Rates of spread of marine pathogens

Ecology Letters, (2003) 6: 1062–1067 doi: 10.1046/j.1461-0248.2003.00545.x LETTER Rates of spread of marine pathogens Hamish McCallum1*, Drew Harvell2 and Andy Dobson3 1 Department of Zoology and Entomology, The University of Queensland, Brisbane, Australia 2 Department of Ecology and Evolutionary Biology, Cornell University, Ithaca, NY, USA 3 Department of Biology, Princeton University, Princeton, NJ, USA *Correspondence: E-mail: hmccallum@zen.uq.edu.au Abstract Epidemics of marine pathogens can spread at extremely rapid rates. For example, herpes virus spread through pilchard populations in Australia at a rate in excess of 10 000 km year)1, and morbillivirus infections in seals and dolphins have spread at more than 3000 km year)1. In terrestrial environments, only the epidemics of myxomatosis and calicivirus in Australian rabbits and West Nile Virus in birds in North America have rates of spread in excess of 1000 km year)1. The rapid rates of spread of these epidemics has been attributed to flying insect vectors, but flying vectors have not been proposed for any marine pathogen. The most likely explanation for the relatively rapid spread of marine pathogens is the lack of barriers to dispersal in some parts of the ocean, and the potential for long-term survival of pathogens outside the host. These findings caution that pathogens may pose a particularly severe problem in the ocean. There is a need to develop epidemic models capable of generating these high rates of spread and obtain more estimates of disease spread rate. Keywords Epidemic, epizootic, marine pathogens, rate of spread. Ecology Letters (2003) 6: 1062–1067 INTRODUCTION Over the last few years, it has become increasingly clear that pathogen epidemics are as significant a component of the ecology of marine systems as they are for terrestrial systems (Harvell et al. 1999). The degree of connectivity and the modes of dispersal are very different in terrestrial and marine ecosystems. Obviously, a highly connected system will allow wide and rapid spread of pathogenic organisms, which will mean that emerging diseases may pose particularly serious problems. It is very likely that marine systems are more open than their terrestrial counterparts, although it is known that there is huge variation in degree of connectivity between marine regions (Hellberg et al. 2002). In this paper, we examine empirical data to see how the rate of spread of pathogens in marine systems compares with the rate of spread of similar pathogens in terrestrial systems. Two previous studies have examined rates of spread in ways that are pertinent to this study. Grosholz (1996) reviewed rates of spread for introduced species in terrestrial and marine systems. He did not specifically look at pathogens, although his terrestrial data set included rabies in European foxes and Yersina pestis (the black plague bacterium) in human beings. His overall conclusion was that there were no significant differences between the rates of spread of the 10 selected marine and terrestrial introduced species, Ó2003 Blackwell Publishing Ltd/CNRS although he did note that taxonomic differences between them may have confounded the relationship (all Grosholz’s marine introduced species were invertebrate animals, whereas the terrestrial examples ranged from the pathogens mentioned above to a weed and several vertebrates). More recently, Kinlan and Gaines (2003) have undertaken a detailed comparison of propagule dispersal in marine and terrestrial environments using both direct and genetic methods. Their analysis suggests that rates of dispersal in marine systems are two orders of magnitude faster than in terrestrial systems; this result is strongly echoed in our analysis. METHODS We searched the literature, using ISI Web of ScienceÒ (ISI Web of Science, Philadelphia, PA, USA) to locate cases of measured rates of spread