Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 1-11 (e234).

Call survey indicates rainbow trout farming alters glassfrog

community composition in the Andes of Ecuador

1*Katherine L. Krynak, 7Dana G. Wessels, *Segundo M. Imba, *Timothy J. Krynak, 7Eric B. Snyder,

3Jane A. Lyons, and **Juan M. Guayasamin

'Department of Biological and Allied Health Sciences, Ohio Northern University, Ada, Ohio, USA *Department of Biology, Grand Valley State

University, Grand Rapids, Michigan, USA *Reserva Las Gralarias, Province of Pichincha, ECUADOR ‘Natural Resources, Cleveland Metroparks,

Cleveland, Ohio, USA *Universidad San Francisco de Quito, Colegio de Ciencias Biologicas y Ambientales, Instituto Bidsfera-USFO, Laboratorio

de Biologia Evolutiva, Av. Diego de Robles y Via Interocéanica, Campus Cumbaya, Quito, ECUADOR Department of Biology, University of North

Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Abstract.—Aquaculture, the farming of fish for human consumption and/or trade, is a growing industry

throughout the world. The effects of farming on local ecosystems and wildlife are understudied, particularly

in regions where farms are often limited to subsistence practices with little to no government regulation. The

influence of Rainbow Trout (Oncorhynchus mykiss) farms on glassfrog community composition was assessed

in the Mindo and Alambi regions of Ecuador. Call surveys were conducted during the dominant glassfrog

reproductive season (March—May 2017) across 13 sites, six of which were in the immediate proximity of trout

farms. Nonmetric multidimensional scaling ordination analyses and multiple response permutation procedures

indicate that glassfrog communities differed between trout farm and non-trout farm sites (MRPP; A= 0.11, P=

0.04). Differences in glassfrog community composition were significantly or marginally correlated with percent

canopy openness, dissolved oxygen (mg/L), conductivity (uS), and total dissolved solids (mg/L), environmental

characteristics altered by the aquaculture practice. As the prevalence of trout farms increases across this

region, it is likely that the glassfrog community composition will be altered, potentially resulting in a pattern

of decreased species richness. It is also likely that habitat changes associated with trout farming practices

including deforestation, water chemistry changes, and predation pressures by escaped trout will influence

glassfrog species persistence. Mitigation strategies including improved barriers to decrease trout escape, the

incorporation of settling ponds to decrease stream contamination, and the preservation of habitat in areas of

high amphibian species richness are warranted.

Keywords. Amphibians, Anura, aquaculture, cloud forest, conservation, habitat protection, introduced species, land

management, water quality

Citation: Krynak KL, Wessels DG, Imba SM, Krynak TJ, Snyder EB, Lyons JA, Guayasamin JM. 2020. Call survey indicates rainbow trout farming

alters glassfrog community composition in the Andes of Ecuador. Amphibian & Reptile Conservation 14(2) [General Section]: 1-11 (e234).

4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any

medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced, are

as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 10 January 2019; Accepted: 1 March 2020; Published: 1 May 2020

Introduction

Aquaculture, or the farming of fish for human consumption

and/or trade, is a growing industry worldwide (Mantri

et al. 2017). While the practice has economic benefit

(Offemet al. 2010), the long-term costs to local wildlife

and ecosystems are largely understudied and likely

underestimated (Niklitschek et al. 2013). Aquaculture

practices in regions with little to no government regulation

may be particularly detrimental to the surrounding

ecosystem because habitat protection practices are often

not utilized, resulting in increased farmed-fish escapes

and water contamination (Niklitschek et al. 2013).

Correspondence. *k-krynak@onu.edu

Amphib. Reptile Conserv.

Rainbow Trout, Oncorhynchus mykiss (Salmonidae), is

a non-native predatory species currently being cultivated

in Andean streams that also are the habitat for several

of Ecuador’s most endangered amphibians (Vimos et al.

2015), including glassfrogs (Centrolenidae). Whether

introductions of O. mykiss have negatively affected

glassfrog populations is currently not known; however,

multiple studies indicate broad negative effects of this

cultivated fish species on amphibians (Gall and Mathis

2010; Garcia et al. 2012; Ortubay et al. 2006; Pearson

and Goater 2009; Vredenburg 2004). Oncorhynchus

mykiss represents a direct threat to amphibian larvae

via predation due to their biphasic life cycle (Garcia et

May 2020 | Volume 14 | Number 2 | e234

Impacts of fish farming on glassfrogs in Ecuador

Fig. 1. Trout farming in the Mindo region of E

cuador utilizes a flow-through aquaculture technique. Stream water is dive

rted

into tandem raceways/holding reservoirs and then flows through these reservoirs back into the natural stream system. This figure

displays a panoramic view of Finca de Jaime’s (FJ) set-up. Photograph by Katherine L. Krynak.

al. 2012; Pearson and Goater 2009). Many amphibian

species have been shown to demonstrate a lack of

predator avoidance in response to this introduced fish

(Gall and Mathis 2010; Garcia et al. 2012), though

this information is limited to temperate amphibian

larvae and it is unknown whether predator avoidance 1s

demonstrated in tropical amphibian larvae. However, in

a recent laboratory study, Martin-Torrijos et al. (2016)

found that the presence of O. mykiss altered larval

morphology in Nymphargus grandisonae, a glassfrog

species included in this survey. The extent to which O.

mykiss presence may affect the glassfrog larvae in situ

has yet to be determined. Additionally, O. mykiss can

introduce pathogens to naive amphibian communities,

including aquatic fungal pathogens such as Saprolegnia

diclina (Martin-Torrijos et al. 2016) and iridoviruses

like ranavirus, a pathogen that has caused amphibian

population declines and extirpations across the globe

(Miller et al. 2011; Smith et al. 2017). Together, these

studies suggest O. mykiss introductions may negatively

affect glassfrog population persistence by decreasing

larval survival through both direct (predation) and

indirect means (aquatic pathogen introduction).

Farming of O. mykiss has been occurring for over

25 years in the Mindo and Alambi regions of Ecuador

(western slope of the Andes Mountains, Province of

Pichincha) and is increasing in prevalence; several

farms in the region are fewer than 10 years old (Rolando

Sanchez and JAL, pers. comm.). Trout farms in the

Mindo and Alambi regions utilize a flow-through system

of aquaculture. Natural stream water is diverted into

tandem holding reservoirs (and/or raceways; Fig. 1);

water then flows through these reservoirs back into the

natural stream system. The system has no mechanism for

preventing stream contamination other than the limited

settling that occurs in reservoirs prior to outflow. Fish

escapes are largely prevented by size sorting of trout

between reservoirs (smaller fish being held in the first

reservoirs, larger fish nearer the outflow) and wire screen

barriers put in place to limit escape. Interviews of farm

managers indicated that heavy rains (notably during the

months of March—May) often result in large amounts of

debris being swept into the diverted stream channels,

which damages the wire barriers that contain the trout in

the reservoirs. Managers estimated that 2—-10% of farmed

Amphib. Reptile Conserv.

trout escape during these common rain events. Between

1 March 2017 and 22 May 2017, 588.70 mm of rain fell

in this region (HOBO U30 Remote Monitoring System

stationed at Reserva Las Gralarias: 0.0091S, 78.7375W,

elevation 2,068 m).

Particular stream characteristics associated with

stream diversion aquaculture may affect aquatic larval

glassfrog survival and, over time, influence glassfrog

community composition. Total dissolved _ solids,

conductivity, and pH are often altered by trout farming

(Boaventura et al. 1997; McNaughton and Lee 2010)

and are known to influence amphibian fitness correlates

including growth, development, and innate immune

defenses (Krynak et al. 2015, 2016). Trout farming

in this region may be particularly detrimental to water

quality given that multiple trout farms often occupy the

same stream, potentially causing a cumulative effect on

water quality. Increased stream water nutrient loads can

increase periphyton abundance (a larval glassfrog food

source) and subsequently decrease dissolved oxygen

levels (Selong and Helfrich 1998), thereby negatively

affecting larval glassfrog survival (Gillespie 2002;

Tattersall and Ultsch 2008). In temperate systems,

canopy cover (or lack thereof due to deforestation) can

also influence periphyton abundance by changing light

availability, potentially altering available larval food

sources (Skelly et al. 2002), though context dependency

of this relationship may be greater in tropical ecosystems

(Garcia et al. 2015). Furthermore, changes to stream

canopy composition may negatively affect glassfrog

persistence by decreasing suitable egg deposition sites

(Arteaga et al. 2013).

The Mindo region of Ecuador is home to nine species

of glassfrogs (Arteaga et al. 2013) that vary in their 2017

IUCN Redlist conservation status from Data Deficient

(DD) to Critically Endangered (CR): Emerald Glassfrog

(Espadarana prosoblepon; Least Concern [LC]), Red-

spotted Glassfrog (Nymphargus grandisonae;, LC),

Pepper Glassfrog (N. griffithsi; Vulnerable [VU]), Las

Gralarias Glassfrog (N. /asgralarias; DD), Lynch’s

Glassfrog (Centrolene lynchi,; Endangered [EN]),

Golden-flecked Glassfrog (C. ballux; CR), Dappled

Glassfrog (C. peristictum; Near Threatened [NT]), Mindo

Glassfrog (Cochranella balionota; VU), and Bumpy

Glassfrog (C. heloderma; CR) [Table 1; Fig. 2]. Previous

May 2020 | Volume 14 | Number 2 | e234

Krynak et al.

studies have suggested that glassfrog population declines

might be partially associated with the introduction of

predatory fish into streams, though this effect has not been

quantified (Catenazzi et al. 2011; Merino-Viteri 2001). In

comparison extensive work has been done in temperate

systems indicating that introduced trout have devastated

amphibian communities (Bosch et al. 2019; Knapp and

Matthews 2000; Knapp et al. 2007; Pope 2008).

In this study, presence/absence acoustic surveys

were conducted throughout the dominant glassfrog

breeding months of March—-May 2017 (Arteaga et al.

2013), to determine the influence of trout farms on

glassfrog community composition in the Mindo region

of Ecuador. The predictions were that trout farms would

have decreased glassfrog species richness and that

particular environmental characteristics (such as water

chemistry, periphyton abundance, and canopy cover)

would correlate to differences in glassfrog community

composition between trout farms and non-trout farms.

Materials and Methods

Call surveys were conducted across 13 sites in the Mindo

region of Ecuador (six trout farms and seven without

trout; Fig. 3), one of the most amphibian-diverse cloud

forests in South America (Arteaga et al. 2013). Sites

were chosen based upon habitat viability, elevation,

and accessibility. Adult glassfrogs in the region inhabit

forested habitats surrounding creeks, streams, and rivers.

Sites included in the study ranged in elevation from

1,596—2,666 m, and habitat was considered to be viable

for glassfrog presence if at least small remnants of forest

surrounded the streams or their tributaries (for streams

both with and without trout farms). Six sites were located

at trout farms along the Rio Alambi and Quebrada Santa

Rosa waterways. The Rio Alambi water system included

trout farm sites referred to as El Paraiso del Pescador

(EP), Finca de Jaime (FJ), Santa Teresita (ST), La

Sierra (LS), and Verdecocha (VC). A single trout farm

was located on Quebrada Santa Rosa system, the Lower

Rio Santa Rosa (LRSR) site. The trout farms ranged in

age from 6—27 years. The non-trout farm sites included

four sites along the Quebrada Santa Rosa stream system

(upstream of LRSR), referred to as Rio Santa Rosa

(RSR), Michelle’s (M), Five Frog Creek (SF), and Ballux

Creek (Berk). Three additional non-trout farm sites were

chosen that represent headwater streams not connected

with Quebrada Santa Rosa or Rio Alambi: Lucy’s

Creek (LC), Kathy’s Creek (KC), and a small tributary

of the Chalguayacu Grande River (C). Ballux Creek

and Five Frog Creek also represent headwater stream

systems forming Quebrada Santa Rosa (see Appendix 1

for details on site locations). The non-trout farm sites,

with the exclusion of LRSR, are located on privately-

owned protected land. Access to headwaters of trout

farm streams was not possible due to transportation and

permission constraints.

Amphib. Reptile Conserv.

i + “gas: ees ‘aai*, i

; Gy we eee Hl

Fig. 2. Glassfrog species found during surveys. (A) Centrolene

heloderma, (B) Centrolene ballux, (C) Esparana prosoblepon,

(D) Nymphargus lasgralarias, (E) Centrolene peristictum, (F)

Nymphargus grandisonae, (G) Centrolene lynchi, (A) Egg mass

from C. ballux. Nymphargus griffithsi was not encountered

during the 2017 survey but has been documented at Kathy’s

creek in 2012 and 2013 (by Jane A. Lyons). Photographs by

Dana G. Wessels (A—F) and Timothy J. Krynak (G—H).

Overnight call surveys were conducted between 2000

h and 0200 h at each survey site (Mean: 3.38 + 1.9 SD

visits per site) on multiple dates during the rainy season,

when glassfrogs are reproductively active (JMG, pers.

comm.). Surveying included 2-8 visits per site with the

exception of a single site (C) which was only visited on

a single occasion due to safety concerns associated with

heavy rains and steep, eroding terrain. The presence of

each of the documented species was recorded at first visit

at each site, therefore sampling effort did not bias the

detection. It should be noted that the species recorded

at site C on 30 March 2017 were the same as had been

previously observed at the site in March—May in 2012 and

2013 (JAL, pers. comm.). Species presence was assessed

May 2020 | Volume 14 | Number 2 | e234

Impacts of fish farming on glassfrogs in Ecuador

Table 1. Glassfrog species presence/absence data across sites in the Mindo region of Ecuador. Sites: RSR = Rio Santa Rosa, LC =

Lucy’s Creek, Berk = Ballux Creek, KC = Kathy’s Creek, M = Michelle’s, C = Chalguayacu Grande River, 5F = Five Frog Creek,

LRSR = Lower Rio Santa Rosa, ST = Santa Terricita, FJ= Finca de Jaime, LS = La Sierra, VC = Verda Cocha, EP = El Paraiso del

Pescador. Values of 0 indicate those “not detected,” whereas values of 1 indicate those audibly detected. IUCN RedList status codes

are listed below each species name.

Species and IUCN RedList conservation status

Centrolene

lynchi

Nymphargus

grandsiosonae

Nymphargus

lasgralarias

fo)

Co oO Co Oo CO Co OC 8S SO = 82 DS S&S

ooolmlUlUlCOmUllUCUCOCOCCO COT COT RSF RF OCCU Rl

(rp me ED a CR ed a DS

based upon audible detection of calling males across an

approximate 200 m distance (up to 100 m above and 100

m below the stream access point). For all sites, detectable

species richness did not change as a function site visits.

Audible recording of calls used for identification for

each of the observed species can be referenced at http://

lasgralariasfoundation.org/cantos-de-ranas.

The location and environmental characteristics

recorded at each of the sites included water temperature

(°C), conductivity (uS/cm), total dissolved solids

(mg/L), dissolved oxygen (mg/L), pH, canopy cover

(% openness), and average periphyton abundance

(chlorophyll a mg/cm?). Site elevation was determined

by estimation via Google Maps™, which was then

corroborated via topographical maps obtained from

Ecuador’s Ministry of the Environment. Minnow traps

(Grayson and Row 2007) were deployed downstream of

each trout farm (with the exception of LRSR) for 24-hr

periods to assess the abundance of escaped trout (catch

per unit effort).

All statistical analyses were performed in R version

3.4.3 (R Development Core Team 2017). For all sites

with detectable glassfrogs, the influence of trout farming

on glassfrog species richness was assessed using a /-test

(t-test function in the stats package by R Core Team 2014).

Sites without detectable glassfrogs were not included in

this analysis to avoid artificially inflating results of the

test. Glassfrog community composition was assessed

using nonmetric multidimensional scaling ordination

using Jaccard distance (NMDS; metamds function in

the vegan package, Oksanen et al. 2018) to visualize

Amphib. Reptile Conserv.

Centrolene

heloderma

Centrolene

ballux

Espadarana

prosoblepon

Centrolene

peristitum

rs)

Zc occ OcUCcCOCUCOCCcOUUlmrHM CC OCUr HR Or Se

cas ld > oo og Ro

oocorco fe Ge co Ke SO

oo 2 emo uw’ 6.6. S24 a SoS

glassfrog community composition similarities across

the 13 sites. Trout farms El Pariso del Pescador, La

Sierra, and Verdecocha are not included in the analysis

because glassfrogs were not observed at these sites.

Multiple response permutation procedures (MRPP;

999 permutations; mrpp function in the vegan package,

Oksanen et al. 2018) were used to quantify differences

in glassfrog community composition between trout

farms and non-trout farm sites. Pearson correlation

tests (cor.test function in the stats package by R Core

Team 2014) between axis scores and the environmental

measures were performed to assess the potential

influences of these environmental characteristics

on glassfrog community similarity in the NMDS

ordination. Moran’s / was used to assess potential spatial

autocorrelations between the environmental variables

and the GPS locations of the sites (Moran.I function in

the ape package, Paradis and Schliep 2018). Glassfrog

taxonomy follows the proposal by Guayasamin et al.

(2009). All surveys were conducted under permit MAE-

DNB-CM-2015-0017 issued by Ecuador’s Ministry of

the Environment (Ministerio de Ambiente del Ecuador)

and with permission of land owners.

Results

A total of seven glassfrog species was recorded

across the 13 sites (Table 1). Species detected were

Centrolene heloderma, C. ballux, C. peristictum, C.

lynchi, Nymphargus grandisonae, N. lasgralarias, and

Espadarana prosoblepon.

May 2020 | Volume 14 | Number 2 | e234

Krynak et al.

Lower Rio

” Santa Rosa

Lucy's

~~ =n Five Frog

om, oa CimlqueVeeued ~@. Creek

7 epetiay

Creek

Rio Santa

| Rosa e.

Verdecocha, %

La Sierra

Santa Teresita

Finca de

Jaime

Paraiso de

Pescador

Fig. 3. Map of the study area. Inset: Pichincha Province, Ecuador. Main: Blue points indicate non-trout farm sites whereas red points

indicate trout farm sites. Yellow line represents the equator (latitude 0).

Glassfrog species richness differed between trout

farm and non-trout farm sites (tf = 2.94, df = 5.54, P

= 0.03; mean trout farm richness = 1.67 + 0.58 SE

species; mean non-trout farm richness = 3.0 + 0.82 SE

species) and the NMDS analyses indicated a difference

in glassfrog community composition between trout farm

sites and non-trout farm sites (MRPP; delta = 0.59, A =

0.11, P=0.04; Fig. 4). Pearson correlation tests indicated

correlations between multiple environmental variables

and NMDS axis scores (NMDS Stress on 2D solution

was 3% indicating good fit, Table 2, Fig. 4). Site elevation

was found to be positively correlated with NMDS Axis

observed at trout farms versus non-trout farms (Fig. 4).

Moran’s I tests revealed pH, chlorophyll a, canopy cover,

and dissolved oxygen were not spatially autocorrelated

(Moran’s I test P > 0.05); while conductivity, total

dissolved solids, elevation, and temperature did suffer

from spatial autocorrelation (Moran’s I test P < 0.05).

Correlation analyses were conducted on all variables

independently, including spatially autocorrelated

variables, given that it is unknown whether the spatial

autocorrelation was due to exogenous or endogenous

factors and the small sample size.

Traps for quantifying Rainbow Trout abundance were

1 (Tig) = 2.73, P = 0.03). NMDS Axis 2 was correlated —_ not effective at the sites and therefore, this effort was

with canopy cover (T,,,.= 2.73, P = 0.03) and dissolved discontinued after multiple attempts (see Discussion).

oxygen Ces 3.16, P= 0.01; Table 2). NMDS axis 2.‘ These traps did, however, catch a single Astroblepus

was marginally correlated with conductivity Ms, 9 = 2.0, sp., a native (non-predatory) fish from the family

P = 0.08) and total dissolved solids (T.,.. = 2.12, P= = Astroblepidae known for climbing waterfalls in these

(8,9)

0.07). NMDS Axis 2 differentiates glassfrog communities

Andean streams at site FJ.

Table 2. Pearson correlation estimates between NMDS axis scores and environmental variables across sites. Statistically significant

values are in bold (P < 0.05). Values of P < 0.1 are indicated with an asterisk (*), and are designated as such based upon their

biological significance and the small sample size.

Total

dissolved Dissolved Canopy

Conductivity solids oxygen openness’ Elevation Temperature Chlorophyll

pH (uS/cm) (mg/L) (mg/L) (“%) (m) (°C) a (mg/cm’)

NMDS1 0.40 0.08 0.18 0.06 <0.01 0.70 -0.39 0.38

NMDS2 0.39 0.58* 0.60* 0.75 0.69 -0.07 -0.50 0.14

Amphib. Reptile Conserv. 5 May 2020 | Volume 14 | Number 2 | e234

Impacts of fish farming on glassfrogs in Ecuador

E. prosoblepon

1.5

LRSR

1.0

He ST

Me e C. heloderma

N. grandisonae 2F @ Bcrk@

RSR/LCe®

Clynchi c@ —— C. ballux

C. Beecerane e

NMDS2

O05 00 05

-1.0

N. lasgralarias

-1.5

ey 2 = 0 1 2

NMDS1

Fig. 4. Two-dimensional NMDS ordination of survey sites

and glassfrog species based upon presence of frogs audibly

documented in 2017 survey conducted in the Mindo region

of Ecuador (Stress = 3%). Red points and labels represent

glassfrog species; grey points represent trout farms; and black

points represent non-trout farms. RSR = Rio Santa Rosa, LC =

Lucy’s Creek, Berk = Ballux Creek, KC = Kathy’s Creek, M

= Michelle’s, C = tributary of the Chalguayacu Grande River,

5F = Five Frog Creek, LRSR = Lower Rio Santa Rosa, ST

= Santa Teresita, FJ = Finca de Jaime, LS = La Sierra, VC =

Verdecocha, EP = El Paraiso del Pescador. Trout farms EP, LS,

and VC are not included in the analysis because glassfrogs were

not observed at these sites. A significant difference in glassfrog

community composition between trout farm and non-trout

farm sites was indicated by MRPP (delta = 0.59, A= 0.11, P =

0.03). NMDS1 correlated with elevation; NMDS2 correlated

with: percent canopy openness, dissolved oxygen (mg/L), total

dissolved solids (mg/L), and conductivity (uS).

Discussion

Understanding the potential effects that trout farming has

on glassfrog community structure 1s critical for improving

species conservation efforts as this aquaculture practice

is expected to increase in this region of Ecuador, and

throughout the world (Diana 2009). Across the thirteen

sites, the presence of seven glassfrog species is reported,

two of which are listed as Critically Endangered (C.

ballux and C. heloderma) and one as Endangered (C.

lynchi, IUCN Redlist 2017). Notably, at Michelle’s

site, Centrolene lynchi was recorded at a much higher

elevation (2,031 m) than previously documented for

the species (published elevational range 1,520—1,858

m; Arteaga et al. 2013). Additionally, a previously

undocumented population of C. heloderma (20+ calling

males) was recorded between the trout farm sites La

Sierra and Santa Teresita (and at Santa Teresita) along the

Rio Alambi system (Krynak et al. 2018). Nymphargus

griffithsi was not observed at any of the sites, though the

species has been recorded in Five Frog Creek at Reserva

Las Gralarias in previous years (Hutter and Guayasamin

2012).

Amphib. Reptile Conserv.

This survey found that (1) mean glassfrog species

richness nearly doubled in non-trout farm sites compared

to trout farm sites, (i1) glassfrog community composition

differed between trout farm sites and non-trout farm sites

(based on clear separation between these factors along

NMDS2 in the ordination and quantitative confirmation

via MRPP analyses), and (111) multiple environmental

measures (dissolved oxygen, canopy cover, total dissolved

solids, and conductivity) were correlated with observed

differences in glassfrog community composition (Table

2; Fig. 4). There are several possible explanations for

these marked differences, as previous research has

indicated amphibian community composition and larval

performance are associated with water chemistry, riparian

cover, and predator presence and the findings reported

here provide additional support for these hypotheses

(Gonzalez-Maya et al. 2018; Hecnar and M'Closkey

1996; Sebasti and Carpaneto 2004; Watling et al. 2011).

This study indicates that water chemistry (measures of

dissolved oxygen, total dissolved solids, and conductivity)

is associated with the difference in glassfrog community

composition between trout farm and non-trout farm sites.

It is probable that increased nutrient loads associated

with uneaten food and fecal waste from the trout may

be driving the increased total dissolved solids (TDS)

content at trout farms sites (Selong and Helfrich 1998).

Increased nutrients from flow-through aquaculture are

known to negatively affect larval amphibian survival by

increasing periphyton and thereby decreasing dissolved

oxygen (DO) content (Gillespie 2002; Tattersall and

Ultsch 2008). However, the measured DO levels were

slightly higher at trout farm sites compared to non-trout

farm sites (mean + SE DO: trout farms = 7.9 + 0.58 mg/L,

non-trout farms = 7.4 + 0.55 mg/L). This phenomenon

may be common within tropical ecosystems, or context

dependent (Garcia et al. 2015). An expanded sampling

effort will be required to tease apart these possible

relationships. Another possibility is that the cooler water

temperatures associated with slighting higher elevations

of one of the trout farm sites may be driving this difference

in DO (Appendix 2, mean + SE temperature: trout farms

= 14.84 °C + 0.65 °C, non-trout farms = 15.55 + 0.93

°C; elevation range: trout farms = 1,593—2,666 m, non-

trout farms 1,693—2,254 m; mean + SE elevation: trout

farms = 2,013 + 16.7 m, non-trout farms = 2,020 + 12.9

m). Surprisingly, periphyton abundances did not differ

between trout farm and non-trout farm sites in this study.

However, we suspect that the increased nutrient levels (as

suggested by TDS) may be affecting water chemistry in

terms of ammonia and nitrite levels in the system, which

could in turn negatively affect larval glassfrog survival;

although this hypothesis needs to be tested.

The correlation found between canopy cover and

glassfrog community composition differences between

non-trout farm and trout farm sites, as visualized by

the separation along NMDS2 in the ordination and

quantitatively confirmed by the MRPP analyses, may

May 2020 | Volume 14 | Number 2 | e234

Krynak et al.

indicate the deforestation at trout farm sites influenced

which glassfrog species inhabited the sites. Based upon

previous literature, we initially hypothesized that the

mechanism for this correlation is that decreased canopy

cover causes increased periphyton abundance (increased

food availability which may benefit only particular

amphibian larvae); however, the results obtained here

contradict this idea (1.e., periphyton measures not

correlated with NMDS axis 2 scores; Skelly et al. 2002).

The decreased canopy cover at trout farm sites may

instead be detrimental to the glassfrog species because of

the lack of egg deposition sites. Overhanging vegetation

along streams is critical to glassfrog reproductive

success. Glassfrogs of this region lay eggs on leaves

overhanging streams (plant families include Araceae,

Annonaceae, Euphorbiaceae, Capparaceae, Fabaceae,

and Rubiaceae) and upon hatching, the rheophilic

larvae drop into the stream below where they continue

to grow and mature (Arteaga et al. 2013). Therefore,

a decrease in the number or quality of egg deposition

sites (Canopy cover) may result in decreased glassfrog

abundance. Furthermore, decreased canopy cover may

also negatively affect glassfrogs by means of increased

ultra-violet (UV) exposure, as UV radiation is known to

negatively affect amphibians at all life stages (Blaustein

et al. 2003). Finally, while generalized deforestation

(and canopy cover loss) cannot be separated from the

deforestation caused by the creation and maintenance of

the trout farms, this lack of vegetation (or appropriate

vegetation) does seem to negatively affect the glassfrog

community richness.

Lastly, the direct effect of predation and indirect effects

of perceived predation threat by trout on glassfrog larvae

in situ remain in need of assessment. During the surveys,

an attempt was made to quantify trout presence directly

measured by catch per unit effort via direct trapping and

indirectly measured via collection of O. mykiss DNA from

the streams. However, both efforts were discontinued

due to ineffectiveness. Although trout were not directly

observed in the streams, and there was no success in

capturing trout using the minnow traps (despite multiple

attempts and equipment adjustments), local people were

seen pole fishing 1n the streams for the trout at El Paraiso

del Pescador and near Santa Teresita. While the use of

environmental DNA (eDNA) has become a valuable tool

for assessing species presence in stream habitats (Young

et al. 2017), there are limitations which must be addressed

to fully utilize this tool in these fast-flowing Andean

streams. The 10u nylon membranes used to filter the

stream water to collect the DNA samples were found to

clog rapidly, limiting the ability to standardize collection

efforts and obtain enough samples for comparisons

across streams. Such assessments may be better suited

for times of the year when there is less rainfall, when

larval trout are not being washed downstream and stream

water is less turbid. Electrofishing was not used to sample

the trout because this methodology may have negative

Amphib. Reptile Conserv.

effects on small vertebrates, including glassfrog larvae

(Miranda and Kidwell 2010). Nevertheless, changes

in tadpole survival, morphology, behavior, and fitness

when fish predators are present has been documented

extensively (Relyea 2001, 2004; Relyea and Hoverman

2003), and may be a widespread phenomenon in Andean

amphibian communities (Martin-Torrijos et al. 2016).

As such, the effects of O. mykiss presence on Andean

stream inhabitants is deserving of further investigation,

especially when an overall negative effect of trout farms

on amphibian richness has been correlated to multiple

environmental characteristics associated with trout

farming, as demonstrated in this study.

Conclusions

As trout farming increases in the Andean cloud forests,

environmental managers need to be concerned about

direct and indirect effects the practice has on naive

communities. While the persistence of the few glassfrog

populations found at the trout farm sites provides

encouragement, the differences in glassfrog community

composition indicate that areas of high glassfrog

species richness should be protected from the farming

of non-native predatory fish. While minimizing water

contamination (e.g., implementation of settling pools)

and preventing fish escapes may be enough to maintain

the existing populations in the streams currently used

for aquaculture, we suspect that naive communities

may undergo a decrease in diversity 1f new farms are

constructed. The results of this study suggest that

mitigation strategies need to be employed in streams

currently used in aquaculture and that trout farming

should be prohibited in areas of high glassfrog species

richness in order to protect these species.

Acknowledgments.—We thank the Fulbright Commission

of Ecuador (Fulbright Early Career Fellowship issued

to KLK), Grand Valley State University (Presidential

Grant to DGW), and the Michigan Space grant

(issued to DGW) for funding this research. We thank

the Universidad de San Francisco de Quito and

Universidad Tecnologica Indoamérica for hosting

KLK and DGW for this research, and supporting

JMG’s research (Collaboration Grant 11164). We

thank K. Becka for her assistance with data collection;

R. Sanchez for his assistance with transportation and

his knowledge of the people and history of this region.

We also thank the trout farmers of the Rio Alambi

area, most notably the Villalba family at La Sierra,

for their hospitality and site access permissions; and

the staff of Reserva Las Gralarias for their hospitality

and unwavering commitment to habitat protection.

All surveys were conducted under permit MAE-DNB-

CM-2015-0017 issued by Ecuador’s Ministry of the

Environment (Ministerio de Ambiente del Ecuador)

and with permission of land owners.

May 2020 | Volume 14 | Number 2 | e234

Impacts of fish farming on glassfrogs in Ecuador

Literature Cited

Arteaga A, Bustamante L, Guayasamin JM. 2013. The

Amphibians and Reptiles of Mindo. Imprenta Mar-

iscal, Quito, Ecuador. 257 p.

Blaustein AR, Romansic JM, Kiesecker JM, Hatch AC.

2003. Ultraviolet radiation, toxic chemicals, and am-

phibian population declines. Diversity and Distribu-

tions 9: 123-140.

Boaventura R, Pedro AM, Coimbra J, Lencastre E. 1997.

Trout farm effluents: characterization and impact on

the receiving streams. Environmental Pollution 95:

379-387.

Bosch J, Bielby J, Martin-Beyer B, Rincon P, Correa-

Araneda F, Boyero L. 2019. Eradication of introduced

fish allows successful recovery of a stream-dwelling

amphibian. PLoS One 14: e0216204.

Catenazzi A, Lehr E, Rodriguez LO, Vredenburg VT.

2011. Batrachochytrium dendrobatidis and the col-

lapse of anuran species richness and abundance in the

Upper Manu National Park, southeastern Peru. Con-

servation Biology 25: 382-391.

Coloma LA, Ron S, Wild E, Cisneros-Heredia D. 2004.

Centrolene lynchi. The IUCN Red List of Threatened

Species 2004: ¢.T54924A 85877848.

Diana JS. 2009. Aquaculture production and biodiversity

conservation. BioScience 59: 27-38.

Gall BG, Mathis A. 2010. Innate predator recognition

and the problem of introduced trout. Ethology 116:

47-58.

Garcia EA, Townsend SA, Douglas MM. 2015. Context

dependency of top-down and bottom-up effects in a

Northern Australian tropical river. Freshwater Sci-

ence 34: 679-690.

Garcia TS, Thurman LL, Rowe JC, Selego SM. 2012.

Antipredator behavior of American Bullfrogs (Litho-

bates catesbeianus) in a novel environment. Ethology

118: 867-875.

Gillespie GR. 2002. Impacts of sediment loads, tadpole

density, and food type on the growth and development

of tadpoles of the Spotted Tree Frog, Litoria spenc-

eri: an in-stream experiment. Biological Conservation

106: 141-150.

Gonzalez-Maya JF, Gomez-Hoyos DA, Cruz-Lizano I,

Schipper J. 2018. From hope to alert: demography of

a remnant population of the Critically Endangered Ar-

elopus varius from Costa Rica. Studies on Neotropical

Fauna and Environment 53: 194—200.

Grayson KL, Row AW. 2007. Glow sticks as effective

bait for capturing aquatic amphibians in funnel traps.

Herpetological Review 38: 168-170.

Guayasamin JM, Castroviejo-Fisher S, Trueb L, Ayar-

zaguena J, Rada M, Vila C. 2009. Phylogenetic sys-

tematics of glassfrogs (Amphibia: Centrolenidae) and

their sister taxon Allophryne ruthveni. Zootaxa 2100:

1-97.

Hecnar SJ, M'Closkey RT. 1996. Amphibian species

Amphib. Reptile Conserv.

richness and distribution in relation to pond water

chemistry in south-western Ontario. Canada Fresh-

water Biology 36: 7-15.

Knapp RA, Boiano DM, Vredenburg VT. 2007. Removal

of nonnative fish results in population expansion of a

declining amphibian (Mountain Yellow-legged Frog,

Rana muscosa). Biological Conservation 135: 11—20.

Knapp RA, Matthews KR. 2000. Non-native fish intro-

ductions and the decline of the Mountain Yellow-

legged Frog from within protected areas. Conserva-

tion Biology 14: 428-438.

Krynak KL, Burke DJ, Benard MF. 2015. Larval envi-

ronment alters amphibian immune defenses differen-

tially across life stages and populations. PLoS One

10(6): e0130383.

Krynak KL, Burke DJ, Benard MF. 2016. Landscape

and water characteristics correlate with immune de-

fense traits across Blanchard's Cricket Frog (Acris

blanchardi) populations. Biological Conservation

193: 153-167.

Krynak KL, Wessels DG, Imba SM, Lyons JA, Guayasa-

min JM. 2018. Newly discovered Bumpy Glassfrog

(Centrolene heloderma) population, with discussion

of threats to population persistence. CheckList 14(1):

261-265.

Mantri VA, Eswaran K, Shanmugam M, Ganesan M,

Veeragurunathan V, Thiruppathi S, Reddy CRK, Seth

A. 2017. An appraisal on commercial farming of Kap-

paphycus alvarezii in India: success in diversification

of livelihood and prospects. Journal of Applied Phy-

cology 29: 335-357.

Martin-Torrijos L, Sandoval-Sierra JV, Mufioz J, Dié-

guez-Uribeondo J, Bosch J, Guayasamin JM. 2016.

Rainbow trout (Oncorhynchus mykiss) threaten An-

dean amphibians. Neotropical Biodiversity 2: 26-36.

McNaughton KA, Lee PF. 2010. Water quality effects

from an aquaculture operation in a meromictic iron pit

lake in Northwestern Ontario, Canada. Water Quality

Research Journal of Canada 45: 13-24.

Merino-Viter1 A. 2001. Analisis de posibles causas de las

disminuciones de poblaciones de anfibios en los An-

des del Ecuador. M.S. Thesis, Pontificia Universidad

Catolica del Ecuador, Quito, Ecuador.

Miller D, Gray M, Storfer A. 2011. Ecopathology of rana-

viruses infecting amphibians. Viruses 3: 2,351—2,373.

Miranda LE, Kidwell RH. 2010. Unintended effects of

electrofishing on nongame fishes. Transactions of the

American Fisheries Society 139: 1,315—1,321.

Niklitschek EJ, Soto D, Lafon A, Molinet C, Toledo P.

2013. Southward expansion of the Chilean salmon in-

dustry in the Patagonian Fjords: main environmental

challenges. Reviews in Aquaculture 5: 172-195.

Offem BO, Ikpi GU, Ada F. 2010. Fish culture technolo-

gies in south-eastern Nigeria. African Journal of Agri-

cultural Research 5: 2,521—2,528.

Ortubay S, Cussac V, Battini M, Barriga J, Aigo J, Alon-

so M, Macchi P, Reissig M, Yoshioka J, Fox S. 2006.

May 2020 | Volume 14 | Number 2 | e234

Krynak et al.

Is the decline of birds and amphibians in a steppe

lake of northern Patagonia a consequence of limno-

logical changes following fish introduction? Aquatic

Conservation- Marine and Freshwater Ecosystems 16:

93-105.

Pearson KJ, Goater CP. 2009. Effects of predaceous

and nonpredaceous introduced fish on the survival,

growth, and antipredation behaviours of long-toed

salamanders. Canadian Journal of Zoology 87: 948—

955.

Pope KL. 2008. Assessing changes in amphibian popula-

tion dynamics following experimental manipulations

of introduced fish. Conservation Biology 22: 1,572—

1,581.

Relyea RA. 2001. The relationship between predation

risk and antipredator responses in larval anurans.

Ecology 82: 541-554.

Relyea RA. 2004. Fine-tuned phenotypes: tadpole plas-

ticity under 16 combinations of predators and com-

petitors. Ecology 85: 172-179.

