diff --git a/google/cloud/bigquery/_pandas_helpers.py b/google/cloud/bigquery/_pandas_helpers.py
index 2ff96da4d..b381fa5f7 100644
--- a/google/cloud/bigquery/_pandas_helpers.py
+++ b/google/cloud/bigquery/_pandas_helpers.py
@@ -93,6 +93,8 @@ def pyarrow_numeric():
 
 
 def pyarrow_bignumeric():
+    # 77th digit is partial.
+    # https://cloud.google.com/bigquery/docs/reference/standard-sql/data-types#decimal_types
     return pyarrow.decimal256(76, 38)
 
 
diff --git a/tests/data/scalars.jsonl b/tests/data/scalars.jsonl
new file mode 100644
index 000000000..4419a6e9a
--- /dev/null
+++ b/tests/data/scalars.jsonl
@@ -0,0 +1,2 @@
+{"bool_col": true, "bytes_col": "abcd", "date_col": "2021-07-21", "datetime_col": "2021-07-21 11:39:45", "geography_col": "POINT(-122.0838511 37.3860517)", "int64_col": "123456789", "numeric_col": "1.23456789", "bignumeric_col": "10.111213141516171819", "float64_col": "1.25", "string_col": "Hello, World", "time_col": "11:41:43.07616", "timestamp_col": "2021-07-21T17:43:43.945289Z"}
+{"bool_col": null, "bytes_col": null, "date_col": null, "datetime_col": null, "geography_col": null, "int64_col": null, "numeric_col": null, "bignumeric_col": null, "float64_col": null, "string_col": null, "time_col": null, "timestamp_col": null}
diff --git a/tests/data/scalars_extreme.jsonl b/tests/data/scalars_extreme.jsonl
new file mode 100644
index 000000000..ceccd8dbc
--- /dev/null
+++ b/tests/data/scalars_extreme.jsonl
@@ -0,0 +1,5 @@
+{"bool_col": true, "bytes_col": "DQo=\n", "date_col": "9999-12-31", "datetime_col": "9999-12-31 23:59:59.999999", "geography_col": "POINT(-135.0000 90.0000)", "int64_col": "9223372036854775807", "numeric_col": "9.9999999999999999999999999999999999999E+28", "bignumeric_col": "9.999999999999999999999999999999999999999999999999999999999999999999999999999E+37", "float64_col": "+inf", "string_col": "Hello, World", "time_col": "23:59:59.99999", "timestamp_col": "9999-12-31T23:59:59.999999Z"}
+{"bool_col": false, "bytes_col": "8J+Zgw==\n", "date_col": "0001-01-01", "datetime_col": "0001-01-01 00:00:00", "geography_col": "POINT(45.0000 -90.0000)", "int64_col": "-9223372036854775808", "numeric_col": "-9.9999999999999999999999999999999999999E+28", "bignumeric_col": "-9.999999999999999999999999999999999999999999999999999999999999999999999999999E+37", "float64_col": "-inf", "string_col": "Hello, World", "time_col": "00:00:00", "timestamp_col": "0001-01-01T00:00:00.000000Z"}
+{"bool_col": true, "bytes_col": "AA==\n", "date_col": "1900-01-01", "datetime_col": "1900-01-01 00:00:00", "geography_col": "POINT(-180.0000 0.0000)", "int64_col": "-1", "numeric_col": "0.000000001", "bignumeric_col": "-0.00000000000000000000000000000000000001", "float64_col": "nan", "string_col": "こんにちは", "time_col": "00:00:00.000001", "timestamp_col": "1900-01-01T00:00:00.000000Z"}
+{"bool_col": false, "bytes_col": "", "date_col": "1970-01-01", "datetime_col": "1970-01-01 00:00:00", "geography_col": "POINT(0 0)", "int64_col": "0", "numeric_col": "0.0", "bignumeric_col": "0.0", "float64_col": 0.0, "string_col": "", "time_col": "12:00:00", "timestamp_col": "1970-01-01T00:00:00.000000Z"}
+{"bool_col": null, "bytes_col": null, "date_col": null, "datetime_col": null, "geography_col": null, "int64_col": null, "numeric_col": null, "bignumeric_col": null, "float64_col": null, "string_col": null, "time_col": null, "timestamp_col": null}
diff --git a/tests/data/scalars_schema.json b/tests/data/scalars_schema.json
new file mode 100644
index 000000000..00bd150fd
--- /dev/null
+++ b/tests/data/scalars_schema.json
@@ -0,0 +1,62 @@
+[
+  {
+    "mode": "NULLABLE",
+    "name": "timestamp_col",
+    "type": "TIMESTAMP"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "time_col",
+    "type": "TIME"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "float64_col",
+    "type": "FLOAT"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "datetime_col",
+    "type": "DATETIME"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "bignumeric_col",
+    "type": "BIGNUMERIC"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "numeric_col",
+    "type": "NUMERIC"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "geography_col",
+    "type": "GEOGRAPHY"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "date_col",
+    "type": "DATE"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "string_col",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "bool_col",
+    "type": "BOOLEAN"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "bytes_col",
+    "type": "BYTES"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "int64_col",
+    "type": "INTEGER"
+  }
+]
diff --git a/tests/system/conftest.py b/tests/system/conftest.py
index 7b389013f..cc2c2a4dc 100644
--- a/tests/system/conftest.py
+++ b/tests/system/conftest.py
@@ -12,15 +12,20 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import pytest
+import pathlib
 
