The Identification of Cutaneous Bacteria on Salamanders that Inhibit the Chytrid Fungus Batrachochytrium dendrobatidis

Introduction

As I anxiously trekked along the path in the woods of Teatown Lake Reservation, I found myself skipping with a light bounce in my step. Ever since I set up my coverboards in the spring of last year, I’d eagerly waited in anticipation of finding eastern redback salamanders in the fall. After doing extensive research on amphibians for a whole year, it was an exhilarating feeling to finally be able to search for salamanders myself. I slowly made my way to the coverboards, the dry autumn leaves crunching crisply under my feet. Taking a deep breath, I knelt down and carefully flipped over one board, and a gush of elation rushed through me: huddling in the moist soil, with its tail curled in a tiny loop, was an eastern redback salamander, and I couldn’t be more ecstatic. After observing such a beautiful creature up close and in person, I was filled with awe, and I could truly understand the importance of protecting amphibians from population declines, which have put 40% or more of all amphibian species at risk of extinction (2008 IUCN Red List).

Amphibians serve as crucial bioindicators, and these declines suggest that ecosystems worldwide are under severe environmental stress (Welsh & Olivier, 1998), thus revealing the exigency of understanding the causes of these declines. Factors that may exacerbate amphibian declines include habitat fragmentation (Dodd & Smith, 2003) and the fungal disease chytridiomycosis (Myers, 2012). Previous research has suggested that habitat loss through urbanization jeopardizes the defense mechanisms of amphibians, particularly the mutualistic cutaneous bacterial species found on the skin of amphibians that may inhibit the growth of pathogens (Belden & Harris, 2007); therefore, specific habitat conditions may influence an amphibian’s susceptibility to disease. The skin disease chytridiomycosis, caused by the chytrid fungus Batrachochytrium dendrobatidis, has also been associated with mass population declines (Phillips, 2013). The growth of the chytrid fungus may be mitigated by certain cutaneous bacterial species on amphibians; however, specific habitat conditions may influence the diversity and presence of such beneficial cutaneous bacteria. Little research has investigated the relationship between specific environmental conditions and the diversity of cutaneous bacteria on salamanders, which is essential to our understanding of amphibian loss. Therefore, this study aimed to determine if urbanization and habitat types may influence the presence of beneficial cutaneous bacteria on salamanders that inhibit the growth of the chytrid fungus Batrachochytrium dendrobatidis.

Figure 1

An eastern redback salamander


Habitat fragmentation has been directly associated with amphibian declines (Kiesecker, 2003; Czech et al., 2000). Anthropogenic disturbances, particularly in the form of urbanization, result in decreased populations (McKinney, 2002), which limit the genetic variability of myriad different species (Gibbs, 2001). Decreased amphibian populations may result in a loss of the genetic diversity of organisms that form symbiotic relationships with amphibians, such as cutaneous bacterial communities that live on the skin of amphibians. Some species of cutaneous bacteria have mutualistic relationships with amphibian hosts and are able to protect amphibians from pathogens (Brucker, 2008). However, a reduced diversity of cutaneous bacteria may influence the presence of such bacterial species on amphibians. Therefore, a smaller diversity of cutaneous bacteria may limit the amphibians’ ability to adapt to disturbances caused by urbanization, placing amphibians in further jeopardy; however, this has yet to be investigated in the context of pathogenic infections.

The pathogenic chytrid fungus Batrachochytrium dendrobatidis causes the disease chytridiomycosis, where the skin of an amphibian becomes keratinized, leading to death by suffocation (Berger et al., 1998). Although amphibians worldwide are threatened by this disease, some amphibian species, such as the eastern redback salamander (Plethodon cinereus) (Figure 1), are able to resist the presence of this fungus (Harris, 2006; Woodhams et al., 2007). It has been postulated that this resistance comes from cutaneous bacterial species that produce antifungal metabolites fatal to Batrachochytrium dendrobatidis, thereby mitigating the growth of the fungus. One such cutaneous bacterium is Janthinobacterium lividum, previously found on the eastern redback salamander (Brucker, 2008). However, when multiple salamanders were swabbed for cutaneous bacteria, Brucker found that this specific bacterial species was not found on all the salamanders swabbed, suggesting that environmental factors may influence the diversity and presence of cutaneous bacteria on salamanders. Brucker’s findings of Janthinobacterium lividum’s antifungal ability also implicate the exigency of genotyping additional cutaneous bacteria from eastern redback salamanders that may attenuate the fungus’ growth. Therefore, the identification of cutaneous bacteria with inhibitory abilities found on salamanders is vital for a deeper understanding of bacterial species that mitigate the growth of the deadly chytrid fungus.

