The Seasonal Abundance and Distribution of the Bivalve
Lyonsia hyalina (Anomalodesmata: Pandoracea) on a
Disturbed New England Mudflat
Kenneth A. Thomas and Mark D. Clements
Northeastern Naturalist, Volume 24, Issue 3 (2017): 300–316
Full-text pdf (Accessible only to subscribers. To subscribe click here.)
Access Journal Content
Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.
Current Issue: Vol. 30 (3)
Check out NENA's latest Monograph:
Monograph 22
Northeastern Naturalist
300
K.A. Thomas and M.D. Clements
22001177 NORTHEASTERN NATURALIST 2V4(o3l). :2340,0 N–3o1. 63
The Seasonal Abundance and Distribution of the Bivalve
Lyonsia hyalina (Anomalodesmata: Pandoracea) on a
Disturbed New England Mudflat
Kenneth A. Thomas1,* and Mark D. Clements1
Abstract – We investigated a casually recognized pattern of seasonal abundance exhibited
by a population of the simultaneously hermaphroditic anomalodesmatid bivalve Lyonsia
hyalina (Glassy Lysonia) at Bluff Hill Cove, Galilee, RI, and quantified the distribution of
these intertidal to subtidal individuals from June 1993 through April 1994. Density in late
spring averaged less than 2 individuals m-2 and was followed by a summer explosion of up
to 200 individuals m-2 in localized areas, with an observed patchy distribution. A subsequent
massive autumn mortality occurred, when the population returned to the previous low
numbers. We propose several factors including predation, reduced temperature, and natural
senescence as causes of the autumnal decline.
Introduction
Tidal flats have gained increasing attention as an endangered habitat and are commonly
threatened by exploitation, pollution, and over-development (Reise 1985).
Examination of the large numbers of organisms within the sediment indicates that
these areas are important for many species of molluscs and polychaetes. They are
an important food resource and shelter for a variety of temporary residents, act as
nurseries for many fish and crustaceans, support numerous shellfisheries, and serve
as a feeding ground for millions of shorebirds (Wolff 1987). Many tidal flats worldwide
are threatened, such as those of the Yellow Sea where 50–80% of the tidal
flats have declined over the past 50 years (Murray et al. 2015). Tidal flats and their
biotic component thus merit serious attention and should be studied and conserved
for future generations.
The tidal flats of southern New England are subject to moderate tidal exchange
with a large range of associated temperature changes, local disturbances, marked
seasonal changes in weather, and bouts of seasonal predation. An organism's longterm
survival in these areas depends on its ability to respond to these changes. One
species that exhibits such adaptability is the bivalve mollusc Lyonsia hyalina (Conrad)
(Glassy Lyonsia) that has been shown to rapidly colonize recently dredged
habitats (Kaplan et al. 1975). This organism was previously reported to occur along
the eastern North American Atlantic coast from Nova Scotia to South Carolina and
inhabits sandy mud substrata, intertidally to a depth of 30 m (Abbott 1974), while
others have reported distribution as far south as Florida and the Gulf of Mexico
(Pimenta and Oliveira 2013).
1Department of Natural Sciences, Northern Essex Community College, 100 Elliott Street,
Haverhill, MA, 01830-2399. *Corresponding author - kthomas@necc.mass.edu.
Manuscript Editor: Melisa Wong
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
301
The biology of the Glassy Lyonsia is known through the work of Chanley and
Castagna (1966) and Chanley and Andrews (1971) where the development of larvae
from fertilization until settlement is described. During spawning under laboratoryinduced
conditions where water temperature was raised to 24–25 °C, Glassy Lyonsia
released 8000–16,500 eggs during a single spawn (Chanley and Castagna 1966);
this temperature is comparable to maximum summer field temperatures in southern
Rhode Island, where the study described in this paper was conducted. Within 24
hours post-fertilization, larvae develop into the “straight hinge” stage, and later
metamorphose over days 3–5 to lose the velum, gain gills, develop a ciliated foot,
and become capable of attaching themselves to the substrate with a byssal thread
(Chanley and Castagna 1966). These larvae resemble those of Pandora gouldiana
(Dall) (Gould Pandora), but are shorter in length and probably have no need for an
outside food source before settlement because of their short pelagic period, large
egg yolk, and small size at metamorphosis (Chanley and Castagna 1966). Chanley
and Castagna (1966) also showed that this species is capable of self-fertilization
with no apparent developmental issues up to time of metamorphosis. Prezant (1977;
1979a, b) studied aspects of the functional morphology and the ultrastructure of the
arenophilic radial mantle glands, and Thomas (1993, 1996) conducted an investigation
of the microanatomy and histology of the digestive system (Thomas 1993) and
an ultrastructural analysis of gametogenesis (Thomas 1996). Glassy Lyonsia has
been shown to be a first intermediate host of either or both of the bucephalid trematode
fish parasites, Rhipidocotyle transversale (Chandler) and R. lintoni (Hopkins),
where branching sporocysts and furcocercous cercariae have been observed in both
the gonads and digestive glands (Stunkard 1976). Although these biological aspects
of the Glassy Lyonsia are generally understood, the ecological dynamics of this
organism remain generally unknown.
