NITROGEN BUDGETS OF PHLOEM-FEEDING BARK BEETLES WITH AND WITHOUT SYMBIOTIC FUNGI.
RICHARD T. WILKENS 
JONATHAN J. RUEL 
MARIA J. LOMBARDERO 
ERICH VALLERY 
Abstract. The nitrogen content of plant tissue is low relative to that of herbivores; as a consequence, dietary N can limit the growth and reproduction of herbivores and select for attributes that increase N acquisition. Bark beetles face a particularly severe challenge because the phloem that they consume is very low in nitrogen and phosphorus relative to their requirements. We quantified variation in the phloem concentrations of N and P in the host tree, Pinus taeda, and evaluated the following hypotheses regarding the role of symbiotic fungi in nutrient budgets of the herbivore Dendroctonus frontalis: D. frontalis experience variation in phloem nutrient concentrations across several spatial scales ([H.sub.1]); mycangial fungi enhance the diet of D. frontalis larvae by contributing to the acquisition of N and P ([H.sub.2]); Ophiostoma minus, an apparently antagonistic fungal symbiont, hinders D. frontalis larvae because it does not enhance nutrient concentrations of the phloem as much as mycangial fungi do ([H.sub.3]); and larvae of bark beetle species that lack mycangial fungi must consume more phloem to accomplish the same growth as larvae of D. frontalis ([H.sub.4]). In addition, we developed a general model for the N budgets of herbivorous insects that identifies the possible combinations of dietary and physiological parameters that can allow developmental success on low-nutrient diets.
Spatial variation in phloem N was mostly at the level of trees within sites (a scale of meters) while P mostly varied among sites (a scale of kilometers). Trees with higher N content produced larger D. frontalis adults. Prior to infestation by beetles, phloem nutrient concentrations were very uniform within trees and very low relative to that of the bark beetles (N and P concentrations of D. frontalis adults were 28 and 8 times greater, respectively). During infestation, phloem nutrient concentrations increased overall and became highly variable within trees. Nitrogen concentrations increased from 0.40 [+ or -] 0.01% (mean [+ or -] 1 SE) in uninfested phloem to 0.86 [+ or -] 0.03% in the phloeni surrounding successfully developing D. frontalis larvae, which are typically associated with one or two species of mutualistic mycangial fungi. Nitrogen concentrations were intermediate in other microhabitats within infested trees, including regions with no adult colonization, with failed larval development, or colon ized by the antagonistic bluestain fungus O. minus. We parameterized a general nutrient-budget model for D. frontalis and a sympatric non-mycangial bark beetle, Ips grandicollis, which indicated that (1) mycangial fungi provide their benefits by concentrating dietary N for larvae; (2) O. minus may exert its antagonistic effects on D. frontalis larvae by failing to concentrate dietary N as much as mycangial fungi do; (3) non-mycangial bark beetles meet their N budgets through high consumption of unaltered, low-N phloem; and (4) larvae should easily meet their P requirements with any combination of consumption rate and development time that allows them to meet their N requirements. Alternative strategies for N acquisition may have general consequences for the population dynamics and community interactions of bark beetles.
Key words: bark beetles and dietary N; Dendroctonus frontalis; fungi, bluestain and mycangial; herbivory; Ips grandicollis; mutualism; mycangial fungi and N acquisition; nitrogen acquisition by herbivores; Ophiostoma minus; phosphorus; symbiotic fungi.
The nitrogen content of plant tissue is very low relative to that of herbivores. As a consequence, dietary nitrogen can limit the growth and reproduction of herbivores and select for attributes that increase nitrogen acquisition (Mattson 1980). Herbivores can potentially solve the problem of low dietary nitrogen by (1) increasing nitrogen-use efficiency, (2) increasing consumption rate, (3) protracting the duration of the feeding life stage, (4) adjusting life histories to exploit seasonal pulses in plant nutrient content, (5) feeding in tissue microsites of relatively high nitrogen content, and/or (6) manipulating the diet to increase nitrogen content (Forcella 1981, Tabashnik 1982, Robbins 1983, Ayres and MacLean 1987, Clancy et al. 1988, Clancy 1992, Slansky and Wheeler 1992, Slansky 1993, White 1993, Yang and Joern 1994, Trier and Mattson 1997). Phosphorus also occurs in lower concentrations in plants than in herbivores and is also a potentially limiting nutrient for herbivores (Ayres and MacLean 1987, P opp et al. 1989, Skinner and Cohen 1994). Bark beetles (Coleoptera: Scolytidae) face a particularly severe challenge because of the very low nutritional value of phloem (Scriber and Slansky 1981, Slansky and Scriber 1985). For example, nitrogen content of the phloem in healthy loblolly pine is [sim]0.38% (Hodges and Lorio 1969) compared to 1-5% in plant foliage (Mattson 1980). Because insects are typically 6-10% nitrogen themselves, growing bark beetles must therefore concentrate dietary nitrogen by 16-26 fold.
Some bark beetles, including Ips grandicollis (Eichhoff), appear to meet their nutritional needs with high consumption rates. Other bark beetles, including the southern pine beetle (Dendroctonus frontalis Zimmer-man), appear to have relatively low consumption rates, but are thought to benefit from associations with fungi and other microorganisms that improve the nutritional quality of ingested phloem (Henry 1962, Becker 1971, Barras and Perry 1975, Martin 1979). Phloem nitrogen in one loblolly pine increased to 131% of control levels following the introduction of D. frontalis and associated microorganisms (Hodges et al. 1968). Adult female D. frontalis have elaborately modified structures (mycangium) that house and nurture two species of mutualistic fungi: Ceratocystiopsis ranaculosus Perry and Bridges and Entomocorticium sp. A (formerly SJB 122) (Barras and Perry 1972, Hsiau 1996). As the adults construct oviposition galleries, the fungal hyphae are exuded from the mycangium and begin to grow within the phl oem tissue as D. frontalis progeny are hatching. Larvae subsequently feed upon the phloem-fungal complex. D. frontalis adults that lacked mycangial fungi laid half as many eggs as control beetles (Goldhammer et al. 1990) and D.frontalis larvae feeding in the absence of mycangial fungi had reduced survival and increased development time (Barras 1973, estimated potential population growth rate, r, to be 0.026 vs. 0.057 beetles.[beetle.sup.-1].[d.sup.-1] in no-fungi and control populations, respectively).
The nutritional challenge for bark beetles can be compounded by intraspecific and interspecific competition for phloem. Interspecific competitors include other beetles (chiefly Scolytidae, Buprestidae, and Cerambycidae) and phoretic microorganisms. D. frontalis is not only associated with two species of apparently beneficial mycangial fungi but also with numerous other microorganisms such as the bluestain fungus (Ophiostoma minus H. & P. Sydow). Some bluestain fungi are thought to be mutualistic with some bark beetle species (Lewinsohn et al. 1994, Paine et al. 1997, Ross and Solheim 1997), but accumulating evidence suggests that O. minus is an antagonist of D. frontalis. There are strong negative correlations between the abundance of O. minus and the reproductive success of D. frontalis across infestation sites within a region, across trees within infestations, and across phloem patches within individual trees (Bridges 1983, Bridges 1985, Bridges et al. 1985, M. P. Ayres, personal observations). Experimenta l additions of O. minus in the laboratory reduce the growth performance of D. frontalis larvae (Barras 1970, Bridges and Perry 1985). The mechanisms underlying antagonistic effects of O. minus are not clear. It is possible that O. minus impedes beetle growth because it does not elevate nutrients as much as mycangial fungi do. Field and laboratory studies suggest that O. minus strongly outcompetes the mutualistic mycangial fungi (Bridges and Perry 1985, Kiepzig and Wilkens 1997), indicating that the presence of O. minus may preclude the establishment of mycangial fungi in larval feeding chambers. It is also possible that secondary metabolites produced by O. minus (e.g., isocoumarins, Hemingway et al. 1977) have allelopathic effects on mycangial fungi or D. frontalis larvae.
