Right now, a disaster that keeps flowing until August is the worst anyone wants to imagine, but there are worse possibilities-- it could be the start of years of unchecked oil flowing into the Gulf. (See this Bloomberg article suggesting that the relief well could take until December to solve the problem. One expert in the article mentions a decade. Keep that in mind as you read onward.)

There are legitimate concerns about the integrity of the casing. Yesterday, someone asked Admiral Allen about that. He said that concerns about the integrity of the well bore were part of the decision to stop the "Top Kill" a few weeks ago, indicating that there are significant concerns. On April 23, the Coast Guard was aware that the size of the leak could grow from 8000 barrels a day to 64,000 to 110,000 barrels a day if the well completely blew out. That's quite close to the current spill estimates. Does that mean that the well is nearing a full blow out?

The reason the casing's integrity matters is that if it's cracked, oil will push out through the cracks and into the surrounding ground, destabilizing the ground around the casing, and bubbling up from the ocean floor. Here's more, with Senator Bill Nelson's interview a week and a half ago saying just that. A seeping well, of course, will be hard to contain.

And finally, more alarming, and possibly a gross overstatement, there is the possibility that as the ground and the casing shift, the whole thing collapses inward, the giant Blow Out Preventer falls over, the drill pipe shoots out of the remains of the well, or any number of other scenarios that could make it very difficult or impossible to eventually stop the gusher even with the relief wells. (I do not know the author of this post, and cannot vouch for its accuracy. I do not share the author's fear that there is a conspiracy to hide this information. My sense is that it's unthinkable and so no one is asking the questions.) Thus the relief well are being drilled in a race with the integrity of the ground around the well and the casing.

If they don't make it, we're looking at a very very different kind of accident. (This is where the "decade" remark comes in.) And if so, Tony Hayward will look back upon yesterday as a relatively pleasant interlude before the *real* disaster struck.

It's time to start asking for a best guess of what is going on in the well hole. It is dishonest not to.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.