FOREST ENTOMOLOGY The Effects of Eucalyptus Pollen on Longevity and Fecundity of Eucalyptus Longhorned Borers (Coleoptera: Cerambycidae)

1 1 1, 2 3 JOCELYN G. MILLAR, TIMOTHY D. PAINE, ANDREA L. JOYCE, AND LAWRENCE M. HANKS

J. Econ. Entomol. 96(2): 370Ð376 (2003) ABSTRACT The longevity and fecundity of adult recurva and were strongly affected by diet. Female P. recurva fed a diet of Eucalyptus pollen and sucrose solution lived 34Ð56% longer than females fed diets containing other types of pollen, ground dog chow, or sucrose solution alone. Diet had no signiÞcant effect on longevity of P. recurva males. Similarly, longevities of P. semipunctata females were increased 48Ð71% on the Eucalyptus pollen diet compared with the other diets. Male P. semipunctata also lived longer on the Eucalyptus pollen diet than most of the other diets. Fecundity was dramatically affected by diet, with P. recurva females fed the Eucalyptus pollen diet laying Ϸ4Ð8 times more eggs than females on the other diets. Eucalyptus pollen also increased the fecundity of P. semipunctata females Ϸ3Ð5-fold. Diet resulted in only minor effects on egg size and percent egg hatch for both species.

KEY WORDS insecta, Phoracantha semipunctata, Phoracantha recurva, pollen, nutrition, biological control, mass rearing

THE EUCALYPTUS LONGHORNED borers Phoracantha semi- (Hymenoptera: Braconidae), in South Africa (Drink- punctata F. and Phoracantha recurva Newman (Co- water 1975), Israel (Mendel 1987), and more recently, leoptera: Cerambycidae) are native to Australia, California (Hanks et al. 1996). Although evidence of where they develop in dead and dying Eucalyptus breedingpopulations (empty cocoons, adult emer- trees and slash (Wanget al. 1996, Wang1998). Both gence holes) has been found in California (Hanks et species have been introduced accidentally into many al. 1996), it is uncertain whether the larval parasitoids other countries where Eucalyptus is grown, where in have become permanently established in any of the the absence of natural enemies both species have countries in which they have been released. become moderate to severe pests, causingsubstantial Efforts to mass rear and establish natural enemies of mortality to Eucalyptus plantings (Chararas 1969a, Phoracantha in exotic countries depend upon 1969b; Drinkwater 1975; Scriven et al. 1986; Hanks et having a large and continuous source of beetle eggs al. 1993b). Female beetles deposit eggs under exfoli- and larvae as hosts for the parasitoids. EfÞcacious atingbark or in bark crevices, and the developing artiÞcial diets for mass rearingthe beetles have not yet larvae feed in the cambium layers, rapidly girdling and been developed, and because of the beetlesÕ longde- killinginfested trees. In contrast, the adult beetles feed velopment times, artiÞcial diets would probably be on nectar and pollen, causingno apparent damageto logistically impractical for large-scale rearing. Instead, their hosts (Scriven et al. 1986, Hanks et al. 1993a). methods of mass rearingboth species of beetles on Attempts have been made to introduce and estab- Eucalyptus logs have been developed (Hanks et al. lish biological control agents for Phoracantha beetles 1993a). These methods consist of hand-infestingpre- in the countries in which the beetles have become conditioned logs with neonate larvae at a density that signiÞcant pests, including South Africa, the countries maximizes adult emergence and limits larval compe- around the Mediterranean sea, and the United States. tition, storage of the infested logs under controlled For example, the egg parasitoid Avetianella longoi Sis- temperature conditions for several months as the lar- caro (Hymenoptera: Encyrtidae) is now established in vae develop, manipulation of temperatures to induce Italy and other Mediterranean countries (Longo et al. pupation and adult emergence, and collection of 1993), California (Hanks et al. 1995a, 1996), and South adults in emergence cages. The adults are then held Africa (Hanks et al. 1996). Attempts have been made with sucrose solution in mixed-sex cages for egg pro- to introduce and establish parasitoids of Phoracantha duction. larvae, includingthe wasp Syngaster lepidus Brulle´ The production of beetle eggs proved to be a lim- itingfactor in efforts to mass rear the eggparasitoid A. 1 Department of Entomology, University of California, Riverside, longoi. Because of the substantial investment of time, CA 92521. 2 E-mail: [email protected]. labor, and facilities required to produce adult beetles, 3 Department of Entomology, University of Illinois at Urbana- we reasoned that it might be more cost-effective to Champaign, Urbana, IL 61801. test methods of increasingeggproductionper female

