Following Cannings and Angell (2001), we recognize subspecies macfarlanei Brewster 1891, with range "southern British Columbia south to eastern Oregon, Idaho, and western Montana". -- Clements Checklist v2021
I'm little bit disappointed macfarlanei was lumped in the first place (old lump: https://www.inaturalist.org/taxon_changes/59252) because its range is not concordant with bendirei, nor is that type of range consistent with typical biogeographic variation. Even more evidence why this change is laughable is because of Proudfoot (2007), who used mtDNA markers to establish criteria for subspeciation. Guess what, he found 7 populations, concordant with that of the 9 named subspecies. I'll give you a hint which subspecies did not reach the bar... bendirei, and Proudfoot's suggestion was to lump it with nominate kennicottii. In a nutshell, that means macfarlanei is distinct from the Pacific population and the subspecies Clements (2018) lumped it with, is not even real!
Clements, J. F., T. S. Schulenberg, M. J. Iliff, S. M. Billerman, T. A. Fredericks, J. A. Gerbracht, D. Lepage, B. L. Sullivan, and C. L. Wood. 2021. The eBird/Clements checklist of Birds of the World: v2021. Downloaded from https://www.birds.cornell.edu/clementschecklist/download/ (連結)
They were a mix; mostly records from the desert south interior of British Columbia (macfarlanei), along a couple from coastal California (bendirei) to correct mis-IDs (one, two). I'll correct them when the taxon change is committed.
@cgbc You're the top identifier (8 obs), do you know which population they came from?