🎓 Doing Animal Diversity

🖼️ The Specimens

Every image on this page, sorted by phylum and then by group, with every group the page covers listed — including those with no image yet, which show a count of zero. Click an image to enlarge it. An AI-enhanced label means a photograph was cleaned up with Google Gemini; the enlarged view has a button to show the unmodified original. An AI-generated label means the image is an illustration, not a photograph of any specimen.

✏️ Stations

Each station holds specimens from one phylum. Look at the photographs (click to enlarge), then fill in all seven fields and check. Wrong answers are marked, and some come with a hint; the explanation appears once the station is complete. After two checks you may choose to see the answers, but the summary you download records that you did.

Stations 1 through 9 cover the groups found in most introductory courses. Stations 10 through 12 add three that are less often seen: a rotifer group, acorn worms, and tunicates.

❓ Twenty Questions

One of thirteen groups has been picked at random: eleven phyla, with the chordates split into lancelets, tunicates, and vertebrates. Ask yes-or-no questions about its body plan, then guess. The score is the number of questions you needed.

The trick is that the characters are nested. A question whose answer you could already have worked out from earlier answers tells you nothing, and the game will point those out at the end. Asked in a good order, these questions can find any of the thirteen in four.

The answers follow the current tree: flatworms, rotifers, and roundworms count as protostomes. Spiny-headed worms are included with the rotifers.

🧩 Grade or Clade?

Each set of photographs is a group people commonly talk about. Decide what kind of group it is.

The three are drawn side by side on the Terms tab, with a fuller discussion of grades. The test that separates the last two: did the group’s most recent common ancestor have the feature that defines the group? If yes, and some descendants are left out, the group is paraphyletic. If no, the feature arose more than once and the group is polyphyletic. Paraphyletic and polyphyletic groups are often called grades when they are defined by a level of organization, and grades are perfectly good descriptions of bodies. They are just not branches of the tree.

🌳 Characters on Two Trees

Pick a character and a tree. The animals that have the character are marked. Predict whether they form a monophyletic, paraphyletic, or polyphyletic group, and the tree then shows the fewest gains and losses that explain the pattern: purple branches carry the character, a filled box marks a gain, and an open box a loss. The red ring marks the group’s most recent common ancestor.

How the tree decides

The reconstruction uses parsimony: of all the ways the character could have been gained and lost along the tree, it picks one that needs the fewest changes. The kind of group then follows from the test on the Grade or Clade tab. If the fewest-change explanation gives the common ancestor the character, the group is paraphyletic; if it does not, polyphyletic. When both cost the same, the tree alone cannot decide, and the page says so.

Parsimony is a simple rule, not a law of nature, and the trees here are coarse, with one tip standing for a whole phylum. Where the two trees give different answers, the difference comes from the trees, not from the animals: the body-plan tree was built largely from these same characters, so they tend to look tidy on it. That circularity is the main reason the DNA evidence matters.

† extinct. The body-plan tree follows a common mid-twentieth-century arrangement, including the Articulata (annelids with arthropods) and lancelets as the closest relatives of vertebrates. The current tree leaves out comb jellies, placozoans, and several small phyla, whose positions are less settled or which have no specimens here.

🧬 Our Lineage

Most of the characters in this laboratory are innovations that arose along the line leading to us. Reading the tree from our own point of view turns them into a sequence.

Part 1. Put the innovations in the order they arose

Earliest at the top. Use the arrows to move them, then check.

The coelom is left out of this sequence on purpose: on the current tree, where along our line it arose is not settled.

🦴 Vertebrates

Classify each vertebrate specimen by domain, kingdom, phylum, subphylum, class, order, and, where it can be determined, family. Where order or family cannot be told from the photograph, the field is not scored.

Classes here follow the traditional ranks (Actinopterygii, Amphibia, Reptilia, Aves, Mammalia). On the current tree, birds are reptiles, which the ranks hide. Order names follow common usage; Caudata is also called Urodela.

