Â鶹ÒùÔº

Herbaria, or "museums for plants," house millions of samples collected over centuries. Stanford biologist Barnabas Daru shares how modern technology can unlock this treasure trove, providing scientists with vital insights into plants and the ecosystems they support.

A single dried flower pressed flat on a page of white cardboard is a plant sample made using very old technology. People have been collecting these types of samples since the 1600s.

Armed with very new technology, such as artificial intelligence and genomic sequencing, Stanford biologist Barnabas Daru sees tremendous opportunity in studying these collections that span centuries.

Researchers could potentially harvest a wealth of information from the millions of samples housed in the world's 3,000 herbaria, what Daru calls "museums for plants." The data could help identify long-term trends in plant growth and yield insights on our own rapidly changing environment.

"So much data on plants are captured in herbarium specimens, but they are untapped and unused," said Daru, assistant professor of biology in Stanford's School of Humanities and Sciences. "It's the perfect time to explore these resources that have been hidden away in storage cabinets. Given the new technologies that are available, we can really use the specimens to understand plant biodiversity in new and exciting ways."

For instance, genomic sequencing makes it possible to take a small piece of leaf tissue from herbarium specimens and see how the genetics of a plant species have changed over time. AI could be used to help analyze a huge number of samples to look for visual changes in a species, such as differences in leaf size or the number or shape of flowers.

The research potential held in herbaria inspired Daru to write two papers detailing the opportunities for other scientists. One, published in , focused on using herbarium resources to help determine how plant species have shifted to different regions over time—valuable knowledge that could help predict future movement of species under environmental change.

The second paper, published in , describes how herbarium samples might be used to study plant ecological interactions with other living things, such as herbivores, pollinators, microorganisms, or plant competitors.

"Plants are at the beginning of the food chain, so anything that affects plants will have a cascading effect on every organism that depends on them, including humans," Daru said.

A lot can be learned about herbivores from the bite marks on a leaf, for example. Half-moon shapes or a skeletonized leaf can indicate what type of plant-eating insects were present. Insects make up the greatest number of herbivores, but larger herbivores like deer and elk leave marks that can also be identified.

In addition, crystallized nectar can provide clues about past pollinators. Daru's lab team is currently working on a project led by doctoral student Lauren Puleo to rehydrate old nectar from herbarium specimens and then use mass spectrometry to analyze the molecules and identify the types of sugars present.

The sugars can indicate what types of pollinators—such as bees, hummingbirds, and butterflies—might have interacted with the plants 50 or even 100 years ago.

Plant microorganisms like bacteria, fungi, and viruses are also critical to track. While some are beneficial to their plant hosts, many also act as pathogens, like the fungus that caused Dutch elm disease, leading to a mass die-off of the tree in Europe and North America in the last century. When a pathogen affects a major crop, it also threatens people, as did the fungus that caused the Irish potato famine in the 1800s.

Dead microorganisms can be found preserved in herbarium specimens, which makes it possible to learn more about these historical microbial interactions with plants. This type of knowledge from the past might help prevent the next mass die-off of a crop or tree species, Daru said.

There is also simply a need for foundational and baseline information about past plant diversity, distribution, and interactions.

"It's important to understand the state of biodiversity historically, before humans started impacting the environment, because if we can know that trajectory, then it's easier to predict how it's going to look in the future," he said. "Herbarium samples can enable that."