Application of methyl jasmonate reduces growth but increases chemical defence and resistance against Hylobius abietis in Scots pine seedlings


*Correspondence: Juha Heijari, University of Kuopio, Department of Ecology and Environmental Science, PO Box 1627, FIN-70211, Kuopio, Finland. Tel.: +358 17 163199; Fax: +358 17 163230; E-mail:


Scots pine (Pinus sylvestris L., Pinaceae) produces a terpenoid resin which consists of monoterpenes and resin acids that offer protection against herbivores and pathogen attacks. Methyl jasmonate (MJ) is a potential plant elicitor which induces a wide range of chemical and anatomical defence reactions in conifers and might be used to increase resistance against biotic damage. Different amounts of MJ (control, 10 mm, and 100 mm) were applied to Scots pine to examine the vigour, physiology, herbivory performance, and induction of secondary compound production in needles, bark, and xylem of 2-year-old Scots pine seedlings. Growth decreased significantly in both MJ treated plants, and photosynthesis decreased in the 100 mm MJ treated plants, when compared to 10 mm MJ or control plants. The large pine weevil (Hylobius abietis L.) (Coleoptera: Curculionidae) gnawed a significantly smaller area of stem bark in the 100 mm treated plants than in the control or 10 mm treated plants. The 100 mm MJ treatment increased the resin acid concentration in the needles and xylem but not in the bark. Furthermore, both MJ treatments increased the number of resin ducts in newly developing xylem. The changes in plant growth and chemical parameters after the MJ treatments indicate shifts in carbon allocation, but MJ also affects plant physiology and xylem development. Terpenoid resin production was tissue-specific, but generally increased after MJ treatments, which means that this compound may offer potential protection of conifers against herbivores.


Large pine weevil (Hylobius abietis L.) (Coleoptera: Curculionidae) adults emerge from tree stumps and feed on the stem phloem and bark of young seedlings, causing severe problems in cultivated clear-cut areas of Scots pine and Norway spruce (Örlander & Nilsson, 1999). Heavy infestations of this insect can result in the widespread loss of young seedlings for several years after planting. Both sexes of large pine weevil beetles are attracted by volatile compounds, such as the α-pinene and β-pinene emitted by fresh conifer stumps left during clear felling (Zagatti et al., 1997). In Europe, insecticides are commonly used to limit large pine weevil damage. There is a need for alternative chemical or mechanical methods to prevent H. abietis damage, as the commonly used insecticide, permethrin, was prohibited in the European authorities on 31 December 2003.

Conifers possess chemical defences, such as the production of terpenoid resins or phenols, that help to reduce attacks by insects and pathogens (Mattson et al., 1988). Resins are thought to be toxic or repellent compounds to herbivores, but they can also act as attractants and feeding stimulants (Gershenzon & Croteau, 1991). Methyl jasmonate (MJ) has been found to be a potential plant elicitor which induces wide plant defence reactions in conifers, such as the formation of traumatic resin ducts, the accumulation of mono- and diterpenes, induction of enzyme activities of terpene synthases, or the formation of additional polyphenolic parenchyma cells in the cambial zone (e.g., Franceschi et al., 2002; Martin et al., 2002), and MJ has been shown to protect plants against pathogens (Kozlowski et al., 1999). In nature, MJ occurs in many plant species as a signal- or hormone-like compound that is involved in several processes, such as growth, photosynthesis, and insect and disease resistance (Creelman & Mullet, 1997).

Pine species (Pinus spec.) contain an elaborate network of interconnected resin ducts located throughout the wood and bark (Wu & Hu, 1997). The presence of insects, pathogens, or mechanical wounding will induce resin synthesis in the wounded areas (Gref & Ericsson, 1984; Franceschi et al., 2002), and this process can also involve complex stem anatomical changes, such as the formation of resin ducts (Tomlin et al., 1998). Plant chemical and anatomical defence responses can be induced by the exogenous application of MJ in Picea (Franceschi et al., 2002; Martin et al., 2002) and in Pinus (Hudgins et al., 2003), and MJ might therefore be used to reduce insect herbivory in conifer trees.

The objective of this investigation was to determine if MJ could induce defence reactions (chemical and anatomical) in Scots pine seedlings. The effects of MJ applications in bark-feeding H. abietis adults were tested. The monoterpene and resin acid concentrations in the needles, bark, and xylem, and the effects of MJ on the shoot and root growth, physiological parameters, and xylem anatomical properties were all investigated.

