During the last 30 years needle biopsy has become a standard procedure in the diagnosis of human renal disease, particularly diffuse disorders like glomerulonephritis. The technique of percutaneous renal biopsy has been proved safe and reliable in many thousands of patients in the hands of experienced operators. The major post-biopsy complications are severe haemorrhage, subcapsular haematomata and arteriovenous fistulae. Renal biopsy has been used much less in other species. Indeed, in the chimpanzee, horse, ox, pig, sheep and rabbit, sparse reports have been confined almost entirely to pilot studies and experimental investigations. In the dog, major experimental studies in the technique and effects of renal biopsy were conducted in the USA between 1967 and 1971, and during the same period 66 dogs with clinical renal disease were also biopsied. In 1982, a 10 year retrospective survey of renal biopsy in 163 dogs indicated that using a small laparotomy ("keyhole") approach to the right kidney the technique is safe and reliable under general anaesthesia. Franklin-Silverman, Metcoff paediatric Franklin-Silverman and "Tru-Cut" needles were all used in this study. The major post-biopsy complications were self-limiting haematuria, severe haemorrhage and renal subcapsular haematomata. Evaluation of renal biopsy in the cat, has, by comparison with the dog, been very limited. Although the "blind" percutaneous approach is simple to perform in the cat, the only substantial series of biopsied clinical cases hitherto reported contained 34 cats. Moreover, experimental studies into the effects of renal biopsy on the cat kidney, with its unique presence of subcapsular veins, have not been reported. In this study, 131 biopsy specimens were obtained from 53 dogs and 50 cats referred as clinical cases to the University of Glasgow Veterinary School over a 10 year period. In every case 41/2 inch "Tru- Cut" needles were used. The majority of dogs were deeply sedated with the reversible neuroleptanalgesic drug "Immobilon". On 39 occasions the direct percutaneous approach to the left kidney was used, with favourable results in terms of the number of biopsy attempts per adequate sample, lengths of samples obtained and glomerular content, when compared with the less frequently used "keyhole" and laparotomy approaches to the right or left kidney. Two investigations were conducted into the effects of percutaneous renal biopsy on the kidney of normal cats. In the first, 11 cats were subjected to a single biopsy and thereafter monitored until euthanasia at intervals up to 2 months after biopsy. At necropsy, radiographic studies following infusion of a contrast gel demonstrated vascular changes in the biopsied kidneys. Histological examination of these kidneys revealed lesions varying from barely discernible linear scars to extensive haemorrhage and wedge-shaped infarcts. A direct relationship was established between the severe renal lesions in 7 cats and biopsy specimens containing medullary tissue and major renal blood vessels. In the second investigation, 3 consecutive biopsy punctures were made in the left kidneys of 3 cats which were then monitored until euthanasia one month later. A further 3 cats were biopsied at monthly intervals on 3 occasions and then monitored until euthanasia one month after the third biopsy. At necropsy, radiographic and histological examinations revealed similar but less severe changes to those encountered in the first investigation. More cautious biopsy sampling reduced vascular damage at the corticomedullary junction but also reduced the overall quality of specimens for diagnostic purposes. Two disadvantages of the "Tru-Cut" biopsy needle became apparent in the course of these studies. First, the 20 mm. long specimen notch is too long to be safe in the average sized adult cat kidney. Second, the 6 mm. leading tip may cause unrecognised trauma to blood vessels at the corticomedullary junction. In an attempt to overcome these disadvantages, a third investigation was undertaken. Standard 41/2 inch "Tru-Cut" biopsy needles were modified so as to reduce the length of the obturator tip and the exposed length of the specimen notch by half. Four normal adult cats were each subjected to 4 biopsy punctures of each kidney using the modified needles. Satisfactory samples were obtained in 28 out of 32 biopsy attempts and the needles were shown to be more adequate and probably safer for use in the cat than the standard "Tru-Cut" instrument.