Behavioural and Physiological Implications of a Burrow-dwelling Lifestyle for Two Species of Upogebiid Mud-shrimp (Crustacea: Thalassinidea)

Estuarine, Coastal and Shelf Science (1997) 44, 155–168

Behavioural and Physiological Implications of a Burrow-dwelling Lifestyle for Two Species of Upogebiid Mud-shrimp (Crustacea: Thalassinidea) C. M. Astalla, A. C. Taylora and R. J. A. Atkinsonb a b

Institute of Biomedical & Life Sciences, University of Glasgow, Glasgow G12 8QQ, Scotland, U.K. University Marine Biological Station, Millport, Isle of Cumbrae KA28 0EG, Scotland Upogebia stellata and U. deltaura (Crustacea: Thalassinidea) construct burrows in nearshore sediments in U.K. waters. Burrow structure is similar in both species; the basic burrow consisting of a two-opening, U-shaped section with a vertical shaft descending from the mid-point of the U. This structure may be variously elaborated. Burrow cross-section is circular, dilations allow turning by somersaulting and surface openings are often constricted. Conditions within the burrows are usually hypoxic and hypercapnic. Burrow water PO2 in the parts normally occupied by the mud-shrimp was between 80–110 Torr, but was much lower (10–45 Torr) in the deepest, poorly-irrigated parts. Both species irrigate their burrows by episodes of pleopod beating of variable duration (mean=8·5&3·5 min and 2·8&0·5 min for U. deltaura and U. stellata, respectively), which draws oxygenated water into the burrow and also particulate food for suspension feeding. When exposed to hypoxia, U. deltaura and U. stellata were able to maintain their rates of oxygen consumption approximately constant over a wide range of PO2 (Pc=30–50 Torr). Under these conditions, there was a pronounced increase in scaphognathite beat rate but heart rate remained relatively constant. Below the Pc, however, both rates declined. ? 1997 Academic Press Limited Keywords: Crustacea; Thalassinidea; burrows; hypoxia; oxygen consumption; irrigation; Clyde Sea area

Introduction Thalassinidean decapods are an abundant component of littoral and sublittoral inshore benthos (Griffis & Suchanek, 1991). Their often deep burrows and, in the case of upogebiids, small burrow openings, have resulted in thalassinidean abundance being underestimated using remote sampling techniques. Their ecological importance, however, may be considerable. For example, Dworschak (1981) estimated that the entire volume of a neap tide was passed through the burrows of Upogebia pusilla on an Adriatic tidal flat. The importance of various thalassinidean species in sediment turnover and nutrient recycling is well documented (e.g. Aller et al., 1983; Koike & Mukai, 1983; Waslenchuck et al., 1983; Aller & Yingst, 1985; Forster & Graf, 1995; Nickell & Atkinson, 1995; Nickell et al., 1995). They are important determinants of community structure (Suchanek, 1983; Posey, 1986; Posey et al., 1991) and are implicated in the re-distribution of pollutants and trace metals that impinge on the sea-bed (McMurty et al., 1985; AbuHilal et al., 1988; Whitehead et al., 1988; Hughes & Atkinson, 1997). Burrow morphology has been described for many thalassinidean species (see reviews of Dworschak, 0272–7714/97/020155+14 $25.00/0/ec960207

1983; de Vaugelas, 1984; Atkinson & Taylor, 1988; Bromley, 1990) and further information is accumulating rapidly (e.g. Nickell & Atkinson, 1995; Rowden & Jones, 1995). That burrow morphology may reflect the lifestyle of the burrower appears to be a reasonable proposition. Several models have been constructed to relate burrow structure to lifestyle and trophic mode (de Vaugelas, 1984; Suchanek, 1985; Griffis & Suchanek 1991). Problems associated with present models are discussed by Dworschak and Ott (1993), Rowden and Jones (1995), and by Nickell and Atkinson (1995), who suggest a different approach. The present paper adds to the literature on burrow morphology in upogebiids by describing the burrows of U. stellata (Montagu) and U. deltaura (Leach). The only other information on the burrows of these species derives from three laboratory casts of U. deltaura burrows (Dworschak, 1983), and three field casts of U. stellata burrows together with some laboratory observations (Nickell & Atkinson, 1995). Thalassinidean burrows not only provide protection from predators, but also buffer external environmental perturbations and provide a locus for feeding, moulting and breeding (Atkinson & Taylor, 1988; Bromley, 1990). Although burrows may confer these advantages, the occupants are faced with the potential ? 1997 Academic Press Limited

156 C. M. Astall et al.

problems of reduced oxygen tension (hypoxia), elevated carbon dioxide (hypercapnia) (Atkinson & Taylor, 1988) and high sulphide concentrations (Astall, 1993; Johns et al., 1996). Burrow irrigation behaviour influences the degree of oxygenation within the burrow, and the complexity of burrow architecture affects the spatial distribution of oxygen through the burrow. Such irrigation allows the introduction of oxygen deep into otherwise anoxic sediment (Aller & Yingst, 1985). Burrow irrigation studies in burrowing decapods are few in number (Dworschak, 1981; Gust & Harrison, 1981; Aller et al., 1983; Koike & Mukai, 1983; Allanson et al., 1992; Nickell, 1992; Forster & Graf, 1995), and most are based on either theoretical or indirect evaluations rather than direct measurements. In the present study, the degree of water flow and the irrigation profile for U. stellata and U. deltaura have been measured directly using an electromagnetic flow probe. The spatial and temporal distribution of oxygen through their burrows have been investigated, as have their rates of oxygen consumption and respiratory responses during hypoxia. Materials and methods Collection and maintenance of shrimps Upogebia stellata were obtained by anchor dredging (depths to 25 m) around the Isle of Cumbrae (Main Channel and White Bay), Scotland, U.K. (55)9*N, 5)11*W). Upogebia deltaura were collected with an anchor dredge (7–10m) from Plymouth Sound (50)20*N, 47)13*W), Plymouth, U.K. and from the North Sea (54)4*N, 4)50*W). Once captured, individuals were transferred to Perspex tanks (35#25#12 cm) filled to 25 cm depth with sediment from the Cumbrae sample site. These tanks were maintained within a temperature controlled room (12&1 )C). Fully-oxygenated seawater (salinity 32) was constantly circulated through each tank. Although animals often constructed burrows within the first few days of being placed in the tanks, the burrows were left undisturbed for 2–3 months prior to experimentation. Burrows constructed within the tanks often made contact with the clear Perspex walls which allowed behavioural observations to be made and provided a site for water sampling (see below). Burrow morphology Casts of burrows were made at depths of 15–20m at Farland Point, Cumbrae (U. stellata) and at 20 m

