Comparison of parasites of mummichogs and sticklebacks from brackish and freshwater ponds on Sable Island, Nova Scotia.
The impoverished fauna on islands is well documented (MacArthur and Wilson, 1967), and applies also to parasite species richness (Conneely and McCarthy, 1984; Kennedy et al., 1986; Marcogliese and Cone, 1991; Marcogliese, 1992b). While species richness is related to the degree of an island's isolation in time and space (MacArthur and Wilson, 1967), recent findings suggest that intraisland habitat differences are as important as interisland characteristics in determining levels of parasite abundance (Dobson et al., 1992).
Sable Island is an isolated, crescent-shaped sand bar, 35 km long by 1.5 km wide, located 44 [degrees] N 60 [degrees] W, 290 km E of Halifax, Nova Scotia [ILLUSTRATION FOR FIGURE 1 OMITTED]. The closest point on the mainland is near Canso, Nova Scotia, 160 km NW. Numerous brackish and freshwater ponds occur throughout the island. Fishes found therein include mummichogs, Fundulus heteroclitus (L.); blackspotted sticklebacks, Gasterosteus wheatlandi Putnam; threespine sticklebacks, G. aculeatus L.; fourspine sticklebacks, Apeltes quadracus (Mitchill); and ninespine sticklebacks, Pungitius pungitius (L.) (Garside, 1969; Marcogliese, 1992a). Marcogliese (1992b) discussed the parasite fauna of threespine, fourspine and ninespine sticklebacks on Sable Island from a biogeographic perspective. In that study, the parasite fauna of Sable Island fishes was explained in terms of initial colonization events by host fish involving a glacial refugium, and subsequent introduction of parasites by birds and seals. Only parasites with direct life cycles were able to colonize the island with their fish hosts. Those with complex life cycles were introduced by bird or mammalian definitive hosts. The entire parasite fauna of fishes is restricted due to the impoverished free-living fauna and the isolation of the island.
Recently, attention has been directed toward the relative contributions of local vs. regional factors in determining parasite community composition and structure (Aho and Bush, 1994; Hartvigsen and Halvorsen, 1994). Whereas Marcogliese (1992b) discussed the regional factors contributing to parasite species composition in fishes on Sable Island, in this study local processes are examined to account for the distribution of the parasites among fish hosts and habitats on the island. Although impoverished faunas on islands are well-recognized, only limited information exists suggesting the importance of habitat differences on islands in determining parasite abundance (Dobson et al., 1992). Due to its remoteness and limited fish fauna inhabiting the ponds, Sable Island is an excellent location to investigate distribution and abundance of parasites among habitats. In this study, the parasite fauna of the mummichog and the ninespine stickleback are compared between brackish and freshwater habitats on the island, to determine if habitat differences affect parasite abundance and further compromise parasite species richness.
MATERIALS AND METHODS
East Pond is a complex of three large and several small brackish ponds which are inter-connected during winter and periods of high water [ILLUSTRATION FOR FIGURE 1 OMITTED]. Salinity in 1968 was reported to be 16% (Garside, 1969). As measured by a calibrated specific gravity hydrometer, salinity was 25% in 1991 and 10% in 1992. The high salinity in 1991 was probably the result of an extremely dry summer, with little freshwater input and high evaporation rates. Main Pond is composed of a large complex of ponds which are normally interconnected, but may become isolated during dry periods. Lily Pond is a small isolated body of water located in a valley N of Main Pond [ILLUSTRATION FOR FIGURE 1 OMITTED]. Both Main Pond and Lily Pond are freshwater, with a salinity of 0-1% [ILLUSTRATION FOR FIGURE 1 OMITTED]. All ponds are shallow ([less than]1 m), with maximum surface temperatures of 25-28 C. The freshwater ponds are more heavily vegetated than the brackish pond. None of these ponds is visited by seals, but all are frequented by a variety of seabirds and shorebirds.
