The production of Para rubber is largest in the island district, extending from the mouth of the Amazon about 500 miles up the river, and comprising a multitude of low, tide-flooded islands and the low alluvial shores of the main stream and its affluents from the south. Here the principal rubber-yielding tree is the "white hevea," a variation of H evea Brasiliensis, re garded by botanists as having "migrated" down the river from its original habitat in she up land region near the headwaters toward the southwest. Its white color is attributed to the peculiar whitish lichens which grow upon its trunk in the down-river country. On the higher lands the lichens are dark gray or black, giving rise *to the title "black hevea." This explanation, however, does not account for the decided difference in the cortex of the two trees even when found growing close together, nor for the "red hevea" growing in both dis tricts. The trees are scattered through the forest, generally from 200 to 300 feet apart, although occasionally in groups of two to six. A trail is laid out to cover from 130 to 150 trees, a distance of about six miles. The hevea sap flows very slowly, and requires from 100 to 150 tappings a season. In some sections a new tapping is made every day, in others every other day. At the camps the rubber is separated from the sap by drying out the moisture over a smoky fire made in such a way that the smoke contains a considerable proportion of acetic acid and creosote—a slow process. A great saving of time in this coagu lation process has been accomplished by the introduction of a new treatment of the raw sap with a mixture of benzine and wood alcohol.
The output of Amazonian rubber is con trolled primarily by the cost of getting the rubber to navigable water, and also to a large degree by the reluctance of laborers to go far into the forests. The population which is de pended upon for laborers is apathetic, due to an anaemic condition of health, this in turn being the result chiefly of climatic conditions. The rainy season begins in November, and from March to May the entire river country is subject to flooding, in some localities the waters rising as much as 50 feet. Owing to the low gradient the water is very slow in running off. The impossibility of maintaining roads under such conditions operates oppress ively against development of the industry. Whatever rubber collecting is done must begin not earlier than May and be completed by the end of October. On the higher lands toward the western and southwestern part of the basin the rainfall is less and the floods run off more rapidly and the rubber season is longer. Com plaint is made that the average laborer works but 100 days in the season, and that in a large part of the rubber-producing region the work could be and should be carried on for 200 days, adding from 50 to 60 per cent to the annual output.
The figures for the 1916 output of wild rubber in the Amazonian region are only par tially complete. They show that in that year
17,747 tons were shipped from Manaos to New York and European ports, and that 38,682 tons were shipped from Para to the same destina tions, a total from these two principal ports of 56,329 tons. Of the whole, 20,334 tons went to Europe (26,792 tons in 1915) and 35,995 tons to New York (33,597 tons in 1915). The world's yield of plantation (cultivated) rubber for 1916 is officially estimated at 160,000 tons. The latest available complete figures are those for the fiscal year ended 30 June 1913: they are quoted here for purposes of comparison. The output of the Amazon valley in that year was 43,362 tons. Of this 31,362 tons were pro duced iu the upper rivers section, including parts of the Bolivia and Peru rubber districts, and 12,000 tons in the lower Amazon. country. As classified, this output consisted of 16,971 tons (39.12 per cent) of ((Fine Hard Para)); 8,860 tons (20.44 per cent) of 'Entre-fine° and aFracan (medium-fine and weak); 7,400 tons (17.07 per cent) of aSernambyn (scraps and niggerheads); and 10,131 tons (23.37 per cent) of Outside of this yield other Brazilian states shipped about 4,000 tons, mostly Ceara rubber. The total Brazilian output for that year was therefore about 47,000 tons, or about 40 per cent of the world's production.
The territory which produces Ceara rubber includes the Brazilian states of Ceara, Piauhy and Bahia. The Manihot trees from which this kind of rubber is collected thrives on compara tively poor soil, on desert plains and dry hill sides up to an elevation of 3,500 feet above the sea. It withstands the long drought from May to November, and is indifferent to occasional hoar frosts. There are sonic plantations of the Manih.ot in Brazil, and tapping is begun as early as the second year. The yield of this district is about 4,000 tons annually.
North of the Amazon the Hevea Brasiliensis does not grow in commercial quantity, but the Hevea Benthamiana is plentiful and yields an excellent rubber. As a rubber dis trict, however, this territory has not been seri ously exploited, except in the Rio Araguaya district near the Atlantic coast. H. Benthami ana is plentiful also along the river Tocantins, south of the mouth of the Amazon.
In Colombia there are three rubber-produc ing districts: in the southern and southeastern section, along the valleys of the Japura and Ica rivers and the Rio Negro; in the north, along the valleys of the river Magdalena and its tributaries; and in the northwest in the basin of the river Atrato. The production is chiefly acaucho,n although a considerable quan tity of fine Para is collected from the ablack of that section. Heretofore the system of collecting the ecauchos sap has included cutting down the •trees, so that the productive area has been constantly diminished. This prac tice is being regulated by law, and eventually will be abolished. The exports of all rubber from Colombia in 1915 amounted to a value of $91,953.