diff --git a/python/lsst/daf/butler/registry/datasets/byDimensions/_manager.py b/python/lsst/daf/butler/registry/datasets/byDimensions/_manager.py
index 288c452fbe..e9a6a891bb 100644
--- a/python/lsst/daf/butler/registry/datasets/byDimensions/_manager.py
+++ b/python/lsst/daf/butler/registry/datasets/byDimensions/_manager.py
@@ -586,9 +586,9 @@ def _find_storage(self, name: str) -> _DatasetRecordStorage:
         record = self._fetch_dataset_type_record(name)
         if record is not None:
             self._cache.add(record.dataset_type, record.dataset_type_id)
-            return _DatasetRecordStorage(
-                record.dataset_type, record.dataset_type_id, record.make_dynamic_tables()
-            )
+            tables = record.make_dynamic_tables()
+            self._cache.add_by_dimensions(record.dataset_type.dimensions, tables)
+            return _DatasetRecordStorage(record.dataset_type, record.dataset_type_id, tables)
         raise MissingDatasetTypeError(f"Dataset type {name!r} does not exist.")
 
     def getCollectionSummary(self, collection: CollectionRecord) -> CollectionSummary:
diff --git a/python/lsst/daf/butler/registry/tests/_registry.py b/python/lsst/daf/butler/registry/tests/_registry.py
index 4b6334fb18..439414a6a3 100644
--- a/python/lsst/daf/butler/registry/tests/_registry.py
+++ b/python/lsst/daf/butler/registry/tests/_registry.py
@@ -326,6 +326,28 @@ def testDatasetType(self):
         self.assertCountEqual([dt.name for dt in types], ["test", "testNoneTemplate"])
         self.assertEqual(missing, ["notarealdatasettype"])
 
+    def testDatasetTypeCache(self):
+        """Test for dataset type cache update logic after a cache miss."""
+        butler1 = self.make_butler()
+        butler2 = butler1.clone()
+        self.load_data(butler1, "base.yaml")
+
+        # Trigger full cache load.
+        butler2.get_dataset_type("flat")
+        # Have an external process register a dataset type.
+        butler1.registry.registerDatasetType(
+            DatasetType("test_type", ["instrument"], "int", universe=butler1.dimensions)
+        )
+        # Try to read the new dataset type -- this is a cache miss that
+        # triggers fetch of a single dataset type.
+        dt = butler2.get_dataset_type("test_type")
+        self.assertEqual(dt.name, "test_type")
+        self.assertEqual(list(dt.dimensions.names), ["instrument"])
+        # Use the dataset type's tags table.
+        self.assertEqual(
+            butler2.query_datasets("test_type", collections="*", find_first=False, explain=False), []
+        )
+
     def testDimensions(self):
         """Tests for `SqlRegistry.insertDimensionData`,
         `SqlRegistry.syncDimensionData`, and `SqlRegistry.expandDataId`.