Content uploaded by Carli Wagner
Author content
All content in this area was uploaded by Carli Wagner on Jan 17, 2020
Content may be subject to copyright.
J. Aquat. Plant Manage. 58: 7–18
Desiccation tolerance of the invasive alga starry
stonewort (Nitellopsis obtusa) as an indicator of
overland spread risk
WESLEY J. GLISSON, CARLI K. WAGNER, MICHAEL R. VERHOEVEN, RANJAN MUTHUKRISHNAN, RAFAEL
CONTRERAS-RANGEL, AND DANIEL J. LARKIN*
ABSTRACT
Human-assisted transport via recreational boats and
trailers is thought to account for most contemporary
movement of aquatic invasive species (AIS) among lakes.
The ability of invasive macrophytes to survive out of water,
that is, their desiccation tolerance, is an important indicator
of capacity for spread to new water bodies through overland
transport. Invasion by the alga starry stonewort (Nitellopsis
obtusa [Desv. in Loisel.] J. Groves; Characeae) in North
America is likely driven via overland transport, but little is
known regarding its ability to remain viable out of water.
We conducted laboratory and outdoor experiments to
evaluate desiccation tolerance of starry stonewort propa-
gules, including single stem fragments, small and large
clumps of fragments, and bulbils (asexual reproductive
structures). Propagules were removed from water for 15 min
to 5 d to identify desiccation thresholds. Fully desiccated
fragments and clumps lost ~90% of their original mass and
bulbils .60% of original mass. Based on the most
conservative estimates from our experiments, starry stone-
wort was no longer viable at 2 h for single fragments, 24 h
for small (5-g) clumps, 4 d for large (45-g) clumps, and 4 h
for bulbils. Overall, starry stonewort appears less tolerant of
desiccation than other invasive macrophytes that have been
evaluated, which comprise vascular plants rather than
Characean algae. Widely adopted guidance that boats
should dry on land for 5 d to prevent AIS spread should
suffice to prevent most starry stonewort spread between
water bodies, given that boaters comply with inspection
protocols and remove large, readily detected clumps.
Key words: bulbil, charophyte, invasive species, macro-
alga, overland transport, viability
INTRODUCTION
Spread of aquatic invasive species (AIS) via overland
transport is a major driver of new invasions (Johnstone et al.
1985, Buchan and Padilla 1999, Johnson et al. 2001).
Although overland transport can occur via animals (Reyn-
olds et al. 2015, Green 2016), human-assisted transport via
recreational boats and trailers likely accounts for a great
majority of recent AIS dispersal events (Johnson and
Carlton 1996, Buchan and Padilla 1999). Overland dispersal
of aquatic species occurs when a propagule is able to remain
viable until introduction to a new water body (Vander
Zanden and Olden 2008). Hence, for an aquatic invasive
plant species to establish in a new water body, a propagule
must arrive with the ability to sprout, continue growth, or
regenerate from living tissue. Removing plant material from
watercraft is effective for reducing spread risk, but
recreational boaters often fail to check and clean their
boats upon leaving a water body (Rothlisberger et al. 2010,
Cimino and Strecker 2018). Even when watercraft are
inspected and cleaned, a small amount of plant material
typically remains (Rothlisberger et al. 2010). Hence, the
ability of an aquatic invasive plant to survive out of water,
i.e., desiccation tolerance, is an important indicator of its
ability to spread to new water bodies. Invasive macrophytes
differ in their ability to tolerate desiccation (Barnes et al.
2013, Bruckerhoff et al. 2015); thus, individual species need
to be examined to determine spread risk.
Starry stonewort (Nitellopsis obtusa [Desv. in Loisel.] J.
Groves; Characeae) is an invasive macroalga native to
Europe and Asia. The first documented record of starry
stonewort in North America is from the St. Lawrence River
in Quebec, Canada, in 1974 (Karol and Sleith 2017). Starry
stonewort has since been discovered in eight states in the
United States and in two Canadian provinces (Kipp et al.
2018). Starry stonewort can produce substantial biomass
(Schloesser et al. 1986, Glisson et al. 2018) and form tall,
dense beds (Sher-Kaul et al. 1995, Pullman and Crawford
2010, Boissezon et al. 2018). This abundant growth can
hinder recreation, reduce native macrophyte abundance
and richness, and potentially have further ecological
consequences, such as altering water chemistry or fish
habitat (Pullman and Crawford 2010, Brainard and Schulz
2017, Larkin et al. 2018). Starry stonewort’s ecological niche
in North America appears broad and could potentially
expand under future climate change scenarios (Escobar et
al. 2016, Romero-Alvarez et al. 2017, Muthukrishnan et al.
*First, second, third, fourth, fifth, and sixth authors: Research Fellow,
Graduate Student, Graduate Student, Research Fellow, Graduate
Student, and Assistant Professor; Department of Fisheries, Wildlife,
and Conservation Biology & Minnesota Aquatic Invasive Species
Research Center, University of Minnesota, 135 Skok Hall, 2003 Upper
Buford Circle, St. Paul, MN 55108. Current address of fourth author:
Assistant Research Scientist, Environmental Resilience Institute, Indiana
University, 717 East 8th Street, Bloomington, IN 47408. Corresponding
author’s E-mail: wjglisson@gmail.com. Received for publication Febru-
ary 14, 2019 and in revised form May 9, 2019.
J. Aquat. Plant Manage. 58: 2020 7
2018, Sleith et al. 2018). Once starry stonewort is established
in a water body, sustained control of an infestation may be
difficult (Larkin et al. 2018, Glisson et al. 2018).
To guide strategic efforts to prevent starry stonewort
spread, it is important to understand the capacity of starry
stonewort to invade new water bodies via overland
transport. The association of starry stonewort occurrences
with boat launches, high dock density, and nearby infested
lakes (Sleith et al. 2015, Midwood et al. 2016) is consistent
with overland transport via watercraft and trailers being the
dominant means of spread. Desiccation during overland
transport is a critical bottleneck for the spread of AIS.
However, there are currently no studies on the desiccation
tolerance of starry stonewort. Understanding how long
starry stonewort can remain viable out of water, and the
corresponding distance boaters can travel in that time, is
needed to better evaluate spread risk, target interventions
(e.g., watercraft inspection and early detection surveillance),
and prevent spread. Moreover, knowledge of starry stone-
wort desiccation tolerance can determine whether recom-
mended drying times for watercraft are sufficient for starry
stonewort. Species-specific information for starry stonewort
is essential because past studies of macrophyte desiccation
tolerance have been performed on vascular plants, which
are evolutionarily and anatomically distinct from Chara-
cean algae (Raven et al. 2005), and thus may be poor
analogues for estimating drying thresholds.
We conducted controlled experiments, both outdoors
and in the laboratory, to examine desiccation tolerance of
starry stonewort. We tested how long starry stonewort
propagules could remain viable out of water by examining
single fragments (segments of stem-like thalli), small clumps
of fragments, and large clumps of fragments, anticipating
that desiccation would be slower for greater amounts of
material. We also tested desiccation tolerance of starry
stonewort bulbils—asexual reproductive structures formed
belowground that are a key means of spread (Larkin et al.
