Survival and Growth of Freshwater Pulmonate and Nonpulmonate Snails in 28-Day Exposures to Copper, Ammonia, and Pentachlorophenol

Abstract

We performed toxicity tests with two species of pulmonate snails (Lymnaea stagnalis and Physa gyrina) and four taxa of nonpulmonate snails in the family Hydrobiidae (Pyrgulopsis robusta, Taylorconcha serpenticola, Fluminicola sp., and Fontigens aldrichi). Snails were maintained in static-renewal or recirculating culture systems with adults removed periodically to isolate cohorts of offspring for toxicity testing. This method successfully produced offspring for both species of pulmonate snails and for two hydrobiid species, P. robusta and Fluminicola sp. Toxicity tests were performed for 28 days with copper, ammonia, and pentachlorophenol in hard reconstituted water with endpoints of survival and growth. Tests were started with 1-week-old L. stagnalis, 2-week-old P. gyrina, 5- to 13-week-old P. robusta and Fluminicola sp., and older juveniles and adults of several hydrobiid species. For all three chemicals, chronic toxicity values for pulmonate snails were consistently greater than those for hydrobiid snails, and hydrobiids were among the most sensitive taxa in species sensitivity distributions for all three chemicals. These results suggest that the toxicant sensitivity of nonpulmonate snails in the family Hydrobiidae would not be adequately represented by results of toxicity testing with pulmonate snails.

Full text

Survival and Growth of Freshwater Pulmonate
and Nonpulmonate Snails in 28-Day Exposures
to Copper, Ammonia, and Pentachlorophenol
John M. Besser
1
•Rebecca A. Dorman
1
•Douglas L. Hardesty
1
•
Christopher G. Ingersoll
1
Received: 27 July 2015 / Accepted: 13 December 2015 / Published online: 8 January 2016
ÓSpringer Science+Business Media New York (outside the USA) 2016
Abstract We performed toxicity tests with two species of
pulmonate snails (Lymnaea stagnalis and Physa gyrina)
and four taxa of nonpulmonate snails in the family
Hydrobiidae (Pyrgulopsis robusta,Taylorconcha serpen-
ticola,Fluminicola sp., and Fontigens aldrichi). Snails
were maintained in static-renewal or recirculating culture
systems with adults removed periodically to isolate cohorts
of offspring for toxicity testing. This method successfully
produced offspring for both species of pulmonate snails
and for two hydrobiid species, P. robusta and Fluminicola
sp. Toxicity tests were performed for 28 days with copper,
ammonia, and pentachlorophenol in hard reconstituted
water with endpoints of survival and growth. Tests were
started with 1-week-old L. stagnalis, 2-week-old P. gyrina,
5- to 13-week-old P. robusta and Fluminicola sp., and
older juveniles and adults of several hydrobiid species. For
all three chemicals, chronic toxicity values for pulmonate
snails were consistently greater than those for hydrobiid
snails, and hydrobiids were among the most sensitive taxa
in species sensitivity distributions for all three chemicals.
These results suggest that the toxicant sensitivity of non-
pulmonate snails in the family Hydrobiidae would not be
adequately represented by results of toxicity testing with
pulmonate snails.
Freshwater mollusks (snails, Gastropoda, and mussels,
Bivalvia) are an ecologically important component of
freshwater faunas that is both species-rich and subject to a
high rate of endangerment (Cuttelod et al. 2011; Johnson
et al. 2013). Despite their importance and the awareness
that pollution is an important threat to mollusk populations,
freshwater mollusks have long been underrepresented in
toxicity databases used to develop water-quality standards.
Neither published guidelines for development of United
States national water-quality criteria (Stephan et al. 1985)
nor guidelines for European Union risk assessments for
aquatic contaminants (European Commission 2011) require
inclusion of toxicity data for mollusks. Recent standard-
ization of methods has led to rapid expansion in the
availability of toxicity data for freshwater mussels, which
have been found to be highly sensitive to some groups of
aquatic contaminants, notably metals and ammonia (Aug-
spurger et al. 2007; Wang et al. 2007; United States
Environmental Protection Agency [USEPA] 2013), but the
availability of laboratory toxicity data for freshwater snails
remains limited. Historically, most published toxicity tests
with snails have been acute lethality tests performed with
field-collected snails (e.g., Nebeker et al. 1986; Mebane
et al. 2012), although a few long-term tests suggest that
greater sensitivity is likely in chronic tests (Arthur and
Leonard 1970; Hedtke et al. 1986; Reed-Judkins et al.
1997). Air-breathing or pulmonate snails (Heterobranchia:
families Physidae, Lymnaeidae, and Planorbidae) have
been widely used in laboratory toxicity tests due to their
ease of culture, rapid growth, short generation time, and
high reproductive output. However, nonpulmonate snails
(Caenogastropoda: 10 families and 540 species in North
America) are more taxonomically diverse, and their phys-
iology (respiration by way of gills) and life history (slow
growth, low reproductive rate) may make them more
Electronic supplementary material The online version of this
article (doi:10.1007/s00244-015-0255-3) contains supplementary
material, which is available to authorized users.
&John M. Besser
jbesser@usgs.gov
1
United States Geological Survey, 4200 E, New Haven Road,
Columbia, MO, USA
123
Arch Environ Contam Toxicol (2016) 70:321–331
DOI 10.1007/s00244-015-0255-3

susceptible to endangerment and more difficult to culture
and test in the laboratory. Evidence from biomonitoring
studies in the United Kingdom suggest that the nonpul-
monate family Hydrobiidae (Pyrg pebblesnails) is the most
sensitive family of aquatic invertebrates to some contam-
inants (Peters et al. 2014).
Hydrobiidae is the most species-rich snail family in
North America and Europe, and more then half of hydro-
biid species are considered vulnerable, threatened, or
endangered (Cuttelod et al. 2011; Johnson et al. 2013). Our
research with was focused on possible threats to several
species of hydrobiid snails of conservation concern from
the Snake River valley of Idaho, USA (United States Fish
and Wildlife Service 1995). We obtained specimens of
several taxa of pulmonate and nonpulmonate snails,
attempted to culture them in the laboratory, and performed
a series of 4-week toxicity tests with three chemicals. The
objectives of this study were to evaluate the suitability of
nonpulmonate snails for laboratory culture and toxicity
testing and to compare the sensitivity of pulmonate and
nonpulmonate snails and other aquatic taxa to toxicity of
selected chemicals.
