Quantitative ecological inventory of terra firme and várzea tropical forest on the Rio Xingu,Brazilian Amazon

Three hectares of Amazonian terra firme forest and an adjacent one-half hectare of várzea forest were quantitatively inventoried at O Deserto, on the Rio Xingu, Pará, Brazil. In the terra firme forest, 1420 individual trees greater than ten cm dbh, in 39 families and 265 species, were inventoried. In the várzea forest, there were 220 individual trees, in 17 families and 40 species.Cenostigma macrophyllum andOrbignya sp. were the most important species in the terra firme forest;Mollia lepidota andLeonia glycycarpa were most important in the várzea forest. Among one-hectare subplots of the total three-hectare terra firme sample, the number of trees ranged from 393 to 460, the number of families was a constant 33, and the number of species ranged from 118 to 162. This variation indicates that one-hectare samples are too small to be used to estimate the species richness of the total forest. The terra firme forest was richer in species and had a greater stature than the várzea forest.     

Primate assemblage structure in Amazonian flooded and unflooded forests

There is considerable variation in primate species richness across neotropical forest sites, and the richest assemblages are found in western Amazonia. Forest type is an important determinant of the patterns of platyrrhine primate diversity, abundance, and biomass. Here we present data on the assemblage structure of primates in adjacent unflooded (terra firme) and seasonally inundated (várzea and igapó) forests in the lower Purús region of central-western Brazilian Amazonia. A line-transect census of 2,026 km in terra firme, 2,309 km in várzea, and 277 km in igapó was conducted. Twelve primate species were recorded from 2,059 primate group sightings. Although terra firme was found to be consistently more species-rich than várzea, the aggregate primate density in terra firme forest was considerably lower than that in the species-poor várzea. Consequently, the total biomass estimate was much higher in várzea compared to either terra firme or igapó forest. Brown capuchin monkeys (Cebus apella) were the most abundant species in terra firme, but were outnumbered by squirrel monkeys (Saimiri cf. ustus) in the várzea. The results suggest that floodplain forest is a crucial complement to terra firme in terms of primate conservation in Amazonian forests.     

Vertebrate responses to fruit production in Amazonian flooded and unflooded forests

We document patterns of fruit and vertebrate abundance within an extensive, virtually undisturbed mosaic of seasonally flooded (várzea and igapó) and unflooded (terra firme) forests of central Amazonia. Using phenological surveys and a standardised series of line-transect censuses we investigate the spatial and temporal patterns of immature and mature fruit availability and how this may affect patterns of habitat use by vertebrates in the landscape. All habitats showed marked peaks in fruiting activity, and vertebrate detection rates varied over time for most species both within and between forest types. Many arboreal and terrestrial vertebrates used both types of flooded forest on a seasonal basis, and fluctuations in the abundance of terrestrial species in várzea forest were correlated with fruit availability. Similarly, the abundance of arboreal seed predators such as buffy saki monkeys (Pithecia albicans) and macaws (Ara spp.) were closely linked with immature fruit availability in terra firme forest. We conclude that highly heterogeneous landscapes consisting of terra firme, várzea and igapó forest appear to play an important role in the dynamics of many vertebrate species in lowland Amazonia, but the extent to which different forest types are used is highly variable in both space and time.     

Tree Phenology in Adjacent Amazonian Flooded and Unflooded Forests1

Most phenological studies to date have taken place in upland forest above the maximum flood level of nearby streams and rivers. In this paper, we examine the phenological patterns of tree assemblages in a large Amazonian forest landscape, including both upland (terra firme) and seasonally flooded (várzea and igapó) forest. The abundance of vegetative and reproductive phenophases was very seasonal in all forests types. Both types of flooded forest were more deciduous than terra firme, shedding most of their leaves during the inundation period. Pulses of new leaves occurred mainly during the dry season in terra firme, whereas those in the two floodplain forests were largely restricted to the end of the inundation period. Flowering was concentrated in the dry season in all forest types and was strongly correlated with the decrease in rainfall. The two floodplain forests concentrated their fruiting peaks during the inundation period, whereas trees in terra firme tended to bear fruits at the onset of the wet season. The results suggest that the phenological patterns of all forest types are largely predictable and that the regular and prolonged seasonal flood pulse is a major determinant of phenological patterns in várzea and igapó, whereas rainfall and solar irradiance appear to be important in terra firme. The three forest types provide a mosaic of food resources that has important implications for the conservation and maintenance of wide‐ranging frugivore populations in Amazonian forests.     

