Intermediate species are organisms that display a mixture of traits from two distinct groups, revealing the evolutionary steps between an ancestral form and its descendants. Paleontologists and evolutionary biologists sometimes call them transitional forms, and the popular press tends to reach for the phrase “missing link,” though researchers generally avoid that term because it implies a single gap to be filled rather than a branching continuum. The concept is richer than most people realize: intermediates show up not only as fossils but also as living animals caught in the act of diverging, and even as laboratory-evolved microbes splitting into new lineages.
The Mosaic Principle
The most important thing to grasp about intermediate species is that they are not fifty-fifty blends of two groups. They are mosaics. An intermediate form typically has some features that look like its ancestors, some that look like its descendants, and some that are unique to itself. Tiktaalik, the famous fish-to-tetrapod fossil, had a pelvis greatly enlarged compared to other finned relatives, with deep hip sockets rimmed by robust bone, yet it still lacked a connection for a sacral rib and had no ischium, features present in later limbed vertebrates.1PubMed Central. Pelvic girdle and fin of Tiktaalik roseae That combination of old and new in one skeleton is the hallmark of a transitional form.
The evolution of the hominin lineage shows a similar pattern. Rather than all “human” traits appearing at once, different features changed at different times and rates, consistent with a mosaic pattern of change across the lineage.2PubMed Central. Mosaic evolution and the pattern of transitions in the hominin lineage Bipedal walking evolved millions of years before large brains, and tool use predates the anatomically modern human face. No single ancestor woke up one morning as a fully modern human. Each intermediate population had its own combination of ancestral and derived characteristics.
Tiktaalik and the Water-to-Land Transition
If you want a single fossil that captures what “intermediate” really means, Tiktaalik roseae is hard to beat. Discovered in 2004 in the Canadian Arctic, this roughly 375-million-year-old creature had fish-like scales and fin rays alongside a flattened, crocodile-like head and a mobile neck, something no fish before it possessed. Its feeding system tells the story in miniature: Tiktaalik could both generate suction like a fish and snap its jaws shut like a land-living predator. Sliding joints in its skull let the cheek and palate expand sideways, much like a modern alligator gar, while also supporting a biting mechanism. Researchers have described this as a “gar-like” stage in early tetrapod evolution, an important step that maintained suction feeding while simultaneously building the hardware for terrestrial prey capture.3PubMed Central. The feeding system of Tiktaalik roseae: an intermediate between suction feeding and biting
Its pelvic anatomy reinforces the point. Tiktaalik’s pelvis was much larger and more robust than that of other fish in its lineage, with paired bones and deep sockets that hint at weight-bearing capability. Yet it still had fins, not legs. The mosaic of primitive fins and derived pelvic features reveals that hind-limb-based propulsion has antecedents in the fins of the closest relatives of limbed vertebrates, not just in the first animals with actual legs.1PubMed Central. Pelvic girdle and fin of Tiktaalik roseae
Archaeopteryx and the Dinosaur-to-Bird Transition
Archaeopteryx, unearthed in a Bavarian limestone quarry in 1861, remains one of the most iconic transitional fossils. It had feathered wings and a wishbone like a modern bird, but also teeth, a bony tail, and clawed fingers like a small theropod dinosaur. For a long time, people assumed Archaeopteryx grew quickly like living birds, since fast growth and high metabolic rates are among birds’ defining traits. Bone tissue analysis tells a different story. Histological study of Archaeopteryx femora shows that the bone is composed almost entirely of parallel-fibered tissue with sparse vascular canals, a pattern indicating slow growth more similar to some other small dinosaurs than to modern birds.4PLoS ONE. Was Dinosaurian Physiology Inherited by Birds? Reconciling Slow Growth in Archaeopteryx So even in its physiology, Archaeopteryx was intermediate: it had the plumage to fly (or at least glide) but grew at a pace that would make any modern songbird seem like a sprinter by comparison.
