ON VARIATION IN THE SHELL OF THE DOG-WHELK, NUCELLA LAPILLUS (L) 1, PEMBROKESHIRE, By J. H. CROTHERS ‘The Leonard Wills Pield Centre, Nettecombe Court. near Williton, Somerset TA4 4H INTRODUCTION THE Common Dog-Whelk, Nuclla lapillus (L.). is unusual amongst marine animals in having no planktonic phase in its life eyele. This. coupled with its sluggish habits, means that the effective breeding. populations (the panmictic units) are small, and it also makes feasible breeding experiments on the shore and in the laboratory. The species thus lends itself to a study of variation. Unfortunately, most papers on this topic consist of qualitative assessments from one restricted area. Frequently the conclusions of one author conflict with those of another and, in the absence of numerical data, it is difficult to know the meaning of ‘squat’. ‘elongated’ and so on. ‘The only safe conclusion from the literature is that populations of Nucella vary one from another, and thar patterns of variation found in one area will not necessarily appear in others. In an attempt to clarify the situation I propose to describe variation in numerical terms wherever possible, and to relate my observations to the patterns si sn in a type locality. This preliminary paper is an account of some 60) samples collected from my type locality: Pembrokeshire, West Wales. TERMINOLOGY “The term ‘population’, like so may familiar terms, seems to present no difficulty until we have to define it more closely.” (Briggs and Walters, 1969.) In previous papers I have used this term to denote the group of dog-whelks in a given site, assuming each isolated area of rocky shore to support a discrete breeding unit. Nothing, however, is known about the size of the panmictic unit in Nulla and. as an interim measure, | propose to use the -deme terminology to describe variation within and between populations’. This terminology was originally suggested by Gilmour and Gregor (1939) and has been recommended mote recently by Briggs and Walters (1969) from whence the definitions below are taken. The suffix ~ deme sienply means and the terms are constructed by the addition of an group of individuals appropriate prefix as follows: Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Topodeme: 2 group of individuals occuzting in a specified geographical area. Ecodeme: 2 group of individuals occurring in a specified type of habitat. Cytodeme: a group of individuals differing cytologically from other groups (usually in chromosome number). Gamodeme: 8 group of individuals which are so situated spatially and temporally that, within the limits of their breeding system, all can interbreed. Caenogarnodeme: a unit composed of all che gamodemes that are considered capable of exchanging genes to some extent, but not with freedom, ‘The advantages of such terms include the absence of the overtones which cloud the meaning of ‘race and ‘population’ in use in ordinary speech. They are not exclusive: nwo individuals of the same topodeme could belong to different cytodemes. THE BIOLOGY OF NUCELLA LAPILLUS ‘The Common Dog-Whelk is a marine snail found between the tide marks on rocky sea shores. In Pembrokeshire it usually feeds on three species of acom barnacles: Balas balanoides Bruguiere. Chthamalus stellatus Poli, and Eduinius madestus Dazwin. ‘The feeding process, described by Fretter and Graham (1962), takes hours if nor days so that the whelk must remain o prey over the period of low tide - in contrast to most shore predators which forage at high tide or at night and hide away at low tide and by day. Mussels. Mytilus edulis 1... are usually eaten if available and may be preferred to barnacles. Mussel-feeding Nwella were included in my samples from Little Haven, Marloes Sands and Tenby. If the usual food is scarce, dog-whelks attack other barnacles and molluscs. On sheltered shores Flat Winkles, Littorina litoralis(.) and Purple Topshells, Gibbula umbilcals (da Costa) are the most frequent alternative Some topodemes feed on limpets, although | have not found any that do so in Pembrokeshire, and occasionally on cockles Cardium (Cerastoderma) edule L. (Morgan. 