26 CENTRAL AMERICAN RUBBER TREE. 



thieves who let the milk run down into a hole which they dig* in the 

 ground, thus allowing it to take up earth, stones, rotten wood, dead 

 leaves, and other impurities. The planting of rubber trees as a regu- 

 lar industry is just beginning, but plans are being made for several 

 large estates. Trees are said to have grown 21 feet in two years, and 

 one tree (age unknown) is said to have produced 6 pounds of rubber. 

 It is believed in this part of Guatemala that young trees do better when 

 planted in the shade. The dry season in the Pacific coast districts is 

 long and rather severe. 



About Panzos and elsewhere in eastern Guatemala, Castilla had only 

 begun to blossom at the time of our visit (April 4), but about two 

 weeks later (April 20), on the way down from Guatemala Cit} T to 

 the tropical belt of the west coast, the fruit of Castilla was found to 

 be already ripe. The first trees were seen along the railroad below 

 Escuintla, and others were found at San Jose. This would seem to 

 indicate a much earlier season for rubber on the west side. Perhaps 

 Castilla begins to form its blossoms at the ODening' of the dry season, 

 which comes earlier on the Pacific side. 



A few miles below Escuintla is a small orchard of rubber trees 

 which must be at least ten or twelve years old. They had been tapped, 

 but not extensively, and the experiment had apparently not been 

 deemed promising enough to call for extension. A few Castilla trees 

 were seen along the railroad between Ocos and A} T utla, and they 

 became more frequent after crossing the Mexican border on the road 

 between the last-named town and Tapachula. In the vicinity- of Tap- 

 achula and again between Tapachula and La Zacualpa wild Castilla is 

 very frequent in favorable situations in all uncleared tracts. In the 

 forests about La Zacualpa, Castilla is a very common tree. 



Along the road between Tapachula and the port of San Benito many 

 Castilla trees have been planted, and in one place there is quite a long 

 avenue, the rubber trees standing in a hedge of sour orange. These 

 trees, although large and vigorous, are said to } r ield little or no rub- 

 ber. The local explanation is that the soil of this neighborhood is a 

 sort of hardpan not suited to rubber production. 



Wild trees of Castilla were occasionally seen in the rather open for- 

 ests near the coast, about San Benito, where palms of the genera 

 lnodes and Attalea are also abundant. The rubber trees were small 

 and rather spindling. One tree 4 inches in diameter u bled" freely 

 when cut with a pocketknife. There had been rain the previous 

 night. This coast region may have been originally forested with 

 dense tropical growth, but this is probably not the case with the 

 lnodes and Attalea districts. Some desert plants, such as wild cres- 

 centias, are found in the open places, but the woods are often thick. 

 The land is very level, the upper la} T ers probabty washed down from 

 the interior mountains and deposited on the beach sand. 