Impact of Flooding on the Species Richness, Density and Composition of Amazonian Litter-Nesting Ants

Litter-nesting ants are diverse and abundant in tropical forests, but the factors structuring their communities are poorly known. Here we present results of the first study to examine the impact of natural variation in flooding on a highly diverse (21 genera, 77 species) litter-nesting ant community in a primary Amazonian forest. Fifty-six 3 × 3 m plots experiencing strong variation in flooding and twenty-eight 3 × 3 m terra firme plots were exhaustively searched for litter-nesting ants to determine patterns of density, species richness and species composition. In each plot, flooding, litter depth, twig availability, canopy cover, plant density, percent soil nitrogen, carbon, and phosphorus were measured. Degree of flooding, measured as flood frequency and flood interval, had the strongest impact on ant density in flooded forest. Flooding caused a linear decrease in ant abundance, potentially due to a reduction of suitable nesting sites. However, its influence on species richness varied: low-disturbance habitat had species richness equal to terra firme forest after adjusting for differences in density. The composition of ant genera and species varied among flood categories; some groups known to contain specialist predators were particularly intolerant to flooding. Hypoponera STD10 appeared to be well-adapted to highly flooded habitat. Although flooding did not appear to increase species richness or abundance at the habitat scale, low-flooding habitat contained a mixture of species found in the significantly distinct ant communities of terra firme and highly flooded habitat.
     

Interspecific primate associations in Amazonian flooded and unflooded forests

Stable associations between two or more primate species are a prominent feature of neotropical forest vertebrate communities and many studies have addressed their prevalence, and their costs and benefits. However, little is known about the influence of different habitat types on the frequency, seasonality, and composition of mixed-species groups in Amazonian forest primates. Here we examine the features of interspecific primate groups in a large mosaic of flooded (várzea and igapó) and unflooded (terra firme) forest in central Amazonia. In total, 12 primate species occurred in the study area, nine of which were observed in mixed-species associations. Primates were more than twice as likely to form associations in várzea forest than in terra firme forest. Squirrel monkeys were most frequently found in mixed-species groups in all forest types, most commonly in association with brown capuchins. Another frequent member of interspecific associations was the buffy saki, which often formed mixed-species groups with tamarins or brown capuchins. There was no seasonality in the frequency of associations in terra firme forest whereas associations in várzea forest were twice as frequent during the late-dry and early-wet seasons than in the late-wet and early-dry seasons. Interspecific primate associations were common in all forest types, but the degrees to which different species associate varied between these environments. We suggest that the temporal variation of várzea forest associations is connected with seasonal changes in habitat structure and resource abundance. However, more work is needed to pinpoint the underlying causes of mixed-species associations in all forest types and their strong seasonality in várzea forest.     

Seasonal Foliage Changes in the Eastern Amazon Basin Detected from Landsat Thematic Mapper Satellite Images1

Seasonal changes in tropical forests are difficult to measure from the ground, especially in areas of high species diversity and low phenological synchrony. Satellite images, which integrate individual tree canopies and cover a large spatial extent, facilitate tests for stand-level canopy phenology. Variability in near-infrated radiance (TM bands 4 and 5) of several distinct vegetation types was used to detect seasonal changes in a series of three Landsat Thematic Mapper (TM) images from the wet season to the dry season in Marabá, Brazil (eastern Amazon basin). Despite different atmospheric and instrumental conditions among the images, spectral changes were distinguishable. A phenological process (leaf aging, leaf drop, water stress) was determined from the spectral changes for each vegetation type. Changes in the spectral properties suggest that during the dry season, upland terra firme forest increased the rate of leaf exchange and some riparian vegetation was deciduous. Terra firme forest that had been altered by penetration of fires from nearby pastures increased in leaf biomass over a 14-month period. This study shows that a time series of images can provide information on temporal changes in primary vegetation and guide field studies to investigate seasonal changes that may not be detectable from the ground.     

