In this study, we showed that simple modification of leaves, that is, leaf rolling, caused marked changes in the fate of immature attelabid weevils related to natural enemies. In particular, a decrease in the parasitism rate by Eulophidae and the egg predation rate contributed to the increase in the survival rate. The fact that parasitism rate by Eulophidae in experimentally rolled leaves was 0% compared to that in unrolled leaves (36.6%) suggests that leaf roll acted as an “insuperable barrier” against Eulophidae. This result is consistent with our previous study revealing that leaf rolling species in Attelabidae were less parasitized by eulophid wasps16. The reason why eulophid wasps do not parasitize eggs or larvae in experimentally rolled leaves may be explained by two different hypotheses: failure-in-access and failure-in-finding. The failure-in-access hypothesis is that eulophid wasps can find hosts and attempt to parasitize, but the leaf roll acts as a structural barrier and eulophid wasps cannot reach weevil eggs or larvae. Considering that leaf rolls in this experiment were loosely rolled and oviposition sites would be easy to access, the plausibility of this hypothesis seems relatively low. On the other hand, the failure-in-finding hypothesis is that eulophid wasps cannot find hosts in leaf rolls because they cannot recognize a “rolled leaf” as a target structure containing hosts. Eulophidae are known as one of the dominant parasitoids of various leaf miners such as leaf mining moths, flies, sawflies, or beetles20,21. Some parasitoids of leaf miners have been reported to have evolved specific visual searching traits for leaf miners during flight22,23,24,25. For example, parasitoids were more attracted to the leaves with many leaf mines using visual cues25. In addition to visual cues, parasitoids of leaf miners also use chemical and vibrational cues for host searching, similar to other parasitoids26,27,28. Considering that in the present study, chemical and vibrational cues would not differ between experimentally rolled and unrolled treatments, changes in the visual cues likely affected the search success of eulophid wasps. In our study, of the 36.6% of eggs and larvae from unrolled leaves parasitized by Eulophidae, 25.6% was attributable to egg parasitism. This means that mine shape would not be an important visual cue in this case because no mine existed on leaves in the egg stage. Thus, eulophid wasps might have a host searching image of “cut-off leaf on the ground”, which was basically flat and thin, and leaf rolls were not recognized as a host because the shape differed from the searching image. The protective barrier effect of leaf roll for inner insects has been reported previously1,5,6,7,8,9,10,11,12,13,14,15. However, this study suggests that the leaf roll effect is not only a structural barrier but also a “visual modification” itself. In order to confirm this visually protective effect of leaf rolls, further experiments controlling leaf shapes in various patterns and comparing parasitism rates among treatments are needed. In this study, important information on the timing of parasitoid attack was also indicated: Mymarid wasps and Ophioneurus sp. were suggested to attack hosts on the tree shortly after weevil oviposit into the leaf and before the leaf was completely cut from the tree. The time from oviposition to leaf cutting is reported to be approximately 30 min in a rhynchitin non-leaf-roller species, Deporaus septemtrionalis29. It is surprising that parasitoids can finish their host finding and oviposit in such a brief time and probably represents a product of the arms race between parasitoids and weevils16. In addition, the weevil behavior of cutting leaves from the host tree might contribute to avoidance of heavier parasitism rates. If leaves including eggs remain suspended from the tree, the success rate of parasitism would most likely increase due to prolonged opportunity for parasitoids to attack.
