Nick,
Although I personally feel that pictigaster is a very poorly-defined race (it was originally thought to be isolated in the mountains of west Texas), here are the currently recognized characteristics (from Dixon [2000]).
pictigaster - "Venter usually heavily mottled with dark color; subcaudals in males usually 57 or more, females usually 52 or more."
laticinctus - "Venter usually uniform tan or orangish pink, with or without a few dark markings; subcaudals in males usually 54 or fewer, females usually 52 or fewer."
The variation in subcaudals between the two forms is actually clinal: higher in the west to lower in the east. Apparently because of that, Roger Conant later on did not stress the use of subcaudal counts as a diagnosing feature between them.
He preferred instead to rely upon the dark ventral coloration (highly variable) and the supposed presence of a light-colored inverted "U" at the base of the dark crossbands to diagnose pictigaster. Unfortunately, this characteristic occasionally turns up in laticinctus from the southern portion of their range.
So, in summary, if you do not know the locality of your specimen, the subcaudal counts are probably the best character in attempting to separate these two subspecies.
Tom Lott