1 ,

+?+a- United States

5**.4
1:
/1
IL
w
5
.- 5
Department of
Agriculture

Forest Service
Skmpling Methods .fo.r
Pacific Northwest
Research Station Terrestrial A.mphibians
General Technical
Report
PNW-GTR-256
July 1990 /
and R.eptiles
Paul Stephen Corn and R. Bruce Bury
Wildlife-Habitat Relationships:
Sampling Procedures for
Pacific Northwest Vertebrates
Andrew B. Carey and Leonard F. Ruggiero, Technical Editors

Sampling Methods for Terrestrial

Amphibians and Reptiles
Paul Stephen Corn
Zoologist
U.S. Department of the Interior
Fish and Wildlife Service
National Ecology Research Center
4512 McMurray Avenue
Fort Collins, Colorado 80525-3400

R. Bruce Bury
Research Zoologist
U.S. Department of the Interior
Fish and Wildlife Service
National Ecology Research Center
4512 McMurray Avenue
Fort Collins, Colorado 80525-3400

USDA Forest Service

Pacific Northwest Research Station
Portland, Oregon
General ‘Technical Report PNW-GTR-256
1990
Preface Concern about the value of old-growth Douglas-fir forests to wildlife in the Pacific
Northwest began escalating in the late 1970s. The available information on wildlife-
habitat relationships suggested that as many as 75 species including amphibians,
birds, and mammals, could be dependent on old-growth forests. The USDA Forest
Service chartered the Old-Growth Forest Wildlife Habitat Program to investigate the
role old growth plays in maintaining viable populations of wildlife. It was apparent that
broad surveys of vertebrate communities would be necessary to determine which
species were truly closely associated with old-growth forests. Insufficient guidance on
techniques, procedures, and sample sizes was available in the existing literature. We
assembled a team of researchers from universities and Federal agencies to conduct
pilot studies to develop sampling protocols and to test the basic experimental design
for contrasting the wildlife values of young, mature, and old-growth forests. The
sampling protocols resulting from the pilot studies were implemented in 1984-86
across broad areas of the Cascade Range in southwestern Washington and in
Oregon, the Oregon Coast Ranges, and the Klamath Mountains of southwestern
Oregon and northern California. Naturally, improvements were made to the protocols
as time passed. A tremendous amount of experience in sampling was gained.

Our goal in this series is to compile the extensive experience of our collaborators into
a collection of methodology papers providing biologists with pilot study-type informa-
tion for planning research or monitoring populations. The series will include papers
on sampling bats, aquatic amphibians, terrestrial amphibians, forest-floor mammals,
small forest birds, and arboreal rodents, as well as papers on using telemetry for
spotted owl studies and a guide to bird calls.

Andrew B. Carey
Leonard F. Ruggiero
Abstract Corn, Paul Stephen; Bury, R. Bruce. 1990. Sampling methods for terrestrial
amphibians and reptiles, Gen. Tech. Rep.. PNW-GTR-256. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Research Station.
34 p.

Methods described for sampling amphibians and reptiles in Douglas-fir forests in

the Pacific Northwest include pitfall trapping, time-constrained collecting, and surveys
of coarse woody debris. The herpetofauna of this region differ in breeding and non-
breeding habitats and vagility, so that no single technique is sufficient for a com-
munity study. A combination of pitfall trapping and hand collecting is the most
effective approach.

Keywords: Amphibians, reptiles, sampling techniques, pitfall trapping, time-

constrained collecting, downed wood.
Contents 1 Introduction
3 Objectives
3 Overview
5 Time-Constrained Searches
6 Surveys of Coarse Woody Debris
6 Pitfall Trapping
7 Experimental Design
7 Time-Constrained Searches
7 Surveys of Coarse Woody Debris
8 Pitfall Trapping
11 Field Methods
11 Crew Sizes
11 Time Frame and Weather
12 Operating Guidelines
17 Identification
17 Disposition of Specimens
18 Data Analysis
21 Conclusions
21 Acknowledgments
22 Equivalents
22 Literature Cited
27 Appendix 1
28 Appendix 2
28 Data Sheets
32 Appendix 3
32 Materials Needed for TCS or Surveys of CWD
32 Materials Needed for Pitfall Installation and Operation
34 Materials Needed in the Lab
Introduction There is a rich herpetofauna in the Pacific Northwest, with 43 species of amphibians
and reptiles present west of the Cascade Range (appendix 1). Depending on the geo-
graphical area, 19 to 32 species may be present at a given site (fig. 1). The number
of species of amphibians is consistent at 13 to 15 species in most areas in this re-
gion, but reptiles range from 5 species in southwestern Washington to 17 species in
both southwestern Oregon and northwestern California. The difference is due to in-
creased aridity and higher temperatures in the southern locales, which favor reptiles.
Although a diverse reptilian fauna may occur in an area, many species (particularly
snakes) are locally rare or restricted to certain habitats; for example, oak-woodland
(many snakes) or permanent water (turtles).

FS LSn7

i
\ \
Figure l-Number of amphibians and reptiles potentially present
in different regions of the Pacific Northwest west of the crest
of the Cascade Range. The histograms are by major taxonomic
groups: F = frogs, S = salamanders, L = lizards, Sn = snakes,
and T = turtles.
During recent research in western Oregon and Washington, we found few or no
reptiles present in closed-canopy Douglas-fir (Pseudotsuga menziesii (Mirb.) France)
forests (Bury and Corn 1987, 1988; Corn and Bury, in press). Reptiles are usually
encountered in rocky, open areas (for example, cliff faces) or in grasslands and oak
woodlands (Herrington 1988, Nussbaum and others 1983); these habitats were rare
or absent in the forest stands we studied. Thus, reptiles were a small fraction of the
sampled herpetofauna, and they will receive little mention here. Biologists will need
to employ special techniques if reptiles are encountered at a study site (see Bury and
Raphael 1983, Jones 1986, Scott 1982b).

In the Pacific Northwest, amphibians are often found in terrestrial habitats, particular-
ly in forests, and among terrestrial vertebrates may be the most numerous group. Ter-
restrial salamanders, for example, can exceed over five individuals/m* in local aggre-
gations (Bury and Raphael 1983, Jaeger 1979). In 1983, Bury estimated that there
were over 400 salamanders/ha in old-growth redwood forests in northern California
(Bury 1983). In 1984, Raphael reported densities of 10 to 180 salamanders/ha in
Douglas-fir forests in northern California (Raphael 1984). We estimated that mean
density of plethodontid salamanders associated with downed wood ranged from
364/ha in young Douglas-fir forests to 744/ha in old-growth forests (Corn and Bury, in
press). For eastern deciduous forests in New Hampshire, Burton and Likens (1975)
estimated about 3,000 salamanders/ha, and Hairston (1987) estimated that energy
present in salamanders in southern Appalachian forests exceeds that of all other
vertebrate predators combined.

Amphibians are important components of the northwestern fauna in ways other than
numbers or biomass. Of 22 amphibian species inhabiting forest habitats in the Pacific
Northwest, 14 species (64 percent) are endemic (species whose distributions are re-
stricted to the Pacific Northwest). Many of these habitats are affected increasingly by
human activities.

Several species of plethodontid salamanders are more abundant in older forests, or

show relations to habitat features that are prominent in old-growth forests. Ensatinas1
are more abundant in older Douglas-fir forests than in younger stands in northern
California (Raphael 1984). Ensatinas, Oregon slender salamanders, and clouded sala-
manders are often associated with large pieces of downed wood (Aubry and others
1988; Bury and Corn 1988; Corn and Bury, in press). Coarse woody debris (CWD) is
a major component of old-growth forests and is severely reduced by modern forestry
practices (Harmon and others 1986, Maser and Trappe 1984). The plethodontid sala-
manders in general are useful for assessing logging impacts because they have com-
pletely terrestrial life cycles (the eggs are deposited on land and hatch into miniature
individuals), and most species have stable populations (Hairston 1987).

1
Scientific names of reptiles and amphibians are given in
table 6 (appendix 1).
 The relations of frogs and aquatic-breeding salamanders to older forests are more
difficult to explain than are the relations of plethodontids. Most of these species use
terrestrial habitats to a degree, especially for feeding. They also may migrate over-
land to breeding ponds or streams and, thus, temporarily occur in many habitats
during their travels. Tailed frogs previously had been considered to be closely tied to
streams (Metter 1967) but we discovered that they are found in forests long dis-
tances from flowing water (Bury 1988). Our results also suggest that juvenile tailed
frogs disperse into terrestrial habitats away from streams.

Given the diversity of amphibian life histories, habitat preferences, and different
means of locomotion, more than one sampling technique is needed to sample ade-
quately all species of amphibians. We used several methods to sample amphibians;
methods for sampling aquatic species are discussed separately (Bury and Corn, in
press). We sampled the terrestrial herpetofauna in three main ways: (1) time-con-
strained searches (TCS), (2) searching specified numbers of pieces of downed wood
(CWD surveys), and (3) pitfall trapping.