The enigmatic occurrence of chytridiomycosis around the world necessitates a more thorough understanding of the disease in order to control its dissemination. Past research has examined the spread of Batrachochytrium dendrobatidis on a global level (Johnson & Speare, 2005), yet little is known about the spread and virulence of the fungus between varying habitat types. Previous research has suggested that more infections were observed in frogs breeding along streams than those breeding in terrestrial habitats and intermittent streams (Kriger & Hero, 2007). Thus, stream-associated amphibians may be at a greater risk of infection than those of other habitat types, potentially due to fewer cutaneous bacterial species with the ability to inhibit the fungus’ growth. However, research has yet to determine which particular habitat conditions influence the diversity and presence of cutaneous bacteria on salamanders that inhibit the chytrid fungus Batrachochytrium dendrobatidis. Thus, in this study, the relationship between habitat conditions, specifically urbanization and habitat types, and the diversity and presence of cutaneous bacteria with the ability to mitigate the growth of Batrachochytrium dendrobatidis was investigated.

 

Purpose and Hypotheses

The primary purpose of this study was to characterize and identify cutaneous bacterial species present on eastern redback salamanders that are capable of inhibiting the chytrid fungus Batrachochytrium dendrobatidis through DNA sequencing. Additionally, this study aimed to determine if specific habitat conditions, such as urbanization and habitat types, influence the diversity and presence of cutaneous bacteria that inhibit Batrachochytrium dendrobatidis.

It was hypothesized that the number of cutaneous bacterial species with the ability to inhibit Batrachochytrium dendrobatidis decreases with increased urbanization, and that there are fewer cutaneous bacterial species with inhibitory abilities on stream-side salamanders than on terrestrial salamanders. It was also hypothesized that the species diversity of cutaneous bacteria decreases with increased urbanization. Because of the crucial niches occupied by amphibians in ecosystems around the world, understanding the causes of population declines is necessary in order to conserve vulnerable species.

 

Methods

Study Sites

Nine study sites were sampled in the New York City metropolitan area. Sites were classified as rural, suburban, or urban based upon each site’s distance from Central Park in New York City (Table 1).

Study Sites Level of Urbanization Distance from Central Park (km)
Van Cortlandt Park Conservancy Urban 14.76
New York Botanical Garden Urban 11.85
Pelham Bay Park Urban 16.95
Rockefeller State Park Preserve Suburban 37.99
Fordham Louis Calder Center    
Biological Field Station    
Teatown Lake Reservation Suburban 49.10
Harriman State Park Rural 57.64
Black Rock Forest Consortium Rural 70.24
Clarence Fahnestock State Park Rural 77.01

Table 1: Study site locations according to an urban-rural gradient utilized to study areas affected differently by urbanization

 

Figure 2: Map showing the study sites

(adapted from Google Maps)


Study sites included three city parks, three state parks, and three nonprofit reservations (Figure 2, Table 1). All study sites were located in mixed deciduous forests. Sampling of salamanders began in early October 2014, when salamanders become more active due to cooler air and soil temperatures (Vernberg, 1953), and continued until early November. Sampling was also conducted in the spring season from late April 2015 to late May 2015.

Coverboards made of untreated pine were cured for approximately three months and were then introduced at four months, on average, before sampling by the student researcher. The dimensions of each coverboard were 30 cm x 20 cm x 2.5 cm (l x w x h). Sampling locations were chosen based on previous observations of salamanders relative to the location of the streams at each site. Specifically, three arrays of coverboards were placed along a topography gradient at each location (Figure 3): streamside (≤ 1 m from stream), mid-upland (15 m from stream), and high upland (30 m from stream). These distances from the stream distinguished the three different habitat types examined in this study.

