>Isn't SCAN enough to ensure an update of the pointer in the related table?! This example does not prove that SKIP 0 also updates that pointer. Also, the LOCATE command is nonsense here, or isn't it?
>
>>SKIP 0 && This is zero?
>>

>>SELECT MYParentTable
>>SET RELATION TO Key INTO ChildTable
>>LOCATE
>>SCAN
>>   SKIP 0
>>   ? ChildTable.Key, EOF("ChildTable"), BOF("ChildTable")
>>ENDIF
>>

This looks like there may be some code between Scan and Skip 0 that may leave the record pointer in a limbo. As shown here, it's really unnecessary, as is the Locate. I assume the code shown is just to demonstrate how it works - even though in this case it would work equally nice without the Skip 0.