This article was downloaded by: [USDA National Agricultural Library] On: 24 May 2010 Access details: Access Details: [subscription number 917340536] Publisher Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37- 41 Mortimer Street, London W1T 3JH, UK

Biocontrol Science and Technology Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713409232

Biology and Nymph Host Range of Anchocoema bidentata and Astroma saltense (Orthoptera: Proscopiidae), Potential Biocontrol Agents for Creosotebush, Larrea tridentata (Zygophyllaceae) in the U.S.A. Guillermo Logarzo ;Miguel Casalinuovo

To cite this Article Logarzo, Guillermo andCasalinuovo, Miguel(2004) 'Biology and Nymph Host Range of Anchocoema bidentata and Astroma saltense (Orthoptera: Proscopiidae), Potential Biocontrol Agents for Creosotebush, Larrea tridentata (Zygophyllaceae) in the U.S.A.', Biocontrol Science and Technology, 14: 1, 39 — 50 To link to this Article: DOI: 10.1080/09583150310001506589 URL: http://dx.doi.org/10.1080/09583150310001506589

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Biocontrol Science and Technology (February 2004), Vol. 14, No. 1, 39Á/50

Biology and Nymph Host Range of Anchocoema bidentata and Astroma saltense (Orthoptera: Proscopiidae), Potential Biocontrol Agents for Creosotebush, Larrea tridentata (Zygophyllaceae) in the U.S.A.

GUILLERMO LOGARZO AND MIGUEL CASALINUOVO Agricultural Research Service, US Department of Agriculture, South American Biological Control Laboratory, 3130 Buenos Aires Place, Washington, DC 20521- 3130, USA

(Received 2 August 2002; returned 9 October 2002; accepted 10 April 2003)

Two stick-like acridids (Orthoptera: Proscopiidae) from Argentina, Anchocoema bidentata Mello-Leitao and Astroma saltense Mello-Leitao, were evaluated as potential biological control agents of creosote bush (Larrea tridentata (DC.) Coville) in the southwestern United States. Their biology, behavior and geographic distribution of those were studied. The host ranges for both insects were established through nymph feeding preference and development tests in the laboratory and in the field. A total of 33 species of belonging to 13 families were tested. Anchocoema bidentata and A. saltense are mimetic species, having as many as three generations a year, and exhibit strong sexual dimorphism; females are larger and less mobile than males. In both species, females laid egg masses in the soil. First instars appeared in the field at the end of the spring, the second generation at mid-summer, and a third at the end of the summer. Adults of A. bidentata and A. saltense appeared in the field at the Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 beginning of the spring. The laboratory multiple-choice feeding test showed that A. bidentata preferred Larrea divaricata Cav., whereas A. saltense preferred L. divaricata and L. cuneifolia Cav. In the nymph development test (no choice), A. bidentata was able to complete its development only on L. divaricata and L. cuneifolia, while Astroma saltense completed its development on six plant species: L divaricata, L. cuneifolia, Bulnesia retama (Gillies ex Hooker et Arnott), B. schickendantzi Hieron (all Zygophyllaceae), Zuccagnia punctata Cav., and Prosopis torquata (Cav. Ap. Lag.) (both Fabaseae). We concluded that A. bidentata could be a biocontrol agent for L. tridentata because the first instar can complete its development only on Larrea spp. Regarding A. saltense, this species showed a wide host range and should not be considered as a biological control agent of L. tridentata.

Correspondence to: Guillermo Logarzo, USDA-ARS SABCL, 3130 Buenos Aires Place, Washington, DC 20521-3130, USA. Tel.: /54-11-4452-1882, ext. 111; Fax: /54-11-4452-1882, ext. 104; E-mail: [email protected]

ISSN 0958-3157 (print)/ISSN 1360-0478 (online)/04/010039-12 # 2004 Taylor & Francis Ltd DOI: 10.1080/09583150310001506589 40 G. LOGARZO & M. CASALINUOVO

Keywords: Larrea tridentata, biological control, weed, acridid, South America, creosote bush, Anchocoema bidentata, Astroma saltense, Zygophyllaceae, Orthoptera, Proscopiidae

INTRODUCTION Creosote bush (Larrea tridentata (DC.) Coville) is a common woody, perennial and slow- growing weedy shrub, covering 19 million ha in the United States (Platt, 1959), that competes with rangeland forage plant species. This desert plant has been increasing its geographical range in the last 150 years, replacing grasslands, probably due to overgrazing (York & Dick- Peddie, 1969). In New Mexico, Melgoza et al. (1984) found that good stands of grasses rapidly appeared after Larrea plants were killed with herbicides. The Larrea originated in South America (Hunziker et al., 1972, 1977; Hunziker, 1975; Wells & Hunziker, 1976), and has four South American, and one North American species: L. tridentata (Hunziker et al., 1977). Creosote bush arrived to North America around 11 000 years ago (Van Devender, 1973; Wells, 1976) and is supposed to be a subspecies of the South American L. divaricata (Hunziker et al., 1977). Biological control of creosote bush could be feasible by introducing phytophagous insects from the South American species of Larrea to the United States. Previous studies on the natural enemies of South American species of Larrea determined that the best candidates for biological control of L. tridentata were the proscopiids, Anchocoema bidentata Mello-Leitao (Astroma riojanum originally) and Astroma saltense Mello-Leitao (Cordo & De Loach, 1993) Proscopiidae, endemic to South America, are very similar to walking sticks (Phasmida) in appearance and in their cryptic nature, and all species are apterous or brachypterous. The biology of this family, which contains more than 100 species, is poorly known. Although several studies mention proscopiids of Larrea (Mares et al., 1977; Orions et al., 1977; Rhoades 1977; Schultz et al., 1977; Schultz, 1979), none analyzed their host range in detail. The purpose of this research was to document the biology and assess the natural and potential host plant range of A. bidentata and A. saltense through field and laboratory tests to evaluate their potential as biocontrol agents for creosote bush in the United States.

