Heterapoderopsis bicallosicollis (Coleoptera ... - PubAg - USDA

BIOLOGICAL CONTROLÑWEEDS

Heterapoderopsis bicallosicollis (Coleoptera: Attelabidae): A Potential Biological Control Agent for Triadica sebifera YI WANG,1,2 JIANQING DING,1,3 GREGORY S. WHEELER,4 MATTHEW F. PURCELL,5 2 AND GUOAN ZHANG

Environ. Entomol. 38(4): 1135Ð1144 (2009)

ABSTRACT Native to China, Chinese tallow, Triadica sebifera L. Small (Euphorbiaceae), is an invasive plant in the southeastern United States. The leaf-rolling weevil, Heterapoderopsis bicallosicollis Voss, is a common herbivore attacking this plant in China. To evaluate its potential as a biological control agent of T. sebifera, biology and host speciÞcity of this weevil were studied in China. H. bicallosicollis occurs over a wide, native, geographic range and its immatures successfully develop at 15Ð35⬚C, indicating its physiological potential to establish and persist throughout the range of climatic conditions where the target plant grows in the United States. Adults make feeding holes on leaves. Before oviposition, the female makes a sealed leaf roll called a nidus and then lays one to two eggs inside. Eggs, larvae, and pupae develop within nidi, and larvae survive only when they develop inside the nidi. This requirement makes the weevil highly host speciÞc to T. sebifera. In laboratory no-choice tests of 54 species from eight families, adults fed on only 3 plant species, T. sebifera, Sapium chihsinianum S. K. Lee, and Phyllanthus urinaria L. and only oviposited on T. sebifera. These results were conÞrmed where, in multiple-choice tests, adults only oviposited on T. sebifera. Given that T. sebifera is the only species in the genus Triadica in the United States, the results of this study suggest that H. bicallosicollis is a potential biological control agent of T. sebifera and should be considered to be imported into quarantine in the United States for further tests on native North American species. KEY WORDS biological weed control, leaf-rolling weevil, host speciÞcity, invasive plant, China

Classical biological control has been an effective method for management of invasive plants for ⬎100 yr (Julien and GrifÞths 1998, McFadyen 1998). However, increasing concerns about the potential risk associated with introduced biological control agents challenges biological control practitioners and decision-makers to be more predictive when they propose or approve introductions of agents (Louda et al. 1997, Pemberton 2000, Strong and Pemberton 2000, Louda et al. 2003, Pearson and Callaway 2003, Delfosse 2005, Pearson and Callaway 2005, Sheppard et al. 2005). Before introduction of biological control agents, focusing on native-range studies will assist in predicting safety, abundance and efÞcacy of potential agents in their new environment (Goolsby et al. 2006, Sheppard et al. 2006). Conducting native-range research on biological and ecological characters is the Þrst step in screening potential biological control agents. Host speciÞcity 1 Invasion Biology and Biological Control Laboratory, Wuhan Botanical Garden/Institute, Chinese Academy of Sciences, Wuhan, Hubei 430074, P.R. China. 2 Department of Plant Protection, College of Plant Science and Technology, Huazhong Agriculture University, Wuhan, Hubei 430070, P.R. China. 3 Corresponding author, e-mail: [email protected]. 4 USDAÐARS Invasive Plant Research Laboratory, 3225 College Ave., Ft. Lauderdale, FL, 33314. 5 USDAÐARS Australian Biological Control Laboratory, CSIRO Entomology, 120 Meiers Rd., Indooroopilly, Queensland, 4068 Australia.

tests in a speciesÕ native-range are more convenient and cost-effective than quarantine research in the introduced range. Furthermore, open-Þeld tests that are usually used to show an insectÕs ecological (Þeld) host range can only be done in their native range (Briese et al. 1995, 2002). Chinese tallow (or Chinese tallow tree, tallow), Triadica sebifera L. Small (synonyms: Sapium sebiferum L. Roxb.) (Euphorbiaceae), is a deciduous tree that originates from central and southern China (Zheng et al. 2005). In China, Chinese tallow is cultivated for many purposes including wood for construction materials, furniture, medicine (bark), black dye (leaves), wax for candles (seeds), and soap (seeds). Cultivation of Chinese tallow is most common in central and southern China and also in some northern provinces (Zheng et al. 2005). T. sebifera is a serious invasive plant in many parts of the southern United States (Bruce et al. 1997). Since it was introduced into the United States for agricultural purpose in the late 18th century, this plant has aggressively displaced native plants and formed monospeciÞc stands (Jubinsky and Anderson 1996, Siemann and Rogers 2003). Currently it is listed as a noxious invasive weed in Florida, Louisiana, Mississippi, and Texas (USDA/NRCS 2007). Current management tactics, such as chemical and mechanical means, are not effective long-term solutions because Chinese tallow

1136

ENVIRONMENTAL ENTOMOLOGY

grows rapidly and produces large numbers of seeds (Siemann and Rogers 2003). Risk of nontarget effects is reduced when using biological control against a target that has few taxonomic relatives such as T. sebifera (Pemberton 2000, Ding et al. 2006). T. sebifera is the only species in the genus Triadica in the United States. Literature surveys for natural enemies associated with Chinese tallow were conducted in China by one of the authors and colleagues (Zheng et al. 2005). Because of the economic importance of this species and cultivation throughout southern China, there is considerable interest in the arthropods that may reduce yields of this agricultural commodity. These studies have compiled a list of 115 species of arthropods associated with Chinese tallow in China (Zheng et al. 2005). Of special interest are the expected monophagous or oligophagous mite (Phyllocoptruta sapii Kuang and Zhuo; Eriophyidae); a leaf-rolling weevil [Heterapoderopsis bicallosicollis (Voss, 1932) (synonym Apoderus bicallosicollis Voss); Attelabidae] (Legalov 2007), two leaf beetles [Bikasha collaris (Baly) and Morphosphaera japonica Hornstedt; Chrysomelidae], a cicada (Gaeana muculata consors Distant: Cicadidae), a planthopper (Fulgora watanabei Matsumura; Fulgoridae), and moths/caterpillars (Gadirtha inexacta Walker); Noctuidae; and Gatesclarkeana idia Diakonoff; Tortricidae), as only Chinese tallow was recorded as their host (Zheng et al. 2005). However, few detailed laboratory and Þeld studies have been reported on the biology and host range of these arthropods. In this paper, we describe the distribution, biology, life history, host range, and thermal tolerance of the leaf-rolling weevil, H. bicallosicollis. We also predict the safety and possible distribution of this weevil in the United States if used as a biological control agent.