1972). In Spain I have found holes drilled in the stalked barnacle Pollpes comucnpia (Gmelin). In general the food of each topodeme is related to the available food supply. ‘The species may be cannibalistic, bur whilst many shells show signs of boring few are drilled right through (Moore, 1938). If the attack is made too close to the shell lip the vietim may be able ro withdraw into its shell behind the threatened spot. I have seen this happen in aquarium and the inner end of the hole was sealed over within a few days. Subsequent growth causes the repaired hole to move round the body whorl to sites where an attack would have been fatal. Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Newly-hatched individuals feed on the tube worm Spirorbis (Moore, 1936,1938) - and many certainly do so in Pembrokeshire, where worm tubes showing pin-prick sized drill marks can be found. They are not, however, restricted to this diet (Largen, 1967b) and I have reared Nucella to at least one-year-old on 2 dict of barnacles (Elminius). Adults congregate on the lower shore to breed. Largen (1967a) found that egg capsules may be laid whenever the sea temperature rises above 7°C and, as the water in Milford Haven rarely drops below this figure (Nelson-Smith, 1967), Pembrokeshire Nuvella may breed at almost any time of the year - in contrast to Yorkshire topodemes (Feare, 1970) - but I believe most capsules are laid in February and March. In my aquaria the capsules hatch after two or three months to release berween ten and cwelve snails, ‘There is no planktonic or other dispersal phase in the life cycle, the veliger stage taking place within the egg capsule. so that the young animals must live on the same stretch of shore as their parents. The species is rarely found beneath the tide marks and does not voluntarily crawl across sand or mud, with the result that Nuella is distributed around the coastline of Europe in innumerable discrete gamodemes. This is particularly noticeable along the deeply-indented coastline of Pembrokeshire. ‘The young snails increase in size for about three years until they reach sexual maturity (Moore, 1936: Feare, 1970) when shell growth stops. Low temperatures inhibit feeding (Largen, 1967a) and topodemes in the North Sea probably feed only in summer, spending the winter in clefts where they may form massive aggregations (Feare, 1970, 1971). In Pembrokeshire Nvvella is found on the open rock surface throughout the year and apparently aggregates only to breed. ‘The life-span is unknown in Pembrokeshire but Feare estimates 5-6 years at his site in Yorkshire. Nothing is known of the formal genetics of shell variation in Nucella but it is likely that intrinsic (genetic) and extrinsic (environmental) factors combine to produce variation in size, shape, colour, thickness and shell ornament. If this is the case, an individual may arty genes for white colour and yet appear purple through the growth of algae or lichens on the exterior. Similarly it may be worn smooth by abrasion even though it carries the genes for ridged surface; and the genes controlling shell shape will not find normal expression if the cells of the mantle, the site of shell secretion, are damaged. Damaged shells are repaired bur the later secretion is never identical to the original and invisible damage to the animal, such as a period of starvation, may induce considerable changes in shell deposition which persist afier dhe immediate cause has disappeared. Reference: Crothers, J. (1974), Field Studies Vol 4, No 1“The shell records the history of that individual from the first-formed protoconch, which is carried on emergence from the egg capsule and retained at the tip of the spire, to the most recently formed part around the aperture, MATERIAL AND METHODS Pembrokeshire was chosen as type locality for my study of shell variation in Nwella lapilus because: a. The animal is widespread and abundant, b. Shores of widely differing exposure to wave action occur within a small geographical area c. The biological effects of differential wave action on rocky shores have been described by several authors, especially around the Dale peninsula where Ballantine (1961) devised his exposure scale. 