MATERIALS AND METHODS To identify the insects used in this study, we compared specimens of A. bidentata and A. saltense with the original descriptions made by Mello-Leita˜o and with the type specimens of A. riojanum and of A. saltense deposited in the Museo de Ciencias Naturales de La Plata. Cordo and De Loach provided us with the specimens used in the publications of 1993. The life history and host range studies were carried out in Buenos Aires and Catamarca

Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 Provinces, Argentina from 1991 to 1993. Most of the laboratory studies were carried out at the USDA-ARS SABCL lab in Buenos Aires. All host range studies were carried out at a temporary substation in Catamarca Province.

Proscopiid Biology, Distribution and Host Plant Associations The field work and collection of A. bidentata were carried out mainly in the area of Miraflores, 70 km S of Catamarca City. This area is an ecotone between the phytogeo- graphical regions of Chaco (dry forest ecosystem) and Monte (shrub desert ecosystem). The collection sites of A. saltense were Santa Marı´a (350 km NW of Catamarca City) and Andalgala´ (130 km W of Catamarca City) (Monte phytogeographical region). Adults and large nymphs of both proscopiids were collected by hand, and first instars by sweeping and beating. Adults and nymphs were kept in 350-mL transparent plastic containers at room

temperature (22Á/288C), and fed fresh leaves of Larrea every other day. A maximum of six adults or large nymphs were placed in a single container. BIOLOGY AND HOST RANGE OF TWO ARGENTINE ACRIDIDS 41

During 1992 and 1993, we conducted behavior observations in the field (copulation, feeding, oviposition, and movement) at different times of the day (from 07:00 to 01:00 h) on both adults and nymphs of A. bidentata at Miraflores (almost daily), and of A. saltense at Villa Marı´a (two to three times a month). First instar A. bidentata were sampled by beating 10 plants of each of the following species: L. cuneifolia Cav., L. divaricata Cav., Cercidium praecox (Ruiz & Pav.) Harms, (Hooker et Arnott), Acacia sp., A. aroma Gillies, Prosopis chilensis (Mol). Stunz, P. torquata (Cav. Ap. Lag.) DC., P. nigra Griseb, Cassia aphylla Cav.: Zisyphus mistol Griseb; and Aspidosperma quebracho-blanco. 2 Astroma saltense densities (adults and nymphs) on plants were estimated in a 35/25-m area in Villa Marı´a. Plants sampled were L. cuneifolia, Zuccagnia punctata, Bulnesia schickendantzi, Prosopis sp. and Plectocarpa tetracantha, which made up 90% of the species in the area. Small and medium sized nymphs were captured by beating, and large nymphs and adults were captured by hand. The plants were selected at random and in proportion to their availability. Development. In the laboratory, field-collected nymphs were reared on L. divaricata and the development was monitored. The number of instars of A. bidentata and A. saltense was established by measuring the third femur of field-collected nymphs. Cast skins were observed to confirm that molt had taken place. The femur length was chosen because it was found that it had the lowest variance compared to head width, body length, width and thorax length. Most first instars collected in the field and all nymphs hatching from eggs obtained in the lab were reserved for preference tests. Small nymphs measured under the dissecting microscope were inevitably injured. Oviposition. Oviposition behavior and substrate preference were studied in the laboratory. Two different types of substrata, sandy-clay soil and sand (fine river sand) both dry and

moistened, were offered in trays to females of A. bidentata (n/15) and A. saltense (n/15). Two individuals of each species were placed in a 1-m3 cage with branches of L. divaricata in water vases. The substrata were sieved every 3 days in search of egg masses. When found, they were opened and the number of eggs counted. Length and width of individual eggs of A.

bidentata and A. saltense (n/20) were measured. Diapause. The life cycle of both proscopiids included a winter egg diapause. In order to establish the conditions under which diapause was broken, 25 newly laid eggs of A. bidentata

were placed in rearing chambers at 28C (range 0Á/48C) for 30 days, and then moved to a rearing chamber at 308C. At the same time, another set of 25 eggs was incubated at 308C. The same test was performed on A. saltense. Mechanism of diapause induction was not studied. Geographical distribution. The distribution of both proscopiids in northwestern Argentina was established by collecting nymphs and adults by hand and by sweeping from 1990 to 1993 in the provinces of Catamarca, La Rioja, San Juan, Santiago del Estero, north of Mendoza,

Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 northwest of Co´rdoba, northeast of Tucuma´n and south of Salta. We compared proscopiid distribution maps with climate maps: mean temperature, precipitation, relative humidity, minimun and maximum temperature (Servicio Meteorolo´gico Nacional, 1960) to identify the main climate variables affecting distributions of proscopiids.