Materials and Methods Experimental Organisms. Chinese tallow, T. sebifera, is a common plant that grows in cultivation and wild in southern China. A 20-yr-old tree can produce ⬎100,000 seeds per year; however, this species only reproduces from seed (Renne et al. 2002, Myers 2005). Seedlings appear in late April or May and grow rapidly (50 Ð70 cm in height per year) and mature after 3Ð 4 yrs. The height of mature trees range from 7 to 20 m (Zhang and Lin 1994, Myers 2005, Zheng et al. 2005). Flowering occurs from July to August and seeds mature in late September. All Chinese tallow plants used in our experiment were transplanted seedlings from Þeld populations or cultured from seeds collected from Luotian County, located ⬇260 km east of Wuhan, Hubei Province, China. Other test plants were collected from Wuhan Botanical Garden, Chinese Academy of Sciences, Wuhan, Hubei Province. Most smaller-size plant species in the tests were seedlings transplanted into pots and maintained in the laboratory at 22Ð30⬚C, 50 Ð70% RH, and under a 14-h photoperiod. Cut shoots from larger trees were also used for host range tests.

Vol. 38, no. 4

Adults in the Attelabidae family feed on leaves and lay eggs inside rolled leaves (nidi) constructed by the females. Larvae and pupae live within the nidus until the emergence of the adult. The nidi usually fall to the ground before emergence of adults (Legalov 2001, Kobayashi and Kato 2004). The Attelabid beetles are characterized by their elaborate behavior to construct nidi into which females oviposit (Iwata 1935, Kobayashi and Kato 2004). The host range of leaf-rolling weevils is usually narrow, with most only feeding on plants from one or two genera or, less frequently, on two closely related families (Legalov 2005). In our experiment, H. bicallosicollis adults and nidi (Fig. 1) were collected from T. sebifera in Luotian County, Hubei province from June to September. Adults were cultured in cages in the laboratory at Wuhan Botanical Garden using potted T. sebifera. Field-collected immatures were reared to obtain adults. All laboratory studies and common garden tests were performed at Wuhan Botanical Garden. Field Surveys. In China, extensive Þeld surveys of H. bicallosicollis were conducted from early summer 2006 to fall 2007 to determine its distribution and Þeld host range. The surveyed areas covered most of the plantÕs native-range including Hubei, Hunan, Henan, Guizhou, Guangdong, Guangxi, Fujian, and Jiangxi Provinces in central and southern China. Sites were selected where T. sebifera trees were common, with at least 20 seedlings or trees occurring in close proximity to each other. Surveys were made at six sites in Hubei Province at intervals of 4 Ð 6 wk from June to October in 2006 and 2007. Four to 10 Þeld sites in each of the other provinces were surveyed one to three times during one or two Þeld seasons. At each site, at least 10 plants were sampled during each visit. Weevils were collected from buds, leaves, and branches either by hand or sweep net. We observed the weevilÕs Þeld host range by examining plants growing near H. bicallosicollisÐinfested T. sebifera in Luotian County, Hubei Province. We recorded the presence of H. bicallosicollis adults, rolled-leaves with eggs, larvae, and pupae. Surveys were conducted at four different habitats (crop Þeld, pinelands, riverine, and upland), each with Þve sites. At each site, we searched all plants located within a 10-m radius of 10 randomly selected T. sebifera and included 32 plant species from 17 families. Life History, Oviposition, and Other Biological Characteristics. Adult feeding and oviposition behavior of H. bicallosicollis were monitored in both the laboratory and Þeld. In the laboratory, ⬇50 adults were released into a cage (80 by 80 by 70 cm) containing four to Þve potted T. sebifera plants and held at 22Ð30⬚C, 50 Ð70% RH, and a 14-h photoperiod. In the Þeld, a similar arrangement was maintained under ambient conditions. Newly rolled nidi were collected from Þeld populations daily and were carefully opened to record the number of eggs in each nidus. To determine the durations of eggs, larvae, and pupae, new nidi were collected from the Þeld and immatures reared in the laboratory. Nidi were carefully unrolled every day to identify the insectÕs developmental stage

August 2009

WANG ET AL.: Heterapoderopsis bicallosicollis ON CHINESE TALLOW

1137

Fig. 1. Life stages and characteristics of H. bicallosicollis: (A) rolled leaves (nidi) on Chinese tallow; (B) a single nidus (eggs, larva or pupae live inside); (C) egg from an opened nidus; (D) a third-instar larva; and (E) adult. (Online Þgure in color.)

and carefully rerolled. The duration of each stage was recorded. Observations on the development of Þeld-collected immatures (newly laid eggs, Þrst instars, and new pupae) were conducted in the laboratory under three different conditions: (1) maintained nidi on moist soil, (2) transferred from nidi onto ßat fresh leaves on moist soil, and (3) transferred from nidi onto moist soil. A total of 50 immatures of each stage were tested under each of three conditions. Ten of each life stage were placed in one box (diameter: 9 cm by height: 10 cm). Each treatment by life stage combination was replicated Þve times. Insects were randomly assigned to each replicate and treatment. Insects were checked and their survival and development were recorded daily. Thermal Physiology of H. bicallosicollis. To assess impacts of temperature on egg and larval development, nidi that contained newly laid eggs or Þrst instars were exposed to four different temperatures (20, 25, 30, and 35⬚C). Five nidi (two eggs or larvae in each) were placed on moist soil in a plastic box (diameter: 9 cm by height: 10 cm) with a lid. The box was placed in a growth chamber set at the designated treatment temperature. Each temperature treatment was replicated Þve times. Nidi were carefully opened and examined daily (as in the section Life History, Ovi-

position, and Other Biological Characteristics) for eclosion, pupation, and mortality. Similar tests were designed for pupae but at Þve temperatures (15, 20, 25, 30, and 35⬚C). We recorded the pupation period and mortality. We used the data from the above experiments to calculate the sum of effective temperatures (SET) and the lower development threshold (LDT) of immature stages and the generation time by the least square method (Jarosik et al. 2002, 2004; Trudgill et al. 2005). Using LDT and climatic data of Wuhan (China) and the United States (NOAA/NCDC 2004) where T. sebifera occurs, we calculated accumulated degree-days (K) at each location using the equation as below and predicted the generations per year of H. bicallosicollis in the United States: K⫽