Relyea RA, Hoverman JT. 2003. The impact of larval

predators and competitors on the morphology and fit-

ness of juvenile treefrogs. Oecologia 134: 596-604.

Sebasti S, Carpaneto GM. 2004. An ecological study

on amphibian communities inhabiting the dewponds

of a lowland deciduous forest along the Tyrrhenian

coast (central Italy). /talian Journal of Zoology 71:

135-141.

Selong JH, Helfrich LA. 1998. Impacts of trout culture

effluent on water quality and biotic communities in

Virginia headwater streams. Progressive Fish-Cultur-

reptiles.

Amphib. Reptile Conserv.

ist 60: 247-262.

Skelly DK, Freidenburg LK, Kiesecker JM. 2002. For-

est canopy and the performance of larval amphibians.

Ecology 83: 983-992.

Smith TC, Picco AM, Knapp R. 2017. Ranaviruses infect

Mountain Yellow-legged Frogs (Rana muscosa and

Rana sierrae) threatened by Batrachochytrium den-

drobatidis. Herpetological Conservation and Biology

12: 149-159.

Tattersall GJ, Ultsch GR. 2008. Physiological ecology of

aquatic overwintering in ranid frogs. Biological Re-

views 83: 119-140.

Vimos DJ, Encalada AC, Rios-Touma B, Suarez E, Prat

N. 2015. Effects of exotic trout on benthic communi-

ties in high-Andean tropical streams. Freshwater Sci-

ence 34: 770-783.

Vredenburg VT. 2004. Reversing introduced species ef-

fects: experimental removal of introduced fish leads to

rapid recovery of a declining frog. Proceedings of the

National Academy of Sciences of the United States of

America 101: 7,646—7,650.

Watling JI, Hickman CR, Orrock JL. 2011. Predators and

invasive plants affect performance of amphibian lar-

vae. Oikos 120: 735-739.

Young MK, Isaak DJ, McKelvey KS, Schwartz MK,

Carim KJ, Fredenberg W, Wilcox TM, Franklin TW,

Chandler GL, Nagel DE, et al. 2017. Species occur-

rence data from the Range-Wide Bull Trout eDNA

Project. United States Forest Service, Fort Collins,

Colorado, USA. Available: http://www.fs.fed.us/rm/

boise/AWAE/projects/BullTrout_eDNA.html — [Ac-

cessed: 17 August 2019].

Katherine L. Krynak is an Assistant Professor of Biology at Ohio Northern University (ONU) and

a founding board member of Las Gralarias Foundation (North Olmstead, Ohio, USA). Her research

interests broadly span the fields of biodiversity conservation and eco-immunology. Katherine works

with Las Gralarias Foundation to help protect the biodiversity of the Mindo region of Ecuador

through research, environmental education, and land protection. Katherine was awarded a Fulbright

Early Career Fellowship (Ecuador) in 2016 that supported the research presented in this article.

Katherine teaches an undergraduate Tropical Biodiversity Conservation course in Ecuador for ONU.

Dana G. Wessels is a recent graduate of Grand Valley State University (Allendale, Michigan, USA)

with a Master’s in Biology and emphasis in aquatic science. Dana’s research was supported by the

President’s Grant from Grand Valley State University and by the Michigan Space Grant.

Segundo M. Imba is a passionate conservationist, excellent field assistant, and long-time staff

member at Reserva Las Gralarias (RLG), Ecuador. Segundo has assisted in a wide variety of research

conducted at RLG including studies of birds, bats, terrestrial and aquatic insects, amphibians, and

May 2020 | Volume 14 | Number 2 | e234

Amphib. Reptile Conserv.

Impacts of fish farming on glassfrogs in Ecuador

Timothy J. Krynak is a Natural Resources Manager for Cleveland Metroparks and a founding

board member and President of Las Gralarias Foundation. Tim’s research focuses on biodiversity

preservation and bat ecology. Tim works to protect the environment through his land management,

stewardship, and environmental education practices. As an adjunct professor at Grand Valley

State University (Allendale, Michigan, USA), Tim co-instructs the Tropical Biology and Field

Conservation Course in Ecuador with Eric Snyder.

Eric B. Snyder is a Professor of Biology at Grand Valley State University (GVSU, Allendale,

Michigan, USA). Eric’s research interests include stream ecology, restoration ecology, aquatic

entomology, lotic ecosystem metabolism, and trophic structure. Eric has been teaching a Tropical

Biology and Field Conservation course in Ecuador for GVSU since 2014, and has mentored two

graduate students, including Dana Wessels, in research conducted in Ecuador’s cloud forest streams.

Jane A. Lyons is the founder and land manager of Reserva Las Gralarias, Ecuador. Jane is currently

the majority owner of Mindo Bird Tours and serves as Vice-President of the Board of the Las

Gralarias Foundation, Inc. (North Olmstead, Ohio, USA). Jane was the first President of Jocotours

Cia Ltda., the tour agency of the Jocotoco Foundation (Quito, Ecuador), and also works as a

consultant for birding tourism and supervisor for graduate students and other field research projects.

Jane works daily to preserve and protect the great biodiversity of the Mindo region through her

efforts in eco-tourism, land acquisition, land management, and environmental education.

Juan M. Guayasamin is an Ecuadorean herpetologist and Professor of Evolution, Biology, and

Island Biogeography at Universidad San Francisco de Quito in Ecuador. Juan’s research interests

include taxonomy, ecology, evolution, and neotropical biodiversity conservation. Juan and his

students also investigate the impact of introduced species to Andean cloud forest streams in an

effort to protect these ecosystems.

10 May 2020 | Volume 14 | Number 2 | e234

Krynak et al.

Appendices

Appendix 1. Survey site locations in the Mindo region of Ecuador (Datum WGS 84). Site locations: RSR = Rio Santa Rosa, LC

= Lucy’s Creek, ST = Santa Teresita, FJ= Finca de Jaime, KC = Kathy’s Creek, M = Michelle’s, C = Chalguayacu Grande River,

5F = Five Frog Creek, LRSR = Lower Rio Santa Rosa, Berk = Ballux Creek, EP = El Paraiso del Pescador, LS = La Sierra, VC =

Verdacocha.

Coordinates (decimal degrees) Trout farm (y/n)

Lucy’s Creek 0.00518, 78.7383W N

Michelle’s 0.0215S, 78.7240W

Chalguayacu Grande River 0.02878, 78.7303W

| RSR

pC |

| ST

PKC | Kathy’sCreek_— | 0.01678, 78.7316W

PIG!

| SF

| _LRSR i

[ep | iParaso det Pescador | ___O012N,78.6727W

[is taSiera—————*YSC—i8 85, 78.607

[ve verdecocna ids. 78.6100

Five Frog Creek 0.0315S, 78.7052W

Appendix 2. Environmental characteristics of sites included in the glassfrog call survey conducted March-May, 2017 Mindo

region of Ecuador. Site abbreviations: VC = Verdacocha, LS = La Sierra, ST = Santa Teresita, FJ= Finca de Jaime, EP = El Paraiso

del Pescador, C = Chalguayacu Grande River, LC = Lucy’s Creek, Berk = Ballux Creek, 5F = Five Frog Creek, KC = Kathy’s Creek,

RSR = Rio Santa Rosa, M = Michelle’s, LRSR = Lower Rio Santa Rosa. All measurements were collected during daylight hours.

Stream Total Dissolved Canopy

discharge Conductivity dissolved oxygen openness __ Elevation Temperature

Site type Site code (m/sec) pH (uS/cm) solids (g/L) (mg/L) (%) (m) (°C)

VC 0.132 ERE i: 55 0.047 7415 55.51 2,666 12.255

LS 0.595 7.765 71.5 0.0595 7.245 28.86 2,483 13.585

make ST 13352 8.125 128.5 0.1055 8.245 33.93 2,186 14.39

FJ 1.42] 8.08 131.5 0.107 7.58 60.21 2,160 14.31

EP 281.25 7.74 78 0.0605 7.785 43.29 1,593. 16.595

C 0.004 7.84 25 0.02 7.61 20.54 2015 16

LC 0.115 7.69 35 0.027 FST 0.26 1,814 15.66

Berk — 8.61 43 0.036 741 22.88 2,254 15.16

Non-trout 5F 0.22 7.7 42 0.034 7.48 32.5 2,167 14.86

farm KC 0.23 FAS 28 0.02 6.9 1.04 2,053 15.8

RSR 1.131 E61 43 0.0034 LST 15.08 1,811 15.97

M 1.339 8.06 4] 0.032 7.91 41.08 2,031 15.43

LRSR 4.003 7.63 43 0.034 P92 28.34 1,693 15.83

Amphib. Reptile Conserv. 11 May 2020 | Volume 14 | Number 2 | e234

Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 12—23 (e235).

Comparison of in vitro methods to inhibit growth of a virulent

strain of Batrachochytrium dendrobatidis (Longcore, Pessier,

and Nichols 1999)

1.2Mlarina E. De Leon

'Department of Microbiology and Molecular Genetics, University of California, 451 Health Science Drive, Davis, California 95616 USA

?Department of Biology, California State Polytechnic University, Pomona, 3801 West Temple Avenue, Pomona, California 91768 USA

Abstract.—A highly virulent genotype of Batrachochytrium dendrobatidis (Bd), the Global Panzootic Lineage

(Bd-GPL), is implicated as a major cause of global amphibian population declines. Bd-GPL strain JEL274 was

Originally isolated from the skin of Western Toads (Anaxyrus boreas) in Colorado, USA, where populations are

in decline due to chytridiomycosis, the disease caused by Bd. Here, the difficulty in attempting to control Bd-

JEL274 is shown by challenging the genotype against three methods of inhibition: exposure to the antifungal

drug amphotericin B, recombinant E. coli-violacein, and A. boreas skin-associated bacterial isolates. The

Minimum Inhibitory Concentration (MIC) value of amphotericin B on Bd-JEL274 was 10-fold higher than in

previously tested strains, suggesting that Bd-JEL274 is remarkably drug-resistant. Violacein, an antifungal

secondary metabolite naturally expressed by some proteobacteria, has been shown to inhibit growth of the

fungus. In this study, the difference in fungal inhibition between recombinant E. coli-violacein and natural

antifungal activities of bacteria isolated from captive A. boreas skin was demonstrated using in vitro Ba-

bacterium inhibition assays. The abundant skin-associated bacterium Chryseobacterium indologenes inhibited

Bd-JEL274 significantly better than recombinant E. coli-violacein and this bacterium may have been involved

in the natural clearing of Bd infections in the toads. Larger studies should focus on using the amphibian skin

microbiome for probiotic treatment of chytridiomycosis in A. boreas toads, rather than risking lowered fitness

or increased mortality from drug treatments.

Keywords. Amphibian conservation, Anaxyrus boreas, Anura, chytridiomycosis, microbial ecology, transformation,

violacein, wildlife disease

Citation: De Leén ME. 2020. Comparison of in vitro methods to inhibit growth of a virulent strain of Batrachochytrium dendrobatidis (Longcore,

Pessier, and Nichols 1999). Amphibian & Reptile Conservation 14(2) [General Section]: 12-23 (e235).

International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any me-

dium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced, are as

follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 14 May 2019; Accepted: 16 March 2020; Published: 1 May 2020

Introduction

Loss of biodiversity is one of the most problematic

issues facing humans today, and amphibians are the

most threatened taxonomic class (Howard and Bickford

2014). The cascading implications of losing the majority

of members of a major vertebrate clade are likely to

have effects that disrupt entire ecosystems (Briggs et al.

2005). Amphibian populations around the world have

dramatically declined in the last four decades, with over

3% of amphibian species already extinct (Alroy 2015).

Current projections estimate that more than 7% of all

amphibians will become extinct within the next century

(Alroy 2015). One of the greatest threats to amphibian

populations is infection by the parasitic fungus

Batrachochytrium dendrobatidis (Bd). This pathogen

is easily transmitted between hosts because the aquatic

zoospores are free-living, flagellated, and substrate

independent. An alarming number of amphibian species

are known to be susceptible to infection by Bd (Olson et

al. 2013; Scheele et al. 2019). Bd can cause extremely

damaging effects on wild amphibian populations and

is thought to have been the proximate driver of many

amphibian species extinctions throughout the world since

the 1970s (Kriger and Hero 2009; Scheele et al. 2019).

Most amphibian population declines that are

sufficiently documented are specifically attributed to

the Global Panzootic Lineage of Bd (Bd-GPL) [Farrer

et al. 2011; James et al. 2015]. Batrachochytrium

dendrobatidis is hypothesized to have originated in Asia

and disseminated around the world due to amphibian

trade, including the use of frogs for pets, food, and

medical testing (O’ Hanlon et al. 2018), and it is therefore

considered an invasive species. Of six known Bd lineages,

Correspondence. msdeleon@ucdavis.edu (ORCID 0000-0001-9973-1951)

Amphib. Reptile Conserv.

May 2020 | Volume 14 | Number 2 | e235

De Leon

Bd-GPL is a hypervirulent genotype that represents the

majority of genotypes isolated globally, which may be

outcompeting endemic lineages (Schloegel et al. 2012;

James et al. 2015). Even within certain Bd lineages

such as GPL, there is significant variation in virulence

between the genotypes (Berger et al. 2005; Fisher et al.

2009; Farrer et al. 2011). Bd-JEL274 is responsible for

Western Toad (Anaxyrus boreas) declines in the United

States, therefore this study utilizes Bd-GPL genotype

JEL274 to evaluate the antifungal effects of A. boreas

skin-associated bacteria as well as the drug susceptibility

of this Bd-GPL strain (Muths et al. 2003; Scherer et al.

2005; Pilliod et al. 2010).

Although extensive A. boreas declines have occurred

due to Bd infection, remaining populations that survived

the emergent disease may consist of individuals which

have some sort of resistance mechanism, such as

antifungal skin-associated bacteria or secretions of

bufadienolide chemicals (Park et al. 2014; Barnhart et al.

2017). These studies indicate that the individual survival

probability of each toad seems to be strongly dependent

on the composition of skin microbiota, host stress level,

and virulence level of the Bd genotype. Therefore,

understanding the microbiomes of A. boreas toads, as well

as the drug resistance level of the Bd genotype associated

with this species, could lead to treatment strategies that

may prevent additional Bd-related die-offs.

In areas of high Bd prevalence, surviving populations

of amphibians tend to have higher numbers of antifungal

bacteria in their skin microbiomes (Harris et al. 2009).

Thus, one promising treatment involves augmenting an

amphibian’s skin bacteria to fight pathogens (Harris et al.

2009). Bd is susceptible to the cutaneous, Gram-negative

bacterium Janthinobacterium lividum, one of the native

microbes that can be cultured from skin swabs of some

amphibians (Harris et al. 2009). Janthinobacterium

lividum secretes protective secondary metabolites

violacein and indole-3-carboxaldehyde, which have been

shown to inhibit pathogen growth (Brucker et al. 2008).

Current evidence suggests that epithelial cells of

some amphibians are incapable of supporting J. lividum

or other non-native bacterial strains, therefore simply

inoculating Bd-susceptible amphibians with foreign

bacteria may not be a plausible approach to treating or

preventing chytridiomycosis (Becker et al. 2011). For

example, amphibian skin peptides and alkaloids may be

toxic to invading bacteria if the host and foreign microbe

did not co-evolve. Bacteria that are associated with frogs

which secrete toxins from their skin can tolerate these

toxins, while artificially introduced microbes may not

(Becker et al. 2011).

Although a number of experimental treatments for

chytridiomycosis have been tried, viable treatment

options for Bd remain debatable (Woodhams et al. 2012).

Experimental solutions include the use of antifungal

pharmaceutical drugs, which have their merits, but can

also be toxic and have been known to kill or retard the

Amphib. Reptile Conserv.

growth of treated frogs (Martel et al. 2011). This study

included a minimum inhibitory concentration (MIC)

assay that shows the in vitro effects of the antifungal drug

amphotericin B on Bd-JEL274. Violacein-producing

microbes were not found from swab cultures in this study

therefore, the heterologous host £. coli-violacein was

used to understand the chemical’s potency in comparison

to native skin microbes in Bd inhibition assays. In order to

better understand the basis for bacterial treatment options

for chytridiomycosis, this pilot study also compared

inhibitory capabilities between A. boreas skin-associated

microbes. The results of this study can assist herpetologists

in understanding how the relationships between an

amphibian’s own native skin bacteria and Bd zoospore

levels may be important aspects of chytridiomycosis

management in captive or wild populations, which can

help guide conservation strategies that involve treatment

of Bd-infected amphibians.

Materials and Methods

Sample Collection

During this study, the California State Polytechnic

University, Pomona herpetological vivarium housed

three A. boreas toads. This species was chosen based on

its availability for repeated swabs and its conservation

status. According to the IUCN, this species is declining

in population, but it is not federally listed as Endangered.

Toads were collected from Menifee, California

(33°44’°07.75"N, 117°11°59.17"W, near McLaughlin

Road). One male (#2) was collected in April 2011, and

one male (#1) and one female (#3) were collected in

May 2014 and housed with the original male. Toads were

swabbed for bacterial collection according to Brucker

et al. (2008). Each toad was handled using clean latex

gloves and thoroughly rinsed with a minimum of 10

mL sterile water to remove transient bacteria and soil

contaminants. Using DNA-free rayon swabs (Medical

Wire and Equipment Co., #MW 100-100), toads were

swabbed five times at each of the following locations for

a total of 25 strokes: ventral surface from mid-abdomen

to cloaca; each inner thigh; and one stroke on the ventral

side of webbing between each hind leg toe. Two swabs

were used per individual toad, one for bacterial culturing

and the other for Bd detection. For the bacterial collection

swabs, the Copan Innovation™ swabbing system was

used, in which the swabs were soaked in sterile PBS

and the samples were shaken prior to spreading 50 uL of

solution onto agar plates using a sterile spreader.

Bacterial Culture and Isolation

Swabs were thoroughly streaked onto 1% tryptone agar

and Tryptic Soy Agar (TSA) plates to allow a variety

of bacteria to grow in different nutrient conditions.

Plates were incubated at 25 °C and colony growth

May 2020 | Volume 14 | Number 2 | e235

Inhibition of Batrachochytrium dendrobatidis growth

was characterized after 72 h, when individual colonies

were picked from culture plates using a sterile loop and

streaked for isolation onto new agar plates. Only the most

visually unique and abundant colonies were sub-cultured

and isolated. Visually unique colonies included those

of differing colors, sizes, and textures. Each bacterial

isolate was Gram stained and evaluated microscopically

(Coico 2006). Bacterial taxa were then identified using

16S rRNA gene sequencing analysis.

Genomic DNA Extraction and 16S

Sequencing

rRNA Gene

Genomic DNA was extracted from a total of 18

bacterial isolates using the Invitrogen Easy-DNA Kit

(cat No. K1800-01, Life Technologies™, Carlsbad,

California, USA) for genomic DNA isolation, according

to the manufacturer’s instructions. Extracted DNA was

amplified using the eubacterial 16S rRNA gene primer

set, B27f (5'-AGAGTTTGATCMTGGCTCAG-3')

and B1492r (5'-ACCTTGTTACGACTT-3) [Eurofins

Microbiology, Garden Grove, California, USA]. PCR

parameters were 5 min at 94 °C; 35 cycles of 30 sec at

94 °C, 30 sec at 55 °C, and 2 min at 72 °C; followed

by 2 min at 72 °C. PCR product was purified using

QIAquick PCR purification kit (cat. Nos. 28104 and

28106, Qiagen, Hilden, Germany) and checked for DNA

quality and quantity using an Implen NanoPhotometer®

(Westlake Village, California, USA). Aliquots of

samples were analyzed by gel electrophoresis to ensure

proper amplification of the 16S rRNA gene region. The

amplified products were sent for Sanger sequencing

to Source BioScience (Santa Fe Springs, California,

USA). Forward and reverse sequence reads were then

concatenated using SeqTrace. Identification of the 16S

gene in FASTA format was performed using NCBI

BLASTn based on sequence similarity to GenBank

sequence database entries (Altschul et al. 1990).

Growth and Maintenance of Bd

Bd-JEL274, originally isolated from A. boreas toads in

Clear Creek Co., Colorado (1999), was obtained from

Joyce Longcore of the Maine Chytrid laboratories,

University of Maine. Bd was maintained in Tryptone

Glucose hydroLysate (TGhL) broth or 1% (w/v) tryptone

broth for seven days until active zoospores were visible

under a dissecting microscope, and then sub-cultured

periodically to maintain fresh cultures. Bd grew best in

TGhL broth with incubation at 23—24 °C. Cultures were

grown on agar plates and/or in 25-cm? Corning cell

culture treated flasks and incubated for 5—10 days. Growth

progress was viewed under a dissecting microscope. Once

maximum zoospore production was observed, plates and

flask tops were wrapped in Parafilm®, and stored at 4

°C for up to two months. Zoospores were harvested by

scraping the flask walls before aspirating liquid from the

Amphib. Reptile Conserv.

flasks, and by flooding plates with 3 mL of 1% tryptone

broth before transferring the zoospores and sporangia to

new media.

Production of Zoospores

Bd was grown in broth until clumps of sporangia

were visible to the unaided eye. A sterile serological

pipette was used to add 0.75 mL of this broth culture

to tryptone agar in 9-cm culture dishes. Inoculated

dishes were left open in a laminar flow hood until the

added broth was dry. Covers were replaced on dishes,

which were then incubated at 23—24 °C. After 7—10

days, active zoospores could be observed around the

periphery of the fungal colonies by inverting the dishes

on the stage of a dissecting microscope viewed at 40x.

Zoospore concentration was measured by counting

spores using a hemocytometer. To harvest zoospores,

plates were flooded with 3 mL 1% tryptone broth and

after ~10 min, zoospores in liquid were collected by

pipetting. Glassware was bleached before washing and

sterilization. All materials that contained or came into

contact with the pathogenic fungus were autoclaved

before disposal.

Determination of Minimum Inhibitory Concentra-

tion of Amphotericin B for Bd-JEL274

The MIC of amphotericin B (X-Gen Pharmaceuticals,

Inc., Big Flats, New York, USA) for Bd-JEL274 was

determined using a macrodilution method in 24-well

plates. The final assay concentrations of the drug were

3.2, 1.6, 0.8, 0.4, 0.2, and 0.1 ug/mL. To each well, 200

ul of TGhL culture broth containing one of the serial

dilutions of amphotericin B was added to 200 ul of a

five-day-old growing Bd culture, containing a mixture

of approximately 1 x 10° Bd sporangia and zoospores.

Cells were counted using a hemocytometer, then diluted

in TGhL media until a standard 5 x 10° concentration

of zoospores and sporangia was obtained. The MIC

value was determined as the lowest concentration of

amphotericin B at which no growth of Bd was recorded.

Growth was assessed after 5, 7, and 10 days of incubation

at 24 °C using stereoscopic and compound microscopic

examination of the wells. To obtain final Bd cell counts,

10 uL of spores from the appropriate wells were placed

in a hemocytometer and five out of 25 grids were counted

within the larger grid. The concentration of zoospores

was calculated by multiplying the number of cells by five,

and again by 10,000 to obtain cells/mL. This experiment

was carried out in triplicate.

Recombinant E. coli-Violacein

Plasmid cloning vectors containing the violacein gene

operon copied from Chromobactrium violaceum were

obtained from Derek Sarovich (Sarovich and Pemberton

May 2020 | Volume 14 | Number 2 | e235

De Leon

2007). The pPSX-viot+ vector (pPSX-violacein)

produces moderate amounts of violacein. pJP1000

(pUC18-violacein), a pUC18 derivative, is multi-copy

and produces high levels of violacein. The pPSX-vio+

(pPSX-violacein-opvl) over-produces violacein. The

relative degrees of violacein production were indicated

by color intensity of the cultures (light to dark violet).

The pPSX-vio+, pPSX-vio++, and pJP1000 vectors

were transformed into Escherichia coli NEBS-alpha

(New England BioLabs) using the manufacturer’s

instructions. All plasmids were stable when cloned into

E. coli, even under no selection pressure. Visual evidence

of violacein production by E. coli was observed when

cultures showed the deep violet color characteristic

of violacein after transformation and throughout the

passage of generations.

Bd Inhibition Assays

Inhibition assays followed Harris et al. (2006). Bd was

grown on 1% tryptone, 1% agar plates for five days, or

until maximum zoospore production was observed (up to

seven days). Three mL of sterile water was added to each

plate, and plates were rocked back and forth to loosen the

zoospores from the agar. Plates were tilted so the liquid

pooled on the side, and 0.75 mL was pipetted and spread

onto four new tryptone plates using a sterile spreader.

Each Bd plate was used to make four new plates. Plates

were left to dry in a laminar flow hood for 45—60 min.

After the Bd solution had soaked into the fresh agar plate,

using a sterile loop one colony from each bacterial isolate

(24-h agar cultures) was streaked across the center of the

agar plate. Plates were incubated inverted at 24 °C for

5-10 days.

Scoring: Measurement of Zone of Inhibition

Differences between bacterial isolates and between

treatments for an isolate in inhibition experiments were

tested after 7-10 days growth on agar medium. The

bacterial isolates were tested in triplicate and scored

as inhibitory against a lawn of Bd. Six points around

the bacterial streak reaching to the edge of the zone of

inhibition (ZOI) were measured using digital calipers.

The average zone of inhibition (the distance between the

edge of the bacterial streak where no Bd growth occurred

and the edge of the Bd lawn) and standard deviations

were calculated for all bacteria with inhibitory properties.

Bacterial isolates were scored as “not inhibitory” if no

zone of inhibition developed and the plates were covered

with active zoospores after 7-10 days of incubation (ZOI

= 0). Bacteria were considered “weakly inhibitory” if

a clear ZOI < 10 mm developed between the bacterial

streak and the Bd culture. Bacterial isolates with a ZOI

> 10 mm were considered as having strong antifungal

properties. When a bacterial streak overtook the whole

plate, the assay result was considered indeterminate. If

Amphib. Reptile Conserv.

an indeterminate result was obtained, the experiment

was repeated two more times before being recorded as

indeterminate. The negative control plates containing Bd

showed complete lawns of Bd (Harris et al. 2006; Park

et al. 2014).

Statistical Analysis

A non-parametric Kruskal-Wallis test followed by

Dunn’s multiple comparisons tests were performed using

GraphPad Prism version 6.00 for Windows (GraphPad

Software, La Jolla, California, USA, http://www.

graphpad.com). The Kruskal-Wallis rank test was used

to test for differences in fungal inhibition by bacterial

isolates. Differences were considered significant if P <

0.05. A box plot for the amphotericin B Bd MIC assays

was created in Rstudio (Version 1.0.136, RStudio, Inc.)

to show differential growth between the positive control

and the experimental groups.

Bd Detection in Toads

Toads were swabbed twice for the presence of Bd, nine

months apart. Swabs were taken according to Kriger et

al. (2006), using a sterile DNA-free fine-tipped rayon

swab (Medical Wire and Equipment Co., #MW _ 100-

100). Bd swabs were placed in microcentrifuge tubes

containing 50 uL 70% ethanol and frozen at -20 °C

until they could be brought to the Vredenburg lab at San

Francisco State University (San Francisco, California,

USA) for DNA analysis. Latex gloves were used during

any interactions with housing, tubes, or frogs, and

changed as necessary. DNA was extracted using Prepman

Ultra (Life Technologies, Carlsbad, California, USA) in

accordance with Cheng et al. (2011). DNA extracts were

amplified following a standard, probe-based quantitative

Polymerase Chain Reaction (qPCR) protocol using the

standard Bd primer set ITS1-3-CHYTR and 5.8S-CHYTR

(Boyle et al. 2004; Hyatt et al. 2007).

Results

Amphotericin B: Bd-JEL274 Minimum Inhibitory

Concentration (MIC)

MIC assays of the antifungal drug amphotericin B were

conducted in the range of 0.1—3.2 ug/mL and indicated the

MIC for Bd-JEL274 as 1.6 ug/mL. Little to no difference

in growth was observed in wells with drug concentrations

from 0.1-0.8 pg/mL, however, noticeable inhibition

occurred at 1.6 ug/mL, where there were visibly fewer

viable cells than in the positive control. The fungicidal

effect of amphotericin B was obvious at 3.2 ug/mL, as

the vast majority of cells were dead. The MIC value was

determined as the lowest concentration of amphotericin

B at which no growth of the Bd strain was recorded. A

boxplot of the triplicate results was created to compare

May 2020 | Volume 14 | Number 2 | e235

Inhibition of Batrachochytrium dendrobatidis growth

Bd cell count (x 10°)

0.0 1.6 3.2

Amphotericin B MIC (ug/mL)

Fig. 1. The Kruskal-Wallis rank sum test for the boxplot returned

a chi-squared value = 2, degrees of freedom (df) = 2, and P =

0.3679. The Wilcoxon rank sum test for the positive control and

the highest concentration of amphotericin B returned W =9, P=

0.1. These analyses show that the alternative hypothesis is true

and assumes that there is information in the magnitudes and

signs of the differences between paired observations, because

the mean (location shift) is not equal to 0. Observed differences

in mean cell count between the three groups (0.0, 1.6, and 3.2

ug/mL) are not statistically significant (P > 0.05).

the positive control (0.0 ug/mL amphotericin B) with

concentrations of amphotericin B for which Bd inhibition

was detected by microscopy, namely 1.6 ug/mL and 3.2

ug/mL (Fig. 1). For comparison, the result of the MIC of

amphotericin B on Bd-JEL274 was 10-fold higher than

previously reported for five other Bd strains (Martel et al.

2011). Although mean differences between cell counts in

the experimental groups versus the control group are not

statistically significant, this 10-fold increase in drug dose

relative to that reported in previous studies 1s likely to be

lethal to amphibians, if tested in vitro. Therefore, testing

the amphotericin B MIC for Bd-JEL274 on Bd-infected

animals is not recommended.

Bacterial Preparation and Identification

Bacterial isolates from captive A. boreas skin swabs were

cultured using either 1% tryptone (Trp) agar or Tryptic

Soy Agar (TSA) media (Suppl. Fig. 1). Using BLASTn

Basic Local Alignment Search Tool for nucleotides

(Altschul et al. 1990), B27f and B1492r 16S rRNA gene

sequence reads in FASTA format were queried in the

NCBI sequence database to obtain the closest relative

matches. The identity (Ident %) is recorded as the extent

to which two nucleotide sequences have the same residues

at the same positions in an alignment within the database,

expressed as a percentage. Sixteen bacterial strains were

isolated and identified from swabs. Isolates were Gram

stained and only potential proteobacteria (Gram negative

rods) were identified genetically. Bacterial identities

determined by 16S rRNA gene sequencing and NCBI

BLASTn included members from six distinct classes:

Amphib. Reptile Conserv.

Flavobacteria (3), Gammaproteobacteria (1), Bacilli

(4), Actinobacteria (4), Alphaproteobacteria (1), and

Betaproteobacteria (2) [Table 1]. Two visually distinct

isolates could not be identified by 16S region and

subsequent NCBI BLASTn searches. Some genera were

isolated from colonies which looked unique and identified

more than once using 16S gene results and BLASTn.

Duplicate isolates of genera include Microbacterium

and Paenibacillus, while Chryseobacterium was isolated

three separate times.

No violacein-producing bacteria were isolated from

the skin swabs, consequently, the inhibitory action of

natural violacein producers could not be evaluated using

a Bd inhibition assay. Instead, violacein production was

introduced using recombinant £. coli strains for the

purpose of comparing the fungal inhibition capacity of

violacein to the bacteria isolated from toad skin swabs.

The E. coli NEBS5-alpha strains included the negative

control (unmodified E. coli), E. coli-vio+, E. coli-vio++,

and E. coli-JP1000. Recombinant FE. coli expressed the

characteristic deep violet color of violacein (Suppl. Fig.

2).

Bd Inhibition Assays

To determine if any of the isolates exhibited natural

antifungal activity against Bd, the bacterial isolates and

recombinant E£. coli-violacein were challenged against

Bd-JEL274 and the size of the ZOI around the bacterial

streak was measured (Table 2). The violacein-producing

E. coli strains did not produce significantly larger ZOI

than the E. coli control. Of the 15 bacterial strains (11

skin isolates and four £. co/i strains) challenged against

Bd, 12 had a measurable ZOI surrounding the bacterial

streak and the Bd clearing zones were well-delineated

and clearly visible to the unaided eye (Fig. 2). According

to the Kruskal-Wallis test, strong fungal inhibition

coincided with bacteria that produced significantly larger

ZOI than that of the violacein over-producer E. coli-vio+

(p < 0.05). A ZOI of 0.0 mm was associated with no

inhibition, ZOI < 10 mm with weak inhibition, and ZOI

> 10mm with strong inhibition. Lysinibacillus fusiformis

swarmed over the agar, overtaking the entire plate, so a

ZOI could not be determined, and therefore L. fusiformis

was considered indeterminate (No Data/ND).

The means and standard deviations of the ZOI for all

trials were calculated for each bacterium tested (Table

2). Among them, C. indologenes, exhibited the largest

ZOI of 11.0 mm, more than twice the size of the ZOI

exhibited by the recombinant violacein over-producer E.

coli-vio+ (4.2 mm). Bacillus sp. also exhibited inhibitory

action with an average ZOI of 7.6 mm. Four bacterial

isolates; two Microbacterium isolates, Rhodococcus

sp., and Brevundimonas sp., did not show any distinct

zone of inhibition, as zoospores were distributed directly

adjacent to the bacterial streaks (Fig. 3).

A non-parametric Kruskal-Wallis test followed by

May 2020 | Volume 14 | Number 2 | e235

De Leon

Fillies a ! —

acillus sp. (right bacterial streak),

Fig. 2. (A) and (B) Duplicate Bd negative controls, (C) E. coli-JP1000 (left bacterial streak) and B

(D) E. coli-JP1000 displaying a clear ZOI surrounding the bacterial streak, (E) FE. co/i control.

Table 1. Bacterial isolate identification 16S rRNA gene sequences. *Unclassified Micrococcacae.

LD. Toad # Medium Gram stain iE tt Class Genus/species

2xMD 1 1% Trp - rod 97 Flavobacteria Chryseobacterium

indologenes

4xMD 1 1% Trp - rod 98 Gammaproteobacteria Klebsiella oxytoca

5xMD 3 1% Trp + rod 99 Bacilli Paenibacillus sp.

7xMD 3 1% Trp + rod 97 Bacilli Paenibacillus pabuli

8xMD zy) 1% Trp + rod 95 Actinobacteria Microbacterium

petrolearium

Al 1 TSA + cocci 97 Actinobacteria Micrococcaceae

bacterium*

A2L 1 TSA + rod 98 Actinobacteria Rhodococcus equi

A3 1 TSA + rod 99 Actinobacteria Microbacterium sp.

A5 1 TSA - rod 99 Alphaproteobacteria Brevundimonas sp.

A6(Y) 1 TSA - rod 83 Betaproteobacteria Comamonas sp.

A6P 1 TSA - rod 83 Betaproteobacteria Acidovorax ebreus

A7 1 TSA - rod 92 Betaproteobacteria Ralstonia sp.

A9 1 TSA - rod 98 Flavobacteria Chryseobacterium tenax

Bl 3 TSA + rod 99 Bacilli Bacillus sp.

B3 3 TSA + rod 99 Bacilli Lysinibacillus fusiformis

B5 3 TSA - rod 97 Flavobacteria Chryseobacterium tenax

A2D 1 TSA + rod - Unidentified

A8 1 TSA - rod - Unidentified

Amphib. Reptile Conserv. 17 May 2020 | Volume 14 | Number 2 | e235

Inhibition of Batrachochytrium dendrobatidis growth

Fig. 3. Microscopic images (40x) of Bd challenge assays. (A) Microbacterium, (B) Micrococcaceae, and (C) recombinant E. coli-

vio+. Vertical arrows point to the Bd lawn and horizontal arrows point to the left side of the bacterial streak. Absence of Bd lawn

within field of view indicates the inhibitory effect from the bacterial streak. Approximate diameter of field of view ~0.5 mm.

Dunn’s multiple comparisons tests were performed to test

the significance of the size of the ZOI (Table 3). When

comparing the relative inhibition of the negative control

(E. coli) to the toad bacterial isolates and the genetically

modified E. coli-violacein strains, only C. indologenes

and Bacillus sp. (this strain is closely related to Bacillus

cereus) showed significantly larger ZOI than E. coli (P <

0.0001). Among pairwise comparisons between isolates,

C. indologenes inhibited Bd significantly more than the

violacein overproducer E. coli-viot. Although Bacillus

sp. also had average ZOI greater than E. coli-viot+ there

was no significant difference between their inhibitory

effects, nor was there was a significant difference

between average ZOI of C. indologenes and Bacillus sp.

Bd Detection by qPCR

Toads #1 and #3 (see Materials and Methods) initially

tested positive for Bd, while toad #2 tested negative.

The zoospore equivalent (ZE) scores indicated low level

infections (ZE scores of 37.68 and 4.96, respectively).

Nine months later, the infected toads were swabbed

again and qPCR analysis revealed a complete clearance

of infection in toad #1, and substantial reduction of

zoospore load in toad #3 (ZE scores of 0.0 and 0.0808,

respectively). Toad #2 was euthanized due to extensive

peripheral edema before the second Bd swab was taken.

Discussion

To gain a better understanding of chytridiomycosis

treatment options, three in vitro Bd inhibition methods

were tested: the antifungal drug amphotericin B for which

the MIC value was determined, Bd inhibition assays

using skin-associated microbes, and inhibition assays

using the antifungal compound violacein in recombinant

E. coli. The results of this study favor of augmentation of

the natural skin microbiota as the most feasible, efficient,

Amphib. Reptile Conserv.

and biologically relevant approach for chytridiomycosis

treatment.

Amphotericin B works by binding directly to

ergosterol (a chemical in fungi which is functionally

equivalent to cholesterol in animals), which disrupts

fungal cell membrane permeability and causes leakage.

This drug has a stronger binding affinity for ergosterol

than for cholesterol, but still binds to cholesterol, making

it potentially lethal for host cells as well as fungal cells.

