158 HISTORY OF THE CACTUS WREN 



as the nests which a colony of Marsh Wrens build in the heart of 

 the swaying' reeds. But it is not a globular mass of material* 

 nor yet a cup; it is like a purse or pouch, and also peculiar in 

 its position ; for such nests are usually pensile. In the present 

 case, the nest resembles a flattened flask more exactly, it is 

 like the nursing-bottle, with which all mothers (and I suspect 

 some fathers) are familiar, and this is laid horizontally, on its 

 flat side, in the crotch of a cactus. It is constructed of grasses 

 and small twigs woven or matted together, and lined with 

 feathers. Including the covered way or neck of the bottle 

 leading to the nest proper, the structure is some ten or twelve 

 inches long, and rather more than half as much in breadth. 

 The bird appears to be an early breeder; Dr. Cooper found it 

 preparing to build nests about San Diego so early as the 26th 

 of February. This, however, may have been somewhat excep- 

 tional; for the nests which the same naturalist actually examined 

 in May contained eggs or newly-fledged young, and must, there- 

 fore, have been constructed in April. The eggs, from four to 

 six in number, and an inch long by two-thirds as much in 

 breadth, are white, but so thickly flecked with small salmon- 

 colored spots that a rich cast of this tint is given to the whole 

 surface. 



The first naturalist to fully identify the species as a bird of 

 the United States was Mr. George N. Lawrence, who examined 

 specimens taken in Texas by Capt. J. P. McCown, in 1851. 

 Soon afterward, it was noticed by Dr. A. L. Heermaun in 

 the vicinity of Guaymas ; and that gentleman's observations 

 upon its habits were published by the Philadelphia Academy, 

 in its "Journal". In 1854, Mr. John Cassiu gave a recogniz- 

 able figure of the species, referring it, as others before him had, 

 to the genus Picolaptes, and consequently placing it on his 

 plate in a climbing attitude, which, however suitable for a Pico- 

 lapteSj is not characteristic of a Campy lorhynchus, as these birds 

 have nothing substantially in common with the scansorial 

 nature of the Creepers and Nuthatches. In the mean time, 

 other observers successively contributed their quota to the 

 general fund of our knowledge, gradually establishing the 

 geographical distribution of the species I have given in a pre- 

 ceding paragraph, and affording further insight into its mode 

 of life. Two of the latest items respecting its distribution, 

 given by Mr. tfenshaw in a work just now issuing from the 

 Government press, are specially interesting, as they carry the 



