Jerald Pinson is a Ph.D. candidate at the University of Florida studying long-lived fern gametophytes. More of his writing is available at All is Leaf.


Ferns are one of the oldest groups of plants on Earth, with a fossil record dating back to the middle Devonian (383-393 million years ago) (Taylor, Taylor, and Krings, 2009). Recent divergence time estimates suggest they may be even older, possibly having first evolved as far back as 430 mya (Testo and Sundue, 2016). However, despite the venerable age of the group as a whole, most of the earliest ferns have since gone extinct. Groups like the Rhacophytales, which were possibly some of the earliest progenitors of ferns, the ancient tree ferns Pseudosporochnales and Tempskya, and the small, bush-like Stauropterids have all long ago disappeared. The diversity of ferns we see today evolved relatively recently in geologic time, many of them in only the last 70 million years.


Today, ferns are the second-most diverse group of vascular plants on Earth, outnumbered only by flowering plants. With around 10,500 living species (PPG 1), ferns outnumber the remaining non-flowering vascular plants (the lycophytes and gymnosperms) by a factor of 4 to 1. How did ferns become so diverse, and what are the secrets to their success? What traits do they share in common, and how are they different from other groups of plants? What follows is a short primer on the biology of ferns, starting at the beginning, with how ferns first originated and evolved into the plants we see in the present, making special note of some of the groups that went extinct along the way. There are separate sections that cover topics ranging from fern morphology, phylogenetic relationships, and the fern lifecycle, along with the important role gametophytes play in the biology of ferns.



Figure 1. Stele structure of seed plants and ferns, with extinct relatives, adapted from Kenrick and Crane, 1997. The position of the protoxylem is denoted by circles and lines within the stele. Open circles denote protoxylem lacunae.

Figure 2. Selected anatomical traits of ferns. Species represented is Polypodium remotum Desv., illustrated in Garden Ferns (1862) by William Hooker.

Figure 3. Sporangia grouped into a sorus. Courtesy of Rogelio Moreno.

Partly because of their considerable age, ferns contain a high amount of diversity, with some groups that look nothing like the more common representatives we usually associate with ferns. There is consequently only one anatomical feature that unites them, an inconspicuous trait that requires observing the development of vascular tissue in the stem. According to Kenrick and Crane (1997), the mesarch (derived in the middle) protoxylem (protoxylem = the water-conducting cells that are the first to grow in a developing stem, the result of primary growth) in ferns is confined to lobes of the xylem strand (Fig. 1). This is opposed to the condition in seed plants in which the protoxylem also develops through the midpoints and center of the xylem strand in any given vascular bundle. Fortunately, further sub-divided groups within ferns have shared traits that are easier to observe.


Most ferns have rhizomes, underground stems from which the leaves are produced (Figure 2). Many ferns have long, creeping rhizomes that form intricate networks underground, and while the leaves may senesce and drop off due to age or cold weather, these rhizomes can persist indefinitely, sending up new leaves year after year. An entire leaf is called a frond, while further subdivisions are referred to as pinnae (first division), which grow along the main stem (called a rachis in ferns), and pinnules (subsequent divisions). The portion of the rachis without pinnae is referred to as the stipe (petiole), which attaches directly to the rhizome. Most fern fronds also have circinate vernation, in which the new growth is tightly coiled in a fractal spiral, which gradually unfurls as the leaf develops, protecting the meristem. This curling forms the familiar fiddlehead at the tip of new fronds. Ferns reproduce by spores, which are generally produced on the bottom (abaxial side) of leaves by specialized structures called sporangia. Sporangia can develop in clusters called sori, which can be circular (Figure 3), in distinct rows, or may even cover the entire underside of a leaf (acrostichoid sori) and are sometimes protected by an overhanging structure called an indusium. Other species have a sterile/fertile frond dimorphy, in which spores are produced on only certain leaves and not on others.



Figure 4. Individual sporangium with selective labeling. Courtesy of Rogelio Moreno.


Figure 5. Currently accepted phylogeny of ferns, according to the Pteridophyte Phylogeny Group (PPG 1).