depth at White Bay, Cumbrae (U. deltaura), using low-viscosity polyester resin (Trylon SP701PA, with 1% by volume liquid peroxide catalyst and 10% by volume styrene thinners), following the methodology of Atkinson and Chapman (1984). Resin casts were also made of a number of burrows that had been constructed by animals kept in mud-filled laboratory aquaria. Further information on burrow structure was obtained by observing burrows that impinged on the walls of these aquaria. Burrow water sampling Attempts to obtain water samples from burrows in the field were unsuccessful due to contamination of the sample by the burrow-wall sediment. Therefore, water samples were obtained from burrows constructed in large laboratory aquaria. Samples of burrow water (approximately 1 ml) were taken from the burrow lumen using a method similar to that described by de Vaugelas and Jaubert (1989). Perspex plates (20#20#8mm) equipped with a plastic tube (3 mm internal diameter) were fixed to the wall of the aquarium. A 2 mm hole was then drilled through the tube, plate and wall of the aquarium, and a 3-way luer valve was attached to the end of the plastic tube to prevent water escaping from the burrow. Burrow water was taken from the sampling ports using a l ml plastic syringe. The pH of a 0·5–1·0 ml sample was measured at the environmental temperature (12 )C) immediately after collection using a pH micro-electrode (Russell) connected to a pH meter (PHM83, Radiometer, Denmark). The partial pressure of oxygen (PO2) of the burrow water was measured by injecting a water sample (200 ìl) into a thermostatted (12 )C) microcell (MC100, Strathkelvin Instruments, Glasgow) containing a polarographic oxygen electrode (1302, Strathkelvin Instruments) connected to an oxygen meter (781, Strathkelvin Instruments). The concentration of ammonia in the burrow water was measured using the method of Liddicoat et al. (1975). Long-term measurements of the PO2 of the burrow water were obtained by connecting plastic tubing (internal diameter=0·5mm) from the luer valve (at the sample port) to the microcell. Burrow water was slowly siphoned through the microcell at a constant flow rate (<1 ml min "1) and the PO2 was continuously monitored using the oxygen electrode. Prior to each experiment, the oxygen electrode was calibrated against a solution having a PO2 of zero (sodium sulphite in 0·01M sodium tetraborate) and airsaturated seawater (at 12 )C). During long-term PO2 measurements, the sampling port was positioned in

Burrowing implications for mud-shrimps 157

that part of the burrow in which the occupant had previously been observed to spend the majority of its time. Irrigation activity In addition to measuring the PO2 of the burrow water, movement of water through the same burrow was measured continuously using an electromagnetic blood flow transducer (Spectramed, SP2202 flowmeter) following the method of Atkinson et al. (1987). The flow probe, 1·5 mm or 3·0 mm diameter (chosen to match the diameter of the burrow opening as closely as possible), was mounted onto a plastic filter funnel (4 cm diameter). The filter funnel was inverted, placed over one of the burrow openings and pushed gently into the sediment to form a seal. Any movement of water through the burrow passed through the flow probe, and the rate of water flow was detected by the flowmeter and recorded using a chart recorder (SE120, Belmont Instruments). Prior to use, the flow probe was calibrated by passing seawater through the probe at known flow rates. Measurement of oxygen consumption during hypoxia Measurements of rates of oxygen consumption during hypoxia were made using ‘ closed ’ respirometry. The acrylic respirometers (volume=60 ml) were supplied with seawater from a reservoir containing fully-aerated, ultraviolet-sterilized artificial seawater (Instant Ocean, at a salinity of 32). The respirometer and reservoir were placed within a water bath which was maintained at 10&0·2 )C. The chamber was fitted with an isolated magnetic stirrer bar, and the water was gently stirred to prevent stratification and local depletion of oxygen around the electrode. The PO2 of the seawater was monitored continuously using an oxygen electrode inserted directly into the respirometer and connected to an oxygen meter and a chart recorder. Calibration of the electrode was carried out before each experiment as described above. Since mud-shrimps appear stressed outside of a burrow environment (MacGinitie, 1930; 1934; Witbaard & Duineveld, 1989; pers. obs.), they were placed within perforated glass tubes, 50–100 mm long, 10–18 mm diameter (2 mm size classes) and the ends were covered with plankton gauze before each experiment. Tube diameter has previously been shown to affect the pleopod activity and performance in the burrowing shrimp, Neotrypaea (as Callianassa) affinis (Farley & Case, 1968). Therefore, the diameter of the tube into which each individual was placed was selected according to the relationship between the

burrow diameter and carapace length established previously for U. pusilla (Dworschak, 1983). The tube was then placed into the respirometer and the animal was left to acclimatize for at least 12 h, with aerated seawater pumped through the respirometer. After this time, the flow of water to the respirometer was stopped and, with the use of one-way valves, the chamber was isolated. The mud-shrimp was then allowed to reduce the PO2 of the available oxygen in the respirometer to <10 Torr. During all recordings, the respirometer was maintained in darkened conditions and the mud-shrimp was kept as free as possible from external disturbances. During a number of preliminary trials, the pH and ammonia concentration of the water were monitored (Liddicott et al., 1975). The maximum variation in pH was 0·6 pH units, and changes in the ammonia concentration were negligible. The oxygen consumption of micro-organisms within the respirometer was determined by running a control in which no animal was present in the respirometer. This value was then subtracted from that measured for each individual mud-shrimp. The apparatus was sterilized with a hypochlorite solution between experiments to reduce bacterial contamination. The fresh body weight (0·3– 12·5 g and 1·2–10 g for U. stellata and U. deltaura, respectively) was determined after removing excess water. For all experiments, only individuals in the intermoult stage were used. Scaphognathite and heart rate measurements Heart rate and scaphognathite activity were measured using an impedance technique modified from Hoggarth and Trueman (1967), and Dyer and Uglow (1977). The electrodes were positioned as described by Morris and Taylor (1984) and were connected to an impedance unit and pen recorder (Searle Bioscience). Results Burrow morphology Casts of the burrows of both Upogebia species, taken in the field and from laboratory aquaria, showed similar features. Burrow morphologies, derived from casts, are illustrated in Figure 1. The basic burrow was U-shaped with two openings to the surface. Tunnels and shafts were circular in cross-section and were often constricted where they opened to the surface. Dilations of the burrow lumen were a characteristic feature. Aquarium observations confirmed that these were used as turning chambers as described by

158 C. M. Astall et al.

(a)

(c)

(b)

(d)

(e)

(f)

(g)

(h)

(i)

F 1. Side views of the burrows of Upogebia stellata (a, b, c, e) and U. deltaura (d, f, g, h, i) derived from photographs of resin casts. Incomplete sections of burrows are drawn in dashed lines, the shape reconstructed from field notes. (f) and (g) are views of the same burrow, (g) rotated through 90) relative to (f) to show details of the side branch. (h) and (i) are another burrow displayed in a similar way. All except (d) are from field casts. Stippling represents the sediment. Scale bar=10 cm.

Dworschak (1983) for U. pusilla burrows. Most burrows had a shaft that descended from the base of the U-shaped upper section. The angle of descent was variable and, in the field, appeared to be influenced by the presence of stones and other obstructions which caused this burrow section to be deflected in both horizontal and vertical planes. The burrows of U. deltaura achieved a larger size than those of U. stellata, reflecting the size difference between the two species. The impression gained from field observations in muddy sands (with admixtures of stones and shell-gravel) around the Isle of Cumbrae (see Methods) was that, although the two species overlapped in their distribution, U. deltaura was able to occupy coarser sediments than U. stellata. Its burrow openings, normally two [Figure 1(d,f,h)], sometimes three, per burrow, were larger and therefore more conspicuous than those of U. stellata. The