Amphotericin B has low selectivity and can potentially

cause nephrotoxicity in vertebrate hosts (Odds et al.

2003; Martel et al. 2011). However, this drug has been

Table 2. Zone of inhibition (ZOI) for Bd inhibition assays.

Bacterial strain Avg. ZOI (mm) S.D.

C. indologenes 11.00 4.72

Bacillus sp. 7.60 326

E. coli-viot 421 4.55

E. coli-JP1000 255 2.87

Comamonas sp. Daly 2:95

Micrococcacae bacterium 1.87 2735

E. coli-viot++ 1.16 1.08

A. ebreus 0.82 137,

Ralstonia sp. 0.77 1256

E. coli 0.46 Pill

K. oxytoca 0.21 0.41

Brevundimonas sp. 0.0 0.0

Microbacterium sp. 0.0 0.0

M. petrolearium 0.0 0.0

R. equi 0.0 0.0

L. fusiformis ND ND

May 2020 | Volume 14 | Number 2 | e235

De Leon

Table 3. Dunn’s multiple comparison test for zones of inhibition (ZOI). Symbols and definitions: ns (not significant) = P > 0.05, *

=P 10-052 8 SP a OO seer = P1000] 8 P0001

Bacterium ZOI (mm) + SD E. coli E.cvt+ = Eic.JP E.c.v+ B.sp. Ci.

C. indologenes (C.i.) 11.00 (+ 4.72) bo ey bi i sh ns

Bacillus sp. (B.sp.) 7.60 (+ 3.26) Aci 3 ns ns

E. coli-viot (E.c.v+) 4.21 (44.55) ns ns ns

E. coli-JP1000 (E.c.JP) 255.45 287) ns ns

E. coli-viot+ (E.c.v++) 1.16 (+ 1.08) ns

E. coli 0.456 (+ 1.11)

found to cause fewer grossly observable side effects than

itraconazole, such as reduced growth, which can diminish

host fitness. Carey et al. (2006) showed that Bd zoospore

load caused chytridiomycosis-induced death in A. boreas

toadlets with exposure to Bd-GPL genotype JEL275, and

found that exposure to 1 x 10° zoospores for three days

was sufficient to cause mortality in the toads. Each MIC

assay in this study used a standard starting concentration

of 1 x 10° zoospores, a 10-fold lower concentration than

the Carey study, and this MIC value would be too high

to be safely applied to in vivo animal studies. The MIC

assay included in this study showed that Bd-JEL274 has

a high tolerance for amphotericin B, inhibiting Bd growth

at a dosage of 1.6 ug/mL, and killing Bd at 3.2 ug/mL.

Previous amphotericin B MIC assays using different Bd

strains have shown lower effective MIC levels (Martel

et al. 2011). High resistance to amphotericin B by Bad-

JEL274, in conjunction with previously established

evidence of toxicity of amphotericin B to amphibians,

suggests that this drug may not be a practical treatment

option for in situ treatment of chytridiomycosis in A.

boreas toads, as the dosage needed for significant effects

could cause negative side effects such as nephrotoxicity,

reduced growth, death, or other unforeseen outcomes.

This study included a modest survey of the skin-

associated bacteria from only three toads, and this

approach corroborated the positive findings from

previous amphibian microbiome studies. Although this

study was narrow in terms of subjects, many of the

same genera of skin-associated bacteria from these few

captive toads are those also found throughout wild North

American amphibian skin microflora (Roth et al. 2013).

For example, the assemblage of microbes found was

similar to that of wild-caught Colorado A. boreas toads,

even though this pilot study used captive animals (Park

et al. 2014). Previous studies on A. boreas have found

bacteria closely related to those that were isolated from

the captive toads here, such as Bacillus, Lysinibacillus,

Rhodococcus, and Chryseobacterium, indicating that

established skin-associated bacteria may persist through

environmental changes (McKenzie et al. 2012). Of the

bacterial isolates screened for potential anti-Bd activity

in vitro, one of the 11 toad isolates (C. indologenes)

inhibited Bd significantly better than the recombinant

violacein over-producer, EF. coli-viot+, suggesting that

Amphib. Reptile Conserv.

commonly occurring skin bacteria may be more efficient

for use in chytridiomycosis treatment than the process

of seeking out native violacein-producing bacteria in the

amphibian population of interest. Two species of interest

in this study, C. indologenes and Bacillus sp. (B. cererus),

have previously been isolated from wild North American

amphibians and have shown anti-Bd activity (Roth et

al. 2013; Park et al. 2014). The antifungal properties

of these bacteria provide insight into the ongoing

investigations of using natural skin-associated bacteria

as a bioaugmentation treatment for chytridiomycosis.

Chryseobacterium indologenes is a Gram-negative,

lactose non-fermenting, oxidase-positive, rod-shaped

bacillus with a distinct yellow to orange pigment. This

bacterium is found ubiquitously in soil, on plants, and

in water sources (Wauters et al. 2015). The antifungal

properties of C. indologenes have not been studied;

however, a closely related species, Chryseobacterium

aquaticum, is known to secrete proteases and chitinases,

which break down chitinous fungal cell walls (Gandhi et

al. 2009). Bacillus cereus is known to protect agricultural

crops from fungal infections, and it naturally secretes the

antibiotics zwittermicin A and kanosamine which inhibit

the growth of fungal plant pathogens. The antifungal

compounds of B. cereus strains have been developed

as useful biological control agents in the suppression

of fungi and crop diseases (Silo-Suh et al. 1994). The

synergistic application of C. indologenes and B. cereus

in the biocontrol of Bd should be further explored as a

chytridiomycosis treatment.

This study demonstrates that antifungal micro-

symbionts can be found even within a small subset of

bacteria cultured from amphibian skin, and that large-

scale skin microbiota studies may not be necessary for

finding antifungal bacteria that could potentially be used

for disease treatment.

Chryseobacterium indologenes, and possibly Bacillus

Sp., may have assisted the hosts in clearing Bd infection.

The three toads used in this study were captured from

the wild in 2011 (#2), and 2014 (#1 and #3), and initially

had contact with other A. boreas toads in the vivarium,

preventing a clear determination of whether the Bd

infections were contracted in the wild, or after the toads

were brought into the vivarium. Despite Bd infection, the

toads did not exhibit grossly observable symptoms of

May 2020 | Volume 14 | Number 2 | e235

Inhibition of Batrachochytrium dendrobatidis growth

chytridiomycosis, such as red legs and ventral surface, or

skin sloughing. Two of the toads initially tested positive

for Bd (#1 and #3) and either naturally cleared (#1) or

reduced (#3) their infections. Having cleared the Bd

infection within nine months suggests that the toads were

infected recently at the vivarium and were not harboring

the infection in the wild. The fact that C. indologenes

was abundantly cultured from the skin swabs, and that

it had significantly better capacity to inhibit Bd in vitro

than all other isolates tested, circumstantially suggests

that C. indologenes may have facilitated the observed

clearing of Bd. One possible explanation for this is

that the microbes were providing protection against Bd

infection, but aside from this correlative evidence, other

factors (such as general good health of the individuals or

bufanolide production) could have also contributed to the

clearing of infection.

Before commencement and during the course of the

study, toad #2 had peripheral edema of the left front

leg. Toad #2 initially tested negative for Bd but was

euthanized due to signs of lethargy and a severely swollen

leg before the second swab was taken. As a known

opportunistic pathogen of amphibians, C. indologenes

can cause peripheral edema among other symptoms in

captive and wild animals, especially if the pathogen can

enter the body through a lesion (Olson et al. 1992; Mauel

et al. 2002). The effect of C. indologenes presence in the

enclosure was not measured, however it 1s conceivable

that there may have been an effect of inhibitory bacteria

on the health status of the toads. The peripheral edema

in one toad and the low level Bd infection in the other

two toads that naturally cleared over time may have been

correlated with the abundance of Chryseobacterium that

was cultured from the skin swabs. Such knowledge of

antifungal bacteria as shown in this study is encouraging

for the future of amphibian microbiome research, and for

the control of Bd without introducing pharmaceuticals to

the natural environments, which could negatively affect

non-target organisms.

Chryseobacterium indologenes, a common bactertum

that seems to be highly abundant in the skin microbiome

of North American amphibians, may have a better

potential for use in the treatment of chytridiomycosis

than the widely studied violacein-producing bacteria

such as J. lividum. Additional studies could involve in

vivo bioaugmentation assays using C. indologenes to

understand the mechanism of its capacity for reducing

zoospore load of infected amphibians and identify whether

this bacterium can be safely inoculated onto otherwise

healthy animals. It may be helpful to determine which

antifungal compound(s) Chyrseobacterium secretes,

and test whether these compounds can be isolated and

utilized to treat chytridiomycosis.

The use of single isolates in a sterile experimental

environment can be useful for establishing baseline

observational data, but microbes do not exist in isolation

in the natural world. Recombinant E. coli that expresses

violacein may not fully represent how native microbes

Amphib. Reptile Conserv.

deliver or utilize this compound. For example, violacein

is secreted by C. violaceum in extracellular membrane

vesicles to both solubilize the hydrophobic pigment

and transport the compound to other microorganisms

in aqueous solutions, a mechanism that is partially

responsible for its bactericidal effects (Choi et al.

2019). The secretion mechanism of violacein produced

synthetically by plasmid DNA in recombinant E. coli is

unknown and could impact violacein’s biocidal effects.

Native microbial communities may also shape

chytridiomycosis treatment strategies as amphibian skin

supports a complex microecosystem. Isolated bacteria

may not behave in the same way or excrete the same

chemical compounds that they would in community

settings. Antifungal bacteria have the potential to be

augmented and used as natural biocontrols for Bd

infection, however multi-species communities are

known to inhibit Bd growth more than monocultures

of constituent species, an important consideration for

designing probiotic treatments (Piovia-Scott et al. 2017).

Interbacterial complementarity and synergy are

important for healthy community function, and

multispecies interactions should be studied in depth before

performing in vivo trials of single isolate bioaugmentation.

For example, the presence of C. indologenes has been

shown to increase the survival of C. violaceum when

the two organisms are co-cultured. When grown in co-

culture with other sympatric bacteria, the production of

violacein by C. violaceum can be reduced or decolorized,

though violacein production is not reduced when grown

in co-culture with C. indologenes (Shiau and Lin 2011).

To achieve maximum &d inhibition, future studies should

include the selection of probiotic mixtures of bacteria that

each have antifungal properties and are complimentary

to each other’s growth and production of antifungal

compounds, perhaps a combination of C. indologenes

and J. lividum.

Acknowledgements.—This_ research was _ performed

at California State Polytechnic University, Pomona

(Pomona, California, USA). I thank my mentors Dr.

Wei-Jen Lin and Dr. Jill Adler-Moore for extending

themselves beyond reason to support this project, and

Jon Olson and Dr. A. Christopher Lappin for technical

support. Chatrathip Nsongkla, Liana Ab Samad, Kun

Ho Lee, and Danielle Valencia contributed invaluable

bench work skills. I also thank Dr. John Pemberton and

Dr. Derek Sarovich for the kind donation of the violacein

plasmids and for extensive advice. This research was

supported by the National Institutes of Health Research

Initiative for Scientific Enhancement (NIH MBRS

RISE) award number 5R25GM113748-02, the BioTiER

(Biological Training in Education and Research) scholars

program, and by the Cal Poly, Pomona MENTORES

program (Mentoring, Educating, Networking, and

Thematic Opportunities for Research in Engineering and

Science).

May 2020 | Volume 14 | Number 2 | e235

De Leon

Literature Cited

Alroy J. 2015. Current extinction rates of reptiles and

amphibians. Proceedings of the National Academy

of Sciences of the United States of America 112(42):

13,003-13,008.

Altschul SF, Gish W, Miller W, Meyers EW, Lipman DJ.

1990. Basic local alignment search tool. Journal of

Molecular Biology 215(3): 8.

Barnhart K, Forman ME, Umile TP, Kueneman J,

McKenzie V, Salinas I, Minbiole KPC, Woodhams

DC. 2017. Identification of bufadienolides from the

Boreal Toad, Anaxyrus boreas, active against a fungal

pathogen. Microbial Ecology 74(4): 990—1,000.

Becker MH, Harris RN, Minbiole KPC, Schwantes

CR, Rollins-Smith LA, Reinert LK, Brucker RM,

Domangue RJ, Gratwicke B. 2011. Towards a better

understanding of the use of probiotics for preventing

chytridiomycosis in Panamanian Golden Frogs.

EcoHealth 8(4): 501-506.

Berger L, Hyatt AD, Speare R, Longcore JE. 2005.

Life cycle stages of the amphibian chytrid

Batrachochytrium dendrobatidis. Diseases of Aquatic

Organisms 68(1): 51-63.

Boyle DG, Boyle DB, Olsen V, Morgan JT, Hyatt AD.

2004. Rapid quantitative detection of chytridiomycosis

(Batrachochytrium dendrobatidis) in amphibian

samples using real-time Taqman PCR assay. Diseases

of Aquatic Organisms 60(2): 141-148.

Briggs CJ, Vredenburg VT, Knapp RA, Rachowicz LJ.

2005. Investigating the population-level effects of

chytridiomycosis: an emerging infectious disease of

amphibians. Ecology 86(12): 3,149-3,159.

Brucker RM, Harris RN, Schwantes CR, Gallaher

TN, Flaherty DC, Lam BA, Minbiole KPC. 2008.

Amphibian chemical defense: antifungal metabolites

of the microsymbiont Janthinobacterium lividum

on the salamander Plethodon cinereus. Journal of

Chemical Ecology 34(11): 1,422-—1,429.

Carey C, Bruzgul JE, Livo LJ, Walling ML, Kuehl KA,

Dixon BF, Pessier AP, Alford RA, Rogers KB. 2006.

Experimental exposures of Boreal Toads (Bufo boreas)

to a pathogenic chytrid fungus (Batrachochytrium

dendrobatidis). EcoHealth 3(1): 5—21.

Choi SY, Lim S, Cho G, Kwon J, Mun W, Im H, Mitchell

RJ. 2019. Chromobacterium violaceum delivers

violacein, a hydrophobic antibiotic, to other microbes

in membrane vesicles. Environmental Microbiology

22(2): 705-713.

Coico R. 2006. Gram staining. Current Protocols in

Microbiology 1: A-3C.1—A-3C.2.

Farrer RA, Weinert LA, Bielby J, Garner TWJ, Balloux

F, Clare F, Bosch J, Cunningham AA, Weldon C,

du Preez LH, et al. 2011. Multiple emergences of

genetically diverse amphibian-infecting chytrids

include a_ globalized hypervirulent recombinant

lineage. Proceedings of the National Academy of

Amphib. Reptile Conserv.

Sciences of the United States of America 108(46):

18,732-1 8,736.

Fisher MC, Garner TWJ, Walker SF. 2009. Global

emergence of Batrachochytrium dendrobatidis and

amphibian chytridiomycosis in space, time, and host.

Annual Review of Microbiology 63(1): 291-310.

Gandhi PM, Badri Narayanan KB, _ Ravindra

Naik P, Sakthivel N. 2009. Characterization of

Chryseobacterium aquaticum strain PUPC1 producing

a novel antifungal protease from rice rhizosphere soil.

Journal of Microbiology and Biotechnology 19: 99-

107.

Harris RN, Brucker RM, Walke JB, Becker MH,

Schwantes CR, Flaherty DC, Lam BA, Woodhams

DC, Briggs CJ, Vredenburg VT, et al. 2009. Skin

microbes on frogs prevent morbidity and mortality

caused by a lethal skin fungus. The [SME Journal

3(7): 818-824.

Harris RN, James TY, Lauer A, Simon MA, Patel

A. 2006. Amphibian pathogen Batrachochytrium

dendrobatidis is inhibited by the cutaneous bacteria of

amphibian species. EcoHealth 3(1): 53-56.

Howard SD, Bickford DP. 2014. Amphibians over the

edge: silent extinction risk of data deficient species.

Diversity and Distributions 20(7): 837-846.

Hyatt AD, Boyle DG, Olsen V, Boyle DB, Berger L,

Obendorf D, Dalton A, Heros M, Hines H, Phillot R,

et al. 2007. Diagnostic assays and sampling protocols

for the detection of Batrachochytrium dendrobatidis.

Diseases of Aquatic Organisms 73: 175-192.

James TY, Toledo LF, Rédder D, da Silva Leite D, Belasen

AM, Betancourt-Roman CM, Jenkinson TS, Soto-Azat

C, Lambertini C, Longo AV, et al. 2015. Disentangling

host, pathogen, and environmental determinants of a

recently emerged wildlife disease: lessons from the

first 15 years of amphibian chytridiomycosis research.

Ecology and Evolution 5(18): 4,079-4,097.

Kriger KM, Hines HB, Hyatt AD, Boyle DG, Hero JM.

2006. Techniques for detecting chytridiomycosis

in wild frogs: comparing histology with real-time

Taqman PCR. Diseases of Aquatic Organisms 71(2):

141-148.

Kriger KM, Hero JM. 2009. Chytridiomycosis,

amphibian extinctions, and lessons for the prevention

of future panzootics. EcoHealth 6(1): 6.

Longcore JE, Pessier AP, Nichols DK. 1999.

Batrachochytrium dendrobatidis gen. et sp. nov., a

chytrid pathogenic to amphibians. Mycologia 91(2):

219.

Martel A, Van Rooi P, Vercauteren G, Baert K, Van

Waeyenberghe L, Debacker P, Garner TWJ, Woeltjes

T, Ducatelle R, Haesebrouck F, et al. 2011. Developing

a safe antifungal treatment protocol to eliminate

Batrachochytrium dendrobatidis from amphibians.

Medical Mycology 49(2): 143-149.

Mauel MJ, Miller DL, Frazier KS, Hines ME. 2002.

Bacterial pathogens isolated from cultured Bullfrogs

May 2020 | Volume 14 | Number 2 | e235

Inhibition of Batrachochytrium dendrobatidis growth

(Rana castesbeiana). Journal of Veterinary Diagnostic

Investigation 14(5): 431-433.

McKenzie VJ, Bowers RM, Fierer N, Knight R, Lauber

CL. 2012. Co-habiting amphibian species harbor

unique skin bacterial communities 1n wild populations.

The ISME Journal 6(3): 588-596.

Muths E, Corn PS, Pessier AP, Green DE. 2003. Evidence

for disease-related amphibian decline in Colorado.

Biological Conservation 110(3): 357-365.

O’Hanlon SJ, Rieux A, Farrer RA, Rosa GM. 2018.

Recent Asian origin of chytrid fungi causing global

amphibian declines. Science 360(6389): 621-627.

Odds FC, Brown AJP, Gow NAR. 2003. Antifungal

agents: mechanisms of action. Trends in Microbiology

11(6): 272-279.

Olson DH, Aanensen DM, Ronnenberg KL, Powell

CI, Walker SF, Bielby J, Garner TWJ, Weaver G,

Fisher MC. 2013. Mapping the global emergence

of Batrachochytrium dendrobatidis, the amphibian

chytrid fungus. PLoS ONE 8(2): e56802.

Olson ME, Gard S, Brown M, Hampton R, Mork DW.

1992. Flavobacterium indologenes infection in

Leopard Frogs. Journal of the American Veterinary

Medical Association 201(11): 1,766—1,770.

Park ST, Collingwood AM, St. Hilaire S, Sheridan PP.

2014. Inhibition of Batrachochytrium dendrobatidis

caused by bacteria isolated from the skin of Boreal

Toads, Anaxyrus (Bufo) boreas boreas, from Grand

Teton National Park, Wyoming, USA. Microbiology

Insights 7(2): 1-8.

Pilliod DS, Muths E, Scherer RD, Bartelt PE, Corn PS,

Hossack BR, Lambert BA, McCaffery R, Gaughan

C. 2010. Effects of amphibian chytrid fungus on

individual survival probability in wild Boreal Toads.

Conservation Biology 24(5): 1,259-1,267.

Piovia-Scott J, Reyjmanek D, Woodhams DC, Worth SJ,

Kenny H, McKenzie V, Lawler SP, Foley JE. 2017.

Greater species richness of bacterial skin symbionts

better suppresses the amphibian fungal pathogen

Batrachochytrium dendrobatidis. Microbial Ecology

74(1): 217-226.

Roth T, Foley J, Worth J, Piovia-Scott J, Pope K, Lawler

S. 2013. Bacterial flora on Cascades Frogs in the

Klamath Mountains of California. Comparative

Amphib. Reptile Conserv.

Immunology, Microbiology, and Infectious Diseases

36(6): 591-598.

Sarovich DS, Pemberton JM. 2007. PPSX: a novel

vector for the cloning and heterologous expression

of antitumor antibiotic gene clusters. Plasmid 57(3).:

306-313.

Scheele BC, Pasmans F, Skerratt LF, Berger L, Martel A,

Beukema W, Acevedo AA, Burrowes PA, Carvalho T,

Catenazzi A, et al. 2019. Amphibian fungal panzootic

causes catastrophic and ongoing loss of biodiversity.

Science 363(6434): 1,459-1,463.

Scherer RD, Muths E, Noon BR, Corn PS. 2005. An

evaluation of weather and disease as causes of decline

in two populations of Boreal Toads. Ecological

Applications 15(6): 2,150—2,160.

Schloegel LM, Toledo LF, Longcore JE, Greenspan SE,

Vieira CA, Lee M, Zhao S, Wangen C, Ferreira CM,

Hipolito M, et al. 2012. Novel, panzootic, and hybrid

genotypes of amphibian chytridiomycosis associated

with the Bullfrog trade. Molecular Ecology 21(21):

5,162—-5,177.

Shiau RJ, Lin TW. 2011. Chryseobacterium indologenes

improves survival of the Chromobacterium violaceum

and violacein production. African Journal of

Biotechnology 10(13): 2,486—2,492.

Silo-Suh LA, Lethbridge BJ, Raffel SJ, He HY, Clardy

J, Handelsman J. 1994. Biological activities of two

fungistatic antibiotics produced by Bacillus cereus

UWS85. Applied and Environmental Microbiology

60(6): 2,023—2,030.

Wauters G, Kampfer P, Vaneechoutte M, Cools P,

Nemec A. 2015. Acinetobacter, Chryseobacterium,

Moraxella, and other nonfermentative gram-negative

rods. Pp. 813—837 In: Manual of Clinical Microbiology,

11th Edition. Editors, Jorsensen JH, Pfaller MA. ASM

Press, Washington, DC, USA. 2,571 p.

Woodhams DC, Geiger CC, Reinert LK, Rollins-Smith

LA, Lam B, Harris RN, Briggs CJ, Vredenburg VT,

Voyles J. 2012. Treatment of amphibians infected

with chytrid fungus: learning from failed trials with

itraconazole, antimicrobial peptides, bacteria, and

heat therapy. Diseases of Aquatic Organisms 98(1):

11-25.

Marina E. De Leon is a doctoral researcher in Jonathan A. Eisen’s lab at the University of

California, Davis (USA). Her primary research focuses on chytridiomycosis in amphibians

and the phylogenomics and production of the bacterially-produced pigment violacein.

Marina has been awarded fellowships with the National Science Foundation Graduate

| Research Fellow Program (NSF GRFP), Smithsonian Tropical Research Institute, Panama

(STRI), and the National Institutes of Health (NIH) MBRS RISE. Marina received a B.S.

f in Wildlife Conservation Biology from UC Davis and an M.S. in Biology from California

Bm State Polytechnic University, Pomona (USA).

May 2020 | Volume 14 | Number 2 | e235

De Leon

Supplementary Figures

Fig. $2. E. coli-violacein transformations. vances gene transformations of E. coli. (A) Example of NEB5- alpha-pJP1000 colony

growth after 24 h post heat shock transformation. Each week, transformants were passed by re-streaking onto fresh agar and incubated at

37 °C for 24 h. (B) NEBS5-alpha-pPSXviot, (C) NEB5-alpha-pPS Xviot+.

Amphib. Reptile Conserv. 23 May 2020 | Volume 14 | Number 2 | e235

Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 24—26 (e236).

Predatory behaviors: Pristimantis savagei (Anura:

Craugastoridae) as prey of Trechalea sp. spiders (Araneae:

Trechaleidae) in a sector of the Piedemonte Llanero,

Villavicencio, Colombia

‘Diego Escamilla-Quitian, 7Azarys de Jesus Paternina-Hernandez, and **Juan E. Carvajal-Cogollo

'Grupo de Investigacion Biodiversidad y Conservacion, Programa de Biologia, Universidad Pedagogica y Tecnolégica de Colombia, Tunja,

COLOMBIA *Grupo Biodiversidad y Conservacion, Universidad Nacional de Colombia, Bogota D.C., COLOMBIA

Abstract.—We recorded a predation event by an individual of Trechalea sp. on a Pristimantis savagei frog,

in a zone of the Piedemonte Llanero, Colombia. This is the first record of the depredation of amphibians by

arachnids in the Piedemonte Llanero.

Keywords. Amphibians, interaction networks, interspecific interaction, predator-prey, predatory behavior

Citation: Escamilla-Quitian D, Paternina-Hernandez AJ, Carvajal-Cogollo JE. 2020. Predatory behaviors: Pristimantis savagei (Anura: Craugastoridae)

as prey of Trechalea sp. spiders (Araneae: Trechaleidae) in a sector of the Piedemonte Llanero, Villavicencio, Colombia. Amphibian & Reptile

Conservation 14(2) [General Section]: 24—26 (e236).

[Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction

in any medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced,

are as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 25 June 2019; Accepted: 23 March 2020; Published: 13 May 2020

The Rain Frog, Pristimantis savagei (Anura: Crau-

gastoridae), is a Colombian endemic species distributed

from the sub-Andean forests to the sub-paramo of the

eastern slopes of the Eastern Mountain Range, ranging

from 600 to 3,000 m of elevation (Acosta-Galvis 2018).

It is a nocturnal species that prefers rocks, shrubs, and

tree leaf microhabitats, and in secondary forests it can

be found near water bodies (Van Berkum 1982). There is

currently little knowledge of the biotic interactions and

natural history of this species, as well as its predators,

habitat use, population density, and other aspects of basic

biology.

The arachnids of genus Trechalea (Thorell, 1869)

(Arachnida: Trechaleidae) are aggressive predators of

a large variety of prey, including both vertebrates and

invertebrates, and they can be easily identified because

they are relatively large spiders with a prosoma length

of 6.1-11.0 mm (Carico 1993). They are mainly found

on rocks, trunks, leaves, and on the ground near water

bodies (Van Berkum 1982). They are also reported to

favor specific foraging sites near water bodies, possibly

due to the prey availability in these habitats (Van

Berkum 1982; Adams 2000). As for the Rain Frog in the

Neotropics, limited information is available regarding its

foraging, types of prey, and general resource use (Whiles

et al. 2006; Cortés-Goémez et al. 2015).

During recent nocturnal sampling for characterizing

the amphibian fauna ina sector of the Piedemonte Llanero

Correspondence. *juancarvajalc@gmail.com

Amphib. Reptile Conserv.

in the Municipality of Villavicencio-Meta, Colombia

(Fig. 1), which is characterized by having Pourouma

guianensis, Rinorea macrocarpa, and Cassia moschata

mixed forests (Rangel-Ch and Minorta-Cely 2014),

we recorded a predation event in which an individual

Trechalea sp. spider depredated a Rain Frog Pristimantis

savagei (Fig. 2). The behavioral observation occurred

on 20 March 2018 during the heaviest rain season of the

year, after 2100 hours, near the stream of Cafio Buque, in

the village of El Carmen, municipality of Villavicencio,

Meta, Colombia (4°8.641’N, 73°40.071’W, 737 m asl).

The area of the finding was characterized as a gallery

forest, slightly intervened with canopy cover that

exceeded 60%, and the presence of a large amount of

rocks, and a stream with a low flow.

Specifically, the predation event took place in the

following sequence of events. The spider was situated

on a rock at an approximate height of 150 cm, and less

than 1 m from the water body (Fig. 3), with a passive

foraging behavior (stillness). During the capture of the

prey there were no major movements for more than

10 minutes, from either the spider or the frog. Despite

being alive, the frog offered no resistance to predation,

perhaps due to a substance injected by the arachnid.

The secretion of digestive enzymes could be seen,

which turned the frog into a “broth” (Fig. 4), as was

also described by Toft (2013) as a feeding mechanism

of these spiders.

May 2020 | Volume 14 | Number 2 | e236

Escamilla-Quitian et al.

73° 15' 00"

73° 30° 00" 73° 45' 00"

( ,

AG 4300 a

— \\\\ villavicensiosMeta Colombia ‘\

4° 09" 00"

4° 09' 00"

4° 07" 00"

4° oF" 00"

4° 05" 00"

4? 05" 00"

73° 30° 00"

Fig. 1. Site of Pristimantis savagei predation by Trechalea

sp., in a sector of Piedemonte Villavicencio-Meta, Colombia.

The red dot indicates the exact site of discovery in the Cafio

Buque.

The findings of anuran depredation by spiders have

been well documented in some neotropical regions (e.g.,

Toledo 2005; Pombal, 2007). The works by Costa-Pereira

et al. (2010), found that for some spiders of families

Psauridae and Trechaleidae, their diet included juvenile

Osteocephalus taurinus frogs, and that depredation

was attributed to the corporal size and abundance of

this frog in the juvenile phase, which led to it being a

major component of the diet of the spiders. Moura et

al. (2011) recorded the depredation of Dendropsophus

melanargyreus by the giant spider, Ancyclometes rufus,

in zones near streams. Also, Kirchmeyer et al. (2017)

recorded the depredation of the tree frog, Scinax similis,

by a weaver spider, Eriophora fuliginea, where the prey

was alive wrapped in the spiderweb. While these are just

a few examples, many of these reports are similar to the

event reported here for the Piedemonte Llanero, which

could indicate a relatively common pattern in these

biological interactions.

The importance of this finding lies in its addition

to the known natural history of this predator and the

interspecific interactions of the prey, and it contributes

to the debate of new questions in the investigation of

the trophic network dynamics in an established area. In

addition to Trachaleidae, there are records of seven other

families of Araneae as anuran predators, while most of

the cases where recorded for the fisher spiders from the

family Pisauridae (Toledo 2005). For the Trechaleidae

family, very few cases have been reported on its hunting

behavior (Hofer and Brescovit 2000), foraging modes, or

predation rates, however the observation reported here

adds the first detailed record for the Piedemonte region

in Colombia, and brings to mind new investigations as

documented by Barbo et al. (2009) for other latitudes.

Amphib. Reptile Conserv.

Fig. 2. Record of predation of Pristimantis savagei by Trecha-

lea sp., in a sector of Piedemonte, Villavicencio-Meta, Colom-

bia.

Fig. 3. Warning stance of the spider 7rechalea sp. The indi-

vidual was on a rock at an approximate height of 150 cm, less

than 1 m from the water source.

Fig. 4. 7rechalea sp. individual secreting its digestive juices.

Other aspects of importance for the predatory

event of Trechalea sp. on Pristimantis savagei are the

anthropogenic stressors witnessed in the study area

(habitat loss and fragmentation), because these have been

reported to impact, in either a positive or negative way,

the interspecific interactions and lead to imbalances in

the natural system (Mahecha-J and Diaz-S 2015).

Finally, our results in this work provide a baseline in

the formulation of new questions for investigation that

can lead to a better understanding of the dynamics (on

the interspecific level) of interaction networks in areas

with high ecological and biological value, such as the

Piedemonte Llanero of Colombia.

May 2020 | Volume 14 | Number 2 | e236

Spider predation on anuran in Colombia

Acknowledgment.—The authors thank CONANDINO

and the Universidad Pedagdogica y Tecnoldgica de

Colombia who financed the framework project through

Agreement 057-2017.

Literature Cited

Acosta-Galvis AR. 2018. Lista de los Anfibios de

Colombia. Batrachia, Villa de Leyva, Boyaca,

Colombia. Available: —_ http://www.batrachia.com

[Accessed: 5 May 2020].

Adams MR. 2000. Choosing hunting sites: web site

preferences of the orb weaver spider, Neoscona

crucifera, relative to light cues. Journal of Insect

Behavior 13(3): 299-305.

Barbo FE, Rodrigues MG, Couto FM, Sawaya RJ. 2009.

Predation on Leptodactylus marmoratus (Anura:

Leptodactylidae) by the spider, Ctenus medius

(Araneae: Ctenidae), in the Atlantic Forest, Southeast

Brazil. Herpetology Notes 2(19): 99-100.

Carico JE. 1993. Revision of the genus 7rechalea Thorell

(Araneae, Trechaleidae) with a review of the taxonomy

of the Trechaleidae and Pisauridae of the Western

Hemisphere. Journal of Arachnology 21(3): 226-257.

Cortés-Gomez AM, Ruiz-Agudelo CA, Valencia-Aguilar

A, Ladle RJ. 2015. Ecological functions of neotropical

amphibians and reptiles: a review. Universitas

Scientiarum 20(2): 229-245.

Costa-Pereira R, Martins FI, Sczesny-Moraes EA,

Brascovit A. 2010. Predation on young treefrogs

(Osteocephalus taurinus) by arthropods (Insecta,

Mantodea and Arachnida, Araneae) tn Central Brazil.

Biota Neotropica 10(3): 469-472.

Hofer H, Brescovit AD. 2000. A revision of the

Neotropical spider genus Ancylometes (Bertkau 1880)

(Araneae: Pisauridae). Predation on young treefrog

(Osteocephalus taurinus) by arthropods (Insecta,

Mantodea and Arachnida, Araneae) tn Central Brazil.

Insect Systematics & Evolution 31(19): 323-360.

Kirchmeyer J, Amaral L, Magaldi A, Baptista R, Carvalho-

e-Silva S. 2017. Predation on the treefrog, Scinax similis

(Anura: Hylidae), by the orb-weaver spider, Eriophora

fuliginea (Araneae: Araneidae), in southeastern Brazil.

Phyllomedusa: Journal of Herpetology 16(1): 113-116.

Mahecha JO, Diaz-S V. 2015. Aproximacion ala diversidad

taxonomica de las mariposas diurnas (Lepidoptera:

Papilionoidea) en la Vereda Cafreria, Municipio

Icononzo, Tolima. Revista Cientifica Unincca 2: 83-91.

Moura MR, Azevedo LP. 2011. Observation of

predation of the Giant Fishing Spider Ancylometes

rufus (Walckenaer, 1837) (Araneae, Ctenidae) on

Dendropsophus melanargyreus (Cope, 1877) (Anura,

Hylidae). Biota Neotropica 11(4): 349-351.

Pombal Jr JP. 2007. Notas sobre predacéo em uma

taxocenose de anfibios anuros no sudeste do Brasil.

Revista Brasileira de Zoologia 24(3). 841-843.

Rangel-Ch O, Minorta-Cely V. 2014. Los tipos de

vegetacion de la Orinoquia colombiana. Pp. 533-612

In: Colombia Diversidad Bidtica XIV. La region de la

Orinoquia de Colombia. Editor, Orlando Rangel-Ch

J. Instituto de Ciencias Naturales, Bogota, Colombia.

895 p.

Toft S. 2013. Nutritional aspects of spider feeding. Pp.

373-384 In: Spider Ecophysiology. Editor, Nentwig

W. Springer, Berlin/Heidelberg, Germany. 529 p.

Toledo LF. 2005. Predation of juvenile and adult anurans

by invertebrates: current knowledge and perspective.

Herpetological Review 36(4): 395-400.

Van Berkum FH. 1982. Natural history of a tropical,

shrimp-eating spider (Pisauridae). Journal of

Arachnology 10(2): 117-121.

Whiles MR, Lips KR, Pringle CM, Kilham SS, Bixby RJ,

Brenes R, Connelly S, Colon-Gaud JC, Hunte-Brown

M, Huryn AD. 2006. The effects of amphibian popula-

tion declines on the structure and function of Neotropi-

cal stream ecosystems. Frontiers in Ecology and the

Environment 4. 27-34.

Amphib. Reptile Conserv. 26

Diego Escamilla-Quitian is a last-semester student of Biology at the Universidad Pedagogica y Tecnoldgica

de Colombia and belongs to the Biodiversity and Conservation research group. Diego has great interest in

studying the ecology, conservation, and taxonomy of amphibians and reptiles of the Colombian territory. He

has carried out research work in the Colombian Andes and in the Choco Biogeographic Region.

Azarys Paternina-Hernandez is currently a researcher in the Biodiversity and Conservation research group

at the Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota, Colombia. Azarys obtained

an M.Sc. from the Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota, Colombia, in

the subject of Landscape Ecology and Amphibian ecology. Azarys is particularly interested in amphibians,

focusing on their habitats, ecology, and conservation.

Juan E. Carvajal-Cogollo is currently an Associate Professor in the Biology program, Universidad Pedagégica

y Tecnologica de Colombia, and is the director of the Museo de Historia Natural Luis Gonzalo Andrade,

Tunja, Colombia. Juan obtained his Ph.D. from the Instituto de Ciencias Naturales, Universidad Nacional de

Colombia, Bogota, Colombia, in the subject of Landscape Ecology and reptile ecology. He has a keen interest

in the natural history of neotropical amphibians and reptiles, and has authored several articles in international

journals and book chapters on Colombian biodiversity. Juan is particularly interested in the habitats, ecology,

and conservation of amphibians and reptiles.

May 2020 | Volume 14 | Number 2 | e236

Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 27-46 (e237).

Endemism on a threatened sky island: new and rare species

of herpetofauna from Cerro Chucanti, Eastern Panama

123Abel Batista, ?**Konrad Mebert, 7Madian Miranda, 7Orlando Garcés, 7Rogemif Fuentes,

and °Marcos Ponce

'ADOPTA El Bosque PANAMA 2Los Naturalistas, PO. Box 0426-01459 David, Chiriqui, PANAMA ?2Universidad Autonoma de Chiriqui, Ciudad

Universitaria El Cabrero David, Chiriqui, 427, PANAMA ‘Programa de Pos-graduacgao em Zoologia, Universidade Estadual de Santa Cruz,

Rodovia Jorge Amado, Km 16, 45662-900, Ilhéus, Bahia, BRAZIL *Museo Herpetoldgico de Chiriqui, PANAMA

Abstract.—Cerro Chucanti in the Darien province is the highest peak in the Majé Mountains, an isolated massif in

Eastern Panama. In addition to common herpetological species such as the Terraranas, Pristimantis cruentus,

and P. caryophyllaceus, rare species such as Pristimantis moro and Strabomantis bufoniformis occur as well.

