Lesson 12: Classification (HL: Cladistics)

This lesson covers Theme A — classification, and at HL cladistics (SL and HL): how biologists organise biodiversity and reconstruct evolutionary relationships. The HL cladistics content — clades, cladograms and molecular evidence — is heavily data-based in exams.

The binomial system and taxonomic hierarchy

Every species has a binomial name (Linnaeus): genus (capitalised) + specific epithet (lowercase), both italicised — *Homo sapiens*, *Panthera tigris*. Binomials are universal and unambiguous, unlike common names ("robin" is a different bird on different continents). Species are grouped in a hierarchy of taxa: domain → kingdom → phylum → class → order → family → genus → species. At SL you must classify one plant and one animal from domain to species — standard examples: the wolf *Canis lupus* (Eukarya, Animalia, Chordata, Mammalia, Carnivora, Canidae, *Canis*, *lupus*) and maize or the oak for plants. The three domains (Bacteria, Archaea, Eukarya) rest on ribosomal RNA evidence (Woese) — Archaea, though prokaryotic, are in many molecular respects closer to eukaryotes than to bacteria. Natural classification groups organisms by evolutionary descent; artificial classification (e.g. "flying animals") groups by convenience/convergent features and misleads.

Dichotomous keys

Keys identify specimens through paired statements of contrasting features. Skill points examiners reward: use observable, unambiguous features; each couplet must cover all possibilities; work from the general to the specific. You may have to *use* a key on photographed specimens or *construct* one for a small set of organisms — in construction, features must be present/absent or clearly measurable, never subjective ("large", "pretty").

Clades and cladograms (HL)

A clade is a group consisting of a common ancestor and all its descendants. Cladograms (phylogenetic trees) are hypotheses of clade relationships built from shared derived characteristics — anatomical, biochemical and especially DNA/amino-acid sequence data. Nodes represent common ancestors (speciation events); the branching order, not the drawn lengths, carries the meaning in basic cladograms; rotating branches at a node does not change relationships (a common misreading). Worked logic: for four taxa with a molecular difference matrix — species A and B differ at 2 nucleotide positions, A–C at 9, B–C at 10, and D differs from all by 15–17 — A and B are the closest relatives and join first; C joins next; D is the outgroup, branching earliest. The principle of parsimony says prefer the tree requiring fewest evolutionary changes. Molecular clocks estimate divergence times from accumulated sequence differences, assuming an approximately constant mutation rate — used to date, for example, primate divergences; examiners probe the assumption (rates vary between lineages and genes, so clocks are calibrated against fossils).

Evidence and its revisions

Classification is repeatedly revised by new evidence: giant pandas were placed with raccoons on anatomy until DNA put them firmly in the bears; the three-domain system replaced "prokaryote" as a taxonomic level. Convergent evolution explains misleading similarities (wings of birds, bats and insects are analogous, not homologous) — a reason modern classification weights molecular data heavily. DNA barcoding (short standard gene regions, e.g. COI in animals) now identifies species from tissue fragments — used in food-fraud detection and conservation forensics.

Nature of Science / TOK link

Classification is knowledge organised by human-chosen criteria.TOK: are species "discovered" or "invented"? Biologists debate species concepts (biological, morphological, phylogenetic) because nature supplies populations that grade into one another — ring species and hybridising oaks resist tidy boxes. The decision to split or lump affects conservation law and funding: taxonomy is factual, but it is not value-free in its consequences.

Common mistakes (from examiner reports)

• Ranking taxa in the wrong order, or writing binomials without italics/genus capitalisation in constructed responses.
• Reading cladograms by the left-to-right order of tips rather than by shared nodes.
• Treating shared habitat or similar appearance as evidence of close relationship (analogy vs homology).
• Saying a molecular clock gives exact dates rather than estimates with assumptions.

