diff --git a/src/sourmash/sig/__main__.py b/src/sourmash/sig/__main__.py
index e342f14470..eab898b39b 100644
--- a/src/sourmash/sig/__main__.py
+++ b/src/sourmash/sig/__main__.py
@@ -553,9 +553,9 @@ def extract(args):
 
         notify(f"loaded {len(picklist.pickset)} distinct values into picklist.")
         if n_empty_val:
-            notify(f"WARNING: {n_empty_val} empty values in column '{picklist.column_name}' in CSV file")
+            notify(f"WARNING: {n_empty_val} empty values in column '{picklist.column_name}' in picklist file")
         if dup_vals:
-            notify(f"WARNING: {len(dup_vals)} values in column '{picklist.column_name}' were not distinct")
+            notify(f"WARNING: {len(dup_vals)} values in picklist column '{picklist.column_name}' were not distinct")
         picklist_filter_fn = picklist.filter
     else:
         def picklist_filter_fn(it):
diff --git a/src/sourmash/sig/picklist.py b/src/sourmash/sig/picklist.py
index 6c2594c100..4d07f309dc 100644
--- a/src/sourmash/sig/picklist.py
+++ b/src/sourmash/sig/picklist.py
@@ -85,12 +85,11 @@ def init(self, values=[]):
         if self.pickset is not None:
             raise ValueError("already initialized?")
         self.pickset = set(values)
+        return self.pickset
 
     def load(self, pickfile, column_name):
         "load pickset, return num empty vals, and set of duplicate vals."
-        pickset = self.pickset
-        if pickset is None:
-            pickset = set()
+        pickset = self.init()
 
         n_empty_val = 0
         dup_vals = set()
@@ -98,7 +97,7 @@ def load(self, pickfile, column_name):
             r = csv.DictReader(csvfile)
 
             if column_name not in r.fieldnames:
-                raise ValueError("column '{column_name}' not in pickfile '{pickfile}'")
+                raise ValueError(f"column '{column_name}' not in pickfile '{pickfile}'")
 
             for row in r:
                 # pick out values from column
@@ -113,9 +112,8 @@ def load(self, pickfile, column_name):
                 if col in pickset:
                     dup_vals.add(col)
                 else:
-                    pickset.add(col)
+                    self.add(col)
 
-        self.pickset = pickset
         return n_empty_val, dup_vals
 
     def add(self, value):
diff --git a/tests/test_cmd_signature.py b/tests/test_cmd_signature.py
index c15bd4d724..e243234de0 100644
--- a/tests/test_cmd_signature.py
+++ b/tests/test_cmd_signature.py
@@ -1462,6 +1462,104 @@ def test_sig_extract_9_picklist_md5_ksize_hp_select(runtmp):
     assert actual_extract_sig.minhash.moltype == 'hp'
 
 
+def test_sig_extract_10_picklist_md5_dups_and_empty(runtmp):
+    # test empty picklist values, and duplicate picklist values
+    sigdir = utils.get_test_data('prot/')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=['md5'])
+        w.writeheader()
+        w.writerow(dict(md5='ea2a1ad233c2908529d124a330bcb672'))
+        w.writerow(dict(md5='ea2a1ad233c2908529d124a330bcb672'))
+        w.writerow(dict(md5=''))
+
+    picklist_arg = f"{picklist_csv}:md5:md5"
+
+    runtmp.sourmash('sig', 'extract', sigdir, '--picklist',
+                    picklist_arg, '-k', '19', '--hp')
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig.minhash.ksize == 19
+    assert actual_extract_sig.minhash.moltype == 'hp'
+
+    err = runtmp.last_result.err
+    print(err)
+
+    assert "WARNING: 1 empty values in column 'md5' in picklist file" in err
+    assert "WARNING: 1 values in picklist column 'md5' were not distinct" in err
+
+
+def test_sig_extract_11_picklist_bad_coltype(runtmp):
+    # test with invalid picklist coltype
+    sigdir = utils.get_test_data('prot/')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=['md5'])
+        w.writeheader()
+        w.writerow(dict(md5='ea2a1ad233c2908529d124a330bcb672'))
+
+    picklist_arg = f"{picklist_csv}:md5:BADCOLTYPE"
+
+    with pytest.raises(ValueError):
+        runtmp.sourmash('sig', 'extract', sigdir, '--picklist',
+                        picklist_arg, '-k', '19', '--hp')
+
+    err = runtmp.last_result.err
+    print(err)
+    assert "ValueError: invalid picklist column type 'BADCOLTYPE'" in err
+
+
+def test_sig_extract_12_picklist_bad_argstr(runtmp):
+    # test with invalid argument format to --picklist
+    sigdir = utils.get_test_data('prot/')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=['md5'])
+        w.writeheader()
+        w.writerow(dict(md5='ea2a1ad233c2908529d124a330bcb672'))
+
+    picklist_arg = f"{picklist_csv}"
+
+    with pytest.raises(ValueError):
+        runtmp.sourmash('sig', 'extract', sigdir, '--picklist',
+                        picklist_arg, '-k', '19', '--hp')
+
+    err = runtmp.last_result.err
+    print(err)
+    assert "invalid picklist argument" in err
+
+
+def test_sig_extract_12_picklist_bad_colname(runtmp):
+    # test with invalid picklist colname
+    sigdir = utils.get_test_data('prot/')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=['md5'])
+        w.writeheader()
+        w.writerow(dict(md5='ea2a1ad233c2908529d124a330bcb672'))
+
+    picklist_arg = f"{picklist_csv}:BADCOLNAME:md5"
+
+    with pytest.raises(ValueError):
+        runtmp.sourmash('sig', 'extract', sigdir, '--picklist',
+                        picklist_arg, '-k', '19', '--hp')
+
+    err = runtmp.last_result.err
+    print(err)
+    assert "ValueError: column 'BADCOLNAME' not in pickfile" in err
+
+
 @utils.in_tempdir
 def test_sig_flatten_1(c):
     # extract matches to several names from among several signatures & flatten