=
yy
Cine

Soa
Ate

a

SNIVERSITY oc.
ILLINOIS LIBRARY
AT URBANA-CHAMPAIGN
NATURAL HIST SURVE’

Standard Protocols for Monitoring
and Sampling Zebra Mussels

J. Ellen Marsden

Biological Notes 138
April 1992

ILLINOIS
NATURAL
HISTORY
SURVEY

Illinois Natural History Survey, Lorin I. Nevling, Chief
A Division of the Illinois Department of Energy and Natural Resources

Printed by authority of the state of Illinois
374506-2M-4-92
US ISSN 0073-490X

The author is an assistant professional scientist with the Illinois Natural History Survey and the
director of the Survey’s Lake Michigan Biological Station (P.O. Box 634, Zion, IL 60099).

Cover design: Krista Sunderland
Editor: John Ballenot

A catalog of the publications of the Illinois Natural History Survey is available without charge
from the address below. A price list and an order blank are included with the catalog.

Illinois Natural History Survey
Distribution Center

607 East Peabody Drive
Champaign, !Ilinois 61820

Citation:
Marsden, J.E. 1992. Standard protocols for monitoring and sampling zebra mussels. Illinois
Natural History Survey Biological Notes 138. 40 pp.

Contents

4.6

4.7
4.8

Introduction |

Purpose of This Document |

How to Use This Document |

Need for Zebra Mussel Monitoring Information |
Definition of and Need for Standard Sampling 2
Early Detection of Zebra Mussels 2

Zebra Mussel Biology 3
Morphology 3
Reproduction and Growth 3
Distribution 4

Sampling Veligers 5

Outline 5

General Comments 5

Sampling Period 5

Equipment 6

Open Water 6

Sample site, vertical plankton tow, oblique plankton tow,

pumped sample, scouring pad samplers

Flowing Water 7

Sample site, collection procedure

Water Intakes 7

Collection of Ancillary Data 7

Counting Veligers 8

Veliger identification, presence/absence data, detection limit, veliger densities,
equipment, counting procedure for plankton tow, counting procedure for scouring
pad sampler, calculation of veliger densities

Sampling Settling Juveniles 11
Outline 11

General Comments 11

Sample Site 12

Equipment 12

Collection Procedure 13

PVC plates, slide rack, multiplate samplers
Counting Procedure 14

Slides, plates

Collection of Ancillary Data 14

Data Reporting: Biomass vs. Density 15

5 Sampling Adults 16
5.1 Outline 16

2 General Comments 16

5.3 Sample Site 16

4 Equipment 16

5 Collection Procedures Using Quadrat Frame 17

Procedure | (high adult densities), procedure 2 (low adult densities), procedure 3
(low densities, loose, irregular substrate)

6 Collection Procedures Using a Ponar Grab 17

5.7 Collection of Ancillary Data 17

8 Counting Procedure 18

6 Reporting Results 19
Zebra Mussel Sighting Report Form 20
U.S. Fish and Wildlife Service Nonindigenous Species Database Form 21

7 Acknowledgments 22
8 Literature Cited 23

Appendixes
I Distinguishing Adult Dreissena polymorpha and Mytilopsis
leucophaeata (Reprinted from MacNeill 1991) 24
II Detection of Zebra Mussel Veligers in Plankton Samples Using Sugar Solution
(Reprinted from Schaner 1990) 25
III Collector for Veliger and Drifting Postmetamorphic Zebra Mussels
(By André Martel) 28
IV Equipment and Chemicals 31
V_ Forms and Labels 32
VI Conversion Tables 38

1 Introduction

The accidental introduction of zebra mussels (Dreissena
polymorpha) into the Great Lakes in 1986 is predicted to
have significant effects on the aquatic ecosystem and on
human water users. These mussels’ high fecundity, rapid
growth, and ability to attach strongly to any hard substrate
have already caused many problems, including blockage of
water intake pipes and fouling of docks and boat hulls.
Because zebra mussels are highly efficient filter feeders,
they threaten to deplete the population of microorganisms
that are the base of the aquatic food web. Excess ingested
food is excreted in pseudofecal bundles, which can reduce
benthic dissolved oxygen during decomposition.
Information about the spread of zebra mussels is
critical for industries and public utilities concerned about
water intake systems and for fisheries agencies concerned
about managing the Great Lakes ecosystem. Water users
need an early warning of the arrival of zebra mussels to
their area so they can be prepared to handle biofouling
problems. Biologists interested in the impact of zebra
mussels on the ecosystem likewise need to know when the
mussels have arrived in their study area. Population density
data can be used to track the movement of zebra mussels, to
determine what environmental factors influence local
population densities, and to determine the need for and
efficacy of control measures. In some areas, mussel densi-
ties may never reach levels high enough to affect water
users or the local environment. Understanding the effect of
zebra mussels on the environment, as well as the effect of
the local environment on the zebra mussels, requires
monitoring zebra mussel densities and growth rates.

1.1 Purpose of This Document

As zebra mussels have spread, so has the need for monitor-
ing programs initiated by agencies, researchers, and
industries around the Great Lakes and inland waters. For
each of these groups to develop a sampling protocol de
novo would involve a huge duplication of effort. Even
simply determining the presence or absence of these
mussels can be facilitated by the use of established methods
that minimize sampling effort while maximizing informa-
tion gain. Comparisons of data among sampling stations or
within studies may be unfeasible unless the same monitor-
ing protocols are used at each site. The purpose of this
document, therefore, is threefold:

1. To document methods that fit broad user needs for
information and are known to be effective for sampling
zebra mussels. Among the numerous sampling
methodologies that have been developed for zebra
mussels in North America over the past few years,
those included in this volume were chosen on the basis
of the following criteria:

* Equipment that is inexpensive, readily obtainable, and
simple to build and deploy.

¢ Techniques that can be used in a variety of field

situations.

¢ Straightforward, rapid data collection and analytical
methods.

. To provide a zebra mussel sampling manual for people
who are not familiar with or trained in biological sam-
pling techniques.

3. To describe standard methods that can be used to

collect data that are comparable between sites.

to

1.2 How to Use This Document
There are no absolute rules for how to design a zebra
mussel monitoring program. Individuals or agencies must
decide which sampling methods to use based upon their
own clearly defined information needs. Too often, much
time and effort has been expended to gather information
later found to be useless. Many information needs may not
require either interstudy comparability (“standardized”
methods) or highly quantitative data. For example, careful,
quantitative sampling is unnecessary when monitoring to
detect the first arrival of mussels in an area. The need in
this case for presence/absence data (see below) is most
appropriately met by low-effort, high-volume sampling.
To some extent, therefore, the use of the word stan-
dard in the title of this document is misleading. Several
methods known to work for sampling zebra mussels are
represented, as are the pitfalls of each method. It is up to
the investigator to determine which of these methods fit his
or her information needs, budget, time constraints, equip-
ment availability, and technical expertise. Use of this
document does not obviate the need for careful thought to
determine what information is required.

1.3 Need for Zebra Mussel Monitoring Information
The need for monitoring is most apparent at the edge of the
zebra mussel range, where early warning of the mussels’
arrival is important for water users. A common misconcep-
tion is that monitoring serves no useful purpose once the
mussels are established at a site. Monitoring is probably
needed most at industrial sites and public utilities con-
cerned about infestations of zebra mussels. Responsible,
economically effective mussel control requires continued
monitoring of local population levels. As an obvious
example, treatments to prevent veliger settlement are
wasted if applied before or after veligers are present. Only
periodic monitoring will indicate when such treatments
should begin and end. In addition, monitoring may reveal
sites at which control of mussels is unnecessary. In
previous studies, monitoring within areas of high zebra
mussel densities has revealed that some sites, for reasons
not yet understood, remain relatively free of zebra mussels
without control measures. Finally, almost every zebra
mussel population in Europe underwent a population

2 Illinois Natural History Survey Biological Notes No. 138
decline 5 to 10 years after invasion of a new area (Walz 1600 5
1973, 1975). Only by monitoring local populations can the 1400 4
occurrence of ie geeine be eee The timing of the z pera
population decline is particularly important for biologists % 12005 Y
conducting long-range population studies of zebra mussels Sie ZA lower surface
and their effects on native species. =

& 800+
1.4 Definition of and Need for Standard Sampling E
Fundamentally, any protocol used by consensus can be a e 6004
standard. An ideal standard protocol fits broad user needs 400 4
for information and 1s easy to use. Standardization is most
important for individuals who plan to compare their data 200 4
with results from other studies or from other years within

0

the same study. These individuals include scientists,
biological consultants, fisheries managers, and individuals
from industries and public utilities. Standard methods
should never be used, however, at the cost of data quality.
Priority should be given to the ability to replicate results
within a study, and then to comparability among studies.

Zebra mussel settlement, growth, and density appear to
be affected by light, temperature, depth, currents, substrate
composition, substrate texture, substrate type, pH, ionic
concentrations, and local fauna. For example, because of
substrate preferences, equal numbers of juveniles measured
on different settlement plate materials may not reflect equal
population densities. Use of a standard sampling protocol is
thus essential for comparing data among sites and between
studies. Reporting ancillary data such as temperature,
depth, and substrate types is equally important.

1.5 Early Detection of Zebra Mussels

Personnel at power plants, water treatment facilities,
marinas, and other water-use areas may need to know as
soon as possible that zebra mussels have arrived in their
area. Identification of the “best” method for early detection
of zebra mussels is problematic. The first life stage likely to
colonize a new area is the planktonic veliger, which can
drift into new areas or be transported in bilge water. Adults
may also colonize new areas by being transported on boat
hulls. Sampling for veligers is fairly likely to give “false
negative” results because the distribution of veligers is
highly clumped. Sampling of settled juveniles more reliably
indicates the incipient formation of a local zebra mussel
population, but this requires a sensitive method to maxi-
mize early detection of these microscopic animals. Al-
though concrete blocks are commonly used as a readily
available substrate on which settled juveniles can be
detected, the juveniles may be settled for several weeks
before becoming visible against the blocks’ coarse texture.
In contrast, newly settled juveniles are highly visible on a
smooth settlement plate, which can be examined under a
microscope.

It is impossible to accurately predict which type of
sampling will produce the first evidence of zebra mussels
in a new area. At several sites where all three life stages
were monitored, either veligers, settled juveniles, or adults
were detected several weeks before the other stages were
found. Therefore, early warning of the presence of zebra
mussels can best be achieved by a combination of plankton
sampling, placement of settlement plates, and regular

Cu brass Plexiglas Al Zn

Figure 1. Numbers of settled zebra mussel juveniles on 10 cm x
10 cm plates of various substrate materials (from Walz 1973).

examination of surfaces for settled adults. Veliger sampling
will be most effective in areas where the veligers may be
carried from a distant location by a mild current. Settled
juveniles can best be detected by examining the mussels’
preferred substrates, such as polyvinyl chloride (PVC)
(Figure 1; Walz 1973, 1975). Adults are most likely to be
found in dark areas, in corners or crevices, and in areas
with a gentle current (not more than 2 m/sec). Because the
adults can only attach to hard substrate, in muddy areas
they will be found attached to embedded rocks, native
mussels, or crayfish.

Sampling to detect new infestations does not require
quantitative data. The most important aspect of this type of
sampling is to maximize the probability of detection with
minimal effort. Sampling a large volume of water for
veligers, even if the volume cannot be readily quantified, is
more valuable than spending time trying to carefully
measure 200 liters. After zebra mussels are sighted, the
monitoring program should switch to quantitative methods.

Most first sightings of mussels in new areas have been
made by the informed public—that is, by people who were
aware of the zebra mussel problem from the news or state
agency educational bulletins or by water intake personnel
aware of potential infestations. These people have reported
mussels found on boats, docks, rocks collected for aquaria,
and commercially harvested native mussel shells. Ordinary
citizens are an especially effective monitoring resource
because they greatly exceed, in time and effort spent on and
near the water, the number of trained biologists who can be
deployed in the field. Thus, the most effective programs to
monitor the spread of zebra mussels are those that involve
public education. Public education also has the benefit of
helping to prevent the accidental spread of the mussels.
Nonetheless, all sightings by nonbiologists must be
confirmed by trained personnel who are thoroughly
familiar with the zebra mussel and similar species. Most of
the state natural resource agencies and Sea Grant program
offices around the Great Lakes have produced information
pamphlets about the zebra mussel. These could be distrib-
uted with a reporting form (see example in section 6) to
marinas, bait shops, water-based industries, commercial
shell fishermen, etc., to elicit information on new sitings
and thus enhance an early warning detection program.

2 Zebra Mussel Biology

The life history of zebra mussels is described in detail, with
an annotated bibliography of European literature, in Mackie
et al. (1989). The purpose here is to summarize the biology
of the zebra mussel to point out salient features that affect
sampling.

2.1 Morphology

Zebra mussels are freshwater bivalves (family Dreissen-
idae) native to the Black and Caspian seas. The name
polymorpha (many forms) indicates the large variability in
many of their characteristics. Adult mussels have a distinc-
tively shaped shell (Figure 2) that is variably banded with
black or brown and cream stripes. Most have jagged, lateral
stripes; some have single longitudinal bands; and all-cream
or all-black individuals have been found (Figure 2). The
shell shape is diagnostic for identification.

Two other species in the family Dreissenidae are
present in North America and could be confused with D.
polymorpha. The native false dark mussel, Mytilopsis
leucophaeata, is found in brackish, estuarine waters, the
upper Mississippi and Hudson rivers, the lower Tennessee
River, and in the Ohio River below Cincinnati. A key to
distinguishing between this mussel and the zebra mussel
has been presented by MacNeill (1991) and is reprinted as
Appendix I of this publication. If there is any doubt about
correct identification, a mollusc taxonomist should be
consulted.

In 1991, a second exotic dreissenid mussel was found
in Lake Ontario (E. Marsden and B. May, Cornell Univer-
sity, unpublished data). At the time of publication, the
species identity of this second exotic had not been estab-
lished. The second species, given the working name
“quagga mussel,” is distinguished from D. polymorpha by
the marked absence of a sharp angle between the dorsal and
ventral surfaces (Figure 3). Confirmation of identification
requires genetic analysis. As of late 1991, the distribution
of the quagga mussel appeared to be limited to Lake
Ontario and the Erie Canal.

2.2 Reproduction and Growth
Adult zebra mussels reach a maximum length of 4 cm.
They usually live three to five years, though some survive
as long as nine years in Europe. These mussels typically
mature sexually in the second year of life, but in Lakes Erie
and St. Clair they generally mature in their first year.
Females can mature at 7 mm and males at 6 mm; individu-
als larger than 10 mm are usually mature (Smirnova 1990;
J. Nichols, Great Lakes Fishery Laboratory, personal
communication).

