Locomotory fatigue during moderate and severe hypoxia and hypercapnia in the Atlantic blue crab, Callinectes sapidus.
The Atlantic blue crab, Callinectes sapidus (Rathbun), is an important commercial and recreational species that is of economic interest in the southeastern United States and the Gulf of Mexico (Whitaker et al., 1998). Estuarine habitats, in which blue crabs are found, fluctuate both daily and seasonally in salinity, temperature, dissolved oxygen, and pH (Carpenter and Cargo, 1957; Sanger et al., 2002). These variations are due to phenomena such as tide, microbial activity, human activity, and season (Diaz and Rosenberg, 1995; Burnett, 1997). Crustaceans that live in estuaries must adjust to these changes while finding food, avoiding predators, and reproducing. Blue crabs survive these environmental conditions by using behaviors to escape them and physiological mechanisms to compensate for them.
Among these environmental challenges, coastal development has led to increased eutrophication in coastal waters, driving decreases in dissolved oxygen, with resulting increases in the frequency and severity of hypoxic events (Diaz and Rosenberg, 1995). Long-term hypoxic zones occur most often during the summer and after spring algal blooms; they can last until autumn. Shorter hypoxic events, lasting from days to weeks, can be produced by weather events and tidal cycles (Diaz and Rosenberg, 2008). Increased levels of C[O.sub.2], or hypercapnia, which often occur simultaneously with hypoxia, are produced by respiration and cause a drop in the water pH (Cochran and Burnett, 1996; Burnett, 1997). Hypercapnic water results in elevated partial pressures of C[O.sub.2] in the tissues of organisms, and the ensuing respiratory acidosis can profoundly affect the transport of oxygen and other physiological functions. This can be especially important in coastal waters where the severity of hypercapnia fluctuates diurnally and can be quite severe (Cochran and Burnett, 1996; Wenner et al., 2004).
Blue crabs are highly sensitive to hypoxia: it causes an increase in mortality and a decrease in rates of feeding, growth, and molting (Das and Stickle, 1993, 1994). Crabs caught in crab pots are killed when high levels of hypoxia are reached because they cannot escape to water where oxygen concentrations are higher, such as in shallow water (Carpenter and Cargo, 1957). Thus, their ability to escape hypoxia is important (Taylor and Eggleston, 2000). In a laboratory setting, crabs that are exposed to hypoxia become active and may even try to flee from the tank; however, after a few hours they become inactive and often will bury themselves when they cannot escape (deFur et al., 1990). In the wild, blue crabs move to shallow and well-oxygenated water in response to hypoxia (Loesch, 1960; Das and Stickle, 1994). Wild crabs tracked by biotelemetry traveled continuously at speeds up to 5.4 m [min.sup.-1] while freely moving around the Chesapeake Bay, feeding 2 to 7 times a day (Wolcott and Hines, 1989). When exposed to mild (2 to 4 mg [l.sup.-1]) or severe (<2 mg [l.sup.-1]) hypoxia, the crabs reduced the amount of time spent feeding (Bell et al., 2003). Short-term shoreward migrations to avoid low dissolved oxygen have also been documented in the blue crab by fisherman observers (Loesch, 1960). The Atlantic blue crab routinely walks on the bottom and swims in the water column (Aguilar et al., 2005). In a tracking study. Eggleston et al. (2005) followed free-ranging blue crabs for hundreds of hours and found that to escape a hypoxic event the crabs walk on the bottom, rather than swim to a normoxic environment. However, moving laterally into the shallow oxygenated waters can be dangerous for the crab if overcrowding occurs, since other crabs can inflict carapace wounds and limb loss. There is also an increased risk of predation, which may compel some crabs to return to hypoxic areas.
In addition to short-term migrations to avoid hypoxic events, this crab displays seasonal migratory movements and may travel hundreds of kilometers, with normal daily movements in the range of 400 to 900 m (Schwartz, 1997; Tankersley et al., 1998). Male blue crabs also move from low-salinity tidal creeks where molting occurs, to brackish waters to find a mate (Hines et al., 1987). The ability of the crab to sustain long periods of locomotory activity--that is, its endurance capacity--appears to be important in multiple contexts (Vidal-Gadea et al., 2008). Thus, anything that compromises the ability of a blue crab to become and remain active, such as limb autotomy, disease, or hypoxia, could affect its survival.
It is unknown how hypoxia and hypercapnia affect locomotion and resistance to fatigue in the blue crab. This information becomes more important as the intensity and duration of coastal hypoxic events increase worldwide (Diaz and Rosenberg, 2008), including in the Gulf of Mexico and western Atlantic habitats of blue crabs.
We were interested in determining how water oxygenation and elevated C[O.sub.2] influence important locomotory behaviors and time to fatigue in blue crabs. Using sexually mature male blue crabs, we tested three hypotheses: (1) low oxygen conditions and hypercapnia decreases the time to fatigue during sustained locomotory activity, (2) as the hypoxia and hypercapnia intensify, so too do the behaviors indicating fatigue, and (3) severe hypoxia exacerbates changes in gait as the crab becomes fatigued. Walking performance was evaluated by exercising crabs on an aquatic treadmill to mimic the continuous activity necessary for behaviors such as finding food, migrating, and avoiding predators. We used various measures described by Stover et al. (2013) and the kinematics of leg movements to assess the effects of hypoxia and hypercapnia on blue crab performance (http://www.biolbull.org/content/supplemental).
Materials and Methods
Adult male Atlantic blue crabs, Callinectes sapidus, were collected in Charleston, South Carolina, and held at 25 [degrees]C at a salinity of 30 ppt, in well-aerated, recirculating seawater with a 12 h:12 h light/dark cycle. Intermolt crabs were kept for a minimum of 3 days but no more than 3 weeks before use in experiments. Every other day, crabs were fed either shrimp or squid that had been previously frozen. Food was withheld from crabs at least 24 h prior to an experiment. Animals used in these studies were 119 to 310 g and 120 to 159 mm in carapace width, measured from tip to tip of the lateral spines, well above the 82-mm carapace width of sexually mature males (Van Engel, 1990).
Animal health assessment
Health assessments were made on each crab 24 h prior to the walking exercise to exclude crabs with pre-existing conditions that might affect their ability to complete the exercise (Burnett et al., 2006; Thibodeaux et al., 2009). Animals with missing limbs or significant external lesions were excluded from the study. Subsequently, hemolymph was sampled from the infrabranchial sinus, with a 23-gauge needle and 1-ml syringe, and used for three health evaluations: presence of bacterial infections, counts of circulating hemocyte, and levels of hemocyanin.
