Indian Journal of Marine Sciences Vol. 31(3), September 2002, pp. 218-224

Reproductive effects of heavy metals on the rough periwinkle Littorina saxatilis (Mollusca: Prosobranchia)

Erema R. Daka1* Port Erin Marine Laboratory, University of Liverpool, Port Erin, Isle of Man, IM9 6JA, United Kingdom and S. J. Hawkins2 Biodiversity and Ecology Division, School of Biological Sciences, University of Southampton, Southampton, SO17 1BJ, United Kingdom Received 20 August 2001, revised 10 June 2002

The effects of chronic mine-related heavy metal contamination on reproduction of Littorina saxatilis from sites around the Isle of Man were assessed. Size at birth of young, reproductive output (i.e. number of embryos in brood pouch) and fre- quencies of abnormal embryos in the brood pouch were examined in animals from five sites. Laxey estuary was the most contaminated compared to other sites with less contamination (Peel) and relatively uncontaminated control sites (Castle- town, Ramsey estuaries and the open coast near Derbyhaven). The size at birth of young ranged from 0.365-0.876 mm and the lowest mean size at birth (0.517 + 0.048 mm) was found in the animals from Laxey estuary. Those from Derbyhaven had the highest mean size at birth (0.632 + 0.063 mm). ANOVA showed a significant difference in mean size at birth of young, between individuals from the five sites (p<0.0001). However, the individuals from Laxey also had the highest absolute mean count of embryos in their brood pouch implying that there might be some trade-off between reproductive output and size of young as a result of metal contamination. No heavy metal related differences in embryo abnormality were apparent. Rather, a high proportion of abnormal embryos in animals from Derbyhaven suggested that disease conditions, genetic factors or unidentified toxins might account for the differences observed.

[ Key words: Littorina saxatilis, Isle of Man, heavy metals, fecundity, size at birth, embryo abnormality ]

Sub-lethal effects of pollutants such as depression of Channel coast of southern England was the result of growth and reproduction occur at concentrations TBT and virtually all populations in England showed where acute toxic effects may not be apparent1,2. The visible effects of TBT pollution6,7. embryonic and larval stages of aquatic animals are It has been suggested that the rough periwinkle generally the most sensitive stages of the life cycle to Littorina saxatilis Olivi (Mollusca: Prosobranchia) heavy metals and other toxicants3. Hunt & Anderson4 could be a useful organism for monitoring the established that at concentrations above 19 μg l-1 zinc, reproductive effects of contaminants because embryos larval metamorphosis in the abalone Haliotis are found in the brood pouches all year round8,9. rufescens was impaired. Oshida & Word5 reported Embryonic development is internal and takes place in that the number of offspring produced by the the brood pouch of the female, hence local pollution polychaete Neanthes aranaceodentata was reduced by events would affect all stages of embryonic exposure to 39 μg l-1 of dissolved chromium (VI). The development, and in turn the number and size of dogwhelk, Nucella lapillus exhibits “imposex”, the juveniles may be affected. Most previous studies on imposition of male sexual characteristics on females, the inter-site differences in numbers of embryo and as a result of exposure to tributyltin (TBT), and the the size of juveniles at birth have concentrated on the decline of populations of Nucella along much of the effects of natural environmental gradients such as 8 ______shore height or, the degree of exposure or shelter of *Author to whom correspondence should be addressed the shores10-12. All stages of development from Present address: uncleaved egg to pre-emergent fully formed juveniles 1 Department of Biological Sciences, Rivers State University of can be found in the brood pouch of female L. saxatilis Science and Technology, P.M.B. 5080, Port Harcourt, Nigeria 2 Marine Biological Association of the United Kingdom, The at the same time. However, in addition to normal Laboratory, Citadel Hill, Plymouth, PLI 3PB, United Kingdom stages of development, the brood pouch may contain a Daka & Hawkins : Effects of metals on Littorina 219