Habitat Selection and Use of Space by Bald-Faced Sakis (Pithecia irrorata) in Southwestern Amazonia: Lessons from a Multiyear, Multigroup Study

Population density and distribution in tropical forest vertebrates are directly linked to patterns of use of space relative to habitat structure and composition. To examine how forest type may explain the ranging behavior and high variance in group density observed within the geographic range of bald-faced saki monkeys (Pithecia irrorata), we monitored habitat use of 5 neighboring focal groups of this species in southwestern Amazonia over 3 yr. To test whether sakis are unflooded (terra firme) forest specialists, we compared home range (HR) use to the corresponding availability of 4 main forest types and quantified HR size and activity budgets as a function of forest type. HR size varied from 16 to 60 ha, and saki population density at this scale (12.5 ± 6.4 SD individuals/km²) was more closely related to forest type than to group size. Although sakis were not obligate habitat specialists, groups clearly avoided bamboo forest and preferred terra firme forest. Terra firme forests were associated with small HRs, intensive use, high HR overlap, and territorial defense, all of which suggest that saki densities will be higher in areas dominated by terra firme forest where large patches of bamboo (Guadua spp.) are absent. The increased desiccation and subsequent forest fires expected in this region from the combined impacts of climate change and human land use potentially threaten the long-term viability of old-growth terra firme forest specialists such as sakis. Regional-scale conservation efforts should ensure that extensive blocks of terra firme forest are protected in areas that remain relatively free of bamboo.     

Seasonal patterns of spatial variation in understory bird assemblages across a mosaic of flooded and unflooded Amazonian forests

We examined seasonal patterns of spatial variation in understory bird assemblages across a mosaic of upland and floodplain forests in central Amazonia, where variation in flooding patterns and floodwater nutrient load shapes a marked spatial heterogeneity in forest structure and composition. Despite great differences in productivity due to flooding by either nutrient-rich “white waters” (várzea) or nutrient-poor “black waters” (igapó), bird assemblages in the two floodplain forest types were relatively similar, showing lower abundances than adjacent upland forests (terra firme) and sharing a set of species that were absent or scarce elsewhere. Species that breed in pensile nests overhanging water were abundant in floodplain forests, whereas species that feed on the ground were generally scarce. Flooding affected assemblage dynamics in floodplain forests, with some influx of ground-dwelling species such as ant-following birds from adjacent upland during the low-water season, and the occupation by riverine and aquatic species such as kingfishers during floods. Spatial configuration influenced the seasonal pattern of assemblage structuring, with movements from terra firme occurring primarily to adjacent igapó forests. No such influx was detected in várzea forests that were farther from terra firme and isolated by wide river channels. Results support the view that habitat heterogeneity created by flooding strongly contributes to maintain diverse vertebrate assemblages in Amazonia forest landscapes, even in the case of largely sedentary species such as understory forest birds. Including both upland and floodplain forests in Amazonia reserves may thus be essential to preserve bird diversity at the landscape scale.     

Comparison Between Two Methods for Measuring Fruit Production in a Tropical Forest1

We compared fruiting data derived simultaneously from fruit traps placed on the ground and from canopy‐surveyed plots in a terra firme rain forest, Colombian Amazonia. Values derived from the canopy‐surveyed plots were higher than fruit‐trap estimates. Fruiting patterns obtained throughout both methods were not correlated. Our results showed that the fruit‐trap method does not accurately reflect fruiting patterns occurring at the highest levels of the forest, while the canopy‐surveyed plots provided both quantitative and qualitative information on canopy fruit production, and each species contribution.     

Patterns of plant phenology in Amazonian seasonally flooded and unflooded forests

Few studies have successfully monitored community‐wide phenological patterns in seasonally flooded Amazonian várzea forests, where a prolonged annual flood pulse arguably generates the greatest degree of seasonality of any low‐latitude ecosystem on Earth. We monitored the vegetative and reproductive plant phenology of várzea (VZ) floodplain and adjacent terra firme (TF) forests within two contiguous protected areas in western Brazilian Amazonia, using three complementary methods: monthly canopy observations of 1056 individuals (TF: 556, VZ: 500), twice monthly collections from 0.5‐m² litterfall traps within two 100‐ha plots (1 TF, 1 VZ; 96 traps per plot), and monthly ground surveys of residual fruit‐fall along transect‐grids within each 100‐ha plot (12 km per plot). Surveys encompassed the entire annual flood cycle and employed a floating trap design to cope with fluctuating water levels. Phenology patterns were generally similar in both forest types. Leaffall peaked during the aquatic phase in várzea forest and the dry season in terra firme. Flowering typically followed leaffall and leaf flush, extending into the onset of the terrestrial phase and rainy season in várzea and terra firme, respectively. Abiotic seed dispersal modes were relatively more prevalent in várzea than terra firme; the main contrast in fruiting seasonality was more likely a result of differences in community composition and relative abundance of seed dispersal modes than differences within individual genera. We emphasize the difficulty in distinguishing the role of the flood pulse from other seasonal environmental variables without multiannual data or spatially replicated studies across the spectrum of Amazonian forest types.     