Recent expeditions to Cerro Chucanti revealed a remarkably rich diversity of 41 amphibian (19% of the total

in Panama) and 35 reptile (13% of the total in Panama) species, including new and endemic species such as

a salamander, Bolitoglossa chucantiensis, a frog Diasporus majeensis, and a snake, Tantilla berguidoi. Here,

an up-to-date summary is presented on the herpetological species observed on this sky island (an isolated

mountain habitat with endemic species), including several species without definitive taxonomic allocation, new

elevation records, and an analysis of species diversity.

Keywords. Amphibians, community, diversity, evaluation, integrative taxonomy, premontane, reptiles, surveys

Resumen.—El Cerro Chucanti en la provincia de Darién es el pico mas alto de las montanas de la serrania de

Majée, un macizo aislado en el este de Panama. Ademas de las especies herpetologicas comunes como las

ranas, Pristimantis cruentus, y P. caryophyllaceus, tambien ocurren especies raras, p. ej. Pristimantis moro y

Strabomantis bufoniformis. Las recientes expediciones al Cerro Chucanti revelaron una gran diversidad de 41

especies de anfibios (19% del total en Panama) y 35 especies de reptiles (13% del total en Panama), incluidas

especies nuevas y endémicas, como la salamandra, Bolitoglossa chucantiensis, la rana Diasporus majeensis

y la serpiente, Tantilla berguidoi. Presentamos un resumen actualizado de las especies herpetologicas

observadas en esta “sky island” (un habitat de montana aislado con especies endemicas), que incluye una

multitud de especies sin asignacion taxonomica definida, nuevos registros de elevacion y un analisis de la

diversidad de especies.

Palabras clave. Anfibios, comunidad, diversidad, evaluacion, integrativa, premontano, reptiles, taxonomia

Citation: Batista A, Mebert K, Miranda M, Garcés O, Fuentes R, Ponce M. 2020. Endemism on a threatened sky island: new and rare species of

herpetofauna from Cerro Chucanti, Eastern Panama. Amphibian & Reptile Conservation 14(2) [General Section]: 27-46 (e237).

4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any

medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced, are

as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 27 March 2019; Accepted: 5 April 2020; Published: 14 May 2020

Introduction

For the last few decades, the knowledge on the

herpetofauna of Panama has experienced a rapid increase,

revealing a high species richness with 219 species of

amphibians (Batista et al. 2016; Hertz 2015) and 263

of reptiles (Lotzkat 2015). Most herpetological studies

have been concentrated in Central Panama and more

recently in Western Panama (Hertz 2015; Lotzkat 2015).

Many areas remain poorly explored, and several remote

sites are threatened and urgently need an evaluation of

their species diversity and population status. The focus

of recent studies has shifted to the herpetofauna of the

far less known eastern half of Panama (Batista et al.

2014a,c, 2015, 2016a,b), with most distinctive areas

lacking representative species check lists. The Chucanti

Private Nature Reserve (CPNR) is one such area without

any comprehensive inventory. The reserve is part of the

Eastern Panamanian montane forests (World Wildlife

Fund 2014) and lies within the Serrania de Majé

(hereafter, Majé Mountains). These mountains contain

many peaks between 800—1,200 m asl, but only Cerro (or

Mount) Chucanti reaches an elevation of >1,400 m asl

(Samudio 2001).

Correspondence. abelbatista@hotmail.com (AB); *konradmebert@gmail.com (KM), mpamela0305@gmail.com (MM);

orlgarces2 1@gmail.com (OG); rogemifdaniel@gmail.com (RF); marcosponce27@gmail.com (MP)

Amphib. Reptile Conserv.

May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

Cerro Chucanti represents a sky island (Doge 1943;

Heald 1967, pp. 114-126), a concept widely used in the

study of island biogeography that describes mountains

as “islands of habitat” (e.g., Quammen 2004, pp. 436—

447, Warshall 1994). Indeed, because of its higher

elevation, Cerro Chucanti is the only area within the

Mayjé Mountains covered predominantly by cloud forest,

and it is isolated from the nearest equivalent cloud forest

of the Cerro Pirre range by a stretch of 120 km which

consists of completely different lowland terrain. More

than a dozen new species of fungi, plants, invertebrates,

and vertebrates have recently been described which are

known only from Cerro Chucanti (see articles listed

at: http://adoptabosque.org/chucanti/newspecies/).

Unfortunately, as much as tropical sky islands have

promoted the evolution of endemic species, and thus,

harbor a disproportionately high diversity, they are under

increasing threat from global warming, causing the

redistribution of biodiversity patterns (Pecl et al. 2017),

and from drastic habitat loss due to upward displacement

of species (Kok et al. 2016; Haines et al. 2017; Mizsei et al.

2020) where the local topography allows this (Sekercioglu

et al. 2008). In addition, the montane rainforest of Cerro

Chucanti is significantly threatened by slash-and-burn

agriculture, as well as logging and cattle ranching on all

but the steepest slopes (http://www.rainforesttrust.org/

expansion-of-the-cerro-chucanti-nature-reserve/). Species

restricted to sky islands like Cerro Chucanti are thus

clearly at risk of extinction, so protecting pristine land

becomes a priority. To address this, Rainforest Trust has

partnered with ADOPTA (http://www.adoptabosque.org)

to purchase land with the long-term aim of creating a

broader government-protected area.

Furthermore, there is an urgent need to evaluate the

exact species composition and status of the amphibians

and reptiles of the Cerro Chucanti sky island, because high

altitude ectothermic animals are particularly vulnerable

to climate change due to their low dispersal ability, a

high level of habitat specialization, and fragmented

distributions (Davies et al. 2004; Sinervo et al. 2010).

A few herpetological studies have been conducted in the

Majé Mountains; including the first amphibian diversity

study in 2007 by Medina et al. (2019), and some effort

focusing on the entire herpetofauna of the Cerro Chucanti

by the current authors since 2012 (e.g., Batista et al.

2014b), resulting in one new species of salamander and

a snake described for this peak area (Batista et al. 2014a,

2016b). This article presents the first check list on the

herpetofauna of CPNR, including the recently described

species and those in anticipation of a formal description.

Furthermore, analyses of abundance and diversity are

presented.

Methods

Located in the southeastern part of the Majé Mountains,

the higher elevation Cerro Chucanti (1,439 m_ asl,

Amphib. Reptile Conserv.

8.8046°N, 78.4595°W; Fig. 1A,B) is part of the eco-region

“Eastern Panamanian Montane Forests” (World Wildlife

Fund 2014). There is no particular climatic information

available for this mountain range (Samudio 2001), but

according to the eco-region, the precipitation varies

between 3,000 and 4,000 mm/year, and the temperature

between 20 and 27 °C (World Wildlife Fund 2014). The

vegetation belts of Cerro Chucanti are Tropical Lowland

Wet/Moist Forest (0-600 m asl), Premontane Moist

Forest (S00—1,000 m asl), and a small area of Premontane

Wet Forest above 1,000 m asl (Holdridge 1967; with

modifications from ANAM 2010; Ramirez 2003). In this

region, rainfall occurs mostly during April—December

(Rio Majé Meteorological Station, 70 m asl; http://www.

hidromet.com.pa/, accessed on 19 Sep 2015). To evaluate

the status of the CPNR herpetofauna, surveyors walked

along transects with 1-2 m width by applying VES

(Visual Encounter Survey) and some voice recording

of the anurans. Only post-metamorphic life stages of

anurans were sampled. The two investigated areas within

the reserve are described below.

Premontane Moist Forest (PMF, Fig. 2). PMF is the

area surrounding the Chucanti Biological Station (~800

m asl), with the forest continuing along a trial above the

station up to ~950 m asl. The forest is pristine with an

open understory and variably-sized boulders scattered

across the floor, which provide many holes and crevices

that can be used as shelter for amphibians and reptiles.

The leaf litter is scarce, mainly due to the inclination

of the terrain (with some slopes > 40°). The area below

the biological station is structured by two trails and the

Chucanti Creek. The main trail, labeled “Entrance” in

Fig. 1B, leads after a few hundred meters downward to

the exit of the reserve, and into an open pasture land. This

area consists mainly of secondary forest (~15 years old).

The second trail continues along “Helicopteros trail”

into a loop to the top of Cerro Chucanti and back along

“Chucanti” and “Escalera trails” to the biological station.

Most areas below 900 m asl are covered by old growth

secondary forest. The PMF zones of both trails, including

the short trail labeled Cascada (Fig. 1B), and Chucanti

Creek were sampled along eight transects during three

visits to the reserve on 1-5 December 2012, 2—5 April

2015, and 8-16 October 2016 (Table 1).

Premontane Wet Forest (PWF, Fig. 2). This forest

begins as low as 950 m asl, where the PMF changes

gradually into the PWF as one ascends. It is a cloud forest

that covers the mountain ridges and several smaller peaks

around Cerro Chucanti. The understory is filled with

palm trees, ferns, and plenty of epiphytes, whereas tree

bark and branches are copiously covered by moss. The

trail up the slope from the biological station passes by

two helicopter wreckages (which crashed decades ago)

and leads to Camp Site | at 1,200 m asl. The understory

and forest floor of the highest areas surrounding Cerro

Chucanti (1,350-1,439 m asl) and the ridge following

east to a second peak (1,290 m asl) are extensively

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Helicopters:

Pp ~LY

Ons

Holdridge life zones

Lowland Moist Forest

{__] Premontane Moist Forest

{| Premontane Wet Forest

—— Trails

—— Streams

0 0

250 500 m

Google Earth

8°45'04.21"N 78°28'28.11"W elev 424m eye alt 2.52km >

@ Bolitoglossa chucantiensis

OB. aff. biseriata

* Diasporus majeensis

® Dermophis aff. glandulosus

% Lachesis acrochorda

®@ Zantilla berguidoi

axe

Entrance ——_

™»

Datum WGS 84, EPSG: 4326 ‘

March, 2019

-78.475

Abel Batista

650

-78.450

Fig. 1. (A) Satellite map and (B) abstract digital map showing the trails used on Cerro Chucanti for transects in the PWF (beige,

black, partly blue, and red dashed trails) and PMF (orange, dark-grey, and green dashed trails). Records of some species are also

shown (see Materials and Methods for details).

covered with bromeliads and moss (Fig. 2). Twelve

transects were laid out in this forest, some of which were

sampled repeatedly during 2012—2016, whereas others

were visited only once (Table 1). Transects all started or

ended at Camp Site 1 and were set up along the following

five trails/areas (Fig. 1B): (1) Chucanti top: covers the

peak; (2) Helicopteros trail: expands along the main trail

down to the biological field station; (3) Chucanti SSW

trail: follows the ridge south-south-west; (4) Chucanti

NNE trail: follows the ridge north-north-east; and (5)

Short Loop and Cristalita: includes two transects on

slopes down to 1,200 m asl and 1,040 m asl, respectively,

which lead to the “Cristalita” stream (Spanish for little

Amphib. Reptile Conserv.

glass), so-named during our surveys because of the loud

choir of calling glass frogs when we arrived there the

first time (see details in Fig. 1 and Table 1). Georeference

points were recorded using a Garmin GPSmap 60CSx

in the WGS 1984 datum format, changed into decimal

degrees, and maps created in QGIS 2.18.0.

Species identification mainly used the keys by Kohler

(2008, 2011), augmented by specialized publications

on the amphibians and reptiles of Eastern Panama.

Specimens that could not be positively identified in the

field, or for which taxonomic allocation was doubtful,

were collected with the permission of the Ministerio

de Ambiente (permits for 2012: SC/A-33-12; for

May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

861 oum ut [eJ0],

8 007 79/-ObL O€ISp8L €SL06L'8 S60Sr8L IS06L'8 ISIOJ] OULJUOW Ig 910T “Yeoso MURONYD OZL

v 007 00¢‘ 1-006 0679V'8L TLS6L'8 Y9ITSP8L ET6BL'B ISIOJ| OULJUOWIg S107 “UOHe}s play svoqy and

Z 00r OrZ—-008 LETSP'8L = OELBL'8—s«OITSH'BL = EZOHBL'B ISIOJ] OURJUOWIIg WOHRIS PJP $,dAJOso oY} 0} 9DURIUY SLL

9 00P 0S8-008 99€Sp 8L POl6l’8 9ITSP8L €768L'8 ISIOJ] OULJUOW Ig 9107 “Yoold HuRONYD 0} dooy ‘uoneys poly Ain

8 O0€ SLO'I-O07'1 OCh9r'8L OF008'8 S60Sr'8L — IS06L'8 19M OULIUOWOI JSOMYLOU UMOP | os dweD OLL

v7 O0€ OSS I-pLZ1 0609V8L TLS6L'8 S60Sr8L — IS06L'8 19M ULILOWOI 9107 ‘Ssajdoorjay 0} | ays dues ysosoy pnojy SLL

87 009°T 6TrI-00E1 968S78L 9LP088 6779P8L 89L6L'8 19M SURUOUIOIg 9107 ‘| doy HueonyD_ Isos0J pno[D ri

07 008°T PLTI-ZSO'T «= OCT OV'BL = BOLOL'B~—sEI9OH'BL ~— BBE08'8 JOM SURIUOWISIg ] Sus dures 0} vuyeysti CLL

v 001 O8O'I-zso'l - 9TS9V'BL = TIPOB'BS = EL9NDV'BL ~—s BB8E08'8 JOM OUL]UOWIAI WeI]S BUTEISLID ZL

Ol Tr6 PLTI-O0E'1 80S9P8L 669648 6779V8L 89L6L'8 JOM OURIUOWIOI JSOMYINOS UMOP | ays dud ILL

8 00s 6TrI-OOE'T 968SP8L 97088 6729V8L 89L6L'8 JOM SULIUOWAIg 9107 *] doy WueonyD Isos0J pno[D OLL

9 007 00¢‘ 1-006 0679P'8L TLS6L'8 Y9ITSP8L = ET6BL'8 JSIOJ| OULJUOWIg S107 “UONe}s play sA0qy al

9 007 79/-9bL O€ISp8L SLO6L’8 S60Sr8L IS06L'8 ISIOJ] OULJUOW Ig 9107 “Yeoso MURONYD SI.

6 O00€ LTSIPLT1 «= COLPLTBL COS PLH'8 = 6TZOV'BL — 889L6L'8 JOM OULJUOWIOI SLOT “1 ONS dures jsoroj pno[D ‘at

6 002 79/-9bL O€ISp8L SLO6L’8 S60Sr8L IS06L'8 ISIOJ] OULJUOWII S107 “Yoosd NURONYD OL

ZI 00s 6TrI-OOE'T 968S7'8L 9LP08'8 67~9V'8L = 89L6L'8 JOM SURIUOUISIg S107 “[ doy HueonyD Isosoy pnoyD SL

8 00€ LTSI-PLT1 «= OOLPLTBL~—COSHLH'B8 = OTZOV'BL = B9LEL'S JOM OULILOWOI ZIOT “YdIvas [eUINIP | ONS dures ysor0J pnojD rl

9 00s 6TYI-OOE'T 968S7'8L 9LP088 67C9V8L 89L6L'8 19M OULILOWOI ZIOT “1 doy yueonyD IsosoJ pnopD EL

Ol 00€ LTEI-PLT1 «= COLPLTBL COS LHS = OTTOV'BL — BOLEL'S 19M OULILOWOI ZLOT “I OHS dues ysos0F pnojD ral

Ol 007 y9L-9PL O€ISp'8L SLO6L8 S60Sr8L IS06L’8 ISIOJ] OULJUOWAIg ZLOZ “Yaeso MURONYD LL

(ui) (Ise ur) M N M N

siy/uvul =. 0B SIG] WONVAIA Sd} VUIP100) Puy SI} VUIP1OOD J.1B}S 3d 4} 489.104 podures zvak UOHBIOT = =—-| DOSUBA],

May 2020 | Volume 14 | Number 2 | e237

‘POA[OAUT SIOADAINS Ploy Jo Joquinu ay} Aq

pordnynu JoAdAIns ou Aq JD9SUBT] B SUTYOIIS PdJSOAUT SINOY JO JOQUINU :IY/UCUT “SIOJOW UI DOULISIP SUTY[LM :9dURISIC, OTOT 01 TINT WOT, UNdD OY} Je poydures sjoosuvy Jo sjiejoq *T aquy

Amphib. Reptile Conserv.

Batista et al.

Fig. 2. Study area at CPNR. (A) Cerro Chucanti; (B) entrance to the reserve; (C) Camp Site 1, on 2016; (D) modern toilet at Camp

Site 1; (E—-F) forest above 1,300 m asl; (G) view of the secondary forest around the biological station up to the cloud forest on the

ridge; (H) the field team, at 1,200 m asl from left: Madian Miranda, above Rogemif Fuentes, Orlando Garces, below Abel Batista,

and lower right Konrad Mebert.

Amphib. Reptile Conserv. May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

2013: SEX/A-7-13; for 2016: SE/A-60-16). Collected

specimens were euthanized by applying T61 solution,

preserved with 5 ml formalin (36%) in 1 L ethanol

(94%), and subsequently stored in ethanol (70%).

Specimens were deposited in the Senckenberg Museum

of Frankfurt (SMF), Museo Herpetologico de Chiriqui

(MHCH), or Museo de Vertebrados de la Universidad

de Panama (MVUP). The nomenclature used to make

the check lists corresponds to the taxonomy obtained

from AmphibiaWeb (http://amphibiaweb.org/) and The

Reptile Database (http://reptile-database.org/).

To evaluate the abundance and diversity status of

amphibians and reptiles in the CPNR, diversity analyses

were performed using PAST software (Hammer et al.

2001). Dominance (D), Shannon (H’) and Equitability

(J) were used as metrics for community evenness

inferences. Relative richness was calculated dividing the

number of taxa per hour of survey time in a transect. To

predict the maximum number of reptile and amphibian

species by forest type, we applied ESTIMATES (Colwell

2006), using the sample-based estimation with 100 runs

of randomization and extrapolated to a factor of doubling

the effort made.

Results

The surveys accumulated a sampling effort of 198 man/

hr and covered 11,793 m of trails (Table 1). A total of

41 amphibian and 35 reptile species were found for the

CPNR (Figs. 3—5; Appendix 1). These numbers consist

of: total of observed species, including those newly

described since 2014 (29 amphibians, 24 reptiles);

potentially new species (nine amphibians, three reptiles):

species likely to occur on Cerro Chucanti, but found in

similar habitats and elevations of an adjacent area within

the Majé Mountains (two amphibians, eight reptiles):

and one non-collected species of a Marsupial Frog heard

from the canopy tree at Camp Site 1 during the 2012

survey. A video-recording captured the brief calls of at

least three specimens, with identifications that they most

likely refer to Gastrotheca nicefori corroborated among

experts (including W.E. Duellman and U. Sinsch).

Among the transects, amphibians were more diverse

at lower elevations with 27 species in the PMF versus

20 species higher up in the cloud forest or PWF (Fig.

6; Appendices 2-3). This diversity is also reflected in

the higher relative species richness index in the PMF

versus PWF for amphibians (average richness index

across transects per forest type: 1.42 versus 0.84).

According to the estimations, a slight increase in the

number of species may occur if the hourly effort made

were doubled, with three more species in the PMF

and one in the PWF. Reptiles showed a comparatively

smaller general species richness (Fig. 6; Appendices

2-3), but a much greater increase would be expected

than in amphibians if the sampling effort were doubled,

with nine more species in the PMF and seven in the

Amphib. Reptile Conserv.

PWF (Fig. 6). The relative richness in reptiles was also

greater in the PMF than in PWE (average richness index

across transects per forest type: 0.46 versus 0.24). The

PME showed also a more evenly distributed community

for both taxonomic groups and most transects (Fig. 6;

Appendices 2-3). The amphibian community in the PWF

was dominated by three species with the highest relative

abundance (Pristimantis caryophyllaceus, Diasporus

majeensis, and Pristimantis cruentus), which together

contributed 75% of all amphibian individuals seen in

that forest (Figs. 7-8). Among reptiles, a potentially new

lizard, Ptychoglossus aff. plicatus, and snake, Geophis

aff. brachycephalus, were the species with the highest

relative abundance in the PWF, contributing 62% of all

reptiles (Figs. 9-10). Across both forest types, amphibian

and reptile species numbers were higher in aquatic than

terrestrial habitats (Fig. 11).

Of the defined species from the CPNR, only one is

listed in any threat category of the IUCN, Pristimantis

pardalis, which is listed as “Near Threatened.” Most of

the remaining species are listed in the “Least Concern”

category or have not been evaluated yet (see Appendix 1).

However, according to the environmental vulnerability

score (EVS), three amphibian and 12 reptile species

are listed in the “high” EVS category (Johnson et al.

2015), although most amphibians scored in the “low”

EVS category and most reptiles in the “medium” EVS

category (see Appendix 1).

Discussion

The high endemism of Cerrro Chucanti, and Eastern

Panama (EP) in general, is a consequence of the complex

geological history of the Isthmus of Panama, with

EP representing the northernmost block of the Choco

biogeographical region (Duque-Caro 1990). This block

began with a widespread uplift as a result of the collision

of the Panama Arc with South America around 20 Mya

(Montes et al. 2012). The uplift continued and formed

regional mountain ranges, including Cerro Chucanti

in the Majé Mountains, that became isolated through

eustatic fluctuations during the middle and late Miocene

(as early as 11 Mya), such as the flooding of the Atrato

and Chucunaque Basins (Duque-Caro 1990; Coates et al.

2004). This isolation promoted speciation events and an

increased species diversity in this region from 5—8 Mya

(Batista et al. 2014a,b, 2016a; Coates et al. 2004).

Currently, the Majé Mountains are separated by at least

100 km from the nearest mountain ranges with elevations

reaching > 1,000 m asl (Cerro Pirre, Cerro Darien). Cerro

Chucanti is situated at the easternmost portion of the

Mayjé Mountains and represents a sky island, as it is the

only site in the region with elevations of more than 1,200

m, providing the topographic and climatic conditions

for a largely isolated cloud forest. This restricted habitat

is home to several endemic species (e.g., Bolitoglossa

chucantiensis, Bolitoglossa aff. biseriata, Diasporus

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Fig. 3. Selected amphibian species found during the 2012—2016 su o Chucanti. (A) Dermophis aff. glandulosus, (B)

second known specimen of Bolitoglossa chucantiensis, recently described and endemic (Batista et al. 2014a); (C) Bolitoglossa aff.

biseriata, (D) Oedipina aff. complex; (E) Strabomantis bufoniformis, (F) Diasporus majeensis, recently described and endemic

(Batista et al. 2016a); (G) Pristimantis gaigei, (H) Pristimantis moro.

Amphib. Reptile Conserv. 33 May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

Fig. 4. Selected amphibian species found during the 2012—2016 surveys on Cerro Chucanti that await formal description or

clarification of relationships. (A) Colostethus aff. pratti, (B) Silverstoneia sp.; (C) Pristimantis aff. latidiscus, (D) Pristimantis aff.

ridens.

majeensis, Geophis aff. brachycephalus, and Tantilla

berguidoi), that have only been found within this area

of about 3 km? calculated with Google Earth (Batista et

al. 2014a,b, 2016a). Its high diversity and endemism are

surprising, generally comparable with other highlands in

the Neotropics, yet slightly higher than the average for

either Central America (Batista and Ponce 2002; Chavez

and Morales 2014; Duellman 1988; Veith et al. 2004;

http://www.somaspa.org/) or the tepuis in South America

(Aubrecht et al. 2012).

In general, amphibians are easier to detect than

reptiles (Myers 2003; Savage 2002; Yeo and Peterson

1998), which is consistent with our results (Fig. 6). The

observed species richness at Cerro Chucanti, at least in

the cloud forest (see potential species in Appendix 1, Fig.

6), appears to approach its current maximum, although

its canopy was not explored. Furthermore, additional

reptile species are expected to be discovered in the future,

because reptiles usually yield lower detection probabilities

(see Fig. 6 for estimating richness). The species richness

in the PMF was consistently larger than in the higher-

altitude PWF (Premontane Wet Forest or cloud forest)

for both amphibians and reptiles, even though a greater

effort was invested in the PWF. This confirms the general

rule of diversity patterns in the American tropics (Wilson

et al. 2010), with decreasing diversity from intermediate

elevations where diversity 1s highest (Whitfield et al.

Amphib. Reptile Conserv.

2016), to higher altitudes where cooler temperatures

physiologically restrict ectothermic animals. Medina et

al. (2019) suggested that the reduced number of observed

species at the highland site on Cerro Chucanti may reflect

the absence of streams and ponds, and the reduced patch

size of the cloud forest. However, the altitude range at

the CPNR was too small to experience large climatic

differences, and the area available for the cloud forest

(PWF) is very limited, roughly estimated at only 1.8

km? above 1,200 m asl (Batista et al. 2016; Medina et al.

2019). These factors may possibly dampen the effect of

altitudinal zonation due to overlap between the two forest

types. However, the J evenness index for PWF versus

PMEF (see Appendices 2-3) indicates that the zonation is

sufficiently accentuated so that a few species dominate,

or have been recorded only, which are particularly well-

adapted to the cooler environment in the PWF. For

example, Pristimantis cruentus, P. caryophyllaceus, and

Diasporus majeensis, accounted for 75% (according to

recorded advertisement calls) of the total relative species

abundance in the PWF. Similarly, only one lizard,

Ptychoglossus sp., dominates the reptilian composition

with a relative abundance of 42% among all species.

Amphibian species richness in Cerro Chucanti was

higher in this study (combined 15 days survey for the

three years) than in a 10-day survey conducted in 2007

(95% CI, see Table 2 in Medina et al. 2019). The observed/

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Fig. 5. Selected reptile species found during the 2012-2016 surveys on Cerro Chucanti. (A) Echinosaura aff. palmeri; (B)

Ptychoglossus aff. plicatus, (C) Anolis aff. fuscoauratus, (D) Geophis aff. brachycephalus, (E) Corallus annulatus, highest

elevation record; (F) Tantilla berguidoi, recently described and endemic (Batista et al. 2016b); (G) Bothrops asper, 1,273 m asl,

highest elevation record for Panama; (H) Lachesis acrochorda juvenile, 1,011 m asl, highest elevation for this species in Panama.

Amphib. Reptile Conserv. 35 May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

Amphibians

30 25 obs.

29 est.

Species richness

Premontane Moist Forest

Premontane Wet Forest

io 8 F @ dlei315 17 19 21 23

Transects

30) Reptiles

25 est.

Species richness

13 5 7 9 11 13 15 17 19 21 23

Transects

Fig. 6. Species accumulation curves for amphibians (left) and reptiles (right) for transects in the PMF and PWF on Cerro Chucanti.

obs: observed species on transects; est: estimated species if the sampling effort is doubled. Details on the transects are shown in

Table 1.

Premontane Moist Forest

Diasporus diastema**

Pristimantis taeniatus*

Craugastor raniformis**

Cochranella granulosa*

Dermophis sp.

Smilisca phaeota

Boana rosenbergi

Diasporus majeensis

Craugastor opimus

Rhinella horribilis

Caecilia isthmica

Leptodactylus savagei

Allobates talamancae

Agalychnis callidryas

Colostethus aff. pratti

Pristimantis pardalis

Pristimantis aff. ridens

Engystomops pustulosus

Dendrobates auratus

Hyalinobatrachium colymbiphyllum

Craugastor fitzingeri

Silverstoneia sp.

Pristimantis gaigei

FAyalinobatrachium chirripoi

Pristimantis cruentus

Rhinella alata

Strabomantis bufoniformis

Espadarana prosoblepon

Smilisca sila*

Rhaebo haematiticus

0.000 0.050 0.100 0.150

Relative abundance

Fig. 7. Relative abundance of amphibians found during

surveys from 2012-2016 in the Premontane Moist Forest

(PMF) on Cerro Chucanti. * Species observed in PWF and

PMEF on Cerrro Chucanti, but outside a transect; ** Species of

likely occurrence that have been found in similar habitat and

elevations in adjacent peaks of Majé Mountains.

maximum-estimated species richness in 2007 was

21-obs/26-est (PMF) and 13-obs/17-est (PWF) species,

which is 6—7 fewer observed species than were reported

for the same elevation herein (25-obs/29-est, respectively

20-obs/22-est species). The difference may be attributed

to the 30% greater survey time in this study, or to the

erratic (although not recorded) weather fluctuations

affecting amphibian activity in the respective study

Amphib. Reptile Conserv.

Premontane Wet Forest

Smilisca sila

Gastrotheca nicefori*

Pristimantis pardalis

Pristimantis cerasinus

Craugastor aff. longirostris*

Bolitoglossa aff. biseriata

Bolitoglossa chucantiensis

Pristimantis gaigei

Oedipina aff. complex

Craugastor crassidigitus

Silverstoneia sp.

Fyalinobatrachium chirripoi

Cochranella euknemos

Rhinella alata

Rhaebo haematiticus

Pristimantis moro

Pristimantis aff. latidiscus

Colostethus aff. pratti

Espadarana prosoblepon

Pristimantis cruentus

Diasporus majeensis

Pristimantis caryophyllaceus

0 0.1 0.2 0.3

Relative abundance

Fig. 8. Relative abundance of amphibians found during surveys

from 2012-2016 in the Premontane Wet Forest (PWF) on Cerro

Chucanti. * Species observed in PWF and PMF on Cerrro

Chucanti, but outside a transect.

periods. No comparative data are available for reptiles

in the 2007 survey (Medina et al. 2019), however, we

would expect an increase of nine and seven species for

the PMF and PWE, respectively.

As aresult, from the first expedition to Cerro Chucanti

in 2012, we have described Diasporus majeensis,

Bolitoglossa chucantiensis, and Tantilla berguidoi. In

addition, several collected specimens represent potentially

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Premontane Moist Forest

Porthidium lansbergii*

Bothriechis schlegelii*

Bothrops asper

Sibon nebulatus**

Rhadinaea decorata

Oxyrhopus petolarius

Pliocercus euryzonus

Spilotes pullatus*

Drymobius margaritiferus*

Dendrophidion percarinatum

Corallus annulatus

Holcosus leptophrys*

Holcosus festivus

Sphaerodactylus lineolatus*

Lepidoblepharis sanctaemartae

Iguana iguana**

Leposoma southi**

Anolis apletophallus**

Anolis biporcatus

Anolis aff. fuscoauratus

Anolis frenata

Corytophanes cristatus**

Thecadactylus rapicauda

Anolis vittigerus

Leptodeira ornata

Imantodes cenchoa

Basiliscus basiliscus

0.000 0.050 0.100 0.150 0.200

Relative abundance

0.250

Fig. 9. Relative abundance of the reptiles found during

surveys from 2012-2016 in the Premontane Moist Forest

(PMF) on Cerro Chucanti. * Species observed in PWF and

PMF on Cerrro Chucanti, but outside a transect; ** Species of

likely occurrence that have been found in similar habitat and

elevations in adjacent peaks of Majé Mountains.

new species of amphibians (Figs. 3-4): Oecedipina

aff. complex, Colostethus aff. pratti, Silverstoneia

sp., Pristimantis aff. latidiscus, and Pristimantis aff.

ridens, and reptiles (Fig. 5): Echinosaura aff. palmeri,

Ptychoglossus aff. plicatus, Anolis aff. fuscoauratus, and

Geophis aff. brachycephalus. Currently, an integrative

taxonomy approach (de Queiroz 2007) is being applied

to determine the taxonomic status or relationships of

these specimens. The more recent expeditions in 2016

have also provided new material, e.g., Dermophis aff.

glandulosus and Bolitoglossa aff. biseriata, that was

assessed only morphologically, but for which molecular

results are not yet available.

Several aspects call for urgent conservation strategies

to preserve this high and important diversity hot spot in

Eastern Panama, including the endemism of species and/

or isolated populations reported for the Majé Mountains,

and in particular for Cerro Chucanti’s CPNR (Batista et

al. 2014a,b, 2016a,b), the accelerated deforestation rate

observed for the region in recent years (G. Berguido

pers. comm.), and the latent threat due to climate

change (Marchese 2015). The conservation status of the

newly described species from Cerro Chucanti and those

awaiting description are unknown, whereas others were

listed according to IUCN (2016) in Appendix 1.

This survey spent 147 man/hr in the Premontane

Wet Forest around the type locality of Bolitoglossa

chucantiensis, but only two specimens were found. This

salamander is rare or occurs at low density, and is probably

Amphib. Reptile Conserv.

Premontane Wet Forest

Lachesis acrochorda*

Bothrops asper

Sibon lamari ' *

Pliocercus euryzonus

Leptodeira ornata

Tantilla berguidoi

Oxybelis brevirostris**

Dendrophidion percarinatum

Enyalioides heterolepis**

Ptychoglossus festae

Echinosaura aff. palmeri

Corytophanes cristatus**

Imantodes cenchoa

Geophis aff. brachycephalus

Ptychoglossus aff. plicatus

0 0.1 0.2 0.3

Relative abundance

0.4 0.5

Fig. 10. Relative abundance of the reptiles found during

surveys from 2012—2016 in the Premontane Wet Forest (PWF)

on Cerro Chucanti. ' indicates provisional identification; *

Species observed in PWF and PMF on Cerrro Chucanti, but

outside a transect; ** Species of likely occurrence that have

been found in similar habitat and elevations in adjacent peaks

of Majé Mountains.

mg Amphibia

@ Reptilia

{= oO) oe)

Average species/transect

Aquatic Terrestrial

Habitat

Fig. 11. Species richness across habitats, as average species per

transect (#species/transect).

restricted to the peak area of Cerro Chucanti. The Dink

Frog (Diasporus majeensis) is common, but as with the

salamander, it has been reported only from the restricted

area above 1,300 m asl. Our decade-long study of these

two genera in most of Eastern Panama, combined with

intensive surveys elsewhere in the Majé Mountains (e.g.,

around Ambroya), as well as most other cloud forests of

Eastern Panama, lead us conclude that these two species

are endemic to Cerro Chucanti (Batista et al. 2014a,

2016a). Since they are each restricted to a small area of

less than 5 km? of a high elevation habitat around the peak

May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

of Cerro Chucanti, we can argue for their classification

as Critically Endangered, applying the IUCN Red List

2012]. Anthropogenic pressure around Cerro Chucanti

probably will lead to further declines of populations

through habitat deterioration and reduction of the area of

occupancy. These species will also be allocated to “high”

levels of environmental vulnerability score (EVS) by

Johnson et al. (2015).

Furthermore, any outbreak of the fungal disease

caused by Batrachochytrium dendrobatidis (Bd) and/

or in combination with climate change is an imminent

threat to the amphibians of Cerro Chucanti. Although

Bd has been reported from the lowlands nearby (King et

al. 2014; Rebollar et al. 2014) and through a cascading

mechanism has negatively affected the reptilian fauna

in Panama (Zipkin et al. 2020), thus far, no amphibian

population decline has been detected on Cerro Chucanti.

However, the threat is latent, as Bd has already expanded

across most of Panama (Lips et al. 2008; Rebollar et al.

2014) and can be expected to affect Cerro Chucanti in

the foreseeable future. Hence, bio-security protocols

must be applied for visitors to the CPNR reserve (Dood

2010) and Bd surveys are urgently required. Since

slight temperature fluctuations due to climate change

could affect the biology of amphibian and reptile

species on Cerro Chucanti, we also recommend that

thermoecological experiments be conducted within the

habitat of the endemic mountain species.

Despite the sampling effort made in this and previous

studies, acaveat aboutthe true species richness remains due

to uncertainties regarding species detectability. Despite

the combined sampling effort conducted throughout the

two studies (Medina et al. 2019 and herein: April, June,

July, October, and December), neither study followed a

systematic and prolonged sampling method, and they only

included visual encounter and opportunistic surveys

for a few short duration excursions. For example, this

sampling could easily have missed amphibian species

with an explosive and temporally limited reproductive

mode (e.g., Hyloscirtus colymba or H. palmeri). Many

fossorial snakes (e.g., Atractus, Micrurus), \izards

(e.g., Bachia, Diplogossus, Lepidoblepharis), frogs

(e.g., Strabomantidae group), and caecilians (e.g.,

Dermophis, Oscaecilia) are difficult to detect in the

leaf litter and require more elaborate techniques, such

as pitfall trapping, to record and properly evaluate their

population status. Species living in the canopy which

rarely come down to the ground (e.g., Cruziohyla

calcarifer, Ecnomiohyla sp.) are usually missed during

short inventories unless they are recorded calling like

the Marsupial Frog, Gastrotheca nicefori (for which

a short calling sequence from this study is accessible

at: https://youtu.be/H4QKsMQDWVQ). Therefore,

to achieve a broader and more comprehensive insight

into the herpetofauna diversity on Cerro Chucanti,

additional studies must increase seasonally repetitive

Amphib. Reptile Conserv.

surveys, canopy inspections, and pitfall trapping.

Nonetheless, the studies conducted on Cerro Chucanti

thus far have increased public awareness through the

publication of various articles and online promotion to

increase the value and enlarge the protection of the “Cerro”

Chucanti Private Nature Reserve and its surroundings. In

this context, Rainforest Trust has purchased three legally

titled properties in order to establish an important buffer

zone that will act as a barrier for preventing squatters

from moving into the extensive public wilderness areas,

and will discourage poachers from hunting in the vicinity

(http://www. rainforesttrust.org/expansion-of-the-cerro-

chucanti-nature-reserve/). The land to be purchased is

part of the very limited high-elevation cloud forest where

many new species have been discovered, in particular

amphibians and reptiles described by the authors. As

a gateway to over 60,000 acres of public lands, Cerro

Chucanti Nature Reserve lays the foundation for the

designation of a government national park, an effort that

Panamanian NGO ADOPTA (http://adoptabosque.org/)

is working hard to achieve.

Acknowledgements.—With this article, we hope to assist

important organizations, in particular Guido Berguido

with adoptabosque.org, in generating attention for the

preservation of this unique and beautifully diverse

paradise, the Chucanti Private Nature Reserve. We also

would like to thank Jesus Pérez, Benjamin, Luis de Leon,

and Sr. Juan Zarzavilla from Rio Pavo for field assistance.