In this study we evaluated the following hypotheses: ([H.sub.1]) D. frontalis experience variation in phloem nutrient concentrations across many scales (within the bole of individual trees, among trees within infestations, and among infestation sites); ([H.sub.2]) mycangial fungi enhance the diet of D. frontalis by contributing to their intake of nitrogen and phosphorus; ([H.sub.3]) O. minus negatively impacts D. frontalis larval development because it does not enhance the nutrient concentrations of the phloem to the same degree as the mycangial fungi do; ([H.sub.4]) larvae of I. grandicollis, which lack mycangial fungi, must consume more phloem to attain the same mass as D. frontalis larvae. In addition, we develop a general model for the nitrogen budgets of herbivorous insects that identifies the possible combinations of dietary and physiological parameters that can allow developmental success on low-nutrient diets.
Phloem nitrogen and phosphorus
Phloem samples were collected from five natural infestations of Dendroctonus frontalis within the Kisatchie National Forest of Louisiana and the Indian Mounds wilderness area of Texas (USA). Infestation sites were separated by 10-200 km. Within each infestation site, phloem samples were collected from five infested trees and five uninfested trees (all loblolly pine, Pinus taeda L., 15-30 cm in diameter at breast height). The infested trees that we sampled were all at the same stage of attack (most D. frontalis were late in the final larval stadium). The uninfested trees that we sampled within each site were within 40 m of the infested trees but were just outside the perimeter of the infestation at the time of sampling. Within a site, infested and uninfested trees appeared to be comparable in all respects except for the presence or absence of D. frontalis. Samples from each infested tree were drawn from each of four mutually exclusive phloem classes (Fig. 1): (1) within 7 mm of adult galleries with successful brood (= good brood); (2) within 7 mm of adult galleries with unsuccessful brood (= failed brood); (3) separated from adult and larval galleries by [greater than]10 mm (= no brood); and (4) covered with fruiting bodies (perithecia) of O. minus (= bluestain). Phloem samples from the good brood, failed brood, and sometimes from the bluestain classes included the tissue that D. frontalis larvae were consuming. Good brood was defined by the presence of well-developed, oval, feeding galleries, while failed brood was defined by narrow, meandering feeding galleries that typically indicate larval mortality. We never observed successful larval development in regions with O. minus perithecia. Two replicate tissue samples from each phloem class in each tree were drawn from regions of phloem separated by 40-150 cm. Similarly, two replicate samples were drawn from each of the unattacked trees, but in this case there was only one phloem class (unattacked healthy phloem). Thus, each infestation site yielded a possible tota l of 50 phloem samples (40 from attacked trees and 10 from unattacked trees). In a few trees, not all possible phloem classes were represented (3 of 25 trees lacked O. minus samples and 5 of 25 trees lacked "no brood" samples). In 10 other trees, we could only obtain one sample from one of the phloem classes. In total, we analyzed 224 phloem samples.
Phloem samples were freeze-dried, ground, then analyzed for total nitrogen and phosphorus with a standard micro-Kjeldahl procedure (sulfuric acid digestion followed by analysis with a Technicon Auto-Analyzer [Pulse Instrumentation, Ltd., Saskatoon, Saskatchewan, Canada]). We also analyzed D. frontalis larvae (feeding final instars) and adults ([sim]2 d after eclosion) that had been feeding upon the phloem. Plant tissue of known nitrogen concentration (National Institute of Standards and Technology, United States Department of Commerce) was included with each digestion lot to ensure quality control. Because results showed a surprising increase in the nitrogen concentration of adults relative to larvae, we made additional measurements on larvae, pupae, and adults removed from their feeding chambers (larvae) or pupation chambers (pupae and callow adults). Subsets of the callow adults were allowed to mature prior to measurements for 2 or 4 d within humidified petri dishes at [sim]24[degrees]C with no food (adult s do not normally feed between emergence from one tree and attack of another tree).
Beetle success as a function of phloem nutrients
We evaluated patterns of variation in the size of adult beetles emerging from different trees and different infestation sites. D. frontalis adults from each of the attacked trees were collected as they emerged. A subsample of these beetles (8-10 of each sex from each tree) were measured for total length using an ocular micrometer ([n.sub.total] = 5 infestation sites X 5 trees X 2 sexes X 8-10 beetles = 446 total adults sampled).
We used an ANOVA to test for variation in phloem N and P content among infestation sites, trees nested within infestation sites, and (in the case of attacked trees) phloem classes within trees. Infestation sites and trees were treated as random effects, while phloem classes were treated as a fixed effect. We estimated the variance attributable to random effects (Sokal and Rohlf 1981). Infested and uninfested trees were analyzed separately because the phloem classes only applied to infested trees. A similar ANOVA model was used to evaluate variation in adult beetle size attributable to infestations, trees nested within infestations, and beetle sex (sex treated as a fixed effect). Correlation analyses tested for associations across trees between the average size of emerging beetles and the nitrogen and phosphorus content of phloem from the tree in which each developed; these analyses used nitrogen and phosphorus concentrations from the "good brood" phloem class because by definition most emerging beetles fed o n phloem of this class.
Consumption and nitrogen-use efficiency
We estimated the consumption of phloem by D. frontalis larvae by measuring the size of successful larval galleries within five trees at each of two natural infestations within unthinned loblolly pine stands in the Homochitto National Forest of Mississippi ([n.sub.total] = 100). We traced the galleries on clear mylar and calculated the area of phloem consumed using a video capture system. Some simpler measurements (with calipers) of D. frontalis gallery size in infestations within Lousiana, East Texas, and Alabama (USA) verified that there was no conspicuous variation in gallery size among forests. Virtually every successful D. frontalis gallery is in the range of 3 X 5 mm to 4 X 7 mm (see Fig. 1). We also measured the larval galleries of Ips grandicollis (3-13 individual galleries from each of four loblolly pine logs encompassing a range of phloem thicknesses). I. grandicollis galleries are long (40-130 mm) and narrow ([less than]2 mm), not spherical like those of D. frontalis. So we measured initial and fin al gallery widths with digital calipers, measured the lengths with a digital planimeter after tracing the galleries on clear mylar, and estimated gallery size as the area of a trapezoid of these dimensions. For both species the mass of consumed phloem was calculated by multiplying the gallery area by phloem mass per unit area. Our primary estimate of phloem specific mass was based on two 1.25-[cm.sup.2] phloem samples collected from each I. grandicollis log during early larval development (logs: 29.8 [+ or -] 7.4 mg/[cm.sup.2] [mean [+ or -] 1 SD]). Because phloem specific mass can vary depending upon tree physiological status (Wilkens et al. 1997), and because we could not measure phloem mass in the same individual trees where we measured D. frontalis galleries, we performed sensitivity analyses to evaluate the effects of phloem mass on estimates of D. frontalis nitrogen budgets. For this, we used phloem measurements from a survey of 11 loblolly pine stands that were chosen to span the range of growth condit ions in our study area (12 trees/stand, 8 measurements/tree; M. P. Ayres, unpublished data).
We estimated nitrogen-use efficiency (NUE, percentage N consumed that is converted to insect tissue) for each beetle species using measurements of adult mass ([M.sub.adult]), neonate mass ([M.sub.neonate]), nitrogen concentration of adults ([N.sub.adult]), total consumption (TC) and dietary nitrogen concentration ([N.sub.diet]):
NUE = ([M.sub.adult] - [M.sub.neonate]) X [N.sub.adult]/TC X [N.sub.diet] X 100. (1)
For I. grandicollis, estimates of [M.sub.neonate] came from drying and weighing 4-7 eggs from each of 3-4 oviposition galleries on each of two logs, and estimates of [M.sub.adult] and [N.sub.adult] came from measurements of 3-10 adults from each of four logs and two multi-bolt emergence traps. For D. frontalis, estimates of [M.sub.neonate] came from 5-10 eggs from each of three oviposition galleries in naturally infested trees, and estimates of [M.sub.adult] came from measurements of 446 adults that emerged from the bark samples collected for phloem nutrient analyses (lengths were converted to dry mass as [M.sub.dry] = (0.76 X length) - 1.45; Clarke et al. 1979). Eq. 1 assumes that the percentage nitrogen in eggs is similar to that of adults; this assumption has little consequence because the mass of eggs is so small that they contain very little N regardless of the concentration.
