Several methods have been described to quantify the first ray mobility. They all have certain disadvantages (great size, sophistication, or lack of validation). The objective of this work was to study the validity and reliability of a new instrument for the measurement of first ray mobility. Anterior-posterior radiographs were obtained from 25 normal feet and 24 hallux valgus feet, with the first ray in a neutral position, maximally dorsiflexed and maximally plantarflexed. The first ray mobility was radiographicaly measured in both groups, and was also manually examined with the new device. A cluster analysis determined whether normal and hallux valgus feet were correctly classified, and a graphic analysis of Bland-Altman was performed to compare the radiographic and manual measurement techniques. Based on the radiographs, the first ray mobility only showed significant differences in dorsiflexion between both groups (P = 0.015). First ray dorsiflexion, plantarflexion and total range of motion measured with the new device were different between both groups (P = 0.040, P = 0.011 and P = 0.006, respectively). The silhouette measure of the cohesion and separation coefficients from the cluster analysis was greater than 0.50 for the dorsiflexion, plantarflexion and total range of motion obtained from the radiographs and from the new device. The Bland-Altman graph suggested that 96% of the data presented agreement between both measurement methods. These results suggested that the new instrument was valid and reliable.