for marine pathogens causing emergent disease. Our search focused on papers published since 1980, although some of these reported earlier epidemics. Our examples are specific to cases in which a pathogen has spread from an identifiable initial point to at least three locations, or where spread along a ÔfrontÕ over at least three time intervals was recorded. Although there may have been anthropogenic involvement in the initial introduction of the pathogen, or anthropogenic influence on environmental conditions or host susceptibility that may have facilitated the Rates of spread of marine pathogens 1063 spread of the pathogen, we have excluded cases where there is clear human involvement in the transport of the pathogen within the epidemic under consideration. We then searched the literature for examples of measured rates of spread for terrestrial pathogens that satisfied the same conditions, and attempted to match the marine and terrestrial epidemics as closely as was possible, by including as hosts vertebrates, invertebrates and (to correspond with sessile corals) plants (Table 1). We excluded cases where anthropogenic factors were significantly involved in the continuing propagation of Table 1 Rates of spread of epidemics in marine and terrestrial environments, sorted by decreasing order of rate of spread Habitat Pathogen Host Max. rate of spread (km year)1) Marine Herpes virus Pilchards 11 000 Marine Marine Herpes virus Bacterium Pilchards Diadema (sea urchin) 5480 4990 Terrestrial Terrestrial Marine RHV Myxoma Morbillivirus Rabbits Rabbits Seals 4970 4740 3970 Marine Morbillivirus Striped dolphins 3230 Terrestrial Marine Marine Terrestrial Terrestrial West Nile Virus Amoeba White pox bacterium Myxoma Mycoplasmal conjunctivitis Rinderpest RHV Rabies Rabies Nuclear polyhedrosis virus Baculovirus Rickettsia Dutch elm disease Sarcoptic mange Birds Sea urchin Coral Rabbits House finch 1150 936 600 600 512 Bovine tuberculosis Black band disease cyanobacterium Nuclear polyhedrosis virus Phytophthera cinnamoni fungus Phytophthera cinnamoni fungus Armillaria ostoyae fungus Inonotus tomentosus fungus African Buffalo Coral Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Marine Terrestrial Terrestrial Terrestrial Marine Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Source Date and location Murray et al. (2001b), Hyatt et al. (1997), Jones et al. (1997), Murray et al. (2000, 2001a) As above Lessios et al. (1984), Lessios (1988) Kovaliski (1998) Ratcliffe et al. (1952) Heide-Jorgensen & Härkönen (1992), De Koeijer et al. (1998), Swinton et al. (1998) Aguilar & Raga (1993), Cebrian (1995) http://cindi/usgs/gov Miller & Colodey (1983) Richardson et al. (1998) Fenner & Fantini (1999) Dhondt et al. (1998) 1995: Australia 1999: 1980: 1995: 1952: 1994: USA Nova Scotia Florida France Eastern USA Dobson & May (1986) Villafuerte et al. (1995) Anderson et al. (1981) Childs et al. (2000) Entwistle et al. (1983) 1887: 1988: 1930: 1991: 1950: Africa Spain Europe North America Canada 1970: 1985: 1944: 1993: Tonga California Quebec Spain 6 1.2 Entwistle et al. (1983) Lafferty & Kuris (1993) Gibbs (1978) Fernandez-Moran et al. (1997) De Vos et al. (2001) Bruckner et al. (1997) Spruce sawfly 0.75 Entwistle et al. (1983) 1972: Scotland Eucalyptus 0.4 Weste & Marks (1987) c. 1960: Victoria Banksia 0.0012 Hill et al. (1994) Douglas fir Spruce 0.0013 0.0002 Peet et al. (1996) Hunt & Peet (1997) c. 1960: Western Australia 1986: Canada c. 1890: Canada Ungulates Rabbits Canids Racoons Forest tent caterpillar Rhinoceros beetle Abalone American elm Chamois 500 180 60 47 32 26.7 20.7 11.5 7.65 1997: Australia 1983: Caribbean 1995: Australia 1951: Australia 1988: Baltic 1990: Mediterranean 1990: Africa 1992: Jamaica We have reported the upper limit, where the original source gives the rate of spread as a range. The dates given are the year that the epidemic commenced. Multiple rates