No evident sex-biased herbivory but phenological and interannual variation in Fuchsia parviflora (Onagraceae) a dioecious Neotropical shrub

Herbivory may have a great impact in plants, reducing plant biomass and fitness. Herbivory patterns may differ among genders in dioecious species, affecting male and female plants differentially, and it has been suggested this is a consequence of differences in resource allocation among genders. In this study, we evaluated herbivory intensity and mechanical defenses among genders during two years and during different phenological stages in populations of Fuchsia parviflora, a dioecious shrub. We also evaluated generalist and specialist herbivore gender preferences through a cafeteria experiment. No evident herbivory patterns were detected among genders using both the proportion of leaves affected by herbivory and the proportion of leaf area removed, except during one year in two populations where females had a significantly higher proportion of leaves affected by herbivory. This pattern is similar to the specialist herbivore preference detected in the cafeteria experiment. We also found significant differences in herbivory intensity among phenological stages, populations, and years, highlighting the relevance of continuous sampling to depict the general patterns of herbivory in dioecious plants.


Introduction
Herbivory, the removal of leaves or other plant tissues by animals, is one of the most widespread and oldest interactions that may have a great impact in reducing plant biomass and fitness (Hawkes and Sullivan 2001;Turcotte et al. 2014). In contrast to the great majority of flowering plants, which bear hermaphrodite flowers, dioecious species have individuals with only male or female flowers. Genders of dioecious plants, besides differing in primary and secondary sexual traits, may differ in less obvious characters such as phenology, growth rate, secondary compounds, nitrogen content, floral display, floral scent, nectar volume and concentration among others (Obeso 2002;Ashman 2009;González et al. 2018). Moreover, such differences may also affect biotic interactions such as pollination and herbivory (Ågren et al. 1999). Sex-biased herbivory (generally male biased) has been widely documented in the literature (reviewed in Cornelissen and Stiling 2005; but see Avila-Sakar and Romanow 2012), and has been suggested to be a consequence of differences in resource allocation among genders (Lloyd and Webb 1997; but see Midgley et al. 2019). For example, females may have slower growth rates than males since they generally invest more in reproduction (flower, fruit, and seed production ;Obeso 2002;Cepeda-Cornejo and Dirzo 2010), and therefore, the cost of photosynthetic tissue loss would be higher compared to males. As a consequence, females would invest more in chemical and/or mechanical defenses (Lloyd and Webb 1997;Delph 1999;Wang et al. 2014). However, if females reallocate more resources to tolerance (e.g., activation of apical meristems) to mitigate negative effects by herbivores, we may expect they would have higher herbivory levels than males (Mutikainen et al. 1994).
Although reproductive allocation in females is generally higher, physiological mechanisms involved in resource acquisition and allocation would evolve in such a way that differences in reproductive effort among genders would be minimized (Watson 1995;Avila-Sakar and Romanow 2012). For example, females need additional carbon for fruit and seed production, but at the same time, males require more nitrogen for pollen production especially in wind pollinated plants, which is costly (Harris and Pannell 2008). In some species, males can have many times more flowers than females, possibly approximating the cost of fruit production (Bullock 1984;Queenborough et al. 2007). However, if most resources are allocated to reproduction in females, they could have lower allocation not only to growth but also to defense (Avila-Sakar and Romanow 2012).
It is currently thought that females have developed more resistance traits than males, and studies where herbivory patterns among genders in dioecious plants have been addressed indicate that generally males have higher herbivory levels. Although some years ago, male-biased herbivory was proposed by some authors as a generalized pattern in dioecious plants (Ågren et al. 1999;Cornelissen and Stiling 2005), a recent meta-analysis found no sex-biased herbivory or secondary defense when controlling for the taxonomic group (Sargent and McKeough 2022).
Studying spatiotemporal variation of herbivory could help to have a more general understanding of plant-herbivore interactions, since the intensity of herbivory may change depending on the population, plant phenological stage, or year, probably differing due to precipitation patterns or other environmental conditions (Filip et al. 1995). However, this component has been rarely evaluated in dioecious species (Ågren 1987;Alliende and Harper 1989). Moreover, herbivory studies in dioecious species have been conducted mainly in temperate zones in a few plant families. In order to reduce this bias and incorporate the inter-annual and phenological variation, in this study we addressed the following questions: (1) Is there a difference in the herbivory intensity suffered by male and female plants in Fuchsia parviflora? (2) How do the patterns of leaf herbivory changes over plant phenological stages and between years and genders? (3) How much variation can be detected in leaf herbivory and mechanical defenses (trichomes) across populations? and finally (4) Do specialist and generalist herbivores differ in their preference for leaf tissue from male and female plants?