The Glassy Lyonsia is an opportunistic species that will populate recently disturbed
areas (Kaplan et al. 1975). This species has very short siphons and settles
within the top layer of sandy sediment, often with the valves partially exposed
(Virnstein 1979). This proximity to the sediment surface makes it vulnerable to
bottom-feeding fish and crab predators such as Leiostomus xanthurus (Lacepède)
(Spot Croaker) and Callinectes sapidus (Rathbun) (Blue Crab) in Chesapeake Bay
(Virnstein 1977, 1979), Pogonias cromis (L.) (Black Drum) in Texas coastal bays
(Cate and Evans 1994), and Pseudopleuronectes americanus (Walbaum) (Winter
Flounder) in the Lower Hudson-Raritan Estuary (Steimle et al. 2000).
It was anecdotally reported that during the summer months numerous Glassy
Lyonsia were located on the tidal mudflat at Bluff Hill Cove, Galilee, RI. This population
could be easily accessed during middle to low tide by simply walking out
on the mudflat and sieving the sediment (R. Bullock, University of Rhode Island,
Kingston, RI, 1988 pers. comm.). A repeating pattern of summer abundance was
consistently observed from 1988 to 1992 (K.A. Thomas, pers. observ.). Specifically,
an increase in abundance began in late spring to peak in the summer, followed by
a period of mortality in September and October with a huge population decrease
through November and December, to remain low overwinter until the cycle was
Northeastern Naturalist
302
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
repeated the following year. Typically, in New England, this organism is found in
subtidal areas and is regionally localized (R. Prezant, Southern Connecticut University,
New Haven, CT, 1990 pers. comm.), so this readily accessible population
provided us with the unique opportunity to easily examine the population dynamics
in this species.
This study was conducted in a temporally stratified manner to quantitatively
describe the seasonal population shifts in abundance of this species at Bluff Hill
Cove. This information contributes to the limited body of ecological knowledge
available for members of the Anomalodesmata, and provides an important
description of an organism well-suited to opportunistically colonize disturbed
intertidal habitats. These data provide a basis for comparison to Lyonsia populations
in similar but undisturbed habitats, which would aid in assessing the impact
of human activity. Thus, the Glassy Lyonsia may be a potentially useful species in
environmental assessment.
Field-Site Description
The study area at Bluff Hill Cove, Galilee, RI, is located at 41°38'N, 71°50'W
(Fig. 1A–C). This ~26-ha mudflat is located in the southeastern corner of Point Judith
Pond, a tidal inlet close to Block Island Sound, off the coast of southern Rhode
Island (Fig 1A). It is a relatively shallow flat partially isolated by several islands
to the north with northeastern access to Point Judith Pond, and a narrow western
channel that leads to the south (oceanic) end of Point Judith Pond (Fig. 1B). The
mudflat is bordered on the east by the mainland and south by a saltmarsh and is
proximal to, though protected from, the seaward opening of its encompassing salt
pond. As a result, this mudflat is regularly flushed with the semidiurnal tides. An
access road atop a berm borders the southern portion of the mudflat, and several
small outlets pass under this road to an isolated marshy region that serves as a bird
sanctuary and borders a barrier beach 0.6 km to the south. The surrounding area
is well-developed, and is a popular tourist destination that contains many summer
residences. Before, during, and after this study, the mudflat was subject to frequent
small-scale sediment disturbances caused by recreational clam fishing, especially
during the summer months. During these events, the sediment was often raked,
dug 10–30 cm, and piled into mounds by clammers, the result of the search for
Mercenaria mercenaria (L.) (Quahog) and Mya arenaria (L.) (Soft-Shell Clam).
During the late fall, winter, and early spring, the area had little fishing activity .
Methods
This study of abundance of the Glassy Lyonsia was conducted from June 1993 to
April 1994. We established 3 parallel NNE/SSW transects (Fig. 1C) 35 m apart on
the mudflat along the tidal gradient (labelled I, II, and III from west to east). Along
each transect, we sited four 15 m x 15 m quadrats. Transect I contained quadrats
1–4, transect II contained quadrats 5–8, and transect III contained quadrats 9–12.
Three quadrats high on the tidal gradient and closest to the shoreline —numbers
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
303
Figure 1. Study area
at Bluff Hill Cove,
Galilee, RI (41°38'N,
71°50'W). (A) Map
showing study site
in relation to Massachusetts,
Connecticut,
Rhode Island (dark
gray), and New York.
(B) Lower Point Judith
Pond; study site
indicated by arrow.
(C) Mudflat and sandbar
at Bluff Hill Cove,
illustrating exposure
at mean low water.
Transects I–III were
established along a
NNE/SSW line, with 4
quadrats per transect.
Quadrats 1, 5, and 9
were abandoned early
in the study due to excessive
disturbances
caused by recreational
clam fishing. During
mean high tide, all
areas shown except
“Saltmarsh” became
inundated with seawater.
On the flood
tide, water flow was
generally west to east,
and opposite that on
the ebb tide. At low
water, flow became
more tortuous; during
spring low tides,
quadrats 3, 7, and 11
often became an isolated
shallow pool.
