The Memory of Nature: Rats, Crystals & Learning

If morphic resonance is real, it makes a clear and testable prediction: a genuinely new pattern should be hard to establish the first time and steadily easier thereafter — everywhere at once, independent of distance or any physical connection, with the ease tracking only how often the pattern has formed before. Rupert Sheldrake has pursued that prediction through three kinds of evidence: animals learning, chemicals crystallising, and people remembering.

Rats that learned faster — everywhere

In the 1920s the Harvard psychologist William McDougall set rats to escape a water maze, swimming away from a brightly lit exit that gave a mild shock toward a dim one that was safe. Over many generations his rats made far fewer errors — early rats blundered into the lit exit dozens or hundreds of times before learning; later ones learned quickly. McDougall read this as Lamarckian inheritance: the parents' learning passed to their pups. But there was a wrinkle he noticed himself — the improvement appeared even in his untrained control line.

The most thorough test came from W. E. Agar's team in Melbourne, who ran the experiment for fifty generations across twenty years. They confirmed the trend — and confirmed the wrinkle decisively: the trained line sped up over the generations, and so did the never-trained controls, by just as much. To a Lamarckian that was a failure, since the trained line had no inherited advantage. To Sheldrake it is the signature of morphic resonance: as more rats anywhere learn the task, a strengthening behavioural field makes it easier for all later rats, related or not. He proposes the decisive next step — run the same task on rats hundreds of miles away, with no connection to the first, and see whether they too begin faster.

Crystals that "learn" to form

A newly synthesised compound has no pre-existing field for its crystal lattice, so — Sheldrake predicts — the first crystals may be very hard to grow, and then form ever more readily everywhere as the compound is crystallised again and again. The chemical folklore fits: substances like turanose and xylitol were known as stubborn liquids for decades until they first crystallised, after which they crystallised readily the world over.

He adds a sharper, falsifiable claim: the melting points of synthetic compounds should creep upward over the twentieth century as their fields strengthen, while compounds that have crystallised in nature for ages should hold steady. His own tables show just that — saccharin, phenolphthalein, aspirin and cortisone each rising several degrees across the century, while natural controls like salicin stayed flat.

He is candid about the difficulty, and this candour is characteristic. Chemists offered him a ready answer — purer samples melt higher, and samples grew purer over time — and he grants the point is reasonable, though he notes it is also somewhat circular. He flags an example that cuts against him (riboflavin, which crystallises in nature yet rose). And he concedes the historical record "can never be conclusive," proposing instead a clean prospective test: make several brand-new compounds, store all but one, crystallise that one many times in a single place, and watch whether its melting point — and only its — rises everywhere.

The phenomenon he points to is real and well documented. Drug manufacturers know the problem of "disappearing polymorphs": once a new, more stable crystal form of a compound appears somewhere, the old form can become hard to make anywhere — the antiviral ritonavir is the textbook case. Chemists explain it by invisible seed crystals drifting between laboratories; Sheldrake proposes a strengthening field, and notes that his version predicts the effect should persist even when stray seeds are filtered from the air. Which is right is an open question — and, he insists, a testable one.

Things we have already learned

If millions of people have already mastered something, a newcomer should — on this hypothesis — find it slightly easier. Sheldrake tested the idea with "old-field" experiments that pit genuine, established material against freshly invented look-alikes. With the Japanese poet Shuntarō Tanikawa he compared a real nursery rhyme known to generations of Japanese children against newly composed imitations; people remembered the genuine rhyme significantly better. The Yale psychologist Gary Schwartz found that people felt more confident rating real Hebrew words than scrambled ones, even when they knew no Hebrew at all; others reported the same with Persian and Arabic words, with real versus invented Morse code, and with Chinese characters. Sheldrake is the first to name the catch — perhaps real words and rhymes spread because they were intrinsically easier to learn — though he suggests that very memorability might itself be a product of morphic resonance.

He points, too, to the Flynn effect: the steady rise of raw IQ scores across the twentieth century, which James Flynn himself called baffling. As tens of millions sit the same tests, Sheldrake suggests, the tests may simply become easier to do — not because people are growing more intelligent, but through resonance from all who took them before.

Same data, opposite conclusions

In 1988 the neuroscientist Steven Rose challenged Sheldrake to test "this seemingly absurd hypothesis," and the two designed a joint experiment on day-old chicks learning to avoid a distinctively coloured bead. After a confounding factor was removed, the later batches did show a rising aversion — and the two men published opposite conclusions about the very same data. "In my view, the data were consistent with the operation of morphic resonance," Sheldrake wrote; "in Rose's view, they were not." It is as honest a snapshot as this field offers of why such questions stay open — and why Sheldrake keeps proposing experiments designed to settle them. The underlying idea is laid out in Morphic Resonance & Formative Causation, and its boldest reach in Laws of Nature as Habits.

Sources & talks

A recent discussion of the newer evidence:

New Evidence of Morphic Resonance — Rupert SheldrakeYouTube

Primary sources: Rupert Sheldrake, Morphic Resonance (2009), Appendix A, "New Tests for Morphic Resonance" (the melting-point tables and learning experiments) · his essay "Rat Learning and Morphic Resonance" · and The Presence of the Past (1988).