0022-0493/03/0370Ð0376$04.00/0 ᭧ 2003 Entomological Society of America April 2003 MILLAR ET AL.: EFFECTS OF Eucalyptus POLLEN 371 than to rear larger numbers of beetles. In the labora- 1 cm diameter (Patterson Dental Supply Co., St. Paul tory, female beetles lay as many as 30 eggs per day MN). Two vials of the solution were used in each cage duringtheir lifetimes of several weeks (Hanks et al. of the sucrose solution only treatment. The Pedigree 1993a). We hypothesized that adult nutrition would dogchow diet (Kal Can Foods Inc., Vernon CA; 21% be important in maximizingreproductive output, with protein) was ground brießy in a blender, then sieved ingested nutrients enhancing egg production. Adults through a Tyler Standard Screen Scale size seven into of Phoracantha species, as with many species of the screen size eight to produce pellets averaging 3 mm in , feed on nectar and pollen (Hanks diameter, similar in size to the pollen granules (see 1999). Furthermore, frass from Þeld-collected beetles below). The pollen diets consisted of granules of consists primarily of exines of Eucalyptus pollen honey bee-collected pollen Ϸ3 mm in diameter, in- (Hanks et al. 1993a, 1998), indicatingthat pollen con- cludinga mixed-composition pollen from the south- stitutes a signiÞcant proportion of the adult diet. We western deserts of the United States (High Desert Bee report here the results of a study designed to test the Pollen Granules, C.C. Pollen Co., Phoenix AZ; 17% null hypotheses that dietary protein did not affect the protein), a mixed composition pollen from Canada longevity and fecundity of P. semipunctata and P. re- (Chilliwack River Valley Raw Bee Pollen, Agnes curva under laboratory rearingconditions, and that all Coutts, Sardis, B.C., Canada; protein content not list- pollen sources, includinghost Eucalyptus pollen, were ed), and Australian jarrah tree pollen (Eucalyptus equivalent as sources of dietary protein for adult bee- marginata Sm.; Jarrah Tree Pollen, Great Health, Brea tles. CA; 28.6% protein). Dogchow or pollen (0.13 g; Ϸ16Ð20 granules) and one vial of sucrose solution were placed into the small Petri dish in each cage. All Methods and Materials diets were replenished every second day for the du- Adult beetles were collected daily as they emerged ration of the study. All Þve diets were fed to both P. from logs held in large cages, with collected adults semipunctata and P. recurva, with 15 replicates of each being segregated by species and sex. Newly emerged diet per species. Diets were randomly assigned to adults were held in groups of Þve females and three cages. It was not possible to measure the amount of males in cylindrical hardware cloth cages (15 cm di- each diet consumed because the pollen and dogchow ameter ϫ 15 cm high) with plastic Petri dish tops and granules rapidly disintegrated as the beetles crushed bottoms (Hanks et al. 1993a). This sex ratio had been them with their mandibles and the residue was mixed found to produce a vigorous, fertile colony (Hanks et with beetle frass and dampened by ßuid from the vials al. 1993a). Because beetles hide under loose bark dur- of sucrose solution. ing daylight hours, a refuge consisting of a square of Cages were checked every second day to assess cardboard (6.5 cm ϫ 6.5 cm) was clipped on the inside beetle survival. Eggs were collected from the cages wall of each cage. The cage ßoor was lined with a and counted every second day, and the mean numbers 12.5-cm diameter circle of Þlter paper. Each cage con- of eggs laid per 2-d period were calculated for females tained a 9-cm diameter plastic Petri dish with a Þlter of both P. semipunctata and P. recurva. At the end of paper wrapped under the bottom and around the sides the experiments, average lifetime fecundities were of the dish. The tight space between the paper-cov- determined for females of each species. ered Petri dish and the Þlter paper ßoor provided the Every 2 wk, percent egg hatch was assessed by thigmotactic stimulus that females prefer for oviposi- taking a subset of 10 eggs from each cage, still attached tion, and eggs were laid in batches sandwiched be- to slips of Þlter paper, and placingeach slip into a tween the two papers. Cages were maintained in the plastic Petri dish sealed with ParaÞlm. The dishes were laboratory under ambient light and temperature con- held under ambient laboratory light and temperature ditions (Ϸ25ЊC and 44%RH). To distinguish males conditions for 10 d, by which time all viable eggs had within a cage, the three males were marked with hatched. The numbers of hatched and nonviable eggs different colors of Liquid Paper correction ßuid. Males were counted. To determine egg size, subsets of eggs that died were replaced with fresh males as necessary (10 per cage) from randomly selected egg masses were to maintain the 5:3 ratio. Females that died were not gently separated, and the length and width of each egg replaced. The experiments were continued until all was measured with a dissectingmicroscope Þtted with females had died. a micrometer. Diet treatments consisted of: (1) sucrose solution Data Analysis. Two-way analysis of variance (0.4 M in distilled water); (2) sucrose solution and (ANOVA) was used to test the effects of diet treat- granules of dog chow; (3) sucrose solution and mixed- ments and cage on adult longevity (SAS Institute composition bee pollen from the desert areas of the 1996). Because it was not possible to determine which southwestern United States which could contain some female in a cage laid a speciÞc egg mass, differences in Eucalyptus pollen; (4) sucrose solution and mixed total fecundity amongdiet treatments were deter- composition bee pollen from British Columbia, Can- mined usingone-way ANOVA usingthe cageas the ada, which would not contain any Eucalyptus pollen; replicate. Eggs were randomly selected from all cages (5) sucrose solution and Eucalyptus pollen. Sucrose and pooled for analysis usingone-way ANOVA to test solution was dispensed from 10-ml glass vials, 1.5 cm for differences in egg size and percent hatch. Differ- diameter by 5 cm long, which were plugged with ences between means were assessed with TukeyÕs tests nonsterile cotton dental rolls, size #2, 3.5 cm longby (Sokal and Rohlf 1995). For each species, male and 372 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 96, no. 2

