The June to September [quarter] was 17 million to 14 million. The first thing to note there is that as we talked about in the July call, the June quarter contained a 1.2 million increase in channel inventory. And so the actual sell-through sequentially in comparison looks very different than our reported sell-in numbers do.

The second thing is that the 14 million exceeded what we had expected to do in iPad. And the reason we had expected it to decline is that we believed that, based on the two or three years of results that we've got, normally we would see a seasonal reduction in the September quarter versus the June quarter. Part of this is because K-12 heavily buys in the June quarter; K-12 doesn't buy very much in the September quarter, it becomes a higher-ed kind of move, and higher-ed is still buying notebooks for the most part. There's some kind of normal seasonal. That's exaggerated further when we announce a new product in March and have an enormous full quarter of demand in the June quarter; when you compare that to the September quarter there would be a natural phase down. In addition to all of that, it's clear that customers delay purchases of tablets due to new product rumors, and these intensified in August and September.

Some of that was anticipated and some of it I wish wouldn't occur, but it did occur. And so, that's how I'd explain the sequential difference. On a year over year basis, because of the year-ago quarter having also a channel inventory build as we stocked the channel to the proper level, the sell-through, year over year, actually grew 44 percent. The underlying sell-through was extremely strong. We continue to feel great about how iPad has done, and with announcing the fourth-generation iPad just this week and adding iPad mini to the family, we think it's going to be an incredible holiday season.

What we did initially was we planned the first 30 to 40 countries prior to introducing the product and rolled those out across September in two different dates. The balance of the quarter, we planned with an eye toward the supply and what we think the demand would be, but we do plan these in advance and so it is not a precise science. We obviously have to plan those with several weeks of notice, and so occasionally it can be different than what we think.