Fisheries and Oceans Canada, Gulf Fisheries Centre, PO Box 5030
(343 Ave. Université Ave.), Moncton, NB E1C 9B6 (E1C 5K4), Canada
Effective disease management and risk analyses rely on accurate data and information. For aquatic organisms, much of this information has been derived from a relatively narrow array of diagnostic tools, most of which are either non-pathogen-specific or have undergone "the test of time" rather than methodical validation or standardization procedures. Recently, however, this range of diagnostic methods has expanded to encompass the molecular expertise pioneered by human health and agricultural food production needs. The complexity of many of these techniques, and their rapid adaptation to "field kits" for use by non-specialist personnel, has prompted a serious re-evaluation of what we use in aquatic animal health management and why (Hiney, 1997).
This paper is aimed at determining which areas of aquatic animal health management are limited by diagnostic and pathogen detection technology, and which are adequately met by traditional methods. Specific disease examples are described elsewhere in these proceedings by the specialists working with them, thus, the points below are deliberately general and provided as "food for thought".
Aquatic animal health management needs arise from two separate situations:
1. The proliferation of opportunistic pathogens in physiologically stressed or immunologically compromised host populations, requiring sensitive, early, detection of potential pathogens.
2. The spread of a primary infectious organisms between infected and uninfected individuals, stocks or populations, requiring accurate identification of the pathogens responsible for disease outbreaks, sensitive detection of pathogens in sub-clinical carriers or abnormal hosts and accurate differentiation between benign and significant infectious organisms.
Disease diagnosis - identification of the cause of a disease outbreak.
Some diseases can be diagnosed in the field with minimal technology or the need to isolate the causative agent, e.g., bacterial gill disease in association with stressful rearing conditions (Thoesen, 1994). Others present clinical signs which defy rapid or conclusive diagnosis, e.g., Malpeque disease of American oysters, Crassostrea virginica (McGladdery, 1993). Yet other diseases are caused by a range of different infectious agents e.g., chitinolytic fungal and bacterial shell diseases of crustaceans (Brock and Lightner, 1990). These situations can lead to diagnostic confusion (misdiagnosis) and ineffective management. First time disease outbreaks may (and should) require sample referrals to laboratories or diagnosticians which have experience with the putative pathogen - experience often being as important as technology for rapid and accurate disease diagnosis.
What do we have?
The tools available for disease diagnosis differ between the types of aquatic organisms being examined. For finfish, there is a relatively broad range of diagnostic techniques, many of which can be used as cross checks for diagnosis of a single disease. For example, epizootic haematopoietic necrosis of redfin perch (Perca fluviatilis) and rainbow trout (Oncorhynchus mykiss) can be confirmed by: i) conventional isolation on BF-2 (bluefin gill 2) or FHM (fathead minnow) cell lines with serological identification of the iridovirus agent; ii) an indirect immunofluorescence antibody test (IFAT); iii) an enzyme-linked immunosorbent assay (ELISA); or iv) polymerase chain reaction (PCR) amplification and subsequent sequencing of the iridovirus DNA, using two published primers (OIE, 1997). Viral encephalopathy and retinopathy (viral nervous necrosis) virus and related nodaviruses are detectable using a range of specific and less specific techniques including: i) ultrastructural confirmation of virus-induced histopathology; ii) immunohistochemistry; iii) DFAT; iv) ELISA; and v) PCR amplification and sequence analysis (OIE, 1997). Some diseases with a more limited range of diagnostic options can be diagnosed accurately using the techniques available e.g. whirling disease of salmonids, caused by Myxosoma cerebralis, is presumptively diagnosed by behaviour, with confirmatory observation of the myxosporean spores in cartilage digests or histology preparations. Furthermore, few finfish diseases with a single aetiology, have defied conclusive diagnosis for long periods.
The role of multiple infectious agents in a disease can usually be resolved through experimental research and verification using Koch's postulates. One example for finfish (which has an as yet unidentified aetiologic agent) is erythrocytic inclusion body syndrome (EIBS). Although the causative agent is believed to be viral in nature, secondary infections by bacteria and fungi can confound diagnosis (Thoesen, 1994; Jarp et al., 1996). Once the primary infectious agent for such diseases is identified, subsequent diagnosis is simplified and, generally, ignores the presence of the secondary pathogens. The classic example of one such multi-factorial disease is epizootic ulcerative syndrome (EUS) which is described in detail elsewhere in these proceedings.