Bacterial infections and hemocyte counts
Pre-existing bacterial infections were determined by diluting hemolymph (20 [micro].1) 1:10 in 10 m[mol.sup.-1] HEPES-buffered 2.5% NaCl, suspended in marine agar, and overlaid onto a sterile tryptic soy agar (TSA; Difco, Becton Dickinson and Co., Sparks, MD) microbial culture plate supplemented with 2.5% NaCl. Plates were incubated for 24 h at 25 [degrees]C; only crabs with no evidence of culturable bacteria in the hemolymph were used in this study. In doing this, we avoided using crabs whose oxygen transport system may have been compromised by an extensive bacterial infection (Burnett et at., 2006).
Similarly, another measure of the presence of an active immune response in blue crabs is the number of hemocytes circulating in the hemolymph. Total circulating hemocyte counts (THC) were performed by diluting hemolymph 1:15 in 20% neutral buffered formalin and then counting the number of hemocytes per volume with a hemocytometer (Mix and Sparks, 1980). Normal blue crab hemocyte counts are highly variable and fall into the range of 4 x [10.sup.6] to 128 x [10.sup.6] THC [ml.sup.-1], with an approximate average of 60 x [10.sup.6] THC [ml.sup.-1] (Sawyer et al., 1970; Holman et al., 2004). All crabs used in the walking exercise had at least 12.3 x 106 THC [ml.sup.-1].
Low hemocyanin levels
Hemolymph was allowed to clot, and a pestle was used to break the clot apart in a microcentrifuge tube. After centrifuging at 1300 x g for 6 min, the supernatant was diluted 3:100 in 10 mmol [l.sup.-1] EDTA with 2.5% NaCl at pH 10, and the absorbance was measured at 335 nm (Beckman DU 530, Life Sciences UV/Vis spectrophotometer, Atlanta, GA). Hemocyanin concentrations were calculated using the extinction coefficient from the related crab species Carcinus maenas (Nickerson and Van Holde, 1971). Crabs with hemocyanin concentrations below 3 g/100 ml of hemolymph were excluded from this study.
Exercise using a treadmill
All experiments were performed on crabs in seawater that was between 24 and 25.5 [degrees]C at 30 ppt salinity. Crabs were placed on a treadmill driven by a variable-speed DC motor and housed in a clear acrylic 130-1 chamber. Crabs were kept on the treadmill by an additional four-sided acrylic box (35 cm x 29.5 cm x 10 cm to the top of the water) in which they could move freely. Water was pumped into the box from the direction in which the crab was traveling to maintain circulation while water was bubbled with different gas mixtures (see below) at both ends of the tank. Animals walked at a speed of 8 m [min.sup.-1] until fatigue was reached. The treadmill speed of 8 m [min.sup.-1] is based on previous studies of blue crabs performing steady-state exercise for an extended period of time (Thibodeaux et al., 2009).
Measures of fatigue
We developed a "pull test" to simulate the reproductive behavior of a male crab holding onto a female crab. This pull test is designed to measure fatigue in the walking legs (Stover et al., 2013), simulating a male grabbing and holding onto a female both in precopula and in mating, a behavior exhibited by crabs during reproduction. In this test, a crab was allowed to clasp onto a stiff black plastic mesh net (cells 2.54 cm x 2.54 cm) in an underwater tank. The crab was then pulled by hand upward from the mesh, and the maximum force required to cause the crab to release the mesh was measured. This process was repeated two more times and the three measurements were averaged. The vertical force generated by the crab holding onto the mesh was measured using a Kistler force transducer 9300 Series (Kistler Instrument Corporation, Amherst, NY) and Kistler charge amplifier type 5995 (Kistler Instrument AG, Winterthur, Switzerland) connected by nonexpanding whipping twine to an underwater platform (Stover et al., 2013). With the crab centered over the transducer, the four contact points of the whipping twine allowed the force to be measured straight down. The entire apparatus was submerged in seawater identical to our treatment conditions (see below). An individual crab was considered fatigued when the maximum force causing the crab to release the mesh was 67% or less of the initial pull force for 2 sequential hours, such as h 5 and 6 (Fig. 1). This criterion for fatigue was chosen by preliminary experiments with animals walking in fully aerated seawater in which 80% of the initial pull force was reached at 3 h of walking and 60% was achieved by 5 to 6 h of walking. Before the pull force dropped to 50% of the initial value, most crabs would no longer walk. Force was standardized by calculating the percent pull force relative to the pull force measured at the start of the experiment. Thus, pull force was tested prior to walking and each hour during the walking exercise until the crab was considered fatigued (Fig. 1).
Two additional behaviors associated with fatigue were assessed for a period of 5 min at specific times during each hour of walking in the exercise trials (Fig. 1). These behaviors were 180[degrees] turns (a turn allows the crab to switch leading and trailing legs) and stopping--that is, riding to the end of the treadmill or holding at the end of the treadmill and not walking for 5 s. This type of intermittent walking has been shown to delay fatigue (Weinstein and Full, 1992). Each behavior took about 5 s to complete, so no more than 60 such behaviors could be counted during a 5-min period. Fatigue-indicating behaviors were recorded 10 min into each hour to give the crabs plenty of time back on the treadmill after they had been removed for the pull force assessment on the hour (Fig. 1). Fatigue-indicating behaviors were totaled for each 5-min assessment period during the exercise trial.
Crabs were subjected to one of three dissolved oxygen environments in the treadmill tank: (1) the control oxygen condition was normoxia or 100% air saturation (fully aerated water is 21.2 kPa [O.sub.2]); (2) moderate hypoxia, or 50% air saturation (10.4 [+ or -] 0.7 kPa [O.sub.2]); and (3) severe hypoxia, or 20% air saturation (4 [+ or -] 0.7 kPa [O.sub.2]). An additional treatment of moderate hypoxia, or 50% air saturation (10.4 [+ or -] 0.7 kPa) with hypercapnia (P[co.sub.2] = 2 kPa), was also completed in the treadmill tank. In a typical experiment, crabs were transferred from their holding tanks to the treadmill tank where the partial pressure of oxygen was maintained at a particular treatment level. Hypoxia was achieved by continuously gassing the water with a mixture of [O.sub.2] and [N.sub.2] and hypercapnic hypoxia was achieved by gassing with a mixture of [O.sub.2], [N.sub.2], and C[O.sub.2]. All mixtures were made using a gas mixer (Pegas 4000 MF gas mixer, Columbus Instruments, Columbus, OH). The oxygen pressure was monitored with an oxygen electrode (YSI Model 58, Yellow Springs, OH). The treatment conditions were established in the treadmill tank prior to the experiment and maintained in the treadmill tank, holding tank, and pull test tanks while the crab performed the exercise challenge. The pH of the water in the hypercapnic treatments was 6.7 to 6.8.
