Spatial and temporal distributions of larval seal-worm , Pseudoterranova decipiens ( Nematoda : Anisakinae ) , in Hippoglossoides platessoides ( Pleuronectidae ) in the Canadian Maritime Region from 1993 to 1999

Spatial and temporal trends of larval sealworm ( Pseudoterranova decipiens ) infection in eastern Canadian groundfish were monitored in an indicator host, Canadian plaice ( Hippoglossoides platessoides ), in the 31 to 40 cm length range. Between February 1993 and September 1999, a total of 8,482 plaice were collected from 33 locations in Canadian Maritime waters (NAFO Subdivisions 4TVWX-5ZE), and their fillets and napes were examined for sealworm. Prevalence (P) and abundance (A) of the parasite were greatest (P ranging from 95 to 100%, A from 7.48 to 15.60) in fish collected from the central Scotian shelf (4VSW) near Sable Island, site of the largest grey seal (Halichoerus grypus ) colony in the northwest Atlantic, and from Jordan Basin in the northeastern Gulf of Maine (4X). The infection of greatest intensity (I=158) occurred in a fish from “The Gully” slopewaters of Banquereau (4VS), a few kilometres northeast of Sable Island. By 1995-99, sealworm prevalence and/or abundance had increased significantly in plaice from most locations where stable or declining infection parameters were observed from 1989 to 1993, but abundance of the parasite continued to decline in the Sable Island area. While spatial and temporal distributions of larval sealworm in plaice seemed largely related to the distribution and growth of grey seal populations, the influence of definitive hosts was probably mitigated by other factors such as changes in environmental temperature and parasite density limiting effects in the indicator host.

A AB BS ST TR RA AC CT T Spatial and temporal trends of larval sealworm (Pseudoterranova decipiens) infection in eastern Canadian groundfish were monitored in an indicator host, Canadian plaice (Hippoglossoides platessoides), in the 31 to 40 cm length range.Between February 1993 and September 1999, a total of 8,482 plaice were collected from 33 locations in Canadian Maritime waters (NAFO Subdivisions 4TVWX-5ZE), and their fillets and napes were examined for sealworm.Prevalence (P) and abundance (A) of the parasite were greatest (P ranging from 95 to 100%, A from 7.48 to 15.60) in fish collected from the central Scotian shelf (4VSW) near Sable Island, site of the largest grey seal (Halichoerus grypus) colony in the northwest Atlantic, and from Jordan Basin in the northeastern Gulf of Maine (4X).The infection of greatest intensity (I=158) occurred in a fish from "The Gully" slopewaters of Banquereau (4VS), a few kilometres northeast of Sable Island.By 1995-99, sealworm prevalence and/or abundance had increased significantly in plaice from most locations where stable or declining infection parameters were observed from 1989 to 1993, but abundance of the parasite continued to decline in the Sable Island area.While spatial and temporal distributions of larval sealworm in plaice seemed largely related to the distribution and growth of grey seal populations, the influence of definitive hosts was probably mitigated by other factors such as changes in environmental temperature and parasite density limiting effects in the indicator host.McClelland, G. and Martell, D.J. 2001.Spatial and temporal distributions of larval sealworm, Pseudoterranova decipiens (Nematoda: Anisakinae), in Hippoglossoides platessoides (Pleuronectidae) in the Canadian Maritime Region from 1993 to 1999.NAMMCO Sci.  S Sp pa at ti ia al l a an nd d t te em mp po or ra al l d di is st tr ri ib bu ut ti io on ns s o of f l la ar rv va al l s se ea al l-w wo or rm m, , P Ps se eu ud do ot te er rr ra an no ov va a d de ec ci ip pi ie en ns s ( (N Ne em ma at to od da a: : A An ni is sa ak ki in na ae e) ), , i in n H Hi ip pp po og gl lo os ss so oi id de es s p pl la at te es ss so oi id de es s ( (P Pl le eu ur ro on ne ec ct ti id da ae e) ) i in n t th he e C Ca an na ad di ia an n M Ma ar ri it ti im me e R Re eg gi io on n f fr ro om m 1 19 99 93 3 t to o 1 19 99 99 9 T he "sealworm index" time series was established to monitor spatial and temporal trends of larval sealworm (Pseudoterranova decipiens) infection in eastern Canadian groundfish, through surveys of a single "indicator" host (McClelland et al. 2000).Although Atlantic cod (Gadus morhua) has been frequently surveyed for larval sealworm in the North Atlantic (Scott and Martin 1957, Templeman et al. 1957, Young 1972, McClelland et al. 1983a, 1983b, 1985, Platt 1975, Hauksson 1984, Brattey et al. 1990, McClelland and Marcogliese 1994, des Clers and Andersen 1995, Boily and Marcogliese 1995), Hippoglossoides platessoides, commonly known as Canadian (McAllister 1990) or American plaice (retained by Robins et al. 1991) was chosen as the indicator host following the rationale of McClelland et al. (1983bMcClelland et al. ( , 1985McClelland et al. ( , 2000)).