🐾 Telling the Mammal Orders Apart

Mammals are the animals most of us know best, and the order is where the familiar ones are easiest to confuse. The table gives, for each group, the features to look for on a living animal or a skeleton, in the order of the tree below it. The Specimens tab shows how many of each are pictured here.

How the orders are related

Names link to their Wikipedia articles (↗). Where three or more branches leave one point, the order in which they split is not settled: at the base of the placentals (afrotherians, xenarthrans, and the rest), and among bats, carnivorans, odd-toed ungulates, and even-toed ungulates.

Two results of the tree run against appearances. First, a squirrel is a closer relative of yours than a cat or a cow is: rodents, rabbits, and primates form one branch of the placentals, and carnivorans and hoofed mammals another. Second, whales are even-toed ungulates. Their closest living relatives are hippopotamuses, and the even-toed ungulates without the whales are a paraphyletic grade. Two of the puzzles on the Grade or Clade tab, “Even-toed ungulates, whales left out” and “Marine mammals,” turn on these relationships, and a third, “Pachyderms,” on an older grouping that mixed the branches together.

🎥 Videos

Those marked core are the place to start; additional ones fill in background, and optional ones are for fun.

The animals themselves

▶

Simple Animals: Sponges, Jellies, & Octopuses — Crash Course Biology #22

CrashCourse · about 12 minutes · core

Sponges, cnidarians, flatworms, rotifers, and mollusks, the last of which are not all that simple. One statement in it is wrong: flatworms do not have a mouth and an anus at opposite ends of the body. Their gut has a single opening that serves as both.

watch on YouTube

▶

Complex Animals: Annelids & Arthropods — Crash Course Biology #23

CrashCourse · about 13 minutes · core

Annelids and arthropods, both segmented. The video treats segmentation as a trait the two share by descent, following the older Articulata view; the current tree places them on separate branches. Try the “Segmented animals” item on the Grade or Clade tab after watching.

watch on YouTube

▶

Chordates — Crash Course Biology #24

CrashCourse · about 12 minutes · core

An overview of the chordates, only a few of which are invertebrates: the lancelets and the tunicates.

watch on YouTube

For fun

▶

Exploring Invertebrates | Compilation

About 30 minutes · optional

More a series of ecological stories than a survey of anatomy or life history. Its aim is to convince you that invertebrates are ecologically important, or at least worth thinking about.

watch on YouTube

▶

10 Giant Invertebrates You Don’t Want to Meet

About 14 minutes · optional

A “biggest” list.

watch on YouTube

📖 Body-Plan Terms

Each term links to a Wikipedia article (↗) for more.

🧬 The Animal Phyla: Phylogeny and Body Plans

All 32 animal phyla recognized here. The tree is a phylogeny: it shows how the phyla are related by descent, as currently inferred mainly from DNA sequences, and not how similar their bodies are. (The Trees tab has an older tree built from body plans, for comparison.) The table gives each phylum’s body plan, using the features the Stations tab asks about, and lists the phyla in the same order as the tree, so that phyla near each other in the table are each other’s closer relatives. The Images column counts the photographs and illustrations on this page; click a count to see them. Phylum and clade names, in the table and in the tree, link to their Wikipedia articles (↗), which go into far more depth than this page can.

The point at the far left is the common ancestor of all animals, and every phylum descends from it. Where three or more branches leave a single point, as at that ancestor and within the molting animals, the order in which those lineages split from one another is not settled. The positions of comb jellies, acoels and their kin (Xenacoelomorpha), arrow worms (Chaetognatha), and two groups of greatly simplified parasites (Dicyemida and Orthonectida) are the most debated.

Counts of animal phyla run from about 30 to 35 depending on the authority, mostly because of a handful of judgment calls: whether spiny-headed worms (Acanthocephala) stand apart from the rotifers, whether peanut worms (Sipuncula) stand apart from the annelids, and whether priapulids, mud dragons, and loriciferans are three phyla or one (Scalidophora). This page folds the first two in and keeps the last three separate.