2018). Because only male starry stonewort has been found in
North America (Larkin et al. 2018), sexual reproduction is
not a known means of spread and oospores were not
evaluated in this study. We evaluated all material over a
wide range of drying times (from 15 min to 5 d). We then
estimated the duration of time required for starry stone-
wort propagules to become fully desiccated and no longer
viable. We conducted two experiments to examine desicca-
tion tolerance of starry stonewort: 1) a climate-controlled
laboratory experiment, and 2) an outdoor experiment
subject to natural weather variation. We conducted the
outdoor experiment first and made minor changes for the
laboratory experiment to improve precision of our results.
As the methods for the laboratory experiment largely
comprise those for the outdoor experiment, we present
the methods for the laboratory experiment first.
MATERIALS AND METHODS
Laboratory desiccation experiment
Assessment of viability. To determine desiccation tolerance
of starry stonewort fragments in the laboratory experiment,
we used three approaches to assess viability. First, we
evaluated the tendency of starry stonewort to lose mass
following removal from water, that is, mass loss following
desiccation, by comparing mass loss of samples removed at
increasing drying-time endpoints to those of fully desiccat-
ed dry controls. Additionally, we examined the rate of mass
loss following desiccation. Next, we determined the ability of
starry stonewort to recover lost mass via rehydration, that is,
mass recovery following rehydration, by comparing unrecovered
mass of rehydrated samples from each drying-time endpoint
to both dry controls (maximum unrecovered mass) and
never-dried wet controls (minimum unrecovered mass).
Last, to supplement the previous two approaches, we
determined the ability of starry stonewort to recover its
physical condition (i.e., color and turgor) via rehydration,
that is, physical recovery following rehydration,byvisually
comparing rehydrated samples to wet controls. We used
the most conservative drying-time endpoint determined by
the first two approaches as the time at which starry
stonewort was no longer viable, and used the third approach
to confirm this result. We did not assess viability of starry
stonewort fragments using direct indicators of growth (e.g.,
thallus growth or rhizoid formation), as it has proven
difficult to grow and maintain starry stonewort cultures
from thalli. Mass loss following desiccation has been shown
to be a significant predictor of, and therefore, a reliable
proxy for macrophyte viability (Evans et al. 2011, Barnes et
al. 2013, Bickel 2015). We directly assessed the viability of
starry stonewort bulbils following desiccation by comparing
sprouting of bulbils removed at each drying-time endpoint
to wet controls.
Experimental setup. We collected live starry stonewort and
bulbils for the laboratory experiment on October 1, 2017
from Lake Koronis in Minnesota (Stearns and Meeker
counties) using a rake attached to a telescoping pole.
Collected material was stored in lake water during transport
and overnight prior to experiments beginning the following
day (all material was used within 24 h of collection).
Experiments were conducted at the Minnesota Aquatic
Invasive Species Research Center’s Containment Laboratory
(MCL) on the University of Minnesota’s St. Paul campus.
Fragments. We examined starry stonewort fragments of
three different size classes: single fragments, small clumps,
and large clumps. Single fragments included three nodes (an
apical node and two nodes along the thallus), were 1020 cm
long, and weighed ~0.5 g (mean wet weight ¼0.49 60.01 g
[1 SE]). We gently aggregated starry stonewort fragments
into small and large clumps of ~5 g (4.84 60.08 g) and ~45
g (45.00 60.26 g), respectively. Small and large clumps were
ovoid; small clumps were approximately 6 33 cm and large
clumps 9 36 cm.
We allowed starry stonewort samples to dry under
controlled conditions and then removed samples at 11
drying-time endpoints (time treatments): 0.25, 0.5, 1, 2, 4, 6,
12, 24, 48, 72, and 120 h. We included a negative control
(never dried, i.e., wet control) and a positive control (dried
to constant mass, i.e., dry control) for each size class. For
each of the 13 total treatments (11 time treatments and wet
and dry controls), we used 15 replicates for single fragments
(N¼195) and 10 replicates for small and large clumps (N¼
8J. Aquat. Plant Manage. 58: 2020
130 each). We desiccated dry controls in paper bags in a
drying oven at 60 C for 3 d.
We conducted the laboratory desiccation experiment in
the MCL inside a 3 33 m mesh tent with a solid floor to
maintain containment. We weighed all samples prior to the
start of the experiment to determine initial weight and then
attached samples to strings strung between two upright PVC
frames within the tent. Samples from each time treatment
were randomly assigned to two strings each and strings were
separated by a height of 3035 cm. We attached single
fragments to strings with a paper clip fastened to the section
of the thallus below the third node (similiar to Bruckerhoff
et al. 2015) and small and large clumps using cable ties
fastened around the center of each clump. Mean temper-
ature in the MCL during the experiment was 24.0 60.17 C
and mean relative humidity (RH) was 43.85 61.05%.
Temperature and RH were similar to those used in other
laboratory-based macrophyte desiccation studies (Jerde et
al. 2012, Mcalarnen et al. 2012, Barnes et al. 2013).
At each drying-time endpoint, we removed samples from
strings, recorded their dried weight to determine mass loss
following desiccation, and immediately placed them in dechlo-
rinated water in glass jars. We recorded the dried weight of
dry controls following removal from the drying oven and
also placed them in water-filled jars. Wet controls were
placed in water-filled jars immediately after initial weighing.
Jars containing the rehydrated samples were randomly
placed on a lab bench and kept under low-light conditions
for 6 h (photosynthetically active radiation [PAR] ¼37 lmol
m
2
s
1
). We then pulled samples from jars and removed
excess water by gently shaking single fragments for 5 s and
placing small and large clumps in a fine-mesh strainer for
510 s. Lastly, we weighed all samples and recorded their
rehydrated weights to determine mass recovery following
rehydration.
After weighing rehydrated samples, we retained a subset
to evaluate physical recovery following rehydration using six 24-h
samples and six wet controls for each size class. We
examined samples from only the 24-h treatments to provide
a confirmation of the other two approaches, mass loss
following desiccation and mass recovery following rehydration, while
balancing the logistical constraints of examining samples
from all treatments. We placed the samples in jars filled with
5 cm of a 50 : 50 potting soil : sand mix and water, covered
the jars with fine mesh, and placed them in clear tanks filled
with water to 10 cm above the tops of the jars. Water was
continuously circulated through the tanks and ~50% of
water volume was replaced each day. After 3 wk, we
removed the jars and visually inspected each sample for
the presence of green, turgid thalli, as an indicator of
viability (following Barnes et al. 2013 for Ceratophyllum
demersum, and Baniszewski et al. 2016 for Hydrilla verticillata).
Bulbils. We separated bulbils from rhizoids of starry
stonewort collected from Lake Koronis and selected those
that appeared robust and were large enough to observe
easily (mean wet weight ¼0.02 60.01 g [1 SE]). We used 12
bulbils as replicates for each of the 11 time treatments and
wet and dry controls (N¼156). We placed bulbils into 50-
mm petri dishes on a table next to the tent used for
fragments and randomly assigned petri dishes to time
treatments. We desiccated dry controls in paper coin
envelopes in a drying oven at 60 C for 3 d.