proportion of abnormally developed embryos. Littorina saxatilis of reproductive size (> 6 mm Anthropogenic contamination has also been reported shell height) were collected from Castletown, Derby- as increasing the frequency of abnormalities9. haven, Laxey, Peel and Ramsey (from 8-10 May Despite being contaminated with heavy metals 1995; 1 July 1996), relaxed in 7.5% magnesium chlo- from previous mining activity13, the Laxey estuary, ride and fixed in 10% formaldehyde-seawater solu- Isle of Man supports a large population of Littorina tion18. Animals were then preserved in 70% ethanol saxatilis; these have been shown to possess enhanced until dissected. The shell height of each individual tolerance to acute levels of zinc and lead14. Sublethal was measured before cracking on a bench vice and concentrations capable of inducing tolerance to metals separating the tissue from the shell. The presence of a may exact reproductive effects. Also tolerance may have energetic costs, which might indirectly affect reproductive output15. The occurrence of such effects in L. saxatilis from sites around the Isle of Man with different levels of heavy metal contamination was studied. Inter-site differences in the proportions of abnormal embryos were examined to assess useful- ness of reproductive indices in monitoring the effects of sub-lethal concentrations of contaminants.

Materials and Methods The sites from which samples were collected around the Isle of Man (Castletown, Derbyhaven, Laxey, Peel and Ramsey) are shown in Fig. 1. The catchments leading into the estuaries at Laxey and Peel are still contaminated by past mining activi- ties13,16. Details of heavy metal concentrations in the environment and bioindicators are given elsewhere14. Metal burdens in the soft tissue of Littorina saxatilis from the five sites are presented in Table 1. The ani- mals from Laxey are exposed to highest concentra- tions of heavy metals, especially zinc but high levels are also present at Peel. In contrast the estuaries at Castletown and Ramsey have no mining contamina- tion, but there is little industrial contamination of Cas- tletown from boating and an industrial estate and Ramsey has been shown to be affected by TBT con- Fig. 1 ⎯ Locations of sampling sites around the Isle of Man and 17 past mining activities. Filled triangles and circles represent major tamination . The Castletown Bay side of Langness and minor producing mines respectively with an indication of the near Derbyhaven was presumed to be a clean coastal order of importance of the ores produced (Cu=copper, Fe=iron, reference site. Pb=lead, Zn=zinc).

-1 Table 1 ⎯ Concentrations (mean + s.d., n=5) of zinc, lead, copper and cadmium (μg g dry tissue) in Lit- torina saxatilis from five sites around the Isle of Man. Zinc Lead Copper Cadmium Dec. June June Dec. June Dec. June 1994 1995 1995 1994 1995 1994 1995