冘

共T i ⫺ LDT兲

where, Ti represents the daily mean temperature (Ti) above LDT. Host Specificity Tests. As female adults of H. bicallosicollis lay eggs in nidi and all immatures (eggs, larvae, and pupae) only survive in nidi (see Results), host range of this weevil is entirely determined by adults. No-choice feeding and oviposition tests were conducted to determine which test plants were acceptable to adults. Choice feeding and oviposition

1138


tests were conducted to determine the preference of adults when the target and test plants grew together. An open-Þeld test determined its ecological host range under natural conditions. Selection of test plant species was based on the modiÞed centrifugal phylogenetic method (Wapshere 1974, Briese and Walker 2002) and plant availability in central China. Euphorbiaceae is a large family that includes Þve subfamilies: Euphorbioideae, Phyllanthoideae, Acalyphoideae, Crotonoideae, and OldÞeldioideae (Webster 1994). The OldÞeldioideae subfamily is overwhelmingly from the Southern Hemisphere and representatives are not included here. The Triadica genus is assigned to the Hippomaneae tribe of the Euphorbioideae subfamily. A total of 54 plant species in eight families were tested and included 39 key species from four subfamilies of Euphorbiaceae. The species and genera selected were either closely related to the target plant (species of Sapium, Excoecaria, Euphorbia, and Pedilanthus) or of economic importance (species of Ricinus, Vernicia, and Croton). No-Choice Adult Feeding and Oviposition Tests. Fifty-four species were included in no-choice feeding tests. For small- to medium-size plant species, potted seedlings were tested; cut shoots were used for larger tree species. Each potted plant was enclosed in a Þne nylon gauze bag. Cut shoots were placed in plastic cylinders (diameter: 9 cm by height: 10 cm). The base of shoots was wrapped with moist cotton and sealed with ParaÞlm. The number of shoots in each cylinder varied according to different species (to achieve similar resource availability for all test plants). Each potted plant or cylinder received two pairs of adults. Five replicates were conducted for each plant species. After 10 d, adults were removed, and we recorded feeding damage and the number of eggs. Four feeding categories were assigned to assess the adult feeding damage on plants: ⫺, no feeding damage; ⫹, area of feeding damage ⱕ20%; ⫹⫹, area of feeding damage between 20 and 50%; ⫹⫹⫹, area of feeding damage ⬎50%. The same feeding categories were used for all other tests. Multiple-Choice Adult Feeding and Oviposition Tests. Test plant species that were fed on by adults in no-choice tests (see Results; T. sebifera, Sapium chihsinianum, and Phyllanthus urinaria) were subjected to multiple-choice feeding and oviposition tests. Because potted plants of S. chihsinianum were not available and this plant does not occur in the Untied States, only T. sebifera and P. urinaria were used in choice tests. Two similar-sized potted plants of each species were placed in a nylon cage (40 by 30 by 40 cm) in the laboratory, and four pairs of adults were added. The test was replicated Þve times (Þve cages). After 2 wk, adults were removed, and we recorded the area of feeding and number of eggs. Open-Field Tests. To conÞrm the host range of H. bicallosicollis under natural conditions, a two-phase open-Þeld multiple-choice test (Briese et al. 2002) was conducted using T. sebifera and P. urinaria. During the Þrst phase of the experiment, six equal-sized potted

Vol. 38, no. 4

plants of each species were randomly placed in a 2 by 3-m plot in an open Þeld. Fifty pairs of adults were randomly released and dispersed into the plot on 20 August 2007. After 7 d, we recorded the number of adults and eggs and the amount of feeding on each plant. In the second “no-choice” phase, all plants of T. sebifera were removed from the plot, and all remaining adults were placed in the pots of P. urinaria, forcing H. bicallosicollis to use P. urinaria within the plot. We recorded the number of adults remaining on P. urinaria and feeding damage 7 d later. Statistical Analysis. One-way analyses of variance (ANOVAs) were used to analyze effects of temperature on egg, larval, and pupal development and to analyze effects of plant species on insect performance in adult no-choice feeding and oviposition tests. When the ANOVA was signiÞcant (P ⫽ 0.05), we compared treatment means using a protected least signiÞcant difference (LSD) test. Dependent t-test was used in adult multiple-choice feeding and oviposition tests. All data analyses were performed using SPSS 11.0 (SPSS 2002). Results Native Distribution and Field Status. Heterapoderopsis bicallosicollis is widely distributed in China and was abundant in all eight surveyed provinces (Fig. 2). At all Þeld sites, H. bicallosicollis was only found feeding on T. sebifera. No damage or nidi of H. bicallosicollis were found on any of the 32 plant species that grew near weevil-infested T. sebifera in Luotian County, Hubei Province (Table 1). Biology. The adults of H. bicallosicollis are brown or red-brown and they most commonly feed on newly formed leaves and create feeding scars (Fig. 1). Female adults are characterized by their elaborate behavior constructing nidi into which they oviposit. When ovipositing, the female Þrst incises the leaf base, and then chews the basal tissue, which softens it. After the leaf has wilted, the female folds it along the midvein and rolls it from the leaf apex toward the base. Once the leaf is rolled, the female creates a hole through which she lays eggs (mostly two eggs, rarely one or three eggs) and seals the leaf. Before completing the nidus, the female also carefully folds over the leaf edge, which prevents the nidus from unrolling. The nidus remains attached to the plant for ⬇1 wk and falls to the ground. Adults mate and oviposit many times, and each female can lay 40 Ð 60 eggs (in 20 Ð30 nidi) in a lifetime. Eggs of H. bicallosicollis are canary color and incubation time is 4 Ð5 d (Table 2). Eggs could only hatch inside nidi (Table 3). Usually all eggs in a nidus would hatch. The Þrst instars are white and become yellow in the third instar. There are three instars, and the larval duration is 5Ð7 d. Larvae could only complete development inside the nidus (Table 3). Pupae are initially yellow and become brown, and pupal duration is 3Ð 4 d (Table 2). Pupae could develop inside the nidus or on moist soil and fresh leaves (Table 3). In Wuhan, the duration from egg to adult was 12Ð16 d,