Peruvian Red Uakaris (<Emphasis Type="Italic">Cacajao calvus ucayalii</Emphasis>) Are Not Flooded-Forest Specialists

In the literature, particularly in primatological books, the Peruvian red uakari (Cacajao calvus ucayalii) is generally considered as a species that is specialized on living in flooded forest, despite existing evidence to the contrary. Here we review all available information on habitats where Cacajao calvus ucayalii have been observed. Most sightings are from terra firme, including palm swamps, or from mixed habitats, including terra firme and flooded forest. Therefore, we conclude that the species is not a flooded-forest specialist, but is flexible in its habitat requirements and generally uses terra firme forests or a mixture of habitats. Proper recognition of habitat requirements is important for understanding the ecoethological adaptations of a species and for appropriate conservation measures.     

Spatial, Temporal, and Economic Constraints to the Commercial Extraction of a Non–timber Forest Product: Copaíba (Copaifera spp.) Oleoresin in Amazonian Reserves

Spatial, Temporal, and Economic Constraints to the Commercial Extraction of a Non?CTimber Forest Product: Copaíba (Copaifera spp.) Oleoresin in Amazonian Reserves. The increasing prevalence of government?C and NGO?Csponsored programs to encourage commercial non?Ctimber forest product (NTFP) extractivism in the humid tropics has highlighted the need for ecological and socioeconomic appraisal of the viability of extractive industries. We adopted a novel integrative approach to examine NTFP resource potential and produced credible landscape?Cscale estimates of the projected value of an economically important Amazonian NTFP, the medicinal oleoresin of Copaifera trees, within two large contiguous extractive reserves in Brazilian Amazonia. We integrated results derived from previous spatial ecology and harvesting studies with socioeconomic and market data, and mapped the distribution of communities within the reserves. We created anisotropic accessibility models that determined the spatial and temporal access to Copaifera trees in permanently unflooded (terra firme) and seasonally flooded (várzea) forests. Just 64.9?% of the total reserve area was accessible, emphasizing the distinction between the actual resource stock and that which is available to extractors. The density of productive tree species was higher in the várzea forests, but per?Ctree productivity was greater in the terra firme forests, resulting in similar estimates of oleoresin yield per unit area (64?C67?ml?ha^?C1) in both forest types. A greater area of the várzea forests was accessible within shorter travel times of ??250 minutes; longer travel times allowed access to increasingly greater volumes of oleoresin from the terra firme forests. The estimated total volume of oleoresin accessible within the two reserves was 38,635 liters for an initial harvest, with projected offtake for a subsequent harvest falling to 8,274 liters. A household that extracted just 2 liters of oleoresin per month could generate 5?% of its mean income; market data suggested that certification could increase the value of the resource fivefold. Our approach is valuable in that it incorporates a range of methodologies and quantitatively accounts for the numerous constraints to the commercial viability of NTFP extraction.     

宁夏黄土丘陵区不同生态恢复生境地表甲虫多样性

为了解宁夏黄土丘陵区不同生境地表甲虫群落多样性变化规律及与环境因子的关系, 并探讨不同生态恢复措施对维持地表甲虫群落多样性的影响, 2013年7-8月, 作者利用陷阱法调查了该区6种生境内的地表甲虫群落多样性。结果表明: 灌草混交林地、乔灌混交林地、生态薪炭林地中地表甲虫物种丰富度和个体数量均较高, 天然封育草地、生态经济林地和水平农田中物种丰富度较低, 生态经济林地和水平农田中甲虫个体数量显著高于天然封育草地。不同生境间, 灌草混交林地、乔灌混交林地与生态薪炭林地之间、生态经济林地与水平农田之间甲虫群落组成相似性较高。多元回归分析表明, 草本层生物量、灌木层盖度及土壤含水量是影响甲虫物种丰富度的决定因素, 林冠层盖度和枯落物厚度是决定地表甲虫个体数量的重要因素。CCA分析表明, 枯落物盖度、枯落物厚度、林冠层盖度及草本层盖度是影响地表甲虫群落组成的重要环境因子。研究表明, 灌草混交林地为地表甲虫群落多样性维持较好的生境类型, 是宁夏黄土丘陵区典型生态恢复的最优模式。     