Fig. 1. Effects of diet on longevity of Phoracantha recurva (A) and P. semipunctata beetles (B). Data (mean Ϯ SE) for each species and sex were analyzed separately by two-way ANOVA, and differences between means were determined with TukeyÕs tests (␣ ϭ 0.05). Bars surmounted by different letters are signiÞcantly different. female survival on individual diets were compared by (Fig. 1A and F ϭ 11.30, df ϭ 4, 282, P Ͻ 0.0001). Mean oneÐway ANOVA. InterspeciÞc comparisons of lon- longevities of females fed the other proteinaceous gevity of male or female beetles on a particular diet, diets were not signiÞcantly different than the mean and fecundity of females on a particular diet, also were longevity of females fed only sucrose solution, sug- determined by one-way ANOVA. gesting that the other two types of pollen and the dog chow did not contribute signiÞcantly to the nutrition of females. In contrast, southwestern desert pollen was Results the only diet that resulted in reduced male P. recurva Diet had a signiÞcant effect on longevity of both longevity compared with the Eucalyptus pollen diet beetle species (Fig. 1). Phoracantha recurva females (F ϭ 2.51, df ϭ 4, 146, P ϭ 0.044). The effect of cage lived longer (mean Ϯ SE, 93.0 Ϯ 5.8 d) on the Euca- was signiÞcant for both female (F ϭ 4.03, df ϭ 12, 282, lyptus pollen diet than on any of the other four diets P Ͻ 0.0001) and male (F ϭ 2.08, df ϭ 12, 146, P ϭ 0.02)

Table 1. Comparisons of adult beetle longevity on different diets, between Phoracantha species (ANOVA)

Diet additivea Females F df P Males F df P Eucalyptus pollen recurva Ͼ semipunc 46.28 1, 109 Ͻ0.0001 recurva ϭ semipunc 1.29 1, 53 0.26 Desert pollen recurva Ͼ semipunc 48.95 1, 117 Ͻ0.0001 recurva ϭ semipunc 2.07 1, 58 0.15 Canadian pollen recurva Ͼ semipunc 60.08 1, 117 Ͻ0.0001 recurva Ͼ semipunc 8.80 1, 56 0.004 Dogchow recurva Ͼ semipunc 56.87 1, 133 Ͻ0.0001 recurva Ͼ semipunc 8.55 1, 63 0.005 Sucrose solution only recurva Ͼ semipunc 46.06 1, 123 Ͻ0.0001 recurva Ͼ semipunc 5.70 1, 55 0.02

a All diets contained a vial of sucrose solution. April 2003 MILLAR ET AL.: EFFECTS OF Eucalyptus POLLEN 373