August 2009


1139

Fig. 2. The natural distribution (shadow) of T. sebifera in China and the surveyed areas (2006Ð2007) where H. bicallosicollis occurs (Œ).

there are six to seven generations per year, and these may overlap. Adults overwinter in the litter around T. sebifera and begin to mate and oviposit in May. Thermal Physiology. Eggs successfully developed at 20 Ð35⬚C, and hatching rates were not signiÞcantly affected by temperature (F3,16 ⫽ 1.367, P ⫽ 0.289; Table 4). However, the incubation was signiÞcantly affected by temperature (F3,120 ⫽ 1135.107, P ⬍ 0.001). Table 1.

Field plants surveyed adjacent to H. bicallosicollis–infested T. sebifera plants in Luotian County, Hubei Province, China

Family Alangiaceae Araceae Compositae

Convolvulaceae Ebenaceae Ginkgoaceae Gramineae

Leguminosae

The eggs required fewer days to complete development as temperature increased: ⬍0.4-fold fewer days at 35⬚C compared with 20⬚C (Table 4). Temperature had a signiÞcant impact on larval survival (F3,16 ⫽ 17.677, P ⬍ 0.001) and development time from Þrst instar to pupa (F3,135 ⫽ 1,554.532, P ⬍ 0.001). The rates of larval survival were not signiÞcantly different at 20, 25, and 30⬚C (all ⱖ76%), although only

Species Alangium chinense (Lour.) Harm Pistia stratiotes L. Artemisia carvifolia Buch.-Ham. ex Roxb. Artemisia lavandulaefolia DC. Prodr Cirsium albescens K. Xanthium sibiricum Patrin Ipomoea batatas L. Lam. Diospyros kaki L.f. Ginkgo biloba L. Oryza sativa L. Pennisetum alopecuroides L.Spreng Phyllostachys heterocycla cv. Pubescens (Carr.) Matsum. Phragmites hirsuta Kitag. Glycine max L. Merr. Glycine soja Sieb. et Zucc. Lablab purpureus L. Sweet

None of the plants were damaged by H. bicallosicollis.

Family Leguminosae Moraceae Oleaceae Phytolaccaceae Polygonaceae Rosaceae Salicaceae Solanaceae Taxodiaceae Verbenaceae

Species Lespedeza bicolor Turcz. Robinia pseudoacacia L. Vigna unguiculata L. Walp. Morus alba L. Osmanthus fragrans (Thunb.) Phytolacca americana L. Polygonum hydropiper L. Polygonum perfoliatum L. Polygonum persicaria L. Amygdalus davidiana (Carr.) C. de Vos Rubus coreanus Miq. Populus tomentosa Carr Capsicum annuum L. Metasequoia glyptostroboides Hu et Cheng Vitex negundo L. Vitex negundo L. var. heterophylla (Franch.) Rehd.

1140


Table 2. Mean ⴞ SE duration and range of life stages of H. bicallosicollis in the laboratory at 22–30°C Life stage

Mean (d)

Range (d)

N

Egg First instar Second instar Third instar Pupa

4.1 ⫾ 0.1 1.6 ⫾ 0.1 2.2 ⫾ 0.1 2.3 ⫾ 0.1 3.8 ⫾ 0.1

4Ð5 1Ð2 2Ð3 2Ð3 3Ð4

50 35 39 33 63

44% of the larvae completed development at 35⬚C. However, the larvae required signiÞcantly fewer days to complete development as temperatures increased: ⬍0.4-fold fewer days at 35⬚C compared with 20⬚C (Table 4). Temperature also had a signiÞcant impact on pupal survival (F4,20 ⫽ 7.045, P ⫽ 0.003) and pupal duration (F4,156 ⫽ 113.706, P ⬍ 0.001). Pupae successfully developed between 15 and 35⬚C, although survival was greatest at 20⬚C (Table 4). The pupae required fewer days to complete development with increased temperature: ⬍0.3-fold fewer days at 35⬚C compared with 15⬚C (Table 4). The SET of egg, larva, and pupa is given in Table 5. The SET from egg to adult was 260.63 DD and excludes the adult preoviposition period (⬇5Ð 6 d at 25Ð30⬚C). The LDT of an entire generation was 11.9⬚C. With this LDT and combined with Wuhan climatic data (Anonymous 2007), the accumulated degreedays (K) in Wuhan is ⬇2,370 DD. No-Choice Adult Feeding and Oviposition Tests. Feeding by adults of H. bicallosicollis only occurred on three plant species: the target (T. sebifera) and two related species (S. chihsinianum and P. urinaria; Table 6). The feeding category was ⫹⫹⫹ on T. sebifera but ⫹⫹ on both S. chihsinianum and P. urinaria. The female adults only rolled leaves and laid eggs on T. sebifera (Table 6). Multiple-Choice Adult Feeding and Oviposition Tests. Adults preferred strongly to feed on T. sebifera compared with P. urinaria (Table 7). Feeding was ⫹⫹⫹ on T. sebifera and only ⫹ on P. urinaria. Adults only laid eggs on T. sebifera (F1,8 ⫽ 17.515, P ⬍ 0.001; Table 7). Open Field Tests. In phase 1, 16 of the 100 adults released in the open Þeld test were recovered: 14 of these were on T. sebifera and 2 were on P. urinaria. Adults left feeding scars on ⬎40% of the leaves of T. sebifera but no feeding was found on P. urinaria. A total of 17 eggs were found on T. sebifera, and no eggs were laid on P. urinaria (Table 7). Table 3. treatments Treatments On moist soil On fresh leaf In rolled leaf a