Ninespine sticklebacks were collected in baited minnow traps from Lily Pond (n = 30) and Main Pond (n = 30) in August 1991 and from East Pond (n = 12) and Lily Pond (n = 17) in August 1992, sampling sites hereafter referred to as LP91, MP91, EP92 and LP92, respectively. Mummichogs were collected in baited minnow traps from East Pond (n = 30) and Main Pond (n = 36) in August 1991, and from East Pond (n = 17) in August 1992, sampling sites hereafter referred to as EP91, MP91 and EP92, respectively. Due to time constraints, the sample of mummichogs from MP91 was frozen for necropsy at a later date. Of the 17 mummichogs from EP92, five were examined completely for parasites, only the alimentary canal being examined in the remainder.
Before necropsy, fish were pithed, measured and weighed. The external surface, gills, viscera, alimentary canal and musculature were examined with a stereomicroscope. Larval parasites found include the nematodes Cosmocephalus obvelatus (Creplin), Paracuaria adunca (Creplin) and Contracaecum sp.; the cestodes Diphyllobothrium ditrimum (Creplin), Schistocephalus pungitii (Dubinina) and Proteocephalus sp.; and the acanthocephalan Neoechinorhynchus cylindratus (Van Cleave). Adult parasites include the monogeneans Gyrodactylus canadensis Hanek and Threlfall and G. avalonia Hanek and Threlfall, and the copepod Thersitina gasterostei (Pagenstechner). Specimens of Diphyllobothrium were fixed live in 10% buffered formalin at room temperature (Andersen et al., 1987). Other cestodes were relaxed in tap water and fixed in 10% buffered formalin. All cestodes were transferred to 10% glycerol in 70% ethanol after 4 wk. Acanthocephalan cystacanths were placed in tap water 12-24 h to induce eversion of the proboscis for identification purposes, and fixed in 10% glycerol in 70% ethanol. Monogeneans were either fixed directly in 10% buffered formalin, or frozen overnight while attached to a piece of host tissue, removed and fixed in 10% glycerol in 70% ethanol. Copepods were fixed in 10% glycerol in 70% ethanol. All nematode juveniles were extremely small and usually heavily encapsulated. They were freed of host tissue by dissection and fixed in hot 10% glycerol in 70% ethanol. Due to their small size and fragile nature, not all nematodes could be identified after being processed. Counts were made of all nematodes combined for each fish. Only those nematodes successfully extracted from capsules were identified. These worms were used to estimate proportions (percent composition) of individual species among the total population of nematodes in a host species. The percent composition of each species was multiplied by the total number of nematodes to derive an estimate of the abundance of each nematode species in each pond. Reliable data were obtained for prevalence and abundance of total nematodes, but only an estimate of abundance was obtained for the individual species. The terms prevalence, abundance and intensity are used in accordance with the definitions outlined in Margolis et al. (1982). Prevalence is the number of infected fish divided by the total number of fish in a sample, expressed as a percent. Abundance is the mean number of parasites per host, including infected and uninfected fish. Intensity is the number of parasites in an individual infected fish. Autogenic species are defined as those which complete their life cycles within the aquatic environment, and allogenic ones are those which use fish or other aquatic vertebrates as intermediate hosts, but require an avian or terrestrial definitive host (Esch et al., 1988).
Mean sizes of fish of each species were statistically compared among ponds using ANOVA (SAS PROC GLM). Spearman's rank correlation was performed on each host species from each pond to determine if individual parasite species abundances were correlated with fish length. If no correlation was found, [log.sub.10](n + 1) transformed abundances of particular parasites within host species were compared statistically across the different ponds and years using one-way analysis of variance (SAS PROC GLM). Student-Newman-Keuls (SNK) multiple-range test was used post hoc to compare mean abundances where significance was detected in the ANOVA. If significant correlations between parasite abundance and length of fish were found in any pond, a one-way analysis of covariance was performed using host length as a covariate to compare [log.sub.10](n + 1) transformed abundances of particular parasites within host species across the different ponds and years, with subsequent pairwise comparisons of standardized means (SAS PROC GLM). Proportions of nematodes within host species were compared across sampling sites using G-tests (Sokal and Rohlf, 1981). Similarity in species composition within each host species was compared across sites with Jaccard's index of community similarity, which is a qualitative measure that does not take abundance into account. Sorenson's index of community similarity, which accounts for abundance of individual species, was used to quantitatively compare the parasite communities within each host species across sites (Magurran, 1988).
Voucher specimens of all parasites were deposited in the Canadian Museum of Nature: from mummichogs (No. CMNP1993-0001 to -0005, -0021), and ninespine sticklebacks (except Gyrodactylus spp.) (No. CMNP19930006 to -0020, NMCC19930034).