Methods
Four species of hydrobiid snails were collected from the
field and shipped to the United States Geological Survey’s
Columbia Environmental Research Center in Columbia,
Missouri, USA. Three species were collected in 2005 and
2006 from habitats in the Snake River valley of Idaho, USA:
Bliss Rapids snail (Taylorconcha serpenticola), Jackson
Lake springsnail (Pyrgulopsis robusta), and an unnamed
species of pebblesnail (Fluminicola sp.). Bliss Rapids snail
is listed as threatened under the United States Endangered
Species Act, and P. robusta was formerly listed as ‘‘Idaho
springsnail’’ (P. idahoensis) before the Snake River popu-
lation was lumped into the nonlisted species, P. robusta.A
fourth hydrobiid species, Ozark springsnail (Fontigens
aldrichi) was collected from southern Missouri USA in fall
2006. Two species of pulmonate snails were also cultured in
the laboratory: the pond snail, Lymnaea stagnalis, was
obtained from a laboratory culture and the pouch snail,
Physa gyrina, was collected from Big Creek, Taney County,
Missouri, USA. In the following text, we refer to these six
taxa by their genus names except where species names are
needed for clarity.
Before the first round of testing, snails were held in 40-L
aquaria with well water (hardness 280 mg/L as CaCO
3
)at
18 °C. Aquaria were aerated and one-half of the water
replaced twice per week. Ambient laboratory lighting
(400–800 lux) was provided by fluorescent bulbs on a 16-h
light-to-8-h dark cycle. Before the second round of testing,
snail cultures were moved to 9-L aquaria in a recirculating
culture system with diluted well water (hardness 100 mg/
L). Culture aquaria were equipped with spawning sub-
strates, i.e., natural rock and/or stainless steel half-round
fish spawning tiles. Nonpulmonate snails were primarily
fed a combination of Instant Algae (Nanochloropsis) and
Shellfish Diet (Reed Mariculture, Campbell, California).
These diets were painted onto stainless steel substrates and
allowed to air dry before they were placed in the culture
tanks to decrease flushing of suspended algae from the
culture tanks. Cultures of pulmonate snails were primarily
fed organically grown carrots and lettuce.
Cohorts of snails for testing were obtained by moving
adults to clean aquaria to deposit eggs then removing adults
to allow eggs to hatch. The length of time adults were
isolated for egg deposition differed among species and
cohorts depending on the number of adults and the rate of
egg deposition. Lymnaea produced enough eggs for a test
cohort within 24 h, whereas Physa typically required
approximately 7 days; offspring of both species were
reared for approximately 7 days before testing. Nonpul-
monate snails typically required about 4 weeks for egg
deposition, and offspring were reared for 4 weeks or more
before testing. Of the nonpulmonate snails, only Pyrgu-
lopsis and Fluminicola regularly produced enough off-
spring for testing in even-aged cohorts. Neither Fontigens
nor Taylorconcha reproduced well in our laboratory, and
these species were tested only in mixed-age groups of field-
collected snails. To collect neonates, 90 % of the water was
siphoned from the tank and the sides and the bottom of the
tank were sprayed gently with well water to loosen snails
and detritus. Rinse water containing neonates and detritus
was passed through a 150-lm stainless steel sieve, and
sieve contents were examined with a dissecting micro-
scope. Neonate snails were collected using a capillary tube
syringe system, which consisted of a glass capillary tube
(1.17-mm inner diameter) connected by vinyl tubing (1.0-
mm inner diameter) to a 2.5-cm, 1-mL syringe with
16-gauge needle (ASTM International 2015a). After col-
lection, young snails were placed into a 4-L beaker con-
taining 2 L of well water. These beakers were aerated
gently; a small amount of food was offered; and the snails
were acclimated to the test water over a 48-h period.
Test conditions for 28-day toxicity tests with freshwater
snails (Online Resource No. 1) were adapted from standard
methods for testing with juvenile mussels (ASTM Interna-
tional 2015a; Wang et al. 2007). Tests were performed under
flow-through conditions in automated proportional diluters
in hard reconstituted water [hardness 160–180 mg/L as
CaCO
3
, dissolved organic carbon (DOC) approximately
0.5 mg/L] at 20 °C (ASTM International 2015b). Test
chambers were screened 300-mL beakers with 200 mL
water, which received 4 volume additions/d. Ten snails were
322 Arch Environ Contam Toxicol (2016) 70:321–331
123

added to each replicate chamber with 4–6 replicates/expo-
sure concentration in survival and growth tests and 2 repli-
cates/concentration for survival-only tests. The 28-day
survival and growth tests with pulmonate snails (Lymnaea
and Physa) were performed with uniform-aged cohorts, but
age/size groups used for tests with nonpulmonate snails
were more variable. Mean shell length and approximate age
range of test cohorts at the start of the exposures are listed in
Table 1. Tests with Lymnaea were started with neonates
produced from eggs laid within 24 h and reared to 7–8 days
post-hatch, and tests with Physa were performed with
cohorts from eggs laid within a 7-day period, and reared to
2–3 weeks post-hatch. Laboratory-cultured cohorts of Pyr-
gulopsis and Fluminicola were less uniform in age and size
due to variation in the productivity of cultures and time to
hatching as well as difficulties encountered with testing
small juveniles. The first two 28-day toxicity tests per-
formed with Pyrgulopsis (tests 6 and 7; Table 1), started
with small juveniles approximately 2–3 week post-hatch
(3–6 weeks after egg deposition) failed due to high control
mortality. Subsequent tests with these taxa were started with
older juveniles (5–13 weeks post-hatch for Pyrgulopsis;
8–13 weeks post-hatch for Fluminicola). During the first
round of testing, additional 28-day survival-only tests (tests
12 to 19) were performed with mixed-age snails from field
collections (Taylorconcha and Fontigens) or with older,
mixed-age snails from cultures (Pyrgulopsis and Flumini-
cola). Field-collected snails used in these tests had been held
for periods from 4 (Fontigens) to 6 months (Taylorconcha)
before they were used in toxicity tests, and laboratory-cul-
tured snails in these tests were approximately 4–8 months
old for Pyrgulopsis and 6–12 months old for Fluminicola.