The range of diagnostic techniques available for molluscan and crustacean diseases is narrower than that for finfish. Most significantly, aquatic invertebrates lack self-replicating cell lines for isolation and identification of intracellular pathogens. Finfish cell lines can be used, but the nature of the isolated viruses is often subject to question, since they could be vertebrate contaminants rather than primary invertebrate pathogens (Hill et al., 1986). In addition, Koch's postulates have rarely been fulfilled or replicated for molluscan or crustacean isolates from fish cellm lines. Thus, most intracellular infections which cause overt disease in crustaceans and molluscs require histopathology for presumptive diagnosis, with ultrastructural confirmation of viral or bacterial aetiology. Histology, although laborious, has the advantage that it provides a permanent record of the pathogen in situ and can be used to assess focal or systemic histopathology. Conversely, it is limited in sensitivity to infectious agents which can be detected, and identified, at the light microscope level, eliminating most viruses, many bacteria, protists and even some metazoan parasites (which require whole mount or adult-stage identification).
As with finfish, multiple diagnostic techniques are available for a number of shrimp diseases (Lightner, 1996) but most clinical cases can be presumptively diagnosed using non-specific techniques (gross observation, histology and tissue smears). Confirmatory diagnosis is then achieved using culture e.g. crayfish plague (Alderman and Polglase, 1986) or electron microscopic examination of ultrastructural features (Lightner, 1996). Pathogen culture is rarely used for diagnosis of molluscan diseases with the exception of two groups of significant disease agents: i) Perkinsus spp. (Ray, 1966; Gauthier and Vasta, 1993; LaPeyre et al., 1993); and ii) Labyrinthuloides-like protists (Bower, 1987; Kleinschuster et al., 1998). In clinical cases, however, these are also readily diagnosed using standard histology. Most other significant disease agents of molluscs are difficult to culture, but can be isolated under certain conditions (Hervio et al., 1993; 1995) i.e. acute infections. However, pathogen isolation is normally reserved for development of more specific detection and identification techniques rather than clinical diagnoses.
What are the limitations?
Speed of diagnosis is always a concern, especially with acute losses relying on histopathology, ultrastructural confirmation or long periods of tissue/media culture. The time span required for confirmatory diagnosis is frequently overcome by remedial action being based on presumptive diagnoses such as tissue smears, gross pathology or behavioural changes. This is most effective in areas with a well defined history of the disease e.g. Denman Island disease which is caused by Mikrocytos mackini in Pacific oysters (Crassostrea gigas) on the west coast of Canada (Bower, 1988). First time disease outbreaks in new species to culture, or appearing at a location for the first time, can undergo protracted periods of non-diagnosis or, worse, misdiagnosis. Examples include a serious disease of hard shell clams, Mercenaria mercenaria, caused by an unidentified Labyrinthuloides-like organism, "QPX". This may have been causing mortalities in pre-culture history (Drinnan and Henderson, 1963) but its significance was not fully realised until hatchery broodstock and cultured stock on grow-out beds started to die (Whyte et al., 1994; Ragone Calvo et al., 1997; Smolowitz et al., 1996). Similarly, infectious salmon anaemia of Atlantic salmon (Salmo salar) was described as haemorrhagic kidney syndrome (HKS) when first detected in Atlantic Canada (Getchell, 1997). It took over a year before the causative agent was recognised as a virus and identified as ISAV, an agent previously known only from Norway (Hastein, 1997).
Where can molecular techniques enhance disease diagnosis?
Significant pathogens that require long, complex culture or histology-based confirmatory diagnosis are prime candidates for rapid, pathogen-specific diagnostic kits. This applies predominantly to microbial pathogens, but may be equally appropriate for protists which are difficult to distinguish morphologically at the light microscope level or which have a diverse host-range. Rapid, pathogen-specific diagnostics would be particularly appropriate for disease management and control when diseases emerge in new geographic locations or host species, as described under limitations. An additional application for molecular techniques is for research into the pathogenesis of a disease via non-lethal sampling e.g. of haemolymph, fin- or gill-clips. This would provide useful information on pathogen proliferation, haemolymph profiles etc. but negates examination of the physical host-pathogen interface.
Pathogen screening - detecting infectious agents in sub-clinical or healthy organisms.
