diff --git a/doc/whats-new.rst b/doc/whats-new.rst
index b3130a1a4ab..fcfe4887485 100644
--- a/doc/whats-new.rst
+++ b/doc/whats-new.rst
@@ -64,6 +64,10 @@ Bug fixes
 - Fixed a bug in decode_cf_datetime where ``int32`` arrays weren't parsed
   correctly (:issue:`2002`).
   By `Fabien Maussion <https://github.com/fmaussion>`_.
+- When calling `xr.auto_combine()` or `xr.open_mfdataset()` with a `concat_dim`,
+  the resulting dataset will have that one-element dimension (it was
+  silently dropped, previously) (:issue:`1988`).
+  By `Ben Root <https://github.com/WeatherGod>`_.
 
 .. _whats-new.0.10.2:
 
diff --git a/xarray/core/combine.py b/xarray/core/combine.py
index 8c1c58e9a40..430f0e564d6 100644
--- a/xarray/core/combine.py
+++ b/xarray/core/combine.py
@@ -340,7 +340,8 @@ def _dataarray_concat(arrays, dim, data_vars, coords, compat,
 
 
 def _auto_concat(datasets, dim=None, data_vars='all', coords='different'):
-    if len(datasets) == 1:
+    if len(datasets) == 1 and dim is None:
+        # There is nothing more to combine, so kick out early.
         return datasets[0]
     else:
         if dim is None:
diff --git a/xarray/tests/test_backends.py b/xarray/tests/test_backends.py
index e0f030368bb..a58104deef5 100644
--- a/xarray/tests/test_backends.py
+++ b/xarray/tests/test_backends.py
@@ -2039,6 +2039,20 @@ def test_open_dataset(self):
                 self.assertIsInstance(actual.foo.variable.data, np.ndarray)
                 assert_identical(original, actual)
 
+    def test_open_single_dataset(self):
+        # Test for issue GH #1988. This makes sure that the
+        # concat_dim is utilized when specified in open_mfdataset().
+        rnddata = np.random.randn(10)
+        original = Dataset({'foo': ('x', rnddata)})
+        dim = DataArray([100], name='baz', dims='baz')
+        expected = Dataset({'foo': (('baz', 'x'), rnddata[np.newaxis, :])},
+                           {'baz': [100]})
+        with create_tmp_file() as tmp:
+            original.to_netcdf(tmp)
+            with open_mfdataset([tmp], concat_dim=dim,
+                                autoclose=self.autoclose) as actual:
+                assert_identical(expected, actual)
+
     def test_dask_roundtrip(self):
         with create_tmp_file() as tmp:
             data = create_test_data()
diff --git a/xarray/tests/test_combine.py b/xarray/tests/test_combine.py
index 09918d9a065..482a280b355 100644
--- a/xarray/tests/test_combine.py
+++ b/xarray/tests/test_combine.py
@@ -377,3 +377,22 @@ def test_auto_combine_no_concat(self):
         data = Dataset({'x': 0})
         actual = auto_combine([data, data, data], concat_dim=None)
         assert_identical(data, actual)
+
+        # Single object, with a concat_dim explicitly provided
+        # Test the issue reported in GH #1988
+        objs = [Dataset({'x': 0, 'y': 1})]
+        dim = DataArray([100], name='baz', dims='baz')
+        actual = auto_combine(objs, concat_dim=dim)
+        expected = Dataset({'x': ('baz', [0]), 'y': ('baz', [1])},
+                           {'baz': [100]})
+        assert_identical(expected, actual)
+
+        # Just making sure that auto_combine is doing what is
+        # expected for non-scalar values, too.
+        objs = [Dataset({'x': ('z', [0, 1]), 'y': ('z', [1, 2])})]
+        dim = DataArray([100], name='baz', dims='baz')
+        actual = auto_combine(objs, concat_dim=dim)
+        expected = Dataset({'x': (('baz', 'z'), [[0, 1]]),
+                            'y': (('baz', 'z'), [[1, 2]])},
+                           {'baz': [100]})
+        assert_identical(expected, actual)