The evidence of nature and the nature of evidence.

s c a l h r e t s s “Thousands of tired, nerve-shaken, over-civilized people,” wrote conservationist John Muir over a century ago, “are beginning to fınd out that going o the mountains is going home; that wilderness is a ecessity; and that mountain parks and reservations are seful not only as fountains of timber and irrigating rivrs, but as fountains of life.” Muir was not breaking new round. Since ancient times, people intuitively have been rawn to places of natural beauty, seeking tranquility, estoration, spiritual fulfıllment, and even better health. Modern research offers tantalizing clues that corroborate his tradition. Consider the benefıts of trees. Studies have ound that post-surgical patients recover more smoothly hen they can view trees through their hospital windows; that public housing residents enjoy stronger social bonds, less violence, and (among girls) greater concentration, impulse control, and delayed gratifıcationwhen there are trees near their apartments; and that in neighborhoods with trees,newbornsare less likely tobe small forgestational age, children are less prone to become overweight, and elders ive longer. An article in this issue of the American Journal of Preventive Medicine adds to the literature on the health benefıts of trees. A beetle—the emerald ash borer—has wiped out millions of ash trees across 15 U.S. states over the last decade. Donovan and colleagues hypothesized that the loss of trees might threaten health—specifıcally, by increasing mortality from cardiovascular and lower respiratory disease. Their independent variables were the presence andduration of emerald ash borer infestation, at the county level, and the extent of ash tree loss (estimated based on county-level tree canopy data and state-level data on the proportion of ash in each state’s canopy, a parameter that nowhere exceeded 7.9%). The dependent variable was cause-specifıc mortality, also at the county level. Regression analysis controlled for such potential confounders as race and ethnicity, income, and education but not for smoking, diet, cholesterol, family history, or air quality. The authors found that ash tree loss was