Vol. 38, no. 4

In phase 2, after the T. sebifera plants were removed, H. bicallosicollis adults disappeared rapidly from the plot, and no weevils were found on P. urinaria after 7 d. Discussion The results of this study conÞrm an earlier literature-based prediction that the leaf-rolling weevil, H. bicallosicollis, is host speciÞc (Zheng et al. 2005). In no-choice tests with 54 plant species in eight families, the adults only fed on T. sebifera, S. chihsinianum, and P. urinaria, and eggs were only laid on T. sebifera. Although the adults fed on P. urinaria in laboratory choice tests, there was no oviposition on this plant. Because the adults only laid eggs in nidi, and eggs only developed in nidi, this behavior became critical in host range evaluations. All laboratory and Þeld research indicated that nidi were only found on T. sebifera; and therefore, it is the only known host of H. bicallosicollis. Host range tests indicated that H. bicallosicollis is a potential biological control agent of T. sebifera in the United States. T. sebifera is the only species in the genus Triadica in North America. It was recently reassigned from the genus Sapium to Triadica (Esser 1999). Sapium is very closely related to Triadica, with four species reported to occur in the United States [S. glandulosum L. Morong, S. laurifolium (A.Rich.) Griseb., S. laurocerasus Desf., and possibly S. japonicum (Siebold and Zucc.) Pax and K.Hoffm] (USDA/ NRCS 2007). From our results, another member of this genus, S. chihsinianum, was fed on by adults but no oviposition occurred. S. chihsinianum is native to China and not distributed in the United States (Ma and Cheng 1997, USDA/NRCS 2007). Among the four North American Sapium species, S. glandulosum is a Caribbean, Central, and South American species only reported from one county (Escambia) in Florida (USDA/NRCS 2007). The other species, S. laurifolium and S. laurocerasus only occur in Puerto Rico, and the presence of S. japonicum in the United States is not conÞrmed (USDA/NRCS 2007). Genera other than Triadica in the subtribe Hippomaninae that occur near its adventive range include Bonania, Gymanthes, Grimmeodendron, Hippomane, Sebastiania, and Stillingia (USDA/NRCS 2007). Members of these genera are considered the most at-risk species and thus a priority of future risk assessment tests. For a long time, prerelease studies in the native range have been important in implementing classical biological control programs. With the increasing concerns for nontarget effects of introduced natural en-

Mean ⴞ SE survival rates (%) and development times (d) of H. bicallosicollis eggs, larvae, and pupae exposed to different Egga

Larvae

Pupae

Survival rate (%)

Duration (d)

Survival rate (%)

Duration (d)

Survival rate (%)

Duration (d)

0A 0A 74.0 ⫾ 4.0 B

Ñ Ñ 4.2 ⫾ 0.4

0A 0A 76.0 ⫾ 5.1 B

Ñ Ñ 6.2 ⫾ 0.2

70.0 ⫾ 3.2 A 68.0 ⫾ 3.7 A 72.0 ⫾ 3.7 A

3.9 ⫾ 0.1 A 4.0 ⫾ 0.1 A 4.0 ⫾ 0.1 A

Means within a column followed by the same letter are not signiÞcantly different according to a protected LSD (P ⫽ 0.01).


August 2009 Table 4.

Mean ⴞ SE survival and development time of H. bicallosicollis grown at different temperatures

Temperature

Egga

b

Larval Duration (d)

Survival rate(%)

Duration (d)

Survival rate(%)

Duration (d)

Ñ 58 ⫾ 5.8 A 72 ⫾ 3.7 A 66 ⫾ 5.1 A 56 ⫾ 9.3 A

Ñ 9.9 ⫾ 0.1 A 5.8 ⫾ 0.1 B 4.7 ⫾ 0.1 C 3.6 ⫾ 0.1 D

Ñ 76 ⫾ 2.5 A 78 ⫾ 3.7 A 80 ⫾ 4.5 A 44 ⫾ 5.1 B

Ñ 12.6 ⫾ 0.1 A 8.4 ⫾ 0.1 B 5.9 ⫾ 0.0 C 5.0 ⫾ 0.0 D

50 ⫾ 3.2 B 90 ⫾ 4.5 A 72 ⫾ 5.8 AB 58 ⫾ 8.6 B 52 ⫾ 5.8 B

9.3 ⫾ 0.2 A 7.8 ⫾ 0.2 B 5.7 ⫾ 0.2 C 3.3 ⫾ 0.2 D 2.5 ⫾ 0.2 D

Sample size is 50. Means within a column followed by the same letter are not signiÞcantly different according to a protected LSD (P ⫽ 0.01).

emies, native range studies are becoming critical in explicitly predicting the safety and efÞcacy of potential biological control agents (Goolsby et al. 2006, Sheppard et al. 2006). For invasive plants like T. sebifera having no congeneric species in the introduced range, host range tests in the native range will assist in providing a rapid and effective assessment of a natural enemyÕs potential for biological control. Only supplementary quarantine testing of additional plant species not available in the native range would be required preserving limited resources and improving success in a biological control program (Pemberton 2000). Biological control practitioners are constantly improving procedures for host testing of biological control agents (Briese 2004, 2005; Sheppard et al. 2005). Appropriate interpretation and analysis of no-choice, choice, and open-Þeld tests conducted in the native range is essential when predicting safety of agents (Briese et al. 1995, 2002; Goolsby et al. 2006). Results of no-choice cage tests help eliminate obvious nonhosts and allow the subsequent test plant list to be shorter. Choice cage tests clarifying the preference of agents, whereas open-Þeld tests and Þeld survey assist in understanding the ecological host range (Kluge 2000, Briese et al. 2002, Briese 2005, Sheppard et al. 2005). In this study, no-choice adult feeding and oviposition tests identiÞed the possible hosts of H. bicallosicollis, only 3 species (T. sebifera, S. chihsinianum, and P. urinaria) of the 54 plant species tested. Because potted plants of S. chihsinianum were unavailable (and this species is not found in the United States), only limited choice tests on the remaining two hosts was required and determined that T. sebifera was the preferred host for feeding and the only host for oviposition. Open-Þeld tests and Þeld surveys veriÞed that T. sebifera was the only ecological host plant, because P. urinaria was not accepted in the Þeld. Therefore, the Table 5. Sum ⴞ SE of effective temperatures and the lower development threshold of immature stages of H. bicallosicollis

