134 



I showed that in Porto Rico a common tree of this genus (C. 

 peltata), though fully equipped with the so-called myrmecophilous 

 adaptations {i.e. the hollow stems, internodal prostomia, and 

 Müllerianbodies),is never tenanted byAztecas, because there are no 

 Aztecas on the island, nor, in fact, on any of the larger Antilles.^ 

 It is also known that some of the Brazilian Cecropias are free 

 from these ants. Several species of Cecropia {C. humholdtiana 

 Klotzsch, insignis Liebm., ohtusifolia Bert., polyphlebia Don. 

 Smith, mexicana Hemsl., and its variety macrostachya Don. Sm.) 

 occur throughout Central America up to altitudes of about 

 3,500 ft. Some of them are really tree-weeds, which spring up 

 in great numbers, like their herbaceous allies, our northern nettles, 

 in all clearings. After examining hundreds of young and old trees 

 I have come to essentially the same conclusions as von Ihering 

 and Fiebrig. As young trees 3 to 6 ft. in height, the Cecropias 

 contain only isolated queen Aztecas in their hollow internodes, 

 just as the young acacias contain only isolated Pseudomyrma 

 queens in their thorns, and are therefore quite as defenceless 

 as other plants which have no thorns or stinging hairs. Never- 

 theless these young Cecropias are very rarely attacked by the 

 leaf-cutters, and their foHage is really in much better condition 

 than that of the older trees, in which the queens have produced 

 hordes of belligerent workers. The foliage of such old trees is 

 often much eaten by sloths, caterpillars, and Chrysomelid beetles 

 or their larvse, as has also been noticed by the observers in South 

 America. On one occasion at Gatun, in the Canal Zone, I saw 

 a female sloth, with a young one on her breast, leisurely devour- 

 ing the large palmate leaves of a tall, ant-infested Cecropia. 

 The foliage of this tree seems, in fact, to be the favourite food 

 of these extraordinary mammals. Fiebrig has also described 

 complete defoliation of Cecropias in Paraguay by grasshoppers. 

 These few notes will suíñce to show that the famous case of these 

 plants is no more clearly established in favour of the Belt-Delpino 

 hypothesis than is that of the acacias. 



^ Recently I have made the same observation on the Cecropias which 

 abound in the hilly regions of the provinces of Havana, Matanzas, and 

 Santa Clara, Cuba. These trees, though closely resembling the conti- 

 nental species of the genus in the structure of the internodes, trichilia 

 etc., have no ant inhabitants. 