Study system
Fuchsia parviflora (Onagraceae) is an endemic species from Mexico that inhabits ravines in pine-oak forests between 1550 and 2500 m asl, encompassing the states of Michoacán, Guerrero, Jalisco, Nayarit and Estado de México. It is a hummingbird and bee pollinated dioecious shrub of 1.5-4 m (Breedlove 1969;González et al. 2018), with smaller tubular red flowers in females (3.2-5.7 mm long) than in males (8.5-11.2 mm). Fruits are rounded berries with 14-20 small seeds 1.9-2.3 mm long. Opposite, oblanceolate to ovate leaves (30-75 mm long) are pubescent in both surfaces (Breedlove 1969). The flowering season is from June to December, with a peak in August (González et al. 2018). Their fructification period is from August to December, with a peak during October (González et al. 2018), and their leaves are shed from January to March, suffering a total or partial lack of leaves in March.

Study site and field procedures
Four populations of Fuchsia parviflora were located in the central part of Michoacán, Mexico within three municipalities which are part of the Pátzcuaro and Zirahuén lake basins: Pátzcuaro (Residuos Sólidos), Salvador Escalante (Agua Verde and Zirahuén), and Quiroga (Chupícuaro; SI. Fig. S1).
Twenty females and 20 males from 1 to 4 m tall were tagged in each population. Herbivory intensity was evaluated as the proportion of leaves affected by herbivory per plant and as the proportion of leaf area removed by herbivores. We decided to estimate leaf herbivory using both methods because they offer complementary information, besides, even whether the proportion of removed area was higher for one gender, it might be lower at the individual plant level because there might be fewer leaves that have been damaged in total. To estimate the proportion of leaves affected by herbivory per plant, the total number of leaves and the total number of damaged leaves were quantified in each plant. We considered leaves with herbivory as those with missing parts of the leaf blade such as holes, damage to the leaf edge or skeltonization leaving only the veins. To estimate the total number of leaves in plants higher than two meters, the total number of leaves on a branch was multiplied by the number of branches. Surveys were performed during the five phenology stages during 2017, approximately every two months: before flowering (June), during peak flowering (August) during peak fruit production (October), after flowering and fruit production (January), and during leaf shedding (March).
To determine the proportion of leaf area removed, 10 damaged leaves per plant were collected at random (from the top, intermediate, and bottom part of each shrub) during each census and transported in a cooler to the laboratory. To obtain the proportion of area removed, the collected leaves were scanned to estimate their total area and the leaf area removed with the aid of Image J 1.x (Schneider et al. 2012). During 2018 herbivory was estimated with both methods only during October, as it was the month with highest herbivory during 2017 (Fig. 1). We understand that the two measures we used to estimate herbivory levels are not completely precise, as all the used methods to estimate herbivory have some inaccuracy, as far as we know. In addition, in contrast to most studies where only one proxy is used, we used two proxies to estimate herbivory.
Trichome density was estimated in young leaves only (2 months old approx.), because in older leaves they are lost. To this end, a leaf from 10 males and 10 females was collected in each population and the number of trichomes was estimated in a 1 × 1 cm square, both in the adaxial and the abaxial leaf surfaces using a magnifying glass.

