Alan,

I am not familiar with anything to count words in a string.  I suppose you
could evaluate each character and count up the spaces plus 1 or you could
use this:

sWords = Mid(oRg, sPos, ePos - sPos)
sWordsArray = Split(sWords, " ")
oCount = UBound(sWordsArray) + 1

Alan,

I monkeyed around with this code a little futher :-)

What if you have:
highlight "indexed" terms.{ XE "higlight "indexed" terms" \b \i }

The code would only highlight the: " terms

I added some code to handled "one" pair of quotes in the indexed term:

Sub HighightBeforeXE()
Dim ofld As Field
Dim oRg As Range
Dim oTempRg As Range
Dim oCount As Integer
Dim ePos As Integer
Dim sPos As Integer
Dim midPos1 As Integer
Dim bQuotedIndex As Boolean
Dim midPos2 As Integer
Dim sWords As String
Dim sWordsArray As Variant

For Each ofld In ActiveDocument.Fields
 bQuotedIndex = False
 If ofld.Type = wdFieldIndexEntry Then
 Set oRg = ofld.Code
 sPos = InStr(5, oRg, ":", 1) + 1 'start position with subentry
 If sPos < 6 Then 'no subentry
   sPos = InStr(5, oRg, """") + 1 'start position
 End If
 midPos1 = InStr(sPos, oRg, "\""")
 If midPos1 > 0 Then
   bQuotedIndex = True
   midPos2 = InStr(midPos1 + 2, oRg, "\""")
 End If
 If midPos2 > 0 Then
   ePos = InStr(midPos2 + 2, oRg, """", 1)
 Else
   ePos = InStr(sPos + 1, oRg, """", 1) 'end pos
 End If
 sWords = Mid(oRg, sPos, ePos - sPos) 'words between the start" end"
marks.
 sWordsArray = Split(sWords, " ")
 If Not bQuotedIndex Then
   oCount = UBound(sWordsArray) + 1
 Else
   oCount = UBound(sWordsArray) + 3
 End If
 With oRg
   .Move unit:=wdWord, Count:=-1
   .MoveStart unit:=wdWord, Count:=-oCount
   .HighlightColorIndex = wdYellow
 End With
 Set oRg = Nothing
 End If
Next ofld
End Sub

I had worked out something similiar using the Find.Execute method
before Jay introduced us to the the oFld.Code.Text range:

Public Sub HighlightIndexedTerms()
Dim oRg As Range
Dim oTempRg As Range
Dim i As Integer

Set oRg = ActiveDocument.Range
oRg.TextRetrievalMode.IncludeHiddenText = True
With oRg.Find
 .ClearFormatting
 .Text = " XE " ' find index entry field code
 .Format = False
 .Forward = True
 .Wrap = wdFindStop
 .MatchCase = False
 Do While .Execute
   'At this point, oRg covers the XE in an index field.
   'Make another range that duplicates oRg...
   Set oTempRg = oRg.Duplicate
   'Use the duplicate to determine the indexed term.
   With oTempRg
     .Move unit:=wdWord, Count:=2 'Move range passed first "
     Do
     .MoveEndUntil Cset:=Chr(34), Count:=wdForward 'range the XE field
term
       If oTempRg.Characters.Last = "\" Then
         .MoveEnd unit:=wdCharacter, Count:=2
         .MoveEndUntil Cset:=Chr(34), Count:=wdForward 'range the XE
field ter
       End If
     Loop Until oTempRg.Characters.Last <> "\"
     i = oTempRg.Words.Count 'count words in indexed term
   End With
   Set oTempRg = oRg.Duplicate 'Again match oTempRg to oRg
   With oTempRg
     .Move unit:=wdWord, Count:=-1 'move range to left edge of indexed
term
     'Expand the range to cover the indexed term
     .MoveStart unit:=wdWord, Count:=-i
     'Highlight it.
     .HighlightColorIndex = wdYellow
   End With
   'The loop now continues from the current position of oRg, which
   'hasn't been changed by fiddling with oTempRg.
   Set oTempRg = Nothing
 Loop
End With
End Sub

This would handle multiple quote pairs in the indexed term, but I
haven't attempted to address the subentries and references in the field
code.  

HTH

So an old dog can learn new tricks ;-)

many thanks again - til next time!

Kind regards - Alan F.