1。如果有ID字段，就是具有唯一性的字段  
   
  delect   table   where   id   not   in   (  
   
      select   max(id)   from   table   group   by   col1,col2,col3...  
  )  
  group   by   子句后跟的字段就是你用来判断重复的条件，如只有col1，那么只要col1字段内容相同即表示记录相同。  
   
   
   
  2,如果是判断所有字段也可以这样  
      select   *   into   #aa   from   table   group   by   id1,id2,....  
      delete   table    
      insert   into   table    
      select   *   from   #aa  
   
   
   
  3，没有ID的情况  
   
  select   identity(int,1,1)   as   id,*   into   #temp   from   tabel  
  delect   #   where   id   not   in   (  
      select   max(id)   from   #   group   by   col1,col2,col3...)  
  delect   table  
  inset   into   table(...)  
        select   .....   from   #temp   
    
    
4.  col1+','+col2+','...col5   联合主键  
   
   
  select   *   from     table   where   col1+','+col2+','...col5   in   (  
   
      select   max(col1+','+col2+','...col5)   from   table    
  where   having   count(*)>1  
  group   by   col1,col2,col3,col4    
  )  
  group   by   子句后跟的字段就是你用来判断重复的条件，如只有col1，那么只要col1字段内容相同即表示记录相同。   
    
  5   
  select   identity(int,1,1)   as   id,*   into   #temp   from   tabel  
  select   *   from     #temp   where   id   in   (  
      select   max(id)   from   #emp   where   having   count(*)>1   group   by   col1,col2,col3...)  

6.
select   distinct   *   into   #temp   from   tablename    
  delete   tablename    
  go  
  insert   tablename   select   *   from   #temp  
  go  
  drop   table   #temp