Host Range Host ranges of A. bidentata and A. saltense were established by performing feeding preference tests in the laboratory, and nymphal development tests in the field and laboratory. Both field-collected and laboratory-reared nymphs (first instar) were used in the tests. A total of 33 plant species were selected to be used in the tests with the following criteria: (a) Phylogenetic relatedness. Plants phylogenetically close to the genus Larrea that shared the same habitat as the proscopiids were selected using Soltis et al. (2000) angiosperm phylogeny, based in molecular traits (18S rDNA, rbcL, and atpB sequences). Zygophyllacea: L. cuneifolia Cav., L. divaricata Cav., L. nitida Cav., Bulnesia schickendantzii Hieron, B. 42 G. LOGARZO & M. CASALINUOVO

retama (Gillies ex Hooker et Arnott), B. bonariensis Griseb, B. foliosa Griseb, Plectocarpa tetracantha Gillies, Plectocarpa sp., Porlieria microphylla (Baill.), Tribulus terrestris L.; : Cercidium praecox (Ruiz & Pav.) Harms, Zucagnia punctata Cav., Geoffroea decorticans (Hooker et Arnott), Acacia furcatispina Burk., A. aroma Gillies, Prosopis chilensis (Mol). Stunz, P. torquata (Cav. Ap. Lag.) DC., P. nigra Griseb, Senna aphylla Cav., Mimosa sp. Malpighiacea: Tricomaria usillo Gillies, Oxalidaceae: Oxalis sp. y. (b) Plants used in the field. Plant species that were used in the field by at least one of the proscopiids studied, but were not closely related to Larrea: Apocynaceae: Aspidosperma quebracho-blanco, Olacaceae: Ximena americana L., Celtidaceae: Celtis tala Gillies, Rhamnaceae: Zizyphus mistol Griseb. (c) Plants of economical importance. Plants of economical value related to Larrea: Rutaceae: Citrus sp., Ruta chalepensis L., Meliaceae: Melia azederach L., Tropaeolaceae: Tropaeolum majus L, Geraniaceae: Geranium sp., Linaceae: Linum perenne L. Nymph feeding preference (multiple choice tests). Feeding preference was estimated based on the consumption of leaves by first instars. Eight Zygophyllaceae and four species of Fabaceae occurring in the Monte-Chaco ecotone were used with A. bidentata. In the case of A. saltense, 32 plant species from the Monte community and ornamental/cultivated plants, nine Zygophyllaceae, 11 Fabaceae and 12 species from 11 other families were used. Between two and 14 leaves of each species (according to the leaf size in order to offer equivalent leaf area) were pined in two pins and randomly placed and in an 800-mL plastic container with a transparent lid. Leaves of the same species were placed opposite each other. Five first instars that had fasted for 24 h were placed in each container with the leaves. After 48 h, the insects were removed and the leaf area consumed of each plant was measured with a square millimeter grid. Nine replicates were performed with A. bidentata and 10 with A. saltense, using different insects in each replicate. Nymph development test. (non choice test). The tests were carried out either on potted plants purchased from local nurseries, or on plants growing close to the laboratory. Ornamental and cultivated plants were tested only on potted plants. The field-plants were transplanted in December (end of spring), and tests were started at the beginning of February (mid-summer). Nine species of Zygophyllaceae, nine species of Fabaceae and 11 species from 10 other families were selected for both proscopiids. Ten replications were carried out per plant species except where stated otherwise. For each plant, a branch was covered with a mesh bag that enclosed one first instar. The bags were checked every 3 days to monitor nymph development.

Statistical Analysis Consumption results of the multiple choice tests were compared with an ANOVA, followed by LSD paired comparisons, to determine the order of plant preferences. Results are Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 reported as mean (9/SD).

RESULTS All proscopiid specimens collected on Larrea spp. at the 19 sampled sites were identified as A. bidentata or A. saltense. Morphological comparison among specimens collected at Miraflores, specimens identified by Cordo and De Loach (1993) as Astroma riojanum, the original description of Astroma riojanum, and the type held in the Museo de La Plata, showed major differences. The specimens collected at Miraflores had an arolium present in the tarsus, characteristic of the genus Anchocoema, but not in the genus Astroma. Alba Bentos (Universidad de la Repu´blica, Uruguay) identified our specimens of Anchocoema as A. bidentata. The other proscopiid specimens collected at Andalgala´ and Santa Marı´a, Catamarca Province, matched with the original description of Astroma saltense. We conclude BIOLOGY AND HOST RANGE OF TWO ARGENTINE ACRIDIDS 43

that the proscopiid species studied by others in the same area were misidentified. Astroma saltense was misidentified as Astroma quadrilobatum (Rhoades, 1977; Schultz et al., 1977), and A. riojanum (Mares et al., 1977; Schultz, 1979). Anchocoema bidentata was named by Cordo and De Loach (1993) as Astroma riojanum, and probably in Mares et al. (1977) as Anchocoema subalata.

Proscopiid Biology Anchocoema bidentata and A. saltense resemble the stems of Larrea species and other desert shrubs. Adults of both proscopiids exhibit sexual dimorphism (males are smaller than females), and adults have slow movements, as opposed to nymphs, which are very active jumpers. Adult females hardly jump, but the males retain the jumping ability as adults. Anchocoema bidentata:development. Of 250 A. bidentata nymphs collected in Miraflores, five instars were observed (Table 1). Males and females showed development differences; for example, the length of the third leg femur of a male is similar to that of a female of the previous instar. Adult males femur lengths were similar to those of fifth instar female.