Parasites in mummichogs. - The mean length of mummichogs from freshwater MP91 was larger than that from the two brackish ponds EP91 and EP92 (ANOVA, df = 2, F = 16.28, P [less than or equal to] 0.0001). Mummichogs in brackish EP91 and EP92, and freshwater MP91 were infected by the spirurid nematodes Cosmocephalus obvelatus and Paracuaria adunca, the anisakid nematode Contracaecum sp. and the acanthocephalan Neoechinorhynchus cylindratus. The cestode Proteocephalus sp. was also found in EP92. Prevalence, abundance and intensity range are presented in Table 1.
Nematodes were usually encapsulated on the viscera, and occurred free in the body cavity only rarely. Total nematode abundance was positively correlated with host length in EP91 and EP92 ([r.sub.s] = 0.91 and 1.00, P [less than or equal to] 0.0001, respectively). Combined nematode abundance was statistically higher in mummichogs from brackish EP91 and EP92 than in those from freshwater MP91 (ANCOVA, df = 2, F = 95.41, P [less than or equal to] 0.0001). Cosmocephalus obvelatus was the most common of the three nematodes in EP91-92 mummichogs, whereas Paracuaria adunca was more common in mummichogs from MP91. Contracaecum sp. was the rarest of the three nematodes in mummichogs from all ponds (Table 1). Proportions of the nematode species were not independent of sampling sites (G-test, P [less than] 0.005).
Neoechinorhynchus cylindratus was most prevalent in mummichogs from EP91-92, being rare in MP91 (Table 1). Abundances of N. cylindratus were positively correlated with host length in MP91 ([r.sub.s] = 0.23, P [less than] 0.05) and EP91 ([r.sub.s] = 0.65, P [less than or equal to] 0.0001). The abundances of N. cylindratus from MP91, EP91 and EP92 were significantly different (ANCOVA, df = 2, F = 62.51, P [less than or equal to] 0.0001). All N. cylindratus were cystacanths in the liver.
The cestode Proteocephalus sp. occurred as plerocercoids or immature stages inhabiting the posterior region of the intestine. It was not observed in mummichogs from MP91 and EP91, but was prevalent in those from EP92 (Table 1), where its abundance was negatively correlated with host length ([r.sub.s] = -0.89, P [less than or equal to] 0.0001).
Jaccard's index of community similarity showed MP91 and EP91 to be identical, with EP92 being very similar to them. Sorenson's index showed that EP91 and EP92 are more similar to each other than they are to MP91 (Table 2).
Parasites in ninespine sticklebacks. - The mean length of sticklebacks from freshwater MP91 was smaller than those from all other sites (ANOVA, df = 2, F = 14.42, P [less than or equal to] 0.0001). In all ponds ninespine sticklebacks were infected by Cosmocephalus obvelatus, Paracuaria [TABULAR DATA FOR TABLE 1 OMITTED] adunca, Contracaecum sp., the cestode Diphyllobothrium ditremum and the mongeneans Gyrodactylus spp. (including G. canadensis and G. avalonia). The cestode Schistocephalus pungitii was only found in the freshwater ponds (MP91, LP91-92). The copepod Thersitina gasterostei and Proteocephalus sp. only infected ninespine sticklebacks in EP92 and MP91. Prevalence, abundance and intensity range are presented in Table 3.
TABLE 2. - Jaccard's index of community similarity based on presence of parasite species, and Sorenson's index of community similarity based on abundance of parasite species in Fundulus heteroclitus in Sable Island ponds
Pond MP91 EP91 EP92
MP91 1.00 1.00 0.80 EP91 1.00 0.80 EP92 1.00
MP91 1.00 0.30 0.11 EP91 1.00 0.77 EP92 1.00
Nematodes were significantly more abundant in ninespine sticklebacks from LP91 and EP92 than in those from MP91 and LP92 (ANOVA, df = 3, F = 14.26, P [less than or equal to] 0.0001; SNK). As in mummichogs, Cosmocephalus obvelatus was the most common nematode in ninespine sticklebacks in EP92. This same parasite was predominant among the nematodes in sticklebacks from LP91, but in LP92 Paracuaria adunca was present in similar proportions. Contracaecum sp., the rarest nematode in EP92 and LP91-92, was predominant in sticklebacks from MP91 (Table 3). The percent composition of nematodes was not independent of sampling site (G-test, P [less than] 0.005). Estimated abundance of each nematode in ninespine sticklebacks was much lower than in mummichogs from the same pond, with the exception of Contracaecum sp. in MP91 (Tables 1, 3).