The sexes are separate, and gametes are released
synchronously into the water column for external fertiliza-
tion. Reproduction begins when the water temperature

remains above 10°C (50°F) for one to two weeks. Spawn-
ing is stimulated by water temperatures of 12°C (54°F) and
by the presence of gametes in the water. Nonetheless,
veligers have been found in the plankton when spring water
temperatures were still as low as 8.5°C (Joe Leach, Ontario
Ministry of Natural Resources, personal communication).
Females can spawn throughout the year at 8- to 10-week
intervals in warm-water areas (Nichols and Kollar 1991).
Individual females usually release 30,000 to 40,000 eggs
each year during a period of several weeks. As many as
1,000,000 eggs per female per year have been noted (Walz
1978).

In Lake Erie, two spawning peaks have been noted, in
late July and late August (Garton and Haag 1990). The egg
hatches within a few days to release a veliger 40—70 um in
diameter. The veliger is a ciliated, free-swimming plank-
tonic stage that is readily transported in water currents.
Veliger densities as high as 400,000/m* have been observed
in Europe (Smirnova 1990), and densities of 1,000,000
veligers/m* have been observed in Lake Erie.

In 8 to 15 days, the veligers grow to 150—250 tum and
develop a clamlike shell. As the shell develops, the ciliated
velum is lost, and the veliger becomes too heavy to swim
and settles onto the substrate. Settlement may occur at a
range of sizes (Lewandowski 1982); in the Great Lakes,
settling juveniles are generally 180-250 um (G. Mackie,
University of Guelph, personal communication), although
individuals up to 2 mm have been found in the plankton
(see Appendix III).

The settled juveniles, also referred to as postveligers or
spat, initially have a round, symmetrical shell. Within a few
days the juveniles begin to elongate into their adult shape
and develop pigmentation in the shell. If settlement occurs
on a hard substrate, the juveniles may crawl around for
several days before extruding gluelike fibers called byssal
threads to attach to the substrate (Lewandowski 1982). The
juveniles prefer dark areas, such as the underside of
suspended surfaces, with currents that will transport food
organisms. In studies by Walz (1973), settlement on the
undersurface of test plates was more than an order of
magnitude greater than that on the upper surface (Figure 1).
Settlement is inhibited in currents greater than 2 m/sec (6.6
ft/sec), and feeding slows in currents of 1—1.5 m/sec.
Textured substrates appear to provide a good surface for
attachment. The juveniles aggregate in cracks and corners
and adjacent to each other. Photographs of the various
juvenile stages are shown in Hopkins (1990).

Although attached zebra mussels are difficult to
displace, due to the strength of the byssal attachment, they
can voluntarily detach from the substrate and move around
using their muscular foot. Juveniles can also disperse by
drifting on mucus threads (Prezant and Chalermwat 1984)

4 Illinois Natural History Survey Biological Notes No. 138

Figure 2. Adult zebra mussels, showing range of shell patterns and
axes used for length and height measurement.

or by displacement during storms (see Appendix III). Thus,
it is not unusual to suddenly find adult mussels in an area
presumed to be free of mussels. Live mussels that have
accidentally or purposefully become detached will readily
reattach to the substrate by producing new byssal threads.
In dense colonies mussels are often anchored in place by
their neighbors.

Settled juveniles may grow 5—20 mm in their first year
of life, and under optimal conditions they can grow up to
0.21 mm/day (J. Nichols, personal communication). They
are highly efficient filter feeders, removing any particles
between 15 and 450 um from the water column (Sprung
and Rose 1977). Items that are not suitable for ingestion are
aggregated into a mucoid ball and ejected as pseudofeces.
Zebra mussel colonies can create sizeable accumulations of
humus due to the production of this pseudofecal matter.

A brief word on life cycle terminology is appropriate
here. The nomenclature for early life history stages of zebra
mussels is complex. Many terms, such as trocophore,
pediveliger, and plantigrade veliger, are primarily useful
for biologists who need exact descriptors of particular life
stages. For the general purposes of zebra mussel monitor-
ing (and ease of discussion in this document), the lite cycle
can be divided into three stages: veligers, settled juveniles,

Figure 3. Comparison of the shell shape of zebra mussels (below)
and quagga mussels (above).

and adults. Veligers are planktonic and smaller than
approximately 250 um. Settled juveniles, as the term
suggests, have settled onto a substrate, but they may still be
highly motile on the substrate. The term adult usually
denotes a reproductively mature individual; however,
because the reproductive status of a mussel is not readily
apparent without dissection, the term will be used here to
indicate a settled mussel easily identified with the naked
eye (1.e., longer than 2 mm).

2.3 Distribution

Veligers can be found at depths of 0 to 10 m, and their
maximum density occurs at depths of 3 to 7 m (10—23 ft)
along the perimeter of lakes (Mackie et al. 1989). The
juveniles tend to be clumped rather than randomly distrib-
uted through the water column. Veligers may undergo a
diurnal vertical migration, remaining in deeper water (8—10
m) during the day and ascending in the water column
during the night (Zhadanova and Gusynskaya 1985, Mackie
et al. 1989). Adult zebra mussels can be found as deep as
55 m (180 ft), but the depth of maximum abundance is
usually 2—4 m (6-13 ft; Mackie et al. 1989). Maximum
densities have occurred in some European lakes at depths
of up to 15 m (49 ft).

3 Sampling Veligers

3.1 Outline

¢ Sample at a minimum of three sites in open water at
least | km (0.6 miles) apart; sample at a minimum of
two sites in flowing water, one in center of current and
one near edge.

¢ Sampler: 63- or 64-ttm-mesh plankton net with 30-cm-
diameter opening, 1:3 diameter:length bias; or |2-liter
Schindler-Patalas trap.

« Method: open water—vertical tow from 3 m depth or
0.5 m above the bottom, or oblique tow starting at 3 m;
flowing water—pump or pour 200 liters (53 gallons)
through net, less in highly eutrophic waters.

¢ Sampling interval: once per week while water tempera-
ture is above 10°C.

¢ Preserve sample in 5% buffered sugar formalin or
ethanol (see Appendix IV).

* Count veligers in five subsamples from each sample.

* Report mean number of veligers/m* and variance.

¢ Ancillary data: water temperature at depth of sample,
Secchi disk depth, direction and rate of current.

¢ Optional data: water temperature at 3, 5, and 10 m;
adjacent substrate type; Ca** concentration; organic
carbon and chlorophyll concentrations; identification
and enumeration of other planktonic organisms in
samples.

3.2 General Comments
Veligers are microscopic, planktonic organisms. To
estimate their abundance in the water column, they must
first be concentrated by sieving a known volume of water
through a plankton net. The net mesh must be sufficiently
small to retain the veligers. The distribution of veligers can
be highly nonrandom in time and in space due to the
synchronous release of eggs within a colony (Stanczy-
kowska 1964 and references therein). Consequently, there
is a high probability that a single sample from a single
location will not contain veligers, even if veligers are
present in an adjacent area or were present one week
previously (Figure 4). The challenge is to determine the
optimal intervals of distance and sampling frequency that
will maximize the probability of finding veligers while
minimizing the sampling effort. Replicate samples must be
taken to estimate the variance in the veliger densities.
Otherwise, a single sample taken by chance within a clump
of veligers could yield a significant overestimation of
veliger abundance. A large number of small samples will
give better results than subsampling a few large samples.
To maximize the potential to detect veligers in a new
area, use of a pump or Clarke-Bumpus sampler to filter
large volumes of water from several depth strata is optimal.
Pumped samples are also the most quantitatively accurate
samples because of the ease of measuring the volume of

water that passes through the net. Plankton net flowmeters
are notoriously inaccurate (except for the very expensive
models). Time spent on quantifying densities and ensuring
replication, however, is less important than sampling large
volumes of water in multiple areas.

Veliger densities may range over several orders of
magnitude. When densities are low, large volumes of water
must be sampled to detect the veligers. One method to
sample large volumes is to pump water slowly through a
plankton net for several hours. When veliger densities are
high, other planktonic organisms will also be abundant. The
volume of water sampled should therefore be reduced to
facilitate counting the veligers and to avoid clogging the
plankton net. The sampling volumes recommended below
should be adequate for detecting and counting veligers until
densities become very high (>10,000/m‘’). If net clogging
becomes a problem or veliger densities are too high to
count without diluting the sample, reduce the sample
volume. Record the volume sampled to estimate veliger
density.

Remember that veligers clinging to plankton nets can
be readily transmitted to an uninfected body of water.
Disinfect all water-contact equipment when transferring
equipment between infested and noninfested sites. Disinfect
by rinsing with ethanol or formalin in a ventilated area or
by drying equipment thoroughly. A 5% chlorine rinse can
be used for most pieces of equipment but may harm the
Nitex mesh of a plankton net.

3.3 Sampling Period

Zebra mussels release their gametes (eggs and sperm) after
the water temperature has been maintained at 12°C (54°F)
for one to two weeks, though exceptions to this pattern
have been noted (Joe Leach, personal communication).
This temperature normally occurs in mid-May in the lower
Great Lakes but can vary considerably from year to year
and between different locations. Therefore, the water
temperature should be monitored starting in the early
spring. Sampling the water column for veligers is usually
unnecessary until the local water temperature has reached
10°C (50°F).

Sampling frequency depends upon the cost of each
sampling trip versus the need to have frequent estimates of
veliger densities. Veligers are present in the water for
approximately one month after each mass spawning event,
so the density of veligers over time at a given site probably
approximates a standard curve (Figure 4). Samples taken
every two weeks could miss the peak density and thus
underestimate maximum veliger counts. Again, decisions
about the frequency of sampling required must be based on
an assessment of information needs.

6 Illinois Natural History Survey Biological Notes

(x1,000)

Veligers/m?

Time (weeks)

Figure 4. Possible distribution of zebra mussel veligers over time
at a given location. Note that sampling once every two weeks
could miss the peak veliger density.

3.4 Equipment

¢ One plankton net with a 30-cm-diameter opening, 63-
or 64-um mesh (= size 25; 200 meshes per inch), and
1:3 bias. Attach a wide-mouth mason jar screw lid rim
into the end of the net using a hose clamp. Alterna-
tively, a mesh-lined plankton bucket can be used.
Attach small lead weights to the hose clamp to ensure
rapid sinking of the net. A 12-liter Schindler-Patalas
trap can be used for shallow water samples.

¢ (Optional) Electronic or digital flowmeter mounted in
the mouth of the net. If the net clogs at all during
sample collection, the flowmeter will give a better
estimate of how much water actually flowed through
the net than could be derived from calculations using
the length of the tow. However, only the expensive
flowmeters (>$200 U.S.) tend to be very reliable.

¢ One case of wide-mouth 1|-pint mason jars with lids.

¢ Eight liters (2 gallons) of 10% buffered sugar formalin
or ethanol.

¢ (Optional) Sieve made from a 250-ml plastic beaker
with the bottom cut off and replaced with 64-l1m
plankton net mesh glued across the bottom.

¢ Squirt bottle or water pump.

¢ Clipboard, pencils, standard forms, and sample labels
(see Appendix V).

¢ Thermometer or temperature probe.

¢ Secchi disk, 20 cm diameter.

3.5 Open Water

Sample Site
Sample sites should be chosen to maximally represent the
various strata that affect veliger distribution or the logistics
of sampling. Such strata include inshore areas, offshore
areas, inlets, outlets, sheltered bays, areas of high current,
areas with high boat traffic, etc. Sampling results from one
stratum cannot necessarily be used to infer veliger distribu-
tions in other strata. Maintaining the same sites in each
stratum is less important than sampling as many strata as
possible.

Where possible, nearshore sites should be at least 5 m
deep to avoid collecting sediment in the plankton net. Net

No. 138

clogging can also be avoided by sampling over hard
substrates or by sampling after a period of clear weather
when disturbed sediments have had time to settle.

The Schindler-Patalas plankton sampler is a good
alternative to a plankton net for shallow water samples.
Because it is more expensive, more delicate, and somewhat
more complicated to deploy than a plankton net, its use is
not described here. Investigators familiar with the
Schindler-Patalas sampler are encouraged to use it.

Vertical Plankton Tow

1. Prepare the net by screwing a wide-mouth 1-pint
mason jar or plankton bucket into the jar lid attached to
the end of the net.

Record the sample number, date, time, station identifi-

cation, and bottom depth on the sample form. Label

sample jars (see example, Appendix V).

3. Drop the net to 3 m depth or to 0.5 m above the
bottom, whichever is shallower. Retrieve the net by
pulling it vertically through the water column with a
steady, unhurried hand-over-hand motion. Retrieval to
the surface should take approximately 10 seconds (0.5
m/sec).

4. Calculate the volume of water sampled as follows:

N

volume sampled = rt X length of tow x radius of net opening?

5. Wash down the sides of the net from the outside (to
avoid adding additional organisms to the sample) to
wash any organisms on the netting into the jar.
Washing can be done with a water pump, hand bilge
pump, hand pumped fire hose, squirt bottle, or with
water in any small, clean container. When the net is
clean, carefully remove the jar.

6. Drain the jar to one-half full by pouring its contents
through a section of the plankton net mesh. Wash any
sample accumulated on the mesh into the jar using a
squirt bottle. Alternatively, use the sieve described
under “Equipment” by pouring excess sample through
the sieve. Flush the sieve into the sample jar.

7. Preserve the sample by filling the sample jar with 10%
buffered sugar formalin. Invert the jar several times to
ensure mixing. This will dilute the preservative to the
required 5%. Alternatively, add 95% ethanol to a one-
third full jar. (Note: Use of other preservatives such as
Lugol’s solution will damage the veligers and hinder
identification.)

8. After use, rinse the plankton net to remove any
remaining organisms and allow the net to dry to kill
any remaining veligers.

If observation of live specimens is required, divide the
fresh sample into two equal parts after shaking to evenly
distribute the contents in the water. Keep one subsample
cool to increase the longevity of the organisms, and
preserve the other subsample by adding buffered sugar
formalin or ethanol.

Oblique Plankton Tow
Large-volume sampling for presence/absence data, or
where veliger densities are extremely low, can be

April 1992

accomplished by doing multiple vertical plankton tows,
pumping water through a net (see below), or with an
oblique plankton tow. An oblique plankton tow consists of
setting the net at 3 m (or 0.5 m above the bottom if water
depth is less than 3 m) and pulling it to the surface while
towing it behind a boat. The boat speed must be slow
(approximately 2 knots). Problems that may be encountered
when doing an oblique tow include snagging the net on the
bottom, clogging the net with disturbed sediments, and
estimating the volume of water sampled. Measuring the
volume of water sampled requires a flowmeter; however, as
mentioned above, even a flowmeter may not accurately
measure the volume of water sampled. Even when veligers
are absent, other plankton may be abundant, and the net is
likely to clog rapidly. A horizontal tow, in which the net is
kept just below the surface, can also be used. The volume
of water sampled can then be calculated from the tow
distance, estimated from the boat speed.