Fig. 2. Mean numbers of eggs produced every8dbyP. recurva females fed on different diets. Eggs were collected every 2 d, but are shown in the Þgure as mean values calculated over8dtosimplify the graphical presentation. longevity. Overall, P. recurva females lived signiÞ- lives, while others had individuals which were very cantly longer (68.2 Ϯ 2.1 d) than males (41.9 Ϯ 1.9 d, long-lived, resulting in a signiÞcant cage effect. The F ϭ 71.9, df ϭ 1, 460, P Ͻ 0.0001). overall longevities of male (31.4 Ϯ 1.7 d) and female Phoracantha semipunctata females also survived P. semipunctata (33.2 Ϯ 1.1 d) were not signiÞcantly longer when fed the Eucalyptus pollen diet than any different (F ϭ 0.84, df ϭ 1, 440, P ϭ 0.36). of the other diets (Fig. 1B; F ϭ 11.9, df ϭ 4, 305, P Ͻ Comparisons of the longevities of female beetles 0.0001). Unlike P. recurva males, the longevity of P. between Phoracantha species fed a particular diet re- semipunctata males was affected by diet, with males vealed that P. recurva females lived signiÞcantly longer fed the Eucalyptus pollen diet livingsigniÞcantly than P. semipunctata females on all of the Þve diets longer than males fed diets containing Canadian pol- (Table 1). Male P. recurva lived signiÞcantly longer len or dogchow ( F ϭ 4.05, df ϭ 4, 127, P ϭ 0.004). The than P. semipunctata males on three of the Þve diets effect of cage was signiÞcant for both female (F ϭ 3.41, (sucrose solution alone, and sucrose solution plus ei- df ϭ 12, 293, P Ͻ 0.0001) and male (F ϭ 2.97, df ϭ 12, ther dogchow or Canadian pollen; Table 1). 115, P ϭ 0.0012) longevity. For both species, some Diet had a major effect on oviposition for both P. cages contained individuals that had relatively short recurva (Fig. 2) and P. semipunctata (Fig. 3), and on

Fig. 3. Mean numbers of eggs produced every8dbyP. semipunctata females fed on different diets. Eggs were collected every 2 d, but are shown in the Þgure as mean values calculated over8dtosimplify the graphical presentation. 374 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 96, no. 2

