diff --git a/src/lexer.test.ts b/src/lexer.test.ts
index ddcc3fe..08578e7 100644
--- a/src/lexer.test.ts
+++ b/src/lexer.test.ts
@@ -81,6 +81,10 @@ test('Lexer passes ident fixtures', () => {
 
     // Escaped idents
     fix('a\\ b\\$c\\.', [t.Ident('a b$c.'), t.EOS()], false),
+    fix('\\$a\\ b\\$c\\.', [t.Ident('$a b$c.'), t.EOS()], false),
+
+    // Underscores
+    fix('_a_b_c', [t.Ident('_a_b_c'), t.EOS()], false),
   ])
 })
 
diff --git a/src/lexer.ts b/src/lexer.ts
index c3c6222..dc7db18 100644
--- a/src/lexer.ts
+++ b/src/lexer.ts
@@ -78,7 +78,7 @@ export class Lexer {
         }
       }
 
-      if (isAlpha(char) || char === '!' || char === '=' || char === '>' || char === '<') {
+      if (isAlpha(char) || char === '!' || char === '=' || char === '>' || char === '<' || char === '\\' || char === '_') {
         tokens.push(this.lexOperatorOrConditional(char))
         continue
       }
@@ -235,9 +235,13 @@ export class Lexer {
   }
 
   private lexIdent(previous: string): Token {
+    /* this function differs from the go implementation in that the go impl
+     hasn't yet advanced the character so will do a peek/next loop.  here we
+     have already done a next() so need to check to see if the first char is
+     an escape sequence */
     let str = ''
-    while (isIdent(this.peek())) {
-      let { char } = this.next()
+    let char = previous
+    while (true) {
 
       // Allow escaping of any character except EOS
       if (char === '\\') {
@@ -252,6 +256,11 @@ export class Lexer {
       if (str.length >= MAXIMUM_INDENT_LENGTH) {
         throw new LexerError('unreasonable literal length', this.cursor)
       }
+
+      if (!isIdent(this.peek())) {
+        break
+      }
+      char = this.next().char
     }
 
     const comingUp: string = this.peek()
@@ -271,7 +280,7 @@ export class Lexer {
       )
     }
 
-    return t.Ident(previous + str)
+    return t.Ident(str)
   }
 
   /**