Modeling insect nitrogen budgets
Eq. 1 can be rearranged to show that the minimum total consumption required for insect development ([TC.sub.min]) is a function of adult mass ([M.sub.adult]), neonate mass ([M.sub.neonate]), adult nitrogen concentration ([N.sub.adult]), nitrogen-use efficiency (NUE) and dietary N concentration ([N.sub.diet]). Given a specified larval development time ([T.sub.dev]), [TC.sub.min] can be used to calculate the minimum relative consumption rate (RCR, in milligrams of biomass consumed per milligram of larva per day) as a function of minimum total consumption (Eq. 2), which itself is a function of dietary nitrogen given NUE (Eq. 1). Eq. 2 follows Gordon (1968) where [M.sub.exp] = mean exponential mass = ([M.sub.adult] - [M.sub.neonate])/ln([M.sub.adult]/[M.sub.neonate]). Alternatively, if RCR is specified, minimum development time ([T.sub.dev,min]) can be calculated as a function of minimum total consumption (Eq. 3) or dietary nitrogen given NUE (Eq. 1).
[RCR.sub.min] = [TC.sub.min]/[M.sub.exp].[T.sub.dev] (2)
[T.sub.dev,min] = [TC.sub.min]/(RCR/ln([M.sub.adult]/[M.sub.neonate]) X [M.sub.neonate])-(RCR/ln([M.sub.adult]/[M.sub.neonate]) X [M.sub.adult]). (3)
Calculations used mean values of adult mass, neonate mass, and adult N concentration reported in this study. Larval development time for D. frontalis larvae (egg hatch to pupation) ranges from [sim]17 to 40 d at 25[degrees]C (Barras 1973, Gagne et al. 1982, Wagner et al. 1984). Relative consumption rate and nitrogen-use efficiency have not been measured for D. frontalis larvae, but, based on measured values for other insects (and theoretical limits of 0 to 1 for efficiencies), probably fall within the range of 1.5-3.5 mg.[mg.sup.-1].[d.sup.-1] for RCR (at 25[degrees]C) and 30-90% for NUE (Scriber and Slansky 1981, Slansky and Scriber 1985).
Nitrogen concentration of fungi associated with D. frontalis
Replicated pure cultures of Entomocorticium, Ceratocystiopsis ranaculosus, and Ophiostoma minus were grown in a malt-extract medium (15 g dehydrated malt extract/L distilled water), in a shaking water bath (150 rpm), for 14 d at 25[degrees]C. The resulting fungal tissue was isolated by filtration (0.2-[micro]m pore size), lyophilized, and analyzed for nitrogen content.
Nitrogen and phosphorus concentrations of phloem
The N concentration of phloem around successfully developing Dendroctonus frontalis larvae was more than twice as high as that of the phloem in uninfested trees (0.86 [+ or -] 0.03% vs. 0.40 [+ or -] 0.01% [mean [+ or -]1 SE], respectively, P [less than] 0.0001; Fig. 2). The N concentration of phloem within infested trees that was associated with failed D. frontalis brood, bluestain, or no D. frontalis gallery was significantly higher than that of uninfested trees but significantly lower than that found in regions of good brood development (0.69-0.73%; Fig. 2, Phloem class in Table 1). Phosphorus concentrations in phloem averaged 6-7 fold less than N concentrations; the pattern across phloem classes matched that for nitrogen (Fig. 2, Table 1). Because D. frontalis attack virtually every pine tree along the advancing front of an outbreak (Thatcher et al. 1980), differences in N and P concentrations between infested and uninfested trees were almost certainly due to the introduction of microbes rather than to selection by the beetles for trees that were already high in phloem nutrients. Indeed, many of the trees included in our sample of uninfested phloem were subsequently infested by D. frontalis.
Within phloem classes, N concentrations varied considerably among trees within sites (Tables 1 and 2: Tree within infestation site). Among 18 infested trees, the N concentration in phloem with successful brood ranged from 0.52 to 1.08%. Variance among trees accounted for 76% of the total random variance in N concentration in infested trees and 60% in uninfested trees. The variance in N concentration was dramatically higher among infested trees than uninfested trees ([sigma] among trees = 0.196 vs. 0.047% N; [F.sub.13,20] = 17.36, P [less than] 0.001). The pattern of N concentrations among phloem classes within infested trees remained quite constant; although the phloem class X tree interaction was significant, it accounted for only 8% of the total random variance. There was no significant variation in phloem nitrogen attributable to infestation site (Tables 1 and 2).
In contrast to nitrogen, phloem concentrations of phosphorus varied markedly among infestation sites but relatively little (although still detectable) among trees within infestations (Tables 1 and 2). Variance among infestation sites accounted for 44% of the total random variance in P concentrations in infested trees and 74% in uninfested trees (6 among infestations = 0.0149 and 0.0203% phosphorus, respectively). Variance among trees within infestation sites accounted for 11% and 15% of the total random variance in P concentrations among infested and uninfested trees, respectively (Tables 1 and 2). As with N concentrations, the variance in P concentrations among infested trees was significantly greater than among uninfested trees ([sigma] = 0.0153 vs. 0.0057% phosphorus; [F.sub.13,20] = 7.20, P [less than] 0.001). The pattern of P concentrations among phloem classes within infested trees remained quite constant across infestation sites and trees within infestations (phloem class X infestation site interacti on was nonsignificant and phloem class X tree interaction, although statistically significant, accounted for only 16% of the total random variance). The variance among replicate phloem samples within trees (i.e., the error term in Tables 1 and 2) was also low, accounting for only 9-19% of the total random variance in N and P concentrations in infested and uninfested trees.
Concentrations of N and P in phloem seemed to vary independently of each other across uninfested trees, but were positively correlated across infested trees (Fig. 3). Among infested trees, there was a significant positive correlation between phloem nitrogen and phosphorus for all phloem classes except those without any D. frontalis gallery (r = 0.58, 0.58, and 0.62 for good brood, poor brood, and bluestain, respectively; r = 0.25 for no gallery).
Beetle size and phloem nutrition
Beetle size varied among trees within infestations ([F.sub.18,400] = 6.77, P [less than] 0.0001), and infested trees with higher N concentrations tended to produce larger beetles of both sexes (Fig. 4). Given the relationship between female adult size and fecundity (Clarke et al. 1979), the range in beetle size among trees translates into an 18% difference in fecundity (152 vs. 180 eggs/ female at 0.5% vs. 1.1% N, respectively). Beetle size did not vary between sexes ([F.sub.1,4] = 5.39, P = 0.081) or among infestation sites ([F.sub.4,18] = 2.49, P = 0.080). Phosphorus concentration was weakly related to male size ([r.sup.2] = 0.25, P 0.035) and unrelated to female size ([r.sup.2] = 0.15, P = 0.12). Multiple regressions that included both N and P did not provide a better fit than models that included only N.
