The widespread use of transgenic animals in research, and especially for generating models of human disease, has given rise to calls for a careful documentation of transgenic strains with particular emphasis on characterizing phenotypes and their animal welfare aspects. Based upon several published suggestions for general health monitoring of laboratory animals and upon a questionnaire we developed for the characterization of transgenic strains, score sheets for litter-wise and individual data collection were tested in a pilot study. The documentation concept and some of the most important parameters are presented and discussed with respect to their feasibility and usefulness. The use of score sheets can help in determining the application of humane end points at the individual level, phenotype characterization and assessment of animal welfare for the whole strain, and the development of better breeding strategies - apart from being of direct scientific interest. We suggest that these score sheets, available via the Internet at: www.bzl.unizh.ch/de/database/formtransg/index2.html, should be applied for routine and comprehensive phenotype characterizations of transgenic mouse strains used and kept for animal experimentation.

The use of genetically modified (GM) animals in biomedical research has increased during recent years and its ethical aspects have been subject to ongoing academic discussion. In order to reinforce this discussion, we analysed applications submitted to animal ethics committees in Sweden during 2002. The aim was to investigate the researchers' statements concerning the production and use of GM animals, as well as the committees' assessments of the applications. For our analysis, we constructed an analytic form. In part, we included the questions and categories of the mandatory application form, noting for example species, degree of severity regarding pain and distress, the management of pain, and endpoints. In addition, we included our own specific questions and categories, and classified the applications accordingly. In particular we focused on the methods of GM animal production and on the expected clinical symptoms attributable to genetic modification and experimental use. Our analysis, which was partly quantitative and partly qualitative, revealed that applications were often approved by the committees despite containing insufficient information regarding ethically relevant aspects, that the arguments for using GM animals were often unclear, and that some applicants indicated awareness of possible unintentional welfare effects attributable to genetic modification. In more than 36% of the applications, obvious or minor clinical symptoms attibutable to genetic modification were expected. However, we also noted that many applicants emphasised that certain GM animals were to be used without the expectation that the animals would display any clinical symptoms. This was obviously viewed as an ethical advantage.

In the early 1990s, a set of new techniques for manipulating mouse DNA allowed researchers to ‘knock out’ specific genes and observe the effects of removing them on a live mouse. In animal behaviour genetics, questions about how to deploy these techniques to study the molecular basis of behaviour became quite controversial, with a number of key methodological issues dissecting the interdisciplinary research field along disciplinary lines. This paper examines debates that took place during the 1990s between a predominately North American group of molecular biologists and animal behaviourists around how to design, conduct, and interpret behavioural knockout experiments. Drawing from and extending Harry Collins’s work on how research communities negotiate what counts as a ‘well-done experiment,’ I argue that the positions practitioners took on questions of experimental skill reflected not only the experimental traditions they were trained in but also their differing ontological and epistemological commitments. Different assumptions about the nature of gene action, eg., were tied to different positions in the knockout mouse debates on how to implement experimental controls. I conclude by showing that examining representations of skill in the context of a community’s knowledge commitments sheds light on some of the contradictory ways in which contemporary animal behaviour geneticists talk about their own laboratory work as a highly skilled endeavour that also could be mechanised, as easy to perform and yet difficult to perform well.

In mammals ‘maleness’, i.e. the presence of testes, is thought to be controlled by a single gene on the Y chromosome. Recently, a candidate gene termed the SRY (sex-determining region Y) gene has been located. If the SRY gene is the gene causing maleness then a transgenic male with the SRY gene on an autosome would produce a greater proportion of male offspring than a normal male. This would be advantageous in situations where male offspring are more valuable than females. Such transgenic males have a reduced probability of propagating their genotype and an effort has to be made to avoid their extinction. This is at the cost of genetic progress which must be made to enable the transgenics to remain competitive with normal males.

In a simulated beef cattle breeding scheme if half of the annual matings were made to transgenics then after 15 years of selection the transgenic males fell the equivalent of 2·6 years of selection behind males in a traditional herd. If all matings were made to transgenics they fell over 9 years behind. Selection for lean food conversion ratio was considered as an example. After 15 years of selection the gain in biological efficiency from more male offspring outweighed the loss from reduced genetic progress only when more than 0·5 of the bulls used in the breeding scheme were normal males. In practice, the difficulty of maintaining a small population of transgenic males along with other costs not included in the calculations suggest that breeding schemes in beef cattle with an SRY transgene would not be practicable without further technology.