Description of mating behavior, life cycle, and antennal sensilla of Cyclocephala putrida Burmeister, 1847 (Coleoptera, Scarabaeidae, Dynastinae) and of

: The genus Cyclocephala is common in Brazil (Coleoptera, Scarabaeidae, Dynastinae). The adults of some species are important pollinators, and the larvae develop in the soil, feed on organic matter, and contribute to nutrient cycle, but immatures of some species feed on plant roots, and some were registered causing damage in crops. The mating process of some phytophagous scarab beetles has a chemical recognition step, and the antenna is the main structure involved in the detection of odorants associated with insect communication. In the present study the mating behavior, life cycle, and antennal sensilla of C. putrida are described. The study was conducted at the Universidade Estadual de Mato Grosso do Sul, Cassilândia, Brazil. Adults were collected by a light trap installed from January 2016 to December 2017 and were taken to the laboratory for studies. Adults swarms are brief and were registered from January to February, and specimens were mostly collected at 20:00 to 22:00h. Chemical recognition may occur at least during one of the mating steps, during which the couple kept their antennae moving and the lamellae open, while females select males. In laboratory, the mating process lasted 7.5 minutes on average. The antennae of females have about 3399 sensilla and males about 4229 sensilla. Sensilla placodea types I, II, and III are the most abundant, and sensilla ampullacea, basiconica, and coeloconica are also present. The embryonic period lasted 16.0 days; first, second and third instars lasted 16.0, 48.3, and 165.3 days, respectively. The pupal period lasted 24.0 days. The period between egg deposition and adult emergency is about 271.5 days.

Some Scarabaeidae species release sexual pheromones to chemical recognition, and females select males before copulation (e.g. Facundo et al. 1999, Rodrigues et al. 2014. As well as plant volatiles, insect structures involved in detecting pheromones are antennal sensilla, which are mainly distributed on the surface of lamellae (e.g. Leal & Mochizuki 1993, Kim & Leal 2000. Before copulation, females return to the soil for egg deposition, and their biological cycle varies with species. The time between egg deposition and adult emergence is about 229 days for C. verticalis Burmeister, 1847 (Rodrigues et al. 2010), about 164 days for C. celata , and about 108 days for C. distincta Burmeister, 1847 (Souza et al. 2015).
The larvae of Cyclocephala use soil with organic matter as food, therefore being important for nutrient cycling such as C. paraguayensis Arrow, 1913). On the other hand, these larvae may consume plant roots, and some species can cause damage to crops such as C. forsteri which has been reported as potential pest to soybean crops ("soja", Glycine max (L.) Merr, Fabaceae) (Santos & Ávila 2007), or such as Cyclocephala flavipennis Arrow, 1914, which has been registered damaging pastures (Duchini et al. 2017).
Although studies have been conducted on Cyclocephala putrida Burmeister, 1847 biology in Cassilândia, state of Mato Grosso do Sul (Brazil), little information is known for this species. Cherman et al. (2014) found larvae of the species in cultivated and non-cultivated areas in the state of Rio Grande do Sul (Brazil), and Bonivardo et al. (2015) reported immatures associated with pasture and maize ("milho", Zea mays L., Poaceae) in Argentine.
Regarding the above-mentioned biological and economic importance of the genus Cyclocephala, the present work aims to describe the mating behavior, the antennal sensilla, and life cycle of C. putrida.

