Had never seen anything like this before, so I grabbed some quick shots. But figured it looked too weedy to be native. Right about that: turns out to be an ornamental, native to Argentina, naturalized in the Southwest US.

It was growing roadside in a bleak, infrastructure-type dead end right off the freeway. Don't quite remember exactly where; haven't been able to relocate the spot on Google Maps -- I suspect it's changed too much to be recognizable. But it had to be somewhere around Castaic, because that's where I got off I-5 for the first part of my wildflowering mission: took the Old Ridge Route parallel with I-5 to the next exit, Templin Hwy, & got back on the freeway there.

Couldn't find it -- or anything like it -- in any of my field guides, although I was pretty sure it was in family Fabaceae. Not common -- as of Jan 09, there was only a single herbarium specimen from Los Angeles County. (See ucjeps.berkeley.edu/cgi-bin/get_county_map.pl?taxon_id=16... -- county dots are clickable.) Also explains why I didn't succeed in IDing it via a search of the Jepson Manual for Fabaceae in Western Transverse Ranges (not used to the brute force method failing!). Turns out that as of that writing it had been documented only in the Central Valley: ucjeps.berkeley.edu/cgi-bin/get_JM_treatment.pl?3691,3879.... Finally got it via a Google Images search for "Fabaceae red stamens"!

I remembered learning that there are three very different-looking subfamilies within Fabaceae: the pea-like flowers, the mimosa-like, & an oddball 3rd kind. This is the 3rd kind. But it turns out that modern DNA analysis has shown that it's not actually a distinct subfamily; instead it's an ancestral type: loco.biosci.arizona.edu/astragalus/images/phylogenies/fab...