diff --git a/lib/iris/_lazy_data.py b/lib/iris/_lazy_data.py
index 824b8ed27b1..6d668ea554a 100644
--- a/lib/iris/_lazy_data.py
+++ b/lib/iris/_lazy_data.py
@@ -174,12 +174,20 @@ def wrap_lazy_elementwise(lazy_array, elementwise_op):
     * elementwise_op:
         The elementwise (numpy) array operation to apply.
 
+    .. note:
+
+        A single-point "dummy" call is made to the operation function, to
+        determine dtype of the result.
+        This return dtype must be stable in actual operation (!)
+
     """
-    # This is just an Iris wrapper for the Dask operation.
-    # For now, we support only a single argument array, and assume that the
-    # output dtype is the same as the input.  This scope is intentionally
-    # minimal : we can extend it later as needed.
-
-    # Note: pass dtype, to prevent Dask making a test call to work it out.
-    return da.map_blocks(elementwise_op,
-                         lazy_array, dtype=lazy_array.dtype)
+    # This is just a wrapper to provide an Iris-specific abstraction for a
+    # lazy operation in Dask (map_blocks).
+
+    # Explicitly determine the return type with a dummy call.
+    # This makes good practical sense for unit conversions, as a Unit.convert
+    # call may cast to float, or not, depending on unit equality : Thus, it's
+    # much safer to get cf_units to decide that for us.
+    dtype = elementwise_op(np.zeros(1, lazy_array.dtype)).dtype
+
+    return da.map_blocks(elementwise_op, lazy_array, dtype=dtype)
diff --git a/lib/iris/tests/unit/lazy_data/test_wrap_lazy_elementwise.py b/lib/iris/tests/unit/lazy_data/test_wrap_lazy_elementwise.py
new file mode 100644
index 00000000000..01f8ce16795
--- /dev/null
+++ b/lib/iris/tests/unit/lazy_data/test_wrap_lazy_elementwise.py
@@ -0,0 +1,68 @@
+# (C) British Crown Copyright 2018, Met Office
+#
+# This file is part of Iris.
+#
+# Iris is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the
+# Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Iris is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public License
+# along with Iris.  If not, see <http://www.gnu.org/licenses/>.
+"""Test function :func:`iris._lazy data.wrap_lazy_elementwise`."""
+
+from __future__ import (absolute_import, division, print_function)
+from six.moves import (filter, input, map, range, zip)  # noqa
+
+# Import iris.tests first so that some things can be initialised before
+# importing anything else.
+import iris.tests as tests
+
+import numpy as np
+
+from iris._lazy_data import as_lazy_data, is_lazy_data
+
+from iris._lazy_data import wrap_lazy_elementwise
+
+
+def _test_elementwise_op(array):
+    # Promotes the type of a bool argument, but not a float.
+    return array + 1
+
+
+class Test_is_lazy_data(tests.IrisTest):
+    def test_basic(self):
+        concrete_array = np.arange(30).reshape((2, 5, 3))
+        lazy_array = as_lazy_data(concrete_array)
+        wrapped = wrap_lazy_elementwise(lazy_array,
+                                        _test_elementwise_op)
+        self.assertTrue(is_lazy_data(wrapped))
+        self.assertArrayAllClose(wrapped.compute(),
+                                 _test_elementwise_op(concrete_array))
+
+    def test_dtype_same(self):
+        concrete_array = np.array([3.], dtype=np.float16)
+        lazy_array = as_lazy_data(concrete_array)
+        wrapped = wrap_lazy_elementwise(lazy_array,
+                                        _test_elementwise_op)
+        self.assertTrue(is_lazy_data(wrapped))
+        self.assertEqual(wrapped.dtype, np.float16)
+        self.assertEqual(wrapped.compute().dtype, np.float16)
+
+    def test_dtype_change(self):
+        concrete_array = np.array([True, False])
+        lazy_array = as_lazy_data(concrete_array)
+        wrapped = wrap_lazy_elementwise(lazy_array,
+                                        _test_elementwise_op)
+        self.assertTrue(is_lazy_data(wrapped))
+        self.assertEqual(wrapped.dtype, np.int)
+        self.assertEqual(wrapped.compute().dtype, wrapped.dtype)
+
+
+if __name__ == '__main__':
+    tests.main()