To assess their kinematic performances in the 100% and 20% air saturation treatments, the crabs were placed on a treadmill and walked until fatigue was reached, as described above. The 100% and 20% air saturation treatments were chosen for kinematic analysis because the first treatment represents the ideal condition a crab would see on a daily basis and the second the most extreme treatment that represents a realistic hypoxic event (Carpenter and Cargo, 1957). In these treatments, gait parameters were assessed using high-speed video capture, which was done with Casio EX-FH25 cameras, shooting from the sides and above, at 120 frames [s.sup.-1]. Three-dimensional (3D) kinematics and a high image-sampling rate were necessary because crab limbs move rapidly in more than one plane. The cameras were offset by 900 for an accurate 3D position of the subject, with a side and top view of the treadmill tank. At the beginning of walking and at fatigue, the cameras were calibrated with a 3D calibration cube roughly the size of the crab, with points of known coordinates in both planes measured with calipers. The calibration cube was made of colored plastic blocks and had 13 points (LEGO Systems, Inc., Billund, Denmark). The coordinate system used the convention of Full et al. (1991), with the positive x-axis directed horizontally and laterally toward the right of the animal, the positive y-axis directed horizontally and anteriorly, and the positive z-axis directed vertically and dorsally.
Video was recorded initially when the crab first began to walk on the treadmill and after the crab was considered fatigued (indicated as HS Video Acquisition in Fig. 1). Three consecutive strides of the trailing fourth pereopod were used for each measurement of gait parameters. Stride measurements were deemed acceptable when the crab kept pace with the treadmill, kept away from the walls of the enclosure, and moved sideways. Stride frequency, stride length, and duty factors were measured for each treatment. Stride length is the distance between two successive placements of the dactyl (Fig. 2A), stride frequency is the strides per unit time, and duty factor is the fraction of a stride when the foot is in contact with the surface. The crab's carapace and chelae concealed a large portion of the walking legs from view during locomotiop, and the leading legs often touched the edge of the enclosure or were blocked by the water circulator, so the trailing fourth pereopod was chosen as the focus for kinematics because most of the joints were exposed at multiple camera angles. Points on the fourth pereopod, or third walking appendage, were digitized by marking anatomical landmarks along the joints by latex squares glued onto the exoskeleton (Fig. 2A). Strides at the beginning of walking and at fatigue were chosen for digitization. The markers were digitized in each video frame using DTLDV3 (Hedrick, 2008) in Matlab ver. 7.12. Three-dimensional joint angles were then calculated to determine the maximum angular extension of each joint by using the dot product:
[theta] = arc cos(([x.sub.1] x [x.sub.2]) + ([y.sub.1] x [y.sub.2]) + ([z.sub.1] x [z.sub.2]) / ([square root of [x.sub.1.sup.2] + [y.sub.1.sup.2] + [z.sub.1.sup.2]]) x ([square root of [square root of [x.sub.2.sup.2] + [y.sub.2.sup.2] + [z.sub.2.sup.2]]])), (1)
where points 1 and 2 are centered around the marker on the joint in 3D space, as a point of reference, creating the angle.
The movement of a leg requires coordinated and synchronized movements of the joints. Synchrony is a measure of the limb joints performing the same function at the same time, such as extending the limb maximally. Synchrony has been used as a measure of inter-joint coordination (Sainburg et at., 1995) and may change with fatigue. The synchrony for each joint in a single stride was measured as the time of maximum extension after the time of liftoff of the dactyl.
Statistical analyses were performed with SigmaPlot ver. 11.0 statistical software and JMP ver. 8 (SAS Institute Inc., Cary, NC). An ANOVA on Ranks was used to test for differences in the duration of walking until fatigue for each of the treatments under low C[O.sub.2] conditions; subsequent pairwise analysis was done using the Dunn's method. Differences in the duration of walking at 50% air saturation with and without elevated C[O.sub.2] were assessed using a Mann-Whitney rank sum test.
An ordinary least-squares regression was used to describe the relationship between time and percent of initial pull force, as well as between time and fatigue behaviors for all individual animals in all treatments. Pull force data were normalized to the initial pull force because of the size range of crabs used in this study and the known relationship between pull force and size (Stover et al., 2013).
The linear relationships between the pull force and time and between fatigue behaviors and time for individual animals allowed us to test for significant differences among the slopes of the relationships (described above) by ANOVA (comparing three levels of dissolved oxygen) or Student's t-test (comparing two levels of C[O.sub.2]).
A two-way repeated measures analysis of variance was used to test parameters of stride frequency, stride length, duty factor, synchrony, maximum extension, and maximum excursion of the joint angles for differences between the beginning of walking and at fatigue, and between treatments (100% and 20% air-saturation).
Animal health assessments
Of 105 crabs tested, 82 (78%) were deemed healthy; 58 of these healthy crabs were used in the present study. These animals were caught during spring, summer, or early fall. In contrast, in a previous report by Stover et al. (2013), only 65% of blue crabs caught in spring passed the same battery of health assessment tests.
Time to fatigue
Crabs walked until they reached fatigue, defined as the maximum pull force generated prior to releasing a stiff mesh net (see above and Stover et al., 2013) that was 67% or less of their initial maximum pull force for 2 hours in a row. Time to fatigue decreased significantly (ANOVA on Ranks and Dunn's method, P < 0.001) from a mean of 6.12 h ([+ or -] 0.118 SEM, n = 17) at 100% air saturation to 4 h (all values = 4, n = 12) at 50% air saturation, and to 2.07 h ([+ or -]0.07 SEM, n = 14) at 20% air saturation. The addition of C[O.sub.2] at 50% air saturation significantly increased the time to fatigue (Mann-Whitney rank sum test, P = 0.005) to a mean of 4.58 h ([+ or -]0.19, n = 12) compared with 50% air saturation at low C[O.sub.2] (4 h, all values = 4, n = 12).