Average length of the femur III for adult males was 15.29/1.1 mm, and for adult females it was 20.69/2.0 mm (Table 1). Apparently, the sampling method used to collect nymphs missed first instar males because only four first instar males of A. bidentata and seven of A. saltense were sampled in about 60 first instars sampled. Although Orions et al. (1977) stated that herbivores of Larrea copulate and oviposit at night, we observed mating during the daytime. At night, females were very active; they left the plants and moved on to the ground looking for oviposition sites and/or to change plants. Egg laying was observed only at night, females

oviposited in the ground by making a 4.29/0.4-cm deep hole (n/4) with their ovipositor. At the bottom of the hole, the females formed a chamber where the eggs were laid. Oviposition. In the laboratory, oviposition occurred in sandy clay soil sufficiently moist to prevent the chamber from caving in. The egg mass was cylindrical and the average length and

diameter were 9.699/1.21 and 6.189/0.62 mm, respectively. Females laid on average 27.69/6.4 eggs per mass (n/ 7). The eggs were cylindrical, with a pointed end (average length 5.69/0.2 mm, mean maximum width 1.19/0.1 mm). The chorion was light brown, stiff and finely punctate. A total of 64% (16/25) of the eggs hatched after chilling treatment. In contrast, the eggs held at a constant temperature of 308C and 80% RH did not hatch. Seasonal cycle. In the field we observed that A. bidentata overwinters as eggs; however, on occasion the adults could survive until spring in warm winters. In the field, first instars appeared from mid-November to mid-December (spring), depending on the spring rainfalls, suggesting that the rainfall triggers hatching. In 1992, they appeared at mid-December, a second hatch was observed at the end of February (summer) and a third hatch took place at the end of April (fall). Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 Hosts plants. Adults and medium and large sized nymphs of A. bidentata were observed on several plant species (Prosopis nigra, P. torcuata, P. chilensis, Mimosa sp., Acacia aromo and A. aphylla) in Miraflores. First instars were collected only on L. divaricata (21 nymphs), L. cuneifolia (16), G. decorticans (one), and P. torcuata (one). None were collected on the other plants sampled. Astroma saltense. The biology of A. saltense was similar to that of A. bidentata. Females

laid an average of 20.29/3.3 eggs per mass (n/5). Egg masses were found under the same conditions as for A. bidentata. Astroma saltense eggs and egg masses were similar to those of A. bidentata. Under the same conditions as in the diapause test for the other proscopiids, 20 nymphs emerged from the chilled eggs, and no nymphs emerged from the control treatment. Approximately 200 nymphs of A. saltense were collected in Andalgala. Table 1 shows the average and SD of femur III lengths for each of the five instars in males and females. The relationship between the female and male lengths mentioned for A. bidentata holds for A. saltense. 44 G. LOGARZO & M. CASALINUOVO

TABLE 1. Measure of the third femur of the inmature stages and adults of male and female Anchocoema bidentata and Astroma saltense

Mean length of femur (mm) Anchocoema bidentata Astroma saltense

Male Female Male Female

Stage N 09/SD N 09/SD N 09/SD n 09/SD st 1 instar 4 3.49/0.1 22 4.19/0.2 7 3.59/0.2 27 3.99/0.2 nd 2 instar 52 4.19/0.2 62 5.59/0.3 56 4.19/0.2 22 5.49/0.3 rd 3 instar 23 5.69/0.4 40 8.09/0.5 58 5.39/0.4 13 7.99/0.6 th 4 instar 28 8.09/0.5 31 10.49/0.5 23 8.09/0.7 10 10.69/0.5 th 5 instar 32 10.69/0.5 29 14.79/1.2 14 10.59/1.2 23 15.79/1.1 Adult 22 15.29/1.1 24 20.69/2.0 22 13.09/0.7 20 18.49/1.3

Host plants. In December 1992, first instars of A. saltense were collected on B. schickendantzi (70 nymphs/plant), L. cuneifolia (11 nymphs/plant), Z. punctata (seven nymphs/plant), and Prosopis sp. (five nymphs/plant) in Santa Marı´a, Catamarca. Although L. cuneifolia was the plant species with higher plant cover (56%), we collected 7 times more first instar nymphs per plant on B. schickendantzi than on L. cuneifolia. Throughout the summer, we observed adults and nymphs of A. saltense on the above-mentioned plants. In Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010

FIGURE 1. Distribution of Anchocoema bidentata and Astroma saltense on Larrea in Northwestern Argentina The sampling areas shown are: (1) Miraflores; (2) Andalgala´; and (3) Santa Marı´a; (----) annual rainfall (mm). BIOLOGY AND HOST RANGE OF TWO ARGENTINE ACRIDIDS 45

TABLE 2. Mean leaf area consumed by first instar nymphs of Anchocoema bidentata

2 1 2 Plant species 09/SD (mm )%TCLSD test

Larrea divaricata 32.99/16.3 40.2 a Bulnesia foliosa 17.39/5.3 21.1 bcd Bulnesia schickendantzi 11.99/7.9 14.5 bcd Bulnesia retama 8.99/7.3 10.9 cdef Larrea cuneifolia 6.29/7.7 7.5 cdef Larrea nitida 2.99/3.9 3.5 defg Zuccagnia punctata 1.29/1.2 1.5 defg Porlieria microphylla 0.29/0.7 0.2 efg Tribulus terrestris 0.29/0.7 0.2 efg Acacia aroma 0.29/0.7 0.2 efg Prosopis torquata 0.0Á/ 0.0 Á/ Cassia aphylla 0.0Á/ 0.0 Á/

1%TC, percentage of total consumption. 2 Means followed by the same letter are not significantly different, P B/0.05).

