The vertical distribution of phytal ostracods in the intertidal zone at Gore Point, Bristol Channel, U.K.

The first detailed investigation of the vertical distribution of ostracods inhabiting intertidal algae is presented herein. A study was made of the ostracod faunas obtained from samples of Corallina collected at various levels along transects of the intertidal zone at Gore Point in the Bristol Channel. A total of 28 species were recorded, including four new species presently left in open nomenclature. Both the composition of the ostracod fauna and the population age structure of individual species were found to vary considerably according to tidal level. The development of instars apparently proceeds at a faster rate at higher tidal levels, probably due to higher mean annual water temperatures. Among the factors which may determine the percentage composition of the ostracod fauna at a given level are temperature, pH, and the life-cycles and migratory habits of individual species.


INTRODUCTION
T h e relationship between ostracods and algae, and its importance in influencing ostracod distribution, has been demonstrated by several authors, such as Athersuch (1.979), Hagerman (1966Hagerman ( ,1968Hagerman ( ,1969, Whatley &. Wall (1 969,1975) and Williams (1969). This study, forming part of a wider investigation of intertidal alstracods in the Severn Estuary and Bristol Channel (Horne, MS 1980), is concerned with the vertical distribution of the ostracods inhabiting intertidal rock-pool algae at o n e site in the Bristol Channel.
Gore Point is situated at the western end of Porlock Bay, just outside the downstream limit of the Severn Estuary ( Fig. 1). Backed by steep wooded slopes and low cliffs of Devonian grits and shales, the upper shore consists of a shingle bank which gives way at its base to a zone of large boulders extending t o about mid-tide level. From mid-tide to low-tide level, the shore slopes more gently and is composed mainly of small boulders and cobbles, amongst which are many shalllow pools. Occasional small pockets of sand and mud are to be found, especially around low-tide level. Westwards from Gore Point, the rocks forming the cliffs are also exposed over much of the intertidal zone, resulting in a more rugged shore profile. A s the upper shore supports little algal growth, the lower shore is the main intertidal seaweed zone, with a cover of Fucus spp on rocks exposed at low water and an abundant and varied flora in the rock pools. Around the low-water mark of spring tides (LWST) a dense forest of Laminaria digitata and L. saccharina is exposed for brief periods. The: rock pool flora is dominated by Corallina officinalis, Chondrus crispus, Ceramium rubrum and Laurencia pinnatifida, with Ulva lactuca bcing abundant in some of the very shallow pools. T h e intertidal flora of Gore Point has been described in detail by Smith (MS 1978).
T h e maximum spring tidal range at Gore Point is approximately 10 m. Tidal and seasonal fluctuations of salinity are minimal (see Bassindale, 1943). In the main water body, salinity, temperature and p H values obtained at low water of spring tides in March, June and September, 1 9 7 8 varied from 31.0-33.8"/00, 9.0-16.0°C, and 8.2-9.6 respectively; greater variations were observed in rock pools isolated from the main water body at low tide, details of which are given in Fig. 2.
In order to investigate the vertical distribution of the ostracods inhabiting the algae in the rock pools, samples were collected at intervals along levelled transects measured from the high-water mark of spring tides (HWST) to the low-water mark of spring tides (LWST), o n three separate occasions.

METHODS
To ensure constancy o f sample type and t o allow comparisons between samples taken o n different occasions and at different localities. only the alga Corallina officinalis was collected. Sampling was not quantitative, but all the material was collected in plastic screw-topped jars of 125 ml capacity, so that approximately equal volumes of weed were obtained. Algae were collected by placing a jar in the water adjacent t o the weed which was then detached from the substrate and transferred to the jar as quickly as possibile t o minimise any loss of material.
Within a few hours of collection, 30% alcohol was added to each sample t o fix the meiofauna. Samples were then washed through 2.00 m m and 63 p m sieves, the finer fraction being retained and all the ostracods

LOCATION OF THE STUDY
picked from it. Only "live" ostracods, those carapaces containing appendages and the soft, often coloured central part of the body, were picked. All specimens were mounted on standard micropalaeontological slides for identification and counting.

THE TRANSECTS
Transects were levelled using a cross-staff, consisting of a 1 m staff with a cross-bar at the top to which a spiritlevel was attached, in conjunction with a 2 m pole, graduated in 10 cm intervals. Vertical distances were measured bv sighting along the cross-bar of the staff, horizontal distances were measured with a 3 0 m tapemeasure. Water temperature, salinity and p H were measured in the field using a mercury thermometer, an MC5 Salinity Bridge (made by Electronic Switchgear

A R E A
Ltd) and a Mach Portable p H meter (Model 16400).
Transect A was surveyed and sampled o n 14.9.77; samples were collected at five levels (A-E), all below mid-tide as Corallina was confined to less than 3 m. above LWST on the part of the shore chosen (Fig.2)     Physico-chemical data collected on transect B on 9.3.78 and 24.6.78 are shown in Fig. 2. The data obtained at low tide o n the latter occasion should be compared with salinity, temperature and pH values recorded at high tide about six hours previously of 31.5"/00, 14.0" C and 8.4 respectively. Only slight variations in salinity were observed on each occasion; pH, on the other hand, showed considerable increases on 24.6.78, the highest value (10.8) being recorded in pool A which was isolated from the main water body for seven hours. The above data, though limited, are in agreement with the findings of other workers such that in the intertidal rock-pool environment, salinity variations tend to be minimal whila temperature, and especially pH, may show considerable diurnal fluctuations (Daniel & Boyden, 1975;Pyefinch, 1943).