4. Talzeady had collections of shells from the County ©. [know the area well The observations which form che basis of this paper are derived from an examination of some 10,000 shells in 60 samples taken from sites on the Pembrokeshire coast between 1966 and 1972 (Table 1). ‘They fall into three groups: 1. Those collected during 1966/1967 in collaboration with Dr. R. J. Berry for work on stabi ing selection published in 1968 (sce p. 47). These samples vary in size but most exceed 200 shells. A determined effort was made to include individuals of all ages. ‘Those collected during 1969/1970 in collaboration with Mr. F. B. Cowell for work on teeth” published in 1970 and 1971. Each sample contains a minimum of 200 shells, collected as far as possible without bias toward size, shape or colour. In practice the smaller shells were under sampled. ‘Those collected subsequently to augment information in certain critical areas. Each comprises 100 shells, specifically taken for analysis of shell shape. They, too, are probably biased towards larger individuals, The varied origin and purpose of these sa ples would preclude some analyses but all are thought to be representative of the full-sized component of the topodeme concerned: and most of the work described here is concentrated on these individuals. The animals were killed very shortly after collection and extracted from their shells. which were cleaned in a liquid houschold bleach and stored dry. Most of the samples were dried in an oven and as, unfortunately. shell colour is altered by heat the collection can only be used for analyses of shell colour in avery general way. Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Pembrokeshire THE INVESTIGATIONS The Dale Fort Marine Fauna (Bassindale and Barrett, 1957: Crothers, 1966) contains the observation shat “as exposure increases so the shell (of Nucelli) thickens. the ornamentation and che overall length decrease, and the proportion of the shell occupied by the aperture increases”. This observation appeared to be generally true elsewhere in Pembrokeshire but has not previously been described in numerical Where possible 100 shells from cach of my samples were measured (o the nearest 0.1mm) for length (L). aperture length (Ap) and thickness (I). The mean with its standard deviation, were obtained for cach sample in respect of L, L/Ap and ‘T ‘THE ASSESSMENT OF WAVE ACTION Direct measurement of wave action on sea shoves is not yet possible but a useful assessment of its biological effects may be made using Ballantine's (1961) scale in which che whole comples of factors that we loosely refer to as “exposure” is estimated through the imegrated effect on the distribution of selected indicator organisms. It anges from the extremely exposed grade 1 to the extremely sheltered grade 8. It was originally devised in relation to the Pembrokeshire coast, so here at least there can be no doubt as to its applicability. In later papers, when T plan to compare variation in other arcas with the data presented here, it will be necessary to examine to what extent this, ale may be relied on because the disuibution and abundance of Ballantine’s indicator species are influenced by many factors in addition to wave action, Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Tables 1-3 list the mean values of L, L/Ap, and T for cach sample, The samples are grouped by ‘exposure but no attempt was made to list the sites in order of exposure within a grade. The same data are displayed graphically (Figs. 2. 4 and 5) in a way which makes no assumptions about the intercepts. between the grades on Ballantine's exposure scale. ‘This is important for, whilst the grades are undoubtedly sequential - thar is, exposure 4 is certainly more exposed than exposure 5 and less exposed than exposure 3 - they are not necessarily equally spaced along the exposure continuum. My own guess is that few errors are incurred by assuming the scale to be linear, and I have done so in some earlier papers (Cowell and Crothers, 1970; Crothers. 1971, 1973) but itis obviously better to avoid making such assumptions where possible. RESULTS Length Length (or height in strict conchological nomenclature) is the easiest character to measure on a dog-whelk shell. I have taken the maximum reading from apex to siphonal notch (Fig. 3). Damaged specimens and all obvious immatures Sree Miosnse were omitted, but itis not easy to identify adults ‘theese nin mmm Asn ec nine pe in Pembrokeshire samples. The extended breeding season, coupled with the widely differing growth rates of individuals from the same brood, make it very difficult to recognize definite age classes. Teeth can no longer be used as a definition of adulthood (Cowell and Crothers, 1970) although it remains trae that an animal having a thin sharp lip to its shell and no ‘teeth’ is (in Pembrokeshire) almost certainly immature. The samples measured were thus a misture of adults and nearly full grown immarures. The mean length