At each drying-time endpoint, we removed bulbils from
petri dishes, weighed them (all 12 bulbils for each treatment
were weighed together for a single treatment weight), and
immediately placed them into 4-oz. glass jars filled with 2 cm
of a 50 : 50 potting soil : sand mix and water. Bulbils were
gently pushed into the sediment until fully covered. Each jar
contained four bulbils from a given treatment (three jars
per treatment). Jars containing bulbils were covered with
fine mesh and placed into three clear plastic tanks filled
with water to 10 cm above the tops of the jars. We recorded
the dried weight of dry controls following removal from the
drying oven and placed them in jars in the same manner as
time treatments. Wet controls were placed in jars immedi-
ately after initial weighing. We randomized the order of jars
in tanks such that each of the three tanks contained one jar
from each time treatment and wet and dry control. We
maintained tanks under a 14 h/10 h light/dark schedule (PAR
¼37 lmol m
2
s
1
at the water’s surface). Mean water
temperature in tanks during the experiment was 16.9 C.
We checked bulbils for sprouting every 37 d for a total
of 8 wk (56 d), which previous research had shown to be
sufficient for most bulbils to sprout under similar condi-
tions (Glisson et al. 2018). A bulbil was considered sprouted
when a green shoot emerged from the bulbil. Sprouted
bulbils were removed from jars to avoid duplicate counting.
At the end of the experiment, we were unable to recover all
bulbils that we had initially placed in jars. Unrecovered
bulbils likely broke apart or decomposed over the course of
the experiment and were thus considered nonviable
(following Glisson et al. 2018).
Outdoor desiccation experiment
Assessment of viability. To determine desiccation tolerance
of starry stonewort fragments in the outdoor experiment,
we used two approaches to assess viability. First, we
determined mass loss following desiccation by comparing mass
loss of samples removed at increasing drying-time end-
points to those of 72-h treatments, which we assumed to be
fully desiccated. Next, we determined mass recovery following
rehydration by comparing unrecovered mass of rehydrated
samples from each drying-time endpoint to both 72-h
treatments that served as dry controls (maximum unrecov-
ered mass), and never-dried wet controls (minimum
unrecovered mass). We used the most conservative drying-
time endpoint determined by these approaches as the time
at which starry stonewort was no longer viable. To assess
viability of starry stonewort bulbils following desiccation,
we used a chemical stain (following Gottschalk and Karol
2020, this issue; see the following) to compare viability of
bulbils from each drying-time endpoint to those in the 72-h
treatments.
Fragments. The methods for the outdoor desiccation
experiment were generally similar to those used in the
laboratory experiment; only methods that differed from the
laboratory experiment are described here. We collected live
starry stonewort and bulbils for the outdoor experiment on
August 7, 2017 from Lake Koronis. Single fragments
J. Aquat. Plant Manage. 58: 2020 9
weighed ~0.5 g (mean ¼0.58 60.02 g [1 SE]), small clumps
~5 g (5.64 60.09 g), and large clumps ~20 g (20.41 60.19
g). We used eight drying-time endpoints (time treatments):
1, 2, 4, 6, 12, 24, 48, and 72 h. We employed only a wet
control and we used 10 replicates for each treatment (N¼90
for each size class).
We conducted the outdoor desiccation experiment on a
paved surface outside the MCL within the same mesh tent
used in the laboratory experiment. We weighed all samples
initially and then randomly assigned and attached them to
strings, separated by a height of 40 cm, tied to PVC frames
within the tent. Temperature and relative humidity fluctu-
ated over the course of the experiment (Appendix). It
rained several times during the experiment and samples
were not protected from getting wet; however, all time
treatments 24 h were removed prior to any rain. At each
drying-time endpoint, we removed samples from strings,
recorded their dried weight to determine mass loss following
desiccation, and placed them in water in glass jars for 6 h
under low-light conditions (PAR ¼37 lmol m
2
s
1
). Wet
controls were placed in water-filled jars immediately after
initial weighing. We then pulled samples from jars, removed
excess water, and weighed all samples and recorded
rehydrated weight to determine mass recovery following
rehydration.
Bulbils. We used 10 bulbils as replicates for each of the
eight time treatments and wet controls (N¼90). We placed
bulbils into 50-mm petri dishes on a table within the tent
and haphazardly assigned petri dishes to time treatments. At
each drying-time endpoint, we removed bulbils from petri
dishes and immediately placed them into 4-oz. glass jars
filled with water. Jars were randomly placed on a lab bench
and kept under low-light conditions for 6 h (PAR ¼27 lmol
m
2
s
1
).
To determine bulbil viability, we used a staining
compound commonly used for assessment of vascular plant
tissue viability: 2,3,5-triphenyl-tetrazolium chloride (TZ). In
live cells, TZ is reduced to formazan, a reddish compound
that is not water soluble; this results in live cells being
stained red (Parker 1953). Staining with TZ is commonly
used to assess angiosperm seed and tissue viability, has
previously been used to assess algae viability (Nam et al.
1998, Calomeni et al. 2014), and has been effective to
determine viability of starry stonewort bulbils (following
Gottschalk and Karol 2020, this issue). Following rehydra-
tion, we placed all bulbils from each treatment into amber
glass jars with 25 ml of 0.1% TZ solution. Jars containing
bulbils were stored in a refrigerator for 1824 h, after which
bulbils were examined under a dissecting scope for signs of
red staining indicative of live cells. Starry stonewort bulbils
are composed of multiple cells; hence, we considered bulbils
with red stain visible on at least one cell to be viable.
Statistical analysis
Fragments. All analyses were conducted using R statistical
software, version 3.4.3 (R Development Core Team 2017).
Analyses were identical between the laboratory experiment
and outdoor experiment, except that 8 time treatments
were used in the outdoor experiment versus 11 in the
laboratory experiment, and the 72-h treatments in the
outdoor experiment were used as proxies for dry controls.
Lastly, only the first two approaches described below, mass
loss following desiccation and mass recovery following rehydration,
were used to assess viability in the outdoor experiment.
To determine mass loss following desiccation, we compared
the proportion of mass lost following desiccation for each
time treatment ([initial weight – dried weight]/initial weight)
to dry controls. We used analysis of variance (ANOVA) to
examine differences among treatments, with proportion of
mass lost as the response variable and treatment (12
levels : 11 time treatments and dry control) as the predictor
variable. We performed separate analyses for single frag-
ments, small clumps, and large clumps. We then used
Tukey’s honest significant differences (Tukey’s HSD) to
compare each time treatment to the dry control. This
allowed us to identify the time point(s) when starry
stonewort in our time treatments no longer significantly
differed from fully desiccated controls.
To determine the rate of mass loss following desiccation,we
used nonlinear least squares (NLS) regression to fit a three-
parameter asymptotic exponential function to our data. We
chose this function because we expected starry stonewort
fragments in our experiment to reach an asymptote at
which increased drying time does not lead to further mass
loss, and because it has been effectively employed in
previous macrophyte desiccation experiments (Bickel
2015). We used the equation y¼a–be
cx
, where ais the
asymptote, abis the intercept, and cis the rate constant.
Starting values are required for NLS models; we provided a
starting value of 1 for both aand b. To determine a starting
value for c, we used xand yvalues for a point likely to fall
along the line of best fit and rearranged the equation above
to solve for c(Crawley 2007). We fit NLS models for single
fragments, small clumps, and large clumps, with the
proportion of mass lost following desiccation as the
response variable, and time (in minutes) as the continuous
predictor variable. Hence, we did not include controls for
this analysis. We determined starting values for cusing mean
mass loss (y) at 30 min, 12 h, and 48 h (x) for single
fragments, small clumps, and large clumps, respectively.