Castletown 121 ± 5 102 ± 3 11.5 ± 2.5 11.5 ± 2.5 11.5 ± 2.5 3.2 ± 0.7 2.5 ± 0.3

Derbyhaven 89 ± 3 103 ± 7 2.3 ± 0.9 2.3 ± 0.9 2.3 ± 0.9 1.9 ± 0.3 1.8 ± 0.2

Laxey 438 ± 42 504 ± 28 16.6 ± 0.7 16.6 ± 0.7 16.6 ± 0.7 3.7 ± 0.3 4.0 ± 0.3

Peel 157 ± 6 128 ± 9 16.6 ± 2.2 16.6 ± 2.2 16.6 ± 2.2 2.3 ± 0.3 1.8 ± 0.1

Ramsey 148 ± 7 135 ± 4 7.7 ± 1.7 7.7 ± 1.7 7.7 ± 1.7 1.5 ± 0.2 2.0 ± 0.3

220 Indian J. Mar. Sci., Vol. 31, No. 3, September 2002

brood pouch and penis was used to identify females normal embryos in each female were calculated for and males, respectively. The brood pouch of each fe- each of the two categories of abnormality (i.e. un- male was removed into a gridded Petri dish, teased shelled and shelled) and for total abnormalities. Sig- open and the contents examined under a stereomicro- nificant inter-site differences in the proportions of scope and counted. unshelled, shelled and total abnormality were tested About 50 animals of reproductive size (> 6 mm using the Kruskal-Wallis test and the Dunn test where shell height) from each site were placed in white plas- applicable. tic containers after acclimatisation to 16°C for 6 to 7 days in other tanks. The individuals transferred were Results thoroughly cleaned to avoid transfer of juveniles on the shells. The warm temperature was a means of in- Fecundity and size of juveniles at birth ducing the free young to emerge from the brood The absolute numbers of embryos in the brood pouch of the parents, which is entirely a function of pouch L. saxatilils from the five sites were variable the juvenile activity8. Juveniles that emerged were with total counts ranging overall from 24 at Castle- removed from the containers every 24 hours for five town to 1117 at Laxey (Table 2). Animals from days and preserved in 70% ethanol prior to measure- Laxey, the contaminated site, had the highest mean ment. The measurements (accurate to + 0.001 mm) number of embryos in both the May 1995 and July were made on a stereomicroscope (Wild Heerbrugg, 1996 samples. The mean absolute brood pouch con- Finlay Microvision Ltd.) equipped with a video cam- tent recorded in July 1996 was generally similar to era. Images were captured and calibrated to a set values for May 1995 except for numbers in samples magnification using a NIH Image Analysis software from Peel and Castletown. In animals from these two (1.58 VDM) on a Macintosh Computer. sites marked increases in mean number of embryos Embryo abnormalities were determined in samples were recorded, which were coincident with the rela- collected on 1 July 1996 during the middle of the re- tive increase in the shell height of females from which productive season. After collection, animals were the embryo were obtained. Kruskal-Wallis tests on stored at ambient temperature in the laboratory for a median total brood pouch content showed significant few days before dissection. The shell height of each differences for both sets of samples, but at different animal was measured as above before cracking on a levels (df = 4, H = 89.27, P < 0.0001 for May 1995 and bench vice. The brood pouch content of females were df = 4, H = 9.86, p = 0.043 for July 1996). Dunn multi- placed in a Petri dish and viewed under a stereomicro- ple comparisons on the May data showed that pair- scope for counting and examination of embryo ab- wise significant differences in numbers of embryo normality. Total number of embryos in each pouch followed the form: Laxey > Ramsey = Derbyhaven = was counted and each embryo was examined for ab- Castletown > Peel. However, the Dunn test could not normality. Abnormalities involving unshelled em- detect any significant differences between any pair of bryos (mainly characterised by the physical disruption sites for the July data in spite of the significant differ- of soft tissues, resulting in the presence of a number ences shown by the Kruskal-Wallis test. of separate, floating, cell masses within the egg cap- The relationships between shell height and number sule; double embryos) and shelled embryos (multiple of embryos in brood pouch were very weak. When embryos, mostly twins; abnormal shells, mostly un- regressions were drawn (loge number of embryos coiled shells) were counted separately. against shell height), the coefficients of determination Differences in the absolute number of embryos be- ranged from 0.050 (Derbyhaven) to 0.471 (Laxey) for tween sites were compared by means of Kruskal- May, 1995 and 0.072 (Derbyhaven) to 0.484 (Peel) Wallis one-way analysis of variance using MINITAB for July, 1996. This precluded the application of for Windows. Where significant differences were ob- Analysis of Covariance to test for differences in re- tained with the Kruskal-Wallis test, the Dunn non- gression coefficients as way of determining signifi- parametric multiple comparison test was used to make cant differences in size-specific number of embryos. pairwise comparisons between average ranks19. Sig- The range and mean sizes of newly emerged juve- nificant differences in size at birth between animals niles (0-24 hrs old) for the five study sites are given in from the five populations were tested by means of Table 3. Animals from Laxey (with the highest abso- one-way analysis of variance (ANOVA) and Tukey lute pouch content) had the smallest juveniles while HSD multiple comparisons. The proportions of ab- those from Derbyhaven were the largest. One-way