Figure 3

Figure 3: Diagram showing location of the coverboards along the topography gradient


 

Habitat Variables

To investigate the relationship between salamanders and habitat characteristics, soil characteristics were measured. To measure characteristics of the soil, the O horizon, which is the layer of soil containing high levels of organic matter, and the A horizon, which is the layer of soil below the O horizon containing accumulations of humified organic matter mixed with mineral material, of the soil were collected at each coverboard array. A subsample of soil was set aside for further analysis of soil carbon.

 

Salamander Sampling

Figure 4: Soon Il checks for salamanders under the coverboards.


Permission to sample salamanders was obtained through the New York State Department of Environmental Conservation and the New York City Department of Parks and Recreation before any data collection was conducted. Sampling was conducted after periods of rainfall to increase the probability of encountering salamanders (Jaeger, 1972). Each site was visited no more than one to two times during the sampling seasons in order to limit disturbance. At each site, no more than five eastern redback salamanders were sampled in order to limit disturbance and consistent with the DEC permit. If more than five salamanders were observed at a site, a headcount of the total number of salamanders at that location was taken, but only five were sampled.

Coverboards and salamanders were handled using nitrile examination gloves that were disposed of after handling each salamander and after sampling each coverboard array in order to minimize cross-contamination (Figure 4). At each sampling, coverboards were gently lifted with an even application of pressure to avoid injuring any salamanders. Salamanders found under the boards were captured and placed in a container. Prior to swabbing, each salamander was rinsed with autoclaved well water to avoid swabbing transient bacteria, thus ensuring that only cutaneous bacteria were sampled from the salamander’s skin. To further minimize cross-contamination, after sampling and replacing each salamander under a coverboard (Figure 5), the container was rinsed three times with autoclaved well water.

Figure 5

Figure 5: Salamanders were swabbed using a non-invasive sampling method.


Bacterial swabbing was conducted using a non-invasive sampling method (Boyle, 2004). In brief, each salamander was swabbed three times, using a sterile cotton swab, on the ventral, left, and right sides (Harris, 2006). These swabs were streaked onto petri dishes with nutrient agar and immediately sealed on site with parafilm. After swabbing, the swabbed salamander was carefully placed adjacent to the coverboard from which it was gathered, unharmed. No salamanders were removed from the natural environment in this study.

 

Laboratory Challenge Assays

Figure 6

Figure 6: Liquid Batrachochytrium dendrobatidis culture


In this study, challenge assays were conducted in order to determine if a bacterial isolate could inhibit the growth of Batrachochytrium dendrobatidis and may thereby protect amphibians from infection by the chytrid fungus. A challenge assay is a test performed where two organisms are placed in the same medium and are given time to grow in order to determine if one organism is able to inhibit the growth of the other organism. Each challenge assay consisted of a petri dish with the chytrid fungus Batrachochytrium dendrobatidis fully colonized and a streak of a bacterial isolate. Streak plates were incubated at room temperature (23°C) until distinct colonies were visible and could be isolated to inoculate pure cultures. These cultures were used for two purposes: 1) to perform challenge assays to test for inhibitory abilities, and 2) to identify bacterial isolates through DNA sequencing using the Sanger sequencing method.

Cultures of Batrachochytrium dendrobatidis, strain JEL 423, were purchased from Dr. Joyce Longcore at the University of Maine, and cultured in a 1% tryptone broth until colonies were evident (Figure 6). Two millimeters of this broth were pipetted onto 10% tryptone agar plates and left under a laminar flow hood to dry. The bacterial isolates were then streaked across these fungal plates and given two weeks to grow out at 23°C.

Inhibitory abilities were classified as either non-inhibitory or inhibitory (Harris, 2006). Non-inhibitory isolates were characterized by no zones of inhibition between the bacterial streak and the fungus, while inhibitory isolates were characterized by zones of inhibition between the bacterial streak and the fungus. Control plates were plates with only the chytrid fungus. Bacterial isolates demonstrating inhibitory abilities were identified via DNA sequencing. Plates that came into contact with B. dendrobatidis were autoclaved after use.

 

Statistical Analysis

Statistical analyses were conducted in Minitab 15. ANOVAs were performed to compare habitat variables, including soil pH and soil carbon, among the nine sites to assess if habitat conditions varied between locations.

