Young Readers

The first Star Stuff zine written to be read aloud, and the sourcing does not relax for it. Every spread carries two layers: large plain type for the child, and a boxed grown-up layer holding the species names, the figures, the papers and the places the evidence runs out — so a caregiver can answer is that really true? without leaving the page. The frog and the axolotl are not a metaphor we imposed. They are the pairing on Lilypad Library’s own logo — Kiersten Case’s low-stimulation puppet series for neurodivergent kids, which Stimpunks fiscally sponsors — and the biology underneath the pairing turned out to be better than it looked. The IUCN’s own Red List assessment describes Ambystoma mexicanum as “a paedomorphic species, living permanently in water, and does not undergo complete metamorphosis”: a description of an adult, not of a delay. The strongest spread is a population-genetics result. Everson and colleagues found the tiger salamander complex holds populations that metamorphose, populations that never do, “and those that do both” — with genetic clusters that “often contain a mixture of paedomorphic and metamorphic taxa” and gene flow between adjacent metamorphic and paedomorphic populations. Geography divides these salamanders; the route through childhood does not. Then the timing, and its price. Newman measured Couch’s spadefoot completing development in eight to sixteen days — the widely-quoted “eight days” is the floor of a range, not the fact. Denver, Mirhadi and Phillips found tadpoles accelerate as their pond dries in “a continuum of response,” not a switch, and decelerate again when the water is restored: the pace was never only a property of the animal. Both studies, one in the field and one in the lab, found accelerated development produced smaller metamorphs. Five refusals, aimed at the grown-up, the first being that “the axolotl never grows up” is the deficit reading wearing a nicer coat. A rhyme, not a proof.

The second read-aloud Star Stuff zine, and the sourcing does not relax for it. Two layers on every spread: large plain type for the child, and a boxed grown-up layer holding the mechanisms, the papers and the limits, so a caregiver can answer is that really true? without leaving the page. Four different kinds of waiting, and they are mechanisms rather than moods. Physiological dormancy broken by a long cold period — the seed is not counting days, it is waiting for evidence that winter has happened and is over. Serotiny, where cones are held shut by a resin bond that only fire's heat will break, releasing seed onto ground the fire has just cleared. Gut passage, which both moves a seed away from its parent and thins its coat. And physical dormancy, where a water-impermeable shell has to be worn through before a seed can even begin. Then the spread the zine is built to reach: look at what each seed waits for and then at where it lives. A seed that germinated before winter would freeze; one that germinated before the fire would burn; one that germinated under its parent would find no room. Every one is waiting for the moment when starting would actually work. A requirement is not a defect. And the longest wait anybody has directly dated: Sallon and colleagues germinated a date seed excavated at Masada and radiocarbon-dated to the first century Common Era — the authors' phrase is “the oldest, directly dated, viable seed,” and directly dated is the load-bearing part, ruling out claims that rest on the age of surrounding material. Its seedling grew normally over 26 months. Seven more were later germinated and sequenced. Five refusals, including that dormancy is a mechanism and not a virtue, that a fire-adapted seed is no argument that wildfire is good, and that two thousand years is a preservation accident rather than the point.

The third read-aloud zine, and it is built on a correction rather than a wonder. Many flowers carry ultraviolet patterns — the UV bull’s eye, which Lunau, Camargo and Ren describe as “a UV-absorbing centre of a UV-reflecting flower” acting as a guide toward a reward. Human eyes have no receptor tuned there. The pattern is physically present, in full daylight, on the front of the flower, and permanently unavailable to us. Then the spread that asks how anybody knows. In 1914 von Frisch trained bees to a coloured card with sugar water, then took the reward away and put that card out beside grey cards of every brightness he could make — so a bee choosing the brightest thing rather than the colour would have failed. The greys are the control, and a six-year-old can hold why they matter. And the correction the whole zine turns on: the popular version says bees see more colours than we do. The review says ultraviolet is “merely one component of colour vision” and that humans are more sensitive to red than bees are. The window is shifted, not widened — roughly 300–650 nm against our 390–750 — and writing it the popular way turns a difference into a ranking. Neither animal gets the whole ribbon; nobody does. The one step it takes toward people is deliberately the small one: a real signal can be present and undetectable to a particular observer, and the one who can detect it is not making it up. Five refusals, including that nobody here has a superpower, and that a UV photograph does not show anyone what a bee sees — the receptors, the brain and the photographer’s false colours are all in the way.

