Cochranella litoralis ( Ruiz-Carranza and Lynch, 1996 )

Maynard, Ross J., Tinajero-Romero, José, Kohn, Sebastian, Pellet, Christophe & Culebras, Jaime, 2024, Conservation status, range extension, and call analysis of the Littoral Glassfrog, Cochranella litoralis (Ruiz-Carranza and Lynch 1996), Amphibian & Reptile Conservation (e 329) 18 (1), pp. 1-9 : 3-6

Cochranella litoralis ( Ruiz-Carranza and Lynch, 1996)

New record. Adult male from Los Laureles, Cotopaxi, Ecuador (0°51’18.2232” S, 79°11’25.926” W, 407 m; Fig. 1), 16 March 2019 at 0400 h; Christophe Pellet and Jaime Culebras leg.; photo voucher CJ12588 ( Fig. 2); uncollected. The specimen was observed calling shortly after a light rain, perched on a leaf 4.5 m high within an abandoned banana plantation adjacent to secondary forest. Other males have been observed at this same location, the first being on 19 February 2017 at 2100 h. Males of Hyalinobatrachium tatayoi have also been observed calling nearby .

New record. Adult male, 20.6 mm snout-urostyle length (SUL), recorded from a fragmented forest adjacent to Cristobal Colón Quininde, Esmeraldas, Ecuador (0.45213°N, 79.14919°W, 178 m; Fig. 1); 21 August 2021 at 2310 h; Ross Maynard and Sebastian Kohn leg.; photo vouchers CJ12587a–d ( Fig. 2); uncollected. The male was observed in a clearing 3 m from the forest edge, calling on the upper surface of a leaf within sparse herbaceous vegetation, perched 1.0 m high. Slow, shallow water was channeled just below the vegetation due to steady rain earlier that evening, which was flowing towards a small stream (about 2–3 m wide and 0.5 m deep) a few meters away. Two additional males were heard calling nearby, one from just within the forest and the other also in the clearing near the forest edge, however their exact locations were not observed. Other glassfrogs recorded along the stream adjacent to where the C. litoralis was observed, but from within the secondary forest, were Sachatamia ilex and Teratohyla spinosa .

Distribution. A review of the literature yielded seven verified localities for C. litoralis : two localities from Colombia in extreme southwest Nariño Department, and five localities in Esmeraldas Province, Ecuador ( Table 1; Fig. 1). The purported localities in Cauca, Colombia, and in Los Ríos, Pichincha, and Santo Domingo de los Tsáchilas, Ecuador, are either unverified or were reported in error (see Discussion). Nonetheless, a search of public-sourced and museum databases identified an additional locality from Los Ríos Province, Ecuador, at the Río Palenque Research Center in August 2021 (http://iNaturalist.org/ observations/90596035; D. Weaver and E. Osterman, pers. comm.). Since the coordinates of the holotype provided by Ruiz-Carranza and Lynch (1996) are imprecise, the placement of the type locality on the map is approximated ( Fig. 1). The new record from Los Laureles, Cotopaxi, expands the elevational range from near sea level to 407 m asl, and extends the known distribution of C. litoralis by abount 175 km south-southeast from the previous southernmost locality at Tsejpu, Río Zapallo, Esmeraldas.

Extinction risk. Despite the new records, the extinction risk for C. litoralis remains relatively high. With the additional localities reported herein, and assuming each of the seven localities where the species had previously been reported represent extant populations, the extent of occurrence (EOO) of the species is about 8,308 km 2 and the area of occupancy (AOO) is 40 km 2. However, the only other reported observations over the past decade are from Tundaloma Lodge, Esmeraldas, Ecuador in 2014 ( Guayasamin et al. 2020) and Tumaco, Nariño, Colombia, in 2015, 2016, and 2020 ( Table 1; IUCN SSC Amphibian Specialist Group 2019; Pinto-Erazo et al. 2020; iNaturalist. org). Except for the latter locality in Colombia, whether there have been subsequent sampling efforts for C. litoralis at the remaining localities in Esmeraldas Province, Ecuador is unclear. Although the status of these subpopulations cannot be verified at this time, we suspect that there has been recent and ongoing decline in the extent and quality of its habitat, given that northwest Ecuador has been a hotspot of deforestation over the past three decades ( Sierra 2013; Kleeman et al. 2022). Logging and agriculture are the main drivers of deforestation in the region, which have resulted in severely fragmented forests throughout its range. As a result of these ongoing pressures, C. litoralis is currently known only from threat-defined locations (sensu IUCN 2012, 2022). While the observations reported here are the first to suggest that the species can tolerate altered habitat adjacent to forest, at least to some degree, the natural history and habitat requirements of the species remain poorly understood. Accordingly, and like the recent threat assessment for its national status in Ecuador ( Ortega-Andrade et al. 2021), we recommend a global threat status of Endangered (EN) for C. litoralis following IUCN criteria B2ab(iii).

Call analysis. The call of C. litoralis consists of a short, single tonal note ( Fig. 3). The call duration was 88.51– 177.17 ms (x - = 132.84 ± 44.33; N = 4), the dominant frequency ranged from 5,210 –5,304 Hz (x - = 5,257 ± 47; N = 4), and the call bandwidth ranged from 738–1,729 Hz (x - = 1,265 ± 527; N = 4).

Compared to the available call descriptions of other species in the genus, C. nola and C. mache have similar call structures and parameter metrics. While C. nola exhibits a simple, non-pulsed note with comparable metrics (call duration: x - = 95 ms ± 11.97, dominant frequency: x - = 5,460 Hz ± 221; Lötters and Köhler 2000; Köhler et al. 2006), C. mache has a call with two pulsed notes, a call duration of - x = 38 ms ± 8, and a dominant frequency of x - = 5,410.2 Hz ± 17.9 ( Ortega-Andrade et al. 2013). Other Cochranella spp. that have described calls, such as C. granulosa ( Fig. 3), have pulsed notes.

The call metrics measured from seven call recordings of C. granulosa observed at Jardín de los Sueños, Cotopaxi, Ecuador are as follows. Calls consisted of 1–4 notes (- x = 2.34), with 8–15 pulses per note; the first notes in multi-note calls are more pulsated than subsequent notes (first note: x - = 15 ± 2 pulses; second note: x - = 12 ± 2 pulses; third note x - = 8 ± 5 pulses); call duration was 150–1,437 ms (- x = 790 ms ± 640); single-note duration varied from 130–260 ms, with the first note generally being longer than subsequent notes, similar to the calls of individuals from Costa Rica ( Ibáñez et al. 1999; Kubicki 2007). The note interval was 45.49–179.31 ms (- x = 112 ± 66.91), and the dominant frequency was measured at 3,943 –4,119 Hz (x - = 4,031 ± 88). Comparable metrics can be found in C. guayasamini , as it also exhibits a high-pitched, pulsed trill with two notes, with the first note having substantially more pulses than the second note ( Twomey et al. 2014). Similar to the lack of phenotypic variation observed between populations of C. granulosa in Ecuador and Central America ( Culebras et al. 2020), call variation also appears to be minimal.