Screening for infectious agents in healthy hosts is probably the most controversial area of aquatic animal health management. This is due to: i) inconclusive negative results; ii) the potential disease risks; and iii) the difficulty of controlling a disease outbreak in naïve and/or open-water populations. Since pathogen screening is frequently a pivotal part of disease risk assessments prior to transboundary transfers, the techniques used can also be "politically sensitive". More recently, pathogen and/or disease screening is being used to define aquatic zones (intra-national and, more rarely, international) based on health profiles. This is usually limited to specific pathogens of commercially important host species (OIE, 1977). These zones can then be used for management purposes, to allow movement of pathogen carriers between non-confluent waters where the pathogen is known to occur ("like-to-like" transfers). Pathogen detection in healthy (carrier) hosts never assures 100% confidence, statistically, therefore negative samples all have a level of error which can be directly related to the sensitivity of the screening technique(s) applied.
Since disease risk analyses have been, and continue to be, well-debated (DeVoe, 1992), more effort has been focussed on epidemiological principals in an effort to quantify and standardise the broad range of qualitative-based risk evaluations (Hiney, 1997; Thorburn, 1999). This has revealed a broad gap between the probability of detection of a single pathogen in a given sample and the statistical confidence in that detection. This reflects non-survey-based assumptions for pathogen prevalence in wild or open-water populations, as well as detection sensitivity, since prevalence is one of the critical factors determining confidence of detection of a single pathogen in any given sample size (Ossiander and Wedemeyer, 1973).
What do we have now?
Pathogen screening of aquatic animals involves the same techniques described above for disease diagnosis. In addition to culture-based techniques, immunoassays (fluorescent antibody tests, agglutination tests, ELISA) and nucleic acid probes have been available for finfish pathogens for years, and some form the basis of kits now used for pathogen screening (e.g. Aeromonas salmonicida which causes furunculosis, and Renibacterium salmoninarum which causes bacterial kidney disease). Pathogen-sensitive techniques for molluscs pathogens have been developed more recently e.g. immunoassays for Perkinsus marinus (Dungan and Roberson, 1993), Bonamia ostreae (Mialhe et al., 1988), Vibrio tapetis (causative agent of brown ring disease of Manila and Portuguese carpet clams, Ruditapes philippinarum and R. descussatus, respectively) (Castro et al., 1995) and giant rickettisia of the sea scallop Pecten maximus (Le Gall et al., 1992). However, none of these techniques has yet been transformed from research to routine diagnostic application and histology remains the most broadly used detection/diagnostic method applied to molluscs. Detection of sub-clinical infections in shrimp is limited to infectious hypodermal and hematopoietic necrosis virus (IHHNV), using bioassays as well as in situ and dot blot hybridisation or PCR of viral product in haemolymph or tissue homogenates, and baculoviral midgut gland necrosis virus (BMNV) using bioassays in susceptible Penaeus japonicus and a fluorescent antibody test (Lightner, 1996). Stress-induced enhancement of infections is another procedure used to enhance some sub-clinical viral infections in shrimp (and other aquatic species) which cannot be detected by histology (Lightner, 1996).
What are the limitations?
For cryptic infectious agents (mainly microbial) routine diagnostic procedures on healthy animals, especially non-culture-based techniques, are particularly weak in detection sensitivity. Thoesen (1994) lists several diseases caused by primary pathogens for which there are no procedures documented for detecting sub-clinical infections (e.g. Vibrio salmonicida, cold-water vibriosis or "Hitra disease"; channel catfish virus, CCV; Haplosporidium nelsoni, MSX; and H. costale, SSO, of American oysters, Crassostrea virginica). Lightner (1996) lists very few pathogens of shrimp for which there no methods to detect sub-clinical infections (hepatopancreas parvovirus, HPV). However, some pathogens can only be detected following stress-testing (e.g. monodon baculoviruses, MBV).
For most disease agents in sub-clinical or abnormal "carrier" hosts, this means that sample size or sampling frequency has to be increased to enhance the level of confidence in detection. For techniques such as histology and ultrastructure, this frequently involves compromise between sample size (confidence level) and resource capability (time and manpower). For other more sensitive techniques (tissue culture, immunoassays and nucleic acid probes), the compromise may involve time, expense and specialist resource factors.
Where can molecular techniques enhance zonation establishment and surveillance or transfer disease risk analysis?
