I'm assuming this is an empidid fly due to the following features: long proboscis; "hump-backed" thorax; 3rd segment of antennae bulb-based, elongate, and tipped with a long thin terminal style; very "spherical" head with thin "neck" and distinctive hemispheric red eyes (a male?) and maxillary palps (the two short protuberances between the eyes and the base of the proboscis); the elongate, tapered-cylindrical abdomen; and the general black coloration. It looks similar to this post by Ken-ichi Ueda from (not so far off) coastal Sonoma County...though the antennal styles in the fly here seem longer and thinner to me. I'd love to be able to get to genus. Tried using the old 1934 tome by Curran(1)...but wasn't satisfied with my results. Any help, comments, or corrections will be much appreciated, as always.

This fly was photographed perched on a manzanita flower (Arctostaphylos sp.). The series of images here shows the empidid "robbing nectar" through a hole at the base of the urn-shaped corolla. "Nectar robbery" refers to animals that obtain nectar through a hole pierced in the corolla, and thus bypass "going through the normal channels" which presumably arose via co-evolution between plant and pollinator(s) to bring about effective pollination. Inspection of manzanita inflorescences will often yield a number of punctured corollas, indicating widespread nectar robbery.

I'm speculating that this is an instance of "secondary nectar robbery" as briefly described on pg 171 of "The Natural History of Pollination"(2) and discussed more completely in this journal article. Basically, "secondary" here refers to a nectar robber that uses a pre-existing hole, e.g. one made previously by something like a short-tongued bumblebee with mandibles strong enough to tear open the base of the corolla. (There were a number of male bumble-bees energetically patrolling the gloriously flowering manzanita shrub in these images.) I'm assuming the relatively delicate proboscis of this dance fly wouldn't be up to the task of puncturing the corollas here...but I've been mistaken before :-)