diff --git a/slides_template/mknotes b/slides_template/mknotes
index 7ecb1f4..571ec99 100755
--- a/slides_template/mknotes
+++ b/slides_template/mknotes
@@ -44,6 +44,7 @@ class Page:
         self.section = ""
         self.subsection = ""
         self.notes = []
+        self.items = []
 
 
 def parse_notes_file(path):
@@ -77,9 +78,10 @@ def parse_notes_file(path):
             i += 1
             continue
 
-        match = re.match(r"^NOTE FRAME (\d+)\|SPEC (.*)$", line)
+        match = re.match(r"^NOTE FRAME (\d+)\|SPEC (.*)\|KIND (.*)$", line)
         if match:
-            frame, spec = int(match.group(1)), match.group(2).strip()
+            frame = int(match.group(1))
+            spec, kind = match.group(2).strip(), match.group(3).strip()
             body = []
             i += 1
             while i < len(lines) and lines[i].strip() != "ENDNOTE":
@@ -90,10 +92,10 @@ def parse_notes_file(path):
             text = " ".join(part.strip() for part in body).strip()
             # A note is re-run for every slide of its frame, so the same record
             # arrives once per slide. Keep the first occurrence only.
-            key = (frame, spec, text)
+            key = (frame, spec, kind, text)
             if key not in seen:
                 seen.add(key)
-                notes.setdefault(frame, []).append((spec, text))
+                notes.setdefault(frame, []).append((spec, kind, text))
             continue
 
         i += 1
@@ -141,9 +143,12 @@ def attach_notes(pages, notes):
         last_slide[page.frame] = max(last_slide.get(page.frame, 0), page.slide)
 
     for page in pages:
-        for spec, text in notes.get(page.frame, []):
+        for spec, kind, text in notes.get(page.frame, []):
             if spec_matches(spec, page.slide, last_slide[page.frame]):
-                page.notes.append(text)
+                if kind == "item":
+                    page.items.append(text)
+                else:
+                    page.notes.append(text)
 
 
 # ---------------------------------------------------------------- typesetting
@@ -243,14 +248,25 @@ def build_notes_pdf(pages, width, height, thumb_w, thumb_h, workdir,
         body.append(r"\noindent\colorbox{notegrey}{%s}%%" % band)
 
         body.append(r"\par\vspace{%.2fbp}%%" % gap)
+        # Paragraph spacing is set here rather than in the preamble so that it
+        # separates the notes without also loosening the heading band above.
         body.append(r"\noindent\hspace{%.2fbp}\begin{minipage}[t]{%.2fbp}"
+                    r"\setlength{\parskip}{0.7em}"
                     % (hmargin, textwidth))
+        # As in beamer, plain \note commands run together as text and
+        # \note[item] ones become a numbered list printed after it. beamer
+        # concatenates the plain ones with no separator at all; a paragraph
+        # break between them is easier to read and is the one deviation.
         if page.notes:
-            body.append(r"\begin{itemize}\setlength{\itemsep}{0.6em}"
-                        r"\setlength{\leftmargini}{1.2em}")
-            for note in page.notes:
-                body.append(r"\item %s" % escape_for_context(note))
-            body.append(r"\end{itemize}")
+            body.append("\n\n".join(
+                escape_for_context(n) for n in page.notes))
+        if page.items:
+            body.append(r"\par\begin{enumerate}\setlength{\itemsep}{0pt}"
+                        r"\setlength{\parskip}{0pt}"
+                        r"\setlength{\leftmargini}{1.4em}")
+            for item in page.items:
+                body.append(r"\item %s" % escape_for_context(item))
+            body.append(r"\end{enumerate}")
         body.append(r"\end{minipage}")
 
         if index != total - 1:
diff --git a/slides_template/talk-notes.sty b/slides_template/talk-notes.sty
index 25e9403..456b668 100644
--- a/slides_template/talk-notes.sty
+++ b/slides_template/talk-notes.sty
@@ -58,21 +58,25 @@
       { SUBSECTION ~ \tl_to_str:N \g__talknotes_subsection_tl }
   }
 
-% \note[<overlay spec>]{text}
+% \note<overlay spec>[item]{text}
+%
+% Follows beamer's interface. A plain \note contributes running text; \note[item]
+% contributes an entry to a numbered list printed after that text.
 %
 % The body is written out stringified, so LaTeX markup inside a note survives
 % to be typeset again on the notes page. A note is recorded once per slide of
 % its frame (the frame body is re-run for each), so mknotes deduplicates.
 % Records are bracketed by ENDNOTE because a long note may be broken across
 % output lines.
-\NewDocumentCommand \note { d<> +m }
+\NewDocumentCommand \note { d<> o +m }
   {
     \iow_now:Nx \g__talknotes_iow
       {
         NOTE ~ FRAME ~ \int_use:N \g__talk_frame_int |
-        SPEC ~ \IfNoValueTF {#1} { - } { \tl_to_str:n {#1} }
+        SPEC ~ \IfNoValueTF {#1} { - } { \tl_to_str:n {#1} } |
+        KIND ~ \IfNoValueTF {#2} { text } { \tl_to_str:n {#2} }
       }
-    \iow_now:Nx \g__talknotes_iow { TEXT ~ \tl_to_str:n {#2} }
+    \iow_now:Nx \g__talknotes_iow { TEXT ~ \tl_to_str:n {#3} }
     \iow_now:Nn \g__talknotes_iow { ENDNOTE }
   }