diff --git a/lib/time.rb b/lib/time.rb
index b157e60..776c82c 100644
--- a/lib/time.rb
+++ b/lib/time.rb
@@ -658,7 +658,7 @@ def xmlschema(time)
     def rfc3339(time)
       pattern = /\A\s*
           (-?\d{4})-(\d\d)-(\d\d)
-          [T\s]
+          [T ]
           (\d\d):(\d\d):(\d\d)
           (\.\d+)?
           (Z|[+-]\d\d:\d\d)
diff --git a/test/test_time.rb b/test/test_time.rb
index 53ac856..4054bc4 100644
--- a/test/test_time.rb
+++ b/test/test_time.rb
@@ -607,6 +607,29 @@ def test_huge_precision
     define_method(test.sub(/xmlschema/, 'rfc3339')) {__send__(sub, :rfc3339)}
   end
 
+  def test_rfc3339_separator
+    # RFC 3339 section 5.6 defines the separator as "T", and only notes that an
+    # application may use a space for readability. The other \s characters are
+    # not permitted, and Time.xmlschema has never accepted any of them.
+    t = Time.utc(2011, 10, 5, 22, 26, 12)
+    assert_equal(t, Time.rfc3339("2011-10-05T22:26:12Z"))
+    assert_equal(t, Time.rfc3339("2011-10-05 22:26:12Z"))
+
+    ["\t", "\n", "\v", "\f", "\r"].each do |sep|
+      s = "2011-10-05#{sep}22:26:12Z"
+      e = assert_raise(ArgumentError, "separator #{sep.inspect}") { Time.rfc3339(s) }
+      assert_match(/invalid rfc3339 format/, e.message, "separator #{sep.inspect}")
+    end
+
+    # Time.xmlschema keeps rejecting every separator but "T".
+    assert_equal(t, Time.xmlschema("2011-10-05T22:26:12Z"))
+    ["\t", "\n", "\v", "\f", "\r", " "].each do |sep|
+      assert_raise(ArgumentError, "separator #{sep.inspect}") do
+        Time.xmlschema("2011-10-05#{sep}22:26:12Z")
+      end
+    end
+  end
+
   def test_parse_with_various_object
     d  = Date.new(2010, 10, 28)
     dt = DateTime.new(2010, 10, 28)