Mannonphryne olmonae: An Ecological Assessment in Tobago, Republic of Trinidad and Tobago Project Number: 300206

Project L.E.A.P.:

Jahson B. Alemu I, Michelle N. E. Cazabon, Lean Dempewolf, Ryan P. Mannette, Kerrie T. Naranjit and Alicia Schimdt-Roach.

Assisted by Dr. Adrian Hailey © J. Alemu I, M. Cazabon, L. Dempewolf, R. Mannette, K. Naranjit and A. Roach 2007.

Not to be reproduced without permission from Project L.E.A.P.

Contact: Jahson Alemu I: [email protected] Table of Contents

1 Introduction...... 2 1.1 Background...... 3 1.2 Species description...... 4 1.3 Habitat description...... 4 1.4 Batrachochytrium dendrobatidis...... 5 2 Aims and Objectives...... 6 2.1 General Aim...... 6 2.2 Objectives: ...... 6 3 Site Description...... 7 4 Field Method...... 10 4.1 Distribution, Presence and Abundance...... 10 4.2 Ecology and Habitat Assessment...... 10 4.3 Chytrid Survey...... 11 4.4 Statistical Analysis...... 12 5 Results...... 13 5.1 Presence Survey...... 13 5.2 Population Structure...... 18 5.3 Habitat Assessment...... 20 5.3.1 Tadpole Microhabitat...... 20 5.3.2 Adult Microhabitat...... 21 5.4 Batrachochytridium dendrobatidis d Survey ...... 24 6 Discussion...... 26 6.1 Habitat, Abundance and Population dynamics ...... 26 6.2 Batrachochytrium dendrobatidis...... 27 7 Measures for Conservation ...... 29 7.1 Future of Mannophryne olmonae...... 29 7.2 Education and Public Awareness programme ...... 30 7.3 Follow up Activities...... 30 8 References...... 33

1 List of Tables.

Table number Table Name Page 5-1 Presence vs. Stream order 13 5-2 Presence vs Watershed 15 5-3 Stream Order 1 Comparison 16 5-4 Comparison of watersheds using second and third 16 orders streams only 5-5 Frog presence vs. Presence of tributaries 17 5-6 Predators vs. Frog Presence 17 5-7 M. olmonae abundance along survey rivers 18 5-8 Physical characteristics of pools with and without M. 20 olmonae tadpoles 5-9 Measured Environmental Variables 21 5-10 PCR Positives for B. dendrobatidis 24

List of Figures

Figure Figure Name Page 3-1 Study Rivers 8 3-2 Watersheds system of the Main ridge Forest Reserve 9 5-1 Surveyed Areas 14 5-2 Size frequency distributions (numbers observed) of 19 Mannophryne olmonae: a) juveniles and females (n = 110); b) males (n = 16). 5-3 Frequency distributions (numbers observed) of: 23 distances of M. olmonae frogs from the river or stream edge (n = 68) 5-4 Areas screened for B. dendrobatidis 24

List of Appendices

Appendix A: Tadpole Key Appendix B: Batrachochytrium dendrobatidis sampling Appendix C: Raw Data Appendix D: PCR Results

2 1 Introduction

1.1 Background

The Bloody Bay Poison Frog, Mannophryne olmonae is endemic to the forests of the Main Ridge Forest Reserve on the island of Tobago, of the Republic of Trinidad and Tobago, and listed as Critically Endangered (Baillie, 2004; IUCN, 2004). Hardy (2004) estimated an 80% population decline over the last three generations, with the population limited to an area less than 100 km2. Hardy’s assessment also indicated that M. olmonae’s had limited geographic range (restricted to north-east Tobago), a severely fragmented distribution, with continuing decline in the number of mature individuals may have also contributed to this decline. All these factors contributed to its designation as critically endangered.

From a global perspective, this decline is another example of the phenomenon affecting amphibian populations worldwide, potentially leading to amphibian species extinction. The major threats to amphibian populations include habitat loss and urbanisation, disease, global warming, pollution, unsustainable use, predation and unknown declines (Gibbons 2000).

All species of the genus Mannophryne are found in Venezuela and the island state of Trinidad and Tobago. Most of the geographical range is dominated by montane areas, although some species reach down to sea level. Members of this genus inhabit humid forests with a mean annual precipitation of 1100 to 2200mm, and mean annual temperatures of 18 to 24°C, parameters that may constitute useful predictors of the occurrence of these frogs in the northern mountains of South America (La Marca 1992).

3 1.2 Species description

The Bloody Bay frog Mannophryne olmonae was described by Hardy (1983), who separated the populations on Tobago from M. trinitatis of Trinidad and Venezuela with which it was formerly included. Subsequently the Venezuelan populations of these aromobatid (formerly dendrobatid – (Grant 2006)) stream frogs have also been separated into a new species (Barrio-Amorós et al., 2006; M. J. Jowers, pers. comm.), leaving endemic Mannophryne species in Tobago.

Mannophryne olmonae belongs to the family Arombatidae, but unlike many close relatives it is non-poisonous and dull coloured. This species can be distinguished from the other locally found brown frogs, M. trinitatis, by a narrow collar which is uniformly coloured and by its call (La Marca 1992). The genus is diurnal and males typically call throughout the day. Males may reach 25mm and females 28mm in length, (Hardy 1983). Sexes are distinguished by the bright yellow chin and black collar in females and grey/black chin in males, however, during the breeding season, males turn jet black while calling.

The eggs are laid on land; hatching tadpoles (11 to 19 in number) are carried to streams by the male where they complete their development (IUCN, 2004). Almost nothing is known about natural prey items of Mannophryne; unpublished data indicate that some Mannophryne species feed on beetles, large ants and mites (La Marca 1992).

1.3 Habitat description

The Bloody Bay Frog is a tropical forest species usually found in upland areas of north- eastern Tobago at elevations above 120m along streams (typically clear-water rivers, of fast velocity in the upper streams, with substrates of stones, gravel and silt and banks of boulders and gravel ) between the river edge and forest edge (Hardy, 2004). Although,

4 the geographical range is dominated by mountains, M. olmonae has been noted to reach down to sea level (La Marca 1992).

1.4 Batrachochytrium dendrobatidis

The chytrid fungus Batrachochytrium dendrobatidis (B. dendrobatidis) is the pathogen responsible for the amphibian disease chytridiomycosis (a disease of the keratinised skin). It is thought to be one of the main causes of the global decline in frog populations since the 1960s, and the dramatic population crashes from the 1970s onwards (Parris and Beaudoin, 2004). In areas such as Australia and Central America, chytridiomycosis is thought to be causing the decline of anurans (Berger, et al. 1998; Lips 1999); and may also be responsible for the loss of otherwise healthy animals (Daszak and Cunningham 1999). Some amphibian biologists consider B. dendrobatidis as a major cause of population declines of amphibian species confined to most montane rain forests worldwide (Daszak 1999; Weldon et al. 2004). B. dendrobatidis has been found in most amphibian populations on every continent, including Asia: Africa, Australasia-Pacific, North America, and South America (Lips et al. 2003; Weldon et al. 2004). To date, there has been limited assessment of the spread of the pathogen thoroughly in the Caribbean, but has been detected in countries such as Puerto Rico (Burrowes 2004) and Dominica (Fa 2004)). Worldwide, extensive research is being done to track the spread and attempts have been made to curb the spread of the pathogen.

Amphibian species likely to decline from B. dendrobatidis are stream-associated (McDonald and Alford 1999), endemic, have a large body size and occur at high elevations (Lips, Reeve et al. 2003). This pathogen is of particular concern to Tobago, a relatively small island (300 km²) with a high potential to spread to the other 14 species of amphibians on the island.

5 2 Aims and Objectives

2.1 General Aim

To determine the ecological status of M. olmonae, while assessing possible threats facing the species.

2.2 Objectives:

1. To assess the population status (numbers and range), ecological needs, and microhabitat and habitat status using rapid assessment protocols.

2. To develop a viable GIS database allowing the continued study of M. olmonae and other amphibians.

3. To determine the presence of the chytrid fungus and the extent of its spread.

4. To promote long term conservation of local endangered species (and habitats) and biodiversity in and around the Main Ridge Forest Reserve. The use of educational and training workshops, geared toward producing conservation leaders/managers will provide a sustainable pillar for further research and development in the environmental sector.

6 3 Site Description

The Main Ridge extends from the Northeast tip of the island Southwest for about two thirds of the length of Tobago, achieving a maximum altitude of 580m. The forested ridge rises steeply from the North coast and the gentler southern slopes are deeply indented by valleys which run down to a narrow but fertile coastal plain. The tropical rain forest is restricted to the sheltered mountain valleys of the Main Ridge. Lower montane forest, xerophytic rain forest, evergreen formations and some elfin woodland also occurs (Davis et al. 1986; Thelen and Faizool 1980). The rainfall regime is seasonal, with less during the dry season (January to May) and more during the wet season (June to December).

Six major rivers with headwaters in the Main Ridge Forest Reserve, as well as several short streams along the Northeast coast of Tobago were identified and assessed for the presence and density of M. olmonae. The criteria for which rivers were selected were based on the type localities of the frog established by Murphy (1997). The study rivers selected along Northside Road were not listed as type localities but were discovered to have M. olmonae present

Sample rivers included: Doctor’s River, Louis D’or River, Argyle River, Roxborough River, Kings Bay River, Bloody Bay River, and several small rivers along the Northside Road (see Figures 3.1 and 3-2).

7

Figure 3.1: Study Rivers

8 Figure 3-2: Watersheds system of the Main ridge Forest Reserve

9 4 Field Method

4.1 Distribution, Presence and Abundance

The local distribution of M. olmonae along each river was determined using visual and aural encounter methods (walking the length of the river). The distances from the mouth of the river at which any M. olmonae were observed along the respective watercourses were noted along with the distance along the river walked. Rock pools were also examined for M. olmonae tadpoles. Tadpoles were compared against an identification key adapted from Kenny (1983), (see Appendix A). When calling was heard, the presence and the number of discernable frogs heard were noted. In conjunction with presence, an abundance survey was conducted (adults and tadpoles) along the main river and its tributaries. Further, the calling range was estimated by moving away from a calling male, until it was no longer heard, and then moving back toward the point that it was heard again. This distance was noted as the audibly detectible calling radius.

4.2 Ecology and Habitat Assessment

The narrowed study area surveyed included a further 250m on either end of the range of detected presence. Where adult M. olmonae were encountered, the following data was collected: activity before disturbance, distance along river, distance from the river, distance from the forest edge, sex, presence of tadpoles on the male’s back, deformities, injuries or parasites and snout-vent length (SVL) to the nearest 0.1mm.

Where frog choruses were heard, the estimated number of frogs calling was recorded as well as the distance along the river that the chorus was heard. 1. If M. olmonae adults were encountered between quadrats, they were omitted from the density survey, but were recorded as part of the presence survey. 2. Frogs were released at the place of capture after examination

10 Where tadpoles were encountered the following data was collected: the number and size M. olmonae tadpoles, the approximate area and depth of the isolated pools (to the nearest 0.1 cm), the temperature, pH and Dissolved Oxygen content in the pools, the distance of the pool from the main river, GPS coordinates of isolated pools, the type of the substratum at the bottom of the pools, the relative abundance of other macrofauna (mosquito larva, crayfish etc) in the pools, the state of development of tadpoles (e.g. with or without legs).

Habitat data collected included: the percentage canopy cover, the river velocity, the substrate type, the dominant vegetation, the width of river, and bank, and the physico- chemical parameters in the river (temperature, pH and dissolved oxygen level).

All other frog species seen or heard during the survey were recorded, but no quantitative data were collected. This data was also used to develop and update species notes on the species (to be published)

4.3 Chytrid Survey

Skin swabs were collected from some of the frogs caught, as described below, for obtaining Batrachochytrium dendrobatidis (B. dendrobatidis) samples for PCR testing.

