Thank you very much.

>Dennis,
>
>You can do this in a couple of steps.
>
>Try the folloowing code:
>

clear
>
>create cursor t1 (ival1 I NULL)
>
>insert into t1 values (.NULL.)
>insert into t1 values (1)
>
>select nvl(iVal1, 000000000000) as iVal1 from t1 into cursor noint
>
>disp stru
>
>select int(nvl(iVal1, 000000000000)) as iVal1 from t1 into cursor withint
>
>disp stru
>
>create cursor appended  (ival1 I NULL)
>
>append from dbf('withint')
>
>disp stru

>notice that the structure of the cursor 'appended' has an integer.
>
>It just takes an extra step to make a receptacle for the integers..
>
>Notice also that I put all the 0s in the NVL() call. This is to ensure that the cursor created has a sufficiently wide numeric field.
>
>Cheers,
>
>Andrew
>
>>Actually I find much solace in this because I know that I have done as much as I can. I can accept the situation. I just want to know if there are obvious solutions or improvements beyond what I have implemented. You have confirmed that there are not. I am very comforted by that. (Of course, I would also be pleased to learn of solutions that I had not considered.)
>>
>>Thanks, Ed.