of topodemes from exposed shores is appreciably less than that of those inhabiting sheltered shores; the mean length of dog-whelk shells increases with increasing shelter from wave action. Shape: as indicated by L/Ap A variety of measurements have been used to describe the shape of dog-whelk shells but those involving maximum breadth, maximum distance across the body whorl or internal measurements of the aperture are all, to some extent, subjective. Measurement of the operculum is often the best indication of aperture shape but this is only really suitable for use on fresh, dead material - and some topodemes from the south-east of the County contain 2 high proportion of individuals in which the operculum is reduced or absent (e.g, Amroth and Lydstep Haven - see Cooke, 1917). Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Thave taken length (L) divided by aperture length (Ap) as defined in Fig. 3 as being the most convenient and objective assessment of shell shape in this species which can be made over the whole range of forms encountered, ‘The values of L-/Ap increase with increasing shelter, indicating the occurrence of a short squat form on exposed sites which gradually gives way to a more clongated varicty in shelter. A more detailed analysis of this data (Crothers, 1973) indicates continuous variation within and between topodemes. Thickness When growth stops, on cessation of feeding or at maturity. the shell lip of Nulla is thickened and a row of white dentiform tubercles (commonly referred to as ‘teeth’) are laid down along the inner margin. Should growth subsequently continue the thickened, ‘toothed’, area remains and fresh shell of normal thickness is deposited around the lip. ‘The process repeats itself whenever growth is seriously checked and exceptional shells have six or more rows of these ‘teeth’. In these individuals i is impossible to determine the true thickness of the shell but fortunately few Pembrokeshire Nucella show more than two rows, Even so the thickened lip prevents the use of normal callipers. Dr. K. Bignell of Imperial College designed and made an ingenious spring calliper which measures thickness inside the body whorl and data obtained with this instrument are preferred to estimates based on weight and size such as have been used in the past. Obviously damaged shells, and those bored by the worm Pobdors or badly corroded by lichen and algal attack were not used. As mentioned above it is not always possible to recognize adults but as there seems to be litde change in thickness, apart from the lip. on reaching maturity perhaps this does not matter. The thickest shells were found on shores of intermediate exposure with the thinnest on the most exposed and most sheltered shores. DISCUSSION ‘The results confirm that “as exposure increases so... the overall length decreases and the proportion occupied by the apermure increases”. A short squat form occurs on exposed shores. which grades completely into a longer. thinner variant found in sheler. ‘There is continuous variation berween the very well marked extremes. Reference: Crothers, J. (1974), Field Studies Vol 4, No 1The data as presented do not, however, establish whether the correlations are with wave action itself or with some other associated factor. Fortunately there are two further sets of observations which help to clarify the situation, Stabilizing Selection \ decrease in variability with age (that is, where the shells of adults are less variable than those of juveniles in the same ropodeme) may indicate the operation of natural selection, presumably eliminating those shell types least firted for survival in that locality. ‘The phenomenon of linkage’ makes it difficult to determine just how the selection operates but the fact that it occurs is relatively ensy to demonsteate. To investigate this phenomenon some parameter must be chosen which does not itself vary with age or size, and Berry and Crothers (1968, 1970) used length divided by the cube root of the drv weight. This, at first sight, unlikely choice is justified because lengch and dry weight are the least inaccurate measurements of small shells and a graph of these nvo parameters plotted against each other showed that length plotted against the cube root of the weight would be a straight line. In other words. length divided by the cube root of the dry weight does not show any trend with increasing shell size. Shells were grouped into five