Each NLS model estimated an intercept, asymptote, and a
rate constant. We considered starry stonewort fully desic-
cated when the proportion of mass lost predicted from the
model was within 1 SE of the asymptote. We compared rate
constants for each size class and considered these signifi-
cantly different if their 95% confidence intervals (CI) did
not overlap.
Next, to determine mass recovery following rehydration,we
compared the proportion of mass not recovered following
rehydration (i.e., unrecovered mass) for each time treatment
([initial weight – rehydrated weight]/initial weight) to wet
and dry controls. The two controls served as minimum (wet
control) and maximum (dry control) benchmarks for
unrecovered mass. The same ANOVA and Tukey’s HSD
analyses described above were repeated for this analysis (for
single fragments, small clumps, and large clumps) with
proportion of unrecovered mass as the response variable
and treatment (13 levels: 11 time treatments, wet control,
and dry control) as the predictor variable. This allowed us to
10 J. Aquat. Plant Manage. 58: 2020
identify time point(s) when unrecovered mass in rehydrated
starry stonewort was significantly greater than wet controls
and as much as (i.e., not significantly different from) fully
desiccated dry controls.
Last, to determine physical recovery following rehydration,we
visually compared rehydrated 24-h treatments for single
fragments, small clumps, and large clumps to wet controls.
We scored each sample (consisting of either a single
fragment, a small clump, or a large clump) as containing
either 1 green, turgid fragment(s) (viable), or no green,
turgid fragment(s) (nonviable). We scored and photo-
graphed each sample to document its condition (images
available upon request).
Bulbils. We determined the viability of starry stonewort
bulbils following desiccation in the laboratory experiment
by identifying the time point(s) when bulbils no longer
sprouted and when fewer time treatment bulbils sprouted
than did wet control bulbils. We analyzed bulbil-sprouting
data with a generalized linear model (GLM) with binomial
errors and logit link to determine the time point(s) when
significantly fewer time treatment bulbils sprouted than
wet control bulbils. We used the proportion of bulbils
sprouted as the response variable (expressed as counts of
sprouted and unsprouted bulbils with n¼12 for each
treatment) and treatment (13 levels: 11 time treatments,
wet control, and dry control) as the predictor variable. We
also examined the proportion of bulbil mass lost following
desiccation to estimate the mass loss associated with no
further sprouting.
RESULTS AND DISCUSSION
Laboratory desiccation experiment
Fragments. In the laboratory experiment, starry stonewort
fragments in fully desiccated dry controls lost 90% of
their mass: 90% for single fragments, 91% for small clumps,
and 91% for large clumps (Figure 1). Following rehydration,
dry controls regained 38% of that lost mass: 55% for
single fragments, 50% for small clumps, and 38% for large
clumps (Figure 2).
Based on mass loss following desiccation, single fragments
were fully desiccated (i.e., the proportion of mass lost no
longer significantly differed from dry controls) at the 2-h
time point, small clumps at the 24-h time point, and large
clumps at the 72-h time point (Figure 1).
The rate of mass loss following desiccation significantly
differed by size class, with single fragments having the
greatest rate of mass loss, followed by small clumps and then
large clumps (Table 1, Figure 3). Based on rate constant and
asymptote estimates from NLS models, single fragments
were fully desiccated at 1 h 25 min, small clumps at 20 h 4
min, and large clumps at 3 d 23 h 57 min (Table 1, Figure 3).
Based on mass recovery following rehydration, single frag-
ments could no longer recover from desiccation (i.e., the
proportion of unrecovered mass was significantly greater
than wet controls, but not significantly different from dry
controls) by the 30-min time point, small clumps by the 12-h
time point, and large clumps by the 72-h time point (Figure
2).
In terms of physical recovery following rehydration, none of
the 24-h single-fragment or small-clump treatments con-
tained viable material, whereas all wet controls did. For
large clumps, five of six (83%) 24-h treatment jars and all six
wet control jars contained viable material.
Based on the three approaches used to assess viability,
starry stonewort in the laboratory desiccation experiment
was no longer viable at 2 h for single fragments, 24 h for
small clumps, and 4 d for large clumps (Table 2).
Bulbils.After8wk,1 bulbil(s) sprouted in the wet control
(9 sprouted bulbils) and 15-min (5), 30-min (11), 1-h (9), and 2-
h (1) treatments (Figure 4a). We observed no sprouting in any
time treatments 4 h or the dry controls. For the treatments
in which bulbils did sprout, sprouting in the 2-h treatment was
significantly lower than the wet control (P¼0.0062). Hence,
Figure 1. Proportion of starry stonewort mass lost following desiccation at
each drying-time endpoint (treatment) in the laboratory experiment for (a)
single fragments, (b) small clumps, and (c) large clumps. Points and error
bars are means 61 SE. Dry controls (DC) were fully desiccated samples.
Dashed gray line is the proportion of mass lost in the dry controls. An
asterisk indicates a significant difference between that time treatment and
the dry control.
J. Aquat. Plant Manage. 58: 2020 11
starry stonewort bulbils in the laboratory desiccation exper-
iment were no longer viable at 4 h. Starry stonewort bulbils in
the dry controls lost 74% of their mass; no bulbils sprouted
when 59% of their mass was lost (Figure 4b).
Outdoor desiccation experiment
In the outdoor experiment, starry stonewort fragments in
the fully desiccated 72-h treatments lost 93% of their
mass: 94% for single fragments, 93% for small clumps, and
93% for large clumps (Figure 5). Following rehydration, the
72-h treatments regained 36% of that lost mass: 36% for
single fragments, 50% for small clumps, and 48% for large
clumps (Figure 6).
Based on mass loss following desiccation, single fragments
were fully desiccated (i.e., the proportion of mass lost no
longer significantly differed from the 72-h treatments) at
the 1-h time point, small clumps at the 2-h time point, and
large clumps at the 6-h time point (Figure 5). For both small
and large clumps, the 48-h treatment significantly differed
from the 72-h treatment (Figure 5); however, the 48-h
treatment was removed after it had rained (see Appendix).
Based on mass recovery following rehydration, for single
fragments, unrecovered mass of rehydrated starry stonewort
was significantly greater than wet controls in the 1-h
Figure 2. Proportion of starry stonewort mass not recovered following
rehydration at each drying-time endpoint (treatment) in the laboratory
desiccation experiment for (a) single fragments, (b) small clumps, and (c)
large clumps. Points and error bars are means 61 SE. Wet controls (WC)
were samples that were never dried, and dry controls (DC) were fully
desiccated samples. Dashed gray lines are the proportion of mass not
recovered in the dry controls (top line) and wet controls (bottom line). One
asterisk indicates a significant difference between that time treatment or
control and the dry control; two asterisks indicate a significant difference
between that time treatment and both the wet control and dry control.
Treatments significantly different from wet controls, but not significantly
different from the dry controls (no asterisks), were considered unable to
recover from desiccation.