small, inconspicuous openings of individual U. stellata burrows mostly had two [Figure 1(b,c,e)] or three [Figure 1(a)] openings, but some were more complex (see Discussion). It was difficult to cast the burrows in the field for three reasons. Firstly, the restricted openings made it difficult to introduce resin. Secondly, the occupant often blocked the burrow lumen with sediment, thus isolating itself and preventing resin penetration. Thirdly, in the case of large U. deltaura, some animals were able to expel resin from their burrows in a pleopod-generated current. Such complications resulted in most casts of this species being incomplete [e.g. Figure 1(f–i), cf. a complete burrow taken from an aquarium, e.g. Figure 1(d)]. Some casts of U. deltaura burrows were difficult to excavate from the sediment because lower burrow sections penetrated through the interstices of a bed of large stones. Thus, in the burrow illustrated in Figure 1(h–i), the lowest section could not be recovered. The basic structure of the burrow of U. deltaura is, however, clear (reconstructed sections are based on field notes at the time of casting). Casts of U. deltaura burrows indicated that the basic U-shaped section penetrated the sediment to depths of 15–24 cm. Openings were usually 15–20 cm apart. Burrow diameter was generally between 17 and 23 mm and was fairly constant for the main shaft and tunnel sections of a given burrow. The dilated turning chambers were usually deeper than wide (widths 22– 30 mm, depths 24–50 mm). Openings constricted to a minimum of 8 mm diameter. Maximum burrow depth was >28 cm. The U-shaped section of the burrows of U. stellata penetrated the sediment to depths of 7–20 cm. Openings were either spaced closely [3 cm, Figure 1(c)] or widely [up to 22 cm, Figure 1(a)]. Burrow diameter was generally from 12 and 16 mm and, as with U. deltaura burrows, was fairly constant for the main sections of a given burrow. Turning chambers were generally 15–18 mm wide and 20–40 mm deep. Openings were constricted to a minimum of 6 mm in the examples measured. Maximum burrow depth was 22 cm. In addition to the descending shaft which forms the deepest component of most burrows, some burrows also had short branch tunnels, sometimes originating from a turning chamber. In laboratory aquaria, prior to moulting, some animals (both species) were observed to construct a side branch from the main U-shaped burrow section. The occupant would retreat into this secondary tunnel during moulting and, once moulting had finished, the tunnel was backfilled so burying the exuvium.

Burrowing implications for mud-shrimps 159

Burrow irrigation Irrigation of the burrow was achieved by pleopod beating. Pereiopods 3–5 were held against the burrow wall and served to stabilize the mud-shrimp during irrigation activity. The metachronal beat of the pleopods created a posteriorly-directed water current through the burrow. During periods of burrow irrigation, both Upogebia species assumed a characteristic feeding position (MacGinitie, 1930; Dworschak, 1987; Nickell & Atkinson, 1995); pereiopods 1 and 2 were held upwards and outwards such that the numerous long setae on the propodi and meri overlapped

Time spent irrigating (%)

100

(a)

80 60 40 20

0 50 Pleopod activity (beats min–1)

Measurements of tunnel diameter in burrows constructed in the laboratory by both Upogebia species indicated that the diameter of the lumen closely reflected the diameter of the tail fan (telson and uropods) of the occupant. Measurements of individuals trapped within the resin casts, together with laboratory observations, indicated that the diameter of the burrow was always smaller than the carapace length of the occupant. Since a mud-shrimp has a carapace length larger than the burrow diameter, the turning chamber allows the animal to change direction within its burrow; this is done by performing a forward somersault. Observations of a number of mud-shrimps within burrows constructed in the laboratory have shown that the mud-shrimps spend a large proportion of their time tending to their burrows. This often involved cleaning, compaction and maintenance of the burrow wall which, in combination with the movement of the mud-shrimp through the burrow, probably accounts for the smooth burrow wall. An oxidized layer of conspicuous light brown sediment surrounded the burrow lumen and varied in thickness from approximately 2 to 13mm, contrasting with the surrounding grey/black reduced sediment. In burrows that had been occupied by the same individual for long periods of time (6–18 months), the gross structure changed very little once the burrow had been constructed. However, the mud-shrimps constantly reworked the sediment within the burrow. Mud collected from the burrow was either carried to the sediment surface and expelled, or retained and used as building material to maintain the walls of the burrow. This suggests that the burrows of upogebiids are relatively stable, semi-permanent structures. Two burrows in which the inhabitant had died were still recognizable after 6 months under the low energy conditions of laboratory aquaria, despite the gradual accumulation of sediment from the unmaintained burrow walls.

25

50

25

50

75

100

125

150

75 100 PO2 (Torr)

125

150

(b)

40 30 20 10

0

F 2. The effect of progressive hypoxia on (a) the amount of time (%) spent irrigating the burrow, and (b) on the rate of beating of the pleopods of Upogebia stellata. Irrigation activity was measured over three consecutive 15-min periods at various oxygen partial pressures. Values are means&SE.

and formed a basket. Suspended particles entering the burrow with the water current were trapped by the setate basket, and the trapped material was removed from the basket by the maxillipeds and transferred to the mouth where it was ingested. Irrigation response to progressive hypoxia The effect of hypoxia on the pleopod activity of three individual U. stellata (mean fresh weight=1·64& 1·08 g) was measured by exposing the mud-shrimps to progressive hypoxia in a respirometer. The frequency of pleopod beating (30·4&2·4 beats min "1) remained remarkably constant over a wide range of PO2 (Figure 2). There was no significant difference (P>0·05) in the amount (%) of time spent irrigating the burrow (40·3&7·7%), measured over three consecutive 15-min periods, over the PO2 range 150– 60 Torr. At oxygen tensions below 60 Torr, however, there was a significant (P<0·05) increase in the % time spent irrigating, reaching a maximum of

160 C. M. Astall et al. T 1. Upogebia deltaura and U. stellata: Summary of the burrow irrigation profiles for two thalassinidean mud-shrimps

Burrow No. U. deltaura 1 2 3 4a 4b 5a 5b U. stellata 1 2 3 4 5 6 7

Carapace length (mm)

Flow rate (ml min "1)

Time irrigating (%)

Pumping time (min)

Time to next event (min)

Pumping episodes

Recording duration (h)

10 20 10 25 25 11 11

4·6&2·7 13·9&11·7 34·1&11·7 25·1&13·4 13·6&8·0 18·0&10·0 19·9&7·6

9·1 4·1 32·6 8·2 2·3 9·9 5·9

3·6&2·8 7·7&4·3 29·3&40·5 5·6&7·3 6·3&6·6 4·8&4·9 2·5&2·0

35·6&35·4 176·0&92·4 60·5&109·1 61·6&53·9 265·0&358·0 36·2&52·9 57·9&46·6

87 13 30 51 13 63 29

56·8 39·7 44·9 57·0 58·8 42·9 20·6

8 4 8 8 10 11 10

4·6&1·5 4·2&3·4 7·4&6·6 27·1&22·8 18·0&10·0 31·6&17·9 15·4&16·0

16·1 7·5 9·1 38·6 6·2 6·7 5·7

3·6&7·2 2·1&1·3 1·6&0·6 5·8&1·2 2·2&3·1 1·4&1·0 2·7&1·9

18·5&19·4 26·1&40·0 16·5&34·9 8·9&4·6 33·3&53·3 21·9&15·2 46·7&47·0

177 94 21 39 89 24 37

54·8 44·1 6·3 9·5 53·3 9·4 29·6

a

and b are recordings made from the same burrow but on two separate occasions. The size of the shrimp was estimated using carapace length (posterior edge of the orbit indent to the mid-posterior edge of the cephalothorax).

93·6&6·5% at 35 Torr (Figure 2). Below a PO2 of 35 Torr, the amount of time spent irrigating decreased significantly (P<0·05) to 55&25% at PO2 =6 Torr but, even at a PO2 of approximately 0 Torr, pleopod irrigation did not cease for several hours. Burrow irrigation profiles Continuous recordings of burrow irrigation activity were made for U. stellata (n=7 individuals) and U. deltaura (n=5 individuals) for periods of time ranging from 6·5 to 57 h, producing a total of 519 h of recordings. In all individuals, episodes of burrow irrigation occurred at regular intervals, although the duration of each bout of irrigation, the flow rate, the amount of time spent irrigating the burrow and the time between consecutive irrigation events varied between individuals in both species (Table 1). A typical recording of water movement through the burrow of U. stellata is shown in Figure 3. In both U. stellata and U. deltaura, periods of irrigation were interspersed with periods during which pleopod beating and burrow irrigation ceased. The duration of each irrigation episode varied from 1·4 min to 29·3 min (mean=8·5&3·5 min and 2·8&0·5 min for U. deltaura and U. stellata, respectively). The amount of time spent irrigating the burrow during the recordings ranged from 2·3 to 38·6% with mean values of 10·3&3·8% and 12·8&4·5% for U. deltaura and U. stellata, respectively.