March 1993, we sampled an average of 4.44 (9/3.54) adults per plant on B. schickendantzi (n/10), and 4.06 (9/4.08) adults per plants on L. cuneifolia(n/10). Geographical distribution. The distribution of A. bidentata and A. saltense barely overlap (Figure 1), and they were never collected both at the same site. Astroma saltense were collected west of the 200-mm rainfall line (annual rainfall lower than 200 mm), where L. cuneifolia was the dominant plant and L. divaricata was scarce and riparian (Morello, 1955, 1958); in the Monte phytogeographic region. Anchocoema bidentata, on the contrary, were collected east of the 200-mm rainfall line

where annual rainfall is higher (200Á/400 mm per year) (Morello, 1955, 1958) and L. divaricata was the dominant plant; in the Espinal and the ecotone between the Monte and Chaco phytogeographic regions.

Host Range Anchocoema bidentata: nymph feeding preference (multiple choice test). The first instar of A. bidentata showed significant differences between plant species in the consumption of leaves 2 (F/ 20.03; df/11, 108; P B/0.05) (Table 2). Larrea divaricata (32.99/16.3 mm ; TC/40.2%) was the preferred species, and accounted for more than 40% of total 2 consumption (TC). Consumption of B. foliosa (17.39/5.3 mm ;TC/21.1%) and B. 2 schickendantzi (11.99/7.9 mm ;TC/14%) was almost 50% lower. The third group of 2 plants in order of preference were, B. retama (8.99/7.3 mm ;TC/10.9%), and L. cuneifolia Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 2 (6.29/7.7 mm ;TC/7.5%). The remaining seven plant species tested formed the lowest consumption group. More than 75% of the total consumption by A. bidentata was of L. divaricata, B. foliosa, and B. schickendantzi (Table 3). Nymph development test (non choice test). A. bidentata completed its development only on L. divaricata and L. cuneifolia (Table 4). Thirty-five percent of the nymphs reached the adult stage on L. divaricata, and 30% on L. cuneifolia. Although nymphs survived up to 6 weeks on seven plant species, none of these reached the adult stage: 22% of nymphs survived for 6 weeks on P. torquata, 20% on B. foliosa, 10% on B. schickendantzi, 10% on B. retama, 10% on P. tetracantha, 10% on Porlieria microphylla, and 10% on Acacia sp. (Table 4). The nymphs did not survive more than 5 days on the remaining 20 plant species tested. Astroma saltense: nymph feeding preference (multiple choice test). The first instar nymphs of A. saltense showed significant differences between plant species in the consumption of

leaves (F/6.89; df/31, 274; P B/0.05), resulting in four groups (LSD test) (Table 3). The 2 most preferred species were L. cuneifolia (68.759/53.00 mm ;TC/25.3%) and L. divaricata 46 G. LOGARZO & M. CASALINUOVO

TABLE 3. Mean leaf area consumed by first instar nymphs of Astroma saltense

2 1 2 Plant species 09/SD (mm )%TCLSD test

Larrea cuneifolia 68.759/53.00 25.3 a Larrea divaricata 63.009/54.63 23.6 a Larrea nitidai 32.409/47.78 11.9 b Bulnesia schickendantzi 25.559/38.09 9.4 b Bulnesia retama 10.159/16.92 3.7 b Cercidium praecox 10.059/7.06 3.7 b Prosopis chilensis 9.209/10.36 3.4 b Zucagnia punctata 7.809/12.50 2.9 bc Tricomaria usillo 6.659/8.89 2.5 c Geoffroea decorticans 6.359/7.69 2.3 c Bulnesia bonariensis 5.609/14.53 2.0 c Bulnesia foliosa 5.309/7.82 1.9 c Acacia furcatispina 4.009/11.31 1.6 c Acacia aroma 3.959/8.82 1.5 c Prosopis torquata 3.909/11.64 1.4 c Plectocarpa tetracantha 3.209/4.40 1.1 c Prosopis nigra 2.059/2.98 0.7 c Porlieria microphylla 1.209/1.53 0.4 c Oxalis sp. 1.009/2.53 0.4 c Mimosa sp. 0.859/2.68 0.3 c Tribulus terrestris 0.00Á/ 0.0 Cassia aphylla 0.00Á/ 0.0 Citrus sp. 0.00Á/ 0.0 Ruta chalepensis 0.00Á/ 0.0 Celtis tala 0.00Á/ 0.0 Zisyphus mistol 0.00Á/ 0.0 Melia azederach 0.00Á/ 0.0 Tropaeolum majus 0.00Á/ 0.0 Aspidosperma quebracho-blanco 0.00Á/ 0.0 Linum perenne 0.00Á/ 0.0 Pelargonium sp. 0.00Á/ 0.0 Ximena americana 0.00Á/ 0.0

1%TC, percentage of total consumption. 2 Means followed by the same letter are not significantly different, P B/0.05).