TABLE 1. ESTIMATES OF RATE CONSTANTS (61 SE), 95% CONFIDENCE INTERVALS (CI) FOR RATE CONSTANTS,AND ASYMPTOTES (PROPORTION OF MASS LOST 61 SE) FROM
NONLINEAR LEAST-SQUARES REGRESSION MODELS FOR STARRY STONEWORT DESICCATION IN THE LABORATORY EXPERIMENT.DESICCATION THRESHOLD (ASYMPTOTE – 1 SE) IS THE
PROPORTION OF MASS LOST AT WHICH THE MATERIAL WAS CONSIDERED NONVIABLE.DIFFERENT LETTERS INDICATE SIGNIFICANT DIFFERENCES.
Sample Rate Constant 95% CI Asymptote Desiccation Threshold
Single fragment 0.0733 (0.0036)
a
0.066–0.081 0.901 (0.003) 0.898
Small clump 0.0034 (0.0002)
b
0.003–0.004 0.894 (0.013) 0.881
Large clump 0.0008 (0.0000)
c
0.001–0.001 0.913 (0.008) 0.905
Figure 3. Rates of starry stonewort mass loss over time in the laboratory
desiccation experiment for single fragments (light gray; diamonds), small
clumps (gray; circles), and large clumps (black; triangles). Rate of mass loss
was determined by a three-parameter asymptotic exponential function fit
by nonlinear least squares regression.
12 J. Aquat. Plant Manage. 58: 2020
treatment but, unexpectedly, not in the 2-h treatment; all
single-fragment time treatments 1 h did not significantly
differ from the fully desiccated 72-h treatment (Figure 6).
For small clumps, unrecovered mass of rehydrated starry
stonewort was significantly greater than the wet controls in
all treatments 1 h. Except for the 12-h treatment, small-
clump time treatments 6 h did not significantly differ
from the fully desiccated 72-h treatment (Figure 6). For
large clumps, unrecovered mass of rehydrated starry
stonewort was significantly greater than the wet controls
in all time treatments 1 h. Large-clump time treatments
6 h did not significantly differ from the fully desiccated
72-h treatment, except for the 48-h treatment that was
exposed to rain (Figure 6, Appendix).
In general, for unrecovered mass (i.e., mass recovery
following rehydration), there was greater variance within
treatments (Figure 6) and less clarity regarding the time at
which starry stonewort was no longer viable compared to
the examination of mass lost (i.e., mass loss following
desiccation: Figure 5). Variation among treatments could be
explained by changes in weather conditions over the course
of the experiment (e.g., rain prior to removal of the 48-h
treatments), whereas variation within or among treatments
could be explained by inconsistency in removing excess
water from samples before weighing. Given the overall
pattern of responses to desiccation, we believe the
inconsistent data for unrecovered mass (2-h treatment for
single fragments and 12-h treatment for small clumps;
Figure 6) reflects the latter explanation—variation in how
these samples were handled in the experiment—rather than
actual starry stonewort response to desiccation. Hence, we
chose to consider the first treatment for which the
unrecovered mass criteria were met (1-h treatment for
single fragments and 6-h treatment for small clumps) as the
time point to assess viability using the mass recovery following
rehydration approach.
Based on the two approaches to assess viability, starry
stonewort in the outdoor desiccation experiment was no
longer viable at 1 h for single fragments, 6 h for small
clumps, and 6 h for large clumps (Table 2). We observed red
staining from the reduction of TZ on 6 of 10 wet control
bulbils (60%) and none of the time treatment bulbils (1
h), indicating bulbils were no longer viable at 1 h.
Synthesis
Compared to other aquatic invasive plant species that
have been evaluated, starry stonewort appears to have lower
desiccation tolerance. Single fragments of starry stonewort
in both the laboratory and outdoor experiments lost
viability more quickly than similarly sized fragments of
Eurasian watermilfoil (Myriophyllum spicatum; Evans et al.
2011), Carolina fanwort (Cabomba caroliniana; Bickel 2015),
curly-leaf pondweed (Potamogeton crispus; Bruckerhoff et al.
2015), and hydrilla (Hydrilla verticillata; Baniszewski et al.
2016) in passive desiccation studies (i.e., studies that did not
simulate wind exposure associated with overland travel).
The duration of time that single fragments of starry
stonewort remained viable was similar to that reported for
other invasive macrophytes subjected to active drying (i.e.,
fan-dried at 7.5 mph), including Eurasian watermilfoil,
Carolina fanwort, and curly-leaf pondweed (Jerde et al.
2012, Barnes et al. 2013). We would expect starry stonewort
to desiccate substantially faster if actively dried, as wind
TABLE 2. RESULTS FROM LABORATORY AND OUTDOOR DESICCATION EXPERIMENTS ON STARRY STONEWORT,INCLUDING THE EXPERIMENT,TYPE OF SAMPLE EXAMINED,MEAN WEIGHT OF
THE SAMPLES (61 SE), THE APPROACH USED TO ASSESS VIABILITY,AND TIME AT WHICH STARRY STONEWORT WAS DETERMINED TO BE NO LONGER VIABLE.
Experiment Sample Type Mean Weight (g) Approach Time At Which No Longer Viable
Laboratory Single fragment 0.49 (0.01) Mass loss 2 h
Rate of mass loss 1 h 25 min
Mass recovery 30 min
Physical recovery 24 h
Small clump 4.84 (0.08) Mass loss 24 h
Rate of mass loss 20 h 4 min
Mass recovery 12 h
Physical recovery 24 h
Large clump 45.00 (0.26) Mass loss 72 h
Rate of mass loss 3 d 23 h 57 min
Mass recovery 72 h
Physical recovery .24 h
Bulbils 0.02 (0.01) Sprouting 4 h
Outdoor Single fragment 0.58 (0.02) Mass loss 1 h
Mass recovery 1 h
Small clump 5.64 (0.09) Mass loss 2 h
Mass recovery 6 h
Large clump 20.41 (0.19) Mass loss 6 h
Mass recovery 6 h
Bulbils NA TZ staining 1 h
J. Aquat. Plant Manage. 58: 2020 13
speed can greatly increase desiccation rate (Bickel 2015).
Regardless, passively dried single fragments of starry
stonewort in our laboratory experiment became fully
desiccated faster than five other aquatic plant species
exposed to 7.5-mph wind speed by Barnes et al. (2013),
and fragments in our outdoor experiment desiccated faster
than all nine species examined in their study.
Comparison of starry stonewort desiccation to other
species should be interpreted in light of macrophyte
desiccation experiments varying in the type (e.g., apical vs.
basal), length, node number, and mass of fragments
evaluated, all of which can influence desiccation rate and
viability (Mcalarnen et al. 2012; Bickel 2015, 2017; Li et al.
2015). These factors may affect comparability among studies
and species. Our single fragments likely had lower mass than
fragments used in similar studies with other species, but
they were similar in length to fragments of Eurasian
watermilfoil (Jerde et al. 2012, Mcalarnen et al. 2012, Barnes
et al. 2013) and similar in length and number of nodes to
fragments of Carolina fanwort (Bickel 2015) used in other
studies. We also used apical stems in our experiment and
these stems can have greater viability and slower desiccation
rates than basal stems (Mcalarnen et al. 2012). Thus, the
estimates for desiccation times and loss of viability we
determined may be conservative; that is, basal starry
stonewort fragments may have indicated even lower
desiccation tolerance.
