FIRST ACCOUNT OF CASTILLA. 19 



district has been described and named as a new species distinct from 

 the original Hevea hraziliensis, which came from the upper Orinoco. 



THE ORIGINAL DESCRIPTION OF CASTILLA. 



The rubber tree of Mexico received a botanical description and 

 name in a paper read by Cervantes before the Real Jardin Botanico 

 (Royal Botanic Garden) of the City of Mexico, July 2, 1794, and 

 printed with an engraved plate a month later in the Suplemento a la 

 Gaceta de Literatura. a publication now very rare. According- to 

 Collins the British Museum copy lacks the illustration of the plant, but 

 that of the Library of Congress at Washington is complete, and the 

 figures are shown in photographic reproduction as Plate II of the pres- 

 ent bulletin. An English translation of Cervantes's account of the rub- 

 ber tree was published anonymously in 1805, but is said to have been 

 the work of Charles Koenig, keeper of the mineralogical department 

 of the British Museum. In this the name of the plant was changed 

 from Castilla to Castilloa, an amendment which has become generally 

 current, although justified by no recognized rule of botanical nomen- 

 clature. The tree was named Castilla in honor of Castillo, a Spanish 

 botanist who died in 1793 while engaged in the preparation of a flora 

 of Mexico. To modify Castilla into Castilloa was not the first change 

 suggested, since a Mexican botanist had already proposed the word 

 Castella in the same year in which Castilla was published. The ques- 

 tion is, of course, of the slightest importance, and turns on whether 

 the personal name should be latinized or not in forming from it the 

 name of a plant. Castella and Castilloa would represent extremes of 

 opinion, but few botanists, if any, would hold that Castilla was incor- 

 rect, and fewer still would recognize the right of anybody to change it. 

 It will be apparent from comparison of our illustrations (see Plates 

 IV to VII) that Cervantes's plate looks little enough like our photo- 

 graphs of the flowers and fruit of Castilla. Indeed, it need not be a 

 matter of surprise if it should be found that they were taken from some 

 different tree." though there seems to be none known at present from 

 Mexico from which they seem likely to have been made. The rounded 

 clusters of fruits pointed with long recurved styles have considerable 

 resemblance to those of some of the species of the South American 

 genus Perebea, and the long, loosely-scaled staminate flower is very 

 unlike those of the true rubber tree, though all these discrepancies may 

 be due merely to careless drawing. That Cervantes was not personally 

 acquainted with the rubber tree in nature seems to be indicated by his 

 saying that the tree is "one of the tallest and most leafy which grow 

 on the hot coasts of New Spain," and again that "the trunk is 3 or 1 

 yards in thickness." 



« Further studies show that more than one species of Castilla i< being cultivated 

 in Mexico and Central America, but the detailed results can not lie included in the 

 present report. 