2 (63.009/54.63 mm ;TC/23.6%). The second most preferred (TC/35.0%) were L. nitida, B. schickndantzi, B. retama (Zygophyllacea) and Z. punctata, C. praecox, P. chilensis (Leguminosae). Astroma saltense also fed on 12 other plant species, which accounted for 17.2% of the total consumption (Table 3). Of 32 plant species tested, A. saltense consumed leaves of 12 different genera in four families. More than 60% of the total consumption was on Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 Larrea species. Nymph development test (non choice tests). A. saltense completed its development on L. cuneifolia (40% adults produced), L. divaricata (30%), B. retama (20%), B. schickendantzi (20%), all Zygophyllaceae, and on the Leguminosae Z. punctata (30%) and P. torquata (11%) (Table 5). Twenty percent of the nymphs survived on G. decorticans and 10% on Mimosa sp. for 6 weeks, but no adults were obtained. Nymphs did not survive more than 1 week on the remaining species.

DISCUSSION Anchocoema bidentata and A. saltense were able to feed on a wide range of plant species (Tables 2 and 3) as shown by field observations and laboratory results. Results of laboratory feeding tests (multlipe choice test) were highly variable. Consumption of leaves of a same plant species exhibited dramatic variations along the test replications. Leaves were almost BIOLOGY AND HOST RANGE OF TWO ARGENTINE ACRIDIDS 47

TABLE 4. Survival of first instar nymph Anchocoema bidentata to the adults stage

Nymphs that reached at least:

Plant species No ind. 2 weeks 4 weeks 6 weeks Adult sage Larrea divaricata 20 8 7 7 7 Larrea cuneifolia 10 7 7 7 3 Mimosa sp. 10 8 6 2 0 Prosopis torquata 10 9 3 2 0 Tricomaria usillo 10 4 3 1 0 Porlieria microphylla 10 10 2 1 0 Bulnesia retama 10 8 1 1 0 Acacia sp. 10 10 3 0 0 Bulnesia foliosa 10 9 1 0 0 Plectocarpa tetracantha 10 7 1 0 0 Bulnesia schickendantzi 10 5 0 0 0 Geoffroea decorticans 10 7 0 0 0 Zucagnia punctata 10 5 0 0 0 Prosopis chilensis 10 5 0 0 0 Zisyphus mistol 10 4 0 0 0 Cercidium praecox 10 3 0 0 0 Acacia aroma 10 2 0 0 0 Bulnesia bonariensis 10 0 0 0 0 Tribulus terrestris 10 0 0 0 0 Cassia aphylla 10 0 0 0 0 Citrus sp. 10 0 0 0 0 Ruta chalepensis 10 0 0 0 0 Celtis tala 10 0 0 0 0 Oxalis sp. 10 0 0 0 0 Melia azederach 10 0 0 0 0 Tropaeolum majus 10 0 0 0 0 A. quebracho-blanco 10 0 0 0 0 Pelargonium sp. 10 0 0 0 0 Ximena americana 10 0 0 0 0

completely eaten in one replication, and not eaten at all or barely eaten in other replications. This was observed with both proscopiids and on most plant species, even on those on which the proscopiids completed their development. Suitability differences between leaves of a same plant species chosen for the experiment may be interfering with the insects’ feeding behavior. Despite the high variability of the multiple choice test results, there seemed to be a feeding pattern: each proscopiid utilized the same three plant species in all replications and varied in

Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 their utilization of the remaining plant species. Anchocoema bidentata utilized only L. divaricata, B. foliosa and B. retama, and A. saltense utilized only L. divaricata L. cuneifolia and C. praecox in all replications. The proscopiids under study differed in the complexness of their development on the tested plants. Anchocoema bidentata only completed its development (from first instar to adult) on L. divaricata and L. cuneifolia, while A. saltense did so on six plant species from two families. The host range tests showed high nymph mortality on all plants tested, even on the natural host plants (Tables 4 and 5). This inexplicable mortality could be due to natural causes like weather or predation or be induced by the experiment itself. The presence of a bag could have prevented the proscopiids from seeking protection from the rain or predators. Rhoades (1977) and Schultz et al. (1977) reported that A. saltense (as A. quadrilobatum) fed on the leaves of five woody perennials in Andalgala, both Larrea species, B. retama, Z. punctata, and T. usillo. In our studies, first instar A. saltense fed and developed on all those plants except T. usillo, where the insect fed, but was not able to develop. 48 G. LOGARZO & M. CASALINUOVO

TABLE 5. Survival of first instar nymph Astroma saltense to the adult stage

Nymphs that reached at least:

Plant species No Ind. 2 weeks 4 weeks 6 weeks Adult stage Larrea cuneifolia 10 9 9 9 4 Larrea divaricata 10 10 6 3 3 Zucagnia punctata 10 7 4 4 3 Bulnesia schickendantzi 10 10 10 6 2 Bulnesia retama 10 9 6 4 2 Prosopis torquata 9741 1 Geoffroea decorticans 10 3 2 2 0 Mimosa sp. 10 7 7 1 0 Bulnesia foliosa 10 4 1 0 0 Oxalis sp. 10 1 1 0 0 Plectocarpa tetracantha 10 10 0 0 0 Acacia sp. 10 6 0 0 0 Prosopis chilensis 10 5 0 0 0 Porlieria microphylla 10 1 0 0 0 Tribulus terrestris 10 0 0 0 0 Bulnesia bonariensis 10 0 0 0 0 Acacia aroma 10 0 0 0 0 Cassia aphylla 10 0 0 0 0 Cercidium praecox 10 0 0 0 0 Citrus sp. 10 0 0 0 0 Ruta chalepensis 10 0 0 0 0 Celtis tala 10 0 0 0 0 Tricomaria usillo 10 0 0 0 0 Zisyphus mistol 10 0 0 0 0 Melia azederach 10 0 0 0 0 Tropaeolum majus 10 0 0 0 0 A. quebracho-blanco 10 0 0 0 0 Pelargonium sp. 10 0 0 0 0 Ximena americana 10 0 0 0 0

