James Croll and the Astronomical Theory of the Ice Ages

The Theory From the Library

Reading voraciously in physics and astronomy, Croll became fascinated by a problem that had stumped the geological establishment: the Swiss-born naturalist Louis Agassiz and others had by then firmly established that great ice sheets had once covered large parts of the northern hemisphere, but no one had a convincing explanation of why such ice ages came and went. Croll found his answer in the sky rather than in the rocks. Drawing on the precise orbital calculations of the French astronomer Urbain Le Verrier — whose mathematics, famously, had already led to the discovery of the planet Neptune — and building on Joseph Adhémar's 1842 book Révolutions de la mer, Croll set out his answer in an 1864 paper for the Philosophical Magazine, titled in full "On the Physical Cause of the Change of Climate during Geological Epochs." The paper was modestly signed "James Croll, Anderson's Institution." The scientific establishment, when it took notice, was astonished to discover that the author was a museum janitor.

Croll was not working in a vacuum. Adhémar had already proposed, in 1842, that the precession of the equinoxes could be a cause of glaciation, and the astronomer John Herschel had mused, around 1830, on the possible climatic effects of eccentricity. What Croll did that neither of them had done was combine the two orbital elements into a single, worked-out mechanism, calculate in detail how each affected the solar radiation received by a hemisphere through the year, predict that ice ages should recur multiple times rather than occur once, and — his single most original contribution — introduce a physical feedback mechanism that could turn a comparatively modest astronomical nudge into a full continental glaciation.

The Mechanism

Croll's theory rested on two orbital ingredients. The first was orbital eccentricity: Earth's orbit around the Sun slowly shifts, over a cycle of roughly 100,000 years, between being nearly circular and noticeably more elliptical, changing how much the Earth–Sun distance varies through the year. The second was the precession of the equinoxes — Earth's slow wobble on its rotational axis, completing a cycle on the order of 22,000 to 26,000 years — which determines whether a given hemisphere's winter falls when Earth happens to be closest to the Sun (perihelion) or farthest away (aphelion).

Croll's key mechanism combined the two: when eccentricity is high and a hemisphere's winter coincides with aphelion, that hemisphere receives markedly less solar radiation during its winter months, allowing snow and ice to accumulate rather than melt away. Croll spelled out just how dramatic he thought the effect could be, writing in 1864 that at latitudes of around 54–55 degrees the ground might, "at the very height of summer, be almost covered with frozen snow, in some places many fathoms deep."

Most presciently of all, Croll proposed a genuine positive feedback loop: once snow and ice begin to accumulate, they reflect substantially more sunlight back into space than bare ground or open ocean does — what climate scientists today call the ice-albedo feedback — cooling the local climate still further and reinforcing the glaciation that produced it in the first place. He layered further amplifying effects on top of this, arguing that an expanding ice sheet would alter prevailing trade winds and deflect warm ocean currents such as the Gulf Stream away from the affected hemisphere, starving it of additional heat. This coupling of a small astronomical trigger to powerful terrestrial feedbacks is why modern historians of science regard Croll as an early, largely self-taught practitioner of what we would now call Earth-system science — thinking about climate as an interconnected system of feedbacks rather than a simple, linear response to sunlight.

Croll's work landed in the middle of a golden age of climate physics on the other side of the Channel and in London. His contemporary John Tyndall was, from 1859 onward, running celebrated experiments at the Royal Institution demonstrating that certain gases — notably water vapour and carbon dioxide — absorb and re-emit radiant heat, laying the laboratory foundation for what we now call the greenhouse effect. Croll and Tyndall knew one another personally; Tyndall exchanged scientific correspondence with Croll and later lent his support to Croll's pension application, a rare instance of one largely self-taught and one formally trained scientist recognising each other's work as part of the same emerging field.

What Croll Did — and Did Not — Claim

It is tempting, in retrospect, to credit Croll with the full modern three-cycle scheme of eccentricity, precession and obliquity — the complete "Milankovitch" picture — simply because he is so clearly its forerunner. That temptation should be resisted, and this article deliberately does not give in to it. Croll's landmark 1864 paper, and his 1875 masterwork Climate and Time, were built on eccentricity and precession alone. The Croll specialist Kevin Edwards, writing in the Journal of Quaternary Science in 2022, is explicit on this point: obliquity and albedo considerations, he notes, "make their first appearance in other papers (viz. Croll 1867b, c) and Croll (1870a)... though not in the 1864 paper or in Climate and Time." In plainer terms: Croll's core, famous theory did not include axial tilt at all. Obliquity was a later addition to his thinking, developed in separate papers from 1867 onward, and it remained a secondary rather than central element of his overall framework — quite unlike the role it would eventually play for Milutin Milankovic.