wide, respectively. Although the eggs from females on the Eucalyptus pollen diet were signiÞcantly longer (F ϭ 81.81, df ϭ 4, 2730, P Ͻ 0.0001) and wider (F ϭ 41.44, df ϭ 4, 2730, P Ͻ 0.0001) than the other eggs, the relatively small variations in egg size with diet indi- cates that egg size was affected only slightly by diet. Similar results were obtained with P. semipunctata eggs (Table 2), which varied in mean length from 2.34 to 2.49 mm, and in width from 0.67 to 0.71 mm. The three pollen diets resulted in eggs that were signiÞ- cantly larger than those produced by females on the dog chow or sugar water diets (length, F ϭ 25.24, df ϭ Fig. 4. Mean (ϮSE) lifetime fecundities of P. recurva and 4, 1,710, P Ͻ 0.0001; width F ϭ 14.77, df ϭ 4, 1,710, P Ͻ P. semipunctata fed on different diets. Data for each species 0.0001), but within the three pollen diets, eggs pro- were analyzed separately by ANOVA, and differences be- duced by females on the Eucalyptus pollen diet were ␣ ϭ tween means were determined with TukeyÕs tests ( 0.05). similar in width to those produced by females fed Bars surmounted by different letters are signiÞcantly differ- southwestern desert pollen and similar in length to ent. those produced by females fed Canadian pollen. The percentages of egg hatch were uniformly high the total lifetime fecundities per female (Fig. 4). Fe- on all diets for both species (Table 3), and were male P. recurva laid more than four times as many eggs minimally affected by diet. For P. recurva, mean egg when fed the Eucalyptus pollen diet throughout their hatch rates ranged from 87 to 93% on the Þve diets lives than when fed any of the other diets (F ϭ 67.00, (F ϭ 2.57, df ϭ 2, 287, P ϭ 0.038). The mean hatch rate df ϭ 4, 54, P Ͻ 0.0001). The total fecundities of females (93%) for P. recurva eggs from females fed on the maintained on any of the other four diets were not Eucalyptus pollen diet was signiÞcantly higher than signiÞcantly different. Analogous results were ob- the hatch rate of eggs from the southwestern desert tained with P. semipunctata; females fed the Eucalyp- pollen treatment (87%), but there were no other sig- tus pollen diet produced three to Þve times more eggs niÞcant differences between these two treatments and than females fed any of the other diets (F ϭ 25.48, df ϭ the other three treatments. For P. semipunctata, mean 4, 58, P Ͻ 0.0001). There were no signiÞcant differ- egg hatch rates ranged from 86 to 91% on the Þve diets ences in total fecundities of females maintained on any (Table 3), with no signiÞcant differences between any of the other four diets. In comparisons between spe- of the treatments (F ϭ 0.78, df ϭ 4, 150, P ϭ 0.54). cies fed a speciÞc diet, P. recurva females laid signif- icantly more eggs than P. semipunctata females when Discussion fed on the Eucalyptus pollen (F ϭ 19.53, df ϭ 1, 21, P ϭ 0.0002), the southwestern desert pollen (F ϭ 7.61, df ϭ There is abundant evidence that nectar and pollen 1, 22, P ϭ 0.011), the Canadian pollen (F ϭ 14.97, df ϭ constitute a major portion of the diet of adult Phora- 1, 22, P ϭ 0.0008), and the sucrose solution diet (F ϭ cantha (Chararas 1969b, Scriven et al. 1986, Hanks et 6.86, df ϭ 1, 23, P ϭ 0.015). al. 1993a). Adult Phoracantha have been observed For both species, egg size was relatively insensitive feedingat Eucalyptus ßower clusters without causing to diet (Table 2). The mean lengths and widths of eggs damage to the inßorescences, indicating that they laid by female P. recurva on the different diets varied were feedingon nectar and/or pollen rather than the from 2.42 to 2.57 mm longand from 0.58 to 0.61 mm ßower tissues. Examination of the frass of Þeld-col-

Table 2. Mean egg sizes for eggs laid by Phoracantha recurva and P. semipunctata females fed on different diets. Within each species, egg size data were analyzed by ANOVA, and means were separated with Tukey’s test. Values followed by different letters are significantly (0.05 ؍ different from each other (P

Width (mm) Length (mm) Dieta N Mean Ϯ SE Range Mean Ϯ SE Range P. recurva Eucalypt pollen 1067 0.61 Ϯ 0.001a 0.47Ð0.73 2.57 Ϯ 0.006a 1.83Ð3.23 Desert pollen 501 0.58 Ϯ 0.002b 0.47Ð0.70 2.42 Ϯ 0.008b,c 2.00Ð3.00 Canadian pollen 518 0.58 Ϯ 0.002b 0.50Ð0.70 2.45 Ϯ 0.008b,c 1.97Ð3.00 Dogchow 287 0.58 Ϯ 0.003b 0.47Ð0.70 2.46 Ϯ 0.011b 2.00Ð2.97 Sucrose solution only 362 0.58 Ϯ 0.003b 0.47Ð0.70 2.42 Ϯ 0.011c 2.00Ð3.17 P. semipunctata Eucalypt pollen 673 0.69 Ϯ 0.002b 0.50Ð0.83 2.49 Ϯ 0.006a 2.00Ð2.90 Desert pollen 366 0.69 Ϯ 0.002b 0.57Ð0.87 2.39 Ϯ 0.008b,c 2.00Ð2.87 Canadian pollen 161 0.71 Ϯ 0.004a 0.57Ð0.87 2.42 Ϯ 0.016a,b 2.00Ð3.00 Dogchow 249 0.68 Ϯ 0.003c 0.57Ð0.83 2.36 Ϯ 0.012c,d 2.00Ð2.83 Sucrose solution only 266 0.67 Ϯ 0.003c 0.50Ð0.83 2.34 Ϯ 0.011d 1.83Ð2.73

a All diets contained sucrose solution. April 2003 MILLAR ET AL.: EFFECTS OF Eucalyptus POLLEN 375