Sampling was conducted during the months May to August using the protocol outlined by Lauren J. Livo (Methods for obtaining Batrachochytrium dendrobatidis (BD) samples for PCR testing) (see Appendix B). Each frog was handled with a new pair of disposable latex gloves covering the hands in order to avoid cross-contamination between individuals. The body surfaces, especially the posterior ventral surface of the frogs were swabbed 25 times. Swabs were then stored in individual tubes with 75% ethanol, and kept out of direct contact with any nucleotide denaturing agents. The diagnostic PCR tests were performed using the Taqman RT-PCR by Pisces Molecular (5311 Western Avenue, Suite E, Boulder, CO 8030).

11

4.4 Statistical Analysis

All statistical permutations were done using the statistical software MINITAB Release 13.32 for Windows. Binary logistic regressions were conducted to examine the relationship between the measured habitat parameters which were assumed to potentially affect the presence of M. olmonae and the presence/absence data (binary response). All means are followed by ± SE. Alpha was set at 0.05.

12

5 Results

5.1 Presence Survey

M. olmonae was detected along all rivers over a total distance of 22.05 km. Figure 5-1 shows the general upland reaches within which they were detected. a) Stream Order

Chi squared association test was conducted to determine the relationship between stream order and frog presence.

Table 5-1: Presence vs Stream order Stream Order Frogs Quadrats Total % Presence Absent Present 1 17 16 33 48 2 86 10 96 10 3 250 39 289 13 4 23 0 23 0

Combining stream orders 3 & 4 for analysis due to small sample of 4th order streams, chi-sq = 32.58, 2 df, P < 0.001.

13 Figure 5-1: Surveyed Areas

14 b) Watersheds

Further comparison of presence was conducted within watersheds of the Main Riodge Forest Reserve (MRFR). North coast tributaries were considered as one watershed.

Table 5-2: Presence vs Watershed Watershed Frog Quadrats Total % Presence Absent Present 1(DR) 24 22 46 48 2(LD) 142 21 163 13 3(BB) 88 5 93 5 4(RR) 80 2 82 2 5(KB) 27 3 30 10 6(AR) 6 2 8 25 7(NCT) 9 10 19 53 1: Doctor’s River; 2: Louis D’Or River; 3: bloody Bay River; 4: Roxborough River; 5: king’s Bay River; 6: Argyle River; 7: North Coast Rivers

Excluding Argyle River as sample too small, chi-sq = 80.0, 5 df, P < 0.001.

c) Interaction between watershed and stream order.

This is to see if stream order explains variation among watersheds or whether there was significant variation after taking differences in stream orders into account. Doing a 3-way analysis results in too small samples, so this was done in two stages:

First stage, compare first order streams which were part of rivers (only Doctor’s River & Roxborough River had first order streams sampled), with the North Coast streams which were not part of larger rivers:

15 Table 5-3: Stream order 1 comparison Stream type Frogs Quadrats Total % Presence Absent Present Stream order 1 8 6 14 43 (river) Stream order 1 9 10 19 53 (Stream)

Chi-sq = 0.31, 1df, P = 0.579. Therefore, first order streams had high presence of frogs, whether they were isolated streams or tributaries of rivers.

Second stage, compare the watersheds using only second and third order streams, which had similar overall levels of 10-13% presence, but exclude fourth order streams which had no presence (analysis a) above):

Table 5-4: Comparison of watersheds using second and third orders streams only Watershed Frog Quadrats Total % Presence Absent Present 1 DR 2+3 22 17 39 44 2 LD 2+3 142 21 163 13 3 BB 2+3 68 5 73 7 4 RR 2+3 73 1 74 1 5 KB 2+3 25 3 28 11 6 AR 2+3 6 2 8 25

Excluding Argyle River as sample too small, chi-sq = 45.2, 4 df, P < 0.001

16 d) Presence of tributaries.

Was presence of frogs associated with presence of a tributary in higher order streams?

Table 5-5: Frog presence vs. Presence of tributaries Stream Order Frogs Quadrats Total % Presence 2-4 Absent Present No tributary1 336 43 379 11 Tributary 23 6 29 21 present

Chi-sq = 2.23, 1df, P = 0.136

e) Predators

Did the presence of predators affect the presence of these stream frogs.?:

Table 5-6: Predators vs. Frog Presence Stream Order Frogs Quadrats Total % Presence Absent Present Abesnt 32 13 45 29 Present 59 35 94 37

Chi-sq = 0.94, 1 df, P=0.333.

17 5.2 Population Structure Table 5-7 below shows the abundance of frogs found within the surveyed areas.

Table 5-7: M. olmonae abundance along survey rivers.

River Survey Area (m2) Abundance

Doctor’s River 2536.5 125

Louis D’Or river 20586.9 28

Bloody Bay river 2240.0 8

Roxborough River 1312.5 2

Argyle River 3000.0 50

King’s bay River 820.0 1

T2 280.0 1

T4 610.0 1

T5 65.0 7

T18 170.0 8

T20 2536.5 9

A total of 126 M. olmonae were measured (Fig.5-2),. The smallest juvenile had a SVL of 9.6 mm. Male M. olmonae ranged from 18.3-25.7 mm. Females also had a maximum size of 25.7 mm

18

Figure 5-2. Size frequency distributions (numbers observed) of Mannophryne olmonae: a) juveniles and females (n = 110); b) males (n = 16)

19 5.3 Habitat Assessment

5.3.1 Tadpole Microhabitat

Eighteen small rock pools close to the river were inspected, nine of these contained M. olmonae tadpoles. All M. olmonae tadpoles were found in rock crevices adjacent to the river or stream with the exception of one locality, a first-order stream near Bloody Bay, where they were found in water-filled tyre ruts. Despite the fact that all of the collection localities of tadpoles were directly adjacent to streams, no M. olmonae tadpoles were seen in the streams themselves. None of the physical characteristics measured differed significantly between pools with and without M. olmonae tadpoles.

Tadpole pools were identified at three locations: Doctor’s River, Bloody Bay River and Argyle River. The dimensions and physico-chemical characteristics of these pools are summarised in Table 5-8 below.

20 Table 5-8: Physical characteristics of pools with and without M. olmonae tadpoles. Values are mean ± SD (with number of pools), and range for pools with tadpoles, and the P value from an unequal-variance t test comparing the means.

Pool Pools with M. Pools without Overall P Parameter olmonae M. olmonae Length (m) 73 ± 69 (8) 20-200 58 ± 35 (9) 0.586 Width (m) 43 ± 33 (8) 11-100 19 ± 8 (9) 0.085 Depth (m) 9.9 ± 7.2 (8) 3.0-25.0 8.0 ± 6.6 (9) 0.581 pH 7.7 ± 0.7 (7) 6.2-8.0 8.1 ± 0.3 (8) 0.156 Temperature 27.4 ± 1.3 (7) 25.2-28.6 27.2 ± 1.0 (8) 0.751 (ºC) DO (mg L–1) 3.2 ± 1.8 (7) 1.72-6.95 3.3 ± 3.0 (8) 0.942 Distance from 2.9 ± 1.3 (6) 1.2-5.1 1.7 ± 1.2 (8) 0.113 River (m) Distance from 1.9 ± 1.4 (6) 0-4.0 1.7 ± 1.9 (8) 0.831 Forest (m)

5.3.2 Adult Microhabitat

In order to assess the effect of different habitat variables on the presence of M. olmonae regression analyses were conducted on those environmental factors. The environmental variables were selected on the assumption that they had the greatest potential of affecting the presence and number of M. olmonae. Table 5-9 below shows the mean of measured environmental variables in which mature frogs were found across all the rivers.

21 Table 5-9: Measured Environmental Variables Environmental variables Frog Present Frog Absent River Width (m) 2.09 ± 3.02(65) 2.94 ± 3.33 (119) River Depth (m) 0.21 ± 0.18 (61) 0.25 ± 0.25 (117) Velocity (m/sec) 0.32 ± 0.19 (58) 0.35 ± 0.19 (116) % Canopy Cover 55.45 45.67 ± (118) pH 8.25 ± 0.87 (42) 8.19 ± 0.92 (77) Temperature (°C) 27.03 ± 1.09 (60) 26.52 ± 2.41 (119) Dissolved Oxygen (mg L–1) 6.97 ± 0.65 (60) 7.19 ± 0.50 (119) Bank Width 3.0 ± 3.64 (51) 3.29 ± 5.29 (87) Distance to Forest Edge 3.74 ± 2.91 (63) 4.30 ± 4.30 (116) Vegetation dominance Heliconia balisier, Heliconia balisier, Bambusa vulgaris Main substrate description stones, gravel and silt and Gravel and silt with banks of boulders and large boulders gravel Main associated fauna Crayfish and fish Mountain Mullet, crayfish

A binary logistic regression was conducted to determine the significance of the synergistic effects of the different parameters on the presence of M. olmonae. Inconsistencies in data collection demanded the occlusion of pH and bank width data, from the analysis (see Appendix C). Major data gaps in these parameters would have meant the inclusion of which would have significantly skewed the outcome of the regression analysis. The analysis showed that temperature (Z= 2.37, P = 0.018) and dissolved oxygen (Z= - 2.03, P = 0.043) were the only significant factors. When temperature and dissolved oxygen were removed from the analysis, stream velocity was the only significant factor (Z= - 2.08, P = 0.038). When dissolved oxygen, temperature and stream velocity were removed from the model, all other factors did not significantly affect the presence of frogs.

22

In order to confirm findings of the binary logistic regression, single regressions were conducted on the significant factors. The analysis showed that dissolved oxygen (T = - 2.54, P = 0.012) was the only factor that significantly affected the presence of M. olmonae at these streams.

All M. olmonae that were located individually were within 10 m of the edge of the river or stream (Figure 5-3). However calling males were occasionally heard farther away from the water. The mean distance from the river or stream was 2.0 ± 2.6 m (n = 68), and more than half of the individuals were within 1.0 m of the water’s edge. The mean distance from the edge of the forest was 1.5 ± 1.1 m (n = 26), with a range from 0-5 m. No M. olmonae were observed within the forest.

Figure 5-3: Frequency distributions (numbers observed) of: distances of M. olmonae frogs from the river or stream edge (n = 68)

23 5.4 Batrachochytridium dendrobatidis d Survey

A total of 124 skin swabs were collected for analysis, of which 91 were from M. olmonae and 40 from other species1 (see Appendix D). Of these, a total of 21 PCR positive samples for B. dendrobatidis were detected, and they were all detected on M. olmonae. Overall, three sites were found to be positive for B. dendrobatidis. Table 5-10 below summarises the results of the chytrid survey.

Figure 5-4: Areas screened for B. dendrobatidis

1 Other species included Bufo marinus and Leptodactylus sp.

24

Table 5-11: PCR Positives for B. dendrobatidis

Species Location Number of Total Percentage of positive Swabs number of swabs that swabs were positive taken Weak = 1 M. olmonae Doctor’s River Positive = 4 Strong = 9 72 26.4 Very strong = 5 Total = 19 North Coast Total = 1 (strong) 9 11.1 Road Argyle River Total = 1 (strong) 5 20.0 Louis D’or Total = 0 3 0.0 Bloody Bay Total = 0 2 0.0 Bufo marinus Man O War Bay Total = 0 26 0.0 Windward Rd Total = 0 6 0.0

25 6 Discussion

6.1 Habitat, Abundance and Population dynamics Mannophryne olmonae were recorded in all surveyed areas, suggesting that this species is still widely distributed in north eastern Tobago. The greatest abundance was noted in the upper areas of the survey streams, where frogs were significantly more associated with first order streams (P < 0.001) rather then higher order streams. When first order streams which are part of rivers (Doctor’s River and Roxborough River), were compared with North Coast streams (not part of larger rivers), there was no significant difference in frog presence between the two, as such this further affirmed that first order streams had high presence of frogs, regardless if there were isolated streams or tributaries of rivers.

