any given field in a table ideally would only have one kind of data because it makes it easier to do more with a query. other data types like timestamps and datestamps are a little different...you generally have to parse them a bit sometimes. there's no crime against putting multiple types of data in a single db field, you just have to be ready to parse out the data yourself.
but as i understand it, dmacman might want to do 2 different types of pricing.
one is a flat rate no matter what quantity. that's easy. if red t-shirts were $5 each, you might have a table with records like this
id=1
name='t-shirt'
color='red'
cost=5.00
suppose he also had blue t-shirts. they are overstocked so he's offering bulk discounts. since the bulk discount has several prices points, you are going to need a separate table. by convention, he could say that if the price for a product is empty then look in the price_break_table for the price breaks.
in the product table, a record might be:
id=2
name='t-shirt'
color='blue'
cost=
suppose that up to 9 t-shirts, they are $5. 10-19 s-shirts would be $4 and 20 or more would be $3. price_break_table needs to associate a quantity and cost with the blue t-shirt. price_break_table in that case might have 3 records that look like:
prod_id, count, cost
2, 1, 5.00
2, 10, 4.00
2, 20, 3.00
the tricky part now is setting up the database queries and store logic to decipher your price_break_table records for a particular item and figure out which record applies.