Daka & Hawkins : Effects of metals on Littorina 221

ANOVA showed a significant difference in the mean proportion of unshelled abnormals; those from Peel length at birth (F4,1412 = 123.0, p < 0.001). Tukey mul- had the least with Laxey site showing intermediate tiple comparisons indicated that significant differ- values. The mean proportions of shelled abnormalities ences conformed to the following order: Derby- were similar in all sites. There was very high intra-site haven > Castletown > Ramsey > Peel = Laxey. variability in the proportion of abnormals. Coeffi- cients of variation ranged between nearly 100% to Embryo abnormality well over 200%. The Kruskal-Wallis test showed a The percentage of females with abnormal embryos significant difference in median unshelled abnormali- was very high, ranging from 92% in samples from ties between sites (df = 4, H = 33.87, p < 0.001); with Castletown to 100% in those from Peel and Ramsey. individuals from Derbyhaven showing significantly The percentages of females with abnormal embryos higher abnormalities than those from all other sites from Derbyhaven and Laxey were 98% and 94% re- (Dunn tests, Table 4). No significant differences were spectively. found for shelled abnormals between sites. The ani- Much of the abnormalities in animals from all sites mals from Derbyhaven also had the highest propor- were found in the unshelled embryonic stages (Table 4). tions of total abnormality. The Kruskal-Wallis test Individuals from Derbyhaven had by far the highest showed a significant difference in median total

Table 2 ⎯ Number of embryos in brood pouches of Littorina saxatilis from five sites in the Isle of Man in May, 1995 and July, 1996 [ Highest mean values are in bold ] Site N Min. Max. Median Mean C.V. (%) May, 1995 Castletown 45 24 652 256 225 58.3 Derbyhaven 84 54 731 262 277 49.5 Laxey 45 52 955 398 435 44.7 Peel 73 27 396 131 145 46.6 Ramsey 44 39 1018 269 316 58.1

July, 1996 Castletown 39 46 798 338 359 45.7 Derbyhaven 63 38 690 278 289 47.3 Laxey 52 80 1117 366 432 59.5 Peel 46 70 835 320 357 56.0 Ramsey 50 115 580 287 305 34.3

Kruskal-Wallis One-way Analysis of Variance (H) and Dunn Tests (p < 0.05): May 1995: H = 89.27, p < 0.001 Laxey ≠ Ramsey = Derbyhaven = Castletown ≠ Peel July 1996: H = 9.86, p = 0.043 Laxey = Castletown = Peel = Ramsey = Derbyhaven

Table 3 ⎯ Size at birth (mm) of juveniles of Littorina saxatilis from five sites in the Isle of Man [ Lowest mean size is in bold ] Site N Minimum Maximum Mean S.D. Castletown 186 0.405 0.876 0.574 0.071 Derbyhaven 103 0.457 0.776 0.632 0.063 Laxey 419 0.356 0.676 0.518 0.048 Peel 417 0.375 0.700 0.525 0.053 Ramsey 292 0.394 0.732 0.547 0.047

Analysis of Variance F4,1412 = 123.0, p < 0.001. Multiple Comparisons (Tukey Tests, p < 0.05): Derbyhaven ≠ Castletown ≠ Ramsey ≠ Peel = Laxey

222 Indian J. Mar. Sci., Vol. 31, No. 3, September 2002

Table 4 ⎯ Proportions of abnormal embryos (unshelled, shelled and total abnormals) in Littorina saxatilis from five sites in the Isle of Man. [ N.B. Large coefficients of variation. N=number of animals. Results of Kruskal-Wallis One-way Analysis of Variance and Dunn tests are shown ] Unshelled abnormals Shelled abnormals Total abnormals Site N Median Mean C.V. (%) Median Mean C.V. (%) Median Mean C.V. (%) Castletown 39 2.6 3.3 120.9 0.8 1.3 108.5 3.3 4.6 106.1 Derbyhaven 63 3.6 14.3 108.4 0.4 1.4 232.9 5.2 16.0 104.6 Laxey 52 1.6 4.1 170.1 0.6 1.4 188.7 2.0 5.5 171.5 Peel 46 1.4 2.3 131.1 0.6 1.0 117.8 2.3 3.2 99.7 Ramsey 50 1.9 3.0 100.8 0.7 1.0 124.1 2.6 3.9 99.9