Within the wider area, watersheds were compared using only second and third order streams. There was also significant variation in frog presence between watersheds, even when only considering second and first order streams. Overall, both stream order and watershed independently had significant effects. First order streams had highest presence of frogs, whether they were isolated streams or parts of rivers. Second and third order streams had lower presence of frogs, and this varied significantly among watersheds.

The highest abundance was noted within the Doctor’s River watershed and the north coast watersheds. The characteristic of these areas are shown in table 5-9. These streams are generally rocky, heavily shaded narrow and relatively undisturbed by human activities or encroachments. However, lower stream areas, where higher order streams are found exhibited signs of habitat degradation through sand mining, agriculture, river dredging, dumping of garbage and eco-tourism.

A total of 223 frogs were encountered during this survey, and 126 were measured for a population structure assessment. The majority of M. olmonae captured were immature juveniles. The dominance of females in the sample reflected the difficult with which it was to capture males due to camouflage and location. Generally, males tended to be farther from the water, where their black camouflage allowed them to seamlessly blend

26 into the rocky forest background. Furthermore, the adult sex ratio of approximately 2:1 did not vary significantly at the different rivers. There was no evident sexual sexual dimorphism, where mean adult sizes were not significantly different: females 21.6 ± 2.2 mm (n = 33); males 21.0 ± 2.4 mm (n = 15) (t = 0.82, 24 df, P = 0.42).

6.2 Batrachochytrium dendrobatidis

These results demonstrate conclusively that Batrachochytrium dendrobatidis occurs in Tobago, and in the Critically Endangered frog Mannophryne olmonae; a worrying finding on both counts. Over 90% of the frogs surveyed were M. olmonae, with the other species consisting of Bufo marinus, and Leptodactylus sp. Overall, 16.9 % of the samples were PCR positive for B. dendrobatidis endrobatidis. However, infected frogs showed neither clinical signs nor symptoms of infection (lethargy, skin lesions) nor did they behave abnormally, which suggests B. dendrobatidis may have no effect on this species. The lack of these clinical signs, suggest that chytridiomycosis may not be pathogenic and responsible for the recent population declines.

One school of though have suggested that B. dendrobatidis may not be pathogenic to some species of infected frogs due to host-pathogen co-evolution or the production of antimicrobial skin peptides (which some studies have linked to chytridiomycosis resistance). It is known, however, that amphibian species vary in their susceptibility to the disease (Blaustein et al., 2005; Garcia et al., 2006). Additionally, even susceptible species can recover to leave B. dendrobatidis isolated to this species rather than an epidemic disease (McDonald et al., 2005; Woodhams and Alford, 2005; Kriger and Hero, 2006; Puschendorf, Bolanos et al., 2006).

Infections were not evenly distributed across all the sampled water courses, where it was detected at three of the survey areas (see Figure 5-4). The highest incidences of infections were detected at Doctor’s River. The vast majority of infections were limited to the south sloping side of the Main Ridge Forest reserve (MRFR). This physical barrier may retard the spread of the disease to the north sloping side of the MRFR.

27

There are no historical population data on M. olmonae, so it is therefore unknown whether current populations have recovered to previous levels, or remain at a reduced density. It is also unclear whether any population decline was caused by epidemic chytridiomycosis. Hardy’s 2004 assessment of M. olmonae in the IUCN Red List reported an estimated 80% population decrease in recent years however the basis for this is unclear. It is clear, however, that widely spread populations of M. olmonae currently harbour endemic chytridiomycosis, apparently without clinical signs. This finding raises several questions and implications for amphibian health in Tobago, and other Caribbean islands.

Pending a further study looking at the degree of spread of the fungus, chytridiomycosis is now endemic the Mannophryne olmonae population, rather than epidemic (affecting other amphibians). It remains possible that increased mortality could result from an interaction of chytridiomycosis with other factors (Lampo, Rodríguez-Contreras et al., 2006). The most likely interaction, with temperature, would however act against increasing severity of the disease. Chytridiomycosis is both more prevalent and more dangerous at lower temperatures in the laboratory and natural environments (Berger et al., 2004; Piotrowski et al., 2004; Drew et al., 2006). Thus although global warming might increase the susceptibility of montane amphibians by bringing their environments within the optimum temperature range of B. dendrobatidis (Pounds et al., 2006), this is unlikely in the lowland and lower-montane frog M. olmonae. This species was associated with a range of water temperatures in both isolated pools (Table 5-8) and in rivers and streams (Table 5-9), already slightly above the optimum for B. dendrobatidis (17-25ºC; Piotrowski et al., 2004), and the altitude of the highest peak in Tobago is only 549 m (Murphy, 1997). By itself, any increase in environmental temperature would seem to make epidemic chytridiomycosis less rather than more likely in M. olmonae and other amphibians in Tobago. Increased susceptibility to chytridiomycosis from climate change is unlikely in amphibian populations in Tobago, as this island does not have high montane environments, but remains a possibility in the sister island of Trinidad

28

7 Measures for Conservation

7.1 Future of Mannophryne olmonae

The survey highlighted the importance of first order steams and watershed to M. olmonae. Currently, M. olmonae is afforded protection, under the umbrella designation of the Main Ridge Forest as a forest reserve. However the relatively low numbers of frogs recorded is cause for concern. Very little is known about the exact niche (role that M. olmonae plays in the environment) or its biology. In order to effectively manage this species, further study is required to monitor the population status and trends of this species.

The only major threat facing the frog is the frog disease chytridiomycosis, (a skin disease caused by Batrachochydium dendrobatidis). B. dendrobatidis has been implicated as the cause of frog decline in Australia, New Zealand, Central America countries and elsewhere (Berger et al., 1998, Berger et al., 1999) (see Section 1.4). This is the first know incidence of the fungus in Tobago, also raises conservation concerns for the other amphibian fauna on the Trinidad and Tobago. The species of most concern in this respect would be the montane golden tree frog (Phyllodytes auratus) of Trinidad, which occurs only above 800 m (Murphy, 1997). A survey of chytridiomycosis in Trinidad, and preventing the possible introduction of B. dendrobatidis to high altitude sites in that island, is clearly a key requirement for conservation of amphibians in Trinidad and Tobago.

The exact cause of the population decline of M. olmonae is not known, and maybe as a result of one cause or a synergy of causes. However, this does not negate the fact that the species is need of conservation. This need has been recognised by the Environmental Management Authority, where it has designated M. olmonae 5th on the list of Environmentally Sensitive Species to be addressed by the EMA for 2007

29 7.2 Education and Public Awareness programme Project L.E.A.P. has made conservation education is a major aim of this project and continues to work with schools. To date the Project LEAP has been involved several formal and informal education programmes and workshops where the theme has been one of conservation of biodiversity. The target demographic so far has been 18 and under, (typical demographic found within the group found within primary and secondary schools). Wider audiences have been targeted through various public symposiums on the state of vanishing species in Trinidad and Tobago. It is hope that other conservation based interest groups and stakeholders will get involved in this endeavour.

A series of posters has also been developed and distributed to schools and through the Environmental Management Authority about the Bloody Bay Poison frog, local biodiversity in Trinidad and Tobago, threats facing biodiversity and how everyday citizens can assist in conservation of the natural environment

During the course of this study, presentations have been held a various youth summer camps and schools on both Trinidad and Tobago. A volunteer programme of the group has proved to be quite successful. Project LEAP has so far recruited and trained 16 volunteers in the basics of conducting amphibian studies ecological surveys. Some of these were even teacher and have incorporated conservation education into their teaching programmes.

7.3 Follow up Activities

The follow-up activities will include the following:

1. Annual monitoring of M. olmonae population as well as further screening on the frog to determine the extent of spread of B. dendrobatidis. Currently, remaining funds from the project (see Appendix E) are being used to fund additional B. dendrobatidis screening on Trinidad.

30 2. Conservation actions will focus on different awareness-raising programs to reach all groups in the communities where the frogs are found and all the stakeholders that use the resources (radio, TV, etc).

3. Strengthening and capacity building of park wardens for working with amphibians, conducting population surveys and B. dendrobatidis screening

4. Recommending the establishment of amphibian conservation zones within the MRFR, which should ensure the protection of key areas within the MRFR. Further, these designations should afford protect to the watersheds and the fragile tributaries that make up these watersheds where the frog is found. Many areas of frog habitat under direct human impact and are readily accessed. These activities should be regulated.

31 Acknowledgements: Fieldwork was supported by a 2006 Future Conservationist Award from the Conservation Leadership Programme (Mannophryne olmonae: An ecological study in Tobago - Project L.E.A.P.), and seed grants from Conservation International. We thank Dave Hardy for information on locations of M. olmonae, Giancarlo “Gian” Lalsingh and Pat Turpin of Environment Tobago for accommodation and logistical help, numerous assistants for help with fieldwork, and three referees for useful comments. Handling frogs was under permit from the Wildlife Section of the Forestry Division of Trinidad and Tobago, and the Department of Natural Resources and the Environment of the Tobago House of Assembly, courtesy of Nadra Nathai-Gyan and Angela Ramsey, respectively.

32 8 References

Baillie, J. E. M., Hilton-Taylor, C. and Stuart, S.N., Hardy, J. (Assessor) (2004). 2004 IUCN Red List of Threatened Species. A Global Species Assessment. Berger, L., R. Speare, et al. (1998). "Chytridiomycosis causes amphibian mortality associated with population declines in the rainforests of Australia and Central America." Proceedings of the National Academy of Science USA 95: 9031-9036. Berger, L., Speare, R., Hyatt, A. Chytrid fungi and amphibian declines: Overview, implications and future directions. Declines and Disappearances of Australian Frogs. Ed. A. Campbell. Environment Australia: Canberra. 1999:23-33. Berger, L., Speare, R., Hines, H., Marantelli, G., Hyattm A. D., McDonald, K. R., Skerratt, L. F., Olsen, V., Clarke, J. M., Gillespie, G., Mahony, M., Sheppard, N., Williams, C., Tyler, M. Effect of season and temperature on mortality in amphibians due to chytridiomycosis. Australian Veterinary Journal 2004;82:31- 36. Blaustein, A. R.; Romansic, J. M., Scheessele,E. A., Han,B. A., Pessier, A. P., Longcore J. E. (2005). Interspecific Variation in Susceptibility of Frog Tadpoles to the Pathogenic Fungus Batrachochytrium dendrobatidis. Conservation Biology 19 (5), 1460–1468. Burrowes, P. A., Joglar, R., Green, D. E. (2004). "Potential causes for amphibian declines in Puerto Rico." Herpetologica 60(2): 141-154. Daszak, P. and Cunningham, A. A., (1999). "Extinction by infection." Trends in Ecology and Evolution 14(7): 279. Drew, A., Allen, E. J., Allen, L. J. Analysis of climatic and geographic factors affecting the presence of chytridiomycosis in Australia. Diseases of Aquatic Organisms 2006;68(3). Fa, J., Hedges, B., Ibéné, B., Breuil, M., Powell, R. & Magin, C. (2004). Leptodactylus fallax. 2006 IUCN Red List of Threatened Species. Garcia, T. S., Romansic, J. M., Blaustein, A. R.. Survival of three species of anuran metamorphs exposed to UV-B radiation and the pathogenic fungus

