A review of Northern Peruvian glassfrogs (Centrolenidae), with the description of four new remarkable species Author Twomey, Evan Author Delia, Jesse Author Castroviejo-Fisher, Santiago text Zootaxa 2014 3851 1 1 87 journal article 10.11646/zootaxa.3851.1.1 5e639fea-f2a5-4151-9914-307ec6cc8335 1175-5326 286921 9840D64B-F08C-44E7-B2DC-4818F8FFDD4F Hyalinobatrachium pellucidum ( Lynch & Duellman 1973 ) Figure 26 Centrolenella pellucida Lynch and Duellman 1973 . Holotype KU 143298, an adult female collected from “Rio Azuela, 1740 m , Quito-Lago Agrio road, Provincia Napo, Ecuador ”. Hyalinobatrachium pellucidum— Ruiz-Carranza and Lynch 1991 Hyalinobatrachium lemur Duellman and Schulte 1993 Background information. This species was described from a single adult female ( Lynch & Duellman 1973 ) based on the following characters: (1) heart visible in life (at the time, a state known only in two other species of Ecuadorian centrolenids), and (2) ulnar, anal, and tarsal folds present. Twenty years later, Duellman and Schulte (1993) described Hyalinobatrachium lemur from the east-Andean versant of San Martín, Peru . They diagnosed H. lemur from H. pellucidum on the basis of skin texture (smooth in H. lemur vs. shagreened in H. pellucidum ) and the extent of webbing on Finger IV (webbing more extensive in H. lemur ). Castroviejo-Fisher et al. (2009) reviewed the type series of both H. pellucidum and H. lemur , as well as additional fresh specimens of the former species both from Ecuador and southern Peru . They concluded that the characters originally used to diagnose H. lemur fell within the range of variability exhibited by H. pellucidum . Thus, they designated H. lemur as a junior synonym of H. pellucidum and identify the paratype as a H. carlesvilai . New data. Specimens have now been found in the Cainarachi valley near Tarapoto, San Martín, Peru ( 6°25'16.7''S , 76°17'' 28.5W , 523 m ). Wen et al. (2012) referred these specimens to H. pellucidum and also provided a description of the advertisement call, which is described as a single, monotone note without amplitude modulation. Also, a gravid female MNCN 45953 was collected on 28 March 2010 by M. Guerra Panaijo and Santiago Castroviejo-Fisher at Puente Chino on the Tocache-Pucallpa road, approximately 29 km NE from Tingo Maria ( 9°8'9.03"S , 75°47'15.78"W , 1147 m ), Ucayali, Peru , where it is sympatric with H. carlesvilai . The specimen was found on the upper side of a tree leaf ~ 4 m above the water level. This is the first report of this species for Ucayali. The identity of this specimen was confirmed by morphology and phylogenetic analysis ( Fig. 2a ). FIGURE 26. Spider predation of Hyalinobatrachium pellucidum embryos. a) A spider was located near a male and three clutches. b) The spider was lightly tapped with a blade of grass, at which point it moved towards the clutch and removed an embryo. c) The male appeared to notice the predation, and d, e) began moving over the clutch towards the spider, however, the spider continued consuming the embryo. f) Several days later it appeared that the male had abandoned his territory. The spider had built a web over all three clutches and was seen to consume embryos over the course of several nights. Natural history. Spider predation was observed in several clutches from a single male’s territory at the San Jose locality mentioned above. During a brief visit to the study stream on 15 May 2011 a male was encountered brooding a newly deposited clutch at 22:57. We returned again on May 24th , finding that the male had sired an additional 2 clutches all positioned next to each other on the lower surfaces of a leaf. At 22:00 a small spider was observed on the same leaf as the male and his egg clutches (see Fig. 26 for the following sequence). We lightly tapped the spider with a blade of grass on the abdomen. The spider then moved towards the oldest clutch, immediately removed an embryo from the clutch with its chelicerae and began consuming the embryo. The male, who was positioned on the opposite side of the clutches, moved over the eggs towards the spider with roughly half of his body positioned over two clutches. The male remained in this position for ca. 2 minutes, even when the spider came in contact with the male’s hind limb. After ca. 2 min the male moved back on top of a clutch and began brooding while the spider continued consuming the embryo. We returned to this site again on 30 May , finding that the male had abandoned his territory (or was predated), and that the spider had built a web over the clutches ( Fig. 26 ). The spider was observed to consume embryos during 4 nights until 2 June , when all embryos had either hatched or were predated. Over this period the spider was found at the nest site during both day and night, but was only observed consuming embryos at night.