From brain storming in different contexts I would suggest :
sex determination as factor with levels=c(1, 2, 3), labels=c('m', 'f', 'indet') nothing else.
measures, especially the names of the bones, differ in labeling: e.g. Martin 1928 (Humerus, H), Trotter & Gleser 1952 (HUM, RAD, ULNA, FEM, ...), Siegmund (H1, F1, Fib1, ...). I would suggest: 3 letters for the bone + 1 digit & 1 letter for the additonal specification, no blanks : Tib1b, Fib1, ...
From brain storming in different contexts I would suggest :