 

Results
Salamanders Swabbed and Bacterial Swabs Collected

In total, 23 eastern redback salamanders were observed across all nine sites, with salamanders observed at all but one rural and one suburban site (Figure 7). Urban sites accounted for about 70% of observed salamanders. Because more than five salamanders were found at one urban location, and juvenile salamanders were observed but not swabbed at another urban site, the number of salamanders and the number of bacterial swabs collected differ. Fourteen bacterial swabs were obtained from salamanders (n = 14) at five out of nine study sites—one rural, one suburban, and three urban sites. In addition to containing more salamanders, urban areas yielded a larger number of distinct bacterial isolates compared with suburban and rural areas (Figure 8). Currently, 54 visually distinct bacterial isolates have been identified and given temporary species identification letters based on distinct morphological characteristics until all can be properly sequenced (Table 2).

Figure 7

Figure 7: This graph illustrates the number of salamanders observed at study sites according to level of urbanization


 

Figure 8

Figure 8: This graph illustrates the number of bacterial isolates that were visually identified from study sites according to level of urbanization.


Table 2: Habitat conditions and temporary species identification letters of bacterial isolates

Study Site and Swab # Level of Urbanization Habitat Isolate # ID Letter
Black Rock 3 Rural Mid-Upland 1 K
Black Rock 3 Rural Mid-Upland 3 AK*
Black Rock 3 Rural Mid-Upland 5 P
Black Rock 3 Rural Mid-Upland 6 M
Black Rock 1 Rural Stream-Side 7.2 V*
Black Rock 1 Rural Stream-Side 7.3 I
Black Rock 1 Rural Stream-Side 7.4 AA
Black Rock 1 Rural Stream-Side 8 J
Black Rock 2 Rural Stream-Side 13 AA
Black Rock 2 Rural Stream-Side 14 AM*
Black Rock 2 Rural Stream-Side 17 P
Black Rock 2 Rural Stream-Side 18.1 U*
Black Rock 2 Rural Stream-Side 18.2 AJ*
Teatown 1 Suburban Stream-Side 21 AH*
Teatown 1 Suburban Stream-Side 22 AL*
Teatown 1 Suburban Stream-Side 23 AF
Pelham Bay 1 Urban Mid-Upland 25.1 H
Pelham Bay 1 Urban Mid-Upland 25.2 G*
Pelham Bay 1 Urban Mid-Upland 26 AG*
Pelham Bay 1 Urban Mid-Upland 29 AF
Pelham Bay 1 Urban Mid-Upland 30.3 AN*
Van Cortlandt 3 Urban Mid-Upland 33.2 Y
Van Cortlandt 3 Urban Mid-Upland 34 I
Van Cortlandt 3 Urban Mid-Upland 35.1 L*
Van Cortlandt 3 Urban Mid-Upland 35.2 M
Van Cortlandt 3 Urban Mid-Upland 35.3 C*
Van Cortlandt 4 Urban Mid-Upland 37 N*
Van Cortlandt 4 Urban Mid-Upland 38.2 H
Van Cortlandt 4 Urban Mid-Upland 39.1 S*
Van Cortlandt 4 Urban Mid-Upland 39.2 AD*
Van Cortlandt 4 Urban Mid-Upland 39.3 P
Van Cortlandt 4 Urban Mid-Upland 39.4 A*
Van Cortlandt 4 Urban Mid-Upland 40 K
Van Cortlandt 4 Urban Mid-Upland 42.1 B*
Van Cortlandt 4 Urban Mid-Upland 42.2 M
Van Cortlandt 4 Urban Mid-Upland 42.3 E*
Van Cortlandt 4 Urban Mid-Upland 42.4 J
Van Cortlandt 4 Urban Mid-Upland 43 D
Van Cortlandt 4 Urban Mid-Upland 44 F*
Van Corltandt 5 Urban High Upland 48 AB*
Van Cortlandt 5 Urban High Upland 51.1 H
Van Cortlandt 5 Urban High Upland 51.2 R*
Van Cortlandt 5 Urban High Upland 51.3 J
Van Cortlandt 5 Urban High Upland 51.4 O
Van Cortlandt 5 Urban High Upland 51.5 AC*
Van Cortlandt 5 Urban High Upland 52 D
Van Cortlandt 5 Urban High Upland 53 T*
NYBG 2 Urban Mid-Upland 58 AE*
NYBG 3 Urban Mid-Upland 59 K
NYBG 3 Urban Mid-Upland 60 AA
NYGB 4 Urban Mid-Upland 62 AF
NYBG 4 Urban Mid-Upland 63 AI*

*Indicated unique isolates not found at other sites

Figure 9

Figure 9: This graph illustrates the number of bacterial isolates identified at each habitat site.