Materials and methods


Scots pine seedlings (2 years old) were obtained from the Suonenjoki Research Nursery of the Finnish Forest Research Institute, Suonenjoki. Seedlings were individually planted in 7.5 l plastic pots in 2 : 1 (v/v) quartz sand (diam. 0.5–1.2 mm, SP Minerals Partek, Finland) and fertilized sphagnum peat (Kekkilä PP6, Finland) on 19 May 2003. After planting, the seedlings were fertilized with a slow-release fertilizer (Taimiston kestolannos: 9% nitrogen, 5% phosphorous, 5% potassium and 5% magnesium, 4% sulphur, and micronutrients; Kemira Co, Finland) on 22 May. Fertilizer was added at 4 g per pot into the surface layer of the growth medium. The seedlings were grown under natural rainfall, and watered as needed.

A split-plot design was used in a field experiment (University of Kuopio Research Garden; 62°13′N, 27°35′E, 80 m above sea level) where Scots pine seedlings grew on 12 separate wooden trays with 16 seedlings on each tray. The trays were grouped into four blocks each with all three treatments. Each tray (16 seedlings) was treated with control, 10 mm, or 100 mm methyl jasmonate (MJ) solution. Each seedling within each treatment received 30 ml (fully coating needles, branches, and stem) of the control solution, 10 mm MJ (Bedoukian Research Inc., Danbury, CT, USA) solution, or 100 mm MJ solution on 27 May. The MJ solution contained (v/v) 5% ethanol and 0.1% Tween 20 detergent; the control solution contained only distilled water with 5% ethanol and 0.1% Tween 20.

Feeding tests

A total of six seedlings per treatment were randomly selected from replicated blocks for insect feeding tests. To compare bark and xylem monoterpene and resin acid concentrations in seedlings not exposed to insects with the concentrations in insect treated seedlings, two seedlings were randomly selected per treatment. The seedlings were transported to a greenhouse, enclosed in ventilated plastic chambers (31 cm in diameter and 68 cm in height, polymethylmeta-acrylate, Gammacril, Nerviano, Italy) and kept under a L16:D8 photoperiod during the 4-day experiment on 25–28 August (96 h). Two weighed H. abietis adults were added to the test chambers. Randomly selected sexes were included in the test, based on the earlier experiments, where no differences were observed in their feeding ability of equal weight individuals (R. Toivonen and H. Viiri, pers. comm.). The mean (±SD) temperature during the experiment was 21.8 ± 2.2 °C and the relative humidity 87.4 ± 12.9%. After the feeding tests, the experimental insects were re-weighed and seedlings were harvested for shoot fresh weight determination and to measure the area of the gnawed feeding scars. The roots were washed and the fresh weight and main root length were measured.

Chemical analyses

For the chemical analyses, current year (C) needles, bark (+phloem), and xylem, and second year (C + 1) bark and xylem (C + 1 samples included current year tissue) were collected in liquid nitrogen on 28 August [from the same seedlings which were used (n = 6) or not used (n = 2) in the insect tests] and stored at −80 °C. For monoterpene analyses, needle, bark, and xylem samples were extracted with n-hexane as previously described in Manninen et al. (2002). Resin acids were extracted from freeze-dried and powdered needle, bark, and xylem samples with petroleum ether-diethyl ether (Manninen et al., 2002). Monoterpene and resin acid extracts were analysed by gas chromatography–mass spectrometry (Hewlett Packard GC type 6890, MSD 5973; Hewlett Packard, Wilmington, DE, USA) using a 30-m long HP-5MS (0.25 mm ID, 0.25 µm film thickness, Hewlett Packard) capillary column. Helium was used as the carrier gas. The temperature program for monoterpenes rose from 50 °C to 250 °C, and for resin acids from 50 °C to 270 °C. The heating rate was 5 °C min−1. The SCAN technique (mass numbers from m/z 30 to 350 were recorded; signal ions in monitoring; 93, 133, 136, 161, 204 m/z) was used for monoterpene samples, and the technique of selected ion monitoring (SIM) 299, 301, 314, 316 m/z for resin acid samples. For the quantification of resin acids and terpenes, calibrations were made using known amounts of available pure compounds relative to known amounts of the internal standard (1-chloro-octane for monoterpenes and heptadecanoic acid for resin acids).