The first read-aloud translation on this site, and it sets the rule for them. A translation takes the next number in sequence — the numbers here record when, and this was made now, so No. 1b would make the number lie — carries a read-aloud of No. N marker in the same slot as companion to No. N, and states in its colophon that the original owns the argument. No. 1 makes the case that a nothing found is not a nothing there; this piece makes the two facts reachable by somebody who cannot yet read them, and does not compress the argument into shorter words. Two facts. Squeeze a bone and it produces a small voltage — Fukada and Yasuda measured it and published in 1957, potentials proportional to the load — and with the effect appearing only when collagen fibres are sheared past each other, with sign and size depending on the angle of the pressure. And the calcium a bone is built from was fused inside stars and scattered when they ended, long before the Sun existed. And a caveat the popular version drops, which this zine puts on the child’s page rather than burying. Fukada and Yasuda measured dried bone. Piezoelectric response falls as tissue is hydrated, and in a living wet body there is a second mechanism — streaming potentials from fluid squeezed through tiny channels — which some researchers think matters more for cell signalling. Recent work finds collagen’s piezoresponse still measurable at physiological humidity. It is not settled, and a spread is given to saying so: nobody has finished working it out, and that is not a problem with the science, that is the science. Writing it turned up two corrections in this house. The effect is Fukada and Yasuda’s; this site credited Yasuda alone for a long time, fixed it on two sibling pages, and never swept it back to the founding zine. And nothing here had ever mentioned the wet-bone question at all. Five refusals, the first being that “every step you take is a current telling your bones where to grow” is a step past the evidence.

Two measurements that disagree about what should be possible. Mäthger, Barbosa, Miner and Hanlon built checkerboards whose two squares differed only in wavelength, matched for brightness against the cuttlefish’s single visual pigment, so the board would look uniform to the animal. A cuttlefish that could see colour would break into its bold disruptive pattern; these ones stayed plain. Their conclusion, quoted on the page: cuttlefish must be color blind — and the same assay showed they resolve contrast differences of about 15%, so the null is not a story about an animal that cannot see. Then Chiao, Wickiser, Allen, Genter and Hanlon photographed live camouflaged cuttlefish with a camera that records every wavelength, modelled how the pictures would look to the fish that hunt them, and found the just-noticeable differences relatively small and distributed randomly, an indication of good color match. Both halves measured, and they should not both be true. Four explanations, and a 2026 review that closes each one. Reiter works through passive skin reflectance, the chromatic-aberration idea built on the animal’s W-shaped pupil, polarized light, and light-sensing skin, and finds each insufficient or weakly supported — noting that the pupil idea has had no physiological tests at all. Then he suggests something better and stranger: maybe there is no colour perception here, and the animal predicts colour from light and dark, the way you know a grey photograph of grass should be green. The one measured imperfection gets its own spread rather than a footnote: cuttlefish reflect a peak near 800 nanometres in the infrared that the sand does not, in a light nothing that hunts them can see. And the belonging claim is about the test, not the animal. Nobody found any of this out by asking a cuttlefish a question; they built a channel it could answer through. Five refusals, the first being that “colour blind and it manages anyway” is the superpower reading with a sympathetic face.

Every other explanation of a dog’s nose starts with how many smell-receptors it has. This one starts with where the air goes. Craven, Paterson and Settles measured seven dogs — a 6.8-kilogram Pomeranian, a beagle, a sheltie–husky, a border collie cross, a German shepherd and two Labradors, the largest 52.9 kilograms — sniffing in masks that recorded the airflow, then built a computational model from MRI scans of a real nasal cavity. Every one of the seven sniffed between four and seven times a second, whatever its size. And on the way out, the air does not go forward. “A ventral-laterally directed air jet is ejected from the nose” — two sideways puffs through the slits at the edges of the nostrils — which “promotes disturbance and mixing of ambient odorants that may be subsequently inspired, while minimizing sample ‘blow-off’ directly ahead of the nostril.” The dog is stirring the scent up instead of scattering it, and the child finds this out with a feather on a table before it is ever explained: blow at it and it goes, blow past it and it stays. Two more things the nose does that yours does not. Each nostril draws from a patch of air the other cannot reach, because “the reach of a nostril is smaller than the internostril separation” — two samples at one moment, which the authors propose is used to work out direction and did not demonstrate, and the zine says so on a spread of its own. And inside, a shelf of bone splits the flow: only about 12–13 per cent of a sniff turns off into the olfactory recess, a room at the back that is, in a 1958 phrase the paper quotes, “seen at its height of perfection in the dog” — and largely absent in primates. Then the comparison, which runs the way nobody expects. Human sniffing yields “dramatically different intranasal airflow patterns in the right and left nasal cavities of the same subject”, which the paper calls “undesirable for consistent odorant deposition.” Your nose is not a broken dog nose. It was never trying to be one. It closes on something unfinished: three dogs called Kevin, Delfi and Charlie picked a warm patch from a cool one across a room, using a nose-tip too cold to be a heater, and their scientists wrote that the limits of it “remain to be elucidated.” Twelve spreads, three pictures, five refusals for the adult, and no claim anywhere about children.