Stage

Sum of effective temperatures (degree-days)

Lower development threshold (⬚C)

Egga Larva Pupa Entire generation

87.62 ⫾ 13.50 121.78 ⫾ 7.17 51.24 ⫾ 3.17 260.63 ⫾ 23.84

10.84 ⫾ 0.61 10.22 ⫾ 0.47 14.60 ⫾ 0.53 11.90 ⫾ 1.61

a

Pupal

Survival rate (%)b

15⬚C 20⬚C 25⬚C 30⬚C 35⬚C a

1141

Sample size is 50.

entire host range test procedure was efÞcient and the outcome was unambiguous. EfÞcacy is very important when assessing the potential of an agent for biological control of invasive plants (McClay and Balciunas 2005, Haßiger et al. 2006). Introduction of ineffective agents is an unnecessary risk to the ecosystem through potential indirect effects on its native components in the system (Pearson and Callaway 2005). EfÞcacy is affected by plant responses to herbivory, resulting in variable control of weeds. In addition, invasive plant resistance and tolerance may evolve in the area of introduction (Mu¨ llerScharer et al. 2004). Rogers and Siemann (2004) reported that invasive ecotypes of T. sebifera tolerated simulated root herbivory more effectively than its native ecotypes. Further studies showed that initial populations of T. sebifera were less vigorous and more susceptible to North American herbivores than populations introduced later (Siemann et al. 2006). Therefore, to predict control efÞcacy of H. bicallosicollis, plant performance and response to herbivory should be studied using different T. sebifera populations in the United States. Plans are presently underway to directly measure the impact of herbivory by this insect on T. sebifera. Leaf-rolling weevils have great potential as biological control agents of weeds because of their narrow host range and signiÞcant damage caused by feeding and unique oviposition behavior (Iwata 1935; Legalov 2001, 2005; Kobayashi and Kato 2004). Legalov (2005) summarized the trophic links of leaf-rolling weevils (Rhynchitidae and Attelabidae) and found that they are usually oligophagous species. In Attelabidae, 75% of these insects are associated with a single plant genus, and only 18% were found on two or three genera (Legalov 2005). Their unique oviposition behavior and speciÞc conditions required for egg and larva (Table 3) may explain their narrow host range. Leaf-rolling weevils suppress plant growth through adult feeding and the creation of a nidus (rolled leaf) for oviposition and development of immatures. We observed that one female of H. bicallosicollis could construct 20 Ð30 nidi (killing an equivalent number of leaves) for oviposition causing signiÞcant damage to plants, especially seedlings. However, the impact of this type of damage toward the plantÕs reproductive Þtness needs to be assessed. The thermal physiology results of this study suggest that H. bicallosicollis will adapt to most of the climate

1142 Table 6.


Vol. 38, no. 4

Results of H. bicallosicollis adult feeding and oviposition no-choice tests

Subfamily

Test speciesa,b

Tribe

Euphorbioideae

Hippomaneae Hippomaneae Hippomaneae Hippomaneae Hippomaneae

Euphorbioideae

Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae Euphorbieae BischoÞeae Epiprineae Phyllantheae Phyllantheae Phyllantheae Acalypheae Acalypheae Acalypheae Acalypheae Alchorneae Epiprineae Epiprineae Aleuritideae Aleuritideae Codieae Codieae Crotoneae Jatropheae

Phyllanthoideae

Acalyphoideae

Crotonoideae

Buxaceae

Celastraceae Daphniphyllaceae Leguminosae Meliaceae Pittosporaceae Rosaceae

Species most closely related to Triadicad Triadica sebifera L. Small Sapium chihsinianum S. K.Lee Sapium sp. Excoecaria cochinchinensis Lour. Excoecaria formosana (Hayata) Hayata Species from other taxa within the Euphorbiaceae Euphorbia aeruginosa Schweick. Euphorbia ammak f.variegata Schweinf. Euphorbia bergeri N.E. Br. Euphorbia decaryi A.Guill Euphorbia horrida Boiss. Euphorbia humifusa willd.ex Schlecht Euphorbia hylonoma Hand.-Mazz. Euphorbia lactea f.cristata Haw. Euphorbia milii ch.des Moulins Euphorbia millii Desm. Euphorbia piscidermis f.cristata Euphorbia pulcherrima Willd.ex Klotzsch Euphorbia tirucalli L. Euphorbia tortirama R.A.Dyer Euphorbia trigona Haw. Euphorbia turbiniformis Chiov. Pedilanthus tithymaloides L. Poit Pedilanthus tithymaloides L. Poit spp. Smallii cv. Nana Compacta Bischofia javanica Blume. Cleistanthus sumatranus (Miq.) Mull. Arg Phyllanthus cheklangensis Croiz.et Mete Phyllanthus urinaria L. Glochidion wilsonii Hutch Acalypha australis L. Claoxylon indicum (Reinw. Ex Bl.) Hassk Mallotus barbatus (Wall.) Mull. Arg Ricinus communis L. Alchorneae trewioides (Benth.) Mull. Arg Cephalomappa sinensis (Chun et How) Kosterm Cleidiocarpon cavaleriei (Levl.) Airy-Shaw Vernicia fordii (Hemsl.) Airy-Shaw Vernicia montana Lour Codiaeum variegatum L. A.Juss. Codiaeum variegatum L. A.Juss.var. Croton lachnocarpus Benth Jatropha podagrica Hook

Species from other families related to Euphorbiaceae Buxus sinica (Rehd.et.Wils.) Cheng Buxus bodinieri Levl. Buxus ichangensis Hatusima Pachysandra axillaris Franch. Sarcococca ruscifolia Stapf Euonymus fortunei (Turcz.) Hand䡠-Mazz. Euonymus bungeanus Maxim. Euonymus Japonicus cv.Aureo-ma Daphniphyllum calycinum Benth. Robinia pseudoacacia L. Aglaia odorata Lour. Pittosporum tobira (Thunb.) Ait. Amygdalus persica L. var. densa Makino

