Plants & People

Historical trends in herbarium specimens could cause misleading observations

Extra care should be taken to avoid misinterpretation of data when looking at centuries-spanning specimens.

Over the past few decades, herbarium specimens have increasingly been used to study long-term changes in plant traits, phenology, and insect herbivory. In order to do this, recent specimens must be compared to older ones, and differences in key features tracked such that patterns can be discerned. But what if other factors are at play? Are there trends in the creation of herbarium specimens over the years that could skew these patterns? And does the role of early herbaria as objets d’art of sorts play a part?

In a new article published in Annals of Botany, lead author Mikhail V. Kozlov and colleagues considered whether trends and practices in the creation of herbarium specimens could be having an unanticipated effect on global change research. The authors measured nine mostly aesthetic characteristics from over 500 specimens of 20 species of common European trees and shrubs. These specimens were collected over more than four centuries, from 1558 to 2016. The researchers then had 23 botanists and 21 artists rank the specimens by scientific and aesthetic value, respectively. To avoid bias, those ranking the specimens were not aware of specimen ages or the purpose of the study.

Solanum lycopersicum var. lycopersicum. Detail of a herbarium sheet with the oldest conserved tomato plants of Europe. Italian collection, 1542-1544. En Tibi herbarium at Naturalis, Leiden. Image: Naturalis Biodiversity Center/Wikimedia Commons.

The authors found that a number of characteristics of herbarium specimens have changed “systemically and substantially” over the course of almost 500 years of pressed-plant preservation. For example, the average number of leaves mounted per sheet increased threefold, likely in large part because of the increasing recognition of the amount of plasticity within the leaves of an individual.

Conversely, the quality of specimen preparation, as judged by the number of folded leaves, decreased. “The overall impression is that, although the folded leaves decreased both the scientific and the aesthetic value of the herbarium specimens, many botanists were not motivated to invest more time in this process, as indicated by the increase in the proportion of folded leaves that paralleled the increase in leaf numbers,” write the authors.

Rankings of the artistic value of herbarium specimens did not change with time, but there was a positive, though weak, correlation between the scientific and artistic value of a given specimen. The proportion of specimens that included reproductive structures increased until the mid-1800s and then decreased again over the past 150 years for reasons that are not clear.

Finally, the increase in number of leaves per sheet over several centuries seems to have led to a preference for branches with smaller leaves. In fact, several of the species tested had smaller average leaf sizes on herbarium specimens than is given as an average for the species in guidebooks, pointing to a systemic bias. This bias could affect estimates of insect herbivory because a greater number of leaves will have a greater chance of displaying insect damage than a small number that can be found in pristine condition, thus mimicking the expected effect of increased insect damage due to climate change.

“This discovery has direct implications for global change research, because the historical patterns in both plant traits and levels of herbivory, derived from studies of herbarium specimens, may reflect shifts in plant collection and preservation practices rather than the effects of past environmental changes on plant characteristics,” write the authors. “Similarly, the changes in plant collection practices may prevent the identification of actual temporal trends in plant traits.”