Between June 1983 and October 1990, index surveys were conducted biennially throughout eastern Canada, from Labrador (NAFO subdivision 2J) to the Gulf of Maine (4X-5ZE) (McClelland et al. 1985(McClelland et al. , 1987(McClelland et al. , 2000)).Samples were taken from 38 to 55 locations in the 1983-84, 1985-86, 1987-88 and 1989-90 surveys.Also included in the time series were earlier  samples from 11 locations in the southern Gulf of St. Lawrence (4T), and on the Breton (4VN), and central and eastern Scotian shelves (4VSW) (McClelland et al. 1983a(McClelland et al. , 1983b)).Prevalence (P) and abundance (A) of P. decipiens were greatest (P ranging from 96 to 100%, A from 12.47 to 22.32) in 1989-90 samples from the central Scotian Shelf near Sable Island (4W), while plaice from northeastern Newfoundland and the Grand Banks (3KLNO) were seldom infected (P<1%) (McClelland et al. 2000).Temporal trend analyses revealed that, between 1980 and 1990, sealworm infection levels increased significantly in plaice from 33 of 41 locations in the Gulf of St. Lawrence, on the Breton and Scotian shelves, and in the Gulf of Maine.Spatial and temporal distributions of larval P. decipiens in plaice seemed to reflect the distribution and growing abundance of grey seals (Halichoerus grypus), important definitive hosts of the parasite, but may also have been influenced by relatively warm near-bot-tom temperatures prevalent in eastern Canada during the late seventies and early eighties (McClelland et al. 2000).
As a consequence of declining plaice stocks and changing research priorities, it became increasingly difficult to obtain samples from Labrador, northeastern and southern Newfoundland, and the northern Gulf of St. Lawrence, and, since 1990, surveys were confined to the Canadian Maritime region (4TVWX-5ZE).The present study documents infection parameters of larval sealworm in Canadian plaice sampled in the Maritimes between February 1993 and September 1999 along with, for the sake of continuity, 1987-90 data from McClelland et al. (2000).Changes in sealworm prevalence and abundance since the late 1980s are analysed, and biotic and abiotic factors that may have influenced infection parameters in our indicator host are also discussed.

M MA AT TE ER RI IA AL LS S A AN ND D M ME ET TH HO OD DS S
Canadian plaice were sampled from the 'Alfred E. Needler' during Department of Fisheries and Oceans (DFO) groundfish surveys conducted in the southern Gulf of St. Lawrence (NAFO subdivision 4T) in September 1995September , 1998September and 1999.Samples were also collected on the Breton (4VN) and Scotian (4VS-W-X) shelves, and in the Gulf of Maine (4X-5ZE) from groundfish cruises in February, March and July of 1993and 1996, and September 1995, and during dedicated surveys of 4VWX plaice in August 1993and October 1996 (Fig.1).Samples consisted primarily of fish in the 31-40 cm length range (n = 7,158), but "extended" samples, including fish <30, and ≥41 cm total length (TL) (n = 1,324), were taken at six 4VS-W locations in 1996, and from Georges Bank (5ZE) in 1993 and 1996.Plaice were bagged, labelled, frozen at sea, and stored in a -17°C walk-in freezer on return to the Halifax Fisheries Research Laboratory, Halifax, Nova Scotia.Prior to inspection, the fish were allowed to thaw for 24 h at room temperature (18 to 22°C).After TL, weight, and sex were recorded, each fish was boned, and fillets, napes (hypaxial musculature enclosing the body cavity) and flesh which remained on the frame were examined for nematodes by slicing, candling and mechanical destruction of tissues (Power 1961).As a rule, nematodes were identified by eye and counted, but inspection with a dissecting microscope was sometimes necessary in order to identify smaller worms and specimens damaged during boning and slicing of flesh.Because larval sealworm is seldom found in the body cavity of plaice (McClelland et al. 1983a), the viscera were not inspected.It was assumed that all sealworm larvae encountered in this survey were P. decipiens sibling species B, or, as recently proposed by Paggi et al. (2000), P. decipiens (sensu stricto).While both P. decipiens species B and C occur in pinnipeds and fish in eastern Canada (Paggi et al. 1991), species C, or P. bulbosa (Mattiucci et al. 1998), has been reported only from the more northerly waters of Labrador and northeastern Newfoundland (Brattey and Davidson 1996).Moreover, the larvae of species C (P. bulbosa) are found primarily in the body cavity of fish hosts (Bristow and Berland 1992).