Feeding categoryc ⫹⫹⫹ ⫹⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺

a

Sample size is Þve. The no. (mean ⫾ SE) of eggs laid on T. sebifera was 4.6 ⫾ 0.6, the only species on which eggs were laid. Adult feeding categories: ⫺, no feeding damage; ⫹, area of feeding damage ⱕ20%; ⫹⫹, area of feeding damage between 20 and 50%; ⫹⫹⫹, area of feeding damage ⬎50%. The same feeding categories were used for all other tests. d Systematics follows Webster 1994. b c

where T. sebifera occurs in the southern United States. SET and LDT results assist in predicting where this weevil will thrive in the introduced range. Using the climatic data of Þve key invaded states of the United

States (NOAA/NCDC 2004), it is estimated that the accumulated degree-days (K) in Houston (Texas), Jacksonville (Florida), New Orleans (Louisiana), Columbia (South Carolina), and Atlanta (Georgia) are


August 2009

Table 7. Adult H. bicallosicollis feeding and oviposition on the target weed T. sebifera and test plant P. urinaria in laboratory choice and open-field choice tests Choice test Plant species

Feeding category

No. of eggs

Triadica sebifera Phyllanthus urinaria

⫹⫹⫹ ⫹

6.80 ⫾ 1.62 0

Open-Þeld test Percent No. of No. of damaged eggs adults leaves ⬎40% 0

17 0

14 2

3,111, 3,096, 3,309, 2,385, and 2,169 DD, respectively. In Wuhan, China, this weevil has six to seven generations per year (2,370 DD). Thus, it may be able to complete Þve to eight generations per year and produce high populations on this plant if released in these Þve states. Additionally, given the signiÞcant damage by this weevil to T. sebifera in its native range, promising laboratory and Þeld host range test results in China, and because there is no congener of the target plant in North America, this study indicates that A. bicallosicollis will be a high-priority candidate for importation into U.S. quarantine, where its speciÞcity will be further tested against North American plant species. Acknowledgments We thank Y. Sun, W. Huang, K. Wu, J. Zhang, Z. Lu, X. Jin, and Y. Wang for laboratory and Þeld assistance and C. OÕBrien, Greenvalley, AZ, and R. Zhang for the determinations of H. bicallosicollis. Voucher specimens of leaf-rolling weevil are preserved in Wuhan botanical garden, Chinese Academy of Sciences, and at the Museum of Entomology, Florida State Collection of Arthropods, Florida Department of Agriculture and Consumer Services, Gainesville, FL. We appreciate the advice of M. Ren on data analysis. We also thank J. Wu, S. Zhang, and S. Feng for assistance in helping to Þnd plant material. E. Delfosse, R. Wiedenmann, and two anonymous reviewersÕ comments improved the manuscript. This research was funded by Florida Department of Environmental Protection and USDAÐARS.

References Cited Anonymous. 2007. China meteorological data sharing service system. (http://www.cdc.cma.gov.cn/index.jsp_. Briese, D. T. 2004. Weed biological control: applying science to solve seemingly intractable problems. Aust. J. Entomol. 43: 304 Ð317. Briese, D. T. 2005. Translating host-speciÞcity test results into the real world: the need to harmonize the yin and yang of current testing procedures. Biol. Control 35: 208 Ð 214. Briese, D. T., and A. Walker. 2002. A new perspective on the selection of test plants for evaluating the host-speciÞcity of weed biological control agents: the case of Deuterocampta quadrijuga, a potential insect control agent of Heliotropium amplexicaule. Biol. Control 25: 273Ð287. Briese, D. T., A. W. Sheppard, and J. M. Reifenberg. 1995. Open-Þeld host-speciÞcity testing for potential biological control agents of Onopordum thistles. Biol. Control 5: 158 Ð166. Briese, D. T., M. Zapater, A. Andorno, and G. Perez-Camargo. 2002. A two-phase open-Þeld test to evaluate the host-