Kruskal-Wallis Analysis H=33.87, p < 0.001 H=4.16, p=0.386 H=25.03, p < 0.001 of Variance Dunn Tests (p<0.05): D C R L P D C R P L

Underlined=no difference C=Castletown, D=Derbyhaven, L=Laxey, P=Peel, R=Ramsey

10 abnormalities between sites (df = 4, H = 25.02, environment. On the contrary, Faller-Fritsch p < 0.001) and Dunn tests found a significantly higher reported that investigations of Littorina rudis abnormality in animals from Derbyhaven in compari- (=L. saxatilis) in many parts of Britain support the son with those from Ramsey, Peel and Laxey. How- view that differences in reproductive energy expendi- ever, no significant difference in total abnormality ture were general in occurrence; such that fewer, lar- was found between animals from Derbyhaven and ger embryos were characteristic of sheltered popula- those from Castletown. Spearman’s rank correlation tions while a large number of smaller offspring are coefficients indicated that no significant correlation favoured in exposed shores. Similarly, Roberts & existed between shell height and the proportion of Hughes11 found from studies of three Welsh sites abnormal embryos in the brood pouches of female (a sheltered saltmarsh, a semi-sheltered boulder shore winkles at any of the sites studied [Castletown: and an exposed cliff) that the largest juveniles were rs (39) = 0.046, Derbyhaven: rs (63) = −0.002, Laxey: produced by the population from the most sheltered rs (52) = −0.121, Peel: rs (46) = −0.206, Ramsey: site. However, the smallest juveniles were not rs (50) = 0.044, in all cases p > 0.05]. obtained from the exposed shore but rather from the semi-sheltered boulder shore and they did not find Discussion any significant differences in the size-specific number of embryo in the brood pouch. They concluded that Fecundity and size at birth of juveniles there was no significant trade-off between the size of Intraspecific differences in numbers of embryo and offspring at birth and parental fecundity in L. rudis size of juveniles at birth in Littorina saxatilis have (=L. saxatilis). been reported to vary with the level of exposure of the shore from which samples are collected10,11,20. Such The locations from which samples were collected variation should be taken into account when interpret- for this work were essentially sheltered, efforts having ing the likely effects of pollution on these reproduc- been made to control for the effect of exposure. How- tive parameters. Conflicting reports have been found ever, some differences were apparent in the degree of in the literature with respect to how exposed/sheltered exposure of the locations from which samples were shore types affect these reproductive parameters. For taken with Peel and Ramsey being slightly more ex- example, Janson20 compared the sizes of newly born posed, the former being more so than the latter. Over- juveniles of two morphs of L. saxatilis: exposed (E- all, Castletown and Derbyhaven were the most shel- morph) and sheltered (S-morph) from Swedish coast tered and Laxey was intermediate between the above populations and found that despite being smaller, E- four sites. The size structure of adults at these sites morph females on the average gave birth to larger reflects the above pattern, the animals from Peel being juveniles than did S-morph females; and concluded the smallest and those from Castletown the largest14. that this difference in hatching size is likely to be of The degree of exposure of shore may therefore not be genetic origin and an aspect of the adaptation to the responsible for the variations in the size of juveniles