Broadly speaking, ferns can be divided into two groups, the eusporangiates and leptosporangiates, with most of the diversity occurring in the latter. These terms refer to how sporangia develop and mature. In eusporangiates, a given sporangium develops from multiple initial cells on the surface of stems or leaves and consists of several cell layers in the early stages of development. Each sporangium can go on to produce several hundred spores. In contrast, leptosporangia arise from just one initial cell, which produces a stalked capsule that is just one cell layer in thickness. These sporangia also have a row of hollow cells arranged along two-thirds of the upper surface that fill with water (Figure 4). The thin, membranous cells are highly permeable, easily allowing water to evaporate.  This causes tension to build within the water column inside, forcing the remaining water to contract, which causes the annulus to slowly pull open the thin-walled sporangium, exposing the spores within. But at some point, the tension in the column becomes stronger than the adhesion properties of water, and the column snaps, which jettisons the spores at high speeds into the surrounding environment.


The eusporangiates are comprised of the horsetails (Equisetales), whisk ferns (Psilotales), moonworts (Ophioglossales), and marattioid ferns, which altogether number about 255 species (PPG 1) (Figure 5). The exact relationships of the first three groups were for a long time unknown; it was unclear whether they represented true ferns or were actually the last vestiges of ancient plant groups that were entirely separate from ferns. For this reason, these groups were often referred to as the fern allies. Recent molecular work, however, has demonstrated that the whisk ferns (Psilotales) and moonworts (Ophioglossales) are unequivocally ferns and that the horsetails are sister to all other species within the fern clade (Knie et al., 2014; Rothfels et al. 2015). Many researchers now use the term ‘monilophyte’ to encompass all of these groups, including all eusporangiate and leptosporangiate clades.


The leptosporangiates contain the bulk of fern diversity, comprised of some 10,323 species, grouped into 44 families (PPG 1). Most leptosporangiate ferns, as well as all eusporangiates, are homosporous, meaning that each species produces spores of only one size. The aquatic ferns in the order Salviniales are the only exception to this rule, having heterosporous spores. In this condition, a single plant produces both small microspores, which develop male gametophytes, and a few much larger megaspores, which develop into endosporic female gametophytes. Whereas the gametophytes of most species will break open the spore casing upon germination, becoming independent and photosynthetic, the female gametophytes of heterosporous species are retained within the megaspore and are dependent on stored lipids and carbohydrates for nutrition. It’s likely that the retention of the female gametophyte in a heterosporous lineage of plants led to the evolution of the first seeds.



Table 1. List of ferns and extinct relatives with their associated fossil record dates. Those groups that still exist are marked as ‘present’ (from Taylor, Taylor, and Kring, 2009).

The Devonian was a period of major change for the planet. The ancestors of green algae had migrated from their marine and freshwater environments onto land earlier in the Paleozoic era and began to evolve stems and roots to enable their survival in the harsh conditions they faced in Earth’s prehistoric terrestrial environments. These first plants, however, lacked true leaves for millions of years, instead possessing chloroplasts in their stems to enable photosynthesis. In the middle and late Devonian, however, as plants began to spread throughout the world’s ecosystems, they locked up a significant amount of CO2 through burial and the weathering of bare rock, causing the planet to cool (Mora, Driese, and Calarusso, 1996). This meant that plants could evolve bigger structures to intercept more light without overheating, and it is during this time that the first leaves begin to appear (Beerling, Osborne, and Chaloner, 2001). While leaves likely evolved multiple times in land plants, the earliest ancestors of ferns were some of the first to possess them.


Some extant species of ferns have either extremely small and specialized leaves (horsetails) or even no leaves at all (whisk ferns), but there is evidence to suggest that this wasn’t always the case for these groups. Some of the ancestors of modern horsetails, the Calamites, grew to the size of trees and had leaves with prominent vascular bundles (Taylor, Taylor, and Krings, 2009).


While ferns first evolved in the Devonian, they became one of the most dominant groups of plants on the planet during the Carboniferous (299-369 mya). Growing alongside the giant tree lycophytes (e.g., Lepidodendron) in vast swamps, ferns thrived and diversified for several million years. Leptosporangiate ferns evolved during this time and underwent the first of three major radiations, giving rise to several families (Rothwell and Stokey, 2008).


When these plants died, they sank into the anoxic swamps, where the lack of oxygen prevented bacteria from degrading dead tissue. The rampant growth in these swamps, and their subsequent burial, created most of the coal and natural gas deposits we have today. Every time you drive your car, you’re using fossilized ferns to reach your destination.