Two things stand out when the table is read against the tree. First, the body-plan features do not line up neatly with the branches: acoelomates and pseudocoelomates turn up among the protostomes, between phyla with a true coelom, and placozoans, without true tissues, are closer relatives of cnidarians and bilaterians than sponges are. Second, the “Protostome or deuterostome” column follows the tree rather than the embryo: arrow worms and some lamp shells develop in the deuterostome fashion yet are protostomes by descent. Both are the grade-versus-clade distinction again, seen across the whole animal kingdom.

🧩 Clades, Grades, and the Groups In Between

Any group of animals can be laid on a tree, and how it sits there decides what kind of group it is. The three panels below show the same tree of vertebrates three times, each with a different familiar group picked out.

branches leading to members of the group a branch left out where the group’s defining feature arose the group’s most recent common ancestor

The single-cut test

Imagine cutting one branch of a tree and lifting off everything above the cut. Whatever comes away is a clade: a common ancestor and all of its descendants. A paraphyletic group starts from such a cut but then has one or more smaller clades cut back out of it, as the birds are cut out of traditional “reptiles.” A polyphyletic group cannot be collected from a single cut at all: its members sit on separate branches, and their most recent common ancestor lacked the feature that unites them.

Grades

A grade is a group defined by a level of organization — a body plan or a way of life — rather than by descent: fish, traditional reptiles, invertebrates, acoelomates. Grades are usually paraphyletic, and for a simple reason. They are defined by features their members kept from a common ancestor (shared ancestral traits), while one branch of that ancestor’s descendants changed enough to be called something else. “Fish” are vertebrates that kept fins and gills; the four-limbed vertebrates are the branch that changed. “Invertebrates” are animals that kept the ancestral condition of having no backbone; the vertebrates are the branch that changed.

Polyphyletic groups arise in a different way, from features that evolved more than once. Wings, a worm-like shape, and radial symmetry as adults have each appeared on separate branches, and a group defined by any of them gathers animals whose likeness is convergence rather than kinship.

So the kind of feature used to define a group predicts what kind of group results. Shared derived features, inherited from the ancestor in which they first arose, mark clades. Shared ancestral features mark grades, which are usually paraphyletic. Features that arose more than once mark polyphyletic groups.

Why the difference matters

Grades remain useful words for describing animals, and this page uses several. But a grade says nothing reliable about who is related to whom, and reading one as a branch of the tree leads to mistakes. A crocodile looks more like a lizard than like a bird, yet crocodiles and birds are each other’s closest living relatives. A lungfish looks more like a goldfish than like you, yet a lungfish is more closely related to you than to a goldfish: lungfishes are lobe-finned fishes, the branch of fishes from which four-limbed vertebrates arose. In both cases the grade (“reptile,” “fish”) groups animals by what they kept, and the tree groups them by what they share through descent.

The Grade or Clade tab has nineteen groups to try, and the Trees tab lets you put a feature on a tree and see which kind of group it produces.

🦇 Homology and Convergence

Two animals can share a feature for two quite different reasons. They may have inherited it from a common ancestor that had it: the feature is then homologous in the two. Or each lineage may have arrived at it separately: the feature is then the product of convergent evolution, and the two versions are analogous (the general term for such independently evolved resemblance is homoplasy). The tree logic of the previous section applies directly: features that are homologous across a group mark a clade, while groups defined by convergent features are polyphyletic.

Homologous at one level, analogous at another

Whether two features are homologous depends on what they are compared as. A bat’s wing and a bird’s wing are homologous as forelimbs: both descend from the forelimbs of the first four-limbed vertebrates. They are analogous as wings: their common ancestor had forelimbs but did not fly, and each line turned its forelimbs into wings in its own way, the bat with skin stretched between long fingers, the bird with feathers on a shortened hand. An insect’s wing and a bird’s wing, by contrast, are analogous at every level: an insect’s wings are outgrowths of its body wall, not modified legs.

