diff --git a/HISTORY.rst b/HISTORY.rst
index de6ba5c4..4694e495 100644
--- a/HISTORY.rst
+++ b/HISTORY.rst
@@ -10,6 +10,15 @@ dev
 
 **Bugfixes**
 
+- Parsing empty ``Link`` headers with ``parse_header_links()`` no longer return one bogus entry
+
+2.18.4 (2017-08-15)
++++++++++++++++++++
+
+**Improvements**
+
+- Error messages for invalid headers now include the header name for easier debugging
+
 **Dependencies**
 
 - We now support idna v2.6.
diff --git a/requests/__version__.py b/requests/__version__.py
index 2d60bcc0..dc33eef6 100644
--- a/requests/__version__.py
+++ b/requests/__version__.py
@@ -5,8 +5,8 @@
 __title__ = 'requests'
 __description__ = 'Python HTTP for Humans.'
 __url__ = 'http://python-requests.org'
-__version__ = '2.18.3'
-__build__ = 0x021803
+__version__ = '2.18.4'
+__build__ = 0x021804
 __author__ = 'Kenneth Reitz'
 __author_email__ = 'me@kennethreitz.org'
 __license__ = 'Apache 2.0'
diff --git a/requests/utils.py b/requests/utils.py
index 1e4960d7..c52ce2d0 100644
--- a/requests/utils.py
+++ b/requests/utils.py
@@ -743,7 +743,7 @@ def default_headers():
 
 
 def parse_header_links(value):
-    """Return a dict of parsed link headers proxies.
+    """Return a list of parsed link headers proxies.
 
     i.e. Link: <http:/.../front.jpeg>; rel=front; type="image/jpeg",<http://.../back.jpeg>; rel=back;type="image/jpeg"
 
@@ -754,6 +754,10 @@ def parse_header_links(value):
 
     replace_chars = ' \'"'
 
+    value = value.strip(replace_chars)
+    if not value:
+        return links
+
     for val in re.split(', *<', value):
         try:
             url, params = val.split(';', 1)
@@ -868,8 +872,8 @@ def check_header_validity(header):
         if not pat.match(value):
             raise InvalidHeader("Invalid return character or leading space in header: %s" % name)
     except TypeError:
-        raise InvalidHeader("Header value %s must be of type str or bytes, "
-                            "not %s" % (value, type(value)))
+        raise InvalidHeader("Value for header {%s: %s} must be of type str or "
+                            "bytes, not %s" % (name, value, type(value)))
 
 
 def urldefragauth(url):
diff --git a/tests/test_requests.py b/tests/test_requests.py
index 7ef6bfee..a2b2213f 100644
--- a/tests/test_requests.py
+++ b/tests/test_requests.py
@@ -1401,14 +1401,17 @@ class TestRequests:
         headers_list = {'baz': ['foo', 'bar']}
 
         # Test for int
-        with pytest.raises(InvalidHeader):
+        with pytest.raises(InvalidHeader) as excinfo:
             r = requests.get(httpbin('get'), headers=headers_int)
+        assert 'foo' in str(excinfo.value)
         # Test for dict
-        with pytest.raises(InvalidHeader):
+        with pytest.raises(InvalidHeader) as excinfo:
             r = requests.get(httpbin('get'), headers=headers_dict)
+        assert 'bar' in str(excinfo.value)
         # Test for list
-        with pytest.raises(InvalidHeader):
+        with pytest.raises(InvalidHeader) as excinfo:
             r = requests.get(httpbin('get'), headers=headers_list)
+        assert 'baz' in str(excinfo.value)
 
     def test_header_no_return_chars(self, httpbin):
         """Ensure that a header containing return character sequences raise an
diff --git a/tests/test_utils.py b/tests/test_utils.py
index b3f398ee..32e4d4a5 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -498,6 +498,10 @@ def test_iter_slices(value, length):
                 {'url': 'http://.../back.jpeg'}
             ]
         ),
+        (
+            '',
+            []
+        ),
     ))
 def test_parse_header_links(value, expected):
     assert parse_header_links(value) == expected