33 Batrachochytrium dendrobatidis. Diseases of Aquatic Organisms 2006;72:163- 169 Grant, T, Frost, D. R.,.Caldwell, J. P., Gagliardo, R., Haddad, C. F. B., Kok, P. J. R., Means, D. B., Noonan, B. P., Schargel, W. E., and Wheeler, W. C. (2006). "Phylogenetic systematics of dart-poison frogs and their relatives (Amphibia: Athesphatanura: Dendrobatidae). ." Bull. Amer. Mus. Nat. Hist. 299: 1-262. Gibbons, J. W., Travis, D. E. S., Ryan, J. , Bulmann K. A., Tuberville T. D., Metts, B. S., Greene, J. L., Mills, T., Leiden, Y., Poppy, S., and Winne C. T. (2000). "The Global Decline of Reptiles, Déjà Vu Amphibians." Bioscience 50(8): 653-666. Kriger, K. M., Hero, J. M., Ashton, K. J.; Cost efficiency in the detection of chytridiomycosis using PCR assay. Diseases of Aquatic Organisms 2006;71:149- 154. La Marca, E. (1992). Amphibian Species of the World - Mannophryne. Lampo, M., Rodríguez-Contreras, A., La Marca, E., and Daszak, P.. 2006 A Chitridiomycosis epidemic and a severe dry season precede the disappearance of Atelopus species from the Venezuelan Andes.Herpetological Journal 16(4): 395- 402 Lips, K. R. (1999). "Mass Mortality and Population Declines of Anurans at an Upland Site in Western Panama." Conservation Biology 13(1): 117-125. Lips, K. R., Green, D. E. et al. (2003). "Chytridiomycosis in wild frogs from southern Costa Rica." Journal of Herpetology 37(1): 215-218. Lips, K. R., Reeve, J. D. et al. (2003). "Ecological traits predicting amphibian population declines in Central America." Conservation Biology 17(4): 1078-1088. McDonald, K. R., Mendez, D., Muller, R., Freeman, A. B., Speare, R.. Decline in the prevalence of chytridiomycosis in upland frog populations in North Queensland, Australia. Pacific Conservation Biology 2005;11(2):114-120. Piotrowski, J. S., Annis, S. L. and Longcore, J. E. (2004). Physiology of Batrachochytrium dendrobatidis, a chytrid pathogen of amphibians. Mycologia, 96(1), 2004, pp. 9-15. Pounds, A. J , Bustamante, M. R , Coloma, L. A., Consuegra, J. A., Fogden, M. P. L., Foster, P. N., la Marca, E., Masters, K. L., Merino-Viteri, A., Puschendorf, R,

34 Ron, S. R., Sanchez-Azofeifa, G. A, Still, C. J., Young, B. E. Widespread amphibian extinctions from epidemic disease driven by global warming. Nature 2006:439:161-167 Puschendorf, R., Bolanos, F., Chaves, G. The amphibian chytrid fungus along an altitudinal transect before the first reported declines in Costa Rica. Biological Conservation 2006;132(1):136-142 Weldon, C.,du Preez, L. H., et al. (2004). "Origin of the amphibian chytrid fungus." Emerging Infectious Diseases 10(12): 2100-2105. Woodhams, D. C., Alford, R. A., Marantelli, G. Emerging disease of amphibians cured by elevated body temperature. Diseases of Aquatic Organisms 2003;55:65-67.

35

APPENDIX A

TADPOLE KEY Simplified key to identify Mannophryne tadpoles (2) (Modified from Kenny, 1969)

Mannophryne Others a) Tail shape Tapered Rounded (e.g. Bufo marinus) b) Tail colour Prominent spotting Unspotted or diffuse pigment c) Spiracle Sinistral (left side) Ventral or rectal d) Eyes Dorsolateral Lateral e) Dorsal : Ventral fin width 1:1 Not 1:1 f) Upper lip 2 tooth rows 1, 3 or 4 tooth rows g) Lower lip 3 tooth rows 4 tooth rows

APPENDIX B

Batrachochytrium dendrobatidis sampling Methods for obtaining Batrachochytrium dendrobatidis (BD) samples for PCR testing

Revised 2004 by Lauren J. Livo Department of Integrative Physiology, CB-354 University of Colorado Boulder, CO 80309-0354 Email: [email protected]

General considerations

In terms of PCR samples, bleach and flame destroy DNA, while alcohol preserves DNA. Collection of samples for PCR testing requires that equipment used to collect samples not inhibit DNA detection while also not contaminating the current sample with DNA from a previous sample. In addition, equipment should be decontaminated so that it does not spread BD (or other pathogens) from one animal to another.

The PCR sample collection methods described here have been used for obtaining samples from Colorado amphibians for BD (Batrachochytrium dendrobatidis) detection. These methods should apply generally to collection of samples for BD testing from amphibians.

Steps for obtaining samples

Collect animals

Animals should be collected with clean, decontaminated equipment, individually handled with fresh disposable gloves, and placed in individual containers prior to obtaining the samples. Although using Purell or other hand decontamination solutions may prevent the spread of live BD from one animal to another, it is likely to allow contamination of samples with BD DNA (in other words, if you handle a BD-negative animal after handling a BD-positive animal, the PCR samples you obtain may both appear to be positive for BD).

Do not place multiple animals in the same container prior to sampling. In this situation, a single infected animal could infect others, and PCR tests could have inflated numbers of positive test results.

Equipment (such as individual containers for holding animals) can be cleaned and bleached so that they can be reused. However, this equipment must be rinsed well and allowed to dry prior to reuse so that there is no residual bleach (note that even parts per million bleach in/on/around a sample could possibly destroy all of the DNA in a sample over the course of a few weeks.)

Obtaining samples

Skin swabs or skin scrapes are the preferred methods of collecting samples from live individuals as the same individual can be tested repeatedly over time. Skin swabs appear to be more gentle to the animal than skin scrapes. Both skin swabs and skin scrapes have similar rates of false negative results when tested on known BD-positive individuals (2 of 15 skin swab and 3 of 15 skin scrape false negative rates in one experiment).

Obtain the PCR sample before doing other procedures with the animal (for example, before weighing, checking PIT tags, and so on). Samples require the following equipment:

• Swabs or pointed sticks: o Swabs: use cotton swabs on 2mm-diameter wood without adhesive (such as Puritan Cotton-Tipped Applicators, #VWR 10806-005, or equivalent) cut to lengths (ca. 3-cm) that fit into 2-ml tubes.

o Skin scrapes: small (ca. 3-cm) lengths of wooden applicators cut so one end is pointed (6-inch long, 2mm-diameter wooden Puritan Applicators, cut into about 3- cm lengths, VWR catalog # 10805-018, or equivalent)

• 2-ml screw-cap tubes containing 1 ml of 70 percent ethanol (2.0 ml screw cap tube with cap/500 per bag, VWR catalog # 20170-217). (Note: Although other tubes are available, the VWR tubes have been very good at avoiding leakage.)

To obtain the sample, hold the animal (using fresh gloves) in one hand, and gently but firmly swab (with the cotton swab) or scrape (using the pointed end of the stick) the ventral surface 25 times; for large animals, you may swab or scrape the ventral surface 20 times and the feet and webbing 5 times.

Figure 1. Swabbing ventral surface of amphibian.

Figure 2. Scraping ventral surface of amphibian.

Place the swab (cotton side down) or stick (pointed end down) in the tube. Secure the lid and place in a rack or other container so that the tube remains upright. (Leakage from one tube with BD may get on other tubes and result in contamination of your samples.)

Figure 3. Insert swab or stick into tube with sample at bottom of tube.

Other skin tissues (such as toe clip samples or samples of ventral skin from dead animals) may also be collected for PCR testing. Use fine scissors to obtain the tissue. Between each sample, clean the scissors with an ethanol-soaked swab or tissue, and then hold the blades over an open flame to destroy any DNA from the previous sample. Place each sample in a 2 ml tube containing 1 ml of 70 percent ethanol.

Figure 4. Cleaning scissor blades with alcohol.

Figure 5. Passing scissor blades through flame to destroy residual DNA.

Toe clipping: If you collect toe clips from live individuals, use fine scissors to amputate the toe tip. When selecting a toe to amputate, you should avoid especially important digits such as the thumb, to avoid having an undue effect on the ability of the animal to feed, reproduce, and so on. Toe clips have rates of false-negative results similar to skin scrapes, but may have more potential for false-positive results through contamination. To obtain a toe clip, cut off the toe tip with the scissors. If bone protrudes from the wound, trim the bone further back (preferably to a joint) so that skin covers the wound, then dab a drop of Vet-bond or other sealant on the wound. In my study, I selected the right rear toe, and continued to encounter individuals with this digit missing 2 to 3 years after the initial sampling.

Labeling samples

Label each tube with a unique sample number. Other information that should be associated with each sample includes:

• solution (for example, 70 percent ethanol)

• sample type (for example, skin swab, skin scrape, pelvic patch sample, or toe clip)

Figure 6. Label on tube.

Do not place sample information inside the tube. (It can be difficult to extricate, may contaminate other samples through handling, and as paper may contain bleaching agents, may inhibit detection of the target DNA).

Designate a plastic bag for the disposal of gloves and other materials (for example, alcohol wipes) to minimize the possibility of contamination.

Shipping samples

Samples to be shipped should be in a box or mailing tube, not a bubble wrap envelope. The Post Office canceling machine can exert enough force on samples in an envelope to cause the tubes to leak.

Acknowledgments

Seanna Annis (Department of Biological Sciences, University of Maine, Orono, Maine 04469) helped develop initial sampling techniques, which were further refined by discussions with John Wood (Pisces Molecular, 2200 Central Avenue, Suite F, Boulder, CO 80301). Alex Hyatt (Australian Animal Health Laboratory, Commonwealth Scientific Industrial Research Organization, Geelong, Victoria, Australia) recommended using swabs for PCR sampling of amphibians for Batrachochytrium dendrobatidis. Cynthia Carey (Dept. Integrative Physiology CB- 354, University of Colorado, Boulder, CO 80309-0354) provided amphibians used in some laboratory evaluations of methods. Funding for PCR testing was provided by Great Outdoors Colorado through a grant to the Colorado Division of Wildlife.

APPENDIX C

Raw Data raw data

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 ID Code River Code Stream order Frog Present Tributary present River Width River Depth velocity pH temperature DO Bank Width Dist. To forest edge % canopy cover 1 DR-100 1 3 0 0 4.2000 0.0350 0.1356 8.59 27.30 7.32 1.90 3.650 80.0 2 DR-150 1 3 0 0 1.1200 0.2700 0.1784 8.69 27.60 7.37 2.30 5.900 20.0 3 DR-200 1 3 0 0 0.5600 0.1400 0.2716 8.72 26.50 7.33 2.55 3.250 80.0 4 DR-250 1 3 1 0 0.5600 0.1500 0.7647 8.75 26.30 5.92 3.70 5.350 85.0 5 DR-300 1 3 1 0 1.8700 0.3600 * 8.74 26.20 7.69 0.00 3.850 60.0 6 DR-350 1 3 1 0 1.6200 0.3400 0.2503 8.68 26.30 7.74 2.25 5.600 50.0 7 DR-400 1 3 1 0 1.7300 0.2900 0.2066 8.30 28.00 6.96 2.55 2.550 45.0 8 DR-450 1 3 1 0 1.2600 0.3500 0.5322 8.45 28.90 6.37 1.15 2.300 90.0 9 DR-600 1 3 0 1 1.1700 0.0400 0.2915 8.48 29.20 6.72 4.00 6.650 0.0 10 DR-650 1 3 1 0 1.8400 0.0500 0.3763 8.52 29.90 6.50 5.10 3.600 0.0 11 DR-700 1 3 1 0 1.7200 0.0400 0.4039 8.26 28.50 6.96 4.05 0.000 10.0 12 DR-750 1 3 0 0 2.0800 0.0500 0.1152 8.44 28.30 6.57 2.60 7.400 40.0 13 DR-800 1 3 0 0 2.7600 0.0400 0.2933 8.46 28.60 6.19 0.00 2.600 80.0 14 DR-850 1 3 1 0 2.4000 0.1400 0.1264 8.47 28.10 6.08 1.70 0.000 95.0 15 DR-900 1 3 0 0 1.2800 0.2800 0.3035 8.51 27.60 6.36 4.30 6.050 90.0 16 DR-950 1 3 1 1 1.5100 0.0900 0.1262 8.53 27.20 6.33 4.70 2.900 35.0 17 DR-1100 1 3 1 1 2.5000 0.0900 0.1143 8.49 26.90 6.24 3.00 2.300 20.0 18 DR-1150 1 3 0 0 1.0500 0.0900 0.4512 8.37 26.70 6.27 0.00 2.750 95.0 19 DR-1200 1 3 0 0 1.3000 0.1200 0.0694 8.23 26.20 5.48 6.45 6.000 40.0 20 DR-1250 1 2 0 0 2.0000 0.0600 0.2783 8.42 26.60 6.27 4.35 3.750 5.0 21 DR-1300 1 2 0 1 3.5000 0.0700 0.2533 8.36 26.90 5.47 0.00 0.000 100.0 22 DR-1350 1 2 1 0 0.6700 0.0600 0.2635 8.45 26.50 6.14 2.20 1.900 60.0 23 DR-1400 1 2 1 0 1.5600 0.2000 0.0721 8.50 27.20 6.11 3.80 4.900 80.0 24 DR-1450 1 2 0 0 0.9000 0.1400 0.1784 8.42 26.90 6.45 0.90 2.950 85.0 25 DR-1500 1 2 1 1 0.8100 0.0700 0.1408 8.37 26.90 6.11 3.50 1.100 65.0 26 DR-1550 1 2 1 1 1.0000 0.1900 0.2279 8.37 27.30 5.88 1.65 1.200 35.0 27 DR-1600 1 2 0 1 1.5800 0.2400 0.0397 8.52 26.30 6.05 0.80 0.900 25.0 28 DR-1650 1 2 0 0 0.9000 0.0600 0.2245 8.55 26.60 6.75 0.00 1.000 80.0 29 DR-1700 1 2 0 0 0.7100 0.1000 0.1589 8.49 26.80 6.06 0.00 0.900 90.0 30 DR-1750 1 2 0 0 0.9000 0.0700 0.0701 8.57 26.80 6.49 1.00 2.700 60.0 31 DR-1800 1 2 1 0 0.9000 0.0800 0.3685 8.54 26.00 6.46 1.85 2.200 90.0 32 DR-1850 1 2 1 0 1.0000 0.0900 0.1153 8.56 27.00 6.00 1.95 3.500 85.0 33 DR-1900 1 2 1 0 0.1200 0.2500 0.1104 8.56 26.20 6.25 0.00 1.750 95.0 34 DR-1950 1 1 1 0 0.9000 0.7000 0.2639 8.50 26.60 6.53 21.45 1.600 90.0 35 DR-2000 1 1 1 1 2.0100 0.1600 0.2261 8.61 26.50 6.84 0.50 2.300 85.0 36 DR-2050 1 1 1 0 0.9000 0.8000 0.0536 8.64 26.40 6.34 0.00 1.100 80.0 37 DR-2100 1 1 0 1 0.6500 0.0000 * 8.49 26.30 6.16 0.00 0.900 85.0 raw data

