Introduction
Reproductive phenology, defined as the relationship between environmental variation and the reproductive cycle of organisms, is modulated by abiotic (e.g., temperature, precipitation, and photoperiod) and biotic (e.g., availability of food, predation risk, and reproductive opportunities) factors that mark the beginning and end of the life cycle phases (Varley et al. Reference Varley, Gradwell and Hassell1973; Forrest Reference Forrest2016). Environmental variation and resource availability influence the behaviours of foraging, migration, courtship, and copulation and determine the duration of each physiological stage (e.g., larval development, sexual maturation, and diapause; Varley et al. Reference Varley, Gradwell and Hassell1973; Takeda Reference Takeda2004; Bonal et al. Reference Bonal, Munoz and Espelta2010; Chuine Reference Chuine2010). By describing the reproductive phenology of species, it is possible to identify natural history traits related to the species’ mating system, patterns of emergence of the sexes, competition for resources, and sexual selection pressures, as well as to determine the reproductive strategy (r or K) and the number of generations presented per year (voltinisms).
In seasonal environments, the reproductive cycle and development of ectothermic organisms such as insects are associated with climatic changes over the course of the year, generating variation in the abundance and spatial distribution of the individuals among seasons (Chuine Reference Chuine2010; Encinas-Viso et al. Reference Encinas-Viso, Revilla and Etienne2012). In these environments, the reproductive activities of the species are limited to the season of the year with the most favourable conditions for seeking a mate, copulation, and development and survival of the new generation. These species are classified as univoltine – that is, having a single reproductive peak in the year. In insects, the onset of the rains marks the season of production and mating of new individuals (Menu Reference Menu1993; Bonal et al. Reference Bonal, Hernández, Espelta, Muñoz and Aparicio2015). In species considered univoltine, overlaps of generations or ages are not expected, and it is possible that all individuals enter diapause during the stage of the year with suboptimal environmental conditions, such as the period of lower precipitation or lower temperatures (Clark et al. Reference Clark, Geier, Hugues and Morris1978; Ishihara Reference Ishihara1999).
In annual insects, the males and females do not emerge simultaneously (Rhainds Reference Rhainds2010), thus influencing the population density, sexual proportion, and mating system of the species (Thornhill and Alcock Reference Thornhill and Alcock1983). For example, the phenological cycle of species in which the females emerge before the males (i.e., protogyny) is characterised by the emergence of sexually immature individuals, females with a longer time of sexual development that require various copulations over the course of their lives, as well as males that present a higher mortality (Van Timmerman et al. Reference Van Timmerman, Switzer and Kruse2000; Degen et al. Reference Degen, Hovestadt, Mitesser and Hölker2015). Differential emergence has also been associated with synchrony between the stages of sexual maturity in males and females. Reproductive synchrony is an evolutionary strategy that favours encounters between sexually mature individuals during the period that offers optimal environmental conditions for mating and development of the progeny (Thornhill and Alcock Reference Thornhill and Alcock1983; Van Timmerman et al. Reference Van Timmerman, Switzer and Kruse2000; Calabrese et al. Reference Calabrese, Ries, Matter, Debinski, Auckland and Roland2008).
In general, the reproductive phenology of the coprophagous beetles has been studied under controlled laboratory conditions, through observation of morphological changes in the reproductive structures of males and females during the process of sexual maturation (Martínez and Montes de Oca Reference Martínez and Montes de Oca1988; Hernández-Martínez and Martínez Reference Hernández-Martínez and Martínez2003; Huerta et al. Reference Huerta, Halffter and Halffter2005), while few studies have been conducted under natural conditions (Martínez and Montes de Oca Reference Martínez and Montes de Oca1994). Although variation in reproductive behaviour among species occurs (Halffter and Edmonds Reference Halffter and Edmonds1982), studies conducted in seasonal environments show that the greatest abundance and the reproductive period are associated with the warmest and most rainy season of the year (Kingston and Coe Reference Kingston and Coe1977; Martínez et al. Reference Martínez, Montes De Oca and Cruz1998; Bang et al. Reference Bang, Crespo, Na, Han and Lee2008; Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018). In order to reach sexual maturity, female coprophagous beetles, in addition to feeding, require an initial copulation to receive seminal liquid from the male that stimulates the physiological mechanism of maturation of the basal oocyte and provokes nesting behaviour (Cruz and Martínez Reference Cruz and Martínez1998). This suggests that the females require more time than the males to reach sexual maturity, as has been observed for various coprophagous beetles (Martínez et al. Reference Martínez, Montes De Oca and Cruz1998).
