diff --git a/lib/spack/spack/cmd/create.py b/lib/spack/spack/cmd/create.py
index cf23c4191c..6d0b3f095e 100644
--- a/lib/spack/spack/cmd/create.py
+++ b/lib/spack/spack/cmd/create.py
@@ -155,7 +155,7 @@ def create(parser, args):
     else:
         urls = [spack.url.substitute_version(url, v) for v in versions]
         if len(urls) > 1:
-            tty.msg("Found %s versions of %s." % (len(urls), name),
+            tty.msg("Found %s versions of %s:" % (len(urls), name),
                     *spack.cmd.elide_list(
                     ["%-10s%s" % (v,u) for v, u in zip(versions, urls)]))
             print
diff --git a/lib/spack/spack/package.py b/lib/spack/spack/package.py
index f733c5cbd2..e96da1a26a 100644
--- a/lib/spack/spack/package.py
+++ b/lib/spack/spack/package.py
@@ -394,8 +394,6 @@ def version(self):
     @property
     def stage(self):
         if not self.spec.concrete:
-            print self.spec
-            print self.spec.concrete
             raise ValueError("Can only get a stage for a concrete package.")
 
         if self._stage is None:
@@ -803,8 +801,7 @@ def fetch_available_versions(self):
                 self._available_versions = find_versions_of_archive(
                     self.url,
                     list_url=self.list_url,
-                    list_depth=self.list_depth,
-                    wildcard=self.default_version.wildcard())
+                    list_depth=self.list_depth)
 
                 if not self._available_versions:
                     tty.warn("Found no versions for %s" % self.name,
@@ -834,26 +831,31 @@ def available_versions(self):
 def find_versions_of_archive(archive_url, **kwargs):
     list_url   = kwargs.get('list_url', None)
     list_depth = kwargs.get('list_depth', 1)
-    wildcard   = kwargs.get('wildcard', None)
 
     if not list_url:
         list_url = os.path.dirname(archive_url)
-    if not wildcard:
-        wildcard = url.wildcard_version(archive_url)
 
-    versions = VersionList()
-    url_regex = os.path.basename(wildcard)
+    # This creates a regex from the URL with a capture group for the
+    # version part of the URL.  The capture group is converted to a
+    # generic wildcard, so we can use this to extract things on a page
+    # that look like archive URLs.
+    url_regex = url.wildcard_version(archive_url)
 
+    # We'll be a bit more liberal and just look for the archive part,
+    # not the full path.
+    archive_regex = os.path.basename(url_regex)
+
+    # Grab some web pages to scrape.
     page_map = get_pages(list_url, depth=list_depth)
 
+    # Build a version list from all the matches we find
+    versions = VersionList()
     for site, page in page_map.iteritems():
-        strings = set(re.findall(url_regex, page))
-
-        for s in strings:
-            match = re.search(wildcard, s)
-            if match:
-                v = match.group(0)
-                versions.add(Version(v))
+        # extract versions from matches.
+        matches = re.finditer(archive_regex, page)
+        version_strings = set(m.group(1) for m in matches)
+        for v in version_strings:
+            versions.add(Version(v))
 
     return versions
 
diff --git a/lib/spack/spack/url.py b/lib/spack/spack/url.py
index 8e4f39899d..0bc7ff53b4 100644
--- a/lib/spack/spack/url.py
+++ b/lib/spack/spack/url.py
@@ -204,4 +204,6 @@ def wildcard_version(path):
     v = Version(ver)
     parts = list(re.escape(p) for p in path.split(str(v)))
 
-    return  v.wildcard().join(parts)
+    # Make a group for the wildcard, so it will be captured by the regex.
+    version_group = '(%s)' % v.wildcard()
+    return version_group.join(parts)
diff --git a/lib/spack/spack/version.py b/lib/spack/spack/version.py
index c433fa8962..8108c0f60b 100644
--- a/lib/spack/spack/version.py
+++ b/lib/spack/spack/version.py
@@ -163,7 +163,10 @@ def a_or_n(seg):
         wc = seg_res[0]
         for i in xrange(1, len(sep_res)):
             wc += '(?:' + sep_res[i] + seg_res[i]
-        wc += ')?' * (len(seg_res) - 1)
+
+        # Add possible alpha or beta indicator at the end of each segemnt
+        # We treat these specially b/c they're so common.
+        wc += '[ab]?)?' * (len(seg_res) - 1)
         return wc