Left, Gymnocarpium dryopteris; right, Cystopteris fragilis; middle, their love child ×Cystocarpium roskamianum" Fig. A1 from Rothfels et al. 2015. Click image for link.

An unassuming little fern has left scientists scratching their heads at the feat of reproductive hijinks it apparently represents.

The fern, xCystocarpium roskamianum (the prefix 'x' indicates it is a hybrid), collected in the French Pyrenees, appeared to be a blend of two ferns they know well. Although this fern is infertile as many hybrids are (think mules), it propagates asexually with ease via underground stems called rhizomes.

Plant hybrids are common, and plant breeders take advantage of plants' apparently more laissez-faire attitude to species boundaries to create hybrids all the time. And as you can see in the photographs above, xCystocarpium roskamianum looks very much like an intermediate between the two ferns sandwiching it: Gymnocarpium dryopteris (oak fern), and Cystopteris fragilis (fragile fern).

But unlike most plant hybrid parents, these two ferns are only distantly related -- so distant, in fact, they had even been placed in different families, mighty evolutionary gulfs in the plant world. Was this new fern indeed the offspring of these two ferns? And if so, just how unlikely were these two parents? A team of American, Canadian, and Dutch scientists wanted genetic proof that this was what had happened, and they also wondered just how distantly related the two putative parents were.

As reported in the March issue of The American Naturalist, the scientists compared the sequence of a marker gene previously shown to be a good choice for studying fern evolution and confirmed that Cystocarpium's parents were indeed the ferns Gymnocarpium dryopteris and an undetermined fern in the Cystoperis fragilis complex (there are many confusing fern populations in this lineage so it is still impossible to tell exactly which one). They were even able to tell which parent was which: because only eggs contribute the cellular organelles called mitochondria and chloroplasts to offspring and these organelles contain their own DNA, they could tell that Gymnocarpium was the mother, and Cystopteris the father.

After comparing the DNA of the parent ferns, the scientists calculated that the most likely divergence date was about 57.9 million years ago, give or take a few ten million years. This is amazing. It is the equivalent, the authors claim, of an elephant producing offspring with a manatee (a manaphant?) or a human with a lemur (a humur?).

On the other hand, the divergence of Cystocarpium from its parents was slight -- it differed at most by a single point mutation from one of its parents in the gene the scientists examined. The freak fertilization that produced the hybrid happened very recently indeed.

As far as the scientists know, Cystocarpium represents the deepest natural hybridization event on record in either the plant or animal kingdom. There are examples that are close, though. Interestingly, they all seem to occur in plants that rely on non-living intermediaries to help sperm meet egg, namely, wind and water. Conifers and ferns are two examples of plants that use these systems, but notably, so do the flowering plants called grasses. That is intriguing because, as mentioned above, grasses are also the current deep hyridization record holders among the flowering plants. There is one possible animal exception: viable hybrids exist both in captivity and possibly in the wild between guinea fowl and chickens, whose last common ancestor lived 30-70 million years ago.

The little fern's existence also touches on one of the larger questions in evolutionary biology: what are the relative contributions of gene flow (i.e. interbreeding) vs. the evolution of reproductive barriers to the generation of new species and thus biodiversity? Reproductive barriers are mechanisms that prevent interbreeding like pollinators that are specific to a species, particular mating dances or other rituals, or even cellular incompatibilities that terminate the formation of an embryo.

Interbreeding/gene flow tends to homogenize populations, while the formation of reproductive barriers permits local adaptation and the evolution of new species. Scientists already knew that in general (and unsurprisingly), the longer it has been since two groups of organisms separated, the greater the reproductive barriers between them. Usually, incompatibility occurs within a few million years. But the pace is not the same between different groups of organisms. The "incompatibility clock" ticks faster for fruit flies (no more than 4 million years to reproductive incompatibility) than for sun fish (37 million years under laboratory conditions; 15 million years in the wild).

Flowering plants seem to hybridize more readily than animals, but their incompatibility clocks seem to tick at about the same rate. Two grasses -- which are indeed flowering plants in spite of their lack of obvious flowers -- managed to hybridize after 14 million years apart, but that is the current record. Hence the scientists' surprise upon encountering Cystocarpium, whose parents' estrangement seemed to encompass a much longer span.

In contrast to plants that rely on indifferent forces of nature to mate, flowering plants often rely on animal pollinators, while animals rely on complex mate-recognition systems, both of which themselves are subject to rapid natural selection. Wind and water are not. The result is that organisms like ferns that rely on aquatic or aeronautic matchmaking are more subject to interbreeding (i.e. gene flow) and thus less likely to speciate. As a consequence, they are much less biodiverse. Indeed, several studies have found lower speciation rates in abiotically pollinated flowering plants compared with their biotically pollinated kin.

This raises the intriguing possibility that the patterns of diversity we see today are not just driven by selection at the individual level, but also by selection operating at a higher level, a concept called "species selection" championed by Stephen Jay Gould, among others. If the idea has merit, the reason there are so many flowering plants and so few non-flowering plants may have nothing to do with any innate superiority or special fitness on the part of flowering plants'. Rather, it may simply be that the non-flowering plants have a lower birth rate of species.

As the authors put it, "One reason we live in a world with more than 250,000 species of flowering plants but only around 10,000 fern species (and approximately 1,000 gymnosperms, 1,200 lycophytes, 12,000 mosses, 9,000 liverworts, and 100 hornworts) may just be that populations of non-flowering lineages take longer to achieve complete genetic separation from one another because they have fewer mechanisms to prevent the sperm of one species from encountering the egg of another."

As such, Cystocarpium may provide strong evidence that more evolved and biodiverse is not necessarily better -- just different.


Rothfels C.J., Peter H. Hovenkamp, David L. Swofford, Harry C. Roskam, Christopher R. Fraser-Jenkins, Michael D. Windham & Kathleen M. Pryer (2015). Natural Hybridization between Genera That Diverged from Each Other Approximately 60 Million Years Ago, The American Naturalist, 185 (3) 433-442. DOI: http://dx.doi.org/10.1086/679662