Invertebrate predators and parasitoids are among the most important natural enemies of insect herbivores. Yet, the strength of natural enemy pressure along an altitudinal gradient and interactions between the groups of natural enemies (such as predation on parasitized prey) are not well known. Various methods are used to reveal the mortality factors of herbivores. Predation pressure is usually assessed through exposure of artificial prey. However, this method cannot provide information about the attacks of parasitoids, or their eventual interactions with predators. Furthermore, artificial or dead prey might not attract predators because they do not show expected host behavior, and this method mostly cannot distinguish between predation and scavenging. For the first time in a tropical rainforest, we quantified elevational contrast in mortality factors using exposure of live caterpillars. We exposed a total of 800 live caterpillars of Talanga excelsalis moresbyensis Strand (Lepidoptera: Crambidae) on saplings of Ficus copiosa Steud. (Moraceae) at two elevations in primary tropical rain forest in Papua New Guinea (200 and 1 200 m a.s.l.). We exposed the caterpillars in two treatments: exposed to and protected from invertebrate predators and parasitoids. Disappearance of caterpillars was significantly higher in the exposed treatment. Furthermore, caterpillar disappearance was significantly higher in lowlands than in highlands (43 vs. 12%). We consider the vast majority of the disappearance to be due to predation, as migration of the caterpillars from the focal trees was not observed (except one caterpillar). This estimate of invertebrate predation rate corresponds with studies which used artificial caterpillar models. No significant difference in parasitism rate between the two elevations was observed (12 vs. 13%). The combination of the disappearance and parasitism rate patterns means that larval parasitoids face stronger pressure from invertebrate predators through higher predation of their hosts in the lowlands than in the highlands.

Papua New Guinea (PNG) has the greatest diversity of dacine fruit fly species (Diptera: Tephritidae: Dacinae) in the world, many of which are significant agricultural pests. Although their taxonomy is relatively well known, there is limited research on their geographical and elevational distribution. We undertook a survey of PNG's fruit fly fauna along a complete elevational gradient on Mt Wilhelm (175–3700 m a.s.l.) to determine the elevational species distribution and turnover of fruit fly communities. Fruit flies were sampled using a selection of male parapheromone lures, including Zingerone, a new and promising male attractant. In total, we collected 10 700 fruit flies representing 77 species. The total abundance and species richness of fruit flies decreased with increasing elevation. Species richness was similarly high at 175 and 200 m a.s.l. (30 and 27 species) and at 700 and 1200 m a.s.l. (16 and 20 species, respectively) but dropped suddenly to only two species at 1700 m a.s.l. and to zero at all four sites above this elevation. There were no flies attracted exclusively to Zingerone. One species (Bactrocera recurrens) exhibited dramatic, temporal changes in abundance during the study period. Fruit fly assemblages were significantly influenced by both the lure type and elevation. Similarity of fruit fly communities decayed linearly with increasing elevational distance. We concluded that the upper limit of fruit fly distribution in PNG occurs between 1700 and 2200 m a.s.l. and the centre of diversity occurs between 0 and 700 m a.s.l.