Table 3. Effect of diet on percent egg hatch for eggs laid by Phoracantha recurva and P. semipunctata females. Within each species, egg hatch data were analyzed by ANOVA, and means were separated with Tukey’s tests. Values followed by different letters are significantly (0.05 ؍ different from each other (P

P. recurva P. semipunctata Dieta Nb Percent hatch Range Nb Percent hatch Range Eucalypt pollen 108 93 Ϯ 1a 60Ð100 61 91 Ϯ 2 43Ð100 Desert pollen 51 87 Ϯ 3b 9Ð100 26 91 Ϯ 2 64Ð100 Canadian pollen 50 93 Ϯ 1a,b 63Ð100 11 91 Ϯ 4 64Ð100 Dogchow 43 91 Ϯ 2a,b 57Ð100 24 86 Ϯ 4 25Ð100 Sucrose solution only 40 93 Ϯ 1a,b 60Ð100 33 90 Ϯ 2 70Ð100

a All diets also contained sucrose solution. b The total number of egg masses that were sampled for each species and diet. lected beetles determined that it was largely com- Koveos 2000), whereas generalists tend to be more posed of the exines of Eucalyptus pollen, clearly in- catholic in their acceptance and utilization of pollen dicatingthat it constitutes a major portion of the diet from a variety of sources (Kirk 1985). However, the of adult beetles (Hanks et al. 1993a, 1998). Further- exact mechanisms that are involved in assessingand more, although P. semipunctata adults in a laboratory choosingpollens as nutritional sources are not well colony were observed to feed on pollen collected by understood. There is evidence to suggest that honey bees (mostly from plants of the Asteraceae), use pollen odors to locate pollen (Dobson 1994, Dob- they apparently were unable to assimilate the nutri- son and Bergstrom 2000), and speciÞc pollens also ents efÞciently because a diet of honey bee pollen and have been shown to contain feedingstimulants (Lin sucrose solution did not improve survivorship over and Mullin 1999, Hollister and Mullin 1999, Dobson sucrose solution alone (Hanks et al. 1993a). and Bergstrom 2000). However, the relative impor- In the study described here, the Eucalyptus pollen tance of olfactory versus gustatory cues for pollen diet clearly increased longevity and fecundity of P. feedinginsects, and the interplay between the two recurva and P. semipunctata in comparison to the other types of cues, remains largely unknown (Lin and Mul- diets. It is unlikely that the effects were a result of lin 1999). differences in the ability of the beetles to locate the The improved performances of the Phoracantha various pollens or the dogchow granulesusingolfac- tory cues because the beetles were conÞned in small beetles when fed Eucalyptus pollen are most easily cages, so that they inevitably contacted the food understandable when placed in the context of their sources while movingaround the cages.Conse- native habitat, Australia, where eucalypts dominate quently, the effects of diet were most likely a result of the temperate and subtropical forest. In California and differences in feedingstimulation or nutritional qual- other areas into which both Eucalyptus and Phoracan- ity, or both. Either factor would be predicted to in- tha beetles have been introduced, the close associa- crease fecundity, the Þrst by increasingthe intake of tion between the beetles and their host trees has nutrients for assimilation (assumingthat each of the persisted. The beetles apparently have not expanded Þve diets would have some nutritional value), and the their range of acceptable hosts in areas where they second by providingnutrients that were assimilated have been introduced (Hanks et al. 1995b), despite most efÞciently into eggs. What was not expected was the fact that Eucalyptus trees represent a scarcer re- that the differences between the different pollen source in these novel countries. types would be so dramatic. Our discovery that Eucalyptus pollen greatly in- Pollen feedingis common in invertebrates, with creases the fecundity and longevity of Phoracantha pollen providingamino acids and proteins, lipids, car- beetles has increased the efÞciency of our mass rear- bohydrates, sterols, vitamins, and minerals (Dobson ingof both eggandlarval parasitoids of the beetles. and Peng1997, Roulston and Cane 2000). Indeed, The approximate quadrupling of the number of eggs many insects require pollen feedingfor survival and obtained from each female beetle has enabled us to maturation of eggs. For example, pollen feeding has rear and release Ͼ200,000 egg parasitoids and 10,000 been shown to inßuence both the longevity and life- larval parasitoids annually, with minimal increase in time fecundity of insects as diverse as Heliconius but- the labor or materials required. terßies (Boggs 1987), weevils (Dobson 1994), and thrips (Kirk 1985, Leskey et al. 1997). However, pol- lens may vary widely in their suitability and accept- ability as sources of nutrients. In particular, some in- Acknowledgments sects and mites that specialize on only a few host plant This work was supported in part by grant #99-35302Ð8188 species undergo optimal development and reproduc- from the United States Department of Agriculture. We grate- tion on pollen from their speciÞc hosts (Kirk 1985, fully acknowledge the assistance of California State Univer- Grout and Richards 1992, James and Whitney 1993, sity at San Diego Santa Margarita Ecological Reserve, Te- Yue et al. 1994, Bruce-Oliver et al. 1996, Leskey et al. mecula, CA, in providing Eucalyptus trees for Phoracantha 1997, Grafton-Cardwell et al. 1999, Broufas and breedingmaterial. 376 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 96, no. 2