N and P concentrations of bark beetles
The nitrogen content of Dendroctonus frontalis adults averaged 11.53 [+ or -] 0.13% (mean [+ or -] 1 SE n = 45 individuals), which is 13.4-fold higher than the concentration of phloem in regions of successful brood. The phosphorus content of D. frontalis adults averaged 0.749 [+ or -] 0.058% (n = 17 individuals), which is [sim]5.8-fold higher than the concentration of phloem in regions of successful brood. Males and females did not differ in their concentrations of N or P (P [greater than] 0.19). D. frontalis larvae had significantly lower concentrations of N than adults (8.13 [+ or -] 0.35% N [mean [+ or -] 1 SE], [t.sub.59] = 3.40, P [less than] 0.0001), but similar concentrations of P (0.672 [+ or -] 0.026% P, [t.sub.19] = 0.62, P = 0.54). The increase in N concentration from larvae to adults was due to decreases in total dry biomass (with no change in total N content) during developmental processes following adult eclosion (Fig. 5). Pupae, callow adults, and mature adults (2-4 d after eclosion) all contain ed an average of [sim]105 [micro]g of N, but the total biomass decreased from an average of 1.3 mg in pupae and callow adults to 0.96 mg two days after eclosion (N concentration: 7.8 [+ or -] 0.2% N [mean [+ or -] 1 SE] for pupae, 8.2 [+ or -] 0.3% N for callow adults, 11.7 [+ or -] 0.4% N for adult at 2 d, and 13.4 [+ or -] 0.3% N for adults at 4 d). This developmental change (also reported by Barras and Hodges ) does not affect the nitrogen-budget model so long as [M.sub.adult] and [N.sub.adult] in Eq. 1 were measured on the same developmental stage (the numerator of Eq. 1 equals [sim]105 [micro]g N in any case). Our N budgets were based on 2-d-old adults, which were estimated to be 11.49 [+ or -] 0.13% N in one sample (biomass: 0.93 [+ or -] 0.05 mg) and 11.67 [+ or -] 0.46% N in another sample (biomass: 0.95 [+ or -] 0.07 mg). These measurements indicate that the outer bark through which prepupae and adults bore before emergence is nutritionally inert. The N concentration of 2-d-old adult Ips. grand icollis was 8.61 [+ or -] 0.27% N, significantly less than that of D. frontalis ([t.sub.72] = 10.81, P [less than] 0.0001 for comparison between species).
Nitrogen concentration of fungi
Nitrogen concentrations of fungi were 8.7-11.6 fold higher than that of uncolonized phloem and the N concentration of Entomocorticium was significantly higher than that of the other two species: 4.62 [+ or -] 0.07% N, 3.86 [+ or -] 0.23% N, and 3.55 [+ or -] 0.07% N (mean [+ or -] 1 SE) for Entomocorticium, Ceratocystiopsis ranaculosus, and Ophiostoma minus, respectively ([F.sub.2,9] = 21.95, P = 0.0006).
N and P budgets
Although D. frontalis adults were larger than I. grandicollis adults, D. frontalis larvae appeared to acquire their mass by consuming only 21% as much phloem as I. grandicollis (5.7 vs. 27.4 mg; Table 3). Analysis of nitrogen budgets indicated that I. grandicollis could meet their physiological demands with a realistic nitrogen-use efficiency of 46% (Table 3). The requisite consumption rate for I. grandicollis to complete development in 25-35 d is 3.5-4.0 mg biomass.[(mg larva).sup.-1].[d.sup.-1] (Fig. 6). This consumption rate is in the upper range of those reported for insects (Scriber and Slansky 1981, Slansky and Scriber 1985) and this development time is somewhat longer than the 20-30 d that appeared to be required for our colony, but in general it seems easy to reconcile the N budgets of I. grandicollis with an empirically and physiologically reasonable parameterization of Eqs. 1-3.
In contrast, the nitrogen budgets of D. frontalis could not be immediately reconciled (note impossible nitrogen-use efficiency in Table 3; see also analyses of Mishra et al.  for a cerambycid beetle). D. frontalis larvae appeared to accumulate over twice as much nitrogen as was present in the phloem that they consumed from their feeding chamber (NUE = 216%, Table 3). Because a nitrogen-use efficiency [greater than]100% is impossible, one or more of the estimates used in our initial parameterization of Eq. 1 must be incorrect. If NUE is 80%, D. frontalis N budgets could be explained if the actual dietary nitrogen content is 1.5-2.0% (vs. estimate of 0.86%) and the actual total consumption is 8.8-6.6 mg (vs. estimate of 5.7 mg). This scenario is consistent with the hypothesis that D. frontalis feed extensively on fungal hyphae that
grow within the feeding chamber and that import biomass and N into the feeding chamber. Any realistic budgets also require higher NUE for D. frontalis than the 46% estimated f or I. grandicollis. The N budgets for D. frontalis cannot be reconciled by any realistic adjustments of phloem specific mass. Even if phloem mass was 45 mg/[cm.sup.2], which is 50% higher than our best estimate, and 15% higher than the maximum from a survey of 11 stands, the calculated nitrogen-use efficiency is still 143%. We could not identify any plausible N budgets for D. frontalis that do not require higher dietary nitrogen content and higher biomass consumption than the estimates in Table 3.
What are the consequences of variation in dietary nitrogen for bark beetle larvae? If consumption rate and NUE are held to some constant upper limit (as expected if larvae are routinely consuming and assimilating at their physiological limits; Ayres and MacLean 1987), then larvae would be forced to compensate for reduced dietary N with increased larval development time (Figs. 6 and 7, lower panels). For example, if the dietary nitrogen of I. grandicollis is 0.8% N in one tree vs. 0.5% N in another tree, then the minimum development time increases from 23.8 to 38.1 d (with NUE = 46% and RCR = 3.7 mg.[mg.sup.-1].[d.sup.-1]; Fig. 6, lower panel). If the dietary nitrogen of D. frontalis larvae decreases from 1.5% to 0.5% (e.g., due to the absence of mycangial fungi) the minimum development time increases from 19.9 to 59.6 d (with NUE = 80% and RCR = 2.0 mg.[mg.sup.-1].[d.sup.-1]; Fig. 7, lower panel).
Similar calculations with phosphorus indicate that larvae should easily meet their P requirements with any combination of consumption rate and development time that allows them to meet their N requirements (because the factor by which insects must concentrate N is 2.3 times greater than the factor by which they must concentrate P). Phosphorus requirements of D. frontalis could theoretically be met even from the 5.7 mg of phloem that is consumed out of their feeding chamber (phosphorus-use efficiency calculated from Eq. 1 = 92%). If I. grandicollis have the same P concentration as D. frontalis, their P budgets could be met with a phosphorus-use efficiency of only 21%.
[H.sub.1]: Dendroctonus frontalis experience variation in phloem nutrient concentrations across many scales
D. frontalis experience significant variation in the nitrogen and phosphorus content of loblolly pine phloem (Fig. 2, Tables 1 and 2). However, the spatial scale of variance in these two dietary elements differed, For N, variance among trees within sites dominated the random sources of variation (a scale of meters), while for P, variance among infestation sites dominated the random variation (a scale of kilometers). There was great additional variation in both N and P content across phloem classes within infested trees (scale of centimeters). Our data provide strong correlative evidence for the importance of dietary nitrogen to D. frontalis larvae. Regions of phloem where larvae survived to pupate were those with the highest N content (Fig. 2) and trees that produced the largest adults were those with the highest N content (Fig. 4). For both bark beetle species, analysis of N budgets provides additional evidence that dietary N is likely to be limiting for larval development.
[H.sub.2]: Mycangial fungi enhance the diet of D. frontalis by increasing the concentrations of N and P
Elevation of dietary nitrogen provides a mechanism to explain the beneficial effects of mycangial fungi for D. frontalis. Phloem with the highest N content (Fig. 2) was in regions of infested trees with small oval larval feeding chambers that are, characteristically associated with thriving colonies of mycangial fungi (Barras and Perry 1972). We propose that the hyphae of mycangial fungi extract N from the phloem surrounding larval feeding chambers and concentrate it within the hyphae and conidia that grow into the feeding chamber (Barras and Perry 1972). When grown in culture, N concentrations in the hyphae of Entomocorticium are 4.6%, which is higher than the other fungal associates of D. frontalis, and very high relative to the 2.8% N that has been measured in soil-dwelling fungal species (also grown in culture, Laursen 1975). High N content may be an adaptation of Entomocorticium that is favored because of its benefits for D. frontalis, on which it depends. The high N content of Entomocorticium compared to the other mycangial associate, Ceratocystiopsis ranaculosus, may explain the higher rates of population growth, and higher lipid content, of D. frontalis populations with a high abundance of Entomocorticium relative to C. ranaculosus (Bridges 1983, Goldhammer et al. 1990, Coppedge et al. 1995). Also, transport of amino acids and proteins from the phloem to the feeding chamber could be facilitated by the multinucleate cellular morphology of Entomocorticium (Happ et al. 1975). This hypothesis to explain the elevation in phloem N predicts a depletion of N in the phloem tissue outside feeding chambers, especially when the mycangial fungi within the chamber is Entomocorticium rather than C. ranaculosus. It further predicts that the mid-guts of D. frontalis larvae contain significant amounts of fungal tissue; specifically, 2-3 mg of fungal tissue (at 4-3% N, respectively) added to 5.7 mg of phloem tissue at 0.86% N can reconcile the nitrogen budgets of D. frontalis larvae (see Fig. 7: ellipse of realized solutio ns). Additional tests of this hypothesis would benefit from protocols that isolate and identify the fungi occupying microsites within and around individual feeding galleries.