A constant flow rate was often maintained throughout the irrigatory event [Figure 3(a)], although on other occasions, considerable variations in flow velocity during each irrigation episode were observed [Figure 3(b)] with flow rates varying between 4·2 and 34·1 ml min "1. Burrow irrigation was not restricted to one particular direction [Figure 3(b)] since, by utilizing enlargements of the burrow lumen, the occupant was able to change direction within the burrow and, as a result, movement of water through the burrow could occur in either direction. Activity of the occupant and changes in direction often resulted in the movement of water through the burrow [Figure 3(c)]. Long-term irrigation rates (ml h "1) of U. deltaura and U. stellata were estimated by multiplying the mean flow rate recorded during the irrigation episodes by the average time (min) that the mud-shrimps spent irrigating the burrow during a 1-h period (determined from the % time spent irrigating). The mean (&SD) long-term pumping rates were 149·5&55·7 ml h "1 and 139·7&35·5 ml h "1 for U. deltaura and U. stellata, respectively. Patterns of water flow through the burrow were, at times, quite complex depending upon the behaviour of the mud-shrimp. Figure 3(c) shows the complex irrigation recordings from the burrow of U. deltaura where water movement is caused by burrow maintenance activity, animal movements and active irrigation (Ir). During burrow maintenance, loose sediment

Flow rate (ml min–1)

Burrowing implications for mud-shrimps 161

20

PO2 160 (100%) pH 8.1

10 0

PO2 110–140 (80%) pH 7.8–8.0

0.5 Time (h)

0.0

80 Flow rate (ml min–1)

(a)

1.0

PO2 80–110 (60%) pH 7.7–7.8

(b)

60

0

40 cm

20 0

5

20

PO2 50–60 (30%) pH 7.6–7.7

40 60 0.5 Time (h)

Flow rate (ml min–1)

0.0

80

E

(c)

1.0

E Ir

60 40

M

20

F 4. Diagrammatic representation of a burrow of Upogebia stellata showing the range of PO2 and pH values recorded in different parts of the burrow and in the water above the sediment surface.

0

M 0.0

0.5 Time (h)

1.0

F 3. Parts of three records of the flow rate of water through burrows of Upogebia deltaura constructed in the aquarium showing (a) the typical pattern of intermittent irrigation behaviour, (b) the change in the direction of water flow through the burrow resulting from the mud-shrimp turning round inside the burrow, and (c) the complexity of some irrigation recordings resulting both from pleopod beating and from movements of the mud-shrimp within the burrow. Ir, periods of irrigation; M, movements of the animal; E, expulsion of sediment by pleopod activity creating water flow through the burrow.

within the burrow was either expelled or retained and used to repair the burrow wall. Expulsion of sediment from the burrow, which was achieved by rapid irrigation (E), and by movements of the shrimp through the burrow (M) were also shown on the flow recordings [Figure 3(c)]. Spatial variation of pH, PO2 and ammonia concentration within the burrow The spatial variation of the pH and oxygen tension within the burrows of U. deltaura and U. stellata

was examined by measuring the PO2 and pH of the burrow water at various positions throughout a number of burrows constructed in aquaria. There was considerable variation in the PO2 of the burrow water within a burrow; higher PO2 values (110–140 Torr) were recorded near the burrow openings whereas, in the deeper parts of a burrow, the PO2 was much lower (50–60 Torr) (Figure 4). The pH profile of the burrow water was similar to that of PO2, with a lower pH recorded in the deeper parts of the burrow probably reflecting the hypercapnic conditions in these regions. Ammonia concentrations in the water in the burrows of U. stellata and U. deltaura were very variable ranging from 1·2 to 60·3 ìg at NH4-N l "1, and were much higher than the values recorded for the overlying seawater (0·1–0·4 ìg at NH4-N l "1). Temporal variation of PO2 within the burrow Long-term recordings of the PO2 of the burrow water were made in conjunction with flow rate measurements for a number of U. stellata and U. deltaura burrows. Considerable temporal variation was recorded in the PO2 of the water within the burrows examined (n=20). During periods when the shrimp was not irrigating the burrow, the PO2 of the burrow

1.5

80 40

–1

.

0 0

1

2

3

–1

120

30

80

0

–1

40

PO2 (Torr)

Flow rate (ml min )

–1

Frequency (pauses min ) fh and fsc (beats min )

Time (h)

(b)

0

1

2

3

0

Time (h)

F 5. Parts of two simultaneous recordings of the water flow rate and the PO2 of the water in a burrow of Upogebia deltaura constructed in the laboratory. (a) Trace showing a single irrigatory event and its effect on the PO2 of the burrow water, and (b) trace showing two irrigatory events.

0.9 0.6 0.3 0.0 125

Duration (min)

(b)

100 75 50 25 0 4.0

(c)

3.0 2.0 1.0 0.0 0.4

water declined steadily. Movement of the shrimp through the burrow and active irrigation activity resulted in rapid increases in the PO2 of the burrow water as fresh, oxygenated seawater was drawn into the burrow. A typical recording of irrigation activity and of the PO2 of the burrow water is shown in Figure 5(a). The PO2 of the burrow water steadily decreased from 50 to 25 Torr over the first 1·5 h of recording. During the 8 min of irrigation activity, the PO2 of the burrow water increased to a maximum of 126 Torr. This was followed by a rapid decrease in the PO2 of the burrow water in the 15 min following cessation of irrigatory activity, probably due to the re-equilibration of the oxygen gradient through the burrow. After this time, the PO2 continued to decline, but at a slower rate reflecting the rates of oxygen consumption of the mud-shrimp and of the meiofauna and micro-organisms associated with the burrow. Another example of the effect of irrigatory activity on the PO2 of the burrow water is shown in Figure 5(b). This is a recording of the same individual but shows two irrigatory events, resulting in a two-step increase in the PO2 of the burrow water. Both periods of irrigation activity were of short duration and, because complete exchange of the burrow water did not occur, the PO2 of the burrow water remained low. The degree of exchange of burrow water and hence the large temporal changes in PO2, is dependent,

(a)

1.2

–1

0

h )

120

(a)

MO2 (µ mol g

30

PO2 (Torr)

–1

Flow rate (ml min )

162 C. M. Astall et al.

(d)

0.3 0.2 0.1 0

40

80 PO2 (Torr)

120

160

F 6. The effect of declining oxygen tension PO2 on (a) oxygen consumption (M ~ O2), (b) heart rate (fh) ( ) and scaphognathite rate (fsc) ( ), (c) the frequency of ventilatory pausing, and (d) the mean duration of the ventilatory pauses for Upogebia deltaura (mean fresh weight=8·4&0·4 g, n=5). All recordings were carried out at 10 )C. Values are means&SE.

therefore, upon the amount of time the occupant spends irrigating the burrow and the flow rate during the period of irrigation activity. Oxygen consumption during hypoxia Both U. stellata and U. deltaura were able to maintain an approximately constant rate of oxygen consumption over a wide range of oxygen tension, until a critical oxygen tension (Pc) was reached. Below the Pc, the rate of oxygen consumption decreased sharply (Figure 6). The Pc range for U. stellata was

Burrowing implications for mud-shrimps 163

30–40 Torr (n=10, fresh weight range 1·60–2·18 g) and for U. deltaura was 35–50 Torr (n=10, fresh weight range 1·70–9·61g).

continued at a lower rate (10–20 beats min "1) for up to 10 h in all of the individuals studied. Dicussion