References Cited the Eucalyptus longhorned borer (Coleoptera: Ceram- bycidae). Ann. Entomol. Soc. Am. 88: 666Ð671. Boggs, C. L. 1987. Ecology of nectar and pollen feeding in Hanks, L. M., J. G. Millar, and T. D. Paine. 1995b. Biological lepidopteran species, pp. 369Ð391. In F. Slansky Jr. and constraints on host-range expansion by the wood-boring J. G. Rodriguez [eds.], Nutritional ecology of insects beetle Phoracantha semipunctata (Coleoptera: Ceramby- mites, spiders, and related invertebrates, Wiley, NY. cidae). Ann. Entomol. Soc. Am. 88: 183Ð188. Broufas, G. D., and D. S. Koveos. 2000. Effect of different Hanks, L. M., T. D. Paine, and J. G. Millar. 1996. Tiny wasp pollens on development, survivorship, and reproduction helps protect eucalypts from eucalyptus longhorned of Euseius finlandicus (Acari: Phytoseiidae). Environ. En- borer. Calif. Agric. 50 (MayÐJune): 14Ð16. tomol. 29: 743Ð749. Hanks, L. M., J. G. Millar, and T. D. Paine. 1998. Dispersal Bruce-Oliver, S. J., M. A. Hoy, and J. S. Yaninek. 1996. Effect of the Eucalyptus longhorned borer (Coleoptera: Ceram- of some food sources associated with cassava in Africa on bycidae) in urban landscapes. Environ. Entomol. 27: the development, fecundity, and longevity of Euseius 1418Ð1424. fustis (Acari: Phytoseiidae). Exp. Appl. Acarol. 20: 73Ð85. Chararas, C. 1969a. E´ tude biologique de Phoracantha semi- Hollister, B., and C. A. Mullin. 1999. Isolation and identi- punctata F. (Cole´opte`re Cerambycidae xylophage) Þcation of primary metabolite feedingstimulants for adult speciÞque des Eucalyptus en Tunisie et recherches sur la western corn rootworm, Diabrotica virgifera virgifera Le- vitalite´ et lÕadaptation de ces essences. Comptes Rendus cont, from host pollens. J. Chem. Ecol. 25: 1263Ð1280. Acad. Agric. France 55: 47Ð57. James, D. G., and J. Whitney. 1993. Cumbungi pollen as a Chararas, C. 1969b. Biologie et e´cologie de Phoracantha laboratory diet for Amblyseius victoriensis (Womersley) semipunctata F. (Cole´opte`re Cerambycidae xylophage) and Typhlodromus doreenae Schicha (Acari: Phytosei- ravageur des Eucalyptus en Tunisie, et methodes de pro- idae). J. Austral. Entomol. Soc. 32: 5Ð6. tection des peuplements. Annales de lÕInstitut National Kirk, W.D.J. 1985. Pollen feedingand the host speciÞcity de Recherches Forestieres de Tunie 2: 1Ð37. and fecundity of ßower thrips (Thysanoptera). Ecol. En- Dobson, H.E.M. 1994. Floral volatiles in biology, pp. tomol. 10: 281Ð290. 47Ð81. In E. A. Bernays (ed.), Insect-Plant Interactions, Leskey, T. C., D. A. Teulon, and E. A. Cameron. 1997. Ef- vol. V. CRC, Boca Raton, FL. fects of temperature and sugar maple pollen on oviposi- Dobson, H.E.M., and G. Bergstrom. 2000. The ecology and tion and longevity of pear thrips (Thysanoptera: Thripi- evolution of pollen odors. Plant Syst. Evol. 222: 63Ð87. dae). Environ. Entomol. 26: 566Ð571. Dobson, H.E.M., and Y.-S. Peng. 1997. Digestion of pollen Lin, S., and C. A. Mullin. 