Snails were fed with suspensions of flake fish food
(Tetramin) during tests with rations expressed as dry mass
per chamber. Tests starting with the smaller snails, Pyr-
gulopsis and Fluminicola (at least 4 weeks post-hatch) and
Table 1 List of 28-day snail toxicity tests showing age and size of snails at the start of tests and control survival on day 28
Test Species Chemical Sequence Starting age Shell diam. (mm) Control survival (%)
Test group 1
1LSCu A \1 week 2.41 93
2 LS PCP B \1 week 1.33 98
3LSCu 1 \1 week 1.65 95
4 LS PCP 2 \1 week 1.83 100
5LSNH
3
3\1 week 1.74 98
6 SS Cu A 2–3 weeks 0.46 60
7 SS PCP B 2–3 weeks 0.59 5
8 SS Cu 1 5–7 weeks 0.49 98
9 SS PCP 2 6–9 weeks 0.93 100
10 SS NH
3
3A 11–13 weeks 1.11 100
11 SS NH
3
3B 7–9 weeks 0.94 100
12 SS Cu 4 Mixed 2.00 100
13 SS NH
3
5 Mixed 2.00 100
14 BR Cu 4 Mixed 1.98 100
15 BR NH
3
5 Mixed 1.98 100
16 PS Cu 4 Mixed 2.69 90
17 PS NH
3
5 Mixed 2.69 100
18 OZ Cu 4 Mixed 1.31 100
19 OZ NH
3
5 Mixed 1.31 100
Test group 2
20 PS NH
3
1A 6 months 1.35 75
21 PS NH
3
1B 6 months 1.81 93
22 PS NH
3
2 8–13 weeks 1.34 55
23 PS Cu 1 9–11 weeks 1.33 40
24 PG NH
3
2 2–3 weeks 0.95 95
25 PG Cu 1 2–3 weeks 0.79 100
26 PG Cu 2 2–3 weeks 0.85 98
Cu, copper; NH
3
, ammonia; LS,L. stagnalis;SS,P. robusta;BR,T. serpenticola; PS,Fluminicola sp.; OZ,F. aldrichi;PG,Physa gyrina
Arch Environ Contam Toxicol (2016) 70:321–331 323
123

Physa (7–14 days post-hatch), included feeding 1.5 mg on
Monday, Wednesday, and Friday for the first 14 days, then
1.5 mg daily for the remainder of test. Tests with larger
snails, Lymnaea and mixed-age groups of nonpulmonate
snails, included feeding 6 mg/d. Chambers for tests with
Lymnaea were covered with screened caps to prevent the
snails from escaping. Survival was estimated by visual
observation on day 4 or day 7 and was confirmed on day 28
by observation of movement with a microscope. Growth
was evaluated by digital measurement of maximum shell
diameter of each survivor on day 28, except the growth of
Lymnaea in tests 1 and 2 was quantified as average wet
weights for each replicate. Growth was not measured at the
end of the survival tests performed with the mixed-age
snails. Raw growth data were converted to growth incre-
ments by subtracting the mean wet weight or mean shell
length measured on day 0 for each test.
Toxicity tests were performed with copper (copper
sulfate), pentachlorophenol (PCP; stock prepared in tri-
ethylene glycol), and ammonia (ammonium chloride).
These chemicals were selected based on their differing
modes of toxic action (Dwyer et al. 2005) and on the
sensitivity of snails in acute range-finding tests. Each
chemical was tested at five concentrations in a 50 %
dilution series plus a control. The control water for PCP
tests included 0.26 mL/L of triethylene glycol, the same
concentration as in the highest PCP treatment. Tests with
ammonia were performed at a target pH of 8.2 using pH-
adjusted ammonia stock solutions (mixtures of ammonium
chloride and ammonium hydroxide) and adjusting the pH
of incoming dilution water to 8.2 using automated pH
controllers. Water samples were collected every 2 weeks
for analysis of filterable copper (0.45-lm pore diameter)
or total PCP and weekly for analysis of total ammonia.
Samples for analysis of copper were filtered using a
polypropylene syringe filter (0.45-lm pore diameter) and
analyzed without concentration or cleanup steps. Copper
concentrations were determined by inductively-coupled
plasma-mass spectroscopy (May et al. 1997). PCP con-
centrations were determined by high-performance liquid
chromatography using a C18 column, a mobile phase of
acetonitrile/water, and UV detection (Orazio et al. 1983).
Total ammonia concentrations were determined on unfil-
tered samples by ion-selective electrode and are expressed
as total ammonia nitrogen (mg TAN/L). Dissolved
organic carbon (DOC) and major ions were analyzed
periodically in test water from supply lines, and routine
water-quality parameters were monitored biweekly in test
chambers.
Toxicity values for survival and growth endpoints were
evaluated by two methods. Statistical differences among
treatments were determined by analysis of variance of
rank-transformed data using SAS/STAT software (version
9.2) using Dunnett’s test to determine significant differ-
ences between treatments and controls. Concentration-re-
sponse curves were characterized using Toxicity Response
Analysis Program (TRAP; version 1.2) to estimate effect
concentrations for survival (median lethal concentration
[LC50] and 20 % lethal concentration [LC20]) and growth
(20 % effect concentration [EC20]). Concentration-re-
sponse models were based on mean measured toxicant
concentrations for each treatment (log-transformed) and
toxicity endpoints measured by replicate. Concentration-
response analysis of growth data were performed after the
data were converted to growth increments. Most models of
survival data were fitted to Gaussian tolerance distribu-
tions, and models for growth data were fit using logistic
regression. If data did not adequately define sigmoidal
concentration response curves, they were fit to linear trends
with survival data fit to rectangular tolerance distribution
and growth data fit by piece-wise linear regression
(Erickson et al. 2010).
Results
Characteristics of test waters are listed in Table 2. Con-
centrations of hardness cations and other major ions in test
waters were similar among tests and were consistent with
guidance from American Society for Testing and Materials
(ASTM International 2015b) specifications except for
slightly greater than expected pH (mean 8.33) and alka-
linity. Mean temperatures remained within the nominal
range (20 °C±1°C). DOC concentrations averaged
slightly greater in group-1 tests reflecting the greater
detection limit for these analyses (1.0 mg/L), but improved
analytical sensitivity during group-2 tests was reflected in a
lower mean DOC concentration (0.6 mg/L), which is
consistent with recent analyses of comparable test waters in
our laboratory (Wang et al. 2016). Dissolved oxygen
concentrations averaged [75 % of saturation in all tests.