The only other study on proscopiid biology (Aguilar, 1974), carried out on the Peruvian polyphagous Anchotatus peruvianus Caudell does not provide much information in general and gives no data on immature stages. To our knowledge, this is the first report that includes substantial information on bionomics, including immature stages of proscopiids. Insect plant mimicry has been associated with adaptation to hiding substrates and to plant utilization patterns (Orians, 1977; Schultz et al., 1977). Insects cryptic on just one plant species supposedly evolved specificity towards that species, whereas insects cryptic on many Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 plant species evolved to be non-specific. Specialized mimicry is thought to respond to a morphological or chemical defense independent of the host plant species. In the system we studied, A. saltensis responds to the literature pattern since it is a generalist and able to camouflage on several plant species (B. retama, Z. punctata, L. cuneifolia, L. divaricata) that showed morphological and chemical defense similarities (Schultz et al., 1977). Anchocoema bidentata, on the contrary, exhibited mimicry on several plant species, but developed only on Larrea spp. However, chemical plant defense on plant species associated with A. bidentata habitat have not been studied. An insects’ ability to camouflage in one or several plant species might not only be related to plant defense but also with hiding substrate for predation escape. Anchocoema bidentata occurred between the annual rainfall lines of 200 and 400 mm and was associated with L. divaricata. Astroma saltense occurred between the annual rainfall lines of 100 and 290 mm and was associated with L. cuneifolia, B. retama, and Z. punctata. Orions et al. (1977) suggested that there is an allopatric displacement from north to south BIOLOGY AND HOST RANGE OF TWO ARGENTINE ACRIDIDS 49

between Astroma saltense and Anchocoema bidentata. However, we observed that the separation between these species coincides approximately with the 200-mm rainfall line, the displacement being east to west in most of the range (Figure 1). Morello (1958) describes the Western border of the Monte as the 200-mm isohyet, except at two sites, Cafayate and Andalgala´ areas, where the Monte distribution falls outside the 200-mm isohyet. A.saltense followed the same distribution pattern, as shown by the collection sites (Figure 1). We conclude that A. bidentata has some biological attributes that make it a good candidate for biological control of Larrea tridentata: (1) it is restricted to Larrea spp.; (2) has two to three generations per year; (3) overwinters as eggs, with a diapause mechanism regulated by rainfall. Studies are recommended to establish the behavior of A. bidentata on L. tridentata. Other Zygophyllaceae species close to the genus Larrea,asPorleria, Bulnesia, Plectocarpa, and Guaiacum (Porter, 1979) should be included in the host range tests to avoid effects on non target native plants (Stiling & Simberloff, 2000) related to Larrea. Also, climatic studies of the distribution areas of A. bidentata and L. tridentata should be considered to find out the potential distribution of the proscopiids in the northern hemisphere. Regarding A. saltense, the host range tests confirmed Schultz’s (1977) field observations. This species has a wide host range, so we recommend that it should not be considered as a biological control agent of L. tridentata.

ACKNOWLEDGEMENTS We are very grateful to Alejandra Carminati, Marı´a Teresa Amela, and Alicia Liva for field and laboratory assistance. We also thank Karen Braun, Juan Briano, Arabella Bugliani, and Willie Cabrera Walsh for comments to improve the manuscript. Finally, we want to thank the late Ricardo Ronderos and Alba Bentos for helping in the proscopiid identifications.