Material and Methods
The study was conducted at the Universidade Estadual de Mato Grosso do Sul (UEMS), Campus of Cassilândia, MS (19° 06' 48" S; 51° 44' 03" W). Sampling and rearing procedures, as well as analysis strategies, were modifications of Rodrigues et al. (2008Rodrigues et al. ( , 2010. Adults of C. putrida were collected using a light trap model "Luiz de Queiroz" from January 2016 to December 2017. The trap was installed in a pasture area (Urochloa decumbens Stapf cv Basilisk, Poaceae) and was turned on daily at 18:00 and turned off at 6:00 of the next day. The insects were collected every 60 minutes, and collected adults were sexed by observing male dilated protarsomeres (Figure 1), and couples were placed in 500 mL plastic containers with half-filled with the same soil from where adults were collected. The adults were reared with slices of bananas (fruit of Musa sp., Musaceae) as food source (see Ferreira et al. 2016). Each couple was observed from February 2016 to February 2017 to behavior observations. The containers were checked weekly when the soil was sieved to egg separation. After removed, the eggs were individualized and placed in other 250 mL plastic containers one-third filled whit soil, which were maintained in a climatized chamber (26 ± 1 ºC and 12 hours scotophase).
Each newly hatched larva was transferred to a 250 mL plastic container with soil and U. decumbens seedlings. Cranium width and body total length were weekly measured ( Figure 2).
Flight activity was observed from February 2 to 27, and data were transformed into 1 + x and submitted to analysis of variance (ANOVA). Means were grouped and compared by the Scott-Knott test (p< 0.05) using SISVAR software. Data on average temperature (•C), precipitation (mm), and solar radiation (kJ/m 2 ) in Cassilândia, were obtained from the Instituto National de Meteorologia (INMET).
The study of antennal sensilla was based on antennal clava of 10 dissected specimens. Antennal segments were formerly treated in a 10% potassium hydroxide solution at 80°C for 60 minutes. The pieces were then successively washed in distilled water, 70% ethanol, 80% ethanol, and 90% ethanol, respectively (Romero-Lópes et al. 2010). Images were taken by a scanning electron microscope (model JSM 5410) at the Laboratório de Microscopia Eletrônica of the Universidade Estadual Paulista, campus of Jaboticabal, state of São Paulo, Brazil. Part of dissected lamellae was mounted on slides with Hoyer's solution (Johnson & Triplehorn 2005) and observed under a Nikon microscope model E2000. Sensilla terminology follows Keil (1999). Meinecke (1975) was used as an alternative sensilla terminology (in brackets) to easily comparison of terms. The insects were deposited in the UEMS entomological collection, campus of Cassilândia.

Results
Mating behavior. Four hundred and fourteen (414) adults were collected in 2016, 132 in January and 282 in February, which indicates a short swarming period. From 18:00 to 19:00 h the brightness decreased from 131.8 KJ/m² to 0, beetles flew from dusk till dawn (from 18:00 to 6:00 of the next day), and swarm activity was high at nightfall (from 20:00 to 22:00 h). Temperature ranged from 28.5 ºC to 23 ºC during the flight (from 18:00 to 6:00 of the next day; Figure 3). Adults remained buried in the soil during the day and started flying from 18:00 onwards. After leaving the soil, the beetles flew for 12.6 ± 1.86 min (11-15) when landed on plants for feeding and mating. Then, they started walking while their antennae kept moving and lamellae remained open. Afterwards, some couples were formed ( Figure 4) and mated, with antennae movement being maintained during all stages.
Six of the 30 couples formed mated. At the first stage, males walked over females from behind (n=5) or side (n=1) and repeatedly touched female pygidium or apex of elytra with their antennae and protarsi. After, males climbed on females dorsum (n = 6) using their protarsal claws to seize females, remaining in such position for 0.55 ± 0.22 min (0.39-1.0). Sometimes, female refused the male for mating (n=4) and walked away from it ( Figure 4). If accepted by the female (n=6), the male griped the female elytra, using all six legs. Then, rhythmical movements were observed in the male abdomen while its aedeagus was exposed, and the copulation began.
Antennal sensilla (Table 1). The entire outer surface of the antennal clava is covered by sensilla chaetica (long and short setae) and sensilla trichodea (minute setae, Figure 5), which were identified mainly as tactile (mechanoreceptors) but also as gustatory (contact chemoreceptors) sensilla (Keil 1999). Whereas the inner surface of clava (the lamellar side in contact with another lamella when the clava is close) was observed covered by sensilla ampullacea, basiconica, coeloconica, and placodea.
Sensilla ampullacea were identified as pores on the surface of the lamella. In Hymenoptera (Insecta), antennal pores were identified as CO 2 receptors (Keil 1999); therefore, their function in scarab beetle need to be checked. Sensilla basiconica were identified as type I (Meinecke 1975: L4) and resembles minute setae enclosed in a depressed pit, with the apex projected over pit limits. Two types of sensilla coeloconica were observed, and both are ovoid and enclosed into a depressed pit, the type I (Meinecke 1975: L1) had a pointed apex and the type II (Meinecke 1975: L2) had blunt apex.  Life cycle. In the laboratory, 22 eggs were obtained in February and 14 eggs in March. The eggs showed 100% viability ( Table 2). The mean duration of egg stage was 16.0 days, first-instar larvae 16.0 days, second-instar larvae 48.3 days, third-instar larvae 165.3 days, and pupae 24 days. The first and second instars showed 80.5 and 60% viability, respectively. The initial development stages had lower viability, as they are the most sensitive during the larval phase. The mean period between egg deposition and adult emergence was 271.5 days, and viability was 41.6%. Cranium width of first-instar larvae was 1.3 mm, of second-instar larvae was 2.4 mm, and of third-instar was 4.17 mm. The body length of first-instar larvae was 4.4 mm, of second-instar larvae was 7 mm, and of third-instar larvae was 15.91 mm. The third-instar larvae had a light grey colour that changed to white when pre-pupal stages started. Pre-pupae stop eating and make pupa chamber, possibly with aid of saliva or moist soil, manipulated with mouth appendages.