Ammonia concentrations in non-ammonia tests remained
lower than ASTM (2015b) guidelines (unionized ammonia
B0.035 mg N/L or total ammonia B0.4 mg N/L at pH 8.4
and 20 °C) except in the two highest copper concentrations
from tests 12, 14, 16, and 18, which experienced high
mortality and decomposition of larger snails leading to
total ammonia concentrations that averaged from 0.7 to
1.2 mg N/L.
Exposure concentrations for toxicity tests (Online
Resources No. 2 through 4) were close to nominal con-
centrations with a few exceptions. Copper samples col-
lected on day 28 from tests 23 and 26 (with Fluminicola
and Physa, respectively) had measured copper concentra-
tions less than one half nominal concentrations, suggesting
324 Arch Environ Contam Toxicol (2016) 70:321–331
123

a diluter malfunction. For these tests, means of four sam-
ples from days 0 to 21 (which were close to nominal) were
used to estimate copper exposure concentrations because
averaging data from all five sampling dates would have
decreased toxicity values by approximately 13 % and
resulted in greater differences between repeated tests for
both species. Measured copper concentrations at the lowest
exposure levels were usually greater than nominal reflect-
ing a background copper concentration in our test systems
(average for controls =0.87 lg Cu/L). Measured PCP
concentrations in tests 4 and 9 averaged approximately
50 % of nominal across all concentrations due to an
apparent problem with the toxicant delivery pump.
Ammonia concentrations in the lower treatments of tests
13, 15, 17, and 19 were less than nominal apparently due to
loss of ammonia to microbial nitrification in the test sys-
tem. The nominal ammonia concentrations were adjusted
upward in subsequent tests to offset these losses of
ammonia in lower treatments.
Snails generally adapted well to toxicity test conditions.
Snails in controls and low-chemical treatments moved
throughout the test beakers and fed actively (as evidenced
by removal of food and biofilm). Snails in some high-
chemical treatments were less active often remaining on
the bottoms of the test chambers, thus allowing food and
biofilm to accumulate. Lymnaea sometimes crawled above
the water level especially in high-copper treatments.
Growth of snails during 28-day tests varied widely among
species. Pulmonate snails grew more rapidly with shell
length in control groups increasing by factors of 3.3 for
Physa and 4.3 for Lymnaea over 28 days. Nonpulmonate
snails grew more slowly with control shell length
increasing by an average of 21 % for Fluminicola and
63 % for Pyrgulopsis.
Treatment means and results of analysis of variance
(ANOVA) for 28-day tests are presented in Online
Resources no. 2 (ammonia), no. 3 (copper), and no. 4
(PCP). All tests performed with pulmonate snails had
control survival greater than our test acceptability criterion
of 90 %, and tests with the field-collected Taylorconcha
(tests 14 and 15) and Fontigens (tests 18 and 19) had
100 % control survival, but tests with laboratory cultured
nonpulmonate snails had more variable control survival
(Table 1). Control survival of Pyrgulopsis was 60 % in test
6 and only 5 % in test 7, which had survival \25 % in all
treatments. Subsequent tests with Pyrgulopsis were stocked
with older snails and had control survival of 98–100 %.
Similarly, tests performed with larger Fluminicola (shell
diameter 1.8–2.7 mm) from field collections (tests 16 and
17) or laboratory cultures (test 21) had control survival
from 90 to 100 %, but tests started with smaller Flumini-
cola (1.33–1.35 mm) had control survival of 40–75 %
(tests 20, 22, and 23).
Table 2 Summary of test water characteristics during group-1 and group-2 toxicity tests with freshwater snails
Parameter Sodium
(mg/L)
Calcium
(mg/L)
Magnesium
(mg/L)
Potassium
(mg/L)
Chloride
(mg/L)
Sulfate
(mg/L)
DOC
(mg/L)
pH Hardness
(mg/L as
CaCO
3
)
Alkalinity
(mg/L as
CaCO
3
)
Ammonia
(mg/L)
Dissolved
oxygen
(mg/L)
Group 1 tests
Mean 50 30 22 4.0 4.9 167 0.72 8.33 172 123 0.22 7.9
Minimum 47 28 21 3.4 1.7 136 0.50 8.17 160 116 0.10 6.9
Maximum 57 34 23 4.5 7.9 189 1.80 8.48 191 128 1.20 8.5
N7 7 7 7 7 7 12 42 42 42 30 42
Group 2 tests
Mean 61 31 23 4.7 6.2 145 0.58 8.32 170 122 0.08 7.8
Minimum 58 29 23 4.5 6.1 141 0.17 8.18 159 111 0.07 7.5
Maximum 63 31 24 4.9 6.2 151 1.22 8.45 178 128 0.10 8.0
N4 4 4 4 3 4 34 24 24 24 12 24
Major ions and DOC analyzed in unspiked source water only; other parameters analyzed in test chambers and averaged across treatments
Arch Environ Contam Toxicol (2016) 70:321–331 325
123

The survival endpoint differed significantly among
treatments for almost all successful tests with the exception
of three ammonia tests. Two of these were tests with
Lymnaea and Physa that had no decreases in survival at the
highest concentrations tested; the third test (with Flumini-
cola) had 100 % mortality at the highest concentration but
had highly variable survival (between only two replicates)
at intermediate concentrations. Effects on growth also
differed among chemicals, with significant ANOVAs for
all successful tests with copper and PCP but for only two of
four tests with ammonia.
Concentration-response data and plots for 28-day tests
are presented in Online Resource no. 5, and effect con-
centrations for survival and growth are listed in Table 3.
Survival data produced LC50s that were well supported by
the data for all tests with significant decreases in survival,
but few tests (5 of 25) had 2 treatments with partial mor-
talities, a requirement to estimate statistically reliable 95 %
confidence limits for LC50s. The uncertainty associated
with LC50s from these tests was estimated by the range of
concentrations between treatments with [90 % survival
and those with \10 % survival. Tests with fewer than 2
treatments with partial mortalities also did not produce
reliable estimates of LC20s. Effects on growth were esti-
mated by EC20s because few tests had growth decreases of
C50 % relative to controls. Effects on survival were not
evident in most tests after 4 or 7 days, with acute LC50s
estimated for only 8 of 25 tests (Online Resource no. 6).