Pumped Sample

Sampling with a pump is useful in shallow water and other
areas where disturbed sediments or plankton blooms may
clog a plankton net. Pumping is also the best method for
sampling in large rivers, where currents and high plankton
densities prohibit use of plankton tows. In strong currents, a
weight may be needed to hold the pump or pump hose at
the desired depth. By pouring or pumping water through
the net, the flow of water through the net can be controlled
to prevent overflow and consequent overestimation of the
volume of water sampled. Water-use facilities may choose
to pump a sample of water through a plankton net in
preference to deploying the net in a water intake. For on-
shore samples, the plankton net can be suspended by its rim
in a 55-gallon drum fitted with an outflow tube near the
bottom. Water is poured or hosed through the net until the
drum is filled to a known volume, and then the drum is
emptied through the outlet. This procedure can be repeated
as many times as needed until the desired volume has been
sampled.

Note that both centrifugal and diaphragm pumps can
be used for sampling veligers. Field data from several
sources indicate that even high-velocity/small-orifice
pumps do not appear to destroy a significant proportion of
veligers. Installing a screen, such as fiberglass window
screening, over the pump opening is advisable to prevent
the pump from clogging with debris.

Scouring Pad Samplers

These samplers, described by Martel in Appendix III,
provide an integrated plankton sample over time and thus
sample more water than can be accessed by a single
plankton tow. They are broadly applicable for zebra mussel
sampling but are especially useful in areas where severe
weather or vandalism are problems.

1. Assemble three samplers and deploy with the top
sampler at a depth of 3 m.

2. Retrieve sampler 24 hours after deployment. Place
sampler in water or preservative for transport to the
laboratory.

3. Repeat the procedure once per week, or as desired.

Marsden: Zebra Mussel Protocols

3.6 Flowing Water

Sample Site

Samples should be taken from at least two sites, one near-
shore and one in the open current. Slower currents near the
shore will be more conducive to veliger settlement,
whereas veligers in the center of the current will be carried
downstream. Take care to avoid areas where silt and debris
may clog the net. Mark the sites using landmarks or stakes
so that the same sites can be used repeatedly.

Collection Procedure
Use the procedures described under “Pumped Samples” or
“Scouring Pad Samplers,” above.

3.7 Water Intakes

Water intake samples will primarily be used by individuals
concerned with veligers infesting water works or water
cooling stations. When information on the efficacy of
control measures is required, set up two sampling stations,
one at the water intake upstream of the control initiation
point (e.g., chlorine injection) and one at a point within the
plant at which veligers must not be found.

Sampling can be done using any of the three methods
described above. A vertical plankton tow can be used in a
wet well or settling tank, a plankton net can be deployed
near an intake pipe so that water flows through the net, or
water can be pumped or poured through a suspended net.
Turbulence is a problem for within-plant sampling: care
must be taken to locate sampling devices to minimize
damage due to high flow. Vertical plankton tows are
frequently unfeasible because turbulence may drag the net
sideways, impeding estimation of the volume of water
passing through the net. Many plants have installed
bioboxes to monitor settling juvenile mussels. Results from
some bioboxes, however, have been equivocal: mussels
have settled elsewhere within a plant, but not in the biobox.
Bioboxes are clearly useful for straightforward monitoring,
but their efficiency should not be assumed under all
circumstances.

3.8 Collection of Ancillary Data
At each sample site:

1. Measure water temperature at the sample depth. In
turbulent water, the surface and sample depth tempera-
tures may be the same, in which case surface tempera-
tures can be used. Report water temperatures in
degrees Celsius.

Measure and record the Secchi disk depth. Attach a
Secchi disk to a line and lower it into the water until
the white quarters are no longer visible. The depth at
which the white disappears is the Secchi disk depth.
For accurate estimation of depth, the line must be
vertical when the measurement is taken; additional
weight may be necessary to hold the disk down ina
current.
Record the depth at which the sample was taken.
4. Record the direction and velocity (m/sec) of local
currents. In open water, this will vary according to
weather patterns. In intake pipes, current is equivalent

tre

Ww

8 Illinois Natural History Survey Biological Notes

to the intake velocity. For flowing water samples,
carefully measure the current at each of the sample
sites. If possible, also measure Ca** concentration,
dissolved oxygen, chlorophyll content of the water,
and organic carbon using American Public Health
Association (APHA) standard techniques (APHA
1989).

3.9 Counting Veligers

Veliger Identification
Veligers collected in the plankton tow will usually range
from 40 to 250 um, although mussels up to 2 mm long may
be found (Appendix III; Lois Deacon, Ontario Ministry of
Natural Resources, personal communication). Veligers can
generally be observed only by using a microscope with at
least 40x magnification, and preferably 50—100x magnifi-
cation. Photographs of veligers at various life stages are
shown in Hopkins (1990); a sketch of a veliger, as well as
of common plankton of similar size, is shown in Figure 5.
Post-trochophore larvae (= 60 um) can be readily identified
because they look like microscopic clams. Native clams,
except for the false dark mussel, have a glochidia larva,
which is not planktonic and has a readily distinguishable
flagellum or whiplike appendage protruding from the shell.
The only organisms that could be confused with zebra
mussel veligers are the veligers of the introduced Asian
clam, Corbicula fluminea, and the false dark mussel,
Mytilopsis leucophaeata. Corbicula are found as far north
as western Lake Erie and southern Lake Michigan (Scott-
Wasilk et al. 1983, White et al. 1984). The most northern
populations are usually associated with warm-water
effluents from power plants. In areas where the species
overlap, differentiation of Corbicula and D. polymorpha
veligers may be difficult or impossible; the primary

Bue

Scenedesmus
(Green alga)

Dreissena
veliger

Cyclops
(Copepod)

Pediastrum
(Green alga)

No. 138

distinction is the slightly flattened hinge in the Corbicula
veliger, which is larger than the D-shaped Dreissena
veliger (Figure 5). M. leucophaeata occurs in brackish,
estuarine waters and in the upper Mississippi and Hudson
rivers. Correct identification of veligers is painstaking and
requires extensive experience; basically, the three veliger
types should be considered indistinguishable.

Until veligers are seen and identified for the first time,
anxiety about misidentification can be great. The best
solution is to obtain a sample of known veligers from
someone familiar with zebra mussels; however, be aware
that supplying such samples can place a considerable
burden on the relatively few investigators and consulting
firms who make regular collections of veligers. Potentially
confusing components of the plankton can be eliminated
rapidly by a mental review of the following characteristics:
veligers lack legs, antennae, eye-spots, or stalklike append-
ages; they have a crisp outline (many rotifers seem to have
a fuzzy outline, even if their stalk is not visible); and a dark
line is usually apparent within the shell. To assure correct
identification, use a probe to turn the veliger on its side so
that the two shells are clearly apparent.

Cross-polarized light can be used to facilitate detection
and identification of veligers. Cross-polarization is accom-
plished by using a microscope with a polarizing filter above
and below the sample. One filter is rotated until the only
light passing through both filters is that which is refracted
by certain substances. Veligers appear irridescent under
cross-polarized light (E. Marsden, unpublished observa-
tions; Johnson 1992).

Newly hatched trochophore larvae may lose several
features such as the cilia and velum during preservation. If
identification of this brief stage is required, observation of
live veligers is advisable. Veligers can be kept alive for up
to a week if the sample is placed in an ice chest immedi-
ately after collection and is maintained at about 5°C.

a 200 itm

Re
ASS

Ceratium
(Dinoflagellate)

Dreissena
postveliger
Daphnia
\ (Cladoceran)
Corbicula
veliger

Figure 5. Comparison of zebra mussel veligers with other common planktonic organisms.

April 1992

Presence/Absence Data

Prior to the first detection of zebra mussel veligers in a
local area, portions of each sample can be pooled to look
for veligers. Large volumes of water can be scanned by
pouring 25 ml into a Petri dish and examining the dish
under a dissecting microscope. Once veligers are seen in
one of these pooled samples, go back and count the veligers
in each original sample from that date to obtain quantitative
data. To maximize the probability of detecting veligers in a
plankton sample, veligers can be concentrated in the
sample by either of two methods:

1. Add a few drops of ethanol to the fresh (unpreserved)
sample. This will slightly anesthetize the veligers,
which will sink to the bottom. A drop of water can
then be pipetted from the bottom of the sample jar and
placed on a microscope slide for observation. Other
narcotizing agents include magnesium chloride, MS-
222, CO,, and chlorotone.

2. Use Schaner’s sugar solution method described in
Appendix II.

These methods should only be used for rapid detection of
the presence of veligers; they should not be used if the
density of veligers is to be calculated.

Detection Limit

Your ability to detect veligers is limited by the volume of
water sampled, the volume of the concentrated sample, and
the number of subsamples examined. For example, if you
sampled 200 liters, concentrated the sample to 200 ml in
the sample jar, and then counted five 1-ml subsamples, then
the lowest detection limit (one veliger) would be

(1 veliger/total subsample vol.[ml]) x vol. conc. sample (ml)
vol. water sampled (ml)
_ (1 veliger/5 ml) x 200 ml

300 liters = 0.2 veligers/liter

If a lower detection limit is desired, as in areas where
imminent infestation is predicted, more subsamples should
be examined. For perspective, the first sighting of veligers
at one location in Ontario was at densities of 0.002 veligers
per liter (Lois Deacon, personal communication).

Veliger Densities

Veliger densities may vary by several orders of magnitude
among sample sites. Densities as high as 1,000,000/m‘
(1,000/liter) have been reported in Lake Erie (Joe Leach,
personal communication). A vertical tow sample from 3 m
in this density of veligers would collect

depth x 7 x radius of net mouth’= vol. of water sampled
3 mx 3.14 x (0.15 m)? = 0.212 m’

vol. of water sampled x veliger density = # veligers in jar
0.212 m? x (1,000,000 veligers/m*) = 212,000 veligers

If 1-ml subsamples are taken from 200 ml of concentrated
sample to count veligers, each will contain

# in sample/vol. of sample jar (ml) = #/unit vol.
212,000 veligers/200 ml = 1,060 veligers/m]

Marsden: Zebra Mussel Protocols

Counting this many animals under a microscope is not only
time consuming but also difficult to do accurately. An
efficient counting protocol will minimize the number of
organisms to be counted while also minimizing the
variance among samples. Counting precision can only be
estimated by counting replicate subsamples, then calculat-
ing the variance among the counts. To minimize variation
among subsamples, each subsample count should include
approximately 60 individuals. If there are extremely few
veligers in the sample, so that some |-ml subsamples have
no veligers at all, examine the total contents of 10 ml of
sample (= 10 Sedgewick-Rafter cells). The sample can be
further concentrated by sieving through 63-um mesh.

Equipment

¢ Stereomicroscope (dissecting microscope) with
magnification to at least 40x, preferably to 50—100x.

¢ Ocular micrometer for microscope if veliger size
measurements are required.

¢ Sedgewick-Rafter counting cell (Figure 6) or plankton
wheel.

¢ Disposable Pasteur pipettes and rubber suction bulbs;
calibrate pipettes by marking them at | ml volume.

¢ If available: magnetic stirrer, Hensen-Stempel pipette.

¢ Dissecting probes.

Counting Procedure for Plankton Tow

1. Mix the sample completely by swirling the jar or using
a magnetic stirring plate. Remove | ml of sample from
the center of the jar using a Pasteur pipette or Hensen-
Stempel wide-bore pipette. Place a cover glass
diagonally across the Sedgewick-Rafter cell. Fill the
cell slowly until the water is evenly in contact with the
cover slip. As the water comes in contact with the
cover slip, the slip will swivel until it covers the cell.
You may wish to dispense with the cover slip if you
need to manipulate organisms on the slide.

2. Examine the slide to familiarize yourself with common
organisms in the sample, and get a feel for scale and
for nonveliger shapes. At this point, evaluate whether
the sample needs to be concentrated or diluted. High
plankton densities may obscure your ability to see
veligers; low densities decrease your chances of
finding veligers.

3. Scan the Sedgewick-Rafter cell at 50x to detect
veligers. If necessary, verify veliger identification at
higher magnification.

Low veliger densities. Count the number of veligers
in the entire Sedgewick-Rafter cell by scanning evenly
back and forth over the cell. Count the cell at least
twice, or until the same total is reached twice.

High veliger densities. If the samples are crowded
with veligers and other planktonic organisms, you can
dilute the sample to reduce veliger densities (record the
dilution!). Counting large numbers of veligers in the
Sedgewick-Rafter cell can be simplified by placing a
grid under the cell so that the veligers can be counted
in smaller units. A reverse grid (white lines on black
background) is preferable because the pale veligers are
more highly visible against a dark background.

10 Illinois Natural History Survey Biological Notes

pipette

cover slip

SS MSSSTAN NSS

NS

ZX

Figure 6. Sedgewick-Rafter cell for counting veligers.

4. Repeat the procedure for a total of five samples from
each sample jar. Record each count separately on the
sample form.

Counting Procedure for Scouring Pad Sampler
1. Rinse each sampler three times by forcing a jet of

water (from a hose or spray bottle) through the pad,

working the spray across the entire pad. Rinse into a

shallow pan. Check efficiency of rinsing by noting

how many organisms are removed from the pad by a

fourth rinse. This check will become unnecessary with

practice.

Sieve the collected water, including liquid from the

container used for transportation, through a 63-um-

mesh sieve, until a workable volume of liquid is
achieved. Preserve with ethanol or buffered sugar
formalin, unless observation of live juveniles is
desired.

3. Subsample, if necessary, by serially diluting the
sample into equal volumes. Check that the split is
equal by examining the density of organisms in more
than one subsample. Place successive volumes of
sample in a Petri dish and count all the juveniles in the
dish using a dissecting microscope; repeat until the
entire sample or subsample has been processed.
Counting can be facilitated by placing graph paper
underneath the Petri dish and scanning rows of
squares.

4. If size measurements are needed, measure the length of
approximately 5O juveniles. Choose juveniles ran-
domly by selecting a subset of the Petri dish (e.g., the
right half) from which to measure juveniles.

. Report data as the number of juveniles collected per
cubic centimeter of pad per day deployed. For ex-
ample: 350 juveniles collected ina 10cm x 10cm x 5
mm pad in 24 hours = 350/50 cm*/day = 7/em*/day.

N

Nn

Calculation of Veliger Densities
Determine the number of veligers per liter of lake water as
follows:

veliger density = # in sample jar/vol. of water sampled,

where # in sample jar = (#/unit vol.) x vol. of sample

The number of veligers per milliliter of sample (number per
unit volume) is the number counted in a Sedgewick-Rafter

No. 138

cell, multiplied (if necessary) by the number of dilutions
used; for example, multiply by 2 if the sample was diluted
by one-half.

The volume of lake water sampled during a vertical
tow is calculated as follows:

depth of tow x m x radius of net mouth? = vol. of water
sampled
3 mx 3.14 x (0.15 m)? = 0.212 m°

Calculate the mean number of veligers per cubic meter by
averaging over the five samples. Calculate the variance
among the samples using the following equation:

X(x-x)?
n—-|

where x = number of veligers in each of the five samples;
x = the mean number of veligers in the five samples; and n =
the number of samples = 5

Some people prefer to report veliger densities as number
per liter. To convert from veligers per cubic meter to
veligers per liter, multiply the number of veligers by 0.001.