Proteocephalus sp. infected sticklebacks in MP91, but was not found in those from LP91 or LP92 (Table 3). It was significantly more abundant in sticklebacks from EP92 (ANOVA, df = 3, F = 43.46, P [less than or equal to] 0.0001; SNK).
Diphyllobothrium ditremum was much more prevalent and occurred at higher intensities in ninespine sticklebacks from LP92 than in any other site (Table 3). Abundance of D. ditremum was significantly higher in LP92 than in LP91, MP91 and EP92 (ANOVA, df = 3, F = 29.81, P [less than or equal to] 0.0001; SNK). All parasites were plerocercoid stages inhabiting the liver and body cavity.
Highest prevalence and intensities of Schistocephalus pungitii in ninespine sticklebacks occurred in LP92 (Table 3). Abundance of this cestode in LP92 was significantly greater than in LP91, MP91 and EP92 where it was not found (ANOVA, df = 3, F = 50.30, P [less than or equal to] 0.0001; SNK). Parasites were in the plerocercoid stage and inhabited the body cavity.
Abundance of Thersitina gasterostei was positively correlated with fish length in EP92 ([r.sub.s] = 0.47, P [less than] 0.05). Highest infection levels occurred in ninespine sticklebacks from EP92, being rare in those from MP91 and absent from LP91-92 (Table 3; ANCOVA, df = 3, F = 982.51, P [less than or equal to] 0.0001). Copepods were found primarily on the gills and opercula, and occasionally on the fins of heavily infected fish.
Gyrodactylus avalonia and G. canadensis were present on the gills and fins of ninespine [TABULAR DATA FOR TABLE 3 OMITTED] sticklebacks from all ponds sampled. Abundances and prevalences were similar across sampling sites (Table 3).
TABLE 4. - Jaccard's index of community similarity based on presence of parasite species and Sorenson's index of community similarity based on abundance of parasite species in Pungitius pungitius in Sable Island ponds
Pond MP91 LP91 LP92 EP92
MP91 1.00 0.75 0.75 0.75 LP91 1.00 1.00 0.71 LP92 1.00 0.71 EP92 1.00
MP91 1.00 0.39 0.16 0.06 LP91 1.00 0.31 0.15 LP92 1.00 0.07 EP92 1.00
Jaccard's index of community similarity showed MP91 was equally similar to LP91, LP92 and EP92. LP91 has the same species composition as LP92, and the lowest values were between LP91-92 and EP92. Sorenson's index was low between all ponds. The highest values occur between LP91 and MP91, and LP91 and LP92, the lowest between EP92 and MP91, and EP92 and LP92 (Table 4).
On Sable Island, Cosmocephalus obvelatus, Paracuaria adunca, Contracaecum sp., Diphyllobothrium ditremum, Schistocephalus pungitii, Gyrodactylus canadensis and Thersitina gasterostei were reported from ninespine sticklebacks inhabiting Main Pond (Marcogliese, 1992b). Parasites not previously observed from sticklebacks on Sable Island include the monogenean G. avalonia and the cestode Proteocephalus sp. Mummichogs on Sable Island have not been surveyed previously for parasites. They were infected with Cosmocephalus obvelatus, Paracuaria adunca, Contracaecum sp., Proteocephalus sp. and Neoechinorhynchus cylindratus. Infections in mummichogs with both spirurids are new host records, and that of N. cylindratus a new Canadian host record.
Comparisons of parasite faunas and their abundances in the same host species from different habitats indicate environmental effects on both host and parasite. Because most parasites occurred in all ponds, Jaccard's index of community similarity is high for all pair-wise comparisons in both mummichogs and sticklebacks. Only one species was confined to either brackish or freshwater ponds, that being Schistocephalus pungitii in the latter. Sorenson's index of community similarity showed most pairwise comparisons within the host fishes are low, reflecting both temporal and spatial variations in abundance, because the species composition of the ponds are similar. Many parasites, including Proteocephalus sp., Cosmocephalus obvelatus, Paracuaria adunca, Neoechinorhynchus cylindratus and Thersitina gasterostei, were more common in brackish East Pond than in the freshwater ponds. However, temporal variability was also high, as reflected by the low similarity of LP91 and LP92 sticklebacks.