The measurement the zine is built on is a measurement of emptiness. Cummins, Seale, Macente, Certini, Mastropaolo, Viola and Nakayama put dandelion seeds through X-ray micro-tomography and a vertical wind tunnel, and counted the plume: 100 filaments [95, 106], each 7.41 millimetres long and 16.3 micrometres thick. Its porosity — “the ratio of the empty projected area to the plan area of the enclosing disk” — came out at 0.916. Nine parts in ten of a dandelion’s flight is nothing at all. And the nothing is the mechanism rather than a saving. Because the plume is full of gaps, the air does not have to escape round the edges; it passes straight through, and that detachment leaves “an extraordinary type of vortex” sitting in the air above the seed and never touching it — “a ring of recirculating fluid, which is detached owing to the flow passing through the pappus.” They proved it was the holes by building plastic discs across a porosity range, and finding that a disc matched to the dandelion behaved like one. Then the number that makes the demonstration work. A solid disc starts fluttering above a critical Reynolds number of 149; a dandelion seed holds steady to 429, and every seed they tested flew below its own. So a paper circle the width of your thumbnail wobbles all the way down and a dandelion seed does not, which is the finding at kitchen scale and the two spreads a child does with scissors and one seed. The zine also prints the correction against itself. A solid disc supplying the same drag at the same speed would be 38 per cent smaller across — per unit of width, solid is the better brake. What the holes buy is four times the load for the same material, and freedom from fluttering. The first plan for this zine had the demonstration as the solid one falls faster, which is false, and the colophon says so rather than quietly fixing it. Twelve spreads, four pictures, five refusals for the adult — including a refusal of the emptiness metaphor, which is the one this piece most needed.

The demonstration is the experiment, at kitchen scale, and it needs nothing at all. Shut your eyes, have somebody clap, point — you are close. Cover one ear with a hand and do it again, and you point wide. Knudsen and Konishi ran that on the owl: “Occluding the right ear caused the owl to orient below and to the left of the sound source; occluding the left ear caused it to orient above and to the right.” Read it twice, because the owl’s version has an extra half — you lose left and right, and it loses left-and-right and up-and-down at once. Which is the whole subject. Human ears are level, so a sound directly above and a sound directly below reach both of them identically; there is no difference to compare. The barn owl’s are not level: “the left ear opening and flap are located high in the left trough, whereas the right ear opening and flap are centered in the right trough” — so up and down produces a difference too. And it is the direction an owl needs most, for a reason Norberg argues geometrically: prey direction “usually forms a shallow angle with the ground”, so the same angular error costs far more distance vertically than horizontally. The strongest spread is a dissociation. With the facial ruff feathers moved aside, the owl “continued to localize sounds accurately in azimuth, but failed to localize sounds in elevation” — one ability destroyed, the other untouched, which is much better evidence than everything getting slightly worse. Then the finding the zine is really for. This did not happen once: ear asymmetry “evolved independently in at least five lines”, and the anatomy differs every time — soft tissue over a symmetrical skull in the barn owl, actual skull bone in others, a bigger opening on one side in others again. Some owls have none at all, and the little owl grows the asymmetry as an embryo and evens back up before it hatches. There is no standard owl ear. How many times it happened is itself given as four, five and seven by three different papers, and the zine prints the disagreement rather than the tidiest number. Thirteen spreads, four pictures, five refusals for the adult — including a refusal of the difference metaphor, which is the one this piece most needed.