1, 5, and 9—were abandoned early in the study because of excessive human disturbance
due to local clamming activity and a subsequent lack of intact specimens
in these areas. Quadrats 2, 6, and 10 were established adjacent to one another and
located in the mid-tidal region at the 250-m mark on each transect. Quadrats 3, 7,
and 11 were situated in a subtidal channel that ran diagonally across the transects
and located at the 305-m, 312.5-m, and 322.5-m marks, respectively. Quadrats 4,
Northeastern Naturalist
304
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
8, and 12 were located at the 373.3-m mark of each transect, adjacent to each other
on a sandbar. This area became exposed unevenly at ebb tide beginning 30–45
minutes previous to quadrats 2, 6, and 10, exposing quadrat 4 first, then quadrat
8, and finally quadrat 12. Water flow was generally perpendicular to the transects
during flooding and ebbing, moving west to east on the incoming tide (transects I to
III) and east to west (transects III to I) on the outgoing tide. The area surrounding
quadrats 3, 7, and 11 (in the subtidal channel) sometimes becomes a large isolated
pool during spring low tides; however, during lesser tides, flow through this channel
continued during low tide, often in a tortuous manner.
We marked quadrats at each corner with a 1-m length of PVC pipe (1.25 cm
diameter) submerged into the substratum until 4–5 cm remained exposed. Many of
the wooden stakes (2 cm x 2 cm x 1 m long) we deployed outside each corner for
quick quadrat identification were lost during the study period; however, all PVC
pipes were retained and allowed for exact quadrat location. We subdivided quadrats
into 900 subsequently numbered 0.5 m x 0.5 m sample plots. On sampling days,
we used a random number generator to select four 0.5 m2 plots for each quadrat
and sampled without replication during the study. Note that 3 plots/quadrat were
sampled on 27 June 1993, but we increased the sample size to 4 plots/quadrat thereafter
due to initially low collection numbers. For each quadrat, an area of 19.75 m2
was surveyed during the study over the course of 20 separate sampling days and
covered 8.78% of the total area within each quadrat. With all quadrats combined
over the entire study, a total area of 177.75 m2 was sampled. We estimated that this
mudflat covers an area of 260,000 m2 (using Google Earth’s polygon area tool);
thus, we examined approximately 0.068% of the entire mudflat.
Samples (including live organisms and shell fragments) were collected in a temporally
stratified manner, so that the most extensive sampling was performed during
the known period of peak abundance (i.e., weekly in summer with a gradual tapering
in frequency during autumn, until a monthly sampling period was established
in November and maintained until March, for the remainder of the study). We
recorded water temperature using an alcohol-based field thermometer in an area of
flowing water near the study transects on 17 sampling days (water temperature was
also recorded for several periods prior to, and after this study). We surveyed the
plots with the use of a 0.5 m x 0.5 m (internal dimensions) PVC square, placed on
the sediment surface. The top 5–6 cm deep layer was collected and passed through a
2.5-mm–mesh sieve, outside and down-current from the quadrat to minimize sediment
disturbance within the quadrats.
Retained fractions were bagged and preserved with a 5% formalin/seawater
mixture. Sampling was tidally coordinated over the study area so that on sampling
days we usually took collections when the quadrats were covered with 0.25
to 0.5 m of water. Care was taken to create minimal disturbance when measuring
and sampling within the quadrats. Selected sample plots were occasionally subject
to recent human disturbance (most frequently during the su mmer months), in
which case we substituted the nearest undisturbed plot in place of the disturbed
plot. Samples were brought into the laboratory where we sorted them under a
dissecting microscope to isolate whole Glassy Lyonsia and any shell fragments.
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
305
We recorded the location, number, and overall length (measured to the nearest 1
mm using an ocular micrometer and dissecting microscope) of individuals in all
plots. Collected individuals were stored in 70% ethanol following the initial fixation
in 5% formalin/seawater, usually within 1–2 weeks. We noted the presence
and number of shell fragments collected simply as a gross estimation of mortality
and counted equally all fragments, regardless of size. Some fragments were nearly
complete shells and occasionally attached to the complementary shell, while others
represented only a small portion of the individual from which it came. It is
unknown if multiple fragments collected in the same sample were from the same
individual, but it was assumed this would not have readily occurred, because Lyonsia
shells are fragile and likely break down quickly into smaller, less noticeable
fragments in situ that would not be retained during collection. We consider the
naturally occurring presence of noticeable shell fragments in the field samples to
be direct evidence of recent mortality.
We tabulated specimen size and relative age (assuming individual Glassy Lyonsia
grow at equal rates, and that larger individuals are older) using 4 separate size
classes marked as categories 1–4 as follows: category 1 = 3–4 mm long, category
2 = 5–7 mm in length, category 3 = 8–10 mm long, and category 4 were Glassy
Lyonsia ≥ 11 mm in length.