Effect concentrations for six snail species and three
chemicals are listed in Table 3. Ammonia toxicity differed
widely between pulmonate and nonpulmonate snails. Sur-
vival of pulmonate snails was not significantly decreased
by ammonia exposures at the highest concentrations tested,
7.9 mg/L for Lymnaea (test 5) and 9.4 mg/L for Physa in
(test 24). There was a small but significant decrease in
growth of Lymnaea in test 5, which produced an EC20 of
7.7 mg N/L. In contrast, survival of nonpulmonate snails
was significantly decreased by ammonia in seven of eight
Table 3 Effects concentrations for ammonia, copper, and PCP in 28-day toxicity tests with freshwater snails
Species Test Survival Growth Test CV Species mean CV
LOEC LC50 CL LC20 CL LOEC EC20 CL
Total ammonia (mg N/L)
LS 5 [8.0 [7.9 – 1.8 7.7 4.4–13 7.7 7.7
PG 24 [9.4 [9.4 –[9.4 [9.4 [9.4 [9.4
PS 17 [7.9 2.2 [1.0–8.0] 1.0 – – 2.2 2.2
PS 21 3.7 3.2 [3.7–8.8] 2.9 [3.7 2.2 1.2–4.0 2.2
SS 10 3.6 3.2 2.3–4.4 [7.9 [7.9 3.2 3.3
SS 13* 7.9 4.9 3.9–6.2 3.4 2.4–4.8 – – 3.4
BR 15 1.7 5.6 4.0–7.9 2.9 2.1–4.0 – – 2.9 2.9
OZ 19* 1.7 0.93 0.76–1.1 0.68 0.52–0.90 – – 0.68 0.68
Copper (lg Cu/L)
LS 1 33 22 [15–33] 18 15 10 [6.1–15] 18 22
LS 3 87 33 [17–87] 29 32 27 3.0–248 27
PG 25 60 42 [29–69] 36 27 24 [[14] 24 20
PG 26 27 34 [13–60] 25 27 16 13–19 16
PS 16 13 13 [6.6–26] 12 – – 13 13
SS 8 17 13 [8.4–32] 10 17 8.2 1.8–37 8.2
SS 12* 13 22 17–28 14 10–19 – – 14
BR 14 13 15 [6.6–26] 13 – – 15 15
OZ 18* 13 23 16–33 14 7.9–23 – – 14 14
Pentachlorophenol (lg/L)
LS 2 377 391 [[178] 306 41 36 10–127 36 66
LS 4 224 210 [[96] 190 224 122 88–168 122
SS 9 224 145 [96–227] 117 24 42 (=EC50) 25–706 42
Italics indicate values considered unreliable. Asterisks (*) indicate \20 snails/treatment
LS,L. stagnalis;PG,P. gyrina;BR,T. serpenticola;SS,P. robusta; PS,Fluminicola sp.; OZ,F. aldrichi;CL, 95 % confidence limits [estimated
values in brackets]; CV, chronic value
326 Arch Environ Contam Toxicol (2016) 70:321–331
123

tests, with LC50s ranging from 0.93 mg/L for Fontigens
(test 19) to 5.6 mg/L for Taylorconcha (test 15). The two
tests with the most robust concentration-response models
produced LC20s of 2.9 mg/L for Taylorconcha (test 19)
and 3.4 mg/L for Pyrgulopsis (test 13), and LC20s esti-
mated for other tests and species ranged from 0.68 to
3.2 mg/L. Ammonia did not significantly decrease growth
of nonpulmonate snails in any of these tests, but growth of
Fluminicola in test 21, although variable, produced a well-
defined EC20 of 2.2 mg/L.
Effect concentrations for copper toxicity were more
consistent among snail taxa and between survival and
growth endpoints (Table 3). Copper caused significant
decreases in both survival and growth of all six snail taxa
tested. Copper LC50s ranged from 13 lg/L (Fluminicola in
test 16 and Pyrgulopsis in test 8) to 42 lg/L (Physa in test
25) and copper LC20s were 14 lg/L for both Pyrgulopsis
(test 12) and Fontigens (test 18). Repeated 28-day copper
tests with three snail species produced consistent results
with the relative percent difference for duplicate LC50s
ranging from 21 to 51 %. Differences between LC50s for
repeated tests were lowest for Physa in tests 25 and 26
(both started with 2- to 3-week-old juveniles) and greatest
for Pyrgulopsis in tests 8 (started with 5- to 7-week-old
juveniles) and 12 (started with [6-month-old juveniles).
Although most growth EC20s were not well defined, these
values suggested a narrower range of sensitivity of this
endpoint among Pyrgulopsis (8.2 lg/L in test 8), Lymnaea
(10 and 27 lg/L in tests 1 and 3), and Physa (16 and 24 lg/
L in tests 26 and 24).
Responses in limited testing of snails with PCP differed
between species, tests, and end points (Table 3). Two tests
with Lymnaea produced very different results: test 2 pro-
duced little lethality (LC50 =391 lg/L) but had sensitive
sublethal effects on growth (EC20 =36 lg/L), whereas test
4 produced greater lethality (LC50 =210 lg/L) but fewer
effects on growth (EC20 =122 lg/L). The single success-
ful test with Pyrgulopsis (test 9) had sensitive responses of
both lethality (LC50 =145 lg/L) and growth. Although the
growth EC20 in this test was poorly defined, a growth EC50
was estimated at 42 lg/L.
Discussion
Median lethal concentrations, LC20s, and EC20s for growth
in successful tests are summarized in Fig. 1. Differences
among snail taxa were most pronounced for ammonia
(Fig. 1a). No effects of ammonia on survival and minimal
effects on growth were observed for the pulmonate snails,
Lymnaea and Physa, at the highest concentrations tested
(7.9–9.4 mg TAN/L). In contrast, tests with four species of
nonpulmonate snails produced LC50s B5.6 mg/L and
LC20s or growth EC20s B3.4 mg/L. Results of copper tests
showed a similar pattern with effect concentrations for
nonpulmonates (8.2–14 lg/L) being consistently lower than
those for pulmonates (16–27 lg/L; Fig. 1b). Tests with PCP
suggested fewer differences between Lymnaea and Pyrgu-
lopsis, although both the LC50 and the growth EC50 for one
test with Pyrgulopsis were less than comparable values from
two tests with Lymnaea (Fig. 1c).
Toxicity thresholds for each test, species, and chemical
are listed in Table 3. The sensitivity of the six snail species
to ammonia, copper, and PCP was compared by deter-
mining the lowest reliable chronic effect concentration
(LC50, LC20, or growth EC20) for each test and averaging
values from multiple tests per species when possible to
produce a mean chronic value for each taxon. Effect con-
centrations for endpoints that had both significant ANO-
VAs and well-defined TRAP models were assumed to be
most reliable. Effect concentrations from tests with control
survival\90 % were not used for these determinations, but
tests with fewer than the recommended number of test
organisms (\20/treatment) were used as long as these tests
had acceptable control survival. In a few cases, we used
effect concentrations from end points with significant
ANOVAs but less-reliable TRAP models (e.g., ammonia
growth EC20 for Lymnaea in test 5) or end points that did
not have significant ANOVAs but had reliable TRAP
models (e.g., ammonia LC50 for Fluminicola in test 17 and
ammonia growth EC20 for Fluminicola in test 21). In most
cases, the lowest effect concentrations from repeated tests
with the same chemical and test organism were similar.
Only Lymnaea in tests 2 and 4 had growth EC20s for PCP
that differed by more than a factor of 3 despite having
similar LC50s. Effect concentrations estimated from tests
that failed due to low control survival were similar (within
a factor of 2) to values determined from successful tests.
Chronic effect concentrations for toxicity of ammonia,
copper, and PCP to freshwater snails from the present study
are compared with mean chronic values for other fresh-
water genera in Fig. 2. For ammonia and PCP, these spe-
cies sensitivity distributions (SSDs) are based on genus
mean chronic values from water-quality criteria documents
(USEPA 1996,2013). For copper, Wang et al. (2007)
identified published chronic effect concentrations that had
the supporting water chemistry data needed for BLM
models. Effect concentrations were adjusted to standard
water chemistry conditions to facilitate comparisons
among studies: pH 8.0 and 20 °C for total ammonia
(USEPA 2013), 85 mg/L hardness and 0.5 mg/L DOC for
copper (USEPA 2007; Hydroqual 2007), and pH 7.8 for
PCP (USEPA 1996). In our tests, chronic toxicity values
for all four nonpulmonate taxa were at or below the 50
th
percentile, and those for both pulmonate taxa were at or
above the 50
th
percentile of species sensitivity distributions
Arch Environ Contam Toxicol (2016) 70:321–331 327
123

A
Ammonia
LS PG PS* PS SS SS* BR* OZ*
Effect concentration, mg TAN/L
0
2
4
6
8
10
Growth EC20
Survival LC20
Survival LC50
>
>
B
Copper
LS LS PG PG PS* SS SS* BR* OZ*
Effect concentration, µg/L
0
10
20
30
40
C
Pentachlorophenol
LS LS SS
Effect concentration, µg/L
0
100
200
300
400
500
Fig. 1 Comparison of effect concentrations for freshwater snails
from 28-day tests with three chemicals. aAmmonia. bCopper.
cPCP. Each bar represents a separate test with the total height of bar
indicating the LC50, with the solid portion of the bar indicating the
LC20 and the hatched portion of the bar indicating the EC20 for
growth. Confidence intervals for individual effect concentrations are
listed in Table 3. Species tested included L. stagnalis (LS) and P.
gyrina (PG); Bliss Rapids snail (BR; T. serpenticola), Jackson Lake
springsnail (SS; P. robusta), Fluminicola sp. (PS), and Ozark
springsnail (OZ; F. aldrichi). Symbol ([) indicates that the LC50
was greater than the highest concentration tested. Asterisk (*)
indicates test that did not have a growth end point
A. Ammonia (mg TAN/L)
1 10 100
0.0
0.2
0.4
0.6
0.8
1.0
Fontigens
Taylorconcha
Pyrgulopsis
Fluminicola
Lymnaea
Physa
C Pentachlorophenol ( µg/L)
Genus mean chronic value
1 10 100
0.0
0.2
0.4
0.6
0.8
1.0
Pyrgulopsis
Lymnaea
Physa
B Copper ( µg/L)
0.1 1 10
Proportion of taxa affected
0.0
0.2
0.4
0.6
0.8
1.0
Pyrgulopsis
Taylorconcha
Fluminicola
Lymnaea
Physa
Fontigens
Lymnaea*
Fig. 2 Species sensitivity distributions for chronic toxicity of atotal
ammonia, bcopper, and cPCP. Genus mean chronic values for
freshwater snails (squares =pulmonate snails; triangles =nonpul-
monate snails) and other taxa (circles) from the present study (hollow
symbols; Table 3) and from previous publications were adjusted to
common water chemistries to facilitate comparisons: pH 8.0 and
25 °C for ammonia; standard moderately hard test water for copper
(hardness 85 mg/L as CaCO
3
and DOC 0.5 mg/L USEPA 2007); and
pH 7.8 for PCP (USEPA 1996). Vertical dashed lines indicate
USEPA ambient water-quality criteria. Two copper chronic values are
presented for Lymnaea to compare of results of the present study with
other published studies (*)
328 Arch Environ Contam Toxicol (2016) 70:321–331
123

for copper, ammonia, and PCP. For ammonia, the most
sensitive snail taxon was Fontigens, which ranked second
of 21 genera tested. Pyrgulopsis was the most sensitive
snail we tested to both copper and PCP. For copper, Pyr-
gulopsis was ranked fourth of 19 genera. The sensitivity of
hydrobiid snails to copper in our 28-day tests is consistent
with results presented by Reed-Judkins et al. (1997) from
114-day exposures to the nonpulmonate snail, Leptoxis
praerosa (Pleuroceridae), which experienced decreased
survival at copper concentrations in the range of
6.3–8.8 lg/L in stream water with similar hardness.
However, recent tests with Lymnaea spp. generated chronic
toxicity values for copper substantially less than those
derived from the present study (Brix et al. 2011; Das and
Khangarot 2011). To illustrate this difference, the mean
copper effect concentration for Lymnaea from these studies
is plotted separately from our data on the copper SSD; this
value makes Lymnaea the most sensitive species tested
with copper (Fig. 2b). Genus mean chronic values for all
three chemicals from the present study would rank one or
more nonpulmonate snails among the four most sensitive
genera that are used to calculate final chronic value in the
development of national water-quality criteria in the USA,
and the derivation of the USEPA (2013) chronic criterion
for ammonia used our (then unpublished) chronic toxicity
data for Pyrgulopsis in the calculation of the final chronic
value.
