American Society of Naturalists

A membership society whose goal is to advance and to diffuse knowledge of organic evolution and other broad biological principles so as to enhance the conceptual unification of the biological sciences.

“Why are there so many flowering plants? A multi-scale analysis of plant diversification”

Posted on

Tania Hernández-Hernández and John J. Wiens (June 2020)

A new study helps explain the extraordinary richness of flowering plants (angiosperms) relative to other plant groups

Read the Article (Just Accepted)

Solving Darwin’s “abominable mystery”: why are there so many flowering plants?

A butterfly (Nymphalidae) and flowering plant (Asteraceae) from southeastern China.<br />(Credit: John J. Wiens)
A butterfly (Nymphalidae) and flowering plant (Asteraceae) from southeastern China.
(Credit: John J. Wiens)

For hundreds of years, researchers have tried to understand why most land plant species are flowering plants, or angiosperms. Roughly 90% of land plants are angiosperms, even though angiosperms are young relative to other major groups, such as mosses, ferns, and liverworts. More than 200 years ago, Darwin considered the cause of the rapid proliferation of angiosperm species to be an “abominable mystery.”

A new study may have finally solved this mystery. Working at the University of Arizona, Tania Hernández-Hernández and John J. Wiens generated a massive dataset of 31 traits for all 678 families of land plants. They then analyzed which traits best explained why some groups proliferated rapidly (like angiosperms) whereas others did not (like mosses). In contrast to earlier studies, they included dozens of traits and compared angiosperms to other groups to understand their rapid proliferation, rather than comparing a few traits within angiosperms.

A pincushion cactus (<i>Mammillaria</i> sp.) in flower from southern Arizona.<br />(Credit: John J. Wiens)
A pincushion cactus (Mammillaria sp.) in flower from southern Arizona.
(Credit: John J. Wiens)

They find that fertilization of plants by animals (e.g. insect pollination) was the most important trait for explaining the rapid radiation of angiosperms. This result suggests that flowers are they key trait that explains the incredible success of angiosperms.

They also find that many other patterns of diversity across plants are explained largely by how widely distributed each group is (range size). Widely distributed families radiate more rapidly. This pattern helps explain why some plant families have been more successful than others, across plants and within angiosperms, ferns, mosses, and gymnosperms.

Finally, the study shows how traits associated with interactions between species (like insects pollinating plants) can be important for explaining diversity patterns at very deep timescales, among clades that are hundreds of millions of years old. In contrast, large-scale geographic factors (like range size) are more important among younger clades. This pattern may apply to many other groups of organisms.


A club moss (Lycopodiophyta) from southern China.<br />(Credit: John J. Wiens)
A club moss (Lycopodiophyta) from southern China.
(Credit: John J. Wiens)

Abstract

The causes of the rapid diversification and extraordinary richness of flowering plants (angiosperms) relative to other plant clades is a long-standing mystery. Angiosperms are only one among 10 major land plant clades (phyla), but include ~90% of land plant species. However, most studies that have tried to identify which traits might explain the remarkable diversification of angiosperms have focused only on richness patterns within angiosperms, and tested only one or a few traits at a single hierarchical scale. Here, we assemble a database of 31 diverse traits among 678 families and analyze relationships between traits and diversification rates across all land plants at three hierarchical levels (phylum, order, family) using phylogenetic multiple regression. We find that most variation (~85%) in diversification rates among major clades (phyla) is explained by biotically mediated fertilization (e.g., insect pollination) and clade-level geographic-range size. Different sets of traits explain diversification at different hierarchical levels, with geographic-range size dominating among families. Surprisingly, we find that traits related to local-scale species interactions (i.e. biotic fertilization) are particularly important for explaining diversification patterns at the deepest timescales, whereas large-scale geographic factors (i.e. clade-level range size) are more important at shallower timescales. This dichotomy might apply broadly across organisms.