The pull force decreased with walking time in all treatments (Fig. 3); there was a significant negative and linear relationship between the percent of initial pull force and walking time for the 100% air saturation treatment ([R.sup.2] = 0.613, P < 0.001), 50% air saturation treatment without additional C[O.sub.2], ([R.sup.2] = 0.931, P < 0.001), 50% air saturation and hypercapnia treatment ([R.sup.2] = 0.884, P < 0.001), and 20% air saturation treatment ([R.sup.2] = 0.839, P < 0.000 (Fig. 3), confirming that pull force is a reasonable proxy for fatigue of the walking leg muscles.
The slopes of the three hypoxia treatment groups (100%, 50%, and 20% air saturation) were significantly different (ANOVA on Ranks; P < 0.001), with all pairwise combinations of treatments showing differences (Dunn's method; P < 0.05). In addition, the slopes of the relationship between pull force and time at 50% air saturation with and without the addition of C[O.sub.2] were significantly different (t-test; P = 0.027). Thus, the pull force decreased more rapidly as the level of hypoxia increased, by 8.0% [h.sup.-1], 11.4% [h.sup.-1], and 27.7% [h.sup.-1], with 100%, 50%, and 20% air saturation treatments, respectively. The addition of hypercapnia to the 50% air saturation treatment slowed the rate of decrease (from 11.4% [h.sup.-1] to 9.9% [h.sup.-1]).
The behaviors indicating fatigue increased with the time spent walking, and also as the level of hypoxia became more severe. There was a significant positive linear relationship between the fatigue behaviors (180[degrees] turns and stops) and walking time for all treatments: 100% air saturation ([R.sup.2] = 0.107, P < 0.001), 50% air saturation without additional C[O.sub.2] ([R.sup.2] = 0.514, P < 0.001), 20% air saturation treatment ([R.sup.2] = 0.515, P < 0.001), and 50% air saturation plus hypercapnia ([R.sup.2] = 0.199, P < 0.001) (Fig. 4).
The stops and turns characterizing fatigue behavior of the three hypoxia treatment groups were significantly different (ANOVA on Ranks; P < 0.001), with all pairwise combinations of treatments showing differences (Dunn's method; P < 0.05). However, the slopes of the relationship between behaviors and time at 50% air saturation with and without the addition of C[O.sub.2], were not significantly different (t-test: P = 0.121). Thus, the number of fatigue behaviors increased as the level of hypoxia increased by 0.9 behaviors [h.sup.-1], 4.1 behaviors [h.sup.-1], and 13.8 behaviors [h.sup.-1], for the 100%, 50% air saturation without additional C[O.sub.2], and 20% air saturation treatments, respectively. The addition of C[O.sub.2] at 50% air saturation resulted in 2.7 behaviors [h.sup.-1], but this was not different from that without the addition of C[O.sub.2] (4.1 behaviors [h.sup.-1]).
Stride frequency (Table 1) was sensitive to hypoxia ([F.sub.1,18]= 5.555, P = 0.030) and the time of exposure ([F.sub.1,18]= 15.822, P < 0.001), both independently and interactively ([F.sub.1,18] = 6.53, P = 0.020). Mean stride frequencies ([s.sup.-1]) were equivalent initially in the 100% and the 20% air saturation treatments (Holm-Sidak method, P = 0.925). At fatigue, the mean stride frequency remained about the same in the 100% air saturation treatment but decreased significantly in the 20% air saturation treatment (Holm-Sidak method, P = 0.001).
Table 1 Mean gait parameters [+ or -] SEM (n = 10 at the beginning of walking and at fatigue in two air saturation treatments Parameter Time % Air Saturation * 100 20 Stride Frequency (str Initial 1.26 [+ or 1.26 [+ or -] [s.sup.-1]) -] 0.06 0.06 (a) (a) Fatigue 1.19 [+ or 0.90 [+ or -] -] 0.07 0.04 (b) (a) Stride Length (m) Initial 0.108 [+ or 0.109 [+ or -] -] 0.006 0.006 (a) (a) Fatigue 0.115 0.152 [+ or -] 0.006 (a) 0.008 (b) Duty Factor (fraction of a Initial 0.48 [+ or 0.45 [+ or -] stride when loot contacts -] 0.02 0.02 (a) the surface) (a) Fatigue 0.44 [+ or 0.45 [+ or -] -] 0.03 0.01 (a) (a) * Different superscript letters denote significantly different means in pairwise comparisons within time or treatment for each parameter (Holm-Sidak method).
Oxygenation ([F.sub.1,18] = 7.103, P = 0.016) and fatigue ([F.sub.1,18] = 21.390, P < 0.001) also significantly affected stride 1,18 length interactively ([F.sub.1,18] = 10.92, P = 0.004) (Table 1). Initial mean stride length (m [str.sup.-1]) was similar in both the 100% and the 20% air saturation treatments (Holm-Sidak method, P = 0.907). At fatigue, the mean stride length remained similar to initial values in 100% air saturation but increased significantly in 20% air saturation (Holm-Sidak method, P < 0.001). Duty factor (Table 1) did not significantly change with fatigue ([F.sub.1,18] = 0.829, P = 0.375), treatment ([F.sub.1,18] = 0.209, P = 0.653), or interactively ([F.sub.1,18] = 6.53, P = 0.423).
The basi-ischium-merus (J1) maximum joint angle extension increased significantly with fatigue ([F.sub.1,18] = 33.43, P < 0.001; Table 2). This increase occurred at 100% air saturation (Holm-Sidak method, P = 0.013) and at 20% air saturation (Holm-Sidak method, P < 0.001). There was no difference overall between treatments ([F.sub.1,18] = 0.788, P = 0.368). In contrast, the merus-carpus (J2) maximum joint angle extension increased significantly with the 20% air saturation treatment ([F.sub.1,18] = 7.97, P = 0.011; Table 2), but not as a result of fatigue overall ([F.sub.1,18] = 0.480, P = 0.497). However, the difference between treatments was significant only at fatigue (Holm-Sidak method, P = 0.007). In preliminary studies we found that the carpus-propodus (J3) joint is a very stiff joint with little or no flex during a stride, so no measurements were taken of this joint angle. Neither fatigue nor treatment affected the maximum extension of the propodus-dactyl (J4) joint angle extension (Table 2).
