There's no smoking gun in this study. The answers are all in there — but we're just not smart enough to figure them out yet.

This result was an impressive and completely unexpected finding because the study was not focused on helping smokers quite smoking. In fact, to participate in the study, smokers had to attest that they did not have a planned quit date for the next six months.

In the 30 years that this study now spans, we have seen some wide fluctuations in the smoking habits of American young people. We are still seeing some residual declines in smoking in the upper grades, as the lower-smoking birth cohorts make their way up the age spectrum...But, even in the upper grades a slowdown is occurring, and we believe the declines are likely to end very soon.

This changes how we think about evolution and, in some respects, disease. That's the part that's exciting.

If you think of the genome as the book of life, then the differences people have been looking at over the past 10 years have been typographical errors. But what we are looking at are redundancies in the text -- places where entire pages [genes] have been reprinted multiple times, or the page is missing.

That gene went through a selective sweep.