4 Sampling Settling Juveniles

4.1 Outline

¢ Locate sample at sites of interest.

¢ Sampling plates: 15-cm-square PVC plates, slide rack
containing microscope slides, or multiplate sampler.

* Deploy settlement plates, slide rack, or multiplate
sampler at 3 m. Retrieve and replace plate, slides, or
multiplate sampler at each sampling interval.

* Sampling interval: once per week while water tempera-
ture is above 10°C.

¢ Preserve slides and plates in buffered sugar formalin or
ethanol.

¢ Scrape one side of each plate clean with razor; count
settled juveniles in 1-cm squares on each plate until
>60 juveniles have been counted, or the entire surface
of the plate has been examined.

¢ Calculate surface area counted on each plate; report
data as number of juveniles per square meter and
volume of juveniles per square meter.

¢ Ancillary data: Water temperature at surface, Secchi
disk depth, direction and rate of current.

¢ Optional data: Water temperature at 3, 5, and 10 m;
adjacent substrate type; Ca** concentration; organic
carbon and chlorophyll concentrations; identification
and enumeration of other organisms settled on slides.

4.2 General Comments

Juvenile settlement is affected by substrate type, substrate
texture, depth, light, water currents, proximity to other
mussels or adjacent surfaces, and ionic concentrations.
Figure | gives an indication of how dramatically settling
densities can be affected by substrate material. Settling
veligers select substrate both during and after settlement.
The settled juveniles can crawl up to 3.8 cm/hr for several
days before making a permanent attachment to the substrate
(Lewandowski 1982). They may be seeking a better surface
than the one on which they originally settled. Because
juveniles may move off a sampling plate after settlement,
the numbers of juveniles estimated may be affected by the
period of plate immersion. Juvenile counts will also be
affected by the following factors:

* Sampling plate orientation. If the plate is horizontal,
different numbers of veligers will settle on the upper
and lower surfaces, due to light avoidance. Juvenile
settlement will also be different on horizontal versus
vertical surfaces.

* Edge effects. Juveniles appear to be thigmotaxic: they
aggregate near corners or each other.

¢ Turbulence. Juvenile settlement appears to be inhibited
in areas of high turbulence.

* Biofilm. Juveniles tend not to settle on “fresh” surfaces
that have not acquired a microscopic biofilm; some
species of algae and diatoms may also inhibit settle-
ment (C. Brousseau, Ontario Ministry of Natural
Resources, personal communication).

For presence/absence sampling, the sampling substrate
should be one that is favored by settling mussels. Juveniles
prefer to settle on horizontal, shaded surfaces with a lot of
surface irregularity—that is, corners and crevices. Rough-
ened surfaces and PVC are also favored by settling mus-
sels. One of the simplest techniques for monitoring is to
suspend concrete blocks in the water; although the earliest
settlement stages will not be noted against the coarse
texture of the block, larger juveniles can be seen on the
block and on the line that holds it.

Microscope slides are optimal standard settling plates
because (1) they are readily available, (2) they are uniform
from source to source, (3) they are inexpensive and
disposable, (4) juveniles are readily observed against a
smooth, uninterrupted background, and (5) slides can be
examined under a microscope if detection of the smallest
settled juveniles is important. Lewandowski (1982), D.
Garton (Ohio State University, personal communication),
and others report good settlement of juveniles on slides.
Microscope slides are also an accepted standard for
limnological studies involving settlement (Lind 1979,
APHA 1989). The smallness of microscope slides relative
to other samplers does not affect the settlement of juve-
niles; the surface available to examine is merely reduced.
Slides do present difficulties, however, because of their
fragility; this can be overcome by combining sampling
methods, as suggested under “PVC Plates,” below.

Whatever type of settlement plate is used, vertical
deployment is preferable to horizontal for several reasons:
(1) sediment and organic matter will settle on the upper
surface of horizontal plates, making it difficult to discern
settled juveniles; (2) use of horizontal surfaces adds the
onus of having to mark which surface was on top, and then
counting both surfaces; (3) juveniles that land on the upper
surface of a plate tend to migrate to the lower surface
(Lewandowski 1982), so the counts on the upper and lower
surfaces must somehow be integrated; (4) horizontal plates
will tend to “kite” in currents, thus changing the depth of
the sampler and increasing abrasion on the lines. Fewer
Corbicula juveniles may be found on vertical surfaces
because they generally secrete only a few byssal threads for
attachment and may fall off these surfaces.

Of the various methods and substrates that have been
used to sample settling juveniles, three are described here
because they fit the criteria of ease of construction, ease of
use, and applicability in diverse field situations. Juvenile
settlement on any of the samplers is enhanced if the plates
have been “prepared” by immersion in fresh water for one
to two weeks, during which time the plates become
colonized by a microscopic layer of biological material.
Obviously, the longer the samplers are immersed, the larger
the juveniles that will have settled on the plates. The
immersion time should be chosen to balance how rapidly

12 Illinois Natural History Survey Biological Notes

information is needed with the increased ease of seeing and
counting larger juveniles.

4.3 Sample Site

Choose sites based on the comments on strata in section
3.5. Avoid sites near chemical discharge points, which
could adversely affect mussel settlement. Also avoid sites
in the path of boat traffic or near popular fishing areas,
where samplers could be inadvertently snagged. Label each
sampler, as needed, with your agency’s name, a sign stating
“Scientific Equipment—Do Not Disturb,” and a contact
number or address in case the sampler is inadvertently
damaged or washed ashore.

Deployment of artificial substrates can be the most
difficult part of zebra mussel sampling because the appara-
tus may be vulnerable to storms, currents, turbulence, and
vandalism. If you are monitoring in the open waters of the
Great Lakes, or in areas of high human activity, expect to
lose several samplers, no matter how well you have
anchored or protected them. Losses may be reduced if you
follow the following suggestions:

¢ Choose a protected site, such as a bay.

¢ Deploy replicate sets at each site whenever possible.

¢ Useat least '/4-in. line for buoys, anchor attachments.

¢ Wherever line may chafe, such as where it runs
through a buoy ring or where it is attached to a cinder
block anchor, protect the line with a rubber sleeve such
as garden hose or Tygon tubing, or use chain.

¢ Use twice as much weight for an anchor as seems

necessary. Three cinder blocks chained together is a

minimal anchor for areas where waves may reach

1-2 m high.

main buo

= —— —_— ————— _

satellite buoy

settlement
samplers

anchor

Figure 7. Suggested methods for deployment of settlement plates.

No. 138

« Maximize the scope on the buoy (1.e., length of line
relative to water depth). Use at least a 5:1 to 7:1 ratio
of line to depth.

¢ Deploy a lighter, secondary anchor between the
sampler and the main anchor, approximately 3 m
(10 ft) from the main anchor (Figure 7). This light
anchor acts as a shock absorber to reduce sudden shock
loads on the main anchor and buoy attachment.
Alternatively, use a satellite buoy (Figure 7).

Near-shore monitoring can take advantage of fixed struc-
tures such as piers, docks, and buoys (if permission is
sought first!) to attach sampling equipment. Remember,
however, that shore access increases the probability that
your equipment will be vandalized. Samplers can be
deployed under a submerged buoy to avoid wave and
human damage. During deployment, run a line from the
anchor to shore; during retrieval, the line can be followed
out from shore to the sampler using a gaff to hold the line.
In areas where water levels may vary considerably due to
floods, water draw-downs, etc., use of fixed structures to
deploy samplers is inadvisable because settling juveniles
will not tolerate repeated drying periods.

4.4 Equipment
¢ Settlement sampler of choice.
¢ 2.5 x 7.5 cm microscope slides.
¢ Indelible pen or paint for labeling slides.
¢ 2-3 liters 5% buffered sugar formalin or ethanol.
¢ Ropes, anchors, and buoys to suspend slide rack at
desired depth.

Z
i 4) buoy

3m
settlement
samplers
main anchor
satellite
anchor

~

April 1992

¢ Razor for scraping slides.
* Clipboard, pencils, standard forms, and sample labels
(Appendix V).

4.5 Collection Procedure

PVC Plates

1. Attach three 15-cm-square PVC plates in series (see
Figure 8). (Note: Plexiglas, which is often used for
settlement plates, has two major disadvantages: it
dissolves slightly in ethanol, and its transparency may
discourage juvenile settlement. Gray PVC is preferable
to white because of the preference of zebra mussel
juveniles for dark surfaces.) Attach a microscope slide
to each plate using a binder clip or similar device.

2. Deploy the plates so that the upper plate is suspended
at a depth of 3 m. Deploy the plates at least one week
before you anticipate settling may begin, in order to
condition the plates.

3. Once per week, or at longer, regular intervals deter-
mined by your sampling plan, remove the bottom plate
and put a fresh plate between the remaining two plates.
Thus, each plate will have been in place for two weeks
(one week for conditioning) before retrieval, except for
the first plate retrieved each season. Place the retrieved
plate in a container with buffered sugar formalin or
ethanol, such that the sides of the plate are protected
from accidental scraping. A plastic food container in
which the plate is supported diagonally works well.
Handle the plates by the sides to avoid damaging
settled juveniles. If observation of live juveniles and
other settled animals is desired, place the plate in water
for transportation and keep it cool. Examine “fresh”
plates within a day of collection to avoid decomposi-
tion of the settled animals. On the next sampling date,
the new bottom plate is retrieved, having been in place
for two weeks. At the end of the sampling period,
remove the top plate to estimate the seasonal accumu-
lation of juvenile mussels.

Slide Rack

Periphyton sampling racks that hold a number of micro-
scope slides can be purchased ready-made (Figure 9; see
Appendix IV). The interval between sampling dates should
be determined by access to the sample site. Weekly
sampling, suggested here, is ideal but may not be practical
at distant sites. Most important is to sample at regular
intervals.

1. Label microscope slides using an indelible marker.
Place slides into the rack.

2. Deploy the rack so that it is suspended at a depth of 3
or 5 m. A choice of depths is given because one or the
other depth may be inaccessible at some sites. Ideally,
suspend a rack at each depth from the same buoy-
anchor line. Deploy the rack at least two weeks before
settling may begin, in order to condition the slides.
Place the rack so that the slides are vertical.

3. Once per week remove five slides (every alternate
slide) from the rack and place them in a slide box.

Marsden: Zebra Mussel Protocols 13

3 m to surface

removed at

PVC plate
end of season

cable tie

moved down
after 2 weeks

removed after
2 weeks

microscope slide

Figure 8. Vertical PVC plates for collecting settling zebra mussel
juveniles.

Immerse the slide box in 70% ethanol or buffered
sugar formalin. Handle the slides by the sides to avoid
damaging settled mussels. Place five new slides in the
rack. On the next sampling date, remove the next set of
slides. This way, each slide will be in place for two
weeks before collection.

4. Once per month, remove two slides that have been in
place since the rack was deployed. These slides permit
observation of long-term settlement and will not be
replaced.

5. Prior to examining the slides, scrape one side clean
with a razor blade. Avoid “selecting” which side to
clean based on the number of mussels observed. For
example, always scrape the unlabeled side.

Multiplate Samplers

Multiplate samplers (Figure 10) have the advantage that,
like microscope slides, they are traditionally used for
periphyton sampling. They also can be designed to conve-
niently fit into a mason jar for preservation and shipping.
Their major disadvantages are the horizontal orientation of
the plates and the need to retrieve an entire sampler for
disassembly to collect data. Multiplate samplers are
commonly constructed of tempered hardboard, but they can
also be manufactured from PVC. The design suggested by
the U.S. Fish and Wildlife Service (William Mason,
personal communication) consists of eight PVC and two
hardboard disks, 7.5 cm in diameter and 0.3 cm thick
(Figure 10). The disks are held together on a 17-cm length
of threaded rod. The spacers can be cut lengths of small-
diameter PVC tubing or stacks of plastic or stainless steel
washers. A turnbuckle is then screwed onto either end of
the threaded rod to provide attachment points for deploy-
ment lines.

1. Deploy two multiplate samplers at a depth of 3 m at
each site. Deploy the samplers at least two weeks
before you anticipate settling may begin, in order to
condition the plates.

14 Illinois Natural History Survey Biological Notes

microscope
eyebolts slides
0) for harness ©)

wood _ ie
or plastic ,
frame

SSS
NN
NN

vie
Cia

3cm

Figure 9. Slide rack for collecting settling zebra mussel juveniles.

2. Retrieve one sampler every two weeks, replacing it
with a new sampler; or use longer, regular intervals as
dictated by your sampling plan.

3. Remove turnbuckles, then insert entire sampler into a
mason jar and fill with 70% ethanol or 5% buffered
sugar formalin.

Loss of some settled juveniles from the plates after retrieval
is inevitable and difficult to quantify. Juveniles that have
settled on a surface but not yet attached themselves with
byssal threads are especially vulnerable to displacement.
Juveniles that have fallen off inside the container used to
transport slides or settlement plates can be counted by
sieving the preservative through plankton netting. The
mesh size should be slightly smaller than the smallest
juveniles you wish to count. If the proportion of “lost”
juveniles to total juveniles is consistent, a figure can be
obtained and used subsequently to estimate juvenile loss.
Juveniles lost during plate retrieval may be impossible to
estimate unless the sampling plates are placed in a bag in
situ before retrieval. Whether the effort is worth the result
depends, again, upon the information needed.

4.6 Counting Procedure

When counting juveniles settled on plates, do not count
near the edges of the plates where mussels are likely to
have been dislodged during handling. Once the settled
juveniles have begun to develop pigment and assume their
adult shape, the only other organisms likely to be confused
with them are ostracods. In live samples, ostracods are
readily distinguished when they scuttle rapidly across the
settlement plate; dead ostracods often fall off settlement
plates. Ostracods are bean-shaped, often have fine hairs on
the shell, and vary from 0.5 to 3 mm. Close examination
and manipulation with a probe will reveal the presence of
leglike appendages or antennae.

No. 138

Slides
1. In the laboratory, gently scrape one surface of each
slide clean using a razor blade. Place this side down on
the dissecting microscope stage. Place a piece of graph
paper with |-cm? divisions under the slide.

. Scan the plate or slide at 30—-40x to find settled
juveniles; confirm identification by turning the
organism on its edge so that the two shells can be seen.
Encrusted algae and other organisms may need to be
probed and teased apart to detect newly settled zebra
mussels. Newly settled juveniles will look like
planktonic veligers; they will be white and approxi-
mately round, with a distinct umbo. As the settled
juveniles grow, they will begin to elongate into the
adult shape. Dark stripes generally begin to appear on
the shell after the shell has begun to elongate.