Abundances of the spirurid nematodes varied spatially among ponds and temporally within ponds (LP91-92), but no clear distributional patterns related to habitat were evident except that they were low in MP91 and high in EP92 in both fish hosts. The temporal variation observed in Lily Pond was due to fluctuations in abundance of Cosmocephalus obvelatus. The spirurids were much more abundant in mummichogs than in sticklebacks from the same ponds, which is attributed to the mummichog's larger size, and, presumably, much greater consumption of the crustacean prey which serve as intermediate hosts. Contracaecum sp. was relatively rare in all hosts from all areas. Like the spirurids, this nematode matures in piscivorous birds.
In contrast to the nematodes, Diphyllobothrium ditremum and Schistocephalus pungitii were most common in the ninespine sticklebacks from LP92. Both parasites use cyclopoid copepods as intermediate hosts and piscivorous birds as definitive hosts. The differences in abundances of the two cestodes between LP91 and LP92 reflects the temporal instability which also occurs among other parasite populations in Sable Island ponds. Schistocephalus pungitii was absent from EP92, which is expected, as members of the family Ligulidae are found exclusively in freshwater fish (Dubinina, 1966). Neither cestode was found in the mummichogs, which is not surprising as D. ditremum is confined to sticklebacks, salmonids and osmerids (Andersen et al., 1987), while S. pungitii is host-specific to ninespine sticklebacks (Dubinina, 1966).
The presence of plerocercoids of Proteocephalus sp. and cystacanths of Neoechinorhynchus cylindratus contradicts the conclusion (Marcogliese, 1992b) that the only autogenic parasites on Sable Island are those with direct life cycles. Given that no adult parasites of these two species were seen in either sticklebacks or mummichogs, they probably mature in another fish species. Some species of Proteocephalus can survive for a short time in the digestive system of atypical fish hosts (Scholz, 1991), whereas N. cylindratus encysts in the liver of a variety of paratenic hosts (Schmidt, 1985). No fishes other than sticklebacks and mummichogs were captured on the island despite intensive trapping, seining and deployment of lines with baited hooks. However, large American eels (Anguilla rostrata) were observed in Main Pond (Garside, 1969; B. Beck, Dep. Fisheries and Oceans; G. Forbes, Environment Canada, pers. comm.) on a number of occasions. American eels are infected with P. macrocephalus and other unidentified species in the genus (Margolis and Arthur, 1979), and adult N. cylindratus (Ward, 1940; Bullock, 1970). Conceivably, eels may occasionally visit Sable Island during their breeding migration from mainland fresh waters to the Sargasso Sea, periodically seeding the ponds with freshwater parasites imported from the mainland. These populations may be subjected to periodic extinctions and recolonizations. During a survey of Sable Island fishes for protozoan parasites in 1994, no acanthocephalans were observed in 15 mummichogs from East Pond nor in 15 from Main Pond (Marcogliese, pers. observ.). Until the definitive hosts are verified, the presence of both Proteocephalus sp. and N. cylindratus remains problematic.
Proteocephalus sp. and Neoechinorhynchus cylindratus were much more common in EP92 than in MP91. It is unlikely that differences in development of Proteocephalus sp. from those of the other two cestodes can explain its distribution, because all use cyclopoid copepods as an intermediate host. However, East Pond is more accessible to eels than is Main Pond, possibly explaining its observed distribution, and that of N. cylindratus.
Prey spectra may differ between the brackish and freshwater habitats. Both mummichogs and ninespine sticklebacks consume insects and crustaceans among other prey (Scott and Scott, 1988), but insects are absent from the brackish ponds on Sable Island (Wright, 1989). This implies that brackish fish eat relatively more crustaceans than do freshwater fish. Such differences could affect transmission of parasites in subtle ways not completely discernible from the data. For example, the parasites using benthic crustaceans as intermediate hosts, Cosmocephalus obvelatus, Paracuaria adunca and Neoechinorhynchus cylindratus (Ward, 1940; Anderson and Wong, 1982; Wong and Anderson, 1982; Marcogliese, 1992c), were more common in brackish EP91-92 than in freshwater MP91.