The objective of this study was to describe the annual reproductive phenology of Dichotomius satanas (Harold, 1867) (Coleoptera: Scarabaeidae, Scarabaeinae), a dominant species of the community of coprophagous beetles of the eastern Mexican cloud forest (Pineda et al. Reference Pineda, Moreno, Escobar and Halffter2005; Halffter et al. Reference Halffter, Pineda, Arellano and Escobar2007; Rös et al. Reference Rös, Escobar and Halffter2012), where it is particularly abundant during the rainy season and is thus considered a univoltine species (Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018). Dichotomius satanas is distributed from Colombia to Mexico (Pardo-Diaz et al. Reference Pardo-Diaz, Lopera Toro, Peña-Tovar, Sarmiento-Garcés, Sánchez-Herrera and Salazar2019) and is frequently found in forested areas with different degrees of conservation, as well as in environments derived from human activity such as pastures, secondary vegetation, and plantations (Pineda et al. Reference Pineda, Moreno, Escobar and Halffter2005; Halffter et al. Reference Halffter, Pineda, Arellano and Escobar2007; Rös et al. Reference Rös, Escobar and Halffter2012). This species is characterised as a generalist coprophagous tunneller (Bourg et al. Reference Bourg, Escobar, MacGregor-Fors and Moreno2016) of nocturnal habits, large corporal size (>10 mm), and evident sexual and intrasexual dimorphism. This study addressed the following specific questions for one population of D. satanas in the mountainous central region of Veracruz, Mexico: (1) What is the annual reproductive phenological pattern of D. satanas? (2) How are temperature and precipitation related to the reproductive phenology of the females and males? and (3) Does reproductive synchrony exist in the stages of sexual maturity of females and males throughout the year? Given that, in the cloud forest, D. satanas has a single peak of highest abundance in the year (Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018), we expect its reproductive phenology to be markedly seasonal (prediction I), such that climatic factors such as temperature and precipitation will similarly influence the process of emergence of the adults and activation of sexual maturation in both sexes (prediction II), for which reason we also expect that the stages of sexual maturity in females and males will be synchronised throughout the year (prediction III).
Materials and methods
Study area
This study was conducted in a well-conserved fragment of remnant cloud forest (19° 30ʹ 55.81″ N; 97° 0ʹ 19.88″ W) of approximately three hectares in area, surrounded by pastures and cultivated land to the west of the city of Xalapa, Veracruz, Mexico. The site is located at 1600 m above sea level; the climate is temperate humid (annual precipitation of 1500–2000 mm and mean annual temperature of 17–20 °C; Williams-Linera and Vizcaíno-Bravo Reference Williams-Linera and Vizcaíno-Bravo2016). In this region, three climatic seasons are distinguished in the year: cold–dry, from November to March (temperature ± standard deviation: 16 ± 1.7 °C; precipitation: 283 mm); warm–dry, from April to May (temperature: 21 ± 0.6 °C; precipitation: 173 mm); and warm–rainy, from June to October (temperature: 19.4 ± 0.7 °C; precipitation: 1131 mm; Williams-Linera Reference Williams-Linera2002).
Sampling
Beetles were captured monthly between October 2015 and September 2016 using 10 pitfall traps located at a distance of 50 m apart along a transect in the central part of the forest fragment. Each trap consisted of a 1-L–capacity disposable cup, half filled with soil and buried to soil level with a plastic funnel in the upper part to prevent the trapped beetles from escaping (Escobar and Chacón de Ulloa Reference Escobar and Chacón de Ulloa2000). Each trap was baited with 50 g of human excrement for 24 hours, at the end of which all captured individuals of D. satanas were transported to the laboratory for extraction of their reproductive systems.