We will discuss the objectives, sampling design, and techniques specific to each
method separately. We will then discuss techniques common to all the methods we
used and make recommendations for effectively sampling the herpetofauna in the
Pacific Northwest.’ The methods described here were used by the Old-Growth Forest
Wildlife Habitat Program (Ruggiero and Carey 1984) in field work from 1983 to 1985.
This program included studies of vertebrates in Douglas-fir forests in California,
Oregon, and Washington west of the Cascade Range (Ruggiero and others, in press).
With the exception of experiments to determine the most effective design for pitfall
trapping (Bury and Corn 1987), these methods were not rigorously tested against
alternatives (field methods, particularly hand-collecting techniques, have rarely been
subjected to experimental comparisons). Rather, they reflect our current professional
judgment and draw heavily from other recent descriptions of field methods (Campbell
and Christman 1982, Jones 1986, Raphael and Barrett 1981, Vogt and Hine 1982).

Objectives The primary objective of our study was to identify species associated with old-growth
Douglas-fir forests (Ruggiero and Carey 1984), and so the techniques we used were
Overview slanted to favor survey methods. Pitfall trapping and CWD surveys will provide some
information on populations. These data can be used to analyze habitat use by individ-
ual species and the patterns shown by groups of species in different habitats. Coarse
woody debris surveys and TCS can also provide detailed information on the use of.
microhabitats by various species. Basic ecological data are needed that can be ap-
plied to recommendations for management of specific habitats.

There are marked differences in catch between hand collecting (TCS and CWD ‘-
surveys) and pitfall trapping (table 1). Species such as clouded salamanders and
Oregon slender salamanders are closely associated with CWD and were frequently
caught by hand but were trapped infrequently. Tailed frogs, newts and other migratory
species were trapped effectively in pitfalls but rarely were caught by hand.

The choice of a specific method to achieve stated objectives depends on the species
under study as well as the scope of the objectives. If a small-scale study on one or a
few species is intended, then only one method may be needed. A survey of commu-
nity structure over a large geographic area will likely require all three methods.

3
Table 1-Comparison of captures of amphibians and reptiles by pitfall trapping
and time-constrained searches (TCS), H.J. Andrews Experimental Forest, 1983

Number of captures

Pitfalls b

Species TCSa Summerc Fallc

Total 281 206 822

Percent of captures (rank)

Pitfalls

TCS Summer Fall

Salamanders:
Northwestern salamander 0 (-) 0 (-) 5 (5)
Pacific giant salamander 0 (-) 1 (13) 2 (6)
Clouded salamander 28 (2) 3 (10) 1 (8)
Oregon slender salamander 22 (3) 4 (7) 1 (10)
Ensatina 43 (1) 24 (1) 25 (2)
Dunn’s salamander 1 (5) 1 (14) 1 (9)
Rough-skinned newt 1 (6) 15 (2) 37 (1)
Frogs:
Tailed frog 0 (-) 9 (5) 19 (3)
Pacific treefrog 2 (4) 4 (8) 1 (7)
Red-legged frog 0 (-) 1 (12) 6 (4)
Lizards:
Western skink 1 (8) 11 (4) 1 (12)
Northern alligator lizard 1 (9) 13 (3) 1 (11)
Western fence lizard 1 (7) 3 (11) 1 (15)
Snakes:
Rubber boa 1 (15) 0 (-)
Northwestern garter snake 7 (6) 1 (14)
Common garter snake 4 (9) 1 (13)
a
TCS were done for 8 staff hours in 18 study areas in April.
b
Arrays of pitfall traps with drift fences (Bury and Corn 1987) were operated in the same areas for 180
days from late May to November.
c
The results of pitfall trapping are divided into the first 90 days of trapping (summer) and the second 90
days (fall).

Throughout this paper, we will use the terms study site and stand interchangably.
This is due to the bias of working in forests, where study sites tend to encompass
areas of more or less uniform habitat, which are referred to as stands. Stands in the
old-growth studies were patches of forest of uniform age with a minimum area of
10 ha (Carey and Spies, in press).

-
Time-Constrained Time-constrained searches involve searching study areas for amphibians and rep-
Searches tiles, which are immediately collected by hand (Bury and Raphael 1983, Campbell
and Christman 1982). Equal effort is expended in each area searched, as measured
by the number of staff hours spent searching. Thus, each search will have a specific
time limit, dependent on the prescribed effort and the crew size. Time-constrained
searches are most useful for determining presence or absence of species and for
providing initial data on the types of microhabitats occupied by individual species.

Time-constrained searches are not suitable for providing population data beyond pres-
ence or absence. Because this is a “plotless” technique, the same amount of potential
habitat tends to be searched in each study area: however, amounts of suitable hab-
itat differ among study areas. Results from some TCS may show habitat-poor areas
yielding similar numbers of animals as habitat-rich areas, even though the population
sizes may be quite different. Indeed, evidence is that salamanders are more clumped
in areas with less habitat, which will increase the bias in favor of these areas. In the
Coast Ranges of Oregon, we found the density in downed wood (number per m3) of
ensatinas was significantly higher in young and mature stands compared to old
growth (fig. 2) (Corn and Bury, in press). In this case, TCS could possibly result in an
inverse relation of numbers caught to’ actual population size.

If population estimates are an objective, then other techniques need to be applied.

We used CWD surveys effectively (explained below), but another common method is
complete removal of all residents of a predetermined area (Bury 1983; Campbell and
Christman 1982; Jaeger 1979; Raphael 1984; Scott 1976, 1982a). Plot searches are
labor intensive: Bury (1983) required 20 to 44 staff hours to search 0.125-ha plots in
old-growth redwood forests in northern California. For surveys of several study areas,
plot searches may require too much effort to produce sample sizes large enough for
statistical analysis.

6 -
l

5-

4 -

Figure 2-Salamander density in

downed wood. The density of
salamanders (number/m3) in
downed logs is inversely related
to the amount of downed wood
present in the study areas.
Salamanders appear to be less
clumped as more habitat is avail-
VOLUME OF DOWNED WOOD (CUBlC METERS/HECTARE) able.

5
 For initial surveys of presence or absence, TCS are more effective than plot searches
because collectors are free to examine large objects over a wide area, and usually
more amphibians are found in large objects than in the leaf litter, at least in the
Pacific Northwest. This method is efficient because the objects searched are most
likely to yield animals. In northwestern forests, TCS may produce as much as a 10
times greater yield than will area-constrained collecting (Bury and Raphael 1983). In
recent studies, capture rates of TCS have ranged from one to two animals per staff
hour in the Cascade Range in Oregon and Washington (Aubry and others 1988, Bury
and Corn 1988), to over eight animals per staff hour in northern California (Welsh
and Lind 1988). Time-constrained searches are best employed when several study
areas need to be surveyed in a short time.

Surveys of Coarse In 1985, we were confronted with the choice of initiating TCS in the Oregon Coast
Woody Debris Ranges or developing a technique to quantify habitat use and estimate density of
selected species of salamanders. We chose the latter and developed a technique
involving searches of predetermined numbers of pieces of downed wood. Numbers of
animals caught were then related to the amounts of CWD in the stand, and minimum-
density estimates were calculated.

Surveys of CWD are operationally similar to TCS; but to estimate animal densities,
the density of CWD must be known. Knowing the amount of CWD present also allows
for quantifying microhabitat use and drawing meaningful comparisons of microhabitat
use among species.

The primary drawback of surveys of CWD is that density estimates apply to only
one feature of the habitat. Surveys of CWD underestimate density of species using
downed wood only occasionally; for example, most species of woodland salamanders
(Plethodon spp.) frequent rocky soils, but an unknown fraction of a population may
occur in CWD. For species strongly associated with CWD (for example, the clouded
salamander or the Oregon slender salamander), surveys of CWD should provide
general estimates of population sizes.

Pitfall Trapping Pitfall trapping is a flexible technique that can be used to achieve several objectives;
for example, drift fences with pitfall traps have been used to encircle specialized hab-
itats such as amphibian breeding ponds (Gibbons and Semlitsch 1981, Shoop 1968,
Storm and Pimentel 1954). This technique can be used for complete enumeration of
breeding populations. Pitfall trapping also has been employed widely for surveys of
amphibian and reptile diversity and abundance in different habitat types (Bury and
Corn 1987; Campbell and Christman 1982; Friend 1984; Jones 1981, 1986; Raphael
1984; Vogt and Hine 1982; also see selected papers in Ruggiero and others, in press;
and Szaro and others 1988). The main drawback of pitfall trapping is that trapability
differs widely among species (Bury and Corn 1987, Campbell and Christman 1982,
Gibbons and Semlitsch 1981). A survey of ail species of herpetofauna in an area
therefore requires more than one technique.
 Pitfall trapping provides data on the presence or absence of species, and because
the trapping effort can be quantified and standardized across study areas, relative
abundances can be calculated. Estimates of actual population size may be possible,
though probably only for abundant species. Pitfalls may be used as live traps if
checked frequently, and mark and recapture techniques also may be used. If pitfalls
are used as a removal method to estimate density, then the area being trapped must
be known. This is extremely difficult to determine for most herpetofauna and is
something we have not done in any of our studies.

Pitfall trapping is also useful for investigating seasonal activity patterns. Traps can be
operated continuously, so that variation in activity due to weather can be detected
(Bury and Corn 1987). Pitfall traps are permanent structures, so long-term monitoring
can be accomplished by operating the same trap array or grid periodically over
several years. Trapping has unknown effects, however, on population structure due
to the removal of resident individuals.

