It is a beautiful adventure that Frédéric Aitken and Jean-Numa Foulc have undertaken, using physical data from the Challenger expedition, the first major oceanographic expedition, sponsored by the British Admiralty in the 1870s. Indeed, this data, temperature and pressure readings at various depths and at multiple points of the world, was relatively little used at the time despite the visionary intuition of one of the initiators of the expedition, Professor Carpenter, that this data would allow for the reconstruction of ocean circulation. The authors attribute this relative lack of interest to the fact that most scientists on the expedition were naturalists, and that from the point of view of biology, the total benefits were already huge, with, for example, the discovery of life at a great depth.

Exploiting data is not the least interesting of the physicist’s tasks. To deal with the problem, we simplify the situation and try not to delete anything essential. The terms of the equations are evaluated, keeping only the most important, and then two situations may arise. Let us say that the discrepancy with the data is clear: we are generally convinced that it has been oversimplified, but where? We are tempted in bad faith to defend our idea, even if it means becoming the Devil’s advocate and destroying what we have built. We go back to the overlooked terms one by one, and, with some luck, this may lead to a new effect. We make do with what we know; the battle is tough, and this is its appeal.

Let us say that the similarity is acceptable. This is when a good physicist is suspicious: is it not a coincidence that two important effects are not offset by any chance? It would be necessary to make a prediction, and to repeat the experiment in different conditions, but it is not always possible. Another boat was not sent out with 200 people around the world for three years! The rigor with which experiments have been conducted, and the confidence that can be placed in the measures, are essential. The experimenters have had to multiply the situations blindly, without knowing which ones would be used as a test, with the sole aim of doing their best every time, by describing their protocol for future use.

The development of the measurement protocol is part of the experiment’s design, as was instrument construction. At that time, a physicist worth his salt would never have used an instrument that he did not know how to build. How can one measure a temperature in a place that one cannot reach oneself (2,000 m below the surface of the sea, for example)? We can record the maximum and minimum temperatures reached during the descent (I found, with much emotion, the description of the maximum and minimum thermometer used by my grandfather in his garden). But what to do for intermediate temperatures? How to make sure that the line does not break in bad weather under the boat’s blows? How to decide the real depth despite currents, and the fact that the line continues to run under its own weight once the sensor is at the bottom? The design phase of the experiment can be exciting: I knew a physicist who was ready to sabotage a barely built experience (under the pretext, of course, of improving it) to be able to move more quickly to the design of the following experiment.

Despite all the attention given to the design, sometimes an error is suspected in the measurements. This is the case here. Having reached unexpected depths (they discovered the Mariana Trench), the Challenger scientists wondered if their measurements had not been distorted by contraction of the glass envelopes. After their return, they assigned Peter Tait, a physicist from Edinburgh, the task of assessing these errors. One thing leading to another, he raised questions about the compressibility of seawater, and other liquids, and so about their equation-of-state, connecting pressure, temperature and density (and even salinity). The result of his studies left a lasting mark on the physics of liquids. Estimating errors, a task hated and despised by the typical physics student, yielded new knowledge.

From the same period as the van der Waals equation, Tait’s efforts were part of the first trials to represent the equation-of-state of dense, liquid and solid bodies by continuous functions. The goal was twofold: metrological, to interpolate between experimental results, and to provide experimenters and engineers with the most accurate characterization of the thermodynamic and physical properties of the fluids they use. But also more fundamental, in the wish to have a better understanding of the underlying physical mechanisms: formation of molecular aggregates, local crystalline order, shape of interaction potentials, etc. These two interests, pragmatism and rigor, are often in conflict, as is clear from the authors’ account, who apply the ideas from that time to fluids that were not of concern then, such as the fluid phases of the two stable isotopes of helium.

Many aspects of this scientific adventure are thus universal, and it is touching to see how the value codes of the scientific approach have been transmitted over decades, or almost centuries. But our step back in time gives us an advantage: the ability to judge the ideas from that period in light of the extraordinary sum of knowledge that has been accumulated since. However, a direct comparison would be unfair and clumsy. It is much more interesting to put us in the mindset of the players of that era, to share their doubts, their hesitations and even their mistakes. This is an aspect that is too often absent from our education. For the sake of efficiency, we do not mention brilliant ideas that have led to a stalemate. Yet these ideas may contribute elsewhere. There may be some hesitation in mentioning great names such as Clausius, Joule and van der Waals, who fill us not only with humility in the face of the mastery that allowed them to find the right path, but also with confidence when faced with our own doubts. The variety of players and points of view that have marked this period show how much science is a collective adventure.

It is all of this that I found in this book by Frédéric Aitken and Jean-Numa Foulc, and even more: the human adventure that was this trip of three years around the world, the incidents, drama and joys, what it revealed about the personality of each participant, their lives which, for some, are also described, the moving relay that is transmitted when a change of assignment, or worse, death, interrupts a task. There is also the welcome reserved for the expedition, sometimes idyllic (ah! the difficulty of leaving Tahiti), sometimes colder, the importance of the band and personal talent of the participants, not to mention the providence that the Challenger represented for the Robinsons, abandoned on an island by a boat that was unable to come back for them. After reading the story based on the logbook, how can we not mention Jules Verne’s novels? It is the same period, that of a thirst for knowledge about our environment, accessible to all of us, acquired by real yet so human adventurers, so close to us. The credit goes to the authors for having dedicated so much time, energy and enthusiasm to this humanist and complete book, with the spirit of this laboratory where I had the pleasure to come for discussions during my years at Grenoble.

Bernard CASTAING

Member of the French Academy of Sciences