Ventilatory and cardiac responses during hypoxia The effects of progressive hypoxia on the mean rate of oxygen consumption (M ~ O2) and on the heart rate and scaphognathite rate of five individual U. deltaura (fresh body weight; 8·4&0·4 g) are shown in Figure 6. Under normoxic conditions, ventilation was frequently interrupted by periods of scaphognathite pausing (apnoea) which were usually coupled with bradycardia or cardiac arrest. Although coupling between the scaphognathite and heart rate was regularly observed, periods of cardiac arrest lasting longer than 7 s were infrequent. The duration of respiratory pausing was variable and lasted from several seconds to 1–2min. During long periods of apnoea (>60 s), therefore, the heart may arrest a number of times. No periods of reversed beating of the scaphognathites (reversals) were recorded using the impedance technique in any of the individuals examined. Unfortunately, the small size of the mud-shrimps precluded measurements of branchial chamber pressure which is a more effective method for recording scaphognathite reversals (e.g. Hughes et al., 1969; Taylor, 1976). The scaphognathite rate was maintained over the PO2 range 155–80 Torr at a mean frequency of 52·2&5·6 beats min "1 [Figure 6(b)]. Below 80 Torr, the scaphognathite rate increased progressively until a maximum rate of 103·7&16·8 beats min "1 was recorded at a PO2 of 45 Torr. Below this PO2, however, there was a pronounced reduction in scaphognathite rate. During anoxia, scaphognathite activity did not cease entirely, but was maintained at a rate of 20–30 beats min "1. This rate was maintained for up to 10 h of exposure to anoxia. The increase in scaphognathite activity below 80 Torr [Figure 6(b)] resulted not only from an increase in the frequency of beating, but also from a reduction in the number and duration of the ventilatory pauses [Figure 6(c,d)]. This resulted in a more continuous pattern of scaphognathite beat which was maintained down to a PO2 of approximately 40 Torr. The scaphognathite beat frequency declined below 40 Torr and the rate and duration of pausing increased, resulting in an overall decrease in scaphognathite activity. The heart rate of U. deltaura showed very little variation during exposure to hypoxia, declining only slightly at low oxygen tensions [Figure 6(b)]. Under anoxic conditions, cardiac activity

Burrow morphology The relatively simple burrow structure of U. stellata and U. deltaura described in the present study is fairly typical of the upogebiids as a group (Dworschak, 1983). The burrows of both species were similar in structure, consisting of a U-shaped section and a descending blind-ending shaft. There is little published information on the burrows of U. stellata and U. deltaura. Dworschak (1983) provided an illustration of three burrows of U. deltaura constructed in an aquarium. Nickell and Atkinson (1995) present a brief description of several burrows of U. stellata from field resin casts and laboratory observations. Interestingly, one of the burrows described by Nickell and Atkinson (1995) bifurcated a number of times giving rise to a structure made up of multiple Us with a shaft originating from the deepest U, but having a total of three openings, with a further four tunnels that terminated just below the sediment surface and may once have opened to the surface. The burrows of U. stellata and U. deltaura were typically enlarged at a number of places, especially at points where the burrow bifurcated. These enlargements are turning chambers and appear to be a common feature of upogebiid burrows (MacGinitie, 1930; Ohshima, 1967; Ott et al., 1976; Frey & Howard, 1975; Dworschak, 1983; de Vaugelas, 1990; Nickell, 1992). Another feature common to upogebiid burrows is the constriction of the upper 5–10 mm of the main shaft (Smith, 1967; Frey & Howard, 1975; Dworschak, 1983). The function of the constriction is unclear, but it may act as an anti-predator device (preventing the shrimp from being sucked from the burrow) or it may affect the velocity of water flow and mixing patterns within the burrow (Vogel & Bretz, 1971). Upogebiid burrows are formed by excavation and compression of sediment (Dworschak, 1983; Nickell, 1992). For U. pusilla, Dworschak (1983) reported that less than half of the sediment corresponding to the burrow volume was transported to the surface and ejected during burrow construction. Once constructed, a burrow appears to be maintained for long periods with little change (MacGinitie, 1930; Dworschak, 1983; Swinbanks & Murray, 1987; Nickell, 1992). The oxidized burrow-wall sediment is a characteristic feature of the burrows of thalassinideans and other burrowing organisms (e.g. Ott et al., 1976; Dworschak, 1983).

164 C. M. Astall et al.

Burrow irrigation Both U. stellata and U. deltaura showed intermittent patterns of burrow irrigation. The amount of time spent irrigating the burrow was, however, greater in these and other suspension-feeding species (see Dworschak, 1981) than has been recorded for some primarily deposit-feeding species such as Callianassa subterranea and Jaxea nocturna (Nickell, 1992; Forster & Graf, 1995), Calocaris macandreae (Anderson et al., 1991) and Neotrypaea (as Callianassa) californiensis (Torres et al., 1977). The burrow irrigation rates recorded for different species of mud-shrimp vary considerably. For example, the mean (&SD) irrigation rates recorded during the present study were 149·5&55·7 ml h "1 and 139·7&35·5 ml h "1 for U. deltaura and U. stellata, respectively. Nickell (1992), using a similar procedure, recorded rates of 50·3&33·6 ml h "1 and 96·3&37·1 ml h "1 for C. subterranea and J. nocturna, respectively. Dworschak (1981) found the irrigation rate of U. pusilla, determined using an overflow apparatus and from respiration measurements, ranged from 5 to 900 ml h "1. The irrigation rates of the deposit-feeder Callianassa japonica and the suspension-feeder U. major, estimated from the oxygen balance of the burrow, were 29–63 ml h "1 and 14–33 ml h "1, respectively (Koike & Mukai, 1983). Using an alternative approach based on solute flux rates, Aller et al. (1983) calculated the in situ irrigation rate for the intertidal U. affinis as 174 ml h "1. Some of the variation between species can be attributed to the different experimental procedures used. Nevertheless, the available data indicate that the suspensionfeeding upogebiids have a higher irrigation rate, in accordance with their feeding style, than those thalassinideans that are primarily deposit-feeders. Although both U. stellata and U. deltaura exhibited some degree of regular periodic irrigation activity, it was not clear whether the stimulus to irrigate was respiratory or nutritional. Since upogebiids are primarily suspension-feeders, periods of irrigation activity probably fulfil both the individual’s feeding and respiratory requirements simultaneously. During exposure to hypoxia, there was a pronounced increase in burrow irrigation activity in U. stellata and U. deltaura, especially at oxygen tensions below 50 Torr. This was achieved mainly as a result of an increase in the amount of time spent irrigating the burrow since the actual rate of pleopod beating did not change significantly during hypoxia. Similarly, Farley and Case (1968) observed an increase in burrow irrigation at low oxygen tensions in N. (as Callianassa) affinis, with pleopod beat frequencies as high as 120 beats min "1