Experimental This technique is a quick survey method requiring few restrictions on the approach.
Design Three points need to be considered: (1) collecting should be done away from forest
edges; (2) aquatic habitats, such as breeding ponds or creeks should be avoided-
Time-Constrained these are covered by a separate protocol (Bury and Corn, in press); and (3) collecting
Searches should cover as much of the stand as possible. There are two ways to accomplish
this last point. One is to devote enough time to the search to be able to collect across
the entire study area. The second is to restrict the search to a fairly small area (for
example, a circle with a radius of 25 m) and restrict the amount of time spent collect-
ing. The number of smaller areas that can be searched in each study area depends
on the amount of time devoted to the TCS. We found that 6 or 8 staff hours of collect-
ing were sufficient; few additional species were detected by collecting for longer than
that. If 1 hour is spent in each of the subsamples, then six to eight areas can be
searched in each study area.

Surveys of Coarse This technique is somewhat more complicated than TCS in that it involves systemat-
Woody Debris icaily searching a predetermined number of logs in each study area. Several ques-
tions must be addressed when a study is designed, including how many logs to
sample, how to apportion the sample among the different decay states of downed
wood, and how to select the logs sampled.

in 1985 in the Oregon Coast Ranges, we conducted CWD surveys in 18 study areas.
Each survey included 30 logs greater than 10 cm in diameter. We selected 10 logs
in decay classes 1 and 2, 10 in decay class 3, and 10 in decay classes 4 and 5 (see
Franklin and others [1981] or Maser and Trappe [1985] for methods of classifying
CWD). The three decay categories that we used reflected natural divisions of the
five-class scale. Class 1 and 2 logs are intact with more or less complete bark cover.
Class 3 logs have decaying sapwood, and the bark is beginning to slough off. Class
4 and 5 logs are thoroughly decayed, have little bark, and are disintegrating. We
recommend sampling equal numbers of logs in each of these decay categories. We
searched each log for a maximum of 20 staff minutes.

7
 We found salamanders in only 37 percent of the logs (198 of 536) that we examined,
so a sample size of 30 logs per stand should probably be considered the minimum. if
few logs are sampled and salamanders occupy a small percentage of these, then the
estimates of salamander density will be based on minimal information.

Logs to be sampled are best selected by a systematic sampling scheme (Mendenhaii

and others 1971). If the study area has not been mapped and the locations of ail logs
determined, it will not be possible to draw a random sample. A systematic sample
involves selecting logs in a specified order as they are encountered while the crew
moves through the stand.

Pitfall Trapping Planning pitfall trapping mainly involves selecting the appropriate trap design. We
used two different pitfall designs in our old-growth studies (fig. 3). in 1983, we used
arrays of pitfall traps with aluminum drift fences (Bury and Corn 1987). in 1984 and
1985, we used grids of single pitfall traps without fences. There were quantitative and
qualitative differences in the yield of each technique that must be considered in plan-
ning a project.

PITFALL ARRAY

DETAIL OF ONE ARM:

Fr

Sm I 3m
t -I

6 X 6 GRID OF PITFALL TRAPS

COLUMN
ABCOEF
lx x xx x x

2x x x x x x

ROW 3 x x x x x x

4x x xxx x

5x x xx x x

6x x XX x x I 15m

llSm

Place traps within 2 m of flagged station:

use natural fenws (logs) where available

Figure 3-Designs for arrangements of pitfall traps either with or without

drift fences. PT = pitfall trap, FT = funnel trap.
Trap rates for salamanders are similar for both arrays and grids, but arrays caught
considerably more frogs and reptiles than grids did (table 2). The differences were
due, in part, to the absence of drift fences in pitfall grids and the season when trap-
ping was done. Pitfall arrays were open continuously for 180 days in 1983, from May
to November. Grids were open for 30-50 days, beginning in October 1984. The grids
were operated too late in the year to capture reptiles and large numbers of postmeta-
morphic juvenile frogs, which disperse from breeding sites in late summer or early
fall. These frogs composed the majority of frogs caught by the arrays.

Although arrays catch more animals than grids do, arrays are not necessarily better
for determining presence or absence of amphibians. Grids caught few reptiles but
were able to detect amphibians, including frogs, as well as or better than arrays
(table 3).

Table 2-Capture totals and trap rates (captures/100 trap nights) for major
groups of amphibians and reptiles for 30 pitfall arrays in the Cascade Range of
Oregon and Washington, 1983 (180 days), and for 48 pitfall grids in the Oregon
Coast Ranges, 1984 and 1985 (80 days)

Pitfall arrays Pitfall grids

,.
Group Species Captures Trap rate Species Captures Trap rate

Salamanders 10 1145 1.77 8 1762 1.27

Frogs 3 915 1.41 2 103 .07
Lizards 3 79 .12 <.Ol
Snakes 3 41 .06 0

Table 3-Species of amphibians and reptiles inhabiting Douglas-fir forests at

Old-Growth Wildlife Habitat Program study areas in Washington and Oregon

Pitfall Pitfall
array? grid?

Central Southern Oregon

Oregon Washington Coast
Species Cascade Range Cascade Range Ranges

Amphibians:
Northwestern salamander C C C
Long-toed salamander P?b P? -*.
Cope’s giant salamander P
Pacific giant salamander c
Olympic salamander :
Clouded salamander - E
Oregon slender salamander
Ensatina c c
Dunn’s salamander C
Larch Mountain salamander c -
Van Dyke’s salamander - - P?
Western red-backed salamander P? C
Rough-skinned newt C C :
 Table 3-continued

Pitfall Pitfall
arraysa gridsa

Central Southern Oregon

Oregon Washington Coast
Species Cascade Range Cascade Range Ranges

Tailed frog C C C
Boreal toad P
Pacific tree frog rz cp
Red-legged frog C C :
Yellow-legged frog P? - P?
Cascades frog P? P? -
Spotted frog P P -
Detection EfficiencyC 59-77 62-77 77-83

Reptiles:
Western skink C
Northern alligator lizard C c
Southern alligator lizard P?
Western fence lizard
Rubber boa
c
C
P
P
F
P
Sharp-tailed snake - - P
Ring-necked snake P - P
Gopher snake - - P
Sierra water snake - - P?
Terrestrial garter snake P? P? P
Northwestern garter snake C P
Common garter snake C : P
Western rattlesnake - - P
Detection Efficiency 75-86 43-60 8
b P= potential occurrence, C = captured in pitfall traps, - = species does not occur in the area.
Potential occurrerm of a species in our study areas was uncertain.
’ Species captured + species potentially present x 100.

The choice of whether to install arrays or grids ultimately depends on the needs of
the study. Arrays are superior for catching reptiles, but reptiles may not be abundant
in forest habitats or of interest to the goals of a study. Arrays can provide large
sample sizes in relatively shot-l periods. Grids remove fewer animals than arrays and
may be more suitable for long-term monitoring. Both techniques are effective for
catching small mammals as well as amphibians.

Arrays may be placed in pairs, as we operated them in 1983 (fig. 3), or single arrays
may be placed at more than one location within a stand. Three or four single arrays
scattered throughout the stand may better assess the variation within study areas, but
this approach requires significantly more time for checking the traps in each area.

10

L
 Cost may be part of the decision on whether to install arrays or grids. Grids are not
substantially cheaper in cost of materials, because more pitfall traps can supplant the
cost of fencing. Grids took only about one-half the effort to install as the arrays did.
When personnel costs are high, this can result in a large difference in cost between
the two methods. The cost involved in checking the traps is similar and depends
mainly on the number of stands and the travel time between them.

Field Methods This section provides instructions for carrying out TCS, surveys of CWD, and pitfall
trapping. We will not discuss selection of study areas. If the study is an integrated
wildlife survey, then study areas for mammals or birds can be used just as well for
studying the herpetofauna. All the techniques discussed here require small areas as
compared to bird or mammal studies.

Crew Sizes Optimal crew sizes depend on the technique being used. Time-constrained searches
and surveys of CWD use the same collecting techniques, and three to four persons
are suitable for both. In both crews, one person is the data recorder, and the remain-
ing people do the collecting. A 6-staff-hour TCS, done with a two-person crew plus a
recorder who does not collect, requires 3 hours, plus the time for breaks.

For pitfall trapping, a large crew is generally necessary to install traps, but only one
or two people are needed to check the traps once they are open. Installation of either
arrays or grids is relatively fast with a crew of six. Crews of this size can install two
arrays or grids per day. Two people can check a grid of 36 traps in an hour or less.
Several sites can be checked in one day, depending on the travel time between
study areas.

Time Frame and Hand collecting (TCS and surveys of CWD) should be done when amphibians are
Weather most likely to be active; that is, in the rain. In the Pacific Northwest, this is either in
spring or fall (it rains in winter also, but low temperatures inhibit surface activity by
amphibians). If there are several study areas, then the primary consideration is that
the weather be as consistent as possible throughout the collecting period. Activity of
amphibians is highly dependent on weather, and comparisons between areas of col-
lection under radically different weather conditions may not be valid. Collecting there-
fore should begin as early as possible in spring or as late as possible in fall, but still
avoiding lengthy periods of cold and snow. Collecting should not be done in heavy
snow; light snowfall in a period of wet weather probably will not seriously affect am-
phibian activity. Two TCS can usually be done in one day, but one survey of CWD
requires most of a day. It is possible, but not recommended, to split a survey between
two days.

