my XML file is fairly small. just a small ascii file say 500 lines long
max.

Note this snippet of code:

 public static void traverse(Node node, int level) {

     // Process node here, then move on to the next child
     // for example -

     if (node.getParentNode().getNodeName().equals("b")) {    <<<<<
         // or whatever,  blah blah blah                        <<<<<
     }

     // If there are any children, visit each one
     NodeList list = node.getChildNodes();
     for (int i = 0; i < list.getLength(); i++) {
         // Get child node
         Node childNode = list.item(i);

         // Visit child node
         traverse(childNode, level + 1);
     }
 }

Where you see the comment // Process node here ... is where you would
put code to handle your data.

I still seriously suggest that you investigate JDOM - it was designed
specifically for _Java_ handling of XML documents (unlike DOM or SAX),
and so its methods are far easier to work with.

Some (quick and hasty) code I wrote using JDOM might show where this is easier:

  public QuakeData(String rssURL, Point p)
        throws IOException, JDOMException, ParseException {

        // retrieve xml and build document in memory
       SAXBuilder builder = new SAXBuilder();
       Document doc = builder.build(rssURL);
        // that's it - the entire xml file is now in memory.

        // this xml has three namespaces in all, so set up the extra two
       Namespace geo =
               Namespace.getNamespace("geo",
"http://www.w3.org/2003/01/geo/wgs84_pos#");
       Namespace dc =
               Namespace.getNamespace("dc",
"http://purl.org/dc/elements/1.1/");

        // some working variables
       List subjects;
       Element eventElement, subjectElement;

        // find the root
       Element root = doc.getRootElement();

        // we're only interested in the the branch of the tree with "item"
        // notice we can specify by _name_ here
       List events = root.getChild("channel").getChildren("item");

        // now that we have only the branch of the tree with "item", iterate through
        // and work the data the way we want. In this case, I'm creating an
new object
        // of class Event, and then setting its attributes based on the xml
data. Notice
        // again I just use the name specified in the xml document - very simple:
       for (Object event : events) {
           QuakeEvent quakeEvent = new QuakeEvent();
           eventElement = (Element) event;
           quakeEvent.setPubDate(eventElement.getChildText("pubDate"));
           quakeEvent.setTitle(eventElement.getChildText("title"));
           
quakeEvent.setDescription(eventElement.getChildText("description"));
           quakeEvent.setLink(eventElement.getChildText("link"));
           quakeEvent.setLocation(eventElement.getChildText("lat", geo),
                   eventElement.getChildText("long", geo));
           subjects = eventElement.getChildren("subject", dc);

        // subject as multiple occurrences, all named the same thing. I need to
        // isolate them differently then previously.  Depending upon their
array index,
        // I set the class attribute accordingly.
           for (int i = 0; i < subjects.size(); i++) {
               subjectElement = (Element) subjects.get(i);
               switch (i) {
                   case 0: {
                       quakeEvent.setMagnitude(subjectElement.getText());
                       break;
                   }
                   case 1: {
                       quakeEvent.setPeriod(subjectElement.getText());
                       break;
                   }
                   case 2: {
                       quakeEvent.setDepth(subjectElement.getText());
                       break;
                   }
                   default:
                       break;
               }
           }

        // all the class attributes have been set, so now I store this object in my
        // ArrayList.
           quakeList.add(quakeEvent);
       }

All the best to you if you keep using DOM - I gave up on it in favor of JDOM :)

Kevin