Introduction

A substantial proportion of deciduous tree species growing in the temperate zone of the Northern Hemisphere produce seeds with large energy reserves that allow their seedlings to successfully compete with other plants. However, such heavy seeds are also a plentiful food resource used by many animals, both invertebrate and vertebrate, which results in high pre- and post-dispersal seed losses (e.g., Jimenez-Pino et al. 2011; Bonal et al. 2012; Bieberich 2016). High seed predation together with abiotic factors in some habitats are able to prevent successful regeneration of forest trees (Pulido and Díaz 2005). Oaks (Quercus spp.) are widely distributed Palearctic species which have large seeds in the form of acorns. The main pre-dispersal predators which strongly limit the reproductive success of oaks are insects, mainly Curculio beetles (Coleoptera: Curculionidae) and Cydia moths (Lepidoptera: Tortricidae) which infest a large proportion of the produced seeds (e.g., Gibson 1981; Jimenez-Pino et al. 2011; Bonal et al. 2012). Even if the seeds do not completely lose their ability to germinate, the fitness of seedlings from such infected seeds is significantly reduced by insect feeding compared to sound seeds (Bonal et al. 2007; Perea et al. 2012). Such significant losses during reproduction have caused a number of morphological and chemical characteristics to evolve that serve as defense mechanisms against seed predators. One of these basic tree strategies is the occurrence of mast years, alternating with periods of very low seed production (Selås 1997), which are spatially synchronized over large areas (Koenig et al. 2013). This strategy reduces the availability of food in the years of limited seed production, which results in a decrease in the number of seed predating animals.

However, such a strategy is not fully effective against polyphagous species that in the absence of one species are able to develop in the seeds of other species. Examples of such species are the moths Cydia splendana and Cydia fagiglandana which can develop in the seeds of relatively numerous trees such as oaks (Quercus spp.), Sweet Chestnut (Castanea sativa) and European Beech (Fagus sylvatica) (Bovey 1966; Bradley et al. 1979). North of the geographical distribution of Sweet Chestnut, the larvae of C. fagiglandana feed on beech nuts and the larvae of C. splendana on acorns. Both moth species are found on Sweet Chestnut in southern Europe, where they cause significant damage (Bovey 1966; Bradley et al. 1979).

In their initial colonising phase, invasive species often take advantage of the inability of existing herbivores and pathogenic species to control them. During this period, invasive species should impact negatively on herbivores, because the invasive species compete for resources and space with native plants which are the natural food resources for the herbivores (Agrawal et al. 2005; Liu and Stiling 2006; Huang et al. 2010; Dai et al. 2014). Some interactions with local animals may be positive for invasive plant species and facilitate faster colonization, for example the long distance dispersion of Northern Red Oak (Quercus rubra) by the European Jay (Garullus glandarius) (Myczko et al. 2014). But, the longer they have been in an area the greater the chance that local enemies will have adapted to use the invasive species as a food source (Shea and Chesson 2002). Additionally the probability of successful colonization an exotic tree in the second range by the native insects increase with presence of congeneric native trees in environment (Roques et al. 2006; Branco et al. 2015). A large proportion of new enemies are recruited from polyphagous species (Roques et al. 2006; Branco et al. 2015). Therefore, the appearance of additional non-native tree species in the environment does not always result in a reduction in niche quality arising from a decrease in natural food resources replaced by invasive species. However, if herbivores adapt to feeding on invasive plant species the quality of the niche can increase because of an increased niche breadth. Adapting to a new food source can reduce competition for nutrients between species that compete for traditional food resources provided by the old host. Such a situation exists in Central Europe where the invasive species Northern Red Oak competes with native oaks (Major et al. 2013; Woziwoda et al. 2014). The occurrence and natural regeneration of Northern Red Oak in forests significantly reduces native species richness and abundance. No vascular plant species is known to benefit from co-occurring with Northern Red Oak, moreover the abundance of this invasive oak strongly limits natural restocking of all native woody species (Woziwoda et al. 2014).

The main aim of our study was to compare the use of acorns of two species of oak, one native and one invasive, to determine if there has been an adaptation to use the new oak species as a foraging source. We hypothesize that they may share the same acorn herbivores. Additionally we suggest that this information has important ecological implications for the population dynamics of the native herbivorous species by increasing their niche breadth. Moreover, indirect effects on native plant species are also possible by increasing the infestation of seeds because of a higher more stable feeding niche of herbivores.