THE OSTRACODA
O n all three occasions, three 12Sml Corallina samples were collected at each of five levels. The numbers of ostracods given in Table 1 for each sampling pool are the combined totals from all three samples collected at that level. A total of fifteen 125 ml samples were thus examined from each transect on each occasion. From transect A on 14.9.77, a total of 1204 ostracods were obtained; the samples from transect B yielded only 756 ostracods c?n 9.3.78, but 5597 o n 24.6.78. The greatest numbers occurred at the lower levels o n both transect A on 14.9.77 and transect B on 9.3.78, while on transect B on 24.6.78 the maximum abundance occurred rather higher up, at level D. At level E on transect B on 9.3.78, only 29 individuals were obtained. A possible explanation for the extreme paucity of ostracods in that particular pool became apparent on 23.6.78, when it was observed that while the falling and rising tides were at that level wave action was focussed on the pool by the form of the surrounding rocky shore; the repeated impacts of large volumes of water and the resultant excessive turbulence must have rendered the environment of pool E extremely inhospitable to any free-living organisms, despite the rich flora of Corallina. When sampling of transect B was repeated on 24.6.78, another pool (D) was selected in place of pool E. (Brady, 1868) Semicytherura nigrescens (Baird, 1838) Semicytherura striata (Sars, 1866) Semicytherura tela Horne & Whittaker, 1980 Ten of the commoner species are illustrated in PI. 1. Taxonomic notes. Two forms of Aurila convexa were flound at Gore Point, differing in details of carapace morphology and colour. Both forms have been recorded bty other authors as A . convexa from N.W. European localities. No adult males have been found, and the taxonomic status of the two forms remains unclear; they are therefore considered as one species for the purpose of this paper.

Vertical distribution
The vertical distribution of the more important species from each transect is shown in the form of kite-diagrams in Figs. 3, 5 and 7 (note that different horizontal scales are used, although the vertical scale is the same in all three). Caution must be exercised in the interpretation of these diagrams, since the samples from which the data were obtained, while of similar size, were not truly quantitative. As an alternative, therefore, the percentage compositions of the faunas at each level are also shown, in Figs. 4, 6 and 8. A study and comparison of all six figures shows clearly that on each transect, both the overall faunal composition and the abundance of individual species varied considerably with respect to tidal level.
On transect A (14.9.77), Loxoconcha rhomboidea was most abundant at the lower levels (pools D and E), with decreasing importance up the transect, both numerically and in terms of percentage composition of the fauna. Similarly, Aurila convexa was abundant at lower levels, but was not found at the highest level sampled; numerically, its maximum was in pool D, but in terms of percentage composition it. reached its greatest abundance higher up, in pool C. Conversely, Paradoxostoma variabile was of minor importance at lower levels, but relatively abundant at the top of the transect.
The data presentation for transect B (9.3.78) (Figs. 5 & 6) may be somewhat misleading due to the previously mentioned sparse fauna obtained at level E, In terms of percentage composition, the faunal variations on transect B on 24.6.78 were essentially similar to those seen on that transect three months previously, withP. variabile dominating the low-level faunas ( F & G) and H . viridis the higher levels ( A , B & D). In terms of actual numbers, H . viridis, Heterocythereis albomaculata a n d P . sp. nov. A all exhibited maxima a t level D; P. variabile was most abundant at the bottom of the transect (G), with a general decrease in numbers upwards, especially above level D.

