	3190	c	c	c	c	1.1	0.2	1.1	0.2
New York B	1039	c	c	c	c	1.1	0.2	1.1	0.2
New York C	512	c	c	c	c	0.6	0.2	0.6	0.2
1990
Florida B	2280	6.7	0	17.4	0	c	c	12.0	0
Georgia B	2022	c	c	3.7	0	c	c	3.7	0
Georgia C	1006	c	c	1.8	0	c	c	1.8	0
Maryland A	3599	c	c	0.3	0	0.3	0	0.3	0

a Values listed are [(# insects/# plants inspected) x 100]

b Other insects included imported cabbageworm, cabbage looper and cabbage webworm

c No transplants intercepted from source during that particular month

Table 2. Susceptibility to permethrin and methomyl of III instar Plutella xylostella larvae obtained from US companies producing southern transplants

		Permethrin			Methomyl
Company	n	Slope±SE	LC50 (90% CL) mg (AI)/mg	RRa	Slope±SE	LC50 (90% CL) mg (AI)/mg	RR
Virginia (standard)	105	1.39±0.27	0.083 (0.035–0.166)	1.0	0.88±0.17	0.26 (0.11–0.52)	1.0
1989
Florida A (site 1)	105	1.72±0.30	0.687 (0.285–1.814)	8.3	2.11±0.49	28.5 (18.0–47.5)	109.6
Maryland A	105	1.77±0.30	0.326 (0.176–0.863)	3.9	0.90±0.15	3.35b	12.9
Georgia C	105	0.74±0.25	0.029b	0.3	0.34±0.14	1.24b	3.8
Florida A (site 2)	105	2.03±0.30	0.607 (0.381–0.982)	7.3	2.70±0.20	8.46 (5.00–14.2)	32.5
1990
Florida B	175	1.55±0.22	1.662 b	20.0	0.85±0.150	52.76b	202.7
Georgia B	175	2.43±0.36	0.349 (0.236–0.528)	4.2	0.76±0.11	2.32 (0.68–9.54)	8.9

^aRR is the resistance ratio determined by dividing the LC₅₀ for a population by the LC₅₀ for the standard population (i.e. Virginia)
^bThe 90% CL could not be determined because g > 0.5 (Russell et al. 1977)

Favourable climatic conditions lead to outbreaks of P. xylostella (Shelton et al. 2000)

Climatic conditions, including higher temperatures and decreased rainfall, have been cited as major factors which regulate the population dynamics of P. xylostella (Harcourt 1986). Increased temperatures can lead to the production of more generations per season and increased rainfall can lead to increased incidence of fungal diseases (Talekar & Shelton 1993), direct mortality of small larvae, or perhaps even to mating disruption (Tabashnik & Mau 1986). In fact, sprinkler irrigation has been shown to reduce damage by P. xylostella in cabbage (McHugh & Foster 1995). However, because of the history of P. xylostella resistance, outbreaks are often attributed to insecticide resistance rather than favourable conditions. Beginning in 1997 we had a chance to assess an outbreak of P. xylostella in California to determine its cause(s).

California is the leading U.S. producer of broccoli where nearly 50,000 ha are grown annually with a farm gate value of nearly $0.5 billion. The major production areas extend from Monterey Bay to the Imperial Valley (Figure 1). In 1997 there was an outbreak of P. xylostella, which resulted in crop losses estimated to be > $6 million (Sances 1997). In the central valley and north coast regions of California, P. xylostella is not considered a major pest, although it was in 1997. To help assess the possible causes of the 1997 outbreak we conducted a survey of P. xylostella in the principal broccoli growing regions of the state. The survey consisted of evaluating populations for their susceptibility to three commonly used insecticides (methomyl, permethrin and Bacillus thuringiensis subsp. kurstaki). A single collection of approximately 300 P. xylostella larvae was made from each of nine locations (Figure 1) between October and November of 1997. All insects were transported to the New York State Agricultural Experiment Station where bioassays were performed. An insecticide susceptible population of P. xylostella, Geneva 88 (Shelton et al. 1993a), was also reared at the same laboratory for comparison. Populations were cultured on rape seedlings (Shelton et al. 1991) and toxicity of the insecticides was measured using a cabbage leaf dip bioassay similar to that reported previously (Shelton et al. 1993a,b).

Figure 1. The major production areas for cole crops in California and the nine collection sites of Plutella xylostella in 1997.

In addition to the assays, insecticide records for 1997 were collected for each field and weather data were collected from stations near the sites identified in Figure 1. The latter was done to assess whether climatological data may help explain the population outbreaks observed in 1997. Precipitation and average daily temperatures were summarised for each month for the following locations, which constitute the principal broccoli production areas where outbreaks occurred: Watsonville, Salinas, King City, Coalinga, Santa Maria and Oxnard. To provide a comparison of the climatic conditions in 1997 with previous years, we averaged the monthly temperature and precipitation data of these five sites and compared these averages to the 30-year average (1961–1990). All data were compiled using information from weather observing sites supervised by the National Oceanic and Atmospheric Administration/National Weather Service and received at the National Climatic Data Center (Ashville, NC).