This finding matters because it pushes back against the idea that an intermediate species is simply the “halfway point” on a straight path. Archaeopteryx had an unexpected combination of bird-like anatomy and dinosaur-like physiology, a package that no living animal replicates exactly. Transitional forms are not neat averages; they often combine features in ways that surprise researchers.
Walking Whales
The origin of whales is sometimes called the best-documented major evolutionary transition in the vertebrate fossil record, largely because paleontologists have found so many intermediate species bridging the gap between a four-legged land mammal and the fully aquatic giants we know today. The key fossils come from Eocene sediments on the Indian subcontinent and span several families: Pakicetidae (wolf-sized, land-dwelling, with ears already adapted for underwater hearing), Ambulocetidae (crocodile-like ambush predators that could swim and walk), Remingtonocetidae (long-snouted, probably semi-aquatic), Protocetidae (more aquatic, some retaining small hind limbs), and Basilosauridae (fully marine, with tiny vestigial hind legs).5PubMed. The origin and early evolution of whales: macroevolution documented on the Indian subcontinent
What makes the whale series so compelling is its density. Instead of a single “missing link,” there are at least five distinct families of intermediate whales, each with its own mosaic of terrestrial and marine features. The nostrils migrated from the tip of the snout to the top of the skull across multiple stages. The hind limbs shrank gradually. The tail changed shape from propulsion on land to propulsion in water. Each family represents a separate snapshot in this process, and together they form a sequence that leaves very little to the imagination about how a land mammal became a whale.
Hominins Between Apes and Humans
Human evolution is another transition rich with intermediate forms, and Australopithecus afarensis, the species that includes the famous “Lucy” skeleton, sits squarely in the mosaic zone. Its foot anatomy provides clues about the origins of upright bipedal walking, one of the earliest defining traits of the human lineage.6PubMed Central. Visualization of a Juvenile Australopithecus afarensis Specimen: Implications for Functional Foot Anatomy Analyses of the postcranial skeleton show clear evidence that these hominins had undergone selection for habitual bipedality, yet they retained a number of primitive, ape-like features in their upper bodies, including long arms and curved finger bones.7PubMed. Interpreting the posture and locomotion of Australopithecus afarensis: where do we stand? Whether those features meant they still climbed trees regularly or were simply evolutionary leftovers remains debated.
Even the spine of A. afarensis captures the mosaic theme. Synchrotron scanning of a 3.3-million-year-old juvenile skeleton from Dikika, Ethiopia, revealed 12 thoracic vertebrae (the same number as most modern humans, rather than 13 like most African apes) but with a thoracolumbar transition one segment higher than in either humans or apes. That distinctive pattern of spinal segmentation appears in all other early hominins examined so far, suggesting a shared body plan that was genuinely its own thing, neither fully ape-like nor fully human.8PubMed Central. Thoracic vertebral count and thoracolumbar transition in Australopithecus afarensis
Why “Missing Link” Gets It Wrong
The phrase “missing link” implies a chain with a single broken connection, as though evolution proceeds in a straight line from inferior to superior. That picture owes more to the pre-Darwinian concept of the Scala Naturae, or “ladder of life,” which ranked all organisms from simple to complex with humans at the top, than it does to actual evolutionary biology. The ladder of life was popular among naturalists until around 1850, but Darwin replaced it with a branching tree in which no living species is “higher” or “lower” than another.9PubMed Central. From the scala naturae to the symbiogenetic and dynamic tree of life
When someone asks “where is the missing link between fish and amphibians?” the implied assumption is that one fossil can close the case. In reality, every new discovery creates two new “gaps” on either side of it. Tiktaalik filled a gap, and then researchers naturally wanted to know what came just before and just after Tiktaalik. This is not a flaw in the evidence; it is how resolution improves. The whale series went from zero known intermediates in the 1970s to at least five distinct families by the 2000s, and each new find refined the picture rather than “completing” it.
