386 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1952 



The term "bromeliad malaria" has been coined (Downs and Pitten- 

 drigh, 1946) to cover this situation, and it is hard to see how it can be 

 improved, although the bromeliad is neither disease, carrier, nor host. 

 It is the host of the carrier and thereby adds another set of variables 

 to the study of malaria. 



There are about 1,600 species of Bromeliaceae, all but one being 

 natives of the New World. They are almost exclusively herbs, and 

 the more primitive types are terrestrial and store water only in the tis- 

 sues in the manner with which we are familiar. However, the more 

 advanced types have adapted themselves to life above ground. They 

 are not parasitic but are capable of living independently on bare rocks, 

 trees, and even telegraph wires. This they have accomplished by two 

 types of development. In one the narrow leaves have developed a 

 dense scaly covering that soaks up water like a blotter and then trans- 

 mits it to the tissues within. In the other the broad flat leaves overlap 

 in a watertight spiral forming a tank (pi. 1). According to the spe- 

 cies, the tank may hold a thimbleful to over a gallon of water. 



Like the entomologist, the botanist has the problem of identifying 

 species and determining their habits in order to discover their impor- 

 tance in the spread of malaria. The types of bromeliads without 

 tanks do not enter the question of malaria, but there are several hun- 

 dred species with tanks that must be taken into account. Of these 

 the rarer in any given region may be ignored, as Pittendrigh has shown 

 (1950c) that it is the frequency of a species rather than the size of its 

 tank that determines its importance in the spread of malaria. Other 

 factors are the shape of the tank and the habitat of the species. Tanks 

 with narrow openings harbor few or no mosquito larvae, while those 

 with broad tops hold many (Downs and Pittendrigh, 1946). Both 

 mosquitoes and bromeliads have height ranges in the forest. In order 

 to be effective in malaria spread a bromeliad must have the same 

 height range as some mosquito that is a carrier. Again suspicions 

 must be varified, this time by a careful census of the different species 

 of mosquito larvae found in the bromeliad tanks. Finally, although 

 the bromeliad is not a parasite, it does show a preference in trees, or 

 rather it finds it easier to secure a perch on some species than on 

 others, and thereby adds yet another variable to the over-all problem. 



In fact the whole cycle of malaria is so complex and presents so 

 many vulnerable spots, that the surprising thing is that it survives 

 at all ; and yet it is one of the most successful of all disease organisms. 

 Another surprise is that bromeliad malaria with all its additional 

 complexities was understood almost as early as the commoner type. 



THE DISCOVERY OF BROMELIAD MALARIA 



In 1897 Ronald Ross in India demonstrated that malaria is carried 

 by mosquitoes. In 1898, Adolpho Lutz, who was to gain fame later as 