The method is the reason the finding exists. Mattila and colleagues marked individual leaves on rowan, Norway maple, silver birch and bird cherry and measured chlorophyll in each of them every morning and every evening for a whole autumn, with an instrument that reads a leaf without damaging it — so the same leaf could be measured hundreds of times. A method that destroys its sample can only ever compare different leaves to each other. What they found, verbatim: “the chlorophyll content of each individual leaf remained constant until a phase of rapid degradation commenced. The fast phase lasted only ~1 week and ended with abscission.” And the leaves are not synchronised. Chlorophyll degradation “started at different times in different B. pendula leaves, and the leaves belonging to same branches did not behave similarly” — so a tree turning gradually over weeks is hundreds of fast changes, staggered, and the smooth fade everybody pictures is an artefact of averaging them together. Then two colours that arrive by completely different routes. The yellow carotenoids are in the leaf all summer, doing photosynthetic work. The red anthocyanins are not there at all until the tree builds them, during senescence — Keskitalo and colleagues caught the switch in aspen, where stressed leaves accumulated anthocyanins after the programme began “but at the beginning of September the leaves did not.” The zine also prints a correction against its own tidiest sentence. Everybody says the green leaves and reveals the yellow that was hiding. The one source that separated the individual carotenoids found “lutein and beta-carotene were degraded in parallel with chlorophyll”, with only some xanthophylls retained longer — so the unmasking story is part of the answer and not all of it, and even the paper stating the standard version hedges it with partly. Why the red gets made is being argued in print right now, with a 2021 test and a 2022 reply. And before it lets go the tree recovers about 80 per cent of the leaf’s nitrogen and phosphorus. Thirteen spreads, four pictures, five refusals for the adult — the first being a refusal of the masking metaphor, which is the one this piece most needed.

The science borrowed the fairy tale’s name and then said nearly the opposite of it. Kopparapu and colleagues put the conservative limits for our own system at 0.99 and 1.70 astronomical units, and draw the consequence in their own abstract: this suggests “the present Earth lies near the inner edge”. So Earth sits about 1.4 per cent of the way across a band 0.71 wide — not the just-right bowl, but the one that only just fits. And we did not arrive there by moving. The young Sun was roughly a third dimmer than it is now, and a dimmer star holds a smaller, closer band; scaling those limits by the square root of luminosity puts Earth about 13 per cent across the band two billion years ago and 23 per cent four billion years ago. The planet stayed exactly where it was. The edge came to it. Then the zine goes somewhere else to show the band is not a fact about distance at all. TRAPPIST-1e completes a year in 6.1 days at 0.02925 astronomical units, because its star is small and cool and keeps its warm band tucked in close — more than thirty times nearer in than ours, and nothing has gone wrong there. Every star gets its own band, computed from its own output, and astronomers have never once called a dim star defective for having a close one. Which is the turn the whole piece is built to reach. A band of conditions in which a living thing can persist is its niche, in Hutchinson’s 1957 sense; a tulip bulb needs twelve to sixteen weeks of proper cold or the flower never leaves the bulb, and that is not fussiness but a location. There is no standard band, so “just right” is never just right on its own — it is always just right for somebody, and you have to say who. The answer is therefore not one room set more kindly, but a place with a cave and a campfire and a watering hole in it, which is a Cavendish Space and the first time this collection has named one. Spread eight carries the correction that made the zine: proximity to the inner edge does not mean carbon dioxide can push us over it — Goldblatt and Watson are explicit that it cannot — but the headroom does shrink as the Sun brightens, and the band a human world needs is nested inside the planet’s and far thinner. Thirteen spreads, four figures, six refusals for the adult.

One paper carries most of this, and it is carrying two findings rather than the famous one. The first is the constraint everybody knows: “These pairs are: adenine (purine) with thymine (pyrimidine), and guanine (purine) with cytosine (pyrimidine)” — fixed, and fixed by size, since one of any pair must be the larger purine and the other the smaller pyrimidine. The second sits a few lines later in the same 1953 paper and is the one this zine is built on: “The sequence of bases on a single chain does not appear to be restricted in any way.” Every constraint in the system runs across the two strands and none runs along one. That is not a gap somebody will close later; it is the property that lets DNA carry information at all, since a sequence permitted only one arrangement could carry none. The demonstration has a child perform the paper’s closing inference. Write a row of letters, write its partner row beneath, cover the first row, put it back — and it comes back exactly, because it never could have come back otherwise. That is what Watson and Crick meant by the sentence they ended on: the specific pairing “immediately suggests a possible copying mechanism for the genetic material.” Then the zine goes to a pond rather than a laboratory, and that is a deliberate refusal. This piece began from a September 2026 result in which an ordinary bacterial enzyme transcribed a synthetic eight-letter alphabet as readily as the natural four — and that finding is nowhere in it, because our own weekly scan had graded it and said the argument needs the biology to stay in the test tube. The wild exception does the same work and never leaves the water. Cyanophage S-2L substitutes 2-aminoadenine for adenine throughout its genome, reported in Nature in 1977 and not explained until 2021 — and a virus is not a cellular organism, so the universality claim survives, which the grown-up layer says rather than letting the child layer overclaim. Twelve spreads, four figures, five refusals for the adult, including a refusal of the genetic-verdict frame and a refusal of worth built on rarity.