Pitfall trapping has more flexibility, because all traps are open at the same time;
thereby reducing variability among study areas due to weather. The best season for
operating pitfalls depends on the animals being trapped. For amphibians, spring and
fall are again the periods of highest activity and will result in the largest catch. If
reptiles are being sampled, then early summer is the best time to open pitfall traps.
Pitfall installation can be done at any time, but data (Bury and Corn 1987) suggest
that pitfalls should be in the ground at least 1 month before trapping begins.

11
Operating Guidelines Time-constrained searches-Determine the number of 1-staff-hour searches that
can be done in the amount of time allotted to each study area. On a topographic map
or aerial photo of the study area, distribute the 1-hour searches for maximum cover-
age of the study area. The crew should enter each TCS with a map of the study area
that shows the approximate location of each l-hour search and the path to follow be-
tween searches, with compass headings and approximate distances. Each 1-hour
search should be confined to an area with a radius of about 25 m, and the center of
each 1-hour search should be at least 75 m from any forest edge.

Each TCS is a survey of as much habitat as possible within each study area. Move
from one object to the next after a few minutes. It is possible to spend over an hour
at one large log, but a maximum of 10 minutes per object should suffice. Assuming a
crew of two collectors and one recorder, each staff hour of search takes 30 minutes
of actual time. When an animal is found, time is spent by the collector in assisting
the recorder. The recorder should keep track of this time, and the total amount of
data recording time is added to the end of the search, so that 1 full hour of collecting
is achieved. This becomes more important in searches yielding many animals,
because data recording will require more time.

Surveys of CWD-The techniques involved here are more precise than those used
in TCS. Logs are chosen by a systematic sampling scheme. Specifically, a choice is
made to sample one log out of a certain number of logs encountered. In most hab-
itats, choosing one out of every three logs will produce a survey covering a large
proportion of the study area. Further, logs are divided into subsamples based on the
decay state of the log. We compressed the standard five-point decay classification
into three categories: category A-decay classes 1 and 2, category B-decay class
3, and category C-decay classes 4 and 5. Sample 10 logs in each category (one of
every three logs encountered in each category) for a total of 30 in each study area.

Plot a path through the study area that will cover a large portion of the area but will
not intersect itself. For each decay category, choose a random number from one to
three. Begin following the designated path. At every downed log, determine the decay
category and whether the log should be sampled. The recorder keeps a running tally
of the number of logs encountered in each category. Each category of log accumu-
lates at its own pace, and whether a log is sampled depends on the number of logs
encountered in that decay category. The decision may be, for example, to sample
every second category-A log, every third category-B log, and every first category-C
log. For this example, table 4 shows which logs will be selected from the first 20 logs
encountered.

When a log is selected, measure the total dimensions (see appendix 2 for data forms
and a description of the data to be recorded). Determine the tree species, if possible,
and the slope and aspect of the site where the log occurs. Search the log for a maxi-
mum of 20 staff minutes. Carefully remove any bark and tear into the decayed wood
layer by layer. If the entire log cannot be sampled within the time limit, search a por-
tion of the log as completely as possible. This is very important, because salamander
densities are based on the volume of wood actually searched.

12
Table 4-A hypothetical example of log selectlon In surveys of CWD
Number encountered
in decay category

Decay
Log number category A B C Sample log?=

1 Ab 1 No
2 A 2 Yes
3 C Yes
4 B 1 No
5 B 2 No
6 A 3 No
7 A 4 No
8 3 Yes
9 : No
10 A Yes
11 C No
12 A No
13 No
14 :: No
15 B No
16 Yes
17 iii No
18 C Yes
19 A 8 Yes
20 B 8 No
and so forthC

’ Assume that 1 out of every 3 logs is to be sampled, and the following sampling scheme is to be followed:
category A, log number 2 of 3, category B, log number 3 of 3, and category C, log number 1 of 3.
b Decay categories: A = decay classes 1 and 2; B = decay class 3; and C = decay classes 4 and 5.
’ Continue selecting logs until 10 logs in each decay category have been sampled.

Collecting tips-We have several pointers for more effective collecting for both TCS
and surveys of CWD. Tools needed for both techniques include potato rakes and
crowbars. It is necessary to purchase high-quality potato rakes; the less expensive
ones cannot withstand extensive use. Crowbars are handy for peeling bark and
breaking up the less-decayed logs. (See appendix 3 for a complete list of materials
and tools needed to take samples.)

Large logs and bark piles adjacent to these or large, well-decayed snags are the
most productive sites for TCS. Follow the instructions above for sampling logs. Cther
habitats should not be ignored during TCS, however. Moderate-sized debris (10 cm
or more in diameter) on the forest floor should be turned over; often two people are
needed to roll logs. In general, avoid raking through leaf litter or turning very small
objects, but search piles of bark, slash, or mounds, because these often house am-
phibians. Rocks or boulders, if present, should be turned. Exercise caution when turn-
ing rocks on steep slopes. Be alert; searches often occur on rainy days when visi-
bility is poor, especially under closed canopies. Salamanders can flee rapidly down

13
 a crevice, so grab them by cupping your hand on top of them. Frogs are elusive, and
to catch them you may need the cooperation of two or three people to surround the
quarry. Collectors should scrutinize the area under turned objects. Salamanders often
freeze and most are cryptically colored.

Some species have special traits. Ensatinas are commonly found, and they rarely
move once exposed. They are easily captured but must be picked up carefully or
else they will autotomize (spontaneously amputate) their tail. Newts are slow moving
but possess a highly toxic skin poison. This poison typically is released only if the
animal is under attack but may show up during rough handling (for example, if the
newt is hit by a rake tine). All terrestrial salamanders have some toxic secretions, but
they rarely exude these substances when being handled.

The Oregon slender salamander and the Larch Mountain salamander often coil up,
an apparent mimicry of distasteful millipedes that also curt up. Check any coiled
animal closely. Clouded salamanders and western redback salamanders can move
rapidly and need to be grabbed quickly. At least one hand should be bare to capture
animals; gloves are usually too awkward for collecting agile species.

Snakes might be encountered during searches. Rattlesnakes occur at low elevations

in Oregon and California, especially around rock outcrops. We recommend no collect-
ing of rattlesnakes. Other snakes or lizards can be grabbed or, if fleeing, stepped on
gently. Reptiles should be sluggish in cool, wet weather.

Habitat destruction can be minimized by returning cover items to their original posi-
tions. Roll small logs and rocks back and replace large pieces of bark slabs. Rake
decayed logs back together and replace as much bark as possible. Some habitat de-
struction is unavoidable, but the organic material remains, and at least a portion of
the log-soil interface can be restored by careful replacement of disturbed objects.

Installation of pitfalls-Place pitfall arrays and grids in spots representative of the

study area. If single arrays are to be placed around the study area, the locations
should be preselected from maps or aerial photos. The array or grid location should
be at least 75 m from any forest edges (the farther, the better). For arrays, establish
the center point of the first array at random. If a pair of arrays is used, measure 25 m
from this point in a random direction for the center of the second array. For a grid,
select one corner at random for the location of the grid. The grid is then laid out by
using handheld compasses and 15-m tapes or measured ropes (necessary in dense
brush). Installation of grids is generally fast with a six-person crew; four people lay
out the grid, and two people begin installing traps. Two-person teams are best for
grid layout. One person pulls the tape or rope until stopped by the second person,
who remains at the previous station. Flag the new point and continue.

14
Pitfalls are constructed by fastening the open ends of two number 10 tin cans togeth-
er with duct tape and then cutting the bottom out of one end (fig. 4). Traps are in-
stalled flush with the ground and have a plastic collar inserted at the top. This collar
functions to keep animals from crawling out of the trap and is constructed by cutting
the bottom out of a l-lb plastic margarine tub. When not being used for trapping, the
traps should be closed; use the plastic lids from the margarine tubs. In grids, place
the trap within 2 m of the station flag. If possible, place the trap next to a cover ob-
ject, such as a rock or downed log. Traps next to logs should be placed on the down-
hill side of the log. The hole for the trap is dug most easily with a posthole digger,
which creates a hole with the correct diameter. A tile spade can also be used. Traps
have an additional optional wood cover. When the trap is open, the cover is suspend-
ed above the opening. This functions in part as a rain cover and partly to attract
animals.

If an array design is being used, drift fences are constructed from 50-cm-tall alumi-
num valley roofing metal. This comes in 15.2-m rolls, which should be cut into 5-m
sections before it is taken to the study area. We placed fences pointing away from
the center of the array at equal (120o) intervals. The interior end of each fence began
3 m from the center of the array (fig. 3). There are many other possible arrangements
for placing pitfall arrays; see figures in Campbell and Christman (1982), Jones (1981,
1986), and Vogt and Hine (1982).

Figure 4-Construction and placement of a pitfall trap.

 Use a mattock or hoe to dig a trench 20 cm deep and 5 m long, stand the fence into
the trench, and back fill with soil. Occasionally an axe is needed to cut large roots.
Tamp down the loose dirt so that the fence is self-supporting (stakes are not neces-
sary for these relatively short fences), and smooth the dirt alongside the fence to
create a runway. Move small obstacles (twigs, rocks) away from the fence. Traps are
placed at the ends of the fence so that no gaps occur between the fence and the rim
of the trap.

