Noninvasive early recognition and treatment of temporomandibular joint dysfunction remains a diagnostic challenge. This pilot study evaluated the use of phosphorus 31 magnetic resonance spectroscopy with magnetic resonance imaging to measure alterations in pH and high-energy phosphate metabolite ratios of muscle that is adjacent to an inflamed temporomandibular joint. Ten New Zealand white rabbits were used in this study. Two animals were used to develop signal acquisition protocols and to ensure that stable baseline data could be measured. In each of the eight animals used in the experiment, one temporomandibular joint was injected with a suspension of silica particles and the contralateral joint served as a control. Data were collected from control and experimental joints on days 0, 7, 14, 21, and 28, after the injection. At the end of the study, temporomandibular joints were block resected and histologically examined to confirm the presence of an inflammatory response. Results indicated that pH and metabolite ratios could be obtained by 31P-magnetic resonance spectroscopy. Changes in pH and some metabolite ratios in experimental joints showed statistical significance ( p < 0.001). Differences were seen on day 2 and day 7 ( p = 0.040 and p = 0.008, respectively) in the phosphocreatine/α-adenosine triphosphate ratios. This contrasts with phosphocreatine/β adenosine triphosphate ratios that showed significance that began at day 7 ( p = 0.022) and continued to day 14 ( p = 0.025). Histologic examination indicated that the tissue response within the joint capsule was less than the granulomatous reaction expected. This pilot study clearly demonstrates that pH changes that occur in muscle that is adjacent to an inflamed joint are detectable by 31P-magnetic resonance spectroscopy.