Environmental variables
We obtained the monthly temperature (maximum, mean, and minimum) and mean precipitation values of the sampling period from an automated meteorological station (EMA 30452) located approximately three kilometres from the study zone. The data were provided by the Comisión Nacional del Agua and are available at https://smn.conagua.gob.mx/es/observando-el-tiempo/estaciones-meteorologicas-automaticas-ema-s.
Extraction of the reproductive system
The individuals captured every month were transported to the laboratory and euthanised, and the female and male reproductive systems were extracted following the protocol proposed by Martínez (Reference Martínez2002). This is one of the most simple and suitable in toto techniques for studying morphological variation in the reproductive systems of insects and is widely used in coprophagous beetles. The procedure consists of removal of the ovary from the females and the reservoir of accessory glands from the males by dissection using a stereomicroscope and subsequent fixing in Carnoy’s solution, consisting of ethyl alcohol, chloroform, and acetic acid, for a period of 24 hours. The morphological variation in size, colour, and other characteristics of these structures is used as an indicator of the stage of sexual maturity of the individuals (Halffter and Lopez Reference Halffter and Lopez1977; Martínez Reference Martínez2002; Sánchez-Carrillo et al. Reference Sánchez-Carrillo, Huerta, Carrillo-Ruiz and Escobar2017). The samples were then stored in plastic vials with 96% ethyl alcohol until subsequent analysis. This technique guarantees the conservation of the tissues and cells in a reproductive (physiological) state closest to that found in the live individual.
Stages of sexual maturity
In order to identify the stage of sexual maturity of the individuals, we used the total length (mm) of the basal oocyte of the females and the volume (mm3) of the right-side accessory glandular reservoir in the males. Measurements were obtained from a digital image using the software of the microscope Z16 APO – Leica® (Leica, Wetzlar, Germany). To obtain a complete characterisation of the reproductive stage of each sex, the degree of sclerotisation of the exoskeleton (soft or rigid), levels of intestinal content (0–5: 0 = absent; 5 = present in large quantities), and the relative quantity of body fat (0–3: 0 = absent; 3 = present in large quantities) were recorded through visual comparison. In addition, the colour of the basal oocyte (0–3; 0 = white and immature, 3 = yellow and mature) was considered in the females, because this is also an indicator of the stage of sexual development.
Data analysis
To characterise the individuals according to the stage of sexual maturity, we used the Random Forest Algorithm (Breiman Reference Breiman2001), which uses non-linear regression models to create classification trees of statistically homogeneous groups (Moisen Reference Moisen, Jorgensen and Fath2008). The predictors (independent variables) determine the formation of the groups according to the values of the dependent variable, which in our case was either the basal oocyte length (mm) in the females or the glandular reservoir volume (mm3) in the males. The predictors used for both sexes were (1) intestinal content and (2) body fat. For the females, we also considered (3) the colour of the basal oocyte. We conditioned the grouping according to the proximity or closeness of the values of predictors among individuals (Liaw and Wiener Reference Liaw and Wiener2018). For the formation of the groups, the importance value of the variable was used; this quantifies both the mean decrease of the precision of the predictions in the absence of the predictor (%IncMSE) and the decrease in the impurity of the nodes in the formation of each group (IncNodePurity; Liaw and Wiener Reference Liaw and Wiener2018). In this way, the most important predictors are those with the greatest values of precision and impurity. Finally, for a more precise classification of the groups, an average tree was obtained through the random combination of all of the trees generated by Bootstrap (number of trees created = 500; Breiman Reference Breiman2001; Cutler et al. Reference Cutler, Edwards, Beard, Cutler, Hess and Gibson2007; Liaw and Wiener Reference Liaw and Wiener2018).