Table 2 Mean joint angles [+ or -] SEM (n = 10) at the beginning of walking and at fatigue in two air saturation treatments Parameter Time % Air Saturation * [dagger] 100 20 Max Extension Initial 155.1[degress] [+ 154.8[degress] [+ or J1 or -] 3.0 (a) -] 2.4 (a) Fatigue 161.9[degress] [+ 168.2[degress] [+ or or -] 2.9 (b) -] 2.4 (b) Max Extension Initial 156.4[degress] [+ 163.9[degress] [+ or J2 or -] 3.5 (a) -] 2.3 (a), (b) Fatigue 152.1[degress] [+ 164.8[degress] [+ or or -] 3.7 (a) -] 2.6 (b) Max Extension Initial 145.5[degress] [+ 146.2[degress] [+ or J4 or -] 1.6 (a) -] 4.1 (a) Fatigue 138.7[degress] [+ 148.5[degress] [+ or or -] 3.8 (a) -] 4.6 (a) Synch al LO J2 Initial 17 [+ or -] 6 22 [+ or -] 5 (a), (ms) (a) (b) Fatigue 42 [+ or -] 11 33 [+ or -] 6 (b) (b) Synch at LO J4 Initial 21 [+ or -] 3 24 [+ or -] 4 (a) (ms) (a) Fatigue 48 [+ or -] 14 32 [+ or -] 9 (a) (a) Max Excursion Initial 33.8[degress] [+ 30.2[degress] [+ or J1 or -] 2.7 (a) -] 3.3 (a) Fatigue 39.5[degress] [+ 42.2[degress] [+ or or -] 3.1 (a), -] 4.8 (b) (b) [dagger] Different superscript letters denote significantly different means in pairwise comparisons within time or treatment for each parameter (Holm-Sidak method). * Max Extension is the peak extension of a particular joint angle at liftoff, in degrees; Synch at LO is the synchrony at liftoff of the angle for joints 2 (merus-carpus) and 4 (propodus-dactyl), in milliseconds; Max Excursion J1 is the difference between the minimum (touchdown) and maximum (liftoff) joint angles for joint J1, the basi-ischium-merus joint.
The synchrony at liftoff for the basi-ischium-merus joint angle (J1) did not vary from 0 in any of our measurements, and it was unaffected by time or treatment. The synchrony of the merus-carpus joint angle (J2) changed significantly with fatigue ([F.sub.1,18] = 5.66, P = 0.029; Table 2), but only at 100% air saturation (Holm-Sidak method, P = 0.031), while there was no significant difference with fatigue or treatment for the propodus-dactyl (J4) joint angle synchrony at liftoff. As stated above, the carpus-propodus (J3) joint angle is a stiff joint during a stride and no measurements of synchrony were made. The maximum excursion at the basi-ischium-merus joint (J1) increased significantly over time for both treatments ([F.sub.1,18] = 10.76, P = 0.004; Table 2). Maximum excursion was calculated only for J1 because this is the angle closest to the body with the largest muscles associated with it and likely has the most influence on the movement of the limb.
Hypoxia has profound effects on the locomotion of the Atlantic blue crab. Crabs fatigued more quickly and displayed more fatigue-indicating behaviors when hypoxia intensified (Figs. 3, 4). However, the addition of hypercapnia in moderate hypoxia (50% air saturation) slightly delayed the time to the onset of fatigue while the crabs walked on a treadmill. Gait changed with fatigue in both the 100% air saturation and severe hypoxia treatments (the only treatments where gait was assessed), but the changes were more extreme in severe hypoxia (Fig. 2B: Table 2). These findings indicate that the ability of blue crabs to perform continuous activity such as walking is compromised in hypoxic waters.
Not only would hypoxia endanger the crab's ability to forage and avoid predators, it might affect its ability to escape a hypoxic event altogether. The hypoxic "Dead Zone" in the Gulf of Mexico can reach 20,700 [km.sup.2] (Rabalais et al., 2002), and while other hypoxic events along the southeastern coast of the United States are much smaller, they may still pose a challenge for crabs to escape. The crabs that walked for the longest duration in the present study walked a distance of 3.36 km, while the crabs that walked the shortest time could go only 0.96 km, which may not get them to the shallow, well-oxygenated waters. Not only must a crab be able to escape an event, it must forage and avoid predators on a daily basis. Clearly, impaired locomotory ability during low oxygen conditions may affect the survival of this and other crustacean species.
Even if a crab can escape hypoxic conditions, low dissolved oxygen may have sublethal effects on the animal. The present study documents large decreases in pull force in blue crabs forced to walk in moderate or severe hypoxia (Fig. 3). The pull force test is designed to measure the capacity of a male crab to clasp a stiff mesh net, mimicking the clasping of a female crab during copulation and mate guarding (Stover et al., 2013); this capacity decreases with the duration of walking and the intensity of hypoxia. The pull forces of crabs walking in normoxic water decreased very little in the first 4 h, but rapidly declined (28%) between hours 4 and 5. That the largest decrease in pull force was not seen until after 4 h of walking suggests that blue crabs are able to sustain constant and moderate levels of performance during this time under ideal oxygen conditions. This performance will, of course, decline as oxygen levels fall (Figs. 3 and 4) or possibly when crabs walk against a strong current or at a faster speed. Interestingly, this period of optimal performance is slightly under the 6-h period of tidal inflow or outflow in the Charleston Harbor estuary. In natural conditions, crabs are unlikely to maintain a constant walking speed. What is more likely is that they will engage in combinations of walking, with stops and turns, at varying speeds and swimming.
In walking animals, behaviors indicating fatigue (1800 turns and stops) increased more rapidly as the level of hypoxia intensified, with more than an 18-fold increase in rate for crabs in severe hypoxia compared to normoxia. This increase in the behaviors indicating fatigue was similar to the decrease in pull force exhibited as the crabs approached exhaustion. In severe hypoxia, crabs became dependent on intermittent walking and on their ability to switch leading and trailing walking legs by performing 180[degrees] turns. Ghost crabs also use intermittent walking to delay fatigue; endurance capacity is 4.5 times greater during intermittent walking than during continuous walking (Weinstein and Full, 1998). Blue crabs may be more likely to use intermittent walking to mitigate fatigue, or they may simply bury themselves in the sediment during a severely hypoxic event when escape is not achieved quickly enough and they can no longer locomote (deFur et al., 1990).