are given for the same host–pathogen interaction only where the epidemics are demonstrably separate events. Ó2003 Blackwell Publishing Ltd/CNRS 1064 H. McCallum, D. Harvell and A. Dobson the epidemic. For example, the most dramatic recent disease outbreak in domestic animals has been the foot and mouth disease epidemic in Britain in 2001, which spread very rapidly over most of the UK. However, whilst there was certainly continuing local spatial spread (Ferguson et al. 2001; Keeling et al. 2001), the broad spatial extent of the epidemic was determined by movements of animals between markets and abattoirs before the epidemic was detected (Gibbens et al. 2001). Similarly, the fungus Phytophthora cinnamomi has spread through forests in Australia, but much of the spread is thought to have been on mud attached to vehicles (Weste & Marks 1987). The figure for rate of spread of this pathogen we include is for smaller scale ÔnaturalÕ transmission. RESULTS Figure 1 compares the rates of spread of epidemics in both marine and terrestrial populations. Several patterns are immediately apparent. Recorded rates of epidemic spread in both types of environment vary by several orders of magnitude. This means that any simple two sample test used to compare rates of spread (along the lines used by Grosholz 1996) will have very low power to detect differences between terrestrial and marine environments. As the data cannot be regarded as a truly random sample, statistical analysis of the data has to be interpreted cautiously, but can show whether the mean difference in spread is consistent with a null hypothesis that our observed rates of spread have been allocated randomly into ÔmarineÕ and ÔterrestrialÕ categories. A two-way analysis of variance on log rates of spread, when corrected for the effect of host 6 5 Terrestrial 4 3 Number of cases 2 1 0 5 4 Marine 3 2 1 0 1e – 4 1e – 3 1e – 2 1e – 1 1e + 0 1e + 1 1e + 2 1e + 3 1e + 4 1e + 5 Maxmimum rate of spread (km/year) Figure 1 Maximum rates of spread of pathogens in terrestrial (top) and marine (bottom) environments. Rates of spread for pathogens with highly mobile vertebrate hosts are shown unshaded and rates of spread for pathogens with sessile (or nearly sessile) nonvertebrate hosts are shown with diagonal hatching. Ó2003 Blackwell Publishing Ltd/CNRS mobility, suggests that marine pathogens spread significantly faster than terrestrial pathogens (F ¼ 13.02, d.f. ¼ 1,25, P ¼ 0.001). When corrected for the effect of habitat, diseases with sessile (or nearly sessile) invertebrate or plant hosts spread significantly more slowly than diseases with mobile, vertebrate hosts (F ¼ 21.69, d.f. ¼ 1,25, P ¼ 0.00001). The two order of magnitude difference in rate of spread (estimated mean difference in log10 rate of spread ¼ 2.1600, SE ¼ 0.5983) between marine and terrestrial systems is very similar to that found by Kinlan and Gaines (2003) in their survey of propagule dispersal in these systems. It is also clear that some viral pathogens of marine vertebrates have extraordinary rates of spread: more than 12 000 km year)1 for the pilchard epidemics on the Australian coast, and 3000–6000 km year)1 for morbillivirus epidemics in seals and dolphins. These pathogens are believed to be spread by contact, yet their rates of spread may exceed (in the case of the pilchard virus) the mean swimming speed of their hosts. Amongst terrestrial epidemics, viral pathogens of vertebrates also have the highest rates of spread. The rates of spread of epidemics in Australia of the introduced rabbit pathogens myxomatosis (4700 km year)1) and rabbit calicivirus (5000 km year)1) are of the same order of magnitude as the marine viral epidemics. The rate of spread of these rabbit viruses is usually attributed to long distance dispersal of insect vectors by high altitude winds (Fenner & Fantini 1999; Cooke & Fenner 2002). Similarly, the recent spread in North