The other factor related to the increase in survival rate of immature weevils was the decrease in mortality due to egg predation. Our results indicate the presence of predators on the ground that feed primarily on eggs in leaf tissue instead of on eggs in leaf tissue within leaf rolls. Few studies have revealed predators of leaf miners in the egg stage, and most studies of leaf miner mortality focus on larvae30. Digweed31 reported potential egg predators of a leaf mining sawfly as spiders, staphylinid beetles, coccinellid beetles, Hemiptera, and thrips. However, in this study, we should consider potential egg predators not on the tree but on the ground. From the soil meso-organisms and macro-organisms lists, predators and opportunistic predators in the litter would be mites, Opiliones, Isopoda, centipedes, millipedes, ground beetles, and spiders32. Furthermore, ants and dipteran larvae could also be considered as potential predators or opportunistic predators33,34. Eggs in unrolled and rolled leaves were protected in the leaf tissue, but leaf decomposition or egg dislodgement by soil organisms might occur more easily in unrolled leaves than in rolled leaves. Such decomposition or dislodgement causes exposure of eggs and a higher risk of predation. Thus, eggs in rolled leaves might show lower predation rates than those in unrolled leaves.
In contrast to predation, weevil mortality due to herbivory, especially by moth larvae, increased in experimentally rolled leaves compared to unrolled leaves. Thus, leaf rolls are not only protective refuges but also potentially risky hiding places for immature weevils. Our observations could be attributed to the fact that leaf rolls were preferred by herbivorous moth larvae as shelters; leaf shelters, that is, leaf rolls, leaf galls, leaf folds, or leaf ties, are often preferred and secondarily used by several insect species, sometimes providing them with a protective effect13,35,36,37,38. The reason why previous studies on the effect of leaf rolls did not detect the negative effect of herbivory could be that the observed leaf rolls were constructed by lepidopteran larvae that could escape herbivory and construct new leaf rolls. In the litter on the forest floor in Japan, lepidopteran larvae, such as those belonging to Blastobasidae or Tineidae, crawl while searching for fallen leaves to feed on. In an attelabid species, Cycnotrachelus roelofsi (Attelabinae), Neoblastobasis spiniharpella (Blastobasidae) larvae were found to infest the inside of leaf rolls; as a result, weevil larvae sometimes died of direct infestation or the lack of food34. In such cases, leaf roll construction had a negative effect on immature survival. However, in the species of Attelabinae, leaf roll construction is crucial to avoiding egg parasitoids, while mortality by herbivory is not so high34. Thus, the protective effect of leaf rolling against egg parasitoids exceeds the negative effects of herbivory. Further, leaf roll construction using excessive leaves by some Byctiscini species (Attelabidae, Rhynchitinae) may be a counter evolution to avoid mortality by herbivory. Various lepidopteran species and dipteran species emerge from leaf rolls of some Byctiscini species consuming leaf rolls (Kobayashi C, unpublished data), and competition for leaf roll resources sometimes causes larval death because weevil larvae cannot exit leaf rolls and search for new leaves. Thus, excessive leaves in the leaf roll may save weevil larvae from dying from food loss or infestation because of herbivory.
Regarding environmental stress, we did not detect any effect of leaf rolls. This may be because immature weevils in unrolled leaves were not directly exposed to environmental stresses due to their leaf mining habit. Therefore, both rolled and unrolled treatments experienced the same environmental conditions.
In summary, the survival rate of the attelabid weevil in this study was significantly increased by leaf rolling. Thus, this study suggests that selection pressure to evolve leaf rolling behavior still exists in the natural population, at least in Attelabidae. However, whether the leaf rolling behavior evolves will depend on the balance of positive and negative effects of leaf roll, that is, the degree of pressure exerted by parasitoids, predators and herbivores. Furthermore, constructing leaf rolls incurs energy costs and time for oviposition. Considering that most Deporaini species are less than 5 mm in length, the behavioral costs for rolling leaves would be high. In Deporaini, few species evolved leaf rolling traits independently, while the others were leaf miners in cut leaves and did not roll leaves17. Although the total survival rate was higher in rolled leaves than in unrolled leaves, contradictory effects of leaf rolls added to construction costs may explain this sporadic evolutionary pattern in leaf rolls of Deporaini. If leaf rolling traits have a mostly positive effect on egg and larval survival, leaf rolling behavior may have evolved more frequently or further diversification of leaf rolling species may have occurred.
Source: Ecology - nature.com
