bit of an issue with data order guys, hoping you could help out.

I'm selecting from one table where a value is in annother, all seems well, I then select with an order by to another table and in 4 places in a dbf of over 100k it has jigged the records around a bit, not so that the order by is wrong but just within the groups of records that have the same order number. I was under the impression that the result data would be in the sequence of the original just minus those that dont match. however just occasionally it mixes a few up.

is this standard or an issue? I know I could always add an overall sequencial number and add a second parameter to the order by but I didn't think it was needed.

ok so an eg.

order  id  
1      1
2      1
3      3
4      2
5      3
6      2

I decide to do a

select * from table into sorttable order by id

and I'd expect the new order sequence to be 1,2,4,6,3,5 and 99% of the time it is, thats why I thought it worked that way, but just sometimes I'll get a 1,2,6,4,3,5 I'm talking about a few occurances in a hundred thousand. is it just me being silly to assume? like I said I can always make it

select * from table into sorttable order by id,order

but would like to knowwhy this happens at all.

cheers in advance.

Ken.