Little penguin
Eudyptula minor
See how 18 penguin species differ in beaks, wings, legs, and tails.
Try the project ↓Help
Penguins share the same basic body plan, but their beaks, wings, legs, and tails do not have the same proportions. Museum and field measurements let us compare those shapes without judging them from photographs.
Compare the little penguin with the emperor penguin to see the full wing-length range in this small set of measured birds.
Select a beak, wing, leg, or tail measurement. The cards and species ranking update together.
Choose any point to see which species has the most similar eight-measurement profile.
Finish with the coverage table. Each species is represented by only four to ten measured birds, and some measurements are missing.
See how 18 penguin species differ in beaks, wings, legs, and tails.
Researchers measured 93 penguins from museum collections and field projects. Start with the smallest and largest wing averages in this extract, then look beyond size.
Choose two species, then select a measurement. Each value is the mean among the recorded birds with that measurement.
Eudyptula minor
Aptenodytes forsteri
How to read this: The recorded little penguin average is 112.2 mm shorter than the emperor penguin average for wing length. The silhouettes use a few measurements for visual emphasis; they are not anatomical drawings.
The map compresses all eight specimen averages into two directions. Nearby points have more similar standardized profiles. Select any species to carry it through the page.
How to read this: Left-to-right mainly follows overall measured size because every measurement contributes positively. Up-and-down separates wider, deeper beaks and longer tails from longer beaks and lower legs. Distance summarizes the chosen measurements; it does not make one species a miniature version of another.
Choose one measurement to rank every species. This separates a real value in millimetres from the compressed pattern above.
Wing bend to the tip of the longest primary feather. Values are shown in millimetres.
How to read this: Longer bars have larger species averages for wing length. A highlighted outline marks the two species in the comparison above. Select any row to make it the first species.
Each cell shows how many recorded birds supplied that measurement. Darker cells are fuller, but even the fullest species has only ten birds in this extract.
| Species | Birds | beak | nostril | width | depth | leg | wing | wingtip | tail |
|---|---|---|---|---|---|---|---|---|---|
| 5 | 5 | 4 | 5 | 5 | 5 | 5 | 4 | 5 | |
| 5 | 5 | 4 | 5 | 5 | 5 | 4 | 4 | 5 | |
| 8 | 8 | 4 | 8 | 8 | 4 | 4 | 4 | 8 | |
| 5 | 5 | 5 | 5 | 5 | 4 | 5 | 4 | 5 | |
| 4 | 4 | 4 | 4 | 3 | 4 | 4 | 4 | 3 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 1 | 4 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
| 5 | 5 | 4 | 5 | 5 | 4 | 4 | 4 | 5 | |
| 5 | 5 | 4 | 5 | 5 | 4 | 4 | 4 | 4 | |
| 7 | 7 | 4 | 7 | 7 | 6 | 4 | 4 | 6 | |
| 5 | 5 | 4 | 5 | 5 | 5 | 4 | 4 | 5 | |
| 5 | 5 | 4 | 5 | 5 | 4 | 5 | 4 | 5 | |
| 5 | 5 | 4 | 5 | 5 | 5 | 4 | 4 | 5 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
| 10 | 10 | 9 | 10 | 10 | 10 | 8 | 8 | 10 | |
| 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 3 |
How to read this: Little penguin has 7 recorded birds. Its cells range from 4 to 7 usable measurements. A low count makes its average more sensitive to the few birds that happened to be measured.
Researchers measured preserved museum skins and living birds. This extract includes 93 birds from 18 species, with four to ten birds per species. It leaves out collection place, collection date, age, and body mass.
The shape map starts with each species' mean for eight measurements. It log-transforms and standardizes those means, then uses principal component analysis to fit them into two directions. Standardizing gives a millimetre-scale beak measurement the same influence as a much longer wing measurement.
The averages describe this extract, not every living member of a species. Sex, age, geography, preservation, measurement practice, and the birds available for study can all affect the pattern. The map cannot show ancestry or prove that one body feature caused another.
Why this exists
Which penguin species have similar measured body proportions, which traits separate them, and how much of the apparent pattern rests on sparse specimen measurements?
Compare two penguin species, select one of eight body measurements, and follow that choice through the specimen cards, shape map, species ranking, and measurement-coverage table.
Method and limits
I discovered the Many penguins extract in TidyTuesday's July 14, 2026 catalog and traced it to AVONET, a peer-reviewed open dataset assembled by Joseph A. Tobias and collaborators. Figshare identifies version 7 as a dataset under Creative Commons Attribution 4.0. I retrieved TidyTuesday's 9,140-byte CSV at commit 7489888351e5cc61e04ae6ad300884a4c7f37166 on October 4, 2026. Its SHA-256 is 96def70a189d831a6dae515ce250b3f5df641cb74c19f12915d1cf8a704112a3. The file contains 93 individual birds from 18 species and six genera.
The reproducible Python reducer verifies the source hash, row count, and complete species set. It retains eight direct measurements in millimetres: two beak lengths, beak width, beak depth, lower-leg length, wing length, wingtip extension, and tail length. It calculates a mean, observed range, and non-missing count for every species and measurement. Every species has at least one value for all eight measurements, but 63 of 744 possible cells across individual birds are blank.
For the shape map, the reducer takes the natural logarithm of each species mean, standardizes each measurement across the 18 species, and performs principal component analysis on the resulting correlation matrix. The first two components explain 48.4% and 23.9% of the standardized variation. The horizontal component loads positively on all eight measurements and therefore reads mainly as overall measured size. The vertical component contrasts wider, deeper beaks and longer tails with longer beaks and lower legs. Nearest-shape matches use Euclidean distance across all eight standardized species means; ties favor the pair with more shared measurements. Repeated reduction produces the same 10,612-byte static browser file at SHA-256 8140099c1a4b46d58b96098bef6c66f485f51fdc546a72bf6298b8c8355ea6f2. The page imports that file and makes no runtime request.
AVONET measurements come from preserved museum skins and living birds measured in field projects. This extract contains only four to ten individuals per species, does not identify collection place or date, and has no body-mass measurement. Species averages may reflect sex, age, geography, preservation, measurement practice, or which individuals were available. The shape map is descriptive, not a family tree, a population estimate, or proof that one body feature caused another. Scientific names come from the source extract; the English common names are interface labels paired with those 18 scientific names.