Methods

We selected five areas in central Wielkopolska, Poland (52°26′–52°36′N; 16°48′–17°03′E) which were separated from one another by 3.6–21.4 km. In each area, mature trees of Pedunculate Oak and Northern Red Oak both grew together, and in each area we randomly chose 10 fruiting trees of each species and collected 100 acorns from each tree. We collected fallen acorns on 21 September 2014 during the period of maximum seed rain. We decided on this strategy because despite the different development strategy of acorns (taking one or two years) seed rain takes place at the same time. Acorns from each species and each area were bulked into a container of 1000 acorns. Acorns where kept in the laboratory at 10 °C and each day we removed larvae that had emerged from them. Larvae were collected from each container until two consecutive days produced no larvae; this took 41 days. For each larvae we recorded the species of oak, the number of days since the beginning of the experiment, their fresh weight and determined the genus or species. Larvae were weighed to a accuracy of 0.1 mg using an analytical balance (Sartorius® BP 160P), no later than 24 h after leaving the acorn. Because of the absence of morphological features to determine to species level, Curculio larvae were recorded to genus level only. The species of genera Curculio occurring in Poland and related to oak species are: Curculio glandium, Curculio venosus, Curculio pellitus and Curculio elephas (Burakowski et al. 1995). However, in the case of Cydia caterpillars we determined the species using the colour of the body and the number of crochets on prologs according to Brown and Komai (2008).

Statistical analysis

For each insect species/group, Chi square contingency tests were used to compare numbers of larvae between the two Quercus species. To test differences in the time to emerge for larvae between Pedunculate Oak and Northern Red Oak acorns we used t tests of each species (Cydia) or genus (Curculio). We used GLMs with a Gaussian error distribution to analyse larval weight in relation to two predictor variables: oak species and days until emergence, and their interaction. Analysis and graphs were produced using R 3.2.3 (R Core Team 2015) using the ggplot2 package (Wickham 2009).

Results

Cydia fagiglandana showed a significant preference to develop in acorns of Pedunculate Oak, however a substantial number of individuals also developed in Northern Red Oak acorns. In contrast, the number of C. splendana larvae did not differ significantly between the two oak species. However, almost all Curculio species developed in Pedunculate Oak acorns; less than 0.4% of the total identified Curculio larvae developed in Northern Red Oak acorns (Table 1).

Table 1 The number of larvae developing in Pedunculate Oak and Northern Red Oak acorns from 5000 acorns of each species
Full size table

There were no significant differences between the two oak species in the speed of larval emergence of C. fagiglandana. In contrast, C. splendana and Curculio species larvae emerged from Northern Red Oak acorns significantly later than from Pedunculate Oak acorns (Table 2). The larval weights (mean ± SE [mg]) developed in Pedunculate Oak acorns were as follows: C. fagiglandana (48.0 ± 0.6), C. splendana (36.1 ± 1.1) and Curculio species (35.2 ± 0.4). In Northern Red Oak the mean larval weights were: C. fagiglandana (46.0 ± 0.9), C. splendana (33.7 ± 0.9) and Curculio species (31.5 ± 4.3).

Table 2 The mean number of days spent post-collection by larvae in Pedunculate Oak and Northern Red Oak acorns from 5000 acorns of each species
Full size table

With the exception of C. fagiglandana, larval body weight wasn’t dependent on the number of days until emergence and in no insect species was there a significant difference in mean larval weight between the oak species (all p > 0.05, Table 3). For C. fagiglandana the average larval body weight significantly decreased with increasing days to emergence (Table 3).

Table 3 Factors affecting the body weight of the larvae of moths and beetles
Full size table