Alternative mechanisms to explain the elevation of N in colonized phloem include: (1) a relative increase due to depletion of carbon by microbial respiration, (2) a relative increase due to carbohydrate export from the phloem to support secondary metabolism in resin ducts during beetle attack, and (3) an increase due to activity of N-fixing bacteria. Alternative 1 must be true to some extent (Martin 1979) because there is considerable [CO.sub.2] evolution from microbially infested phloem (M. P. Ayres, unpublished data), but neither it, nor alternative 2, can easily explain the differences in N content between phloem infested with mycangial fungi and Ophiostoma minus. Alternative 3 has been suggested (Peklo and Satava 1949, Peklo 1968), but discredited because of the very low abundance and metabolic activity of nitrogen-fixing bacteria associated with D. frontalis (Bridges 1981). Assuming that actual dietary N is 1.5-2.0%, a nitrogen budget model (Eq. 3) with nitrogen-use efficiency (NUE) of 80% and relative consumption rate (RCR) of 2.0 mg biomass.[(mg larvae).sup.-1].[d.sup.-1] accurately predicts that the minimum development time for D. frontalis larvae in the presence of mycangial fungi would be [sim]18 d vs. [greater than]40 d in the absence of mycangial fungi (Fig. 7; Barras 1973, Wagner et al. 1984).
[H.sub.3]: Ophiostoma minus negatively impacts the larval development of D. frontalis
At the scale of our sampling (Fig. 1), phloem nitrogen concentrations were significantly lower in patches colonized by O. minus than in patches presumably dominated by mycangial fungi (Fig. 2). This difference in N content (0.86% vs. 0.70% N, respectively) would be enough to increase minimum development time of larvae from 34.7 d to 42.6 d (with RCR of 2.0 mg.[mg.sup.-1].[d.sup.-1] and NUE of 80%; Fig. 7). There are no apparent concentrations of O. minus tissue analogous to the feeding chambers of D. frontalis. Instead, O. minus appears to be very evenly dispersed within patches of [sim]1 [dm.sup.2] or more. D. frontalis larvae feeding in phloem infested with O. minus create meandering tunnels (and almost always die) rather than the small feeding chamber typical of successful larvae (Fig. 1). Thus the larvae in phloem infested with O. minus are feeding at a similar scale to our sampling and the dietary nitrogen of these larvae is probably close to the 0.70% that we measured. If, as we hypothesize, the dietar y N in bluestain vs. mycangial feeding sites is 0.70 vs. 1.5-2.0%, then the nutritional benefit of the mycangial fungi vs. O. minus is to more than halve larval development time from 42.6 d to [sim]20 d (with RCR of 2.0 mg.[mg.sup.-1].[d.sup.-1] and NUE of 80%; Fig. 7). We cannot reject the hypothesis that O. minus further impedes D. frontalis reproduction by allelopathic effects of isocoumarins (Hemingway et al. 1977) or other secondary metabolites. However, Yearian et al. (1972) found no negative effects of O. minus on larvae of Ips avulsus, I. calligraphus or I. grandicollis. In any case, O. minus appears to be an antagonist of D. frontalis, and the antagonism is strengthened by strong competitive inhibition of mycangial fungi by O. minus (Klepzig and Wilkens 1997). Consequently, any factors that favor colonization of phloem by O. minus will have a negative impact on the potential reproduction of D. frontalis. Factors that might affect O. minus abundance and growth include phloem chemistry, temperature, re lative humidity, and the abundance of Tarsonemus mites that are phoretic on D. frontalis adults and act as vectors of O. minus spores (Bridges and Moser 1986, Bridges 1987, Cook and Hain 1987, Lieutier and Yart 1989, Paine et al. 1997).
[H.sub.4]: Ips grandicollis, which lack mycangial fungi, must consume more phloem to attain the same mass as D. frontalis
D. frontalis and I. grandicollis employ different strategies to meet their nitrogen budgets. I. grandicollis consumes large quantities of low-N phloem, while D. frontalis modifies the diet by introducing mutualistic fungi and consumes much less of a relatively high-N diet. We hypothesize that these two species represent alternative nutritional strategies of bark beetles. The Scolytidae includes [sim]6000 species worldwide (1430 species in the Nearctic). Based on the morphological diversity of specialized anatomical structures that house and nurture fungal associates, close mutualistic relationships between scolytids and fungi have evolved at least 6 times (Wood 1982). Genetic studies of the fungal associates also indicate multiple evolutionary origins (Cassar and Blackwell 1996). It seems unlikely that nutritional strategies intermediate to that of D. frontalis and I. grandicollis would be viable because the high consumption rates necessary without mycangial fungi are incompatible with the small feeding cham bers needed to maintain physical associations with slow-growing mycangial fungi (Klepzig and Wilkens 1997). In this case, evolutionary transitions between strategies should be relatively rapid. Estimates of relative consumption provide one easily obtained measure for comparing the nutritional strategies of species; I. grandicollis larvae consume [sim]35.9 mg/mg compared to [sim]6.2 mg/mg for D. frontalis (Table 3). We predict that similar measurements of additional species will reveal a bimodal frequency distribution with the two peaks corresponding to alternative strategies. Sterilization studies allow a more direct, but technically challenging, means to test for effects of microbial associates on bark beetle development and reproduction (Barras 1973, Fox et al. 1992, Colineau and Lieutier 1994, Six and Paine 1998).
Population and community consequences of alternative nutritional strategies
The suite of traits that are correlated with mycangial vs. non-mycangial strategies may have general consequences for population dynamics and community interactions. For example, we predict that nonmycangial species (e.g., I. grandicollis) will generally be more sensitive than mycangial species (e.g., D. frontalis) to variation among trees in phloem nitrogen (because a unit change in dietary N has the greatest impact on larval nutrition at low dietary N; Figs. 6 and 7). Thus, one effect of mycangial associations may be a damping of ecological effects for the consumer of variation in host plants (Auerbach and Strong 1981).
While populations of mycangial beetle species might be less sensitive than their non-mycangial counterparts to exogenous forces, they may be more influenced by complex endogenous dynamics because they exist within a community matrix that involves numerous strong interactions and therefore a greater potential for delayed density dependence (Berryman 1979, Turchin 1991, Hanski and Henttonen 1996). In fact, D. frontalis populations tend to cycle with a period of [sim]8 years and an amplitude of [sim]50X, but the cause of the cyclicity is unknown (Turchin et al. 1991, Reeve et al. 1995). We hypothesize that these cycles are produced by delayed density dependence resulting from the positive effect of growing beetle populations on the transmission of O. minus, which has an inverse effect on beetle population growth because O. minus outcompetes the mycangial fungi on which beetles depend to meet their nitrogen budgets. Similar community interactions may be common in mycangial species. At least it seems that a dispr oportionate number of the most economically important bark beetle species are mycangial: e.g., D. ponderosae, D. jeffreyi, D. brevicomis, D. adjunctatus, Ips acuminatus, I. sexdentatus, and Scolytis ventralis (Francke-Grosmann 1963, Barras and Perry 1971, Livingston and Berryman 1972, Paine and Birch 1983, Berryman 1986, Liebhold et al. 1986, Berryman and Ferrell 1988, Raffa 1988, Levieux et al. 1991, Six and Paine 1996). In contrast, non-mycangial beetles exist within a weaker matrix of community interactions and should be less likely to experience endogenous feedback from the rest of the community (Hochberg and Holt 1990, Wilson et al. 1996).