Living Intermediates
Intermediate forms are not confined to the fossil record. Some living animals effectively model earlier evolutionary stages, and researchers study them to understand how structural change influenced locomotor function during major transitions. Mudskipper fish, for instance, haul themselves across mudflats using their pectoral fins, and biomechanical studies compare their ground reaction forces with those of tiger salamanders walking on land. Mudskipper fins bear lower vertical forces relative to body weight and push more sideways compared with salamander forelimbs, giving scientists a window into how the shift from fin-based to limb-based land locomotion worked.10Integrative and Comparative Biology. Propulsive Forces of Mudskipper Fins and Salamander Limbs during Terrestrial Locomotion: Implications for the Invasion of Land Some researchers argue that early tetrapods may have moved on land with a gait resembling that of a mudskipper before evolving the more efficient limb mechanics seen in salamanders.11PubMed. Kinematic comparisons between mudskipper fins and salamander limbs during terrestrial locomotion
Ring species offer another kind of living intermediate, this time in the speciation process rather than in anatomy. The Ensatina eschscholtzii salamander complex in western North America forms a ring of subspecies around California’s Central Valley. Populations along the ring interbreed with their neighbors, but where the two ends of the ring meet in southern California, they overlap without interbreeding, behaving like separate species.12PubMed. Geographic variation in allozymes in a “ring species,” the plethodontid salamander Ensatina eschscholtzii of western North America The complex presents a full array of conditions between well-marked species and geographically variable populations, with differentiated segments reflecting various depths of time of isolation and mixing.13PubMed. Incipient species formation in salamanders of the Ensatina complex Fine-scale genetic work on the hybrid zone where the terminal forms meet has confirmed asymmetric reproductive isolation, meaning the history of species formation can be traced geographically, back through the connecting forms.14PubMed Central. Asymmetric reproductive isolation between terminal forms of the salamander ring species Ensatina eschscholtzii revealed by fine-scale genetic analysis of a hybrid zone In a sense, the intermediate subspecies along the ring are living transitional forms between two incipient species.
Speciation in the Lab
You do not even need millions of years to watch intermediate stages of speciation unfold. Experimental evolution with lambda bacteriophage, a virus that infects E. coli, has demonstrated the process in real time. When researchers provided two different host cell types in the same environment, the phage population quickly split into two lineages with different host preferences. Genetic exchange was limited to within host cells, so the ecological preference of each lineage became tightly linked to which lineage it could swap genes with. Some phage lineages even evolved genetic incompatibility with each other after diverging, adding a genetic barrier to the ecological one.15Evolutionary Journal of the Linnean Society. A microbial perspective on speciation Before incompatibility was complete, those partially diverged lineages occupied exactly the intermediate zone between one species and two.
The phage experiment is useful because it collapses timescales. In a few hundred viral generations, what normally takes thousands or millions of years in animals played out in a flask. It also refutes the old assumption that microbes do not really speciate because they swap genes too freely. The ecological and genetic mechanisms that drive apart animal populations, habitat preference, assortative mating, and eventual reproductive isolation, have clear analogues even in organisms with no sex life in the traditional sense.
Stem Groups and Crown Groups
Evolutionary biologists organize the concept of intermediate species using two categories that are worth knowing if you read anything about paleontology. The “crown group” of any lineage is the set of all living members plus their last common ancestor and everything descended from it. The “stem group” includes all the extinct relatives that branch off before the crown group’s common ancestor. Most of the famous intermediate species are stem-group members: they sit on branches that lead toward, but fall short of, the modern crown group.
Modeling work on how stem and crown groups interact over time shows a recurring pattern: the stem group diversifies rapidly until the crown group emerges, at which point stem diversity collapses and the stem lineages soon go extinct.16PubMed Central. The dynamics of stem and crown groups This helps explain why so many intermediate species are extinct. They are not failures; they are the experimental phase of a lineage, exploring different combinations of features until one combination succeeds well enough to found the modern group. Tiktaalik, Archaeopteryx, and the walking whales are all stem-group members of their respective lineages.
