To clean gunk off your teeth, toothpastes have to be a little abrasive. But that sandpapery action can also wear down the enamel. Most of the studies on tooth-scratching have been done in the lab, which is unrealistic—your mouth has bacteria, saliva, and other stuff floating around. So in this study, some researchers at Unilever in the U.K. tried out a method for testing abrasion on bits of teeth stuck in volunteers' mouths. They were mostly interested in the methodology, but the results are kind of cool, too.

What the researchers wanted to know: To find out how abrasive different toothpastes are, can you use pieces of tooth in the mouths of human volunteers?

What they did: Twelve people with full sets of dentures had a three-millimeter-thick slice of baby tooth attached to their dentures. Each tooth slice had a tiny indentation made in it, so the researchers could measure wear on the tooth. For 28 days, the subjects took their dentures out and brushed them with a soft toothbrush and the toothpaste they'd been assigned. There were three different toothpastes, unhelpfully described in the study as A, B, and C. The volunteers weren't allowed to use any other toothpaste, dental floss, denture cleaner, or mouthwash during the study.

What they found: Yep, the study worked: it found a difference. Toothpaste C wore enamel significantly more than toothpastes A and B.

What the study means to you: Your enamel knows all toothpastes are not the same. But since the researchers aren't naming names here, you can't tell which is which.

Caveats: It's a small study, but it was big enough to find a statistical significance between the different toothpastes.

Find out more: On buying toothpaste: www.dentalreference.com

Read the article: Joiner, A., Pickles, M.J., Tanner, C., Weader, E., and P. Doyle. "An in Situ Model to Study the Toothpaste Abrasion of Enamel." Journal of Clinical Periodontology. June 2004, Vol. 31, pp. 434-438.

Free abstract online: http://www.ncbi.nlm.nih.gov