As described above, most agents of significant infectious diseases are difficult to detect using routine diagnostic techniques in healthy, sub-clinical hosts. This means that establishing an area which is designated free of a specific pathogen, inherently, includes a degree of error. Molecular screening techniques for specific pathogens could reduce this error margin by increasing confidence of detection. This would be especially important for areas that export live aquatic animals on a regular basis ("uninfected" zone to "uninfected" zone transfers). However, the pathogen specificity of these screening techniques negates detection of any other pathogenic or potentially significant organisms in the same specimens. Additional non-specific, but less sensitive, screening techniques may, therefore be required to give a true health "profile". In addition, full-scale molecular-based testing of populations for a given pathogen, especially where there has been no history of the disease, could meet with varying degrees of resistance on both a practical and political level. Interpretation of low positive results from such an area would be especially problematic and difficult to resolve. In conclusion, molecular techniques might best serve as confirmatory screening to reinforce/refute results from general screening methods both for establishing zones and for certifying stocks free of specific pathogens. This would reduce the sample size and frequency required for high-technology screening, making their application more practical and easy to justify. Ideally the confirmatory screening should be on the same specimens (or sub-samples) from the same collections to ensure cross-reference validity.
Certification of stocks as free of a specific pathogen could also benefit from the application of molecular-based detection techniques, especially where transmission is direct and negative result accuracy is imperative to prevent the spread of an endemic disease. Again, however, use of molecular probes or other pathogen-specific assays would mean any other infectious organisms would be undetected. Thus, as with zonation and surveillance, this pathogen-specific technology may best be applied as a confirmatory detection method, especially for certification of transfers from disease endemic areas.
Epizootiological research - determining the factors that trigger pathogen transmission and proliferation.
Since pathogen eradication is rarely achieved in open-water or flow-through production systems, this is a crucial area of scientific research. It provides the information essential for effective reduction of disease losses to a negligible or economically acceptable level. Epizootiology is a complex science, involving detailed research into host immunity, physiology, genetics and environmental influences. Therefore, it requires a complex battery of techniques that range in application from controlled laboratory experiments to field observations.
What do we have now?
The methods available for epidemiological research are the same as those described above for pathogen screening and disease diagnosis
What are the current limitations to epizootiological research?
The difficulty of direct observation, handling stress and duplication of environmental variables in laboratory investigations often complicates the process of quantifying qualitative clinical and sub-clinical disease observations. In addition, despite extensive and well studied physiological and immunological parameters for finfish host-environment-pathogen research (Thomas and Woo, 1995; van Muiswinkel, 1995), a lack of standardisation and validation of routine diagnostic procedures has negated their direct application to epidemiological investigations of finfish diseases (Klontz, 1993; Thorburn, 1999). Research into host-pathogen interactions is further complicated for molluscs and crustaceans, where molecular immunology has only come under close scrutiny relatively recently (Bachere et al., 1995). One notable exception is research on Gaffkemia (caused by Aerococcus viridans var. homari) of lobsters (Homarus americanus) (Stewart and Zwicker, 1972; 1974). Sadly, however, this case bears little extrapolation to other crustacean host-pathogen interactions since lobster and bacterium have a rather unique association, as summarised by Stewart (1984).
Another limitation to epidemiological research into aquatic animal pathogens is the inability to easily detect abnormal hosts (carrier, reservoir, accidental) of significant pathogens, especially those with low or unknown host-specificity. Abnormal hosts may demonstrate non-characteristic lesions or harbour the agents in tissues that are not infected in the "normal" host. This makes both detection and identification difficult, or even impossible, using routine diagnostic techniques.
Where can molecular techniques enhance epizootiological research?
As for screening, detection of sub-clinical (pre- and post-clinical) infections is imperative for understanding the dynamics of the pathogen, the factors that trigger pathogenicity and determining optimum management strategies. This includes detection of the pathogen in the environment or "abnormal" host species. In order to improve confidence in screening such samples, pathogen-specific detection or isolation techniques are required. To date, few probes which show consistent sensitivity have been developed for such broad screening (Hiney, 1997) and this appears to be an area which merits further study and development (Stokes et al., 1997).
In general the range of tools available and under development show different advantages and disadvantages for a range of different aquatic animal health applications. No one technique shows a replacement advantage over another, and none appear sufficient to merit "stand-alone" application, with the possible exception of pathogen-specific research.
Alderman, D.J. and Polglase, J.L. (1986). Aphanomyces astaci: isolation and culture. Journal of Fish Diseases 9, 367-379.
Bachere, E., Mialhe, E., Noel, D., Boulo, V., Morvan, A. and Rodriguez, J. (1995). Knowledge and research prospects in marine mollusc and crustacean immunity. Aquaculture 132, 17-32.
Bower, S.M. (1987). Artificial culture of Labyrinthuloides haliotidis (Protozoa: Labyrinthomorpha), a parasite of juvenile abalone. Canadian Journal of Zoology 65, 2013-2020.