America of West Nile Virus in birds at 1200 km year)1 and mycosal conjunctivitis amongst house finches at about 500 km year)1 are rapid, and also involve flying as a means of pathogen dispersal. The fastest terrestrial epidemic in which flight is not a clear mode of dispersal is the spread of rinderpest through migratory African ungulates at the end of the 19th century, at c. 500 km year)1. The rate of spread of rabies, both in Europe and North America, has been relatively modest, at 30–60 km year)1. In both marine and terrestrial environments, infections of relatively sessile hosts (corals, plants and invertebrates) tend to have lower rates of spread than infections of mobile, vertebrate hosts. Even amongst relatively sessile invertebrates, some marine epidemics have spread at extremely high rates: for example, at least 2800 km year)1 in an epidemic (suspected to be bacterial; Lessios 1988) amongst sea urchins (Diadema) in the Caribbean, 900 km year)1 in an amoebal epidemic amongst sea urchins in Nova Scotia, and 600 km year)1 in a bacterial white pox (Aurantimonas) epidemic amongst corals in the Caribbean. In contrast, rates of pathogen spread amongst terrestrial invertebrates are typically slower. Plant epidemics are perhaps the closest terrestrial analogues to epidemics amongst sessile corals. It is clear from Table 1 that epidemics of soil borne fungal pathogens may be very slow moving indeed. Rates of spread of marine pathogens 1065 DISCUSSION In the absence of flying vectors, to which the rapid spread of some terrestrial pathogens has been attributed, how can rates of epidemic spread in the oceans in excess of 2000 km year)1 be explained? There is good evidence that pathogens may be transported on an intercontinental scale by high-level winds (Brown & Hovmøller 2002), but these are rare, one-off events that relate to colonization, rather than spread through a continuum. At least three hypotheses to account for the rapid spread of marine epidemics can be proposed. First, the approximately one-dimensional (along coastlines) nature of some marine epidemics (e.g. the two independent epidemics of herpes virus amongst pilchards in Australia) may mean that the linear rate of spread is greater than in comparable terrestrial epidemics, which are more characteristically two-dimensional. Second, marine environments may lack barriers to dispersal of pathogens that characterize terrestrial environments (see, e.g. Smith et al. 2002). Although there has been significant new work emphasizing effective barriers to gene flow in the ocean generated by small-scale currents and eddies (Hellberg et al. 2002; Palumbi et al. 2003), many locations are effectively open (Hellberg et al. 2002) and nothing is known about barriers to pathogen flow. Third, the strongly directional ocean currents that run along many coastlines may be responsible for epidemic spread. Although high altitude jet streams may act in a similar way for some insect vectored terrestrial pathogens, only the myxoma virus and rabbit calicivirus are likely to have been influenced by this effect. The simplest model of spatial spread of an invading species is a diffusion model (Skellam 1951; van den Bosch et al. 1992). These models often represent spread in twodimensions. Comparison of one- and two-dimensional diffusion models shows that the rate of linear spread in a one-dimensional model is the same as that in any direction for a symmetric two-dimensional model, given the same diffusion coefficient and intrinsic growth rate (Murray 1989). If the process can be described by a diffusion model, there is nothing intrinsic to one-dimensional spread that should increase the rate of linear spread. As various authors have pointed out, diffusion may not necessarily be a good model for the spread of organisms, including pathogens. One problem is that diffusion models assume that organisms have a finite (although