Cafeteria experiment
In October 2018, 10 caterpillars from an unidentified butterfly but recognized as the main herbivore of F. parviflora and 10 generalist crickets (Pyrgomorphidae), that were also observed feeding at the leaves from the study species, were collected from the Chupícuaro population. All caterpillars were of similar size, belonging to the same larval stage. Before the experiment, both caterpillars and crickets were starved for 12 h. Then, caterpillars were placed in Petri dishes with humid filter paper to keep moisture, and crickets in plastic jars with little holes in the covers. Two leaves of approximately the same size, one from a female and one from a male plant, were placed at the extremes of each Petri dish or plastic jar. We used five male and five female plants for the experiment previously extracted from three populations (4 plants from R. Sólidos, 4 from Chupícuaro, and 2 from Agua Verde) and kept in pots inside a shade house. Leaves were cut with a scalpel and collected with forceps and gloves to avoid any contact that could have an affect on the herbivore's preference. Caterpillars and crickets were placed in the center and observed until they chose one of the leaves. Once the insect's selection was observed, they were allowed to feed on the leaves to calculate the total area removed. Each caterpillar and cricket was offered leaves of a different female and male plant every day for 5 days.

Statistical analyses
To determine differences between males and females, phenological stages and populations in the proportion of leaves affected by herbivory and the proportion of area removed per leaf during 2017, we conducted generalized linear mixed models (GLMMs) using binomial distribution and logit function due to the non-constant binomial variance of proportions. The first group of models were developed for: (i) the proportion of leaves affected by herbivory and (ii) the proportion of area removed as dependent variables. The affect of gender, phenological stage (census at which the data were taken) and population, as well as all their interactions were included as fixed factors, while plant nested within population was included as a random effect (Pinheiro and Bates 2000). In a second group of models, we considered only data from October 2017 and 2018, as this month was the one with the highest herbivory intensity (see "Results" section). The GLMM included the affect of the year, gender, and population, as well as their interactions as fixed factors, while the proportion of leaves affected by herbivory and the proportion of leaf area removed were the dependent variables. We also included the plant nested within population as a random effect, while the proportion of leaves affected by herbivory and the proportion of leaf area removed were the dependent variables. In both models, we included population as fixed factor because we were interested to test differences among populations. For all response variables, we developed saturated models which included all main terms and their interaction. We then remove the non-significant terms and obtained the minimal adequate model (Crawley 2012) and finally, performed multiple comparison tests.
To determine whether the number of trichomes differed among genders, a general linear model was performed considering gender, population, and the side of the leaf (adaxial and abaxial surfaces) as fixed factors, as well as the number of trichomes as the dependent variable. The number of trichomes were square root transformed to normalized the data before the analysis.
Finally, differences between genders in the number of leaves chosen by each caterpillar and cricket and in the proportion of area removed were analyzed by means of a paired sample t-test (males vs females). All analyses were conducted in R v. 3.3.2 (R Development Core Team 2016), while for mixed models we used the lme4 package (Bates et al. 2015). Multiple comparisons were conducted with the multcomp package v1.4-16 (Hothorn et al. 2008).

