diff --git a/qurator/dinglehopper/extracted_text.py b/qurator/dinglehopper/extracted_text.py
index 46c8fec..c492e7e 100644
--- a/qurator/dinglehopper/extracted_text.py
+++ b/qurator/dinglehopper/extracted_text.py
@@ -194,7 +194,7 @@ class ExtractedText:
                 try:
                     return int(index)
                 except TypeError:
-                    return None
+                    return float('inf')
             textequivs = sorted(textequivs, key=get_index)
 
             return textequivs[0].find('./page:Unicode', namespaces=nsmap).text
diff --git a/qurator/dinglehopper/tests/extracted_text_test.py b/qurator/dinglehopper/tests/extracted_text_test.py
index 0d59c99..0caf4f3 100644
--- a/qurator/dinglehopper/tests/extracted_text_test.py
+++ b/qurator/dinglehopper/tests/extracted_text_test.py
@@ -93,3 +93,26 @@ def test_textequiv_index():
     expected = "gefahren zu haben, einzelne Bemerkungen und Beobach-"
 
     assert expected == result
+
+
+def test_textequiv_no_index():
+    """
+    Test that extracting text from a PAGE TextEquiv ignores missing indices.
+    """
+
+    textline="""<?xml version="1.0"?>
+      <TextLine id="l3" xmlns="http://schema.primaresearch.org/PAGE/gts/pagecontent/2018-07-15">
+        <TextEquiv>
+          <Unicode>gefahren zu haben, einzelne Bemorkungen und Beobäch-</Unicode>
+        </TextEquiv>
+        <TextEquiv index="1">
+          <Unicode>gefahren zu haben, einzelne Bemerkungen und Beobach-</Unicode>
+        </TextEquiv>
+      </TextLine>
+    """
+    root = ET.fromstring(textline)
+    nsmap = {'page': "http://schema.primaresearch.org/PAGE/gts/pagecontent/2018-07-15" }
+    result = ExtractedText.from_text_segment(root, nsmap, textequiv_level='line').text
+    expected = "gefahren zu haben, einzelne Bemerkungen und Beobach-"
+
+    assert expected == result
\ No newline at end of file