Results
Overall size frequency distribution
Overall, we collected a total of 3284 individual specimens of Glassy Lyonsia,
varying from 3–19 mm in length, with a mean size of 6.99 ± 0.04 mm (all means
are reported with ± 1 SE) (Fig. 2). Of all individuals collected, 65.23% were ≤7 mm
in size, with approximately equal numbers of 5 mm (21.22%) and 6 mm (20.1%)
individuals, and fewer 7 mm (14.83%) individuals. Larger individuals were less
common (8 mm = 11.88% of total; 9 mm = 8.80% of total; 10 mm = 6.15% of total;
11 mm = 3.53% of total; 12 mm = 1.95% of total; 13 mm = 1.00 % of total). Glassy
Lyonsia >13 mm were rare and comprised 1.46% of total observations.
Mean size vs. density
Examination of the mean size vs. density of Glassy Lyonsia (Fig. 3) shows that
in late June average size was large (10.78 ± 1.16 mm) while density was low (2.07
m-2). As density began to increase in mid-July (8.22 m-2), average size declined
(5.57 ± 0.21 mm). Density then sharply increased over summer and peaked in August
at 59.44 m-2, with localized densities as high as 200 m-2, while size increased
slowly to an average of 6.80 ± 0.09 mm. A sharp decline in density followed in September
(24.33 m-2), concurrent with continued average size increase (8.02 ± 0.18
mm), followed by a sharp decline (6.50 ± 0.15 mm). Density continued to decline
steadily during fall while average size increased (9.69 ± 0.66 mm). This density
decline continued (2.89 m-2 in mid-December) with a winter low in February (0.78
m-2, 7.00 ± 1.27 mm) and March (1.33 m-2, 9.33 ± 1.14 mm), with sizes approaching
that of individuals observed the previous spring.
Northeastern Naturalist
306
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
Water temperature
Water temperature (Fig. 4) warmed during the month of June, ranged from 22
°C to 26 °C during July–September, and droppped through the fall and early winter
to reach a minimum of 2 °C from January through March. Water temperature then
began an annual spring increase during the month of April.
Shell-fragment analysis
Analysis of Glassy Lyonsia shell fragments (Fig. 5) shows that the lowest density
occurred in late June (10.50 ± 2.84 fragments m-2). This number increased during
the next sampling period (20.78 ± 4.12 fragments m-2) and remained steady for
several weeks. A rapid increase followed over the next 8 weeks, to peak at 162.56
± 15.68 fragments m-2 during mid-September, following the peak in live individual
density by 1 month. Density of shell fragments declined steeply, then steadily, to
11.89 ± 1.68 fragments m-2 in March.
Frequency distribution by quadrat
The number of individuals varied across quadrats (n = 190–593; Table 1) with a
patchy distribution and highest relative abundances in quadrats 11 (n = 593) and 12
Figure 2. Frequency distribution by size. In this study, Lyonsia hyalina (Glassy Lysonia)
varied from 3 mm to 19 mm in length, and over half collected were 5–7 mm. The abundance
of animals in the 3–4 mm range probably are an underrepresentation due to a lack of sieve
retention, with actual numbers likely exceeding those in the 5–6 mm range.
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
307
(n = 551), both located on transect III (Fig. 1C). A large number of individuals were
also observed in quadrats 2 (transect I; n = 420) and 7 (transect II; n = 396) with
lower relative abundances on either side of this patch. On the sandbar, abundance
fell dramatically toward quadrat 4 (quadrat 4: n = 190, quadrat 8: n = 324) where
aerial exposure was the greatest.
Figure 3. Average size vs. density of Lyonsia hyalina (Glassy Lysonia). At the beginning of
this study, size (average length) was highest, while density was very low. Density abruptly
increased during the summer months, then dropped sharply in late summer. During this initial
increase in density, size dropped quickly, indicating that a recruitment event took place.
Table 1. Number of Lyonsia hyalina (Glassy Lysonia) a per quadrat. Some quadrats had many fewer
organisms than others. The abundance was highest in quadrats 11 and 12, showing the patchy nature
of this species.
Quadrat n
2 420
3 226
4 190
6 292
7 396
8 324
10 292
11 593
12 551
Northeastern Naturalist
308
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
Figure 4. Water temperature at Bluff Hill Cove during the study period (June 1993–March
1994). Water temperature the month prior to the beginning of the study was 19 °C (11 June
1993), followed by a warming trend over the summer, then a slow steady cool down during
autumn to reach winter lows in January. The beginning of the subsequent spring warming
period can be seen at the end of the study.
Figure 5. The number of Lyonsia hyalina (Glassy Lysonia) shell fragments per m2. Fragments
were initially low in the spring and mirrored the increase in live animal abundance,
but ~1 month later, likely to be a reflection of major mortality of this species which increased
in summer and peaked in late September.
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
309
Frequency distribution by size class
By individual quadrat, Category 1 Glassy Lyonsia (3–4 mm) averaged 9.01%
(± 0.44%) of the total Glassy Lyonsia collected (Table 2), with a narrow range, and
were observed most often in quadrats 11, 2, 7, and 12. Category 2 (5–7 mm) was
the most abundant size class and these Glassy Lysonia averaged 55.82% (± 3.32%)
of total found per quadrat, with a wide range, and were seen most often in quadrats
11, 2, 12 and 7. Category 3 individuals (8–10 mm) collected averaged 26.94%
(± 2.08%) of total, with a moderate range, and were found most often along rransect
III in quadrats 11, 12 and 10. Category 4 individuals (≥11 mm) averaged 8.23%
(± 1.75%) of total, also with a moderate range, and were collected in greatest numbers
from quadrats 12, 8, 11, and 4.