being recorded under hypoxia. The PO2 below which irrigation activity rapidly increased was less than 28 Torr. This value is lower than that recorded for U. stellata and U. deltaura, and probably reflects the more hypoxic burrow environment and the greater tolerance of N. affinis to hypoxia. A similar response was recorded for Lepidophthalmus louisianensis (as Callianassa jamaicense), where pleopod beat rate rapidly increased below a PO2 of 30 Torr (Felder, 1979). Paterson and Thorne (1993) reported, however, that Trypaea (as Callianassa) australiensis increased its rate of pleopod beating only when the PO2 of the water was reduced to below 1 kPa (approximately 7·5 Torr). Observations on U. stellata and U. deltaura held under anoxia for >12 h have shown that pleopod activity still occurred, although at very low frequencies (10–15 beats min "1), for up to 3–4 h. When exposed to longer periods of anoxia, two of the four individuals studied were observed to eventually leave their burrows, probably in an attempt to move away from the localized anoxic conditions. Although many species of burrowing decapods do actively irrigate their burrows, it must be recognized that burrows situated in an environment where water flows over the sediment surface may be subject to a considerable degree of passive irrigation (Vogel & Bretz, 1971; Allanson et al., 1992). In a laboratory simulation using artificial burrows to simulate those of U. africana, Allanson et al. (1992) found that currentinduced flow through the burrow could considerably supplement active irrigation if one opening were higher than the other, resulting in a significant saving of energy by the occupant. Further experimentation with more accurate simulations (openings of differing diameter rather than differing height above the general sediment surface) or using natural burrows are needed to fully evaluate the importance of passive ventilation in upogebiids. Spatial and temporal variation of PO2 within the burrow Within the burrows of U. stellata and U. deltaura there is a gradient of PO2 with depth, with higher PO2 values being recorded near the burrow openings. A similar gradient of oxygen tension with depth has been recorded in burrows of C. macandreae with the deeper levels being severely hypoxic (<20 Torr) (Anderson et al., 1991). In the more complicated burrow system of C. subterranea, however, the PO2 of the burrow water varies throughout the burrow (Witbaard & Duineveld, 1989; Nickell, 1992). Oxygen concentrations of 0·6–88% of air-saturation (PO2 approximately 1–136 Torr) have been recorded from the burrow waters of C. subterranea (Witbaard & Duineveld,

Burrowing implications for mud-shrimps 165

1989; Nickell, 1992; Astall, unpubl. obs.) and burrow water oxygen levels of between 6·4 and 89·7% air saturation (PO2 approximately 10–139 Torr) have been found for the deposit-feeder, J. nocturna (Nickell, 1992). Clearly, hypoxic conditions are a regular occurrence in the burrows of these mudshrimp species. The basic U-shaped burrow of upogebiids is ideal for the efficient unidirectional through-flow of water (de Vaugelas, 1990). The higher oxygen tensions of the water within upogebiid burrows are primarily a result of the periodic irrigatory activity of the mudshrimps, which not only provides a feeding current but also serves to replenish the oxygen-depleted burrow waters. In contrast, the burrows of the deposit- feeders can be quite complex structures (de Vaugelas, 1990) with some burrows having numerous shafts and a matrix of interconnecting galleries. As a result of their continual digging through the largely anoxic sediment as part of their feeding strategy, the burrow waters of the deposit-feeders often become severely hypoxic and have high concentrations of sulphide (Johns et al., 1996). Due to the complex nature of the burrow, only those tunnels directly connecting the burrow openings may be regularly irrigated and, as a result of poor circulation, extreme hypoxia or even anoxia may exist for long periods in some sections of the burrow. In addition, the burrow waters of intertidal species are likely to become severely hypoxic during tidal emersion since the burrows cannot be irrigated at this time, and it is in these burrows that some of the most hypoxic conditions have been recorded (Thompson & Pritchard, 1969; Torres et al., 1977; Felder, 1979; Koike & Mukai, 1983). Temporal variations in the PO2 of the burrow water such as those recorded during the present study are also a feature of thalassinidean burrows (Koike & Mukai, 1983; Nickell, 1992). These temporal fluxes in oxygen tension are dependent upon the respiratory and irrigatory behaviour of the occupant, the oxygen demands of the burrow wall biota, the decomposition of organic matter and the diffusion of oxygen through the burrow. The burrows of both U. stellata and U. deltaura were enriched in terms of ammonia, with burrow-water values being 10-fold those of overlying waters. Accumulation of ammonia in burrow waters has also been reported for C. japonica, C. subterranea, U. major, U. affinis and J. nocturna (Aller et al., 1983; Koike & Mukai, 1983; Witbaard & Duineveld, 1989; Nickell, 1992). Ammonia enrichment in the burrow is the result of the metabolism of both the mudshrimps and the other burrow-associated fauna and micro-organisms.

Oxygen consumption during hypoxia Compared with other decapod Crustacea, thalassinideans appear to have lower rates of oxygen consumption than those of many non-burrowing species. These low rates have been interpreted both as an adaptation to the hypoxic conditions often prevalent in the burrow habitat (Thompson & Pritchard, 1969; Felder, 1979; Hill, 1981) and to the reduced metabolic demands associated with low levels of activity (Anderson et al., 1991). Under conditions of declining oxygen tension, U. stellata and U. deltaura, like many other burrowing decapods (Atkinson & Taylor, 1988; Taylor & Atkinson, 1990), showed a high degree of respiratory independence. The Pc values recorded (30–40 Torr in U. stellata and 35–50 Torr in U. deltaura) are not as low, however, as those recorded for J. nocturna and C. subterranea (10–20 Torr) using identical procedures (Astall, 1993). These Pc values were similar to those reported for other thalassinideans; 10–25 Torr in N. (as Callianassa) californiensis (Thompson & Pritchard, 1969; Miller et al., 1977; Torres et al., 1977), 10–25 Torr in L. louisianensis (as Callianassa jamaicense) (Felder, 1979), 45–50 Torr in U. pugettensis (Thompson & Pritchard, 1969), 37 Torr in T. australiensis (Paterson & Thorne, 1995) and 10 Torr in C. macandreae (Anderson et al., 1991). It is interesting to note that deposit feeding species tend to have lower Pc values than those species that are primarily suspension-feeders, and this is correlated with the more severe hypoxia to which they may be exposed (see above). Respiratory responses to hypoxia The respiratory responses to hypoxia shown by U. stellata and U. deltaura were similar to those shown by other thalassinideans such as C. macandreae (Anderson et al., 1991) and by T. australiensis (Paterson & Thorne, 1995), namely, an increase in ventilation rate and the maintenance of the heart rate over a wide range of oxygen tension. Such responses have now been established as being typical of most other decapods (McMahon & Wilkens, 1983). In U. stellata and U. deltaura, the increase in ventilation rate was brought about not only by an increase in the rate of beating of the scaphognathites but also by a progressive reduction in the number and duration of ventilatory pauses (see also Morris & Taylor, 1985; Anderson, 1989). During hypoxia, oxygen uptake at the gills can also be facilitated by an increase in gill perfusion. It was not possible during the present study to measure cardiac output in U. deltaura. Although heart rate remained constant over a wide range of

166 C. M. Astall et al.

oxygen tension, cardiac output may have increased as a result of an increase in stoke volume (Johansen et al., 1970; McMahon & Wilkens, 1975; Airriess & McMahon, 1994). Oxygen uptake during hypoxia can also be facilitated by the possession of a respiratory pigment having a high oxygen affinity. The haemocyanin of U. stellata and U. deltaura, like that of a number of other thalassinideans (Miller & Van Holde, 1974; Miller et al., 1977; Anderson et al., 1994), has been shown to have a higher oxygen affinity than that of many non-burrowing species. The P50 values (the PO2 at which the haemocyanin is 50% saturated) at the in vivo pH (7·8) were 6·3 and 7·5 Torr for U. deltaura and U. stellata, respectively (Astall, 1993). This high oxygen affinity may be an important adaptation to the prolonged periods of hypoxia to which these species are exposed. More information on the oxygen transporting properties of the haemolymph of these and other thalassinidean species will be published elsewhere. In conclusion, for U. stellata and U. deltaura, the burrow-dwelling lifestyle affords protection from predators and provides a protected feeding environment but, at the same time, exposes the mud-shrimps to physicochemical conditions that may differ considerably from those in the overlying water. The present paper provides data on burrow irrigation and the respiratory responses of these species to hypoxia. Further work is needed on the role of irrigation in chemical flux (including nutrients and sulphide) across the sediment–water interface, on morphometric relationships between mud-shrimps and their burrows, and on the influence of upogebiid burrow form on active and passive through-flow of water. Acknowledgements This study was carried out whilst CMA was in receipt of an NERC Research Studentship. The authors are also grateful to Dr A. Rowden for supplying most specimens of U. deltaura. The authors also acknowledge the assistance of the dive team and crew of RV Aplysia at the University Marine Biological Station, Millport. References Abu-Hilal, A., Badran, M. & Vaugelas, J. de 1988 Distribution of trace elements in Callichirus laurae burrows and nearby sediments in the Gulf of Aqaba, Jordan (Red Sea). Marine Environmental Research 25, 233–248. Airriess, C. N. & McMahon, B. R. 1994 Cardiovascular adaptations enhance tolerance of environmental hypoxia in the crab Cancer magister. Journal of Experimental Biology 190, 23–41. Allanson, B. R., Skinner, D. & Imberger, J. 1992 Flow in prawn burrows. Estuarine, Coastal and Shelf Science 35, 253–266.