As an alga, starry stonewort has morphological charac-
teristics that likely make fragments less tolerant to
desiccation than other macrophytes. Namely, starry stone-
wort lacks the multicellular tissues found in vascular plants
(i.e., vascular, dermal, and ground tissues; Raven et al. 2005).
Figure 4. Response of starry stonewort bulbils to desiccation in the
laboratory experiment, including proportion of (a) bulbils sprouted, and (b)
bulbil mass lost following desiccation at each drying-time endpoint
(treatment). Wet controls (WC) were bulbils that were never dried, and
dry controls (DC) were fully desiccated bulbils. Dashed gray line is the
lowest proportion of bulbil mass lost at which bulbils no longer sprouted (6-
h treatment). Wet control bulbils were weighed prior to the experiment
only, and hence, not included in (b). For (a), an asterisk indicates a
significant difference between that time treatment and the wet control.
Figure 5. Proportion of starry stonewort mass lost following desiccation at
each drying-time endpoint (treatment) in the outdoor desiccation
experiment for (a) single fragments, (b) small clumps, and (c) large clumps.
Points and error bars are means 61 SE. Dashed gray line is the proportion
of mass lost in the 72-h treatment. An asterisk indicates a significant
difference between that time treatment and the 72-h treatment.
14 J. Aquat. Plant Manage. 58: 2020
The presence of these multicellular tissues in the vascular
macrophytes typically evaluated in similar desiccation
studies (e.g., Barnes et al. 2013) likely allows these plants
to retain moisture longer. Starry stonewort has only a single
elongated cell between nodes (up to 24 cm long; Steudle et
al. 1977, Bharathan 1983). The branchlets of starry
stonewort, which form whorls at each node (similar in
appearance to the leaves of a vascular plant), are also
primarily composed of single elongated cells (Bharathan
1983, Boissezon et al. 2018). Hence, the bulk of starry
stonewort cells have very high surface area–to–volume
ratios; this characteristic, along with the high water
permeability of internodal starry stonewort cells (Yoshioka
and Takenaka 1979), likely facilitates rapid desiccation
following removal from water. Moreover, the single elon-
gated cells of starry stonewort are easily broken or crushed
when handled, releasing the cells’ cytoplasm and rendering
them nonviable. We handled fragments carefully in our
experiments, but it is likely that fragments would encounter
greater force when snagged on a motor or trailer.
Starry stonewort desiccated approximately twice as
quickly in the outdoor experiment than the laboratory
experiment. Although the large clumps in the outdoor
experiment weighed an average of 25 g (56%) less than
those in the laboratory experiment, they became fully
desiccated in 10% of the time. The temperature on the
first day of the outdoor experiment—when starry stonewort
of all size classes was fully desiccated—was only slightly
higher than the mean temperature in the laboratory
experiment (25.2 versus 24.0 C). Relative humidity was also
only slightly higher on the first day of the outdoor
experiment (47.0 versus 44.9%). Thus, it is unlikely that
temperature and humidity account for the differences in
desiccation tolerance we observed between the outdoor and
laboratory experiments. Rather, exposure to wind and
sunlight (Appendix), that is, conditions more representative
of those encountered during overland transport, are likely
the main factors that lowered desiccation tolerance in the
outdoor experiment. Future controlled experiments that
vary weather conditions (e.g., wind, humidity, temperature,
and sunlight exposure) could help determine the role of
these factors on starry stonewort desiccation tolerance and
the ability to survive overland transport under realistic
scenarios.
Single fragments of starry stonewort desiccated quickly,
but clumps of starry stonewort remained viable for far
longer. This is consistent with other desiccation studies that
have examined clumps or coils of macrophyte stems. For
example, coiling of Eurasian watermilfoil stems increased
desiccation times from 24 to 72 h (Bruckerhoff et al. 2015),
and clumping of Carolina fanwort stems increased desicca-
tion times from 9 to 42 h (Bickel 2015); we observed a
similar pattern for starry stonewort. This is unsurprising,
given the capacity of larger clumps to retain more moisture
because of lower surface area–to–volume ratio.
Starry stonewort bulbils dried quickly in our experi-
ments; bulbils were no longer viable within 1 h in the
outdoor experiment and 4 h in the laboratory experiment.
These times are substantially less than those reported for
the asexual reproductive structures (turions) of curly-leaf
pondweed, which can survive .4wkoutofwater
(Bruckerhoff et al. 2015) or seeds of Eurasian watermilfoil,
which can survive .8 mo out of water (Standifer and
Madsen 1997). It is not surprising that starry stonewort
bulbils dried more quickly than these reproductive struc-
tures, as the regenerative cells of bulbils lack the hard
protective coatings found in seeds or turions of other
aquatic plants (Bharathan 1987). Starry stonewort bulbils
sprout quickly and in high proportion (this study; Bhar-
athan 1987, Glisson et al. 2018), but their susceptibility to
desiccation may limit their potential to spread and form
large infestations. However, we did not examine desiccation
Figure 6. Proportion of starry stonewort mass not recovered following
rehydration at each drying-time endpoint (treatment) in the outdoor
desiccation experiment for (a) single fragments, (b) small clumps, and (c)
large clumps. Points and error bars are means 61 SE. Wet controls (WC)
were samples that were never dried. Dashed gray lines are the proportion of
mass not recovered in the 72-h treatments (top line) and the wet control
(bottom line). One asterisk indicates a significant difference between that
time treatment or control and the 72-h treatment; two asterisks indicate a
significant difference between that time treatment and both the wet control
and the 72-h treatment; ns indicates a time treatment that was not
significantly different from the wet control or the 72-h treatment.
Treatments significantly different from wet controls, but not significantly
different from the 72-h treatment (no asterisks), were considered unable to
recover from desiccation.
J. Aquat. Plant Manage. 58: 2020 15
tolerance of starry stonewort bulbils embedded in clumps,
attached to rhizoids, or associated with sediment. Bulbils
under such conditions can survive out of water longer (12–
24 h; Gottschalk and Karol 2020, this issue), likely because of
buffering of moisture loss.
CONCLUSION
Our findings provide estimates for how long starry
stonewort fragments and bulbils can remain viable out of
water, a key indicator of overland spread risk. These
estimates can guide prevention protocols and guidelines
for watercraft leaving starry stonewort–infested water
bodies. The relatively low desiccation tolerance we observed
is encouraging in terms of preventing the spread of starry
stonewort via overland transport. The smallest pieces of
starry stonewort—that is, single fragments and individual
bulbils most likely to be missed during visual inspection—
quickly lost viability. Boaters traveling to multiple lakes on
the same day may unintentionally move viable single
fragments and bulbils to new water bodies, but it is unlikely
that these propagules would survive if a boat were kept out
of water for a day or more. Although small and large clumps
remained viable for longer, the size of these clumps makes
them easier to locate during inspection—particularly the
large clumps that retained viability longest. Furthermore,
our findings from the outdoor experiment demonstrate that
even these large clumps of starry stonewort can dry rapidly
under warm, sunny conditions with natural air movement.
Fragments and clumps would likely desiccate even faster if
exposed to the wind experienced during vehicle travel
(Bickel 2015). For the masses of starry stonewort and
conditions evaluated in this study, adherence to the 5-d
drying time recommended by state agencies such as the
Minnesota Department of Natural Resources (Minnesota
Department of Natural Resources [DNR] 2018a) should
ensure that starry stonewort is no longer viable before
entering another water body.