 

Out of the 54 bacterial isolates visually identified to date, only 13 were from swabs collected from salamanders in stream-side habitats (Figure 9). When comparing urban, suburban, and rural locations, 18 unique isolates were identified from urban sites, while only two and five isolates were found in suburban and rural sites, respectively. Of the isolates observed across sites, seven were observed at both urban and rural sites.

 

Challenge Assays

Figure 10: Isolate A exhibiting inhibitory abilities


Bacterial isolate A, which was obtained from a salamander found in an urban, mid-upland site at Van Cortlandt Park, demonstrated inhibitory potential, showing a zone of inhibition after two weeks in the challenge assay with Bd (Figure 10). The initial streak of the isolate was visible, and bacterial growth was evident within the streak, indicating that A may have inhibitory abilities that mitigate the growth of the chytrid fungus. Multiple challenge assays must be conducted in order to determine the inhibitory abilities of the remaining bacterial isolates.

 

Discussion

This study was conducted to understand the effects of specific habitat conditions on the diversity and presence of cutaneous bacteria with the ability to inhibit the pathogenic fungus Batrachochytrium dendrobatidis. It was hypothesized that the number of cutaneous bacterial species with the ability to inhibit Bd decreases with increased urbanization. According to my current results, urban sites actually yielded a larger diversity of bacterial isolates than rural sites did, contrary to what was hypothesized. This finding also contrasts with previous research, which has found that urbanization is associated with a restricted genetic variability among species (McKinney, 2002; Gibbs, 2001). This may suggest that urbanization plays a role in natural selection: the bacteria at urban sites may be more inclined to compete and adapt to environmental changes caused by anthropogenic disturbances, making these bacteria more resilient to urbanization. Thus, bacteria from rural areas may be less exposed to environmental changes and less likely to successfully adapt to environmental disturbances when compared with bacteria from urban areas, potentially explaining why bacterial diversity increased with increased urbanization. However, when examining these findings, it is essential to note that more salamanders were found and swabbed in urban areas than in rural areas; this difference in sample sizes between urban and rural locations may provide an explanation for the discrepancy between this study’s findings and that of past research. Thus, more accurate and conclusive statements about the diversity of cutaneous bacteria along the urban-rural gradient may be drawn from an equal sample size across all study sites.

In addition, it was predicted that there are fewer cutaneous bacterial species that can inhibit Batrachochytrium dendrobatidis found on stream-side salamanders than on terrestrial salamanders. It was found that only 13 out of 54 bacterial isolates were from amphibians in stream-side habitats. This may suggest that amphibians in stream-side habitats may have fewer cutaneous bacterial species than those in terrestrial habitats. However, out of the 14 salamanders swabbed, five salamanders were found in streamside habitats, seven in mid-upland habitats, and only one in high upland habitats, evincing the uneven sample size between salamanders in the three types of habitats. Once again, comparing the diversity of cutaneous bacteria between an equal number of samples per habitat type may impart a more reliable conclusion. Plans to sample salamanders at all nine study sites in the spring season are underway in order to obtain a more even sample size of salamanders and to compare bacterial diversity of cutaneous bacteria between habitat types more accurately.

It was hypothesized that the number of cutaneous bacteria with inhibitory abilities decreases with increased urbanization. However, the results remain inconclusive about the inhibitory abilities of cutaneous bacteria. Currently, challenge assays are being performed by the student researcher, which may provide sufficient results about inhibitory abilities that will allow me to arrive at a more sound statement. This research is ongoing; once challenge assays are fully performed, DNA sequencing will also be conducted by the student researcher in order to identify bacterial isolates with inhibitory abilities. 