Growth and physiological measurements

Seedling height growth was measured from all seedlings (n = 16) per tray once a week from 27 May to 11 September 2003. Seedling diameter (1 cm from ground level) was measured on all seedlings (n = 16) per tray on 26 May and 11 September 2003. Net photosynthesis and stomatal conductance were measured from the current year needles of the same four seedlings per tray, three times, on 25 July, 13–14 August, and 5 September 2003. For net photosynthesis measurements, the needle length of 10 two-needle fascicles per annual shoot was measured. The total area (At) of needle was calculated using the model At = 4.2235 (needle length) – 15.6835 (Flower-Ellis & Olsson, 1993). Net photosynthesis and stomatal conductance measurements were made in saturating sunlight, between 10:00 and 14:00 hours with a CI-510 Ultra-light Portable Photosynthesis System (CID Inc., Camas, WA, USA).

Xylem and bark anatomical structure

Xylem cross-section samples (including bark and phloem) were collected on 28 August (from the same six seedlings that had been used in the insect tests) to anatomical analyses for 1.0 cm height of ground level and stored at −80 °C. Samples were embedded in OCT compound (Tissue-Tik, Sakura-Finetek, Zoeterwoude, The Netherlands) in isopentane (2-Methylbutan) and frozen in liquid nitrogen. A cryomicrotome was used to cut 16 mm thick cross-sections at −20 °C. The cross-sections were stained for 2 min with a solution of safranin 1% and alcian blue 1% (1 : 2 vol/vol) and rinsed with distilled water. Water was removed from the cross-sections with increasing ethanol series (50%, 70%, 94%, and twice in 100%, 2 min in each), after which the samples were kept in xylene (2 × 2 min) and mounted on glass slides with DePeX (BDH Gurr, Poole, UK) mounting medium. The cross-section samples were photographed with a light microscope using an Olympus C-5060 (Japan) digital camera and analysed with PC ImageJ v1.30 (NIH, Bethesda, MD, USA). From the digital pictures, current year xylem growth and bark (including phloem) thickness from four different sides of the cross-section, cell lumen area from two different pictures (average number of cells per seedling was 300), and number of resin ducts per mm2 from half of the current year xylem area were measured.

Statistical analysis

The General Linear Models (GLM) repeated-measures procedure was used for the analysis of seedling height growth, net photosynthesis, and stomatal conductance, with time as a within-subject variable and MJ treatments as a between-subject factor. The GLM Univariate procedure was used for seedling diameter, shoot and root fresh weights, root length, secondary compounds, insect feeding, and microscope analysis. If the main effects were significant, Tukey's multiple range test was done for a comparison of treatment means. If the parameters were not normally distributed they were analyzed with a Kruskall–Wallis test followed by a Mann–Whitney test with Bonferroni correction of P-values. A t-test was used when differences between insect gnawed and non-gnawed seedlings were analyzed. Correlation coefficients (r) were tested by Pearson correlations. Block means (n = 4) were used in the seedling height growth, net photosynthesis, stomatal conductance, and seedling diameter. Plant means (n = 6) were used in the shoot and root fresh weights, root length, secondary compounds, insect feeding tests (non-gnawed, n = 2), and microscope analysis. The statistical analyses were carried out using SPSS 11.5.1 for Windows (SPSS Inc., Chicago, IL, USA).


Herbivory and secondary chemistry

The amount of C and C + 1 bark gnawed by H. abietis was significantly affected by the 100 mm MJ treatment (Table 1). The H. abietis adults consumed 62% less of the bark in the 100 mm MJ treated seedlings than in the control seedlings. If only C or C + 1 bark was considered, the gnawed area did not differ between treatments (Table 1). No differences were observed in the weights of H. abietis between treatments after the feeding test (F2,18 = 0.477, P = 0.630). No correlations were observed between the C + 1 bark total monoterpenes (r = 0.110, P = 0.665; n = 18) or total resin acids (r = 0.075, P = 0.768; n = 18) and the gnawed C + 1 bark area. The gnawing response of male and female beetles did not differ (t = 0.228, P = 0.823; n = 18) from each other, and gnawing was similar if two males (n = 6), two females (n = 4), or two insects of opposite sexes (n = 8) were together in the chamber (F2,18 = 0.510, P = 0.611).