Literature Cited

ANAM (Autoridad Nacional del Ambiente). 2010. At/as

Ambiental de la Republica de Panama. Autoridad

Nacional del Ambiente, Panama, Panama. 190 p.

Aubrecht R, Barrio-Amoréds CL, Breure ASH, Brewer-

Carias C, Derka T, Fuentes-Ramos OA, Gregor

M, Kodada J, Kovacik L, Lanczos T, et al. 2012.

Venezuelan Tepuis: Their Caves and Biota. Acta

Geologica Slovaca Monograph. Comenius University,

Bratislava, Slovakia. 168 p.

Batista A, Ponce M. 2002. Abundancia, riqueza y

distribucion de especies de anfibios en el Distrito de

Mirono, Comarca Ngdbe Buglé, Panama. Tesis de

Licenciatura, Universidad Autonoma de Chiriqui,

David, Chiriqui, Panama.

Batista A, Kohler G, Mebert K, Vesely M. 2014a. A new

species of Bolitoglossa (Amphibia: Plethodontidae)

from eastern Panama, with comments on other species

of the adspersa species group from eastern Panama.

Mesoamerican Herpetology 1: 97-121.

Batista A, Hertz A, Kohler G, Mebert K, Vesely M. 2014b.

Morphological variation and phylogeography of

frogs related to Pristimantis caryophyllaceus (Anura:

Terrarana: Craugastoridae) in Panama. Salamandra

50: 155-171.

Batista A, Hertz A, Mebert K, Kohler G, Lotzkat S,

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Ponce M, Vesely M. 2014c. Two new fringe-limbed

frogs of the genus Ecnomiohyla (Anura: Hylidae)

from Panama. Zootaxa 3826: 449-474.

Batista A, Ponce M, Vesely M, Mebert K, Hertz

A, Kohler G, Lotzkat S. 2015. Revision of the

genus Lepidoblepharis (Reptilia: | Squamata:

Sphaerodactylidae) in Central America, with the

description of three new species. Zootaxa 3994: 187—

221.

Batista A, Kohler G, Mebert K, Hertz A, Vesely M.

2016a. An integrative approach to reveal speciation

and species richness in the genus Diasporus

(Amphibia: Anura: Eleutherodactylidae) in eastern

Panama. Zoological Journal of the Linnean Society

178(2): 267-311.

Batista A, Mebert K, Lotzkat S, Wilson LD. 2016b. A

new species of centipede snake of the genus Zantilla

(Squamata: Colubridae) from an isolated premontane

forest in eastern Panama. Mesoamerican Herpetology

3(4): 949-960.

Chavez G, Morales V. 2014. Evaluacion de la riqueza

de anfibios y reptiles de la Reserva comunal E] Sira.

Pp. 34-35 In: CoGAP. Proyecto Cogestion Amazonia.

Transecto Altitudinal Yuyapichis: Monitoreo de

la Biodiversidad y del Cambio Climdatico-FM 4.

GIZ (Deutsche Gesellschaft fir Internationale

Zusammenarbeit), Bonn, Germany and _ Servicio

Nacional de Areas Naturales Protegidas por el Estado

(SERNANP), Lima, Peru.

Coates AG, Collins LS, Aubry MP, Berggren WA. 2004.

The geology of the Darién, Panama, and the late

Miocene-Pliocene collision of the Panama Arc with

northwestern South America. Geological Society of

American Bulletin 116: 1,327—1,344.

Colwell RK. 2006. EstimateS. Statistical Estimation of

Species Richness and Shared Species from Samples.

Robert K. Colwell, Boulder, Colorado, USA.

Available: http://purl.oclc.org/estimates [Accessed:

22 March 2019].

Davies KF, Margules CR, Lawrence JF. 2004. A

synergistic effect puts rare, specialized species at

greater risk of extinction. Ecology 85: 265-271.

Dodd CK. (Editor). 2010. Amphibian Ecology and

Conservation: a Handbook of Techniques. Oxford

University Press, Oxford, United Kingdom. 585 p.

Dodge N. 1943. Monument in the mountain. Arizona

Highways 19(3): 20-28.

Duellman WE. 1988. Patterns of species diversity in

anuran amphibians in the American tropics. Annals of

the Missouri Botanical Garden 75: 79-104.

Haines ML, Stuart-Fox D, Sumner J, Clemann N,

Chapple DG, Melville J. 2017. A complex history of

introgression and vicariance in a threatened montane

skink (Pseudemoia cryodroma) across an Australian

sky island system. Conservation Genetics 18: 939-

950.

Hammer 0, Harper DAT, Ryan PD. 2001. PAST:

Amphib. Reptile Conserv.

Paleontological _ statistics

education and data

Electronica 4(1): 1-9.

Heald W. 1967. Sky Island. Van Nostrand, Princeton,

New Jersey, USA. 166 p.

Hertz A. 2015. Integrative taxonomy and conservation

status of amphibians in western Panama with an

emphasis on the highlands of the Cordillera Central.

Doctoral Dissertation, Goethe University, Frankfurt

am Main, Germany. 294 p.

IUCN. 2012. IUCN Red List Categories and Criteria:

Version 3.1. Second Edition. YUCN, Gland,

Switzerland and Cambridge, United Kingdom. tv +

32 p.

IUCN. 2016. IUCN Red List of Threatened Species.

Version 2016.2 [Online]. Available: http://www.

iucnredlist.org [Accessed: 27 October 2019].

Johnson JD, Mata-Silva V, Wilson LD. 2015. A

conservation reassessment of the Central American

herpetofauna based on the EVS measure. Amphibian

& Reptile Conservation 9(2) [General Section]: 1-94

(e100).

Kung D, Bigler L, Davis LR, Gratwicke B, Griffith

E, Woodhams DC. 2014. Stability of microbiota

facilitated by host immune regulation: informing

probiotic strategies to manage amphibian disease.

PLoS One 91): e87101.

Lips KR, Diffendorfer J; Mendelson JR III, Sears MW.

2008. Riding the wave: reconciling the roles of

disease and climate change in amphibian declines.

PLOS Biology 6(3): e72.

Lotzkat S. 2015. Diversity, taxonomy, and biogeography

of the reptiles inhabiting the highlands of the

Cordillera Central (Serrania de Talamanca and

Serrania de Tabasara) in western Panama. Doctoral

Dissertation, Goethe University, Frankfurt am Main,

Germany. 931 p.

Marchese C. 2015. Biodiversity hotspots: a shortcut for

a more complicated concept. Global Ecology and

Conservation 3: 297-309.

Medina D, Ibafiez R, Lips KR, Crawford AJ. 2019.

Amphibian diversity in Serrania de Majé, an isolated

mountain range in eastern Panama. ZooKeys 859:

117-130.

Mizsei E, Szabolcs M, Szab6 L, Boros Z, Mersini Kk,

Roussos SA, Dimaki M, Ioannidis Y, Végvari Z,

Szabolcs L. 2020. Determining priority areas for

an endangered cold-adapted snake on warming

mountaintops. Oryx, in press.

Myers CW. 2003. Rare snakes. Five new species from

eastern Panama: reviews of northern Atractus and

southern Geophis (Colubridae: Dipsadinae). American

Museum Novitates 3391: 1-47.

Pecl GT, Araujo MB, Bell JD, Blanchard J, Bonebrake

TC, Chen I-C, Clark TD, Colwell, RK, Danielsen F,

Evengard B, et al. 2017. Biodiversity redistribution

under climate change: impacts on ecosystems and

software package for

analysis. Palaeontologia

May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

human well-being. Science 355: eaai9214.

Quammen D. 2004. The Song of the Dodo: Island

Biogeography in an Age of Extinctions. Scribner, New

York, New York, USA. 702 p.

Ramirez CA. 2003. Estado de la Diversidad Bioldgica

de los Arboles y Bosques de Panama. Documentos

de Trabajo: Recursos Genéticos Forestales FGR/S0S.

Servicio de Desarrollo de Recursos Forestales,

Direccion de Recursos Forestales, FAO, Rome, Italy.

Available: http://www.fao.org/3/j0604s/j0604s00. htm

[Accessed: 3 May 2020].

Rebollar EA, Hughey MC, Harris RN, Domangue RJ,

Medina D, Ibafiez R, Belden LK. 2014. The lethal

fungus Batrachochytrium dendrobatidis is present in

lowland tropical forests of far eastern Panama. PLoS

One 9(4): e95484.

Samudio R. 2001. Panama. Pp. 371-396 In: Bosques

Nublados del Neotrdpico. Editors, Kappelle M,

Brown AD. INBio, Heredia, Costa Rica. 698 p.

Sekercioglu CG, Schneider SH, Fay JP, Loarie SR. 2008.

Climate change, elevational range shifts, and bird

extinctions. Conservation Biology 22: 140-150.

Sinervo B, Méndez-de-la-Cruz F, Miles DB, Heulin B,

Bastiaans E, Villagran-Santa Cruz M, Lara-Resendiz

R, Martinez-Méndez N, Calderon-Espinosa ML,

Meza-Lazaro RN, et al. 2010. Erosion of lizard

diversity by climate change and altered thermal

niches. Science 328: 894-899.

Veith M, Lotters S, Andreone F, Rédel MO. 2004.

amphibians in captivity.

groups such as scorpions.

Amphib. Reptile Conserv.

Measuring and monitoring amphibian diversity in

tropical forests. II. Estimating species richness from

standardized transect censing. Ecotropica 10(2): 85—

99.

Warshall P. 1994. The Madrean Sky Island Archipelago:

A planetary overview. Pp. 6-18 In: Biodiversity

and Management of the Madrean Archipelago.

The Sky Islands of Southwestern United States and

Northwestern Mexico. General Technical Report

RM 264. United States Forest Service, Fort Collins,

Colorado, USA. 669 p.

Wilson LD, Townsend JH, Johnson JD. 2010.

Conservation of Mesoamerican Amphibians and

Reptiles. Eagle Mountain Publishing, Eagle Mountain,

Utah, USA. 812 p.

Whitfield SM, Lips KR, Donnelly MA. 2016. Amphibian

decline and conservation in Central America. Copeia

104: 351-379.

World Wildlife Fund. 2014. Eastern Panamanian Montane

Forests [Online]. Available: http:/www.eoearth.org/

view/article/151914 [Accessed: 18 March 2019].

Yeo JJ, Peterson CR. 1998. Amphibian and Reptile

Distribution and Habitat Relationships in the Lost

River Mountains and Challis-Lemhi Resource Areas.

Bureau of Land Management, Idaho State Office,

Boise, Idaho, USA. 40 p.

Zipkin EF, DiRenzo GV, Ray JM, Rossman S, Lips

KR. 2020. Tropical snake diversity collapses after

widespread amphibian loss. Science 367(6479): 814—

816.

Orlando Ariel Garcés R. is a graduate Biologist with orientation in Animal Biology, and interests in

herpetology, conservation, taxonomy, exploration, and photography. Orlando is currently a member

of the Mesoamerican and Caribbean Network for the Conservation of Amphibians and Reptiles,

as well as the Mesoamerican Society for Biology and Conservation, chapter in Panama. He is

experienced in monitoring amphibians and reptiles, as well as the conservation and management of

Rogemif Daniel Fuentes Magall6én is a Zoologist who is passionate about nature and neotropical

herpetology, and graduated from the University of Panama. With interests in research, rescue, and

conservation, he currently collaborates on projects primarily involving amphibians and ophidians,

and actively participates in wildlife rescue and relocation projects across the country. Rogemif has

previously worked with anurans, snakes, and crocodiles, and collaborated on research with other

Abel Batista is a Biologist by profession and nature lover, who completed a master's degree at

the Universidad de los Andes Bogota, Colombia, and a doctorate at the Senckenberg Institute (in

association with Goethe University), Frankfurt, Germany, both focused on the study of amphibians

and reptiles in Panama. With 20 years of field experience in Panama, Costa Rica, and Colombia,

Abel has carried out various fauna rescue works, monitoring, and research. His main interests are

_ bioacoustics, interactions between anuran communities, and the biogeography, phylogeny, and

taxonomy of amphibians in Panama.

Madian Pamela Miranda is a graduate from the Autonomous University of Chiriqui, Panama, where

she obtained her Bachelor's Degree in Biology, and she is a biologist by profession and nature lover

§ with seven years of field experience in Panama. She has recently joined the Los Naturalistas (NGO)

; team. Madian is passionate about amphibians and reptiles, and has collaborated on various amphibian

and reptile monitoring and conservation projects. She currently participates in environmental

consulting projects, expeditions in search of new species, and conservation.

May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Marcos Ponce, Lic., is a Zoologist who has been working with amphibians in Panama for 15 years,

and has participated in the descriptions of more than 10 new species of amphibians from Panama.

Marcos has visited some of the less-explored places in Panama, and currently is the head of a

herpetological group that is evaluating the biodiversity of different national parks in Panama.

Konrad Mebert is an independent researcher and international project coordinator based in

Switzerland, who conducts studies globally on amphibians and reptiles in general, but with an

emphasis on vipers for the past 15 years. After completing a Master's degree at the University of

Zurich, Switzerland, on geographic variation and the effects of inbreeding on the Dice Snake, and

a doctoral degree on hybrid zones in North American water snakes at Old Dominion University,

Virginia, USA, he currently is associated with the State University of Santa Cruz, Ilhéus,

Bahia, Brazil and the IDECC (Institute of Development, Ecology, Conservation and Cooperation) in

Italy. He has produced more than 140 professional and popular publications/reports, and two books

on water snakes, covering a range of topics on evolution, ecology, biodiversity, and conservation.

Many expeditions and a passion of photography have led him to all continents except Australia, and

his current (2019) principal study sites are located in Panama, Turkey, Brazil, Slovenia, Ecuador,

and China.

Appendix 1. Amphibians and reptiles recorded from the Chucanti Private Nature Reserve (and adjacent peaks of Majé Mountains)

and their conservation status. NE indicates Not Evaluated, with other abbreviations for IUCN categories: DD (Data Deficient),

LC (Least Concern), NT (Near Threatened), VU (Vulnerable), and EN (Endangered). Abbreviations for EVS (environmental

vulnerability score, sensu Johnson et al. 2015) are L (Low: scores from 3—9), M (Medium: scores from 10—13), H (High: scores

from 14-19), and N (Not scored). Two species received an additional classification in the Panama Conservation Status system: VU

for Dendrobates auratus and CR for Sibon lamari. Red List status designations of potentially new species are listed according to

the species of their morphological affinity, and labeled with a respective ‘aff.’ and epithet. * = Species of likely occurrence that have

been found in similar habitat and elevations in the adjacent peaks of Majé Mountains.

Scientific name DE A pe

status

Class Amphibia (41 species)

Order Caudata (3 species)

Family Plethodontidae

Bolitoglossa chucantiensis NE H

Bolitoglossa aff. biseriata NE NE

Oedipina aff. complex NE NE

Order Gymnophiona (2 species)

Family Caeciliidae

Caecilia isthmica DD H

Dermophis aff. glandulosus NE NE

Order Anura (36 species)

Family Bufonidae (3 species)

Rhaebo haematiticus LC L

Rhinella horribilis LC

Rhinella alata LC Ly

Family Centrolenidae (5 species)

Cochranella euknemos LC L

Cochranella granulosa LC L

Amphib. Reptile Conserv. 41 May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

Appendix 1 (continued). Amphibians and reptiles recorded from the Chucanti Private Nature Reserve (and adjacent peaks of

Majé Mountains) and their conservation status. NE indicates Not Evaluated, with other abbreviations for IUCN categories: DD

(Data Deficient), LC (Least Concern), NT (Near Threatened), VU (Vulnerable), and EN (Endangered). Abbreviations for EVS

(environmental vulnerability score, sensu Johnson et al. 2015) are L (Low: scores from 3—9), M (Medium: scores from 10-13),

H (High: scores from 14—19), and N (Not scored). Two species received an additional classification in the Panama Conservation

Status system: VU for Dendrobates auratus and CR for Sibon lamari. Red List status designations of potentially new species are

listed according to the species of their morphological affinity, and labeled with a respective ‘aff.’ and epithet. * = Species of likely

occurrence that have been found in similar habitat and elevations in the adjacent peaks of Majé Mountains.

Scientific name ST Ee

status

Espadarana prosoblepon LC

Fyalinobatrachium chirripoi LC M

Hyalinobatrachium colymbiphyllum LC L

Family Craugastoridae (15 species)

Craugastor crassidigitus LC M

Craugastor fitzingeri LC L

Craugastor aff. longirostris LC M

Craugastor opimus EC M

Craugastor raniformis* Le L

Pristimantis caryophyllaceus LC L

Pristimantis cerasinus Le M

Pristimantis cruentus LC L

Pristimantis gaigei LC M

Pristimantis moro LC L

Pristimantis pardalis NT H

Pristimantis taeniatus Le |

Pristimantis aff. latidiscus NE NE

Strabomantis bufoniformis LC M

Pristimantis aff. ridens LC M

Family Dendrobatidae (4 species)

Allobates talamancae Le M

Colostethus aff. pratti LC M

Dendrobates auratus LC L

Silverstoneia sp. NE

Family Eleutherodactylidae (2 species)

Diasporus diastema* LC L

Diasporus majeensis NE H

Family Hylidae (5 species)

Gastrotheca aff. nicefori NE

Boana rosenbergi LC L

Smilisca phaeota LC L

Smilisca sila LC L,

Agalychnis callidryas LC L

Family Leptodactylidae (1 species)

Leptodactylus savagei LC L

Family Leiuperidae (1 species)

Engystomops pustulosus LC L

Amphib. Reptile Conserv. 42 May 2020 | Volume 14 | Number 2 | e237

Batista et al.

Appendix 1 (continued). Amphibians and reptiles recorded from the Chucanti Private Nature Reserve (and adjacent peaks of

Majé Mountains) and their conservation status. NE indicates Not Evaluated, with other abbreviations for IUCN categories: DD

(Data Deficient), LC (Least Concern), NT (Near Threatened), VU (Vulnerable), and EN (Endangered). Abbreviations for EVS

(environmental vulnerability score, sensu Johnson et al. 2015) are L (Low: scores from 3—9), M (Medium: scores from 10-13),

H (High: scores from 14-19), and N (Not scored). Two species received an additional classification in the Panama Conservation

Status system: VU for Dendrobates auratus and CR for Sibon lamari. Red List status designations of potentially new species are

listed according to the species of their morphological affinity, and labeled with a respective ‘aff.’ and epithet. * = Species of likely

occurrence that have been found in similar habitat and elevations in the adjacent peaks of Majé Mountains.

Scientific name DT Ee

status

Class Reptilia

Order Squamata (35 species)

Family Corytophanidae (2 species)

Basiliscus basiliscus LC

Corytophanes cristatus* LC M

Family Dactyloidae (5 species)

Anolis frenata NE H

Anolis apletophallus* LC H

Anolis biporcatus LC L

Anolis fuscoauratus NE M

Anolis vittigerus LC H

Family Gymnophthalmidae (4 species)

Echinosaura aff. palmeri DD M

Leposoma southi* LC H

Ptychoglossus aff. plicatus NE H

Ptychoglossus festae LC

Family Hoplocercidae (1 species)

Enyalioides heterolepis* LC M

Family Iguanidae (1 species)

Iguana iguana* LC M

Family Mabuyidae (1 species)

Marisora unimarginata* LC H

Family Phyllodactylidae (1 species)

Thecadactylus rapicauda LC L

Family Sphaerodactylidae (2 species)

Lepidoblepharis sanctaemartae LC H

Sphaerodactylus lineolatus LC H

Family Teiidae (2 species)

Holcosus festivus LC M

Holcosus leptophrys LC H

Family Boidae (1 species)

Corallus annulatus DD M

Family Colubridae (4 species)

Dendrophidion percarinatum LC

Drymobius margaritiferus LC IL

Oxybelis brevirostris* LC M

Spilotes pullatus LC L

Tantilla berguidoi NE M

Amphib. Reptile Conserv. 43 May 2020 | Volume 14 | Number 2 | e237

Endemism on a Threatened Sky Island

Appendix 1 (continued). Amphibians and reptiles recorded from the Chucanti Private Nature Reserve (and adjacent peaks of

Majé Mountains) and their conservation status. NE indicates Not Evaluated, with other abbreviations for IUCN categories: DD

(Data Deficient), LC (Least Concern), NT (Near Threatened), VU (Vulnerable), and EN (Endangered). Abbreviations for EVS

(environmental vulnerability score, sensu Johnson et al. 2015) are L (Low: scores from 3—9), M (Medium: scores from 10—13),

H (High: scores from 14-19), and N (Not scored). Two species received an additional classification in the Panama Conservation

Status system: VU for Dendrobates auratus and CR for Sibon lamari. Red List status designations of potentially new species are

listed according to the species of their morphological affinity, and labeled with a respective ‘aff.’ and epithet. * = Species of likely

occurrence that have been found in similar habitat and elevations in the adjacent peaks of Majé Mountains.

Scientific name See

status

Family Dipsadidae (8 species)

Imantodes cenchoa LC L

Leptodeira ornata LC L

Pliocercus euryzonus LC M

Oxyrhopus petolarius LC M

Rhadinaea decorata Le L

Sibon lamari EN H

Sibon nebulatus* LC L

Geophis aff. brachycephalus

Family Viperidae (4 species)

Bothrops asper LC M

Lachesis acrochorda DD H

Bothriechis schlegelii LC M

Porthidium lansbergii NE H

NE=8 NE=5

LC =31 LC =27

VU=1 EN = 1

DD=1 DD =3

EVS Amphibians EVS_ Reptiles

L= 20 L=8

M=3 M=14

H= 3 H=12

Amphib. Reptile Conserv. 44 May 2020 | Volume 14 | Number 2 | e237

Batista et al.

00: cOT

L10

190

eZ I

ve 0

9IV

008¢c 000¢

IcO scO

p90 CLO

vil OTT

6£€ 0 6€ 0

S8l vv

9 ¢

910C

00 1S 008 007

eo 0 €9°0 001

S06 0 €LL0 C968 0

€86C svc | cv I

8c£90 0 6L€ 0 Eee 0

[81 Ic 9

LT ¢ v

00 OI 008 00 Ic

090 SLO LS 0

89 0 060 98 0

cc I I9'T Sic.

9€0 CC 0 S10

vv vv vL

9 9 cl

1S9.10,] 194 JULITOWIIAg

YUNdD Je S}so10F JO sodA} OM} OY} UIYIIM S}OOsSUBI] SUOTS PUNO; suviqiyduwe Jo ssouyos sordods pue soxopul AjISIOAIC *Z xIpusoddy

910C

JSI1OW JSIOP, VUBJNOUIIIg

006

001

06 0

86 |

910

[Ee

SOC

00 CI

SLO

£80

981

610

00 VC

850

08 0

ole

910

009

L90

€c6 0

8c |

CLOC

006

8Z I

6076 0

ESS

8860 0

9€

S10C

009

Lol

98 0

L6 |

910

00 OI

O€ 1

L9Vv6 0

8CV Cc

8600

CLO?

yoyo Surjdues

SSOUYIII SANLIOY

f Aupigennby

H uouueys

q doueulW0g

s[enprAlpu]

S PXxe],

JOOSUBIT,

Teoh

yoyo Surjdues

SSOUYII SATILOY

f Ariqenby

H uouueys

q 9ouvulWiog

s[enprarpuy

S bXxP],

yOOsUBIT,

Teo

May 2020 | Volume 14 | Number 2 | e237

Amphib. Reptile Conserv.

Endemism on a Threatened Sky Island

I¢ 8 4 ‘a 9 9 9 6 Ol yoyye surjdures

1€0 E10 S70 00.1 ooo LTO €£0 8/0 001 SSOUYOI SATC[OY

760 00'0 000 760 000 00'I 760 001 ¢ Aupiqeynby

SST 00'0 000 790 000 69'0 €8'1 OT H uouueys

O10 00'1 00'I 9¢°0 00'I 0s'0 610 Ome) qd souvurlwog

LZ I I € I Z 6 Ol s[enplArpuy

S exe],

JOOSUBL

9107 STOZ Z10Z TeaK

JSXIOY JSIOJA, VUBJUOWIAIG

4 Ol 8 IZ 6 ral VT 8 9 Ol yoyjo surjdures

SL0 070 E10 O10 ITO L10 S70 S70 €€£0 070 SSOUYOI DATIL[OY

L830 00'1 00'0 980 000 L60 760 180 00'T 760 ¢ Ayyiqeynby

660 690 00'0 09'0 00°0 L9'0 89'| 9¢°0 69:0 790 H uouueys

rr 0s 0 001 650 00'1 750 170 €9'0 0s 0 9¢°0 q souvulw0g

¢ Z I L Z ¢ 6 4 j € sjenprAIpuy

€ ‘G I j I Z 9 Z Z z S exel,

JOOSUBLT,

SIOZ Z10Z TeX

189.10, J9AQ DUBJUOWIIAg

YNdD 1e S}sosOF JO SadA} OM} DY} UTYJIM S}OOSUBI] SUOTe PUNO} SofIWdoI Jo ssouYoII sardads pue soxapul AVISIOAIG *¢ xIpusddy

May 2020 | Volume 14 | Number 2 | e237

Amphib. Reptile Conserv.

Amphibian & Reptile Conservation

14(2) [General Section]: 47-63 (e238).

Official journal website:

amphibian-reptile-conservation.org

urn:lsid:zoobank.org:pub:C255BCBD-BEF5-444C-BD05-F2F16C9828FC

An endemic new species of Andean lizard of the genus

Liolaemus from southern Peru (Iguania: Liolaemidae) and its

phylogenetic position

12.*Juan C. Chaparro, *Aaron J. Quiroz, '?Luis Mamani, *Roberto C. Gutiérrez,

'2Peter Condori, ‘Ignacio De la Riva, ‘*Gabriela Herrera-Juarez, “José Cerdena,

3Luis P. Arapa, and °*Cristian S. Abdala

'Museo de Biodiversidad del Peru, Urbanizacién Mariscal Gamarra A-61, Zona 2, Cusco, PERU *Museo de Historia Natural de la Universidad

Nacional de San Antonio Abad del Cusco, Paraninfo Universitario (Plaza de Armas s/n), Cusco, PERU *Museo de Historia Natural. Universidad

Nacional de San Agustin de Arequipa, Av. Alcides Carrion s/n, Arequipa, PERU *Department of Biodiversity and Evolutionary Biology, Museo

Nacional de Ciencias Naturales, CSIC, C/José Gutiérrez Abascal, 2, 28006, Madrid, SPAIN °Consejo Nacional de Investigacién Cientificas y

Técnicas (CONICET)—Unidad Ejecutora Lillo (UEL), San Miguel de Tucuman, ARGENTINA °Facultad de Ciencias Naturales e Instituto Miguel

Lillo (IML), Universidad Nacional de Tucumdn, San Miguel de Tucuman, ARGENTINA

Abstract.—Integrative evidence of several external morphological characters and molecular phylogenetic

analyses of mitochondrial DNA (12S, cyt-b) are used to place a new species of Andean lizard of the genus

Liolaemus (|Iguania: Liolaemidae) in the Liolaemus montanus group and as sister group of the clade formed

by L. signifer. The new species is characterized by a unique combination of morphometric characteristics,

scalation, and color pattern. The L. montanus group now contains seventeen species in southern Peru,

distributed along the eastern and western slopes of the Andes.

Keywords. Andes, Apurimac, Eu/aemus, Puna, reptile, systematics, taxonomy

Resumen.—utilizamos evidencia integradora de varios caracteres morfoldgicos externos y analisis filogenéticos

moleculares de ADN mitocondrial (12S, cyt-b) que ubican una nueva especie del género Liolaemus (Iguania:

Liolaemidae) en el grupo de Liolaemus montanus y como grupo hermano del clado formado por L. signifer. La

nueva especie se caracteriza por una combinacion unica de patron morfometrico, escamacion y color. El grupo

montanus del género Liolaemus en Peru contiene diecisiete especies, distribuidas a lo largo de la vertiente

oriental y occidental de los Andes en el sur del pais.

Palabras clave. Andes, Apurimac, Eu/aemus, Puna, reptiles, sistematica, taxonomia

Citation: Chaparro JC, Quiroz AJ, Mamani L, Gutiérrez RC, Condori P, De la Riva |, Herrera-Juarez G, Cerdefa J, Arapa LP, Abdala CS. 2020. An

endemic new species of Andean lizard of the genus Liolaemus from southern Peru (Iguania: Liolaemidae) and its phylogenetic position. Amphibian &

Reptile Conservation 14(2) [General Section]: 47-63 (e238).

tion 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any

medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced, are

as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 9 April 2020; Accepted: 4 May 2020; Published: 21 May 2020

Introduction

Recent taxonomic and systematic studies of lizards of

the genus Liolaemus in the traditional biogeographic

regions of Peru, such as the coast and the Andes, have

increased the diversity of Peruvian Liolaemus to 24

currently recognized species (Gutiérrez et al. 2018;

Uetz et al. 2020). The genus Liolaemus was divided by

Laurent (1983) in two main subgroups, Liolaemus sensu

stricto (L. chiliensis group) and Eulaemus (Argentine

group). The L. alticolor-bibroni group is placed within

the chiliensis group (Aguilar et al. 2013; Barbour 1909;

Gutiérrez et al. 2018; Lobo et al. 2007; Quinteros 2013;

Shreve 1938, 1941), which includes seven Peruvian

species (L. alticolor, L. chavin, L. incaicus, L. pachacutec,

L. tacnae, L. walkeri, and L. wari). On the other hand, the

Argentine group includes the most diverse L. montanus

group (Abdala et al. 2019a), which has sixteen Peruvian

species (L. annectens, L. balagueri, L. chiribaya, L.

etheridgei, L. evaristoi, L. insolitus, L. melanogaster,

Correspondence. *jchaparroauza@yahoo.com, juan.chaparro@mubi-peru.org

Amphib. Reptile Conserv.

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

L. nazca, L. ortizi, L. poconchilensis, L. polystictus, L.

robustus, L. signifer, L. thomasi, L. victormoralesii, and

L. williamsi), and the less diverse L. boulengeri group,

represented only by one species in Peru (L. ornatus:

Carrillo and Icochea 1995). All Peruvian Liolaemus

species are distributed along the southern coast and the

Andes (Abdala and Quinteros 2014; Abdala et al. 2019a;

Aguilar et al. 2016; Gutiérrez et al. 2018). However,

although studies on the Liolaemus of Peru have increased

considerably in recent years, important knowledge gaps

remain, and the diversity of the genus 1s underestimated

(Abdala et al. 2019a; Aguilar-Puntriano et al. 2019).

In 2000, a series of expeditions began along the

highlands of southern Peru, and several localities were

surveyed in the departments of Apurimac, Arequipa,

Ayacucho, Cusco, and Puno. Preliminary analyses reveal

a much richer fauna of reptiles than previously known,

including several new species that are in the process

of description. During expeditions in the Puna areas of

the Apurimac department, populations were found of a

distinctive species of Liolaemus that belongs to the L.

montanus group. These populations are characterized by

the presence of a blade-like process of the tibia associated

with hypertrophy of the muscle tibialis anterior and by

having keeled dorsal scales. The new species is described

here, and it is endemic to the southeast of the department

Apurimac, in southern Peru.

Materials and Methods

Fieldwork procedures. Field work was conducted in

2011, 2012, 2016, and 2019 in the Dry Puna (3,740-

4,615 m asl) of the central cordillera of the Andes in

southern Peru, at the localities of Chila, Ccomerococha,

Chequello, Choaquere, Ccosana, Huanacopampa,

Huanquere, Kuchuacho, Nahuinlla, Progreso,

Pumamarca, and Punchayoc Ccasa, all in Apurimac

department. Individuals of Liolaemus qalaywa sp. nov.

were collected by hand and euthanized with a lethal dose

of Halatal 1%. Tissue samples (muscle or liver) were

taken and stored in microtubes (2 ml) containing 48%

ethanol, and specimens were fixed over 24 h in 10%

formalin and preserved in 80% ethanol. The examined

material is listed in the Appendix.

Images and maps. Photographs of live specimens were

taken using a digital camera (Canon EOS 70D). Close-

up photographs of the holotype (preserved) were taken

with a Stereo Microscope Optics Nikon SMZ25. The

distribution map was elaborated in ArcMap 10.3, using

coordinates previously used by Aguilar et al. (2016)

and Gutiérrez et al. (2018). Type localities were taken

from the original manuscripts of species descriptions.

Coordinates of our records were taken with a GPS device

(datum WGS84), Garmin Etrex 30.

Morphology. Qualitative and quantitative morphological

characters of Liolaemus follow Abdala et al. (2019a,b).

All bilateral characters were measured on the right side.

Color in life was described based on field observations

and photographs of captured specimens. Squamation

was examined under a binocular microscope, and body

measurements were taken with a precision caliper

(+0.01 mm). The terminology of scale descriptions

follows Abdala (2007), Abdala et al. (2019b), Etheridge

(1993, 1995, 2000), and Laurent (1985). Color

pattern terminology follows Abdala (2007), Abdala

et al. (2019a), and Lobo and Espinoza (1999). Some

specimens were dissected to determine sex and maturity;

sex was determined using morphological characters

such as snout-vent length, color and size of anal pores,

and body color patterns. Museum acronyms are: Museo

de Biodiversidad del Pert (MUBI); Museo de Historia

Natural, Universidad Nacional de San Agustin, Arequipa,

Peru (MUSA); and Fundacion Miguel Lillo (FML).

DNA extraction, amplification, and sequencing.

Corresponding voucher specimens are listed in Table

1. Total DNA was extracted from tissue samples at the

Laboratory of Molecular Systematics of the Museo

Nacional de Ciencias Naturales-CSIC (Madrid, Spain)

using the Qiagen DNeasy extraction kit and protocol

(Qiagen Inc., Hilden, Germany). Fragments of the

mitochondrial small subunit rRNA gene (12S) and the

mitochondrial cyt-b genes were amplified by polymerase

chain reaction (PCR). Purified PCR products were sent to

Macrogen Inc. (Seoul, Republic of Korea) for sequencing

in both directions with the amplification primers. The

primers 12SA4 L/12SB-H for 12S (Hillis et al. 1996)

and Glu/CB3/F1/C2 were used for cyt-b (Morando et

al. 2004). Raw sequence chromatographs for sequences

generated in this study were edited using Sequencher

4.9 (Gene Codes Corporation 2009). Two new gene

sequences of these loci were produced, with GenBank

accession numbers for 12S of MT371370, and for cyt-b

of MT366060—MT366062. Voucher information for new

sequences is provided in Table 1.

Phylogenetic analyses. The sequences obtained in this

study (Table 1) were compared with the homologous

Table 1. GenBank codes and voucher information of Liolaemus sequenced for this study. Localities and geographical coordinates

are listed in the type material sections.

Species names Voucher code

Liolaemus qalaywa sp. nov. MUBI 12081

Liolaemus qalaywa sp. nov. MUBI 12099

Liolaemus thomasi MUBI 5925

Amphib. Reptile Conserv.

GenBank cyt-b GenBank 12S

MT366061 MT371370

MT366062

MT366060

May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

sequences held in GenBank (Table 2) for the species of

the L. montanus group that are registered for Peru and

used in the latest studies by Aguilar et al. (2016) and

Villegas-Paredes et al. (2020). This analysis used two

samples of the new species (each one with 12S and cyt-b)

and one sample from L. ornatus from the L. boulengeri

group as “outgroup” (MUSM 31438), according to

Aguilar et al. (2016, 2018) and Villegas-Paredes et al.

(2020). The nucleotide sequences (79 cyt-b, 68 12S)

were aligned using the MUSCLE algorithm (Edgar 2004)

in MEGA version X (Kumar et al. 2018). Mitochondrial

fragments (12S and cyt-b; 1303 nt, 79 individuals)

were concatenated using the MESQUITE version 3.2

program (Maddison and Maddison 2017), and were run

in Mr. Bayes v.3.2.2 (Ronquist and Huelsenbeck 2003).

Bayesian Analyses were performed with the nucleotide

evolution model GTR+ G+ I, for ten million generations

each; an average standard deviation of the frequencies

Table 2. GenBank codes for sequences of Liolaemus lizards and the outgroup used in this study.

Species names

Amphib. Reptile Conserv.

Voucher code

L. ornatus (outgroup) MUSM 31438

L. annectens BYU 50489

L. annectens BYU 50486

L. annectens BYU 50491

L. annectens “Lampa” MUSM 31433

L. balagueri MUSA 5575

L. balagueri MUSA 5576

L. etheridgei BYU 50494

L. etheridgei BYU 50495

L. etheridgei BYU 50497

L. etheridgei MUSM 31493

L. etheridgei BYU 50493

L. etheridgei BYU 50499

L. etheridgei MUSM 31494

L. insolitus MUSM 31490

L. insolitus BYU 50462

L. nazca (L. “Nazca’) BYU 50472

L. nazca (L. “Nazca’) BYU 50507

L. nazca (L. “Nazca’) BYU 50508

L. nazca (L. “Nazca’) MUSM 31523

L. nazca (L. “Nazca’) MUSM 31524

L. ortizi MUSM 31513

L. ortizi MUSM 31514

L. poconchilensis MUSM 31545

L. poconchilensis MUSM 31543

L. poconchilensis MUSM 31544

L. polystictus MUSM 31451

L. polystictus MUSM 31446

L. robustus MUSM 31504

L. robustus MUSM 31508

L. robustus MUSM 31505

L. robustus BYU 50483

L. thomasi BYU 50469

L. thomasi BYU 50466

cyt-b 128 Source

KX826632 KX826732 Aguilar et al. 2016

KX826616 KX826717 Aguilar et al. 2016

KX826615 Aguilar et al. 2016

KX826617 KX826718 Aguilar et al. 2016

KX826618 KX826719 Aguilar et al. 2016

MK568539 Villegas-Paredes et al. 2020

MK568538 Villegas-Paredes et al. 2020

KX826620 KX826721 Aguilar et al. 2016

KX826621 KX826722 Aguilar et al. 2016

KX826622 Aguilar et al. 2016

KX826624 KX826724 Aguilar et al. 2016

KX826619 KX826720 Aguilar et al. 2016

KX826623 KX826723 Aguilar et al. 2016

KX826625 KX826725 Aguilar et al. 2016

KX826627 KX826727 Aguilar et al. 2016

KX826626 KX826726 Aguilar et al. 2016

KX826673 KX826768 Aguilar et al. 2016

KX826674 KX826769 Aguilar et al. 2016

KX826675 KX826770 Aguilar et al. 2016

KX826676 KX826771 Aguilar et al. 2016

KX826677 KX826772 Aguilar et al. 2016

KX826633 KX826733 Aguilar et al. 2016

KX826634 KX826734 Aguilar et al. 2016

KX826637 KX826736 Aguilar et al. 2016

KX826635 Aguilar et al. 2016

KX826636 KX826735 Aguilar et al. 2016

KX826642 KX826740 Aguilar et al. 2016

KX826641 KX826739 Aguilar et al. 2016

KX826646 KX826743 Aguilar et al. 2016

KX826648 KX826744 Aguilar et al. 2016

KX826647 Aguilar et al. 2016

KX826643 Aguilar et al. 2016

KX826680 KX826775 Aguilar et al. 2016

KX826678 KX826773 Aguilar et al. 2016

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

Table 2 (continued). GenBank codes for sequences of Liolaemus lizards and the outgroup used in this study.

