diff -r e281a57d5b29 Doc/library/urllib.request.rst
--- a/Doc/library/urllib.request.rst	Tue Apr 19 08:53:14 2016 +0200
+++ b/Doc/library/urllib.request.rst	Wed Apr 20 20:44:39 2016 +0200
@@ -166,6 +166,8 @@
    in a case insensitive approach, for all operating systems first, and when it
    cannot find it, looks for proxy information from Mac OSX System
    Configuration for Mac OS X and Windows Systems Registry for Windows.
+   If both, lowercase and uppercase environment variables exist (and disagree),
+   lowercase is preferred.
 
 
 The following classes are provided:
diff -r e281a57d5b29 Lib/urllib/request.py
--- a/Lib/urllib/request.py	Tue Apr 19 08:53:14 2016 +0200
+++ b/Lib/urllib/request.py	Wed Apr 20 20:44:39 2016 +0200
@@ -2455,10 +2455,19 @@
 
     """
     proxies = {}
+    # in order to prefer lowercase variables, process environment in
+    # two passes: first matches any, second pass matches lowercase only
     for name, value in os.environ.items():
         name = name.lower()
         if value and name[-6:] == '_proxy':
             proxies[name[:-6]] = value
+    for name, value in os.environ.items():
+        if name[-6:] == '_proxy':
+            name = name.lower()
+            if value:
+                proxies[name[:-6]] = value
+            else:
+                proxies.pop(name[:-6], None)
     return proxies
 
 def proxy_bypass_environment(host):
@@ -2467,6 +2476,7 @@
     Checks the environment for a variable named no_proxy, which should
     be a list of DNS suffixes separated by commas, or '*' for all hosts.
     """
+    # unlike no_proxy, the other vars are handled as mixed case variants also
     no_proxy = os.environ.get('no_proxy', '') or os.environ.get('NO_PROXY', '')
     # '*' is special case for always bypass
     if no_proxy == '*':