There are two important safety rules to follow when installing arrays. First, always
wear gloves to handle the aluminum. The sharp edges can inflict serious cuts on un-
protected hands. Second, exercise extreme caution in wet weather. The tools quickly
become coated with slick mud, and a mattock or axe flying out of someone’s hands
is a lethal weapon.

Funnel traps will need to be constructed if reptiles are a target of the study (see
Jones 1986, Vogt and Hine 1982). Funnel traps are constructed from window screen,
which comes in rolls 76 cm wide. Cut a piece 90 cm long, and staple the ends togeth-
er along the cut edge. Fold back the stapled edge so that you have a tube 25 cm in
diameter by 76 cm long. Construct funnels by rolling square pieces of screen into a
cone and stapling. Fold back the edge and attach to the tube. One end is fastened
permanently with staples, and paper clips are used at the other end so that animals
may be easily removed. Funnel traps are placed midway on both sides of each drift
fence. Shape the trap and fill in with dirt so that no gap occurs between the fence
and the trap. Shade the trap by placing loose bark or litter over the trap.

Pitfall operation-operating pitfall traps is a simple task. Techniques do not differ

between arrays and grids. The primary decision is how frequently the traps should be
checked. Check the traps every other day, if possible, but if there are a large number
of study areas, then traps may have to be checked lessfrequently. Intervals of more
than 5 days between checks should be avoided. Checking traps more frequently pro-
duces better specimens, particularly among the mammals that will be caught. If the
number of study sites is such that all traps cannot be opened on the same day, care
must be taken that all traps are closed in the same order they were opened in. This
ensures the same trapping effort for each area.

Each time a trap is checked, remove debris that has fallen into the trap, and bail out
excess water. A small amount of water should be placed in traps when they are
opened, but in wet weather, most traps will accumulate more water than is desired. It
has previously been recommended that water be placed in pitfall traps (Raphael and
Barrett 1981, Williams and Braun 1983), and this is probably the quickest, most hu-
mane way to kill small mammals. Current guidelines for using pitfall traps to kill trap
small mammals (American Society of Mammalogists 1987) specify drowning as the
only acceptable method. But drowning is a slow and inhumane way to kill amphibians,
and it has been prohibited in the current guidelines for field methods for herps (ASIH
and others 1987). A generally acceptable compromise between these apparently in-
compatible recommendations is to keep a small amount of water (2 to 5 cm) in traps
and check them frequently. Small mammals, particularly shrews, will become hypo-
thermic and drown in this amount of water, but most amphibians should be able to
survive.

16
 All animals trapped in pitfalls are to be returned to the laboratory for processing. Sep-
arate mammals, live herps, and dead herps, but otherwise place all animals from the
same trap in one plastic bag. Carry a field notebook with waterproof paper to record
the number of individuals, species, and trap number of all animals caught. This record
is important and should become a permanent part of the data set. It provides critical
information during the initial processing of specimens and is a valuable reference for
the questions that inevitably arise even after the data have been processed. Record
the study area, date, and trap number in pencil on a small piece of waterproof paper
and place in each bag of specimens. Bag all the specimens from a single study area
together in a large plastic bag. Keep the specimens in a cooler with reusable ice con-
tainers while in the field. On returning to the lab, place dead specimens in a freezer
and live herps in a cool space or refrigerator.

Identification Accurate identification of specimens in the field is critical for TCS and surveys of
CWD. Field identification is less important for pitfalls, because all specimens are
examined later in the laboratory. The field notes listing the specimens caught in each
trap are more valuable, however, if they are accurate. To increase accuracy, it is
helpful for team members to examine series of specimens at a museum before field
work begins. An additional field practice session is recommended to catch animals
alive and to practice field identification. Most forms have distinct shapes or colors,
but some species present problems. Most people have difficulty with woodland
salamanders (Plethodon spp.), ranid frogs (Rana spp.), and juvenile salamanders.
References for identification of northwestern herps are Nussbaum and others (1983)
and Stebbins (1985). Other useful regional references are by Green and Campbell
(1984) and Gregory and Campbell (1984).

Disposition of ’ All animals captured in pitfall traps are routinely euthanized and preserved (special
Specimens consideration will need to be given to species with special status, such as those list-
ed by the Federal or State governments as threatened or endangered). Specimens
from TCS or CWD surveys may be treated in the same manner, or they may be
released after the surveys near points where they were captured. If specimens are
released, then positive identification is absolutely necessary (see above). Also, if ani-
mals are released, a representative series of voucher specimens should be retained
from each study area and preserved. Capturing animals and retaining specimens
requires valid scientific collecting permits from the appropriate State wildlife agency,
and arrangements should be made before the study begins to deposit the specimens
in an appropriate museum.

Process all specimens from a given survey, or all specimens collected from a pitfall
site on a given day, together. This will provide for the most accurate recordkeeping,
and it helps in solving the mystery of the occasional unlabeled specimen. Thaw any
frozen specimens, and kill the live ones. Be sure to keep the label identifying the
specimen closely associated with each specimen. Kill by relaxing amphibians in a
dilute solution of Chloretone and by injecting reptiles with aqueous sodium pento-
barbital. Chloretone is a saturated solution of hydrous chlorobutanol in 95 percent

17
 ethanol. An effective dilution is 2 ml per 570 ml of water. Sodium pentobarbital
(Nembuto12 is one trade name) is a restricted drug and may be difficult to obtain.
Reptiles may also be killed by injecting 95 percent ethanol into the heart region.

After the animal is dead, weigh and measure it (see appendix 2), tie a numbered
tag to the right hind leg, and preserve in formalin. Create a lo-percent solution of
buffered formalin by diluting commercial formalin to 10 percent and adding 4 g of
baking soda or sodium carbonate per 400 ml of solution. Amphibians that appeared
dead may begin to move when placed in the formalin. These should immediately be
rinsed in water and returned to the Chloretone until dead. Amphibians and lizards
should be laid out ventral side down in a shallow pan with a tight-fitting lid; for
example, a plastic freezer container. Line the bottom of the pan with commercial
paper towels (household towels have “dimples” that become imprinted on the skins
of the animals), and pour a small amount of formalin into the pan. Snakes should be
folded into an oblong coil with the head on the inside. The coil should be short
enough to fit in the storage jars. Reptiles also must have formalin injected into the
body cavity, limbs, and tail. Do not inject so much that a balloonlike specimen is
created. If injection is not possible, then the body cavity, limbs, and tail must be slit
to allow the formalin to enter the body. Body cavities of large Pacific giant salaman-
ders should also be slit for thorough preservation. Pisani (1973) provides a thorough
discussion of preservation techniques. Let the specimens fix in the formalin for at
least 24 hours, then store in 50 percent isopropyl alcohol.

If specimens are released, then reasonably accurate measures of snout-vent and

total lengths can still be made. Place the animal in a plastic bag and restrain it
against the bottom of the bag. When the animal is quiet and relatively straight,
measure to the nearest millimeter with a ruler. Mass can also be measured in the
field with spring scales available in forestry supply catalogs.

The investigator should be aware that in northwestern forests, twice as many small
mammals as herps generally are captured in pitfall traps. If a study is planned that
uses pitfall traps, provision should be made for preserving the mammals. Neglecting
this would be a criminal waste of valuable data.

Data Analysis Numerous analyses can be done on the types of data collected from surveys of
amphibian occurrence and abundance (see papers in Szaro and others [1988] and
Ruggiero and others [in press] for examples). We will give a couple examples of the
types of analyses that can be done, and we will discuss any special analyses that
need to be performed.

All the techniques are excellent at providing data on presence or absence of species,
and two or more techniques can be combined to provide a complete assessment of
all the species potentially present. One example is provided by considering amphib-
ians and reptiles detected by pitfall trapping with arrays and TCS at 18 study areas

2
The use of trade, firm, or corporation names in this publica-
tion is for the information and convenience of the reader.
Such use does not constitute an official endorsement or ap-
proval by the U.S. Department of Agriculture of any product
or service to the exclusion of others that may be suitable.

18
 in the Oregon Cascade Range in 1983 (table 5). Presence-absence data can be ana-
lyzed by calculating measures of similarity and then using a clustering procedure to
look for patterns among groups of study areas (Pielou 1984). From the data matrix in
table 5, similarities were calculated for every pair of stands by using Jaccard’s index
(Pielou 1984), which is the percentage of species both areas have in common com-
pared to the total number of species present at either area. Clustering was accom-
plished by using the nearest-neighbor technique. One group of five old-growth and
mature stands cluster together above the 60-percent level of similarity, but in general,
there are few recognizable patterns related to habitat type (fig. 5). Pielou (1984) and
Gauch (1982) are valuable sources of techniques for analyzing the structure and
organization of communities.
Table 5-List of species of amphibians and reptiles present (P) at 18 study areas in and near
the H.J. Andrews Experimental Forest, Lane County, Oregon, 1983