Mycangial species should generally have higher resource-use efficiency (which translates into lower relative consumption, Table 3). Bark beetles frequently experience strong intra- and inter-specific competition (Raffa and Berryman 1983, Miller 1984, Flamm et al. 1987, Rankin and Borden 1991, Zhang et al. 1992, Schlyter and Anderbrandt 1993, Gara et al. 1995, Reeve et al. 1998). When phloem resources are limiting, and other factors are equal, mycangial species should be able to sustain larger populations than non-mycangial species and be favored in interspecific competition (Tilman 1982). Also because of their higher resource-use efficiency, populations of mycangial species may be less likely to experience intense intraspecific competition (larger numbers of beetles could successfully complete development in the same tree bole), and therefore be less likely to experience instantaneous density dependence and stable population dynamics. Studies that characterize the N acquisition strategies of additional bark beetle species will allow tests for the population and community consequences of alternative nutritional strategies.
Bill Mattson generously assisted with the nitrogen and phosphorus analyses. John Reeve provided access to his Ips grandicollis. Stan Barras, Bob Bridges, Greg Eaton, Jane Hayes, Kier Klepzig, Pete Loria, John Moser, and John Reeve participated in many useful conversations. Mark Abrahamson, Bruce Ayres, Heather Govenor, K. Kleprig, Frank Siansky, Mac Strand, and two anonymous reviewers improved the manuscript. Research was supported by NRI CGP number 9835302 and the Southern Global Change Research Program of the U. S. Forest Service.
(1.) Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire 03755 USA
(2.) Department of Biology, Dowling College, Oakdale, New York 11769-1119 USA
(3.) Department of Biological Sciences, Northern Arizona University, Flagstaff Arizona 86011-5640 USA
(4.) Southern Research Station, USDA Forest Service, 2500 Shreveport Highway, Pineville, Louisiana 71360 USA
(5.) E-mail: Matthew.P.Ayres@Dartmouth.Edu
Auerbach, M. J., and D. R. Strong. 1981. Nutritional ecology of Heliconia herbivores: experiments with plant fertilization and alternative hosts. Ecological Monographs 51:63-84.
Ayres, M. P., and S. F MacLean, Jr. 1987. Development of birch leaves and the growth energetics of Epirrita autumnata (Geometridae). Ecology 68:558-568.
Barras, S. J. 1970. Antagonism between Dendroctonus frontalis and the fungus Ceratocystis minor. Annals of the Entomological Society of America 63:1187-1190.
Barras, S. J. 1973. Reduction of progeny and development in the southern pine beetle following removal of symbiotic fungi. Canadian Entomologist 105:1295-1299.
Barras, S. J., and J. D. Hodges. 1974. Weight, moisture, and lipid changes during life cycle of the southern pine beetle. USDA Forest Service Research Note SO-178.
Barras, S. J., and T. Perry. 1971. Gland cells and fungi associated with prothoracic mycangium of Dendroctonus adjunctus (Coleoptera: Scolytidae). Annals of the Entomological Society of America 64:123-126.
Barras, S. J., and T. Perry. 1972. Fungal symbionts in the prothoracic mycangium of Dendroctanus frontalis (Coleopt.: Scolytidae). Zeitschrift fur angewandte Entomologie 71:95-104.
Barras, S. J., and T. J. Perry. 1975. Interrelationships among microorganisms, bark or ambrosia beetles, and woody plant tissue: an annotated bibliography, 1965-1974. United States Department of Agriculture Forest Service, Southern Forest Experiment Station, General Technical Report SO-10.
Becker, G. 1971. Physiological influences on wood-destroying insects of wood compounds and substances produced by microorganisms. Wood Science and Technology 5:236-246.
Berryman, A. A. 1979. Towards a theory of insect epidemiology. Research on Population Ecology 19:181-196.
Berryman, A. A. 1986. Forest insects: principles and practice of population management. Plenum, New York, New York, USA.
Berryman, A. A., and G. T. Ferrell. 1988. The fir engraver beetle in western states. Pages 556-578 in A. A. Berryman, editor. Dynamics of forest insect populations: patterns, causes, and implications. Plenum, New York, New York, USA.
Bridges, J. R. 1981. Nitrogen-fixing bacteria associated with bark beetles. Microbial Ecology 7:131-137.
Bridges, J. R. 1983. Mycangial fungi of Dendroctonus frontalis (Coleoptera: Scolytidae) and their relationship to beetle population trends. Environmental Entomology 12:858-861.
Bridges, J. R. 1985. Relationship of symbiotic fungi to southern pine beetle population trends. Pages 127-135 in S. J. Branhan and R. C. Thatcher, editors. Integrated pest management research symposium: the proceedings. United States Department of Agriculture Forest Service, Southern Forest Experiment Station, General Technical Report SO-56.
Bridges, J. R. 1987. Effects of terpenoid compounds on growth of symbiotic fungi associated with the southern pine beetle. Phytopathology 77:83-85.
Bridges, J. R., and J. C. Moser. 1986. Relationship of phoretic mites (Acari: Tarsonemidae) to the bluestaining fungus, Ceratocystis minor, in trees infested by southern pine beetle (Coleoptera: Scolytidae). Environmental Entomology 15: 951-953.
Bridges, J. R., W. A. Nettleton, and M. D. Connor. 1985. Southern pine beetle (Coleoptera: Scolytidac) infestations without the bluestain fungus, Ceratocystis minor. Journal of Economic Entomology 78:325-327.
Bridges, J. R., and T. J. Perry. 1985. Effects of mycangial fungi on gallery construction and distribution of bluestain in southern pine beetle-infested pine bolts. Journal of Entomological Science 20:27 1-275.
Cassar, S., and M. Blackwell. 1996. Convergent origins of ambrosia fungi. Mycologia 88:596-601.
Clancy, K. M. 1992. Response of western spruce budworm (Lepidoptera: Tortricidae) to increased nitrogen in artificial diets. Environmental Entomology 21:33 1-344.
Clancy, K. M., M. R. Wagner, and R. W. Tinus. 1988. Variation in host foliage nutrient concentrations in relation to western spruce budworm herbivory. Canadian Journal of Forest Research 18:530-544.
Clarke, A. L., J. W. Webb, and R. T. Franklin. 1979. Fecundity of the southern pine beetle in laboratory pine bolts. Annals of the Entomological Society of America 72:229-321.
Colineau, B., and F. Lieutier. 1994. Production of Ophiostoma-free adults of Ips sexdentatus Boern. (Coleoptera: Scolytidae) and comparison with naturally contaminated adults. Canadian Entomologist 126:103-110.
Cook, S. P., and F. P. Ham. 1987. Four parameters of the wound response of loblolly and shortleaf pines to inoculation with the blue-staining fungus associated with the southern pine beetle. Canadian Journal of Botany 65:2403- 2409.
Coppedge, B. R., F. M. Stephen, and G. W. Felton. 1995. Variation in female southern pine beetle size and lipid content in relation to fungal associates. Canadian Entomologist 127:145-154.
Flamm, R. O., T. L. Wagner, S. P. Cook, P. E. Pulley, R. N. Coulson, and T. M. McArdles. 1987. Host colonization by cohabiting Dendroctonus frontalis, Ips avulsus, and I. calligraphus (Coleoptera: Scolytidae). Environmental Entomology 16:390-399.
Forcella, F. 1981. Twig nitrogen content and larval survival of twig girdling beetles Oncideres cingulata (Coleoptera: Cerambycidae). Coleopterists Bulletin 35:211-212.
Fox, J. W., D. L. Wood, R. P. Akers, and J. R. Parmeter. 1992. Survival and development of Ips paraconfusus Lanier (Coleoptera, Scolytidae) reared axenically and with tree-pathogenic fungi vectored by cohabiting Dendroctonus species. Canadian Entomologist 124:1157-1167.
Francke-Grosmann, H. 1963. Some new aspects in forest entomology. Annual Review of Entomology 8:415-438.
Gagne, J. A., T. L. Wagner, P. J. H. Sharpe, R. N. Coulson, and W. S. Fargo. 1982. Reemergence of Dendroctonus frontalis (Coleoptera: Scolytidae) at constant temperatures. Environmental Entomology 11:1216-1222.
