In a previous study (Cox, Swinson, Kuch & Reichman, Behaviour Research and Therapy, 31, 427-431, 1993), factor analyses of the responses of 177 Canadian panic disorder with agoraphobia patients to the 'When Accompanied' and 'When Alone' scales of the Mobility Inventory (Chambless, Caputo, Jasin, Gracely & Williams, Behaviour Research and Therapy, 23, 35-44, 1985) revealed three factors in each case: Fears of (1) Public places; (2) Enclosed spaces; and (3) Open spaces. Using two distinct methods of factorial analysis, evidence was found for the cross-national generalizability of the factor model when the responses of Dutch members of a society for individuals suffering from an anxiety disorder (N = 213) were contrasted with the original Canadian findings. Inventory items were distributed in a non-overlapping fashion across the corresponding three subscales. Psychometric properties of the subscales were encouraging, although some difficulties emerged when attempts were made at distinguishing Fears of Enclosed spaces from Fears of Open spaces. This was because of their correlational configurations with other measures. Scores on all scales varied with socioeconomic status (SES); Ss in lower SES groups had significantly higher agoraphobic avoidance scores than their equivalents in higher SES groups. Results of higher-order analysis, which included several state and trait measures of psychological functioning in addition to the Mobility Inventory, revealed two orthogonal, second-order factors which were interpreted as Agoraphobia and Neuroticism/Negative Affect vs Positive Affect. Implications for further studies are briefly outlined.