Overall temporal frequency distribution by size class
The overall incidence of category 1 individuals across the study area gradually
increased during the summer (Fig. 6) with a distinct initial peak demonstrated in
the sample taken on 14 August 1993 (n = 46), and a second spike 1 month later (14
September 1993, n = 23). There was a dramatic increase in the abundance in category
2 individuals (starting in early July to spike on 14 August 1993 (n = 320). A
rapid drop in abundance ensued over the next several weeks, followed by a second
smaller spike in abundance (14 September 1993, n = 138) and further drop, marking
the beginning of a slow steady decline to reach winter lows. Category 3 Glassy
Lyonsia began to increase 2–3 weeks after the category 2 increase, to spike on 27
August 1993 (n = 180), followed by a rapid decline over the next 3 weeks (n = 33 on
14 September 1993). Numbers fluctuated for 2 months, then subsequently dropped
to reach winter lows. A less dramatic increase began in category 4 individuals 3–4
weeks after the category 3 onset and 4–5 weeks after the category 2 onset, with a
27 August peak (n = 59). As with the other categories, there was a sharp drop off in
abundance followed by low numbers in the late fall that tapered in January to reach
winter lows.
Table 2. Overall frequency of Lyonsia hyalina (Glassy Lysonia) by size per quadrat.
Category 1 Category 2 Category 3 Category 4
Quadrat ≤4 mm % 5–7 mm % 8–10 mm % ≥11 mm %
2 44 10.48 277 65.95 84 20.00 15 3.57
3 22 9.73 152 67.26 42 18.58 10 4.42
4 16 8.42 78 41.05 66 34.74 30 15.79
6 21 7.19 185 63.36 71 24.32 15 5.14
7 42 10.61 258 65.15 83 20.96 13 3.28
8 25 7.72 150 46.30 97 29.94 52 16.05
10 29 9.93 139 47.60 103 35.27 21 7.19
11 57 9.61 333 56.16 170 28.67 33 5.56
12 41 7.44 273 49.55 165 29.95 72 13.07
Mean 9.01 55.82 26.94 8.23
SE 0.44 3.32 2.08 1.75
Median 9.61 56.16 28.67 5.56
Northeastern Naturalist
310
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
Temporal frequency distribution by size class within individual quadrats
In the temporal analysis of size classes within individual quadrats (Fig. 7), differences
in category 1 are difficult to resolve because of low abundances. However,
in all quadrats there was a clear initial spike in category 2 individuals, usually between
July and mid-August, often with a subsequent (and sometimes third) peak,
and in some quadrats (2, 4, 6, 8, and 12) this delayed spike exceeded the magnitude
of the first spike in category 2. In quadrats 3, 7, and 11 (all of which lie within
the subtidal channel), this second spike was delayed until mid-September. In all
quadrats, abundance of category 3 individuals increased 1–4 weeks after the initial
spike for category 2 Glassy Lysonia. Category 4 individuals show a clear abundance
spike in only 4 quadrats (4, 8, 11, and 12) in late summer, always after the spike of
category 3 Glassy Lysonia. All quadrats show tapering of abundance in every size
class through the fall and into the winter.
Discussion
This study quantifies a pattern similar to that observed casually over the previous
4 years, and one described anecdotally for several years prior (R. Bullock, 1988
Figure 6 . Relative abundance by size of Lyonsia hyalina (Glassy Lysonia). The number of
individuals collected in each of the size classes (categories 1–4) are shown over the study
period. All categories showed a large increase during the summer which culminated in peak
abundance during middle to late August. Abundance in all size classes decreased to reach
winter lows during November or December.
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
311
pers. comm.). During the spring, there was an initially low abundance of large-sized
Glassy Lyonsia, followed by a water temperature increase and population explosion
during the summer, indicating massive recruitment, with a subsequent crash
in the fall. Over the summer, average individual size slowly increased (Fig. 3), and
size-category peaks shifted sequentially (Fig. 6), suggesting that survivors systematically
grew into the next size class. Although average size of individuals dropped
during July and August, the dramatic increase in population likely attracted
benthic-feeding predators such as Blue Crab, which has demonstrated optimalforaging
type behavior (i.e., marginal value; Charnov 1976) during predation of
the bivalve Macoma balthica L. (Baltic Macoma; Clark et al. 2000). As individuals
grew over the summer, predation pressure, along with intra- and inter-specific
competition and natural senescence, may have caused the sharp mortality increase
that was observed in August and peaked in September. The declining numbers of
Figure 7. Relative abundance of Lyonsia hyalina (Glassy Lysonia) by quadrat. This series
of graphs illustrates the number of individuals collected temporally in each of the size categories
(1–4) per quadrat over the study period.
Northeastern Naturalist
312
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
individuals and presence of shell fragments indicate that a slow but steady die off
continued during the late fall and tapered as winter arrived. During this time, the
average size of individuals increased, leading to a comparatively small population
of larger, overwintering adults that helped repopulate stocks the following spring.