Modeling the vertical foliage distribution of an individual <Emphasis Type="Italic">Castanopsis cuspidata</Emphasis> (Thunb.) Schottky,a dominant broad-leaved tree in Japanese warm-temperate forest

The vertical foliage distribution of Castanopsis cuspidata (Thunb.) Schottky was examined in trees of various sizes to clarify its variation in relation to tree size and the light environment in a stand. As indices of these parameters, we analyzed crown social position (CSP: percent of stand height) and specific leaf area (SLA). The vertical foliage distribution of trees was expressed by a Weibull function. The variation in the vertical foliage distribution of C. cuspidata could be categorized into three types using crown social position and light environment. In the first type, leaves were concentrated to the top 20% of the tree; such trees are canopy trees that can receive full sunlight. The second type had a large relative crown depth and an asymmetric distribution with the maximum foliage located near the top of the tree; such trees are suppressed trees whose crowns do not receive sufficient light. The third type had a large relative crown depth and a symmetric distribution; such trees occur in high light environments, although their crowns are in the understory layer. The differences in the vertical foliage distribution are related to the strategies used to capture light. Multiple regression analysis showed that CSP and SLA at the top layer of the tree explained successive changes in the vertical foliage distribution. These results will contribute to scaling-up the vertical foliage distribution to the community level in pure stands of C. cuspidata using an individual-based model.     

Avian distribution in treefall gaps and understorey of terra firme forest in the lowland Amazon

We compared the bird distributions in the understorey of treefall gaps and sites with intact canopy in Amazonian terra firme forest in Brazil. We compiled 2216 mist-net captures (116 species) in 32 gap and 32 forest sites over 22.3 months. Gap habitats differed from forest habitats in having higher capture rates, total captures, species richness and diversity. Seventeen species showed a significantly different distribution of captures between the two habitats (13 higher in gap and four higher in forest). Gap habitats had higher capture rates for nectarivores, frugivores and insectivores. Among insectivores, capture rates for solitary insectivores and army ant followers did not differ between the two habitats. In contrast, capture rates were higher in gaps for members of mixed-species insectivore flocks and mixed-species insectivore–frugivore flocks. Insectivores, especially members of mixed-species flocks, were the predominant species in gap habitats, where frugivores and nectarivores were relatively uncommon. Although few canopy species were captured in gap or forest habitats, visitors from forest mid-storey constituted 42% of the gap specialist species (0% forest) and 46% of rare gap species (38% forest). Insectivore, and total, captures increased over time, but did so more rapidly in gap than in forest habitats, possibly as a response to gap succession. However, an influx of birds displaced by nearby timber harvest also may have caused these increases. Avian gap-use in Amazonian terra firme forests differs from gap-use elsewhere, partly because of differences in forest characteristics such as stature and soil fertility, indicating that the avian response to gaps is context dependent.     

Communities of insect herbivores foraging on saplings versus mature trees of Pourouma bicolor (Cecropiaceae) in Panama

The arthropod fauna of 25 saplings and of three conspecific mature trees of Pourouma bicolor (Cecropiaceae) was surveyed for 12 months in a tropical wet forest in Panama, with particular reference to insect herbivores. A construction crane erected at the study site provided access to tree foliage in the upper canopy. A similar area of foliage (ca. 370 m²) was surveyed from both saplings and trees, but samples obtained from the latter included 3 times as much young foliage as from the former. Arthropods, including herbivores and leaf-chewing insects with a proven ability to feed on the foliage of P. bicolor were 1.6, 2.5 and 2.9 times as abundant on the foliage of trees as on that of saplings. The species richness of herbivores and proven chewers were 1.5 (n=145 species) and 3.5 (n=21) times higher on trees than on saplings, respectively. Many herbivore species preferred or were restricted to one or other of the host stages. Host stage and young foliage area in the samples explained 52% of the explained variance in the spatial distribution of herbivore species. Pseudo-replication in the two sampling universes, the saplings and trees studied, most likely decreased the magnitude of differences apparent between host stages in this forest. The higher availability of food resources, such as young foliage, in the canopy than in the understorey, perhaps combined with other factors such as resource quality and enemy-free space, may generate complex gradients of abundance and species richness of insect herbivores in wet closed tropical forests.     

