*2.2. Morphological Characters*

We followed Duellman and Lehr [18] and Lynch and Duellman [19] for formats of diagnosis and description, except for using the term "dentigerous processes of vomers" instead of "vomerine odontophores" [20]. For taxonomy we follow Padial et al. [21] and Heinicke et al. [22]. We measured the following variables to the nearest 0.1 mm with digital calipers under a stereomicroscope: snout-vent length (SVL), tibia length (TL), foot length (FL, distance from proximal margin of inner metatarsal tubercle to tip of Toe IV), head length (HL, from angle of jaw to tip of snout), head width (HW, at level of angle of jaw), eye diameter (ED), tympanum diameter (TY), interorbital distance (IOD), upper eyelid width (EW), internarial distance (IND), eye–nostril distance (E–N, straight line distance between anterior corner of orbit and posterior margin of external nares), eye to tympanum distance (E-TY), forearm length (ForL), hand length (HaL), finger I length (FIL), finger II length (FIIL), toe I length (TIL), and toe II length (TIIL). Fingers and toes are numbered preaxially to postaxially from I–IV and I–V, respectively. We determined comparative lengths of toes III and V by adpressing both toes against toe IV; lengths of fingers I and II were determined by adpressing the fingers against each other. We compared the new taxon with all described species. Specimens examined are listed in Appendix A; codes of collections are: CORBIDI = Centro de Ornitología y Biodiversidad, Lima, Peru; MUBI = Museo de Biodiversidad del Peru, Cusco, Peru; MUSM = Museo de Historia Natural Universidad Nacional Mayor de San Marcos, Lima, Peru; MUSA = Museo de Historia Natural de la Universidad Nacional de San Agustín, Arequipa, Peru; UMMZ = University of Michigan Museum of Zoology, Ann Arbor, MI, USA.

Additionally, we used the morphological data to examine if body size and body shape vary across elevations. We used principal components analysis (PCA) to examine morphological variation between males and females, and between low (<300 m) and high elevation (1200–1490 m) populations. To remove the possible confounding e ffect of body size, we performed a body size-correction in which all variables were divided by SVL. We used generalized least-squares (GLS) regression to examine the relationship between SVL and elevation, and fitted a regression line separately for males and females. Subsequently, we calculated PCA using the corrected morphological data. We projected the first two

PC axes on a morphospace using the function princomp in R and displayed differences between male and females taking into account elevation.