Table 1.  The effects of MJ treatments on Hylobius abietis performance, Scots pine growth, and anatomy. C = Current year bark and C + 1 = second year bark (including current year tissue). Means (±SEM) followed by different letters are significantly different (P<0.05, Tukey's test)
Control10 mm100 mm
  • a

    n = 1.

  • b

    Kruskal–Wallis tests χ2 and P values.

 Total gnawed bark area (cm2) 3.1 ± 0.6a 2.6 ± 0.5a  1.5 ± 0.3b3.805 0.048
 C + 1 gnawed bark area (cm2) 2.7 ± 0.5 2.3 ± 0.2  1.4 ± 0.33.065 0.084
 C gnawed bark area (cm2) 0.4 ± 0.1 0.3 ± 0.3  0.1 ± 0.0a0.103 0.905
 Seedling diameter 26 May (mm) 4.2 ± 0.2 4.3 ± 0.1  4.2 ± 0.20.266 0.772
 Seedling diameter 11 September (mm) 6.8 ± 0.5a 6.9 ± 0.3a  5.2 ± 0.3b6.410 0.019
 Shoot fresh weight (g)31.8 ± 4.6a29.9 ± 3.1a  9.6 ± 1.4b25.054 0.0002
 Root fresh weight (g)38.4 ± 4.9a36.1 ± 4.4a  7.6 ± 1.4b19.483 0.001
 Main root length (cm)51.2 ± 1.253.3 ± 2.2 42.2 ± 4.53.500 0.522b
 Number of resin ducts (mm−2)4.12 ± 0.54c6.16 ± 0.40b23.74 ± 2.32a95.004<0.0001
 Xylem annual growth (mm)1.06 ± 0.08a0.83 ± 0.08a 0.27 ± 0.04b34.968<0.0001
 Xylem cell area (µm2)1.81 ± 0.07a1.73 ± 0.10a 1.31 ± 0.04b12.730 0.0002
 Bark thickness (mm)0.81 ± 0.050.88 ± 0.04 0.76 ± 0.032.204 0.135

The total monoterpene concentration in the C xylem was significantly (F2,18 = 17.089, P<0.001) higher in the 100 mm MJ treated seedlings than in the 10 mm MJ or control seedlings (Figure 1). With respect to the individual compounds present in C xylem, only the concentration of α-pinene increased significantly in the 100 mm MJ treated seedlings. No changes in the total monoterpene concentration were observed in the C bark or C needles (Figure 1) or in the C + 1 bark or C + 1 xylem (F2,18 = 1.540, P = 0.246 and F2,18 = 1.930, P = 0.180, respectively). The total monoterpene concentration in the xylem (C + 1) of non-gnawed seedling was 2.2-fold higher (t = −5.620, P<0.001; n = 24) than in the gnawed seedlings (pooled from all gnawed seedlings; n = 18), whereas no differences were found between gnawed and non-gnawed C tissues.

Figure 1.

Total monoterpene concentration (means + SEM) in: (A) current year needles, (B) bark, and (C) xylem of Scots pine seedlings after control, 10 mm, or 100 mm MJ treatments or Hylobius abietis gnawing. Bars capped by different letters are significantly different from the other treatments (P<0.05, Tukey's test); gnawed seedlings (n = 6) and non-gnawed seedlings (n = 2).

The total resin acid concentration in the C needles (F2,18 = 4.369, P = 0.034) and C + 1 xylem (F2,18 = 3.530, P = 0.047) was significantly higher in the 100 mm treated seedlings than in the control seedlings (Figure 2). The concentrations of most individual resin acid compounds in C needles increased in the 100 mm MJ treated seedlings whereas in C + 1 xylem, only the concentration of abietic acid increased in the 100 mm MJ treated seedlings. The resin acid concentration in the bark of non-gnawed seedlings (C + 1) was 5.3-fold lower (t = 7.188, P<0.001; n = 24) and in the xylem (C + 1) of non-gnawed seedlings it was 4.7-fold higher (t = −3.109, P = 0.026; n = 24) than in the gnawed seedlings (pooled from all gnawed seedlings; n = 18) (Figure 2). However, in the non-gnawed seedlings, the differences in the resin acid concentration between treatments were not statistically significant (needles: F2,6 = 0.528, P = 0.636; bark (+ phloem): F2,6 = 0.605, P = 0.605; xylem: F2,6 = 1.084, P = 0.442).

Figure 2.

