diff --git a/Lib/gettext.py b/Lib/gettext.py
index 4c3b80b0239b0ab..daad346718f68d0 100644
--- a/Lib/gettext.py
+++ b/Lib/gettext.py
@@ -385,6 +385,9 @@ def _parse(self, fp):
                     item = b_item.decode().strip()
                     if not item:
                         continue
+                    # Skip over comment lines:
+                    if item.startswith('#-#-#-#-#') and item.endswith('#-#-#-#-#'):
+                        continue
                     k = v = None
                     if ':' in item:
                         k, v = item.split(':', 1)
diff --git a/Lib/test/test_gettext.py b/Lib/test/test_gettext.py
index b5ed05eab7bbe14..4e4d4c20ff8ca63 100644
--- a/Lib/test/test_gettext.py
+++ b/Lib/test/test_gettext.py
@@ -580,6 +580,19 @@ def test_plural_form_error_issue17898(self):
             # If this runs cleanly, the bug is fixed.
             t = gettext.GNUTranslations(fp)
 
+    def test_ignore_comments_in_headers_issue36239(self):
+        """Checks that comments like:
+
+            #-#-#-#-#  messages.po (EdX Studio)  #-#-#-#-#
+
+        are ignored.
+        """
+        with open(MOFILE, 'wb') as fp:
+            fp.write(base64.decodebytes(GNU_MO_DATA_ISSUE_17898))
+        with open(MOFILE, 'rb') as fp:
+            t = gettext.GNUTranslations(fp)
+            self.assertEqual(t.info()["plural-forms"], "nplurals=2; plural=(n != 1);")
+
 
 class UnicodeTranslationsTest(GettextBaseTest):
     def setUp(self):
diff --git a/Misc/NEWS.d/next/Library/2019-03-09-23-51-27.bpo-36239.BHJ3Ln.rst b/Misc/NEWS.d/next/Library/2019-03-09-23-51-27.bpo-36239.BHJ3Ln.rst
new file mode 100644
index 000000000000000..3a7420291513787
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2019-03-09-23-51-27.bpo-36239.BHJ3Ln.rst
@@ -0,0 +1 @@
+Parsing .mo files now ignores comments starting and ending with #-#-#-#-#.