3. Count the juveniles in each of five 1-cm squares that
were previously marked on the graph paper to random-
ize the counts. Include juveniles on the upper and
right-hand lines; ignore those lying on the lower and
left-hand lines. Record the total number of juveniles in
all five squares; repeat the count for the other micro-
scope slides. Avoid the area that was covered by the
clip. If juvenile densities are very high, count only as
many squares as needed to reach a total of approxi-
mately 60 animals. Record the number of squares
counted. If juvenile densities are very low, count the
entire surface of the plate.

4. If a microscope is unavailable, count juveniles in five
1- cm squares using the naked eye. Use of a dark
background, divided into a metric grid, will assist the
count.

5. If size measurements are required, determine what size
square will contain approximately 50 animals. This
may be one |-cm square, or it may be five l-cm
squares on each of two slides, depending on how
densely the veligers have settled. Measure the length of
all animals in the selected area using an ocular
micrometer.

6. Do not reuse the slides. Even if the slides are scraped
clean, their surface will be different (to a juvenile) than
that of a fresh slide.

in)

Plates

The plates can be processed in the same manner as the
slides, except that a grid cannot be used beneath the opaque
PVC. The surface of the PVC can be subdivided for
counting by lightly drawing lines through the accumulated
biofilm (scratches in the PVC will affect subsequent
settlement of the juveniles!). Alternatively, 1-cm strips
across the plate can be isolated by scraping away material
on either side; juveniles within the strip can then be
counted in 1-cm blocks. Clean plates by scraping with a
razor blade, then washing. PVC plates can be reused, but

April 1992

turnbuckle
or eyebolt

hardboard
0.3 cm spacing a

0.6 cm spacing

PVC
disks

7.5cm
diameter

0.9 cm spacing

1.2 cm spacing

EES
_—_
ee hardboard

Figure 10. Multiplate sampler for collecting settling zebra mussel
juveniles.

standardization of the texture of these plates is probably
impossible, especially because the composition of the
plastic itself varies slightly among manufacturers.

4.7 Collection of Ancillary Data
At each sample site:

1. Measure water temperature at the sample depth. In
turbulent water, the surface and sample depth tempera-
tures may be the same, in which case surface tempera-
tures can be used. Report water temperatures in
degrees Celsius.

Measure and record the Secchi disk depth.

Record the depth at which the sample was taken. If
possible, also measure Ca** concentration, chloropkiyll
content of the water, and total organic carbon using
APHA standard methods (APHA 1989). In the
laboratory, you can identify and count other settled
organisms on the slides.

why

Marsden: Zebra Mussel Protocols 15

4.8 Data Reporting: Biomass vs. Density

Reports of high densities of settled juveniles or adult
mussels are impressive and attractive to the popular press,
but they are not very meaningful unless accompanied by
size measurements. For example, 100,000 newly settled
mussels per square meter may only constitute a monolayer
1—2 mm thick, whereas a similar density of 2- to 3-cm
adults would form an encrustation over 10 cm thick.
Unfortunately, acquisition of accurate biomass data
requires time and specialized equipment and is beyond the
scope of most monitoring programs. Details of dry-weight,
wet-weight, and live-weight biomass measurements are
covered by McCauley (1984) and need not be repeated
here. Length and volume measurements may be the most
accessible data with which to estimate biomass for most
investigators. Ideally, a random subset of 50-100 animals
of each age cohort from each sampling location and date
should be measured. Minimally, the largest and smallest
mussels should be measured, and the median size should be
estimated by examination of the whole group. The relation-
ship between free dry weight and shell length can be
estimated using the equation of Bij de Vaate (1991);
however, this equation requires calibration with field data.
A more useful figure for general comparisons is the volume
of mussels per unit area. Volume can be estimated using
the displacement method: place the mussels from a known
area in a beaker, fill the beaker with water to just above the
mussels, then subtract the volume of water added from the
total volume in the beaker. Report data as milliliters of
veligers per square meter.

5 Sampling Adults

5.1 Outline

¢ Locate sample sites in areas of interest.

¢ Sampler: 1-m quadrat square divided into 10-cm
squares, or smaller quadrat square; coring device.

¢ Sampling interval: at investigator’s discretion.

* Quadrat method: for multilayer colonies, select a
random location, take a core sample of known area,
count all adults within core; for low-density colonies,
count mussels in 10-cm-square areas until at least 60
have been counted.

¢ Grab method: use ponar grab to collect three replicate
samples of bottom substrate.

¢ Estimate percent cover by zebra mussels within local
area.

e Preserve adults in ethanol.

¢ Report data as number of mussels per square meter and
volume of mussels per square meter.

¢ Ancillary data: water temperature at surface and
sampling depth, Secchi disk depth, direction and rate
of current, description of substrate type.

¢ Optional data: Ca** concentration, organic carbon and
chlorophyll concentrations, size of adults in 5-mm
intervals.

5.2 General Comments

Sampling of adult zebra mussels provides information on
their settlement and growth rates in a local area. The
sampling protocol is intended for enumeration of animals
large enough to be visible to the naked eye. Adults can also
be counted on substrates such as smooth tiles that have
been left in the water for several months. The substrate on
which the adults are counted must be noted. Substrate
material and texture will affect adult densities. Material
texture will also influence accuracy of counting the
smallest individuals, which may be lost against a high-
relief background.

Sampling adult zebra mussels in situ, on natural
substrates, usually requires scuba divers. Fixed sites such as
water intake structures can only be sampled in situ. Mussel
densities on natural substrates can be measured using a
ponar grab, but ponar grabs do not sample bedrock or large
cobble substrates well because the grab cannot close over
large rocks. Sampling adults on natural substrates may not
provide data that are comparable from site to site, due to
differences in substrate type. For many purposes, use of
artificial substrates will be simpler than sampling on natural
substrates. For example, adult densities can be measured on
concrete blocks or multiplate samplers placed in the water
in the early spring. This measurement provides a good esti-
mate of adult densities on nearby concrete structures; it also
allows comparison of densities between different bodies of
water that may not have similar natural substrates.

Much data can be acquired with the assistance of local
dive clubs, which are often more than willing to find an
excuse to dive. In large river systems, commercial shell
fishermen have greater and more frequent access to the
bottom substrates than anyone else and are therefore a
valuable source of data. As noted in the introduction, these
resources are invaluable for extensive monitoring.

5.3 Sample Site

Choice of sites to calculate adult densities will depend upon
the questions of interest to the investigator. Note that adult
densities will be dramatically affected by sampling loca-
tion. Adult mussels can only settle on hard substrate and
tend to prefer dark areas, corners, crevices, and other zebra
mussels. These factors must be noted to avoid bias when
sampling. As with sampling of veligers and juveniles, the
various types of substrate in the sampling area should each
be sampled to avoid bias. A report of adult zebra mussel
densities must be accompanied by an estimate of the area
represented by the measurement. For example, if you
counted an equivalent of 10,000 zebra mussels on a unionid
mussel isolated on a muddy bottom, report the average
number of unionid mussels per square meter. Artificial
substrates such as concrete blocks should be deployed in
areas where they are unlikely to be interfered with by
curious passersby or vandals and where currents or
turbulence will not result in loss of the substrates.

5.4 Equipment

¢ Small (15-cm-square) ponar grab and a no. 30 sieve; or
a quadrat frame.

¢ Quadrat frame. This can be purchased ready-made or
can be easily constructed. The size of the quadrat
square depends on the population densities of mussels.
A |-m square may be necessary to count reasonable
numbers of mussels in areas of very low density (<100/
m?), whereas a 10-cm square may be sufficient in high-
density areas. A quadrat square for use by divers can
be constructed of 1-in. PVC tubing and right-angle
connectors. Drill small holes in the frame to release
trapped air so that the frame will be negatively
buoyant. On a l-m quadrat square, mark the edges at
10-cm intervals. String wire or line across the quadrat,
tied at opposite edges, to subdivide the frame into
10-cm squares.

* “Coring” device 8-12 cm in diameter. This is not a
true corer but a device for outlining a discrete,
measureable area of hard substrate in which mussels
can be counted. The device can be constructed from an
aluminum food can opened at both ends. Cut the rim
off one end to leave a sharp edge. A dowel passed
through two holes at the upper end of the can will

April 1992

5.5

serve as a handle. Measure and record the opening
diameter.

Fine-mesh bags of material such as cheesecloth
(pantyhose material has also been used effectively).
Large zip-lock bags.

2-3 liters of ethanol.

Razor blades (the type used in paint scraping, mounted
in a handle or having one blunted edge) and paint
scrapers.

Hammer.

Collection Procedures Using Quadrat Frame

Procedure | (high adult densities, >10,000/m? or multi-
layer colonies)

1.

Randomly place the coring device on the surface to be
sampled. Decide on your method for randomizing
before sampling to avoid bias. For example, pick a
square in the quadrat frame, drop the frame so that it
lands at random, then place the corer within the
selected square. Using whatever force is necessary
(this is where the hammer may come in handy), push
the device into the colony of mussels until its edges are
firmly in contact with the substrate.

Using a razor blade or paint scraper, remove mussels
from around the device until there are none within
several centimeters of the device.

Remove the device. If there is more than a single layer
of mussels, measure the height of the isolated plug of
mussels in centimeters. Carefully, to avoid damaging
the animals, use a razor blade to scrape the mussels
into a mesh bag. At the surface, transfer the sample to
a zip-lock bag and add ethanol to preserve the mussels.

. Repeat the procedure until five samples have been

collected in separate bags. This procedure can also be
used to estimate densities of mussels on introduced
substrates such as tiles or on rocks brought to the
surface by divers.

Procedure 2 (low adult densities, <10,000/m? or single-
layer colonies)

1.

Place the 1-m quadrat frame against the substrate to be
sampled. Minimize inadvertent “selection” of the area
to be sampled, which will result in biased data. This
can be accomplished by dropping or throwing the
frame onto a horizontal substrate so that it lands at
random. For vertical substrates, one method is to keep
your eyes closed, move parallel with the vertical
surface, then place the quadrat frame against the
surface. Do not decide to reposition the frame because
there are “too few” mussels within it; replication will
sample both high and low densities to give a represen-
tative average and variance.

. Estimate how many 10-cm squares you will need to

count to find a total of approximately 50 mussels.
Select the required number of squares randomly before
placing the quadrat on the substrate. For example,
number the squares from | to 100, then use a random
number table to select the squares by number.

Marsden: Zebra Mussel Protocols 17

. Count the number of mussels in each of the selected

10-cm squares.

. Remove the mussels from each of the selected squares

using a razor blade; place the mussels in a zip-lock bag
and cover them with ethanol.

. Make five replicate counts (i.e., counts from five

placements of the quadrat square, with the same
number of small squares examined in each).

Calculate the total area counted in each |-m quadrat
square (= number of small squares counted x 100 cm’).
Calculate the adult density as follows:

# of mussels counted x 10,000

adults/m? = =
area counted (cm?)

Procedure 3 (low densities, loose, irregular substrate)

Ile

2

5.6

Sel

Place the 1-m quadrat square against the substrate as
described above.

. Estimate how many 10-cm squares you will need to

count to find a total of approximately 50 mussels.
Randomly select the required number of squares before
placing the quadrat on the substrate. For example,
number the squares from | to 100, then use a random
number table to select the squares by number.

. Within the selected squares, remove rocks or other

particles to which mussels are attached. Remove only
those particles that have mussels attached and that
have 50% or more of their mass within the sample
square. Place all material collected into a zip-lock bag
and add ethanol to cover the sample.

Collection Procedures Using a Ponar Grab

. Select three sampling sites at approximately 3 m depth

that are representative of the area to be sampled.
Lower the cocked (open) grab into the water until it
touches the bottom, close the grab by releasing the
messenger or trip line, then bring the grab to the
surface. Place a bucket underneath the grab as it leaves
the surface to collect small organisms that may be lost
as the water runs out. If the grab is brought to the
surface partially open because a hard object has
jammed in the opening, discard the sample and take
another grab.

Place the full grab into the bucket and wash out the
contents using a bucket of water. Slosh the grab in the
water until clean. Be careful: a grab in the cocked
position can be dangerous if it shuts accidentally.

Pour the sample into a no. 30 sieve and wash it using a
twisting, sloshing motion in the water. Do not let water
slosh over the rim of the sieve.

. Preserve the cleaned material that remains in the sieve

in ethanol or buffered sugar formalin.

Collection of Ancillary Data

At each sample site:

to

Measure water temperature at the surface and the
collection depth. If a temperature probe is available,
measure water temperature at 3, 5, and 10 m. Report
water temperatures in degrees Celsius.

Measure and record the Secchi disk depth.

eS)

5.8

N

Illinois Natural History Survey Biological Notes No. 138

Record the depth at which the sample was taken. If the
sample was taken in an intake pipe, record the depth
and location of the inlet opening, as well as where in
the pipe (relative to the opening) the sample was taken.

. Record the type of substrate sampled—for example,

broken limestone bedrock, vertical concrete pilings,
silty bottom with native mussels, etc.—and the type of
sampler used. The substrate can be quantitatively
described by using the quadrat frame. Place the frame
randomly, then record the most common substrate type
(>50% of the area) in each square. Flip the frame so
that it rests on a new area, then repeat the data record-
ing. Do this for five frame counts.

Estimate the proportion of the area of interest that is
covered by mussels. This can be done in either of two
ways:

Random distribution of mussels. Place quadrat
square randomly (see above) on the substrate, then
record which squares are more than half filled with
mussels. The percent coverage of mussels is then
equivalent to the proportion of squares counted as
more than half full. For example, if 43 of the 100
squares were more than half full of mussels, then 43%
of the area inside the quadrat square was covered with
mussels. Repeat this for a total of five quadrat square
counts.

Clumped distribution of mussels. Measure five
clumps as follows. Measure the longest axis of the
clump, then the length of an axis at right angles to the
long axis. Measure the average distance between
clumps, or the average number of clumps per square
meter.

If possible, also measure Ca** concentration, dissolved
oxygen, chlorophyll content of the water, and organic
carbon using APHA standard methods (APHA 1989).

Counting Procedure

In the laboratory, carefully separate the mussels from
each other and count them. Discard all mussels broken
during the collection process and any dead mussels
(i.e., empty shells). Carefully examine the shells of
each mussel and remove smaller attached mussels.
Record the size of the smallest mussel that you count.
Measure the length of approximately 200 individuals
as illustrated in Figure 2. To select the individuals for
length measurement, subsample by gently breaking the
colony apart before separating individual mussels, so a
random subsample is achieved. A dissection micro-
scope with 10-40x magnification may be useful for
this process.

If possible, count the number of individuals that were
alive during sampling. Dead mussels will tend to have
open, empty shells. Because some closed shells may be
empty, each shell needs to be opened and examined.

If possible, record the wet weight of the sample in
grams.