Stand number

Old-growth stands Mature stands Young stands Clearcut stands

Species 2 3 15 17 24 25 29 33 11 35 39 42 47 40 75 55 92 93

Northwestern salamander P P P P P P P

Pacific giant salamander P P ,;lj c’ P P P P

‘.. I.
Clouded salamander P P P P P P P P P P P P P P

Oregon slender salamander P P P P P P P P P P P P P

Ensatina P P P P P P P P P P P P P P P P P P

Dunn’s salamander P P P

Rough-skinned newt P P P P P P P P P P P P P P

Tailed frog P P P P P P P P P P P

Pacific trecfrog P P P P P P P P

Red-legged frog P P -P P

Reptiles:
.-
Western skink P P P

Northern alligator lizard P P P P P _. P

Western fence lizard P P P

Rubber boa P

Northwestern garter snake P P P P P

Common garter snake P P P P P P P P

Number of species 8 5 6 7 6 7 11 3 5 6 4 6 5 8 9 9 13 6

19
 slMllARlTv
100 80 50 40 20 0
STANDTYPE # I I I I I I I I I I I

OLDQROWlli 24
c4DQrn 3

MAlulE 11
uDQRowM 17
OIDQ~ 2
ct.- 55
MATURE 35
YWNQ 75
OLDQROWTH 29
92

Ot.DQm 15
YOUNQ 40
OLDQf3OWlH 33

Figure 5-Nearest-neighbor clustering of herpetofauna at 18 study areas in the Oregon Cascade

Range in 1983. Similarities were calculated by using presence-absence combined data from
pitfall trapping and TCS.
.ei?)
Surveys of CWD can provide initial estimates of population density. The density in
downed wood of each species of salamander (number per cubic meter) is calculated
as the number caught in each log, divided by the volume of wood sampled in each
log. Mean densities in downed wood in each stand were calculated for each of the
three decay categories (decay classes 1 and 2, class 3, and classes 4 and 5). Use a
nested analysis of variance (stands within forest age classes) to test whether density
(log transformed) in downed wood of any species varies among decay categories or
age class (old growth, mature, and young growth).

We calculated predicted densities of plethodontid salamanders in 45 forest stands

from the following formula:

3
D = C (di*Vi) ,
i=l

where D = number of salamanders per ha, di = density in downed wood in decay

category i, and Vi = m3 of downed wood per ha in category i. See Spies and others
(1988) for techniques to determine the amount of downed wood present in a stand.
Where d varied among age classes, D was calculated by using the mean density in
downed wood for each age class.

20
 -

;. STAND AGE (YEARS)

Figure B-Estimated densities of salamanders relative to forest age

in the Oregon Coast Ranges in 1985. Data are based on surveys
of CWD at 15 study areas. Density was estimated for an additional
30 stands, for which data existed on the volume of downed wood
present in the study area.

Estimated density of plethodontid salamanders was related to stand age for 45 study
areas in the Oregon Coast Ranges in 1985 (fig. 6). There were 15 study areas with
surveys of CWD. Densities in the remaining 36 areas were estimated by using the
average values of d for each habitat type and the measured value of V for each area.

Conclusions There is a vast literature on techniques for sampling and analyzing vertebrate popula-
tions, but it was not our intention to provide a complete overview. Rather, we have
described the specialized methods for sampling herpetofauna that we have used and
refined in 3 years of field work in the forests of Oregon and Washington. Comprehen-
sive references on sampling techniques include Cooperrider and others (1986) and
Schemnitz (1980).

The methods we have described are most appropriate for surveys of forest-dwelling
amphibians. Because these species use several habitats for breeding, feeding, and
cover and differ widely in vagility, no single method is adequate to sample the entire
community. Pitfall trapping needs to be combined with either time-constrained collect-
ing or surveys of coarse woody debris in any comprehensive survey. _.

Acknowledgments We thank Michael Bogan, Andrew Carey, Lawrence Jones, and Leonard Ruggiero for
reading and commenting on this paper. Development of these techniques was aided
by Lawrence Jones and Martin Raphael and a dedicated corps of field biologists. This
is contribution number 68 of the Wildlife Habitat Relationships in Western Oregon
and Washington Project.

21
 When you know: Multiply by: To find:

Centimeters (cm) 0.394 Inches

Meters (m) 3.281 Feet
Square meters (m2) 10.764 Square feet
Hectares (ha) 2.471 Acres
Cubic meters (m3) 35.315 Cubic feet
Grams (g) 0.035 Ounces
Milliliters (ml) 0.035 Fluid ounces

American Society of Ichthyologists and Herpetologists; The Herpetologists’

League; Society for the Study of Amphibians and Reptiles. 1987. Guidelines
for the use of live amphibians and reptiles in field research. [Location of publishers
unknown]: 14 p.

American Society of Mammaloglsts. 1987. Acceptible field methods in mammal-

ogy: preliminary guidelines approved by the American Society of Mammalogists.
Journal of Mammalogy. 68 (suppl.): l-18.

Aubry, Keith B.; Jones, Lawrence L.C.; Hall, Patricia A. 1988. Use of woody
debris by plethodontid salamanders in Douglas-fir in Washington. In: Szaro,
Robert C.; Severson, Keith E.; Patton, David R., tech. coords. Management of
amphibians, reptiles, and small mammals in North America: Proceedings of a
symposium; 1988 July 19-21; Flagstaff, AZ. Gen. Tech. Rep. RM-166. Fort
Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station: 32-37.

Banks, Richard C.; McDiarmid, Roy W.; Gardner, Alfred L. 1987. Checklist of
vertebrates of the United States, the U. S. Territories, and Canada. Resour. Publ.
166. Washington, DC: U.S. Department of the Interior, Fish and Wildlife Service.
79 p.

Burton, Thomas M.; Likens, Gene E. 1975. Salamander populations and biomass
in the Hubbard Brook Experimental Forest, New Hampshire. Copeia. 1975:
541-546.

Bury, R. Bruce. 1983. Differences in amphibian populations in logged and old growth
redwood forest. Northwest Science. 57: 167-l78.

Bury, R. Bruce. 1988. Habitat relationships and ecological importance of amphibians

and reptiles. In: Raedeke, Kenneth J., ed. Streamside management: riparian wild-
life and forestry interactions: Proceedings of a symposium; 1987 February 11-13;
University of Washington, Seattle. Contribution 59. Seattle: Institute of Forest
Resources, University of Washington: 61-76.

Bury, R. Bruce; Corn, Paul Stephen. 1987. Evaluation of pitfall trapping in north-
western forests: trap arrays with drift fences. Journal of Wildlife Management.
51: 112-119.

22
Bury, R. Bruce; Corn, Paul Stephen. 1988. Douglas-fir forests in the Oregon and
Washington Cascades: relation of the herpetofauna to stand age and moisture. In:
Szaro, Robert C.; Severson, Keith E.; Patton, David R., tech. coords. Management
of amphibians, reptiles, and small mammals in North America: Proceedings of a
symposium; 1988 July 19-21; Flagstaff, AZ. Gen. Tech. Rep. RM-166. Fort Collins,
CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and
Range Experiment Station: 11-22.

Bury, R. Bruce; Corn, Paul Stephen. [in press]. Sampling methods for aquatic
amphibians. Gen. Tech. Rep. Portland, OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research Station.

Bury, R. Bruce, Raphael, Martin G. 1983. Inventory methods for amphibians and
reptiles. In: Bell, John F.; Atterbury, Toby, eds. Renewable resource inventories for
monitoring changes and trends: Proceedings of an international conference; 1983
August 15-19; Corvallis, OR. SAF 83-14. Corvallis, OR: Society of American
Foresters: 416-419.

Campbell, Howard W.; Christman, Steven P. 1982. Field techniques for herpeto-
faunal community analysis. In: Scott, Norman J., Jr., ed. Herpetological com-
munities. Wildlife Res. Rep. 13. Washington, DC: U.S. Department of the Interior,
Fish and Wildlife Service: 193-200.

Carey, Andrew B.; Spies, Thomas A. [in press]. Sampling design of the old-growth
community studies. In: Ruggiero, Leonard F.; Aubry, Keith B.; Carey, Andrew B.;
Huff, Mark H., tech. coords. Wildlife and vegetation of unmanaged Douglas-fir
forests: Proceedings of a symposium; 1969 March 29-31; Portland, OR. Gen.
Tech. Rep. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Research Station. .>C‘

Cooperrider, Alien Y.; Boyd, Raymond J.; Stuart, Hanson R., eds. 1986. Inventory
and monitoring of wildlife habitat. Denver, CO: U.S. Department of the Interior,
Bureau of Land Management. 858 p.

Corn, Paul Stephen; Bury, R. Bruce. [in press]. Terrestrial amphibian communities
in the Oregon Coast Ranges. In: Ruggiero, Leonard F.; Aubry, Keith B.; Carey,
Andrew B.; Huff, Mark H., tech. coords. Wildlife and vegetation of unmanged
Douglas-fir forests: Proceedings of a symposium; 1989 March 29-31; Portland, OR.
Gen. Tech. Rep. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. , ,.
Franklin, J.F.; Cromack, K., Jr.; Denison, W. [and others]. 1981. Ecological
characteristics of old-growth Douglas-fir forests. Gen. Tech. Rep. PNW-118. ”
Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest
Forest and Range Experiment Station. 48 p.

Friend, Gordon R. 1984. Relative efficiency of two pitfall-drift fence systems for
sampling small vertebrates. Australian Zoologist. 21: 423-433.

23
 Gauch, Hugh G., Jr. 1982. Multivariate analysis in community ecology. New York:
Cambridge University Press. 298 p.

Gibbons, J. Whitfield; Semlitsch, Raymond D. 1981. Terrestrial drift fences with

pitfall traps: an effective technique for quantitative sampling of animal populations.
Brimleyana. 7: l-16.