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 ID Code River Code Stream order Frog Present Tributary present River Width River Depth velocity pH temperature DO Bank Width Dist. To forest edge % canopy cover 38 DR-2150 1 1 1 0 0.7000 0.8000 0.0979 8.39 26.10 5.73 0.80 0.800 60.0 39 LD-3100 2 3 0 0 8.6600 0.0900 0.3890 8.66 26.80 7.57 4.65 4.650 90.0 40 LD-3150 2 3 0 0 5.4000 0.1550 0.4260 8.80 27.10 7.43 2.30 2.300 0.0 41 LD-3200 2 3 0 0 5.4000 0.3250 0.2920 8.63 27.15 6.99 0.00 0.000 * 42 LD-3250 2 3 0 0 6.2700 0.2000 0.2030 8.69 27.25 7.29 5.01 5.010 30.0 43 LD-3300 2 3 1 0 5.5000 0.2250 0.2240 8.73 28.00 7.22 5.55 5.550 30.0 44 LD-3350 2 3 0 0 7.6000 0.2700 0.6070 8.64 27.70 7.32 4.65 15.000 15.0 45 LD-3400 2 3 0 0 6.8900 0.2300 0.5550 8.69 27.90 7.56 4.60 4.800 65.0 46 LD-3450 2 3 1 0 10.7000 0.0450 0.4440 8.73 27.90 7.04 2.28 9.000 55.0 47 LD-3500 2 3 0 0 11.6000 0.0800 0.2380 8.62 27.60 7.24 1.10 12.000 0.0 48 LD-3550 2 3 0 0 6.4000 0.0700 0.5180 8.61 27.80 7.26 1.30 5.750 50.0 49 LD-3600 2 3 1 0 6.2000 * 0.3710 8.59 27.65 7.19 1.93 3.050 85.0 50 LD-3650 2 3 0 0 3.5200 0.1500 0.4470 8.43 28.10 7.15 2.10 3.250 80.0 51 LD-3700 2 3 1 0 2.2000 * 0.8610 8.55 29.10 7.22 11.30 11.300 80.0 52 LD-3750 2 3 1 0 3.2000 0.1100 0.7160 8.64 28.25 7.49 5.90 5.900 45.0 53 LD-3800 2 3 1 0 12.3000 0.1600 0.3990 8.57 28.80 6.79 2.20 7.050 0.0 54 LD-3850 2 3 1 0 5.5000 * 0.4430 8.57 28.85 6.47 5.20 8.150 55.0 55 LD-3900 2 3 1 0 3.0000 0.1350 0.2820 8.48 28.65 6.77 2.67 3.700 70.0 56 LD-3950 2 3 0 0 4.6000 0.2500 0.5430 8.61 28.30 7.15 0.70 0.700 80.0 57 LD-4000 2 3 0 0 10.0000 0.0460 0.7110 8.69 28.40 7.62 3.75 3.750 80.0 58 LD-4050 2 3 0 0 9.2000 0.0350 0.5420 8.66 28.75 7.41 4.45 7.750 75.0 59 LD-4500 2 3 0 0 0.0620 0.0700 0.5510 8.64 26.60 7.72 1.70 2.900 20.0 60 LD-4550 2 3 0 0 0.0383 0.2200 0.2770 8.72 26.90 7.08 0.82 2.320 40.0 61 LD-4600 2 3 0 0 0.0450 0.1100 0.6760 8.73 27.00 7.20 0.00 1.500 20.0 62 LD-4650 2 3 0 0 0.0350 0.3300 0.2740 8.72 27.10 7.03 6.10 8.400 85.0 63 LD-4700 2 3 0 0 0.0550 0.3650 0.1330 8.73 27.00 7.23 33.50 5.000 60.0 64 LD-4750 2 3 0 0 0.0373 0.2400 0.5010 8.63 26.95 7.34 0.60 3.335 80.0 65 LD-4800 2 3 1 0 0.0570 0.2500 0.4020 8.68 27.15 7.19 2.10 3.600 30.0 66 LD-4850 2 3 0 0 0.0931 0.1600 0.4360 8.54 27.00 7.02 0.70 6.450 20.0 67 LD-4900 2 3 1 0 0.0870 0.2300 0.2230 8.64 27.00 6.93 0.00 6.150 50.0 68 LD-4950 2 3 0 0 0.1080 0.0900 0.7930 8.62 27.20 7.15 2.40 2.850 40.0 69 LD-5000 2 3 0 0 0.0940 0.0750 0.3070 8.64 27.40 7.01 3.95 3.950 45.0 70 LD-5050 2 3 0 0 0.0830 0.0825 0.2920 8.65 27.25 7.23 0.00 4.000 90.0 71 LD-5100 2 3 1 0 0.0345 0.1700 0.4440 8.68 27.25 7.16 10.25 12.950 95.0 72 LD-5150 2 3 0 0 0.0560 0.4700 0.0630 8.52 27.40 6.86 0.00 0.000 60.0 73 LD-5200 2 3 0 0 0.1350 0.1200 0.4470 8.66 27.35 7.07 1.60 2.700 90.0 74 LD-5250 2 3 0 0 0.0380 0.2600 0.1760 8.63 27.40 6.95 2.10 2.950 90.0 raw data

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 ID Code River Code Stream order Frog Present Tributary present River Width River Depth velocity pH temperature DO Bank Width Dist. To forest edge % canopy cover 75 LD-5300 2 3 0 0 0.0600 0.1100 0.4440 8.73 27.70 7.44 2.15 2.150 30.0 76 LD-5350 2 3 0 0 0.0000 0.0750 0.1910 8.75 27.60 7.34 5.50 5.500 60.0 77 LD-5400 2 3 0 0 0.0630 0.1250 0.3240 8.74 27.70 7.28 1.20 1.200 25.0 78 LD-5450 2 3 1 0 0.0355 0.1700 0.6390 8.64 26.40 7.14 1.57 1.570 35.0 79 LD-5500 2 3 0 0 0.0650 0.1000 0.6690 8.55 26.70 7.02 5.32 5.320 55.0 80 LD-5550 2 3 1 0 0.0547 * 0.3450 8.59 26.80 7.11 1.69 1.690 60.0 81 LD-5600 2 3 0 0 0.0295 0.1200 0.1310 8.60 26.90 6.94 9.62 18.850 80.0 82 LD-5650 2 3 0 0 0.1477 0.1400 0.7750 8.26 26.80 7.46 3.00 3.000 5.0 83 LD-6250 2 3 0 0 0.0680 0.0000 0.4040 8.70 27.15 7.18 0.00 0.000 25.0 84 LD-6300 2 3 0 0 0.0700 0.1000 0.7070 8.67 27.55 6.77 0.00 4.500 20.0 85 LD-6350 2 3 0 0 0.0370 0.1500 0.5020 8.62 27.00 7.28 0.90 12.100 25.0 86 LD-6400 2 3 0 0 0.1100 0.0600 0.6760 8.73 27.10 7.13 0.00 19.550 40.0 87 LD-6450 2 3 0 0 0.0600 0.2400 0.2490 8.55 26.90 7.22 1.90 1.900 30.0 88 LD-6500 2 3 0 0 0.0725 0.1500 0.4630 * 26.90 7.10 * 0.000 1.0 89 LD-6550 2 3 1 0 0.0690 0.1600 0.4230 * 26.90 7.11 * 4.300 50.0 90 LD-6600 2 3 0 0 0.0540 0.1200 0.3340 * 27.00 7.03 * 7.000 50.0 91 LD-6650 2 3 0 0 0.0770 0.1200 0.5210 * 27.00 6.99 * 2.900 40.0 92 LD-6700 2 3 0 0 0.0475 0.1300 0.5160 * 27.30 6.88 * 8.600 1.0 93 LD-6750 2 3 1 0 0.0865 0.1300 0.1950 * 27.10 7.10 * 6.200 20.0 94 LD-6800 2 3 1 0 0.0400 0.2100 0.5380 * 27.10 6.86 * 0.000 0.0 95 LD-6850 2 3 0 0 0.0330 0.2200 0.4690 * 27.20 6.84 * 0.000 70.0 96 LD-6900 2 3 0 0 0.0350 0.1700 0.3760 * 27.20 6.79 * 0.000 0.0 97 LD-6950 2 3 1 0 0.0320 0.1600 0.4310 * 27.20 7.01 * 5.100 20.0 98 LD-7000 2 3 1 0 0.0340 0.2100 0.4380 * 27.30 6.75 * 4.100 1.0 99 LD-7050 2 3 0 0 0.0515 0.1400 0.7550 * 27.50 7.14 * 0.000 10.0 100 LD-7100 2 3 0 0 0.0400 0.2500 0.1510 * 27.60 7.10 * 3.600 30.0 101 LD-7150 2 3 0 0 0.0570 1.0000 0.0930 * 27.60 7.17 * 1.000 85.0 102 LD-7200 2 3 0 0 0.0240 0.2400 0.6670 * 27.60 7.07 * 7.300 55.0 103 LD-7250 2 3 0 0 0.0690 0.2200 0.4850 * 27.70 7.23 * 2.800 40.0 104 LD-7300 2 3 1 0 0.0755 0.4200 0.3020 * 27.80 7.11 * 3.500 10.0 105 LD-7350 2 3 1 0 0.0395 0.2100 0.5930 * 27.80 7.50 * 0.000 0.0 106 LD-7400 2 3 1 0 0.0310 0.4000 0.6250 * 28.00 7.57 * 7.550 0.0 107 LD-7450 2 3 1 0 0.0985 0.3000 0.2300 * 27.90 7.83 * 0.000 0.0 108 LD-7500 2 3 1 0 0.0520 0.3000 0.3340 * 27.80 7.55 * 9.200 0.0 109 LD-7550 2 3 1 0 0.0620 0.2000 0.4940 * 27.70 7.47 * 4.400 0.0 110 LD-7600 2 3 0 0 0.0760 0.3300 0.2300 * 27.30 7.54 * 4.500 0.0 111 LD-7650 2 3 1 0 0.0520 0.3700 0.2040 * 27.30 7.49 * 8.500 0.0 raw data

