Re: first minimum date in the column for each distinct column a value

Two ways:

-- SQL Server 2000
SELECT cola, colb, colc, cold
FROM Xyz AS A
WHERE A.cold = (SELECT MIN(B.cold)
                       FROM Xyz AS B
                       WHERE B.cola = A.cola)


-- SQL Server 2005
WITH RankedXyz
AS
(SELECT cola, colb, colc, cold,
            ROW_NUMBER() OVER(
                    PARTITION BY cola
                    ORDER BY cold) AS seq
 FROM Xyz)
SELECT cola, colb, colc, cold
FROM RankedXyz
WHERE seq = 1;

HTH,

Plamen Ratchev
http://www.SQLStudio.com