Despite these findings of starry stonewort’s relatively low
desiccation tolerance, this species is nevertheless expanding
in its invaded range (Kipp et al. 2018, Minnesota DNR
2018b). Although its spread may be constrained by lower
desiccation tolerance, opportunities for spread are clearly
occurring. Rare, long-distance dispersal events can have a
substantial influence on the spread of invasive plant species
(Higgins and Richardson 1999). Such opportunities could
arise when boaters do not comply with requirements to
inspect and remove plants from watercraft and trailers
(Rothlisberger et al. 2010, Cimino and Strecker 2018).
Fragments and bulbils missed during inspection that are
protected from moisture loss or left submerged in residual
water will likely remain viable much longer, providing
increased opportunity for spread. For example, starry
stonewort propagules caught between a boat hull and
carpeted bunk trailer are likely to escape detection and,
due to buffering of moisture loss, could potentially remain
viable for substantially longer periods than the material
observed in our experiments. We did not examine algal
material protected from moisture loss in our experiments,
but this could be examined in future work for a better
understanding of starry stonewort’s ability to spread via
overland transport. Moreover, there is still a chance that
stems left on a boat or trailer that become fully desiccated
and appear nonviable, are in fact viable (e.g., Evans et al.
2011). Thus, relying solely on the tendency of starry
stonewort, or any AIS, to dry out quickly is insufficient for
spread prevention. Visual inspection and hand removal,
however, are effective means to remove invasive aquatic
plants from watercraft (Rothlisberger et al. 2010) and should
be used at a minimum when leaving and entering a water
body. If such inspection protocols are followed, starry
stonewort masses larger than those evaluated in this study
(and hence, potentially able to remain viable longer) should
be readily detected by boaters. More aggressive strategies
such as steam treatments may also be effective at reducing
starry stonewort viability and spread (Gottschalk and Karol
2020, this issue).
Given that starry stonewort infestations can be large and
difficult to manage (Glisson et al. 2018), it is crucial to
inspect and remove this alga before it can be transported to
a new water body. Maps of invasion risk for starry stonewort
have been developed in the upper Midwest United States
(Muthukrishnan et al. 2018) and these can be used with the
findings reported here to focus watercraft inspection
efforts. These efforts, combined with outreach programs
and compliance by individual boaters, are vital for
preventing the spread of starry stonewort.
ACKNOWLEDGEMENTS
Funding for this research was provided through the
Minnesota Aquatic Invasive Species Research Center from
the Minnesota Environment and Natural Resources Trust
Fund. We thank the Associate Editor and two anonymous
reviewers for comments that greatly improved the manu-
script.
LITERATURE CITED
Baniszewski J, Cuda JP, Gezan SA, Sharma S, Weeks ENI. 2016. Stem
fragment regrowth of Hydrilla verticillata following desiccation. J. Aquat.
Plant Manage. 54:53–60.
Barnes MA, Jerde CL, Keller D, Chadderton WL, Howeth JG, Lodge DM.
2013. Viability of aquatic plant fragments following desiccation. Invasive
Plant. Sci. Manage. 6:320–325.
Bharathan S. 1983. Developmental morphology of Nitellopsis obtusa (Desv.)
Groves. Proc. Plant. Sci. 92:373–379.
Bharathan S. 1987. Bulbils of some charophytes. Proc. Indian Acad. Sci.
Plant. Sci. 97:257–263.
Bickel TO. 2015. A boat hitchhiker’s guide to survival: Cabomba caroliniana
desiccation resistance and survival ability. Hydrobiologia 746:123–134.
Bickel TO. 2017. Processes and factors that affect regeneration and
establishment of the invasive aquatic plant Cabomba caroliniana. Hydro-
biologia 788:157–168.
Boissezon A, Auderset Joye D, Garcia T. 2018. Temporal and spatial changes
in population structure of the freshwater macroalga Nitellopsis obtusa
(Desv.) J. Groves. Bot. Lett. 165:103–114.
Brainard AS, Schulz KL. 2017. Impacts of the cryptic macroalgal invader,
Nitellopsis obtusa, on macrophyte communities. Freshwater Sci. 36:55–62.
Bruckerhoff L, Havel J, Knight S. 2015. Survival of invasive aquatic plants
after air exposure and implications for dispersal by recreational boats.
Hydrobiologia 746:113–121.
Buchan LAJ, Padilla DK. 1999. Estimating the probability of long-distance
overland dispersal of invading aquatic species. Ecol. Appl. 9:254–265.
16 J. Aquat. Plant Manage. 58: 2020
Calomeni A, Rodgers JH, Kinley CM. 2014. Responses of Planktothrix agardhii
and Pseudokirchneriella subcapitata to copper sulfate (CuSO4 5H2O) and
a chelated copper compound (Cutrine-Ultra). Water, Air, Soil Pollut.
225:2231.
Cimino SA, Strecker AL. 2018. Boater knowledge and behavior regarding
aquatic invasive species at a boat wash station. Northwest Sci. 92:224–
233.
Crawley MJ. 2007. The R book. John Wiley & Sons, Chichester, England.
Escobar LE, Qiao H, Phelps NBD, Wagner CK, Larkin DJ. 2016. Realized
niche shift associated with the Eurasian charophyte Nitellopsis obtusa
becoming invasive in North America. Sci. Rep. 6:29037
Evans CA, Kelting DL, Forrest KM, Steblen LE. 2011. Fragment viability and
rootlet formation in Eurasian watermilfoil after desiccation. J. Aquat.
Plant Manage. 48:57–62.
GlissonWJ,WagnerCK,McComasSR,FarnumK,VerhoevenMR,
Muthukrishnan R, Larkin DJ. 2018. Response of the invasive alga starry
stonewort (Nitellopsis obtusa) to control efforts in a Minnesota lake. Lake
Reservoir Manage. 34:283–295.
Gottschalk SD, Karol KG. 2020. Survivability of starry stonewort bulbils
using commonly available decontamination strategies. J. Aquat. Plant.
Manage. 58:xx–xx.
Green AJ. 2016. The importance of waterbirds as an overlooked pathway of
invasion for alien species. Diversity Distrib. 22:239–247.
Higgins SI, Richardson DM. 1999. Predicting plant migration rates in a
changing world: the role of long distance dispersal. Am. Nat. 153:464–
475.
Jerde CL, Barnes MA, DeBuysser EK, Noveroske A, Chadderton WL, Lodge
DM. 2012. Eurasian watermilfoil fitness loss and invasion potential
following desiccation during simulated overland transport. Aquat.
Invasions 7:135–142.
Johnson LE, Carlton JT. 1996. Post-establishment spread in large-scale
invasions: dispersal mechanisms of the zebra mussel Dreissena polymorpha.
Ecology 77:1686–1690.
Johnson LE, Ricciardi A, Carlton JT. 2001. Overland dispersal of aquatic
invasive species: a risk assessment of transient recreational boating.
Ecol. Appl. 11:1789–1799.
JohnstoneIM,CoffeyBT,Howard-WilliamsC.1985.Theroleof
recreational boat traffic in interlake dispersal of macrophytes: A New
Zealand case study. J. Environ. Manage. 20:263–279.