1999. Lipid, polyamide, and ßa- components by larvae of the ßower-specialist bee Che- vonol phagostimulants for adult western corn rootworm lostoma florisomne (Hymenoptera: Megachilidae). J. In- from sunßower (Helianthus annuus L.) pollen. J. Agric. sect Physiol. 43: 89Ð100. Food Chem. 47: 1223Ð1229. Drinkwater, T. W. 1975. The present pest status of euca- Longo, S., V. Palmeri, and D. Sommariva. 1993. SullÕattivita lyptus borers Phoracantha spp. in South Africa, pp. 119Ð di Avetianella longoi ooparassitoide di Phoracantha semi- 129. In Proceedings of the 1st Congress of the Entomology punctata nellÕItalia meridionale. Redia 76: 223Ð239. Society of South Africa. Entomology Society of South Mendel, Z. 1987. Major pests of man-made forests on Israel: Africa, Pretoria, South Africa. origin, biology, damage, and control. Phytoparasitica 15: Grafton-Cardwell, E. E., Y. Ouyang, and R. L. Bugg. 1999. 131Ð137. Leguminous cover crops to enhance population devel- Roulston, T. H., and J. H. Cane. 2000. Pollen nutritional opment in Euseius tularensis (Acari: Phytoseiidae). Biol. content and digestibility in . Plant Syst. Evol. 222: Control 16: 73Ð80. 187Ð209. Grout, T. G., and R. I. Richards. 1992. The dietary effect of SAS Institute. 1996. SAS userÕs guide for statistics, release windbreak pollens on longevity and fecundity of a pre- 6.13 edition. SAS Institute, Cary, NC. dacious mite Euseius addoensis addoensis (Acari: Phto- Scriven, G. T., E. L. Reeves, and R. F. Luck. 1986. Beetle seiidae) found in citrus orchards in South Africa. Bull. from Australia threatens Eucalyptus. Calif. Agric. 40 Entomol. Res. 82: 317Ð320. (JulyÐAugust): 4Ð6. Hanks, L. M. 1999. Inßuence of the larval host plant on Sokal, R. R. and F. J. Rohlf. 1995. Biometry, 3rd ed. W. H. reproductive strategies of cerambycid beetles. Annu. Freeman and Co., New York. Rev. Entomol. 44: 483Ð503. Wang, Q. 1998. The phoracanthine beetles (Coleoptera: Hanks, L. M., J. S. McElfresh, J. G. Millar, and T. D. Paine. Cerambycidae: Cerambycine): taxonomic overview and 1993a. Phoracantha semipunctata (Coleoptera: Ceramby- generic relationships. Invert. Taxon. 12: 667Ð684. cidae), a serious pest of Eucalyptus in California: biology Wang, Q., I.W.B. Thornton, and T. R. New. 1996. Biogeog- and laboratory-rearingprocedures. Ann. Entomol. Soc. raphy of the Phoracanthine beetles (Coleoptera: Ceram- Am. 86: 96Ð102. bycidae). J. Biogeog. 23: 75Ð94. Hanks, L. M., T. D. Paine, and J. G. Millar. 1993b. Host Yue, B., C. C. Childers, and A. H. Fouly. 1994. A comparison species preference and larval performance in the wood- of selected plant pollens for rearing Euseius mesembrinus boringbeetle Phoracantha semipunctata F. Oecologia (Acari: Phytoseiidae). Int. J. Acarol. 20: 103Ð108. (Berl.). 95: 22Ð29. Hanks, L. M., J. R. Gould, T. D. Paine, J. G. Millar, and Q. Wang. 1995a. Biology and host relations of Avetianella Received for publication 23 January 2002; accepted 2 Oc- longoi (Hymenoptera: Encyrtidae), an egg parasitoid of tober 2002.