Assortative mating for psychiatric characteristics has significant implications for treatment as well as genetic studies. This paper presents data on spouse similarity for personality and temperament traits as well as psychiatric symptomatology from a non-clinical sample. Second/third and sixth/seventh grade school children of several communities as well as their parents (N=376 couples) completed self-rating scales on personality (EPQ), temperament (DOTS-R) and psychiatric symptomatology (SCL-90R). Significant positive but rather low spouse associations were found for psychiatric symptomatology and the personality trait Psychoticism, while positive statistical trends were observed for several temperament dimensions. Spouse similarity for psychiatric symptomatology was independent of similarity for personality traits and that for personality traits was independent of underlying temperament traits. Spouse similarity was symmetric between husbands and wives and did not significantly depend on psychiatric symptomatology or demographic variables. In conclusion, our non-clinical data support the existence of significant spouse similarity for psychiatric symptomatology and personality as suggested by clinical studies. In addition, a trend for temperament resemblance was also observed. Our finding that spouse similarity for temperament, personality and psychiatric symptomatology were largely independent highlights the necessity of simultaneous assessment of these psychiatric domains in the search for the underlying characteristics conditioning non-random mate selection