diff --git a/generators.html b/generators.html
index 61547f2..5633342 100755
--- a/generators.html
+++ b/generators.html
@@ -283,7 +283,7 @@ rules = []
The build_match_and_apply_functions() function has not changed. You’re still using closures to build two functions dynamically that use variables defined in the outer function.
The global open() function opens a file and returns a file object. In this case, the file we’re opening contains the pattern strings for pluralizing nouns. The with statement creates what’s called a context: when the with block ends, Python will automatically close the file, even if an exception is raised inside the with block. You’ll learn more about with blocks and file objects in the Files chapter.
The for line in <fileobject> idiom reads data from the open file, one line at a time, and assigns the text to the line variable. You’ll learn more about reading from files in the Files chapter.
-Each line in the file really has three values, but they’re separated by whitespace (tabs or spaces, it makes no difference). To split it out, use the split() string method. The first argument to the split() method is None, which means “split on any whitespace (tabs or spaces, it makes no difference).” The second argument is 3, which means “split on whitespace 3 times, then discard the rest of the line.” A line like [sxz]$ $ es will be broken up into the list ['[sxz]$', '$', 'es'], which means that pattern will get '[sxz]$', search will get '$', and replace will get 'es'. That’s a lot of power in one little line of code.
+Each line in the file really has three values, but they’re separated by whitespace (tabs or spaces, it makes no difference). To split it out, use the split() string method. The first argument to the split() method is None, which means “split on any whitespace (tabs or spaces, it makes no difference).” The second argument is 3, which means “split on whitespace 3 times, then leave the rest of the line alone.” A line like [sxz]$ $ es will be broken up into the list ['[sxz]$', '$', 'es'], which means that pattern will get '[sxz]$', search will get '$', and replace will get 'es'. That’s a lot of power in one little line of code.
Finally, you pass pattern, search, and replace to the build_match_and_apply_functions() function, which returns a tuple of functions. You append this tuple to the rules list, and rules ends up storing the list of match and apply functions that the plural() function expects.