Green, David M.; Campbell, R. Wayne. 1984. The amphibians of British Columbia.
Handb. 45. Victoria, BC: British Columbia Provincial Museum. 101 p.

Gregory, Patrick T.; Campbell, R. Wayne. 1984. The reptiles of British Columbia.
Handb. 44. Victoria, BC: British Columbia Provincial Museum. 103 p.

Hairston, Nelson G., Sr. 1987. Community ecology and salamander guilds. New
York: Cambridge University Press. 230 p.

Harmon, M.E.; Franklin, J.F.; Swanson, F.J. [and others]. 1986. Ecology of
coarse woody debris in temperate ecosystems. Advances in Ecological Research.
15: 133-302.

Herrington, Robert E. 1988. Talus use by amphibians and reptiles in the Pacific
Northwest. In: Szaro, Robert C.; Severson, Keith E.; Patton, David R., tech.
coords. Management of amphibians, reptiles, and small mammals in North
America: Proceedings of a symposium; 1988 July 19-21; Flagstaff, AZ. Gen. Tech.
Rep. RM-166. Fort Collins, CO: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Forest and Range Experiment Station: 216-221.
.:rj:;
Jaeger, Robert G. 1979. Seasonal spatial distributions of the terrestrial salamander
Plethodon cinereus. Herpetologica. 35: 90-93.

Jones, K. Bruce. 1981. Effects of grazing on lizard abundance and diversity in

western Arizona. Southwestern Naturalist. 26: 107-l15.

Jones, K. Bruce. 1986. Amphibians and reptiles. In: Cooperrider, Allen Y.; Boyd,
Raymond J.; Stuart, Hanson R., eds. Inventory and monitoring of wildlife habitat.
Denver, CO: U.S. Department of the Interior, Bureau of Land Management:
267-290.

Maser, Chris; Trappe, James M., tech. eds. 1984. The seen and unseen world of
the fallen tree. Gen. Tech. Rep. PNW-164. Portland, OR: U.S. Department of Agri-
culture, Forest Service, Pacific Northwest Forest and Range Experiment Station.
56 p.

Mendenhall, William; Ott, Lyman; Scheaffer, Richard L. 1971. Elementary survey

sampling. Belmont, CA: Duxbury Press. 247 p.

Metter, Dean E. 1967. Variation in the ribbed frog, Ascaphus truei Copeia.
1967: 634-649.

24
Nussbaum, Ronald A.; Brodie, Edmund D., Jr.; Storm Robert M. 1983. Amphib-
ians and reptiles of the Pacific Northwest. Moscow, ID: University Press of Idaho.
332 p.

Pielou, E.C. 1984. The interpretation of ecological data. A primer on classification

and ordination. New York: John Wiley and Sons: 263 p.

Pisani, G.R. 1973. A guide to preservation techniques for amphibians and reptiles.
Society for the Study of Amphibians and Reptiles Herpetological Circular. 1: l-22.
Available from: Dr. Henri Seibert, Department of Zoology, Ohio University, Athens,
OH 45701. [Current (1990) price is $1.OO per copy.]

Raphael, Martin G. 1984. Wildlife populations in relation to stand age and area in
Douglas-fir forests of northwestern California. In: Meehan, William R.; Merrell,
Theodore R., Jr.; Hanley, Thomas A., eds. Fish and wildlife relationships in
old-growth forests: Proceedings of a symposium: 1982 April 12-15; Juneau, AK.
Morehead City, NC: American Institute of Fishery Research Biologists: 259-274.

Raphael, Martin G.; Barrett, Reginald H. 1981. Methodologies for a comprehen-

sive wildlife survey and habitat analysis in old-growth Douglas-fir forests. Cal-Neva
Wildlife Transactions. 1981: 106-l21.

Ruggiero, Leonard F.; Aubry, Keith B.; Carey, Andrew B.; Huff, Mark H., tech.
coords. [in press]. Wildlife and vegetation of unmanaged Douglas-fir forests:
Proceedings of a symposium; 1989 March 29-31; Portland, OR. Gen. Tech. Rep.
Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest
Research Station.

Ruggiero, Leonard F.; Carey, Andrew B. 1984. A programmatic approach to the

study of old-growth forest-wildlife relationships. In: Proceedings of the 1983
national convention; 1983 October 16-20; Portland, OR. Washington, DC: Society
of American Foresters: 340-345.

Schemnitz, Sanford D., ed. 1980. Wildlife management techniques manual. 4th ed.,
rev. Washington, DC: The Wildlife Society. 686 p.

Scott, Norman J., Jr. 1976. The abundance and diversity of the herpetofaunas of
tropical leaf litter. Biotropica 8:41-58.

Scott, Norman J., Jr. 1982a. The herpetofauna of forest leaf litter plots from
Cameroon, Africa. In: Scott, Norman J., Jr., ed. Herpetological communities. es
Wildlife Res. Rep. 13. Washington, DC: U.S. Department of the Interior, Fish and
_.
Wildlife Service: 145-150.

Scott, Norman J., Jr., ed. 1982b. Herpetological communities. Wildlife Res. Rep.
13. Washington, DC: U.S. Department of the Interior, Fish and Wildlife Service.
239 p.

25
 Shoop, CR. 1968. Migratory orientation of Ambystoma maculatum movements near
breeding ponds and displacement of migrating individuals. Biological Bulletin.
135: 230-238.

Spies, Thomas A.; Franklin, Jerry F.; Thomas, Ted B. 1988. Coarse woody debris
in Douglas-fir forests of western Oregon and Washington. Ecology. 69: 1689-1702.

Stebbins, Robert C. 1985. A field guide to western amphibians and reptiles. 2d ed.
Boston, MA: Houghton Mifflin Company. 336 p.

Storm, Robert M.; Pimentel, R.A. 1954. A method for studying amphibian breeding
populations. Herpetologica. 10: 161-l66.

Szaro, Robert C.; Severson, Keith E.; Patton, David R., tech. coords. 1988.
Management of amphibians, reptiles, and small mammals in North America:
Proceedings of a symposium; 1988 July 19-21; Flagstaff, AZ. Gen. Tech. Rep.
RM-166. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station. 458 p.

Vogt, Richard C.; Hine, Ruth L. 1982. Evaluation of techniques for assessment of
amphibian and reptile populations in Wisconsin. In: Scott, Norman J., Jr., ed.
Herpetological communities. Wildlife Res. Rep. 13. Washington, DC: U.S. Depart-
ment of the Interior, Fish and Wildlife Service: 201-217.

Welsh, Hartwell H., Jr.; Lind, Amy J. 1988. Old growth forests and the distribution
of the terrestrial herpetofauna. In: Szaro, Robert C.; Severson, Keith E.; Patton,
David R., tech. coords. Management of amphibians, reptiles, and small mammals
in North America: Proceedings of a symposium; 1988 July 19-21; Flagstaff, AZ.
Gen. Tech. Rep. RM-166. Fort Collins, CO: U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range Experiment Station: 439-458.

Willlams, Daniel F.; Braun, Suzanne E. 1983. Comparison of pitfall and conven-
tional traps for sampling small mammals. Journal of Wildlife Management.
47: 841-845.

26
Appendix 1 Table 6-Scientific and common names of amphibians and reptiles found in the
Pacific Northwest west of the Cascade Range from northern California to
British Columbia
4-letter
Scientific namea code Common namea

Amphibia, order Urodela (salamanders):

Family Ambystomatidae-
Ambystoma gracile AMGR Northwestern salamander
A. macrodactylum AMMA Long-toed salamander
Family Dicamptodontidae-
Dicamptodon copei DICO Cope’s giant salamander
D. ensatus DIEN California giant salamander
D. tenebrosus DITE Pacific giant salamander
Rhyacotriton olmpicus RHOL Olympic salamander
Family Plethodontidae-
Aneides ferreus ANFE Clouded salamander
A. flavipunctatus ANFL Black salamander
A. lugubris ANLU Arboreal salamander
Batrachoseps attenuatus BAAT California slender salamander
B. wrighti BAWR Oregon slender salamander
Plethodon dunni PLDU Dunn’s salamander
P. elongatus PLEL Del Norte salamander
(includes P, stormi)
P, larselli PLLA Larch Mountain salamander
P. vandykei PLVA Van Dyke’s Salamander
P. vehiculum PLVE Western red-backed
salamander
Family Salamandridae-
Taricha granulosa TAGR Rough-skinned newt
T. rivularis TARI Red-bellied newt
T. torosa TATO California newt
Amphibia, order Anura (frogs and toads):
Family Leiopelmatidae, Ascaphus truei ASTR Tailed frog
Family Bufonidae, Bufo boreas BUBO Western toad
Family Hylidae, Hyla regilla HYRE Pacific treefrog
Family Ranidae-
Rana aurora RAAU Red-legged frog
R. boylii RABO Foothill yellow-legged frog
R. cascadae RACA Cascades frog
R. catesbeiana RACT Bullfrog (introduced)
R. clamitans RACL Green frog (introduced)
R. pretiosa RAPR Spotted frog
Reptilia, order Chelonia (turtles):
Family Emydidae-
Chrysemys picta CHPI Painted turtle
Clemmys marmorata CLMA Western pond turtle
Reptilia, order Squamata (lizards and snakes):
Family Anguidae- -.
Gerrhonotus coeruleus GECO Northern alligator lizard
G. multicarinatus GEMU Southern alligator lizard
Family Iguanidae-
Phrynosoma douglassii PHDO Short-horned lizard
Sceloporus graciosus SCGR Sagebrush lizard
S. occidentalis scoc Western fence lizard
Family Scincidae, Eumeces skiltonianus EUSK Western skink
Family Boidae, Charina bottae CHBO Rubber boa

27
 Table 6-continued
4-letter
Scientific namea code Common name’

Family Colubridae-
Coluber constrictor coca Racer
Contia tenuis COTE Sharptail snake
Diadophis punctatus DIPU Ringneck snake
Lampropeltis getulus LAGE Common king snake
L. zonata LAZO California mountain kingsnake
Masticophis taeniatus MATE Striped whipsnake
Pituophis melanoleucus PIME Gopher snake
Thamnophis couchi THCO Sierra garter snake
T. elegans THEL Western terrestrial garter snake
T. ordinoides THOR Northwestern garter snake
T. sirtalis THSI Common garter snake
Family Crotalidae, Crotalus viridis CRVI Western rattlesnake
a
Scientific and common names follow Banks and others (1987).
Sources: Nussbaum and others (1983) and Stebbins (1985).

