The recent sixty-eight day hunger strike by imprisoned British animal rights activist, Barry Horne, brought to the fore once again the fulminating legal and ethical controversy that surrounds animal experimentation. But such events can often obscure the scientific arguments for and against the use of animals in academic and industrial research.

The list of biomedical advances in which animal experimentation has played a key role is a very long one. Indeed, as those on the ?animal testing is necessary? side of this fraught debate are fond of stating, two thirds of all the Nobel prizes that have been awarded for medicine since the beginning of this century have been for research involving animals.

However bio-science research has come along way since Pasteur vaccinated chickens and rabbits against cholera by feeding them the contents of their dead relatives guts. And, although there are still some situations for which the only viable option is the ?whole body? model offered by animals, the molecular-, computing- and imaging revolutions have already rendered many hitherto essential testing and manufacturing techniques obsolete. So that now animal experimentation - which has fallen by almost 50% over the two decades - makes up less than 10% of all medical research.

But does this mean that there are - or that there soon will be - alternatives to all the ways in which animals are currently used in scientific research? Or are there still avenues of inquiry that continue to require, and justify the use of vertebrates in the laboratory?

At present most industrialised nations use between 2 and 4 million animals for research per year. The vast majority of these are rodents, with less than 10% comprising rabbits, fish, reptiles, amphibians, birds, agricultural animals, dogs, cats and - very rarely - primates.

But what are all these animals used for? Well, contrary to what one might conclude from media hype, cosmetics testing makes up only 0.1% of all animal experiments. It is actually the development, toxicity testing and, more recently, the manufacture of new medical and veterinary surgery, treatments and diagnostics (some steps of which legally demand animal tests) that is the largest user of laboratory animals, taking up around 45%. Fundamental scientific research (that which does not necessarily work towards a therapeutic goal, but rather expands our basic knowledge base) comes a close second, using around one third of all lab animals. The rest are used for safety testing of non-medical products such as household, industrial and agricultural chemicals or simply for breeding other animals.

However, over the last twenty or so years modern technology, spurred on by public pressure, scientific fashion and the industrial profit motive, has begun to replace laboratory animals with alternatives that are often not just more ethically acceptable but also quicker, cheaper, and scientifically superior. The most important of these is probably tissue culture.

Tissue culture is basically the technique of keeping cells or pieces of organs alive ?_in vitro_?, or outside the body (where inside is referred to as ?in vivo?). This can range from cell culture involving single layers of one cell type, to organ or spheroid cultures - complex three dimensional structures which closely mimic organs. These cultures, of either human or animal cells, can then be used to observe normal and toxically challenged inter-cellular behavior.

For example, toxicity testing in human spheroid liver and brain cultures (multicellular organ-like balls of cells) is now widely used in the early stages of drug design. ?After the initial expense of starting it up, it costs 90% less than putting putative drugs into animals, and, being on human cells, gives us better predictive capacity? says Dr Chris Atterwill, Head of Pre-Clinical drug Safety at Roche Discovery, Welwyn, UK. ?Although there are still mandatory drug regulation tests in various species,? he explains, ?these kinds of technologies help us to reduce the number of compounds that make it to those tests, and thereby reduce the number of animals used overall.?

Moreover regulators are fast moving towards accepting some of these in-vitro tests as alternatives to final-stage carcinogenicity (cancer-causing) and genotoxicity (genetically damaging) drug trials.

The other major alternative to animal research is computer modelling. Computer molecular modeling is now a corner stone, for example of rapid, so-called ?rational? drug design whereby models of drug candidates for a known target are tried like virtual keys in a virtual lock to see which might fit. Computing power has also opened new windows on the brain, with the advent of neural network modeling and non-invasive imaging. And, more indirectly, the explosion in easily accessible databanks on CD ROM and the Internet has allowed the reduction of duplicate animal testing via the pooling of data.

But amid the overall downward trend in animal experimentation there are two obvious ?growth areas?, both of which involve genetic engineering. The first is the use of cloned ?transgenic? animals - animals into which scientists have put genes from other species, usually humans - to generate ?biopharmaceuticals? such as insulin, monoclonal antibodies (now banned in several EU member countries, though not France) growth hormone, vaccines or even increasingly, tissue and organs for xenotransplantation surgery.

Although animal ?bioreactors?, as these creatures have been dubbed, already make up 13% of all lab animals, their use has begun to be limited by the fact that they have met with considerable public disquiet. ?A 1996 European Commission survey found that people consider this technology both risky and morally wrong,? warns Dr Gill Langley, Scientific Adviser for the Dr Hadwen Trust, a charity which funds research alternatives research in the UK. So with the plant and bacterial production of therapeutic proteins developing so rapidly, Langley hopes that the days of animal biopharmaceuticals will be numbered. That the same will be true of xenotransplants, on the other hand, looks less likely.

The second growth area is the development of genetically mutated animals to study the role of particular genes and/or to model diseases - a technology that few can deny has lead to an exponential growth in our understanding of genetics. ?Virtually every gene in the mouse genome will be knocked out at some stage to find out what it does,? says oncologist and council member of the UK?s Research Defense Society, Terry Hamblin of the University of Southampton. ?We will never understand the human genome without these experiments,? he continues, ?we cant just sit in our offices waiting for rare human knockouts to come along like we had to do in the old days.?

But many on the animal rights side of the debate, such as Caren Broadhead of the UK-based Fund for Research into Development of Alternatives, are more skeptical. ?A lot of these models are pointless,? Broadhead complains, ?people are developing them just because they can.? A view backed up by Professor Michael Balls, Head of Unit at the European Commission?s Centre for Validation of Alternative Methods in Milan. ?The force for scientific progress can overwhelm the ethical consequence of an action,? he cautions. Since there will always be a conflict between human benefit and animal cost, Balls believes that ?as a society we have a duty to ensure that the rate of progress justifies the suffering.?

This standpoint inflames basic scientists such as Hamblin, ?we need to understand how things function; no one knows what the ramifications are going to be of research that may seem to be ?direction-less? now.? And Frankie Trull, President of the USA?s Foundation for Biomedical Research, an organization which champions humane and responsible animal research, agrees, ?the fundamental premise of research is the pursuit of knowledge for knowledge sake.? And since many of the alternatives to animal studies have what Trull describes as, ?serious limitations,? she, speaks for most scientists when she says that, for the foreseeable future at least, ?animal research will remain essential.?