Discussion
Adults of C. putrida occurred in the field from January to February, as with C. paraguayensis found by Riehs (2006) in South of Brazil, but different from C. melanocephala, C. tucumana, C. verticalis, and others, which occur at beginning of rainy season, in September and October (Rodrigues et al. 2010, Nogueira et al. 2013. The occurrence patterns of some Cyclocephala species may be a strategy to avoid competition, but this hypothesis needs to be checked further. Field observation showed that C. putrida start to flight at dusk, as other Cyclocephala species, as C. verticalis (Rodrigues et al. 2010, Barbosa & Rodrigues 2016, and other phytophagous Scarabaeidae, as Anomala testaceipennis Burmeister, 1856 (Rutelinae) (Rodrigues et al. 2014).
In the laboratory, adults started swarming at 00:00 h and copula started at 1:00 h. Adults of C. verticalis, when coupled in the laboratory, started copulating regardless of the day (Rodrigues et al. 2010).
Mating of C. putrida encompassed several stages, including coming out of the soil, start walking, swarming, forming couples, selecting females, and copulating. This process is quite similar to that observed in other Cyclocephala species such as C. melanocephala, C. testaceipennis, and C. verticalis (Rodrigues et al. 2014, Barbosa & Rodrigues 2016. Female selection in Scarabaeidae may be related to the recognition of male sexual maturity as in Canthon cyanellus cyanellus (LeConte, 1859) (Fávila 1988) and C. verticalis (Barbosa & Rodrigues 2016), and is might be intermediated by chemical communication (see Rodrigues et al. 2019).   (Figure 6).
At the beginning of mating behavior, adults showed active movements of antennae and lamellae, probably for detection of pheromones released by other specimens. Some studies with Scarabaeidae have found that lamellar sensilla are involved in the detection of sexual pheromones, and these sensilla are more abundant in males than in females (Romero-López et al. 2004, Tanaka et al. 2006, Mutis et al. 2014. Also, males of C. putrida had more sensilla than females. Sensilla placodea was predominant in males and females of this species, while sensilla coeloconica and sensilla basiconica occurred in smaller quantities. Such dominance of sensilla placodea was also found in other Dynastinae species such as Oryctes rhinoceros (Linnaeus, 1758) (Renou et al. 1998), and in other phytophagous subfamilies such as Anomala iconstans Burmeister, 1844 (Rutelinae, Rodrigues et al. 2019). Larsson et al. (2001) studied the antennal sensilla of Anomala cuprea (Hope, 1839) and concluded that the outer sensilla placodea are responsible to detecting sexual pheromones while the inner ones are responsible for detecting plant volatiles.
The predominance of sensilla placodea may be related to important roles in detection food resources and adult info-chemicals (sexual or aggregative).
Concerning the life cycle of C. putrida, the period between egg deposition and adult emergence was on average of 271.5 days, so it could be considered a long life beetle if compared to other Cyclocephala species. Long cycles were also noticed for C. melanocephala (113 days Table 3). The embryonic phase of C. putrida had a similar length to that other Cyclocephala species; however, the first instar lasted 16 days, a period shorter than in the other species (Table 3). The second and third instars of C. putrida lasted, in days, shorter than did C. tucumana, which lasted 66.1 and 176.8 days, respectively (Table 3). And its pupal phase was also shorter than C. celata (Table 3).  Regarding the immatures, C. putrida larvae fed soil organic matter, so the species is not a potential pest to cultivated plants. White grubs had similar behavior compared to other Cyclocephala species. Its active third instar is yellowish-white, while prepupa is white and builds a pupal chamber, just as observed in C. celata and C. distincta (Souza et al. , 2015.