REFERENCES

AGUILAR, P.G. (1974) Los palitos vivientes de Lima. II: Proscopiidae de las lomas. Revista Peruana de Entomologı´a 17, 107Á/113. CORDO, H.A. & DE LOACH, J.C. (1993) Phytophagous insects that attack Larrea spp. (Zygophyllaceae) in Argentina and their potential for biocontrol of creosotebush (L. tridentata) in North America. Biological Control 3,6Á/16. HUNZIKER, J.H. (1975) On the geographical origin of Larrea divaricata (Zygophyllaceae). Annals of the Missouri Botanical Garden 62, 497Á/500. HUNZIKER, J.H.,PALACIOS, R.A., DE VALESI, A.G. &POGGIO, L. (1972) Species disjunction in Larrea: evidence from morphology, cytogenetics, phenolic compounds and seed albumins. Annals of the Missouri Botanical Garden 59, 224 /233. Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010 Á HUNZIKER, J.H., PALACIOS, R.A., POGGIO, L., NARANJO, C.A. & YANG T.W. (1977) Geographic distribution, morphology, hybridization, cytogenetics, and evolution, in Creosote Bush: Biology and Chemistry of Larrea in New World Desert (MABRY, T.J., HUNZIKER, J.H. & DI FEO, D.R. Jr, Eds.) US/IBP Synthesis Ser. 1-6. Dowden, Hutchinson and Ross, Inc., Stroudsburg, PA, pp. 10Á/47. MARES, M.A.,ENDERS, F.A.,GREEGOR, D.,HULSE, A.C.,HUNT, J.H.,OTTE, D.,SAGE, R.D. &TOMOFF, C.S. (1977) Strategies and community patterns of desert animals, in Convergent Evolution in Warm Deserts (ORIANS, G.H. &SOLBRIG, O.T., Eds.). Dowden Hutchinson & Ross Inc, Stroudsburg, PA, pp. 226Á/255. MELGOZA, A.,MORTON, H.L.,SIERRA, J.S. &MENGOLZA, B. (1984) Botanical changes associated with applications of tebuthiuron in creosotebush (Larrea tridentata) communities. Proceedings-Western Society of Weed Science 37,98Á/112. MORELLO, J. (1955) Estudios bota´nicos en las regiones a´ridas de la Argentina. I. Ambiente, morfologı´a y anatomı´a de cuatro arbustos resinosos de follaje permanente del Monte. Revista de Agronomı´a del. Noroeste Argentino 1, 301Á/370. MORELLO, J. (1958) La provincia fitogeogra´fica del monte. Opera Lilloana II, Tucuma´n, Repu´blica Argentina. pp. 153. ORIONS, G.H.,CATES, R.G.,MARES, M.A.,MOLDENKE, A.,NEFF, J.,RHOADES, D.F.,ROSENZWEIG, M.L., SIMPSON, B.B.,SCHULTZ, J.C. &TOMOFF, C.S. (1977) Resource utilization systems, in Convergent 50 G. LOGARZO & M. CASALINUOVO

Evolution in Warm Deserts. Dowden Hutchinson & Ross Inc, Stroudsburg PA, pp. 226Á/255. PLATT, K.B. (1959) Plant control. Some possibilities and limitations. II. Vital statistics of range management. Journal of Range Management 12, 194Á/200. PORTER, D.M. (1979) The position of Larrea in the Zygophyllaceae, in Serie el Desierto, Vol. 2, Larrea (CAMPOS LOPEZ, E., MABRY, T.J. & TAVIZON, S.F., Eds.). Centro de Investigacion en Quimica Aplicada, Saltillo, Coahuila, Mexico. pp. 27Á/40. RHOADES, D.F. (1977) The antiherbivore chemistry of Larrea,inCreosote Bush: Biology and Chemistry of Larrea in New World Desert (MABRY, T.J., HUNZIKER, J.H. & DI FEO, D.R. Jr., Eds.). US/IBP Synthesis Ser. 1-6. Dowden, Hutchinson and Ross, inc., Stroudsburg, PA, pp. 135Á/175. SERVICIO METEOROLO´ GICO NACIONAL (1960) Atlas Clima´tico de la Repu´blica Argentina. Secretarı´a de Aerona´utica, Argentina. 66 maps. SCHULTZ, J.C. (1979) Asymptotic species richness models and the insect faunas of Larrea,inLarrea (CAMPOS LOPEZ, E., MABRY,T.J.&TAVIZON, S.F., Tech. Eds.). Serie el Desierto, vol. 2. Centro de Investigaciones en Quı´mica Aplicada. Saltillo, Coahuila, Mexico, pp. 101Á/125. SCHULTZ, J.C., OTTE,D.&ENDERS, F. (1977) Larrea as a habitat component for desert arthropods, in Creosote Bush: Biology and Chemistry of Larrea in New World Desert (MABRY, T.J., HUNZIKER, J.H. & DI FEO, D.R. Jr., Eds.). US/IBP Synthesis Ser. 1-6. Dowden, Hutchinson and Ross, inc., Stroudsburg, PA, pp. 176Á/208. SOLTIS, D.E.,SOLTIS, P.S.,CHASE, M.W.,MORT, M.E.,ALBACH, D.C.,ZANIS, M.,SAVOLAINEN, V.,HAHN, W., HOOT, S.B.,FAY, M.F.,AXTELL, M.,SWENSEN, S.M.,PRINCE, L.M.,KRESS, W.J.,NIXON, K.C. &FARRIS, J.S. (2000) Angiosperm phylogeny inferred from rDNA, rbcL, and atpB sequences. Botanical Journal of the Linnean Society 133, 381Á/461. STILING, P. &SIMBERLOFF, D. (2000) The frequency and the strength of nontarget effects of invertebrate biological control agents of plant pests and weeds, in Non Target Effects of Biological Control (FOLLET, P. &DUAN, J., Eds.). Kluwer Academic, Norwell, MA, pp. 31Á/43. VA N DEVENDER, T.R. (1973) Late Pleistocene plants and animals of the Sonoran Desert: a survey of ancient packrat middents in southwestern Arizona. Ph.D. Thesis. University of Arizona, Tucson, AZ, pp. 199. WELLS, P.V. &HUNZIKER, J.H. (1976) Origin of the creosote bush (Larrea) deserts of southwestern North America. Annals of the Missouri Botanical Garden 63, 843Á/861. YORK, J.C. &DICK-PEDDIE, W.A. (1969) Vegetation changes in southern New Mexico during the past 100 years, in Arid Lands in Perspective (MCGINNIES, W.G. &GOLDMAN, B.J., Eds.). University of Arizona Press, Tucson, AZ, pp. 157Á/166. Downloaded By: [USDA National Agricultural Library] At: 12:53 24 May 2010