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 ID Code River Code Stream order Frog Present Tributary present River Width River Depth velocity pH temperature DO Bank Width Dist. To forest edge % canopy cover 112 LD-7700 2 3 0 0 0.0500 0.1900 0.3160 * 27.50 7.40 * 5.750 0.0 113 LD-7750 2 3 0 0 0.0600 0.3000 0.2360 * 22.30 7.57 * 2.000 0.0 114 LD-7800 2 3 0 0 0.0430 0.1900 0.7020 * 27.30 7.39 * 5.750 10.0 115 LD-7850 2 3 0 0 0.0580 0.2000 0.4380 * 26.10 7.57 * 7.100 1.0 116 BB-1350 3 3 0 0 8.9000 0.0700 0.2670 8.24 26.80 7.67 0.75 4.210 60.0 117 BB-1400 3 3 0 0 6.5000 0.1900 0.3280 8.18 26.15 6.80 2.80 4.800 25.0 118 BB-1450 3 3 0 0 3.4000 * 0.1970 8.16 26.40 7.10 1.55 4.325 20.0 119 BB-1500 3 3 0 0 6.6000 0.1200 0.3800 8.26 27.15 6.85 16.40 11.625 0.5 120 BB-1550 3 3 0 0 4.5700 0.1300 0.4050 8.21 26.70 6.40 21.05 13.725 10.0 121 BB-1600 3 3 1 0 11.7000 0.2200 0.0520 8.27 26.90 6.36 5.90 6.130 40.0 122 BB-1650 3 3 0 0 6.2000 0.1400 0.2970 8.36 26.70 6.89 2.80 4.845 5.0 123 BB-1700 3 3 0 0 7.2000 0.2100 0.3470 8.27 27.00 7.23 5.85 6.540 0.0 124 BB-1750 3 3 0 0 4.8000 0.1400 0.5100 6.69 27.10 8.24 1.75 4.995 5.0 125 BB-1800 3 3 0 0 7.4000 0.6700 0.1610 8.26 27.05 6.77 0.00 3.385 10.0 126 BB-1850 3 3 0 0 3.6000 0.7800 0.1730 8.28 27.15 6.70 25.00 * 70.0 127 BB-3350 3 3 0 1 3.5500 1.2000 0.1300 8.65 25.35 7.65 0.75 4.200 70.0 128 BB-3400 3 3 0 0 1.5900 0.2700 0.5620 * 25.60 7.65 9.60 8.625 40.0 129 BB-3450 3 3 0 0 10.8000 0.6100 0.1750 * 2.51 7.57 0.00 3.785 55.0 130 BB-3500 3 3 0 0 9.0000 0.1400 0.1880 * 26.30 7.23 3.25 5.240 50.0 131 BB-3550 3 3 0 0 5.2000 1.2000 0.1750 * 25.70 7.57 3.10 5.335 85.0 132 BB-3600 3 3 0 0 6.0000 0.5300 0.0830 * 26.00 7.68 0.00 3.840 30.0 133 BB-3650 3 3 1 0 2.8300 0.2300 0.3110 * 25.80 7.63 1.00 4.315 60.0 134 BB-3700 3 3 0 0 5.8700 0.9300 0.0950 * 24.90 8.23 3.87 6.048 60.0 135 BB-3800 3 3 0 0 9.7000 0.3000 0.7320 * 25.30 7.62 0.50 4.060 80.0 136 BB-3850 3 3 1 0 8.4000 0.2100 0.3610 * 25.00 7.65 0.00 3.825 40.0 137 BB-3900 3 3 0 0 9.8000 0.3400 0.3230 * 25.20 7.57 0.00 3.785 60.0 138 BB-3950 3 3 0 0 8.5000 0.2200 0.2400 * 25.30 7.56 0.00 * 50.0 139 BB-4000 3 3 1 0 9.2000 0.2500 0.1500 * 25.20 7.65 0.00 * 25.0 140 BB-4050 3 3 1 0 12.7000 0.2700 0.6880 * 25.20 7.50 0.00 * 70.0 141 BB-4100 3 3 0 0 7.6000 0.3500 0.1690 * 25.20 7.45 0.00 * 30.0 142 RR-2650 4 2 0 0 4.8500 0.2000 0.4900 * 26.10 7.37 * 5.100 35.0 143 RR-2700 4 2 0 0 8.2000 0.1600 0.5000 * 26.70 6.84 * 0.900 10.0 144 RR-2750 4 2 0 0 8.1000 0.1700 0.6500 * 26.60 6.90 * 0.450 5.0 145 RR-2800 4 2 0 0 7.7500 0.2000 0.3050 * 26.60 6.95 * 0.300 15.0 146 RR-2850 4 2 0 0 0.6700 0.2950 0.3800 * 26.30 7.25 * 6.420 10.0 147 RR-2900 4 2 0 1 5.5300 0.2050 0.3750 * 26.50 7.23 * 0.000 45.0 148 RR-2950 4 2 0 1 4.1000 0.1600 0.5950 * 26.70 7.21 * 1.750 85.0 raw data

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 ID Code River Code Stream order Frog Present Tributary present River Width River Depth velocity pH temperature DO Bank Width Dist. To forest edge % canopy cover 149 RR-3000 4 2 0 0 5.5000 0.1750 0.3050 * 26.30 7.30 * 0.420 90.0 150 RR-3050 4 2 0 0 8.5000 0.9000 0.2250 * 25.70 7.98 * 4.040 95.0 151 RR-3100 4 2 0 0 3.7800 0.2500 0.4000 * 25.90 8.11 * 3.000 80.0 152 RR-3150 4 2 0 0 2.7700 0.3800 0.2300 * 25.70 7.98 * 0.000 95.0 153 RR-3200 4 2 0 0 4.3000 0.2750 0.1450 * 25.80 7.82 * 1.500 35.0 154 RR-3250 4 2 1 1 5.6000 0.5000 0.1200 * 25.60 7.78 * 5.600 35.0 155 RR-3300 4 2 0 0 3.1900 0.2800 0.4350 * 25.50 8.12 * 0.000 45.0 156 RR-3350 4 2 0 0 9.2600 0.2500 0.2050 * 25.70 7.82 * 2.650 40.0 157 RR-3400 4 2 0 1 4.4000 0.3200 0.4450 * 25.60 7.65 * 9.400 25.0 158 RR-3450 4 2 0 0 7.3000 0.6750 0.7250 * 25.50 7.78 * 1.000 15.0 159 RR-3500 4 2 1 0 3.4400 0.2000 0.3150 * 25.70 7.73 * 2.500 10.0 160 AG-0 6 2 1 0 0.0660 0.5800 0.2930 8.74 25.25 8.17 7.25 6.950 50.0 161 AG-50 6 2 1 0 0.0850 0.2000 0.1750 8.57 25.30 7.91 4.77 7.080 60.0 162 AG-100 6 2 0 0 0.0700 0.3940 * 8.55 25.50 7.95 7.13 7.135 60.0 163 AG-150 6 2 0 0 0.0152 1.5000 0.5630 7.41 25.20 7.94 8.01 28.500 70.0 164 AG-200 6 2 0 1 0.0900 0.6300 0.1930 8.56 25.50 7.71 4.75 6.470 20.0 165 AG-250 6 2 0 0 0.0230 0.3800 0.3530 8.54 25.40 7.64 0.00 4.215 20.0 166 T2-50 7 1 1 0 0.8000 0.0650 0.1930 5.74 26.00 7.25 3.20 3.200 15.0 167 T2-100 7 1 0 0 0.8800 0.0600 0.1820 5.69 26.00 6.95 2.52 2.520 70.0 168 T2-150 7 1 0 0 0.7550 0.1100 0.1420 5.45 26.05 6.92 3.50 3.500 65.0 169 T2-200 7 1 0 0 0.7800 0.1200 0.1690 5.53 25.95 7.64 1.40 1.400 30.0 170 T2-250 7 1 1 0 0.6100 0.1700 0.2330 5.28 25.90 7.64 1.30 1.300 60.0 171 T2-300 7 1 1 0 0.1500 0.0530 0.0310 5.32 25.75 7.59 1.00 1.000 80.0 172 T2-350 7 1 0 0 0.5100 0.0700 0.2780 5.06 25.75 7.10 1.00 1.000 60.0 173 T2-400 7 1 0 0 1.0000 0.0400 0.3210 4.99 26.35 7.38 2.50 2.500 85.0 174 T4-50 8 1 1 0 0.1700 0.0300 * 7.25 26.40 7.60 4.00 4.000 40.0 175 T5-50 9 1 0 0 1.4200 0.2700 * 6.79 26.90 7.34 6.80 0.000 30.0 176 T5-100 9 1 1 0 0.7500 0.1500 0.0530 6.61 27.05 5.24 4.20 0.000 50.0 177 T5-150 9 1 0 0 1.1000 * 0.2370 6.50 26.30 7.00 1.20 0.000 60.0 178 T5-200 9 1 0 0 3.0000 0.0800 0.4360 6.47 25.75 7.50 0.00 0.000 50.0 179 T5-250 9 1 0 0 0.1200 0.3700 0.2160 6.37 25.60 7.71 0.00 0.000 80.0 180 T18-0 10 1 1 0 0.3000 0.0000 * * * * 0.80 0.800 50.0 181 T18-50 10 1 1 0 0.3000 0.0000 * * * * 0.50 0.500 80.0 182 T20-0 11 1 1 0 1.5000 0.0000 * * * * 0.00 3.600 40.0 183 T20-50 11 1 1 0 0.3000 0.0000 * * * * 0.50 1.500 60.0 184 T20-100 11 1 1 1 0.3000 0.0000 * * * * 0.00 0.000 80.0 185 ————— 20/06/2007 09:37:58 PM ————————————————————

Binary Logistic Regression: Frog Present versus Stream order; Tributary pr; ...

Link Function: Logit

Response Information

Variable Value Count Frog Pre 1 44 (Event) 0 118 Total 162

162 cases were used 22 cases contained missing values

Logistic Regression Table Odds 95% CI Predictor Coef SE Coef Z P Ratio Lower Upper Constant -12.806 8.751 -1.46 0.143 Stream o -0.3832 0.3908 -0.98 0.327 0.68 0.32 1.47 Tributar 1 0.4062 0.6604 0.62 0.539 1.50 0.41 5.48 River Wi -0.04555 0.06449 -0.71 0.480 0.96 0.84 1.08 River De 1.3324 0.9155 1.46 0.146 3.79 0.63 22.80 velocity -1.809 1.222 -1.48 0.139 0.16 0.01 1.80 temperat 0.7042 0.2976 2.37 0.018 2.02 1.13 3.62 DO -0.8265 0.4078 -2.03 0.043 0.44 0.20 0.97 Dist. To 0.03784 0.05032 0.75 0.452 1.04 0.94 1.15 % canopy -0.004133 0.006556 -0.63 0.528 1.00 0.98 1.01

Log-Likelihood = -82.391 Test that all slopes are zero: G = 24.710; DF = 9; P-Value = 0.003

Goodness-of-Fit Tests

Method Chi-Square DF P Pearson 183.545 152 0.041 Deviance 164.781 152 0.226 Hosmer-Lemeshow 9.214 8 0.325

Table of Observed and Expected Frequencies: (See Hosmer-Lemeshow Test for the Pearson Chi-Square Statistic)

Group Value 1 2 3 4 5 6 7 8 9 10 Total 1 Obs 2 2 2 2 2 3 3 10 7 11 44 Exp 0.9 1.7 2.3 2.9 3.7 3.9 4.7 5.7 7.4 10.9 0 Obs 14 14 14 14 15 13 13 6 9 6 118 Exp 15.1 14.3 13.7 13.1 13.3 12.1 11.3 10.3 8.6 6.1

Total 16 16 16 16 17 16 16 16 16 17 162

Measures of Association: (Between the Response Variable and Predicted Probabilities)

Pairs Number Percent Summary Measures Concordant 3869 74.5% Somers' D 0.50 Discordant 1293 24.9% Goodman-Kruskal Gamma 0.50 Ties 30 0.6% Kendall's Tau-a 0.20 Total 5192 100.0%

Saving file as: C:\Program Files\MTBWIN\Data\frog3.MPJ * NOTE * Existing file replaced. Saving file as: C:\Program Files\MTBWIN\Data\frog3.MPJ * NOTE * Existing file replaced. Binary Logistic Regression: Frog Present versus Stream order; Tributary pr; ...