Elevated levels of resistance were seen only with permethrin and seven of the nine populations had tolerance ratios (RR) >100 (Table 3). Only the two populations collected from the Imperial Valley area had RR values similar to Geneva 88, indicating that these populations had not been intensively selected for resistance and also that the Geneva 88 population was a realistic standard for these assays. Traditionally, cole crops are grown for only 3–4 months in the Imperial Valley, compared to year-round in the rest of California, so selection pressure would be reduced. A significant difference, based on non-overlapping of the 95% FL of the LC₅₀ values, was observed between Imperial Valley #1 and Geneva 88 populations, but this translated to a RR value of only 1.6. Based on previous reports, this difference would not result in control failures in the field, although RR values for permethrin of >100 would (Shelton et al. 1993a). For methomyl, the three most susceptible populations (Imperial Valley #1, Imperial Valley #2 and Geneva 88) had LC₅₀ values not significantly different from each other, but these three were significantly different from the other populations. However, all populations had RRs <10 and, based on previous reports (Shelton et al. 1993a), these levels would probably not result in control failures in the field. For B. thuringiensis there were significant differences between some populations, but all populations except one had RR values <10. The Santa Maria population had a RR value of 11.1 and this value is probably borderline for control problems in the field (Perez & Shelton 1996).

Table 3. Resistance ratios (RR) of commonly used insecticides against populations of Plutella xylostella collected from California in 1997. For complete table including slopes, LC₅₀ values and 95% CL see Shelton et al. 2000

Population	Btka	Methomyl	Permethrin
G88	1.0	1.0	1.0
Imperial Valley #2	5.1	0.7	1.3
Soledad	6.7	3.9	110.7
Guadalupe	6.0	3.9	121.7
Oxnard	4.8	6.5	206.3
Santa Maria	11.1	6.6	110.3
Sprecklers	5.3	4.5	145.3
Coalinga	1.4	7.0	154.3
Ocean Cliff	3.9	7.1	126.0
Imperial Valley #1	3.6	0.8	1.7

^aBacillus thuringiensis subsp. kurstaki

Weather data provide some insight into possible causes for the outbreak in 1997. Hot, dry conditions are known to favour outbreaks of P. xylostella and the mild winter of 1996/7 and the below normal rainfall and warmer conditions in the growing season of 1997 met these criteria. Temperatures during the winter of 1996/7 were above normal and this continued through until the normal harvest time of late August of 1997. During the month of May when much of the broccoli was in the ground, daily mean temperatures were 110% of normal. Rainfall during the main production period, February–August, was well below normal. In fact, only in August did the precipitation exceed 50% of the 30-year average and then it only reached 67% of the norm.

Our surveys indicated that growers used an array of insecticides in each of the fields from which we obtained populations. At one time or another, however, most growers used either methomyl or a B. thuringiensis product. Based on our assays, these materials should have provided adequate control under normal circumstances, if applied correctly. However, because of favourable climatic conditions for P. xylostella that resulted in much higher than normal populations throughout the year, it appears that the insects could not be controlled to the level required, regardless of which insecticides were used unless, perhaps, they were used more frequently than they were in 1997.

In 1998 there were no reported significant outbreaks of P. xylostella in California cole crops. Most likely this was primarily due to the considerably higher rainfall (300% in some areas) which reduced populations especially during the winter and spring. However, it could also be due to the fact that spinosad became registered in the fall of 1997 and was used in 1998, or that growers were able to get control with other materials such as methomyl or Bacillus thuringiensis. Most likely it was a combination of reduced insect pressure and the use of effective insecticides. In years in which populations will not be so suppressed by environmental conditions, it will be important to know the geographic distribution of resistance to specific classes of insecticides so they can be avoided.

Discussion

The two examples of outbreaks of P. xylostella mentioned above help illustrate the importance of regional management. The transplant growers whose practices led to insecticide resistance within their greenhouses or fields also contributed to a regional problem when they shipped contaminated plants. It is unlikely that federal and state legislation could have prevented this situation with existing resources, although interstate movement of plants is subject to certification. The size of P. xylostella eggs as well as the vast numbers of plants moved makes detection difficult. In the New York example, the ‘solution’ for New York growers was to take their business elsewhere, and many of them have done this by purchasing greenhouse plants from Canada or constructing their own greenhouses. Growers in other regions of the world have become sensitised to the problem of moving contaminated transplants because of the New York example, so perhaps this will decrease the likelihood of movement of contaminated plants being a source of outbreaks of P. xylostella.

Even if contaminated plants are the source of an initial infestation and even if this is exacerbated by the insects being resistant to available insecticides, outbreaks will occur only if favourable weather conditions exist. Outbreaks of P. xylostella are far less likely to occur in cool, wet conditions (Harcourt 1986, Talekar & Shelton 1993), but besides overhead irrigation (often unavailable to many growers and deleterious to disease management programs) there is little that can be done to manipulate the weather. Perhaps the best advice that can be provided to growers is to advise them of the need for enhanced awareness of the potential for regional outbreaks before and during times of favourable climatic conditions and to stress the need to begin their field season with transplants not already contaminated with P. xylostella, especially if they are resistant to the insecticides they intend to use.

Acknowledgements

I thank the many colleagues, past and present, whose work contributes to our present knowledge base of P. xylostella.

References