Publication Date



Dana Royer


Earth & Environmental Sciences


English (United States)


Paleobotanists have debated whether the fossil leaf record is biased towards an overrepresentation of small leaves. Previous studies have found a "small leaf bias" in both present-day leaf assemblages and fossil floras. However, few of these studies exist, especially in fossil floras, due to a lack of reliable methods for reconstructing full leaf area from fragments. Here, using a methodology for inferring leaf area from vein density scaling in fragments, I examine whether a small leaf bias exists in paleofloras from the Fort Union and Hell Creek Formations in the Williston Basin of southwestern North Dakota and northwestern South Dakota. Since paleobotanists use leaf area of complete leaves to estimate mean annual precipitation (MAP), I also tested whether including reconstructed areas from fragments can skew MAP measurements. I found no meaningful difference between the average reconstructed areas of complete and fragmented fossils in 42 morphotype-site pairs, suggesting that these floras are not considerably biased against large leaves. Differences between MAP estimates made using complete leaf areas and those made using reconstructed areas of both completes and fragments did not seem particularly meaningful, though other paleobotanists may disagree. My results indicate that paleobotanists working on floras from these locations can be confident that complete leaf fossils represent average leaf size with reasonable accuracy. Additionally, small leaf bias will probably not skew paleoclimate estimates made using leaves from these floras.



© Copyright is owned by author of this document