Appendix 2 Data sheet for TCS-This data sheet (fig. 7) needs to be on waterproof paper. The
number of data sheets needed will depend on the number of animals captured. Note
Data Sheets that each area search (1 staff hour) is listed separately and there is room for five
animals per search. If more than five animals are captured in one area, then continue
the data in the space for the next area, but if fewer than five animals are captured,
then skip to the space for the next area before recording data from the new area.
Data categories are explained below:

1. Standard header. This will differ by study. We illustrate the information we

recorded in the old-growth study.
2. Weather (WR). Use the codes listed at the bottom of the data sheet.
3. Temperature (oC).
4. Start time, end time. Use 24-hour notation.
5. Crew. List the initials of the other crew members. The recorder should be the
same person for each stand. Note whether or not the recorder participated in
the collecting.
6. Catalog initials. Initials of the collector in whose catalog the specimens will be
recorded.
7. Area. Each-l -staff hour search should be numbered sequentially.
8. Aspect (degrees). Record for each area searched.
9. Slope (percent). Record for each area searched.
10. Specimen number. Each herptile encountered is given a unique number, either
sequentially for the entire stand (1, 2, 3,...n), or sequentially for each area
searched (l-l, 1-2, l-3,...n; 2-1, 2-2, and so forth). Whichever method is used,
the data collected in the laboratory (see below) must be matched to the data
collected in the field.
11. Catalog number. This is the number given to preserved specimens. We use
small, rectangular tags, preprinted with the catalog number.
12. Species. This is the four-digit code for each species (see appendix 1).

28
 TIME-CONSTRAtNED COLLECTING DATA

snout-vent total W&A rree d-v covex object (cm)

catalog number ,, SpCieS ..= ..lagth (mm) ..lw@ (mm) __ m= 0 ~0s SpWkS width ..
;Tpec# II *e= *. leJ%o ..

snout-vent mee decay cover object (cm)

VERTICAL FOSlTlON: WEATHER

l-onsulface 7 - Under Bark on Log l-clear 6 - Showers

2 - Under Rock (not in talus) 8-UnderBarkonGround~ ImfromSnag 2-Paltlycloudy 7-Rain
3-UnderRockinTalus 9-UnderBarkonGround < I mfromLog 3-overcast 8 - Snow
4-UnderlBg 10 - Under Bark on Ground 4-Fogonly
5-InlDg (>lmfromlogorsnag) 5-Dride
6-UnderE~arkonSnag ’ 11 -other

Figure 7-Data sheet for recording information from time-constrained surveys. This sheet should be on
waterproof paper.
 The following data items (13-17) are recorded most accurately from anesthetized
animals in the laboratory. Animals should be placed individually in plastic bags with
the specimen number (item 9) so that the data can be properly recorded. If animals
are released after collecting, these data can still be recorded; they will have slightly
lower accuracy.

13. Sex. M = male; F = female; if unknown, leave blank.

14. Age. A = adult, J = juvenile.
15. Snout-vent length. Record to the nearest 0.1 millimeter.
16. Total length. Record to the nearest 0.1 millimeter. If the tail has been broken or
is otherwise incomplete, leave this blank.
17. Mass. Record to the nearest 0.1 gram.

Items 18-21 are recorded in the field.

18. Vertical position. Use the codes at the bottom of the data sheet.
19. Tree species. The four-digit code for the species (if known) of the snag or log.
20. Decay class. Use the code for either snags or logs, as appropriate.
21. Cover-object dimensions. Record length and width to the nearest centimeter.

Data sheet for surveys of CWD-This data sheet (fig. 8) also needs to be on water-
proof paper. Data for the specimens collected at each log are recorded directly below
the data for the log. At least 15 data sheets will be needed per study area. As with
the data for TCS, there is room for five animals per log. If the number of animals cap-
tured exceeds the space available, then follow the same procedures as for TCS.
Data items are explained below:

1. Standard header. The first two lines at the top of the page are the same as for
TCS. The following items (2-13) are data collected on each log before it is
searched for animals.
2. Log number. Number logs sequentially from the start of each survey.
3. Time. Record the number of minutes required to search the log (20 staff minutes,
maximum).
4. Decay class. Use the five-class scale. Other decay categories can be assigned
during data analysis.
5. Tree species.
6. Aspect.
7. Slope. Record the percent slope over a 10-m run, with the log at the midpoint.
8. Total log: length. Record to the nearest meter.
9. Total log: maximum diameter (cm).
10. Total log: minimum diameter (cm).
11. Portion sampled: length (m). Record the amount of the log that was actually
searched.
12. Portion sampled: maximum diameter (cm).
13. Portion sampled: minimum diameter (cm).

30
 LOG SURVEY DATA

technique province habitat stand # month year temp start time end time crew recorder
JI-llday-l--J&t II-II

uamn-n~ffimmommi3mnmmmm
d-y total log: portion sampled:
1% # , time ,,class , tree species aspect % slope length (m) ,max diam (cm) , min diam (cm) length (m) max diam (cm) min diam (cm)

mmnmnmmim3iin3m2~
specimen
so&men posi-
oosi- depth
death snout-vent total

d-y total log: portion sampled:

log # , time ,,class , aspect % slope length (m) max diam (cm) min diam (cm) length (m) max diam (cm) min diam (cm)

posi- depth snout-vent

catalog number

W
2 POSITION: 1 - Under Bark (on log); 2 - Under Bark (on ground); 3 - Under Log; 4 - In Log; 5 - Other

Figure 8-Data sheet for recording information from surveys of coarse woody debris. This sheet should be on waterproof paper,
 The following items are collected for each animal encountered. Most are the same as
for TCS and may be recorded in the field or in the lab, if all the animals are retained.
Data unique to log surveys that are recorded in the field are:

14. Position (POS). Use the codes at the bottom of the data sheet.
15. Depth in log (cm). Record the distance to the exterior surface of the log.

Pitfall trapping data sheet-These data (fig. 9) are recorded in the lab when the
animals are processed. A waterproof sheet is not necessary. Use a new data sheet
for each time the traps are checked. Most of the data are the same as those
collected for TCS and surveys of CWD. Unique elements are:

1. Trap night. Record the number of nights since the traps were opened; for
example, if the traps were opened October 1, and these are data for animals
picked up when the traps were checked on October 18, then trap nights are 17.
2. Trap number. Record the trap position (column and row) for each animal.

32
 PITFALL DATA

mP snout-vent
catalog number ,, species .. length (mm)

TRAP NIGHT: Number of nigh& since the traps were opened.

SEX: Male, &male (if unknown, leave blank)
AGE: &iult, Subadult, Juvenile

Figure g--Data sheet for recording information on animals collected in pitfall traps. This sheet does not need to be on waterproof paper

33
Appendix 3 Item Number

Materials Needed for Potato rakes (a backup rake is not a bad idea) 2
TCS or Surveys of CWD Crowbar 1
Stopwatch 1
Clipboard 1
Thermometer 1
Plastic bags several
Cloth bags or pillowcases 1 or2
Pencils 2+
Compass 1
Clinometer 1
Short (15 cm) plastic ruler 1
Long (30 cm) plastic ruler 1
10-m measuring tape 1

Materials Needed for Item Number

Pitfall Installation
and Operation Installation
Posthole digger (l/person) I+
15-m tape or measured nylon rope 2+
Plastic flagging (1 roll/pair of people) l+
Waterproof ink marker (l/person) l+
Number 10 tin cans 72/grid, or
24/array
l-lb margarine tubs 36/grid, of
12/array
Wood covers 36/grid, or
12/array

Operation
Waterproof notebook and paper (l/person) l+
6- by 10-inch plastic bags Many
12- by 16-inch plastic bags Many
Plastic cup or long handled spoon (l/person) l+
Small cooler with reusable refrigerant l+

Materials Needed Calipers

in the Lab Plastic ruler (30 cm)
Spring scales (10 g, 50 g, 100 g)
Scissors
Forceps
Tags with preprinted catalog numbers
Paper towels (industrial type)
Plastic trays with lids
Cloretone
Nembutol
Formalin (40 percent formaldehyde solution)
95 percent ethanol
Isopropyl alcohol (dilute to 50 percent)
Jars for specimen storage

34