This distinction matters for reasons beyond pedantry. Popular accounts of Croll frequently compress his career into a single moment — "in 1864 he discovered the Milankovitch cycles" — that flattens a real intellectual development spanning years and, more importantly, overstates what his founding paper actually contained. The accurate account, and the one this article follows, is that Croll pioneered the eccentricity-plus-precession mechanism in 1864, extended his thinking to include obliquity from 1867, and that the fully integrated, mathematically rigorous three-cycle theory — with obliquity restored to a central role, alongside a correct emphasis on summer rather than winter insolation — is properly credited to Milutin Milankovic, working from 1912.

The Vindication of a Janitor

Croll's 1864 paper brought him swiftly to the attention of the leading figures of British science of the day — Charles Lyell, William Thomson (Lord Kelvin), Andrew Ramsay and Archibald Geikie among them. In 1867, Geikie arranged for Croll to join the Edinburgh office of the Geological Survey of Scotland, as secretary and accountant and keeper of maps, a post that gave him a modest but reliable salary and, crucially, time to continue writing. The appointment came with a hitch entirely in keeping with the "genius janitor" character of his story: to join the civil service, Croll had to sit a formal examination — and he failed both the arithmetic and the English papers. His scientific patrons persuaded the authorities that his already-published calculations on the eccentricity of Earth's orbit were more than sufficient evidence of his arithmetical capacity, and the appointment went ahead regardless.

Climate and Time, in Their Geological Relations: A Theory of Secular Changes of the Earth's Climate appeared in 1875, published in London by Daldy, Isbister & Co. of 56 Ludgate Hill, with a simultaneous United States edition from D. Appleton & Co. of New York. It is worth correcting a publishing myth here: popular sources, including at times the Wikipedia article on Croll, have wrongly credited the 1875 first edition to Edward Stanford (who published only some of Croll's later works) or to Adam and Charles Black (who issued a reprint in 1885). The 1875 title page and library catalogues confirm Daldy, Isbister as the original publisher. The book had a genuinely significant effect on contemporary science: Charles Lyell revised his influential Principles of Geology in direct response to Croll's theory, and Croll himself corresponded with Charles Darwin, Alfred Russel Wallace, Joseph Hooker and other leading naturalists of the period.

The year 1876 marked the summit of Croll's public recognition. He was elected a Fellow of the Royal Society of London, his nomination certificate headed by the signature of Charles Darwin, with John Tyndall, Andrew Ramsay, Archibald Geikie and Lord Kelvin among his supporters; he was awarded an honorary LLD by the University of St Andrews; and he was made an honorary member of the New York Academy of Sciences. He also received several monetary awards from the Geological Society of London over the years: the balance of the Wollaston Donation Fund in 1872, the Murchison Fund in 1876, and the Barlow-Jamieson Fund in 1884. A myth worth debunking directly: Croll did not receive the Wollaston Medal or the Lyell Medal, the Geological Society's senior honours — the awards he received were cash fund distributions of similar name, and he was never even elected a Fellow of that particular society. He went on to publish two further books, Discussions on Climate and Cosmology (1885) and Stellar Evolution and Its Relations to Geological Time (1889), the latter containing an early attempt at explaining the chemical composition of the stars.

Overshadowed, Then Confirmed

For all its influence, Croll's theory faded significantly within his own lifetime. It predicted that ice ages should alternate between hemispheres and that the last one ended around 80,000 years ago, but improving geological dating techniques — including, notably, estimates based on the recession rate of Niagara Falls — pointed instead to a much more recent end to the last glaciation. That mismatch, combined with growing doubts about the theory's emphasis on winter insolation, meant the astronomical theory of the ice ages was widely regarded as discredited by the close of the nineteenth century. Croll retired in 1880 after suffering a stroke, living on a meagre superannuation of £75 16s 8d a year; his scientific friends helped arrange a move to a more comfortable house in Perth in 1886. He died there, in relative poverty, on 15 December 1890, his estate valued at only a few hundred pounds.