infinitesimal) probability of moving infinite distances. Telegraph models, which assume that velocity is finite and direction is maintained are alternatives, but in practice the difference between their predictions of rate of spread from those of diffusion models is small (Holmes 1993). At the other extreme, despite the small probability of infinite dispersal, diffusion models assume that the distribution of dispersers through space follows a normal distribution, whereas it is likely that dispersal distributions may have ÔfatterÕ tails. That is, there may be a small number of very long distance dispersers. Obviously, these might have a very major influence on the rate of propagation of an epidemic. In particular, fat tailed distributions increase the sensitivity of the rate of propagation to R0 (Clark et al. 2001). Whether the tails of dispersal kernels for marine organisms should be fatter than those for terrestrial organisms is unclear. It is likely that pathogens (whether viral, fungal or bacterial) may survive better in the sea than in air (Colwell & Huq 2001), which may increase the probability of occasional longdistance dispersal. Not all marine disease outbreaks have been characterized by a spatial spread of disease from one, or a few foci. For example, the Ôwasting diseaseÕ of eelgrass (Zostera marina), which caused a catastrophic decline in eelgrass beds on both sides of the North Atlantic in the 1930s, did not appear to spread in an epidemic manner (Rasmussen 1977). Indeed, it is not entirely clear that a pathogen was the primary cause of wasting disease. There have been mass mortalities of oysters in most parts of the world, attributed to a variety of pathogens (Sindermann 1990), but little information is available to permit an estimate of the spatial spread of these epidemics. There has been great recent interest in measuring degree of connectivity in marine ecosystems. Many of these studies have focused on the relatively few cases where apparently open marine ecosystems are effectively closed by limited gene flow (Hellberg et al. 2002). Even in cases where gene flow is shown to be restricted, it is a very large assumption, and completely untested, that marine pathogens will be captured by the same barriers and processes that restrict fish and invertebrate gene flow. Thus, it is very possible that some pathogens, due to great longevity, rafting ocean currents and diverse vectoring, will experience very few barriers to dispersal. This renders estimates of pathogen longevity and spread rates in the ocean as critical measurements to make, in testing the hypothesis that marine epidemics spread faster than many terrestrial epidemics. ACKNOWLEDGEMENTS This letter arose from the NCEAS Working Group ÔDiseases in the OceanÕ. We are also grateful to Professor Frank Fenner, who provided information about the initial spread of myxomatosis in Australia. REFERENCES Aguilar, A. & Raga, J.A. (1993). The striped dolphin epizootic in the Mediterranean Sea. Ambio, 22, 524–528. Anderson, R.M., Jackson, H.C., May, R.M. & Smith, A.M. (1981). Population dynamics of fox rabies in Europe. Nature, 289, 765–771. Ó2003 Blackwell Publishing Ltd/CNRS 1066 H. McCallum, D. Harvell and A. Dobson van den Bosch, F., Hengeveld, F.R. & Metz, J.A.J. (1992). Analysing the velocity of animal range expansion. J. Biogeogr., 19, 135–150. Brown, J.K.M. & Hovmøller, M.S. (2002). Aerial dispersal of pathogens on the global and continental scales and its impact on plant disease. Science, 297, 537–541. Bruckner, A.W., Bruckner, R.J. & Williams, E.H. (1997). Spread of a black-band disease epizootic through the coral reef system in St Ann’s Bay, Jamaica. Bull. Mar. Sci., 61, 919–928. Cebrian, D. (1995). The striped dolphin Stenella-Coeruleoalba epizootic in Greece, 1991–1992. Biol. Cons., 74, 143–145. Childs, J.E., Curns, A.T., Dey, M.E., Real, L.A., Feinstein, L., Bjornstad, O.N. et al. (2000). Predicting