sequential ‘age’ classes on the basis of Moore's (1936, 1938) accounts of ty the life cycle, and the variance of the mean IW acs for each class in cach sample. At the time of this investigation (1966/1967) the presence of “teeth” was generally considered to indicate maturity (see below) and we used this character to define our oldest age class. This we now know to be unreliable but the main conclusions of the work are not affected by this error for selection was found to ‘operate primarily on younger individuals. We found evidence of 7 considerable variance-reduction Se (presumed selection) in samples a from exposed sites but no Parone obvious pattern could be seen in those from shelter (Fig. 6). We Fe & concluded that selection was Thejntaniy of sabsng etn by wave aon oo J seer Dae: Rembstoi, Th Mak ea anes Poser ee lems a tas Th acting to maintain the “Tad Grothe 1985 ws preponderance of short squat individuals on exposed sites and consider it an example of stabilizing selection maintaining the characteristic ecodeme under these conditions. Reference: Crothers, J. (1974), Field Studies Vol 4, No 1‘This clearly is only part of the story for it does not explain the situation on sheltered shores where the exposed-shore form is progressively replaced by the longer thinner variant. “Teeth” For some 30 years following Moore’s work on dog-whelks nobody questioned his assumption that teeth’ were produced at the onset of maturity and at no other time. Then, in the late 1960s, chree people, working entirely independently, noticed that they are also formed when growth is temporarily interrupted. Bryan (1969) recorded “tooth” formation in immature Nyala at Porthleven (Cornwall) and attributed it to starvation following the Torrey Canyon incident: Feare (1970) showed “tooth” development on the Yorkshire coast to be associated with winter cessation of feeding: whilst I noticed that shells with two or more rows of “teeth” are particularly common in Somerset. The formation of these “teeth” simply indicate a prolonged interruption of growth. All mature individuals can be expected to show a row near the lip because they will permanently have finished growing and additional rows inside the body whorl corresponding to the number of periods of starvation they suffered as juveniles. In some places near the northern and southern limits of the species range the shells of Nucella lapillus ae very thin and “teeth” are not usually detectable, but this thinshelled form does not occur in Pembrokeshire. Mr. E. B. Cowell, then Warden of Orielton Field Centre (South Pembrokeshite) noted a relaconship berween the incidence of juvenile starvation (as revealed by “tooth” formation) and exposure. He and I (Cowell and Crothers, 1970) were able to demonstrate that some 42 per cent of toothed shells on very exposed shores can be expected to show signs of starvation, a proportion which decreases progressively to 7 per cent in extreme shelter. We consider that this relationship is produced directly by the effect of wave action and suggest that individuals on exposed shores are that much mote likely to be dislodged by the waves (and thus swept away from their food supply) than those in shelter. The pattern cannot be explained simply by availability of food for both mussels and bamacles arc most abundant on exposed shores. Starvation can be caused by other factors as well so that unexpectedly high numbers of shells show additional rows of “teeth” in some areas. One such factor is low winter temperature which can prevent feeding (Largen, 19672; Feare, 1970a, 1971) and this may explain the data from the North Sea (Crothers, 1971) bur locally pollution can be more important as may be the case at Polkerris (Cornwall. Reference: Crothers, J.(1974), Field Studies Vol 4, No 1Collections from Lire Haven and further south in the County do in general conform to the expected pattem for south-west England (Fig. 7) but those from Nolton Haven and further north give substandally higher values. The discontinuity between these two sites in St. Bride's Bay compares with those others in Lynmouth Bay, Carmarthen Bay and Lyme Bay described by Crothers (1971) but as yet unexplained. Despite these interesting anomalies. it is clear that young Nucella are much more likely to suffer starvation on exposed shores than in shelter and that there is considerable evidence that wave action itself is affecting the maintenance of the short squat shell