Uncovering the beauty of flowers in the garden — and in 3D

Adam Roddy spends a lot of time among the flowers.


An evolutionary physiologist and assistant professor in FIU’s Institute of Environment, Roddy is fascinated by the structure of plants and how they’ve changed over time — especially their flowers.

Out of the estimated 400,000 flowering plants on Earth that bloom into a kaleidoscope of colors and shapes, each species evolved to be completely unique.

For years, Roddy has been trying to understand how. He’s led and developed innovative methods to understand and uncover some of the mysteries of flowers. He was a part of a team recently awarded a nearly $700,000 grant from the National Science Foundation (NSF) for a 3D modeling project to continue his work.

While there’s a lot of research examining the leaves and stems of plants, flowers have been somewhat ignored–at least in terms of their physiology. As a doctoral student at UC Berkeley, Roddy attempted to measure the way flowers use water. Just as a leaf loses water during photosynthesis, he tried to measure how much water flowers lose. He soon discovered there was almost nothing to measure. Surprisingly, flowers don’t lose water.

“Eventually, I realized this was the story — flowers are very efficient at not losing water. While leaves have evolved to transport lots of water because it enables high rates of photosynthesis necessary for growth, flowers have evolved in the opposite direction,” Roddy said. “Immediately, I knew I had to move beyond physiology to explore the interesting evolutionary questions. What are costs of producing and maintaining a flower, and how have these costs evolved?”

Roddy followed these questions. And those questions laid the groundwork for his research.

3D image of flowers

As a postdoctoral researcher at Yale, Roddy started using 3D imaging technology — something no one else in his field had attempted. Essentially, it’s a medical CAT scan, but much higher resolution. Located at Lawrence Berkeley National Laboratory in California, the 3D scanner takes thousands of images of a single flower petal from every angle. Then, these thousands of images are turned into a 3D image that give a glimpse at the tissue and cells beneath the flower petal’s surface.

The earliest scans were spectacular, if not slightly baffling. Roddy realized a flower’s tissue is incredibly porous with a lot of airspace between the cells. To Roddy, it didn’t make sense how flowers were able to physically hold themselves together.

Knowing this was a question of physics, he collaborated with a team of physicists at Yale. They are currently working together to turn those 3D scans into models that revealed how the cells grow and change shape over time as a bud blooms into a flower. It’s like a dance, of sorts. The cells form the perfect arrangement to support the flower petal while also being cheap to produce.

The NSF grant will allow Roddy and his collaborators to compare different species in an evolutionary context. Plants all have the same cellular building blocks to work with, but end up forming radically different flowers. This means the arrangement of those cells varies from species to species. The 3D models will help illustrate some of the fine scale innovations that have taken place to effectively produce those various types of flowers.

Those innovations have happened because the primary goal of flowering plants is to attract pollinators. If those pollinators are more attracted to big, showy flowers, the plant will respond by producing larger flowers. But, larger flowers require more resources.

Natural selection helps maintain plant diversity, research finds

Roddy thinks that plants have evolved around this potential cost. By producing flowers that are cheap and efficient, larger and more abundant flowers can be produced. The team’s early results suggest that newer flower species, like rhododendron, have evolved to be very cheap to produce, requiring far less water and carbon.

Roddy’s research helps build on what’s currently known about plants and flowers. It also has potential applications. Comparing the 3D modeling to real flowers could help in the development of new biomimetic materials and tissues with tunable properties.

After working in the northeast and western U.S., Roddy can’t wait to work with the big, colorful tropical flowers that grow in South Florida. To him, they are breathtaking, and hold as much mystery as they do beauty. As always, though, he’s ready to follow the questions — wherever they lead.