Calculate the density of the mussels:

# counted x 10,000

density (N/m?) = -
area counted (cm*)

5. Estimate the volume of the mussels in a unit area by
displacement. Place the mussels in a beaker, add water
to just above the mussels, then subtract the volume of
water added from the total volume in the beaker.
Calculate the volume of mussels per unit area as
follows:

volume of mussels (ml)

volume (liters/m*) = 5
area counted (cm*)

Note: The counts from each square within the quadrat
frame are summed to obtain a total, not an average.
Replication is achieved by repetitive placements of the
quadrat frame, not by multiple counts of squares within the
frame.

6 Reporting Results

Several extensive monitoring programs are already under
way in the Great Lakes. These programs provide regional
coordination of sample collection and analysis, with
dissemination of information and sampling results. Results
of independent monitoring can be reported to the project
leader of the regional monitoring program. The U.S. Fish
and Wildlife Service has been involved with tracking the

distribution and status of nonindigenous aquatic species
since 1977. The USF&WS database has a national scope
and will be very useful in tracking zebra mussel infestation
nationwide. The USF& WS reporting form is shown on
page 21. Reporting new zebra mussel sightings is highly
encouraged, as are summaries of seasonal monitoring. A
generalized reporting form is shown on the next page.

State and National Zebra Mussel Information Centers

National U.S. Fish and Wildlife Service New York Charles O’Neill, Jr.
Geographic Information Section New York Sea Grant
National Fisheries Research Center 250 Hartwell Hall
7920 N.W. 7\st Street SUNY College at Brockport
Gainesville, FL 32606 Brockport, NY 14420-2928
904-378-8181 716-395-2638
FAX: 904-378-4956 FAX: 716-395-2466

Illinois, Joseph O’ Leary Ohio Maran Brainard

Indiana I}linois-Indiana Sea Grant Ohio Sea Grant College Program
Purdue University The Ohio State University
Forestry Building 1314 Kinnear Rd.
West Lafayette, IN 47907 Columbus, OH 43212
317-494-0409 614-292-8949

Michigan John Schwartz Ontario Chris Brousseau, Coordinator
Michigan Sea Grant Zebra Mussel Coordination Office
Michigan State University Ontario Ministry of Natural Resources
334 Natural Resources Building P. O. Box 500, Maple, Ontario L6A 1S9
East Lansing, MI 48824 416-832-7275
517-353-9568 FAX: 416-832-7177

Minnesota Jeff Gunderson Wisconsin Clifford Kraft

Minnesota Sea Grant College Program
University of Minnesota-Duluth

208 Washburn Hall

Duluth, MN 55812

218-726-8106

Wisconsin Sea Grant Program
Bldg. ES-105

University of Wisconsin, Green Bay
Green Bay, WI 54311-7001
414-465-2795

20 Illinois Natural History Survey Biological Notes No. 138

Zebra Mussel Sighting Report Form

What Are Zebra Mussels?

Zebra mussels are small brown-and-white-—striped clamlike
animals native to Eastern Europe. They were discovered in
the Great Lakes in 1988 and were probably distributed
from a freighter’s ballast water picked up in a European
port. They grow to about 5 centimeters (2 inches) in length
and breed and spread very quickly. They are causing
significant damage to water intake pipes, where they build
up in large numbers and block water flow. They also foul
boat hulls and may harm fish communities.

How Can You Help?

You can help monitor the spread of the zebra mussel. To
report a sighting in your area, complete this form and
forward it to the address indicated at the bottom of this
page. Your sighting information will form part of a
database that will be used to track the distribution, spread,
and abundance of zebra mussels.

PLEASE PROVIDE THE FOLLOWING INFORMATION (please print):

Name:

Phone: Home

Address:

Work

Date of sighting:

State, county:

Name of body of water:

Distance to and name of nearest town:

Zebra Mussel Information

Approximate number of mussels found: Size range:
Depth of water: River, lake, or pond?:
Type of substrate: mud __; rocks ___; mud and rocks___; sand __; other (describe)

Substrate on which mussel(s) were found:

Additional comments:

Name and phone number of person who confirmed identification of mussel (if any)

Mail this form to:

Zebra Mussel Watch
(Address of local natural
resource agency or state
Sea Grant office)

April 1992 Marsden: Zebra Mussel Protocols

National Fisheries Research Center
7920 N.W. 71st Street
Gainesville, Florida 32606
Telephone (904) 378-8181 FAX (904) 378-4956

NONINDIGENOUS SPECIES DATA BASE

Common Name

Genus Species
PRIDSPOCUGS = 0a ee ee Oecd
State County___ Drainage Basin
Location |

(please be as specific as possible)

Habitat Type: River/Stream____- Natural Lake/Pond____. Marsh/Swamp _____
Canal/Ditch_____ Man Made Reservoir____Estuary/Bay___
Other Habitat
Water Temperature Salinity DO pH Depth
Water Velocity Substrate
Vegetation

(presence, absence, type, species if known)

Identified By

(name, address, and phone number)

Collected By

(name, address, and phone number)

Method of Collection

(cast net, electrofishing, gill net, hook/line, rotenone,seine,trammel net,trawl,trot line ?)
Number Collected Age Class Size

(larvae, juvenile, adult)
Method of Disposal

(discarded, ethanol, formalin, frozen, mounted, released, tagged and released ?)

Specimen Storage
(museum or agency name and collection number)

Comments

At minimum, please provide the name or description of what you have observed,
where it was observed, the date it was observed, your name and how you may be
reached.

7 Acknowledgments

This document was conceived during a workshop on zebra
mussel research sponsored by the Great Lakes Research
Consortium and held in Syracuse, New York, on April 19—
20, 1990. Funds for researching methods and preparing the
document were granted by the Great Lakes Research
Consortium in response to a joint proposal submitted by
Ellen Marsden (then at Cornell University), Dan Molloy
(New York State Museum), Alec Aitken (Queens Univer-
sity), and Ron Engel (State University of New York at
Oswego).

The methods described here were compiled using
information from numerous researchers and agency
personnel already involved in zebra mussel monitoring, and
from standard limnological techniques (e.g., see Lind 1979,
McCauley 1984, APHA 1989). Many individuals contrib-
uted comments, reviews, and helpful discussions of the first
draft of this document, and I hereby gratefully acknowl-
edge their assistance. I would especially like to thank the
following people: Chris Brousseau and Lois Deacon of the

Ontario Ministry of Natural Resources Zebra Mussel
Coordination Office for inviting me to participate in the
development of their Fisheries Assessment Unit sampling
program for zebra mussels; Dan Molloy, Brian Malone, and
Jim Carlton for their extensive comments on the draft;
William Mason, Bob Peoples, and Ken Pickering of the
U.S. Fish and Wildlife Service for coordination with the
USF&WS protocols; Michael J. Keniry for preparation of
figures; Tammy Keniry and Nanette Trudeau for assisting
with monitoring experiments; and the participants of the
round-table discussion on monitoring techniques and
results held at the Second International Zebra Mussel
Conference, Rochester, New York, November 19-22,
1991. I would also like to thank Mary Balcer, Art Brooks,
Thomas Ferro, David Garton, Wendell Haag, Jory Jonas,
Cliff Kraft, Gerry Mackie, Dave MacNeill, Leif Marking,
André Martel, Allen H. Miller, Jerrine Nichols, Hans
Pearson, Charles Ramcharan, Ted Schaner, and Ray Tuttle
for their generous advice and information.

8 Literature Cited

AMERICAN PuBLic HEALTH ASSOCIATION. 1989. Standard
methods for the examination of water and wastewater. 17th
ed. Washington, D.C. 1,268 pp.

Bu be Vaate, A. 1991. Distribution and aspects of popula-
tion dynamics of the zebra mussel, Dreissena polymorpha
(Pallas, 1771), in the lake [Jsselmeer area (The Nether-
lands). Oecologia 86:40—S0.

Garton, D.W., and W.R. Haac. 1990. Seasonal patterns of
reproduction and larval abundance of Dreissena in western
Lake Erie: what a difference a year makes. Page 7 in
Proceedings of the International Zebra Mussel Research
Conference, Columbus, Ohio, December 1990. Ohio Sea
Grant College Program, Columbus.

Hopkins, G.J. 1990. The zebra mussel, Dreissena
polymorpha: a photographic guide to the identification of
microscopic veligers. Water Resources Branch, Limnology
Section, Ontario Ministry of the Environment, London,
Ontario. 8 pp. + figures.

Jounson, L. 1992. New technique for the rapid identifica-
tion of bivalve larvae in plankton samples. Dreisssena
polymorpha Information Review 3(1):2.

LewANbowskI, K. 1982. The role of early developmental
stages in the dynamics of Dreissena polymorpha bivalvia
populations in lakes. 2. Settling of larvae and dynamics of

numbers of settled individuals. Ekologia Polska 30:223-286.

Linp, O.T. 1979. Handbook of common methods in limnol-
ogy. 2nd ed. C.V. Mosby Co., St. Louis, Missouri. 199 pp.

McCau ey, E. 1984. The estimation of the abundance and
biomass of zooplankton in samples. Pages 228-265 in J.A.
Downing and F.H. Rigler, eds. A manual on methods for
the assessment of secondary productivity in fresh waters.
Blackwell Scientific Publications, Oxford, UK.

Mackie, G.L., W.N. Gippons, B.W. Muncaster, and I.M.
Gray. 1989. The zebra mussel, Dreissena polymorpha: a
synthesis of European experiences and a preview for North
America. Water Resources Branch, Great Lakes Section,
Ontario Ministry of the Environment, London, Ontario. 76
pp. + appendices.

MacNEIL, D. 1991. Identification of juvenile Dreissena
and Mytilopsis leucophaeta. Dreissena polymorpha
Information Review 2(1):1—2.

Nicuo.s, S.J., and B. Kottar. 1991. Reproductive cycle of
zebra mussels (Dreissena polymorpha) in western Lake
Erie at Monroe, Michigan. Page 3 in Proceedings of the

Second International Zebra Mussel Research Conference,
Rochester, New York, November, 1991. New York Sea
Grant, Brockport.

PREZANT, R.S., and K. CHALERMWAT. 1984. Floatation of the
bivalve Corbicula fluminea as a means of dispersal.
Science 225:1491—1493.

ScCHANER, T. 1990. Detection of zebra mussel veligers in
plankton samples using sugar solution. Lake Ontario
Fisheries Unit Annual Report, LOA 91.1. Ontario Ministry
of Natural Resources, Picton. 3 pp.

Scott-WasILK, J., G.G. Downina, and J.S. Lierrzow. 1983.
Occurrence of the Asiatic clam Corbicula fluminea in the
Maumee River and western Lake Erie. Journal of Great
Lakes Research 9:9—13.

Smirnova, N. 1990. Biology and ecology of Dreissena
polymorpha from the European USSR. Pages 1-8 in J.D.
Yount, ed. Ecology and management of the zebra mussel
and other introduced aquatic nuisance species. Environ-
mental Protection Agency, Environmental Research
Laboratory, Duluth, Minnesota.

SprunG, M., and U. Rose. 1977. Influence of food size and
food quantity on the feeding of the mussel Dreissena
polymorpha. Oecologia 77:526—532.

STANCZYKOWSKA, A. 1964. On the relationship between
abundance, aggregations and “condition” of Dreissena
polymorpha Pall. in 36 Mazurian lakes. Ekologia Polska A
12:653-690.

Watz, N. 1973. Studies on the biology of Dreissena
polymorpha in the Lake of Constance. Archiv fuer
Hydrobiologie Suppl. 42:452-482.

Watz, V.N. 1975. The settlement of larvae of Dreissena
polymorpha on artificial substrates. Archiv fuer
Hydrobiologie Suppl. 47:423-431.

Watz, N. 1978. The energy balance of the freshwater
mussel Dreissena polymorpha in laboratory experiments
and in Lake Constance. III. Growth under standard condi-
tions. Archiv fuer Hydrobiologie Suppl. 55:121-141.

Wuire, D.S., M.H. WINNELL, and D.J. Jupe. 1984. Discov-
ery of the Asian clam, Corbicula fluminea, in Lake
Michigan. Journal of Great Lakes Research 10:329-331.

ZHADANOVA, G.A., and S.L. GusyNSKAYA. 1985. Distribu-
tion and seasonal dynamics of Dreissena larvae in Kiev and
Kremenchug reservoirs. Hydrobiological Journal 3:35—40.

Appendix I: Distinguishing Adult Dreissena polymorpha and Mytilopsis

leucophaeata

Reprinted with permission from MacNeill, D. 1991. Identification of Dreissena and Mytilopsis, part Il. Dreissena

polymorpha Information Review 2(2):9.

Shell structure Dreissena Mytilopsis

1) Internal Microscopic Features

a) Posterior Retractor Muscle:
does not extend to
anterior shell margin

extends to anterior
shell margin.

may be invaginated
forming a sinus.

b) Pallial Line: rounded at posterior
portion, no sinus.

c) Myophore Plate (Septum):
broad, scars of both
anterior muscles anterior adductor scar
present on septum. present on septum.
no apophysis present. anterior retractor

attached to inward
facing apophysis.

narrowed, only the

2) External Shell

a) Shape: more flattened at more rounded and
anterior marginand _ broad laterally.
ventrally.

Dreissena polymorpha Mytilopsis leucophaeta
Gee ee
b) Markings: _ typically have herring- often have the herring-

bone patterns, may be bone pattern. generally

radially striped or darker coloration.

show diffuse striping.

Posterior

retractor
Posterior ™U
adductor scar
muscle
scar

Anterior
retractor
mu:

Anterior
adductor

Dreissena polymorpha
Posterior Posterior
adductor retractor
muscle muscle
scar scaly

No Anterior
retractor
muscle
scar

Anterior
adductor
muscle

Pallial sinus scar

Pallial line

narrow

Mytilopsis leucophaeta

Dorsal
Anterior

Posterior

Ventral

Appendix II: Detection of Zebra Mussel Veligers in Plankton Samples

Using Sugar Solution
Ted Schaner

Ontario Ministry of Natural Resources, Lake Ontario Fisheries Unit, R.R. #4, Picton, Ontario KOK 2T0

Reprinted with permission from Lake Ontario Fisheries Unit 1990 Annual Report, LOA 91.1 (Chapter 6).

A method is presented to efficiently detect presence of zebra mussel veligers in plankton
samples. The veligers can be separated from many other constituents of the sample by allowing
the sample to settle through a column of sugar solution. Most of the veligers are recovered
within 20 min in a few drops of liquid at the bottom of the column, allowing quick examination.
The method is especially suitable for initial detection of veligers at low concentrations, but

potentially it also has quantitative applications.

Introduction

In 1990, the Ontario Ministry of Natural Resources
started to monitor zebra mussel (Dreissena polymorpha)
in Lake Ontario. Since the mussel was just beginning to
invade the lake, and population densities were expected
to be low, we examined large volumes of water to reliably
detect presence of the mussel’s veliger larvae. The method
described here was developed to speed up processing of
the samples, and to reduce analytical costs.