The effects of monthly temperature (maximum, mean, and minimum) and mean monthly precipitation on the total abundance and the abundance of individuals in each state of maturity and per sex were evaluated using generalised linear models assuming a Poisson distribution. The model was selected considering the significance (P ≤ 0.05) AICc (Akaike information criterion for small samples < 40) values and the normality of the residuals (Shapiro test, α = 0.05) (Burnham and Anderson Reference Burnham, Anderson, Burnham and Anderson2002; Zuur et al. Reference Zuur, Ieno, Walker, Saveliev and Smith2009).
In order to determine differences in the number of females and males at each stage of sexual maturity, we also used generalised linear models assuming a Poisson distribution. The number of individuals was modelled considering (1) stage of sexual maturity, (2) sex, and (3) the interaction between (1) and (2). The optimum model was selected through inspection of the normality of the residuals (Shapiro test, α = 0.05), significance (P ≤ 0.05), and the AIC values (Burnham and Anderson Reference Burnham, Anderson, Burnham and Anderson2002; Zuur et al. Reference Zuur, Ieno, Walker, Saveliev and Smith2009). From the selected model, analysis of deviance was conducted as well as multiple comparisons (contrasts) through Tukey’s honest significance test to determine which groups were statistically different.
We used Kendal’s coefficient of concordance (W; Kendall and Gibbons Reference Kendall and Gibbons1990) to determine whether synchrony between the stages of sexual maturity of the males and females occurred over the course of the 12 months of sampling. The coefficient was estimated considering the number of males and females of each stage of sexual maturity captured in each month. The W values varied from 0, where there was no concordance, to 1, or perfect concordance, among the stages of each sex (Kendall and Gibbons Reference Kendall and Gibbons1990). To corroborate that the values of W differed statistically, we used the chi-square test (χ2). All of the analyses were conducted in the program R (R Core Team 2015).
Results
Stages of sexual maturity
In total, 239 individuals were dissected. Of these, 107 were females and 132 were males. The females presented two to nine oocytes per ovariole, and the length of the basal oocyte ranged from 0.68 to 3.19 mm (mean ± standard deviation: 1.54 ± 0.72, Fig. 1A,B; Supplemental materials, Fig. S1). Of the total number of females captured, 90 presented a rigid exoskeleton and had basal oocytes ranging from 0.70 to 3.19 mm (1.55 ± 0.70) in length. Seventeen females presented a soft exoskeleton (considered as recently emerged), of which seven presented a basal oocyte that ranged from 0.67 to 1.71 mm (1.50 ± 0.69) in length. In the 10 remaining females, no developed oocytes were evident.
Fig. 1. Female (A and B) and male (C and D) reproductive structures of individuals of Dichotomius satanas captured between October 2015 and September 2016 in a cloud forest in Veracruz, Mexico. Basal oocyte length: A, 0.68 mm and B, 3.19 mm; male glandular reservoir volume: C, 0.21 mm3 and D, 6.04 mm3.
The glandular reservoir volume in the males ranged from 0.21 to 6.04 mm3 (mean ± standard deviation: 1.40 ± 1.37). Of the total number of males, 104 presented a rigid exoskeleton, and the volume of their accessory glands ranged from 0.27 to 6.04 mm3 (1.40 ± 1.37; Fig. 1C,D). Eight individuals presented a soft exoskeleton and glandular reservoirs that ranged from 0.21 to 2.47 mm3 (1.26 ± 0.83) in volume. Of the remaining males, 20 individuals (five with rigid and 15 with soft exoskeletons) presented glandular reservoirs with no apparent differentiation. These were attributed a glandular reservoir volume of zero.