Gara, R. I., R. A. Werner, M. C. Whitmore, and E. H. Holsten. 1995. Arthropod associates of the spruce beetle Dendroctonus rufipennis (Kirby) (Col., Scolytidae) in spruce stands of south-central and interior Alaska. Journal of Applied Entomology 119:585-590.
Goldhammer, D. S., F. M. Stephen, and T. D. Paine. 1990. The effect of the fungi Ceratocystis minor (Hedgecock) Hunt, Ceratocystis minor (Hedgecock) Hunt var. barrasii Taylor, and SJB 122 on reproduction of the southern pine beetle, Dendroctonus frontalis Zimmerman (Coleoptera: Scolytidae). Canadian Entomologist 122:407-418.
Gordon, G. T. 1968. Quantitative aspects of insect nutrition. American Zoologist 8:131-138.
Hanski, I., and H. Henttonen. 1996. Predation on competing rodent species: a simple explanation of complex patterns. Journal of Animal Ecology 65:220-232.
Happ, G.M., C.M. Happ, and S. J. Barras. 1975. Bark beetle--fungal symbiosis. III. Ultrastructure of conidiogenesis in a Sporothrix ectosymbiont of the southern pine beetle. Canadian Journal of Botany 53:2702-2711.
Hemingway. R. W., G. W. McGraw, and S. J. Barras. 1977. Polyphenols in Ceratocystis minor infected Pinus taeda: fungal metabolites, phloem and xylem phenols. Journal of Agricultural and Food Chemistry 25:717-722.
Henry, S. M. 1962. The significance of microorganisms in the nutrition of insects. Transactions of the New York Academy of Sciences, Series II 24:676-683.
Hochberg, M. E., and R. D. Holt. 1990. The coexistence of competing parasites. I. The role of cross-species infection. American Naturalist 136:517-541.
Hodges, J. D., S. J. Barras, and J. K. Mauldin. 1968. Amino acids in inner bark of loblolly pine, as affected by the southern pine beetle and associated microorganisms. Canadian Journal of Botany 46:1467-1472.
Hodges, J. E., and P. L. Lorio, Jr. 1969. Carbohydrate and nitrogen fractions of the inner bark of loblolly pines under moisture stress. Canadian Journal of Botany 47:1651-1657.
Hsiau, 0. T.-H. 1996. The taxonomy and phylogeny of the mycangial fungi from Dendroctonus brevicomis and Dendroctonus frontalis (Coleoptera: Scolytidae). Dissertation. Iowa State University, Ames, Iowa, USA.
Klepzig, K. D., and R. T. Wilkens. 1997. Competitive interactions among symbiotic fungi of the southern pine beetle. Applied and Environmental Microbiology 63:621-627.
Laursen, G. 1975. Higher fungi in soils of coastal arctic tundra communities. Dissertation. Virginia Polytechnic Institute and State University, Blacksburg, Virginia, USA.
Levieux, J., P. Cassier, D. Guillaumin, and A. Roques. 1991. Structures implicated in the transportation of pathogenic fungi by the European bark beetle Ips sexdentatus Boerner: ultrastructure of a mycangium. Canadian Entomologist 123:245-254.
Lewinsohn, D., E. Lewinsohn, C. L. Bertagnolli, and A. D. Patridge. 1994. Blue-stain fungi and their transport structures on the Douglas fir beetle. Canadian Journal of Forest Research 24:2275-2283.
Liebhold, A. M., P. Berck, N. A. Williams, and D. L. Wood. 1986. Estimating and valuing western pine beetle Dendroctonus brevicomis impacts. Forest Science 32:325-338.
Lieutier, F, and A. Yart. 1989. Temperature preference of the fungi associated with Ips sexdentatus Boern. and Tomicus piniperda L. Coleoptera Scolytidae. Annales des Sciences Forestieres (Paris) 46:411-415.
Livingston, R. L., and A. A. Berryman. 1972. Fungus transport structures in the fir engraver, Scolytus ventralis (Coleoptera: Scolytidae). Canadian Entomologist 104:1793-1800.
Martin, M. M. 1979. Biochemical implications of insect mycophagy. Biological Reviews 54:1-21.
Mattson, W. J. 1980. Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics 11:119-161.
Miller, M. C. 1984. Effect of exclusion of insect associates on Ips calligraphus Coleoptera Scolytidae brood emergence. Zeitschrift Fur Angewandte Entomologie 97:298-304.
Mishra, S. C., Sarma P. K. Sen, and R. Singh. 1985. Chemical changes in wood during the digestive process in larvae of Hoeplocerambyx spinicornis (Insecta: Coleoptera: Cerambycidae). Material und Organismen (Berlin) 20:53-64.
Paine, T. D., and M. C. Birch. 1983. Acquisition and maintenance of mycangial fungi by Dendroctonus brevicomis (Coleoptera Scolytidac). Environmental Entomology 12: 1384-1386.
Paine, T. D., K. F. Raffa, and T. C. Harrington. 1997. Interactions among scolytid bark beetles, their associated fungi, and live host conifers. Annual Review of Entomology 42: 179-206.
Peklo, J. 1968. Symbiosis of Azotobucter with insects. Nature 158:795-796.
Peklo, J., and J. Satava. 1949. Fixation of free nitrogen by bark beetles. Nature 163:336-337.
Popp, M. P., R. C. Wilkinson, E. J. Jokela, R. B. Harding, and T. W. Phillips. 1989. Effects of slash pine phloem nutrition on the reproductive performance of Ips calligraphus Coleoptera: Scolytidac. Environmental Entomology 18:795-799.
Raffa, K. F. 1988. The mountain pine beetle in western North America. Pages 506-530 in A. A. Berryman, editor. Dynamics of forest insect populations: patterns, causes, and implications. Plenum, New York, New York, USA.
Raffa, K. F., and A. A. Berryman. 1983. The role of host plant resistance in the colonization behavior and ecology of bark beetles (Coleoptera: Scolytidac). Ecological Monographs 53:27-49.
Rankin, L. J., and J. H. Borden. 1991. Competitive interactions between the mountain pine beetle and the pine engraver in lodgepole pine. Canadian Journal of Forest Research 21:1029-1036.
Reeve, J. R., M. P. Ayres, and P. L. Lorio, Jr. 1995. Host suitability, predation, and bark beetle population dynamics. Pages 339-357 in N. Cappuccino and P. W. Price, editors. Population dynamics: new approaches and synthesis. Academic Press, San Diego, California, USA.
Reeve, J. D., D. A. Rhodes, and P. Turchin. 1998. Scramble competition in southern pine beetle (Coleoptera: Scolytidae). Ecological Entomology 23:433-443.
Robbins, C. T 1983. Wildlife feeding and nutrition. Academic Press, New York, New York, USA.
Ross, D. W., and H. Solheim. 1997. Pathogenicity to Douglas fir of Ophiostoma pseudotsugae and Leptographium abietinum, fungi associated with the Douglas fir beetle. Canadian Journal of Forest Research 27:39-43.
Schlyter, F, and O. Anderbrant. 1993. Competition and niche separation between two bark beetles: existence and mechanisms. Oikos 68:437-447.
Scriber, J. M., and F. Slansky, Jr. 1981. The nutritional ecology of immature insects. Annual Review of Entomology 26:183-211.
Six, D. L., and T. D. Paine. 1996. Leptographium pyrinum is a mycangial fungus of Dendroctonus adjunctus. Mycologia 88:739-744.
Six, D. L., and T. D. Paine. 1998. Effects of mycangial fungi and host tree species on progeny survival and emergence of Dendroctonus ponderosae (Coleoptera: Scolytidac). Environmental Entomology 27:1393-1401.
Skinner, R. H., and A. C. Cohen. 1994. Phosphorus nutrition and leaf age effects on sweet potato whitefly (Homoptera: Aleyrodidae). Environmental Entomology 23:693-698.
Slansky, F. 1993. Nutritional ecology -- the fundamental quest for nutrients. Pages 29-91 in N. E. Stamp and T. M. Casey, editors. Caterpillars. Chapman & Hall, New York, New York, USA.