Although most individuals in the Bluff Hill Cove population appear to survive for
only a few months, an annual growth line was observed in the shells of some category
4 individuals, which Prezant (1977; pers. comm.) used to estimate a 2–3 year
lifespan in this species.
In September, secondary spikes in abundance of category 1 and 2 individuals
suggest that another recruitment event occurred in quadrats 2, 3, 7, and 12.
Examination of chlorophyll-a data from nearby Station 2, West Passage, Narragansett
Bay, RI (Narragansett Bay Phytoplankton Time Series, 1959–1997;
NarrBay.org 2004) shows that chlorophyll-a spiked three weeks prior, probably
due to an increase in phytoplankton, and may have contributed to these events.
Hunt et al. (2009) tested predictions that bedload transport could dramatically affect
bivalve distribution up to several kilometers over 1 month. They found their
model to be consistent for the Soft-Shell Clam and a pooled group of bivalve species,
including Glassy Lyonsia, in the Navesink River estuary, NJ. Their model,
however, was inconsistent in the case of Gemma gemma (Totten) (Amethyst Gemclam),
and they suggested that sandy sediment selection and lack of larval stages
in this species influence its distribution (Hunt et al. 2009). Infaunal bivalves
such as Soft-Shell Clams 3–10 mm in length are known to sometimes become
entrained and moved around via bedload transport, so this may also occur with
similarly sized Glassy Lyonsia (H. Hunt, University of New Brunswick, St. John,
NB, Canada, 2017 pers. comm.). Here, localized bedload displacement of some
individuals may have occurred, but the isolated nature of the study site makes it
unlikely that this mechanism brought in a new cohort of individuals from outside
the mudflat (H. Hunt, 2017 pers. comm.). Thus, we believe that the secondary
abundance spikes observed were related to secondary larval recruitment and that
the sequential spikes in size-category abundances reflect growth of individuals
from smaller sizes, rather than entrainment from outside sources.
The disturbed nature and physical characteristics of the mudflat at Bluff Hill
Cove makes this habitat well-suited for Glassy Lyonsia. Interestingly, Prezant et
al. (2008) found that another anomalodesmatan, Laternula rostrata (Sowerby),
(corrected from L. truncata L.) maintained refugia in mangrove roots but were uncommon
in open areas subjected to heavy shellfish harvesting. During the Glassy
Lyonsia density peak in August, there were an average of 59 individuals m-2, with 1
localized, single plot density of 200 m-2, but by January average density had fallen
to 2 m-2 and then to below 1 m-2 in February. The largest individual collected during
the sampling period was 19 mm (n = 1), but the average size was much smaller
(6.98 ± 0.04 mm). Prezant (1977) observed specimens as long as 9.8 mm in a Massachusetts
Bay population of Glassy Lyonsia (sampled from depths of 15–18 m),
although average size in that study was only 3.6–5.6 mm. Shells that washed up in
Zostera marina L. (Eelgrass) mats from a Martha’s Vineyard, MA, salt pond were
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
313
generally twice this size and up to 22.8 mm in length (Prezant 1977). Individuals as
long as 24 mm were observed by Virnstein (1979) during fish-predator–exclusion
experiments, and Prezant et al. (2002) reported a “giant” individual over 25 mm
long in coastal Virginia.
A “scarce to abundant in patches” distribution pattern for this species was described
in a Virginia population by Chanley and Andrews (1971), and data from
this study indicates that patchiness may have a size-dependent component. Smaller
Glassy Lyonsia (3–7 mm) were found in greatest densities diagonally across
the study area and account for more than 65% of all individuals in this study (Table
2, Fig. 7). These smaller individuals were least abundant in quadrat 4, which
had the longest period of aerial exposure as compared with all other quadrats,
suggesting that extended exposure decreases the survivability in juvenile and/or
newly settled individuals.
Glassy Lyonsia was previously described as capable of reproducing multiple
times during its life (i.e., iteroparous) with 1 annual spawn (Chanley and Andrews
1971, Mackie 1984). Spawning is known to occur in March and April and possibly
continues through May, June, and even July in coastal Virginia populations (Chanley
and Andrews 1971). In coastal Massachusetts, spawning occurs in July and
August (Prezant 1977), and evidence from this study indicates a major and minor
recruitment event occurred in Rhode Island. This organism’s high fecundity, ability
to self-fertilize, and short time as pelagic larvae prior to settlement may allow for
a small number of gravid adults to reseed extensive seasonal populations like that
observed on the mudflats at Bluff Hill Cove.