The companion No. 105 promised on its own face. That zine said in its grown-up layer that the genetic code “is not in this zine at all”; this is the promise kept. The letters are read three at a time, each three-letter word — a codon — naming one amino acid, one bead on a protein string. Then the arithmetic does the arguing. Four letters read in threes give sixty-four possible words, and there are only twenty beads to name plus a stop signal, so sixty-one naming codons distribute as five fourfold sets, nine twofold, three sixfold, one threefold and two singletons — which sums to exactly twenty, every amino acid once. Nearly everything therefore has more than one correct spelling: glycine four, leucine and serine and arginine six each, and only methionine and tryptophan one. The demonstration is the mechanism, on paper. GGU is glycine; change the last letter to any of the four and it is still glycine; change the first and it is serine. Every codon on the page was checked against the NCBI standard translation table rather than written from memory, because the game fails if one row is wrong. And the exception is inside the reader this time. NCBI lists thirty-three translation tables, and a human body runs two at once — mitochondria read UGA as tryptophan where the rest of you reads it as stop, which is the sharpest evidence for what No. 7 argued about a bacterium that never quite stopped being itself. What the zine refuses is the reason spread ten is in plain type rather than in a box. Its companion stayed safe by never saying what DNA does; this one cannot, so it corrects the popular meaning of “genetic code” directly: the table names which bead comes next and contains no entry for a trait, an ability or a person. It also declines to call the code kindly arranged — the famous error-minimising result sits against Crick’s frozen accident, and how the code got its shape is an open argument. Twelve spreads, four figures, five refusals for the adult.

The third part of an arc, and the one that answers the question the other two leave open. No. 105 gave the four-letter alphabet and No. 106 gave the code that reads it three letters at a time; both stopped short of asking what the whole thing is. This one asks, and checks rather than asserts. A plan is a drawing of a finished thing made in advance, so a more complicated thing needs a bigger drawing — that is a prediction, and it fails twice. By quantity: an onion carries about 16,400 million bases against a human 3,200 million, roughly five times more. And the objection is handed to the child before it is answered. Counting letters is unfair, because at least 95 per cent of the onion genome is repetitive sequence — so the zine counts instructions instead, and that is worse for the metaphor rather than better. Before anybody had counted, human gene estimates ran to 100,000; the answer came in near 20,000, about the same as Caenorhabditis elegans, a worm a millimetre long with 959 body cells, whose entire genome is thirty-two times smaller. Then the strongest form: within the onion genus alone, genome size varies four and a half fold, so onions disagree with other onions by more than onions disagree with us. What replaces the metaphor is older than the metaphor. Schrödinger coined code-script in 1944 and wrote in the same paragraph that the term is “of course, too narrow,” because chromosomes are “architect’s plan and builder’s craft — in one.” The plan and the building are the same thing, and there is no finished you in there to be a drawing of. The history of how that qualifier got dropped is a chain and is deliberately not in here. Twelve spreads, four figures, five refusals for the adult; the zine stops at onions and worms and never argues about people.

The first piece here about who gets named, and it is built almost entirely out of the discoverer’s own account. Jocelyn Bell spent two years helping build a radio telescope across four and a half acres — “an area that would accomodate 57 tennis courts” — and then read what it drew: “between them they produced 96 feet of chart paper every day. The charts were analyzed by hand—by me.” The group had chosen not to computerise the analysis, because “a human can recognize signals of different character whereas it is difficult to program a computer to do so” — and the discovery is downstream of that sentence. Then three months of failure, which the zine keeps, because a child told only about the moment of discovery learns that finding nothing means you were wrong to look.

And it refuses the easy telling, because she refused it herself. Her name is second of five on the 1968 Nature paper — the widely circulated version in which she was left off the paper is false, and the zine says so. The 1974 Nobel went to her supervisor without her, and in 1977 she gave four reasons she should not have shared it, ending “I am not myself upset about it.” So the zine hands the child the harder and truer distinction: being kind about something is not the same as the something being fair, and a reader may reach their own verdict. What she does record as absurd is the press, who on finding a woman was involved “descended even faster”, photographed her examining bogus records, and asked how many boyfriends she had. Then 2018, and the £2.3 million she gave away entire — to students including disabled students — saying “I don’t need the money myself.” It ends on the one thing a four-year-old can do about any of it, which takes four words.