Materials and methods

Plankton samples were processed through a settling ap-
paratus consisting of a 25 ml pipette fitted with a three-way
rubber pipetting bulb (Fig. 1). The pipette was partly filled
with sugar solution (Table 1), and the plankton sample was
introduced over the top of the solution. Planktonic or-
ganisms were allowed to settle through the sugar solution
for a period of time, and were then collected from the tip
of the pipette.

The pipette was FISHERbrand 25 ml in 1/10 (#13-665
SZ N). The outflow tip of the pipette was sanded off to
the point where the inner diameter was approximately 1.5
mm. This prevented clogging of the opening with large or-
ganisms and filamentous algae. The pipette was filled with
sugar solution up to the "10 ml" mark (slightly more than
15 ml of liquid or a column 195 mm tall). A 10 ml
plankton sample topped up the liquid to the "0 ml" mark.

To set up the settling apparatus, the sugar solution was
first drawn into the pipette using the rubber bulb. The tip
was then sealed off with Parafilm, and the bulb removed.
The plankton sample was introduced from the top using
a syringe, and the timer was started . In a quick succession,
the bulb was fitted back on the pipette, the Parafilm was
removed, and any hanging drops were wiped off. Samples
were periodically withdrawn from the tip of the pipette by
pressing the "E" button (Fig.1) of the rubber bulb gently
and slowly so that a single drop of liquid was released
onto a depression slide.

Rates at which veligers settle through the apparatus
were investigated by processing plankton samples, either

‘o)|| Sess
ml
plankton
sample
oO) So oe
mi
sugar
syrup

————

FIG. 1. The settling apparatus.

containing known numbers of veligers, or representing
batches of known veliger concentration. In all cases the
samples were preserved with buffered formalin (Table 1)
for at least 0.5 h before processing to allow osmotic
equalization. Drops with settled organisms were collected
from the tip of the pipette at 2 or 5 min intervals over a
period of up to 30 min, and examined under a dissecting
microscope at 25x magnification. To indicate efficiency,
the numbers of recovered veligers were expressed as per-
centage of the number introduced into the apparatus.

26 Illinois Natural History Survey Biological Notes No. 138

TABLE 1. Composition of preservative fluid and sugar
solution used in processing of the plankton samples.

Preservative fluid:
(Modified after G.Hopkins

Ontario Ministry of Environment
Rexdale, pers. comm.)

37% formaldehyde 850 ml
Distilled water 1000 ml
Sugar 500 g

Sodium bicarbonate to raise pH to 7.0

Combining 1 part of this solution with 4 parts of
plankton sample results in approximately 4% formal-
dehyde concentration.

Sugar solution:
Sugar 130 g
Distilled water 400 ml
Results
Veligers

The ideal settling rates and efficiencies for veligers were
examined in three trials in which known exact numbers of
veligers were introduced into the apparatus. Few other
particles and organisms were present to interfere with
veliger settling. The first veligers settled at 2 to 4 min (Fig.
2), and settling began to level off at 15 to 20 min. After
30 min the cumulative number of settled veligers repre-
sented 75-94% of the numbers used to seed the samples.

Tests with "real" plankton samples showed similar or
somewhat lower efficiencies. After 20 min of settling, mean
efficiencies ranged between 55 and 85%, and increased up
to 90% at 30 min (Fig. 3A, B, C). The efficiencies appear
to vary depending on the overall particle concentration of
the introduced plankton sample. A very dense plankton
sample from Nanticoke, Lake Erie, was diluted to 20 and
50% of original concentration, and a series of tests were
run with each of the two batches. The more diluted batch
(20%, Fig. 3B) showed efficiencies similar to ones ex-
perienced with veliger-only samples (Fig. 2), while the ef-
ficiencies with the more concentrated batch (50%, Fig. 3C)
were lower, and arrivals at the bottom of the pipette
leveled off only slightly during the 30 min experiments.
This suggests that in highly concentrated samples the
veligers are prevented from settling through the introduced
sample to the top of the sugar column.

Size selectivity of the settling process was investigated
using measured veligers mixed with a veliger-free plankton
sample to simulate a realistic sample. Sizes of settled
veligers were compared with those in the original sample.
It appears that the size composition of the veligers settling
through the apparatus within 25 min tended to be biased
towards larger individuals (Fig. 4), though the difference
from the original size frequency distribution was not statis-
tically significant (chi-square and Kolmogoroy-Smirnov
tests, p>0.1).

ici
ra Efficiency

Number of veligers seeded:

47, — 22 a 14

0 2 4 6 8B 10 12 14 16 18 20 22 24 26 28 30
Time (min)

FIG. 2. Settling rates measured by allowing exactly known numbers of
veligers to settle through the sugar column in almost complete absence of
other organisms. Counts were made at 2 min intervals, and are expressed as
cumulative (over time) percentages of the starting numbers.

Other Organisms and Debris

The settling times for the other plankters varied. Too
few experiments were performed to confidently describe
the generalities. However, two observations appear reli-
able: 1) inorganic debris tended to settle within the first
4 min, and 2) cladocerans started arriving after ap-
proximately 15 min. At 30 min the arrivals of all types of
particles began to decrease, though the liquid above the
sugar column remained turbid.

Discussion

Settling the plankton samples through a column of sugar
solution can be useful in detection of zebra mussel veligers.
A high proportion of the veligers passed through the set-
tling apparatus within the time window of 5 to 20 min,
and allowed separation from both early-settling inorganic
debris, and late-settling planktonic organisms. The bias in
sizes of the settled veligers was small.

The method offers several advantages. Firstly, a bank
of settling pipettes can be set up and operated simul-
taneously, and thus samples representing large volumes of
water can be rapidly processed. In our sampling program,
two people were able to process and microscopically ex-
amine samples representing 840 litres in 40 min. Fatigue
is minimized since very little time is spent in microscopic
examination. Various types of organisms tend to settle at
various times, and if samples of settlers are taken peri-
odically throughout the settling period, then the lesser
variety of organisms in any single sample leads to easier
detection of veligers. Fractions of the plankton sample can
be discarded with little decrease in efficiency: discarding
a drop at 3 min and then stopping the process at 20 min,
will avoid sand and silt particles as well as most
cladocerans, while still capturing at least half of the veliger
larvae.

April 1992 Marsden: Zebra Mussel Protocols 27
Qumul no. veligers Qumul no. veligers Oumul. no. veligers
(mean & Efficienoy (meen & Efficiency (mean & stddev) Effidienoy
20 A 100% 100% 100%
10 26
803 60% 60%
15 20
60% 60% 15 60%
10 6
40% 40% 10 ied
5
20% 20% 6 20%
oO OS o—= Os ie) 0%
o066lh6D 6H UBD CHOC Oo 6 WW 6G 20 2 80 0 6 OW 6G 2 2 80
Time (min) Time (min) Time (min)

FIG. 3. Settling rates of veligers, measured in samples that included other plankton organisms. Each graph represents several replicate runs, and the mean
cumulative (over time) count is shown. The replicates were taken from known concentrations of veligers, but the exact number of veligers in each replicate
was not known. Therefore, some part of the standard deviation is due to variability in the starting numbers. Efficiency is the number of recovered veligers
expressed as percentage of the starting numbers. A: Lake Erie veligers mixed with veliger-free plankton sample from Lake Ontario, six replicate trials. B:
Plankton sample with veligers from Nanticoke, Lake Erie diluted to 20% original strength, nine replicate trials. C: Same as B, but diluted to 50%, eight

replicate trials.

It appears that efficiency decreases when the plankton
sample is too concentrated and veligers are impeded in
reaching the sample-sugar interface. The data presented
here indicate approximate efficiencies, and allow for rough
correction. No guidelines for maximum particle con-
centration are given here, though establishing such
guidelines would render the method more suitable for
quantitative applications.

The most logical application of this method is in the
initial detection of zebra mussel infestation, when veliger
densities are on the order of several veligers per cubic
metre. Here the drawback of incomplete enumeration is
offset by the ability to examine large volumes of water.

Acknowledgments

I thank Joe Leach, Dave Lowther, and Don Lewis for
providing veliger larvae and plankton samples, and Gord
Hopkins for providing the formalin recipe.

Percent frequency
20%

HM orig.eample

Settled veligers
15%

10%

5%

042 014 016 018 O02 0.22 0.24 0.26
Size of veligers (mm)

O%

FIG. 4. Size selectivity of the settling process: size frequency distribution of
introduced sample (400 veligers measured, sample from Fig. 3A) contrasted
with distribution of veligers settled within 25 min (66 veligers pooled from
six replicate runs).

Appendix III: Collector for Veliger and Drifting Postmetamorphic Zebra

Mussels

André Martel

Research Division, Canadian Museum of Nature, P.O. Box 3443, Station D, Ottawa, Ontario, Canada K1P 6P4

One of the major disadvantages of using settlement plates
for collecting juvenile zebra mussels is the long period of
time needed for deployment. The plates are vulnerable to
storm damage and vandalism during this time, and results
are not obtained for weeks to months. Plankton tows to
collect veligers have the disadvantage that they sample a
relatively small, contiguous unit of water. This results in a
high probability of missing veligers if their distribution is
patchy. The purpose of this paper 1s to describe a new
technique for collecting planktonic zebra mussels which
overcomes the disadvantages of existing techniques. The
collectors used in this new technique have two main
advantages over other types of artificial substrate currently
used to monitor and sample zebra mussels:

1. Only a short time period is needed between deploy-
ment and retrieval in the field. High collection rates are
achieved within 2472 hours of deployment. This
means that results are obtained rapidly, and the equip-
ment has a minimal chance of being lost or damaged
due to severe weather or vandalism.

2. The collection rate of both veligers and postmeta-
morphic stages (early juveniles) is higher than that
with other settlement plates. This results in a high
presence-or-absence detection rate in a short period of
time. Also, because a wide size range of zebra mussels
are collected, these collectors integrate the functions of
both plankton nets and settlement plates.

Equipment Construction

In recent studies using the new technique, the active
element of the collector was a 12 cm x 11 cm, 0.6- to 0.8-
cm-thick, white nylon scouring pad. Similar collectors have
been used in studies of Mytilus in Ireland (King et al. 1990)
and in Canada (Martel, unpublished data). The scouring
pads used in this study of zebra mussels were manufactured
by the Fireco Company, Houseware Products, 1280
Courtney Park Dr., Mississauga, Ontario, LST 1N6, and
were purchased at a local grocery store. Similar fine-fiber
scouring pads from other manufacturers may be useable,
but they must be free of detergent. The pads were approxi-
mately $0.50 (Canadian) apiece. To stabilize the pads and
prevent them from folding during deployment, and to
permit comparison of the pads with standard Plexiglas
settlement plates, two Plexiglas strips were attached to the
top and bottom of the pad using hot-melt glue (Figure 1).
Because they make strong joints, yellow or white glue
sticks are preferable to transparent glue sticks for this
purpose. The edges of the Plexiglas strips were roughened
with sandpaper prior to gluing to enhance the strength of
the joint. Glue was allowed to penetrate about 4 mm into
the pad fibers. The strongest joint was achieved when the
pad and Plexiglas were bound together while the glue was
still very hot and liquified. The Plexiglas strips used in this

experiment were 10 cm x | cm (top) and 10 cm x 5 cm
(bottom) (Figure 1A). However, two similar-sized Plexiglas
stabilizing strips can be used if comparison of the scouring
pad with Plexiglas plates is not an objective of the study.
During testing of the scouring pads in Lake Erie,
swivels were attached to the top and bottom of the collector
to allow free rotation of the collector in wind-generated
water currents and waves. This system permitted consistent
exposure of the collector’s broad surface to the water flow,
thus maximizing the amount of water “sieved” by the
collector and ensuring a similar orientation for all the
collectors. The effectiveness of the swivel depended upon
how well the swivel was centered on the collector’s
Plexiglas strips (Figure 1A). Newly constructed collectors
were tested prior to deployment by attaching them to a line
and dragging them through a large tub of water. The
collectors oriented easily and consistently toward the
imposed current, similar to a self-orienting windmill.

Deployment of Collectors

Collectors were deployed in series along a line, using a
3'/-in. galvanized nail attached to each swivel using hot-
melt glue (Figure 1A). A small notch was made 7-8 mm
from the end of each nail, and the swivel ring was posi-
tioned inside the notch before gluing (Figure 1A). The head
of each nail was inserted through the ply of */s-in. or */s-in.
three-ply twisted polypropylene line. This system permitted
rapid addition, replacement, and removal of collectors. The
system was also highly reliable: no collectors were lost
during multiple trials under various weather conditions in
Lake Erie and on the coast of Vancouver Island. For
optimal deployment, the two nails were parallel to each
other and perpendicular to the line before the nail heads
were inserted into the line. Several collectors were de-
ployed on a single anchored and buoyed line. The lowest
collector was at least 40-60 cm above the bottom to avoid
damage and clogging with bottom silts. The distance
between collectors varied according to study needs and the
water depth at a particular site. After initial deployment, the
collectors were checked to ensure that the nails of the
collectors were still aligned in one plane and nearly parallel
(Figure 1A).

An alternative method of attachment and deployment
of the scouring pad collectors consists of simply tying the
two Plexiglas pieces (top and bottom of collector) to the
rope using a stainless steel ring and clip system. The clip is
attached to one of the three major plies of the rope (Figure
1B). This method of attachment may be used in situations
where currents are weak or orientation of the collector to
currents is not a concern. Also, where currents are unidirec-
tional, the collector can be tied by the corners to a fixed
point, such as a float or a wall of a large intake pipe,
allowing the pad to be fully exposed to the flow. Evidently,

April 1992

swivel

Plexiglas holding edge
hot melt glue joint

scouring pad

hot melt glue joint

Plexiglas plate

swivel

hot melt glue attachment clip

notched galvanized nail

peamecceoeree - —s
2cm
Figure 1. Diagrams of two off-bottom scouring pad collectors
with attached Plexiglas stabilizing plates. Collectors are shown
attached to a */16-in. polypropylene rope by (A) a self-orienting
swivel system using a notched galvanized nail and (B) a simple
clip system.

the fixation and deployment of the scouring pads can be
adapted to specific situations and to the needs of the study.

To ensure that the line holding the collectors was
vertical, a small, submerged buoy was attached to the line
20-30 cm below the main buoy on the surface. Alterna-
tively, a small satellite anchor can be attached a few meters
away from the main anchor. This anchor will bounce in
heavy waves and acts as a shock absorber to reduce tension
on the main anchor. For the brief deployment periods
needed for these collectors, cinder block or brick anchors
appear to be quite adequate. For deployment in highly
exposed areas, an anchor made from a section of chain
(e.g., 50-70 cm length of 1- to 2.5-cm-diameter link chain)
offers the advantage of being extremely stable due to its
high density and low profile during strong bottom surge
generated by storms.