Post-settlement growth data on this species appears lacking and growth time
from larval size (150–180 μm) to category 1 size (3–4 mm in length) is previously
unreported. Virnstein (1979) showed that Glassy Lyonsia flourished during
enclosure experiments, and grew to sizes larger than normally found in unenclosed
populations (up to 24 mm in 6 months or less). Assuming that growth was continuous
over 6 months, then the average daily growth rate was 0.13 mm d-1. Using that
rate, we extrapolate that it takes ~ 23 days for post-settled larvae to grow to 3 mm
length. The mid-July drop in average size of Glassy Lysonia observed in this study
(Fig. 3) indicates that a recruitment event had occurred and suggests that newly
settled clams grew quickly over 2–3 weeks. Previously undetected individuals
would have increased in size from 3.5 mm or less (smaller than sieve mesh size) to
~5.6 mm (average size in mid-July). Conservatively, growth calculated using these
data show a rate of 0.10–0.15 mm d-1. However, peak abundance data for category
2 and 3 (Fig. 6) suggest a growth rate of 0.23 mm d-1, where individuals grew 3
mm over 13 days. The growth rate of Glassy Lyonsia appears comparable to that
of Soft-Shell Clams, reported as approximately 0.13–0.20 mm d-1 for individuals
10–24 mm length (Chalfoun et al. 1994), but much greater than those of Corbicula
fluminea (Müller) (Asian Clam) which has been reported as 0.03 mm d-1 in individuals
less than 10 mm in length (Cataldo et al. 2001).
It is known that some bivalve species will grow at reduced rates in response
to the activity of predators. Quahogs, for instance, exhibit reduced growth rate in
Northeastern Naturalist
314
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
response to the presence of the predatory snail Busycon carica (Gmelin) (Knobbed
Whelk; Nakaoka 2000), and increased nutrient load will accelerate growth rate in
Quahogs and Soft-Shell Clams (Chalfoun et al. 1994). If marine phytoplankton
become more abundant, or if predator populations increase, Glassy Lyonsia may adjust
their growth rate accordingly. Predation pressures on this species (from avian,
molluscan, fish predators, etc.) likely occur at Bluff Hill Cove; however, future
studies are needed to elucidate specific predatory patterns and if such predation
reduces growth rate.
Currently we lack clear understanding of the source of recruitment into Bluff
Hill Cove. Based upon geographic proximity, individuals that make up this summer
population explosion could be recruited from 3 sources: first, individuals
that overwinter on the mudflat at Bluff Hill Cove; second, adjacent subtidal areas
within Bluff Hill cove, but not on the mudflat (and hence are never exposed
to direct atmospheric conditions); and third, a stock population(s) that overwinters
in areas outside of Bluff Hill Cove. We know little about the hydrodynamics
of this area, and such understanding would help determine source populations,
perhaps indicating whether or not this population is genetically heterogeneous.
Genetic analyses would help determine whether members of the Bluff Hill Cove
mudflat population are interrelated and indicate if there is a seed population of
overwintering individuals. Such studies may also help to identify if self-fertilization
is a natural strategy employed by this species and should be included in
future investigations.
This study begins to describe what is apparently a complex system. Numerous
ecological questions are generated from these observations regarding this
organism’s recruitment, population dynamics, and role in the food chain. Future
studies should be directed toward these questions including hydrodynamic analyses
through the tidal cycle, quantification of planktonic and newly settled Glassy
Lyonsia larvae, predator exclusion and mark/recapture experiments, and identification
of nearby populations with comparative genetic analysis to the Bluff Hill
Cove population.
Acknowledgments
We thank the following people for their assistance during field collection and help with
specimen sorting: Jane Fieldhouse-Thomas, Joseph Grenier, Julie Hammer, Neil Hurley,
Alex Mooza, Bradley Peterson, and Steve O’Neil. Thanks also to Bradley Peterson, Robert
Prezant, and 1 anonymous reviewer for their critical comments of this manuscript. Heather
Hunt provided insight into bivalve juvenile transport, and David Borkman supplied access
to Theodore Smayda’s extensive Narragansett Bay phytoplankton data. Special thanks to
the Rhode Island Department of Environmental Management for their continued encouragement
and support of this research. This work was financially supported by the National
Capital Shell Club’s Carl I. Aslakson Scholarship, the Western Society of Malacologist’s
Student Research Grant in Malacology, the University of Rhode Island, and Northern Essex
Community College.
Northeastern Naturalist Vol. 24, No. 3
K.A. Thomas and M.D. Clements
2017
315
Literature Cited
Abbott, R.T. 1974. American Seashells. Second Edition. Van Nostrand Reinhold Company,
New York, NY. 663 pp.
Cataldo, D.H., D. Boltovskoy, J. Stripeikis, and M. Pose. 2001. Condition index and growth
rates of field-caged Corbicula fluminea (Bivalvia) as biomarkers of pollution gradients
in the Paraná river delta (Argentina). Aquatic Ecosystem Health and Management
4:187–201.
Cate, A.S., and I. Evans. 1994. Taphonomic significance of the biomechanical fragmentation
of live molluscan shell material by a bottom-feeding fish (Pogonias cromis) in
Texas coastal bays. Palaios 9(3):254–274.
Chalfoun, A., J. McClelland, and I. Valiela. 1994. The effect of nutrient loading on the
growth rate of two species of bivalves, Mercenaria mercenaria and Mya arenaria, in
estuaries of Waquoit Bay, Massachusetts. The Biological Bulletin 187(2): 281.
Charnov, E.L. 1976. Optimal foraging: The marginal value theorem. Theoretical Population
Biology 9:129–136.
Chanley, P., and J.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia.
Malacologia 11:45–119.