Population structure
The abundance of ostracods on transect B on 24.6.78, in comparison with the relatively low numbers found three months earlier, can be largely attributed to reproductive activity during the intervening period by three species, P. variabile, H . viridis and H . albomaculata. The population structures on 24.6.78 at each level of each of these species arc shown in Figs. 9-1 1. The populations of H . albomaculata and H . viridis consisted almost enirely of juveniles, which would have developed from eggs laid in the spring; the instars of the former would be expected to reach maturity by August or September to form an overwintering adult population, while those of the latter would pass the winter as latestage juveniles, maturing and reproducing the following spring (Hagerman, 1969;Horne, MS 1980). A detailed examination of the population structures of the two species shows clearly that they differed according to tidal level. In H . albomaculata, the A-5 and A-4 stages were dominant at the bottom of the transect; upwards, the percentage of larger instars steadily increased, so that at the highest level the A-2 stage was dominant. A similar pattern was exhibited by H . viridis, with the A-3 stage dominant at the bottom of the transect, the A-1 at the top.
The population structures of P . variabile followed a similar, though less clearly defined pattern. At the lower end of the transect, small instars were dominant and adults were few; upwards, adults increased in numbers, reaching a maximum at level D, above which the trcnd towards increasing abundance of larger instars was n o longer apparent, although adults remained dominant. P . vuriubile is apparently capable of reproducing throughout the year, although it reaches a maximum iin the spring or early summer (Elofson, 1941;Hagermann, 1966;Horne, MS 1980).
The populations of all three o f the above-mentioned species appear to have been at a more advanced stage o f development at the higher levels than a t the lower levels of the transect. This pattern could be attributed t o the effects of differing temperature regimes at different tidal levels on the development o f eggs and instars. High-level pools, being more often exposed to direct sunlight, are likely to have higher mean annual water temperatures than those at low levels, affecting the ostracods both directly (the developmenit of eggs in particular is know to be temperature-dependantsee Thiesen (1966)) and indirectly (by influencing food supply, such as diatoms), and resulting in faster growth rates at higher levels. A further contributing factor could be that at higher levels a reduction in the number of species competing for the same food allows faster development, through the absence o r reduced numbers of species found in abundance at lower levels.

DISCUSSION
The above results demonstrate that both the composition of the ostracod fauna and the population structure of individual species vary with tidal level, at least as far as the faunas inhabiting Corallinu at Gore Point are concerned. The reasons for this are unclear; the considerable fluctuations in temperature, p H and O2 levcls that occur in intertidal pools (Daniel & Boyden, 1975) may to some extcnt be responsible, although other factors should also be considered (e.g. competition with other organisms (such as copepods), qualitative and quantitative variations in food supply, and seasonal reproductive activity and migrations by the species concerned). [PI Paradoxostoma n.sp.A.

Home
Seasonal fluctuations due t o reproduction and especially migration by certain species almost certainly have some effect on the faunal composition at different levels. Thcre is evidence (Whatley & Wall, 1975;Horne, MS 1980) to suggest that L . rhomboidea, for example, migrates to low intertidal o r upper sublittoral levels during the winter, this being supported by the present findings. This species has a life-cycle of one generation per year, with overwintering eggs and adults (Elofson, 1941 ;Whittaker, MS 1972;Horne, MS 1980). O n 14.9.77, populations consisting almost entirely of adults were present at every level sampled on transect A, though with a considerable increase in abundance towards the bottom of the transectthese would have been recentlymatured individuals beginning their down-shore migration. O n 9.3.78, only a few specimens, mainly adults, were found at the lower end of transect B, while greater numbers, likewise nearly all adults, were found in a Laminaria holdfast sample taken at level H on that datethis is consistent with the idea of overwintering adults confined to around LWST. Finally, on 24.6.78, popula-tions mainly o f juveniles were found at every level on transect B, representing the new generation hatched from eggs laid the previous autumn before the migration of the adults. Similar seasonal migrations have been postulated by Whatley & Wall (1975) for H . viridis, Cythere lutea and H . albomaculata.
The population structures o f H . albomaculata, H . viridis and P . variabile on all three transects were consistent with what is known of their life-cycles. In the case of P . variabile, its virtual restriction to the top of transect A on 14.9.77 may be indicative of migration to higher levels during the late summer, perhaps in response to competition from other species which reach their rcproductive maxima at that time.
Before any conclusions may be drawn with respect to a possible intertidal zonation scheme for rock-pool ostracods, further study is needed. A future investigation should determine whether o r not the patterns of vertical distribution of ostracods are constant, or vary seasonally, and also whether lateral variations occur along the shore at different levels.  (1936) observed the restriction of some copepods to certain levels in rock-pools in the Isle o f Man, and noted that ostracods (four species) were particularly abundant at mid-tide level. Wieser (1952) found no correlation between the density of the phytal ostracod fauna and tidal level, considering only the overall fauna, but he did note such relationships in several other groups of organisms. Whatley & Wall (1975) recognised a "critical level" at the boundary between the "upper" and "lower" sublittoral o n the basis of phytal ostracod faunas, and some of their results hinted at a preference for certain intertidal levels (eulittoral and littoral fringe) by some species, although thc many different types of algae sampled confused the issue in this respect. Finally, it should be borne in mind that the tidal range in the Severn Estuary and Bristol Channel is the largest in the British Isles (10 m. at Gore Point, 14 m. at Avonmouth), a fact which may have some bearing on the vertical distribution of the ostracods; it has yet to be shown whether or not similar faunal variations with tidal level can be recognized in other localities where the range is smaller.