Quantitative parameters of infection such as prevalence (P), abundance (A), intensity (I), and density (D) are defined according to Margolis et al. (1982).Data from the 1989-90, and, for some locations, from the 1987-88 survey were included in analyses of temporal variations of infection parameters.Variations in prevalence and abundance were analysed by one-way ANOVA (SYSTAT) of time series samples nested within location (McClelland et al. 1987(McClelland et al. , 2000)).In ANOVA of prevalence, individual infected fish were assigned the value "1", and each uninfected fish, the value "0" (Li 1964, Neter et al. 1985).To permit ANOVA of abundance, frequency distributions of seal- worm counts, which were positively skewed to varying degrees, were brought closer to normality by a log (n + 1) transformation (Sokal andRohlf 1969, Platt 1975).Significance of F values for contrasts were tested at 1% Bonferroni probability (Morrison 1976).

R RE ES SU UL LT TS S
Between February 1993 and October 1996, filets and napes of 8,482 Canadian plaice, predominantly in the 31 to 40 cm length range (n = 6,892), were examined for larval sealworm.Plaice examined included 2,549 sampled from the Breton (4VN) and Scotian shelves (4VS-WX), and Gulf of Maine (4X-5ZE) in 1993; 4,427 collected from southern Gulf of St. Lawrence (4T) and 4VWX-5ZE in 1995-96, and 1506 taken from 4T groundfish cruises in September of 1998 and 1999.

Southern Gulf of St. Lawrence (4T)
In 1995, 1,411 plaice in the 31 to 40 cm length range were collected from 12 locations in the southern Gulf of St. Lawrence.Southeastern Gulf fish were generally more heavily infected than those from southwestern 4T, and the most heavily infected fish overall (P ranging from 87 to 97%, A, from 2.73 to 6.94) were found off southeastern Prince Edward Island (PEI), southwestern Cape Breton, and the St. Georges Bay area of mainland Nova Scotia (Table 1).Contrasts of infection parameters in 1995 and 1989 (or 1987-88) samples from 10 southern Gulf locations reveal that sealworm prevalence and abundance were significantly greater in the more recent samples of plaice from Baie des Chaleurs, the Magdelen Shoals, St. Georges Bay and Chéticamp, and that there was also a significant increase in abundance of the parasite in Bradelle Bank fish.Temporal disparities in infection levels in plaice from Banc Miscou, the north Shediac Valley, Port Hood and the Cape Breton Channel were not significant, while declines in prevalence and abundance were confined to fish from the Gaspé Peninsula.
A total of 1,506 plaice were sampled from 13 4T locations in 1998-99.While the heaviest infections occurred in samples from St. Georges Bay, southeastern PEI and southwestern Cape Breton, infection levels in southwestern Gulf plaice (P = 82 to 92%; A = 2.10 to 3.15) now rivalled those found in fish from the southeastern Gulf (P = 87 to 97%; A = 2.27 to 6.94).Contrasts of 1995 and 1998-99 samples at 11 locations reveal that infection levels had increased significantly at five southwestern Gulf locations, including Gaspé, Baie des Chaleurs, North Shediac Valley, northwestern PEI. and Bradelle Bank, while the only significant change observed at southeastern Gulf sites between 1995 and 1998-99 was an increase in sealworm abundance in plaice from St. Georges Bay.Contrasts of 1998Contrasts of -99 and 1989Contrasts of (or 1987-88) -88) samples, however, indicate that infection parameters were significantly greater in the more recent samples at four of five southeastern, and five of six southwestern Gulf locations.