Chanley, P., and M. Castagna. 1966. Larval development of the pelecypod Lyonsia hyalina.
Nautilus 79:123–128.
Clark, M.E., T.G. Wolcott, D.L. Wolcott, and A.H. Hines. 2000. Foraging behavior of an
estuarine predator, the Blue Crab, Callinectes sapidus, in a patchy environment. Ecography
23:21–31.
Hunt, H.L., D.C. Fugate, and R.J. Chant. 2009. Modeling bedload transport of juvenile
bivalves: Predicted changes in distribution and scale of postlarval dispersal. Estuaries
and Coasts 32:1090–1102.
Kaplan, E.H., J.R. Welker, M.G. Kraus, and S. McCourt. 1975. Some factors affecting the
colonization of a dredged channel. Marine Biology 32(2):193–204.
Mackie, G.L. 1984. Bivalves. Pp. 351–418, In A.S. Tompa, N.H. Verdonk, and J.A.M. van
den Biggelaar (Eds.). The Mollusca. Vol. 7. Reproduction. Academic Press, Inc., New
York, NY. 486 pp.
Murray, N.J., Z. Ma, and R. Fuller. 2015. Tidal flats of the Yellow Sea: A review of ecosystem
status and anthropogenic threats. Austral Ecology 40:472–481.
Nakaoka, M. 2000. Nonlethal effects of predators on prey populations: Predator-mediated
change in bivalve growth. Ecology 81(4):1031–1045.
NarrBay.org. 2004. Narragansett Bay Phytoplankton time series data: Plankton and supporting
variables. 1959–1997. Available at online at http://www.narrbay.org/d_projects/
plankton-tsv/plankton-tsv.htm. Accessed 22 February 2017.
Pimenta, A.D., and C.D. Oliveira. 2013. Taxonomic review of the genus Lyonsia (Pelecypoda:
Lyonsiidae) from east coast of South America, with description of a new species and
notes on other Western Atlantic species. American Malacological Bulletin 31(1):75–84.
Prezant, R.S. 1977. The biology of Lyonsia hyalina Conrad with emphasis on the radial
mantle glands. M.Sc.Thesis. Graduate School of Arts and Sciences, Northeastern University,
Boston, MA. 137 pp.
Prezant, R.S. 1979a. Shell spinules of the bivalve Lyonsia hyalina. Nautilus 93:93–95.
Prezant, R.S. 1979b. The structure and function of the radial mantle glands of Lyonsia hyalina
(Bivalvia: Anomalodesmata). Journal of Zoology 187:505–516.
Prezant, R.S., C.L. Counts, and E.J. Chapman. 2002. Mollusca of Assateague Island, Maryland
and Virginia: Additions to the fauna, range extensions, and gigantism. The Veliger
45(4):337–355.
Northeastern Naturalist
316
K.A. Thomas and M.D. Clements
2017 Vol. 24, No. 3
Prezant, R.S., C. Sutcharit, K. Chalermwat, N. Kakhai, T. Duangdee, and P. Dumrongrojwattana.
2008. Population study of Laternula truncata (Bivalvia: Anomalodesmata:
Laternulidae) in the mangrove sand flat of Kungkrabaen Bay, Thailand, with notes on
Laternula CF. corrugata. The Raffles Bulletin of Zoology. Supplement No. 18:57–73.
Reise, K. 1985. Tidal Flat Ecology: An Experimental Approach to Species Interactions.
Springer-Verlag, New York, NY. 191 pp.
Steimle, F.W., R.A. Pikanowski, D.G. McMillan, C.A. Zetlin, and S.J. Wilk. 2000. Demersal
fish and American lobster diets in the Lower Hudson - Raritan Estuary. United States
Department of Commerce, NOAA Tech Memo NMFS NE 161. Woods Hole, MA.106 pp.
Stunkard, H.W. 1976. The life cycles, intermediate hosts, and larval stages of Rhipidocotyle
transversale Chandler, 1935 and Rhipidocotyle lintoni Hopkins, 1954: Life-cycles and
systematics of Bucephalid trematodes. The Biological Bulletin 150(2):294–317.
Thomas, K.A. 1993. The functional morphology of the digestive system of Lyonsia hyalina
Conrad, 1831 (Bivalvia: Anomalodesmata: Pandoracea), Journal of Molluscan Studies
59:177–188.
Thomas, K.A. 1996. The reproductive biology of Lyonsia hyalina (Conrad, 1831) (Mollusca:
Bivalvia: Anomalodesmata): Ultrastructural aspects of spermatogenesis and oogenesis.
Ph.D. Dissertation. University of Rhode Island, Kingst on, RI. 91 pp.
Virnstein, R.W. 1977. The importance of predation by crabs and fishes on benthic infauna
in Chesapeake Bay. Ecology 58(6):1199–1217.
Virnstein, R.W. 1979. Predation on estuarine infauna: Response patterns of component species.
Estuaries 2(2):69–86.
Wolff, W.J. 1987. Flora and macrofauna of intertidal sediments. Pp. 81–105, In J.M. Baker
and W.J. Wolff (Eds.). Biological Surveys of Estuaries and Coasts, Cambridge University
Press, New York, NY. 449 pp.