58 Annals of the Carnegie Museum. 



to assign exact altitudinal limits to these belts, which are of course 

 primarily dependent upon the conditions in question. On the north 

 slope of the Sierra Nevada, for instance, the several zones overlap to 

 an extent unknown in the Andes, all the zones above the Tropical 

 dropping- down considerably below the usual limits, so far as their 

 lower borders are concerned. Reserving this and other features of the 

 case for fuller discussion further on, we would make the following 

 zonal divisions in the Santa Marta region : 



Tropical Zone o to ' 4,500 feet. 



Subtropical Zone 4,Soo to 9,000 feet. 



Temperate Zone . . 9,000 to 11,000 feet. 



Paramo Zone 11,000 feet to the snow-line (15,000 feet). 



These figures are general and average only; they require modifica- 

 tion to allow for local conditions. 



By far the larger number of species are confined to one zone, and 

 few are known to range beyond two zones. Allowance must neces- 

 sarily be made for the cases where the forms normally belonging to 

 'a given zone invade the limits of the adjacent zone and mingle with 

 its constituents. The various zones are of course determined, not by 

 any arbitrary altitudinal limits, but by their several characteristic ele- 

 ments or species; if the ranges of these are not sharply defined, then 

 the delimitation of the zones becomes indefinite to that extent. There 

 is another phase of the distribution of bird-life in the tropics which 

 must be taken into consideration in referring the various species to 

 their respective zones, especially in the case of the three upper zones. 

 This is altitudinal migration in search of food, or a natural tendency 

 to wander about when not confined to one place by the rearing of , 

 young. In the lists which follow the effort has been, so far as our 

 present knowledge and experience permits, to make allowance for 

 these movements and to relegate each species to the zone in which it 

 breeds, for this is its real habitat. With some species this is compara- 

 tively easy, as with the terrestrial or semi-terrestrial dwellers in the 

 forest, for these are almost entirely sedentary in their habits, but there 

 are others, especially among the hummingbirds, fruit-eaters, and the 

 dwellers in the tree-tops, which present a more puzzling problem. To 

 solve this question accurately -would necessitate continuous collecting 

 throughout the year in all the different zones, a matter quite impossible 