Daka & Hawkins : Effects of metals on Littorina 223

observed as L. saxatilis from Laxey had the smallest Pollard9 working on the southern coast of Wales mean embryo size and the highest mean absolute found that individuals from a site which was furthest number of embryos both in May and July. The situa- away from the known sources of pollution had the tion of the animals from Laxey does not also conform lowest frequency of abnormal embryos in comparison to the contention by Roberts & Hughes11 that there is with those from other sites, but there was no correla- no trade-off between size at birth and fecundity, since, tion between the gradient of contamination and fre- the lowest size at birth coincided with the highest fe- quency of abnormality and they suggesting that a cundity in the samples from the Laxey. complex aetiology might be involved. It is possible A likely explanation for the size at birth and fecun- that the animals from Derbyhaven (highest embryo dity of L. saxatilis at Laxey relative to the snails from abnormality but least metal contamination) experience other sites (Tables 2 and 3), is exposure to high levels some form of contamination that has not been de- of heavy metals (mainly zinc but also cadmium and tected in this work, but this is unlikely. Derbyhaven lead). Bodar et al.21 observed from laboratory experi- has full strength seawater so it is not the cause of the ments with Daphnia magna exposed to cadmium that high frequency of abnormalities at this site. concentrations of 0.5, 1.0 and 5.0 ppb caused a sig- Dixon & Pollard9 suggested that abnormalities may nificant increase in the average number of neonates be caused by a latent pathogen which is activated to per female, but the size of the neonates were smaller. disease condition by environmental stress. This asser- They found, however, that at higher concentrations tion was based on the physical features of the most both brood size and body size declined. The lower dominant type of abnormality they observed were size at birth and higher absolute number of embryos reminiscent of the kind of uncontrolled breakdown in L. saxatilis may be similar to the above finding. It that might be expected from a neoplastic disorder, in is possible that the growth rate of embryo in the brood this case confined within the egg capsule. However, pouch is lowered by metal pollution such that it takes they did not find any pathological condition to be as- a longer time for the embryo to develop to the emer- sociated with the somatic tissue of adult snails con- gent stage at which it crawls out of the brood pouch. taining high incidence of abnormality. The predomi- The longer gestation time may have led to the pres- nant type of embryo abnormality observed in L. saxa- ence of large number of embryos in the brood pouch tilis from Derbyhaven bear the above features de- at any given time since the release of eggs is continu- scribed by Dixon & Pollard9 and it is possible that ous. The smaller size of the embryo may then be the some disease condition may be responsible for the result either of a packed brood pouch with limited high incidence of abnormalities at that site. space for growth or the direct effect of the reduced Janson12 argues that genetic factors may play an growth rate of embryos. Notably, the size of young at important role since environmental stress factors do birth followed the reverse order of metal contamina- not unquestionably influence the degree of embryo tion such that the cleanest (Derbyhaven) had the larg- abnormality. Large within-site variations in embryo est and Laxey the smallest with Castletown > Ramsy abnormality were suggested to be due to genetic > Peel. rather than environmental factors. Presumably, in- Other factors which may, however, affect the dif- breeding in small random mating groups was the pri- ferences in embryo numbers of L. saxatilis include the mary genetic factor involved. The possibility of such availability of food and the relative period of tidal an effect occurring in the Isle of Man populations of exposure which could affect foraging time.10 Berry8 L. saxatilis cannot be ruled out as very high coeffi- observed that fecundity in snails was higher for indi- cients of variation were found in the proportions of viduals collected from the upper shore than those abnormals in these sites. from the lower shore. Spatial differences in such fac- Contamination by heavy metals may be exerting tors from our study sites may have contributed to the direct or indirect effects leading to a reduction in the differences observed. size of embryo at birth in L. saxatilis from Laxey es- tuary. No evidence for induction of embryo abnor- Embryo abnormality malities by heavy metals was apparent in this study. The results of this study did not show any correla- The high proportions of abnormal embryos in animals tion in the proportion of abnormal embryos (Table 4) from Derbyhaven which are not known to be exposed in the Isle of Man populations of Littorina saxatilis to high levels of heavy metal pollution may be due to with metal pollution (Table 1). In contrast, Dixon & disease and/or genetic factors. It is concluded that