Bower, S.M. (1988). Circumvention of mortalities caused by Denman Island oyster disease during mariculture of Pacific oysters. American Fisheries Society Special Publication 18, 246-248.
Brock, J.A. and Lightner, D.V. (1990). Diseases of Crustacea. Diseases caused by microorganisms. In: Kinne, O. (ed.) Diseases of Marine Animals. Vol III: Introduction, Cephalopoda, Annelida, Crustacea, Chaetognatha, Echinodermata, Urochordata. Biologische Anstalt Helogoland, Hamburg, Germany, pp. 245-349.
Castro, D., Luque, A., Santamaria, J.A., Maes, P., Martinez-Manzanares, E. and Borrego, J.J. (1995). Development of immunological techniques for the detection of the potential causative agent of the brown ring disease. Aquaculture 132, 97-104.
DeVoe, R.M. (ed.). (1992). Introductions and Transfers of Marine Species: Achieving a Balance Between Economic Development and Resource Protection. South Carolina Sea Grant Consortium, Charleston, South Carolina. 198 pp.
Drinnan, R.E. and Henderson, E.B. (1963). 1962 mortalities and a possible disease organism in Neguac quahaugs. In: Ellerslie Biological Station Annual Report 1962-63, B. Department of Fisheries and Oceans, Canada, 18-20.
Dungan, C.F. and Roberson, B.S. (1993). Binding specificities of mono- and polyclonal antibodies to the protozoan oyster pathogen Perkinsus marinus. Diseases of Aquatic Organisms 15, 9-22.
Gauthier, J.D. and Vasta, G.R. (1993). Continuous in vitro culture of the eastern oyster parasite Perkinsus marinus. Journal of Invertebrate Pathology 62, 321-323.
Getchell, R. (1997). ISA investigation continues; new disease challenges fish diagnosticians. Fish Farming News 5, 12.
Haastein, T. (1997). Infectious salmon anaemia (ISA): a historical and epidemiological review of the development and spread of the disease in Norwegian fish farms. Infectious Salmon Anaemia Workshop, St. Andrews, NB (Canada), Nov 26 1997.
Hervio, D., Bower, S.M. and Meyer, G.R. (1993). Detection, isolation, and host specificity of Mikrocytos mackini, the cause of Denman Island disease in Pacific oysters Crassostrea gigas. Journal of Shellfish Research 12, 136.
Hervio, D., Bachere, E., Boulo, V., Cochennec, N., Vuillemin, V., Le Coguic, Y., Cailletaux, G., Mazurie, J. and Mialhe, E. (1995). Establishment of an experimental infection protocol for the flat oyster, Ostrea edulis, with the intrahaemocytic protozoan parasite, Bonamia ostreae: Application in the selection of parasite-resistant oysters. Aquaculture 132, 183-194.
Hill, B.J., Way, K. and Alderman, D.J. (1986). IPN-like birnaviruses in oysters: infection or contamination? In: Vivares, C.P., Bonami, J.-R. and Jaspers, E. (eds.) Pathology in Marine Aquaculture (Pathologie en Aquaculture Marine). European Aquaculture Society, Special Publication no. 9, Bredene, Belgium. p. 297 (abstract only).
Hiney, M. (1997). How to test a test: Methods of field validation for non-culture-based detection techniques. Bulletin of the European Association of Fish Pathologists 17, 245-250.
Jarp J., Taksdal T. and Torud B. (1996). Infectious pancreatic necrosis in Atlantic salmon Salmo salar in relation to specific antibodies, smoltification, and infection with erythrocytic inclusion body syndrome (EIBS). Diseases of Aquatic Organisms 27, 81-88.
Kleinschuster, S. J., Smolowitz, R., and Parent, J. (1998). In vitro life cycle and propagation of quahog parasite unknown. Journal of Shellfish Research 17, 75-78.
Klontz, G.W. (1993). Epidemiology. In: Stoskopf, M.K. (ed.) Fish Medicine. W.B. Saunders, Philadelphia, US. pp. 210-213.
LaPeyre, J.F., Faisal, M. and Burreson, E.M. (1993). In vitro propagation of the protozoan Perkinsus marinus, a pathogen of the eastern oyster, Crassostrea virginica. Journal of Eukaryotic Microbiology 40, 304-310.
Le Gall, G., Mourton, C., Boulo, V., Paolucci, F., Pau, B. and Mialhe, E. (1992). Monoclonal antibodies against a gill Rickettsiales-like organism of Pecten maximus (Bivalvia): application to indirect immunofluorescence diagnosis. Diseases of Aquatic Organisms 14, 213-217.