A newly described Cambrian fossil, Oreinorema, illustrates the stem-group concept beautifully. Phylogenetic analyses place it as a transitional form between the Radiodonta (anomalocaridids, those bizarre Cambrian predators) and more crownward stem-group euarthropods, clarifying how the head and body plan of arthropods like insects and crabs were assembled step by step.17PubMed. A transitional, early Cambrian species bridges radiodonts and upper stem-group euarthropods Even half a billion years ago, evolution was working by mosaic assembly.
The Fossil Record’s Limits
If intermediate species exist throughout evolution, why are they not everywhere in the fossil record? Part of the answer is simply that fossilization is rare. An animal needs to die in the right kind of sediment, avoid scavenging and decay, and then have those sediments survive millions of years of geological upheaval. Taphonomic research has focused on quantifying the probability of preservation across different groups of organisms, the time resolution of fossil deposits, and how preservation changes across geological history.18Annual Review of Ecology and Systematics. The Quality of the Fossil Record: Implications for Evolutionary Analyses Soft-bodied organisms, small populations, and forest-dwelling species are drastically underrepresented.
Even the way we name species creates an illusion of sharp boundaries. Paleontological species concepts like chronospecies, in which a lineage is chopped into named segments at somewhat arbitrary points in time, have drawn criticism for imposing discrete labels on what was a continuous process. Some taxonomists advocate abandoning such concepts in favor of recognizing species based on unique combinations of characters, which better accommodates the mosaic nature of intermediate forms.19PubMed Central. Taxonomy and fossils: a critical appraisal In other words, the gap between “fish” and “amphibian” looks bigger than it really was, in part because we insist on drawing a line where nature did not.
Intermediate Forms in Plant Evolution
Animals tend to dominate popular discussions of transitional fossils, but the evolution of flowering plants has its own rich set of intermediates. One theory proposes that angiosperms evolved slowly from seed ferns during the Jurassic through three fundamental transitions: the carpel appeared first, then double fertilization, and finally the flower as we recognize it. These three steps may have taken over 100 million years to complete, with the modern combination of all three features not appearing until the Early Cretaceous.20TAXON. A transitional-combinational theory for the origin of angiosperms Just as whale evolution proceeded through multiple families with overlapping trait combinations, flowering-plant evolution involved a long series of partial assemblies before the full angiosperm package came together.
The plant record also highlights a challenge that is less obvious in vertebrate paleontology: many key transitions involve soft tissues, pollen, and microscopic structures that fossilize differently from bones. Recognizing intermediate plant species often requires combining microfossil evidence (like ancient pollen grains) with macrofossils (like leaf impressions or permineralized flowers), and the two lines of evidence sometimes tell conflicting stories about timing. The result is that debates about transitional stages in plant evolution can feel more contentious than equivalent debates about, say, the fish-to-tetrapod transition, even though the underlying principles are the same.
What Molecular Data Add
Fossils give you anatomy, but molecular biology can reconstruct what ancestral proteins and genes looked like. Ancestral sequence reconstruction takes the DNA or protein sequences of living species and infers what the sequences of their common ancestors were. This technique has proven largely robust to many of the modeling assumptions researchers worry about: even relatively simple statistical models produce nearly identical ancestral reconstructions as more complex ones, as long as among-site rate variation is accounted for.21Molecular Biology and Evolution. Robustness of Ancestral Sequence Reconstruction to Among-site and Among-lineage Evolutionary Heterogeneity
Reconstructed ancestral proteins can be synthesized in the lab and tested for function, effectively resurrecting molecular intermediates. Researchers have used this approach to trace how hormone receptors, visual pigments, and metabolic enzymes transitioned between different functional states. In some cases, the reconstructed ancestor works differently from both the ancient and the modern version, confirming that intermediate molecular forms existed and had their own distinct biochemistry. Where the fossil record gives you the external shape of a transitional organism, ancestral sequence reconstruction fills in what was happening inside its cells. Together, the two approaches make the concept of intermediate species far more three-dimensional than either could alone.