This study was performed in order to investigate the difficulty in diagnosing retrobulbar disease with ultrasound. A retrospective study was accomplished to investigate the use of ultrasound in patients with suspected retrobulbar problems, and how often the examination is diagnostic. The computer based patient files, at the University of Agricultural Science in Uppsala, were searched for dogs with suspected retrobulbar disease. During 1999-01-01-2004-04-14, 14 patients fulfilling the search criteria were found. The patient records were studied, and the written reports from the ultrasound examinations were compared with the printed pictures from the examinations. This study indicated that the most common ultrasound diagnoses were abscess and foreign body. Ultrasound was diagnostic regarding prominent abscesses. When the changes were more subtle, diagnoses such as cellulitis, neoplasia and abscess were considered. Foreign bodies were often on the list of differential diagnoses, but seldom verified. The conclusion is that it is difficult to confidently diagnose pathological changes in the retrobulbar area using ultrasound. Another aim was to increase the knowledge of the normal variation of the appearance of the retrobulbar area, examined with ultrasound. Ultrasound examinations of the retrobulbar area was done on four clinically healthy beagles. The examinations showed a very similar retrobulbar appearance, with the characteristic retrobulbar cone, containing nervus opticus, optic disc, fat and muscles. The skeleton was seen medially, and the zygomatic salivary gland laterally. Finally we wanted to investigate if it can be diagnostically valuable to measure resistive index (RI) of the ciliary arteries in dogs whith retrobulbar disease. We measured RI on retrobulbar arteries on four clinically healthy beagles. Two different types of sedation were used, Domitor vet. together with Torbugesic®, and Plegicil® vet.. The RI differed between types and degree of sedation, with a mean RI for Domitor vet. and Torbugesic® of 0,28, and a mean RI of 0,40 for Plegicil. The conclusion is that RI varies with sedation and is time consuming and thus difficult to use in a clinical situation unless the examiner is specially trained in ocular ultrasoud.