Gyrodactylus canadensis and G. avalonia are euryhaline parasites (Cone and Wiles, 1985), and occurred in similar abundances on ninespine sticklebacks in all ponds. In contrast, the euryhaline copepod Thersitina gasterostei was abundant in EP92, but relatively rare in MP91 and absent from LP91-92. Salinity is proposed to be a limiting factor in the distribution of T. gasterostei, as it apparently is not found in waters of salinity less than 0.5% (Walkey et al., 1970). These authors further suggest that an abrupt salinity threshold exists, below which the copepod cannot survive. The presence of T. gasterostei in MP91 does not support this conjecture. The concept of a salinity threshold also contradicts natural records of the copepod's distribution, as it has been reported from fresh water by Threlfall (1968) in Newfoundland and Levsen (1992) in Norway. Its very low abundance in MP91 and absence from LP91-92, but high levels in EP92, indicate that this parasite does fare better under brackish conditions.
Habitat differences did not greatly affect species richness in the two fish hosts. Only one parasite (Schistocephalus pungitii) was confined to fresh waters, and none to the brackish pond. However, habitat apparently contributed greatly to differences in abundances of the various parasites among ponds. Similarly, Dobson et al. (1992) found pronounced differences in abundances of lizard parasites among habitats on islands, which they speculated was due to differential survival of free-living stages of parasites. On Sable Island, salinity effects were manifested directly on the abundances of S. pungitii and Thersitina gasterostei, and possibly indirectly on Cosmocephalus obvelatus, Paracuaria adunca and Neoechinorhynchus cylindratus by influencing the species composition of the invertebrate fauna. Other habitat differences, such as accessibility, may have played a role in the different abundances of N. cylindratus and Proteocephalus sp. among brackish and freshwater ponds. Superimposed on these effects is a temporal instability among the parasite populations, which could result from the uneven distribution of avian definitive hosts or the unstable conditions of a harsh, isolated environment.
AHO, J. M. AND A. O. BUSH. 1994. Community richness in parasites of some freshwater fishes from North America, p. 185-193. In: R. E. Ricklefs and D. Schluter (eds.). Species diversity in ecological communities. University of Chicago Press, Chicago.
ANDERSEN, K., H. L. CHING AND R. VIK. 1987. A review of freshwater species of Diphyllobothrium with redescriptions and the distribution of D. dendriticum (Nitzsch, 1824) and D. ditremum (Creplin, 1895) from North America. Can. J. Zool., 65:2216-2228.
ANDERSON, R. C. AND P. L WONG. 1989. The transmission and development of Paracuaria adunca (Creplin, 1846) (Nematoda: Acuarioidea) of gulls (Laridae). Can. J. Zool., 60:3092-3104.
BULLOCK, W. L. 1970. The zoogeography and host relations of the eoacanthocephalan parasites of fishes, p. 161-173. In: S. F. Snieszko (ed.). A symposium on diseases of fishes and shellfishes. Spec. Publ. No. 5. Am. Fish. Soc., Washington, D.C.
CONE, D. K. AND M. WILES. 1985. The systematics and zoogeography of Gyrodactylus species (Monogenea) parasitizing gasterosteid fishes in North America. Can. J. Zool., 63:956-960.
CONNEELY, J. J. AND T. K. MCCARTHY. 1984. The metazoan parasites of freshwater fishes in the Corrib catchment area, Ireland. J. Fish Biol., 24:363-375.
DOBSON, A. P., S. W. PACALA, J. D. ROUGHGARDEN, E. R. CARPER AND E. A. HARRIS. 1992. The parasites of Anolis lizards in the northern Lesser Antilies. I. Patterns of distribution and abundance.
DUBININA, M. N. 1966. Tapeworms (Cestoda, Ligulidae) of the Fauna of the USSR. Moscow and Leningrad: Izdatelstvo Nauka. (English translation by Amerind Publishing Co. Pvt. Ltd., New Delhi, 1980.) 320 p.
