Vol. XXI"j Breninger, San Clemente Island Birds. 221 



iqo4 J ' ^^ *■ 



Particular interest attaches itself to many of the land birds. 

 Centuries of isolation has developed habits and features quite dif- 

 ferent from the same species or closely related forms of the main- 

 land. From association with most of the geographical races of 

 Melospiza I have learned to frame Song Sparrows in the same 

 scene with rippling brooks, moist meadows, and tule-bordered 

 lagoons. Over the whole length and breadth of San Clemente 

 Island there is no fresh water, except what may gather after a 

 rainfall in the rock basins at the bottoms of the washes. There is 

 absolutely no swamp ground, yet Song Sparrows are there in 

 thousands, from the shores to the highest point of the island, feed- 

 ing and nesting among the bushes of the hillsides, along with 

 Bell's Sparrow {^Amphispiza belli). On the mainland Bell's Spar- 

 row marks the other extreme, making its home on the dry sage- 

 covered mesas. Another departure is that of the San Clemente 

 Wren {Tkryofnanes lettcophrys)., a numerous bird on the island, 

 where it nests in the holes and crevices of the rocks. I am 

 inclined to believe it also places its nest amid the protective arms 

 of the prickly pear. T. beivickii spilurus and T. b. leucogaster, two 

 closely allied forms of the mainland, both nest in holes in trees. 

 The change is probably due to the conditions, for on most of the 

 island there are no trees. 



The same is true of Carpodacus, the form inhabiting the island 

 being known as Carpodacus frontalis denientce. The sheep-sheds 

 at the ranch were lined with nests of this bird, old and new, and 

 at that early date I took several sets of four and five eggs. There 

 were some nests built among the spiny leaves of the prickly pear, 

 but by far the greater number were built in holes in the rocky 

 wall of the sea. A pair built their nest in the interstices between 

 the sticks of an eagle's nest. There were at the time of my visit 

 no eggs in the finch's nest, though the eagle's nest was tenanted. 

 The question naturally arises, does this species pass back and 

 forth from the mainland to the island ? 



To a bird having the power of flight, as in Carpodacus, this is 

 not at all impossible. On clear days Catalina Island is clearly 

 visible from the mainland, only twenty-five miles away, while the 

 channel between Catalina and San Clemente is but twenty-two 

 miles wide. The House Finch nest built in an eagle's nest, of 