the local dynamics of epizootic rabies among raccoons in the United States. Proc. Natl. Acad. Sci. USA, 97, 13666–13671. Clark, J.S., Lewis, M. & Horvath, L. (2001). Invasion by extremes: population spread with variation in dispersal and reproduction. Am. Nat., 157, 537–554. Colwell, R. & Huq, A. (2001). Marine ecosystems and cholera. Hydrobiologia, 460, 141–145. Cooke, B.D. & Fenner, F. (2002). Rabbit haemorragic disease and the biological control of wild rabbits, Oryctolagus cuniculus, in Australia and New Zealand. Wildl. Res., 29, 689–706. De Koeijer, A., Diekmann, O. & Reijnders, P. (1998). Modelling the spread of phocine distemper virus among harbour seals. Bull. Math. Biol., 60, 585–596. De Vos, V., Bengis, R.G., Kriek, N.P.J., Michel, A., Keet, D.F., Raath, J.P. et al. (2001). The epidemiology of tuberculosis in free-ranging African buffalo (Syncerus caffer) in the Kruger National Park, South Africa. Onderstepoort J. Vet. Res., 68, 119– 130. Dhondt, A.A., Tessaglia, D.L. & Slothowa, R.L. (1998). Epidemic mycoplasmal conjuntivitis in house finces from Eastern North America. J. Wildl. Dis., 34, 265–280. Dobson, A.P. & May, R.M. (1986). Disease and conservation. In: Conservation Biology (ed. Soulé, M.E.). Sinauer, Sunderland, MA, pp. 345–365. Entwistle, P.F., Adams, P.H.W., Evans, H.F. & Rivers, C.F. (1983). Epizootiology of a nuclear polyhedrosis virus (Baculoviridae) in European spruce sawfly (Gilpinia hercyniae): spread of disease from small epicentres in comparison with spread of baculovirus diseases in other hosts. J. Appl. Ecol., 20, 473–487. Fenner, F. & Fantini, B. (1999). Biological Control of Vertebrate Pests: The History of Myxomatosis. An Experiment in Evolution. CABI Publishing, Wallingford, Oxon, UK. Ferguson, N.M., Donnelly, C.A. & Anderson, R.M. (2001). The foot-and-mouth epidemic in Great Britain: pattern of spread and impact of interventions. Science, 292, 1155–1160. Fernandez-Moran, J., Gomez, S., Ballesteros, F., Quiros, P., Benito, J.L., Feliu, C. et al. (1997). Epizootiology of sarcoptic mange in a population of cantabrian chamois (Rupicapra pyrenaica parva) in Northwestern Spain. Vet. Parasitol., 73, 163–171. Gibbens, J.C., Sharpe, C.E., Wilesmith, J.W., Mansley, L.M., Michalopoulou, E., Ryan, J.B.M. et al. (2001). Descriptive epidemiology of the 2001 foot-and-mouth disease epidemic in Great Britain: the first five months. Vet. Rec., 149, 729–743. Gibbs, J.N. (1978). Intercontinental epidemiology of Dutch elm disease. Ann. Rev. Phytopathol., 16, 287–307. Grosholz, E.D. (1996). Contrasting rates of spread for introduced species in terrestrial and marine systems. Ecology, 77, 1680–1686. Ó2003 Blackwell Publishing Ltd/CNRS Harvell, C., Kim, K., Burkholder, J., Colwell, R., Epstein, P., Grimes, J. et al. (1999). Emerging marine diseases – climate links and anthropogenic factors. Science, 285, 1505–1510. Heide-Jorgensen, M.P. & Härkönen, T. (1992). Epizootiology of the seal disease in the eastern North Sea. J. Appl. Ecol., 29, 99–107. Hellberg, M.E., Burton, R.S., Neigel, J.E. & Palumbi, S.R. (2002). Genetic assessment of connectivity among marine populations. Bull. Mar. Sci., 70, 273–290. Hill, T.C.J., Tippett, J.T. & Shearer, B.L. (1994). Invasion of Bassendean dune Banksia woodland by Phytophthora cinnamomi. Aust. J. Bot., 42, 725–738. Holmes, E.E. (1993). Are diffusion models too simple? A comparison with telegraph models of invasion. Am. Nat., 142, 779–795. Hunt, R.S. & Peet, F.G. (1997). Annual spread rate of tomentosus root disease. Plant Dis., 81, 1053–1056. Hyatt, A.D., Hine, P.M., Jones, J.B., Whittington, R.J., Kearns, C., Wise, T.G. et al. (1997). Epizootic mortality in the pilchard Sardinops sagax neopilchardus in Australia and New Zealand in 1995. II. Identification of a herpesvirus within the gill epithelium. Dis. Aquat. Org., 28, 17–29. Jones, J.B., Hyatt, A.D., Hine, P.M., Whittington, R.J., Griffin, D.A. & Bax, N.J. (1997). Special