diff --git a/doc/source/release.rst b/doc/source/release.rst
index 0da7337977851..bd3fbf53de039 100644
--- a/doc/source/release.rst
+++ b/doc/source/release.rst
@@ -474,6 +474,7 @@ Bug Fixes
     explicitly passing labels (:issue:`3415`)
   - Fixed wrong check for overlapping in ``DatetimeIndex.union`` (:issue:`4564`)
   - Fixed conflict between thousands separator and date parser in csv_parser (:issue:`4678`)
+  - Fix appending when dtypes are not the same (error showing mixing float/np.datetime64) (:issue:`4993`)
 
 pandas 0.12.0
 -------------
diff --git a/pandas/tools/merge.py b/pandas/tools/merge.py
index 2cc9d586a05a3..5792161e0171e 100644
--- a/pandas/tools/merge.py
+++ b/pandas/tools/merge.py
@@ -17,13 +17,16 @@
 from pandas.core.internals import (IntBlock, BoolBlock, BlockManager,
                                    make_block, _consolidate)
 from pandas.util.decorators import cache_readonly, Appender, Substitution
-from pandas.core.common import PandasError, ABCSeries
+from pandas.core.common import (PandasError, ABCSeries,
+                                is_timedelta64_dtype, is_datetime64_dtype,
+                                is_integer_dtype)
+
 import pandas.core.common as com
 
 import pandas.lib as lib
 import pandas.algos as algos
 import pandas.hashtable as _hash
-
+import pandas.tslib as tslib
 
 @Substitution('\nleft : DataFrame')
 @Appender(_merge_doc, indents=0)
@@ -1128,6 +1131,8 @@ def _concat_blocks(self, blocks):
             return block
 
     def _concat_single_item(self, objs, item):
+        # this is called if we don't have consistent dtypes in a row-wise append
+
         all_values = []
         dtypes = set()
 
@@ -1141,22 +1146,57 @@ def _concat_single_item(self, objs, item):
             else:
                 all_values.append(None)
 
-        # this stinks
-        have_object = False
+        # figure out the resulting dtype of the combination
+        alls = set()
+        seen = []
         for dtype in dtypes:
+            d = dict([ (t,False) for t in ['object','datetime','timedelta','other'] ])
             if issubclass(dtype.type, (np.object_, np.bool_)):
-                have_object = True
-        if have_object:
-            empty_dtype = np.object_
-        else:
-            empty_dtype = np.float64
+                d['object'] = True
+                alls.add('object')
+            elif is_datetime64_dtype(dtype):
+                d['datetime'] = True
+                alls.add('datetime')
+            elif is_timedelta64_dtype(dtype):
+                d['timedelta'] = True
+                alls.add('timedelta')
+            else:
+                d['other'] = True
+                alls.add('other')
+            seen.append(d)
+
+        if 'datetime' in alls or 'timedelta' in alls:
+
+            if 'object' in alls or 'other' in alls:
+                for v, s in zip(all_values,seen):
+                    if s.get('datetime') or s.get('timedelta'):
+                        pass
+
+                    # if we have all null, then leave a date/time like type
+                    # if we have only that type left
+                    elif isnull(v).all():
+
+                        alls.remove('other')
+                        alls.remove('object')
+
+        # create the result
+        if 'object' in alls:
+            empty_dtype, fill_value = np.object_, np.nan
+        elif 'other' in alls:
+            empty_dtype, fill_value = np.float64, np.nan
+        elif 'datetime' in alls:
+            empty_dtype, fill_value = 'M8[ns]', tslib.iNaT
+        elif 'timedelta' in alls:
+            empty_dtype, fill_value = 'm8[ns]', tslib.iNaT
+        else: # pragma
+            raise AssertionError("invalid dtype determination in concat_single_item")
 
         to_concat = []
         for obj, item_values in zip(objs, all_values):
             if item_values is None:
                 shape = obj.shape[1:]
                 missing_arr = np.empty(shape, dtype=empty_dtype)
-                missing_arr.fill(np.nan)
+                missing_arr.fill(fill_value)
                 to_concat.append(missing_arr)
             else:
                 to_concat.append(item_values)
diff --git a/pandas/tools/tests/test_merge.py b/pandas/tools/tests/test_merge.py
index 0eeb68c4691eb..203769e731022 100644
--- a/pandas/tools/tests/test_merge.py
+++ b/pandas/tools/tests/test_merge.py
@@ -742,6 +742,30 @@ def test_merge_nan_right(self):
         assert_frame_equal(result, expected)
 
 
+    def test_append_dtype_coerce(self):
+
+        # GH 4993
+        # appending with datetime will incorrectly convert datetime64
+        import datetime as dt
+        from pandas import NaT
+
+        df1 = DataFrame(index=[1,2], data=[dt.datetime(2013,1,1,0,0),
+                                           dt.datetime(2013,1,2,0,0)],
+                        columns=['start_time'])
+        df2 = DataFrame(index=[4,5], data=[[dt.datetime(2013,1,3,0,0),
+                                            dt.datetime(2013,1,3,6,10)],
+                                           [dt.datetime(2013,1,4,0,0),
+                                            dt.datetime(2013,1,4,7,10)]],
+                        columns=['start_time','end_time'])
+
+        expected = concat([
+            Series([NaT,NaT,dt.datetime(2013,1,3,6,10),dt.datetime(2013,1,4,7,10)],name='end_time'),
+            Series([dt.datetime(2013,1,1,0,0),dt.datetime(2013,1,2,0,0),dt.datetime(2013,1,3,0,0),dt.datetime(2013,1,4,0,0)],name='start_time'),
+            ],axis=1)
+        result = df1.append(df2,ignore_index=True)
+        assert_frame_equal(result, expected)
+
+
     def test_overlapping_columns_error_message(self):
         # #2649
         df = DataFrame({'key': [1, 2, 3],