There are increasing concerns that some pedigree breeding practices compromise canine welfare; with many breeds affected by inherited disorders likely due to adherence to breed standards that encourage exaggerated conformations. Consequently, calls for quantitative limits to be included in breed standards have been advocated to safeguard against conformations that are detrimental to canine welfare. This thesis explores the relationships between extreme conformation and disease-risk, to identify the most at-risk conformations currently being selected for, and recommend healthy 'limits'. Three conformation-related disorders of high welfare importance were investigated; brachycephalic obstructive airway syndrome (BOAS), intervertebral disc extrusion (IVDE) and corneal ulceration. Seven hundred dogs entering the RVe's small animal referral hospital were recruited over 14-months; their conformations quantified using established protocols and disease-status assessed. Generalised linear mixed-model analyses demonstrated for the first time that: (1) relatively shorter muzzles and relatively thicker necks increased the risk of BOAS; (2) relatively longer backs, miniaturisation and being overweight increased the risk of IVDE; and (3) the presence of nasal folds, relatively short muzzles, relatively larger eyelid openings and the presence of exposed sclera increased the risk of corneal ulcers. To test the risk estimates generated from this population, a second study of BOAS was carried out in a non-referral population. Short muzzles and thick necks were confirmed to be risk factors for BOAS, with two further lifestyle risk factors identified: obesity and neutering. A questionnaire-based study discovered an owner-perception of BOAS being 'normal for the breed' that may hamper efforts to reduce its prevalence in the future. As such, educational interventions to change attitudes may be required. The results of these studies are hoped to instigate and inform future breed standard revisions, and influence those within the breeding community to change current breeding practices to improve canine welfare by selecting for more moderate, lower-risk morphologies.

A joint research project was undertaken by the Australian Customs Service (Customs),Royal Guide Dogs Associations of Australia (RGDAA) and The University of Melbourne. The aims were to establish a breeding program for Drug Detector Dogs(Detector Dogs) and to further improve the RGDAA breeding program. Behavioural studies were conducted on the RGDAA breeding population of Labradors(Guide Dog population) and a ‘trial’ breeding population was established by Customs during the research (Detector Dog population). We defined traits important for success as a Detector Dog and designed an aptitude test (CRR test) to measure the traits. We found some sex differences for individual traits within the combined Guide Dog and Detector Dog populations. However, no sex difference was observed in terms of overall performance. Within the Detector Dog population, dogs selected to start a Detector Dog training course performed significantly better on the CRR test than dogs not selected to start a course. (For complete abstract open document)