1143

speciÞcity of candidate biological control agents for Heliotropium amplexicaule. Biol. Control 25: 259 Ð272. Bruce, K. A., G. N. Cameron, P. A. Harcombe, and G. Jubinsky. 1997. Introduction, impact on native habitats, and management of a woody invader, the Chinese tallow tree, Sapium sebiferum (L.) Roxb. Nat. Area. J. 17: 255Ð260. Delfosse, E. S. 2005. Risk and ethics in biological control. Biol. Control 353: 319 Ð329. Ding, J., R. Reardon, Y. Wu, H. Zheng, and W. Fu. 2006. Biological control of invasive plants through collaboration between China and the United States of America: a perspective. Biol. Invasions 8: 1439 Ð1450. Esser, H.-J. 1999. A partial revision of the Hippomaneae (Euphorbiaceae) in Malesia. Blumea 44: 149 Ð215. Goolsby, J. A., R. D. van Klinken, and W. A. Palmer. 2006. Maximising the contribution of native-range studies towards the identiÞcation and prioritisation of weed biocontrol agents. Aust. J. Entomol. 45: 276 Ð286. Hafliger, P., M. Schwarzlander, and B. Blossey. 2006. Impact of Archanara geminipuncta (Lepidoptera: Noctuidae) on aboveground biomass production of Phragmites australis. Biol. Control 38: 413Ð 421. Iwata, K. 1935. On the habits of some Rhynchitinae, Attelabinae and Apoderinae in Japan (in Japanese). Kontyu 9: 261Ð278. Jarosik, V., A. Honek, and A.F.G. Dixon. 2002. Developmental rate isomorphy in insects and mites. Am. Nat. 160: 497Ð510. Jarosik, V., L. Kratochvil, A. Honek, and A.F.G. Dixon. 2004. A general rule for the dependence of developmental rate on temperature in ectothermic animals. Proc. R. Soc. B. 271: S219 ÐS221. Jubinsky, G., and L. C. Anderson. 1996. The invasive potential of Chinese tallow tree (Sapium sebiferum Roxb.) in the southeast. Castanea 61: 226 Ð231. Julien, M., and M. Griffiths. 1998. Biological control of weeds: a word catalogue of agents and their target weeds, 4th ed. CABI Publishing, Wallingford, United Kingdom. Kluge, R. L. 2000. The future of biological control of weeds with insects: no more ÔparanoiaÕ, no more ÔhoneymoonÕ, pp. 459 Ð 467. In N. R. Spencer (ed.), Proceedings of the X international symposium on biological control of weeds, 4 Ð14 July 1999, Montana State University, Bozeman, MT. Kobayashi, C., and M. Kato. 2004. To be suspended or to be cut off? Differences in the performance of two types of leaf-rolls constructed by the attelabid beetle Cycnotrachelus roelofsi. Popul. Ecol. 46: 193Ð202. Legalov, A. A. 2001. Revision der holarktischen Auletini (Coleoptera, Attelabidae). Russ. Entomol. J. 10: 33Ð 66. Legalov, A. A. 2005. Trophic links of leaf-rolling weevils (Coleoptera, Rhynchitidae and Attelabidaes). Entomol. Rev. 85: 361Ð370. Legalov, A. A. 2007. Leaf-rolling weevils (Coleoptera: Rhynchitidae, Attelabidae) of the world fauna. AgroSiberia, Novosibirsk, Russia. Louda, S. M., D. Kendall, J. Connor, and D. Simberloff. 1997. Ecological effects of an insect introduced for the biological control of weeds. Science. 277: 1088 Ð1090. Louda, S. M., R. W. Pemberton, M. T. Johnson, and P. A. Follett. 2003. Nontarget effectsÑthe AchillesÕ Heel of biological control? Retrospective analyses to reduce risk associated with biocontrol introductions. Annu. Rev. Entomol. 48: 365Ð396. Ma, J., and Y. Cheng. 1997. Chinese plant ßora. Science Press, Beijing, China. McClay, A. S., and J. K. Balciunas. 2005. The role of prerelease efÞcacy assessment in selecting classical biologi-

1144


cal control agents for weedsÑapplying the Anna Karenina principle. Biol. Control 35: 197Ð207. McFadyen, R. 1998. Biological control of weeds. Annu. Rev. Entomol. 43: 369 Ð393. Mu¨ ller-Scharer, H., U. Schaffner, and T. Steinger. 2004. Evolution in invasive plants: implications for biological control. Trends Ecol. Evol. 19: 417Ð 422. Myers, R. 2005. Triadica sebifera. Fire effects information system. (http://www.fs.fed.us/database/feis/). [NOAA/NCDC] National Oceanic and Atmospheric Administration/National Cliamatic Data Center. 2004. Climatography of the United States No. 20 1971Ð2000. (http://www. ncdc.noaa.gov/oa/climate/normals/usnormals.html). Pearson, D. E., and R. M. Callaway. 2003. Indirect effects of host-speciÞc biological control agents. Trends Ecol. Evol. 18: 456 Ð 461. Pearson, D. E., and R. M. Callaway. 2005. Indirect nontarget effects of host-speciÞc biological control agents: implications for biological control. Biol. Control 35: 288 Ð298. Pemberton, R. W. 2000. Predictable risk to native plants in weed biocontrol. Oecologia (Berl.) 125: 489 Ð 494. Renne, I. J., W. C. Barrow, L. A. Johnson Randall, and W. C. Bridges. 2002. Generalized avian dispersal syndrome contributes to Chinese tallow tree (Sapium sebiferum, Euphorbiaceae) invasiveness. Divers. Distrib. 8: 285Ð295. Rogers, W. E., and E. Siemann. 2004. Invasive ecotypes tolerate herbivory more effectively than native ecotypes of the Chinese tallow Sapium sebiferum. J. Appl. Ecol. 41: 561Ð570. Sheppard, A. W., R. D. van Klinken, and T. A. Heard. 2005. ScientiÞc advances in the analysis of direct risks of weed biological control agents to nontarget plants. Biol. Control 35: 215Ð226. Sheppard, A., M. Haines, and T. Thomann. 2006. Nativerange research assists risk analysis for non-targets in weed

Vol. 38, no. 4

biological control: the cautionary tale of the broom seed beetle. Aust. J. Entomol. 45: 292Ð297. Siemann, E., and W. E. Rogers. 2003. Herbivory, disease, recruitment limitation and success of alien and native tree species. Ecology 84: 1489 Ð1505. Siemann, E., W. E. Rogers, and S. J. Dewalt. 2006. Rapid adaptation of insect herbivores to an invasive plant. Proc. R. Soc. B. 273: 2763Ð2769. SPSS. 2002. SPSS 11.0 users guide. Lingnan University, Hong Kong, China. Strong, D. R., and R. W. Pemberton. 2000. EcologyÑ biological control of invading speciesÑrisk and reform. Science 288: 1969 Ð1970. Trudgill, D. L., A. Honek, D. Li, and N.M.V. Straalen. 2005. Thermal timeÑ concepts and utility. Ann. Appl. Biol. 146: 1Ð14. [USDA/NRCS] U.S. Department of Agriculture/Natural Resources Conservation Service. 2007. Plants proÞle: Triadica sebifera (L.) small Chinese tallow tree. (http://plants. usda.gov/java/nameSearch?keywordquery⫽sapium⫹ sebiferum&mode⫽sciname). Wapshere, A. J. 1974. A strategy for evaluating the safety of organisms for biological weed control. Ann. Appl. Biol. 77: 201Ð211. Webster, G. L. 1994. ClassiÞcation of the Euphorbiaceae. Ann. Missouri Bot. Gard. 81: 3Ð32. Zhang, K., and Y. Lin. 1994. Chinese tallow. China Forest Press, Beijing, China. Zheng, H., Y. Wu, J. Ding, D. Binion, W. Fu, and R. Reardon. 2005. Invasive plants established in the United States that are found in Asia and their associated natural enemies. Forest Health Technology Enterprise Team, Morgantown, WV. Received 7 March 2008; accepted 17 April 2009.