WOODS HOLE, Mass. -- As biologists know, nature can take its sweet time explaining itself. Andrew Gillis , associate scientist at the Marine Biological Laboratory (MBL), has been investigating how the paired fins of fishes evolved for nearly 20 years - ever since he was a PhD student with Neil Shubin at the University of Chicago.
A new study from Gillis and the MBL-UChicago Graduate Research Fellowship Program doesn't quite put the matter to rest – but it gives important insight to a broader question in evolutionary biology.
Published in Proceedings of the National Academy of Sciences , the study shows that two different cell populations in the embryo can give rise to the same repeating structures (or "serial homologs") in the adult -- in this case, the paired fins and gill arches (repeating bars of cartilage) in the little skate. The developmental equivalency of these different cell populations, the study proposes, is what produces serially homologous structures, which may hold true for the development of other repeating body parts, such as vertebrae, fingers or toes.
"You know it when you see it," Gillis says of serial homology. "It's variations on a theme. For example, cervical and thoracic and lumbar vertebrae all look different, but you see a common pattern. However, trying to explain why something is serially homologous is surprisingly difficult."
A traditional explanation for serial homology is the transformation of one body part to another, over long expanses of evolutionary time. For example, insect wings evolved through the gradual transformation of leg segments, as shown by Heather Bruce and Nipam Patel in 2022 . In the 1870s, biologist Karl Gegenbauer proposed that the paired appendages (fins/limbs) of jawed vertebrates evolved from transformation of gill arches, bony bars supporting the gills in fish and amphibians.
"Our study offers a new way of thinking about serial homology that doesn't necessarily have to invoke one thing transforming into another," Gillis says. "We are trying to define serial homology by explaining it from a developmental perspective."
Exploring evolutionary change
Two decades ago, Gillis set out to explore Gegenbauer's hypothesis for the evolution of paired fins by transformation of a gill arch. Over the course of several years and published papers, Gillis found many common genes and pathways controlling the development of the gill and fin skeletons in the little skate (Leucoraja erinacea), indicating they had common evolutionary trajectories.
"But a question that kept nagging at us was the fact that the gill skeleton and the fin skeleton come from very different parts of the embryo, from different germ layers," Gillis says. Gill arches were thought to derive from the neural crest, while paired fins/limbs derived from lateral plate mesoderm.
However, in a surprising 2020 study , Gillis's former postdoctoral fellow, Victoria Sleight, mapped out the germ layers of the early skate embryo and discovered an overlap in the cell populations that make gills and fins.
"We found there's not this sharp boundary, where these cells make gills and other cells make fins, but actually a blending of the cells," Gillis says. Both structures developed from a common pool of mesoderm and neural crest cells that apparently had the potential to become either body part.
The present paper experimentally tested this finding. First author Michael Wen , a University of Chicago PhD student working with co-advisors Gillis and UChicago's Victoria Prince , carried out the experiments. Wen took cells from the neural crest stage of the embryo, which normally form the gill arch skeleton, and transplanted them into a developing fin. The cells incorporated normally into the fin skeleton. He also transplanted mesoderm cells into a developing jaw (the most anterior serial homologue of the gills), and they incorporated normally as well.
"What this means is the cells making these two body parts are equivalent and interchangeable," Gillis says. "We propose that is why the structures that form from these cells are serially homologous." The cells can receive whatever information is fed to them from the environment, be it "make a fin" or "make a gill."
"The similar response of the cells to the environment may be encoded in the genome," Wen says. "How similar the genomic landscapes are between cells may provide another layer to our explanation of serial homology and is something we are now investigating."
Looking at other examples of serial homology, "I would bet if you transplanted a lower-back skeletal cell to the neck during embryonic development, it would behave like a neck cell," Gillis says.
So, do we now understand how paired fins evolved? That's still an open question, due to lack of fossil evidence.
"Unlike the fin-to-limb transition, where we have all these nice fossils showing the gradual transformation of one part to another, we don't have that for the origin of fins," Gillis says.
Gillis's lab, meanwhile, has moved on to a focus on other developmental investigations. "But I feel like we've made a lot of interesting discoveries around this question," he says.
This research was supported by the National Science Foundation, the Owens Family Foundation, the Biological Sciences Division (BSD) of the University of Chicago, the John W. and Valerie A. Rowe MBL/University of Chicago Graduate Research Fellowship Program, University of Chicago BSD Dean's International Student Fellowship, the UChicago O'Brien- Hasten Fellowship, and an American Heart Association Predoctoral Fellowship.
Citation:
M.T.C. Wen, V.E. Prince, & J.A. Gillis (2026). Shared competence forms the basis of gill arch and paired fin serial homology. Proc. Natl. Acad. Sci. U.S.A, DOI: 10.1073/pnas.2529365123 .
Photo/video captions:
Photo 1: A skeletal preparation of a little skate (Leucoraja erinacea) hatchling, ventral view. This image shows the head in ventral view, with the jaws, gills and fins arranged from the upper left to the lower right. Cartilage is stained blue, and mineralized cartilage is stained pink. Credit: Andrew Gillis, MBL
Photo 2: A neurula-stage skate (Leucoraja erinacea) embryo, with embryonic germ layers false-colored. Tissues shown here include the neural tube (cyan), endoderm (yellow), paraxial mesoderm (magenta), lateral plate mesoderm (green) and tailbud (blue). Credit: Andrew Gillis, MBL
Photo 3: A cyanograph image of a skeletal preparation of a little skate (Leucoraja erinacea) hatchling. Credit: Michael Wen, University of Chicago/MBL
4 Video: A little skate (Leucoraja erinacea) hatchling. Skates are oviparous (egg-laying) cartilaginous fishes, and hatchlings emerge from the egg resembling small adults. Credit: Andrew Gillis, MBL
5 Graphic: Shared competence of neural crest- and lateral plate mesoderm-derived cells forms the basis of gill arch-paired fin serial homology. (A) The neural crest (blue) and lateral-plate mesoderm (red) contribute to a field of mesenchyme (purple) at the head trunk boundary with a common skeletogenic competence in the ~S19/20 skate embryo. (B) Anatomical structures that derive from this field of mesenchyme—the skeleton of the jaw, gill arches, and paired fins (in purple)—are serially homologous because of the reciprocal equivalence and shared response of their generative mesenchyme to local external stimuli. (From Wen, Prince and Gillis, PNAS, 2026)
