diff --git a/lib/galaxy/datatypes/interval.py b/lib/galaxy/datatypes/interval.py
index 1992479e8bb..578c02e748a 100644
--- a/lib/galaxy/datatypes/interval.py
+++ b/lib/galaxy/datatypes/interval.py
@@ -314,14 +314,14 @@ class Interval( Tabular ):
         >>> Interval().sniff( fname )
         True
         """
-        headers = get_headers( filename, '\t' )
+        headers = get_headers( filename, '\t', comment_designator='#' )
         try:
             """
             If we got here, we already know the file is_column_based and is not bed,
             so we'll just look for some valid data.
             """
             for hdr in headers:
-                if hdr and not hdr[0].startswith( '#' ):
+                if hdr:
                     if len(hdr) < 3:
                         return False
                     try:
@@ -504,12 +504,12 @@ class Bed( Interval ):
         >>> Bed().sniff( fname )
         True
         """
-        headers = get_headers( filename, '\t' )
+        headers = get_headers( filename, '\t', comment_designator='#' )
         try:
             if not headers:
                 return False
             for hdr in headers:
-                if (hdr[0] == '' or hdr[0].startswith( '#' )):
+                if hdr[0] == '':
                     continue
                 valid_col1 = False
                 if len(hdr) < 3 or len(hdr) > 12:
diff --git a/lib/galaxy/datatypes/sniff.py b/lib/galaxy/datatypes/sniff.py
index b2a82e9f8bf..2fa4f42c7c8 100644
--- a/lib/galaxy/datatypes/sniff.py
+++ b/lib/galaxy/datatypes/sniff.py
@@ -198,24 +198,34 @@ def convert_newlines_sep2tabs( fname, in_place=True, patt="\\s+", tmp_dir=None,
         return ( i + 1, temp_name )
 
 
-def get_headers( fname, sep, count=60, is_multi_byte=False ):
+def get_headers( fname, sep, count=60, is_multi_byte=False, comment_designator=None ):
     """
-    Returns a list with the first 'count' lines split by 'sep'
+    Returns a list with the first 'count' lines split by 'sep', ignoring lines
+    starting with 'comment_designator'
 
     >>> fname = get_test_fname('complete.bed')
     >>> get_headers(fname,'\\t')
     [['chr7', '127475281', '127491632', 'NM_000230', '0', '+', '127486022', '127488767', '0', '3', '29,172,3225,', '0,10713,13126,'], ['chr7', '127486011', '127488900', 'D49487', '0', '+', '127486022', '127488767', '0', '2', '155,490,', '0,2399']]
+    >>> fname = get_test_fname('test.gff')
+    >>> get_headers(fname, '\\t', count=5, comment_designator='#')
+    [[''], ['chr7', 'bed2gff', 'AR', '26731313', '26731437', '.', '+', '.', 'score'], ['chr7', 'bed2gff', 'AR', '26731491', '26731536', '.', '+', '.', 'score'], ['chr7', 'bed2gff', 'AR', '26731541', '26731649', '.', '+', '.', 'score'], ['chr7', 'bed2gff', 'AR', '26731659', '26731841', '.', '+', '.', 'score']]
     """
     headers = []
     in_file = compression_utils.get_fileobj(fname)
     try:
-        for idx, line in enumerate(in_file):
+        idx = 0
+        for line in in_file:
             line = line.rstrip('\n\r')
             if is_multi_byte:
                 # TODO: fix this - sep is never found in line
                 line = unicodify( line, 'utf-8' )
                 sep = sep.encode( 'utf-8' )
+                if comment_designator is not None and comment_designator != '':
+                    comment_designator = comment_designator.encode( 'utf-8' )
+            if comment_designator is not None and comment_designator != '' and line.startswith( comment_designator ):
+                continue
             headers.append( line.split(sep) )
+            idx += 1
             if idx == count:
                 break
     finally: