1. The Latitudinal Biotic Interaction Hypothesis (LBIH) predicts that the strength of various biotic interactions decreases from low to high latitudes. Inconsistency between studies testing this hypothesis may result from variations among different types of interactions and among study systems. Therefore, exploration of multiple interactions within one system would help to disentangle latitudinal patterns across individual interactions and to evaluate latitudinal changes in the overall impact of enemies on prey.
2. We tested the prediction based on the LBIH that the pressure of natural enemies on herbivorous insects decreases with an increase in latitude across the boreal forest zone. We also asked whether the impacts of major groups of these enemies exhibit similar latitudinal patterns and whether these patterns are consistent across study years. 
3. In 10 forest sites located from 60°N to 69°N in Northern Europe, each summer, from 2016–2019, we measured (i) mortality of three groups of leafmining insects caused by birds, ants, parasitoids, and unknown factors, (ii) bird attacks on caterpillar-shaped plasticine models, and (iii) birch foliar damage caused by defoliators and leafminers.
4. Latitudinal patterns in both insect herbivory on birch and top-down pressure on herbivorous insects varied considerably and inconsistently among the four study years, so that only some of the year-specific correlations with latitude were statistically significant. Nevertheless, meta-analysis combining correlations across years, preys and enemies revealed general decreases in predation by birds (on both natural and model prey) and ants, but an increase in parasitism rates, from low to high latitudes.
5. We found that the direction of latitudinal changes in the strength of biotic interactions was interaction-specific: predation and herbivory supported LBIH, whereas parasitism exhibited an opposite trend. Consequently, the overall impact of natural enemies on herbivorous insects did not change with latitude and was therefore an unlikely reason for the poleward decrease in herbivory observed in our gradient. Considerable among-year variation in the strength of the latitudinal patterns in all the studied interactions suggests that this variation is a widespread phenomenon.

The data are of three types: (1) mortality of leafminers: three data filed and three metadata files; (2) bird attack rates on artificial prey: one data file and one metadata file; and (3) herbivory on birches: one data file and one metadata file.

 1. Mortality of leafminers

We collected birch leaves mined by the larvae of Eriocrania spp. (between 10 and 29 June) and Stigmella spp. (between 8 and 29 August), and rowan leaves mined by the larvae of several moth species (between 8 and 29 August); sampling was conducted at each of our study sites annually from 2016‒2019. The same experienced persons searched for the mined leaves at all sites and during all study years. All discovered mines had been collected; we attempted to obtain 20–50 mines of each of three groups of moths from each study site annually. The samples were classified as either empty mines left by larvae which had successfully completed their development (i.e. survived individuals) or mines containing dead larvae or mines that had been opened by predators. The mines containing dead larvae were classified as killed by parasitoids if a parasitoid larva or pupa was found inside the undamaged mine. If no signs of parasitism were discovered, then the death was attributed to unknown reasons. All larvae that failed to reach 10% of their final size were classified as having died from unknown reasons because they were too small to have been a target for predators or parasitoids. If the mine contained no dead larva and only a small part of a single epidermal wall of the mine was damaged (usually in the form of one or more small round holes), then the larva was classified as having been eaten by ants. If both walls of the mine were punched through and severely damaged, the mortality was attributed to birds.

2. Bird attack rates on artificial prey 

We fabricated our prey models from non-toxic, unscented, soft modelling clay (Chemical plant ‘Luch’, Yaroslavl, Russia) of green and brown colours to imitate the natural colours of palatable caterpillars. Green wire 0.3 mm in diameter was used to attach two plasticine models (28–32 mm length and 3.5–4 mm diameter) of different colours to thin branches of each of five haphazardly selected mature (>3 m height) birches at a height of 1.2–1.8 m and at least 50 cm apart. The models were exposed from 10–12 June and checked on 26–30 June, 8–12 August and 23–29 August in 2016–2019. During each inspection, we counted plasticine caterpillars with beak marks left by birds and then remoulded the attacked models or replaced them if the damage was severe. When a model (including the wire) was not found, this record was considered as missing and a new model was placed on the tree.

3. Herbivory on birches

On 8–16 August 2016–2019, when the majority of insect herbivores had completed their feeding, we haphazardly selected five mature individuals of each of the two birch species at each site. Each individual was located at least 10 m apart from others of its species. From each tree, we collected a branch with approximately 80–120 leaves (median value: 100 leaves). We avoided the impact of unconscious biases on the values of foliar losses by selecting the branches while standing at a distance of 5‒10 m away, which prevented visual evaluation of leaf damage by insects. The branches were labelled with random numbers, so the person assessing herbivory was blinded with respect to the sample origin.

The leaves on each branch (including the petioles of fully consumed leaves) were counted, and each leaf was carefully examined for the presence of insect damage. Each leaf was assigned to one of the damage classes according to the percentage of the area of the leaf that was consumed or otherwise damaged by defoliating insects: 0 (intact leaves), 0.01–1, 1–5, 5–25, 25–50, 50–75 and 75–100%. The plant-specific percentage of foliage lost to insects was calculated as follows: the numbers of leaves in each damage class were multiplied by the respective median values of the damaged leaf area (i.e. 0 for intact leaves, 0.5% for the damage class 0.01–1%, 3% for the damage class 1–5%, etc.); the obtained values were summed for all damage classes and divided by the total number of leaves (including undamaged ones) in a sample. The abundances of Eriocrania and Stigmella leafminers were quantified as their intensities, i.e. the number of mines divided by the number of leaves on the branch.