As the Carboniferous came to a close, most of the first leptosporangiate families to have evolved gradually went extinct. At least one lineage survived, however, to give rise to the second major radiation of leptosporangiate ferns, which began in the late Permian (~250 mya) (Rothwell and Stokey, 2008). Some of the oldest fossils from this diversification are of the Osumundales, which include species such as the cinnamon fern (Osmundastrum cinnamomeum) and royal fern (Osmunda regalis) (see table 1 for a list of families and their ages). Most of the groups that evolved during this time have survived to the present, and while they contain a modest amount of diversity, the third and final radiation gave rise to the greatest bulk of fern species by far.


About 135 mya, during the Cretaceous, a small group of plants evolved that would quickly and drastically change the planet’s ecosystems. Originating in the tropics, flowering plants (angiosperms) rapidly diversified and spread to all major portions of the globe, driving several groups of plants to extinction and severely reducing the diversity in others. Leptosporangiate ferns appear to be the only group of vascular plants that thrived alongside angiosperms, rather than being marginalized (Schuettpelz and Pryer, 2009). The advent of towering, angiosperm-dominated rainforests in the tropics opened up new environments that ferns were able to successfully exploit and diversify in, which led to the third radiation of leptosporangiate ferns. Consequently, most species of ferns today grow in the tropics. Costa Rica, for example, is smaller than the state of West Virginia and yet has nearly 3X as many fern species as the entire continental United States and Canada combined.



Figure 6. Lifecycle of ferns, depicting the various modes of reproduction that can take place, excluding asexual reproduction, such as apomixis (from Sessa, Testo, and Watkins, 2016).


Figure 7. Various morphologies of fern gametophytes (from Pinson et al., 2017 and Paul K).

Across the land plant phylogeny, there is a pattern of increasing sporophyte complexity along with an associated decrease in the independence of the gametophyte portion of the lifecycle. Bryophytes, which were the first plants to colonize land, grow as independent gametophytes that produce nutritionally-dependent sporophytes. Conversely, on the opposite end of the land plant phylogeny, the seed plants have dominant sporophytes with dependent gametophytes that have been reduced to just a few cells. Ferns and lycophytes, which span the evolutionary gap between these lineages, are the only groups of plants in which both the sporophyte and gametophyte are completely independent of each other.


In ferns, a mature sporophyte will develop haploid spores via the process of meiosis. Once these spores mature, they are dispersed into the surrounding environment and will eventually germinate into gametophytes. In eusporangiate ferns, the gametophytes are subterranean (with the exception of Marattioid ferns) and non-photosynthetic, obtaining carbohydrates from a symbiotic relationship with a fungus. In homosporous leptosporangiate ferns, the gametophytes grow above ground and are photosynthetic. Gametophytes may either go on to produce both male (antheridia) and female (archegonia) sex organs, or they may produce them separately. Sperm in all ferns are motile, possessing several flagella that allow them to travel short distances. Many leptosporangiate ferns have small, heart-shaped (cordate) gametophytes that must therefore grow close enough together to allow for sperm to swim between them in order for outcrossing to occur (although some ferns are capable of self-fertilization). Once the sperm has united with the egg, a new diploid sporophyte grows directly from the gametophytic tissue, after which the gametophyte senesces and/or is subsumed within the new growth (Figure 6).


The heart-shaped gametophytes of most leptosporangiate ferns are often found in recently disturbed areas, as spores buried in the soil are then exposed and capable of germinating with little surrounding competition. The gametophytes then grow quickly in order to establish new sporophtyes before the next disturbance (Watkins, Mack, and Mulkey, 2007). A large percentage of ferns (~10%), however, are epiphytic, from having diversified in the canopies of angiosperm-dominated forests in the Cretaceous (Schuettpelz and Pryer, 2009). Life in tropical canopies imposes entirely different constraints on the growth of both sporophytes and gametophytes, requiring that each stage of the lifecycle adapt in order to survive. Tropical canopies support dense vegetation, which makes it hard for new plants to become established and compete for the limited available resources. Dense mats of bryophytes also hinder the growth of the gametophyte mats needed for outcrossing. The gametophytes of many epiphytic ferns consequently have a much more branched and dissected morphology than their terrestrial counterparts (either ribbon-shaped, filamentous, or strap-shaped), which is capable of continued meristematic growth (Figure 7) (Dassler and Farrar, 2001). This not only allows them to compete in a dense network of bryophyte growth, but if two spores land a significant distance from each other on the same tree, rather than having to undergo self-fertilization, these gametophytes can grow until they are close enough for sperm to swim between them.