Body fat content was the predictor of greatest importance in the grouping of the stages of sexual maturity of females and males (Supplemental materials, Fig. S2). For both sexes, the Random Forest Algorithm analysis showed the formation of three groups: immature, maturing, and mature individuals (Supplemental materials, Fig. S2). For grouping of the immature females, the lower limit of the basal oocyte length was ≤ 1.46 mm, for maturing females, it was from 1.47 to 2.21 mm, and for mature females, it was ≥ 2.22 mm (Supplemental materials, Fig. S2). For grouping of immature males, the glandular reservoir volume limit was ≤ 1.76 mm3, for maturing males, it was from 1.77 to 3.21 mm3, and for mature males, it was ≥ 3.22 mm3 (Supplemental materials, Fig. S2).
Reproductive phenology
The number of D. satanas individuals fluctuated throughout the year, with the peak of greatest abundance observed at the end of the warm–rainy season, followed by a drastic decrease in the number of individuals during the cold–dry season and subsequent increase in abundance during the warm–dry season (Fig. 2B). The temporal variation in the total number of individuals was explained by the minimum temperature and precipitation (Supplemental materials, Table S1).
Fig. 2. Number of individuals of Dichotomius satanas captured between October 2015 and September 2016 in a cloud forest in Veracruz, Mexico. A, Monthly variation in the total number of individuals and the monthly minimum temperature (°C) and mean precipitation (mm); B, monthly variation in the number of individuals according to maturation states: immature, maturing, and mature. Climate data were obtained from the National Water Commission (CONAGUA, Comisión Nacional del Agua, Mexico City, Mexico). Seasons: warm–rainy (W–R), cold–dry (C–D), and warm–dry (W–D).
The population of D. satanas comprised 54% immature individuals (128 individuals), followed by 32% maturing individuals (76 individuals) and a lower quantity of mature individuals at 14% (35 individuals). The analysis of deviance from the selected model showed that the total number of beetles differed among stages of sexual maturity, with an effect of the interaction between the stage of sexual maturity and the sex of the individuals (Fig. 3; Supplemental materials, Table S2). The number of immature males was almost twice that of the females (Fig. 3). This contrasts with that observed in the stage of maturation, in which the number of females was greater than that of males (16% more females), although this difference was not statistically significant (Fig. 3; Supplemental materials, Table S2). For the mature individuals, the numbers of females and of males were similar (Fig. 3; Supplemental materials, Table S2).
Fig. 3. Number of individuals of females and males of Dichotomius satanas according to sexual maturation state, captured between October 2015 and September 2016 in a cloud forest in Veracruz, Mexico. The line within each box represents the median, and the height of each box represents the first and third quartiles. Black dots correspond to the observed values, and the dotted lines represent the minimum and maximum variation observed. Different letters indicate statistical differences according to Tukey test of contrasts (see Supplemental materials, Table S2).
The number of individuals of each of the stages of sexual maturation varied throughout the year (Fig. 2B; Supplemental materials, Fig. S3). During much of the year, the number of immature individuals exceeded that of the maturing and mature individuals, apart from at the onset of the rainy season, when the number of maturing and mature individuals (Fig. 2B) increased. The monthly variation in the number of immature individuals (Fig. 4) was explained by the minimum temperature and precipitation. The variation in the number of maturing individuals (Fig. 4) was related to the minimum temperature, and that of the mature individuals was related to precipitation (Fig. 4; Supplemental materials, Table S1).
Fig. 4. Monthly variation in the number of females and males of Dichotomius satanas per maturation state (immature, maturing, and mature), captured between October 2015 and September 2016 in a cloud forest in Veracruz, Mexico. Seasons: warm–rainy (W–R), cold–dry (C–D), and warm–dry (W–D). The dashed line corresponds to the reproductive period of D. satanas (Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018).
The monthly variation in the numbers of immature and maturing females was explained by the minimum temperature, but the abundance of mature females was not related to any of the environmental variables considered (Supplemental materials, Table S1). The abundance of immature and maturing males was explained by precipitation, whereas the variation in mature males was related to both minimum temperature and precipitation (Table S1). It was evident that the mature males were more abundant at the beginning of the rainy season, whereas mature females were more abundant at the end of the rainy season and during the transition to the cold–dry season (Fig. 4). High concordance was found between the number of immature males and females (W 12, 11 = 0.94, χ2 = 20.8; P = 0.03), whereas the values of concordance were lower for the number of maturing (W 12, 11 = 0.31, χ2 = 7.02, P = 0.79) and mature (W 12, 11 = 0.14, χ2 = 3.11, P = 0.98) males and females.
