PITTIER TREATMENT OF THE GENUS CASTILLA. 249 



locality of Motzorongo, whence proceed the samples which are de- 

 scribed here as belonging to the typical Castillo, elastica. Cervantes's 

 critic, J. L. M., also cites the " Departamento de San Bias " as a local- 

 ity where the Castilla trees are especially numerous. If this is the 

 San Bias of Tepic, it is also the northernmost recorded station of the 

 genus on the Pacific coast, but it is by no means certain that the 

 species found there is identical with the one from the opposite coast. 



HISTORY OF SPECIES HITHERTO KNOWN. 



For more than half a century after its establishment by Cervantes 

 the genus Castilla was considered as monotypic, and all specimens 

 met with from Mexico to Peru were referred to his single species C. 

 elastica. Every mention of the rubber trees of the region, also, 

 assumed that there was only the one species. Even to-day, at a time 

 when this genus has acquired a considerable importance as a rubber 

 producer, we find in most treatises and reviews dealing with the eco- 

 nomic features of the rubber question the obsolete notion that one of 

 its main sources is Castilla elastica, from Central and South America. 



In 1851, however, Liebmann published his C. costaricana, founded 

 on incomplete specimens collected by Oersted at Turrialba. Liebmann 

 had familiarized himself with the Mexican species, which he had 

 observed and collected at no less than six different localities in the 

 State of Veracruz," and notwithstanding the few and unsatisfactory 

 differential characters presented by the leaves, he did not hesitate to 

 assign Oersted's material to a distinct species. 



Thus far Castilla elastica had not attracted much attention as an 

 economic tree, but 25 years later rubber exportation from Central 

 America had attained its full swing, and it became more and more 

 urgent to acquire a better knowledge of the sources of supply. In 

 1800 Cavanilles had reported the presence on the Isthmus of Panama 

 of a rubber tree which he identified with Cervantes's species and the 

 existence of which was confirmed in the fifties by Sutton Hayes, 

 botanical explorer of the route between Panama and Colon. In con- 

 nection with this information of Cavanilles and Hayes, which later 

 reports and actual experience have shown to be well founded, there 

 should be mentioned here the singular mistake made by James Collins,*' 

 describing and figuring as " ule-ule," or " Panama rubber," under the 

 name oi^Castilloamai'khamiana" specimens of a Perebea collected by 

 Hayes on the Isthmus. The description, and especially the very good 

 illustration accompanying Collins's report, leave scarcely a doubt as to 

 the identity of the small tree, as has already been acknowledged by 

 Hooker and other authorities. 



"Dansk. Vid. Selsk. Skr. V. 2: 318. 1851. 



'' Report on Caoutchouc of Commerce, p. 12. (1872.) 