A total of 1,053 plaice (31 to 40 cm TL) were taken from the same 9 sites in 1995-96.Again, infection parameters were greatest (P = 85 to 100%; A = 4.49 to 12.43) in Smoky Channel and Banquereau fish, and the most intense infection (I = 158) was found in a plaice taken from "The Gully" slopewaters of Banquereau.Plaice from Misaine Bank had the lightest infections (P = 57%; A = 1.35).1989, 1993 and 1995-96 samples reveals significant variation in prevalence and/or abundance at six of 9 locations in 4V (Table 2).Abundance of larval P. decipiens in plaice from the Smoky Channel (4VN), east Misaine Channel, and central and western Banquereau (4VS) increased between 1989 and 1995-96.Infection parameters in southwestern Breton Shelf plaice fell significantly in 1993, but recovered in 1995.A net decline in sealworm parameters between 1989 and 1996 was evident only in samples from Louisbourg Hole.

Central and western Scotian Shelf and Gulf of Maine (4WX-5ZE)
Samples from 4WX-5ZE included 1,222 plaice (31 to 40 TL) collected from 9 locations in 1993, and 793 taken from eight sites in 1996.In 1993, the heaviest infections (P = 94 to 97%; A = 7.48 to 13.37) occurred in fish from the vicinity of Sable Island (4W), and the lightest (P = 32 to 40%; A = 0.59 to 0.64), in Roseway Basin (4X) and Georges Bank (5ZE) fish (Table 3).In 1996, heavily infected fish (P = 89 to 100%; A = 7.28 to 15.60) were found, not only in plaice from the Sable Island area, but also in those from Jordan Basin in the northeastern Gulf of Maine (4X).The most intense infection (I = 92) occurred in a Jordan Basin fish.
ANOVA of infection parameters in 1989-90, 1993 and 1996 samples from 4WX-5ZE indi-cates significant change in sealworm prevalence and/or abundance at 7 of 10 locations (Table 3).While infection parameters in plaice from banks on the central Scotian Shelf (4W) declined, significant increases in sealworm abundance were observed in fish from deeper waters in Canso Hole (4W), and Roseway and Jordan Basins (4X) between 1993 and 1996.Infection parameters in Georges Bank (5ZE) plaice collapsed in 1993, but rebounded in 1996.

Extended samples
Samples taken from six 4VWX locations in 1996 were "run of the catch", including additional plaice (n = 1,037), in the 5 to 30 cm length range.Plaice in the 8 to 30 and 41 to 61 cm length ranges were also taken with "index" samples from Georges Bank (5ZE) in 1993 (n = 154) and 1996 (n = 133).Sealworm prevalence and abundance increased with host length in fish from each of three 4V locations, Louisbourg Hole, Misaine Bank, and east Banquereau, as well as in Roseway Basin (4X) fish (Fig. 2).Notably, sealworm were not found in fish ranging from 6 to 30 cm in length (n = 127), from a single set (no. 97, July 1996 Groundfish Cruise) on the western half of Misaine Bank.In fish of "index" size (31 to 40 cm in length) from the same set, sealworm prevalence of was 47%, and abundance, 1.00, (n = 30), and 8 of 10 fish (A = 2.30) in the 36 to 40 cm length range were infected.In contrast, sealworm prevalence and abundance in Middle Bank plaice peaked in 16 to 25 cm fish, and there were no consistent relationships between infection parameters and host length in Georges Bank samples.

