Summary: Objectives: To compare accuracy of the ultrasound-guided craniodorsal (CrD) approach with the dorsal (D) approach to the cervical articular process joints, and to evaluate the effect of the transducer, needle gauge, and operator experience. Methods: Cervical articular process joints from 14 cadaveric neck specimens were injected using either a D or CrD approach, a linear (13 MHx) or microconvex transducer (10 MHz), and an 18 or 20 gauge needle, by an experienced or inexperienced operator. Injectate consisted of an iodinated contrast material solution. Time taken for injection, number of redirects, and retrieval of synovial fluid were recorded. Accuracy was assessed using a scoring system for contrast seen on computed tomography (CT). Results: The successful performance of intraarticular injections of contrast detected by CT using the D (61/68) and CrD (57/64) approaches was comparable. No significant effect of approach, transducer or needle gauge was observed on injection accuracy, time taken to perform injection, or number of redirects. The 18 gauge needle had a positive correlation with retrieval of synovial fluid. A positive learning curve was observed for the inexperienced operator. Clinical relevance: Both approaches to the cervical articular process joints were highly accurate. Ultrasound-guided injection of the cervical articular process joints is an easilylearnt technique for an inexperienced veterinarian. Either approach may be employed in the field with a high level of accuracy, using widely available equipment.