The scouring pad collectors worked best when
deployed for short periods of time, between 24 hours to one
week. Longer periods of time may result in fouling of the
pads with detritus and planktonic organisms. The optimum
period for deployment will vary according to the local
turbidity (as measured by Secchi disk). Recent storms will
tend to suspend volumes of sediment, particularly detritus,
which may clog the collectors, although at Wheatley,
Ontario, deployment during such periods usually resulted in
very high numbers of zebra mussels, particularly
postmetamorphic stages, in the scouring pads (Table 1).
Before use, all new collectors were soaked in filtered (64
uum) lake water for 2—3 days. This prewashing assisted in
conditioning the pads and removed any traces of chemicals
that may have been present on the pads.

Marsden: Zebra Mussel Protocols 29

Table 1. Number of larval and postmetamorphic zebra mussels
collected on scouring pad samplers and Plexiglas plates deployed
at several stations near a breakwater at Wheatley, Ontario, in Lake
Erie during August 1991.*

Scouring

Station Date — Collector Plexiglas pad
1 Aug 8 1A 0 368
1D 0 486

2 2A 0 317
2C 0 345

2E 2 527

l Aug 14 1A 0 5
1B 0 9

1c 0 3

1D 0 6

1E 0 2

4 Aug 24 3A 0 151
3¢€ 0 212

4A | 451

4E 2 39

*Collectors were each deployed for 24 hours. Collectors were
progressively deeper from A to E, with the maximum depth at 5
m. Note the high numbers of zebra mussels captured in scouring
pads compared with settlement on the Plexiglas plates.

Extraction, Observation, and Preservation of Mussels
The Plexiglas portion of the collectors was examined prior
to extracting zebra mussels from the pad. The pads were
washed using a low-flow, fairly strong jet of cool tap water,
obtained using a hose equipped with an adjustable nozzle.
This pressurized jet is necessary to remove zebra mussels
that have already secreted new byssal threads during the
24-hour period of deployment. Washing was done over a
shallow pan, with the hose held 1 cm from the pad. Each
collector was washed three times for 30 seconds each. A
quick examination of the filtrate under a stereomicroscope
permitted estimation of whether successive washes were
removing additional material. The filtrate was passed
through two filters to grade the collected material. Both
filters were constructed by removing the bottom from a
250-ml plastic beaker and gluing Nitex mesh across the
hole. A 500-t1m and a 125-1m filter were used. After
extraction, the samples were carefully examined under a
stereomicroscope to determine whether zebra mussels
caught in the pads were alive. This determination is
important because empty shells of small juveniles (2—3 mm
or less) may be resuspended by currents. In areas where
zebra mussels are known to be present, the occurrence of
empty shells or dead mussels may not, for instance,
necessitate any immediate mitigation treatments. With a
high-power bright-field stereomicroscope (40—60x), live
veligers and early postmetamorphic zebra mussels can
easily be distinguished by checking for distinct, well-
defined internal body organs, including the foot and gill
lamellae. In particular, the foot and the long cilia of the
veliger’s velum can be seen, either expanded or withdrawn
into the shell.

30 Illinois Natural History Survey Biological Notes

After initial observation under the microscope, the
samples were preserved in 70% ethanol. The collectors
were then thoroughly washed with a water jet and placed in
55-60°C water for 5 minutes to ensure that any remaining
mussels were killed. Boiling water will melt the hot-melt
glue; ethanol is also an inappropriate disinfectant because it
is a weak solvent for both the Plexiglas and the scouring
pad. The collectors were dried flat, inspected for damage,
repaired with hot-melt glue if needed, and then reused.

Results

During short-term deployment (24-72 hours), collection
rates of zebra mussels on the pads was frequently two
orders of magnitude higher than settlement on the Plexiglas
plates (Table 1). More than 600 individual zebra mussels
were collected in a single pad after 24 hours. Settling
veligers as well as drifting juveniles from 300 to 900+ lum
long were collected in the pads. Factors that may contribute
to the high collection rates obtained with the pads include
the following: (1) the three-dimensional complexity of the
pad substrate, which offers a large surface area for settle-
ment; (2) the large amount of water “filtered” by the porous
structure of the pad; and (3) preferential settlement of
larvae and early postmetamorphic stages onto structurally
complex substrata compared with very smooth surfaces.

Conclusion

The scouring pad collector appeared to act as a passive
plankton net by sieving a large volume of water over the
entire period of deployment. A wide size range of plank-
tonic zebra mussels were collected, including
postmetamorphic stages or early juveniles. These small
juveniles may have been washed off surfaces during severe
weather or currents; possibly, some may have undergone
intentional dispersal by drifting on mucus threads, as
shown in Corbicula (Prezant and Chalermwat 1984),
marine bivalves (Lane et al. 1985, Beukema and de Vlas
1989, Martel and Chia 1991a), and some gastropods
(Martel and Chia 199 1a, b). The collector therefore
effectively integrates the uses of plankton nets and settle-
ment plates. The collectors are inexpensive (less than $2.50
U.S. apiece), easy to build from readily available materials,
reusable, and provide quantitative, reproducible results.
They can be used for short periods of time to avoid severe
weather and vandalism, and they quickly produce informa-
tion about local zebra mussel levels.

Acknowledgments

I am indebted to R. Toonen for his assistance in the original
design and construction of the scouring pad collectors.

J. Fournier, K. Klein, J.E. Marsden, and J. Topping
commented on earlier versions of the manuscript. I also
want to thank the staff of the Lake Erie Fishery Station -
OMNR, at Wheatley, Ontario, especially the director, Dr.
S. Nepszy, as well as Dr. J. Leach, for their support and for

allowing me to use the station facilities during the testing of

the scouring pad collectors in Lake Erie.

No. 138

Literature Cited

BeukeEMA, J.J., and J. be VLAs. 1989. Tidal-current transport
of thread-drifting postlarval juveniles of the bivalve
Macoma balthica from the Wadden Sea to the North Sea.
Marine Ecology Progress Series 52:193—200.

Kina, P.A., D. McGratn, and W. Britton. 1990. The use
of artificial substrates in monitoring mussel (Mytilus edulis
L.) settlement on an exposed rocky shore in the west of
Ireland. Journal of the Marine Biological Association of the
United Kingdom 70:37 1-380.

Lang, D.J.W., A.R. BEAUMONT, and J.R. Hunter. 1985.
Byssus drifting and the drifting threads of the young post-
larval mussel Mytilus edulis. Marine Biology 84:301—308.

MarteL, A., and F.S. Cua. 1991a. Drifting and dispersal of
benthic marine invertebrates: common occurrence of small
bivalves and gastropods with direct development in off-
bottom algal collectors. Journal of Experimental Marine
Biology and Ecology 150:131—147.

MarteL, A., and F.S. CuiA. 1991b. Foot-raising behaviour
and active participation during the initial phase of post-
metamorphic drifting in the gastropod Lacuna spp. Marine
Ecology Progress Series 72:247-254.

PREZANT, R.S., and K. CHALERMWaAT. 1984. Floatation of the
bivalve Corbicula fluminea as a means of dispersal.
Science 225:1491-1493 .

Appendix IV: Equipment and Chemicals

Reagents
A note about preservatives: Formalin (10% formaldehyde)

is a traditional preservative for plankton samples. However,

formalin vapor is irritating to the eyes and nose, and
prolonged exposure is carcinogenic. Great care should be
used to keep formalin contained and to only examine
preserved specimens under an exhaust fan or in a well-
ventilated area. Samples of shelled mussels (settled
juveniles or adults) can be rinsed and transferred to water
for the duration of the counting procedure to minimize
exposure to formaldehyde. This procedure is trickier with
veligers because of the probability of losing veligers during
the transfer to water. Ethanol is also a commonly used
preservative and is gaining popularity because it is more
benign to the user than formalin. However, state and
federal agencies are often required to use alcohol that has
been denatured with compounds such as acetone, which are
irritating to humans and may affect the sample.

Buffered sugar formalin: Dissolve 80 g of granulated sugar
in | liter of 10% formalin. Dilute with sample by one-half
to produce a 5% solution. (Note: 10% formalin is a 9:1
dilution of water and formaldehyde. Formaldehyde, as
purchased, is a 37% solution in water). Buffer the formalin
by adding a handful of marble chips (calcium carbonate) to
each liter of solution. Sodium bicarbonate can also be used
to buffer to pH 7.0

Alcohol: Use 70% ethanol or isopropyl! alcohol.

Equipment Sources

Note: Identification of equipment manufacturers and
distributors is provided for convenience and should not be
taken as an endorsement of any kind.

Design Alliance, 114 East 8th St., Cincinnati, OH 45202.
513-621-9373 (slide racks)

Environmental Research Instruments, 70 Durham St.,
Guelph, Ontario, Canada, NIH ZY3

Forestry Suppliers Inc., P.O. Box 8397, Jackson, MS
39284-8397. 800-647-5368

Ernest A. Case, P.O. Box 45, Andover, NJ 07821. 201-347-

1365

Limnotech, 136 Scarborough Rd., Toronto, Ontario,
Canada, M4E 3M6. 416-698-0978

Research Nets, 23102 55th Ave. W., Mt. Lake Terrace,
WA 98043. 206-821-7345

Robar Maching Inc., 2611 East 40th St. Chattanooga, TN
37407. 615-867-4717 (settlement racks)

Wildco Wildlife Supply Company, 301 Cass St., Saginaw,
MI 48602. 517-799-8100

Estimated Sampling Costs

Prices for various critical pieces of sampling equipment are
provided to assist investigators in estimating the scope and
strategy for their sampling program. Cost estimates, given
in U.S. dollars, are based on 1991 equipment prices.

Sampling Veligers

Plankton net 30 cm, D/L ratio 1:3, 63-64 mm mesh $200
Flowmeter >$200

Secchi disk (20-cm diameter) $50

Sedgewick-Rafter counting cell $25

Dissecting microscope $500—$5,000

Incidental costs for preservative, pipettes, clipboards, etc.

Sampling Settled Juveniles

Microscope slides ($14/gross) $0.10

Periphyton sampling rack for microscope slides
(ready-made) $38

Multiplate sampler (ready-made) $17

PVC plate sampler (home-made) $5

Scouring pad sampler (home-made) $3

Incidental costs for lines, buoys, anchoring material,
preservative, etc.

Sampling Adults

Ponar grab $300-$600

Incidental costs for constructing sampling frame and
coring device

Use of commercial divers can run from $50 to several
hundred dollars per diver and may require a
minimum of three divers per dive.

Appendix V: Forms and Labels

Secchi
depth

water
temp

sampled

sample |volume

direction {depth

current
velocit

bottom |bottom
t e Cua

ZEBRA MUSSEL VELIGER SAMPLING FORM
lift #_|depth

type”

RECORDER:

“sample type: vertical tow (vt), oblique tow (ot), pumped water (pump), flowing water (flow)

“bottom type: rocky (rky), sand, silt/mud, vegetation (veg), artificial (art), unknown (unk)

April 1992

ZEBRA MUSSEL VELIGER COUNTING FORM

RECORDER:

Marsden: Zebra Mussel Protocols

33

34 Illinois Natural History Survey Biological Notes No. 138

SAMPLING FORM FOR SETTLED ZEBRA MUSSEL JUVENILES

nearest
bottom |water |water |Secchi |current |current |mussel
type* _|depth |temp. {depth [velocity |direction |colon

RECORDER:

location type date/hour date/hour |depth

* bottom type: rocky (rky), sand, silt/mud, vegetation (veg), artificial (art), unknown (unk)

April 1992 Marsden: Zebra Mussel Protocols 35

COUNTING FORM FOR SETTLED ZEBRA MUSSEL JUVENILES

vol. of |smallest |largest
sample sampler |retrieval |immersion | twen. oor om square aH = 1 a square __|#juven.|juven. |mussel |mussel
oS ined a La mie per m2 |per m2 |counted_|on slide

(yun) UMOoUuyUN ‘(YWe) jeloie ‘(Ban) Uoe}eBHaA ‘pnu/jIIs ‘pues ‘(Ay1) Ayo :adA} Woyog,

No. 138

} Sal = Sheol ee eee
beat peptides ce ete te ede
[ee a a eS a eT Re es ER ae
[SEX teal Se Pe ee eae Tee as EI ae
(eames) Fics] ie ved eet eel ieee ese ieee
ae Lz ee

:(pasn j!) YAHOO JO HALAWVIG ‘Y430YHOOSY

Illinois Natural History Survey Biological Notes

36

WHO DONIIdNVS TASSNW VHEsaZ LINGV

April 1992

Marsden: Zebra Mussel Protocols

Sample labels - fill in with pencil and insert in sample container with preservative

Collector:

Preservative:

Replicate #:

Collector:

Preservative:

Replicate #:

Sample #:
Replicate #:

Collector:

Preservative:

Site:
Sample #:
Replicate #:

Collector:

Preservative:

Replicate #:

Collector:

Preservative:

Replicate #:

Sample #:
Replicate #:

Site:
Sample #:
Replicate #:

Collector:
Preservative:

Replicate #:

Collector:
Preservative:

Sample #:
Replicate #:

Collector:
Preservative:

Sample #:
Replicate #:

Replicate #:
Date:
Time:

37

Appendix VI: Conversion Table

Convert
from

Centimeters

Cu centimeters

Liters

Meters

Meters/sec

Microns

Kilometers

Sq centimeters

Sq meters

Convert
into

feet
inches

liters

pints (US liq)

milliliters

cu feet
cu meters
milliliters

pints (US liq)
gallons (US liq)

centimeters
feet
inches

feet/min
km/hr

statute miles/hr

inches
millimeters

statute miles

sq meters
sq inches

sq millimeters

sq centimeters

sq yards

Multiply
by

0.0328
0.3937

0.001
0.002
l

0.035
0.001
1000
2s
0.264

100
3.281
39.370

196.85
3.6
2.2369

0.000039
0.001

0.621

0.0001

0.155
100

10,000.
1.196

Reverse
conversion

30.48
2.54

1000
473.2
1

28.316
1000
0.001
0.473
3.785

0.01
0.3048
0.0254

0.00508
0.2778
0.447

25400
1000

1.609
10000

6.4516

0.01

0.0001
0.836

Illinois Natural History Survey
Natural Resources Building
607 East Peabody Drive
Champaign, Illinois 61820
217-333-6880

A Division of the Illinois Department of Energy and Natural Resources

Mt briana ts
lah Cates Th be Me tbs be
Laer ee

(ited

fia
ce PCO Pn e
A heb fig Nabi OPE
“ep *
ThE a at
ative mal bee)
Lieteigwne
aantal,

7 paren
Mie lipmidumaiameests!
Beets

Rn
Bre

shishtintsy ae i
, rhe i ; Y ‘ rH eon rt
; fadtineat

a ee

arn
B Coa
Sa ed ci fal aeh 1
BreDuttonth i , y adi,
“i ebb his
Yi igi

S

eae

dita
retin
fel she

Natt ; a
ret u anerearunty ii
Fe Psa AR aS

nt

34

Ves
coat

eit

it
seat

{total
pees
Seay
rare

3,

eyly
Peters
wee

RO
1 Hb