412 Philippine Journal of Science 1920 
- Most of the endemic species are confined to single islands or 
to groups of islands, while the genera to which they belong are 
represented in other islands by other endemic species. 
Walden ™* long ago remarked that— 
As might be anticipated from analogy with other isolated areas, some 
of the Philippine Islands, although only separated by narrow seas, possess 
species peculiar to themselves. Although well defined, these are strictly 
representative forms. 
Steere has expressed this phase of distribution in the follow- 
ing words: 
The law of distribution of non-migratory land-birds of the Philippines 
may be stated as follows:—Every genus is represented by only a single 
species in one place. Or, in more general terms, as follows:—No two 
species structurally adapted to the same conditions will occupy the same 
area. [Italics are given as in the original.]™ 
The first statement of Steere’s law seems rather unfortunate, 
for “place” is a word of too general meaning. It is hard to un- 
derstand why he did not use “island,” which seems from the 
context to be what he meant. If the law were restated with 
more precise words it would read: “(No genus is represented by 
more than one species in any island.” Asa matter of fact there 
are many genera of which this is not true. This was recognized 
by Steere, and he makes the following statement: 
In 17 genera and 74 species each genus is represented in the islands 
by several species, two or more of which may be found inhabiting the 
same island; but the species thus found together, with the same generic 
name, differ greatly in size, colouring, or other characteristics, and belong 
to different natural sections or subgenera. 
Authors have already attempted in several cases to raise the natural 
sections of these genera [Steere’s list D] to generic rank. 
By “other characteristics” Steere seems to have intended food, 
kind of country inhabited, and whether a species is social or soli- 
tary. Although these characteristics, as well as size and color- 
ing, may be made use of in generic descriptions they are not. 
considered to be valid generic characters. 
Steere admits that the two species of Philippine bee eaters 
“probably exist together on every island of the group.” He then 
says that they differ in habits and food; therefore they belong 
to different subgenera; hence, they are not an exception to 
“Trans. Zool. Soc. London 9? (1875) 131. ? 
“Ibis VI 6 (1894) 419. In the Auk 11 (1894) 239, Steere gave his 
law in a slightly different form, as follows: “The genus is represented by 
but a single species in a place.” and “No two species near enough alike 
structurally to be adapted to the same conditions will occupy the same area. 