A similar study conducted by Culp, Falkinham, and Belden in 2007 identified cutaneous bacteria on the skin of eastern newts (Notophthalmus viridescens), bullfrogs (Lithobates catesbeianus), and redback salamanders (Plethodon cinereus). While that study was able to identify cutaneous bacterial species from redback salamanders, only three salamanders at a single terrestrial site were sampled for cutaneous bacteria. In addition, that study was limited in that it did not investigate the identified cutaneous bacterial species’ ability to inhibit the chytrid fungus Batrachochytrium dendrobatidis. My research is the first to compare multiple habitat types and a larger sample size of salamanders in order to determine the effects of habitat conditions on the diversity of cutaneous bacteria found on eastern redback salamanders that have inhibitory abilities.

Once additional challenge assays are conducted to assess for inhibitory abilities, bacterial isolates displaying inhibitory abilities are to be identified through a DNA analysis using the Sanger sequencing method. In addition, soil samples collected from each study site are currently under examination in order to investigate how additional habitat characteristics, such as soil carbon levels and soil pH, may correlate with bacterial isolates. Future research is necessary to sample a wider range of amphibian species in order to broaden the current understanding of inhibitory abilities of cutaneous bacteria on additional amphibian species. Additionally, it is recommended that other environmental variables, such as water chemistry and heavy metals, be studied to gain a more comprehensive understanding of specific habitat conditions that serve as environmental stressors for both amphibians and cutaneous bacteria (Ometo et al., 2000). A better understanding of these environmental conditions may help conservation efforts by expanding knowledge of methods to ameliorate amphibian declines, such as bioaugmentation, and ultimately mitigate the impact of chytridiomycosis on declining amphibian populations.

 

Conclusion

The results of this study indicate that cutaneous bacterial diversity may actually increase with increased urbanization. In addition, diversity of bacteria from stream-side amphibians may be smaller than the diversity of bacteria from terrestrial amphibians. This research is the first to examine the combined effects of urbanization and habitat types on the diversity and presence of cutaneous bacteria on eastern redback salamanders that inhibit the growth of the chytrid fungus Batrachochytrium dendrobatidis, and may help conservationists to focus on increasing the presence of cutaneous bacteria with inhibitory abilities among declining amphibian species, thereby attenuating the impact of chytridiomycosis on amphibian populations.

 

Acknowledgements 

I would like to thank my mentor, Dr. James Lewis at Fordham University, for his indispensable guidance, which allowed me to conduct this research project. I would also like to express my gratitude to Dr. Joyce Longcore of the University of Maine, who graciously provided the chytrid fungus and made it possible for me to perform the challenge assays that were a crucial component of this study. I would also like to thank my science research teachers, Mr. Angelo Piccirillo and Ms. Valerie Holmes, for their endless support, which has helped me throughout the course of this project. I would also like to thank my family and friends for their constant guidance throughout my research process.

 

Bibliography

The IUCN Red List of Threatened Species. The IUCN Red List of Threatened Species, n.d. Web.

Becker, M.H., and R.N. Harris. “Cutaneous bacteria of the redback salamander prevent morbidity associated with a lethal disease.” PLoS One 5.6 (2010): 10957.

Belden, Lisa K., and Reid N. Harris. “Infectious diseases in wildlife: The community ecology context.” Frontiers in Ecology and the Environment 5.10 (2007): 533-539.

Berger, L., et al. “Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America.” Proceedings of the National Academy of Sciences 95.15 (1998): 9031-9036.

Boyle, D.G., D.B. Boyle, V. Olsen, J.A.T. Morgan & A.D. Hyatt. “Rapid quantitative detection of Chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay.” Diseases of Aquatic Organisms 60 (2004): 141-148.

Brucker, R.M., et al. “Amphibian chemical defense: Antifungal metabolites of the microsymbiont Janthinobacterium lividum on the salamander Plethodon cinereus.” Journal of Chemical Ecology 34.11 (2008): 1422-1429.

Cushman, S.A. (2006). “Effects of habitat loss and fragmentation on amphibians: A review and prospectus.” Biological Conservation 128.2 (2006): 231-240.

