After tobacco the most important product of the Cumanacoa valley is indigo, whose intense color makes it the equal of Guatemalan indigo. All the indigo factories that we visited are constructed along the same principles. Two vats, where the plants 'rot', are placed together. Each one measures 15 feet square and 2. feet deep. From these upper vats the liquid passes into beaters where the water-mill is placed. The axle-tree of the great wheel crosses the two beaters. It is nailed with ladles, fixed to long handles, for the beating. From another percolating vat the coloured starch passes to the drying-boxes, spread on planks of Brazil-wood on small wheels so that they can be pushed under a roof in case of sudden rain. These sloping and low roofs give the drying-boxes the appearance of hothouses from a distance. In the Cumanacoa valley the fermentation of the plant takes place amazingly quickly; usually it does not take longer than four or five hours. This can be attributed to the humidity and the absence of sun during the plant's development.

The look of the sky, the movement of electricity, and the downpour of the 28th March announced the start of the rainy season: we were still advised to go to San Fernando de Apure by San Francisco de Capanaparo, along the Sinaruco river and the San Antonio hato to the Otomac village recently founded on the banks of the Meta river, and to embark on the Orinoco a little above Carichana. This land road crosses an unhealthy, fever-ridden country. An old farmer, Don Francisco Sanchez, offered to lead us. His clothes revealed how simply people live in these far-off countries. He had made a fortune of 100, piastres yet he rode on horseback barefoot with large silver spurs. We knew from several weeks' experience how sad and monotonous the llanos are and so we chose the longer route along the Apure river to the Orinoco. We chose one of the long pirogues that the Spaniards call lanchas. A pilot and four Indians were sufficient to drive it. On the poop a cabin covered with corypha leaves was built in a few hours. It was so spacious that it could have held a table and benches. They used oxhides stretched and nailed to frames of Brazil-wood. I mention these minute details to prove that our life on the Apure river was very different from the time when we were reduced to the narrow Orinoco canoes. We packed the pirogue with provisions for a month. You find plenty of hens, eggs, bananas, cassava and cacao at San Fernando. The good Capuchin monk gave us sherry, oranges and tamarinds to make fresh juices. We could easily tell that a roof made of palm leaves would heat up excessively on the bed of a large river where we would be always exposed to the sun's perpendicular rays. The Indians relied less on our supplies than on their hooks and nets. We also brought some weapons along, whose use was common as far as the cataracts. Further south the extreme humidity prevents missionaries from using guns. The Apure river teems with fish, manatees (91) and turtles whose eggs are more nourishing than tasty. The river banks are full of birds, including the pauxi and guacharaca, that could be called the turkey and pheasant of this region. Their flesh seemed harder and less white than our European gallinaceous family as they use their muscles more. We did not forget to add to our provisions fishing tackle, firearms and a few casks of brandy to use as exchange with the Orinoco Indians.

Two main aims guided my travels, published as the Relation historique. I wanted to make known the countries I visited, and to collect those facts that helped elucidate the new science vaguely named the Natural History of the World, Theory of the Earth or Physical Geography. Of these two aims, the second seemed the more important. I was passionately keen on botany and certain aspects of zoology, and flattered myself that our researches might add some new species to those already known. However, rather than discovering new, isolated facts I preferred linking already known ones together. The discovery of a new genus seemed to me far less interesting than an observation on the geographical relations of plants, or the migration of social plants, and the heights that different plants reach on the peaks of the cordilleras.

We observed that fern trees are usually far rarer than palm trees. Nature has limited them to temperate, humid and shady places. They shun the direct rays of the sun and while the pumos, corypha of the steppes, and other American palms prefer the naked, burning plains these tree fern, which seen from afar look like palms, maintain the character and habits of cryptogams. They prefer solitary places, shade, humidity and damp. Sometimes you find them on the coast, but only when protected by thick shade.

Travelers know by experience that views from the summits of high mountains are neither as beautiful, picturesque, nor as varied as those from the heights of Vesuvius, Righi or the Poy-de-Dôme. Colossal mountains such as Chimborazo, Antisana or Monte Rosa compose such a huge mass that the richly cultivated plains are seen only at a great distance where a bluish and watery tint spreads over the landscape. The Tenerife peak, due to its narrow shape and local position, combines the advantages of the less high summits with those of the very high. From its top we can see not only the sea to the horizon, but also the forests of Tenerife and the inhabited coastal strips, which seem so close that their shapes and tones stand out in beautiful contrasts. It could be said that the volcano crushes the little island that serves as its base, and that it shoots up from the depths of the seas to a height three times higher than cloud level in summer.

In our publications Bonpland and I have considered every phenomenon under different aspects, and classed our observations according to the relations they each have with one another. To convey an idea of the method followed, I will outline what we used in order to describe the volcanoes of Antisana and Pichincha, as well as Jorullo, which on the night of the 20th of September 1759 rose 1, feet up from the plains of Mexico. We fixed the position of these remarkable mountains in longitude and latitude by astronomical observations. We took the heights of different parts with a barometer, and determined the dip of the needle and magnetic forces. We collected plants that grew on the slopes of these volcanoes, and specimens of different rocks. We found out the exact height above sea-level at which we made each collection. We noted down the humidity, the temperature, the electricity and the transparency of the air on the brinks of Pichincha and Jorullo; we drew the topographical plans and geological profiles of these volcanoes by measuring vertical bases and altitude angles. In order to judge the correctness of our calculations we have preserved all the details of our field notes.