Species names

. thomasi

. signifer

melanogaster

. melanogaster

melanogaster

. williamsi

““AbraA pacheta”

. “AbraA pacheta”

““AbraA pacheta”

Be S S S? B ISP a? ERPS GS oS cS, CRS BS ithe hae cee DUIS. ES GY Besse GSS FR tee iS cBetauibe: CS che IN CRS IS ES IS SoS

. victormoralesii (L.

“AbraToccto’)

. polystictus “Castrovirreyna”

. robustus “MinaMartha”

Amphib. Reptile Conserv.

Voucher code

MUSM 31516

BYU 50467

MUSM 31443

MUSM 31434

BYU 50444

BYU 50357

BYU 50350

MUSM 31437

BYU 50355

MUSM 31447

MUSM 29110

BYU 50151

MUSM 31472

MUSM 31475

BYU 50154

MUSM 31371

MUSM 31374

MUSM 31373

BYU 50426

MUSM 31461

BYU 50430

MUSM 31462

BYU 50431

BYU 50428

MUSM 31464

MUSM 31465

MUSM 31468

BYU 50463

MUSM 31485

BYU 50143

BYU 50464

BYU 50144

MUSM 31486

BYU 50465

MUSM 31481

BYU 50145

BYU 50148

MUSM 31454

BYU 50630

BYU 31455

BYU 50438

MUSM 31439

cyt-b

KX826681

KX826679

KX826656

KX826654

KX826652

KX826651

KX826649

KX826655

KX826650

KX826657

KX826653

KX826628

KX826630

KX826631

KX826629

KX826665

KX826667

KX826666

KX826661

KX826668

KX826663

KX826669

KX826664

KX826662

KX826670

KX826671

KX826672

KX826684

KX826687

KX826682

KX826685

KX826683

KX826688

KX826686

KX826660

KX826658

KX826659

KX826639

KX826638

KX826640

KX826644

KX826645

128

KX826776

KX826774

KX826752

KX826750

KX826748

KX826747

KX826745

KX826751

KX826746

KX826753

KX826749

KX826728

KX826730

KX826731

KX826729

KX826757

KX826762

KX826758

KX826763

KX826760

KX826764

KX826761

KX826759

KX826765

KX826766

KX826767

KX826778

KX826587

KX826779

KX826777

KX826781

KX826780

KX826756

KX826754

KX826755

KX826738

KX826737

KX826570

KX826741

KX826742

May 2020 | Volume 14 | Number 2 | e238

Source

Aguilar et al.

2016

Chaparro et al.

go Liolaemus_poconchilensis_MUSM_31544

Liolaemus_poconchilensis_MUSM_31543

Liolaemus_ornatus_MUSM_31438

Liolaemus_nazca_BYU_50472

Liolaemus_nazca_MUSM_31524

iooLiolaemus_nazca_BYU_50507

Liolaemus_nazca_BYU_50508

100

100 100

Liolaemus_nazca_MUSM_31523

Liolaemus_balagueri_MUSA5575

— Liolaemus_balagueri_MUSA5576

100

Liolaemus_robustus_MUSM_31504

Liolaemus_robustus_MUSM_31508

, Liolaemus_robustus_MUSM_31505

* Liolaemus robustus_BYU_50483

yo) Liolaemus_robustus_MinaMartha_BYU_50438

Liolaemus_robustus_MinaMartha_MUSM_ 31439

Liolaemus_polystictus_ MUSM_31451

Liolaemus_polystictus_MUSM_31446

AbraApacheta_MUSM_31481

AbraApacheta_BYU_50145

100

AbraApacheta_BYU_50148

Liolaemus_polystictus_Castrovirreyna_MUSM_31454

Liolaemus_polystictus_Castrovirreyna_BYU_31455

Liolaemus_polystictus_Castrovirreyna_BYU_ 50630

Liolaemus_victormoralesii_MUSM_31371

Liolaemus. victormoralesii_MUSM_31374

Liolaemus_victormoralesii_MUSM_31373

Liolaemus_victormoralesii_ MUSM_31461

400L jolaemus_victormoralesii_BYU_50430

Liolaemus_victormoralesii_ BYU_50428

Liolaemus_victormoralesii_MUSM_31464

Fa Liolaemus_victormoralesii_BYU_50431

Liolaemus_victormoralesii_MUSM_31468

Liolaemus_victormoralesii_MUSM_31465

Liolaemus_victormoralesii_MUSM_31462

Liolaemus_victormoralesii_BYU_50426

Liolaemus_melanogaster_BYU_50151

a Liolaemus_melanogaster_MUSM_31472

Liolaemus_melanogaster_MUSM_31475

Liolaemus_melanogaster_BYU_50154

97 Liolaemus_williamsi_BYU_50463

Liolaemus_williamsi_BYU_50464

Liolaemus_williamsi_BYU_50144

5, Liolaemus_williamsi_MUSM_31486

Liolaemus_williamsi_BYU_50143

Liolaemus_williamsi_BYU_50465

Liolaemus_williamsi_MUSM_31485

, Liolaemus_ annectens_BYU_50489

Liolaemus_annectens_BYU_50491

Liolaemus_annectens_BYU_50486

Liolaemus_etheridgei_BYU_50494

%3Liolaemus_etheridgei_BYU_50493

Liolaemus_etheridgei_BYU_50497

;, Liolaemus_etheridgei_MUSM_31493

Liolaemus_etheridgei_MUSM_31494

Liolaemus_etheridgei_BYU_50499

Liolaemus_etheridgei BYU 50495

Liolaemus_annectens_Lampa_MUSM_31433

, Liolaemus. _ signifer_MUSM_31443

“Liolaemus signifer_BYU_50444

Liclaemus_signifer_ MUSM_31434

Liolaemus_signifer_BYU_50357

Liolaemus_signifer_BYU_50350

Liolaemus_signifer_BYU_50355

Liolaemus_signifer_MUSM_31447

Liolaemus_signifer_MUSM_31437

“9 Liclaemus_qalaiwa_MUBI_12081

Liolaemus_qalaiwa_MUBI_12099

Liolaemus_signifer_MUSM_29110

5) Liolaemus_ortizi_MUSM_31513

Liolaemus_ortizi_ MUSM_31514

Liolaemus_thomasi_BYU_50469

100

Liolaemus_thomasi_BYU_50466

sg Liolaemus_thomasi_BYU_50467

sa Liolaemus_thomasi_MUSM_31516

Liolaemus_thomasi_MUBI_5925

Liolaemus_poconchilensis_MUSM_31545

Liolaemus_insolitus_MUSM_31490

Liolaemus_insolitus_BYU_50462

0.03

Fig. 1. Phylogenetic tree obtained using Bayesian Methods (BM).

divided below 0.05 was obtained, trees were sampled

every 1,000 generations from the Markov Chain Monte

Carlo (MCMC) output and using four simultaneous

chains (one cold and three hot) in each run. The

convergences of the chains to the stationary distribution

were confirmed using Tracer v1.7.1 (Rambaut et al. 2018)

and the first 25% of generations that were not within the

stationary distribution of the registration probabilities

were conservatively discarded. The trees and subsequent

probabilities were summarized using the “50% majority”

Amphib. Reptile Conserv.

consensus method (Huelsenbeck and Ronquist 2001;

Wilcox et al. 2002). The resulting consensus tree was

edited using FigTree v1.4.3 (Rambaut 2014).

Results and Discussion

Phylogenetic analyses. Our Bayesian phylogenetic

analysis (Fig. 1) shows that the terminals MUBI 12081

and MUBI 12099 form a monophyletic subclade, sister

of a terminal identified as L. signifer sensu lato (MUSM

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

29110) from Desaguadero, in Puno department, near

the Bolivian border. The clade that contains these three

terminals is deeply separated from its sister clade, which

contains L. signifer, L. annectens, and L. etheridgei.

Therefore, we consider that the new populations of

Liolaemus studied herein represent a species different

from those previously described for the Z. montanus

group. Likewise, in addition to L. signifer, L. annectens

appears also to be paraphyletic, as the sample MUSM

31433 from Lampa (Puno department) identified as such

is sister to a clade containing both L. annectens and L.

etheridgei, a result already found by Aguilar et al. (2016).

Thus, the two divergent lineages represented by

MUSM 29110 and MUSM 31433 might be two potential

new species of the LZ. montanus group from the Andean

highlands of southern Peru.

Taxonomy

Liolaemus qalaywa sp. nov.

(Figs. 2-3)

urn: lsid:zoobank.org:act:047FCD4E-8717-4378-B130-0B1BB633D187

Holotype. MUBI 13286, an adult male (Figs. 2-3)

from Choaquere, District of Challhuahuacho, Province

of Cotabambas, Department of Apurimac, Peru,

(14°7°20.32”S, 72°13’29.27°W) at 3,740 m above sea

level (m asl), collected on 15 December 2011, by L.

Mamani and J.C. Chaparro.

Paratypes. Four adult males: MUBI 13265, from

Nahuinlla, District of Chorrillos, Province of Cotabambas,

Department of Apurimac, Peru (13°57’20.64’S,

72°23’59.12”W) at 4,010 m asl, collected on 25 April

2016 by FP. Condori; MUBI 12981, from Progreso,

District of Progreso, Province of Grau, Department

of Apurimac, Peru (14°5’16.07”S, 72°27°32.23”W)

at 4,180 m asl, collected on 25 April 2016 by A.J.

Quiroz; MUBI 17621, from Punchayoc Ccasa, District

of Cotabambas, Province of Cotabambas, Department

of Apurimac, Peru (13°47’23.55”S, 72°18’15.19”"W) at

4,290 masl, collected on 15 October 2019 by L. Mamani;

MUBI 17622, from Ccosana, District of Haquira,

Province of Cotabambas, Department of Apurimac,

Peru (14°217°58.53”S, 72°20°40.10”W) at 4,615 m asl,

collected on 16 October 2019, by L. Mamani. One

subadult male: MUSA 5600, from Pumamarca, District

of Challhuahuacho, Province of Cotabambas, Department

of Apurimac, Peru (14°2’56.27”S, 72°19°46.10”W) at

4.615 m asl, collected on 10 December 2011 by A.J.

Quiroz. Nine adult females: MUSA 5601 (MUBI 12080),

MUBI 12081, and MUBI 12084, from Chila, District of

Challhuahuacho, Province of Cotabambas, Department

of Apurimac, Peru (14°7’5.65”S, 72°13’48.97°W)

at 3,750 m asl, collected on 24 May 2012 by J.C.

Chaparro; MUBI 12100, from Choaquere, District of

Amphib. Reptile Conserv.

Challhuahuacho, Province of Cotabambas, Department

of Apurimac, Peru (14°7°20.32”S, 72°13’29.27’W)

at 3,740 m asl, collected between 30 May and 6 June

2012 by J.C. Chaparro; MUBI 13260 (Fig. 3B,D,F,H)

and MUBI 13264, from Nahuinlla, District of Chorrillos,

Province of Cotabambas, Department of Apurimac,

Peru (13°57°20.64”S, 72°23’59.12”W) at 4,010 m asl,

collected on 25 April 2016 by F.P. Condori; MUBI

13287 from Pumamarca, District of Challhuahuacho,

Province of Cotabambas, Department of Apurimac, Peru

(14°2’56.27°S, 72°19’46.10”W) at 4,615 m asl, collected

on 10 December 2011 by L. Mamani; MUBI 15900 and

MUBI 15903, from Ccomerococha, District of Coyurqui,

Province of Cotabambas, Department of Apurimac,

Peru (13°50°44.99"S, 72°21°24.14"W) at 4310 m

asl, collected on 13 July 2016 by A. Quiroz. Thirteen

immatures: MUBI 12982-83, from Progreso, District

of Progreso, Province of Grau, Department of Apurimac,

Peru (14°5’16.07”S, 72°27°32.23”W) at 4,180 m asl,

collected on 25 April 2016 by A.J. Quiroz; MUBI 12096—

99 and MUBI 12101—04, from Choaquere, District of

Challhuahuacho, Province of Cotabambas, Department

of Apurimac, Peru (14°7’20.32”S, 72°13’29.27°W) at

3,740 m asl, collected on 30 May 2012 by J.C. Chaparro;

MUBI 15901-02, from Ccomerococha, District of

Coyurqui, Province of Cotabambas, Department of

Apurimac, Peru (13°50°44.99”S, 72°21°24.14"W) at

4,310 m asl, collected on 13 July 2016 by A.J. Quiroz;

MUBI 17623, from Ccosana, District of Haquira,

Province of Cotabambas, Department of Apurimac,

Peru (14°217°58.53”S, 72°20°40.10”W) at 4,615 m asl,

collected on 16 October 2019 by L. Mamani.

Diagnosis. We assign L. galaywa sp. nov. to the L.

montanus group because it presents a blade-like process

on the tibia, associated with the hypertrophy of the tibial

muscle tibialis anterior (Abdala et al. 2019b; Etheridge

1995) and based on molecular phylogeny (Fig. 1). The

species of the ZL. montanus group differ from those of

the L. boulengeri group by the complete absence of

patches of enlarged scales in the posterior part of the

thigh (Abdala 2007). Compared to the species of the L.

montanus group, L. galaywa sp. nov. is a robust lizard

differing by its larger size (max SVL = 96.06 mm) from

L. andinus, L. audituvelatus, L. balagueri, L. cazianiae,

L. chiribaya, L. duellmani, L. eleodori, L. erguetae,

L. erroneus, L. etheridgei, L. evaristoi, L. fabiani, L.

famatinae, L. fittkaui, L. foxi, L. gracielae, L. griseus,

L. hajeki, L. halonastes, L. huacahuasicus, L. insolitus,

L. islugensis, L. molinai, L. montanus, L. multicolor,

L. nazca, L. omorfi, L. orko, L. ortizi, L. pantherinus,

L. poconchilensis, L. poecilochromus, L. porosus, L.

pulcherrimus, L. reichei, L. robertoi, L. rosenmanni, L.

ruibali, L. schmidti, L. stolzmanni, L. tajzara, L. thomasi,

L. torresi, L. vallecurensis, and L. williamsi (all with

SVL 50-80 mm). The presence of imbricate dorsal

scales with keels differentiates L. galaywa sp. nov.

May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

Fig. 2. Details of the holotype of Liolaemus qalaywa sp. nov. MUBI 13286 (SVL = 85.54 m

ventral views of body, (C) lateral, (D) dorsal, and (E) ventral views of head, (F) ventral view of precloacal pores, (G) ventral aspect

of right hand, (H) ventral aspect of right foot, (I) keeled dorsal body scales, (J) ventral body scales. Scale = 5 mm.

Amphib. Reptile Conserv. 53 May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

Fig. 3. (A, C, E, G) Adult male of Liolaemus qalaywa sp. nov. (unvouchered specimen; SVL = 91.9 mm, Tail = 121.1 mm); (B, D,

F, H) Adult female of Liolaemus qalaywa sp. nov. (MUBI 13260 paratype; SVL = 85.53 mm, Tail = 110.78 mm). Both individuals

are from Nahuinlla, Department of Apurimac, 4,010 m asl.

Amphib. Reptile Conserv. 54 May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

from species with smooth juxtaposed or sub-imbricate

scales such as L. andinus, L. audituvelatus, L. balagueri,

L. cazianiae, L. chiribaya, L. eleodori, L. erguetae, L.

fabiani, L. foxi, L. gracielae, L. halonastes, L. insolitus, L.

islugensis, L. jamesi, L. molinai, L. nigriceps, L. omorfi,

L. patriciaiturrae, L. pleopholis, L. poconchilensis,

L. poecilochromus, L. porosus, L. reichei, L. robertoi,

L. robustus, L. rosenmanni, L. ruibali, L. schmidti, L.

scrocchii, L. torresi, L. vallecurensis, L. victormoralesii,

and L. vulcanus.

The new species differs from L. chiribaya, L. evaristoi,

L. etheridgei, L. islugensis, L. insolitus, L. multicolor, L.

omorfi, L. poconchilensis, L. pulcherrimus, L. robertoi,

L. ruibali, and L. schmidti, by the absence of sky blue or

celeste scales on the sides and dorsum of the body and

tail. The number of scales around midbody in L. galaywa

Sp. nov. varies between 52 and 58 (mean = 54.6), which

differentiates it from several species of the group with

more than 65 scales, such as L. andinus, L. annectens, L.

audituvelatus, L. cazianiae, L. duellmani, L. eleodori, L.

erguetae, L. forsteri, L. foxi, L. gracielae, L. halonastes,

L. inti, L. molinai, L. multicolor, L. nigriceps, L.

patriciaiturrae, L. pleopholis, L. poecilochromus, L.

porosus, L. pulcherrimus, L. robertoi, L. rosenmanni,

L. ruibali, L. schmidti, L. signifer, and L. vallecurensis.

The number of ventral scales between the mental

scale and the border of the vent in L. galaywa sp. nov.

varies between 71 and 83 (mean = 75.7), and is lower

than the number in the following species (with more

than 90 ventral scales): L. andinus, L. cazianiae, L.

erguetae, L. foxi, L. gracielae, L. halonastes, L. inti, L.

multicolor, L. nigriceps, L. pachecoi, L. patriciaiturrae,

L. pleopholis, L. poecilochromus, L. porosus, L. robertoi,

L. rosenmanni, L. torresi, and L. vallecurensis, and

higher than the number in the following species (with

less than 70 ventral scales): L. dorbignyi, L. fittkaui, L.

melanogaster, L. polystictus, and L. thomasi. Females of

L. qalaywa sp. nov. exhibit precloacal pores in contrast to

females of L. andinus, L. audituvelatus, L. aymararum, L.

balagueri, L. duellmani, L. fabiani, L. fittkaui, L. griseus,

L. hajeki, L. islugensis, L. jamesi, L. melanogaster, L.

polystictus, L. puritamensis, L. reichei, L. robertoi, L.

rosenmanni, L. ruibali, L. signifer, and L. vallecurensis

(all lack precloacal pores). Additional measurements

of morphometric characteristics in adult specimens are

shown in Table 3.

The coloration patterns of males and females

(especially the deep yellow and orange color around the

eye), of the palpebral scales, and on the posterior inner

edge of the auditory meatus in females are character

states that have not been reported in Liolaemus. This

exclusive coloration pattern in both sexes was seen in

different individuals throughout the year. Therefore, they

differ from all known species in the L. montanus group.

Description of the holotype. Adult male (MUBI 13286).

SVL 85.54 mm. Head 1.08 times longer (20.24 mm) than

Amphib. Reptile Conserv.

wide (18.71 mm). Head height 15.9 mm. Neck width 20.5

mm. Eye diameter 4.67 mm. Interorbital distance 9.99

mm. Orbit-auditory meatus distance 8.71 mm. Auditory

meatus 4.51 mm high, 1.53 mm wide. Orbit-commissure

of mouth distance 7.35 mm. Internasal width 1.59 mm.

Subocular scale length 5.3 mm. Trunk length 37.68 mm,

width 28.7 mm. Tail length 110 mm. Femur length 16.03

mm, tibia 16.47 mm, and foot 22.8 mm. Humerus length

12.04 mm. Forearm length 10.91 mm. Hand length

14.11 mm. Pygal region length 7.67 mm, and cloacal

region width 6.37 mm. Dorsal surface of head rough,

with 13 scales, rostral 2.62 times longer (3.25 mm) than

wide (1.24 mm). Mental larger (4.06 mm) than rostral,

trapezoidal, surrounded by four scales. Nasal separated

from rostral by two scales. Two internasals, longer than

wide. Nasal surrounded by seven scales, separated from

canthal by two scales. Six scales between frontal and

rostral. Frontals divided into four scales. Interparietal

larger than parietal, in contact with six scales. Preocular

separated from lorilabials by one scale. Five superciliaries

and 15 upper ciliaries scales. Three differential scales

at anterior margin of auditory meatus. Six temporal

scales. Five lorilabials in contact with subocular. Ten

supralabials, which are not in contact with subocular.

Seven supraoculars. Seven lorilabials. Six infralabials.

Five chin shields, 4" pair separated by six scales. Fifty-

six scales around half of body. Fifty-four triangular

dorsal body scales, imbricated, with an evident keel; fore

and hind limbs and tail with lamellar scales, imbricated

and keeled. Seventy-one ventral scales, from the mental

to the cloacal region, following the ventral midline of

the body, laminar, imbricated. Thirty imbricated gulars,

not keeled. Neck with longitudinal fold with 43 granular,

not keeled scales, ear fold and antehumeral fold present.

Gular fold absent. Eighteen subdigital lamellae on the 4"

finger of the hand. Fourth toe with 23 subdigital lamellae

with three keels, plantar scales with keels. Lamellar

ventral tail, scales imbricated with a slight keel. Seven

precloacal pores. Supernumerary pores present.

Coloration

Holotype coloration in preservative. Upper temporal

area dark brown. Lower temporal region, supralabial,

infralabial, lorilabial, and loreal scales white with dark

edges. A transverse dark stain crossing the eye and the

palpebral scales, the rest of 1t white. Lateral color of the

neck white, back of the neck dark brown. Scales of body,

limbs, and tail black with a white distal end. Irregular lines

transverse to the body, white, with black edges, extending

from one side of the body to the other. White spots on the

back and sides of limbs, hands, tail, and scapular region.

No vertebral line, dorsolateral bands, and ante-humeral

arch. Sides of the body lighter below the lateral midline.

Venter uniform, with white scales and black edges.

Color variation in life. Liolaemus galaywa sp. nov.

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

Table 3. Measurements (in mm) of morphological characteristics in adult specimens of Liolaemus qalaywa sp. nov. * broken tail,

** regenerated tail.

MUBI MUSA MUBI MUBI

Museum code 13286 5600 13287 13264

Stage of development Adult Subadult Adult Adult

Sex Male Male Female Female

Type material Holotype Paratype Paratype Paratype

Snout-vent length 85.54 79.25 79.35 77.23

Head length 20.24 18.86 18.09 16.85

Head width 18.71 13.35 17.32 15.46

Head height 15.90 10.64 12.36 12.54

Interorbital distance 9.99 8.25 9.16 8.90

Tail length 110.00 103.44 86.15 (*) 108.57

Cloacal opening width 12.76 9.56 10.74 10.64

Body width 28.70 17.42 26.85 24.41

Width of base of tail 37.68 33.67 40.20 38.28

Femur length 16.03 15.16 13.60 14.32

Tibia length 16.47 15.86 14.41 13.98

Foot length 22.80 19.12 19.67 19.24

ruse Secor 14.39 11.35 12.61 12.53

Humerus length 12.04 10.73 8.74 9.78

Humerus width 7.03 3.9] 5.90 5.45

Radio length 10.91 9.93 9.45 9.68

Hand length 14.11 10.27 14.02 12.88

Tympanum height 4.5] 3-54 3.91 3.85

Tympanum length | ea 1-72 OF 120

Neck width 20.50 13.10 17.40 15.62

shows evident sexual dimorphism. In males, head varies

considerably from brown to black. The back of the head

is generally black or dark gray, in most cases darker than

the sides of the head. The temporal region is similar in

color to the back, but with lighter shades and in some

specimens with white, yellow, or light gray spots. The

supralabial, infralabial, and part of the lorilabial scales

are always lighter than the rest of the head, sometimes

immaculate or tinted with darker colors. The subocular,

preocular, postocular, and loreal scales in most specimens

are faint yellow, white, or light gray with light blue

shades. The eyelid scales are always conspicuous, faint

yellow, as are the posterior internal scales of the auditory

meatus. In some specimens, the atrial scales, or part of

them, also have the same yellow color as the palpebral

scales. The general color of the body varies between

chestnut and dark gray. Most of the dorsal scales of the

body have a posterior end lighter than the anterior end,

with yellow being the predominant color. The design of

the dorsal body color pattern 1s diffuse and variable. Some

males do not have paravertebral or lateral spots, while

others exhibit thin, irregular, dark-colored transverse

lines with a light back trim, with black and yellow being

the most common combination. These lines can thicken

or have a sub-quadrangular shape in the paravertebral

Amphib. Reptile Conserv.

MUBI MUBI MUBI MUBI MUBI

13260 12100 12084 12081 12080

Adult Adult Adult Subadult Subadult

Female Female Female Female Female

Paratype Paratype Paratype Paratype Paratype

85.53 79.43 79.75 68.51 74.78

19.34 17.23 18.20 16.04 16.89

16.71 15.45 15.75 14.10 15.66

12.10 12.50 11.64 9.68 11.03

9.30 8.99 9.06 8.50 8.99

110.78 59.45 (**) 64.44 (**) 87.87 99.88

11.64 9.69 13.14 10.77 11.49

34.34 29.34 30.28 22.06 26.18

37.34 42.76 36.52 28.87 32.61

15.48 13.63 14.36 1237 13.79

14.36 13.21 13.94 13.04 13.41

20.17 18.62 19.03 18.00 19.65

12.92 11.11 12.18 11.50 12.22

10.15 9.26 9362 7.45 9.08

5.93 5.24 5.27 4.65 5.00

9.54 8.63 8.99 8.76 8.84

11.84 11.07 10.96 10.88 12.53

416 3.64 3.72 3.36 3.18

1.66 1.18 1.62 1.49 1.27

17.44 15.11 17.43 14.46 18.00

region, and in some specimens the yellow lines fuse in

the vertebral region. In the scapular region, numerous

circular spots are highlighted, small in size and white,

yellow, or light gray in color; these spots may also be

present on the sides of the body, including some that are

irregular or oblong in shape. The limbs and tail have the

same pattern as the body. On the back of the limbs there

are small light-colored spots, mostly intense yellow. The

sides of the fingers and toes are faint yellow or white.

In the antehumeral, pygal, and femoral regions, yellow

shades stand out. Most specimens are white ventrally, but

several specimens have black or dark gray scale edges in

the mental, gular, pectoral, and abdominal regions. Some

specimens have gray ventral scales with a light blue

or light gray distal end. The tail generally changes to a

lighter color after the first third. In females the coloration

pattern is totally different and, unlike most Liolaemus

species, the females have a more lively and showy

coloration than males. The main difference is in the deep

yellow color of the palpebral scales, those of the internal

auditory meatus, the sides of the fingers and toes, as well

as the back of the thighs and arms. As in males, the back

of the head is darker than the sides. The predominant

color on the back 1s dark brown. Small spots or white

scales are present on the frontal and interparietal regions.

May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

The temporal region varies from brown to light gray. The

supralabial, infralabial, and lorilabial scales are brown

or light gray, with dark edges. The subocular, preocular,

postocular, and loreal scales range from light gray to

deep yellow. The color of the body varies from gray to

brown and, like in the males, there are dark scales with a

light distal end. The paravertebral spots are conspicuous

and evident, generally black and diamond-shaped, which

may be attached to a transverse black line that can extend

to the vertebral zone and to the mediolateral line on the

sides of the body. These paravertebral spots in some

cases have a white anterior border. The sides of the body,

humeral area, limbs, and tail have similar patterns as the

males. Ventrally they are white or gray with some yellow

undertones. In the gular and mental regions there may

be dark spots and nuances, while on the belly some have

iridescent scales that are greenish gray or bluish gray.

The tail becomes darker distally.

Etymology. The specific epithet Qalaywa, refers to the

Quechua word for the Liolaemus lizards from the high

Peruvian Andes.

Distribution and natural history. All known specimens

and observations of L. galaywa sp. nov. come from

the 12 localities of the type series, in the southeastern

Department of Apurimac, Peru, at elevations between

3,740-4,615 m asl (Figs. 4-5). This species inhabits

high Andean puna (Figs. 4-5). Within the geographical

distribution of L. galaywa sp. nov., four vegetation units

were determined: Wetlands, Puna Lawn, Grassland,

and Bushes. The common floristic composition of each

unit is as follows—Wetlands: Distichia muscoides,

Zameoscirpus muticus, Carex sp., Eleocharis

albibracteata, Calamagrostis rigescens, and Oritrophium

limnophyllum, Puna Lawn: Aciachne —acicularis,

Calamagrostis vicunarum, Paranephelius ovatus, and

Trichophorum rigidum; Grassland: Jarava ichu, Festuca

dolichophylla, and Calamagrostis sp.; and Bushes:

Ribes sp., Gynoxis sp., Escallonia myrtylloides, and

Berberis sp. Liolaemus qalaywa sp. nov. inhabits all the

vegetation units, but in wetlands the presence of lizards

is restricted to the edges. During intensive field work

in the study area, the presence of adult individuals was

registered throughout the year; newborns (with presence

of abdominal scar) appeared from November to March,

and immatures were found from December to March. In

December, some couples showing reproductive behavior

were observed. The species presents a viviparous

reproduction; an embryo was found inside the body

of an adult female. Sympatric amphibian and reptile

species include Gastrotheca marsupiata, Pleurodema

marmoratum, Telmatobius ctf. jelskii, Liolaemus. aff.

incaicus, and Tachymenis peruviana.

Individuals were registered and collected during

the dry and wet seasons, in natural rocky areas (under

rocks when hiding and on rocks when active). One of the

Amphib. Reptile Conserv.

authors spent 82 person-hours of search time, between

0815 and 1500 hrs distributed from 2013 to 2019; and the

total captures (catch and release) were 249 individuals,

147 in the wet season and 102 in the dry season (97

males, 113 females, and 39 immatures). The average

results of our records in areas with different degrees of

impact per hour of effort varied: in natural areas without

anthropic intervention, five individuals/hour of search

were registered; in natural areas with little human impact,

three; in agricultural areas (which include sectors with

stones or rocky outcrops) 1.5; and in very impacted areas

(including areas with removal of land and vegetation,

like mining, roads, and buildings) the result was zero

individuals per hour. In general, our observations show

that L. galaywa sp. nov. is abundant in sectors where

the habitat remains intact, especially in areas with the

presence of natural rock outcrops and scattered stones

which serve as permanent or temporary shelters. However,

in areas with the presence of agricultural, livestock, and

open-pit mining activities, abundance tends to decrease,

as the lizards maintain their populations at the perimeters

of zones with anthropic activities.

Individuals of this species are capable of building

their own burrows and taking advantage of the burrows

of other animals, such as tarantulas and mice; in the

same way, arachnids, insects, and small mammals use

the lizard’s shelters. Several individuals were observed

thermoregulating on small and medium-sized stones,

and on stony sandy substrate and grass-like vegetation,

always near their shelters. The peak of activity was

during 0900-1100 hrs.

The presence of ectoparasites of the family

Trombiculidae was recorded throughout the year;

however, the presence and density of ectoparasites

seemed to be lower during December—March, increasing

in number during April-November. The ectoparasites

were found distributed in the mite pocket, tympanic duct,

prefemoral region, inguinal pocket, and axillary pocket.

An adult male individual of L. galaywa sp. nov.

was registered eating a frog of the species Pleurodema

marmoratum, when it was under a stone. Other

individuals were seen consuming insect larvae.

Male-male interactions were observed, including

ageressive behavior by back arching, followed by a

mutual lateral presentation that ends with one lizard

charging and biting the other, and the persecution from

the winner to the loser. Also, when the lizards sense

danger, such as when researchers try to capture them,

they show back arching and opening of the mouth,

producing an exhalation-like sound.

Acknowledgments.—We are grateful to several

institutions for allowing the review of specimens from

their museum collections: Evaristo Lopez Tejeda

(MUSA), Sonia Kretzschmar and Esteban Lavilla

(FML), and the staff of Museo de Biodiversidad del

Peru (MUBI). The comments of Claudia Koch, Mika

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

Fig. 4. Habitat of Liolaemus galaywa sp. nov. from localities in the Department of Apurimac: (A) Quequello; (B) Nahuinlla; (C)

Huanquere; (D) Choaquere; (E) Queufia; (F) Ccomerococha; (G) Huanacopampa.

Amphib. Reptile Conserv. 58 May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

16°0'0"S 15°0'0"S 14°0'0"S 13°0'0"S 12°0'0"S 11°0'0"S

17°0'0"S

PACIFIC OCEAN

18°0'0"S

OQ 250 500 1,000 km

19°0'0"S

75°0'0"W

70°0'O"W

Legend

@ L. polystictus * L. signifer

@ L. robustus

oO L. victormoralesii sd a

cc) bia Apacheta’ [=f] L. annectens

@ “Castrovirreina’ CAL. evaristoi

@ ‘Minas Martha” (2) Type localities

AL balagueri /\, Type localities

A L-chiribaya ae Type localities

A. L- insolitus ce Type localities

A L-nazca [J] 0 - 1,000

ZL. poconchilensis — 4,001 - 2,000

AL. williamsi 2,001 - 3,000

He L etheridge WN 3,001 - 4,000

4,001 - 5,000

* L. melanogaster =

MMMM 5,001 - 6,891

* L. ortizi

* L. qalaywa sp. nov.

70°0'0"W

Fig. 5. Geographic distribution of 17 formally described species, and three candidate species of Liolaemus. Symbols with a black

dot in the middle represent the type locality of each species. Species with quotation marks in names belong to the candidate species.

Jormakka, and two anonymous reviewers improved our

manuscript considerably, and we are grateful to Lucy

Vargas for providing botanical information. JCC is

grateful to Javier Bustamante for technical aids to cover

obtaining molecular information and to Fernando Rojas,

and to Kelly Preciado for field assistance. Sequencing

expenses were funded by project CGL2011-30393 of

the Spanish Government (PI: I. De la Riva). CSA thanks

the Cerdefia family and the Hotel Princes from Arequipa

and the Gutiérrez family in Lima for their friendship and

support. Thanks to the Agencia de Investigacion Cientifica

y Tecnologica, Argentina (PICT 2015-1398). Collection

permits for specimens held at the Museo de Biodiversidad

del Pert (MUBI) were issued and recognized by

SERFOR through Resolucion de Direccion General N°

024-2017-SERFOR/DGGSPFFS, and Resolucién de

Direccion General N° 369-2019-MINAGRI-SERFOR-

DGGSPFFES.

References

Abdala CS. 2007. Phylogeny of the L. boulengeri

group (Iguania: Liolaemidae, Liolaemus) based on

morphological and molecular characters. Zootaxa

1538: 1-84.

Abdala CS, Quinteros AS. 2014. Los ultimos 30 afios

Amphib. Reptile Conserv.

de estudios de la familia de lagartijas mas diversa de

Argentina. Actualizacion taxonomica y sistematica

de Liolaemidae. Cuadernos de Herpetologia 28:

55-82.

Abdala CS, Quinteros AS, Semhan RV, Bulacios-Arroyo

AL, Schulte J, Paz MM, Ruiz-Monachesi MR,

Laspiur A, Aguilar-Kirigin AJ, Gutiérrez-Poblete R,

et al. 2019a. Unravelling interspecific relationships

among highland lizards: first phylogenetic hypothesis

using total evidence of the Liolaemus montanus

group (Iguania: Liolaemidae). Zoological Journal of

the Linnean Society 114: 1-29.

Abdala CS, Aguilar-Kirigin AJ, Semhan RV, Bulacios-

Arroyo AL, Valdés J, Paz MM, Gutiérrez RC,

Valladares-Faundez P, Langstroth R, Aparicio J.

2019b. Description and phylogeny of a new species

of Liolaemus (Iguania: Liolaemidae) endemic to the

south of the Plurinational State of Bolivia. PLoS

ONE 14(12): e0225815.

Aguilar-Puntriano C, Avila LJ, De la Riva I, Johnson

L, Morando M, Troncoso-Palacios J, Wood PL, Sites

JW. 2018. The shadow of the past: convergence

of young and old South American desert lizards

as measured by head shape traits. Ecology and

Evolution 2018(8): 11,399-11,409.

Aguilar-Puntriano C, Ramirez C, Castillo E, Mendoza

May 2020 | Volume 14 | Number 2 | e238

New Liolaemus species from Peru

A, Vargas VJ, Sites JW Jr. 2019. Three new lizard

species of the Liolaemus montanus group from Peru.

Diversity 11(161): 1-19.

Aguilar C, Wood PL Jr, Belk M, Duff MH, Sites JW

Jr. 2016. Different roads lead to Rome: integrative

taxonomic approaches lead to the discovery of two

new lizard lineages in the Liolaemus montanus

group (Squamata: Liolaemidae). Biological Journal

of the Linnaean Society 120: 448-467.

Aguilar C, Wood PL Jr, Cusi JC, Guzman A, Huari

F, Lundberg M, Mortensen E, Ramirez C, Robles

D, Suarez J, et al. 2013. Integrative taxonomy

and preliminary assessment of species limits

in the Liolaemus walkeri complex (Squamata,

Liolaemidae) with descriptions of three new species

from Peru. ZooKeys 364: 47-91.

Barbour T. 1909. Some new South American cold-

blooded vertebrates. Proceedings of the New

England Zoélogical Club 4: 47-52.

Carrillo N, Icochea J. 1995. Lista taxonomica preliminar

de los reptiles vivientes del Peru. Publicaciones del

Museo de Historia Natural Universidad Nacional

Mayor de San Marcos 49: 1—27.