Total resin acid concentration (means + SEM) in: (A) current year needles, (B) second year bark, and (C) second year xylem of Scots pine seedlings after control, 10 mm, or 100 mm MJ treatments or Hylobius abietis gnawing. Bars capped by different letters are significantly different from the other treatments (P<0.05, Tukey's test); gnawed seedlings (n = 6) and non-gnawed seedlings (n = 2).

Growth, physiology, and xylem anatomy

Both MJ treatments significantly (F2,12 = 16.933, P = 0.001) affected seedling height growth (Figure 3A). Twenty days after the MJ treatments, a significant growth reduction was detected between control, 10 mm, and 100 mm MJ treated seedlings, which increased towards the end of the growing season (Figure 3A). In addition, the 100 mm MJ treatment significantly decreased seedling diameter, and shoot- and root fresh weights when compared to control or 10 mm MJ treated plants (Table 1). No difference was observed in the main root length (Table 1). During the study period, 59.4% of the 100 mm MJ treated seedlings died, but only 1.6% of the control and 10 mm treated seedlings.

Figure 3.

Seedling (A) height growth and (B) net photosynthesis of Scots pine (means ± SEM) in different MJ treatments in 2003; n = 4. Application of methyl jasmonate was done on 27 May.

Net photosynthesis was significantly lower (F2,12 = 5.555, P = 0.027) in the 100 mm seedlings than in control and 10 mm seedlings (Figure 3B), but no significant time × treatment effect was observed. Stomatal conductance (mean ± SEM: 5.42 ± 1.28 mmol m−2 s−1) did not differ (F2,12 = 0.968, P = 0.416) between treatments and no significant time × treatment effect was observed.

Xylem control samples showed relatively normal anatomy with only a few axial resin ducts (Figure 4A). In contrast, both MJ treatments induced the formation of resin ducts in a continuous band in the newly developing xylem (Figure 4B, C), this being highly visible in the 100 mm MJ treated seedlings (Figure 4C). The number of resin ducts was 5.8- and 3.9-fold higher in the 100 mm MJ seedlings than in the control or 10 mm MJ treated seedlings, respectively (Table 1). Furthermore, the 100 mm MJ treatment reduced annual diameter growth and tracheid cell lumen area when compared to control or 10 mm MJ treated seedlings (Table 1). No effects were seen in bark (+phloem) thickness (Table 1).

Figure 4.

Xylem cross-sections of Scots pine seedlings treated with different MJ treatments. (A) Control, (B) 10 mm, and (C) 100 mm. Abbreviations: RD = resin duct, C = cambium, and X = xylem. Bars = 0.5 mm.


MJ treated seedlings experienced lower feeding of Hylobius abietis than control seedlings. It seems that MJ increased the plant chemical defence, which may have reduced the feeding of H. abietis. However, no correlation was observed between gnawed bark and total monoterpenes or resin acid concentrations in the bark. Thus, the reduced insect feeding in the higher MJ treated plants might be due to the increased mono- and diterpene concentration in the xylem, because no changes in bark resin terpenoids were found. In a similar manner, MJ treatments induced an accumulation of terpenoid resins which was most apparent in the xylem but not in the bark of Norway spruce seedlings (Martin et al., 2002). We observed that the resin acid concentration increased in the bark and decreased in the xylem of the H. abietis gnawed seedlings when compared to non-gnawed seedlings. This observation indicates that the resin had migrated from surrounding tissues to the bark, and that this created a chemical and physical barrier to the insects. However, bark tissues might be simultaneously undergoing de novo synthesis of resin terpenoids (Gref & Ericsson, 1984). The sample material was quite low for the non-gnawed seedlings, and therefore the results should be considered tentative.

This is one of the first experiments to demonstrate the potential effect of MJ on plant–herbivore interactions in conifers. Secondary compounds (terpenoid resins) are important in plant–herbivore interactions and act as feeding or oviposition deterrents to generalists and nonadapted specialists. We hypothesize that changes in the anatomy or chemistry of plants after MJ treatments, such as tissue resin acid composition, can have significant effects on herbivore performance. An increase in individual secondary compounds and changes in the tissue quality consumed by the insect may depress the growth of insect populations in their natural environments. The resource balance of the seedlings seemed to be associated with the growth and resin acid concentrations that followed the prediction obtained from the growth-differentiation balance hypothesis (Herms & Mattson, 1992). Furthermore, MJ did not significantly alter the monoterpene composition of the seedlings, so it seems that it is MJ that accounts for the largest variation in the resin acid characteristics, whereas changes in the individual resin acid compounds seem to be tissue-specific. This conclusion is also supported by other studies (Sallas et al., 1999), where the planting of Scots pine seedlings changed resin acid concentration to a greater extent compared to monoterpene concentration in needles.