D DI IS SC CU US SS SI IO ON N
The spatial and temporal distributions of larval Pseudoterranova decipiens in eastern Canadian groundfish seem to be related, primarily, to the geographical distribution and growth of Northwest Atlantic grey seal populations.Sealworm infection parameters recorded in Canadian plaice and other groundfish species over the past two decades (McClelland et al. 1983a, 1983b, 1985, 1987, 2000, Boily and Marcogliese 1995, Marcogliese 1996, Marcogliese 2001, McClelland and Martell 2001) were greatest in fish from southern Newfoundland (3P), the Gulf of St. Lawrence (4RST), the Breton and Scotian Shelves (4VWX), and the northeastern Gulf of Maine (4X-5ZE) (Fig. 1).These areas encompass the majority of grey seal breeding and haulout sites in eastern Canada (Stobo et al. 1990).By 1980-84, sealworm infection parameters in 4RS (Northern Gulf of St. Lawrence), 4VW and 4X-5ZE cod and flatfishes (McClelland et al. 1983a(McClelland et al. , 1983b(McClelland et al. , 1985) ) were higher than those documented three decades earlier ( Prior to 1990, the most dramatic increases in larval sealworm parameters occurred in groundfish from the central and eastern Scotian Shelf (4VW) near Sable Island (McClelland et al. 1983b(McClelland et al. , 1990(McClelland et al. , 2000) ) (Fig. 3).This may be attributable to the fact that the Sable Island component of the eastern Canadian grey seal population grew by more than 20 fold between the early 1960s and 1988, whereas the remainder of the population, partly as a consequence of a DFO bounty program and a cull/controlled commercial hunt, is estimated to have increased by only two-to five-fold (Zwanenburg and Bowen 1990).Zwanenburg and Bowen (1990) speculate that, following the cessation of the annual cull of ice-breeding seals in the southeastern Gulf of St. Lawrence in 1983, the growth of the Gulf grey seal population may have accelerated to an exponential rate.It is, perhaps, no coincidence that sealworm infection parameters in 4T plaice appeared to remain stable from the 1950s to the early 1980s (McClelland et al. 1985), but increased significantly in the mid-to late 1980s (McClelland et al. 2000) after the suspension of the cull.As evident in the present study (Fig. 3), sealworm infections appeared to collapse or plateau in plaice from many sites in eastern Canada during the late 1980s to early 1990s.A resumption of positive trends in sealworm prevalence and/or abundance has since been observed in plaice from most the latter locations with the exception of those sampled from the central Scotian Shelf near Sable Island.
While grey seals are the most important definitive hosts of P. decipiens in Atlantic Canada (Marcogliese et al. 1996, Marcogliese 2001), other seal species may also have a significant impact on larval sealworm levels in groundfish.
Pseudoterranova decipiens B (sensu stricto) commonly occurs in harbour seals (Brattey and Stenson 1993), and, as there is no evidence of reproductive isolation, cross infection between harbour and grey seals and interbreeding between progeny of nematodes from the two host species may occur.The parasite is usually far less abundant in harbour seals than in grey seals (McClelland 1980a, Brattey, Stobo et al. 1990, Brattey and Stenson 1993, Hauksson and Ólafsdóttir 1995;Marcogliese et al. 1996), and the former seem less suitable hosts than the latter in terms of nematode survival, growth, gravidity, and the magnitude of the host tissue response (McClelland 1980b, Aspholm et al. 1995).
Nevertheless, in areas where they outnumber grey seals, e.g.inshore waters of the southern Gulf of St. Lawrence and southwestern Nova Scotia (Scott and Martin 1959), the lower Bay of Fundy (McClelland 1985), and southern Norway (Aspholm et al. 1995, des Clers andAnderson 1995), harbour seals may have considerable influence on larval sealworm abundance in groundfish.Although few P. decipiens reach maturity in harp seals (Phoca groen-landicus) (Brattey andNi 1992, Marcogliese et al. 1996), the cumulative progeny of adult nematodes hosted by large harp seal breeding colonies, especially during the pre-and postwhelping periods (Scott and Fisher 1958), may account for a significant proportion of the larval sealworm in Newfoundland and Gulf of St. Lawrence groundfish (McClelland et al. 1983a).