The theory did not die with him. It was revived and mathematically transformed by Milutin Milankovic (1879–1958), a Serbian mathematician and engineer who began his own climate work in 1912 and published his complete synthesis, the Canon of Insolation of the Earth and Its Application to the Ice-Age Problem, in 1941. Milankovic's contribution was not a mere restating of Croll's ideas; it involved two genuinely crucial corrections. First, he calculated insolation for specific latitudes rather than treating the globe as a single unit, giving the theory far greater precision. Second, and most importantly, he argued that the decisive factor for glaciation was northern-hemisphere summer insolation — whether summers are cool enough to leave the previous winter's snow unmelted — rather than the winter insolation that Croll had emphasised. Milankovic also restored obliquity to a central place in the theory. It is because of this mathematical rigour and this crucial correction that the orbital cycles now bear his name — "Milankovitch cycles" — even though Croll had proposed the essential astronomical mechanism roughly half a century earlier. The honest verdict, and the one this article adopts, is that Croll's priority for the underlying mechanism is genuine, but Milankovic's mathematical completeness and his summer-insolation correction were essential to the theory's eventual scientific acceptance. Some scholars accordingly refer to the combined framework as "Croll–Milankovitch cycles," a label that fairly credits both men.

Final vindication came in 1976, when J. D. Hays, John Imbrie and Nicholas Shackleton published "Variations in the Earth's Orbit: Pacemaker of the Ice Ages" in Science (volume 194, issue 4270, pages 1121–1132). Analysing deep-sea sediment cores recovered from the southern Indian Ocean, they identified dominant climate cycles at roughly 23,000, 41,000 and 100,000 years — matching precession, obliquity and eccentricity almost exactly. The orbital cycles that Croll had first identified in outline, and that Milankovic had mathematically refined, did indeed pace the rhythm of the ice ages, more than a century after Croll's original 1864 paper.

Why Croll Still Matters

The dominant glacial cycle of roughly the last 800,000 years runs at approximately 100,000 years — closely matching the eccentricity cycle Croll identified, although the underlying physics of why eccentricity, the astronomically weakest of the three forcings, comes to dominate the geological record remains an active area of research known as the "100,000-year problem," and this article does not present it as settled. Understanding these natural orbital rhythms remains fundamental to modern palaeoclimatology and to climate modelling more broadly.

It is also essential, today, to distinguishing natural climate variation from human-driven change. NASA is explicit that orbital forcing cannot account for present-day warming: across glacial cycles, atmospheric CO₂ fluctuated between roughly 180 and 280 parts per million, whereas human activity has now pushed concentrations past 420 ppm, and the resulting greenhouse warming overwhelms the much slower orbital signal that Croll first described. On orbital grounds alone, Earth's climate would eventually drift toward another glaciation; but Ganopolski, Winkelmann and Schellnhuber, writing in Nature in 2016, calculate that even the cumulative emissions already released are "already sufficient to postpone the next ice age for another 50,000 years," concluding that "we are basically skipping a whole glacial cycle, which is unprecedented." The very orbital framework Croll built in a Glasgow museum library to explain the coming and going of the ice ages now helps scientists measure precisely how far human activity has pushed the climate off its natural, orbit-driven course.

Croll is honoured today by the James Croll Medal, the highest award of the Quaternary Research Association, established in 2010; a matching medal instituted at the University of Edinburgh in 2011; and a memorial in Perth, in a courtyard near the Fair Maid's House. Not a bad legacy for a Perthshire crofter's son who never attended university and spent years as a museum janitor.

Timeline

1. c. 1830

   John Herschel muses on eccentricity's possible climatic effects

   An early, undeveloped suggestion that orbital geometry might matter to climate

2. 1821

   James Croll born at Little Whitefield, Cargill parish, Perthshire

   Son of a stonemason-crofter; formal schooling ends at about age 13

3. 1842

   Joseph Adhémar publishes Révolutions de la mer

   Proposes precession as a cause of glaciation; an important influence on Croll

4. 1859

   Croll becomes janitor of the museum at Anderson's College, Glasgow

   Light duties and library access let him pursue physics and astronomy self-study

5. 1864

   Croll publishes 'On the Physical Cause of the Change of Climate during Geological Epochs'

   Sets out eccentricity plus precession as the driver of ice ages, with ice-albedo feedback as an amplifier