Severe hypoxia (20% air saturation) exacerbated changes in gait as crabs became fatigued. Intermittent walking and 180[degrees] turns initially delay the onset of fatigue and the associated gait changes, as the results of this study show; however, at a certain point these behaviors are not enough to avoid exhaustion. Severe hypoxia not only fatigued the crabs much more rapidly, but also brought about changes in gait. As the crabs reached fatigue, their stride length increased while their stride frequency decreased, so the crabs were making slower and longer strides than at 0 h. The basi-ischium-merus (J1) joint angle extension and maximum excursion increased significantly with time, indicating that the associated muscles of this joint, closest to the body, fatigue rapidly and perhaps explaining increases in stride length (Fig. 2B). The merus-carpus (J2) joint angle extension increased significantly with treatment and also contributed to the increase in stride length in the severe hypoxia treatment. The synchrony of this same joint (J2) decreased at fatigue in 100% air saturation, indicating that the crabs were losing some interjoint coordination as they approached exhaustion.
A decrease in stride frequency with fatigue, as demonstrated in the present study, has been documented in human runners (Mizrahi et al., 2000; Hunter and Smith, 2007). Horses galloping on a treadmill also decreased their stride frequency and increased their stride length as they reached fatigue (Johnston et al., 1999; Colborne et al., 2001), much like the crabs in the present study. Vertical excursions of a trunk marker on the horse also increased as fatigue was reached (Colborne et al., 2001), resembling the basi-ischium-merus joint angle increase in extension and excursion that helped to lengthen stride length with fatigue in the severely hypoxic crabs. The kinematics of fatigue during locomotion is an understudied topic due to issues with collecting data for an extensive period in animals that are seemingly difficult to fatigue, especially in a laboratory setting. Furthermore, a metabolic advantage may be had when approaching fatigue. Hunter and Smith (2007) studied human runners exercising intensely and documented a decreased stride frequency; they suggested that this phenomenon permits operation at a minimum metabolic cost. In another study of humans, Donelan et al. (2001) suggested that adjustments in stride length can minimize metabolic cost. Thus, changing the stride length and frequency may be a metabolic cost advantage in crabs.
The blue crabs used in the present study were healthy wild-caught specimens. The health assays eliminated most of the variability in time to fatigue seen in crabs in normoxic seawater in a preliminary study. Health issues such as limb loss, bacterial infection, or parasitic infection could negatively affect the ability of blue crabs to become and remain active. Limb autotomy occurs in 19% to 39% of the population and has been shown to adversely affect the ability of a male blue crab to defend females from competing males with all of their limbs (Smith and Hines, 1991; Smith, 1992). Crabs are more likely to lose their limbs from fighting when they reach shallow waters during chronic hypoxia because they are more likely to encounter high densities of their own kind (Loesch, 1960; Eggleston et al., 2005). Autotomy of the walking appendages would change a crab's gait and could have a negative impact on locomotion, especially if more than one limb were lost, making it harder for the compromised animal to escape hypoxic waters. Parasites and disease could also negatively impact the crab's locomotion in the short term, with more energy needed to fight the infection and with respiration impaired due to hemocyte aggregates forming in response to pathogens (Burnett et al., 2006). Additionally, low hemocyanin levels in the blue crab could impair oxygen delivery and aerobic respiration, making walking performance more difficult to maintain. In the long term, health issues like bacterial infections, viral infections, limb loss, or anemia could mean an even greater impairment of locomotion when there are low levels of dissolved oxygen.
The addition of hypercapnia to moderate hypoxia delayed the onset of fatigue in crabs with a slower decrease in pull force (Fig. 3). The addition of C[O.sub.2] in moderate hypoxia is associated with several physiological responses, and under those treatment conditions, oxygen transport to the active tissues may be a limiting factor. First, hypercapnia will decrease pH by more than 0.4 units, bringing about a large respiratory acidosis that will take 2 to 3 hours to fully develop (Cameron, 1978). Such an acidosis can have a profound negative influence on oxygen transport. Compensation for the acidosis will occur (Cameron, 1978), but not in the time-frame of our experiments. This acidosis influences oxygen transport because the hemocyanin of the blue crab has a fairly strong Bohr shift (Mangum and Burnett. 1986); any acidosis will decrease hemocyanin oxygen affinity. However, there is a significant C[O.sub.2]-specfic effect on blue crab hemocyanin that results in a large increase in hemocyanin oxygen affinity, with [P.sub.50] decreasing by as much as 1.3 kPa (Mangum and Burnett, 1986) at a resting pH of 7.6 (Booth et al., 1982). The hemolymph pH in a blue crab drops from 7.6 to 7.1 after 25 min of moderate swimming (Booth et al., 1982), and at a pH of 7.1 C[O.sub.2], has a more profound influence on hemocyanin oxygen affinity, decreasing [P.sub.50] by about 4 kPa (Mangum and Burnett, 1986). This counterbalancing C[O.sub.2]-specfic effect on hemo-cyanin oxygen affinity may overcome the negative effect of pH, which may account for the delay in fatigue, as measured by pull force, compared with moderate hypoxia alone. Clearly hemocyanin oxygen affinity is also strongly influenced by lactate in some crabs, including C. sapidus (Booth et al., 1982; Graham et al., 1983; Johnson et al., 1984). Some lactate is produced in C. sapidus during walking (Thibodeaux et al., 2009) and under the hypoxic experimental conditions of the present study, but the differences in lactate concentrations in the two moderate hypoxia treatments are unknown.
Our treatment levels of dissolved oxygen and hypercapnia (50% air saturation with hypercapnia, 50% air saturation. and 20% air saturation) are seen very often during the warmer summer months in estuaries in the western Atlantic and in the Gulf of Mexico, and can even occur daily, especially in the tidal creeks (Hackney et al., 1976; Cochran and Burnett, 1996; Wenner et al., 2004; Diaz and Rosenberg, 2008). The intensity and duration of these hypoxic events are increasing worldwide and will continue to affect the blue crab's physiology and behavior (Diaz and Rosenberg, 1995). Although blue crabs can increase hemocyanin levels to deal with low oxygen levels, this can only be done over a period of days in response to hypoxia (deFur et al., 1990). If oxygen levels in their estuarine habitats continue to decrease, blue crab survival may be at risk due to harmful impacts on the ability to reproduce, to forage, to escape from predators, and to flee from severely hypoxic events.