The distribution and growth of seal populations are clearly not the only factors influencing the dynamics of larval sealworm in groundfish.Despite the continued growth of local seal populations (Mohn and Bowen 1996), infection parameters in plaice stabilised or declined at many sites off southern Newfoundland and in the northern Gulf of St. Lawrence, between 1985and 1990(McClelland et al. 2000).By the early 1990s (this study), sealworm infections collapsed in southern Gulf of St. Lawrence, Breton Shelf, Scotian Shelf and Gulf of Maine plaice (Fig. 3).Between 1990 and 1992, sharp declines in sealworm infection parameters were also recorded in cod, longhorn (Myoxocephalus  al. 1990, Marcogliese and McClelland 1994, McClelland 1995, McClelland et al.2000).The parasite's distribution may also be influenced by abiotic factors such as ocean currents (dispersal of ova passed in seal feces) (McConnell et al. 1997) and environmental temperatures (des Clers et al. 1990, Marcogliese et al. 1996, Marcogliese 2001, Measures 1996).Near-bottom temperature would be particularly important in that it directly regulates the parasite's prehatch developmental rate, post-hatch survi-

86
Sealworms in the North Atlantic: Ecology and Population Dynamics val of the free-living ensheathed larvae, efficiency of transmission of larvae to and between poikilothermic intermediate hosts, and developmental rates of larvae in the these hosts (McClelland 1982(McClelland , 1990)).Temperature may also affect sealworm transmission and survival indirectly by influencing growth rates, abundance and distribution (hence, availability) of important intermediate hosts (Swain and Kramer 1993, Campana et al. 1995, Swain 1996, 1997).Marcogliese (2001) speculates that declines in levels of P. decipiens infection, apparent in 1992 samples of grey seals and groundfish from the Gulf of St. Lawrence, were largely attributable to the relatively low near bottom temperatures in the central Gulf during the late 1980s and early 1990s.Near bottom temperatures in the southern Gulf of St. Lawrence remained > 0 o C over an area of 20,000 km 2 during a cold water anomaly in 1984 (Swain 1993), and an area varying from 12,000 to 25,000 km 2 between 1990 and 1996 (Gilbert et al. 1996; D.P. Swain, Department of Fisheries and Oceans, Gulf Fisheries Centre, Moncton, N.B., personal communication).Evidently, embryonation and larvation of sealworm eggs cease when temperatures fall to the 0 to 1 o C range, and prehatch mortalities may result from prolonged incubation at these temperatures (McClelland 1982, Measures 1996).Low near-bottom temperatures prevalent in Atlantic Canadian waters since the mid 1980s (Campana et al. 1995) may have resulted, not only in declining sealworm abundance in plaice from the Gulf of St. Lawrence in the late 1980s (McClelland et al. 2000), but also in the collapse of infection parameters in plaice from the Breton and Scotian shelves, and Gulf of Maine in the early 1990s (present study).
In the present study, larval sealworm were not found in plaice 6 to 30 cm in length (n = 127), from a single set (no. 97, July 1996 Groundfish Cruise) on the western half of Misaine Bank.
On the other hand, plaice of similar size from adjacent areas of Misaine Bank and nearby Middle Bank, and larger fish (≥31 cm in length) from set 97 were heavily infected (Fig. 2).Much of Misaine Bank lies in a pocket of colder water on the eastern Scotian Shelf (Campana et al 1995), and bottom temperatures recorded in the vicinity of set 97 in recent years have been in the 0 to 2 o C range (Anonymous MS 1993a, 1993b, 1994).Bearing in mind a time lag factor, on which we speculate below, the presence of sealworm in "index" plaice from set 97 may indicate that bottom temperatures found in the area several years ago were warm enough to support the development and hatching of sealworm eggs.The lack of parasites in the smaller (younger) fish might be indicative of the inhibition of development and mortality of ova related to more recent declines in bottom temperatures.It is also possible that "index" plaice, being larger and more mobile than the smaller fish, may have incurred infections through seasonal excursions to deeper, warmer waters, or may simply be immigrants from other areas.A dip in near bottom temperatures in the northeastern Gulf of Maine during the early nineties (Campana et al. 1995), similarly, may have resulted in the decline of sealworm infection levels in Georges Bank plaice in 1993, and contributed to the erratic relationships of infection parameters and host length in both the 1993 and 1996 samples from this location (Fig. 2).
Results presented here, and in other reports (Marcogliese et al. 1996, Marcogliese 2001, McClelland et al. 2000) suggest that low near bottom temperatures experienced in eastern Canadian waters in recent years may have had a mitigating effect on abundance of P. decipiens in both groundfish and seals.Some of the data from plaice in the Canadian Maritime region (herein), however, seem difficult to interpret in terms of distribution and abundance of definitive hosts, or environmental temperature.By 1995-99, infection parameters in Canadian plaice had reached unprecedented highs at most (4TVWX-5ZE) locations where they had stabilised or fallen between 1989 and 1993.For example, infection parameters in Bradelle Bank and Magdalen Shoal plaice in 1998 were the highest yet recorded at these locations, despite the fact that bottom temperatures (<0 o C) prevailing in central 4T since 1990 (Swain 1993, Gilbert et al. 1996) may have been lethal to sealworm eggs (Measures 1996).Moreover, uninterrupted increases in prevalence and abundance of larval sealworm were documented in plaice from many locations (e.g.Banquereau sites on the eastern Scotian Shelf) from 1980 to 1996, regardless of temperature trends.Finally, since 1993, declines in infection parameters were observed in plaice from only a few locations on the central and eastern Scotian Shelf.These fish were collected, not only from isolated cold water areas (Misaine Bank), but also from warmer sites (Middle, Sable Island and Western Banks) frequented by grey seals from the large and rapidly growing Sable Island colony.