Herbivory intensity between female and male plants
The proportion of leaves affected by herbivory in F. parviflora differed between genders and phenological stages (Table 1A; Fig. 1). However, differences between genders were generally minor, with males only statistically more affected by herbivory in the March assessment. A wide variation in the proportion of herbivory at the plant level was observed (10 to 90%) over the time the plants had leaves. During June (before flowering) the proportion of leaves affected by herbivory was close to zero (females 0.01% ± 0.02; males 0.01% ± 0.01; Fig. 1). Damage was greatest in October (peak in fruit production) when there was a higher proportion of leaves affected by herbivory both for females (mean ± SD hereafter, 31% ± 0.26) and males (28% ± 0.25), compared to the rest of the census time. Damage in January and March was intermediate, generally significantly less than in October but significantly more than in June and August.
The area removed between genders was similar (Table 1B), but the proportion of leaf area removed had low, though significant variation among phenological stages ( Fig. 2; Table 1B) and populations but this was not consistent. For example, a single population with a low area removed in a census may have the largest in the following census (Fig. 2). The lowest value was registered at Residuos Solidos in June (0.09 ± 0.02) and the largest at Agua Verde in March (0.21 ± 0.02). There was also an interaction between population and phenological stage.
When comparing the proportion of leaves affected by herbivory per plant in October of each year, significant differences were detected between years, populations, as well as the interactions year × population, and year × population × gender (Table 2A; Fig. 3). In 2017 a higher proportion of leaves affected by herbivory was detected in both genders (females 0.31 ± 0.26; males 0.28 ± 0.25) compared to 2018 (females 0.13 ± 0.18; males 0.09 ± 0.11; Fig. 3). The only difference between genders was observed in R. Sólidos population during 2017 where females had significantly higher proportion of leaves affected by herbivory (female R. Sólidos 0.45 ± 0.33, male 0.22 ± 0.18) and in 2018 at Chupícuaro (females 0.27 ± 0.24, males 14 ± 0.14; Fig. 3). Consistently, we  Fig. 4), as well as the interaction year × population. Although no differences were detected between genders, we found significant interactions of gender × year and gender × population, as well as the three-way interaction (Table 2). This was mainly due to the higher leaf area removed in females than males in 2018 at the Agua Verde (females 0.13 ± 0.01) and at Chupícuaro populations (females 0.12 ± 0.01; Fig. 4).

Cafeteria experiment
Ten caterpillar and 10 crickets chose 26 and 48 times, respectively, the leaf of one gender. Some caterpillars died during the experiment and others pupated. Crickets kept choosing throughout the experiment and only two died on the fourth day. Caterpillars more likely chose females (16 times) than males (10 times, paired samples t-test; t = 2.7, d.f. 8, P < 0.05), however, there were no significant differences in the preference of the crickets for any particular gender (they preferred females 26 times and males 22). There were also no significant differences between genders in the leaf area removed by the caterpillars (females 0.68 ± 0.22; males 0.59 ± 0.22) or by the crickets (females 0.69 ± 0.28; males 0.74 ± 0.28).

