Several research laboratories dealing with aquatic animals are forced to use mains (tap) water as a source of experimental water. This water, which must normally be de-chlorinated before use, introduces further noise to the test system, leading to less reliable results. Pumping of such water to the experimental aquaria, as is often necessary, can also result in problems related to supersaturation of dissolved nitrogen.

Consequently, we find that aquatic test systems, especially those utilising fish, often display high levels of background mortality among the test animals. The problem presently exists however, that high background mortalities are generally more readily accepted in experiments involving fish, than in those involving terrestrial animals. This is probably due to the fact that concerns for animal welfare in terrestrial animals assumes a higher profile than aquatic animals, with the exception of whales.

The health status of many fish acquired for experimental work is often dubious. These animals are frequently found to carry bacterial, viral and/or fungal diseases and as such should be regarded as unsuitable for use as research animals. Access to defined strains of aquatic test animals is almost universally low. Iso-genic strains have been produced in some species, but such fish are not normally declared as specific pathogen free (SPF) animals. Due to both regulatory and economic problems related to the transportation of large numbers of fish between countries, no international trade in laboratory strains of SPF salmon or trout has so far been established.

Another extremely important issue related to fish as test animals is the genetic profile of the fish used. The genetic profile of experimental fish is perhaps the least obvious and most elusive variable among the numerous factors that may influence the outcome of an experiment. Failure to describe and standardise the test population will undoubtedly compromise the value of otherwise diligent work.

Several attempts are being made to replace in vivo test systems in aquatic animals with in vitro tests. This is of course the aim for all scientists presently using live animals in their research. However, with respect to aquatic in vivo test systems, we have a long way to go before we may claim that our systems are valid. Unless such validation is achieved, validation of future in vitro tests will be impossible.