MJ is known to induce the production of defence-related compounds and resistance against several herbivores in agricultural crops, such as tomato (Thaler, 1999; Redman et al., 2001), potato (Rivard et al., 2004), and amaranth (Delano-Frier et al., 2004). Plants are typically challenged to multiple stresses at the same time in natural environments, and it may therefore not be possible to simultaneously increase defense, e.g., by MJ applications against attack by both herbivores and pathogens (Thaler et al., 1999). However, MJ affects both above- and below-ground plant parts and it is known that it also suppresses pathogen growth (Kozlowski et al., 1999); moreover jasmonic acid may assist mycorrhization in Norway spruce (Regvar et al., 1997). Furthermore, jasmonates are involved in many plant processes such as seed germination, fertility, senescence, and yield, as well as responses to environmental stresses, such as drought and salinity. In addition, MJ can up-regulate the genes that are involved in secondary metabolism, cell-wall formation, and jasmonate biosynthesis, whereas the genes involved in photosynthesis are down-regulated (Cheon & Yang, 2003). Therefore, further studies are needed in different plant species to determine how MJ treatments can modulate plant vigour, secondary compounds, and the ecophysiology of plants exposed to several simultaneous stresses under field conditions.

The exogenous application of MJ to Scots pine seedlings evoked major effects, indicating that MJ was responsible for a decline in both growth and physiological parameters. Height and diameter growth, as well as shoot- and root fresh biomasses were observed to be reduced only at the higher MJ concentration, but new root length was less affected. A similar reduction in plant vigour has previously been observed in other plant species (Redman et al., 2001). A decrease in photosynthesis was only observed in the seedlings treated with the higher MJ concentration, whereas no effect was seen in stomatal conductance. This agrees with the studies of Hristova & Popova (2002), who observed a decline in the photosynthesis of barley leaves, but not with the findings of Suhita et al. (2003), who noted that the stomatal opening of tree tobacco was suppressed by the presence of abscisic acid or MJ. These results indicate that a low dose of MJ applied exogenously on seedlings is not detrimental to shoot and root development or to plant physiology. In contrast, seedlings treated with a high dose of MJ show reduced growth and phytotoxic effects, such as needle browning, that might affect insect herbivory secondary to decreased plant vigour.

We found a formation of axial resin ducts on Scots pine xylem after MJ treatments, which has also been observed in other conifer species (Franceschi et al., 2002; Martin et al., 2002; Hudgins et al., 2004). Pinus species generally have four to five axial resin ducts per square mm of xylem (Wu & Hu, 1997), as we also observed in our study, but seedlings treated with higher MJ concentrations had over fivefold that number of resin ducts in the newly developing xylem. These anatomical defence reactions are probably a very important factor contributing to the decreased insect performance, and the anatomical changes may also affect other bark/phloem or xylem feeding herbivores. After the MJ treatments, the amount of secondary xylem and developed tracheids were considerably smaller, especially in the plants treated with the higher MJ concentration. Vascular cambium is thus affected by the MJ, which clearly affected cell expansion and might also influence cell length, cell wall chemistry, and cell wall structure. Therefore, the role of MJ on plant structural characteristics and ecophysiology requires clarification.

We conclude that the exogenous application of MJ to Scots pine seedlings can elicit chemical and structural plant defence by activating complex induced defence reactions at the whole plant level. MJ can provide potential resistance against herbivores; however, the timing and concentration of MJ treatment is crucial for shoot elongation, and seedlings should be sprayed after the annual flush. Acquired insect resistance after MJ application might also be achieved in more mature Scots pine trees and is thus a factor which should be examined in the future.


The Graduate School in Forest Sciences and Academy of Finland (decision no. 202300) have financially supported this research. We thank Mervi Ahonpää and Riitta Toivonen from the Finnish Forest Research Institute for help in the insect tests and Jaana Rissanen, Sari Hyttinen, Maarit Lauronen, and Juhani Tarhanen for their valuable help with the chemical analyses, Eija Rahunen with the anatomical preparation, Toivo Kuronen with the field experiment, and Ewen MacDonald for revising the language.