Czech, B., P.R. Krausman & P.K. Devers. (2000). “Economic associations among causes of species endangerment in the United States.” BioScience 50.7 (2000): 593-601.

Davic, R.D. and H.H. Welsh. “On the ecological roles of salamanders.” Annual Review of Ecology, Evolution, and Systematics (2004): 405-434.

Dodd, C.K., and L.L. Smith. (2003). “Habitat Destruction and Alteration.” Amphibian Conservation. Smithsonian Institution, Washington (1953): 94-112.

Gibbs, J.P. (2001). “Demography versus habitat fragmentation as determinants of genetic variation in wild populations.” Biological Conservation 100.1 (2001): 15-20.

Harris, R.N., T.Y. James, A. Lauer, M.A. Simon & A. Patel. “Amphibian pathogen Batrachochytrium dendrobatidis is inhibited by the cutaneous bacteria of amphibian species.” EcoHealth 3.1 (2006): 53-56.

Jaeger, R.G. “Food as a limited resource in competition between two species of terrestrial salamanders.” Ecology (1972): 535-546.

Johnson, M.L., and R. Speare. “Possible modes of dissemination of the amphibian chytrid Batrachochytrium dendrobatidis in the environment.” Diseases of Aquatic Organisms 65 (2005): 181-186.

Kiesecker, J.M. “Invasive species as a global problem: Toward understanding the worldwide decline of amphibians.” Amphibian Conservation (2003): 113-126.

Kriger, K.M., and J.M. Hero. “The chytrid fungus Batrachochytrium dendrobatidis is non‐randomly distributed across amphibian breeding habitats.” Diversity and Distributions 13.6 (2007): 781-788.

Lips, K.R., J.D. Reeve & L.R. Witters. “Ecological traits predicting amphibian population declines in Central America.” Conservation Biology 17.4 (2003): 1078-1088.

McKinney, M.L. “Urbanization, biodiversity, and conservation.” BioScience 52.10 (2002): 883-890.

Myers, J.M., J.P. Ramsey, A.L. Blackman, A.E. Nichols, K.P. Minbiole & R.N. Harris. “Synergistic inhibition of the lethal fungal pathogen Batrachochytrium dendrobatidis: The combined effect of symbiotic bacterial metabolites and antimicrobial peptides of the frog Rana muscosa.” Journal of Chemical Ecology 38.8 (2012): 958-965.

Noël, S., M. Ouellet, P. Galois & F.J. Lapointe. “Impact of urban fragmentation on the genetic structure of the eastern red-backed salamander.” Conservation Genetics 8.3 (2007): 599-606.

Ometo, Jean Pierre, et al. (2000). “Effects of land use on water chemistry and macroinvertebrates in two streams of the Piracicaba River Basin, South‐East Brazil.” Freshwater Biology 44.2 (2000): 327-337.

Phillips, B.L., and R. Puschendorf. “Do pathogens become more virulent as they spread? Evidence from the amphibian declines in Central America.” Proceedings of the Royal Society B: Biological Sciences 280.1766 (2013): 1290.

Rollins-Smith, L.A., L.K. Reinert, C.J. O’Leary, L.E. Houston & D.C. Woodhams. “Antimicrobial peptide defenses in amphibian skin.” Integrative and Comparative Biology 45.1 (2005): 137-142.

Welsh, H.H. and L.M. Ollivier. “Stream amphibians as indicators of ecosystem stress: A case study from California’s redwoods.” Ecological Applications 8.4 (1998): 1118-1132.

Welsh, H.H., and S. Droege. “A case for using plethodontid salamanders for monitoring biodiversity and ecosystem integrity of North American forests.” Conservation Biology 15.3 (2001): 558-569.

Woodhams, Douglas C., et al. “Symbiotic bacteria contribute to innate immune defenses of the threatened mountain yellow-legged frog, Rana muscosa.” Biological Conservation 138.3 (2007): 390-398.

Wyman, R.L. “Experimental assessment of salamanders as predators of detrital food webs: Effects on invertebrates, decomposition and the carbon cycle.” Biodiversity & Conservation 7.5 (1998): 641-650.

Vernberg, F.J. “Hibernation studies of two species of salamanders: Plethodon cinereus and Eurycea bislineata.” Ecology 34 (1953): 55-62.