The ability to grow continuously, and often asexually, in these gametophytes means that they can live indefinitely. Because of their small size, they can also exploit small, protected microhabitats in areas where conditions are otherwise unfavorable for their growth. Because the sporophytes are much larger, this means that gametophytes can often grow in places where the sporophyte can’t, which has led to a spatial separation of the two generations (Pinson et al., 2017). Gametophytes of some species have also been shown to tolerate a wider range of environmental conditions than their sporophyte counterparts (Watkins and Cardelús, 2009; Sato and Sakei, 1981). Around thirty known species of ferns have distributions in which the gametophyte occupies a wider geographic range than its sporophyte, and at least three fern species have no known sporophyte anywhere on Earth.

Literature cited

Beerling, D. J., Osborne, C. P., & Chaloner, W. G. (2001). Evolution of leaf-form in land plants linked to atmospheric CO2 decline in the Late Palaeozoic era. Nature. 410: 352-354.

Dassler, C. L., & Farrar, D. R. (2001). Significance of gametophyte form in long-distance colonization by tropical, epiphytic ferns. Brittonia. 53: 352-369.

Kenrick, P., & Crane, P. R. (1997). The origin and early diversification of land plants a cladistic study. Washington, DC: Smithsonian Institution Press.

Knie, N., Fischer, S., Grewe, F., Polsakiewicz, M., & Knoop, V. (2015). Horsetails are the sister group to all other monilophytes and Marattiales are sister to leptosporangiate ferns. Molecular phylogenetics and evolution. 90: 140-149.

Mora, C. I., Driese, S. G., & Colarusso, L. A. (1996). Middle to late Paleozoic atmospheric CO2 levels from soil carbonate and organic matter. Science. 271: 1105.

Pinson, J. B., Chambers, S. M., Nitta, J. H., Kuo, L. Y., & Sessa, E. B. (2017). The Separation of Generations: Biology and Biogeography of Long-Lived Sporophyteless Fern Gametophytes. International Journal of Plant Sciences. 178: 1-18.

PPG 1: The Pteridophyte Phylogeny Group*. (2016) A community-derived classification for extant lycophytes and ferns. Journal of Systematics and Evolution. 54: 563-603. *This project was organized by E Schuettpelz, H Schneider, AR Smith, P Hovenkamp, J Prado, G Rouhan, A Salino, M Sundue, TE Almeida, B Parris, EB Sessa, AR Field, AL de Gasper, CJ Rothfels, MD Windham, M Lehnert, B Dauphin, A Ebihara, S Lehtonen, PB Schwartsburd, J Metzgar, L-B Zhang, L-Y Kuo, PJ Brownsey, M Kato, and MD Arana, with 68 additional contributors.

Rothfels, C. J., Li, F. W., Sigel, E. M., Huiet, L., Larsson, A., Burge, D. O., ... & Shaw, S. W. (2015). The evolutionary history of ferns inferred from 25 low-copy nuclear genes. American Journal of Botany. 102: 1089-1107.

Rothwell, G., & Stokey, R. (2008). Phylogeny and evolution of ferns: a paleontological perspective. In T. Ranker & C. Haufler (Eds.), Biology and Evolution of Ferns (pp. 332-366). Cambridge: Cambridge University Press.

Sato, T., & Sakai, A. (1981). Cold tolerance of gametophytes and sporophytes of some cool temperate ferns native to Hokkaido. Canadian Journal of Botany. 59: 604-608.

Schuettpelz, E., & Pryer, K. M. (2009). Evidence for a Cenozoic radiation of ferns in an angiosperm-dominated canopy. Proceedings of the National Academy of Sciences, 106: 11200-11205.

Sessa, E. B., Testo, W. L., & Watkins, J. E. (2016). On the widespread capacity for, and functional significance of, extreme inbreeding in ferns. New Phytologist. 211: 1108-1119.

Taylor, E. L., Taylor, T. N., & Krings, M. (2009). Paleobotany: the biology and evolution of fossil plants. London: Academic.

Testo, W., & Sundue, M. (2016). A 4000-species dataset provides new insight into the evolution of ferns. Molecular phylogenetics and evolution. 105: 200-211.

Watkins Jr, J. E., & Cardelús, C. (2009). Habitat differentiation of ferns in a lowland tropical rain forest. American Fern Journal. 99: 162-175.

Watkins, J. E., Mack, M. K., & Mulkey, S. S. (2007). Gametophyte ecology and demography of epiphytic and terrestrial tropical ferns. American Journal of Botany. 94:701-708.

Banner image of fern sporangia by Rogelio Moreno.