form on the more exposed sites. Shell Thicknes The pattern of variation shown for this character is more complicated than that of length or shape. If all the data are considered together the thickest samples are scen to occur on shores of intermediate exposure which agrees with the situation near Roscoff (Brittany) (Staiger, 1954: 1957 in English), where Nucella lapilles show's chromosomal polymorphism between a = 13 and n = 18. There is no difference in the amount of chromosomal material present so that the two forms are fully interfertile and individuals may be found with any of the possible diploid combinations between 26 and 36. The haploid set of form 13 contains 3 metacentric chromosomes. each of which corresponds to two non-homologous chromosomes in form 18, so that the nwo sets are really n = 8 + 5 in form 13 and n = 8 + (5 x 2) in form 18, Near Roscoff the cytodeme homozygous n = 13 occupies exposed situations and cyrodeme n = 18 sheltered ones. Intermediate sites are occupied by heterozygous cytodemes showing continuous variation (within and between topodemes) in chromosome number. The homozygous forms have thin light shells whilst the heterozygous shells are thick and heavy. (Staiger did not measure thickness directly but estimated it from a combination of weight and linear measurements.) ‘This particular situation may not be generally true. Staiger (1957) could find only the n = 13 form east of the Baie de Morlaix in Britany, which is also the case in Norway (Hoxmark, 1970) and on the Adantic coast of America (Mayr. 1963). Current research in this Country (Dr. C. R. Bantock. personal communication) has shown that the a = 13 form predominates in many areas, though in others both are present. No information is available for Pembrokeshire at present. It may be misleading to lump all the thickness data together. for in some areas a different pattern is seen. In the north-west of the County thickness is related directly 10 exposure, which is as Staiger (1957) Reference: Crothers, J. (1974), Field Studies Vol 4, No 1described when only the n= 13 form was present. But elsewhere the data conform to Staiger’s description where both forms and the intermediates were found. Chromosomal data for Pembrokeshire topodemes is urgently required and without it no further conclusions can be drawn. It should be noted also chat whilst my d: do fit closely with Staiger's in relation to thickness the two sets do not agree in respect of length. He found the longest shells at the extremes of exposure and che shortest ones in the middle. Shell Ornament Colton (1922) wrote that “The shell surface may be smooth or it may have 10 distinct ridges. In some cases alternate ridges become reduced or obliterated, giving the shell six distinct ridges. The most striking variation of all is one called imbricata in which the ten ridges become fluted. E: erience seems to show that all these varieties grade one into another.” Most Pembrokeshire Niclla are ridged but the six ridged form has only been noted at Amroth and the 10 primary ridges may have secondary ridges haw berween them giving a total between 10 and 16. The only smooth sh en in the County had clearly been abraded or corroded by environmental conditions. The variety imbricata is widespread but nor common. Usually shells which appear to have been imbricated when young have lost the fluting on Inter whorls. Largen (1971) is convinced that this is always due to abrasion of the body whorl. Colour Most Pembrokeshire Nucella are white or yellow but dark-coloured shells ean be found at some localities. Non.white shells may be self-coloused - thar is the colour is uniformly distributed - or there may be coloured bands on a white background. A wide banded form, usually with three darker bands. is easily recognized but is much less common than a narrow-banded one in which the ridges on the shell surface are white and the pigment confined to the hollows. In other parts of the country the banding may be superimposed on a (differently) coloured base colour but I have not noticed this in Pembrokeshire, Some shells show a combi tion of wide and narrow bands so that the broad bands are represented by three or four narrow stripes. In a few shells the banding is seen only on the inside, the outside surface being white. Moore (1936) related colours to food supply. showing that purple pigment comes from mussels. Animals feeding on mussels would be pigmented and