Sept.,1914 STATUS OF CERTAIN ISLAND FORMS 017 SALPINCTES 215 culmen. No mainland bird was found with length of culmen equal to the maximu m of San Nicolas Island specimens. Twenty-two examples of pulverius in juvenal plumage are quite indis- tinguishable from young birds from the mainland. There is not the slightest tendency toward the development of any differential features at this stage, such as are so conspicuous in the young Salpinctes guadeloupensis. To sum up, it seems apparent that the only distinctive feature of the San Nicolas Island Rock Wren is the slightly greater average length of culmen. In neither adults nor young are there any characters of color or markings not included in the range of variation found in the mainland bird. It seems advisable to recognize the slight size difference shown in the island series by the use of a separate aame, pulverius, as has been done, but the name should be restricted to the birds from San Nicolas Island. Specimens at hand from others of the Santa Barbara Islands in every respect fall within the range of variation of S. obsoletus obsoletus. In the spring of 1912 Mr. George Willett made a small collection of birds on certain of the islands off the coast of northwestern Lower California. These form part oJ? his collection now on deposit at the Los Angeles Museum of History, Science and .Art. Among the specimens collected on this trip is a single adult Rock Wren from San Martin Island; and it is rather startling to find that this bird is radically different in appearance from the mainland Salpinctes obsoletus, and but slightly distinguished from S. guadloupensis. This wren I propose to call: Salpinctes guadeloupensis proximus, new subspecies San Martin Island Rock Wren ?/'ype.--Adult male; San Martin Island, Lower California; April 10, 1912; collected by George Willerr; original number 1150. Oharacters.--In coloration most nearly like ?. gua?eloupensis gua?eloupensis. Dark brown, as in that race, and with the back rather heavily barred. The most appar- ent co]or difference between the forms is that in proximus the outer webs of the tertials, secondaries, and some of the primaries, are rather conspicuously barred, as fn some ex- amples of obsoletus, while in the five adults at hand from Guadalupe Island. they are almost or quite uniform. As regards measurements, proximus has the long, heavy bill of guadeloupensis. It has not the relatively short wing and tail of the latter race, but in these measurements is more nearly like the mainland form. Remarks.--It is not without reluctance that 'I have decided to attach a name to this supposed island race, for I am aware of the objections that might be made to such a course. San Martin is only about six miles from the main- land. It is of small size, its area comprising but a few square miles, and it is at a comparatively remote distance from Guadalupe. However, similar ap- parent anomalies in distribution are known among other animals of insular distribution on the Pacific coast of North America; and, conceding the pe- culiarities of range and the limited material available, this single specimen still hardly admits of any other treatment. Its characteristics are absolutely unlike S. obsoletus, and point as definitely toward guadeloupensis in affini- ties. This is the more striking in consideration of the uniformly obsoletus- like character of the Rock Wrens of other islands, some near and some re- mote from San Martin. It may be urged that it is sufficient to point out the peculiarities of such a specimen, without attaching a new name to it, but it is doubtful if such procedure emphasizes the case sufficiently. 'In depending upon research and collecting in the future it is far more likely that a definite "type locality"
