diff --git a/importlib_resources/_compat.py b/importlib_resources/_compat.py
index 64773f73..becff6f7 100644
--- a/importlib_resources/_compat.py
+++ b/importlib_resources/_compat.py
@@ -106,9 +106,16 @@ def _native_reader(spec):
             reader = _available_reader(spec)
             return reader if hasattr(reader, 'files') else None
 
+        def _namespace_reader(spec):
+            from . import namespace
+            if 'NamespaceLoader' not in spec.loader.__class__.__name__:
+                return
+            return namespace.Multiplexed.load(spec.submodule_search_locations)
+
         return (
             # native reader if it supplies 'files'
             _native_reader(self.spec) or
+            _namespace_reader(self.spec) or
             # local ZipReader if a zip module
             _zip_reader(self.spec) or
             # local FileReader
diff --git a/importlib_resources/namespace.py b/importlib_resources/namespace.py
new file mode 100644
index 00000000..e664aa87
--- /dev/null
+++ b/importlib_resources/namespace.py
@@ -0,0 +1,70 @@
+import os
+import itertools
+from ._compat import suppress, ZipPath, Path
+from .abc import Traversable
+
+
+def infer_path(path):
+    return resolve_zip_path(path) or Path(path)
+
+
+def resolve_zip_path(candidate, tail=''):
+    if not candidate:
+        return
+    try:
+        return ZipPath(candidate, at=tail)
+    except Exception:
+        new_tail = os.path.basename(candidate) + '/' + tail
+        new_base = os.path.dirname(candidate)
+        return resolve_zip_path(new_base, new_tail)
+
+
+class Multiplexed(Traversable):
+    """
+    Given a series of Traversable objects, implement a merged
+    version of the interface across all objects. Useful for
+    namespace packages which may be multihomed at a single
+    name.
+    """
+    @classmethod
+    def load(cls, paths):
+        return cls(map(cls._infer_path, paths))
+
+    def __init__(self, *paths):
+        self._paths = paths
+
+    def iterdir(self):
+        return itertools.chain.from_iterable(
+            path.iterdir() for path in self._paths)
+
+    def read_bytes(self):
+        return self.open(mode='rb').read()
+
+    def read_text(self, *args, **kwargs):
+        return self.open(mode='r', *args, **kwargs).read()
+
+    def is_dir(self):
+        return any(path.is_dir() for path in self._paths)
+
+    def is_file(self):
+        return any(path.is_file() for path in self._paths)
+
+    def joinpath(self, child):
+        children = (
+            path.joinpath(child)
+            for path in self._paths
+            )
+        existing = (
+            child
+            for child in children
+            if child.is_dir() or child.is_file()
+            )
+        return Multiplexed(*existing)
+
+    __truediv__ = joinpath
+
+    def open(self, *args, **kwargs):
+        for path in self._paths[:-1]:
+            with suppress(Exception):
+                return path.open(*args, **kwargs)
+        return self._paths[-1].open(*args, **kwargs)