Karol KG, Sleith RS. 2017. Discovery of the oldest record of Nitellopsis obtusa
(Charophyceae, Charophyta) in North America. J. Phycol. 53:1106–1108.
Kipp RM, McCarthy M, Fusaro A, Pfingsten IA. 2018. Nitellopsis obtusa
(Desvaux in Loiseleur) J. Groves, (1919): U.S. Geological Survey,
Nonindigenous Aquatic Species Database.
Larkin DJ, Monfils AK, Boissezon A, Sleith RS, Skawinski PM, Welling CH,
Cahill BC, Karol KG. 2018. Biology, ecology, and management of starry
stonewort (Nitellopsis obtusa; Characeae): A red-listed Eurasian green alga
invasive in North America. Aquat. Bot. 148:15–24.
Li F, Zhu L, Xie Y, Jiang L, Chen X, Deng Z, Pan B. 2015. Colonization by
fragments of the submerged macrophyte Myriophyllum spicatum under
different sediment type and density conditions. Sci. Rep. 5:11821.
Mcalarnen LA, Barnes MA, Jerde CL, Lodge DM. 2012. Simulated overland
transport of Eurasian watermilfoil: survival of desiccated plant
fragments. J. Aquat. Plant Manage. 50:147–149.
Midwood JD, Darwin A, Ho Z, Rokitnicki-Wojcik D, Grabas G. 2016.
Environmental factors associated with the distribution of non-native
starry stonewort (Nitellopsis obtusa) in a Lake Ontario coastal wetland. J.
Great Lakes Res. 42:348–355.
Minnesota Department of Natural Resources [DNR]. 2018a. Clean in clean
out: When boating or fishing in Minnesota, protect your waters by
following state aquatic invasive species laws. https://www.dnr.state.mn.us/
invasives/preventspread_watercraft.html. Accessed March 10, 2018
Minnesota DNR. 2018b. Minnesota Department of Natural Resources
infested waters list. http://www.eddmaps.org/midwest/tools/
infestedwaters/. Accessed October 16, 2018.
Muthukrishnan R, Sleith RS, Karol KG, Larkin DJ. 2018. Prediction of starry
stonewort (Nitellopsis obtusa) invasion risk in upper Midwest (USA) lakes
using ecological niche models. Aquat. Bot. 151:43–50.
Nam BH, Jin HJ, Kim SK, Hong YK. 1998. Quantitative viability of seaweed
tissues assessed with 2,3,5-triphenyltetrazolium chloride. J. Appl. Phycol.
10:31–36.
National Oceanographic and Atmospheric Administration. 2017. August
2017 local climatological data: Minneapolis, Minnesota. KMSP.
Parker J. 1953. Some applications and limitations of tetrazolium chloride.
Science 118:77–79.
Pullman GD, Crawford G. 2010. A decade of starry stonewort in Michigan.
Lakeline 36–42.
R Development Core Team. 2017. R: A language and environment for
statistical computing. R Foundation for Statistical Computing, Vienna,
Austria.
Raven P, Evert R, Eichhorn S. 2005. Biology of plants, 7th ed. Freeman and
Company, New York, NY.
Reynolds C, Miranda NAF, Cumming GS. 2015. The role of waterbirds in
the dispersal of aquatic alien and invasive species. Diversity Distrib.
21:744–754.
Romero-Alvarez D, Escobar LE, Varela S, Larkin DJ, Phelps NBD. 2017.
Forecasting distributions of an aquatic invasive species (Nitellopsis obtusa)
under future climate scenarios. PLOS ONE12:e0180930
Rothlisberger JD, Chadderton WL, McNulty J, Lodge DM. 2010. Aquatic
invasive species transport via trailered boats: what is being moved, who
is moving it, and what can be done. Fisheries 35:121–132.
Schloesser DW, Hudson PL, Nichols SJ. 1986. Distribution and habitat of
Nitellopsis obtusa (Characeae) in the Laurentian Great Lakes. Hydro-
biologia 133:91–96.
Sher-Kaul S, Oertli B, Castella E, Lachavanne J-BB. 1995. Relationship
between biomass and surface area of six submerged aquatic plant
species. Aquat. Bot. 51:147–154.
Sleith RS, Havens AJ, Stewart RA, Karol KG. 2015. Distribution of Nitellopsis
obtusa, (Characeae) in New York, U.S.A. Brittonia 67:166–172.
Sleith RS, Wehr JD, Karol KG. 2018. Untangling climate and water
chemistry to predict changes in freshwater macrophyte distributions.
Ecol. Evol. 8:2802–2811.
Standifer NE, Madsen JD. 1997. The effect of drying period on the
germination of Eurasian watermilfoil seeds. J. Aquat. Plant Manage.
35:35–36.
Steudle E, Zimmermann U, L ¨
uttge U. 1977. Effect of turgor pressure and
cell size on the wall elasticity of plant cells. Plant Physiol. 59:285–289
Vander Zanden MJ, Olden JD. 2008. A management framework for
preventing the secondary spread of aquatic invasive species. Can. J.
Fish Aquat. Sci. 65:1512–1522.
Yoshioka T, Takenaka T. 1979. Nitellopsis obtusa internodal cell birefrin-
gence change during action potential. Biophys. Struct. Mech. 5:1–10.
J. Aquat. Plant Manage. 58: 2020 17
APPENDIX.TEMPERATURE,RELATIVE HUMIDITY,WIND SPEED,AND PRECIPITATION DURING THE STARRY STONEWORT OUTDOOR DESICCATION EXPERIMENT CONDUCTED AUGUST 811,
2017. DATA ARE FROM THE AUGUST 2017 LOCAL CLIMATOLOGICAL REPORT FOR MINNEAPOLIS, MN, AND WERE COLLECTED AT THE MINNEAPOLIS–ST.PAUL INTERNATIONAL
AIRPORT (KMSP; 448520N; 938130W; NATIONAL OCEANOGRAPHIC AND ATMOSPHERIC ADMINISTRATION 2017).
Day Hour Temperature (C) Relative Humidity (%) Wind Speed (mph) Precipitation (cm)
August 8, 2017 9 23.3 56 6 0
12 27.2 39 11 0
15 28.3 36 8 0
18 27.2 38 6 0
21 23.9 52 8 0
24 21.1 61 0 0
Mean 25.2 47 6.5 Daily total ¼0.00
August 9, 2017 3 20.0 65 7 0
6 18.9 68 5 0
9 19.4 71 10 0.03
12 18.3 87 8 0.25
15 19.4 87 10 0.03
18 19.4 84 8 0
21 19.4 81 5 0
24 17.2 87 17 0.48
Mean 19.0 78.8 8.8 Daily total ¼1.45
August 10, 2017 3 16.1 93 5 0
6 16.7 90 9 0
9 18.3 81 8 0
12 21.7 66 15 0
15 21.7 73 11 0.02
18 21.7 63 13 0
21 18.9 73 6 0
24 16.7 84 5 0
Mean 19.0 77.9 9.0 Daily total ¼0.03
August 11, 2017 3 15.0 90 6 0
6 15.0 93 5 0
9 20.6 66 7 0
12 25.0 45 11 0
Mean 18.9 73.5 7.25 Daily total ¼0.00
Mean 20.4 70.3 8.1 Total ¼0.81
18 J. Aquat. Plant Manage. 58: 2020