Discussion

Acorns of both Pedunculate Oak and Northern Red Oak are hosts for the development of C. fagiglandana, C. splendana and also Curculio weevil larvae. However, there are significant differences in the prevalence of these invertebrates in the acorns of the two studied oak species. Larval stages of the moth species C. fagiglandana and C. splendana occur in considerable numbers in Northern Red Oak acorns, and for C. splendana there was no significant difference in numbers between the two oak species. The use of Northern Red Oak acorns may help to avoid competition with Curculio weevil larvae, which has been described previously in the case of C. fagiglandana (Jimenez-Pino et al. 2011). In that study, inter-specific competition with the weevil C. elephas was the factor that most affected the spatial distribution of C. fagiglandana larvae (Jimenez-Pino et al. 2011). Apart from reduced competition with other seed eating insects the ability to develop larvae in the tissues of another host plant can increase the stability of the population. Our forest trees are characterized by the presence of mast years and years with poor production of seeds (Kantorowicz 2000). As a result, there are alternate years suitable for the development and increase of the invertebrate population and years when the availability of seeds is very limited causing a decrease in population size (Selås 1997). Of course, animals have evolved mechanisms to survive such adverse periods; an example is prolonged diapause in Curculio weevils which stabilizes the local population by surviving the unfavorable period in the pupal stage and developing into the adult insect in subsequent years (Menu, Roebuck and Viala 2000; Menu and Desouhant 2002). However, a lack of data on the occurrence of this mechanism in the case of Cydia spp. suggests a different strategy based on adapting to a greater number of host plants. More larval host species results in a greater stability of food resources and makes a drastic reduction of the population, due to the lack of fruiting of suitable food plants in some years, much less likely. Thus, the introduction by humans of additional invasive species into the environment has increased the stability of the local population of C. splendana and C. fagiglandana. An advantage for insects able to develop in acorns of the invasive species Northern Red Oak is that the acorns mature after 2 years, unlike native Oaks in this part of Europe where maturation takes place in the same growing season as flowering. This difference has a significant impact on the occurrence of seed years (Koenig and Knops 2000) and the presence of oaks of asynchronous acorn production helps to stabilize populations of generalists (Fukumoto and Kajimura 2011). Thus, the introduction of invasive species into the environment may have a positive impact on acorn-developing insects. Additionally, this process can be reinforced by anthropogenic fragmentation of habitats, significantly increasing seed production and seed size (Bonal et al. 2012). Moreover, climate change may also disrupt the occurrence of mast years (Selås 1997).

The results of our study show that Curculio weevils in Northern Red Oak acorns in Poland are extremely rare. Therefore, seed losses associated with feeding by insects in acorns are several times lower in comparison with the native oak, in which infestation of acorns by Curculio weevils is many times greater than that of other species of insects. The development of larvae in the acorn does not always mean a failure to germinate. However, infestation by an insect causes a significant decrease in the ability to germinate, for example germination of sound Northern Red Oak acorns was 86% but only 26% in the case of weevil-damaged acorns. In addition, seedlings developing from infected acorns had a significantly lower fitness (Lombardo and McCarthy 2009). In the case of natural regeneration, the development of Curculio larvae in the acorn not only directly affects the plant’s survival, but also indirectly reduces its fitness by reducing the likelihood of seed dispersion by animals. Infested acorns are more often dispersed than those from which larvae have emerged, however both are still significantly less dispersed than sound acorns (Perea et al. 2012). However, it should be remembered that dispersion by animals is also affected by other factors, such as the species of acorn (Myczko et al. 2014; Bonacchi et al. 2015).

Marginal infestation of acorns by Curculio weevils supports the enemy release hypothesis during the colonization of new areas by invasive species which is often assisted by the lack of natural herbivores and pathogens, as well as the inability of naturally existing herbivores and pathogenic species to control the invasive organisms (Agrawal et al. 2005; Liu and Stiling 2006; Huang et al. 2010; Dai et al. 2014). An additional argument in the case of Northern Red Oak enemy release hypothesis is that, in its natural range, Northern Red Oak averages over 52% acorns damaged by insects, mainly by Curculio weevils (Gibson 1981). However, these were different weevil species than occur in Europe. Currently available data do not support the conclusion that in the long term European species of Curculio weevils will adapt to develop in the acorns of Northern Red Oak. However, the very few Curculio weevil larvae in our sample of Northern Red Oak acorns suggests the possibility of such a scenario.

One of the most important features that is easy to measure in the case of insect larvae is their weight, this measurement is very strongly correlated with fitness. The higher weight of pre-imaginal stages strongly influences the fecundity of females (Calvo and Molina 2005) and also positively influences reproductive success of males (e.g., Klingenberg and Spence 1997). The weight of larvae emerging from acorns reflects their fitness and also the quality of their food. Our results indicate no significant effect of the species of oak on the larval weight emerging from acorns, which means Cydia species can develop in the non-native host species to reach the same level of fitness as in the native oak. However, achieving this mean weight takes significantly longer in the case of C. splendana and Curculio weevils (Table 2) in the non-native oak species, which means a lower performance. C. splendana and Curculio spp. take longer to develop on Northern Red Oak acorns in comparison with Pedunculate Oak acorns, and they will also then be exposed to increased probability of consumption by animals foraging on acorns, such as wild boars and small mammals (Focardi et al. 2000). The results shown in Table 3 indicate that, with the exception of C. fagiglandana, the final mean weight of insects was not influenced by duration within the acorn. However, in C. fagiglandana the average weight of larvae significantly decreased with the number of days elapsed. This suggests that C. fagiglandana prefers to leave acorns earlier, reducing the risk of being predated by animals feeding on acorns even if the larvae had not reached optimum weight.

In conclusion, we clearly show that a new element of the European flora is currently used as a foraging substrate by local moths and beetles, and can probably stabilize local trophic niches of C. splendana and C. fagiglandana.