Lightner, D.V. (1996). A Handbook of Shrimp Pathology and Diagnostic Procedures for Diseases of Cultured Penaeid Shrimp. World Aquaculture Society, Baton Rouge, Louisiana, US. (loose-leaf non-paginated).
McGladdery, S.E. (1999). Shellfish diseases (Viral, bacterial and fungal). In: Woo, P.T.K. and Bruno, D.W. (eds.) Fish Diseases and Disorders Vol. 3. Viral, bacterial and fungal infections. CAB International, Oxon, UK. pp. 723-842.
McGladdery, S.E., Drinnan, R.E. and Stephenson, M.F. (1993). A manual of parasites, pests and diseases of Canadian Atlantic bivalves. Canadian Technical Report of Fisheries and Aquatic Science 1931, 123 pp.
Mialhe, E., Boulo, V., Elston, R.A., Hill, B.J., Montes, J., van Banning, P. and Grizel, H. (1988). Serological analysis of Bonamia in Ostrea edulis and Tiostrea lutaria using polyclonal and monoclonal antibodies. Aquatic Living Resources 1, 67-69.
OIE (1997). OIE Diagnostic Manual for Aquatic Animal Diseases. 2nd Edition. OIE, Paris, France. 251 pp.
Ossiander, F.J. and Wedemeyer G. (1973). Computer program for sample size required to determine disease incidence in fish populations. Journal of the Fisheries Research Board of Canada 30, 1383-1384.
Ragone Calvo, L.M., Walker, J.G. and Burreson, E.M. (1997). Occurrence of QPX, quahog parasite unknown in Virginia hard clams, Mercenaria mercenaria. Journal of Shellfish Research 16, 334.
Ray, S.M. (1966). A review of the culture method for detecting Dermocystidium marinum, with suggested modifications and precautions. Proceedings of the National Shellfisheries Association 54, 55-69.
Smolowitz, R., Leavitt, D. and Perkins, F. (1996). An important new disease of hard clams, Mercenaria mercenaria, in the Northeast United States. Journal of Shellfish Research 15, 460-461.
Stewart, J.E. (1984). Lobster diseases. Helgolander Meeresuntersuchungen 37, 243-254.
Stewart, J.E. and Zwicker, B.M. (1972). Natural and induced bacterial activities in the hemolymph of the lobster, Homarus americanus: products of hemocyte-plasma interaction. Canadian Journal of Microbiology 18, 1499-1509.
Stewart, J.E. and Zwicker, B.M. (1974). Comparison of various vaccines for inducing resistance in the lobster Homarus americanus to the bacterial infection, Gaffkemia. Journal of the Fisheries Research Board of Canada 31, 1887-1892.
Stokes, N.A., Flores, B.S., Burreson, E.M., Alcox, K.A., Guo, Ximing and Ford, S.E. (1997). Life cycle studies of Haplosporidium nelsoni (MSX) using PCR technology. Journal of Shellfish Research 16, 336.
Thoesen, J.C. (ed.). (1994). Suggested procedures for the detection and identification of certain finfish and shellfish pathogens. 4th Edition, Version 1, Fish Health Section, American Fisheries Society, Bethesda, Maryland, USA (loose-leaf, non-paginated).
Thomas, P.T. and Woo, P.T.K. (1995). Immunological approaches and techniques. In: Woo, P.T.K. (ed.) Fish Diseases and Disorders Vol. 1. Protozoan and Metazoan Infections. CAB International, Oxon, UK., pp. 751-771.
Thorburn, M.A. (1999). Applying epidemiology to infectious diseases of fish. In: Woo, P.T.K. and Bruno, D.W. (eds.) Fish Diseases and Disorders Vol. 3. Viral, bacterial and fungal infections. CAB International, Oxon, UK. pp. 689-722.
van Muiswinkel, W.B. (1995). The piscine immune system: Innate and acquired immunity. In: Woo, P.T.K. (ed.) Fish Diseases and Disorders Vol. 1. Protozoan and Metazoan Infections. CAB International, Oxon, UK., pp. 729-750.
Whyte, S.K., Cawthorn, R.J. and McGladdery, S.E (1994). QPX (Quahaug Parasite X), pathogen of northern quahaug Mercenaria mercenaria from the Gulf of St. Lawrence, Canada. Diseases of Aquatic Organisms 19, 129-136.