those feeding on barnacles white. (Yellow coloration is not apparently related to food. This explanation appears to fit field observations in Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Comwall (where Moore collected many of his samples) and in some other places but in much of Britain colourcd shells are scarce even in topodemes that regularly feed on mussels. This might be accounted for if ir is assumed that the expression of purple pigment in the shell is controlled by three groups of genes. One set would contzol the uptake of the pigment and its deposition in the shell: another would control the intensity, and hence the colour observed: whilst a third would control any banding. Clearly the first set would be critical, for unless the animal is able to take up and deposit pigment in ies shell the other nvo cannot find expression. We my ¢ postulate that this critical set is widespread in the Cornish caenogamodeme(s) but very scarce in Pembrokeshire. Mytilus is generally distributed but abundant at only a few sites (Crothers, 1966) so that the incidence of individuals which carry the requisite set of genes feeding regularly on mussels (and hence the incidence of purple coloured Nulla) is low. Most coloured shells are banded. so perhaps the genes for this character are widespread. Even this explanation cannot be the whole story for I have raised the progeny of coloured parents from egg capsules laid in the laboratory on a diet of only bamacles, and many were coloured. But before Moore's explanation is entirely discredited it must be emphasized that coloured Nivella shells ate only common in areas of mussels. The occurrence of coloured individuals in the field is affected by extrinsic factors in addition to the presumed genetic ones. Not only does the human observer notice conspicuous individuals more than others but birds probably do so too. Selection for protective patterns (crypsis) is thus expected but has not yet been demonstrated for this species. The extreme crypsis of certain colour patterns amongst dense mussel patches is very striking. However, there is litte evidence that avian predation is critical in Pembrokeshire and all topodemes sampled contain numbers of very conspicuous shells. Berry and Crothers (1974) review colour variation in Nacella throughout the British Isles. SUMMARY 1. Nucella lepillus exists in innumerable distinet topodemes around the shores of Pembrokeshire. The very limited dispersion of individuals suggests that each topodeme may be a discrete breeding unit and should properly be considered a gamodeme. 2, Pembrokeshire gamodemes show continuous variation in the length (height) of full-sized shells, correlated with the exposure of the site to wave action so that the mean length of 100 shells decreases with increasing exposure. 3. Pembrokeshire gamodemes show a related variation in shape. A short squat form is typical of exposed situations and a more elongated variety in shelter. Reference: Crothers, J. (1974), Field Studies Vol 4, No 14. This variation in shell size and shape is. in part, maintained by the sclective effect of wave action on exposed shores where the short squar type is less likely to be dislodged. ‘The incidence of juvenile starvation, as indicated by the occurrence of “teeth”. accords with this view: 5. There is little variation berween Pembrokeshire gamodemes in colour or ornament and all are thought to be genetically similar in this respect. The apparent differences arise from the abrasion of che omamentation on exposed shores and/or the colonization of the shell surface by various algae and lichens. 6, Individual gamodemes can exhibit characters not seen in others: for example, the occurrence of reduced or absent opercula at Amroth and Lydstep Haven. >. Pembrokeshire gamodemes do not show a simple pattern of variation in shell thickness. Area effects occur. Shell thickness is thought to be influenced by chromosomal polymorphism in this species and a comparison is mooted berween the situation ia Pembrokeshire and that in Brittany where the chromosomal pattern is known. 