diff --git a/apis/python/src/tiledbsoma/soma_dataframe.py b/apis/python/src/tiledbsoma/soma_dataframe.py
index d39ea5ea70..285829a211 100644
--- a/apis/python/src/tiledbsoma/soma_dataframe.py
+++ b/apis/python/src/tiledbsoma/soma_dataframe.py
@@ -303,6 +303,9 @@ def read_as_pandas_all(
         column_names: Optional[Sequence[str]] = None,
         result_order: Optional[SOMAResultOrder] = None,
     ) -> pd.DataFrame:
+        """
+        This is a convenience method around ``read``. It concatenates all partial read results into a single DataFrame. Its nominal use is to simplify unit-test cases.
+        """
         return pd.concat(
             self.read_as_pandas(
                 ids=ids,
@@ -310,8 +313,6 @@ def read_as_pandas_all(
                 column_names=column_names,
                 result_order=result_order,
             ),
-            ignore_index=True,
-        )
 
     def write_from_pandas(self, dataframe: pd.DataFrame) -> None:
         """