The great British geologist Arthur Hallam has suggested that if James Hutton, the father of geology, had visited Switzerland, he would have seen at once the significance of the carved valleys, the polished striations, the telltale strand lines where rocks had been dumped, and the other abundant clues that point to passing ice sheets. Unfortunately, Hutton was not a traveler. But even with nothing better at his disposal than secondhand accounts, Hutton rejected out of hand the idea that huge boulders had been carried three thousand feet up mountainsides by floods-all the water in the world won’t make a boulder float, he pointed out-and became one of the first to argue for widespread glaciation. Unfortunately his ideas escaped notice, and for another half century most naturalists continued to insist that the gouges on rocks could be attributed to passing carts or even the scrape of hobnailed boots.

Local peasants, uncontaminated by scientific orthodoxy, knew better, however. The naturalist Jean de Charpentier told the story of how in 1834 he was walking along a country lane with a Swiss woodcutter when they got to talking about the rocks along the roadside. The woodcutter matter-of-factly told him that the boulders had come from the Grimsel, a zone of granite some distance away. “When I asked him how he thought that these stones had reached their location, he answered without hesitation: ‘The Grimsel glacier transported them on both sides of the valley, because that glacier extended in the past as far as the town of Bern.’ ”

Charpentier was delighted. He had come to such a view himself, but when he raised the notion at scientific gatherings, it was dismissed. One of Charpentier’s closest friends was another Swiss naturalist, Louis Agassiz, who after some initial skepticism came to embrace, and eventually all but appropriate, the theory.

Agassiz had studied under Cuvier in Paris and now held the post of Professor of Natural History at the College of Neuchâtel in Switzerland. Another friend of Agassiz’s, a botanist named Karl Schimper, was actually the first to coin the term ice age (in German Eiszeit), in 1837, and to propose that there was good evidence to show that ice had once lain heavily across not just the Swiss Alps, but over much of Europe, Asia, and North America. It was a radical notion. He lent Agassiz his notes-then came very much to regret it as Agassiz increasingly got the credit for what Schimper felt, with some legitimacy, was his theory. Charpentier likewise ended up a bitter enemy of his old friend. Alexander von Humboldt, yet another friend, may have had Agassiz at least partly in mind when he observed that there are three stages in scientific discovery: first, people deny that it is true; then they deny that it is important; finally they credit the wrong person.

At all events, Agassiz made the field his own. In his quest to understand the dynamics of glaciation, he went everywhere-deep into dangerous crevasses and up to the summits of the craggiest Alpine peaks, often apparently unaware that he and his team were the first to climb them. Nearly everywhere Agassiz encountered an unyielding reluctance to accept his theories. Humboldt urged him to return to his area of real expertise, fossil fish, and give up this mad obsession with ice, but Agassiz was a man possessed by an idea.

Agassiz’s theory found even less support in Britain, where most naturalists had never seen a glacier and often couldn’t grasp the crushing forces that ice in bulk exerts. “Could scratches and polish just be due to ice?” asked Roderick Murchison in a mocking tone at one meeting, evidently imagining the rocks as covered in a kind of light and glassy rime. To his dying day, he expressed the frankest incredulity at those “ice-mad” geologists who believed that glaciers could account for so much. William Hopkins, a Cambridge professor and leading member of the Geological Society, endorsed this view, arguing that the notion that ice could transport boulders presented “such obvious mechanical absurdities” as to make it unworthy of the society’s attention.

Undaunted, Agassiz traveled tirelessly to promote his theory. In 1840 he read a paper to a meeting of the British Association for the Advancement of Science in Glasgow at which he was openly criticized by the great Charles Lyell. The following year the Geological Society of Edinburgh passed a resolution conceding that there might be some general merit in the theory but that certainly none of it applied to Scotland.

Lyell did eventually come round. His moment of epiphany came when he realized that a moraine, or line of rocks, near his family estate in Scotland, which he had passed hundreds of times, could only be understood if one accepted that a glacier had dropped them there. But having become converted, Lyell then lost his nerve and backed off from public support of the Ice Age idea. It was a frustrating time for Agassiz. His marriage was breaking up, Schimper was hotly accusing him of the theft of his ideas, Charpentier wouldn’t speak to him, and the greatest living geologist offered support of only the most tepid and vacillating kind.

In 1846, Agassiz traveled to America to give a series of lectures and there at last found the esteem he craved. Harvard gave him a professorship and built him a first-rate museum, the Museum of Comparative Zoology. Doubtless it helped that he had settled in New England, where the long winters encouraged a certain sympathy for the idea of interminable periods of cold. It also helped that six years after his arrival the first scientific expedition to Greenland reported that nearly the whole of that semicontinent was covered in an ice sheet just like the ancient one imagined in Agassiz’s theory. At long last, his ideas began to find a real following. The one central defect of Agassiz’s theory was that his ice ages had no cause. But assistance was about to come from an unlikely quarter.

In the 1860s, journals and other learned publications in Britain began to receive papers on hydrostatics, electricity, and other scientific subjects from a James Croll of Anderson’s University in Glasgow. One of the papers, on how variations in Earth’s orbit might have precipitated ice ages, was published in the Philosophical Magazine in 1864 and was recognized at once as a work of the highest standard. So there was some surprise, and perhaps just a touch of embarrassment, when it turned out that Croll was not an academic at the university, but a janitor.

Born in 1821, Croll grew up poor, and his formal education lasted only to the age of thirteen. He worked at a variety of jobs-as a carpenter, insurance salesman, keeper of a temperance hotel-before taking a position as a janitor at Anderson’s (now the University of Strathclyde) in Glasgow. By somehow inducing his brother to do much of his work, he was able to pass many quiet evenings in the university library teaching himself physics, mechanics, astronomy, hydrostatics, and the other fashionable sciences of the day, and gradually began to produce a string of papers, with a particular emphasis on the motions of Earth and their effect on climate.

Croll was the first to suggest that cyclical changes in the shape of Earth’s orbit, from elliptical (which is to say slightly oval) to nearly circular to elliptical again, might explain the onset and retreat of ice ages. No one had ever thought before to consider an astronomical explanation for variations in Earth’s weather. Thanks almost entirely to Croll’s persuasive theory, people in Britain began to become more responsive to the notion that at some former time parts of the Earth had been in the grip of ice. When his ingenuity and aptitude were recognized, Croll was given a job at the Geological Survey of Scotland and widely honored: he was made a fellow of the Royal Society in London and of the New York Academy of Science and given an honorary degree from the University of St. Andrews, among much else.