Results

We recorded a total of 91 adult individual of P. rusticus, 69 males and 22 females (Table 1). In the 2015/16 breeding season, we recorded 34 individuals, with a recapture rate of 61% (n = 21 individuals); of these, 34.7% (n = 12) were recaptured more than five times in the sampling period (Figure 2 (A)). The Schumacher method estimated a total of 28.9 individuals (25.4–33.8) for this breeding season. In the 2017/18 breeding season, we recorded 21 individuals, with a recapture rate of 57% (n = 12 individuals), and of these, 28.5% (n = 6) were recaptured more than five times in the sampling period (Figure 2 (B)). Some males were recaptured in the following days during the same effort (month) and in subsequent months and always in the same location. The Schumacher method estimated a total of 21.0 individuals (19.2–23.2) for this breeding season. Individuals captured on the pond were never recaptured in the marsh 100 m away and vice versa. Most recaptures occurred within the same breeding season. In the 2018/19 and 2020/21 breeding seasons, we recorded 13 and 23 individuals, respectively. However, in November 2018, we recaptured two individuals that were tagged during at the previous breeding season (October 2017).

Pithecopus rusticus showed a pattern of seasonal activity (p <.001 according to Rayleigh’s test), with a peak mean abundance in November (Figure 3). The degree of seasonality (r) was high for both breeding seasons 2015/16 and 2017/18 (r = 0.90 and r = 0.85, respectively). Males called from October to January, with greater activity at the beginning of the period (Figure 4). Variation in abundance of P. rusticus among the sampled nights was significantly related with RH (R 2 = 0.58; adjusted R 2 = 0.26; F (4,35) = 4.49; p <.01 and β = 0.57; p <.01). There was no significant influence of rainfall (β = −0.19; p =.24), minimum temperature (β = 0.13; p =.40) or maximum temperature (β = −0.11; p =.64) on the abundance of individuals.

Calling activity also varied throughout the night (F(2,66) = 28.57; p <.01), since a lower frequency of males called at the beginning of the nocturnal sampling (7pm) than in the middle (9pm) or at the end of this period (11pm) (Tukey p <.01 for both comparisons). The frequency of calling males did not varied between the middle (9pm) and the end of the nocturnal sampling period (Tukey p <.01). In addition, the highest frequency of calling males was observed at temperatures between 14 and 16.9°C (61.8%; n = 175), with a peak of activity between 14.0 and 14.9°C (29.3%; n = 83) (Figure 5). The majority of calling males used mainly leaves of Fimbristylis ( Cyperaceae;>50%) and Juncus ( Juncaceae; ~20%) from the edges of ponds and marsh (Table 2). However, males also used leaves of Senecio ( Asteraceae) and Eriocaulon ( Eriocaulaceae) as calling sites (<30%). Calling males perched at heights of 16.00 ± 10.09 cm (1–33 cm; n = 17) and at distances of 155.50 ± 393.02 cm (5– 1400 cm; n = 12) from the edges to the centre of the pond. Non-calling males showed the same pattern in vegetation use and were perched at heights of 18.50 ± 9.89 cm (10– 30 cm; n = 4; Table 2). The distance from the edges to the centre of the pond was 6 ± 4.24 cm (3–9 cm; n = 2).

10 (N = 1)

Females also used mainly leaves of Fimbristylis (>50%) and Juncus (about 15%) at the edge of pond and marsh, perched at 24.80 ± 17.20 cm (0–54 cm; n = 13) high. Likewise, in a smaller proportion (>20%), they used leaves of Senecio ( Asteraceae) and Eriocaulon ( Eriocaulaceae). Most of the females’ perches were established at 173.70 ± 433.00 cm (2– 1400 cm; n = 10) from the edges to the centre of the pond, and few (n = 2) were found outside the edges at 7.00 ± 9.90 cm (0–14 cm; n = 2). Recaptured males and females showed the same pattern of microhabitat use as in the first captures (Table 2).

Females were larger and heavier than males (Table 3; Figure 6 (A)). The OSR was male-skewed, with 0.27 (± 0.22, 0–0.50, n = 4) females to each male in the 2015/16 breeding season and 0.17 (± 0.21, 0–0.44, n = 4) females in the 2017/18 breeding season. We observed spawning in December 2015 (n = 16), October 2017 (n = 6), October and December 2020 (n = 2 and n = 7, respectively) and January 2021 (n = 1) (Figure 6 (B)). The spawning consisted of yellowish eggs, surrounded by gelatinous capsules. The spawns contained a mean of 43.00 ± 5.28 eggs (26–46; n = 13), with a diameter of 3.26 ± 0.45 mm (2.88–4.19 mm; n = 13). Most of the spawns were found on leaves of Senecio (n = 26) and some on leaves of Eriocaulon (n = 5). Pithecopus rusticus used only one folded leaf for each egg-laying (Figure 6 (B); Table 4). The spawns were hanging over the water (88%; n = 14) or on the water’s edge (13%; n = 2). The mean distance from spawns to the edge was 19.00 ± 22.18 cm (1–2500 cm; n = 10) for those located internally to the edge and 17.00 ± 16.45 cm (2–50 cm; n = 6) for those located externally. The spawns were at a height of 23.50 ± 6.41 cm (15–36 cm; n = 16) above the water surface (81.25%; n = 13) or wet ground (18.75%; n = 3).