In summary, our study has five main findings. First, we have shown that hypoxia fatigues crabs more rapidly as the levels of oxygen fall. Second, the pull forces measured by our pull force test suggest that the ability of male crabs to clasp onto a female crab during copulation and mate-guarding decline with time spent walking, and that this rate of decline is more severe as water becomes more hypoxic. Third, fatigue-indicating behaviors (180[degrees] turns and stopping) increase with time spent walking, and these behaviors increase as water becomes hypoxic. Fourth, in severe hypoxia, crabs change their gait in a manner similar to that of other animals by decreasing stride frequency and increasing stride length. This may result in a metabolic cost advantage. And fifth, the addition of C[O.sub.2] to moderate hypoxia improves crab performance; crabs take slightly longer to fatigue and maintain greater pull forces compared to their controls.
We thank Darwin Jorgensen for his help and expertise on this project. We also thank Tom Stover for his artistic assistance. This is contribution No. 402 of the Grice Marine Laboratory, College of Charleston. This project was supported with funding from the National Science Foundation (IOS-0725245 to L.E.B. and K.G.B.).
Received 17 July 2012; accepted 5 April 2013.
* To whom correspondence should be addressed. E-mail: email@example.com
Aguilar, R., A. H. Hines, T. G. Wolcott, D. L. Wolcott, M. A. Kramer, and R. N. Lipcius. 2005. The timing and route of movement and migration of post-copulatory female blue crabs. Callinectes sapidus Rathbun, from the upper Chesapeake Bay. J. Exp. Mar. Biol. Ecol. 319: 117-128.
Bell, G. W., D. B. Eggleston, and T. G. Wolcott. 2003. Behavioral responses of free-ranging blue crabs to episodic hypoxia. II. Feeding. Mar. Ecol. Prog. Ser. 259: 227-235.
Booth, C. E., B. R. McMahon, and A. W. Pinder. 1982. Oxygen uptake and the potentiating effects of increased hemolymph lactate on oxygen transport during exercise in the blue crab Callinectes sapidus. J. Comp. Physiol. 148: 111-121.
Burnett, L. E. 1997. The challenges of living in hypoxic and hypercapnic aquatic environments. Am. Zool. 37: 633-640.
Burnett, L. E., J. D. Holman, D. D. Jorgensen, J. L. Ikerd, and K. G. Burnett. 2006. Immune defense reduces respiratory fitness in Callinectes sapidus, the Atlantic blue crab. Biol. Bull. 211: 50-57.
Cameron, J. N. 1978. Effects of hypercapnia on blood acid-base status, NaCl fluxes, and trans-gill potential in freshwater blue crabs Callinectes sapidus. J. Comp. Physiol. 123: 137-141.
Carpenter. J. H., and D. G. Cargo. 1957. Oxygen Requirement and
Mortality of the Blue Crab in the Chesapeake Bay. Chesapeake Bay Institute Technical Report 13, Johns Hopkins University, Baltimore.
Cochran, R. E., and L. E. Burnett. 1996. Respiratory responses of the salt marsh animals, Fundulus heteroclitus, Leiostomus xanthurus, and Palaemonetes pugio to environmental hypoxia and hypercapnia and to the organophosphate pesticide, azinphosmethyl. J. Exp. Mar. Biol. Ecol. 195: 125-144.
Colborne, G. R., D. M. Birtles, and I. C. Cacchione. 2001. Electromyographic and kinematic indicators of fatigue in horses: a pilot study. Equine Vet. J. 33: 89-93.
Das, T., and W. B. Stickle. 1993. Sensitivity of crabs Callinectes sapidus and C. similis and the gastropod Stramonita haemastoma to hypoxia and anoxia. Mar. Ecol. Prog. Ser. 98: 263-274.
Das, T., and W. B. Stickle. 1994. Detection and avoidance of hypoxic water by juvenile Callinectes sapidus and C. similis. Mar. Biol. 120: 593-600.
deFur, P. L., C. P. Mangum, and J. E. Reese. 1990. Respiratory responses of the blue crab Callinectes sapidus to long-term hypoxia. Biol. Bull. 178: 46-54.
Diaz, R. J., and R. Rosenberg. 1995. Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of benthic macrofauna. Pp. 245-303 in Oceanography and Marine Biology: An Annual Review, A. D. Ansell, R. N. Gibson, and M. Barnes, eds. University College of London Press.
Diaz, R. J., and R. Rosenberg. 2008. Spreading dead zones and consequences for marine ecosystems. Science 321: 926-929.
Donelan, J. M., R. Kram, and A. D. Kuo. 2001. Mechanical and metabolic determinants of the preferred step width in human walking. Proc. R. Soc. B Biol. Sci. 268: 1985-1992.
Eggleston, D. B., G. W. Bell, and A. D. Amavisca. 2005. Interactive effects of episodic hypoxia and cannibalism on juvenile blue crab mortality. J. Exp. Mar. Biol. Ecol. 325: 18-26.
Full, R. J., R. Blickhan, and L. H. Ting. 1991. Leg design in hexapedal runners. J. Exp. Biol. 158: 369-390.
Graham, R. A., C. P. Mangum, R. C. Terwilliger, and N. B. Terwilliger. 1983. The effect of organic acids on oxygen binding of hemocyanin from the crab Cancer magister. Comp. Biochem. Physiol. A Physiol. 74: 45-50.
Hackney, C., W. Burbanck, and 0. Hackney. 1976. Biological and physical dynamics of a Georgia tidal creek. Chesapeake Sci. 17: 271-280.
Hedrick, T. L. 2008. Software techniques for two- and three-dimensional kinematic measurements of biological and biomimetic systems. Bioinspir. Biontim, doi:10.1088/1748-3182/3/3/034001.
Hines, A. H., R. N. Lipcius, and A. M. Haddon. 1987. Population dynamics and habitat partitioning by size, sex. and molt stage of blue crabs Callinectes sapidus in a subestuary of central Chesapeake Bay. Mar. Ecol. Prog. Ser. 36: 55-64.
Holman, J. D, K. G. Burnett, and L. E. Burnett. 2004. Effects of hypercapnic hypoxia on the clearance of Vibrio campbellii in the Atlantic blue crab, Callinectes sapidus Rathbun. Biol. Bull. 206: 188-196.
Hunter, I., and G. A. Smith. 2007. Preferred and optimal stride frequency, stiffness and economy: changes with fatigue during a 1-h high-intensity run. Ear. J. Appl. Physiol. 100: 653-661.
Johnson, B. A., C. Bonaventura, and J. Bonaventura. 1984. Allosteric modulation of Callinectes sapidus hemocyanin by binding of L-lactate. Biochemistry 23: 872-878.
