We trapped 656 Common Terns (Sterna hirundo) and measured five body dimensions and body mass for each bird; 313 birds were of known age, and 229 were sexed by DNA. Males were larger than females in all five dimensions, but were smaller in body mass. Early-nesting birds were larger than late-nesting birds in all five dimensions: at least for wing length, this difference was related to both laying date and age. Head length (from back of skull to tip of bill) was the most useful measure for sexing Common Terns in the field. Discriminant functions indicated that 75.9% of single birds and 84.5% of pairs could be sexed correctly by head length alone. We present rules and nomograms for field sexing of Common Terns; these provide trade-offs between sensitivity (proportion of birds classified) and specificity (proportion of birds correctly sexed). Three of 80 pairs (4%) included two females: these pairs nested early and were at least as successful as male-female pairs. Within pairs, tarsus lengths were negatively correlated; we found no evidence for positive assortative mating by linear dimensions or body mass. This study confirms some previous reports of sexual dimorphism in this species based on less reliable methods of sexing, but fails to confirm other reports of sexual dimorphism and assortative mating.