Link Function: Logit

Response Information

Variable Value Count Frog Pre 1 44 (Event) 0 118 Total 162

162 cases were used 22 cases contained missing values

Logistic Regression Table Odds 95% CI Predictor Coef SE Coef Z P Ratio Lower Upper Constant -0.3823 0.9372 -0.41 0.683 Stream o 0.1067 0.3019 0.35 0.724 1.11 0.62 2.01 Tributar 1 0.6154 0.6008 1.02 0.306 1.85 0.57 6.01 River Wi -0.07500 0.06180 -1.21 0.225 0.93 0.82 1.05 River De -0.1493 0.8037 -0.19 0.853 0.86 0.18 4.16 velocity -2.387 1.147 -2.08 0.038 0.09 0.01 0.87 Dist. To 0.01795 0.05057 0.35 0.723 1.02 0.92 1.12 % canopy -0.000698 0.006215 -0.11 0.911 1.00 0.99 1.01

Log-Likelihood = -90.691 Test that all slopes are zero: G = 8.109; DF = 7; P-Value = 0.323

Goodness-of-Fit Tests

Method Chi-Square DF P Pearson 162.243 154 0.309 Deviance 181.382 154 0.065 Hosmer-Lemeshow 9.851 8 0.276

Table of Observed and Expected Frequencies: (See Hosmer-Lemeshow Test for the Pearson Chi-Square Statistic)

Group Value 1 2 3 4 5 6 7 8 9 10 Total 1 Obs 4 1 2 2 6 5 4 5 9 6 44 Exp 1.9 2.7 3.2 3.6 4.3 4.4 4.9 5.3 6.0 7.8 0 Obs 12 15 14 14 11 11 12 11 7 11 118 Exp 14.1 13.3 12.8 12.4 12.7 11.6 11.1 10.7 10.0 9.2

Total 16 16 16 16 17 16 16 16 16 17 162

Measures of Association: (Between the Response Variable and Predicted Probabilities)

Pairs Number Percent Summary Measures Concordant 3330 64.1% Somers' D 0.29 Discordant 1813 34.9% Goodman-Kruskal Gamma 0.29 Ties 49 0.9% Kendall's Tau-a 0.12 Total 5192 100.0%

Binary Logistic Regression: Frog Present versus Stream order; Tributary pr; ... Link Function: Logit

Response Information

Variable Value Count Frog Pre 1 50 (Event) 0 122 Total 172

172 cases were used 12 cases contained missing values

Logistic Regression Table Odds 95% CI Predictor Coef SE Coef Z P Ratio Lower Upper Constant -0.3798 0.7587 -0.50 0.617 Stream o -0.1831 0.2527 -0.72 0.469 0.83 0.51 1.37 Tributar 1 0.6349 0.5510 1.15 0.249 1.89 0.64 5.56 River Wi -0.07841 0.06053 -1.30 0.195 0.92 0.82 1.04 River De -0.1350 0.7710 -0.18 0.861 0.87 0.19 3.96 Dist. To 0.00163 0.04763 0.03 0.973 1.00 0.91 1.10 % canopy 0.001674 0.005829 0.29 0.774 1.00 0.99 1.01

Log-Likelihood = -101.296 Test that all slopes are zero: G = 4.763; DF = 6; P-Value = 0.575

Goodness-of-Fit Tests

Method Chi-Square DF P Pearson 173.817 165 0.304 Deviance 202.591 165 0.025 Hosmer-Lemeshow 18.851 8 0.016

Table of Observed and Expected Frequencies: (See Hosmer-Lemeshow Test for the Pearson Chi-Square Statistic)

Group Value 1 2 3 4 5 6 7 8 9 10 Total 1 Obs 4 1 7 8 5 1 1 7 8 8 50 Exp 2.8 3.4 4.3 4.7 5.1 4.9 5.1 5.6 6.2 7.8 0 Obs 13 16 10 9 13 16 16 10 9 10 122 Exp 14.2 13.6 12.7 12.3 12.9 12.1 11.9 11.4 10.8 10.2

Total 17 17 17 17 18 17 17 17 17 18 172

Measures of Association: (Between the Response Variable and Predicted Probabilities)

Pairs Number Percent Summary Measures Concordant 3505 57.5% Somers' D 0.16 Discordant 2512 41.2% Goodman-Kruskal Gamma 0.17 Ties 83 1.4% Kendall's Tau-a 0.07 Total 6100 100.0%

Saving file as: C:\Program Files\MTBWIN\Data\frog3.MPJ * NOTE * Existing file replaced. Saving file as: C:\Program Files\MTBWIN\Data\frog3.MPJ * NOTE * Existing file replaced. Binary Logistic Regression: Frog Present versus Stream order; Tributary pr; ...

* NOTE * Algorithm has not converged after 20 iterations. * Convergence has not been reached for the * parameter estimates criterion. * The results may not be reliable. * Try increasing the maximum number of iterations.

Link Function: Logit

Response Information

Variable Value Count Frog Pre 1 44 (Event) 0 118 Total 162

162 cases were used 22 cases contained missing values

Logistic Regression Table Odds 95% CI Predictor Coef SE Coef Z P Ratio Lower Upper Constant -12.068 9.995 -1.21 0.227 Stream o -1.1216 0.6722 -1.67 0.095 0.33 0.09 1.22 Tributar 1 0.0384 0.7662 0.05 0.960 1.04 0.23 4.67 River Wi 0.00948 0.08486 0.11 0.911 1.01 0.85 1.19 River De 0.655 1.078 0.61 0.543 1.93 0.23 15.92 Dist. To 0.01735 0.05600 0.31 0.757 1.02 0.91 1.14 % canopy -0.007509 0.007182 -1.05 0.296 0.99 0.98 1.01 River Co 2 -0.6889 0.6899 -1.00 0.318 0.50 0.13 1.94 3 -0.896 1.166 -0.77 0.442 0.41 0.04 4.01 4 -2.699 1.377 -1.96 0.050 0.07 0.00 1.00 6 -0.155 1.443 -0.11 0.914 0.86 0.05 14.48 7 -22 9944 -0.00 0.998 0.00 0.00 * 9 -1.841 1.660 -1.11 0.267 0.16 0.01 4.10 velocity -1.666 1.318 -1.26 0.206 0.19 0.01 2.50 temperat 0.6130 0.3496 1.75 0.080 1.85 0.93 3.66 DO -0.1576 0.5090 -0.31 0.757 0.85 0.31 2.32

Log-Likelihood = -76.427 Test that all slopes are zero: G = 36.637; DF = 15; P-Value = 0.001

Goodness-of-Fit Tests

Method Chi-Square DF P Pearson 154.898 146 0.291 Deviance 152.854 146 0.332 Hosmer-Lemeshow 13.906 8 0.084

Table of Observed and Expected Frequencies: (See Hosmer-Lemeshow Test for the Pearson Chi-Square Statistic)

Group Value 1 2 3 4 5 6 7 8 9 10 Total 1 Obs 0 3 1 1 2 2 8 8 9 10 44 Exp 0.2 1.0 1.9 2.6 3.5 3.8 4.7 6.0 8.2 12.0 0 Obs 16 13 15 15 15 14 8 8 7 7 118 Exp 15.8 15.0 14.1 13.4 13.5 12.2 11.3 10.0 7.8 5.0

Total 16 16 16 16 17 16 16 16 16 17 162

Measures of Association: (Between the Response Variable and Predicted Probabilities)

Pairs Number Percent Summary Measures Concordant 4132 79.6% Somers' D 0.59 Discordant 1047 20.2% Goodman-Kruskal Gamma 0.60 Ties 13 0.3% Kendall's Tau-a 0.24 Total 5192 100.0%

Saving file as: C:\Program Files\MTBWIN\Data\frog3.MPJ * NOTE * Existing file replaced.

————— 21/06/2007 08:24:59 AM ————————————————————

Welcome to Minitab, press F1 for help. Retrieving project from file: C:\Program Files\MTBWIN\Data\frog3.MPJ ————— 26/06/2007 07:37:08 PM ————————————————————

Welcome to Minitab, press F1 for help. Retrieving project from file: C:\Program Files\MTBWIN\Data\frog3.MPJ Results for: frog.MTW

Regression Analysis: Frog Present versus temperature

The regression equation is Frog Present = - 0.374 + 0.0266 temperature

179 cases used 5 cases contain missing values

Predictor Coef SE Coef T P Constant -0.3741 0.4563 -0.82 0.413 temperat 0.02657 0.01704 1.56 0.121

S = 0.4715 R-Sq = 1.4% R-Sq(adj) = 0.8%

Analysis of Variance

Source DF SS MS F P Regression 1 0.5403 0.5403 2.43 0.121 Residual Error 177 39.3479 0.2223 Total 178 39.8883

Unusual Observations Obs temperat Frog Pre Fit SE Fit Residual St Resid 129 2.5 0.0000 -0.3074 0.4137 0.3074 1.36 X

X denotes an observation whose X value gives it large influence.

Regression Analysis: Frog Present versus DO

The regression equation is Frog Present = 1.46 - 0.158 DO

179 cases used 5 cases contain missing values

Predictor Coef SE Coef T P Constant 1.4559 0.4419 3.29 0.001 DO -0.15752 0.06192 -2.54 0.012

S = 0.4663 R-Sq = 3.5% R-Sq(adj) = 3.0%

Analysis of Variance Source DF SS MS F P Regression 1 1.4069 1.4069 6.47 0.012 Residual Error 177 38.4814 0.2174 Total 178 39.8883

Unusual Observations Obs DO Frog Pre Fit SE Fit Residual St Resid 19 5.48 0.0000 0.5927 0.1071 -0.5927 -1.31 X 21 5.47 0.0000 0.5943 0.1076 -0.5943 -1.31 X 38 5.73 1.0000 0.5533 0.0926 0.4467 0.98 X 176 5.24 1.0000 0.6305 0.1212 0.3695 0.82 X

X denotes an observation whose X value gives it large influence.

Regression Analysis: Frog Present versus velocity

The regression equation is Frog Present = 0.409 - 0.222 velocity

174 cases used 10 cases contain missing values

Predictor Coef SE Coef T P Constant 0.40895 0.07385 5.54 0.000 velocity -0.2215 0.1892 -1.17 0.243

S = 0.4723 R-Sq = 0.8% R-Sq(adj) = 0.2%

Analysis of Variance

Source DF SS MS F P Regression 1 0.3057 0.3057 1.37 0.243 Residual Error 172 38.3610 0.2230 Total 173 38.6667

Unusual Observations Obs velocity Frog Pre Fit SE Fit Residual St Resid 4 0.765 1.0000 0.2395 0.0878 0.7605 1.64 X 51 0.861 1.0000 0.2182 0.1047 0.7818 1.70 X 68 0.793 0.0000 0.2333 0.0927 -0.2333 -0.50 X 82 0.775 0.0000 0.2373 0.0895 -0.2373 -0.51 X

X denotes an observation whose X value gives it large influence.