topic review: Australasian pilchard mortalities. World J. Microbiol. Biotech., 13, 383–392. Keeling, M.J., Woolhouse, M.E.J., Shaw, D.J., Matthews, L., ChaseTopping, M., Haydon, D.T. et al. (2001). Dynamics of the 2001 UK foot and mouth epidemic: stochastic dispersal in a heterogeneous landscape. Science, 294, 813–817. Kinlan, B.P. & Gaines, S.D. (2003). Propagule dispersal in marine and terrestrial environments: a community perspective. Ecology, 84, 2007–2020. Kovaliski, J. (1998). Monitoring the spread of rabbit hemorrhagic disease virus as a new biological agent for control of wild European rabbits in Australia. J. Wildl. Dis., 34, 421–428. Lafferty, K.D. & Kuris, A.M. (1993). Mass mortality of Abalone Haliotis cracherodii on the California Channel-Islands: tests of epidemiologic hypotheses. Mar. Ecol. Prog. Ser., 96, 239–248. Lessios, H.A. (1988). Mass mortality of Diadema antillarium in the Caribbean: what have we learned? Ann. Rev. Ecol. Syst., 19, 371– 393. Lessios, H.A., Robertson, D.R. & Cubit, J.D. (1984). Spread of Diadema antillarum mass mortality through the Caribbean. Science, 226, 335–337. Miller, R.J. & Colodey, A.G. (1983). Widespread mass mortalities of the Green Sea-Urchin in Nova-Scotia, Canada. Mar. Biol., 73, 263–267. Murray, J.D. (1989). Mathematical Biology. Springer-Verlag, Berlin. Murray, A.G., O’Callaghan, M. & Jones, B. (2000). The development of a model of the spread of the pilchard fish kill events in southern Australian waters. CSIRO Marine Research, Hobart. Murray, A.G., O’Callaghan, M. & Jones, B. (2001a). A model of transmission of a viral epidemic among schools within a shoal of pilchards. Ecol. Model., 144, 245–259. Murray, A.G., O’Callaghan, M. & Jones, B. (2001b). Simple models of massive epidemics of herpesvirus in Australian (and New Zealand) pilchards. Env. Internat., 27, 243–248. Palumbi, S.R. (2003). Population genetics, demographic connectivity, and the design of marine reserves. Ecol. Appl., 13, S146– S158. Rates of spread of marine pathogens 1067 Peet, F.G., Morrison, D.J. & Pellow, K.W. (1996). Rate of spread of Armillaria ostoyae in two Douglas-fir plantations in the southern interior of British Columbia. Can. J. Forest Res., 26, 148–151. Rasmussen, E. (1977). The wasting disease of eelgrass (Zostera marina) and its effects on environmental factors and fauna. In: Seagrass Ecosystems: A Scientific Perspective (ed. McRoy, C.P. & Helfferich, C.). Marcel Dekker, New York, pp. 1–51. Ratcliffe, F.N., Myers, K., Fennessy, B.V. & Calaby, J.H. (1952). Myxomatosis in Australia: a step towards biological control of the rabbit. Nature, 170, 7–13. Richardson, L., Goldberg, W., Kuta, K., Aronson, R., Smith, G., Ritchie, K. et al. (1998). Florida’s mystery coral-killer identified. Nature, 392, 557–558. Sindermann, C.J. (1990). Principal Diseases of Marine Fish and Shellfish. Diseases of Marine Shellfish, Vol. 2, 2nd edn. Academic Press, San Diego, pp. 364–388. Skellam, J.G. (1951). Random dispersal in theoretical populations. Biometrika, 38, 196–218. Smith, D.L., Lucey, B., Waller, L.A., Childs, J.E. & Real, L.A. (2002). Predicting the spatial dynamics of rabies epidemics on heterogeneous landscapes. Proc. Natl. Acad. Sci. USA, 99, 3668– 3672. Swinton, J., Harwood, J., Grenfell, B.T. & Gilligan, C.A. (1998). Persistence thresholds for phocine distemper virus infection in harbour seal Phoca vitulina metapopulations. J. Anim. Ecol., 67, 54–68. Villafuerte, R., Calvete, C., Blanco, J.C. & Lucientes, J. (1995). Incidence of viral hemorrhagic disease in wild rabbit populations in Spain. Mammalia, 59, 651–659. Weste, G. & Marks, G.C. 1987. The biology of Phytophthora cinnamomi in Australasian forests. Ann. Rev. Phytopathol., 25, 207– 229. Editor, Peter Thrall Manuscript received 29 August 2003 First decision made 25 September 2003 Manuscript accepted 13 October 2003 Ó2003 Blackwell Publishing Ltd/CNRS