IB Exam Experience

Paper 1 style. 1. Which is the correct sequence of taxa from largest to smallest? A. Kingdom → phylum → class → order → family → genus B. Phylum → kingdom → order → class → family → genus C. Class → order → phylum → family → kingdom → genus D. Kingdom → class → phylum → order → genus → family 2. In a cladogram, a node represents… A. an extinct species B. a common ancestor of the groups branching from it C. a mutation D. the most advanced species 3. Two species have very similar DNA sequences in a gene. The best conclusion is that they… A. are the same species B. share a recent common ancestor C. live in the same habitat D. evolved by convergence Answer key. 1 — A. 2 — B. 3 — B (similarity of sequence reflects recency of shared ancestry, not identity of species). Paper 2 style. The table shows amino-acid differences in cytochrome c between humans and four species: chimpanzee 0, horse 12, kangaroo 10, tuna 21. (a) Deduce which of horse and kangaroo is more closely related to humans, explaining your reasoning. [2] (b) Suggest why a protein such as cytochrome c is suitable for such comparisons. [2] (c) State one limitation of using a single protein to build a phylogeny. [1] *Marking notes.* (a) Kangaroo (10 differences < 12) [1]; fewer differences mean less time since common ancestry for changes to accumulate [1]. (b) Universal in aerobic organisms [1]; functionally constrained so changes accumulate slowly and measurably / length suitable for alignment [1]. (c) Single gene may not reflect whole-genome history (e.g. gene duplication, horizontal transfer, convergent substitutions) [1]. IA idea. Build a cladogram from published cytochrome c or haemoglobin sequences for 8–10 species using database alignments, and test how well it matches the accepted taxonomy — quantifying agreement and discussing discrepancies.

Extended drill: building and reading a cladogram from raw data

Suppose DNA sequences (20 bases of one gene) give these difference counts: human–chimpanzee 1, human–gorilla 3, chimpanzee–gorilla 3, human–orangutan 6, chimpanzee–orangutan 6, gorilla–orangutan 7. Build the tree by clustering the smallest distances first: human and chimpanzee join (1); gorilla joins that pair next (3 to either); orangutan is the outgroup of this set (6–7). The resulting cladogram asserts: humans and chimpanzees share a more recent common ancestor with each other than either does with gorillas. Two habits protect marks: the tree is a *hypothesis* — a different gene might (rarely) disagree because of incomplete lineage sorting, so strong phylogenies use many genes; and branch *lengths* in a phylogram carry the difference counts, while in a pure cladogram only the branching order matters — check which diagram you have been given. Molecular clock application: if the human–orangutan split is calibrated by fossils at ~14 million years for 6 differences, the rate is ~0.43 differences per million years, estimating the human–chimp split (1 difference) at ~2.3 million years by this gene — crude (single gene, small numbers, rate variation), and real estimates use genome-scale data to place it around 5–7 million years. Showing you understand why the crude estimate is unreliable earns more than a lucky number.

Dichotomous keys in practice — and their limits

A serviceable key for five British trees might run: 1. Leaves needle-like or scale-like → 2; leaves broad and flat → 3. 2. Needles in pairs → Scots pine; needles single, with cones hanging → Norway spruce. And so on. Construction rules that examiners apply: every couplet must be a true dichotomy covering all remaining options; features must be visible on the specimen provided (do not use "flowers pink" in winter material); avoid overlapping quantitative ranges; each step should reduce the candidates decisively. Keys are artificial classification tools — excellent for identification, silent about evolution — which is why the same syllabus pairs them with cladistics: one answers "what is it?", the other "what is it related to?". Confusing those two purposes in an essay is a category error the TOK strand of this lesson exists to prevent.

One more application: when classification changes the law

Taxonomy has teeth. When genetic evidence split the African elephant into savanna and forest species, conservation status and ivory-trade rules had to be reconsidered for each; when a "species" of endangered frog is reclassified as two, each half has a smaller range and a stronger claim to protection. DNA barcoding now audits the seafood trade (studies repeatedly find a substantial fraction of restaurant "snapper" or "cod" is something else) and identifies CITES-protected timber from wood shavings. The IB point is not cynicism but precision: a classification is a hypothesis with consequences, maintained by evidence and revised by it — which is why museums keep voucher specimens and why the barcode library is tied to identified specimens, not just sequences. In TOK terms, the category system is a human instrument pointed at a real, branching world; the instrument can be well or badly made.

Examiner's lens: final drills (classification)

• Mnemonic order check: Domain, Kingdom, Phylum, Class, Order, Family, Genus, Species — decreasing size, increasing relatedness. The wolf's full classification (Lesson 12) is your rehearsed example; learn one plant to the same depth.
• Cladogram reading rule: two species are close relatives if they share a *recent node*, regardless of how far apart their names sit at the tips or how the branches are rotated.
• Homology vs analogy in one contrast: bat wing and bird wing — homologous as forelimbs (same bones), analogous as wings (flight evolved separately). Examiners accept either framing if you state which character you are classifying by.
• Self-test: why is "fish" not a valid clade? (It excludes tetrapods, which descend from within the fish lineage — a group missing some descendants of its ancestor is paraphyletic, not a clade.)