Room for Cotopaxi

Cotopaxi stands unusually close to Chimborazo in Alexander von Humboldt's printed Andes. The broad snowy dome and the smoking cone make an effective pair. Humboldt knew that the arrangement was geographically wrong. He explained why he had chosen it.

In February 1803, while sketching the first version at the Pacific port of Guayaquil, he reported hearing Cotopaxi's subterranean detonations. They sounded, he wrote, like repeated discharges from a battery. The volcano was far away, but it earned a place in the drawing. Cariguairazo, the mountain that properly belonged beside Chimborazo, was passed over.

The resulting Tableau physique des Andes et pays voisins was designed to hold much more than a view. Plant names occupy a cutaway mountain; columns on either side connect altitude with crops, weather, air pressure and other physical conditions. Its title credits Humboldt and the French botanist Aimé Bonpland, and specifies observations made between ten degrees north and ten degrees south of the equator in 1799–1803. That is a substantial stretch of tropical America to fit inside one outline.

Plants needed paper

Humboldt, a Prussian naturalist trained in mining, and Bonpland travelled through Spanish America between 1799 and 1804. Collecting plants required room, labour and supplies. During an excursion inland from Cumaná in present-day Venezuela in 1799, their plant tins and handkerchiefs filled. Humboldt recorded that Bonpland's stock of paper could not accommodate the abundance. They sent a messenger back for eight hundred more sheets.

Plant geography added another demand to the work of identifying and preserving specimens. Where did a plant grow, how common was it, and what conditions accompanied it? A mountain offered an especially compact comparison: a climb passed through changes of temperature and vegetation that a traveller might otherwise encounter over a long journey towards colder latitudes.

Humboldt's notes on the Río Magdalena distinguished common plants by underlining their names, using more underlines for greater prevalence. On the subsequent route from Honda to Bogotá in 1801, a table joined barometric heights to characteristic vegetation, including sugar cane, bamboo and trees of the genus Weinmannia. The plant-name mountain was taking shape before the famous attempt on Chimborazo.

The travellers were entering a region with an existing scientific community. In Bogotá they worked with José Celestino Mutis, whose botanical investigations extended across decades. In the Quito region they met Francisco José de Caldas, who was measuring heights and studying the distribution of Andean plants. Caldas shared measurements and discussed instruments and methods with Humboldt.

Humboldt later sent Mutis a copy of the first Guayaquil sketch. He considered a botanist who had collected at many different elevations particularly well placed to judge it. Mutis's work, Caldas's measurements and Bonpland's collecting helped supply the knowledge behind the two names printed beneath the mountain.

A height had to be calculated

A barometer measured atmospheric pressure. As a traveller climbed, the diminishing weight of air above generally produced a lower reading. Turning that reading into altitude required calculation, including allowance for temperature. A plant specimen and its eventual elevation were therefore not always two facts that could simply be written together at the instant of collection.

The calculations continued after the journey. In his preface Humboldt thanked the French engineer Gaspard de Prony for having more than four hundred height measurements calculated under his supervision, using Laplace's formula and allowing for gravity. The physicist Jean-Baptiste Biot calculated tables for refraction and the weakening of light through the atmosphere. Preparing the sheet involved numerical work as well as drawing.

The margins invite a reader to compare those different kinds of information at a common height. Follow a level across the paper and the mountain can be read beside pressure, cultivation and the appearance of the sky. Heights elsewhere in the world provide reference points for readers who know Mont Blanc better than the Andes.

Even blue had a scale. Humboldt carried a cyanometer made by Paul in Geneva after the instrument used by the Alpine investigator Horace-Bénédict de Saussure. It allowed the observer to compare the sky with graded blues. Humboldt made his observations overhead, at the zenith; the plate's column gives numbered values for the sky's intensity. The apparent colour of the mountain air had become something to set alongside a barometer reading.

Antisana inside Chimborazo

The plants demand particular care. In March 1802 Humboldt and Bonpland spent four days based high on Antisana, another Ecuadorian volcano. Bonpland collected along its slopes, including a plant recorded as Senecio nivalis at about 4,860 metres. The snow-covered Chimborazo ascent in June supplied a very different opportunity for botanical work.

In 2019, Pierre Moret and colleagues reconstructed the expedition's high-altitude collecting from journals, specimen labels and later botanical publications. Much of the alpine material led back to Antisana, rather than the mountain whose outline became famous. The printed ordering also failed to incorporate some of the expedition's own highest plant records. Chimborazo's prominence on the sheet cannot establish where a particular specimen was collected.

Humboldt's explanatory text was often more particular than a neat succession of belts might suggest. He described most palms as plants of low, warm country, then singled out the wax palm of the Andes, which he and Bonpland knew from the Quindío and Tolima regions. It grew far higher, its trunk coated with wax. The exception required an explanation rather than removal from the scheme.

There were practical consequences to getting a plant's identity right. Discussing the bark-producing Cinchona, Humboldt rejected reports that placed the genuine trees close to the permanent snow. Other trees had bark used against fever too; similarity of use did not make them the same botanical genus. A familiar product name could conceal different plants and different growing ranges.

The sheet grows in Paris

The sketch made at Guayaquil travelled into a larger production. In Paris, the artist Lorenz Adolf Schönberger enlarged Humboldt's drawing. According to Humboldt, Schönberger lacked time to complete the detail required for engraving, so the botanical artist Pierre Jean François Turpin carried the work forward.

The design imposed a basic difficulty. If horizontal distance and height used the same scale, the mountains would scarcely rise from the page. If the landscape retained the scale chosen for height, the paper would have to be vastly longer. Humboldt estimated that even the limited section he wanted would require a sheet forty times the length of the book's format. He exaggerated the slopes and compressed the ground between them.

The forest could be drawn as vegetation, with recognisable palms near the sea, but the mountain also needed space for words. Dense Latin names replace solid rock. Larger diagonal labels gather plants into regions, while smaller ones occupy the available surface. Turpin had to make a drawing that could be inspected as information and still hold together as a landscape.

Humboldt acknowledged the conflict. Some viewers, he wrote, would have preferred the observations moved away from the mountain into the margins. He had tried to serve both pictorial effect and accuracy. Many newly collected plants were still awaiting settled names, so even the vocabulary to be engraved remained incomplete.

Bonpland's role continued through the checking. Humboldt asked him to have the engraving reviewed by Candolle and Biot after his own inspection. In the letter reproduced by later historians, Humboldt insisted that his reputation was at stake. A sheet intended to make a large body of knowledge visible at once was being checked while that knowledge was still being worked out.

The drawing bears the date 1805. The book's title page also gives 1805, but publication history places the public appearance of the work and its accompanying plate in 1807. Field observation, drawing and publication belong to different stages of its manufacture.

Its vegetation zones remained open to correction. The high grass belt on the famous plate did not agree with the sequence of grassland below sparse alpine vegetation described in earlier observations. Humboldt's comparative diagram of 1815 put those zones in the more defensible order. The imposing mountain-shaped version was not his final arrangement of the plants.

At the foot of the French sheet, beneath the two travellers' names, a smaller line records the people who turned it into print: Schönberger and Turpin drew it, Bouquet engraved it, Beaublé supplied the lettering, and Langlois printed it. The tiny plant names had their own specialist.