Results of the present study indicate that freshwater
nonpulmonate snails of the family Hydrobiidae may be
relatively sensitive to aquatic contaminants. Tests with
hydrobiid snails generally produced lower chronic effect
concentrations than tests with pulmonate snails. However,
we encountered mixed success in our attempts to culture
hydrobiid snails in the laboratory with two species (P.
robusta and Fluminicola sp.) producing enough offspring
for testing and two other species (T. serpenticola and F.
aldrichi) producing few or no offspring. We also had dif-
ficulty achieving adequate control survival in tests with
juvenile hydrobiid snails, but this problem was generally
solved by starting tests with older juveniles. Both tests with
older juveniles and tests with adults or mixed-age groups of
hydrobiid snails produced lower effect concentrations than
tests with younger pulmonate snails. Tests with field-col-
lected F. aldrichi, which did not include a growth end point
due to high variation in starting size, produced some of the
lowest lethal concentrations for ammonia and copper. Even
when laboratory-cultured juvenile hydrobiid snails were
available, the nonuniform starting size (due to low repro-
ductive output) and slow growth of these snails hampered
efforts to quantify potentially sensitive growth responses to
chemical exposure. Because the increase in shell length of
hydrobiid snails during 28-day toxicity test (20–30 %) was
much less than that achieved by pulmonate snails
([300 %), it was necessary to express growth as the
incremental increase in shell length, relative to the average
starting shell length, to accurately model percent decreases
in growth. Although shell length was a precise and repro-
ducible measure of growth, measurement of wet weight
may provide a wider range of growth and may be a better
measure of true growth rates of hydrobiid snails. For
example, the apparent differences in sensitivity of the
Lymnaea growth end point between repeated tests with
copper and PCP may reflect in part the greater sensitivity of
growth measured as weight in tests 1 and 2 compared with
measurements of growth in shell length in tests 3 and 4
(and in subsequent tests).
Several characteristics of pulmonate snails make them
easy to use for toxicity testing. Taxa such as Lymnaea and
Physa are readily cultured in the laboratory, and their high
reproductive output allows toxicity tests to be started with
younger and more uniform-sized juveniles. The rapid
growth rates of these juveniles allow easier measurement
of wet mass after 28 days, and their large scope for growth
may make calculation of growth increment unnecessary.
Rapid growth and short generation times also make
determination of reproductive end points logistically easier
in tests with genera such as Physa and Lymnaea. Results of
the present study suggest that the advantages of pulmonate
snails as test organisms may be outweighed by the greater
sensitivity of hydrobiid snails, but other studies (Brix et al.
2011; Das and Khangarot 2011) suggest that some pul-
monate snails may be more sensitive to some chemicals
than is indicated by our test results. The sensitive response
of Lymnaea growth to copper and other metals apparently
reflects ionoregulatory disruptions that affect shell forma-
tion during rapid growth of young snails (Grosell et al.
2006; De Schamphelaere et al. 2008; Brix et al. 2011,
2012). These effects may have been less severe in our
Lymnaea tests because we started tests with older snails
(7–8 days old) compared with the\24-h old neonates used
to start exposures in other published studies. We did not
compare copper sensitivity of Lymnaea among different
size/age classes of Lymnaea, but we did observe differ-
ences in sensitivity to copper between different cohorts of
the hydrobiid Pyrgulopsis. Copper effect concentrations for
Pyrgulopsis from tests 6 and 8, which were started with
known-age juvenile snails, were lower than those from test
10, which was started with larger mixed-age snails
(Table 3). Similar trends were evident between ammonia
tests with juvenile and adult Pyrgulopsis. Toxic effects on
snail reproduction may also be easier to determine for some
taxa of pulmonate snails, such as Lymnaea and Physa, due
to their rapid growth and maturation. Hedtke et al. (1986)
reported that P. gyrina produced eggs within 2 weeks of
hatching and that egg production was the most sensitive
end point in 36-day chronic exposures to PCP. However,
Arch Environ Contam Toxicol (2016) 70:321–331 329
123

some hydrobiid snails, notably the parthenogenic and
ovoviviparous strain of the Potamopyrgus antipodarum,
are also well suited for reproductive toxicity studies (Duft
et al. 2007). The sensitivity of reproductive end points
apparently differs among taxa and among chemicals in
tests with both pulmonate and nonpulmonate snails (Gomot
1998; van Wijngaarden et al. 1998).
Results of the present study indicate that snails of the
family Hydrobiidae are sensitive to contaminants and may
not be adequately protected by toxicity tests performed
with pulmonate snails. We had mixed success when cul-
turing hydrobiid snails in the laboratory using relatively
unspecialized methods, but tests with both laboratory-cul-
tured and field-collected snails provided evidence of the
sensitivity of hydrobiid snails to the chemicals tested.
Other laboratories have developed more sophisticated
facilities and methods for the mass culture of nonpul-
monate snails for the purpose of restoration of endangered
species (Paul Johnson, Alabama Aquatic Biodiversity
Conservation Center, Marion AL, personal communica-
tion). Presumably these methods can be adapted to culture
neonate or juveniles snails for toxicity testing. Future
efforts in our laboratory will focus on the development of
better methods for quantifying effects on survival, growth,
biomass, and possibly reproduction of this diverse and
ecologically important group with the goal of better char-
acterizing the sensitivity of different end points and species
across a wider range of environmental contaminants that
may adversely affect snail populations.
References
Arthur JW, Leonard EN (1970) Effects of copper on Gammarus
pseudolimnaeus, Physa integra, and Campeloma decisum in soft
water. J Fish Res Board Can 27:1277–1283
ASTM International (2015b) Standard guide for conducting acute
toxicity tests on test materials with fishes, macroinvertebrates,
and amphibians (E 729-96 (2014)). ASTM Annual Book of
Standards Volume 11.06. ASTM International, West Con-
shohocken, PA
ATSM International (2015a) Standard guide for conducting labora-
tory toxicity tests with freshwater mussels (E2455-06 (2013)).
