Summary I. Reliability of the results of bio‐medical research clearly depends upon the animals used showing as standard responses as is possible. 2. The majority of animals used in this field are small, homoiothermic mammals which have sensitive and strong homeostatic mechanisms. If a change in ambient conditions is of sufficient magnitude to unbalance homeostasis, then the neuroendocrine system is stimulated so as to restore it, and this can interfere with the response to test conditions or agents. 3. The homeostatic effectors involved are diverse and can include both physiological and behavioural changes in the animal. These can affect metabolic rate, body temperature, activity, food consumption, hormone concentration, wake/sleep patterns, maturation, posture, lactation and many other bodily functions. Any of these changes is potentially capable of influencing experimental results. 4. The evidence presented shows how environmental factors may affect the outcome of experiments in the fields of animal behaviour, cancer research, immunology, pathology, pharmacology, psychology, reproduction, teratology and toxicology; particular attention is paid to the effects of ambient temperature, relative humidity, air movement and quality, light and sound. 5. While a constant, reproducible environment would be ideal, there is little possibility of controlling all the variables; nevertheless all investigators should minimize those environmental variables that have been shown to be important. 6. To enable other investigators to repeat experiments or carry out comparative studies, environmental conditions pertaining during an experiment should be adequately described in any publications.

Some common laboratory species are now considered unsuitable for the study of the toxicity of inhalants because of the natural incidence of abnormal respiratory histology. Levels of ammonia commonly encountered in animal boxes and cages have been shown to cause histopathological changes in the tracheal epithelium of rats, and it is suggested that 'abnormal respiratory histology" could be a reflection of the standard of husbandry employed before and during an experiment.

Most local barrier systems are designed primarily either to protect animals from airborne contamination (exclusion) or to ensure the safety of personnel (hazard containment). Few, other than isolators, are able to cope with the often conflicting demands to do both. The Positive Individually Ventilated system (PIV), which provides pressure adjustable, individual supply and exhaust ventilation to each cage has been tested with this in mind as well as the need to ensure that environmental conditions comply with the requirements of the Home Office Code of Practice (CoP). The results indicate that when compared with traditional open racking the system can reduce both the risk of animals becoming contaminated by airborne infection from the room and the risk of aeroallergens escaping from the cages into the room air. At the same time, environmental conditions within the cages are both less variable and less stressful to the occupants. Conditions of air temperature, relative humidity, ventilation rates, light intensity and (with appropriate air handling) sound levels, all comply with or are better than those required by the CoP. Even with the room air change rate set at only 8/h (8 ac/h) the air distribution system results in draught-free cage ventilation rates of around 50-120 ac/h. This means that with at least the PIV exhaust(s) linked directly into the air conditioning system of the building, the size of the latter and its associated running costs can be reduced by around 50%; this energy saving concept has been approved in principle by the Home Office. Additional benefits include the fact that bedding is kept much drier allowing further cost savings in bedding and associated labour costs. The system is thus beneficial to the animals in protecting them from airborne infection and other stresses. By providing a less variable environment it also helps to minimise the sort of interference with experiments which can arise from that source. Similarly, in accordance with the aims of 'The Control of Substances Hazardous to Health Regulations' (COSHH 1988) and the Health & Safety Executive (HSE 1990), by reducing dust levels in the room air, including allergens, it is also beneficial to the personnel working in the animal rooms.