Aller, R. C. & Yingst, J. Y. 1985 Effects of the marine deposit feeders Heteromastus filiformis (Polycheata), Macoma baltica (Bivalvia) and Tellina texana (Bivalvia) on averaged sedimentary solute transport, reaction rates and microbial distributions. Journal of Marine Research 43, 615–645. Aller, R. C., Yingst, J. Y. & Ullman, W. J. 1983 Comparative biogeochemistry of water in intertidal Onuphis (polychaeta) and Upogebia (crustacea) burrows: temporal patterns and causes. Journal of Marine Research 41, 571–604. Anderson, S. J. 1989 Physiological Ecology of the Mud-burrowing Shrimp Calocaris macandreae Bell. Ph.D. Thesis, University of Glasgow, Scotland, U.K. Anderson, S. J., Atkinson, R. J. A. & Taylor, A. C. 1991 Behavioural and respiratory adaptations of the mud-burrowing shrimp Calocaris macandreae Bell (Thalassinidea: Crustacea) to the burrow environment. Ophelia 34, 143–156. Anderson, S. J., Taylor, A. C. & Atkinson, R. J. A. 1994 Anaerobic metabolism during anoxia in the burrowing shrimp Calocaris macandreae Bell (Crustacea: Thalassinidea). Comparative Biochemistry and Physiology 108, 515–522. Astall, C. M. 1993 Comparative Physiological Ecology of Some Mudburrowing Shrimps (Crustacea, Decapoda, Thalassinidea). Ph.D. Thesis, University of Glasgow, U.K. Atkinson, R. J. A. & Chapman, C. J. 1984 Resin casting: a technique for investigating burrows in sublittoral sediments. Progress in Underwater Science 9, 15–25. Atkinson, R. J. A., Pelster, B., Bridges, C. R., Taylor, A. C. & Morris, S. 1987 Behavioural and physiological adaptations to a burrowing lifestyle in the snake blenny Lumpenus lampretaeformis and the red band fish Cepola rubescens. Journal of Fish Biology, 31, 639–659. Atkinson, R. J. A. & Taylor, A. C. 1988 Physiological ecology of burrowing decapods. Symposium of the Zoological Society of London 59, 201–226. Bromley, R. G. 1990 Trace fossils: biology and taphonomy. Special Topics in Palaeontology, Vol. 3, Unwin Hyman, London, pp. 1–280. Dworschak, P. C. 1981 The pumping rates of the burrowing shrimp Upogebia pusilla (Petagna) (Decapoda: Thalassinidea). Journal of Experimental Marine Biology and Ecology, 52, 25–35. Dworschak, P. C. 1983 The biology of Upogebia pusilla (Petagna) (Decapoda: Thalassinidea) I. The burrows. Marine Ecology, 4, 19–43. Dworschak, P. C. 1987 Burrows of Mediterranean Decapoda. Investigacion Pesquera, 51, 264 Dworschak, P. C. & Ott, J. A. 1993 Decapod burrow in mangrovechanel and back-reef environments at the Atlantic Barrier Reef, Belize. Ichnos 2, 277–290. Dyer, M. F. & Uglow, R. F. 1977 On a technique for monitoring heart and scaphognathite activity in Natantia. Journal of Experimental Marine Biology and Ecology 27, 117–124. Farley, R. D. & Case, J. F. 1968 Perception of external oxygen by the burrowing shrimp, Callianassa californiensis Dana and C. affinis Dana. Biological Bulletin of the Marine Biological Laboratory, Woods Hole, 134, 261–265. Felder, D. L. 1979 Respiratory adaptations of the estuarine mud shrimp, Callianassa jamaicense (Schmitt, 1935) (Crustacea, Decapoda, Thalassinidea). Biological Bulletin of the Marine Biological Laboratory, Woods Hole, 157, 125–138. Forster, S. & Graf, G. 1995 Impact of irrigation on oxygen flux into the sediment: intermittent pumping by Callianassa subterranea and, piston-pumping, by Lanice conchilega. Marine Biology Berlin 123, 335–346. Frey, R. W. & Howard, J. D. 1975 Endobenthic adaptations of juvenile thalassinidean shrimp. Bulletin of the Geology Society of Denmark, 24, 283–297. Griffis, R. B. & Suchanek, T. H. 1991 A model of burrow architecture and trophic modes in thalassinidean shrimp (Decapoda: Thalassinidea). Marine Ecology Progress Series, 79, 171–183.

