The last solution I posted uses a series of updates. Choose one of them and run it four times, something like this:

DECLARE @pairs TABLE (Duplicate_ID INT, Master_ID INT)

INSERT INTO @pairs

SELECT 16492, 23011 UNION ALL

SELECT 23011,24014 UNION ALL

SELECT 23011,24951 UNION ALL

SELECT 16492,24951 UNION ALL

SELECT 16492,29381 UNION ALL

SELECT 24951,29381 UNION ALL

SELECT 23011,29381 UNION ALL

SELECT 24014,66665 UNION ALL

SELECT 3344,66665;

-- preprocess the sample data: there isn't a dupe and a master,

-- they're simply members of a set of two.

IF object_id('TempDB..#Temp') IS NOT NULL DROP TABLE #Temp;

WITH SequencedData AS (

SELECT *, SetNo = ROW_NUMBER() OVER (ORDER BY Duplicate_ID, Master_ID)

FROM @pairs

)

SELECT SetNo, Duplicate_ID

INTO #Temp

FROM SequencedData

UNION ALL

SELECT SetNo, Master_ID

FROM SequencedData;

-- (102 row(s) affected)

---------------------------------------------------------------------------

-- FOUR updates have to be executed with the sample data to complete.

---------------------------------------------------------------------------

DECLARE @RowsUpdated INT

SET @RowsUpdated = 1

WHILE @RowsUpdated > 0

BEGIN

UPDATE t SET SetNo = c.newset

FROM #Temp t

INNER JOIN (

SELECT d.SetNo, newset = MIN(newset)

FROM (

SELECT Duplicate_ID, SetNo, newset = MIN(SetNo) OVER (PARTITION BY Duplicate_ID)

FROM #Temp

) d

GROUP BY d.SetNo

HAVING MIN(newset) <> d.SetNo

) c ON c.SetNo = t.SetNo;

SET @RowsUpdated = @@rowcount

END

---------------------------------------------------------------------------

-- Final result set

---------------------------------------------------------------------------

SELECT t.Duplicate, x.Duplicate_ID

FROM (

SELECT SetNo, Duplicate = MIN(Duplicate_ID)

FROM #Temp

GROUP BY SetNo

) t

CROSS APPLY (

SELECT DISTINCT *

FROM #Temp ti

WHERE ti.SetNo = t.SetNo

AND ti.Duplicate_ID <> t.Duplicate

) x;