diff --git a/README.md b/README.md
index 5111fb1..9a59ef6 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,7 @@ The fips module simply wraps a huge pile of static data from the US Census Burea
 
 It is named after the FIPS codes used by the USCB and some other federal entities to refer to specific geographic regions.
 
-Everything worth doing in the module goes through its `states` table: `fips.states`.  By default, that table is indexed by states' postal abbreviations/codes (AL, AR, MS, etc.), which enables quick, human access to a state's info via `fips.states.[postal code here]`. The index is case-insensitive by default; so `fips.states.ak` works just as well as `fips.states.AK`, for example.  Each state is a dict that treats keys case-insensitively, treats unknown attributes as keys, and has a key `COUNTIES` that maps to a table of the county information for all the counties in that state. Those county tables are, by default, indexed by the counties' FIPS codes (`'001'`, `'003'`, etc.). Those tables use the same sort of case-insensitive, attribute-defering dict that the overall `states` table uses, but one other feature of these tables is that calling the table with a single parameter will look that parameter up in the index, which gets around the inability to use digits at the start of attribute names.  Each county dict has a `STATE` key that points back to the state dict that the county (or county-equivalent) is part of.
+Everything worth doing in the module goes through its `states` table: `fips.states`.  By default, that table is indexed by states' postal abbreviations/codes (AL, AR, MS, etc.), which enables quick, human access to a state's info via `fips.states.[postal code here]`. The index is case-insensitive by default; so `fips.states.ak` works just as well as `fips.states.AK`, for example.  Each state is a dict that treats keys case-insensitively, treats unknown attributes as keys, and has a key `COUNTIES` that maps to a table of the county information for all the counties (or county-equivalents) in that state. Those county tables are indexed by the counties' FIPS codes (`'001'`, `'003'`, etc.). Those county tables use the same sort of case-insensitive, attribute-defering dict that the overall `states` table uses, but one other feature of these county tables is that calling the table with a single parameter will look that parameter up in the index, which gets around the inability to use digits at the start of attribute names.  Each county dict has a `STATE` key that points back to the state dict that the county (or county-equivalent) is part of.
 
     #For instance, 
     tangipahoa_parish = fips.states.la.counties('105')