ESCH, G. W., C. R. KENNEDY, A. O. BUSH AND J. M. AHO. 1988. Patterns in helminth communities in freshwater fish in Great Britain: alternative strategies for colonization. Parasitology, 96:519-532.
GARSIDE, E. T. 1969. Distribution of insular fishes of Sable Island, Nova Scotia. J. Fish. Res. Board Can., 26:1390-1392.
HARTVIGSEN, R. AND O. HALVORSEN. 1994. Spatial patterns in the abundance and distribution of parasites of freshwater fish. Parasitol. Today, 10:28-31.
KENNEDY, C. R., D. D'A. LAFFOLEY, G. BISHOP, P. JONES AND M. TAYLOR. 1986. Communities of parasites of freshwater fish of Jersey, Channel Islands. J. Fish Biol., 29:215-226.
LEVSEN, A. 1992. Parasites of the three-spined stickleback Gasterosteus aculeatus from Norway, with emphasis on trichodinid ciliates. Fauna, 45:40-48. (In Norwegian with English summary.)
MACARTHUR, R. H. AND E. O. WILSON. 1967. The theory of island biogeography. Princeton University Press, Princeton, N.J. 203 p.
MAGURRAN, A. E. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, N.J. 179 p.
MARCOGLIESE, D. J. 1992a. First report of the threespine stickleback, Gasterosteus aculeatus, from Sable Island. Can. Field-Nat., 106:264-266.
-----. 1992b. Metazoan parasites of sticklebacks on Sable Island, Northwest Atlantic Ocean: biogeographic considerations. J. Fish Biol., 41:399-407.
-----. 1992c. Neomysis americana (Crustacea: Mysidacea) as an intermediate host for sealworm, Pseudoterranova decipiens (Nematoda: Ascaridoidea), and spirurid nematodes (Acuarioidea). Can. J. Fish. Aquat. Sci., 49:513-515.
----- AND D. K. CONE. 1991. Do brook charr (Salvelinus fontinalis) from insular Newfoundland have different parasites than their mainland counterparts? Can. J. Zool., 69:809-811.
MARGOLIS, L. AND J. R. ARTHUR. 1979. Synopsis of the parasites of fishes of Canada. Bull. Fish, Res. Board Can. No. 199. 269 p.
-----, G. W. ESCH, J. C. HOLMES, A. M. KURIS AND G. A. SCHAD. 1982. The use of ecological terms in parasitology (report of an ad hoc committee of the American Society of Parasitologists). J. Parasitol., 68:131-133.
SCHMIDT, G. D. 1985. Development and life cycles, p. 273-305. In: D. W. T. Crompton and B. B. Nickol (eds.). Biology of the Acanthocephala. Cambridge Univ. Press, Cambridge, U.K.
SCHOLZ, T. 1991. Studies on the development of the cestode Proteocephalus neglectus La Rue, 1911 (Cestoda: Proteocephaliidae) under experimental conditions. Fol. Parasitol., 38:39-55.
SCOTT, W. B. AND M. G. SCOTT. 1988. Atlantic fishes of Canada. Can. Bull. Fish. Aquat. Sci. No. 219. 731 p.
SOKAL, R. R. AND F. J. ROHLF. 1981. Biometry, 2nd ed. W. H. Freeman and Co., San Francisco, Calif. 859 p.
THRELFALL, W. 1968. A mass die-off of three-spined sticklebacks (Gasterosteus aculeatus L.) caused by parasites. Can. J. Zool., 46:105-106.
WALKEY, M., D. B. LEWIS AND H. J. G. DARTNALL. 1970. Observations on the host-parasite relations of Thersitina gasterostei (Crustacea: Copepoda).J. Zool. Lond., 162:371-381.
WARD, H. L. 1940. Studies on the life history of Neoechinorhynchus cylindratus (Van Cleave, 1913) (Acanthocephala). Trans. Am. Microsc. Soc., 59:327-347.
WONG, P. L. AND R. C. ANDERSON. 1982. The transmission and development of Cosmocephalus obvelatus (Nematoda: Acuarioidea) of gulls (Laridae). Can. J. Zool., 60:1426-1440.
WRIGHT, B. 1989. The fauna of Sable Island. Curatorial rep. No. 68. Nova Scotia Museum, Halifax. 93 p.