--- a/imagej2_analyze_skeleton.xml	Tue Aug 04 11:14:30 2015 -0400
+++ b/imagej2_analyze_skeleton.xml	Tue Aug 04 13:19:34 2015 -0400
@@ -78,18 +78,28 @@
         </test>
     </tests>
     <help>
+
+.. class:: warningmark
+
+The underlying AnalyzeSkeleton Fiji plugin works on binary images, so other image types will
+automatically be converted to binary before they are analyzed.  This step is performed using the
+ImageJ2 **Make Binary** command with the following settings: **Iterations:** 1, **Count:** 1,
+**Black background:** No, **Pad edges when eroding:** No.  If these settings are not appropriate,
+first manually convert the image to binary using the **Convert to binary (black and white) with
+ImageJ2** tool, which allows you to change them.
+
+.. class:: warningmark
+
+The **Skeletonize images** tool produces skeleton images that can be used as input to this tool.
+Images that have not been skeletonized will not result in errors, but the results may or may not
+be useful.
+
 **What it does**
 
 <![CDATA[
 Tags all pixel/voxels in a skeleton image and then counts all its junctions, triple and quadruple
-points and branches, and measures their average and maximum length.  The "Skeletonize images" tool
-produces skeleton images that can be used as input to this tool.  Images that have not been
-skeletonized will not result in errors, but the results may or may not be useful.  The underlying
-AnalyzeSkeleton Fiji plugin works on grayscale images, so other image types will automatically be
-converted to grayscale before they are analyzed.  This step is performed using the ImageJ **Make
-Binary** command.
-
-The voxels are classified into three different categories depending on their 26 neighbors:
+points and branches, and measures their average and maximum length.  The voxels are classified into
+three different categories depending on their 26 neighbors:
 
 - **End-point** voxels: if they have less than 2 neighbors.
 - **Junction** voxels: if they have more than 2 neighbors.