A dispute over water vapour

John Tyndall (c.1822–1893), Irish physicist, mountaineer, and public intellectual, is best known in scientific circles for his work on the absorption of heat by gases such as carbon dioxide in the atmosphere (and for explaining why the sky is blue).

But it was not carbon dioxide that most interested him. On 20th November 1860 he determined for the first time the substantial absorption of heat by water vapour compared to dry air. As he refined his apparatus for measuring the absorption of heat by gases, the strong absorption by water vapour struck him ever more forcibly.

Tyndall’s radiant heat apparatus from the frontispiece of

his book Contributions to Molecular Physics in the Domain of

Radiant Heat (Roland Jackson)

It was immediately apparent to him not only that this absorption could explain differences between the temperatures at midday and evening, or the temperature at the top of a mountain compared to the bottom, but also that changes in the amount of water vapour, carbon dioxide, or hydrocarbons, all of which absorbed heat, could have climatic effects. He wrote: ‘if, as the above experiments indicate, the chief influence be exercised by aqueous vapour, every variation of this constituent must produce a change of climate. Similar remarks would apply to the carbonic acid [carbon dioxide] diffused through the air’.

But the path of scientific discovery is not always smooth. Starting in March 1861, one of the most respected experimental physicists of the day, the German Gustav Magnus (1802–1870) challenged Tyndall's findings. With his apparatus, he claimed to show that water vapour did not significantly absorb heat.

That autumn, Tyndall had time to follow up his interest in water vapour, and to start to address Magnus’s challenge. On 13th September he recorded that ‘aqueous vapours in the atmosphere this day exerted an absorption of at least 25 times the absorption of the atmosphere itself’, and by 10th October the multiplier had gone to 40.

Who was right, and whom should others believe? It was not just a matter of personal reputations. For Tyndall, it was the significance of the absorption of heat by water vapour for meteorology that particularly exercised him. He thought that meteorologists would be seriously misled by believing otherwise. The consequences of this disagreement preoccupied him for 20 years.

Over his lifetime Tyndall gathered rather a reputation for scientific feuding (though it takes two to feud). But this dispute with Magnus was no feud. Tyndall had worked in Magnus’s private laboratory in Berlin, like many other leading German physicists. He held Magnus in the greatest personal respect, and gave pride of place to a photograph of Magnus over his chimney piece. The respect was mutual. Magnus saw Tyndall as one of the ablest young scientific men of Britain and regarded him with great personal friendship.

During the next few years, they worked away at the problem in their own laboratories. Both privately and publicly they exchanged letters and papers. The issue was so significant for Tyndall that he later said that he effectively suspended other work for two years from August 1862 while he sought to meet Magnus’s objections. In hindsight, it seems surprising that Magnus went to his death in 1870 apparently unconvinced, because Tyndall had plenty of evidence. He met Magnus’s criticisms of his experimental arrangement by redesigning it (finding the same results), and pointed out the likely fault in Magnus’s (Tyndall suggested that Magnus allowed his gas sample to come into direct contact with his heat source, thereby chilling it). One argument that Tyndall considered particularly compelling was the well-established strong absorption of heat by liquid water. He had shown in many experiments that liquids that were strong absorbers had vapours that behaved likewise. For Tyndall, this was explicable on the basis that absorption was a property of the molecules themselves. But still Magnus disagreed.

This continuing challenge from so eminent a physicist as Magnus meant that Tyndall could never feel secure. Yet he knew his findings were important for meteorology. Given its capacity to absorb and radiate heat, Tyndall could see the importance of the movement of water vapour in the atmosphere, and its complexity. He wrote to the great geologist Charles Lyell in 1865, in response to a question from him on the possible differential heating and cooling of the Earth’s hemispheres: ‘the existence of our atmosphere & the transport of water in the shape of snow from the equatorial regions to the polar ones, render the actual problem a complicated one’.

Even as late as 1880, 10 years after Magnus’s death, objections to Tyndall's findings were current, so Tyndall recommenced experiments. In his fourth and final Bakerian Lecture to the Royal Society in 1881 (no one since has given as many of this prestigious lecture) Tyndall reiterated his claims, with some further compelling evidence. He had recently explored the emission of sound by gases, by subjecting them to pulses of heat. His strong absorbers, from previous experiments, gave loud sounds. To his delight, but not to his surprise, water vapour gave a loud sound too.

Reproducibility of scientific results is a perennial issue. It was exacerbated in Tyndall's day because of the individual and unique design of apparatus by different experimenters. Tyndall repeatedly stressed the importance of the sustained and specific experimental knowledge and skill that he had built around his apparatus. He argued that someone less skilled, less conscious of all the possible sources of error, could not have achieved his results. Not even Magnus could repeat the experiments, even after he had seen them with his own eyes in London.

In the final part of his lecture, Tyndall drew on the significance for meteorology of the absorption and emission of radiant heat by water vapour, with a side-swipe at meteorologists who refused to accept the validity or implications of laboratory experiments. He had long argued that high enough in the atmosphere, above significant absorption by water, the solar spectrum would show a greater proportion of invisible to visible solar rays. A letter from Samuel Langley, from 12,000ft up Mount Whitney in October 1881, had verified his prediction.

After he gave this lecture, the physicist George Stokes told him: ‘I had indeed not examined, except very imperfectly, Magnus’s work, and I had heard the opinion expressed that you had best shut up. But I could not imagine how what you had done could be upset’. Tyndall was grateful, writing: ‘I have sometimes been disheartened by the efforts of really eminent men to reconcile . . . Magnus’s results and mine . . . Of course I knew that it would all come right in the end. But it is gratifying to have it set right in one’s own lifetime’. Lyon Playfair had a similar response to Stokes, writing: ‘I was very loth to doubt your previous results, but your present experiments seem to me to place them beyond all question and to establish a chief factor in the solution of many of our Meteorological difficulties’.

In the end, Tyndall's views, so central to understanding the role of water vapour in the atmosphere, prevailed.

We have Tyndall to thank both for a physical understanding of the properties of water vapour in the atmosphere, something relevant to weather, climate, and the safety of mariners (as covered in Part 1). Tyndall showed that laboratory experiments too could help us understand weather in the uncontrollable atmosphere.

A version of this post appears as a guest piece on Roger Turner's Picturing Meteorology blog.