Johnston, C., M. Gottlieb-Vedi, S. Drevemo, and L. Roepstorff. 1999. The kinematics of loading and fatigue in the Standardbred trotter. Equine Vet. J. 31: 249-253.
Loesch, H. 1960. Sporadic mass shoreward migrations of demersal fish and crustaceans in Mobile Bay, Alabama. Ecology 41: 292-298. Mangum, C. P., and L. E. Burnett, Jr. 1986. The C[O.sub.2] sensitivity of the hemocyanins and its relationship to [Cl.sup.-] sensitivity. Biol. Bull. 171: 248-263.
Mix, M. C., and A. K. Sparks. 1980. Tanner crab Chionocetes bairdi Rathbun haemocyte classification and an evaluation of using differential counts to measure infection with a fungal disease. J. Fish Dis. 3: 285-293.
Mizrahi, J., O. Verbitsky, E. Isakov, and D. Daily. 2000. Effect of fatigue on leg kinematics and impact acceleration in long distance running. Hum. Mov. Sci. 19: 139-151.
Nickerson, K. W., and K. E. Van Holde. 1971. A comparison of molluscan and arthropod hemocyanin. 1. Circular dichroism and absorption spectra. Comp. Biochem. Physiol, B 39: 855-872.
Rabalais, N. N., R. E. Turner, and W. J. Wiseman, Jr. 2002. Gulf of Mexico hypoxia, a.k.a. "The Dead Zone." Anna. Rev. Ecol. Syst. 33: 235-263.
Sainburg, R. L., M. F. Ghilardi, H. Poizner, and C. Ghez. 1995. Control of limb dynamics in normal subjects and patients without proprioception. J. Neurophsiol. 73: 820-835.
Sanger, D. M., M. D. Arendt, Y. Chen, E. L. Wenner, A. F. Holland, D. Edwards, and J. Caffrey. 2002. A Synthesis of Water Quality Data: National Estuarine Research Reserve System-wide Monitoring Program (1995-2000). National Estuarine Research Reserve Technical Report Series 2002:3. South Carolina Department of Natural Resources. Marine Resources Division, Charleston, SC.
Sawyer, T. K., R. Cox, and M. Higginbottom. 1970. Hemocyte values in healthy blue crabs, Callinectes sapidus, and crabs infected with the amoeba. Paramoeba pemiciosa. J. Invertebr Pathol. 15: 440-446.
Schwartz, F. J. 1997. Distance movements of fishes, white shrimp, and blue crabs tagged in or near the estuarine Cape Fear River and adjacent Atlantic Ocean, North Carolina, 1973 through 1978. J. Elisha Mitchell Sci. Soc. 113: 123-132.
Smith, L. D. 1992. The impact of limb autotomy on mate competition in blue crabs. Callinectes sapidus Rathbun. Oecologia 89: 494-501.
Smith, L. D., and A. H. Hines. 1991. Autotomy in blue crab (Calli-nectes sapidus Rathbun) populations: geographic, temporal, and ontogenetic variation. Biol. Bull. 180: 416-431.
Stover, K. K., K. G. Burnett, E. J. McElroy, and L. E. Burnett. 2013. Locomotory fatigue and size in the Atlantic blue crab, Callinectes sapidus. Biol. Bull 224: 63-67.
Tankersley, R. A., M. G. Wieber, M. A. Sigala, and K. A. Kachurak. 1998. Migratory behavior of ovigerous blue crabs Callinectes sapidus: evidence for selective tidal-stream transport. Biol. Bull. 195: 168-173.
Taylor, D. L., and D. B. Eggleston. 2000. Effects of hypoxia on an estuarine predator-prey interaction: foraging behavior and mutual interference in the blue crab Callinectes sapidus and the infaunal clam prey Mya arenaria. Mar. Ecol. Frog. Ser. 196: 221-237.
Thibodeaux, L. K., K. G. Burnett, and L. E. Burnett. 2009. Energy metabolism and metabolic depression during exercise in Callinectes sapidus, the Atlantic blue crab: effects of the bacterial pathogen Vibrio camphellii. J. Exp. Biol. 212: 3428-3439.
Van Engel, W. A. 1990. Development of the reproductively functional form in the male blue crab. Callinectes sapidus. Bull. Mar. Sci. 46: 13-22.
Vidal-Gadea, A. G., M. D. Rinehart, and J. H. Belanger. 2008. Skeletal adaptations for forwards and sideways walking in three species of decapod crustaceans. Arthropod Struct. Dev. 37: 95-108.
Weinstein, R. B., and R. J. Fun. 1992. Intermittent exercise alters endurance in an eight-legged ectotherm. Ain. J. Physiol. 262: R852--R859.
Weinstein, R. B., and R. J. Full. 1998. Performance limits of low-temperature, continuous locomotion are exceeded when locomotion is intermittent in the ghost crab. Physiol. Zool. 71: 274-284.
Wenner, E., D. Sanger, M. Arendt, A. F. Holland, and Y. Chen. 2004. Variability in dissolved oxygen and other water-quality variables within the national estuarine research reserve system. J. Coast. Res. SI (45): 17-38.
Whitaker, J. D., L. B. Delancey, J. E. Jenkins, and M. B. Maddox. 1998. A review of the fishery and biology of the blue crab, Callinectes sapidus, in South Carolina. J. Shellfish Res. 17: 459-463.
Wolcott, T. G., and A. H. Hines. 1989. Ultrasonic biotelemetry of muscle activity from free-ranging marine animals: A new method for studying foraging by blue crabs (Callinectes sapidus). Biol. Bull. 176: 50-56.
KRISTIN K. STOVER *, KAREN G. BURNETT, ERIC J. McELROY, AND LOUIS E. BURNETT
Hollings Marine Laboratory, 331 Fort Johnson., and Grice Marine Laboratory, College of Charleston, 205 Fort Johnson, Charleston, South Carolina 29412
|Printer friendly Cite/link Email Feedback|
|Author:||Stover, Kristin K.; Burnett, Karen G.; McElroy, Eric J.; Burnett, Louis E.|
|Publication:||The Biological Bulletin|
|Date:||Apr 1, 2013|
|Previous Article:||Locomotory fatigue and size in the Atlantic blue crab, Callinectes sapidus.|
|Next Article:||Energetics of byssus attachment and feeding in the green-lipped mussel Perna canaliculus.|