Slansky, F., Jr., and J. M. Scriber. 1985. Food consumption and utilization. Pages 87-163 in G. A. Kerkut and L. I. Gilbert, editors. Comprehensive insect physiology, biochemistry, and pharmocology. Pergamon, Oxford, UK.
Slansky, F., and G. S. Wheeler. 1992. Caterpillars compensatory feeding response to diluted nutrients leads to toxic allelochemical dose. Entomologia Experimentalis et Applicata 65:171-186.
Sokal, R. R., and F. J. Rohlf. 1981. Biometry. W. H. Freeman & Company, San Francisco, California, USA.
Tabashnik, B. E. 1982. Responses of pest and nonpest Colias butterfly larvae to intraspecific variation in leaf nitrogen and water content. Occologia 55:389-394.
Thatcher, R. C., J. L. Searcy, J. E. Coster, and G. D. Hertel. 1980. The southern pine beetle. Technical Bulletin 1631. USDA Forest Service, Combined Forest Pest Research and Development Program, Pineville, Laouisiana, USA.
Tilman, D. 1982. Resource competition and community structure. Monographs in population biology. Princeton University Press, Princeton, New Jersey, USA.
Trier, T. M., and W. J. Mattson. 1997. Needle mining by the spruce budworm provides sustenance in the midst of privation. Oikos 79:241-246.
Turchin, P. 1991. Nonlinear modeling of time series data: limit cycles and chaos in forest insects, voles and epidemics. Pages 39-62 in J. A. Logan and F. P. Ham, editors. Chaos and insect ecology. Virginia Polytechnic Institute and State University, Blacksburg, Virginia, USA.
Turchin, P., P. L., Lorio, Jr., A. D. Taylor, and R. F. Billings. 1991. Why do populations of southern pine beetles (Coleoptera: Scolytidae) fluctuate? Environmental Entomology 20:401-409.
Wagner, T. L., J. A. Gagne, P. J. H. Sharpe, and R. N. Coulson. 1984. A biophysical model of southern pine beetle, Dendroctonus frontalis Zimmermann (Coleoptera: Scolytidae), development. Ecological Modelling 21:125-147.
White, T. C. R. 1993. The inadequate environment: nitrogen and the abundance of animals. Springer-Verlag, New York, New York, USA.
Wilkens, R. T., M. P. Ayres, P. L. Lorio, Jr., and J. D. Hodges. 1997. Environmental effects on pine tree carbon budgets and resistance to bark beetles. Pages 591-616 in R. A. Mickler and S. Fox, editors. The productivity and sustainability of southern forest ecosystems in a changing environment. Springer-Verlag, New York, New York, USA.
Wilson, H. B., M. P. Hassell, and H. C. J. Godfray. 1996. Host-parasitoid food webs: dynamics, persistence, and invasion. American Naturalist 148:787-806.
Wood, S. L. 1982. The bark beetles and ambrosia beetles of North and Central America (Coleoptera Scolytidac), a taxonomic monograph. Great Basin Naturalist Memoirs 6:1-1359.
Yang, Y., and A. Joern. 1994. Compensatory feeding in response to food quality by Melanopus differentiallis. Physiological Entomology 19:75-82.
Yearian, W. C., R. J. Gouger, and R. C. Wilkinson. 1972. Effects of the bluestain fungus, Ceratocystis ips, on development of ips bark beetles in pine bolts. Annals of the Entomological Society of America 65:48 1-487.
Zhang, Q. H., J. A. Byers, and F. Schlyter. 1992. Optimal attack density in the larch bark beetle Ips cembrae (Coleoptera Scolytidae). Journal of Applied Ecology 29:672-678.
Results from ANOVAs of phloem and phosphorus concentration, for four phloem classes within loblolly pines frontalis (see Fig. 2). Nitrogen (%) Source of variation df MS X [10.sup.4] F Phloem class 3 2029 9.64 [**] Infestation site 4 627 0.82 Phloem class X Infestation site 21 210 1.76 Tree (Infestation site) 13 763 16.44 [***] Phloem class X Tree (Infestation site) 39 120 2.58 [***] Error 63 46 Phosphorus (%) Source of variation MS X [10.sup.6] F Phloem class 7250 11.88 [***] Infestation site 11092 20.46 [***] Phloem class X Infestation site 610 1.61 Tree (Infestation site) 542 5.38 [***] Phloem class X Tree (Infestation site) 379 3.76 [***] Error 101 Notes: The F-test denominator for Phloem class was [MS.sub.phloem class X Infestation site]; for Infestation site, it was [MS.sub. Tree(Infestation site)]; and for Phloem class X Infestation site, it was [MS.sub. Phloem class X Tree(Infestation site)]' Other were tested over [MS.sub.error]' (Table 2 shows corresponding analysis of uninfested trees.) (*.)P [less than] 0.05; (**.)P [less than] 0.01; (***.)P [less than] 0.001. Results from ANOVAs of phloem nitrogen and phosphorus concentration in healthy uninfested loblolly pines adjacent to those infested with Dendroctonus frontalis (see Fig. 2). Nitrogen (%) Phosphorus (%) Source of variation df MS X [10.sup.4] F MS X [10.sup.6] Infestation site 4 83 1.61 2211 Tree (infestation site) 20 52 7.58 [***] 109 Error 25 7 66 Source of variation F Infestation site 20.22 [***] Tree (infestation site) 2.51 Error Note: The F-test denominator for infestation was [MS.sub.Tree(Infestation)], and for Tree, it was [MS.sub.error]. (*.)P [less than] 0.05; (**.)P [less than] 0.01; (***.)P [less than] 0.001. Estimated consumption and nitrogen- use efficiency (and values used in the estimates) for larvae of Dendroctonus frontalis and Ips grandicollis. D. frontalis Variable Mean 1 SE n Egg mass (mg) 0.015 0.001 10 Adult mass (mg) 0.93 0.05 446 Adult N content (%) 11.56 0.13 45 Adult N content ([micro]g) 107 Area of feeding gallery ([mm.sup.2]) [+] 19.1 1.1 10 Specific mass of phloem (mg/[cm.sup.2]) [+] 29.8 Mass of consumed phloem (mg) [+] 5.7 Relative consumption (mg/mg) [++] 6.2 Nitrogen in consumed phloem (%) [+] 0.86 0.03 25 Nitrogen in consumed phloem ([micro]g) 49 Nitrogen-use efficiency, NUE (%) [ss] 216 I. grandicollis Variable Mean 1 SE n Egg mass (mg) 0.018 0.001 37 Adult mass (mg) 0.78 0.09 35 Adult N content (%) 8.61 0.27 29 Adult N content ([micro]g) 67 Area of feeding gallery ([mm.sup.2]) [+] 92.8 13.5 4 Specific mass of phloem (mg/[cm.sup.2]) [+] 29.8 3.9 4 Mass of consumed phloem (mg) [+] 27.4 5.0 4 Relative consumption (mg/mg) [++] 35.9 Nitrogen in consumed phloem (%) [+] 0.52 0.03 26 Nitrogen in consumed phloem ([micro]g) 142 Nitrogen-use efficiency, NUE (%) [ss] 46 (+.)n = number of trees (based on 5-10 replicate measurements per tree). (++.)Consumed phloem mass/(adult mass -- egg mass). (ss.)Calculated from Eq. 1. Note that the estimated NUE for I. grandicollis is reasonable, but that of D. frontalis is impossibly high.
|Printer friendly Cite/link Email Feedback|
|Author:||AYRES, MATTHEW P.; WILKENS, RICHARD T.; RUEL, JONATHAN J.; LOMBARDERO, MARIA J.; VALLERY, ERICH|
|Article Type:||Statistical Data Included|
|Date:||Aug 1, 2000|
|Previous Article:||CAN HOST-PARASITOID METAPOPULATIONS EXPLAIN SUCCESSFUL BIOLOGICAL CONTROL?|
|Next Article:||EFFECTS OF GRASSLAND FRAGMENTATION ON INSECT SPECIES LOSS, COLONIZATION, AND MOVEMENT PATTERNS.|