224 Indian J. Mar. Sci., Vol. 31, No. 3, September 2002

embryo abnormality in L. saxatilis may not be a good 9 Dixon D R & Pollard D, Embryo abnormalities in the peri- indicator of sub-lethal effects of heavy metals in the winkle, Littorina saxatilis, as indicators of stress in polluted marine environments, Mar Pollut Bull, 16 (1985) 29-33. field. 10 Faller-Fritsch, R J, Reproductive strategies of the winkle Lit- torina rudis in relation to population dynamics and size Acknowledgement structure, in Biology of benthic organisms, edited by B F We are grateful to Candace Browne and Harry Keegan, P O’Ceidigh & P J S Boaden, (Pergamon Press, Ox- Sealy for their assistance in the laboratory. The finan- ford) 1977, pp. 225-231. 11 Roberts J & Hughes R N, Growth and reproduction rates of cial assistance of the Commonwealth Scholarship Littorina rudis form three contrasted shores in North Wales, Commission is acknowledged. U.K., Mar Biol, 58 (1980) 47-54. 12 Janson K, Growth and reproduction rates of Littorina rudis References from three contrasted shores in North Wales, U.K., J Moll Stud, 51 (1985) 64-68. 1 Vranken G & Heip D, Toxicity of copper, mercury and lead 13 Southgate T, Slinn J D & Eastman J F, Mine-derived metal to a marine nematode, Mar Pollut Bull, 17 (1986) 453-457. pollution in the Isle of Man, Mar Pollut Bull, 14 (1983) 2 Verriopoulous G & Hardouvelis D, Effects of sublethal con- 137-140. centration of zinc on survival and fertility in four successive 14 Daka E R, Population differences in the toxic effects of heavy generations of Tsibe, Mar Pollut Bull, 19 (1988) 162-166. metals to Littorina saxatilis Olivi (Prosobranchia: Mol- 3 Leland H V & Kuwabara J S, Trace metals, in Fundamentals lusca), Ph D thesis, University of Liverpool, UK, 1996. of aquatic toxicology, methods and applications, edited by G 15 Posthuma L, Verweij R A, Widianarko B & Zonneveld C, M Rand,. & S R Petrocelli, (Hemisphere, Washington) 1985, Life-history patterns in metal-adapted collembola, Oikos, pp. 374-415. 67 (1993) 235-249. 16 Gibb J O T, Allen J R & Hawkins S J, The application of 4 Hunt J W & Anderson B S, Sublethal effects of zinc and mu- biomonitors for the assessment of mine-derived pollution of nicipal effluents on larvae of the red abalone Haliotis rufes- the West Coast of the Isle of Man, Mar Pollut Bull, cens, Mar Biol, 101 (1989) 545-552. 32 (1996) 512-519. 5 Oshida P S & Word L S, Bioaccumulation of chromium and 17 Proud S V, Tributyltin pollution and the bioindicator Nucella its effects on reproduction in Neanthes arenaceodentata lapillus: Population recovery and community level re- (Polychaeta), Mar Environ Res, 5 (1982) 41-49. sponses, Ph D thesis, University of Liverpool, UK, 1994. 6 Bryan G W, Gibbs P E, Hummerstone L G & Burt G R, The 18 Reid D G, Barnacle-dwelling ecotypes of three British Lit- decline of the gastropod, Nucella lapillus around south-west toina species and the status of Littorina neglecta Bean, England: Evidence for the effect of tributyltin from antifoul- J Moll Stud, 59 (1993) 51-62. ing paints, J Mar Biol Ass UK, 66 (1986) 611-640. 19 Zar J H, Biostatistical analysis (Prentice-Hall Inc., New Jer- 7 Bryan G W & Langston W J, Bioavailability, accumulation sey) 1984, pp. 176. and effects of heavy metals in sediments with special refer- 20 Janson K, Genetic and environmental effects on the rate of ence to United Kingdom estuaries: a review, Environ Pollut, Littorina saxatilis growth, Mar Biol, 69 (1982) 73-78. 76 (1992) 89-131. 21 Bodar C W M, Leeuwen C J V, Voogt P A & Zandee D I, 8 Berry A J, J Some factors affecting the distribution of Lit- Effect of cadmium on the reproductive strategy of Daphnia torina saxatilis (Olivi), Anim Ecol, 30 (1961) 27-45. magna, Aquat Toxicol, 12 (1988) 301-310.