diff --git a/src/dask_nested/accessor.py b/src/dask_nested/accessor.py
index 11082fe..ceb8fb9 100644
--- a/src/dask_nested/accessor.py
+++ b/src/dask_nested/accessor.py
@@ -35,7 +35,7 @@ def _check_series(series):
     def fields(self) -> list[str]:
         """Names of the nested columns"""
 
-        return self._series.head(0).nest.fields  # hacky
+        return list(self._series.dtype.fields)
 
     def to_lists(self, fields: list[str] | None = None) -> dd.DataFrame:
         """Convert nested series into dataframe of list-array columns
diff --git a/src/dask_nested/backends.py b/src/dask_nested/backends.py
index 991d182..bd589f2 100644
--- a/src/dask_nested/backends.py
+++ b/src/dask_nested/backends.py
@@ -21,7 +21,9 @@
 @make_meta_dispatch.register(npd.NestedFrame)
 def make_meta_frame(x, index=None) -> npd.NestedFrame:
     """Create an empty NestedFrame to use as Dask's underlying object meta."""
-    result = x.head(0)
+
+    dtypes = x.dtypes.to_dict()
+    result = npd.NestedFrame({key: pd.Series(dtype=d) for key, d in dtypes.items()})
     return result
 
 
diff --git a/src/dask_nested/core.py b/src/dask_nested/core.py
index 2f3cc80..a93b3da 100644
--- a/src/dask_nested/core.py
+++ b/src/dask_nested/core.py
@@ -83,7 +83,9 @@ def from_nested_pandas(
             the size and index of the dataframe, the output may have fewer
             partitions than requested.
         chunksize: `int`, optional
-            Size of the individual chunks of data in non-parallel objects that make up Dask frames.
+            The desired number of rows per index partition to use. Note that
+            depending on the size and index of the dataframe, actual partition
+            sizes may vary.
         sort: `bool`, optional
             Whether to sort the frame by a default index.
 
@@ -133,7 +135,7 @@ def nested_columns(self) -> list:
                 nest_cols.append(column)
         return nest_cols
 
-    def add_nested(self, nested, name) -> NestedFrame:  # type: ignore[name-defined] # noqa: F821
+    def add_nested(self, nested, name, how="outer") -> NestedFrame:  # type: ignore[name-defined] # noqa: F821
         """Packs a dataframe into a nested column
 
         Parameters
@@ -142,13 +144,29 @@ def add_nested(self, nested, name) -> NestedFrame:  # type: ignore[name-defined]
             A flat dataframe to pack into a nested column
         name:
             The name given to the nested column
+        how: {‘left’, ‘right’, ‘outer’, ‘inner’, ‘cross’}, default ‘outer’
+            How to handle the operation of the two objects.
+
+            * left: use calling frame’s index (or column if on is specified)
+
+            * right: use other’s index.
+
+            * outer: form union of calling frame’s index (or column if on is
+            specified) with other’s index, and sort it lexicographically.
+
+            * inner: form intersection of calling frame’s index (or column if
+            on is specified) with other’s index, preserving the order of the
+            calling’s one.
+
+            * cross: creates the cartesian product from both frames, preserves
+            the order of the left keys.
 
         Returns
         -------
         `dask_nested.NestedFrame`
         """
         nested = nested.map_partitions(lambda x: pack_flat(x)).rename(name)
-        return self.join(nested, how="outer")
+        return self.join(nested, how=how)
 
     def query(self, expr) -> Self:  # type: ignore # noqa: F821:
         """