diff --git a/include/mrk/lexer.h b/include/mrk/lexer.h
index ae4baa3..d8782fa 100644
--- a/include/mrk/lexer.h
+++ b/include/mrk/lexer.h
@@ -19,6 +19,7 @@ typedef enum mrk_token_type {
   mrk_token_type_dashes,
   mrk_token_type_underscores,
   mrk_token_type_stars,
+  mrk_token_type_equals,
   mrk_token_type_blank_line,
   mrk_token_type_space,
   mrk_token_type_line_break,
diff --git a/src/lexer/lexer.c b/src/lexer/lexer.c
index 7d56aa9..3dd2463 100644
--- a/src/lexer/lexer.c
+++ b/src/lexer/lexer.c
@@ -124,7 +124,8 @@ mrk_lexer_err mrk_lexer_next(mrk_token *out, mrk_lexer *lexer) {
     char c = mrk_lexer_advance(lexer);
     switch (c) {
     // All these characters have multiple meanings depending on their location
-    // in the file and how many there are
+    // in the file and how many there are, so the lexer can only match them as
+    // one or more grouped characters
     case '#':
       mrk_lexer_advance_eq(lexer, c);
       mrk_lexer_emit(out, lexer, mrk_token_type_pounds);
@@ -145,6 +146,10 @@ mrk_lexer_err mrk_lexer_next(mrk_token *out, mrk_lexer *lexer) {
       mrk_lexer_advance_eq(lexer, c);
       mrk_lexer_emit(out, lexer, mrk_token_type_stars);
       break;
+    case '=':
+      mrk_lexer_advance_eq(lexer, c);
+      mrk_lexer_emit(out, lexer, mrk_token_type_equals);
+      break;
       // Two consecutive newlines constitute a blank line, otherwise they're
       // ignored as whitespace
     case '\n':