ASTM Annual Book of Standards Volume 11.06. ASTM
International, West Conshohocken, PA
Augspurger T, Dwyer FJ, Ingersoll CG, Kane CM (2007) Editorial:
advances and opportunities in assessing the contaminant sensi-
tivity of freshwater mussel early life stages. Environ Toxicol
Chem 26:2025–2028
Brix KV, Esbaugh AJ, Grosell M (2011) The toxicity and physio-
logical effects of copper on the freshwater pulmonate snail,
Lymnaea stagnalis. Comp Biochem Physiol C: Toxicol Phar-
macol 154:261–267
Brix KV, Esbaugh AJ, Munley KM, Grosell M (2012) Investigations
into the mechanism of lead toxicity to the freshwater pulmonate
snail, Lymnaea stagnalis. Aquat Toxicol 106:147–156
Cuttelod A, Seddon M, Neubert E, Commission ISS (2011) European
red list of non-marine molluscs. Publications Office of the
European Union, Luxembourg
Das S, Khangarot B (2011) Bioaccumulation of copper and toxic
effects on feeding, growth, fecundity and development of pond
snail Lymnaea luteola L. J Hazard Mater 185:295–305
De Schamphelaere KAC, Koene JM, Heijerick DG, Janssen CR
(2008) Reduction of growth and haemolymph Ca levels in the
freshwater snail Lymnaea stagnalis chronically exposed to
cobalt. Ecotoxicol Environ Saf 71:65–70
Duft M, Schmitt C, Bachmann J, Brandelik C, Schulte-Oehlmann U,
Oehlmann J (2007) Prosobranch snails as test organisms for the
assessment of endocrine active chemicals—an overview and a
guideline proposal for a reproduction test with the freshwater
mudsnail Potamopyrgus antipodarum. Ecotoxicology 16:169–182
Dwyer FJ, Mayer FL, Sappington LC, Buckler DR, Bridges CM,
Greer IE et al (2005) Assessing contaminant sensitivity of
endangered and threatened aquatic species: part I. Acute toxicity
of five chemicals. Arch Environ Contam Toxicol 48:143–154
Erickson RJ (2010) Toxicity Relationship Analysis Program (TRAP),
version 1.21, EPA/600/C-11/002. USEPA, Washington, DC
European Commission (2011) Technical guidance for deriving envi-
ronmental quality standards. Guidance document No. 27. Com-
mon implementation strategy for the Water Framework Directive
(2000/60/EC). Technical Report 2011-055. EC, Luxembourg
Gomot A (1998) Toxic effects of cadmium on reproduction,
development, and hatching in the freshwater snail Lymnaea
stagnalis for water quality monitoring. Ecotoxicol Environ Saf
41:288–297
Grosell M, Gerdes RM, Brix KV (2006) Chronic toxicity of lead to
three freshwater invertebrates—Brachionus calcyfloris,Chirono-
mus tentans, and Lymnaea stagnalis. Environ Toxicol Chem
25:97–104
Hedtke SF, West CW, Allen KN, Norberg-King TJ, Mount DI (1986)
Toxicity of pentachlorophenol to aquatic organisms under
naturally varying and controlled environmental conditions.
Environ Toxicol Chem 5:531–542
Hydroqual (2007) Biotic ligand model, Windows interface, version
2.2.3. User’s guide and reference manual. Hydroqual Inc.,
Mahwah, NJ
Johnson PD, Bogan AE, Brown KM, Burkhead NM, Cordeiro JR,
Garner JT et al (2013) Conservation status of freshwater
gastropods of Canada and the United States. Fisheries
38:247–282
May TW, Wiedmeyer RH, Brumbaugh WG, Schmitt CJ (1997) The
determination of metals in sediment pore waters and in 1 N HCl-
extracted sediments by ICP-MS. At Spectrosc 18:133–139
Mebane CA, Dillon FS, Hennessy DP (2012) Acute toxicity of
cadmium, lead, zinc, and their mixtures to stream-resident fish
and invertebrates. Environ Toxicol Chem 31:1334–1348
Nebeker AV, Stinchffield A, Savonen C, Chapman GA (1986) Effects
of copper, nickel and zinc on three species of Oregon freshwater
snails. Environ Toxicol Chem 5:807–811
Orazio C, Kapila S, Manahan S (1983) High-performance liquid
chromatographic determination of phenols as phenolates in a
complex mixture. J Chromatogr 262:434–440
Peters A, Simpson P, Moccia A (2014) Accounting for both local
aquatic community composition and bioavailability in setting
site-specific quality standards for zinc. Environ Sci Pollut Res
21:105–117
Reed-Judkins DK, Farris JL, Cherry DS, Heath AG, Cairns J (1997)
Functional responses in Leptoxis praerosa to increasing metal
concentration and exposure duration. Environ Toxicol Chem
16:1666–1676
Stephan CE, Mount DI, Hansen DJ, Gentile JH, Chapman GA,
Brungs WA (1985) Guidelines for deriving numerical national
330 Arch Environ Contam Toxicol (2016) 70:321–331
123

water quality criteria for the protection of aquatic organisms and
their uses, EPA-PB85-227049. USWPA, Duluth, MN
United States Environmental Protection Agency (1996) 1995 updates:
water quality criteria for the protection of aquatic life in ambient
water. EPA-820-B-96-001. USEPA, Duluth, MN
United States Environmental Protection Agency (2007) Aquatic life
ambient freshwater quality criteria—copper, 2007 revision.
EPA-822-R-07-001. USEPA, Washington, DC
United States Environmental Protection Agency (2013) Aquatic life
ambient water quality criteria for ammonia—freshwater 2013.
EPA 822-R-13-001. USEPA, Washington, DC
United States Fish and Wildlife Service (1995) Snake River Aquatic
Species Recovery Plan, December 1995. USFWS, Boise, ID
van Wijngaarden RPA, Crum SJH, Decraene K, Hattink J, van
Kammen A (1998) Toxicity of derosal (carbendazim) to aquatic
invertebrates. Chemosphere 37:673–683
Wang N, Ingersoll CG, Greer IE, Hardesty DK, Ivey CD, Kunz JL
et al (2007) Contaminant sensitivity of freshwater mussels:
chronic toxicity of copper and ammonia to juvenile freshwater
mussels (Unionidae). Environ Toxicol Chem 26:2048–2056
Wang N, Dorman RA, Ingersoll CG, Hardesty DK, Brumbaugh WG,
Hammer EJ et al (2016) Acute and chronic toxicity of sodium
sulfate to four freshwater organisms in water-only exposures.
Environ Toxicol Chem 35:115–127
Arch Environ Contam Toxicol (2016) 70:321–331 331
123