Discussion
Stages of sexual maturity
The variation found in the basal oocyte of the D. satanas females was within the range of values reported for other species of Scarabaeinae (Copris diversus: 0.85–3.25 mm; Tyndale-Biscoe Reference Tyndale-Biscoe1983; Canthon indigaceus chevrolati: 0.1–4.2 mm, Canthon cyanellus cyanellus: 0.1–3.1 mm; Martínez Reference Martínez2002). The number of oocytes per ovary observed in the D. satanas females is similar to that reported for other tunneller species (Halffter and Lopez Reference Halffter and Lopez1977; Martínez Reference Martínez1992; Martínez and Huerta Reference Martínez and Huerta1997). In tunneller species, females tend to produce more eggs per nest than females of roller or resident species do, and this life history trait has been related to body size and the number of generations per year (Hanski and Cambefort Reference Hanski and Cambefort1991). The nesting pattern of some tunneller species, as seems to be the case of D. satanas, is therefore characterised by high energetic investment in the production of nesting chambers and provision of sufficient food to ensure survival of the progeny, compensating thus for the reduced or null parental care (Halffter and Edmonds Reference Halffter and Edmonds1982). In univoltine species of large and medium size, such as Heliocopris dilloni, Copris diversus, and C. tripartitus, numbers of 2–10 eggs have been reported (Kingston and Coe Reference Kingston and Coe1977; Tyndale-Biscoe Reference Tyndale-Biscoe1983; Hanski and Cambefort Reference Hanski and Cambefort1991; Huerta and Bang Reference Huerta and Bang2004), whereas in multivoltine species of small size, such as Digitonthophagus gazella, this number can reach up to 44 eggs per nest (Blume and Aga Reference Blume and Aga1975). However, the number of eggs laid per nest by D. satanas females remains unknown, as is the number of nests they produce over the course of their reproductive lives.
The stages of sexual maturation of D. satanas coincide with those observed for various species of coprophagous beetles, for example, Phanaeus daphnis, P. mexicanus (Halffter and Lopez Reference Halffter and Lopez1977), Canthon cyanellus cyanellus (Martínez and Cruz Reference Martínez and Cruz1999), Copris ochus, and C. tripartitus (Bang et al. Reference Bang, Crespo, Na, Han and Lee2008). Laboratory studies report a fourth stage of sexual maturity that corresponds to the phase of reabsorption of the oocytes and sexual inactivity, with such individuals classified as old females (Halffter and Edmonds Reference Halffter and Edmonds1982; Martínez Reference Martínez2002). In very old females, the oocytes are not apparent and the ovary is similar in size and shape to that of the immature females (Halffter et al. Reference Halffter, Halffter and Huerta1983; Martínez Reference Martínez1992). It is therefore possible that a fraction of the females classified as immature in this study were old females. The number of old females during each reproductive period varies among species; for example, in Onthophagus merdarius and Onitis ion, González-Megías and Sánchez-Piñero (Reference González-Megías and Sánchez-Piñero2004) found that old females corresponded to 40 and 20%, respectively, of the total number of females captured. However, to identify the old females in the population with certainty, histological analysis of the ovaries (Halffter and Lopez Reference Halffter and Lopez1977) is necessary, although body wear and quantity of fat content can be considered indirectly as evidence of female agedness, as has been conducted for two roller species of the genus Canthon (Martínez and Montes de Oca Reference Martínez and Montes de Oca1994).
The differences observed in the number of individuals in each stage of sexual maturity suggest that the population dynamic of D. satanas is characterised by high migration (exit of individuals), as has been observed at landscape scale (Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018). Given that approximately 15% of the total number of individuals dissected were mature, we believe that immature and maturing individuals have a greater probability of dispersion at different times throughout the year. It is possible that the immature males disperse first during the rainy season, followed by the maturing females just before the beginning of winter when the environmental temperature decreases. This could be explained by the differential effect of the environmental variables on each sex and stage of sexual maturation.