Edgar RC. 2004. MUSCLE: multiple sequence

alignment with high accuracy and high throughput.

Nucleic Acids Research 32(5): 1,792—1,797.

Etheridge R. 1993. Lizards of the Liolaemus darwinii

complex (Squamata: Iguania: Tropiduridae) in

northern Argentina. Bollettino del Museo Regionale

di Scienze Naturali, Torino 11: 137-199.

Etheridge R. 1995. Redescription of Ctenoblepharis

adspersa Tschudi, 1845, and the taxonomy of

Liolaeminae (Reptilia: Squamata: Tropiduridae).

American Museum Novitates 3142: 1-34.

Etheridge RE. 2000. A review of the Liolaemus

wiegmannii group (Squamata, Iguania, Tropiduridae),

and a history of morphological change in the sand-

dwelling species. Herpetological Monographs 4:

293-352.

Gene Codes Corporation. 2009. Sequencher. Version

4.9. Gene Codes, Ann Arbor, Michigan, USA.

Gutiérrez RC, Chaparro JC, Vasquez MY, Quiroz AJ,

Aguilar-Kirigin A, Abdala CS. 2018. Descripcion

y relaciones filogenéticas de una nueva especie de

Liolaemus (Iguania: Liolaemidae) y notas sobre

el grupo de L. montanus de Peru. Cuadernos de

Herpetologia 32: 81-99.

Hillis DM, Moritz C, Mable BK. 1996. Molecular

Systematics. Sinauer, Sunderland, Massachusetts,

USA. 655 p.

Huelsenbeck JP, Ronquist F. 2001. Mr. Bayes: Bayesian

inference of phylogenetic trees. Bionformatics 177:

754-755.

Kumar S, Stecher G, Li M, Knyaz C, Tamura K. 2018.

MEGA X: Molecular Evolutionary Genetics Analysis

across computing platforms. Molecular Biology and

Evolution 35: 1,547-1,549.

Amphib. Reptile Conserv.

Laurent RF. 1983. Contribucion al conocimiento de

la estructura taxonomica del género Liolaemus

Wiegmann (Iguanidae). Boletin de la Asociacion

Herpetologica Argentina |: 16-18.

Laurent RF. 1985. Segunda contribucidn al conocimiento

de la estructura taxonodmica del género Liolaemus

Wiegmann (Iguanidae). Cuadernos de Herpetologia

1(6): 1-37

Lobo F, Espinoza RE. 1999. Two new cryptic species

of Liolaemus (Iguania: Tropiduridae) from

northwestern Argentina: resolution of the purported

reproductive bimodality of Liolaemus alticolor.

Copeia 1999(1): 122-140.

Lobo F, Quinteros S, Diaz JM. 2007. Description ofanew

species of the Liolaemus alticolor group (Iguania:

Liolaemidae) from Cuzco, Peru. Herpetologica 63:

537-543.

Maddison W, Maddison D. 2017. Mesquite: a modular

system for evolutionary analysis. (Version 3.2).

Available: http://mesquiteproject.org [Accessed: 5

May 2020].

Morando M, Avila LJ, Baker J, Sites JW. 2004.

Phylogeny and phylogeography of the Liolaemus

darwinii complex (Squamata: Liolaemidae):

evidence for introgression and incomplete lineage

sorting. Evolution 58(4): 842-861.

Quinteros AS. 2013. A morphology-based phylogeny

of the Liolaemus alticolor-bibronii group (Iguania:

Liolaemidae). Zootaxa 3670: 1-32.

Rambaut A. 2014. FigTree v1.4.2. Available: http://tree.

bio.ed.ac.uk/software/figtree/ [Accessed: 15 March

2020].

Rambaut A, Drummond AJ, Xie D, Baele G, Suchard

MA. 2018. Posterior summarization in Bayesian

phylogenetics using Tracer 1.7. Systematic Biology

67(5): 901-904.

Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian

phylogenetics inference under mixed models.

Bioinformatics Applications Notes 12: 1,572—1,574.

Shreve B. 1938. A new Liolaemus and two new

Syrrhopus from Peru. Journal of the Washington

Academy of Sciences 28: 404—407.

Shreve B. 1941. Notes on Ecuadorian and Peruvian

reptiles and amphibians with description of new

forms. Proceedings of the New England Zoélogical

Club 18: 71-83.

Uetz P, Freed P, HoSek J. (Editors). 2020. The Reptile

Database. Available: http://www.reptile-database.

org [Accessed: 30 April 2020].

Villegas-Paredes L, Huamani-Valderrama L, Luque-

Fernandez C, Gutiérrez RC, Quiroz AJ, Abdala CS.

2020. Una nueva especie de Liolaemus (Iguania:

Liolaemidae) perteneciente al grupo L. montanus

en las lomas costeras del sur de Pert. Revista de

Biologia Tropical 68(1): 69-86.

Wilcox T, Zwickl D, Heath T. 2002. Phylogenetic

relationship of the Dwarf Boas and a comparison of

May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

Bayesian and bootstrap measures of phylogenetic

support. Molecular Phylogenetics and Evolution

25(2): 361-371.

Amphib. Reptile Conserv.

Juan C. Chaparro is a Peruvian Biologist with extensive experience in studying the fauna of all

the traditional geographic regions of Peru. Juan graduated in Biological Sciences from Universidad

Nacional Pedro Ruiz Gallo, Lambayeque, Peru; received a Master’s degree in Biodiversity in Tropical

Areas and Conservation in 2013, from an institutional consortium of the International University

of Menendez Pelayo (UIMP-Spain), Universidad Tecnolégica Indoamérica (UTI-Ecuador), and

Consejo Superior de Investigaciones Cientificas (CSIC-Spain). He is currently the president of the

Herpetological Association of Peru (AHP), director and curator of the Herpetological Collection

of the Museo de Biodiversidad del Peru (MUBI, https://mubi-peru.org/herpetologia/p-mubi), and

he works as a consultant in environmental studies. Juan has authored or co-authored 50 peer-

reviewed scientific papers, notes, book chapters, and books on fauna, especially in herpetology and

arachnology, on topics such as taxonomy, biodiversity, systematics, phylogeny, conservation, and

biogeography in South America. He is interested in those topics, as well as life history, distributional

patterns, and evolution using amphibian and reptiles as biological models. Four species have been

named in his honor: Phyllomedusa chaparroi (Amphibia), Phrynopus chaparroi (Amphibia),

Hadruroides juanchaparroi (Arachnida), and Chlorota chaparroi (Insecta).

Aaron J. Quiroz graduated in Biological Science, and is currently a Research Associate of the

Museum of Natural History of the National University of San Augustin, Arequipa, Peru. Aaron

is a co-author and collaborator on publications which focus on the taxonomy and conservation of

amphibians and reptiles in Peru. He is currently developing a career as an independent professional

in the direction and design of amphibian and reptile research and conservation projects.

Luis Mamani is a Biologist who graduated from the Universidad Nacional de San Antonio Abad

del Cusco, Peru, and he obtained an M.Sc. degree from the Universidad de Concepcién (UdeC) in

Chile. Luis is currently a researcher at the Museo de Biodiversidad del Peru (MUBI) and the Museo

de Historia Natural de la Universidad Nacional de San Antonio Abad del Cusco (MHNC). His

current research includes systematics, taxonomy, and biogeography of lizards from the Cordillera

de los Andes.

Roberto C. Gutiérrez is a Biologist who graduated from the National University of San Augustin

de Arequipa of Peru. Roberto is currently the Curator and Principal Researcher of the Herpetological

Collection, Museum of Natural History, National University of San Augustin de Arequipa, Peru,

and Vice President and Founding Member of the Herpetological Association of Peru (AHP). He is

interested in the herpetofauna of the tropical Andes and the coastal desert, with a special focus on

lizards of genus Liolaemus, developing studies in the systematics of amphibians and reptiles, ecology,

and conservation. Roberto has conducted several biodiversity inventories, biological assessments,

and biodiversity monitoring programs, and is currently working at the Natural Protected Areas

Service of the Peruvian Ministry of Environment.

Peter Condori is a Biologist who graduated from the Universidad Nacional de San Antonio Abad

del Cusco (Peru). Peter is currently a researcher at Museo de Biodiversidad del Peru (MUBI) and

Museo de Historia Natural de la Universidad Nacional de San Antonio Abad del Cusco (MHNC).

His current research interests include systematics, taxonomy, and biogeography of the lizards and

amphibians from the Cordillera de los Andes.

61 May 2020 | Volume 14 | Number 2 | e238

Amphib. Reptile Conserv.

New Liolaemus species from Peru

Ignacio de la Riva is a tenured Scientist and Curator of Herpetology at the Museo Nacional de

Ciencias Naturales-High Council of Scientific Research (MNCN-CSIC) in Madrid, Spain. Ignacio

graduated and obtained his Ph.D. in Biology from the Universidad Complutense (Madrid, Spain),

before taking postdoctoral and visiting scholar positions at The University of Kansas (USA) and

James Cook University (Australia). His main lines of research include systematics, life history,

thermal ecology, biogeography, and the roles of emergent diseases and climatic change in the

conservation of several groups of tropical amphibians and reptiles.

Gabriela Herrera-Judarez studied Biology at the Universidad Nacional de San Agustin de

Arequipa, Peru. Gabriela has participated in several biodiversity inventories, biological assessments,

and biodiversity monitoring programs. She is interested in taxonomy, natural history, ecology,

and conservation of mammals and herpetofauna; and is currently a researcher at the Museo de

Biodiversidad del Per (MUBI).

José Cerdejia is a Biologist who graduated from the Universidad Nacional de San Agustin de

Arequipa (Peru), and a researcher at Museo de Historia Natural de la Universidad Nacional de

San Agustin de Arequipa (MUSA) in Peru. José’s research includes systematics, taxonomy, and

biogeography of Lepidoptera, but with recent interest in the taxonomy and ecology of the genus

Liolaemus in southern Peru.

Luis P. Arapa is a Biologist who graduated from the Universidad Nacional San Agustin de

Arequipa (UNSA), Peru, in 2019, and is now a volunteer researcher at the Museum of Natural

History of the UNSA (MUSA). For his degree thesis, Luis studied the diversity of amphibians and

reptiles in Protected Natural Areas of the coastal desert of southern Peru. He is currently working on

environmental assessments of coastal and highland areas, and his research interest is the taxonomy

of reptiles, focusing on lizards.

Cristian S. Abdala is an Argentinian Biologist, researcher at CONICET, and professor at the

National University of Tucuman in Argentina. Cristian received his Ph.D. degree from the

Universidad Nacional de Tucuman (UNT), and is a herpetologist with extensive experience in the

taxonomy, phylogeny, and conservation of Liolaemus lizards. He has authored or co-authored over

70 peer-reviewed papers and books on herpetology, including the descriptions of 49 recognized

lizard species, mainly in genus Liolaemus. One species, Liolemus abdalai, has been named in his

honor. He has conducted several expeditions through Patagonia, the high Andes, Puna, and salt flats

of Argentina, Chile, Bolivia, and Peru. Since 2016, Christian has been the president of the Argentine

Herpetological Association.

62 May 2020 | Volume 14 | Number 2 | e238

Chaparro et al.

Appendix. Specimens examined.

Liolaemus annectens (n= 15): PERU. Arequipa: Sumbay, MUSA 4114, 4265-66; Caylloma, MUSA 4344-4348, MUSA 1591-97.

Liolaemus etheridgei (n = 17): PERU. Arequipa: Cabrerias, Cayma, MUSA 501; Cerro Uyupampa, Sabandia, MUSA 549-54;

Monte Riberefio de la Quebrada de Tilumpaya Chiguata. Pocsi, MUSA 1113-14, 1116, 1264-68, 1353; Anexo de Yura Viejo, Yura,

MUSA 1229.

Liolaemus evaristoi (n = 16): PERU. Huancavelica: Los Libertadores, Pilpichaca, Huaytara, MUSA 2841 (holotype), 2781-85,

2840, 2842—45, MUBI 10474—78 (paratypes).

Liolaemus insolitus (n = 9): PERU. Arequipa: Quebrada Quialaque, Lomas de Challascapa, Mejia, Dean Valdivia Islay, MUSA

313-315, 320-324; Lomas de Mejia, Dean Valdivia Islay, MUSA 448.

Liolaemus melanogaster (n = 12): PERU. Arequipa: Laguna de Corococha, Orcopampa, MUSA 372-376; Huancavelica:

Huancavelica, 1 km Southwest of Betania, MUSA 2762-2767. Ayacucho: 45 km East of Puquio, FML 2491 (paratype).

Liolaemus ortizi (n= 3): PERU. Cusco: Huacoto, MUSA-CSA 1432; Santa Barbara, MUSA 1443, 1511.

Liolaemus poconchilensis (n = 4): PERU. Tacna: MUSA 1428-29, MUSA 1638-39.

Liolaemus polystictus (n = 13): PERU. Huancavelica: Mountain close of Rumichaca, Pilpichaca, MUSA 1337-1338; Santa Inés,

Castrovirreyna, MUSA 2448-2457; Santa Inés, FML 1683 (paratype).

Liolaemus robustus (n = 11): PERU. Lima: Surroundings of Huancaya, Reserva Paisajistica Nor Yauyos Cochas, MUSA 1693-

1702; Junin: Junin, FML 1682 (paratype).

Liolaemus signifer (n = 12): PERU. Puno: Titicaca Lake, 3,840 m, FML 1434; Titicaca Lake, road to Puno, FML 1557; near

Tirapata, MUSA 1415; Huancané, Comunidad Taurahuta, MUSA 1441-43; Huerta Huayara community, 3 km before Puno, MUSA

1483-87.

Liolaemus thomasi (n= 15): PERU. Cusco: After Mahuayani pass, MUSA 1398-1412; Pampacancha, Quispicanchi MUBI 5925.

Liolaemus williamsi (n = 15): PERU. Ayacucho: Surroundings of Pampa Galeras, MUSA 1519-1531, FML 1701 (paratype);

Lucanas, Pampa Galeras, FML 13403 (paratype).

Amphib. Reptile Conserv. 63 May 2020 | Volume 14 | Number 2 | e238

Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 64—72 (e239).

The effect of biological microreserves in a highly anthropized

environment on the biology of Natrix maura (Linnaeus, 1758)

1*Robby M. Drechsler, 7Pablo Vera, ‘Daniel C. Martinez, and ‘Juan S. Monros

'Cavanilles Institute of Biodiversity and Evolutionary Biology, University of Valencia, c/ Catedrdtico José Beltran Martinez 2, 46980 Paterna,

Valencia, SPAIN *SEO/BirdLife, Delegation of the Valencian Community, c/ Tavernes Blanques 29, 46120 Alboraia, SPAIN

Abstract.—Human disturbance in highly anthropized areas is known to have many negative effects on

biodiversity. Due to their close relation to the environment (e.g., thermoregulation) and their limited movement

Capacities, reptiles (and snakes in particular) are excellent bioindicators of environmental quality. In addition,

they carry out essential ecological functions, acting as prey and predator simultaneously. However, studies

focusing on the effects that conservation efforts have on their populations after implementation are scarce.

Hence, the aim of this study was to evaluate the effect of small protected areas on one of the most common

snake species in the Iberian Peninsula, Natrix maura. During 2015 and 2018, samplings were conducted at eight

different points in the Albufera de Valencia Natural Park in Spain, three of them being biological reserves. The

results show that such protected areas have positive effects on the N. maura populations at different levels. In

protected areas, population structures are more complex and the body condition of large individuals is better.

The study also examined predatory pressure, but did not find a significant correlation between the abundance

of predatory birds (herons) and injured or “scared” individuals of N. maura. In conclusion, such areas can

be essential for allowing snakes to maintain their biological cycles, and in some cases even to prevent their

disappearance from highly anthropized environments. Therefore, we strongly recommend the creation of more

protected areas in order to promote the conservation of biodiversity in these areas.

Keywords. Conservation, ecology, human disturbance, snakes, Spain, wetlands

Citation: Drechsler RM, Vera P, Martinez DC, Monrés JS. 2020. The effect of biological microreserves in a highly anthropized environment on the

biology of Natrix maura (Linnaeus, 1758). Amphibian & Reptile Conservation 14(2) [General Section]: 64-72 (e239).

tion 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any

medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced, are

as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 4 June 2019; Accepted: 12 March 2020; Published: 2 June 2020

Introduction

In a more and more anthropized world, the conservation

of biodiversity is becoming an increasingly important

challenge. However, there is an imbalance in the

partitioning of conservation efforts among different

animal groups (Czech et al. 1998; Gerber 2016).

Especially in society in general, poorly accepted groups

tend to receive less attention when planning and carrying

out conservation and management projects (Czech et

al. 1998). One group where this result is most evident

is snakes, as their image is affected very negatively by

many cultural factors, such as religious beliefs and the

general media, leading to a learned fear and aversion

(Ballouard et al. 2013). However, acting as predators and

prey simultaneously, snakes are of vital importance to the

function of many ecosystems. Moreover, given their tight

relationship with environmental conditions and _ their

limited movement capacities, snakes represent an ideal

group for use as bioindicators of environmental quality

and for evaluating the success of conservation measures

already implemented (Beaupre and Douglas 2009).

Natrix maura is one of the most common snake

species of the Iberian fauna, and clearly the most

abundant snake in water-related habitats (Pleguezuelos

et al. 2002). In addition, a great ecological plasticity has

allowed this species to persist in highly human disturbed

areas, such as agricultural areas (Santos 2015). However,

there is still a lack of knowledge regarding the effect of

human disturbances on this species. This is concerning,

as in areas where its population status had been studied

previously, very negative effects of human activity on N.

maura populations have been detected (Miras et al. 2009;

Santos and Llorente 2009; Santos 2015).

The threats N. maura populations are facing are

diverse. In addition to natural factors which negatively

affect N. maura populations, including the improvement

of natural predator populations such as herons (e.g.,

Garrido et al. 2012), the great majority of the threats are

related to human activity. Road mortality is an important

Correspondence. *robbymar@wv.es (RMD), pvera@seo.org (PV); daniel.comes@uv.es (DCM); Juan. Monros@uv.es (JSM)

Amphib. Reptile Conserv.

June 2020 | Volume 14 | Number 2 | e239

Drechsler et al.

Figure 1. Satellite image of the study area, the Albufera de

Valencia Natural Park, obtained from Google Maps (accessed:

25 January 2019), indicating the locations of the eight sampling

points.

factor, especially in wetlands, as roads usually run

parallel to water channels (Llorente et al. 2005; Santos

2015). In addition, the destruction, fragmentation, and

disturbance of N. maura habitats are important threats

(Santos et al. 2002) and, given that it 1s a predator, the

reduction of prey availability has also negative effects on

the species (Filippakopoulou et al. 2014). The massive

use of plant protection products in agriculture and

contaminants being dumped into rivers from urbanization

have noticeably reduced the water quality in this species’

habitats. Given that the prey items of N. maura are

mainly aquatic, studies have shown that NV. maura has a

great capacity for the bioaccumulation of contaminants

(Santos et al. 1999). In addition, ecological alterations

of the ecosystems by the introduction of exotic species

has also exerted direct or indirect negative effects on N.

maura populations (e.g., Alarcos et al. 2009). Finally,

the fear and negative image that ophidians have in the

general population leads to their encounters with humans

often ending with the snakes’ death, even for those that

are absolutely inoffensive (Hailey and Davies 1987). The

effects of all these factors were seen, for example, in the

Ebro Delta of Spain, when Santos and Llorente (2009)

detected a dramatic reduction in the abundance of N.

maura in a span of only 13 years.

As in the Ebro Delta, in the Albufera de Valencia

Natural Park (Spain) the use of the major terrain is for rice

plantations (Sanchez 2008). This intensive agriculture has

promoted the massive use of fertilizers, herbicides, and

insecticides, which in the past decades has led to massive

Amphib. Reptile Conserv.

eutrophication problems and a drastic decrease in general

water quality (e.g., Soria 2006). To confront this problem,

several natural reserves were created between 2006 and

2011 in order to foment biological conservation and the

recovery of water quality. These reserves are former rice

fields which were transformed into artificial wetlands,

recovering the marshy habitats and the associated plant

and animal biodiversity, and restoring (at least in small

areas) the natural state of the Albufera wetland. The

importance of this type of biological reserve has already

been reported for various groups of organisms, including

birds (Rodrigo et al. 2018; Sebastian-Gonzalez et al.

2013), amphibians (Reques 2004), and turtles (Drechsler

et al. 2018).

Fauna conservation studies often focus on “umbrella”

or “indicator” species or zoological groups, for example

birds. However, this implies a simplification of the

ecological relationships and has already been criticized,

as the conservation of one group does not necessarily

entail the conservation of the others (e.g., Figuerola and

Green 2003). In the case of reptiles (with the exception

of turtles with conservation problems), there are very

few studies which evaluate the effects of conservation

measures on their target populations (Shine et al. 1998;

Webb et al. 2005; Schoemaker 2007).

The aim of this study is to evaluate the effect of small

biological reserves in highly anthropized environments

on several aspects of the biology of N. maura, especially

population dynamics. Given that this species is the most

common reptile in aquatic environments in the Iberian

Peninsula, these data will be of great use when comparing

biological quality of coastal wetlands throughout the

Iberian Peninsula, and at the same time, for evaluating the

success of conservation measures that have been taken.

Materials and Methods

Study Area

The study area is the Albufera de Valencia Natural Park,

located about 10 km south of Valencia city in Spain

(39°21718.7"N 0°21°38.4”"W). It is one of the most

important wetlands of the Iberian Peninsula, with an

area of ~21,000 ha. Approximately 3,000 ha correspond

to the central lagoon, while the remaining 18,000

ha mainly correspond to intensive rice fields (Oryza

sativa) [Fig. 1]. The water supply of this environment

has several different origins, including the Jucar

river, rain water from highly urbanized watersheds,

springs from subterranean aquifers, urban or industrial

effluents (depurated wastewater), and irrigation returns

from the agriculture in the area. The study area is

highly anthropized for the agricultural use, so that the

natural marsh vegetation is restricted to canals and the

perimeter of the central lagoon, and is dominated by

Common Reed (Phragmites australis) and Southern

Cattail (Thypha dominguensis).

June 2020 | Volume 14 | Number 2 | e239

Natrix maura in an anthropized environment

Figure 2. Satellite images of the three natural reserves: Pipa (A), Milia (B), and Illa (C), obtained from Google Maps (accessed: 5

December 2019).

Embedded in this anthropized environment are the

three studied biological reserves: Tancat de la Pipa,

Tancat de Milia, and Tancat de Illa (Fig. 2). All three

reserves are former rice fields that have been transformed

into artificial wetlands in order to restore water quality,

foment the conservation of biodiversity, and provide

areas for environmental education. The access to these

areas is restricted to the fulfilment of these specific

objectives. The Tancat de la Pipa (Pipa) is the oldest

and largest reserve, created in 2006 with an area of

about 40 ha, and different types of shallow marshy

habitats have been recreated there, including canals with

dense vegetation on the shores and permanent lagoons.

The other two reserves were created in 2011 and are

characterized by more active management of the water

levels and vegetation. The Tancat de Milia (Milia) has

an intermediate area of 33.4 ha and the Tancat de Illa

(Illa) is the smallest, covering only 16 ha. In addition to

the reserves, five additional points were sampled: (i) the

Catarroja port and the surrounding rice fields (AR), an

area far away from the central lagoon and high anthropic

pressure due to the presence of bars, frequently used

roads, fishing, and boat traffic; (11) rice fields near the

Silla port (LR), an area close to the central lagoon and

with less anthropic pressure; (111) the surroundings of El

Palmar (Palmar), an area with a relatively high anthropic

pressure, abundant vehicle traffic, and fishing activity;

(iv) Socarrada (SC), an area close to Illa; and (v) the

Muntanyeta dels Sants (MS), a small urbanized area

embedded in the rice field environment (Fig. 1).

Fieldwork

Between March and November in both 2015 and 2018,

one researcher carried out random searches for NV. maura

for 1.5—2 hours at the eight sampling points. All censuses

were carried out by the same researcher and were initiated

2-3 hours after sunrise. The sampling was repeated at

each point every two weeks in 2015 and once per week

in 2018. In each survey, adverse weather conditions were

annotated, such as clouds, wind, or rain. All individuals

of N. maura observed were recorded, indicating whether

Amphib. Reptile Conserv.

they were adults or juveniles when possible. Individuals

were considered adults based on SVL > 250 mm (males)

and SVL > 300 mm (females) [Santos 2015]. In parallel

to the counts, whenever possible, all individuals were

captured by hand and placed in individual cloth bags

until posterior measurements could be taken. Once the

censuses were finished, the captured individuals were

processed, measuring basic biometry with the use of a

measuring tape and a weighing scale: snout-vent length

(SVL, to the nearest 0.5 mm), total length (to the nearest

0.5 mm), and weight (with a precision of 0.01 g). In

addition, records were made of the presence of injuries,

scars, or a broken tail, which could be interpreted as the

result of a predation event, or if there were gut contents

(these individuals were excluded from body condition

analysis). Each individual was marked with a unique

code of small notches on the ventral scales in order to

control for recaptures (Lang 1992). After measurements

were taken, each individual was released at the capture

point.

Data Analysis

To determine the abundance, the numbers of individuals

observed were standardized for the duration of each

census (in hours), obtaining the values in individuals/

hour. Data were grouped by areas, years, seasons (as

“Spring” for March—May, “Summer” for June-August,

and “Autumn” for September—November), whether the

area was protected or not, and whether it was close to

the central lagoon (<2 km) or far from the lagoon (>

2 km). In order to determine statistical significance, a

Poisson Logit Linear Regression Model (Poisson GLM)

was performed. Effects of the presence of clouds or other

adverse meteorological conditions on the snake censuses

were eliminated by comparing the snake counts on the

days with such conditions with the days immediately

before and after which had favorable conditions (Poisson

GLM, z = 1.043, p =0.297).

In order to analyze predation risk, a Binomial Logit

Linear Regression Model was performed considering the

following variables: SVL, Type of dorsal coloration (zig-

June 2020 | Volume 14 | Number 2 | e239

Drechsler et al.

zag vs. bilineata, Santos et al. 2017), Sex, Year, and Site.

For this analysis, individuals with SVL < 250 mm (n =

305) were excluded, because large individuals are known

to be more prone to predation (Santos et al. 2011). In

addition, in the case of the reserves, the proportions of

injured individuals were calculated (Santos et al. 2011),

and these data were compared with the abundances of

birds that are potential predators of N. maura, especially

large herons (Great Egret, Egretta alba; Grey Heron,

Ardea cinereal,; and Purple Heron, Ardea purpurea).

The number of birds was obtained by conducting weekly

counts during the same period as each snake census.

The bird counts were accomplished with binoculars

and telescopes, always during the first four hours after

sunrise and avoiding adverse meteorological conditions

such as wind, rain, or fog. In the cases of Pipa and Illa,

these counts were made on foot, recording all individuals

seen; while in the case of Milia, with less vegetation

surrounding open water areas, counts were made from

inside a parked car.

Data from both years were used for constructing the

population pyramids, without discriminating periods.

The captured individuals for each area were grouped

in intervals of 150 mm SVL and the proportions

of individuals corresponding to each interval were

calculated with respect to the total area. A chi-square

test was performed to determine statistically significant

differences.

Finally, to study the body condition, the relation

between weight (g) and size (SVL, in mm) of the

individuals was represented and the tendency line was

drawn. Individuals with an SVL above 600 mm were

excluded, as such individuals appeared in only two

areas (Pipa and Milia). The tendency line of the global

data cloud served as the reference function. Afterwards,

the data were separated by areas and the correlation

coefficient (R) and determination coefficient (R?) of

the corresponding data cloud were calculated with the

reference function. A low coefficient for an area indicates

that the gain of weight of the individuals in this area does

not follow the mean increase of the general population.

In order to analyze the evolution of body condition in

the different areas in greater detail, the SVL and weight

values were log-transformed and the residuals of the

correlation between these two variables were calculated.

To determine statistical differences between protected

and non-protected areas for the residuals, Kuskal-Wallis

tests were performed for each SVL interval.

Results

A total of 721 snakes was recorded in 459 sampled hours.

The abundance analysis showed an almost significant

reduction of abundance values between 2015 (1.76 +

1.80 ind/h) and 2018 (1.45 + 1.55 ind/h) [Table 1]. This

decrease is especially evident in Illa for both adults and

juveniles and in AR for adults (Fig. 3). In addition, areas

Amphib. Reptile Conserv.

a 230

2.00

Ah 1.80

1.60

2 140

= a0

S 1.00

S 0.80

= 0.60

0.40

0.20

0.00 Li yy Yip

LR Tila Milia Pipa

2.00

1.80 B

~ 1.60

= 1.40

Ef 1.20

Z 0.80

B 060

0.40

0.20

0.00 Z li

L Illa Milia Pipa

2.00

1.80

1.60

= 140

ae)

| 120

—

2 1.00

3 080

te 0.60

<= o40

0.20

0.00 ZZ Li ey

LR Illa Milia Pipa

Figure 3. Abundances of Natrix maura in the different study

areas in 2015 (filled bars) and 2018 (striped bars), for the total

population (A), adults only (B), and juveniles only (C).

located near the central lagoon and in protected areas

showed abundances which seem to be significantly lower

than in areas far from the lagoon and non-protected areas,

respectively (Table 1).

Significant differences were found in the distributions

of sizes between areas (X? = 362.2, df=35, p< 0.001). The

population pyramids (Fig. 4) show that the most mature

populations, 1.e., those with the highest proportions of

large individuals, correspond to Pipa and Milia, which

presented ~10—15% of individuals above 600 mm SVL.

In the other areas, the populations are mainly formed

Table 1. Results of the GLM for Poisson distributions

evaluating the effect on the abundance of N. maura of five

different variables: Year, Season, Area, Protected/non-protected

and Close/far from the central lagoon.

; Coefficient

vocab estimate P

Year -0.318 -1.937. 0.053

Season -0.122 -1.329 0.184

Area 0.315 3.843 <0.001

Protected/non-protected -0.775 -3.364 <0.001

Close/far from lagoon 0.784 3.835 <0.001

June 2020 | Volume 14 | Number 2 | e239

Natrix maura in an anthropized environment

Table 2. Values of the correlation and determination coefficients for the increase of weight with body size with respect to the

reference function for each sampling area.

Sampling area

SC 0.982

LR 0.953

AR 0.950

Pipa 0.942

Palmar 0.939

Milia 0.929

Illa 0.888

MS 0.886

by immature individuals, specifically individuals with

150-299 mm SVL. The second most abundant group had

300-449 mm SVL, and is formed by young adults. In

these areas, very few individuals (~5%) reached an SVL

above 450 mm.

The results obtained regarding predation indicate that

none of the variables considered significantly affected

predation risk (z = 1.929, p = 0.054, for SVL; z= -0.774,

p = 0.439, for Coloration; z = -1.782, p = 0.075, for

Year; z = -0.326, p = 0.744, for Sex; and z = -0.857, p =

0.392, for Site). Significant correlations were not found

between either the abundance of potential predator birds

(expressed as number of birds detected per census and

ha of the reserve) or the proportion of injured N. maura

individuals (Lineal Regression Model, F, ,, = 4.424, p =

0.126, considering only the three mentioned large heron

species, and Lineal Regression Model, F, |, = 2.757, p =

0.195 for all heron species).

Pipa

Palmar

Milia

Illa

ag!

hao

co)

Correlation coefficient (R)

Determination coefficient (R’)

0.965

0.908

0.903

0.888

0.881

0.863

0.789

0.785

The analysis of body condition showed that in the

sampled population of N. maura in the Albufera de

Valencia Natural Park, weight (W, in g) varies with body

size (SVL, in mm) according to the following function

(n= 649):

W- 0.787328 Bon ae (R° z 0.95)

The comparison of the data clouds of each area with

this reference function showed that MS and Illa were

the areas with the lowest correlation coefficients for

these variables, and SC and LR were the areas with

the highest correlation values (Table 2). Among the

biological reserves, Illa was the least correlated and

Pipa the most correlated. Comparing protected areas

with non-protected areas, a clear difference was seen

regarding body condition, with large individuals (>

450 mm SVL) presenting significantly higher body

ca)

co)

ras)

Figure 4. Population pyramids of each area representing the proportion of individuals for each SVL interval in the population:

< 150 mm (Bi. 150-299 mm (@): 300-449 mm ("= ); 450-599 mm (—_); 600-749 mm (__—); and >750 mm (___).

Amphib. Reptile Conserv.

June 2020 | Volume 14 | Number 2 | e239

Drechsler et al.

Table 3. Sample size (in parentheses), and values of mean + standard deviation of weight (in g) for the individuals in different body

size ranges for each area, comparing the protected (Illa, Milia, and Pipa) and non-protected areas (AR, LR, MS, Palmar, and SC);

and the results of the Kruskal-Wallis test comparing the residuals of the correlations between the log-transformed SVL and weight

data from protected and non-protected areas.

SVL (mm)

< 150 150-299 300-449 450-599

SC (22) 2.31 +0.51 (6) 3.67 + 0.97 (2) 26.50 + 4.38 (1) 80.80 + 0.00

LR (31) 2.74 +0.55 (71) 6.68 + 5.12 (19) 31.51 + 10.98 (2) 91.92 + 54.28

AR (15) 2.51 +0.33 (72) 9.50 + 6.04 (42) 30.04 + 10.34 (5) 60.28 + 9.75

Pipa (41) 2.53 +0.51 (67) 6.45 + 4.68 (28) 34.064 14.15 (44) 131.70 + 43.04

Palmar (5) 2.96 + 0.61 (9) 11.09 + 7.49 (12): 37-57 4 21.26 -

Milia (4) 2.39+0.14 (24) 8.16 + 5.84 (8) 30.08 + 7.81 (5) 100.71 £35.15

Illa (31) 2.64 + 0.64 (53) 6.92 + 4.98 (19) 35.47 + 14.49 (2) 91.17 + 18.17

MS (1) 2.30 + 0.00 (6) 15.39 + 7.48 (4) 30.64 + 4.85 ~

Protected (76) 2.57 + 0.56 (144) 6.91 + 5.00 (55) 33.974 13.45 = (51) 127.14 + 42.88

Non-protected

(74) 2.57 £0.53

(164) 8.38 + 6.05

(79) 31.48 + 12.51

Kruskal-Wallis results for residuals

(8) 70.75 + 26.38

Mean (Protected) (76) 0.047 + 0.224

Mean (Non-protected) (74) 0.099 + 0.213

DF 1

xX 3.102

P 0.078

condition values in protected areas than in non-

protected areas (Table 3).

Discussion

The results reported here show that biological reserves

are local areas of importance for the conservation of

snakes in the Albufera de Valencia Natural Park. The

population structure in protected areas is more complex,

individuals reach larger body sizes, and their body

condition is better in comparison to non-protected areas.

The effect of such reserves as natural environments can

noticeably reduce some threats and pressures on snakes,

such as agricultural activities and human presence, which

impart higher mortality (due to machinery, roadkill, or

ageression caused by the aversion to snakes in many

people) [e.g., Whitaker and Shine 2000]. It is precisely

expected that this mortality is accentuated in larger

individuals (Shine and Koenig 2001). Moreover, in non-

protected areas, the low environmental quality and human

disturbance also have been shown to negatively affect the

availability of prey, such as fish and amphibians (Lawler

2001; Marco 2002). Finally, it has been demonstrated that

the bioaccumulation rate of toxics in N. maura is more

important in large individuals than in small ones (Santos

et al. 1999; Lemaire et al. 2018). While not directly

measured here, all these factors could play synergistic

roles in explaining the results of this study. Although a

decrease in abundance was found in protected areas in

relation to non-protected ones, this could be related to

differences in detectability (e.g., dense vegetation in

reserves, abundant junk items serving as hiding spots

in anthropized areas, etc.). This could also explain the

differences between areas close to and far from the

Amphib. Reptile Conserv.

(144) -0.045£0.211

(164) -0.011 + 0.225

(55) -0.081 + 0.225

(79) -0.103 + 0.167

(51) 0.109 + 0.262

(8) -0.164 + 0.163

1 1 1

1.782 0.056 Halt 7

0.182 0.812 0.007

lagoon, as the latter tended to be more anthropized.

The results of this study also indicate that

characteristics of protected areas, such as size and

management, could play important roles in their ability

to conserve NV. maura populations, as not all protected

areas showed the same results. For instance, although Illa

is a protected area, its reduced size and the fact that it

is surrounded by a heavily-travelled road could prevent

the development of a population as complex as the ones

observed in Milia and Pipa. Another possible explanation

could be the presence of predators, especially birds, and

in particular herons. These birds also tend to congregate

in the reserves, using it for reproduction, feeding, and as

temporary roosts during migration (e.g., Gosalvez et al.

2012; Pérez-Granados et al. 2013). Many birds, especially

herons, predate on N. maura (e.g., Amat and Herrera

1977, Gonzalez and Gonzalez-Solis 1990). In fact, the

proportions of injured N. maura individuals were higher

in Pipa and Milia than in the other areas. However, in this

study we did not find evidence for a significant effect of

bird density on the status of NV. maura populations, and

in most cases the correlations were even negative. This

could be explained by the greater density of vegetation

in the protected areas, providing the snakes with more

refuges and possibilities for hiding and escaping from

predators. We also have to take into account the effect of

the “bias of the survivor,” as we can only see and measure

what correspond to failed predation attempts (Gregory

and Isaac 2005). Finally, the coloration of the snakes has

been found to have a significant effect on predator attack

rate, where in dense habitats individuals with a zig-zag

pattern (Batesian mimicry) are attacked significantly less

than individuals with striped dorsal patterns (Santos et

al. 2017). However, this study did not detect a significant

June 2020 | Volume 14 | Number 2 | e239

Natrix maura in an anthropized environment

effect of dorsal coloration on predation risk.