Burrowing implications for mud-shrimps 167 Gust, G. & Harrison, J.T. 1981 Biological pumps at the sedimentwater interface: mechanistic evaluation of the alpheid shrimp Alpheus mackayi and its irrigation pattern. Marine Biology, Berlin 64, 71–78. Hill, B. J. 1981 Respiratory adaptations of three species Upogebia (Thalassinidea: Crustacea) with special reference to low tide periods. Biological Bulletin of the Marine Biological Laboratory, Woods Hole, 160, 272–279. Hoggarth, K. R. & Trueman, E. R., 1967 Techniques for recording the activity of aquatic animals. Nature 213, 1050–1051. Hughes, G. M., Knights, B. & Scammell, C. A. 1969 The distribution of PO2 and hydrostatic pressure changes within the branchial chambers in relation to gill ventilation of the shore crab Carcinus maenas L. Journal of Experimental Biology 51, 203–220. Hughes, D. J. & Atkinson, R. J. A. 1997 Towed video survey of megafaunal bioturbation in the north-eastern Irish Sea. Journal of the Marine Biological Association, U.K., (submitted). Johansen, K., Lenfant, C. & Mecklenburg, T. A. 1970 Respiration in the crab, Cancer magister. Zeitschrift fu¨r Vergleichende Physiologie 70, 1–19. Johns, A. R., Taylor, A. C., Atkinson, R. J. A. & Grieshaber, M. K. 1996 Sulphide metabolism in thalassinidean Crustacea. Journal of the Marine Biological Association of the U.K. (in press). Koike, I. & Mukai, H. 1983 Oxygen and inorganic nitrogen contents and fluxes in burrows of the shrimps Callianassa japonica and Upogebia major. Marine Ecology Progress Series, 12, 185–190. Liddicoat, M. I., Tibbitts, S. & Butler, E. I. 1975 The determination of ammonia in sea water. Limnology and Oceanography 20, 131–132. MacGinitie, G. E. 1930 The natural history of the mud shrimp Upogebia pugettensis (Dana). Annals and Magazine of Natural History, 6, 36–44. MacGinitie, G. E. 1934 The natural history of Callianassa californiensis Dana. American Midland Naturalist, 15, 166–177. McMahon, B. R. & Wilkens, J. L. 1975 Respiratory and circulatory responses to hypoxia in the lobster Homarus americanus. Journal of Experimental Biology 62, 637–655. McMahon, B. R. & Wilkens, J. L. 1983 Ventilation, perfusion and oxygen uptake. In The Biology of Crustacea, vol. 5. (Mantel, L.H., ed.). New York: Academic Press, pp. 290–372. McMurty, G. M., Schneider, R. C., Colin, P. L., Buddemeier, R. W. & Suchanek, T. H. 1985 Redistribution of fallout radionuclides in Enewetak Atoll lagoon sediments by callianassid bioturbation. Nature 313, 674–677. Miller, K. I., Eldred, N. W., Arisaka, F., Van & Holde, K. E. 1977 Structure and function of hemocyanin from thalassinid shrimp. Journal of Comparative Physiology 115B, 171–184. Miller, K. I. & Van Holde, K. E. 1974 Oxygen binding by Callianassa californiensis hemocyanin. Biochemistry 13, 1668– 1674. Morris, S. & Taylor, A. C. 1984 Heart rate response of the intertidal prawn Palaemon elegans to simulated and in situ environmental changes. Marine Ecology Progress Series, 20, 127–136. Morris, S. & Taylor, A. C. 1985 The respiratory response of the intertidal prawn Palaemon elegans (Rathke) to hypoxia and hyperoxia. Comparative Biochemistry and Physiology 81A, 633–639. Nickell, L. A. 1992 Deep Bioturbation in Organically Enriched Marine Sediments. Ph.D. Thesis, University of London, U.K. Nickell, L. A. & Atkinson, R. J. A. 1995 Functional morphology of burrows and trophic modes of the shrimps Callianassa subterranea (Montagu), Jaxea nocturna Nardo and Upogebia stellata (Montagu) (Decapoda: Thalassinidea) with a discussion on a new approach to the classification of thalassinidean burrow morphology. Marine Ecology Progress Series 128, 181–197. Nickell, L. A., Hughes, D. J. & Atkinson. R. J. A. 1995 Megafaunal bioturbation in organically enriched Scottish sea lochs. In: Biology and Ecology of Shallow Coastal Waters. (Eletheriou, A., Ansell, A. D. & Smith, C. R. eds), Proceedings of the 28th European Marine Biology Symposium, Olsen & Olsen, Fredensborg, Denmark, pp. 315–322.

Ohshima, K. 1967 Burrows of Japanese Thalassinidea. Chiku Kagatu, 21, 11–18. Ott, A., Fuchs, B., Fuchs, R. & Malasek, A. 1976 Observations on the biology of Callianassa stebbingi Borradaille and Upogebia litoralis Risso. Senckenbergiana Maritima, 8, 61–79. Paterson, B. D. & Thorne, M. J. 1993 The effect of oxygen on the swimmeret rate of Callianassa australiensis and C. arenosa (Crustacea, Decapoda, Thalassinidea). Marine Behaviour and Physiology, 24, 15–24. Paterson, B. D. & Thorne, M. J. 1995 Measurements of oxygen uptake, heart and gill bailer rates of the callianassid burrowing shrimp Trypaea australiensis Dana and its responses to low oxygen tensions. Journal of Experimental Marine Biology and Ecology, 194, 39–52. Posey, M. H. 1986 Changes in benthic community associated with dense beds of a burrowing deposit feeder Callianassa californiensis. Marine Ecology Progress Series, 31, 15–22. Posey, M. H., Dumbald, B. R. & Armstrong, D. A. 1991 Effects of a burrowing mud-shrimp, Upogebia pugettensis (Dana), on abundance of macro-infauna. Journal of Experimental Marine Biology and Ecology, 148, 283–284. Rowden, A. A. & Jones, M. B. 1995 The burrow structure of the mud shrimp Callianassa subterranea (Decapoda: Thalassinidea) from the North Sea. Journal of Natural History 29, 1155–1165. Smith, K. L. 1967 Callianassid Crustacean Burrows as Possible Palaeoenvironmental Indicators. M.Sc., University of Delaware, U.S.A. 51 pp. Suchanek, T. H. 1983 Control of seagrass community and sediment distribution by Callianassa (Crustacea: Thalassinidea) bioturbation. Journal of Marine Research 41, 281–298. Suchanek, T. H. 1985 Thalassinid shrimp burrows: ecological significance of species-specific architecture. In: Proceedings of the Vth International Coral Reef Congress, Papeete, Tahiti. Anteene Museum-Ephe, Moorea, French Polynesia, 5, 205–210. Swinbanks, D. D. & Murray, J. W. 1981 Biosedimentological zonation of Boundary Bay tidal flats, Fraser river delta, British Columbia. Sedimentology, 28, 210–237. Taylor, A. C., 1976 The respiratory responses of Carcinus maenas to declining oxygen tension. Journal experimental Biology, 65, 309– 322. Taylor, A. C. & Atkinson, R. J. A. 1990 Respiratory adaptations of decapod Crustacea and fish to a burrowing mode of life. In: Physiological strategies for gas exchange and metabolism. (Woakes, A. J., Grieshaber, M. K. & Bridges, C. R. eds), Society for Experimental Biology Seminar Series No. 41. Cambridge University Press, Cambridge, pp. 211–234. Thompson, R. K. & Pritchard, A. W. 1969 Respiratory adaptations of two burrowing crustaceans, Callianassa californiensis and Upogebia pugettensis (Decapoda, Thalassinidea). Biological Bulletin of the Marine Biological Laboratory, Woods Hole, 136, 274–287. Torres, J. J., Gluck, D. L. &, Childress J. J. 1977 Activity and physiological significance of the pleopods in the respiration of Callianassa californiensis (Dana) (Crustacea: Thalassinidea). Biological Bulletin of the Marine Biological Laboratory, Woods Hole, 152, 134–146. Vaugelas, J. de. 1984 Preliminary observations on two types of Callianassid (Crustacea: Thalassinidea) burrows, Gulf of Aquaba, Red Sea. Proceedings of the Symposium on the Coral Reef Environment in the Red Sea, Jeddah, January 1984, 520–539. Vaugelas, J. de. 1990 Ecologie des Callianasses (Crustacea, Decapoda, Thalassinidea) en Milieu Recifal Indo-Pacifique. Conséquences du Remaniement Sédimentaire sur la Distribution des Matieres Humiques, des Métaux Traces et des Radionucle´ides. Doctorat dHabilitation à Diriger des Recherches, Université de Nice, France, 226 pp. Vaugelas, J. de & Jaubert, J. 1989 An experimental closed circuit limarium for investigating the behaviour and ecophysiology of burrowing crustaceans. Bulletin de l’Institut Océanographique de Monaco 5, 71–77.

168 C. M. Astall et al. Vogel, S. & Bretz, W. L. 1971 Interfacial organisms: Passive ventilation in the velocity gradients near surfaces. Science 175, 210–211. Waslenchuk, D. G., Matson, E. A., Zajac, R. N., Dobbs, F. C. & Tramontano, J. M. 1983 Geochemistry of burrow waters vented by a bioturbating shrimp in Bermudian sediments. Marine Biology, Berlin 72, 219–225. Whitehead, N. E., Vaugelas, J. de, Parsi, P. & Navarro, M. C. 1988 Preliminary study of uranium and thorium redistribution in

Callichirus laurae burrows, Gulf of Aqaba (Red Sea). Oceanologica Acta 11, 259–266. Witbaard, R. & Duineveld, G. C. A. 1989 Some aspects of the biology and ecology of the burrowing shrimp Callianassa subterranea (Montagu) (Thalassinidea) from the southern North Sea. Sarsia, 74, 209–219.