Reproductive phenology
The results indicate that the reproductive phenology of D. satanas follows a seasonal pattern, as proposed in prediction I, but the selected models indicate that temperature and precipitation had a different effect on each sex, depending on the stage of sexual maturity and contrary to that proposed in prediction II. Moreover, contrary to the proposal of prediction III, the mature males and females were not synchronous.
According to the results obtained, the reproductive cycle of D. satanas can be divided into two phases. A phase of intense activity coincides with the moderate to high temperatures and greater precipitation in the months of August to October, when new individuals emerge and feeding, sexual maturation, mating, and nesting occur (Halffter and Edmonds Reference Halffter and Edmonds1982). This is followed by a period of reduced activity in which, possibly due to the low temperatures and potential decrease in available resources during the cold–dry season (specifically, January and February), the beetles reduce their energetic expenditure (physiological pause) and remain buried in the soil. This has been observed for species that inhabit temperate regions or arid–semiarid environments that have marked seasonality in precipitation (Hanski and Cambefort Reference Hanski and Cambefort1991; Bang et al. Reference Bang, Crespo, Na, Han and Lee2008). Various studies show that, in environments with marked seasonal changes, most of the coprophagous beetle species enter a physiological pause (Huerta and Bang Reference Huerta and Bang2004; Bang et al. Reference Bang, Crespo, Na, Han and Lee2008). However, it is necessary to determine the complete life cycle of D. satanas in order to fully understand the effects of environmental conditions on the development and reproductive activity of the species.
In insects, low temperatures reduce the metabolic activity of individuals (Varley et al. Reference Varley, Gradwell and Hassell1973). In the case of coprophagous beetles, low temperatures cause adult beetles to interrupt their feeding and reproduction and reduce or attenuate larval development (Halffter and Matthews Reference Halffter and Matthews1966; Huerta et al. Reference Huerta, Halffter and Halffter2005; Bang et al. Reference Bang, Crespo, Na, Han and Lee2008). On the other hand, increases in temperature and variations in the precipitation regime can modify the adult insects’ period of activity and alter the time taken for larval development, thus influencing the period of emergence (Duchenne et al. Reference Duchenne, Thébault, Michez, Elias, Drake and Persson2020) and number of generations per year (Bonal et al. Reference Bonal, Hernández, Espelta, Muñoz and Aparicio2015; Forrest Reference Forrest2016). It is therefore expected that climatic changes in regions with marked seasonality can have direct consequences for species demographics and population dynamics (Régnière Reference Régnière2009; Duchenne et al. Reference Duchenne, Thébault, Michez, Elias, Drake and Persson2020). Changes in phenology can be a signal that populations are adapting to new environmental conditions, and species with a greater capacity to modify their phenology may be predicted to be more successful in terms of coping with environmental changes than those species that have restricted times of emergence and reproduction (Pozsgai and Littlewood Reference Pozsgai and Littlewood2014). Phenological studies can therefore serve as models with which to understand the effects of environmental variation on the reproduction of species and thereby provide helpful information for monitoring and conservation programmes.
The results of this study suggest that D. satanas is a protogynous species, as was found for Onthophagus incensus under field conditions (Martínez et al. Reference Martínez, Montes De Oca and Cruz1998). According to Degen et al. (Reference Degen, Hovestadt, Mitesser and Hölker2015), protogyny is characteristic in populations in which males present greater mortality, coinciding with that observed for populations of D. satanas, due to intrasexual competition and greater male mobility (Barretto et al. Reference Barretto, Cultid-Medina and Escobar2018). However, although the phenological cycle of insects is usually characterised by differential emergence between sexes, protogyny is rarely observed (Van Timmerman et al. Reference Van Timmerman, Switzer and Kruse2000). For this reason, we highlight the importance of conducting future phenological studies that identify patterns of emergence in the coprophagous beetles, relating this to the biological attributes (e.g., mobility) and demographic parameters (e.g., rates of birth, mortality, and migration) of each species.