Discussion
Although male-biased herbivory has been considered as a "general pattern in dioecious plants" (Cornelissen and Stiling 2005), a recent meta-analysis (adding 67 new studies) found no general sex-biased pattern when control for taxonomic group (Sargent and McKeough 2022). In our study, we only detected male-biased seasonal herbivory in the proportion of leaves affected by herbivory per plant in March (during leaf shedding) in 2017. Conversely, over a year period females had a significantly higher proportion of leaves affected by herbivory in two populations (R. Sólidos in 2017 and Chupícuaro in 2018), which agrees with the specialist herbivore preference (see below). Generally, females invest more resources to reproduction compared to males (Lloyd and Webb 1997;Delph 1999) despite the higher flower production in males (González et al. 2018;Ambriz et al. 2021; but see Midgley et al 2019). In female plants, the highest reproductive investment occurs after flowering, during fruit and seed production, which in terms of biomass represent a higher cost compared to males (Bullock 1984;Allen and Antos 1988;Delph and Herlihy 2012). A higher investment in reproductive biomass in females could compromise defense allocation, so that herbivory intensity would be higher in females as was found in two populations. It is likely that females were more attacked in general but having possible tolerance mechanisms (i.e., compensation through leaves production). This is slightly supported by the results of the proportion of leaf area removed (Fig. 3) but not really for other times and populations. According to our records, F. parviflora begins to shed leaves in January and by March, when higher levels of herbivory were detected in males, many plants had already lost many of their leaves. Although we could not assess differences in resource allocation to defense between genders, we would not expect plants to invest much in defense when leaves are about to be shed naturally. Even though no consistent sex-biased herbivory was detected among populations and phenological stages, a very wide variation in the proportion of herbivory at the plant level was observed (10-90%). For example, in the dioecious shrub Baccharis salicifolia, recent studies found stronger genetic variation effects at plant level, in the herbivore abundance than the affects of gender (Nell et al. 2018). The same working group found that genetic differences at plant level were more relevant than sexual condition in the terpene profiles, a secondary metabolite related to plant defense in plants ( Moreira et al. 2019). The wide variation in herbivory detected at plant level in F. parviflora, could be related to qualitative or quantitative differences in the chemical defenses, surpassing the differences between genders per se. Therefore, it would be useful to quantify the secondary metabolites in F. parvifolia and ideally the genetic variation at plant level. In addition, the micro-environmental conditions of each population could be an important factor in biased herbivory patterns. It is known that dioecious plants are more stress susceptible due to both biotic and abiotic factors (Obeso 2002). Populations where female-biased herbivory was observed, (Chupícuaro and R. Solidos) are located in disturbed sites, from cutting down trees to probably soil pollution, which could differentially influence the herbivore impact in each gender. In fact, we observed some female plants totally defoliated by caterpillars in Chupicuaro, 2-3 years before this study. However, the lack of a consistent male or female-biased herbivory pattern could be a consequence of no differences among genders in leaf nutritional content and/or secondary compounds concentration, as have been documented in other species (Galambosi et al. 2009;Espírito-Santo et al. 2012), since these characteristics are an important factor in the preference of herbivores.
Our cafeteria experiment suggested a female preference by caterpillars (that we suspect is the principal herbivore of F. parviflora). Crickets did not show any preference, which confirms their role as generalists. In a similar experiment in Salix cinerea, the herbivore Phratora vulgatissima and the omnivorous Anthocoris nemorum strongly and significantly preferred females instead of males (Kabir et al. 2014). In this study, most of the caterpillars and their eggs were observed in female plants, however, it is necessary to conduct an intensive monitoring to evaluate eggs and caterpillar's abundance, the adult choice for oviposition, as well as the identification of the herbivore and its interaction with F. parviflora in different populations.
In contrast, we observed temporal variation in herbivory patterns of F. parviflora, where the highest proportion of leaves affected by herbivory was recorded in October, which is related to a higher abundance of herbivores (caterpillars) on plants.
October is the fructification peak period where females allocate more resources to reproduction, which can affect their resource allocation both to defense and/or tolerance. In June there was less herbivory, since during that period leaves are initiated and young leaves are protected with trichomes that are lost when mature. It is known that trichomes protect leaf buds from herbivores, reduce the temperature and increase leaf reflectance, prevent water loss and reduce leaf abrasion (Wagner et al. 2004). However, the number of trichomes did not differ among genders but did among populations, probably due to variable light availability to which plants were exposed, that can affect their number or physical constitution. During 2017, we found a higher herbivory intensity than in 2018, even though the total precipitation was lower in 2017 (792 mm during 2017 vs 1114 mm during 2018). A reduction in rainfall may affect caterpillar's growth and survival in Pararge aegeria due to the decreased moisture content in host plants (Gibbs et al. 2011), so we had expected the opposite inter-annual pattern in herbivory.
Our results mainly suggest spatial and temporal variation in the herbivory intensity patterns in F. parviflora rather than gender differences. Few studies have quantified herbivory intensity throughout a year, it is usually quantified only during a period, which does not show the complete picture. Similarly, those data have been usually taken only in one year, when normal variation in climate and even global climate change would not likely have an effect. In almost all traits measured, there were differences between populations, so future studies need to consider microclimatic data, soil nutrient availability and evaluate secondary metabolites and nutritional content in leaves among genders and populations of F. parviflora, their variation during the year and over longer periods.