Phyllostomid Bat Assemblage Structure in Amazonian Flooded and Unflooded Forests

In Amazonia, the assemblages of several taxa differ significantly between upland terra firme and white‐water flooded várzea forests, but little is known about the diversity and distribution of bats in these two forest types. We compare the spatio‐temporal patterns of bat assemblage composition and structure in adjacent terra firme and várzea forests in the lower Purus River region of central Brazilian Amazonia. Bats were sampled using mist nets at five sites in each forest type during 40 nights (2400 net‐hours). We captured 1069 bats representing 42 species and Phyllostomidae bats comprised 99.3 percent of all captures. The bat assemblages in várzea and terra firme forests were significantly different, mainly due to a marked dissimilarity in species composition and in the number of captures during high‐water season. In addition, bat assemblages within forest types differed significantly between seasons for both terra firme and várzea. Frugivores dominated the bat assemblages in both forest types. Overall guild structure did not change between várzea and terra firme or between seasons, but frugivore and animalivore abundance increased significantly in várzea forest during the inundation. The difference in assemblage structure observed in the high‐water season is probably caused by the annual várzea flooding, which provides an effective barrier to the persistence of many understory bats. We also hypothesize that some bat species may undertake seasonal movements between forest types in response to fruit abundance, and our results further underline the importance of floodplain habitats for the conservation of species in the Amazon.     

Diet and feeding ecology of gray woolly monkeys (lagothrix lagotricha cana) in Central Amazonia: Comparisons with other Atelines

I studied gray woolly monkeys (Lagothrix lagotricha cana)in an undisturbed central Amazonian terra firme forest, near the headwaters of the Urucu river, Tefé, Amazonas, Brazil (5°50’S, 65°16’W). I report the diet and feeding ecology of a group of 39–41 individuals, based on systematic feeding observations obtained during 11 months. Woolly monkeys are primarily frugivorous; mature fruits and young seeds account for 83 and 7% of 3298 feeding records, respectively. On a seasonal basis, however, they relied heavily on young foliage (16%), seed-pod exudates (6%), and flowers (3%), particularly during the greatest annual period of ripe fruit scarcity, as determined by a phenological survey. Animals represent only 0.1% of their year-round diet, and they spent little time capturing arthropods and other prey items. Although at least 225 plant species, belonging to 116 genera and 48 families, are in their diet, the three top-ranking families (Moraceae, Sapotaceae, and Leguminosae) account for 43% of their food species and 63% of the time they spent feeding on a year-round basis. I compare the feeding ecology and diet of L. 1. canain the Urucu and other taxa of Lagothrixin upper Amazonia — the last large-bodied Neotropical primates to be studied — to those of other ateline genera: Atelesand Brachyteles. An erratum to this article is available at .     

Complexity of leaf miner–parasitoid food webs declines with canopy height in Patagonian beech forests

1. Consumer–resource species interactions form complex, dynamic networks, which may exhibit structural heterogeneity at various scales. This study set out to address whether host–parasitoid food web size and topology vary across forest canopy strata, and to what extent foliar resources and species abundances account for vertical patterns in network structure. 2. The vertical stratification of leaf miner–parasitoid food webs was examined in two monotypic beech (Nothofagus pumilio) forests in northern Patagonia, Argentina. Quantitative food webs were constructed for separate canopy layers by sampling foliage from three tree‐height classes at 0.5–1, 2–3 and 5–6 m above ground. 3. Leaf miner abundance per unit leaf mass and foliar damage (%) did not differ across strata, although foliage quality and quantity increased from the understorey to the upper canopy. Parasitism rates and food web complexity decreased with canopy height, as reflected by reduced linkage richness, linkage density, mean interaction strength, and host vulnerability. 4. Null model analyses revealed that food web metrics, especially in the upper canopy, were often lower than expected when compared with randomly structured networks. Overall, these patterns held for two forests differing in vertical structure and in dominant miner morphotype and parasitoid species. 5. These results suggest that vertical declines in network complexity may be driven by the parasitoids' limited functional response to host abundance and dispersal from pupation sites in the forest floor. A broader constraint on food web structure seemed to be imposed by host–parasitoid trait matching, a reflection of large‐scale assembly processes.