-from google.cloud import bigquery
+import pytest
 import test_utils.prefixer
 
+from google.cloud import bigquery
+from google.cloud.bigquery import enums
 from . import helpers
 
+
 prefixer = test_utils.prefixer.Prefixer("python-bigquery", "tests/system")
 
+DATA_DIR = pathlib.Path(__file__).parent.parent / "data"
+
 
 @pytest.fixture(scope="session", autouse=True)
 def cleanup_datasets(bigquery_client: bigquery.Client):
@@ -36,6 +41,11 @@ def bigquery_client():
     return bigquery.Client()
 
 
+@pytest.fixture(scope="session")
+def project_id(bigquery_client: bigquery.Client):
+    return bigquery_client.project
+
+
 @pytest.fixture(scope="session")
 def bqstorage_client(bigquery_client):
     from google.cloud import bigquery_storage
@@ -54,3 +64,37 @@ def dataset_id(bigquery_client):
 @pytest.fixture
 def table_id(dataset_id):
     return f"{dataset_id}.table_{helpers.temp_suffix()}"
+
+
+@pytest.fixture(scope="session")
+def scalars_table(bigquery_client: bigquery.Client, project_id: str, dataset_id: str):
+    schema = bigquery_client.schema_from_json(DATA_DIR / "scalars_schema.json")
+    job_config = bigquery.LoadJobConfig()
+    job_config.schema = schema
+    job_config.source_format = enums.SourceFormat.NEWLINE_DELIMITED_JSON
+    full_table_id = f"{project_id}.{dataset_id}.scalars"
+    with open(DATA_DIR / "scalars.jsonl", "rb") as data_file:
+        job = bigquery_client.load_table_from_file(
+            data_file, full_table_id, job_config=job_config
+        )
+    job.result()
+    yield full_table_id
+    bigquery_client.delete_table(full_table_id)
+
+
+@pytest.fixture(scope="session")
+def scalars_extreme_table(
+    bigquery_client: bigquery.Client, project_id: str, dataset_id: str
+):
+    schema = bigquery_client.schema_from_json(DATA_DIR / "scalars_schema.json")
+    job_config = bigquery.LoadJobConfig()
+    job_config.schema = schema
+    job_config.source_format = enums.SourceFormat.NEWLINE_DELIMITED_JSON
+    full_table_id = f"{project_id}.{dataset_id}.scalars_extreme"
+    with open(DATA_DIR / "scalars_extreme.jsonl", "rb") as data_file:
+        job = bigquery_client.load_table_from_file(
+            data_file, full_table_id, job_config=job_config
+        )
+    job.result()
+    yield full_table_id
+    bigquery_client.delete_table(full_table_id)
diff --git a/tests/system/test_arrow.py b/tests/system/test_arrow.py
new file mode 100644
index 000000000..f97488e39
--- /dev/null
+++ b/tests/system/test_arrow.py
@@ -0,0 +1,88 @@
+# Copyright 2021 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""System tests for Arrow connector."""
+
+import pytest
+
+pyarrow = pytest.importorskip(
+    "pyarrow", minversion="3.0.0"
+)  # Needs decimal256 for BIGNUMERIC columns.
+
+
+@pytest.mark.parametrize(
+    ("max_results", "scalars_table_name"),
+    (
+        (None, "scalars_table"),  # Use BQ Storage API.
+        (10, "scalars_table"),  # Use REST API.
+        (None, "scalars_extreme_table"),  # Use BQ Storage API.
+        (10, "scalars_extreme_table"),  # Use REST API.
+    ),
+)
+def test_list_rows_nullable_scalars_dtypes(
+    bigquery_client,
+    scalars_table,
+    scalars_extreme_table,
+    max_results,
+    scalars_table_name,
+):
+    table_id = scalars_table
+    if scalars_table_name == "scalars_extreme_table":
+        table_id = scalars_extreme_table
+    arrow_table = bigquery_client.list_rows(
+        table_id, max_results=max_results,
+    ).to_arrow()
+
+    schema = arrow_table.schema
+    bignumeric_type = schema.field("bignumeric_col").type
+    # 77th digit is partial.
+    # https://cloud.google.com/bigquery/docs/reference/standard-sql/data-types#decimal_types
+    assert bignumeric_type.precision in {76, 77}
+    assert bignumeric_type.scale == 38
+
+    bool_type = schema.field("bool_col").type
+    assert bool_type.equals(pyarrow.bool_())
+
+    bytes_type = schema.field("bytes_col").type
+    assert bytes_type.equals(pyarrow.binary())
+
+    date_type = schema.field("date_col").type
+    assert date_type.equals(pyarrow.date32())
+
+    datetime_type = schema.field("datetime_col").type
+    assert datetime_type.unit == "us"
+    assert datetime_type.tz is None
+
+    float64_type = schema.field("float64_col").type
+    assert float64_type.equals(pyarrow.float64())
+
+    geography_type = schema.field("geography_col").type
+    assert geography_type.equals(pyarrow.string())
+
+    int64_type = schema.field("int64_col").type
+    assert int64_type.equals(pyarrow.int64())
+
+    numeric_type = schema.field("numeric_col").type
+    assert numeric_type.precision == 38
+    assert numeric_type.scale == 9
+
+    string_type = schema.field("string_col").type
+    assert string_type.equals(pyarrow.string())
+
+    time_type = schema.field("time_col").type
+    assert time_type.equals(pyarrow.time64("us"))
+
+    timestamp_type = schema.field("timestamp_col").type
+    assert timestamp_type.unit == "us"
+    assert timestamp_type.tz is not None