The results presented here do not allow for inferences

on the long-term tendencies of the populations in the

studied areas or the general population of N. maura in the

Albufera de Valencia Natural Park. However, the fact that

a significant decrease of snake abundance was detected

between 2015 and 2018 can be interpreted as an indicator

that the populations may be suffering a decline. In fact,

snake populations in general are known to be declining

considerably, and there are indications that in Europe

the most probable causes are habitat deterioration, low

prey availability, and pollution (Reading et al. 2010). In

the case of the Iberian wetlands, industrialization, the

modernization of sowing and harvesting techniques, the

massive use of fertilizers, habitat loss and fragmentation,

and other factors have already caused an important

decrease in snake populations in the Ebro Delta (Santos

and Llorente 2009), an area very similar to the area of this

study. Specifically, the authors detected a decline in the

Ebro Delta of about 50%, in some cases even 100%, of

the N. maura populations in only 13 years. However, the

lack of data for the Albufera de Valencia Natural Park on

the status of the NV. maura population in the past makes

it impossible to confirm whether a similar situation is

happening in the Albufera de Valencia Natural Park.

In conclusion, the existence of biological reserves in

this highly anthropized environment has had a positive

effect on N. maura populations on different levels.

Considering that human disturbance and the effects of

intensive agriculture (landscape homogenization, loss of

boundaries, temporal disturbance, and use of chemical

products) are growing stronger and that a general decline

of snake populations has already been described, we could

even affirm that such protected areas are essential for

snakes to maintain their biological cycles and, in extreme

cases, even to avoid the disappearance of populations of

species like N. maura in this type of environment. Taking

all of these considerations into account, it is evident that the

creation of more such protected areas is important and has

to be promoted in highly anthropized areas, with the aim

of conserving and protecting snakes, a group of predators

that is essential for the function of many ecosystems and to

which it is increasingly necessary to pay attention.

Acknowledgements.—We would like to thank the

management teams of the biological reserves Tancat de

la Pipa (SEO/BirdLife and Acci6é Ecologista Agro), and

Tancat de Milia and Illa (PAVAGUA and Global Nature

Foundation), for allowing us to collect data inside the

reserves and supporting us during the fieldwork. We

would also like to thank the management staff of the

Albufera de Valencia Natural Park for the sampling

permits. The main author, Robby Drechsler, is supported

by a Val I+D predoctoral grant (ACIF/2016/331) from

the Ministry of Education, Investigation, Culture and

Sport of the Regional Government of Valencia and the

European Social Fund.

Amphib. Reptile Conserv.

Literature Cited

Alarcos G, Alvarez-Collado F, Flechoso ME, Madrigal

J, Lizana M. 2009. Peces exoticos de la familia

Centrarchidae, un peligro para Natrix maura. Boletin

de la Asociacion Herpetologica Espatiola 20: 95-97.

Amat JA, Herrera CM. 1977. Alimentacion de la garza

imperial (Ardea purpurea) en las marismas del

Guadalquivir durante el periodo de_ nidificacion.

Ardeola 24: 95-104.

Ballouard JM, Ajtic R, Balint H, Brito JC, Crnobrnja-

Isailovic J, Desmonts D, ElMouden EH, Erdogan

M, Feriche M, Pleguezuelos JM, et al. 2013.

Schoolchildren and one of the most unpopular

animals: are they ready to protect snakes? Anthrozoés

26: 93-109.

Beaupre SJ, Douglas LE. 2009. Snakes as indicators and

monitors of ecosystem properties. Pp. 244-261 In:

Snakes: Ecology and Conservation. Editors, Mullin

SJ, Seigel RA. Comstock Associates, Ithaca, New

York, USA. 365 p.

Czech B, Krausman PB, Borkhataria R. 1998. Social

construction, political power, and the allocation of

benefits to endangered species. Conservation Biology

12: 1,103—1,112.

Drechsler RM, Fernandez LM, Lassalle M, Monros JS.

2018. Movement patterns and home range sizes of

translocated European Pond Turtles (Emys orbicularis,

Linnaeus, 1758) in eastern Spain. Herpetological

Conservation and Biology 13: |-9.

Figuerola J, Green AJ. 2003. Aves acuaticas como

bioindicadores en los humedales. Pp. 47-60 In:

Ecologia, Manejo y Conservacion de los Humedales.

Instituto de Estudios Almerienses, Almeria, Spain.

237 DP:

Filippakopoulou A, Santos X, Feriche M, Pleguezuelos

JM, Llorente GA. 2014. Effect of prey availability

on growth-rate trajectories of an aquatic predator,

the viperine snake, Natrix maura. Basic and Applied

Herpetology 28: 35—S0.

Garrido JR, Molina B, Del Moral JC. 2012. Las Garzas

en Espatia, Poblacion Reproductora e Invernante

en 2010-2011 y Método de Censo. SEO/BirdLife,

Madrid, Spain. 213 p.

Gerber LR. 2016. Conservation triage or injurious

neglect in endangered species recovery. Proceedings

of the National Academy of Sciences of the United

States of America 113: 3,563—3,566.

Gonzalez M, Gonzalez-Solis J. 1990. Datos sobre

la alimentacion de ardeidos en el delta del Ebro.

Miscel-lania Zoologica 14: 240-244.

Gosalvez R, Gil-Delgado JA, Vives-Ferrandiz C, Sanchez

G, Florin M. 2012. Seguimiento de aves acuaticas

amenazadas en lagunas de la Reserva de la Biosfera

de La Mancha Humeda (Espafia Central). Poligonos.

Revista de Geografia 22: 89-122.

Gregory PT, Isaac LA. 2005. Close encounters of the

June 2020 | Volume 14 | Number 2 | e239

Drechsler et al.

worst kind: patterns of injury in a population of grass

snakes (Natrix natrix). Herpetological Journal 15:

213-219.

Hailey A, Davies PMC. 1987. Growth, movement, and

population dynamics of Natrix maura in a drying

river. Herpetological Journal |: 185-194.

Lang M. 1992. A review of techniques for marking

snakes. Smithsonian Herpetological Information

Service 90: 1-19.

Lawler SP. 2001. Rice fields as temporary wetlands: a

review. [srael Journal of Zoology 47: 513-528.

Lemaire J, Bustamante P, Olivier A, Lourdais O,

Michaud B, Boissinot A, Galan P, Brischoux F. 2018.

Determinants of mercury contamination in viperine

snakes, Natrix maura, in Western Europe. Science of

the Total Environment 635: 20-25.

Llorente GA, Garriga N, Montori A, Richter-Boix A,

Santos X. 2005. Incidéncia de les carreteres sobre

els amfibis i réptils de Catalunya. Departament de

Politica Territorial 1 Obres Publiques, Generalitat de

Catalunya, Barcelona, Spain. 167 p.

Marco A. 2002. Contaminacion global por nitr6geno y

declive de anfibios. Revista Espatiola de Herpetologia

iota My:

Miras J, Cheylan M, Said Nouira M, Joger U, Sa-Sousa

P, Pérez-Mellado V, Schmidt B, Meyer A, Sindaco R,

Romano A, Martinez-Solano I. 2009. Natrix maura.

The IUCN Red List of Threatened Species 2009:

e.T61538A12510365.

Pérez-Granados C, Serrano-Davies E, Noguerales

V. 2013. Avifauna acuatica invernante en lagunas

artificiales: la laguna de Meco. Revista Catalana

d’Ornitologia 29: 60-69.

Pleguezuelos JM, Marquez R, Lizana M. 2002. Atlas

y Libro Rojo de los Anfibios y Reptiles de Espafia.

Organismo Autonomo Parques Nacionales, Ministerio

de Medio Ambiente, Madrid, Spain. 587 p.

Reading CJ, Luiselli LM, Akani GC, Bonnet X, Amori

G, Ballouard JM, Filippi E, Naulleau G, Pearson D,

Rugiero L. 2010. Are snake populations in widespread

decline? Biology Letters 6: 777-780.

Reques R. 2004. Habitats reproductivos de anfibios en la

provincia de Cadiz: perspectivas para su conservacion.

Revista de la Sociedad Gaditana de Historia Natural

4: 83-103.

Rodrigo MA, Valentin A, Claros J, Moreno L, Segura

M, Lassalle M, Vera P. 2018. Assessing the effect of

emergent vegetation in a surface-flow constructed

wetland on eutrophication reversion and biodiversity

enhancement. Ecological Engineering 113: 74-87.

Sanchez OEA. 2008. Evaluacion de riesgos ambientales

del uso de plaguicidas empleados en el cultivo del

arroz en el Parque Natural de La Albufera de Valencia.

Ph.D. Dissertation, Universidad Politécnica de

Valencia, Valencia, Spain. xii + 249 p.

Amphib. Reptile Conserv.

Santos X. 2015. Culebra viperina — Natrix maura. In:

Enciclopedia Virtual de los Vertebrados Espafoles.

Museo Nacional de Ciencias Naturales, Madrid,

Spain. Available: http://www. vertebradosibericos.org

[accessed: 5 December 2019].

Santos X, Llorente GA. 2009. Decline of a common

reptile: case study of the viperine snake, Natrix maura,

in a Mediterranean wetland. Acta Herpetologica 4:

161-169.

Santos X, Azor JS, Cortés S, Rodriguez E, Larios J,

Pleguezuelos JM. 2017. Ecological significance of

dorsal polymorphism in a Batesian mimic snake.

Current Zoology 64: 745-753.

Santos X, Feriche M, Leon R, Filippakopoulou A, Vidal-

Garcia M, Llorente GA, Pleguezuelos JM. 2011.

Tail breakage frequency as an indicator of predation

risk for the aquatic snake, Natrix maura. Amphibia-

Reptilia 32: 375-383.

Santos X, Llorente GA, Montori A, Carretero MA. 2002.

Natrix maura (Linnaeus, 1758). Pp. 289-291 In: Atlas

y Libro Rojo de los Anfibios y Reptiles de Espania.

Editors, Pleguezuelos JM, Marquez R, Lizana M.

Direccion General de Conservacion de la Naturaleza,

Asociacion Herpetolégica Espafiola, Madrid, Spain.

587 p.

Santos X, Pastor D, Llorente GA, Albaigés J. 1999.

Organochlorine levels in viperine snake, Natrix

maura, carcasses from the Ebro Delta (NE Spain):

sexual and size-related differences. Chemosphere 39:

2,641-—2,650.

Sebastian-Gonzalez E, Botella F, Sdanchez-Zapata

JA. 2013. Patrones, procesos y conservacion de

comunidades: el caso de las aves acuaticas en

humedales artificiales. Revista Catalana d’Ornitologia

29: 75-92.

Shine R, Koenig J. 2001. Snakes in the garden: an analysis

of reptiles “rescued” by community-based wildlife

carers. Biological Conservation 102: 271—283.

Shine R, Webb JK, Fitzgerald M, Sumner J. 1998. The

impact of bush-rock removal on an endangered snake

species, Hoplocephalus bungaroides (Serpentes:

Elapidae). Wildlife Research 25: 285-295.

Shoemaker KT. 2007. Habitat manipulation as a viable

strategy for the conservation of the Massasauga

Rattlesnake in New York State. Ph.D. Dissertation,

State University of New York, College of

Environmental Science and Forestry, Syracuse, New

York, USA. x + 99 p.

Soria JM. 2006. Past, present, and future of La Albufera

of Valencia Natural Park. Limnetica 25: 135-142.

Webb JK, Shine R, Pringle RM. 2005. Canopy removal

restores habitat quality for an endangered snake in a

fire suppressed landscape. Copeia 2005(4): 894-900.

Whitaker PB, Shine R. 2000. Sources of mortality of

large elapid snakes in an agricultural landscape.

Journal of Herpetology 121-128.

June 2020 | Volume 14 | Number 2 | e239

Amphib. Reptile Conserv.

Natrix maura in an anthropized environment

Robby M. Drechsler studied biology at the University of the Balearic Islands, Spain. Afterwards,

he acquired his Master’s degree in Biodiversity, Conservation, and Evolution at the University

of Valencia, Spain. He obtained his Ph.D. in Biodiversity and Evolutionary Biology at the same

university, focusing his research on the ecology of reptiles in the Albufera de Valencia Natural Park.

Pablo Vera received his Ph.D. in Biodiversity from the University of Valencia (Spain), and is a

specialist in the management and conservation of wetlands. Pablo leads bird monitoring programs

in the Valencian Community and his research focuses on the role of rice fields and their management

on general biodiversity, especially aquatic birds.

Daniel C. Martinez studied biology at the University of Valencia (Spain) and is currently working

on his Master’s degree in Biodiversity, Conservation, and Evolution, with his work focusing

especially on reptiles.

Juan S. Monrés is an Associate Professor of Ecology at the University of Valencia (Spain) and

Director of the Cavanilles Institute of Biodiversity and Evolutionary Biology. He received his

B.A. in Biology and Ph.D. in Biology from the University of Valencia, studying the ecology of

the Montpellier Snake (Malpolon monspessulanus) in the Eastern Iberian peninsula. Currently,

his studies involve the population ecology of terrestrial vertebrates, mainly birds, focused on the

conservation of small populations. Juan also works on the ecology of other vertebrates, including

mammals (bats), amphibians living in temporal ponds, and reptiles in the Albufera de Valencia

Natural Park.

72 June 2020 | Volume 14 | Number 2 | e239

Abronia fuscolabialis (Tihen 1944). The Mount Zempoaltepec Arboreal Alligator Lizard has an EVS of 18 (Johnson et al. 2017) and

its distribution is restricted to the Sierra Madre de Oaxaca of Oaxaca, Mexico (Mata-Silva et al. 2015). This species is poorly known

since it is represented by only five museum specimens from two different localities in the Sierra Madre de Oaxaca (Cerro Pelon

and Cerro Zempoaltepetl). This individual was observed and photographed in a third (new) locality in the Sierra Juarez of Oaxaca,

Mexico. Photo by César Mayoral Halla.

ok ee, Foe 7

Official journal website:

amphibian-reptile-conservation.org

Amphibian & Reptile Conservation

14(2) [General Section]: 73-132 (e240).

Perspective

Conserving the Mesoamerican herpetofauna: the most critical

case of the priority level one endemic species

‘Eli Garcia-Padilla, 7Dominic L. DeSantis, *Arturo Rocha, 7Vicente Mata-Silva,

2Jerry D. Johnson, and ***Larry David Wilson

‘Oaxaca de Judrez, Oaxaca 68023, MEXICO *Department of Biological Sciences, The University of Texas at El Paso, El Paso, Texas 79968-0500,

USA *Department of Biological Sciences, El Paso Community College, El Paso, Texas 79927, USA “Centro Zamorano de Biodiversidad, Escuela

Agricola Panamericana Zamorano, Departamento de Francisco Morazan, HONDURAS °1350 Pelican Court, Homestead, Florida 33035, USA

Abstract.—Of significant biodiversity importance, the Mesoamerican herpetofauna now increases at a rate of

approximately 35 species annually. As its size increases, however, the global problem of biodiversity decline

continues to worsen with time. Recently, a set of conservation priority levels was established for individual

species based on a combination of physiographic distribution and Environmental Vulnerability Score (EVS).

The 18 such levels identified range from level one, encompassing species that occupy a single physiographic

region and with a high EVS, to level 18, including species that inhabit six physiographic regions and have a low

EVS. For the Mesoamerican herpetofauna, the greatest number of species is placed in level one, amounting to

970 taxa with documentable distributions. From one to 149 priority level one species are found in 20 of the 21

physiographic regions recognized in Mesoamerica. Slightly more than three-quarters of the priority level one

species of anurans, salamanders, and squamates are found in the Baja California Peninsula and six montane

regions in Mexico and Central America. Conservation biology, thus far, has not been successful at reversing

the steady loss of biodiversity nor at placing biodiversity decline on the global agenda. In addition, humans are

becoming increasingly divorced from contact with the natural world and, thus, less aware of the life-threatening

impact they are having on the planet’s life-support systems. Given this situation, the authors of this paper have

become increasingly devoted to trying to understand why humans in general exhibit the highly dangerous

anthropocentric worldview. As have other biologists, the authors ascribe this behavior to what is known as

“the mismanagement of the human mind.” This mismanagement of the human mind is believed to result from

a cascade of psychological ailments giving rise to increasingly restrictive forms of centristic thinking. In the

final analysis, these types of thinking appear likely to doom to failure any efforts to establish for perpetuity

protected areas that can harbor the priority level one species identified in this and earlier papers. Until and

unless the anthropocentric worldview can be transformed into a worldview consonant with the realities of how

life operates on planet Earth, we humans are not only endangering ourselves but also all other life. This article

discusses the implications of this worldview for the potential conservation of the priority level one endemic

species of the Mesoamerica herpetofauna.

Keywords. Amphibia, biodiversity decline, Central America, conservation priority levels, Mexico, Reptilia

Resumen.—De gran significancia en materia de biodiversidad, la herpetofauna Mesoamericana aumenta a una

tasa aproximada de 35 especies anualmente. Sin embargo, asi como aumenta su importancia, el problema de la

disminucion global de la biodiversidad continua empeorando con el tiempo. El trabajo reciente por algunos de

nosotros establecio un numero de niveles de conservacion prioritarios que estan basados en la combinacion

de la distribucion geografica y el indice de Vulnerabilidad Ambiental (Environmental Vulnerability Score = EVS,

por sus siglas en ingles). Dieciocho niveles han sido identificados, que van desde el nivel uno, que incluye las

especies que se encuentran en una sola region fisiografica y con un EVS alto, al nivel 18, que incluye especies

que habitan en seis regiones fisiograficas y con un EVS bajo. El mayor numero de especies se encuentra

en el nivel uno, con 970 taxones. De una a 149 especies en el nivel de prioridad uno, se encuentran en 20 de

las 21 regiones fisiograficas reconocidas en Mesoamerica. Ligeramente mas de tres cuartos de los anuros,

salamandras, y escamosos en el nivel de prioridad uno, se encuentran en la Peninsula de Baja California y

en seis regiones montanosas de Mexico y Centroamerica. A la fecha, la conservacion bioldgica no ha sido

exitosa en revertir la perdida consistente de biodiversidad, ni en establecer la disminucion de la biodiversidad

en la agenda global. Adicionalmente, los humanos cada vez estan mas divorciados del contacto con el mundo

natural, y asi, menos conscientes del impacto mortal que estamos ejerciendo en los sistemas que sostienen

Correspondence. eligarciapadilla86@gmail.com (GPD), didesantis@miners.utep.edu (DLS); turyrocha@gmail.com (AR);

vmata@utep.edu (VMS); jjohnson@utep.edu (JDJ); *bufodoc@aol.com (LDW)

Amphib. Reptile Conserv. 73 June 2020 | Volume 14 | Number 2 | e240

Garcia-Padilla et al.

la vida del planeta. Dada la situacion actual, los autores de este articulo se han dedicado seriamente a intentar

entender por qué los humanos en general demuestran una vision antropocéntrica del mundo muy peligrosa.

En concordancia con otros bidlogos, estos autores atribuyen esta conducta a lo que se conoce como “la

mala conducta de la mente humana”. Esta conducta mental es el resultado de una cascada de problemas

psicologicos que dan origen a una creciente variedad de pensamientos centristas. En el analisis final, son los

tipos de pensamientos centristas los que probablemente aseguran el fallo de los esfuerzos para establecer

areas naturales protegidas perpetuas que pueden albergar a las especies en el nivel uno de prioridad que hemos

identificado en este y otros articulos anteriores. Mientras no sea posible transformar la vision antropocentrica

del mundo en una que vaya acorde con la realidad de como funciona la vida en el planeta Tierra, hasta entonces

los humanos no solo estaremos poniendo en riesgo nuestras propias vidas, si no la de todos los seres vivos.

Este articulo discute las implicaciones de esta cosmovision para la conservacion potencial de las especies

endémicas de primer nivel de la herpetofauna de Mesoamerica.

Palabras Claves. Anfibia, América Central, disminucién de la biodiversidad, México, niveles prioritarios de

conservacion, Reptilia

Citation: Garcia-Padilla E, DeSantis DL, Rocha A, Mata-Silva V, Johnson JD, Wilson LD. 2020. Conserving the Mesoamerican herpetofauna: the

most critical case of the priority level one endemic species. Amphibian & Reptile Conservation 14(2) [General Section]: 73-132 (e240).

tribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in

any medium, provided the original author and source are credited. The official and authorized publication credit sources, which will be duly enforced,

are as follows: official journal title Amphibian & Reptile Conservation; official journal website: amphibian-reptile-conservation.org.

Received: 12 June 2019; Accepted: 11 March 2020; Published: 23 June 2020

“Oxymorons, such as “sustainable development,” are — this decline, we must rapidly accumulate the baseline

strung together by politicians and developers in any data needed to document its nature and extent in order

attempt to make all this destruction and homogenization — to transform the search for ultimate solutions from its

seem less offensive.” current position on “herpetological wish lists” to rapid

Eric R. Pianka (1994) enactment over the long term.

Johnson et al. (2017) and Mata-Silva et al. (2019)

Introduction examined the endemic herpetofaunas of Mexico and

Central America, respectively, in an attempt to establish a

The Mesoamerican herpetofauna is of tremendous set of conservation priority levels based on physiographic

biodiversity significance (Wilson and Johnson 2010; — distribution and Environmental Vulnerability Score

Wilson et al. 2013a,b; Johnson et al. 2015; Johnson et (EVS; Wilson et al. 2013a,b; Johnson et al. 2015).

al. 2017; Mata-Silva et al. 2019), and that significance Calculations in Johnson et al. (2017) and Mata-Silva et

only increases with time due to the continuing discovery __ al. (2019) led to the recognition of a series of 18 priority

of new taxa within the region (see below). Wilson levels ranging from level one (species occupying a single

and Johnson (2010) comprehensively documented a physiographic region and having a high category EVS) to

herpetofauna for the region of 1,879 species. The current —_level 18 (species occurring in six physiographic regions

figure for Mesoamerica is 2,156 species, or an increase —_ and having a low category EVS).

of 277 species over approximately eight years, i.e., 34.6 Johnson et al. (2017) and Mata-Silva et al. (2019)

species per year (http://mesoamericanherpetology.com; — considered the priority level one species to be the most

accessed 9 November 2019). If this rate of discovery in need of conservation attention, due to their limited

were to hold until mid-century, then the total figure distribution and high environmental vulnerability.

for Mesoamerica could be expected to rise to ~3,229 Johnson et al. (2017) listed 490 such species in Mexico,

species. While this increase in our knowledge of the — and Mata-Silva et al. (2019) listed 429 species in Central

Mesoamerican herpetofauna is occurring, the factors that | America, for a total of 919 species. In the interim

exacerbate the overall global problem of biodiversity | beyond the appearance of these two papers, a number of

decline are worsening at an exponential rate, in concert additional species have been described that also qualify

with the rise in human population numbers (Johnson as conservation priority level one species, and we have

et al. 2017; Jarvis 2018). Unfortunately, we know _ incorporated them into our analysis below. In addition,

much more about the growth of our knowledge of the — several corrections to the categorizations that were

Mesoamerican herpetofauna than we do about its decline. | assigned in these two papers have been necessitated by

The rate at which our knowledge of this herpetofauna new information, and these re-classifications are reflected

increases (as indicated above), undoubtedly pales into —_as necessary in the tables accompanying the text of this

virtual insignificance when compared to the probable paper.

(but essentially unknown) rate of herpetofaunal species The purpose of this paper is to examine in detail the

decline over time. What data we do have, however, points future prospects for the preservation of the conservation

to a decline in herpetofaunal diversity that is increasing priority level one species identified by Johnson et al.

ever more rapidly with time. If we have any hope to limit (2017) and Mata-Silva et al. (2019) in Mexico and Central

Amphib. Reptile Conserv. 74 June 2020 | Volume 14 | Number 2 | e240

Perspective: Conserving priority level one endemic species

America, respectively. The approach we have taken is to

examine the distribution of these species in greater detail

than was undertaken in these two previous papers, with

a view to focusing on the relative significance of the

various Mesoamerican physiographic areas.

The “Conservation Priority Level” Concept

The concept of conservation priority levels was developed

for application to the Mesoamerican herpetofauna

by Johnson et al. (2017) and Mata-Silva et al. (2019).

These priority levels are based on a combination of

environmental vulnerability scores (EVS) and occurrence

in physiographic regions. Since these two papers were

published, additional herpetofaunal taxa have been

described, primarily in Mexico. These new taxa are

discussed immediately below.

Recent Changes to the Mesoamerican Herpetofauna

Inthe relatively short time since the publication of Johnson

et al. (2017) and Mata-Silva et al. (2019), a number of

significant additions to the herpetofauna of Mexico and

Central America have appeared. These additional taxa

are listed in Table 1, along with citations of their place of

publication, distribution among physiographic regions,

EVS calculations, and conservation priority levels.

Those that occupy priority level one are incorporated into

the sections below.

The 71 species included in Table 1 comprise 19

anurans, three salamanders, 20 lizards, 27 snakes, and

two turtles. Forty-eight of the 71 species were described

as new and the remainder involved elevations from

subspecies to species level or reports as new for the

herpetofauna of Mesoamerica. Thirty-five of the 48 new

species were described in 2018, one in 2016, four in 2017,

and eight in 2019. Twenty-nine of the 48 species were

described from Mexico, and the other 15 from Central

America; and only nine of the 71 species are known to

occupy more than a single physiographic region (see

Table 1). The physiographic regions (as recognized by

Wilson and Johnson [2010]) involved for all 71 species

are as follows: BC (9 species), CG (1), CP (1), CRP (6),

EP (2), GCR (5), GH (5), HN (8), MC (6), NB (2), NP

(2), OC (4), OR (7), SC (13), SD (1), SU (11), TT (4),

and YP (1). All but six species are placed in the high EVS

category of vulnerability, with scores ranging from 14 to

19; with six exceptions having EVS of 13 (4), 12 (1), and

9 (1). As a consequence, 59 of the 71 species in Table 1

qualify as priority level one taxa and, thus, need to be

included in the following tables.

Priority Levels among the Members of the

Mesoamerican Herpetofauna

As noted in the introduction, Johnson et al. (2017) and

Mata-Silva et al. (2019) developed and utilized a scheme

Amphib. Reptile Conserv.

for assigning conservation priority levels to the members

of the Mexican and Central American herpetofauna.

Given that the herpetofauna of these two regions has

increased considerably in size since these papers were

published, it is necessary to comprehensively summarize

the current data on the diversity and endemicity of this

herpetofauna for all of Mesoamerica.

Thus, Table 2 indicates the diversity of all the

Mesoamerican herpetofauna to the present day, amounting

to a total of 70 families (21 amphibian and 49 reptile),

294 genera (92 amphibian and 202 reptile), and 2,156

species (834 amphibians and 1,322 reptiles). The number

of families was recently augmented by Goicoechea et al.

(2016), which accomplished the erection of the family

Alopoglossidae to include the genera A/opoglossus and

Ptychoglossus, the latter of which contains, among others

in South America, three species that occupy Lower Central

America. The greatest numbers of these taxa at all levels

belong to the Order Anura among the amphibians and the

Order Squamata among the reptiles.

The level of endemicity of the Mesoamerican

herpetofauna is startling and strongly indicative of a

global stature for this group of animals in this region.

The species-level endemicity 1s documented in Table 3.

The total level of herpetofaunal endemism is at 79.0%,

meaning that more than three of every four species in the

region are found nowhere else in the world. Amphibian

endemicity in Mesoamerica is higher, at 84.2%, than that

for reptiles, at 75.8%. The amphibian level indicates more

than eight of every 10 species are endemic to the region;

while slightly more than three of every four reptile species

are endemic. Finally, at the ordinal level, the figure for

salamanders is simply incredible, at 96.0%, indicating

that for every 100 salamander species, only four are not

endemic. In addition, the levels of endemicity for both

anurans and squamates include more than three out of

every four species (77.9% and 76.8%, respectively).

As noted above, Johnson et al. (2017) and Mata-Silva

et al. (2019) constructed a set of conservation priority

levels for the herpetofaunas of Mexico and Central

America, respectively. The results of the categorizations

of these authors, updated to the present time (Table 4),

indicate that of the 18 recognized priority levels, six are

allocated to the high EVS priority levels, eight to the

medium EVS priority levels, and four to the low EVS

priority levels. In general, the total numbers of species

allocated to each level decrease precipitously from levels

one to six among the high EVS levels, and from seven

to 14 among the medium EVS levels, but this pattern is

not seen with the few species (eight in total) placed in

the low EVS levels. This same general pattern is seen

for both Mexico and Central America, when considered

individually (although there is but one low EVS species

in Central America). The total counts for the three EVS

levels decrease markedly from high (1,253) to medium

(216) to low (eight). Thus, the high EVS level species

make up 84.8% (1,253 of 1,477) of the total number

June 2020 | Volume 14 | Number 2 | e240

Garcia-Padilla et al.

: \ oe

xX at

Craugastor daryi (Ford and Savage 1984). Ford’s Robber Frog

has an EVS of 17 (Mata-Silva et al. 2019) and inhabits cloud

forest at elevations of 1,500—2,290 m in the Sierra Xucaneb

and Sierra de las Minas in central Guatemala (Frost 2019). This

individual was found at Purulha, Baja Verapaz, Guatemala.

Photo by Andres Novales.

The Omoa Bromeliad Frog has an EVS of 20 (Mata-Silva et

al. 2019), which lies at the upper limit of the range of values

for this conservation measure. It was described from the

visitors’ center in Parque Nacional El Cusuco in northwestern

Honduras, one of the most significant areas of herpetofaunal

endemicity in the country (Townsend and Wilson 2008). This

individual came from Parque Nacional Cusuco, Honduras.

Photo by Andres Novales.

: a ae

Dendropsophus sartori (Smith 1951). Taylor’s Yellow Treefrog

has an EVS of 14 (Johnson et al. 2017) and a distribution

encompassing the “Pacific slopes of southwestern Mexico

(Jalisco to Oaxaca)” (Frost 2019). These individuals were

found in the Municipality of San Juan Lachao, Oaxaca, Mexico.

Photo by Vicente Mata-Silva.

Amphib. Reptile Conserv.

Eleutherodactylus syristes Hoyt 1965. The Piping Peeping Frog

has an EVS of 16 (Johnson et al. 2017) and occupies the “‘pine-

oak woodland on the Pacific slopes of the Sierra de Miahuatlan

and Mixteca Alta, Oaxaca, east into the Sierra Madre del Sur of

Guerrero, Mexico” (Frost 2019). This individual was located in

the Municipality of San Juan Lachao, Oaxaca, Mexico. Photo

by Vicente Mata-Silva.

Charadrahyla_ sakbah_ Jiménez-Arcos, Calzada-Arciniega,

Alfaro-Juantorena, Vazquez-Reyes, Blair, and Parra-Olea

2019. This recently-described hylid frog has an EVS of 15

(Table 1) and is restricted to cloud forest in the western portion

of the Sierra Madre del Sur of Oaxaca, Mexico, an area of

high herpetofaunal endemicity (Mata-Silva et al. 2015b). This

individual is from Rio Chite ku’e (Rio de las Mil Cascadas),

San Isidro Paz y Progreso, Santa Maria Yucuhiti, Oaxaca.

Photo by Victor H. Jiménez-Arcos.

fr +

Plectrohyla dasypus McCranie and Wilson 1981. The Cusuco

Spotted Treefrog has an EVS of 14 (Mata-Silva et al. 2019)

and occurs in cloud forest at elevations of 1,300—1,990 m in

the Sierra de Omoa of northwestern Honduras (Townsend

and Wilson 2008). This individual was encountered at Parque

Nacional Cusuco, Honduras. Photo by Andres Novales.

76 June 2020 | Volume 14 | Number 2 | e240

Perspective: Conserving priority level one endemic species

Table 1. Mesoamerican herpetofaunal species described or elevated to species level since Johnson et al. (2017) and Mata-Silva et

al. (2019), along with their places of publication, physiographic region(s), EVS calculations, and conservation priority levels. The

abbreviations for regions involved are as follows: BC = Baja California and adjacent islands; NB = Northern Plateau Basin and

Ranges; SD = Sonoran Desert basins and ranges; MC = Mesa Central; SC = Pacific lowlands from Sonora to western Chiapas,

including the Balsas Basin and Central Depression of Chiapas; OC = Sierra Madre Occidental; OR = Sierra Madre Oriental; TT =

Atlantic lowlands from Tamaulipas to Tabasco; YP = Yucatan Platform; SU = Sierra Madre del Sur; GCR = Pacific lowlands from

southeastern Guatemala to northwestern Costa Rica; GH = Caribbean lowlands of eastern Guatemala and northern Honduras; CRP

= Isthmian Central American highlands; CG = western nuclear Central American highlands; HN = eastern nuclear Central American

highlands; CP = Pacific lowlands from central Costa Rica through Panama; NP = Caribbean lowlands from Nicaragua to Panama;

and EP = eastern Panamanian highlands.

—S— eee ian se level

| Craugastor aenigmaticus aenigmaticus Arias et fAriasetal.2018 2018 | S+844=17 17

[crag costae __[Mecranie2oie ‘| _uN | ever [one

22. ee ae Ce

CG, HN, GH, Occupies =a be-

Sceloporus olloporus Solis-Zurita et al. 2019 ee NP 5+1+3=9 See 17 and 18

| Sceloporus schmidti schmidti |McCranie2018 2018 | S+743=15 15

Amphib. Reptile Conserv. 77 June 2020 | Volume 14 | Number 2 | e240

Garcia-Padilla et al.

Table 1 (continued). Mesoamerican herpetofaunal species described or elevated to species level since Johnson et al. (2017) and

Mata-Silva et al. (2019), along with their places of publication, physiographic region(s), EVS calculations, and conservation priority

levels. The abbreviations for regions involved are as follows: BC = Baja California and adjacent islands; NB = Northern Plateau

Basin and Ranges; SD = Sonoran Desert basins and ranges; MC = Mesa Central; SC = Pacific lowlands from Sonora to western Chi-

apas, including the Balsas Basin and Central Depression of Chiapas; OC = Sierra Madre Occidental; OR = Sierra Madre Oriental;

TT = Atlantic lowlands from Tamaulipas to Tabasco; YP = Yucatan Platform; SU = Sierra Madre del Sur; GCR = Pacific lowlands

from southeastern Guatemala to northwestern Costa Rica; GH = Caribbean lowlands of eastern Guatemala and northern Honduras;

CRP = Isthmian Central American highlands; CG = western nuclear Central American highlands; HN = eastern nuclear Central

American highlands; CP = Pacific lowlands from central Costa Rica through Panama; NP = Caribbean lowlands from Nicaragua to

Panama; and EP = eastern Panamanian highlands.

; Physiographic EVS Conservation

Species References : hs

region(s) calculations priority level

, Ramirez-Reyes and Flores- =

, Ramirez-Reyes and Flores- E

ole Ramirez-Reyes and Flores- 7

Ramirez-Reyes and Flores- a

Ramirez-Reyes and Flores- SC 6+843<17 One

Villela 2018

Plestiodon lotus

Aristelliger nelsoni

Lepidophyma inagoi Palacios-Aguilar et al. 2018

Phyllodactylus rupinus

Lampropeltis greeri

Lampropeltis leonis

Sonora annulata Cox et al. 2018 3+74+5=15

Sonora cincta Cox et al. 2018 2+74+5=14

Sonora episcopa Cox et al. 2018 3+7+3=13

Sonora fasciata Cox et al. 2018 5+8+5=18

Sonora straminea

Chersodromus australis

;

Sonora punctatisima Cox et al. 2018 2+8+3=13

Rhadinella dysmica 6+8+2=16

Salvadora gymnorhachis Hernandez-Jiménez et al. 2019 5+8+4=17

Amphib. Reptile Conserv. 78 June 2020 | Volume 14 | Number 2 | e240

Perspective: Conserving priority level one endemic species

Plectrohyla exquisita McCranie and Wilson 1998. The Cusuco

Giant Treefrog has an EVS of 15 (Mata-Silva et al. 2019) and

is distributed from 1,430—1,780 m in cloud forest in the Sierra

de Omoa in northwestern Honduras (Townsend and Wilson

2008). This individual was found at Parque Nacional Cusuco,

Honduras. Photo by Andres Novales.

j > SAS

Bolitoglossa chinanteca Rovito, Parra-Olea, Lee, and Wake

2012. The Chinanteca Salamander has an EVS of 18 (Johnson

et al. 2017) and a distribution within the Sierra Juarez of

Oaxaca, Mexico (Frost 2019). This individual was encountered

in the Municipality of San Felipe Usila, Oaxaca, Mexico. Photo

by Vicente Mata-Silva.

Bolitoglossa oaxacensis Parra-Olea, Garcia-Paris, and Wake

2004. The Atoyac Salamander has an EVS of 17 (Johnson et al.

2017) and is distributed in “humid oak-pine forest in the Sierra

Madre del Sur, specifically from the mountains south of Sola de

Vega, to immediately south of the Atoyac River Basin, in the

vicinity of Puerto Portillo, Oaxaca, Mexico” (Frost 2019). This

individual was encountered in the Municipality of Santa Catarina

Juquila, Oaxaca, Mexico. Photo by Vicente Mata-Silva.

Amphib. Reptile Conserv.

oe M4 no eae

Quilticohyla acrochorda (Campbell and Duellman 2000). The

Warty Mountain Stream Frog has an EVS of 14 (Johnson et al.

2017) and ranges “at elevations from 594—900 m on the Atlantic

slopes of the Sierra Juarez [sic], Oaxaca, Mexico” (Frost 2019).

This individual was found in the Municipality of San Felipe

Usila, Oaxaca, Mexico. Photo by Vicente Mata-Silva.

Bolitoglossa conanti McCranie and Wilson 1993. Conant’s

Mushroomtongue Salamander has an EVS of 16 (Mata-Silva et

al. 2019) and is found at moderate and intermediate elevations

of 1,370-—2,000 m in cloud forest on both versants from

northwestern Honduras to extreme northwestern El Salvador,

as well as adjacent eastern Guatemala (Townsend and Wilson

2008; Frost 2019). This individual was encountered at La

Union, Zacapa, Guatemala. Photo by Andres Novales.

Bolitoglossa helmrichi (Schmidt 1936). The Coba

Mushroomtongue Salamander has an EVS of 16 (Mata-Silva

et al. 2019) and ranges in southwestern Alta Verapaz and Baja

Verapaz, Guatemala, at elevations of 1,000—2,000 m (Frost

2019). This individual was found at Purulha, Baja Verapaz,

Guatemala. Photo by Andres Novales.

June 2020 | Volume 14 | Number 2 | e240