Telling them apart

Three kinds of evidence are used, and the third is decisive.

• Correspondence of parts. Homologous structures tend to share an underlying plan: the same parts, in the same order, with the same connections, however different their shapes or uses. Analogous structures typically share only their outward form or function.
• Development. Homologous structures usually arise from corresponding tissues in the embryo. The acorn worm’s gill slits and a fish’s develop in much the same way; a starfish’s radial symmetry develops in a bilateral larva, quite unlike a cnidarian’s.
• Position on the tree. If a feature is best explained by a single origin in a common ancestor of the animals that share it, it is homologous in them. If it requires separate origins, the resemblance is convergent. This is the same test the Trees tab applies.

Why convergence happens

Similar ways of life pose similar problems, and natural selection often arrives at similar solutions. Flying favors wings, burrowing a long and limbless body, and fast swimming a streamlined one — which is why a shark and a dolphin share a body shape that neither inherited from the other. On this page, the shells of a chiton, a crab, and a turtle; the radial symmetry of cnidarians and adult echinoderms; and the worm shape of a dozen unrelated phyla are all convergent. Whether the segmented bodies of annelids, arthropods, and vertebrates share any common origin is still debated.

The Grade or Clade tab includes several groups built on convergent features: “Worms,” “Animals with shells,” “Flying animals,” and “Animals with radial symmetry as adults.”

🧩 Putting It Together: Features, Groups, and Grades

The two previous sections are one idea seen from two sides. What kind of group a feature defines depends on what kind of feature it is: homologous or convergent, and, if homologous, newly arisen (derived) or carried over from a more distant ancestor (ancestral). The words clade and grade go back to the biologist Julian Huxley, who coined clade in 1957 and contrasted it with grade: groups defined by branching versus groups defined by level of organization.

Derived or ancestral depends on the level

The same homologous feature can be derived at one level and ancestral at another. A backbone is a shared derived trait of vertebrates, marking them as a clade. Among vertebrates, though, it is ancestral: every vertebrate has one, so it says nothing about which vertebrates are closer relatives. A group of vertebrates held together by a backbone plus fins and gills — “fish” — is held together by ancestral features, and is a grade. This is why grades are mostly built on homologies, and yet are still not clades: the homologies are real, but they are ancestral at the level of the group.

Can a grade be convergent?

Yes. A grade is defined by a level of organization, and nothing prevents two lineages from reaching the same level separately. Three examples:

• Radiata, as first drawn. The French anatomist Georges Cuvier grouped echinoderms with cnidarians and comb jellies (along with parasitic worms and other simple organisms) as radially organized animals. The radial symmetry of echinoderms arose separately, from bilateral ancestors, so the group was a convergent grade, and it has been abandoned.
• Pseudocoelomates. On the older body-plan tree, roundworms and rotifers form one branch, and “pseudocoelomate” looks like a clade. On the current tree they sit on separate branches of the protostomes, so a pseudocoelomate body most likely arose more than once: a convergent grade. The Trees tab shows both readings.
• Limbless vertebrates. Snakes, legless lizards, and the amphibians called caecilians all lost their legs separately. A grade defined by absence can be convergent too.

Some grades are a mixture. “Acoelomates” include flatworms and the acoels and their kin (Xenacoelomorpha). Whether each kept a body without a coelom from an ancestor or lost a coelom secondarily is still debated, so the grade may be partly ancestral and partly convergent. Similarly, whether the two groups of greatly simplified parasites once united as “Mesozoa” (dicyemids and orthonectids) share a common simplified ancestor or simplified separately is unsettled.

A caution about absences

Features defined by what is missing — no backbone, no coelom, no legs — need extra care. An absence can be inherited unchanged from an ancestor that never had the feature (as with the invertebrates’ lack of a backbone), or it can be a loss that happened separately in several lines (as with limbless vertebrates). The two look the same in the animal; only the tree tells them apart.

🌳 Tree Terms

🧮 Phage Biology and Phage Therapy Calculators