Recent declines of sealworm infections in plaice from the Sable Island Bank complex might be attributable to parasite density limiting factors in our indicator host.Sealworm infection may prove lethal to fish, through damage to vital tissues and organs by feeding and migrating nematodes, or, perhaps, through chemical impairment of the host's ability to forage and avoid predators (McClelland 1995).
Plaice appear to accumulate sealworm most frequently as 0-group or yearling fish, < 15 cm in length (Martell and McClelland 1995;McClelland 2000, McClelland andMartell 2001), and the likelihood of survival to "index" size (31 to 40 cm TL) may decline as intensity of infection increases.Mortalities among the most heavily infected hosts would result in the attenuation of the tails of worm count distributions.Data from Sable Island and Middle Bank fish seem to support this scenario.While prevalence of the parasite in plaice remained at or near 100%, abundance and maximum intensity of infection fell dramatically between 1989 and '96 (Table 3).Evidence of mortality of heavily infected juvenile hosts from the Sable Island area may also be found in the 1996 "extended" sample from Middle Bank.Whereas data from "extended" samples collected in 1996, and during earlier surveys (McClelland et al. 1983a(McClelland et al. , 1983b(McClelland et al. , 1985(McClelland et al. , 1990) ) show that sealworm infection levels generally increase as plaice mature, sealworm abundance in the 1996 Middle Bank sample peaks in 16 to 25 cm juveniles and declines in larger (older) fish (Fig. 2).
One final factor, which must be taken into consideration when interpreting spatial and temporal variations in infection levels in plaice, is the likelihood that events or processes which result in changes in sealworm levels may predate our surveys by months to years (McClelland et al. 1990).During this time, hosts may migrate some distance from locations where many of their worms were acquired.Embryonation, larvation and hatching of P. decipiens ova occur on the seabed, where the parasites developmental rates are regulated by environmental temperature.In cold waters favoured by plaice in eastern Canada (Scott 1982, Swain 1997), development time to hatch for P. decipiens eggs may range from several weeks to the better part of a year (McClelland 1982, Brattey 1990, Measures 1996).Subsequent growth and development to the infective stage (advanced L3) in poikilothermic intermediate hosts may also require additional months or years (McClelland 1990(McClelland , 1995)).Moreover, upon reaching the infective stage, sealworm may survive indefinitely in a fish host (Hemmingsen et al. 1993), although, the longevity of the P. decipiens L3 in fish seems largely dependent on the host response and varies with host species and age, and density of infection (McClelland 1995).
Evidence of larval sealworm mortality and the host encapsulation response were lacking in naturally infected Canadian plaice held at the Halifax Fisheries Research Laboratory from June 1992 to January 1998 (McClelland 2000).This indicates that infections in plaice are cumulative, with nematodes from the earliest transmissions persisting as the host matures.
Plaice diet studies (Martell andMcClelland 1994, 1995), and analyses of the lengths of P. decipiens recovered from by digestion of juvenile plaice (McClelland 2000, McClelland andMartell 2001) further reveal that larval sealworm accumulate most rapidly in 0-group and yearling fish <15 cm in length.In contrast, the 31-40 cm length range used in our "index" surveys encompasses the mean lengths-at-age for six-to eleven-year-old females, and six-to fifteen-year-old males (Beacham 1982).Hence, a significant proportion of the sealworm found in our indicator hosts may have been acquired years earlier.
Given a lag time of several years between cause and effect, it is possible that relatively high near-bottom temperatures prevalent in eastern Canadian waters during the late seventies and early eighties (Swain 1993, Campana et al. 1995) may have played a significant role in promoting marked increases in levels of P. decipiens observed in Canadian plaice during the eighties (McClelland et al. 2000).This is particularly true in the Gulf of St. Lawrence, where increases in infection levels of the magnitude detected in plaice between 1983 and 1989 (Fig. 3) could not be attributed solely to the growth of definitive host populations.As noted above, growth of the Gulf grey seal population had stalled as a consequence of a bounty program and an annual cull/controlled hunt conducted from 1967 to 1983 (Zwanenburg and Bowen 1990).