Saving file as: C:\Program Files\MTBWIN\Data\frog4.MPJ ————— 10/09/2007 07:29:58 PM ————————————————————

Welcome to Minitab, press F1 for help. Retrieving project from file: C:\Program Files\MTBWIN\Data\frog4.MPJ

APPENDIX D

PCR Results Chytrid Fungus Test Results PCR assay for B. dendrobatidis BP Conservation Project - Tobago November, 2006 samples

Test samples: One hundred twenty five skin swab samples were received 11/20/06 from Jahson Berhane. Note one sample vial (T20-01, Pisces # 67775) did not contain a swab but only a small amount of liquid.

Sample Preparation: For each sample 800uL to 1 ml of 70% ethanol was added to the sample vial, mixed by pipetting the liquid up and down, then the entire volume, including any visible skin pieces, was transferred to a microfuge tube. After spinning at maximum speed in a microcentrifuge (~16,000 x G) for 3 minutes, the supernatant was drawn off and discarded, tissue lysis buffer added, and any pellet resuspended by vortexing. 10 ug of carrier DNA was added to the lysis buffer for the skin swab samples (no carrier was added to the egg mass samples). Total DNA was extracted from all samples using a spin-column DNA purification procedure.

PCR assay: All sample DNA preparations were assayed for the presence of the Batrachochytrium dendrobatidis ribosomal RNA Intervening Transcribed Sequence (ITS) region by 45 cycle single-round PCR amplification using an assay developed by Seanna Annis and modified for greater specificity and sensitivity at Pisces.

Each PCR run included the following controls: Positive DNA: DNA prepared from a laboratory culture of B. dendrobatidis, Strain JEL 270, kindly provided by Joyce Longcore. This sample was previously demonstrated to be positive by PCR. The signal from this sample is the standard for a strong positive (++) signal.

Negative DNA: DNA prepared from a laboratory culture of a non-batracho chytrid fungus, Strain JEL 151, kindly provided by Joyce Longcore This sample was previously demonstrated to be negative (-) by PCR.

No DNA: H2O in place of template DNA. This reaction remains uncapped during addition of sample DNA to the test reactions, and serves as a control to detect contaminating DNA in the PCR reagents or carryover of positive DNA during reaction set-up.

Results: Twenty one samples were positive for the B. dendrobatidis ribosomal RNA ITS region; the remaining 104 swab samples were negative. Individual sample results are shown on the following four pages.

Scoring: +++ = very strong positive signal ++ = strong positive signal + = positive signal w+ = weak positive signal - = no signal/below limit of detection

Pisces Molecular LLC 9/17/2007 Individual sample PCR test results Source Source ID # Source Information Pisces # Sample Form/Condition Tested For Results Comments BPC/JB Doctors R #1 L.V. ? 06/8/06 67770 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Hard to read label BPC/JB Doctors R #2 M.O. 6/8/06 67771 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB Doctors R. - 2 M.O. 9-8-06 67663 swab/EtOH B. dendrobatidis - 1ml 70%EtOH added before prep BPC/JB Doctors R. - 3 M.O. 9-8-06 67664 swab/EtOH B. dendrobatidis ++ 1ml 70%EtOH added before prep BPC/JB Doctors R. - 4 M.O. 9-8-06 67665 swab/EtOH B. dendrobatidis - 1ml 70%EtOH added before prep BPC/JB Doctors R. - 5 M.O. 9-8-06 67666 swab/EtOH B. dendrobatidis w+ 1ml 70%EtOH added before prep BPC/JB Doctors R. - 6 M.O. 9-8-06 67667 swab/EtOH B. dendrobatidis - 1ml 70%EtOH added before prep BPC/JB Doctors R. - 7 M.O. 9-8-06 67668 swab/EtOH B. dendrobatidis ++ 1ml 70%EtOH added before prep BPC/JB Doctors R. - 8 M.O. 9-8-06 67669 swab/EtOH B. dendrobatidis +++ 1ml 70%EtOH added before prep BPC/JB Doctors R. - 9 M.O. 9-8-06 67670 swab/EtOH B. dendrobatidis - 1ml 70%EtOH added before prep BPC/JB Doctors R. - 10 M.O. 9-8-06 67671 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 11 M.O. 9-8-06 67672 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep

BPC/JB Doctors R. - 12 M.O. 10-8-07 67673 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 13 M.O. 10-8-07 67674 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 14 M.O. 10-8-07 67675 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 15 M.O. 10-8-07 67676 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 16 M.O. 10-8-07 67677 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 17 M.O. 10-8-07 67678 swab/EtOH B. dendrobatidis + 850uL 70%EtOH added before prep BPC/JB Doctors R. - 18 M.O. 10-8-07 67679 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 19 M.O. 10-8-07 67680 swab/EtOH B. dendrobatidis +++ 850uL 70%EtOH added before prep BPC/JB Doctors R. - 20 M.O. 10-8-07 67681 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 21 M.O. 10-8-07 67682 swab/EtOH B. dendrobatidis +++ 850uL 70%EtOH added before prep BPC/JB Doctors R. - 22 M.O. 10-8-07 67683 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 23 M.O. 10-8-07 67684 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep

BPC/JB Doctors R. - 24 11.8.06 67685 swab/EtOH B. dendrobatidis + 850uL 70%EtOH added before prep BPC/JB Doctors R. - 25 11.8.06 67686 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 26 11.8.06 67687 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 27 11.8.06 67688 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 28 11.8.06 67689 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 29 11.8.06 67690 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 30 11.8.06 67691 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 31A 11.8.06 67692 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 31B 11.8.06 67693 swab/EtOH B. dendrobatidis ++ 850uL 70%EtOH added before prep BPC/JB Doctors R. - 32 11.8.06 67694 swab/EtOH B. dendrobatidis - 850uL 70%EtOH added before prep BPC/JB Doctors R. - 33 11.8.06 67695 swab/EtOH B. dendrobatidis +++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 34 11.8.06 67696 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 35 11.8.06 67697 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 36 11.8.06 67698 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 37 11.8.06 67699 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 38 11.8.06 67700 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep Pisces Molecular LLC 1 of 4 18/06/2007 Individual sample PCR test results Source Source ID # Source Information Pisces # Sample Form/Condition Tested For Results Comments BPC/JB Doctors R. - 39 11.8.06 67701 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 40 11.8.06 67702 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 41 12.8.06 67703 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 42 12.8.06 67704 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 43 12.8.06 67705 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 44 12.8.06 67706 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 45 12.8.06 67707 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 46 12.8.06 67708 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 47 12.8.06 67709 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 48 12.8.06 67710 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 49 12.8.06 67711 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 50 12.8.06 67712 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 51 12.8.06 67713 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 52 12.8.06 67714 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 53 12.8.06 67715 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 54 12.8.06 67716 swab/EtOH B. dendrobatidis + 800uL 70%EtOH added before prep BPC/JB Doctors R. - 55 12.8.06 67717 swab/EtOH B. dendrobatidis + 800uL 70%EtOH added before prep BPC/JB Doctors R. - 56 12.8.06 67718 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 57 12.8.06 67719 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 58 12.8.06 67720 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 59 12.8.06 67721 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 60 12.8.06 67722 swab/EtOH B. dendrobatidis +++ 800uL 70%EtOH added before prep BPC/JB Doctors R. - 61 12.8.06 67723 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R. - 62 12.8.06 67724 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB Doctors R M. olmonae 08-07-06 67777 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB Doctors R M. olmonae 07-07-06 67778 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R M. olmonae 10-07-06 67779 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R M. olmonae 08-07-06 67782 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R M. olmonae 08-07-06 67783 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB Doctors R 10.7.06 67785 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB B. merinus Man O War 30-8-06 67725 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67726 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67727 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67728 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67729 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67730 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67731 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67732 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67733 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67734 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67735 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

Pisces Molecular LLC 2 of 4 18/06/2007 Individual sample PCR test results Source Source ID # Source Information Pisces # Sample Form/Condition Tested For Results Comments BPC/JB B. merinus Man O War 30-8-06 67736 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67737 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67738 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67739 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67740 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67741 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67742 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67743 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67744 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67745 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67746 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67747 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67748 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67749 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus Man O War 30-8-06 67750 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67762 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67763 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67764 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67765 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67766 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB B. merinus S.C. Road 30-08-06 67767 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB M. olmonae Bloody Bay 070806 67753 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Sprayed when cap opened

BPC/JB M. olmonae Argyle 26-08-06 67756 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep BPC/JB M. olmonae Argyle 26-08-06 67757 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB M. olmonae Argyle 26-08-06 67758 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB M. olmonae Argyle 26-08-06 67759 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB M. olmonae Argyle 26-08-06 67760 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB M.O. #1 Louis Dior 8/8/06 67768 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB M.O. #2 Louis Dior 8/8/06 67769 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB NOR 4.1 T18-01? 67754 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Hard to read label; Sprayed when cap opened

BPC/JB T18 River 18-08-06 67761 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Hard to read label BPC/JB T18 River - 01 18-08-06 67772 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB T18 River - 03 18-08-06 67773 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB T18 River - 04 18-08-06 67774 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB T20-01 19-8-06 67775 no swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; No swab in sample (?), but a little liquid BPC/JB T20-02 19-8-06 67776 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

Pisces Molecular LLC 3 of 4 18/06/2007 Individual sample PCR test results Source Source ID # Source Information Pisces # Sample Form/Condition Tested For Results Comments BPC/JB T20-03 19-08-06 67755 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB DR02 7.7.06 67781 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep BPC/JB DR06 8.11.06 67787 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Hard to read label

BPC/JB Bloody Bay M. olmonae 07-08-06 67784 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB Louis Dior 10.7.06 67786 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB 1 + 800 01 f/s 67751 swab/EtOH B. dendrobatidis ++ 800uL 70%EtOH added before prep

BPC/JB King's Bay toe clip bag 67752 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep

BPC/JB RP 3:45? 09.02.06 67780 swab/EtOH B. dendrobatidis - 800uL 70%EtOH added before prep; Hard to read label

Pisces Molecular LLC 4 of 4 18/06/2007

APPENDIX E

Financial Report Grant Proposal budget Actual Expenditure Difference TTD USD TTD USD TTD USD Phase 1: Project Preparation Administrative TT$8,316.00 $1,320.00 TT$4,100.00 $650.79 TT$4,216.00 $669.21 Scientific Equipment TT$5,499.90 $873.00 Other TT$3,458.70 $549.00 TT$11,723.55 $1,860.88 TT$2,764.95 $438.88

Phase II: Project Implementation Stipend TT$11,340.00 $1,800.00 TT$0.00 $0.00 TT$11,340.00 $1,800.00 Local Guides TT$4,725.00 $750.00 TT$0.00 $0.00 TT$4,725.00 $750.00 Transportation TT$15,120.00 $2,400.00 TT$20,685.75 $3,283.45 -TT$5,565.75 -$883.45 Logistics/ Living Expenses TT$15,246.00 $2,420.00 TT$15,201.40 $2,412.92 TT$44.60 $7.08 Education/Awareness TT$3,811.50 $605.00 TT$10,770.99 $1,709.68 -TT$6,959.49 -$1,104.68 Laboratory Testing TT$3,015.90 $478.71 -TT$3,015.90 -$478.71

Phase III: Post Project Expenses Administration TT$315.00 $50.00 TT$0.00 $0.00 TT$315.00 $50.00 Report preparation TT$1,260.00 $200.00 TT$0.00 $0.00 TT$1,260.00 $200.00 Contingency TT$6,507.90 $1,033.00 TT$0.00 $0.00 TT$6,507.90 $1,033.00

Total CLP Grant TT$78,750.00 $12,500.00

Additional Funding CI Seed grant (used for lab testing TT$14,882.99 $2,362.38 TT$14,882.99 $2,362.38 TT$0.00 $0.00

Exchange Rate 6.3 Total TT$93,632.99 $14,862.38 TT$80,380.58 $12,758.82 TT$13,252.41 $2,103.56