Frank wrote: > Why an unsigned int, instead of simply int? One virtually never has negative values in an identity column. By using an unsigned int instead of an int, you're effectively doubling the number of values a column can have. I have no idea what size data types MySQL uses, but if it's (say) 16-bit, an unsigned int can hold 0-65535; a signed int, -32,768-32,767. Why restrict your upper value to 32,000-odd if you're not going to use any negative numbers, and you can have up to 65,000 without using any further storage? Paul.