8. The Pembrokeshire gamodemes could either be considered as forming several eaenogamodemes on the basis of shell thickness (each containing a closely similar genetic constitution with respect to size. shape, colour and ornament) or as one caenogamodeme encompassing a considerable range of variation in thickness. | favour the latter explanation, 9. ‘The observed variation in Nucella lapilins is a complex of genotypic and phenotypic factors. Some of the more easily studied patterns of variation seen around the Pembrokeshire coast are described in a manner to facilitate comparison with the situation encountered in other parts of Ei ope. a first step in the identification of cacnogamodemes. (ems) qi 2 239 a4 « «8S a6 6 7 3 Expoture edhe ag on tthe ale et SASL Se le ae oe me se mn ain Reference: Crothers, J. (1974), Field Studies Vol 4, No 1exposure Fr, 4 The reaionship between shell shape (LAP) and exponire for Pembrekehire amples of Mace lapis, Sample rom cxpoure grades TRS Snipa Seep ae ae mb recy rd a ea Duo Tae use ceeds a "Er ebered atthe Iercros been expore grades are ot necveacy equal, Dat rom Table 2 i * 2e| * Bi : a “Pas gelone eener ht Mike end sees fy Pern Scat GA then ah a lg Sn sa eae aed Bo eT Peo eae ade ads Se cnet eqs Bata on ate Reference: Crothers, J. (1974), Field Studies Vo! 4, No 1— el bakes 7 2 28 9 4 4 «8 5 Se 6 67 7 = exposure Fest “ne iarp bowen ote svn a inde by the arm sh by aie Mela i, ore a pe ee a a aM la ee aaa ST itRSS oe atm Geen: fot), meen, tom pure grades nd rn a ped Oy Cand a enh SP pees are more rey iS sae carn a ped os ether or ae Coan see See ambsee ar mor ly Sweden te ean aus rhe vars apart ade inks san) Gao ate Se ee Yes eet fiken ls dened mace) ted ie rd epee ce te ecg. 1 exponure for Pembroke should be rene “Tee paige et os (7 tt ec api of Sas te ida oo Een MEA sO Siac pe Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Table 1. Variation i the mean length (L) of Nuclia lapis sel called Pembrokeshire Samples hace been grouped by exposure rade, but ve ised wit eck rade merely in numerical trder of the sample number. The values for the mean and standard deviation of L were calealated “fom nls’ 7 a Gea [Coes [Soa 1 [10 fetiom tae Noe [fe T2747 Java a61 | SS 1 | He sitile tease | i as oom as ise emer as P| pe ii eve | SA 2 |e es | | Fetes | 1 | ae Gemain secs |S aus 4 as | ese 2 | esis tae soa 16) 2 |gseivrias | sc axe | Re 2 passotr B agsidoeetd, | St gaie) Bee st 23 | acre on | ontee| ug 198 3 [usmeatensonte | sae mre ee te) | 3 3 | 123 Greac Castle Head ‘SM 647059 | Feb. 1870 2 3 emer | Sea ie is |e 2 aa 3 eecest Sis |S 3 | 18/2 3 | iene SURGE [OE ie tT e+ | us wera 24 zea | 50 | to | 2 Ee upwcaoas | sk mets | reese So MERA, | RR OSs 1] + | upytnetines | ye gras | uy ee | 00 | 2 4 [upMonkries | sk as F + |B See } { |Besery., | Stee i 1 |Brskiens: |S 8ee i IBReREraS [Ree FRESE | asso. 1 )geower 5s get ed 1 pee, | ae EE 2 [eterna |B See ERE 2 [Bezeeeen | a ey ed $ | HRERE ee | ER tee ee 4 |Meat |S ae ESE +5 | iepaersttcen | ane rons | ser 167| 00 2-c020-21 $5 | ieepematisse | xtra | ate] | soee a 5 erica SURRY ESEEE | scour 5 | izewerancesy | sac aay | ay 198 | 00| 2720.25 A eta A eae iat 3 | ewe SE es [fuk || PER 2 | Rae SRS [PS || EER 4 [inrwarsceetuy | scans | soe 6 | wo] 2-2920-27 co |aiumameey | errepes] ise] an] azar |, & | sycepazey snc sss | ae | ro | 2820 $ fies, [tea | tee sus) pasen § [Saeco | 38 passe tok | ak ee $ [weet | gepimer | set ian |i) dee f ERAN | SUE] Soe Ut) me bo § [RoR Reeiuo | ae seria im) pee B ee SSURIMI/ ER IBH [gt] ates 2 | wacom s2¢ ces | ste 361 | ao | 2 toeo © | gee SUBBHGS ie PBS 6 | gta Si game kee igs | We | pea 5 ese SES | Se | S| ERE | nso Table 1—eontnud Epon Te Gina | Gin | s | Londen 67) 323 St. Justinian’s ‘SM 722253 | Sept. 1971 | 100, 7 al eines liege) | pected ated Pi Bee ie |e | iboe Ee ae ie 7 $M 960360 | Oct. 1970 | 100 2e2e0.15 } ees SBEE SS 8/8 eae 33 ames | om, 19 | ss920 25 | f [sttarptcanmeny |S SE | et] | BEE | eo Soe ae aha arta BR Dee Ege Ah stele Aa eer eT "nt anh winded rer yw eed hte 7) ny tencatnn ad slay aps war nly acted he wet slab mato Reference: Crothers, J. (1974), Field Studies Vol 4, No 1Table 2, Variation in the shape of Nucells lapillus shells collected in Pembrokeshire. The ‘mean and its standard deviation are gicen for the parameter L]Ap—length of the sell divided ‘by maximum length ofthe aperture. From Crathers (1973) veeeee Dee eeeeece 12 sttta tae Rock | 100 1d Sener id 3 313 Stan ito a Grea aie 10 1d Maro Sands weg iis Bactse 8 18 Bese tet io 1) So Pe I 1G Cae Cheret ie 18 Foca io 3 Reesty ie ns War De i Hi Seth is 3 i Baie We Ed 3 119 Monk Haver we | ge up Nes Hives ie pa 2 ey Fe . ig Pearse ter te 8 Ho Bester: Wot me a Reference: Crothers, J. (1974), Field Studies Vol 4, No 1 ademas Uap bazoo1 hareo.oF batons 120-005