18 CENTRAL AMERICAN RUBBER TREE. 



this Department from Mr. Hemy E. Neibert, an American teacher 

 stationed at Jaro, on the island of Leyte: 



There is a rubber plant indigenous to the soil here, the native name of which in 

 the Binasaya dialect is quiliquili (pronounced ke-li-ke-li) . Neither the natives nor 

 the Spaniards seem to be aware of its commercial value, and have cut all easily accessi- 

 ble specimens at an early age for the columns which support their houses. Because 

 it is a prolific plant is the only reason that it still exists. 



It is not known that either Castilla or Hevea has been introduced 

 into the Philippines. Seeds for experimental planting can be secured 

 more easily from Ceylon or the Straits Settlements than from tropical 

 America, but, as soon as the superiority of any one or more of the 

 Mexican or Central American varieties of Castilla has been deter- 

 mined, a new supply of seed should be sent out. The suitability of 

 some of the various soils and climatic conditions found in the Philip- 

 pines for the culture of Castilla is to be expected. The character of 

 the natural vegetation would enable one conversant with the subject 

 to select the most favorable localities for experimental plantings, but 

 until these have given evidence of success extensive undertakings will 

 not be justified. 



BOTANICAL STUDY OF CASTILLA. 



DIFFICULTIES IN STUDYING TROPICAL TREES. 



On account of their larger size it might be supposed that trees would 

 be the easiest of all objects of botanical stud} T , but this is far from 

 being the fact. The size itself makes it difficult to observe a tree as 

 a whole or to bring numerous individuals under the eye at once, as 

 may be done with smaller plants. Moreover, trees can not be pre- 

 served as complete specimens, and only small fragments can be accom- 

 modated for read}^ reference in the herbarium. Nevertheless, the task 

 may be said to be comparatively simple with the trees of temperate 

 climates, where forests are relatively open and are f requently composed 

 of only a few kinds of trees, or perhaps a single species. In the 

 Tropics a natural forest of one species is practicall} T unknown; hun- 

 dreds of kinds grow indiscriminate^ mixed together. Crowded 

 together in tropical forests trees have nothing like the shapes or habits 

 they would assume if standing alone. All are putting forth, as it 

 were, their best efforts to grow tall and thus secure as much sunlight 

 as possible. Their leaves and branches are inextricably confused, 

 interlaced with climbing plants, and encumbered with parasites and 

 epiph} T tes. To cut down a particular tree may be impossible unless 

 one is willing to clear a large neighboring area to permit it to fall. 

 Unless the botanist finds a clearing, his opportunities for securing 

 even the desired fragments of branches with leaves, flowers, and fruit 

 may be extremely few, hence our knowledge of tropical trees is still 

 in the early pioneer stages. Very recently the rubber tree of the Para 