It follows that the impact of the cold water anomaly of 1984 (Swain 1993) may not have been felt until 1992, when sealworm infection parameters in plaice from Bradelle Bank, and elsewhere in the Gulf of St. Lawrence fell precipitously (Boily and Marcogliese 1995).Near-bottom temperatures, which were <0 o C throughout central 4T in 1984 would probably have proven lethal to sealworm eggs (Measures 1996).Yet, levels of larval sealworm infection in plaice sampled from Bradelle Bank and the Magdalen shoals, locations which lay within the 0 o C contour in 1984, increased significantly between 1983 and 1989 (Fig. 3).In Bradelle Bank fish, sealworm prevalence increased more than three-fold, and abundance more than seven-fold.In any event, low near bottom temperatures prevalent in the Canadian Maritimes during the late eighties and early nineties (Campana et al. 1995, Swain 1997) seemed to have resulted in only a brief interruption in the positive trends of sealworm infection parameters observed in plaice over the last two decades.In the present study, sealworm infections in many Maritime plaice populations appeared to reach a plateau or collapse briefly in the late eighties to early nineties, but significant increases in infection parameters were subsequently detected in most of these populations during the mid to late nineties. A

G
. McClelland 1 and D.J. Martell 2 I IN NT TR RO OD DU UC CT TI IO ON N octodecemspinosus) and shorthorn (M.scorpius) sculpin, Canadian plaice, yellowtail floun-der (Pleuronectes ferruginneus) and turbot (Reinhardtius hippoglossoides) (Boily and Marcogliese 1995, Marcogliese 1995) sampled throughout the Gulf of St. Lawrence.Spatial and temporal distributions of P. decipiens in our indicator host may be influenced by various biotic phenomena including changes in distribution and abundance of intermediate hosts, and mechanisms that limit parasite density in intermediate and final hosts (des Clers et Fig. 3 Temporal variations in abundance oflarval sealworm, Pseudoterranova decipiens, in Canadian plaice, Hippoglossoides platessoides, at selected sampling locations in the southern Gulf of St. Lawrence, on the Breton and Scotian shelves and in the northeastern Gulf of Maine, from 1980 to 1999.

Table 1 .
Prevalence (P), abundance (A) and maximum intensity (I max ) of larval Pseudoterranova decipiens infection in Hippoglossoides platessoides, 31-40 cm in length, from the southern Gulf of St. Lawrence, with temporal contrasts of prevalence and abundance.

Table 2 .
Prevalence (P), abundance (A) and maximum intensity (I max ) of larval Pseudoterranova decipiens infection in Hippoglossoides platessoides, 31-40 cm in length, from the Breton and eastern Scotian shelves, with temporal contrasts of prevalence and abundance.

Table 3 .
Prevalence (P), abundance (A) and maximum intensity (I max ) of larval Pseudoterranova decipiens infection in Hippoglossoides platessoides, 31-40 cm in length, from the central and southwestern Scotian Shelf, and the Gulf of Maine, with temporal contrasts of prevalence and abundance.
AC CK KN NO OW WL LE ED DG GE EM ME EN NT TS SThe authors thank Doug Swain and Tom Hurlbut (DFO Gulf Fisheries Centre, Moncton, New Brunswick) for supervising the collection of plaice samples from 1995, 1998 and 1999 groundfish cruises in the southern Gulf of St. Lawrence.Gratitude is also extended to Steve Smith, Bob Mohn, Paul Fanning and Gerri Young of Marine Fish Division (MFD), Bedford Institute of Oceanography, Dartmouth, Nova Scotia, and to Joe Hunt, Mike Strong and Maria Buzeta (MFD, St. Andrews Biological Station, St. Andrews, N.B.) for their efforts in obtaining samples from Scotia-Fundy groundfish cruises in 1993 and 1996.We are especially grateful to Treena Skory, who helped collect samples from a dedicated sealworm research cruise in August 1993.Finally, the authors thank Andrew Bauder (National Research Council, Halifax, NS), who assisted in collection of samples from a dedicated cruise in October 1996, and Jason Melendy (Gulf Fisheries Centre, Moncton, N.B.), who helped with statistical analyses, and redrafting of tables and graphics during recent revision of the manuscript.