6. 1867

   Croll begins extending his theory to obliquity and albedo in further papers

   The three-cycle scheme develops only from this point, not in the original 1864 paper

7. 1867

   Croll appointed to the Geological Survey of Scotland in Edinburgh

   Brought in by Archibald Geikie after the 1864 paper drew wide scientific attention

8. 1870

   Further papers refine Croll's treatment of obliquity

   Continuing development of the theory beyond its 1864 core

9. 1875

   Climate and Time, in Their Geological Relations published (Daldy, Isbister & Co., London; D. Appleton, New York)

   Croll's masterwork; prompts Charles Lyell to revise Principles of Geology

10. 1876

    Croll elected FRS, nominated by Charles Darwin; honorary LLD from St Andrews

    The summit of his public scientific recognition

11. 1880

    Croll retires after a stroke, on a small pension

    His theory is already fading under pressure from conflicting geological dating evidence

12. 1885

    Croll publishes Discussions on Climate and Cosmology

    Continues defending and extending his ideas late in life

13. 1890

    James Croll dies in Perth, in relative poverty

    His astronomical theory is largely out of favour at the time of his death

14. 1912–1941

    Milutin Milankovic develops and publishes his Canon of Insolation

    Recalculates insolation by latitude, emphasises northern-hemisphere summer insolation, and gives the cycles their modern name

15. 1976

    Hays, Imbrie and Shackleton publish 'Pacemaker of the Ice Ages' in Science

    Deep-sea sediment cores confirm dominant climate cycles matching precession, obliquity and eccentricity

16. 2010–2011

    The James Croll Medal and an Edinburgh memorial medal are instituted

    Modern recognition of Croll's pioneering, once-discredited theory

Myths & Facts

Did You Know?

• James Croll left school at about age thirteen and was almost entirely self-taught, reading his way into physics and astronomy from a museum janitor's post in Glasgow.
• Croll failed the arithmetic and English examinations required to join the Geological Survey of Scotland's civil service in 1867 — his published orbital calculations were accepted instead as proof of his mathematical ability.
• Charles Darwin personally headed Croll's nomination certificate for election as a Fellow of the Royal Society in 1876.
• Croll's 1864 theory did not include axial tilt (obliquity) at all — that came only in later papers from 1867 onward.
• Croll never received the Wollaston Medal or Lyell Medal, despite what some popular sources claim; he received Geological Society cash funds of similar names instead.
• The orbital cycles are named 'Milankovitch cycles' after Milutin Milankovic, who corrected Croll's theory by showing that summer insolation, not winter insolation, controls glaciation.
• It took 112 years — from Croll's 1864 paper to the 1976 'Pacemaker of the Ice Ages' study — for the astronomical theory of the ice ages to be confirmed by physical evidence.
• Croll died in relative poverty in Perth in 1890, a decade before his theory would even begin to be taken seriously again.

Honest Caveats

Croll's 1864 theory did not include obliquity. His founding paper and his 1875 book Climate and Time were built on eccentricity and precession only; axial tilt entered his thinking in separate papers from 1867 onward, and it would be inaccurate to credit him with the full three-cycle Milankovitch scheme from the start.

The theory's modern name and mathematical completeness belong to Milutin Milankovic, not to Croll. Milankovic's identification of northern-hemisphere summer insolation, rather than Croll's winter insolation, as the decisive factor was an essential correction, not a mere restatement.

Empirical confirmation is credited specifically to Hays, Imbrie and Shackleton's 1976 sediment-core study, not to Croll or Milankovic themselves, neither of whom had the geological evidence to prove the theory in their own lifetimes.

Croll did not receive the Wollaston or Lyell Medals, contrary to some popular accounts; his Geological Society honours were cash fund awards, and he was never a Fellow of that Society.

The astronomical theory does not explain modern human-driven climate change. Orbital forcing operates far too slowly to account for the rapid CO₂-driven warming of the industrial era, and this article does not suggest otherwise.

Minor biographical discrepancies exist and are not treated as resolved. Croll's exact birthplace is described slightly differently across sources (Cargill parish, near Wolfhill, or St Martin's — neighbouring Perthshire localities), and his mother's maiden name appears as both "Ellis" and "Geddes." The "100,000-year problem" in glacial cycle physics also remains a genuinely unresolved research question, not a settled matter.