It was interesting to observe that the abundance of males and females throughout the year was related to different environmental signals, according to the selected models. Based on this, we suggest that, for the males, precipitation marks the beginning and duration of each physiological stage of reproduction during their adult lives. For the females, however, the beginning and duration of the first two stages of sexual maturity seem to be determined by minimum temperature. It is therefore possible to conclude that immature and maturing females could be physiologically more vulnerable to the minimum temperature because they must invest more metabolic energy than males do in order to reach sexual maturity (Trivers Reference Trivers and Campbell1972). However, for the best model, the abundance of mature females was not related to any of the environmental variables considered. This result corroborates, as has been recorded for other beetle species (Cruz and Martínez Reference Cruz and Martínez1998; Martínez and Cruz Reference Martínez and Cruz1999), that D. satanas females require the presence of mature males and first copulation to complete their sexual development and to be physiologically ready to nest. In general, first copulation most frequently occurs at 30 days after a female emerges (Cruz and Martínez Reference Cruz and Martínez1998), but in some species – such as Onthophagus merdarius and Onitis ion – first copulation occurs much earlier, at two to three days after emergence (González-Megías and Sánchez-Piñero Reference González-Megías and Sánchez-Piñero2004). Although there is no information regarding the factors that trigger first copulation, the difference in the timing of its occurrence could be related to the duration of the optimum period for reproduction. However, for a full understanding of the reproductive phenology of D. satanas and the possible effects of climatic variation, it is necessary to determine in more detail the physiological and behavioural changes in the females that mark the transition between the sexually maturing and mature stages.
The absence of synchrony between the stages in maturing and mature females and males of D. satanas can be explained by the way different environmental signals modulate the phenological reproductive cycle of each sex to determine the different maturation times and the probability of permanence and survival of the sexes throughout the year. The absence of synchrony in this population may be a mechanism that serves to maximise a population’s reproductive success. Such a mechanism would favour the sexual maturation of the females at the end of the warm–rainy period and the beginning of winter, which is considered the optimum period for reproduction and initiation of larval development.
The difference in the number of individuals for each stage of sexual maturity and the absence of reproductive synchrony between the mature individuals are possibly associated with characteristics of D. satanas’s mating system and population dynamics that have yet to be described. However, it is known that, in species with sexual dimorphism such as D. satanas, the males compete for resources (food and mates), triggering adjustments in density and sex ratio that would result in the higher reproductive success of some individuals within the population, as has been observed in other species of coprophagous beetles (Pomfret and Knell Reference Pomfret and Knell2008). However, detailed studies are necessary in order to fully understand the relationships among the phenology, mating system, and population dynamic for this species.
Identification of the reproductive stages and analysis of the factors that determine reproductive phenology under natural conditions are challenging in the study of insect population ecology. This study provides new data about the biology and natural history of D. satanas and, more generally, the population ecology of the subfamily Scarabaeinae. However, given the lack of information regarding the relationship between this species’ life history traits and its demographic parameters, the phenological patterns cannot be completely understood and, in some cases, must be interpreted with caution. We recommend further research both to establish this relationship and to investigate the reproductive phenology of insects, and of organisms in general. The increased understanding that would result would enable more precise estimates of the possible impacts of projected climatic scenarios on insect species and would also serve as a tool for evaluating and monitoring wild populations. Such information is invaluable to predict population decline and the degree of vulnerability of species with precision.
Acknowledgements
The authors thank Ricardo Madrigal Chavero and biologist Jaime Pelayo for their support with fieldwork. The authors appreciate the comments of the evaluators, whose suggestions helped to improve the manuscript. The authors also gratefully acknowledge the financial support and a Ph.D. scholarship (No. 589280) received for this research from the Mexican National Council for Science and Technology (CONACYT).
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.4039/tce.2020.75.
